Natural plasmid transformation in the marine environment

Citation
Natural plasmid transformation in the marine environment

Material Information

Title:
Natural plasmid transformation in the marine environment
Creator:
Frischer, Marc E.
Place of Publication:
Tampa, Florida
Publisher:
University of South Florida
Publication Date:
Language:
English
Physical Description:
xii, 212 leaves : ill. ; 29 cm

Subjects

Subjects / Keywords:
Plasmids ( lcsh )
Genetic transformation ( lcsh )
Marine bacteria ( lcsh )
Dissertations, Academic -- Marine Science -- Doctoral -- USF ( FTS )

Notes

General Note:
Thesis (Ph.D.)--University of South Florida, 1994. Includes bibliographical references (leaves 186-212).

Record Information

Source Institution:
University of South Florida
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
030021416 ( ALEPH )
31202498 ( OCLC )
F51-00187 ( USFLDC DOI )
f51.187 ( USFLDC Handle )

Postcard Information

Format:
Book

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

NATURAL PLASMID TRANSFORMATION IN THE MARINE ENVIRONMENT by MARC E FRISCHER A d i ssertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Marine Science University of South Florida May 1994 Major Professor : John H Paul, Ph.D.

PAGE 2

Graduate School University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Ph.D. Dissertation This i s to certify that the Ph D Dissertat i on of MARC E. FRISCHER with a major in Marine Science has been approved by the Examining Committ ee on December 13, 1993 as satisfactory for the dissertation requirement for the Ph.D degree Examining Committee : Major Professor: John H. Paul Ph D Member : Mvlien Dao, Ph.D Joan B Rose, Ph D. Member= Gregory J. Stewart Ph D Member : Raymond R Wilson Ph.D.

PAGE 3

DEDICATION To the summers in Union Pier and those who shared them

PAGE 4

ACKNOWLEDGMENTS It is wi th deep gratitude that I thank all those who have helped me during my graduate studies and those who have helped me get there First to Dr. John H Paul whose commitment to scientific excellence, no matter how f rustrat i ng made it all worthwh i le I'd also like to thank the rest of my committee for the i r support and guidance A special thanks to Dr. Raymond R. Wilson for allowing me to cultivate my obsess i on with fishing and for putting the "marine" in my marine science degree Also Dr. Joan B Rose, who even makes sewage fun Acknowledgment of Dr. Ro y Curt iss III at Washington University is also necessary for the early insp ira t i on. I would also like to thank all those who had to put up with me on a daily bas i s in the lab Scott "Catfish" Pichard my comrade in arms We shared time ideas, an office and a few beers .. I'll miss you Sunny Jiang Chris Kellogg Alex Miller Haydn Wil liams, Debby Boswell-Lane, the space is all yours (at least until you get to move into the new building). I'll miss you guys I would also like to thank John s students who came before me; Drs Wade Jeffrey and Mary DeFlaun You can't appreciate how comforting it was to know that you guys graduated I'm also indebted to Wade for showing that DI-9 was transformable, even if the rest of the world didn t believe you I did

PAGE 5

Thanks to Lisa Cazares, Jennifer Thurmond, and Debby Boswell-Lane for technical assistance, ordering, lab maintenance and, unfortunately, sometimes for cleaning-up after me. To Cheri Harpole for dishes, East Bedell for secretarial help and looking after me Its good to know you're there even if I'm too busy to stop and chat most of the time. Also to the rest of the Marine Science Department staff who keeps the place running. A hearty thanks to Chad Edmisten for all the graphics and letting me spread out in your space. Thanks to the Crews of the RJV Pelican, RJV Cape Hatteras, the R/V Bellows, and the support staff of the NOAA NURC facility in Key Largo, FL, without whom we could get nothing done in the real world or have visited all the exotic ports of call. To John's uncle Dr. Ralph Slepecky. I'll always remember Bimini, Mongoose ("Maangoos"), and ant tracks on agar plates Many thanks are deserved by those who helped financially Of course to John Pau l who remains committed to keeping his students funded. I know you missed a few golf games to get those proposals in. Thanks to my parents for their understanding and generosity throughout my life Thanks to the Gulf Oceanographic Trust Fund, the Jack B. Lake Fellowship, and the Clearwater Power Squadron for their fellowships Not only did they ease the financial burdens but they gave me confidence to pursue my ideas. No thanks to the US government and their stupid student loan program Get a clue guys, not all full time graduate students are registered for 9 credits a semester. Thanks to my housemates and St. Pete family; Lynn Leonard and Eric Wright. Admit it, you like the Marlin and you threw away my remote control. Finally, and most importantly, I wish to extend my gratitude to my parents and family Without your love, guidance, support, and understanding I wouldn t be who and

PAGE 6

where I am today. I'm proud to be your son (brother) and I hope I've made you proud of me. I love you

PAGE 7

TABLE OF CONTENTS LIST OF TABLES v LIST OF FIGURES vii ABSTRACT v CHAPTER 1 : INTRODUCTION 1 Mechanisms of bacterial gene transfer 6 Gene transfer in the environment 8 Mechanisms of natural transformation 14 Development of competence 15 DNA binding and uptake 19 Recombination and expression of DNA 22 Natural plasm i d transformation 23 Natural transformation in the environment 31 CHAPTER 2 : NATURAL PLASMID TRANSFORMATION IN A HIGH-FREQUENCY -OF TRANSFORMATION MARJNE VIBRIO STRAIN 38 Introduction 38 Methods 40 Bacterial strains 40 Transforming DNA and gene probes 40 Preparation of transforming DNA 42 Culture conditions 44 Filter transformation assay 46 Verification of plasmid transformation 4 7

PAGE 8

Curing of plasmids from natural plasmid transformants 50 DNA binding studies 51 Results 52 Isolation of HFf strains 53 [ 3H]DNA binding studies 62 Discussion 67 CHAPTER 3 : FACTORS AFFECTING COMPETENCE IN A HIGH-FREQUENCY-OF TRANSFORMATION MARINE VIBRiO 71 Introduction 71 TI Strains and plasmids 72 Culture conditions 73 Preparation of plasmid as transforming DNA 73 Transformation assays 75 Results 77 Transforming plasmid DNA 77 Concentration and form of transforming plasmid DNA 79 Time required for transformation 86 Development of competence 88 Maintenance of the competent state 88 Effect of temperature 91 Effect of salinity 91 Effect of nutrients 92 Discussion 92 CHAPTER 4: In situ TRANSFORMATION IN SEAWATER MARINE SEDIMENTS, AND IN A MARINE DEPOSIT FEEDER 100 Introduction 100 Methods 101 Strains and plasmids 101 Culture conditions 102 Transforming DNA and gene probes 103 In situ transformation assays 103 Water column assays 103 ii

PAGE 9

Transformation in sediments 104 In situ sea cucumber assays 105 In situ transformation mating studies 106 In situ matings in water 106 In situ sediment assays 107 In situ sea cucumber assays 109 Disposal of all GEM contaminated material 110 Results 110 In situ transformation assays in water 110 In situ transformation assays in sediment and in sea cucumbers Ill In situ transformation matings in sediments and sea cucumbers 113 Discussion 115 CHAPTER 5 : PLASMID TRANSFER TO INDIGENOUS MARINE BACTERIAL POPULATIONS BY NATURAL TRANSFORMATION 118 Introduction 118 Methods 120 Preparation of transforming DNA and gene probes 120 Purification of transforming plasmid DNA 120 Construction of pQSR50 gene probes 121 Transformation assays 121 Distribution of naturally transformable marine bacteria 124 Transformation of indigenous marine bacterial populations 125 Verification of transformation 126 Restriction analysis 127 PCR amplification of the nptiT gene in plasmid transformants 126 Results 128 Transformation of indigenous marine bacterial populations 128 Distribution of naturally transformable marine bacteria 134 Analysis of plasmid DNA from transformants 134 Restriction analysis 134 Location of transformed DNA 142 PCR amplification 144 lll

PAGE 10

Discussion 144 Significance of natural plasmid transformation in the marine environment 147 CHAPTER 6 : NATURAL TRANSFORMATION AS A POSSIBLE MECHANISM OF PLASMID EVOLUTION 154 Introduction 154 Methods 156 Bacterial strains 156 Preparation of DNA and gene probes 156 Purification of plasmid DNA 156 Construction of pQSR50 gene probes 156 Deletion mapping 158 Results 162 Restriction modification versus genetic rearrangement deletions 162 Methyl modification 162 Deletion of H i ndi.II and Bgni restriction sites 164 Discussion 167 Non-sequence versus sequence modifications 169 The generation of new plasmids by natural transformation 170 CHAPTER 7 : NATURAL PLASMID TRANSFORMATION IN THE MARINE ENVIRONMENT: A DISCUSSION AND SUMMARY 175 LITERATURE CITED 186

PAGE 11

LIST OF TABLES Table 1. Transformation processes of model transformation systems 16 Table 2. Organisms capable of natural transformation 24 Table 3. Bacteria and plasmids 41 Tab le 4. Plasmid curing of transformants 50 Table 5 Natural plasmid transformation of wildtype Vibrio strains 53 Table 6. Natural plasmid transformation of HFT Vibrio strains 61 Table 7 Natural transformation of wild-type and HFT Vibrio strains with homologous chromosomal DNA 62 Table 8 Short-term binding rates of [3H]DNA by wild-type and HFT (WJT-1C) Vibrio strains 66 Table 9. Bacteria and plasmids 74 Tab le 10. Transformation frequency of WJT-1C with various plasmids 78 Tab le 11. Composition of pQSR50 plasmid preparations 81 Table 12. Bacteria and plasmids 102 Table 13. In situ transformation of WJT-lC with pQSR50 multimers at varying cell densities in water 111 Table 14. In situ transformation of WIT -1 C with pQSR50 multimers in sediment and in holothurians 112 Table 15. In situ transformation mating of MFN-1C2 and E coli RM1259(pQSR50) in water 113 v

PAGE 12

Table 16 In situ transformation mating of Vibrio JT-1 and E coli RM1259(pQSR50) in sediments and sea cucumbers 114 Table 17. Natural plasmid transformation of indigenous marine bacterial populations (water column) 129 Table 18 Natural plasmid transformation of indigenous marine bacterial populations (sediment) 130 Table 19. Natural plasmid transformation of indigenous marine bacterial populations (invertebrates) 131 Table 20. Distribution of marine bacteria which can be transformed by plasmid DNA 135 Table 21. Transformation of marine and subsoil isolates 136 Table 22. Estimates of natural transformation and conjugation kinetic transfer coefficients and formation rates from filter assays 152 Table 23. Bacterial strains 157 Table 24. PCR primers 161 Table 25. PCR amplification of Tn5 from pQSR50 and pQSR50 transformants 168

PAGE 13

Discussion 144 Significance of natural plasmid transformation in the marine environment 147 CHAPTER 6: NATURAL TRANSFORMATION AS A POSSIBLE MECHANISM OF PLASMID EVOLUTION 154 Introduction 154 Methods 156 Bacterial strains 156 Preparation of DNA and gene probes 156 Purification of plasmid DNA 156 Construction of pQSR50 gene probes 156 Deletion mapping 158 Results 162 Restriction modification versus genetic rearrangement deletions 162 Methyl modification 162 Deletion of Hindlll and Bglii restriction s it es 164 Discussion 167 Non-sequence versus sequence modifications 169 The generation of new plasmids by natural transformation 170 CHAPTER 7: NATURAL PLASMID TRANSFORMATION IN THE MARINE ENVIRONMENT: A DISCUSSION AND SUMMARY 175 LITERATURE CITED 186

PAGE 14

LIST OF TABLES Table 1. Transformation processes of model transformation systems 16 Tab l e 2 Organisms capable of natural transformation 24 Table 3 Bacteria and plasmids 41 Table 4 Plasmid curing of transformants 50 Tabl e 5. Natural plasm id transformation of wildtype Vibrio strains 53 Tab l e 6 Nat u ral plasm i d transforma t ion of HFT Vibrio strains 61 Tab l e 7. N a tural transformation of wild type and HFT Vibrio s t rains wi t h homologous chromosomal DNA 62 Ta ble 8 Short-term b i nding rates of [3H]DNA by wild-type and HFT (WIT -lC) Vibrio strains 66 Table 9 B ac teria and plasmids 74 Table 10. Transformation frequency of WJT -1 C with various plasm ids 78 T a ble 11. Compo s it i on of pQSRSO plasmid preparations 81 Tab le 12 Bacteria and plasmids 102 Tab l e 13. In situ transformation of WIT -1 C with pQSR50 mu1timers at varying cell densities in water 111 Tab le 14 In situ transformation of WIT -1 C with pQSRSO multimers in sediment and in holothurians 112 Table 15. In s it u transformation ma ti ng of MFN-1 C2 and E. coli RM 1259(pQSR50) in water 113 v

PAGE 15

Table 16 In situ transformation mating of Vibrio IT-1 and E coli RM1259(pQSR50) in sediments and sea cucumbers 114 Table 17 Natural plasmid transformation of indigenous marine bacterial populations (water column) 129 Table 18. Natural plasmid transformation of indigenous marine bacterial populations (sediment) 130 Table 19. Natural plasmid transformation of indigenous marine bacterial populations (invertebrates) 131 Table 20. Distribution of marine bacteria which can be transformed by plasmid DNA 135 Table 21. Transformation of marine and subsoil isolates 136 Tab le 22. Estimates of natural transformation and conjugation kinetic transfer coefficients and formation rates from filter assays 152 Table 23 B acte rial strains 157 Tab le 24 PCR primers 161 Tabl e 25. PCR ampl i ficat i on of Tn5 from pQSR50 and pQSR50 transformants 168

PAGE 16

LIST OF FIGURES Figure 1. Genealogy of plasmids and Riboprobe transcription vectors used in this study 43 Figure 2 Preparation of pGQ3 plasmid multimers 45 Figure 3. Comparison of colony lift hybridization methods 49 Figure 4 Colony morphology of Vibrio strain DI-9, WJT-1 and WJT-lC 54 Figure 5 Scanning electron photomicrographs of Vibrio strain DI-9 and WJT-lC 55 Figure 6. Growth curves of parental Vibrio strain DI-9 and HFT Vibrio strain WJT-IC 56 Figure 7. Direct identification of transformants by colony hybridization 58 Figure 8. Autoradiogram of Southern transfer of strain DI-9, transformant strain WJT-1, and HFT strain WJT-lC 59 Figure 9. Dot blot of total DNA from parental, transformant, and HFT Vibrio strains 60 Figure 10 Typical DNA/binding uptake curves of (3H] lambda DNA by parental Vibrio strain DI-9 and HFT strain WJT-lC 63 Figure 11. DNA binding/uptake of [3H] chromosomal and pQSR50 plasmid DNA by Vibrio strain DI-9, HFT strain WJT-lC, and by an unknown contaminant strain 65 Figure 12. Autoradiograph of a Southern transfer of pQSR50 plasmid preparations hybridized with [35S]labelled RNA probe NPTII 80 Vll

PAGE 17

Figure 13. Transformation frequency of WIT -1 C as a function of amount of plasmid DNA per filter 82 Figure 14. Transformation frequency of WIT-1C as a function of oligomeric plasmid DNA 84 Figure 15. Transformation of Vibrio strain WJT-1C as a function of length of time of exposure to pQSR50 DNA 87 Figure 16. Development of competence in Vibrio strain WIT -1 C at 29 C and 37C. 89 Figure 17. Maintenance of competence in Vibrio strain WIT -1 C in artificial seawater with nutrients and in artificial seawater without nutrients 90 Figure 18. Effect of temperature, salinity, and nutrient concentration on the competence of Vibrio strain WIT-1C 93 Figure 19. Schematic representation of in situ transformation mating experiment conducted in a Bahamas coral reef environment 108 Figure 20. Restriction map of pQSR50 123 Figure 21. Transformation experiment with a concentrated mixed microbial population 132 Figure 22. Transformation experiment with a concentrated mixed microbial population 133 Figure 23. Autoradiograph of Hindiii digestions of pQSR50 transformants T1-T9 and two kanamycin and streptomycin resistant isolates from a calf thymus control 137 Figure 24. Autoradiographs of restriction digestions of purified pQSR50 plasmid DNA from pQSR50 transformants 138 Figure 25. Hoechst 33258 gel of restriction digestions of purified pQSR50 DNA and plasmid DNA from pQSR50 transformant MG-1 140 viii

PAGE 18

Figure 26. Autoradiographs of restriction digestions of pQSR50 and pQSR50 transformants 141 Figure 27. Autoradiograph of a dot blot of purified pQSR50, total genomic DNA preparations of pQSR50 transformant T9, and purified plasmid DNA from T9 143 Figure 28. Autoradiograph of PCR products of nptii coding region from purified pQSR50 and plasmid DNA from pQSR50 transformants 145 Figure 29. Ideal transformation curve 150 Figure 30. Restriction maps of pQSR50, Tn5, and Left Inverted Repeat region of Tn5 160 Figure 31. Autoradiograph of undigested, Dpnl-digested, and Mbol-digested pQSR50 (purified from E. call) and plasmid DNA from pQSR50 transformant strains BS-10, SJl-T, T9 and 14T 163 Figure 32. Undigested and Mboi-digestions of the 762 bp PCR product of the nptii coding region from pQSR50 and transformant strains SJlT and BS-10 165 Figure 33. Autoradiographs of dot blots of pQSR50 transformants 166 Fi g ure 34 Proposed mechanisms for the generation of new plasmids during na t ural plasmid transformation 173

PAGE 19

NATURAL PLASMID TRANSFORMATION IN THE MARINE ENVIRONMENT by MARC E FRISCHER An Abstract Of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Marine Science University of South Florida May 1994 Major Professor : John H. Paul, Ph.D X

PAGE 20

The flow of genetic information through natural microbial communities is a fundamental microbial process, yet little is known concerning bacterial gene flow in nature Natural transformation is the mechanism of procaryotic gene transfer that involves the uptake and integration of genetic information encoded in extracellular DNA and may be a mechanism for the horizontal transfer of non-conjugative plasmids in the marine environment The objectives of this study were three-fold. First, to develop a marine model system for the study of natural plasmid transformation in marine environments and to compare this system to other transformation systems which have been examined in detail. Second, to determine whether marine environments favor transformation, and third; to demonstrate the movement of plasmid-encoded genes to indigenous marine microbial populations by natural plasmid transformation. The estuarine bacterium Vibrio sp., strain DI-9, had been previously shown to be naturally transformable with both broad host range plasmid and homologous chromosomal DNA Growth of plasmid transformants of DI-9 in nonselective medium resulted in cured strains which transformed at significantly higher efficiencies than the parental wildtype strain. These strains were termed high frequency of transformation strains (HFT). The origin of the HFT variant strains is believed to be genetic rather than physiological, since the HFT phenotype was stable and heritable. A transformation filter assay was optimized for the detection of natural plasmid transformation. This assay was used to measure the Xl

PAGE 21

effects of nutrients, salinity, temperature, and the development and maintenance of competence in the HFT strains. Integration of the results of these studies suggest that conditions typical of subtropical marine systems favor the transfer of plasmids by natural transformation. Furthermore, the transformation system of the Vibrio HFT strains was determined to be distinct from other transformation systems, particularly in regards to the development and maintenance of competence The optimized transformation filter assay was also used to document the transfer of plasmid encoded-genes to other marine bacterial isolates (3 of 30) and to mixed microbial populations from a variety of marine environments (5 of 14). In several cases acquisition of the plasmid by marine bacteria was accompanied by some modification of the plasmid that altered its restriction enzyme digestion pattern Estimation of transformation rates in estuarine environments ranged from 5 x 104 to 1.5 transformants/liter per day. Extrapolation of these rates to ecosystem scales suggest that natural transformation may be an important mechanism for plasmid transfer in marine bacterial communities. Abstract Approved: __________ Major Professor: John H Paul, Ph.D. Professor, Department of Marine Science Date Approved: I 1:7\q Z; c xu

PAGE 22

1 CHAPTER ONE: INTRODUCTION Much is known about bacterial genetics at the physiological, biochemical and molecular level. Likewise, the field of microbial ecology has developed rapidly. However, only recently have studies begun to link these two fields; bacterial genetics in the envi ronment. One fundamental aspect of environmental bacterial genetics is bacterial gene ex c hange (prokaryotic sex) has been the subject of intense investigation This has been due primarily to the perceived risk of the release intentional or accidental, of genetical l y-eng i neered microorganisms (GEM's) into the environment. A chief concern is wh e ther recombinant sequences can enter natural microbial populations via some mechanism (s) of prokaryoti c gene transfer. Therefore, studies designed to determine the po t ent ial for ge n e exchange in natural environments as well as the mechanisms of inter s pec ific gene.exchange are particularly relevant. The majority of gene transfer studies concerning marine environments and organisms have been conducted from the perspective of transfer, and risks of terrestrial releases on marine microbial communities (Sizemore and Colwell, 1977 ; Stewart and Koditscheck 1980 ; Gauthier er al. 1985; Sorensen, 1988; Gautier, 1990) Few studies have examined the role that interand intrageneric gene transfer between marine microorganisms may play in the development, maintenance, adaptive potential, and ecology of marine microbial communities Such studies have been hampered by the lack

PAGE 23

2 of marine model systems and the difficulty of working in the marine envi ronment. This is evidenced by the reliance on non-marine organisms (e. g terrestrial Pseudomonads, Bac illus subtilis and Escherichia. coil) in the majority of gene transfer studies conducted in marine environments (Gauthier and Breittmayer, 1990). Therefore, a goal of this project has been the development of marine gene transfer systems suitable for studying the pote n t ial and role of gene transfer among marine microb ial communities, and for studying the mechanisms of gene transfer employed by marine microorganisms. Altho ugh re latively few gene transfer studies have been done in marine systems gen e t r ansf e r s tudies hav e been conducted in many other types of environments and with a lar g e v ariet y of organisms Gene transfer has been examined in soil and freshwater env ironments the root and rhizome environment in animal systems in liquid, on s u rfaces, and in wastewa t er and sewage treatment facilities. Howe v er, the paucity of gene t rans f e r studie s in the marine environment has not inhibited discussions concern ing the s ignificanc e that gen e transfer event s may have in regard to marine microbial communiti es. Potentially, gene transfer affects four aspects of marine mic robial communities: ecology, epidemiology of microbial pathogens risks associated with the use and release of GEM's and the evolution of marine bacterial populations The organization and maintenance of community diversity is an important aspect of mic r o bial ecology Although we are only beginning to realize the diversity of marine microbial commun ities, (using newly-developed technologies which are not dependant on cultivation) it is l ikel y that gene exchange has been important in maintaining community diver s ity. This is probably most true during times of rapid env ironmental change or

PAGE 24

3 stress (Smith eta/. 1991) Several studies have indicated the potential for intergeneric genetic exchange (Levy and Miller, 1989; Thomas, 1989; Fry and Day, 1990) Others have examined the genetic structure of genes found in diverse populations (e g streptomycin resistance) and have concluded that genes have been transferred between microbial populations (Norelli et a/. 1991). These observations suggest that gene transfer events have allowed the maintenance of community diversity by allowing useful genes to be passed between diverse populations. Thus, the sharing of genes between microbial populations alleviates the necessity of a particular gene evolving independently among diverse members of a microbial community Gene transfer mechanisms may provide a plausible explanation for the rapid adaptat i on of microbial populations to stress imposed by pollutant introductions, heavy metal s, or xenob i ot i cs which appear to occur more rapidly than would be expected from rand o m mutat i onal events alone (Amabile-Cuevas and Chicurel, 1992). However, not all genes exhibit the same degree of promiscuity. In general, genes that code for accessory functions (e g. antibiotic resistance) appear to be considerably more mobile than genes coding for essential functions such as those involved in different respiratory pathways (e g nitrification, sulfur reduction, or methanogenesis). Indeed, genealogically tight groupings, made on the basis of respiratory pathways, seem to exist and suggest the existence of genetic barriers that have prevented widespread genetic recombination between distantly related organisms (Smith et al 1991) The epidemiology of microbial pathogens is another area in which gene transfer is believed to have played an important role. Perhaps the most notorious example is the

PAGE 25

4 transfer of antibiotic resistances between clinical pathogens after the widespread introduction of antibiotics in medical practice. Transfer of resistance factors (later determined to be genes encoded on conjugative plasmids and transposons) by conjugation have been determined to be the cause of this phenomenon (Akiba er al., 1960; de Graaff er a/., 1976; Elwell er al., 1977; Roberts 1989). Plasmids carrying antibiotic resistances as well as virulence genes are common in marine bacteria. For example, Vibriosis in salmonid fish attributed to the pathogen Vibrio anguillarum, has been shown to be caused by a plasmid-encoded siderophore, a key component in the iron scavenging system responsible for the pathology of this disease (Crosa J H 1980) Furthermore, plasmid-conferred pathogenicity in Vibrio ordalii is also believed to be caused by the same or similar plasmid It is believed that this plasm i d was transferred between V anguillarum and V. ordalii but the mechanism of transmission remains unknown Over the past decade 267 serotype 01 V. anguillarum isolates from diseased fish around the world have been described; 260 carried a virulence plasmid that was similar or identical to the V anguillarum plasmid pJM1, and only nine of the se isolates carried additional plasmids (Larsen et al. 1988 ; Lemos er al., 1988; Tolmasky er al., 1988; Wiik et al., 1989; Bolinches et al., 1990; Conchas et al., 1991 ; Myhr er al ., 1991 ; Singer et al 1992) Conjugal plasmids encoding for drug resistance have been found in other bacterial fish pathogens (foranzo, 1983; Toranzo 1984). As stated earlier, the risks associated with the release of geneticallyengineered microorganisms has been an area of active investigation One of the concerns is that the introduction of exotic genetic sequences from engineered organisms will be transferred

PAGE 26

5 and incorporated into the genomes of indigenous microbial populations, thereby permanently altering the genetic composition of these communities. Unlike the addition of inert materials, or even of nutrients and living bacteria, the addition of new genetic information to a community would be self-propagating, and once incorporated, would be extremely difficult if not impossible to remove. To address these concerns, and to determine the best strategies for use of GEM's in the environment, it is of importance that we understand the mechanisms of gene transfer, the potential for gene transfer, and the genetic composition of specific microbial communities at risk of interacting with GEM's The goal of determining risks has been the greatest impetus for funding of environmental gene transfer studies in recent years. Whether or not bacterial gene transfer poses a significant risk to natural ecosystems remains an unknown; however the percei ved threat of introducing genetically-engineered organisms in natural settings has, at least contributed greatly to the expansion of our understanding of bacterial gene transfer in the environment. A fourth area of impact is that of the role that gene transfer has played in bacterial evolution. The impact of bacterial gene transfer on the evolution of bacteria is not known. However, given the ubiquity of the biochemical systems for reorganizing bacterial genomes, repeated sequences, transposable elements, plasmids, phages recombination enzymes, and gene transfer mechanisms, it should be apparent that these systems have played a major role in shaping the genetic information that permits bacteria to thrive and carry out so many functions in different ecosystems Given these considerations perhaps bacterial evolution should be understood in terms of information

PAGE 27

6 exchange among many cell types rather than in terms of intracellular modification within isolated lineages In other words, it may make more sense to think of the development of new bacterial abilities as a multicellular phenomenon and of the changing unit as the entire bacterial community rather than the isolated cell and its clonal descendants. Mechanisms of bacterial gene transfer Three different mechanisms of bacterial genetic exchange have been recognized, distinguished by the etiological agent of DNA transfer: conjugation transduction, and transformation. Conjugation was first described by Lederburg and Tatum (1946), and invol v es the transfer of plasmids from donor to host during direct cell-to-cell contact. Conjugation is mediated by plasmid encoded genes, or in the case of conjugative transposons on transposons. Because conjugation primarily involves plasmids, genes that are mob i lized by conjugation generally encode accessory traits which provide, for example resistance to antibiotics, or allow expanded metabolic capabilities. Plasmids that contain information required for self mobilization and transfer are termed conjugative plasmids. Several conjugation systems are described in: Willetts and Wilkins, 1984; Hopwood et al 1984; Bradley, 1985; Clewell and Gawron-Burke, 1986; Ippen-Ihlerand Minkley, 1986 ; Schaberg and Zervos, 1986; Stachel and Zambryski, 1986; Clewell et al., 1987; Lyon and Skurray, 1987; Odelson et al ., 1987; Paranchych and Frost, 1987; Willetts and Skurray, 1980, 1987; Ippen-Ihler, 1989; Mcintire, 1989; Wilkins, 1990. Transduction is the transfer of genetic information between bacterial cells

PAGE 28

7 mediated by bacteriophage (Masters, 1985). As in conjugation, transduction is mediated thmugh an extrachromosomal element. Transduction was flrst recognized by Zinder and Lederburg (1952) as a means of genetic exchange that occurred in the absence of cell-to cell contact. This exchange was mediated by a filterable agent later shown to be a bacteriophage (Margolin, 1987). Transduction appears to be a consequence of the accidental packaging of bacterial host genetic material into bacteriophage particles during the lytic phase of a phage infection. During the transduction process bacterial genes (chromosome or plasmid-encoded) can become incorporated into a phage particle which, when infecting another bacterial host, affects a bacterial gene transfer. Bacteriophage which have incorporated bacterial genetic material are termed transducing phages. Two types of transduction systems have been characterized; specialized and generalized Specialized transducing phages are those phages which integrate into specific sites on their host s genome and therefore always affect the movement of a specific region of DNA. Generalized transducing phages are capable of moving any gene. The process of transduction has been extensively reviewed (Susskind and Botstein, 1978; Sternberg and Hoess, 1983; Masters, 1985; Margolin 1987; Weisberg, 1987). Transformation the first mechanism of bacterial gene exchange to be recognized (Griffith 1928) is a process whereby a cell takes up extracellular DNA and genetically incorporates it as its own genetic material Natural transforrnation is a norrnal physiological process of many bacteria, and unlike conjugation or transduction, is mediated by chromosomal genes. An important distinction must be made between natural transformation and artificial transforrnation Mandel and Higa (1970) frrst

PAGE 29

8 showed that the normally non-transformable bacterium E. coli could be induced to competence by an exposure to low concentrations of CaC12 Since this discovery, artificially-induced transformation has become one of the most commonly used laboratory tools in molecular biology for the introduction of cloned genes and plasmids into bacterial hosts. However, these transformations are induced by artificial means and are quite different from natural transformation which is a normal physiological process associated with specific gene products The process of natural transformation has also been reviewed (Venema, 1979; Smith er a!., 1981; Stewart, 1989; Palmen et al., 1993; Lorenz and Wackernagel, In Press). These three mechanisms of gene transfer have been investigated primarily under laboratory conditions, however, the delineations between these mechanisms may not be so clear-cut in the environment. For example, conjugative plasmids may be introduced into a reci pient by transformation or transduction, or bacteriophages may facilitate transformation by lysing donors In fact, it is possible that yet uncharacterized mechanisms of bacterial gene transfer remain to be discovered. Gene transfer in the environment Until recently, gene exchange between bacteria was thought to be chiefly a laboratory phenomenon Furthermore, most studies have concerned clinically important organisms not necessarily representative of organisms found in other environments. However, in the past few decades, the possibility of intra-and interspecific prokaryote

PAGE 30

9 gene exchange occurring in natural environments has been considered. These invest i gat i ons hav e led to a convincing body of indirect evidence that gene transfer occurs in the environment. Two fundamental questions have been addressed in reference to gene transfer in natural envi ronments ; the role that gene transfer has played in the past, and the potential for gene transfer to occur in the future Past gene transfer events have been inferred by epi demiolog ical studies while the potential for gene transfer can be inferred by demon stratin g that different environments contain the necessary components for gene transfer (e.g conjugative plasmids bacteriophage, extracellular DNA) and by demonstrating gene transfer in environmental simulations Other studies, concerned with risks in the use of GEM' s have focused on the survival of exotic organisms in natural env ironments Epi demiolog i cal studies seem to suggest the horizontal spread of genes within local env ironments (Kelch and Lee 1978; Maimone er al., 1986), particularly those genes which are carried on plasmids such as antibiotic and heavy metal resistances and for the degradation of xenobiotics, pestic ides, and hydrocarbons More recently the genetic struc ture of existing microbial populations have been examined to determine whether horizontal gene transfer events could explain the distribution of specific genotypes Antibiotic resistant genes have been especially useful for this type of analysis. For example Norelli er al. (1991) examined the distribution of the streptomycin resistance gene Psp36 among bacteria isolated from apple leaves and or chard soil in a New York state apple orchard Psp36 was found in all streptomycin

PAGE 31

10 resistant strains of Pseudomonas syringae pv. papulans examined, as well as in 39% of other streptomycin resistant Gram-negative bacteria in the orchard. Probes homologous to Psp36 from the New York orchard did not hybridize to streptomycin resistant P. syringae strains isolated from an Ohio orchard indicating that streptomycin resistance in New York strains arose independently from Ohio strains. Thus, the authors concluded that Psp36 first appeared in P. syringae and had been transferred to other Gram-negative bacteria following the use of streptomycin in New York orchards. In another example, the gene PBP2B, which encodes a membrane-bound penicillin binding protein from 14 penicillin resistant and 6 sensitive strains of pneumococcus were (Dowson et al., 1989, 1990) Penicillin resistance is due to a mutation in this gene and thus, both penicillin resistant and sensitive strains contain the PBP2B gene. The genes from the sensitive strains collected over a period of 50 years from three continents were very similar. The greatest difference between two strains was 14 out of 1 ,453 nucleotides Thus S. pneumoniae can be considered genetically uniform. However comparison of the PBP2B gene from penicillin resistant S. pneumoniae strains fell into two classes (A and B) which were distinct from the sensitive strains. The class B genes were found predominantly in strains collected from Spain and the United Kingdom and were found to differ from the gene of the sensitive strains by 57/274 (20 8%) nucleotides in a central region. Nine class A genes were found in penicillin resistant pneumococci isolated from various countries between 1971 and 1988. They were found to contain a common region of altered sequence and elsewhere, the gene consisted of a mosaic of regions similar to those in sensitive strains. Little similarity

PAGE 32

11 between the altered regions of class A and class B genes was found. These results, recently reviewed by Smith et al. (1991), were interpreted as indicating that class A and B resistances had arisen independently and have been spread among S. pnewnoniae strains by some mechanism(s) of horizontal gene transfer. Similar comparisons of gene sequences from several eukaryotes and their symbiotic bacterial partners have indicated that horizontal gene transfer has not been limited to the prokaryotes (Bannister and Parker, 1985; Houck et al., 1991; Bork and Doolittle, 1992; Amabile-Cuevas and Chicurel, 1993) In the laboratory trans-kingdom gene transfer has been demonstrated from prokaryotes to eukaryotes (Zambrysik et al., 1989; Heinemann and Sprague 1989), although transfer from eukaryotes to prokaryotes has not yet been demonstrated. This type of study clearly demonstrates the importance that horizontal gene transfer has played in bacterial evolution and perhaps, in eukaryotic evolution as well. The potential for bacterial genetic exchange by conjugation, transduction, and natural transformation has been demonstrated in a variety of diverse environments (Graham and lstock, 1978 ; Trevors and Oddie, 1986; Bale et al., 1987; Bale er al., 1988 ; Rochelle et al., 1988 ; O'Morchoe et al., 1988; Van Elsas et al., 1989; Fry and Day, 1990; Henschke and Schmidt, 1990; Paul et al., 1991; Pretorius-Guth, 1990; Saye er a!., 1990; Stewart and Sinigalliano, 1990; K.lingmuller, 1991) Additionally, gene transfer events from defined donors to indigenous microbial populations in a variety of environmental simulations have been documented (Henschke and Schmidt, 1990; Stewart and Sinigalliano 1990; Fulthorpe and Wyndham, 1991; Smit et al., 1991)

PAGE 33

12 Furthermore, it is clear that natural environments contain the necessary componenfs for genetic exchange. Conjugative plasmids are abundant in soil and aquatic environments, especially in environments that have been exposed to antibiotic, heavy metal, or other pollutant-induced stress (Sizemore and Colwell, 1977; Kelch and Lee, 1978; Burton er al ., 1982; Baya et al 1986; Schutt, 1989, 1990; Belliveau et al., 1991). Additionally, environmental isolates have been shown to be capable of mobilizing nontransmissible plasmids, facilitating the transfer of nonconjugative plasmid DNA by conjugation (Mancini et al 1987) Bacteriophage concentrations have been measured to be as high as 2 5 X 108 virus particles per milliliter in natural waters (Proctor and Furman, 1990), but measurements of bacteriophage densities of lOS to 106 per ml are more commonly cited (Bergh, 1989; Bratbak et al., 1990; Paul et al 1991b; Shigemitsu er al 1991; Bratbak er al ., 1992; Wommack et al., 1992) Recently, high numbers of virus like particles have been reported from marine sediments as well (Paul et al., 1993). This has implications for gene transmission by transduction. Lastly, in marine and freshwater environments, the presence of an abundant extracellular DNA pool containing high molecular weight DNA which could act as transforming material (DeFlaun and Paul, 1987), coupled with the existence of bacterial strains competent for natural transformation (Rochelle et al., 1988; Saunders and Saunders, 1988; Jeffrey et al., 1990) suggests the potential for natural transformation. Thus, a considerable amount of data indicates the potential for genetic exchange to occur in these environments. Examination of the known organisms possessing a transformation phenotype suggests that natural competence is widespread among the prokaryotes (Iuni, 1972;

PAGE 34

13 Ehrlich, 1977; Albritton eta/. 1978; Sox eta/. 1979; Juni and Heym, 1980; Saunders and Guild, 1980; Tucker and Pemberton, 1980; VanDen Hondel, 1980; Gromkova and Goodgal, 1981; Macrina et al., 1981; Smith et al., 1981; Carlson et al., 1983; Golden and Sherman, 1984; Stewart and Carlson, 1986; Saunders and Saunders, 1988; Jeffrey, 1990; Stewart, 1989; Frischer et al 1990; Wang and Taylor, 1990; Wahlund and Madigan, 1991; Frischer, unpublished results). Yet, despite the wide distribution of the transformable organisms, our understanding of the mechanisms of natural transformation are the result of studies from a few organisms The most well studied are two representatives from the Gram-positive genera, Streptococcus, and Bacillus, and one Gram-negative genus, Haemophilus. Studies of these genera has revealed that the physiology and genetics of the transformation process of these different systems vary Historically, differences were seen to be greatest between Gram-positive and Gram negative bacteria More recently, however, studies involving other transformation model systems have made this definition less clear, and it now seems that the process of natural transformation of _various different transformable organisms shows more common traits than previously thought (Stewart, 1989; Frischer et al., 1993). Reports of marine microorganisms which can be naturally transformed have recently been made P srurzeri, strain Zobell (formally P. peifectomarina), has been reported to be transformable with homologous chromosomal DNA (Carlson eta/. 1983). P srurzeri strain ZoBell is also transformed by chromosomal DNA from the terrestrial strain of P. sturzeri. Transformation frequencies are only about 10-fold lower than the transformation of recipients with DNA isolated from strains of the same species (Stewart,

PAGE 35

14 1989). This could be highly significant since it suggests the potential for interspecific transport of genes from population to population and from one environment to another. Transformation of P. stutzeri by heterologous plasmid DNA has not been demonstrated in these studies. In our lab we have discovered another marine isolate, tentatively identified as Vibrio sp. which is naturally transformable Natural transformation of this isolate has been detected by homologous chromosomal DNA and by broad host range plasmid DNA (Jeffrey et al 1990; Frischer et al., 1990). The presence of natural transformable microorganisms in marine environments suggests the potential for gene exchange via transformation in these environments, yet the process in marine environments may differ from those of terrestrial or clinical environments The studies pursued in this work develop the marine Vibrio system as a marine model for natural transformation and use the Vibrio as a guide to explore the potential for natural transformation among marine microbial communities. Mechanisms of natural transformation As was stated earlier, our understanding of the process of transformation is derived largely from laboratory examination of three genera of naturally competent bacteria, Streptococcus, Bacillus, and Haemophilus. Based on these organisms, Smith (1981) proposed two general models for transformation, one pertaining to Gram-positive organisms and the other to the Gram-negative ones. Since DNA must penetrate the outer membrane of the recipient, the variations in composition and structure of cell walls

PAGE 36

15 between Gram-negative and Gram-positive bacteria could have caused the evolution of two dist i nct processes of transformation. This was the basis for Smith's dual transformation model. However, examination of the mechanisms of transformation in Haemophilus, Bacillus, and other bacteria suggest that such divisions are not so straightforward. Historically, the analysis of natural transformation has involved separation of the proces s into several discrete steps : competence development, DNA binding and uptake, and DNA integration and expression This format has been used in a number of reviews to discuss the process of natural transformation (Venema, 1979; Smith et al 1981 ; Goodgal 1982; Kahn and Smith, 1984; Stewart and Carlson 1986 ; Stewart, 1989; Sco c ca, 1990 ; Dubnau 1991; Palmen et al. 1993) Table 1 summarizes the trans f ormation systems of these bacteria Development of Competence Bacterial genetic competence is defined as a ph y siological state that permits the uptake of exogenous DNA in macromolecular form. With the except i on of pilliated forms of Neisseria gonorhoeae (Sparling, 1966 ; Gibbs et a! ., 1989 ; Scocca 1990), competence appears to be regulated in all naturally tran sf ormable bacteria (Goodgal, 1982) However, Williams (1993) recently reported that natural genetic competence (henceforth referred to simply as competence) was constitutive in Acinerobacter calcoaceticus, although previous work had shown this organism to regulate competence (Cruze et a/., 1979; Ahlquist et al 1980). Competence in Streptococcus sp (S. pneumoniae and S. sanguis) is induced by small extracellular proteins (competence factors) that are synthesized and exported during

PAGE 37

Table t. Transformation processes of model transformation systems PROCESS STREPTOCOCCUS BACILLUS HAEMOPHILUS Competence Prolein. RNA, a:nd DNA synlhe!i! RNA:proll:in ratio Shift towonb unhal.nced c..-lh. Induction nocluct:d. Critical concmtntion ond DNA ynlheoi i reduced. in cdl compooition, of onluble eKtncellular Criticol concmtntion of At leut i polypqlticles compdaoce factor ruched. ooluhlc utroc
PAGE 38

17 growth Competence occurs when an effective concentration of competence factor has accumulated In culture thi s occurs during exponetial growth at cell densities of about 108 cells /ml. Competence is induced in virtually 100% of the cells, and is correlated with the appearance of a set of new proteins (Competence Induced Proteins CIP's) se v eral of which appear to be membrane localized (Morrison and Guild, 1973; Vijayakumar and Morrison, 1986) Once aquired competence persists for only 10 to 20 minutes and then decays rapidly (Tomasz and Hotchkiss, 1964; Tomasz and Mosser, 1964; Tomasz, 1966 ; Tomasz, 1970; Hui and Morrison 1991) Competence in B. subtilis is usually expressed after rather than during the exponential growth phase At maximal competence only 10-20% of competent cultures consist of competent cells Competence development in B subtilis can be accelerated b y the addition of cell-free spent media from competent cultures. As with Streptococcus, this implies the participation of some extracellular competence factor. Competence in B sub rili s may be related to several other post-exponentially regulated processes of B. subrilis including sporulation, motility, expression of degradative enzymes, and antibiotic production (Dubnau, 1991). Similar to the Streptococcus system competence in Bacillis is a short lived phys i ological condition that disappears rapidly in stationary growth phase (Albano er al., 1987) Competence development of Haemophilus (H. injluenzae and H. parainjluenzae) is internally regulated ; no extracellular competence factor has been detected (Stewart and Carlson 1986) As in the other models, competence is expressed transiently and once achie ved is rapidly lost (Goodgal and Herriott, 1961; Redfield 1991). Many

PAGE 39

18 environmental changes can trigger competence but the mechanisms controlling competence regulation in Haemophilus are not well understood However, shifting of an exponentially growing culture to a medium that does not support growth can cause virtually 100% of the population to become competent (Herriott et al., 1970; Stewart and Carlson, 1986). Competence development also occurs at the onset of the stationary phase in several other Gram-negative bacteria, including Pseudomonas stuzeri (Carlson et al., 1983), and Azotobacter (Page, 1982). Competence in Neisseria meningitidis and N gonorrhoeae is continuously expressed therefore, it is presumably not subject to the growth stage and nutritional forms of control typical of other systems (Gibbs et al., 1989; Scocca, 1990). It has been postulated that transformation in Neisseria has evolved as a mechanism to enhance antigenic variation (Gibbs et al., 1989; Scocca, 1990). Competence development in Acinetobacter calcoaceticus does not appear to be correlated with a metabolic downshift normally associated with the transition from exponential to stationary phase Rather, competence in A. calcoaceticus initiates during log-phase growth (Palmen et al., 1992). Similarly, several other Gram-negative organisms develop competence during logarithmic growth phase including : Deinococcus radiodurans (firgari and Moseley, 1980), Synechococcus (Essich et al., 1990) and Chlorobium (Ormerod, 1988; Wahlund and Madigan, 1991). In all these organisms competence is maximal during logarithmic growth and declines thereafter. The relevance of these examples to competence development in marine heterotrophs is unknown. However, these examples serve to demonstrate that the

PAGE 40

19 regulatory apparatus that controls competence is highly variable among different organisms and evidently has evolved independently to serve the special needs of each organism. Therefore, in order to identify and elucidate the triggers for transfonnation in marine environments, it will be useful to examine the development and maintenance of competence in a naturally transformable marine heterotroph. DNA and Uptake Once competence has been established, binding of DNA to the cell surface occurs (Fox and Hotchkiss, 1957). In general, the DNA binding and uptake process is rapid in competent cells (Garcia et a/., 1978). The DNA binding process in S. pneumoniae is mediated by the presence of one or more membrane-associated proteins termed binding factors (Lacks, 1977). Homologous DNA is not bound preferentially over heterologous DNA indicating that the DNA b i nding proteins do not exhibit 5e4uence specific recognition for S. pneumoniae. DNA saturation studies reveal that there are multiple binding sites on the surface of the competent In the case of S pneumoniae, as many as 80 binding sites may exist on the cell surface (Fox and Hotchkiss, 1957). These sites may correlate with the presence of binding proteins, but there is some evidence suggesting that the binding protein is always present and that autolytic enzymes expressed during competence expose those proteins to extracellular DNA during competence induction (Seto and Tomasz, 1975) The uptake process in Streptococcus involves the transport of a single strand of the bound DNA through the cell membrane with the concomitant degradation of the

PAGE 41

20 complement (Lacks, 1962). There appears to be no strand preference for uptake (Lacks, 1977) The single-strand DNA (ssDNA) that is transported into the cell is protected from further degradation by cytoplasmic nucleases by association with a ssDNA binding protein synthesized at the onset of competence (Morrison and Baker, 1979). Binding and uptake of DNA by Bacillus appears similar to S pneumoniae. Double-stranded DNA (dsDNA) appears to be required for transformation, although a limited amount of single-stranded DNA does appear to be bound by competent cells. The process of DNA binding and uptake in B. subtilis involves several membrane bound proteins that associate with dsDNA forming a membrane bound protein complex As in the Stre pt o c occus transformation system, a single strand of DNA is taken up with the complement being degraded extracellularly In the Bacillus system the release of the ssDNA from the membrane complex seems to occur simultaneously with its incorporation into the endogenote (te Riele and Venema, 1984) Several differences in DNA binding and uptake systems are exhibited by the Gram-negative genus Haemophilus compared to DNA binding and uptake in Streptococcus and Bacillus. One important difference seems to be that DNA binding in Haemophilus is restricted to homologous DNA This specificity discriminates against the entry of at least some foreign DNA The specificity for binding is limited to the same or closely related species However, DNA from other sources can bind to the surface of competent cells but is not tightly absorbed and is thus a poor competitive inhibitor of homologous DNA binding (Smith er al 1981) In Haemophilus the specificity for recognition of binding sites appears to reside in an 11 base pair recognition sequence (5' -

PAGE 42

21 AAGTGCGGTCA-3'; Sisco and Smith, 1979; Smith and Danner, 1981). More recently Tomb er a/. (1991) demonstrated that only the flrst 9 base pairs of this sequence was required for sequence specific uptake. The introduction of this sequence to heterologous DNA allows binding and uptake of that DNA (Danner er al., 1980; Danner et al., 1982) In contrast to the pneumococcal system, where as many as 80 binding sites exist per cell, the Haemophilus cell appears to have only 3 to 8 sites per cell, based on DNA saturation studies (Deich and Smith, 1980) As in the pneumococcal systems, transforming DNA is fust loosely bound and during this stage can be removed by high salt washes or DNase digestions However, shortly after binding the dsDNA is incorporated into membrane vesicles called transformasomes (Kahn er al 1983). This process appears to protect the DNA from DNase activity. Associated with these vesicles is a dsDNA binding protein. This protein appears to be important in the transformation process because mutants that lack it are deficient in transformation (Kahn et al. 1979). Transport of DNA through the periplasmic space varies between the H. influenzae and the H. parainfluenze systems. In H. influenzae DNA sequestered in the transformasome appears to undergo degradation until homologous pairing occurs with the endogenote. At this time the surviving strand is exchanged and is stabilized in the recipient genome. In H. parainj1ue71Zill! the DNA is not degraded while in the transformasome. The process of DNA binding and uptake in other organisms has not been extensively studied. Pilliated strains of Neisseria appear to be constitutive for DNA uptake (Biswas er al., 1977). DNA uptake in Neisseria, like Haemophilus, involves

PAGE 43

22 sequence-specific uptake (Goodman and Scocca, 1991), although the sequence is not the same as that recognized by Haemophilus (Mathis and Scocca, 1982). The DNA elements recognized by competent Neisseria is a 10-residue nucleotide sequence (5'GCCGTCTGAA-3'; Goodman and Scocca, 1988) Similarly, the uptake of DNA in P. srutzeri has been shown to require homologous DNA, but unlike Haemophilus or Neisseria, the demand for homology in P. stutzeri reflects the recombination event rather than the occurrence of sequence specific DNA binding (Carlson er al., 1984). Recombination and Expression of DNA. While the gene products involved in competence development and the binding and uptake of DNA appear to be unique to tran sformation, the integration and expression of transforming DNA seems to be mediated by generalized recombination enzymes. In pneumococcus the integration of the tran sformed DNA involves homologous recombination between the ssDNA protein complex (eclipse complex) and a homologous DNA strand of the endogenote (Fox and Allen 1964). The nature of the integration process is similar to a RecA-mediated exchange pro ces s (Clark, 1973) In B subrilis the recombination process involves a number of gene products At least 12 distinct recombination genes have been mapped by mutation analysis (Piggot and Hoch, 1985). As in the Gram positive organisms integration occurs by homologous recombination in Haemophilus. ssDNA released from the transformasome into the cytoplasm is immediately incorporated into the endogenote The transfer of chromosomally encoded genes is facilitated by homologous recombination in a number of other naturally transformable organisms including: P sruzeri (Carlson er al., 1984), Campylobacrer sp (Wang and Taylor, 1990), and in Acinerobacter (Juni

PAGE 44

23 and Janik, 1969). Natural plasmid transformation In general, bacteria that can be transformed by chromosomal DNA can also be transformed by plasmid DNA. However, the efficiency with which chromosomallyencoded or plasmid-encoded markers are transferred differ. As is the case with chromosomal transformation, what is known concerning the process of natural plasmid transformation comes from studies of relatively few organisms under laboratory conditions Natural plasmid transformation has been studied in greatest detail in Bacillus subrilis, Srreprococcus pneumoniae, Haemophilus injluenzae, Neisseria gonorreae, and to a limited extent in Azotobacter vinelandii. However, plasmid transformation has been demonstrated in a wide varity of organisms. A list of organisms which are naturally transformable with plasmid DNA is presented in Table 2. Plasmid transformation of B. subtilis occurs at a much lower frequency than does transformation with chromosomal DNA. Transformation by chromosomal DNA requires the uptake of approximately one molecule of transforming DNA per transformant while plasmid transformation requires the uptake of 1,000 to 10,000 molecules per transformant (Contente and Dubnau, 1979). Only multimeric forms of plasmids (i.e. plasmids containing self-homology) transform B subtilis (Canosi et al., 1978; Mottes et al 1979; de Vos and Venema, 1981).

PAGE 45

Table 2. Organisms capable of natural transformation Organism Transformable with plasmid DNA A.cinerobacrer calcoaceticus A.gmenellwn quadruplicatwn PR6 (Formally Synechococcus PCC7002) A.nacysris nidulans R2 AzoTobacter vinelandii Bacillus subrilis Campylobacrer coli Campylobacter jujuni Haemophilus injluenzae Haemophilus parainjluenzae Neisseria gonorrhoeae Pseudomonas srurzeri Rhodobacrer sphaerodies (Formally Rhodopseudomonas sphaeroides) Srreprococcus murans Streptococcus pneumoniae Streptococcus sanguis Vibrio sp. Reference Cruze er al., 1979 Buzby et al., 1983 24 Chauvat et al 1983 Golden and Shennan, 1984 Doran et al., 1987 Glick er al., 1985 Ehrlich, 1977 Canosi er al 1978 Albano et al., 1987 Wang and Taylor, 1990 Wang and Taylor, 1990 Albritton et al 1981 Pifer, 1986 Stuy and Walter, 1986 Gromkovaand Goodgal, 1979 Sox er al., 1979 Biswas er al., 1986 Carlson er al., 1984 Tucker and Pemberton, 1980 Kuramitsu and Long, 1982 Murchison et al., 1986 Barany and Tomasz, 1980 Chen and Morrison, 1987 Saunders and Guild, 1980 LeBlanc and Hassel, 1976 Behnke, 1981 Macrina er al., 1981 Jeffrey er al., 1990 Frischer er al. 1990 Other eenera of bacteria capable of natura} transformation Achromobacter sp. Juni and Heym, 1980 Chlorobiwn sp. Ormerod, 1988 Merhylobacreriwn sp. Micrococcus sp. Wahlun and Madigan, 1991 O'Connor er al., 1977 Tigari and Moseley, 1980

PAGE 46

25 Monomers may exhibit transforming activity if they contain regions of homology to the chromosome or to an extrachromosomal element in the recipient (Contente and Dubnau, 1979; Gryczan et al., 1980; Bensi et al., 1981; Canosi et al., 1981; Michel et al 1983; Weinrauch and Dubnau 1987; Dubnau, 1991). The inability of plasmid monomers to transform B subtilis is not a consequence of a physical or genetic deficiency of monomeric plasmid DNA. Rather, the requirement for multimeric forms of plasmid DNA reflects the mode of DNA processing in the transformation in B subrilis (de Vos and Venema, 1981). The process of plasmid transformat ion in B subrilis proceeds similarly to chromosomal transformation in this organism Plasmid DNA is first bound to the cell surface, is nicked, and a single strand is transported through the membrane while the complimentary strand is degraded (de Vos er al 1981). Immediately after uptake, monomer DNA is always found as a single stranded molecule with a shorter length than the parent plasmid. Multimeric plasmid DNA is found in single-stranded, double-stranded, and partially double-stranded forms (de Vos er a/ 1981). The double-stranded form is the result of annealing and synthesis of the complimentary strand Multimers are thought to anneal as follows. Single strands with opposite end polarity damaged during entry into the cell are converted into acid soluble products and fragments that are smaller than the monomer plasmid. These fragments may form partially double-stranded molecules in a recEA-independent reannealing event. DNA synthesis converts them into double stranded form and superhelical turns are introduced resulting in a closed circular monomeric plasmid (de Vos er al 1981). Thus, plasmid DNA that contains self-homology and can successfully

PAGE 47

26 repl i cate in B. subrilis can be naturally transformed into B subtilis Plasmids isolated from Staphylococcus aureus and E coli have be transformed into B. subtilis (Gryczan et al., 1978 ; Randen and Venema 1984) In contrast to B subtilis, S pneumoniae can be transformed by monomeric plasm i d molecules However, this process occurs with two hit kinetics. Two plasmid molecules are carried into a recipient and associate to form a duplex. The remaining g a p s are regene r a ted resulting in an i ntact plasmid (Saunders and Guild, 1981a). The proc ess is not perfect as regenerated plasmids often contain deletions (Saunders and Guild 198la ) M u ltimeric p l asmids transform with first order kinetics (Saunders and Guild 1981 b ) Open circular and linear forms of plasmids also exhibit transforming activity, bu t a t 3 5fold lower frequencies compared to closed circular forms (Saunders and Guild, 1981 b ) As in Bacillus, p l asmid transformation in Streptococcus is relatively inefficient, requ iri ng 10 to 1 ,000 plasmid molecules per transformant. Intera c t i ons between homologous sequences in the rec i pient and transforming DNA in S pneumoniae are s i milar to those observed for B. subtilis. Transformation in the presence of homologous DNA requ i res the uptake of only one plasmid molecule (Lopez er al 1982). Monomeric plasmi d molecules which contain sequence homology to the chromosome of the recipient will transform approximately ten times more efficiently than plasm i d DNA without chromosomal homology. Recombination and exchange between homologous sequences in recipients and transforming DNA is similar to that observed in B subrilis (Lopez er al 1982). The amount of exchange is

PAGE 48

27 proportional to the length of the homologous sequence. Similarly the addition of homologous chromosomal DNA fragments to heterlogous plasmid DNA allows the transformation of plasmid DNA by P. stutzeri (Carlson et al., 1984). While transformation with plasmid DNA for B. subtilis and S. pneumoniae is generally similar in function to transformation with chromosomal DNA, in Haemophilis there appear to be some differences between the transformation process with chromosomal and plasmid DNA. As in the other models, plasmid transformation in Haemophilis is considerably less efficient than with chromosomal DNA. When the transforming plasmid contains no homology to the recipient's genome, approximately one in 1CP cells is transformed (Notani er al., 1981; Pifer, 1986). This is in marked contrast to the high rate of transformation of chromosomal markers which occurs in 1 to 10% of viable competent cells The reason for this differential transformation efficiency is not fully under s tood. One possibility is that heterologous plasmids are inefficiently taken up due to a lack of specific uptake sites present in H injluenzae chromosomal DNA (Danner er al., 1980; Balganesh and Setlow, 1985) Notani et al. (1981) found that chromosomal DNA competed about 100 times more effectively than plasmid DNA for uptake by competent H influenzae cells. However, inefficient uptake may not fully explain the 4 to 5-log difference between plasmid transformation and chromosomal transformation frequencies. Incorporation of homologous chromosomal DNA inserts into transforming DNA allows increased plasmid DNA in some cases (Balganesh and Setlow, 1985) but not in others (Pifer, 1986). Examination of the fate of transforming plasmid DNA that contained chromosomal

PAGE 49

28 insertions to those without, indicate that two types of rec -dependent recombination exist between a plasmid. Genetic infonnation can be transferred from transfonning plasmid DNA to the chromosome (marker rescue) In this case the plasmid is generally not maintained, or the plasmid can be maintained as a autonomously replicating unit (plasmid establishment). Establishment of the plasmid is two orders of magnitude less efficient than is marker transfer to the chromosome (Pifer, 1986). In some cases chromosomal markers have been inserted into the plasmid during recombination, but this type of recombination event appears to be independent of transfer of the plasmid marker to the chromosome (Setlow er a!. 1981 b). Neither of these types of recombination occur after conjugal transfer (Balganesh and Setlow, 1985) Yet another caveat to plasmid transfonnation in Haemophilis is the possibility of plasmid establishment by bypassing any DNA processing. During chromosomal transformation and marker rescue, extensive degradation of donor DNA occurs during transformation. One strand of an entering molecule is completely degraded and the other strand is a least partially degraded during entry into the cytoplasm (Notani and Goodgal, 1966 ; Barnay er al., 1983 ; Pifer and Smith, 1985). A free end of the transfonning molecule is necessary for efficient translocation out of transfonnasomes (Pifer and Smith, 1985). However, Pifer (1986) using plasmid molecules bearing no detectable chromosomal homology, observed two inconsistencies with the accepted mechanism of transfonnation First, intact plasmids could be recovered from cells transfonned by linear plasmid molecules; indicating that no infonnation was lost from the ends of linear molecules during plasmid establishment. Second, it was found that cicular covalently

PAGE 50

29 closed (CCC) molecules were able to carry out both marker rescue and plasmid establishment as efficiently as topological forms which did not have a free end (Pifer, 1986) These observations suggest that the transforming plasmid DNA did not undergo any degradation during the transformation process, leading to the speculation that, in rare cases, donor plasmids may enter the cytoplasm of naturally competent H. injluenzae cells as intact duplexes. Pifer (1986) has proposed that plasmid DNA sequestered in the transformasome in rare cases exits from the transformasome without the usual directional translocation and degradation. In these instances, although most donor molecules are partially degraded and thus prevented from establishing in the cell, an occasional DNA molecule enters the cytoplasm in intact double-stranded form This may come about because a rare trasformasome lacks the normal translocation proteins and allows passive diffusion of the DNA into the cell. This process bears a resemblance to the entry of DNA into artificially competent cells, except that the DNA uptake mechanisms present in the naturally-competent cell are utilized in place of changes in membrane permeability by physical, chemical, or enzymatic means. Another indication that the mechanism of plasmid transfer differs from that of chromosomal transformation in Haemophilis is the differential effects of high concentration of divalent cations (Smith et al., 1981) and temperature shock (Stuy, 1979). Both of these treatments stimulate the uptake of circular or linear plasmid DNA but have no effect on chromosomal DNA uptake in competent Haemophilis cells. It is important to note that these treatments effect uptake in cells that are already competent. Cations do not create competence in Haemophilis as they do for E coli during artificial

PAGE 51

30 transformation (Gromkova and Goodgal, 1981). In P. stutzeri the presence of cations have been demonstrated to decrease transformation efficiences of competent cultures (Carlson et al 1983) The relevance of the transformation process in Streptococcus, Bacillis, and Haemophilis to transformation in marine environments is unknown However, insights gained fro m the study of these organisms guides our ability to predict the potential for transformation to occur in other organisms and in other environments For example, if competence develops only in stationary phase or when cell densities reach critical levels predictions are possible about the circumstances or environments which are likely to facilita t e gene transfer b y transformation Likewise, the nature of the DNA binding process may allow certain environmental parameters (e g salinity and ion concentration, presenc e of surfaces etc ... ) to influence the effective transformation of a cell population The r e f ore the potential for transformation in specific environments may be influenced by the e f fecti v e binding of DNA to cells and by the stability of the DNA-recipient compl e x The sequence specificity of DNA uptake, as are the mechanisms of genetic recomb i nat i on, are also important when considering the potential for natural transformation and the role that transformation may play as a means of inter-or intragenetic exchange amongst natural microbial populations. In most transformation systems competent bacteria will take up heterospecific as well as homospecific DNA. the potential for interspecific gene exchange by plasmid transformation is likely to be greater than interspecific transformation by chromosomal material. Comparison of transformation processes between marine organisms and other systems

PAGE 52

31 enables us to evaluate whether what is known about transformation based on a few nonmarine bacteria may be analogous to the process of transformation of marine bacteria Natural transformation in the environment A limited number of studies have examined transformation in natural env ironments. Graham and !stock (1978 1979, 1981) demonstrated transformation of four different chromosomal markers, three of which were linked, between B subtilis strains in sterile potting soil. After eight days 79 percent of the isolates exhibited a phenotype associated with both parental strains Since this pioneering work the process of natural transformation of B subtilis in soil and sediment environments has been further in v estigated. One interesting observation is that DNA bound to particles is re s istant to nuclease degradation while still retaining its transforming activity (Lorenz et a!. 1981 ; Aardema et al 1983; Lorenz and Wackernagel, 1987; Lorenz et al. 1988; Romano ski eta!., 1992 ; Romanoski et al., 1993). Although these studies are interesting, their rele v ance to transformation in soil and sediments is not clear Not only were the majority of the:;e studies conducted in the absence of ambient microbial cells, but most often they were conducted with clean soil and sediment, which do not occur naturally. In aquatic habitats transformation of auxotrophic Acientobacter strains were shown to be transformed by plasmid or chromosomal DNA on surfaces. The effect of the ambient microbial community was inconsistent on the transformation frequency observed in sterile or nonsterile situations. In some instances the transformation frequencies observed in

PAGE 53

32 situ were higher than in sterile microcosms (Rochelle et al 1988; Williams et al., 1992; Williams 1993; ) Natural transformation of P stutzeri has been demonstrated in sterile and nonsterile marine sediments with homologous chromosomal DNA (Stewart and Sinigalliano 1990; Stewart et al 1991) Due to the rapid rates of DNA turnover in aquatic environments several investigators have speculated that natural transformation is more likely to occur in sediments than in the water column because the interaction between sediment particles and DNA has been shown to inhibit the degradation of extracellular DNA (Aardema et al., 1983 ; Lorenz and Wackemagel 1987; Stewart and Sinigalliano 1990 ; Romanowski et al 1991) Nonetheless, in our laboratory we have demonstrated natural transformation of a marine Vibrio by plasmid DNA in sterile and non-sterile marine water microcosms, as well as in sterile sediment. In this system transformation was not observed in native sediments (Paul et al., 1991a) Currently there is onl y one report of the transformation of a mixed indigenous microbial popu l ation Stewart and Sinigalliano (1990) were able to detect a 3 fold increase in rifampin resistance when non-sterile sediment was exposed to DNA extracted from a mixed population of rifampin resistance organisms isolated from the same location This increase in rifampin resistance was interpreted as indicating that transformation occurred Increased rifampin resistance was reduced by half in the presence of DNase. Early work in this laboratory indicated a similar increase in the expression of antibiotic resistance after the addition of DNA containing genes encoding such resistance was added to ambient marine microbial communities However transformation could not be confirmed by molecular probing for the genes suspected of being transformed (Jeffrey,

PAGE 54

33 1989). Nevertheless, the potential for transformation to occur has been demonstrated in several natural environments but has not been extensively explored. The transfer of plasmids among natural microbial populations, including marine populations, is of special interest since plasmids may increase the metabolic potential of indigenous flora (Novick, 1969; Broda, 1979; Foster, 1983; Trevors et al., 1987; Schutt, 1990). Plasmids are common in marine microbial populations. In 1977 Sizemore and Colwell conducted a survey of antibiotic-resistant marine bacteria off the southeastern coast of the United States and found that from 0 to 55% of the total viable count were resistant to multiple antibiotics. Of these, 60% contained plasmids Further studies revealed that eight out of sixteen plasmid-bearing, antibiotic resistant strains were able to transfer their plasmids and resistant patterns to E coli or Beneckea harveyii (Sizemore and Colwell 1977). In an extensive study Schutt (1990) examined numerous fresh water and marine habitats for bacteria containing plasmids. Schutt reported that approximately 41% of nearly 500 bacterial isolates collected contained plasmids of various molecular weights. Furthermore, in sites that were considered to be highly impacted by human activity as many as 80% of the isolates contained plasmids. This h igh percentage of plasmid-bearing strains suggests an important role of plasmids in aquatic environments. The occurrence and incidence of bacterial plasmids has been reported from numerous other marine environments including from bacterial assemblages in the Antarctic (Kobori et al., 1984; Ray et al., 1991), pristine and polluted freshwater environments (Burton et al., 1982 ; Genthner et al., 1988; Pickup, 1989; Schutt, 1989), from marine waters (Sizemore and Colwell, 1977; Glassman and McNicol, 1981 ; Baya

PAGE 55

34 er al., 1986; Aviles et al., 1993) in marine sediments (Belliveau et al., 1991), and in groundwater (Ogunseitan et al., 1987). Direct evidence for plasmid transfer in natural environments and to the indigenous population is difficult to document. A major constraint is the safety concern involved in a release of trackable plasmids. These plasmids usually contain recombinant sequences and are thus not acceptable for field experiments. Bale et al. (1988), circumvented this problem by demonstrating the transfer of large, naturally-occurring mercury-resistance plasmids between river epilithic bacteria. Fulthorpe and Wyndham (1991), using flowthrough freshwater mesocosms, demonstrated the transfer of a naturally occurring plasmid encoding 3-Chlorobenzoate (3Cba) degradation (pBRC60) from Alcaligenes sp (strain BR60) to naturally occurring bacteria. In these studies it was not possible to determine the mechanism of transfer, although conjugation was suspected (Fulthorpe and Wyndham, 1991) Generally this approach is not feasible and many people have resorted to inferring plasmid transfer events by the examination of plasm i d distribution amongst bacterial populations (Schofield et al., 1987), by the use of self-contained microcosms, or by laboratory simulations, to explore the potential for plasmid transfer in various environments Plasmid transfer has been demonstrated in many different environmental simulations and by all three recognized mechanisms of gene transfer Plasmid transfer has been shown to be facilitated by transduction in waste water treatment facilities (Osman and Gealt, 1988), in animals (Novick and Morse, 1967; Novick et al., 1986), and in aquatic environments (Morrison et al., 1978; Saye et al., 1987; Amin and Day,

PAGE 56

35 1988; Saye eta!., 1990). The transfer of chromosomal markers by transduction has been demonstrated in sterile and nonsterile soil (Zeph et al., 1988) Plasmid transfer by conjugation has been documented in numerous environments under both sterile and non sterile conditions including soil and rhizosphere environments (Trevors and Oddie, 1986; Pretorius-Guth, 1990; Top et al., 1990; Watler et al., 1991; Klingmuller, 1991; Kinkle and Schmidt, 1991; Brokamp and Schmidt, 1991), on plants (for reviews see: Nester and Kosuge, 1981; Hepburn, 1982; Hooykaas and Schilperoort, 1984 ; Hille et al 1984; Gheysen eta!., 1985; Binns and Thomashow, 1988; Zambryski, 1988), in freshwater (Bale er al ., 1987; Bale et al., 1988; Genthner et al., 1988; Kolenc et al 1988; O Morchoe et al 1988; Rochelle eta!., 1989), in wastewater and sludge (Altherr and Kasweck 1982; Mach and Grimes, 1982; Gealt et al 1985; McPherson and Gealt, 1986; Mancini et al., 1987; Nublein eta!., 1992) in drinking water (Sandt and Herson, 1991), in marine environments (Sizemore and Colwell, 1977; Gauthier et al ., 1985; Sorensen 1988; Gauthier and Breittmayer, 1990), and in animal models (for review see Roberts 1989) Additionally plasmids which can mobilize non-conjugative plasmids hav e been isolated from aquatic environments (Hill et al 1992). Although the transfer of plasmids amongst microbial communities is most often associated with conjugation, plasmid transfer can also be accomplished by natural transformation However, few studies have demonstrated natural plasmid transformation in natural environments. Rochelle eta!. (1988) reported that a small plasmid encoding mercury resistance which was isolated from the river Taff, Wales (pQM17; 7.8 Kb) could be transferred by natural transformation between Acientobacter strains and

PAGE 57

36 transformation could occur in aquatic environments In a related study Fry and Day (1990 ) found that 65% of endogenously isolated plasmids from the Taft River were less than 10 Kb, although these plasmids could presumably be mobilized by the presence of conjugative plasmids plasmids in this size range are too small to encode conjugation genes (Thiry et a/., 1984) In our laboratory we have demonstrated natural plasmid transformation in marine environments (Jeffrey et al., 1990; Paul et al., 1991a; Paul et a/., 1992). Thus, transformation may be a mechanism responsible for the dissemination o f small non-conjugative plasmids in aquatic environments. In summary, the exchange of genetic information among microbial communities represents a fundamental although poorly understood process in environmental microbiology today Even before nucleic acids were recogn i zed as the material responsible for inheritance, bacterial gene transfer had been observed. Since Griffith's pioneering work in the 1920's with Streptococcus and mice our understanding of microbial gene transfer has greatly increased Much of the success of modern genetic eng ineering depends on gene transfer techniques. In the laboratory three distinct phy s i olog ic al processes that enable genes to be transferred from one bacterium to another and these gene transfer processes have been examined in great detail, in at least a few systems. It has been demonstrated that bacterial gene transfer is a widespread phenomenon among bacteria and a great deal of data has been obtained that suggest that gene transfer, including natural transformation, may be of significant importance to natural microbial communities. Examination of diverse environments has revealed that natural environments contain all the components necessary for bacterial gene exchange.

PAGE 58

37 As more information about gene transfer and processes is acquired it is realistic to expect that we will have access to the sum of 3.5 billion years of prokaryote evolution in the form of existing gene pools Yet, our understanding of gene transfer in the environment is limited. Not only do risk assessment studies of the application of this new need to be done, but fundamental questions concerning gene transfer in the environment remain to be answered. The of this study were three-fold First, to develop a marine model system for the study of natural plasmid transformation in marine environments and to compare th i s system to other transformation systems which have been studied in greater detail. to determine whether marine environments favor transformation and third, to demonstrate the movement of plasmid encoded genes to marine microbial popu l ations by natural plasmid transformation.

PAGE 59

CHAPTER 2: NATURAL PLASMID TRANSFORMATION IN A HIGH-FREQUENCY-OF TRANSFORMATION MARINE VIBRIO STRAIN. Introduction 38 Unt i l recently gene exchange between bacteria by natural transformation was thoug h t to be largely a laboratory phenomenon restricted to a limited number of bacterial species However, in the past several years the potential for intraand interspecific gene exchange in natural environments has been convincingly demonstrated (Schuberg and Zervos 1989 ; Trevors et al., 1987; Coughter and Stewart, 1989; Levy and Miller, 1989 ; F ry and Day, 1990; Smith eta/. 1991; Veal era/., 1992). As a result of these studies t h e importan c e of horizontal gene transfer and genetic recombination among bacteria is now thought to be more significant for the evolution adaptation and ecology of natural m ic robial assemblages than previously believed (Fry and Day, 1990). The potential for gene transfer in the marine environment has been demonstrated; by conjugation, (Gauthier 1990), transduction (Baross era/. 1973; Keynan eta/., 1974; Levisohn, 1987) and transformation (Stewart 1989; Paul er al., 1991a) and marine environments have been shown to contain the components necessary to support natural transformation That is several marine bacterial isolates have been reported to be naturally transformable (Table 2) and, aquatic environments have been shown to contain an abundance of

PAGE 60

39 dissolved DNA which could potentially act as transforming DNA (DeFlaun et al., 1986, 1987; Paul et a!., 1987) Presently, two primary obstacles hinder the study of natural plasmid transformation in the marine environment. First, there are few reports of naturally transformable marine bacteria It is unclear whether this indicates that few naturally transformable marine organisms exist, or there are too few investigations which have explored the possibility. Second, in the few examples where plasmid transformation has been studied in marine bacteria, the transformation frequencies reported are extremely low (Chauvat eta! 1983; Doran et al. 1987; Frischer et al., 1990; Jeffrey, 1990). In our laboratory plasmid transformation in the marine Vibrio sp strain DI-9 has been reported to be in the range of 10"9 to 10"8 transformants/recipient (Frischer eta!., 1990; Jeffrey et al., 1990). Thus transformation of this organism can only be detected under laboratory conditions requiring cultivation to very high densities. Natural transformation by chromosomal markers in the marine bacterium Pseudomonas stutzeri strain ZoBell, formally P. perfectomarina, has been reported to be as high as 4 x 10"5 transformants per recipient but plasmid transformation has not been reported in this strain. P. stutzeri could be transformed by plasmid DNA when the plasmid contained chromosomal fragments from the recipient. In this case the transformation efficiency increased proportionally with the size of the chromosomal insert (Carlson et al., 1984). In this part of the study the development of a high-frequency-of-transformation (HIT) variant of the marine Vibrio strain DI-9 is described and a preliminary comparison between the parental strain DI-9 and the HIT strains is made. These strains are suitable

PAGE 61

40 for the invest i gation of natural plasmid transformation in the marine environment. Methods Bacterial strains The strains and plasmids used in this study are listed in Table 3 Strain DI-9 has been identified as a Vibrio sp (Frischer eta!., 1990 ; Jeffrey et al 1990). To verify the identity of transformants, HFT strains, and the parental Vibrio strain, biochemical taxonomic tests were performed (Baumann and Schubert, 1984) as well as phenotypic profiling using API 20E test strips (Sherwood Medical, Plainview, N.Y. ) Strains DI-9 and WJT -1 C were also typed by fatty acid analysis (MIDI, Newark, DE) and by the Biolog s y stem (Haywood CA) These strains were found to be identical to each other but they could no t be identified with a high degree of confidence to the species level (data not shown ) Transforming DNA and gene probes The plasmids used in these transformation studies were the broad host range Inc Q / P4 plasmids RSF1010, pGQ3, and pQSR50 RSF1010 encodes for resistance to streptomycin and sulfanilamide while pGQ3 and pQSR50 encode resistance to both streptomycin and kanamycin.

PAGE 62

Table 3 Bacteria and Plasmids Strain or plasmid Strain s Vibrio sp DI-9 Vibrio sp RRVP3 Phenotype or description Estuarine isolate from Davis Island, FL Spontaneous rifampin-resistant mutant of strain DI-9 Source or reference Gift of G Stewart This study 41 Vibrio sp MF-1 V i bri o sp. MF-1C Vibrio sp. MF3C DI9 naturally transformed with pGQ3 MF-1 cured of plasmid pGQ3 This study This study This study MF-1 transformed with pGQ3 and cured V i bri o s p. WJT-1 Vibrio sp WIT -1 C V paraha e molyticus USFS3420 V. p a raha e m o l y ti c us DI-9 transformed with pKT230 WJT -1 cured of the plasmid pKT230 V parahaemolyticus M F -4 USFS3420 transformed with RSF1010 V paraha emo l ytic us M F-4C MF-4 cured of the plasmid RSF1010 E. coli RM1259(pQSR50) F thi lacY thr leuB supE44 trpE5 E coli JL3700(RSF1010) F gal thijhuA endA sbcB15 hsdR4 slu!M+ E coli JlA062(pGQ3) HB101 Pla s mids RSF1010 pQSR50 pGQ3 IncQ / P-4 Sur Sm r Mob + IncQ/P-4 Sur Smr Kmr Mob+ lncQ / P-4 Sur Smr Kmr rdk+ Mob+ Jeffrey et al. (1990) This study Gift of G Stewart This study This study Meyer et al (1982) Gift of G Stewart Carlson et al. (1985) Frey and Bagdasarian (1989) Meyer et al. (1982) Carlson et al. (1985)

PAGE 63

42 RSF1010 is an 8 684 kb plasmid broad host range plasmid which was originally isolated from a terrestrial Pseudomonas sp (Guerry et al., 1974) pGQ3 is a derivative of pKT230 (Bagdasarain et al 1981) that contains the Escherichia coli thymidine kinase (tdk) gene (Carlson et al., 1985; Jeffrey et al., 1990). pKT230 is a derivative of RSF1010 and pACYC177 (Bagdasarian et al. 1981) pQSRSO is a Tn5-containing derivative of the plasmid R1162 (Meyer et al., 1982) RSF1010 and R1162 are believed to be identical. Portions of these plasmids were subcloned into the Riboprobe vector pGEM3Z or pGEM4Z (Promega Biotech Madison, Wis.). e5S]RNA probes were prepared by transcription of the subcloned fragments with T7 or SP6 RNA polymerase, using [35S]UTP (1 ,320 J.LCi/mmol; NEN Research Products Boston, Mass ) as described by Prom ega (Riboprobe system or Riboprobe Gemini system. Transcription of cloned DNA Promega Technical Bulletin 002. Promega Biotech Madison Wis 1988) The genealogy of these plasmids and the construction of the probes are shown in Figure 1. Chromosomal DNA from a spontaneous rifampin-resistant mutant of DI-9 (RRVP3) was used as transforming DNA for the chromosomal transformation assay Spontaneous rifampin-resistant mutants of the wildtype strain DI-9 were obtained by plating 100 J.Ll of a late-log phase culture of DI-9 onto artificial seawater nutrient media (ASWJP+PY; Paul 1982) media containing rifampin (500 J.Lg/ml) and single colonies selected Preparation of transforming DNA Plasmids were amplified in E. coli cultures us i ng and uridine

PAGE 64

E Figure 1. 43 5 7 kb pQSRSO 11. 9 kb 3Z p 13 2 kb Genealogy of plasmids and Riboprobe transcription vectors used in this study. The arrows inside the circles indicate directions of transcription Abbreviations : Su, sulfanilamide; Sm, streptomycin; Km, kanamycin; E, EcoRI; A, Avai; B BamHI; P, Psti; X, Xhoi; 1'7, T7 RNA polymerase promoter recognition site; Sp6, Sp6 RNA polymerase recognition site.

PAGE 65

44 as described by Maniatis et al. (1982) Large-scale plasmid DNA purification was performed by alkaline lysis as described by Griffith (1988). To further separate plasmid from chromosomal DNA, the plasmid extract was passed through a pZ523 column (5 prime __ .,. 3 prime Inc., West Chester, Pa.) (Felsenstein, 1988; Zervos et al., 1988) Plasmid multimers were prepared by cutting to completion at the unique EcoR1 site followed by ligation with T4 ligase (Frischer et al., 1990, Jeffrey et al., 1990, Frischer eta!., 1993). The degree of multimer formation was judged by visualization on a 0.4% agarose gel. Figure 2 shows a 0 4% agarose gel stained with the DNA fluorochrome Hoechst 33258 (DeFlaun et al 1987) with CCC, linear, and multimerized plasmid preparations of pGQ3. Chromosomal DNA was prepared by the method of Mannur (1961). All DNA concentrations were determined by the Hoechst 33258 method (Paul and Meyers 1982) with a Aminco model #J47440 or Perkin-Elmer model #LS-5 fluorescence spectrophotometer Culture conditions Strain DI9 and all the HFf strains were grown in artificial seawater with 5 g of peptone per liter and 1 g of yeast extract (ASWJP+ PY [Paul, 1982]) per liter. When antibiotics (kanamycin or streptomycin) were required they were added at 500 1-'g/ml or 1 000 1-'g/ ml, respectively. Selection for rifampin resistance was at 500 1-'g/ml. These concentrations were determined empirically by testing for growth of a verified pQSR50 plasmid transformant of the HFf strain MF-1 C in the presence of increasing antibiotic

PAGE 66

Figure 2. MW Kb 23. 1 9 4 6.6 4.4 2 3 2.0 ; 45 1 2 3 4 5 6 7 8 9 10 Preparation of pGQ3 plasmid multimers Hoechst 33258 agarose gel (0.4% w/v) of pGQ3 Lanes : 2, undigested pGQ3 ; 2, Hindiii-digested (linearized) pGQ3; 5-10, T4ligase multimerized pGQ3 Hindiii digested lambda molecular weight markers are shown in lane 1

PAGE 67

46 concentrations (data not shown). All antibiotics and DNase I used in these studies were purchased from Sigma Chemical Co. (StLouis, Mo.) For transformation assays ofDI-9 and HFT strains, cells were grown to an optical density of 0 8 at 600 nm at room temperature (26 + 2C) with shaking (150-200 rpm), corresponding to approximately 2 x 109 cells per ml or late log phase. E. coli strains were grown in Luria Broth (LB ; NaCl 10 g /1; Tryptone 10 g /1; Yeast Extract, 5 gll) Kanamycin and streptomycin were added at 50 and 25 J.Lg/ 1 respectively to LB when required. Filter transformation assay One ml of late-log-phase cell culture was immobilized onto a sterile Nuclepore filter (47 mm; 0 .2-J.Lm pore size) (Nuclepore Corp., Pleasanton, Calif.) under mild vacuum (10-15 mm Hg) keeping cells to a spot of no more than 1.5-cm diameter. The filter was then transferred aseptically cell side up onto an ASWJP+ PY agar plate and overlaid with 4 J.Lg of pasteurized plasmid multimers suspended to a final volume of 100 J.Ll in sterile 4.2 mM MgC12 Pasteurization of DNA was accomplished by incubation at 72 o c for 2 hours. Cells were allowed to incubate 16 to 20 hours at room temperature in the presence of transforming DNA. Following incubation, the filter was transferred to 10 ml of ASWJP+PY in a 125 ml flask and allowed to shake at approximately 200 rpm for 1 hour This was necessary to resuspend the cell mat and to allow the cells to recover before exposure to antibiotics. Cells were then serially diluted in sterile ASWJP

PAGE 68

47 (no nutrients) and plated on selective (ASWJP+ PY and 500 p.g of kanamycin per ml, 1,000 p.g streptomycin per ml) and nonselective medium (ASWJP+ PY) and enumerated after 48 to 72 hours of growth. In some cases it was necessary to enrich for transformants to increase the assay sensitivity by liquid selection. In this case the remaining cell suspension, after plating, was added to 100 ml liquid media containing antibiotics and incubated with shaking at 28C. After an overnight incubation, one ml of this culture was transferred to 50 ml of fresh media containing antibiotics and incubation was continued. Following these incubations plasmid mini-preps were analyzed for hybridization to the NPTII gene probe and individual hybridizing colonies were isolated. If transformants were detected, transformation frequencies were reported as greater then 1/total viable count from initial total cfu counts. This protocol was referred to as "liquid enrichment". Chromosomal filter transformation assays were done identically as plasmid filter transformation assays except that 10 p.g of homologous chromosomal DNA from RRVP3 was placed on the cell spot. Where specified 100 Kunitz of DNase I in 4.2 mm was added to transformation filters to inhibit transformation. Selection for transformants was on ASWJP+ PY containing 500 p.g of rifampin per mi. Verification of plasmid transformation Presumptive plasmid transformants, identified as antibiotic-resistant colonies, were transferred to charged nylon circles Due to the textural quality of the HFT strains (hard

PAGE 69

48 and flat) it was found that they were not efficiently transferred to the nylon filters Therefore, it was necessary to allow the colonies to grow on the filters until they were visible (24-48 hours). Colonies were lysed on the filters by the method of Buluwela et al. (1989), and the DNA was denatured and fixed by the method described by Maniatis et al. (1982) We found that the combination of these two procedures yielded a more distinct hybridization signal than either method alone (Figure 3). Filters were hybridized overnight with [35S]RNA probes at 42 oc essentially by the method of Church and Gilbert (1984) and modified as described by Promega Technical Bulletin 002. Filter washing consisted of one wash in 2 X SSC (0.3 M NaCl, 0.03 M sodium citrate [pH 7.0] containing 1 mM dithiothreitol) for 5 min at room temperature followed by three 60-min washes at 65 o c in PSE (0 .25 M sodium phosphate, 2% sodium dodecyl sulfate, 1 mM EDTA, pH 7 4) and three 30 min washes in PES (40 mM sodium phosphate, 1% SDS, 1 mM EDTA, pH 7.4) at 65 C Filters were dried under a heat lamp for approximately 10 min and hybridization was detected by autoradiography. Colonies were also subcultured, and the plasmid DNA was extracted by the miniprep method of Maniatis et al. (1982). Transforming plasmid DNA could be identified in the transformants by Southern blotting and probing with [35S]RNA gene probes as described above.

PAGE 70

Figure 3 49 A B Comparison of colony lift hybridization methods of: (A) Buluwela et al., 1989 and (B) Maniatis et al. 1982 Vibrio HFT strain WJT-1C was transformed with pQSR50. Transformants were transferred from selective plates to a charged nylon filter (MSI, Westboro, MA). Filters were transferred, cell-side-up, to a fresh ASWJP+ PY agar plate supplemented with kanamycin (500 JLg/ml) and streptomycin (1 000 JLg/ml) and allowed to incubate at 28C for 48 hours until colonies were visible. The filter was then cut in half. In treatment A, one half was placed on filter paper soaked in 2 X SSC (0.3M NaCl, 0.03M NaCitrate pH 7.0), 5% SDS for two minutes The dish with the filter was transferred to a microwave oven and treated for 2 5 minutes at full setting (700 watts) In treatment B, colonies were lysed by placing filter on paper soaked in 10% SDS for 3 minutes. Following these lysis procedures both halves of the filters were placed on paper soaked in 1.5 M NaCl, 0 5 M NaOH for 10 minutes, and then transferred to paper soaked in 1.5 M NaCl, 0 5 M Tris (pH 8 0) for 15 minutes. Following the neutralization step the filters were dried under a heat lamp and baked in a vacuum oven for 2 hours. Filters were probed simultaneously with the r5S] labelled probe NPTII.

PAGE 71

50 Curing of plasmids from natural plasmid transformants The pGQ3 transformants ofDI-9 (MF-1 and MF-3), the pKT230 transformant of DI-9 (WJT-1), and the RSF1010 transformant ofUSFS 3420 (MF-4) were cured of their plasmids by growth in ASWJP+ PY without antibiotics After five successive 24-hour transfers to fresh medium (approximately 60 generations), the cultures were serially diluted and plated onto non-selective medium. After growth, the colonies were replica plated to selective medium. Colonies that failed to grow in the presence of antibiotics were selected. The effectiveness of the curing process was approximated by comparing the numbers of colonies on non-selective versus selective media (Table 4). Curing efficiencies ranged from 11 to 81. 3%. Table 4 Plasmid curing of transformants 60 in media Strain Plasmid Total cfu/ml on Abr cfu/ml on % Curing non-selective media selective media MF-1 pGQ3 7 .5 (2. 1) X 108 1.4(0 .1) X 108 81.3 MF-3 pGQ3 8.3(1 3) X 108 4 3(0.1) X 108 48.2 WJT-1 pKT230 7.0(4 0) X 108 5 0(0.7) X 108 28 6 MF-4 RSF1010 1.1(0.3) X 109 9 8(3 .1) X 108 11.0 Numbers in parentheses indicate standard deviations

PAGE 72

51 DNA binding studies eH]DNA was prepared by end labelling Hindffi-digested lambda DNA, Aval digested purified pQSR50 plasmid DNA, or Aval-digested chromosomal DNA from Vibrio strain DI-9 Both Ava! and Hindiii are asymmetric six base cutters which leave five base overhangs containing all four nucleotides (Hindiii, 5'-A AGCTI-3'; A val 5' Thus all four [3H] deoxynucleoside triphosphates were used for labelling to achieve the highest possible specific activities Specific activities of labelled DNA's ranged from 0 0165 1-LCi/ ng DNA for labelled pQSR50 plasmid DNA to 0.045 llCilng DNA for lambda DNA. Aval-digested chromosomal DNA labelled to 0.0275 1-LCi/ ng DNA. All specific activities were adjusted to 0.0165 J.LCi/ng DNA by the addition of the appropriate unlabelled DNA's. Labelling was performed as described by Paul er a!. (1988) and all [3H] deoxynucleosides were purchased from NEN Research Products (Boston, Ma). To measure DNA uptake rates of DI-9 and WJT-lC, cultures were diluted to approximately 2 x 108 cells per ml in freshly autoclaved sterile filtered ASWJP+ PY in an acid-washed polymethylpentene flasks. [3H]DNA was added to the diluted culture (0.25 1-LCi/ml). Triplicate 2-ml samples were flltered onto Nuclepore filters (pore size, 0 2 J.Lm) at time intervals ranging from 0 to 4 hours After the sample had passed through the filter with mild vacuum (10 mm Hg) the filters were immediately washed with 3 ml of sterile filtered ASWJP containing 10 J.Lg of calf thymus DNA per ml to dilute labelled DNA and minimize further binding of labelled [3H]DNA. Filters were then placed in scintillation vials containing 0.5 ml of 0.5 M Protosol (NEN

PAGE 73

52 Research Products, Boston, Mass. ) to solubilize the filter. After the filters were completely dissolved, 25 of glacial acidic acid and 10 m1 of Ecoscint 0 scintillation counting fluid (National Diagnostics, Manville, N.J.) was added to the vial, and the radioactivity associated with the filter determined by liquid scintillation counting. Nonspecific binding of DNA by medium particulates was assessed in cell-free controls. Nonspecific binding generally contributed less than 10% of overall binding compared with the 30-min time points with cells present. Background counts were routinely subtracted from all experiments. Results Table 5 shows the results of plasmid multimer transformation of wild type Vibrio strains DI-9 and V parahaemolyticus strain USFS 3420. Transformation of Vibrio strain DI-9 occurred with plasmid multimers of pQSR50 and pGQ3 at frequencies of 3.5 x 10-9 and 1.44 x IQ-9 transformants per recipient, respectively (Table 5). A strain of V parahaemolyticus (USFS 3420) was also naturally transformed with plasmid DNA at a frequency of 1. 9 x IQ-9 transformants per recipient with pQSR50 plasmid multimers and 3 5 x IQ-7 with multimers of RSF1010.

PAGE 74

53 Table 5. Natural plasmid transformation of wildtype Vibrio strains. Strain Plasmid Transformation ( +/-) Transformation Frequency DI-9 pGQ3 + 1.4 x 10 9 pQSR50 + 3.5 x 109 USFS 3420 pGQ3 < 1.0 X 10"9 pQSR50 + 1.9 x 10 9 RSF1010 + 3.5 x 10 7 Transformation detected by liquid enrichment. Isolation of HFT strains Several DI-9(pGQ3) transformants were cured of their plasmids by successive growth in non-selective medium. The transformants and cured strains possessed a colony morphology distinct from that of the original (parental) strain DI-9 (Figure 4). No obvious differences between DI-9 and the HFT strains were observed by SEM (Figure 5) and no significant differences in their growth rates were measured. Generation times were measured as 1.8 hours for DI-9 and 1.5 hours for WJT-lC (Figure 6). Phenotypic profiles of all the cured strains were identical to that of DI-9 as determined by the API 20E test strip (Sherwood Medical), by the Biolog method (Biolog, Inc Haywood, CA), and by fatty acid analysis (MIDI, Newark DE). DI-9 and all the cured strains are rodshaped, motile (a single polar flagellum), oxidase positive, and able to ferment glucose

PAGE 75

Figure 4. 54 Colony morphology of Vibrio strain DI-9 (A) transformant strain WJT 1 (B), and cured strain WJT 1 C (C) Colonies were grown 7 da y s on artificial seawater agar supplemented with 5 g of pepton e per liter and 1 g of yeast extract per liter (ASWJP+ PY).

PAGE 76

Figure 5. A B Scanning electron photomicrographs of Vibrio strain DI9 (A), and WJT -lC (B). 55

PAGE 77

1.0 56 o.s tA g 0 8 e o.70.8 ..... rn d Cl) ,....... s D'l ,....... ,....... Cl) u 0.5 t-0 4 0.3 12 Time (hours) B 108 ,......,..! I I -108 -I / -! ... 107 8 10 12 0 2 4 6 Time (hours) Figure 6. Growth curves of parental Vibrio strain DI-9 ( ..-.) and HFT Vibrio strain WJT-lC (-). Optical density at 600 nm (A), and plate count s (B ) of cultures grown at 28 C Generation times of DI-9 (1.8 h) and WJT-lC (1.5 h).

PAGE 78

57 The results of natural plasmid transformation assays of the plasmid cured strains appear in Table 6 When exposed to plasmid multimers, these cured strains transformed at significantly higher frequencies than the parental strain DI-9 Transformation frequencies of the cured strains MF-lC, WJT-lC and MF-3C ranged from 9 x 1()9 to 6 7 x 1 o-s transformants per recipient with pGQ3 multimers or were 6 to 48-fold greater than the transformation rate of the parental strain DI-9. The cured strains transformed at freq u encies ranging from 1 x 10-6 to 1.5 x 10__. transformants per recipient with pQSR50 plasmid multimers or 286 to 42,857 more efficiently than the parental strain DI9 The cured RSF1010 transformant of V parahaemolyticus (USFS 3420) MF-4C also transformed with significantly higher frequencies than did the parental strain. Transformation frequencies of MF-4C with pQSR50 multimers were 3.05 x w-s or 3 1 ,443 times grea t er than USFS 3420 with the same plasmid. These strains are referred to a s h i gh-frequency-of tran s formation (HFT) strains. Plasmid transformants could be ident ified d i rectly by colony hybridization (Figure 7) or by Southern hybrid i zation (Figure 8) with the appropriate labelled gene probe. To rule out the possib i l i t y that these high transformation frequencies were caused by homology resulting from transforming plasmid DNA remaining in the HFT strains a dot blot of a total DNA preparation from the HFT strain MF-lC was hybridized with [35S]l abelled probe PJHPII (Figure 1) complimentary to pGQ3 (Figure 9) Concentrations of DNA dotted on the filter were sufficient to detect a single copy of the pla s m i d inserted into the chromosome

PAGE 79

Figure 7 "' 58 Direct identification of transformants by colony hybridization. A colony lift of potential pQSR50 transformants of WJT 1 C hybridi z ed with r ss] labelled probe NPTII.

PAGE 80

Figure 8 59 1 2 3 4 5 6 7 8 Autoradiogram of Southern transfer of strain DI 9, transformant strain WJT -1, and HFT strain WJT -1 C probed with r5S] labelled Riboprobe RNA probe pJHPl. Lanes: 1, undi gested pKT230; 2, Xhol-digested pKT230 ; 3, undigested DI 9 plasmid DNA ; 4, Xhol-digested DI-9 plasmid DNA; 5, undigested WJT-1 plasmid DNA ; 6, Xhol-digested WJT-1 plasmid DNA; 7, undigested WJT-1C plasmid DNA; 8 Xhol-digested WJT-1C plasmid DNA The faint signals i n lanes 3 and 7 are due to contamination from lanes 2 and 6

PAGE 81

Figure 9. 60 D c 8 A 1 2 3 4 Dot blot of total DNA from parental, transformant and HFT Vibrio strains Row A, DI-9 (2,000 [lane 1], 4 000 [lane 2], and 6,000 [lane 3] ng) ; row B, pKT230 plasmid (5 [lane 1], 10 [lane 2], 15 [lane 3], and 50 [lane 4] ng) ; row C MF-1 (2,000 [lane 1], 4,000 [lane 2], and 6,000 [lane 3] ng); row D, MF -1C (2, 000 [lane 1], 4,000 [lane 2], and 6,000 [lane 3 ng). Probed with [35S] labelled Riboprobe RNA probe pJHPII.

PAGE 82

61 Table 6. Natural plasmid transformation of HFf Vibrio strains Transformation HFT/wildtype Strain Plasmid Transformation Frequency ratio MF-1C pGQ3 + 0 9 tO 6.6 X 1Q8 6-47 pQSR50 + 0.2 tO 1.1 X 1Q 5 571-3,143 WJT-lC pGQ3 + 2.3 tO 6. 7 X 1Q 8 16-48 pQSR50 + 0.01 to 1.5 x 104 286-42,857 MF-3C pGQ3 + 1.3 X 10"8 9 pQSR50 + 2. 5 X 10-6 714 MF-4C pQSR50 + 3.1 X 1QS ND ND: Not Determined No hybridization was found with the parental strain DI-9 or the cured strain MF-1C, whereas strong hybridization resulted when the pQG3 transformant of DI-9 (MF-1) (Figure 8). These results indicate that no sequences from the transforming DNA remained in the cured strains Table 7 shows the result of chromosomal transformation studies with the wild-type and HFT strains. WJT IC was transformed at a frequency of 8 3 x 105 transformants per recipient (Table 7). This represents a 244-fold increase in chromosomal transformation frequency compared with that of the parent DI-9. This increase in transformation efficiency cannot be explained by the presence of plasmid DNA residing in the HFT stains, since Vibrio strain DI-9 chromosomal DNA has no homology to these plasmids (Figure 8.)

PAGE 83

Table 7 Strain Natural transformation of wild-type and HFr Vibrio strains with homologous chromosomal DNA. Treatment (Resistance Frequency) 62 CTDNA Rif DNA Rif DNA+ Transformation HFT/DI-9 DNase I Frequency Ratio DI-9 2 7 x w -7 6 1 X I0 -7 2 2 X 10"7 3.4 x w -7 WJT-lC 3.3 x w -7 8 3 X 10S 1.1 X I0-7 8.3 X lQS 244 CT DNA Calf thymus chromosomal DNA; Rif DNA, Chromosomal DNA from RRVP3; Rif DNA+DNase I, chromosomal DNA from RRVP3 and DNase I (100 Kunitz) [ 3H]DNA binding studies The resu l ts of a typical eH] DNA uptake experiment using heterologous lambda phage DNA by the wildtype strain DI-9 and its HFT derivative WJT-lC are shown in Figu r e 10. Binding rate comparisons within each experiment revealed that mid-log-phase cultures of both strains bound heterologous DNA similarly (Figure lOA), while late-logphase cultures of WJT-lC bound heterologous DNA approximately twice as fast as DI-9 during the first 30 minutes of exposure to labelled DNA (Figure lOB). Uptake rates varied dramatically between experiments, but the ratio of the short-term (30 min) uptake rates of DI-9 and WJT-lC remained fairly constant (Table 8). Incubation temperature before and during exposure to DNA may greatly affect the DNA

PAGE 84

63 2 0 1.8 A. tr.1 1.6 --(l) 1.4 T CJ --t -en 0 1.2 ....... 1.0 < 0.8 z , 0 0 6 / .. OD ..;1 c: 0 4 ..... 0 2 0 0 0 30 60 90 120 2.0 1.8 B. tr.1 1.6 A ., (l) ., 1.4 CJ ., m 1.2 0 ., ., ....... 1.0 ., < 0.8 ., z , 0 0 6 OD ' ..... c: 0.4 , 0 2 0.0 0 30 60 9Q 120 Time {min} Figure 10. Typical DNA binding / uptake curves of [3H] lambda DNA by parental Vibrio strain DI-9 (..._....) .and HFf strain WJT-lC ( ......_..) at 25 C. Mid-log phase culture (A); late-log-phase culture (B).

PAGE 85

64 uptake rate DI-9 and HFf strains will grow at temperatures ranging from 15 to 39C, with opt imum growth rates at 37C. Uptake studies performed at 28C yielded 6 to 2,383 times the uptake rate of those performed at room temperature (20 to 23 C) Midlog phase cultures of both strains possessed similar DNA uptake rates The ratio of WJT-1C to DI-9 uptake rate for mid-log-phase cells was 0.918 0.219 (0. 1 < P < 0 025, n = 5). In contrast the initial uptake rates of late-log-phase cultures were approximately two -fol d-greater (2.1 0 .59) for the HFf strain than for the parental DI 9 (0. 02 < P < 0 05, n = 4 [Table 8]). Interestingly when similar uptake experiments were conducted with labelled plasmid DNA (heterologous) or homologous chromosomal DNA, no d i fferences i n uptake was observed between the wildtype (DI-9) and HFT {WJT-1C) strains (Figure llA and liB) In fact no uptake was observed The lack of uptake could not be explained by unsuccessful labelling or experimental protocol since an unknown contamina ting strain (not Vibrio ) was able to take-up the labelled plasmid and chromosomal DNA efficiently (Figure llC). Discus sion In this part of the study a high-frequency-of-transformation (HFT) variant of the marine Vibrio strain DI-9 was identified. These strains were efficiently transformed by heterologous plasmid and homologous chromosomal DNA. A V. parahaemolyticus strain was also shown to be naturally transformable and a HFT variant of this strain was also isolated

PAGE 86

6S 0.0005 o.ooo A 0.0003 0 .0002 0.0001 0.0000 -0.0001 s 80 I'll B -0 .006 -1-=:::::::::::: I l C) Cl) 0 o .oo < z 0.002 0.000 0 us 30 s 80 0 8 o. 0.2 0 15 30 80 Time (min) Figure 11. DNA binding/uptake of [3H] chromosomal ( .-.) and pQSRSO plasmid ( ......-v) DNA by Vibrio strain DI-9 (A), HFT strain WJT-lC (B) and by an unknown contaminant strain (C).

PAGE 87

Table 8 Short-term binding rates of [3H]DNA by wild-type (DI-9) and HFf (WIT -1 C) Vibrio strains Experiment DNA b inding rate (ng of DNA/109 cells/min) no.1 DI-9 WJT-1C WJT-1C/DI 9 Mid log phase2 1 0 233 0 166 0 712 2 0 020 0 016 0.800 3 0 007 0.008 1 130 4 1.350 1 040 0.770 5 1.410 1.670 1.180 Late log phase3 6 0 0006 0.0016 2.67 7 0 0200 0.0430 2.15 8 0.0040 0.0092 2.29 9 1.1280 1.4300 1.28 Experiments 1 to 3 and 6 to 8 were performed at room temperature, and experiments 4,5, and 9 were performed at 28 C. 66 2 3 A v erage ration of WJT-1C / DI-9 = 0.918 0 219, t = 0 837(0 1
PAGE 88

67 per recipient (Jeffrey et a!., 1990). In the present study, the broad host range plasmids pqQ3, pQSR50, and RSF1010 were used as transforming DNA. The HFf strains transformed at significantly greater frequencies than the wild type with both plasmids and with homologous chromosomal DNA. Transformation frequencies of the HFT strains varied by 4 orders of magnitude, depending upon the transforming DNA employed. Use of pQSR50 plasmid multimers as transforming DNA consistently resulted in higher transformation frequencies compared to those obtained with pGQ3 multimers. This was surprising since pGQ3 and pQSR50 are derived from plasmids that are thought to be identical (RSF1010 and R1162, respectively [Christopher et a!., 1989]). Because pQSR50 carries the transposon Tn5, it is conceivable that the transposon played a role in the increased transformation efficiency observed for this plasmid. It may be speculated that illegitimate recombination events encoded for by transposition genes could have accounted for the higher frequency of transformation with pQSR50. Transposons that mediate their own conjugal transfer have been reported previously (Berg, 1989), although this phenomenon has not been reported for transformation Thus, illegitimate recombination events may have circumvented the need for homology required with normal RecA type-mediated recombination, which is believed to occur in natural transformation (Stewart and Carlson, 1986). However, transformation of WJT-1C occurred with R1162, the Tn5-free precursor of pQSR50, at an even greater frequency (Chapter 3; Table 10). Thus, in the absence of further transformation experiments utilizing a variety of other plasmids, it can only be concluded that different plasmids can have different transformation activities

PAGE 89

68 The efficiency of transformation was over two orders of magnitude greater with chromosomal DNA for the HFT variants than for the parental DI-9 strain. This indicates that the HFT phenotype was not associated with a condition that involved only efficient plasmid transformation (i.e., plasmid uptake or recircularization) but is a generalized condition. These results also demonstrate that the high transformation frequencies observed in the HFT strains were not caused by cryptic plasmid DNA either remaining free in the cell or incorporated in the chromosome of the HFT (cured strains) which could have provided DNA homology for plasmid transformation. There was no detectable homology between the transforming chromosomal DNA and the plasmid employed. The nature of the physiological difference between the HFT strain and the parental strain DI-9 that allows efficient transformation is not yet known. This variant may possess a mutation(s) that affects the DNA binding or uptake mechanism(s), because the HFT strains bound [3H] lambda phage DNA at about twice the rate of the parental strain DI-9. However, a similar difference in uptake rates did not exist when biologically active transforming DNA was studied. Furthermore, a two-fold increase in binding rate might not account for the increase by several orders of magnitude observed in transformation frequency for the HFT strains Interestingly, DNA uptake rates were considerably less than might be expected when compared to similar uptake experiments with mixed microbial populations from aquatic environments. Paul and Pichard (1989) reported [3H] lambda phage DNA uptake rates to be on the order of 40 ng DNA/lcf cells*min for mixed water column microbial

PAGE 90

69 populations from a Florida freshwater reservoir Uptake rates by the transforming strains DI-9 and WJT -1C measured in this study were several orders of magnitude lower (0.01 to 1.4 ng DNA/109 cells*min ; Table 8). Thus, compared to other aquatic microbial populations these strains exhibit a poor DNA uptake capacity. Extracellular DNA i s rich in phosphorous and organic nitrogen as well as being a source of salvagable nucleosides that are metabolically expensive to synthesize Therefore extracellular DNA is cons idered to be useful as a nutrient source for marine bacteria and is readily bound, diges ted, and imported (Beebee, 1991; DeFlaun et al. 1987; Paul et al 1988; Paul and Pi c hard 1989; R. Redfield personal communication). The inability of the transformable strains DI-9 and WJT-lC to bind/ import extracellular DNA may, therefore, suggest a link to their genetic competence If these strains do not possess the ability to salvage DNA a s a nutrient extra c ellular DNA may avoid digestion and may be available for genetic transformation. Continued investigation is be required to explore this hypothesis. Although the genetic basis of the difference between the wild type and HFT Vibrio strains has not been ident ified, it seems reasonable that the HFf phenotype is the result o f a s pontaneous mutat ion to a competent phenotype. The above observations are not consistent with a physiological condition that may render a portion of a population competent, since the high frequency of transformation and morphological phenotype are heritable and stable (no reversal of a physiological condition to the DI-9 phenotype has been observed). Further, the frequency of initial transformation of DI-9 (approximately 109 ) is consistent with the frequency of occurrence of spontaneous mutations Mutants of H e m o philus injluenzae and S pneumoniae have been isolated that exhibit enhanced

PAGE 91

70 competence (Redfield, 1991; Lacks and Greenburg, 1973). However the similarity of these mutations to the HFT mutation in Vibrio is unclear. The Vibrio strain DI-9 and the HFT derivative strains fulfill the requirements as a marine model for the investigation of natural plasmid transformation in the marine environment (i.e. a marine organism that is efficiently transformed by plasmid DNA). Additionally, the HFT strains may have applicability for assessing the propensity for gene transfer by natural transformation from a genetically engineered microorganism (GEM). Currently environmental managers are being mandated to assess the risks associated with the deployment of specific genetically engineered microorganisms in a large variety of different environments. Yet, there are few systems available that accurately predict the propensity for gene exchange to occur in natural environments The HFT strains may be useful as test organisms whereby if a GEM does not transfer a particular marker to the HFT strain under defined conditions then it can be declared safe in regards to gene transfer via natural transformation. Although natural transformation in most competent bacteria seems to be limited largely to homologous genomic DNA, many microorganisms are also transformable with heterologous plasmid DNA as well. Less specific natural transformation systems may allow for the spread of plasmids within natural microbial communities. Thus, the finding that a high frequency of transformation strain of a marine bacterium suggests that natural transformation by these strains could be a means of plasmid dissemination in marine and aquatic environments.

PAGE 92

CHAPTER 3: FACTORS AFFECTING COMPETENCE IN A HIGH-FREQUENCY-OF TRANSFORMATION MARINE VIBRIO. Introduction 71 Natural transformation as a mechanism of genetic transfer in marine environments has not been extensively studied As discussed in chapter one, the most extensive transformation studies have focused on the details of transformation in nonmarine organisms whose relevance to the transformation process in other organisms is not clear Furthermore, the effects of dynamic environmental conditions on the occurrence of natural transformation in the marine environment are largely unknown and relevant to our understanding of the process in nature In this part of the study the transformation of the HFT marine Vibrio strain WJT 1 C wa s examined w ith respect to parameters that may modulate the transformation process The development and maintenance of competence was investigated as a function of physiological status for Vibrio WJT-lC. Genetic competence is most stringently defined as the ability to take up (bind) macromolecular DNA. However, in these studies competence is defined experimentally as the ability to be transformed Comparison of competency between cultures were made by comparison of the fraction of a population which was transformed (i. e the transformation #transformants/total

PAGE 93

72 population) Transformation of WJT-IC was evaluated with respect to dissolved DNA pools in marine environments (DNA concentration and turnover times). Finally, the effects of three environmental parameters (nutrients, salinity, and temperature) were examined with respect to transformation in the HFf marine Vibrio strains Methods Strains and plasmids Bacterial strains and plasmids used as transforming DNA are listed in Table 9 The HFT Vibrio strain WJT-IC (Frischer et al., 1990) was used as the recipient in all studies. The IncQ /P4 broad host range, non-conjugative plasmids RSFIOIO, R1162, and pQSR50, were examined for their transforming activity and compared to the previously measured transforming activity of another IncQ/P4 plasmid, pGQ3 (see Table 6; Frischer eta!., 1990). RSF1010 and R1162 are believed to be identical (Frey and Bagdasarian, 1989). For full description of these plasmids see chapter two (Figure 1). The p15A/ pACYC184 derivative plasmid pSU2178 was also used as transforming DNA in the initial studies. After initial experiments the plasmid pQSR50 was chosen for all following investigations.

PAGE 94

73 Culture conditions In all experiments Vibrio strain WIT -1 C was grown in artificial seawater (ASWJP) supplemented with peptone and yeast extract (PY; Difco) at 29C except where noted Unless otherwise specified the concentrations of peptone and yeast were five and one g /L, respectively E coli strains were grown at 37C in Luria Broth (Sambrook et a!., 1989) containing appropriate antibiotics. Preparation of plasmid as transforming DNA Plasmid DNA was purified as described in Chapter 2 by alkaline lysis (Griffith 1988) and passage through a pZ523 plasmid purification column (5'-->3' Inc, West Chester, PA USA) Plasmid preparations enriched in linear or multimer forms were prepared by cutting to completion at a unique restriction site or by restriction digestion followed by ligation w ith T4 ligase, respectively (see Chapter 2 ; Jeffrey et a!., 1990; Frischer era!., 1990) Multimerization was judged by visualization on a 0.4% (w/v) agarose gel stained with the fluorochrome Hoechst 33258. In indicated experiments the composition of plasmid preparations enriched for monomer, linear, or multimer plasmid forms were assayed to determine the percent of each topological form present. To determine the composition of plasmid preparations 100 ng of each plasmid preparation was separated on a 1% (w/v) agarose gel and transferred to a charged nylon filter by Southern transfer.

PAGE 95

74 Table 9. Bacteria and plasmids Strain or plasmid Phenotype or description Source or reference Strains Vibrio WIT-1C HFr derivative of Frischer et al estuarine isolate DI-9 (1990) E coli RM1208(Rl162) F thi lac Y thr leuB supEA4 Meyer et al. (1982) E coli RM1259(pQSR50) F thi lac Y thr leuB supEA4 Meyer et al (1982) E coli JL3700(RSF1010) F gal thi jluA end A sbcB15 G. Stewart hsdR4 shdM+ E. coli JIA062(pGQ3) HB101 Carlson et al. (1985) E coli RR2718(pSU2718) JM109 This study Plasmids R1162 IncQ / P-4 Sur Smr Mob + Meyer et al. (1982) RSF1010 IncQ / P-4 sur smr Mob+ Frey and Bagdasarian (1989) pQSR50 (Rll62:: Tn5) IncQ / P-4 Sur Smr Kmr Mob+ Meyer et al (1982) pGQ3 IncQ/P-4 Sur Smr Kmr tdk+ Carlson et al. Mob+ (1985) pSU2718 cmr Martinez et al. (1988) Antibiotic abbreviations: Su, Sulfanilamide; Km, kanamycin; Sm, streptomycin; Cm, chloramphenicol

PAGE 96

75 The filters were hybridized to [35S] labelled RNA probe NPTII (see Chapter 2; Frischer et al., 1990) Filters were scanned using an AMBIS radio analytic imaging system (6scanner) with a 1031 resolution plate and signals associated with plasmid bands quantified as a percent of the total radioactivity associated with all bands. The topological composition of each enrichment is reported as the average of three samples. Transformation assays The standard filter transformation assay was performed essentially as previously described (Chapter 2; Jeffrey et al., 1990; Frischer et al., 1990), except that stationary phase (16 -20 hour growth period) cultures were used, unless noted otherwise Transformants were enumerated on ASWJP+ PY plates containing kanamycin (500 J,Lg/ml) and streptomycin (1000 J.Lg/ml). Transformation frequency is reported as the number of antibiotic-resistant c f.u. per total c.f u Where specified, transformation was inhibited by the addition of 100 Kunitz of DNase I in 4 2 mm MgC12 to filtered cells. All antibiotics and DNase I were purchased from Sigma. To determine the time required for transformation, cells from one ml of a stationary phase culture of WJT-1C (approximately 109 cells) were immobilized onto a 0.2 J.Lm Nuclepore filter and overlaid with 4 J.Lg of transforming DNA as in the standard transformation assay but, at appropriate intervals DNase I was added to inhibit further transformation. To investigate the development of competence, early log-phase cultures were used

PAGE 97

76 to inoculate fresh media and transformation was assayed as a function of growth phase using the standard assay. Growth was determined by measuring the optical density of the culture at 600 nm. Transformation was determined at discrete growth phases by allowing cell exposure to transforming DNA for two hours after which DNase I was added to the filter. To determine whether transformation was completely inhibited by the DNase treatment, DNase was added immediately following the addition of transforming DNA to replica treatments. In these controls, no transformants were ever detected The approximate concentration of cells was 109 cells/fllter. The maintenance of competence in spent media was investigated by allowing a culture of WJT-1C to incubate shaking (150 -200 rpm) at 29C for 10 days. Competence maintenance under starvation conditions was also determined by incubating washed cells in ASWJP without nutrients under the same conditions for 10 days. In the latter studies the concentration of the culture was maintained at low cell densities (approx 106 cells /ml) to simulate naturally occurring bacterial densities and to prevent nutrient scavenging In both conditions (i.e. spent or starvation media), the culture was sampled daily for 5 days and again after 10 days. Viable counts (total colony forming units / ml), direct cell counts, and the transformation frequency were determined at all sampling times. To investigate the effect of temperature on the transformation frequency, cultures of WJT 1 C were grown to late log-phase (absorbance of 0. 8 at 600 nm) at temperatures ranging from 15 to 37C and were transformed at the same temperatures as growth. For transformation assays at 4 C, cells were grown at 15 oc and were then transferred to 4 oc

PAGE 98

77 after being filtered To determine the effect of salinity media of various salinities were prepared by dilution of double strength artificial seawater (2 X ASWJP) with deionized water The lowest salinity media (6 was prepared by adding PY to deionized water Cultures of WIT 1 C were then grown and transformed using the standard filter assay at salinities ranging from 6 to 63 / oo The effect of nutrient concentration on transformation was evaluated with the standard assa y by growing and transforming cells at nutrient concentrations ranging from ASWJP+ 1% PY (.05 g / 1 P and .01 g / 1 Y) to ASWJP+200 % PY (10 g il P and 2 g il Y). Results Transforming Plasmid DNA The re s u lt s of transformat ion of the HFf Vibrio strain WIT-I C with multimers of four non-conjugative broad-host range Inc Q /P4 plasmids and the plasmid pSU2178 are shown in Table 10. pSU2178 is a derivative of the p15A replicon and pACYC184 (Redfield, 1991; Chandler 1991; Martinez et al 1988) that replicates in both Haemophilus and E. coli (Chandler, 1991). Multimers of the plasmids RSF1010 and R 1162 yielded the highest transformation frequencies, from 4.5 -9 1 x 103 transformants / recipient. Transformation of WIT -1 C with pQSR50 multimers resulted in transformation frequencies approximately two orders of magnitude lower than did

PAGE 99

78 RSF1010 and Rll62 (5.5. 92 x 105 transformants/recipient) pGQ3 multimers had still lower transforming activity (4.4.18 x 10"8 transformants/recipient), approximately three orders of magnitude lower that of pQSR50. Transformation by pSU2718 was not observed Transformation frequencies were compared by analysis of variance and multiple-range testing (Zar, 1984) It was concluded that multimers of RSF1010 and Rl162 had identical transforming activities, while pGQ3 and pQSR50 had significantly different transforming activities, both from RSF1010 and Rl162 as well as from each other (fable 10) Table 10. Transformation frequency of WJT -1 C with various plasmids Selected Transformation No of Plasmid markers frequency (SD) trials RSFlOIO Sm 9 1 (2.2) x 10 3 7 Rl162 Sm 4 5 (1.8) X 10 3 4 pQSR50 Km Sm 5 5 (3.9) X IOSb 19 pGQ3 Km, Sm 4.4 (3.2) X IOSc 2 pSU2718 Cm Not Detected 1 Antibiotic abbreviations: Km, kanamycin; Sm, streptomycin; Cm, chloramphenicol. a indicates results that are statistically identical (P > 0.5); b, indicates significant difference from the transformation frequency ofRSF1010 and pGQ3 (0.01
PAGE 100

79 Although the plasmids Rl162 and RSF1010 yielded the highest transformation frequencies, spontaneous resistance to the single streptomycin marker occurred in WJT1C at a frequency of 3.1 x 10 7 Therefore, we chose to use pQSR50 as transforming DNA in all following investigations since the rate of spontaneous mutation to WJT-1C to both streptomycin and kanamycin was below 109 (data not shown). Concentration and form of transforming plasmid DNA The amount of plasmid DNA required for transformation of WIT 1C and the relationship between the topological form of the plasmid DNA and its transforming activity was investigated. In Figure 12 an autoradiograph of a Southern transfer blot of plasmid preparations that were enriched for monomer, linear, or multimer preparations of pQSR50, respectively is shown. The Southern was hybridized to the [35S] labelled probe NPTII. The percent of each plasmid form in each of these preparations are shown in Table 11. None of the preparations were composed solely of one form of plasmid. Five bands in the three preparations were observed and correspond to forms indicated in Figure 12 (band #1 is presumed, on the basis of size, to be composed of linearized dimer plasmid molecules). The monomer preparation contained 85. 3% monomeric plasmid molecules (relaxed + supercoiled) with the remaining portion composed of linear (8. 8%) and multimeric molecules (6.1 %). The plasmid preparation that was enriched for linear molecules contained 81.3% linear plasmid molecules, 12.1% monomeric molecules, and 6.5% multi mer and dimer plasmid molecules.

PAGE 101

80 A 8 c 23130 9419 6557 Figure 12. Autoradiograph of a Southern transfer of pQSR50 plasmid preparations hybridized with e5S] labelled Riboprobe RNA probe NPTII. Untreated plasmid preparation (A), linear enrichment (B), and multimer enrichment (C). Bands 1 to 5 marked on the autoradiograph correspond to: linear dimer (1), covalently closed relaxed monomer (2), covalently closed multimer (3) linear (4), and covalently closed supercoiled monomer (5) forms of pQSR50. Positions of hindlll-digested lambda phage DNA molecular size markers (in base pairs) are indicated.

PAGE 102

81 The preparation enriched for multimeric plasmid molecules contained 73.9% multimer and d i mer plasmid molecules 20.1% linear form plasmids and 6 .1% monomer form plasmid molecules (Table 11). Exhaustive attempts to purify each of these plasmid forms intact from agarose gels were unsuccessful probably due to the large size of the plasmid molecules (14.4 to > 28. 8 kb). Tabl e 1 1 Compos i t ion of pQSR50 plasmi d preparations Plasmid Preparation Monomer Linear Multi mer Band No Plasm i d Form % of preparation (SD) 1 2 3 4 5 Linear d i mer Monomer (relaxed) Multi mer Linear Monomer ( supercoi led) 0 0 (0) 33.2 (6.1) 6 1 (2.6) 8 8 (4.1) 52. 1 (9.8) Band numbers refer to those marked in Figure 12. 3 8 (3.4) 18.0 (2.0) 0.0 (0) 0 0 (0) 2 7 (3.0) 55.9 (2.4) 81.3 (7.1) 20.1 (2. 2) 12.1 (0.3) 6.1 (2.6) S i nce we were unable to obtain sufficient quantities of pure monomer, linear, or multimers the transformation activities of each preparation enriched for these forms was assayed Transformation was observed in all preparations and increased exponentially with concentrat ion until saturation (F i gure 13). Transformation was observed at the

PAGE 103

82 ./I .... ,{ 0.0 0.5 1.0 1.5 2.0 2.5 3.0 5.5 . o AMOUNT OF DNA (,UG/Filter) Figure 13. Transformation frequency of WJT-lC as a function of the amount of plasmid DNA per filter using an untreated plasmid preparation (v), a unique site restriction enzyme digested plasmid preparation () or a plasmid preparation enriched for multimers ().

PAGE 104

83 lowest concentrations of DNA tested (100 ng/filter or approximately 0.1 fg/cell), even in the linear preparation. Except in the case where 4 0 J.Lg of monomer or multimer plasmid DNA/ filter was used the transformation frequencies were significantly different (P<0.05; i.e multimer gave higher transformation efficiencies than did monomers and monomers higher than linearized plasmid DNA). Dose response curves were generated by plotting the transformation frequency against the concentration of DNA. The slope of this line provides information about the number of molecules required to produce a single transformant (Saunders and Saunders, 1988). For example, if the slope of the curve was one, this would be interpreted as meaning that a single molecule was responsible for each transformation event while if the slope was found to be two, it would suggest that two molecules were responsible for each transformation event (i. e two hit kinetics). Since it has been shown that multimers of heterologous plasmids are required for transformation of B subtilis (Canosi et al., 1978; Mottes et al., 1979), we were interested in determining whether this was also the case for Vibrio Therefore, the log transformation frequency was analyzed as a function of the log concentration of multimers (i.e linear dimers and multimers) in each plasmid preparation (Figure 14) to determine whether the presence of multimeric molecules could account for all the observed transforming activity Similar plots (transformation frequency versus the total DNA concentration) are not shown since the slopes of these lines are identical to those shown. Slopes were calculated by regression over the linear portions of the curves The slope of the regression line generated by the plasmid preparation enriched in linear molecules was 1.74 r = 0 .99, and the slopes of the

PAGE 105

84 10/. .-0 .-CD 10-!5 --/ j CD J... ...... 10-e ./ /0 ""4 ....., Cd 10-7 /. 8 J... 0 ...... 10-8 rn s::= (!j J... E-4 10-8 0.01 0.1 1 10 Amount of oligomers (J.Lg) Figure 14. Transformation frequency of WIT -1 C as a function of mu1timeric plasmid DNA. Untreated plasmid preparation restriction digested preparation C.), multimeric plasmid enrichment (tl}

PAGE 106

85 monomer and multimer plasmid data were 1.69, r = 0.93 and 0.77, r = 0 99, respectively Comparisons of slopes were made by analysis of variance, and the equality of the slopes were made to ideal slopes of either 1 or 2 (Zar, 1984) The slope of the linear plasmid data was equal to 2 (p > .5) and not equal to 1 (0 02 < P < 0 005), indicative of two-hit kinetics. The slope of the monomer regression line (1.69) was not different from either 1 or 2 (P > 0 5), while the slope of the multi mer regression line (0.77)wasdifferentfromboth 1 (0 .05 0.5) suggesting similar transformation processes However, the slope of multimer regression line was different (P < 0 001), indicative of an alternate transformation process These results are consistent with the hypothesis that homologous recombination is required for plasmid establishment. Homology of heterologous DNA can be provided by the presence of multiple plasmid copies, supplied either as mult i ple molecules (linear or monomer molecules) or as a single multimeric molecule These plots show that the transforming activity of linear plasmid preparations has a significantly lower transforming activity than was expected due to the presence of multimers in this plasmid mixture This suggests that linear molecules inhibited the transforming activity of multimeric molecules, probably by competing for uptake sites At higher DNA concentrations (greater than 0 05 p.g multimers / filter) monomer plasmid preparations gave somewhat higher transformation frequencies than could be explained by the presence of multimeric molecules. This may indicate that monomers have transformation activ ity associated with

PAGE 107

86 them. Thus, the observed transformation from monomer enrichments are the additive results of transformation by multimer and monomer molecules. These results demonstrate that multimerization increases the transforming activity of a plasmid preparation. However transformation activity of monomer and linear molecules can not be absolutely determined from this analysis Time required for transformation In Figure 15 the transformation of WJT-lC exposed to pQSR50 multimers as a function of time is shown. Transformants were detected after one hour of exposure (9.2 x 1 o-7 transformants / recipient) but were not detected when the exposure period was limited to 30 minutes. The transformation frequency increased exponentially from one hour to four hours at which point the transformation frequency was 8.3 x 105 transformants per recipient. The transformation frequency was only slightly increased by continuing exposure overnight (22 hours; 1.4 x 104 transformants/recipient). Because these incubations were terminated by DNase I treatment these results indicate that at least the binding and uptake of transforming DNA resulting in a DNase I protected condition occurs within this time period, similar to the transformation process in other organisms (Contente and Dubnau 1979; Albano et al., 1987; Seifert et al., 1988)

PAGE 108

0 1 2 3 4 5 6 ,;t 720 22 87 Exposure time to transforming DNA (hours) Figure 15. Transformation of Vibrio strain WJT-lC as a function of length of time of exposure to pQSR50 DNA.

PAGE 109

88 Development of Competence In Figure 16 the onset of competence in cultures of WIT-1C is shown. Transformation could be detected at the earliest growth phase assayed (early-log) and increased exponentially from 3.5 x 10-7 transfonnants/recipient in early log-phase to 2. 8 X 1 o-s transfonnants/recipient in late log phase. Maximal transformation frequencies occurred during stationary growth phase (1.8 x 10-s transformants/recipient). These results indicate that competence initiates in early-log phase but is maximally expressed in cultures that have reached late-log or stationary growth phase Maintenance of The Competent State In Figure 17 the maintenance of competence in cultures of WIT-1C which were kept for up to ten days in spent rich media is shown (Figure 17a) or in starvation media (Figure 17b), respectively Total viable counts, DAPI direct counts, transformants/ml, and transformation frequency were measured over the incubation period. Competence was maintained at high levels in both spent and nutrient free media for the ten day duration of the experiments In spent nutrient media transformation frequency decreased from 5.5 x 10-s to 6.6 x transformants/recipient over the 10 day period while in nutrient free media the transformation frequency decreased from 3 .59 x 10-s to 2.1 x transformants/recipient over the same time period. In both cultures total viable counts, direct counts and the number of transformants/ml remained relatively unchanged.

PAGE 110

89 1.0 10-s A 0.8 10-4 >-0.8 () 10-6 ........ 8 cu 0 4 ;::j tT 10-e 0 cu 0 J.-4 0 2 co '+-4 10 -'7 ......... >. 0 0.0 o+.J ""'4 0 2 8 8 10 20 ""'4 ....,_) CIJ aj s 1.0 cu "0 J.-4 10-s 0 B '+-4 0.8 en C) 10_, ""'4 aj ....,_) J.-4 0.8 p.. 0 10-6 0.4 10-e 0.2 10 -'7 0.0 0 2 ... 8 8 10 20 Time (hours) Figure 16 Development of competence in Vibrio strain WJT-lC at 29C (A) and 37C (B) (e) Transformation frequency, (o) growth of culture

PAGE 111

......... A ... .... ]r 1 2 3 4 5 6 7 8 9 10 --g--g-cI I -g--g. ______ 0 _,0 0 _,0 -0 ---. .. "'/".-- L f .. . . .! . I.... ....... . ... . . B 1 2 3 4 5 6 7 8 9 10 Time (days) 109 108 107 106 105 90 104 8 rn -107 Figure 17. Maintenance of competence in Vibrio strain WJTlC in (A) artificial seawater with nutrients (5 g peptone and 1 g yeast extract per liter) and in (B) artificial seawater without nutrients. Transformation frequency (..._..); Transformants per m1 (---); total colony forming units (o--o); DAPI direct counts (.--). Bars indicate standard deviation for cells/ml determ inations (not shown where smaller than symbol).

PAGE 112

91 These results indicate that competence is maintained at relatively high levels despite a lack of growth. Effect of temperature The effect of temperature on competence of WIT -1 C after 4 or 20 hours of exposure to plasmid DNA is shown in Figure 18a. WJT-IC can grow at temperatures up to 39o c with fastest growth rates around 37C (data not shown). From 15 to 33C temperature had little effect on the maximum transformation frequency (measured after 20 hours of exposure) but, transformation frequencies measured after 4 hours increased with temperature in this range After 4 hours of exposure transformants could not be detected at either 4 or 15 C. At 37C the maximal transformation frequency was significantly lower (1.5.01 x 10-6 transformants / recipient) compared to maximal transformation frequencies at 29 o c (7 .5 7 x 10-5 transformants / recipient). This was true whether exposure to DNA was for a short period (4 hours), or for a longer one (20 hours). At 37 C competence developed equivalently to competence development at 29 but transformation frequencies were approximately 100 -fold lower during all growth phases (Figure 16). Effect of salinity The effect of salinities ranging from of 61 oo to 63o/ oo on the competency of WJT-

PAGE 113

92 1 C is shown in Figure 18b. Transformants were detected at all salinities, with lowest tnmsformation frequencies at 6 00 (3.9 X 10"7 transformants/recipient) and highest at 22 01oo (8.4 X 10"5 transformants/recipient). From 22 to 51/00 there was a small decrease in transformation efficiency to 3.2 x 105 and a further decrease to 4.85 x 10-6 transformants/recipient at 63 / oo Effect of nutrients Nutrient concentrations ranging from 1% (w/v) of standard PY concentration to 200% (w/v) had no significant effect on the frequency of transformation (Figure 18c) although higher nutrients levels stimulated the growth of both transformants and nontransformants (data not shown). Discussion The induction and maintenance of a competent state are important both for the comparison of the HFT model to other transformation models and to determine favorable conditions for natural transformation in the environment. The development of competence in the HFT Vibrio began during early-log phase and once attained was maintained for at least ten days. Furthermore, competence was unaffected by starvation conditions and could only be reduced by allowing the cells to return to exponential growth.

PAGE 114

10-4 A/.___. 93 1o- 1 o- _... / 10_., to- I I / 4 8 12 u 20 24 28 T-32 3S Temperature CJ d 10=' cT B r.. 10d 0 - 1o- s r.. 0 en 10_., d 0 50 55 so r.. E-c Salinity (ppt) c 10_, 10o 20 40 eo eo 100 120 ao uo uo zoo Nutrient concentration (" PY) F i gure 18. Effect of temperatur e salinity and nutrient concentration on the competence of Vibrio strain WJT-IC. Effect of temperature (A). Transformation determined after 4 hours(....-.), or 20 hours exposure to plasmid DNA ( v--v ) Effect of salinity (B) Effect of nutrient concentration (C). Nutrient concentration expressed as a percentage of the standard peptone (P) and yeast extract (Y) concentration (100% PY = 5 g P, 1 g Y per liter) Bars in (C) indicate standard deviation (not shown where smaller than symbol).

PAGE 115

94 This pattern is in contrast to competence development patterns in other transformation models. Competence in most other organisms appears to be a tightly regulated process, occurring only during late log or stationary growth phase for a short period of time, after which competency declines dramatically (Smith eta/ 1981; Stewart and Carlson, 1986 ; Stewart, 1989). Exceptions includeS. pneumoniae, in which a low level of competence develops in mid-log phase cultures but is not maintained after maximal competency is achieved in late-log phase (Pakula and Walczak, 1963; Tomaz and Hotchkiss, 1964), and Neisseria, in which competence is constitutive in piliated strains (Sparling, 1966). Competence in Acinerobacrer calcoacericus strains derived from strain BD413 (Juni, 1972) has recently been shown to be constitutively expressed (Williams, 1993). However, competence in this strain was not maintained longer than 30 hours in sterile river water (Williams, 1993). This work was in contrast with earlier work with A. calcoacericus strains BD413 and NCIB8250 in which competence was found only during mid-exponential growth phase (Ahlquist et al., 1980; Cruze et a/. 1979). Thus, the HFT Vibrio WJT -1 C seems to be unique with respect to competence development and maintenance Mutants of H. injluenzae and S pneumoniae have been isolated that exhibit enhanced competence (Redfield, 1991; Lacks and Greenberg, 1973). However, the similarity of these mutations with the HFT mutation is unclear since the HFf strains seem not only to exhibit enhanced competence during early-log phase but also during stationary phase (Frischer er al., 1990) Additional investigations are required to determine the nature of these mutations in H. injluenzae, S. pneumoniae, and in Vibrio.

PAGE 116

95 It may be that these marine Vibrio HFf strains do not possess transformation systems that are under the control of a specific regulon Rather, a mutation in any number of cellular functions (e.g. membrane permeability, macromolecular transport, nuclease activities, or any DNA processing function) may give rise to the competent phenotype. If this is the case then the establishment of heterologous DNA by transformation may be more common than is indicated by the rarity of competent organisms, especially if plasmids which provide selective advantages are involved. The effects of temperature, salinity, and nutrient concentration on plasmid transformation frequency in WJT-lC were minimal. Significant reduction from the maximal tran s formation frequency was observed only in conditions that rarely occur in estuarine environments. This observation suggests that estuarine environments provide conditions that facilitate plasmid transformation. Similar conclusions have been reached concerning the process of conjugation in environmental isolates (Rochelle et al., 1989; Sambri and Lovett, 1989). The stability of transformation frequency over a wide temperature range indicates that the transformation process is independent of the growth rate substantiating the observation that transformation occurs normally during periods of unbalanced or stationary growth Interestingly, competence was significantly reduced at 3rc. Cells that were grown and transformed in a variety of different salinities (1250/ oo) exhibited only small differences in transformation frequencies, demonstrating that transformation of this organism could occur equally efficiently under a wide range of salinities that might normally be encountered by this organism As with temperature and salinity, the concentration of nutrients in the growth media had little effect on the

PAGE 117

96 transformation frequency of WIT 1 C, although the nutrient level greatly affected the culture yields. The pattern of competence development and the flexibility in culture conditions which supports transformation suggests that these HFr strains could readily acquire plasmids via natural transformation under a variety of circumstances, especially during starvation or even during the seemingly dormant physiological state termed the "starvation state" (Morita, 1985). Plasmid multimers were used in these studies as transforming DNA because they contain self-homology, believed to be required for homologous recombination (Stewart, 1989; Saunders and Saunders, 1988). Enrichment for multimers in the plasmid preparations increased the transforming activity suggesting that, as in other transformation systems, homologous recombination facilitates plasmid establishment in the HFT marine Vibrio Regardless of the plasmid preparation used, the transformation frequency was dependent on the amount of DNA available and the process could be saturated, possibly indicating the existence of DNA binding sites. Alternatively saturation may be an artifact of the assay which probably limits transformation to organisms on the surface of the cell spot. The slope of lines generated by plotting transformants versus DNA concentration has previously been used to reveal information concerning the number of molecules required to produce a transformant. The comparison of transformation frequency versus the concentration of multimeric DNA in each type of plasmid preparation allowed the testing of the hypothesis that multimers alone were responsible for transformation The slope of the DNA dose response curve for the plasmid preparation enriched in linear

PAGE 118

97 molecules (1. 74) was found to have a slope equivalent to 2 implying that transformation involved two-hit kinetics while the dose response curve generated by multimer plasmid preparation (0 77), although not equal to 1, is suggestive of single-hit kinetics. The dose curve generated by the monomer plasmid preparation (1.69) was not significantly different from either 1 or 2 Interestingly, the slopes of all regression lines were less than expected (i.e multi mer < 1 and monomer and linear < 2). This may imply a systematic error, possibly due to the use of mixtures of plasmid types Alternately, a systematic slope reduction could be the result of the overgrowth of transformants or possible donor mediated transformation (Paul et al ., 1992) occurring during incubation Comparison among slopes indicated that the transforming activity of monomer and linear plasmid preparations were different than that of the multimer preparation. These observations suggest that, as in other models for natural plasmid transformation, homologous recombination is necessary for the expression of transforming plasmid DNA in the HFT Vibrio strains. Since all transformation in the linear preparation could be explained by the presence of multimers with a remaining transformation deficit of approximately two orders of magnitude, it is unlikely that linear molecules possess transforming activity It seems likely that linear plasmid molecules inhibited the transforming act i vity of other molecules, probably by competing for DNA binding sites on the cell surface Conversely, monomers may have transforming activity since not all of the transforming activity in monomer plasmid preparations could be explained by the presence of multimers. A critical question concerning the potential for natural transformation as a route

PAGE 119

98 of gene exchange in the marine environment is whether there are sufficient concentrations .of dissolved DNA in marine environments to act as transforming DNA The results from this study demonstrated transformation even at the lowest DNA concentrations used (100 ng/filter; 109 cells/filter) This concentration is within the range of DNA concentration reported for marine environments (DeFlaun et al 1987; Ogram et al., 1987) However, it is difficult to extrapolate these results to the environment since there are no estimates of the fraction of dissolved DNA that represents DNA with transforming potential and only a few estimates of the fraction of marine bacteria that are competent for natural transformation (Stewart and Cyr, 1987; Frischer, see Chapter 5) Thus, the implications are that sufficient dissolved DNA concentrations for transformation exist in estuarine waters but further characterization of dissolved DNA pools and competency in marine microorganisms with marine plasmids will be necessary before we can better extrapolate our results to the marine environment. Another important consideration concerning the potential for transformation in the environment is the comparison between the rate of degradation of DNA and the rate at which it can be taken-up by transformation The results presented in this work indicate that transformation, or at least the sequestering of transforming DNA into a DNase protected form is rapid, occurring on a time scale of minutes to hours while rates of DNA turnover in marine surface waters and sediments have been reported to be on the order of hours to days (DeFlaun and Paul, 1989; Paul et al 1987; Bazalyan and Ayzatullin, 1979 ; Novitsky, 1986; Maeda and Taga, 1974) Therefore, it may be speculated that the turnover rate of DNA is not the primary limiting factor for

PAGE 120

99 transformation in marine environments and that depending on other parameters (e.g. cell density, DNA availability, presence of inhibitory agents etc. ) natural transformation may occur readily. These studies indicate that the HFf marine Vibrio strains possess transformation characteristics not observed in other transformation systems. The transformation system of the HFT marine Vibrio enabled transformation to occur under variable conditions typical of natural environments The ability of these strains to maintain a competent state for long periods and the speed at which they can sequester DNA support our earlier observations from microcosm studies that natural transformation could allow for the transfer of low molecular weight plasmids in the marine environment.

PAGE 121

CHAPTER 4: IN SITU TRANSFORMATION IN SEAWATER, MARINE SEDIMENTS, AND IN A MARINE DEPOSIT FEEDER Introduction 100 The potential for gene transfer by natural transformation to occur in aquatic environments has not been studied extensively and has been primarily limited to sediment simulations (Graham and lstock, 1978; Lorenz et al., 1981; Aardema et a/., 1983; Lorenz and Wackernagel, 1987; Lorenz and Wackernagel, 1988; Lorenz et al., 1988; Duncan et al., 1989; Lorenz and Wackernagel, 1990; Stewart and Sinigaliano, 1990; Stewart et al. 1991; Lorenz and Wackernagel, 1991; Romanowski eta/., 1991; Paul et al 1991; Cohan et al., 1991). Recently transfer of chromosomal and plasmid markers has been demonstrated in situ in a river epilithon environment utilizing Acinetobacter calcoacericus derivatives of BD413 (Williams, et al., 1992). The high-frequency-of-transformation (HFT) marine Vibrio strains seem to be ideally suited for the investigation of natural plasmid transformation in a wide range o f aquatic environments. As previously discussed, the HFT strains are transformed at high efficiencies with a variety of broad host range plasmids and produce a unique colony morphology that enables rapid and accurate enumeration of recipients against an ambient background flora. Furthermore, plasmid transformants can be easily verified by

PAGE 122

101 molecular probing. The HFf strains have been used to demonstrate the potential for plasmid transfer by transformation in marine sediment and water microcosms (Paul et al., 1991a). Transformation was seen in sterile and nonsterile water, in sterile sediments, but not in non-sterile sediments. We have also shown that plasmids can be transferred between species by natural transformation utilizing the HFT Vibrio strains as recipients for broad host range plasmids contained in an E coli strain (Paul et al 1992). Concurrent with these microcosms studies preliminary in situ studies in enclosed containments were undertaken in water, sediment, and inside invertebrate marine deposit feeders (holothurians) to determine if the HFf system could be used in natural settings. Methods Strains and plasmids Bacterial strains used as plasmid donors or recipients, and plasmids used as transforming DNA are listed in Table 12. The HFT recipient strain JT-1 was a spontaneous nalidixic acid, rifampin-resistant mutant of WJT-1C (Paul et al., 1992). MFN-1C2 was a spontaneous naladixic acid-resistant mutant of MF-1C.

PAGE 123

102 Table 12 B acte ri a and Plasmids Strain or p lasmid P henotype or d escription Source or refere nce Strains Vibrio WJT-1C HFT derivative of Frischer et al. estuarine isolate D I-9 (1990) Vibrio MFN-1C2 Spontaneous naladixic This study acid-resistant mutant of HFT strain MF-1C Vibrio JT-1 Spontaneous n aladixic acid Paul et al. and rifampin-resistant ( 1 99 1 ) mutant of HFT s t rai n W I T -1 C E coli RM1259 F thi lac Y thr leuB Meyer et al. supE44 trpE5 (1982) Plasmids pQSR50 (Rl162::Tn5) IncQ / P-4 Sur Sm r Mob + Frey and Bagdasarian (1989) Culture conditio n s Vibrio s t rains were grown in artificial seawater (ASWJP+ PY) supplemented with peptone and yeast extract (PY) at 29 C as described in the previous chapters. E. coli strains were grown in Luria B roth (LB) at 37 C also as described earlier. Antibiotics were added as necessary Antibiotics were purchased from S igma (St. Louis, MO).

PAGE 124

103 Transforming DNA and gene probes Plasmid DNA was purified as described in Chapter 2 by alkaline lysis (Griffith, 1988) with passage through a pZ523 plasmid purification column (5'-->3' Inc, West Chester, PA, USA). Multimerization of plasmid preparations was accomplished by cutting to completion at the unique EcoRI site of pQSR50 followed by ligation with T4 DNA ligase (see Chapter 2; Frischer et al., 1990) Transformants were verified by hybridization with the ess] labelled RNA Riboprobe NPTII transcript as previously described (Chapters 2 and 3; Frischer er a/., 1990). In situ transformation assays Water column assays. Water column assays were preformed in surface water from the Atlantic, 25'N 76 51 'W (Collected 911/90 on R/V Cape Hatteras, station 4). Cells from 20 ml of a 50 ml overnight culture of WJT-lC grown in ASWJP+PY were harvested by centrifugation and resuspended in 20 ml of sterile ASWJP to remove antibiotics. Washed cells were added to 100 ml of freshly-collected seawater in 300 ml sterile Lifecell tissure culture flasks (Fenwall Laboratories, Morton Grove, 11) at concentrations ranging from 2.5 x lOS to 4 0 x 107 cells / mi. These flasks are flexible gas permeable, plastic bags and have been used in in situ transduction experiments in a freshwater lake (Saye et al., 1990) The flasks were filled with water through the attached tubing and the flasks were then sealed. Access into the bags for inoculating

PAGE 125

104 was done by way of sampling site couplers (Fenwall Laboratories) designed specifically for use with the lifecell chamber pQSR50 plasmid multimers were added to the bags (33 JJ.g DNA/L) and the bags placed in a 100 liter decktop incubator exposed to sunlight with runn i ng seawater. Bags were allowed to incubate overnight (16 to 20 hours) After incubation the contents of each bag were filtered onto a sterile 47 mm 0.2 JJ.m Nuclepore filter (Nuclepore Corp., Pleasanton, Calif ) The filter was then placed in a sterile 15 cc tube containing 5 ml of sterile ASWJP and the cells were resuspeded by vigorous mixing Indigenous and total recipient populations were enumerated on ASWJP+ PY plates (HFT recipients were identified by colony morphology) and transformants were enumerated on ASWJP+ PY containing kanamycin (500 JJ.g/ ml) and streptomycin (1 ,000 JJ.g/ ml ) Transformants were verified as previously described by molecular probing (see Chapters two and three; Frischer et al., 1990 ; Paul et al., 1991; Paul et al 1992; Frischer er al., 1993) Transformation in sediments Sediment transformation assays were conducted with sediments collected from North Shore Beach, St. Petersburg Florida and from Key Largo Florida Sediments from North Shore Beach were collected in 50 cc sterile specimen cups from the top 2 em of beach sediment that was continually submerged by approximately 1 to 3 feet of water Sediments from Key Largo were collected by SCUBA from a depth of approx i mately 5 meters. Sediments were transported back to shore in 5 gallon buckets and experiments were initiated within two hours of collection Twenty cm3 of fresh sediment were added to 45 cm3 sterile disposable conical screw-cap tubes (Sarstedt Inc, Newton, NC) and 40 JJ.g pQSR50 multimers, dissolved in one ml

PAGE 126

105 of sterile ASWJP, were added to the sediment (2 DNA/cc sediment) and thoroughly mixed Tubes were placed in running seawater tank (23C) and allowed to incubate overnight (16 to 20 hours). After incubation 20 ml of fresh sterile ASWJP was added to each tube and vortexed vigorously for 10 minutes Sediment fines were removed by gentle centrifugation from the sediment extract. Total viable bacteria were enumerated on ASWJP+PY media and transformants on ASWJP+PY with kanamycin (500 and streptomycin (1000 Presumptive transformants (antibiotic resistant colonies) were verified as transformants by molecular probing In situ sea cucumber assays. Sea cucumbers (Srichopus sp ) were collected at the Key Largo site for in situ transformation studies. Four animals were collected from Mosquito Bank, Key Largo, Florida (25'.25N 80'.65W) and transported back to the shore lab immediately Two animals were separated and placed into clean 5 gallon buckets with aeration. One animal was inoculated with 40 pQSR50 multimers dissolved in one ml of sterile ASWJP+ PY by force feeding through the mouth using a sterile one ml glass pipet. The second animal was adminstered one ml of sterile ASWJP as a control. The animals were then returned to their aquaria. After 20 hours the animals were sacrificed and the gut sediment was removed to a sterile 45 cc tube. An equal volume of sterile ASWJP was added to the tube and the sediment extracted by vigorous mixing for 10 minutes. Total viable bacteria were enumerated on ASWJP+PY media and transformants on ASWJP+ PY with kanamycin (500 and streptomycin (1000 Presumptive transformants (antibiotic resistant colonies) were verified as transformants by molecular probing.

PAGE 127

106 In situ transformation mating studies Mating studies are distinguished from the standard transformation assays by providing the transforming plasmid DNA intact in viable E coli (RM1259) donors instead of in purified form. Previous laboratory studies demonstrated transformation of HFT Vibrio strains using this system (Paul et al. 1992) In situ matings in water. in situ matings in water were conducted in the Bahamas (25 19 .07N 78 05' .33W) in a coral reef environment. One hundred ml o f non-sterile seawater was added to each of six 300 ml sterile Lifecell tissue culture flasks (Fenwall Laboratories, Morton Grove 11). Treatments included donor only (D), recipient only (R), donor+ recipient (D+R), D+R +colicin, D+R + Membrex concentrated water containing virus like-particles and D+ R + DNasei. E coli RM1259 (donor strain) and V i bri o HFT strain MFN-1 C2 (recipient strain) were grown to stationary phase in LB supplemented with kanamycin (50 Jlg/ ml) and streptomycin (25 Jlg/ ml) or ASWJP+ PY suppleme n ted with nalidix i c acid (500 Jlg/ ml), respectively. Each culture was washed in antibiotic free media and resuspended in an volume of ASWJP resulting in cultures of each strain at approximately 2 x 1
PAGE 128

107 coworkers (1992) prior to initiation of the experiment. The bags were then placed in a I 00 L tank containing seawater and transported to the reef site Bags were suspended in the water column by SCUBA divers, fastening them to a tethered rope which was kept vertical in the water column by a submerged buoy (Figure 19). All bags were suspended in 3 to 5 meters of water. The bags were retrieved by SCUBA after approximately 20 hours and transported back to the ship in the 100 L tank containing seawater The contents of each bag were collected onto a sterile 47 mm 0.2 JLm Nuclepore filter with gentle vacuum (10 15 mm Hg). Cells were resuspended off the filters in 10 ml of sterile ASWJP by shaking at 28 C for one hour. Total viable counts were determined by plating on ASWJP + PY, the E coli donor was enumerated on LB media containing kanamycin (50 JLg/ml) and streptomycin (25 JLg/ml) (K/S), the HFf Vibrio recipient was enumerated on ASWJP+ PY supplemented with nalidixic acid (500 JLg/ml) (N), presumptive transformants were enumerated on ASWJP+ PY containing kanamycin (500 JLg/ ml), streptomycin (1 ,000 JLg/ml), and nalidixic acid (500 JLg/ml) (K/S/N). All plates contained amphotericin B (5x10-6M) to inhibit eucaryotic growth. Presumptive transformants (antibiotic resistant colonies) were verified as transformants by molecular probing. In situ sedi ment assays Sediments for in situ mating assays were collected from Key Largo, Florida as described above E. coli RM1259(pQSR50) was used as the plasmid donor strain and the HFf Vibrio strain IT -1 was used as the transformation recipient. As in the water matings RM1259 was grown to stationary phase, washed, and resuspended in an equal volume of antibiotic free ASWJP.

PAGE 129

Figure 19 108 Schematic representation of in situ transformation mating experiment conducted in a Bahamas coral reef environment. Also shown is a Lifecell tissue culture flask (Fenwall Laboratories Morton Grove, IL)

PAGE 130

109 JT-1 was grown to stationary phase in ASWJP+PY containing rifarnpin (500 JLg/ml) and nalidixic acid (150 JLg/ml) at 29C Treatments included non-sterile sediment + recipients (control) and non-sterile sediment + recipients and donor In the control treatment 0 5 ml of recipient (JT-1) was mixed with 0 5 ml ASWJP and added to 20 cm3 of freshly collected sediment in a sterile 45 cc tube. Cell-to-cell mating in sediment was initiated by mixing 0.5 ml recipient (JT-1) with 0 5 m1 of donor (RM1259) and immediately added to 20 ml of sediment. The incubations were then placed into a running seawater tank (23 C) and allowed to incubate overnight. After incubation the sediment was extracted in 20 ml of sterile ASWJP. Total bacteria, E. coli donors, Vibrio recipients and presumptive transformants were enumerated on ASWJP+PY, LB (K/S), ASWJP+PY (N/R) and ASWJP+PY (K/S/N/R), respectively. As in previous experiments amphotericin B (5x10-6M) was added to all growth media to inhibit fungal growth. Presumptive transformants (antibiotic resistant colonies) were verified as transformants by molecular probing In situ sea cucumber assays. Matings in sea cucumbers, as with sediments, were done concurrently with the transformation studies at Key Largo, Florida Matings in cucumbers were initiated by force feeding freshly collected sea cucumbers with either 0 5 ml of washed HFT recipient Vibrio strain (JT-1) in final volume of one ml ASWJP (control treatment) or with one ml of a one to one mixture of JT-1 and E coli RM1259(pQSR50) (experimental treatment). After feeding, the animals were held in individual seawater aquaria with aeration for 20 hours. Following incubation the animals were sacrificed and the i r gut contents transferred to sterile 45 cc tubes. Gut contents

PAGE 131

110 were extracted in an equal volume of sterile ASWJP by vigorous vortexing and total bacteria, E. coli donors, Vibrio recipients and presumptive transformants were enumerated as described above As in previous studies all presumptive transformants were verified by hybridization with the pQSR50 probe NPTII. Disposal of all GEM-contaminated material All water, sediments, and animals which were in direct contact with either purified pQSR50 DNA or the donor strains were autoclaved before disposal to prevent the release of GEM s or the spread of the engineered plasmid pQSR50 to natural m i crobial populations Results In siru transformation assays in water In Table 13 the results of transformation assays of the HFT strain WIT -1 C done m sterile and non-sterile seawater with decreasing recipient cell concentrations are shown In sterile seawater transformation was only observed in bags containing the highest cell densities (4x107 cells /ml; transformation frequency was 3.4x10 9 transformants / recipient) while in the non-sterile bags no transformation was observed even at the highest cell density Limits of detect ion ranged from 1.3 x 109 to 2 x 10"11

PAGE 132

111 transformants/ml in sterile systems and 2.5 x 108 to 6.1 x 1() transformants/ml in the non-sterile system. Table 13 Sterile / non-sterile Non -s terile sterile Non-sterile Sterile Non-sterile Sterile Non-sterile Sterile Non -s terile Sterile In situ Transformation of WJT-1C With pQSR50 Multimers at Varying Cell Densities in Water. Transformation freq WJT-1C recipient Transformation or Oimit of (cells/ml) detected (yes / no ) detection) 4.0 X 107 NO ( <2.4 X 4.0 X 107 YES 3.4 X 109 1.0 X 107 NO ( < 4.5 X 10"8 ) 1.0 X 107 NO ( < 1.6 X 2.5 X 106 NO ( <5.3 X 2 5 X 106 NO (<3. 1 X 1.0 X 106 NO (<6. 1 X 1.0 X 106 NO ( < 1.3 X 2.5 X 105 NO ( < 2.5 X 10"8 ) 2.5 X 105 NO ( < 2.0 X 10"11) In situ tranformation assays in sediment and in sea cucumbers The results of transformation assays in sterile and non-sterile sediments as well as in the digestive tract of sea cucumbers (marine invertebrate deposit feeders) are shown in Table 14 Transformation was observed only in sterile sediments (7.1 x 108 trans formants per recipient). Transformation was not detected in non-sterile sediments and sea cucumbers. However, in these non-sterile environments, the recovery of the

PAGE 133

112 HFT recipient strain was poor; significantly reducing the transformation dectection limits Detection limits were 1.4 and 2.5 x I0-7 transformants/ml in raw sediment and sea cucumbers respectively These values are approximately two orders of magnitude lower than the transformation frequencies observed in sterile environments Table 14. Sterile / non-sterile Sediment Non-sterile Sterile Sea cucumber Non-sterile In situ Transformation of WJT-lC with pQSR50 Multimers in Sediment and in Holothurians. Transformation freq WJT-lC recipient Transformation or (limit of (cells /ml) detected (yes/no) detection) 7.3 X 106 NO < < 1.4 x w -7 ) 6 4 X 108 YES 7 1 x 10-8 4.1 X 106 NO < < 2 s x w-7 ) In Table 15 the results of transformation mating experiments conducted in a Bahamas coral reef environment are shown Total bacteria (including indigenous bacteria), E coli donors (E), and Vibrio HFT recipients (V), were determined To test whether the presence of high numbers of bacteriophage populations or donor lysis affected transformation efficiencies, concentrated marine phage samples collected from a nearby water location or colicin was added to separate treatments A DNasei control treatment was also included No transformation of the HFT recipient WJT-lC was

PAGE 134

113 observed in any of the treatments. However, in the treatment that recieved the virus sample, transformation of the ambient population was observed Transformation frequencies appeared to be quite high (3.8 x 104 transformants/ambient recipient) as verified by colony hybridization However, the hybridization signal was lost when several of these colonies were grown in culture for further analysis (data not shown), indicating that the plasmid was unstable Interestingly the majority of the hybridizing colonies maintained the ab i lity to grow in the presence of kanamycin and streptomycin desp i te the loss of hybridization signals Table 15. In situ Transformation Mating of MFN-1 C2 and E coli RM1259(pQSR50) in Water Transformation freq Transformation freq Transform or Oimit of detection) or (limit of detection) Trea t ment (yes / no) of MFN1C2 of indigenous pop EXV NO ( < 3.3 X 10'8 ) ( < 6.7 X 10 "7 ) E X V + Colicin NO ( < 5 2 X 10"8 ) ( <2. 2 X 10 "7 ) E X V + Membrex YES ( < 3. 7 X 10"8 ) 3.8 X 104 Concentrate EX V + DNasei NO ( < 2 9 X 10"8 ) ( <2. 0 X JO-<'i) Abbreviations: E, E. coli RM1259(pQSR50) ; V HFf Vibrio strain MFN-1C2 In situ transformation matings in sediments and sea cucumbers The results of transformation matings of E coli RM1259(pQSR50) and HFf

PAGE 135

114 Vibrio strain IT -1 in non-sterile sediments and in the digestive tracts of holothurians are shown in Table 16. Transformation of JT-1 was not observed in these experiments However as with the previous experiments the limits of detection was quite low ranging from 1 X 1() 7 transformants/recipient in sea CUCUmbers to 5.6 X 10 "9 in sediments. Another problem encountered in the sea cucumber system was the detrimental effect of inoculating animals with high doses of bacteria. Sea cucumbers which had been feed either Vibrio alone or both E. coli and Vibrio suffered from severe diarrhoea resulting in almost completely void bowels. Under normal situations approximately 20 to 30 cc of sediment can be collected from a healthy adult animals, however, after 20 hours incubation following inoculation with Vibrio and / or E. coli only one to two ml of runny sediment could be recovered. This significantly reduced the population size that could be examined for the presence of transformants Table 16. Environment Sediment Sea cucumber In situ transformation mating of Vibrio IT -1 and E. coli RM1259(pQSR50) in sediments and sea cucumbers Recovered Recovered Transformation freq. Transform recipients donors or (limit of detection) (yes/no) (cells/ml) (cells/ml) of JT1 NO 1.8 X 108 2.6 X 108 ( <5. 6 x NO 9 5 X 106 3 9 X 107 ( < 1.0 X 10"7 )

PAGE 136

115 Discussion In these studies the high-frequency -of-transformatio n (HFr) marine Vibrio strains were used in transformation and transformation mating experiments under more realistic conditions then have been previously tested Transformation mating studies are distinguished from the standard transformation experiments by using cells with resident plasmids as a DNA source instead of purified plasmid DNA. Ambient cell densities and realistic amounts of dissolved DNA concentrations were used in enclosed containments in natural environments Our dependance on an engineered plasmid (pQSR50) or an engineered plasmid donor strain (RM1259) prevented us from preforming experiments in open systems Transformation in water at realistic cell and DNA concentrations (lOS to 107 cells /ml and 33 J.Lg/ L, respectively) was only observed in sterile water with the highest cell concentration (4 x 107 cells /ml). As reported in previous studies, the presence of the ambient microbial assemblage in seawater inhibited the transformation efficiency of the HFT strains compared with identical treatments in sterile systems (Paul et al 1991). Similar results were seen in sediment environments. Transformation of WJT 1 C was observed in sterile sediments (7 .1 x 10"8 transformants / recipient) but not in nonsterile sediments or non-sterile sea cucumber processed sediments. However, the Vibrio recipient strain did not survive well in natural sediments. Thus, it is not clear from these experiments whether transformation was i nhibited by sediment microbial communities or if we were unable to detect transformation with sufficient sensitivity. In fact, in these

PAGE 137

116 experiments transformation efficiencies would have had to be enhanced by an order of magnitude over that in sterile sediments for us to have detected a single transformant. Plasmid transfer from intact, viable E. coli (RM1259) to an HFf recipient (JT-1) has previously been demonstrated both in filter and liquid matings as well as sediment and water microcosms (Paul er al. 1992) Generally, cell-to-cell transfer frequencies were higher in liquid than on a filter. The opposite trend was observed when transformation by purified plasmid DNA was used In these studies transformation matings were conducted in situ and at realistic cell concentrations. Plasmid transfer between E coli RM1259 and Vibrio HFf strains was not observed. However, as seen in sediment transformation experiments where the detection limits were compromised by the presence of the natural microbial community a similar effect was seen in the mating experiments The problem was further exacerbated due to the relatively greater die-off of E coli strains than the Vibrio strains Therefore, it is difficult to conclude whether transformation d i d not occur or whether our ability to detect transformation events prevented our observation of transformation. Interestingly plasmid transfer to indigenous water populations was observed when a Membrex concentrate containing marine virus sized particles was added to the system (Table 14). This suggests that plasmid transfer was mediated by virus', perhaps by increasing the amount of extracellular DNA available for transformation by promoting bacterial lysis (virus assisted transformation) Alternatively transduction may have occurred. A similar effect was not seen when colicin, which disrupts normal membrane permeability of Gram negative bacteria, especially in E coli was added In other

PAGE 138

117 studies the addition of the coliphage T2 did not enhance transformation matings between E. coli and Vibrio (Paul, personal communication). However, the hybridizing signal was not stably maintained and was lost upon growth of the presumptive plasmid recip ients The results of these studies, although preliminary, illustrate the potential for the HFT strains to be used in situ and the difficulties associated with such experiments. Of principle concern is the low limits of detection that were obtained in these studies. These detection limits appeared to be influenced primarily by the poor survival of E. coli donor strains and the HFT strains, especially in natural sediments. Also of concern was the use of genetically engineered strains and plasmids which preclude their use in open systems. In future studies it will be desirable to develop systems that can be used outside of containments and that are better suited for survival in natural marine environments. Isolation of marine plasmids and marine donor strains should improve the versatility of the HFT marine Vibrio system significantly.

PAGE 139

118 CHAPTER 5 : PLASMID TRANSFER TO INDIGENOUS MARINE BACTERIAL POPULATIONS BY NATURAL TRANSFORMATION Introduction During the past two decades many studies have demonstrated the potential for bacterial gene exchange outside the laboratory (Chapters 1 and 2). Other investigations have suggested that gene transfer events have played a significant role in bacterial evolution (Norelli era!., 1991; Dowson et al., 1990; Smither al., 1991). As a result of these studies the importance of inter-and intraspecific horizontal gene transfer and genetic recombination among bacteria is now thought to be more important in the evolution, adaptation, and ecology of natural microbial assemblages than previously believed (Thomas, 1989). However, the mechanisms responsible for gene transfer in natural environments remain unclear. Considerable circumstantial evidence exists that suggests that low molecular weight non-conjugative plasmids are spread among marine bacterial populations by natural transformation. Small non-conjugative plasmids are common in marine bacterial communities. Schutt (1990) reported screening 492 randomly isolated bacterial strains from a variety of freshwater and marine environments for the presence of plasmids.

PAGE 140

119 Anywhere between 28 to 80 percent of the isolates contained plasmids and the abundance .of plasmid bearing strains was correlated with environments most heavily impacted by human activity. Small plasmids ( < 10 Mdaltons) were commonly found in marine environments Plasmids in this size range are not believed to be conjugative. Although the transfer of plasmids is most often associated with conjugation, the transfer of small plasmids can also be accomplished by transformation Fry and Day (1990) found that 65% of endogenously isolated plasmids from the River Taff Wales were less than 10 Kb, and furthermore, one of these small plasmids (pQM17; 7.8 kb), which encoded mercury resistance, could be transferred by natural transformation between Acinetobacter strains (Rochelle et al., 1988). Several other studies have found small plasmids to be common in aquatic environments (Sizemore and Colwell, 1977; Glassman and McNicol, 1981; Burton et al., 1982; Kobori et al. 1984; Baya er al, 1986 ; Oguseitan er al. 1987; Genthner era/., 1988; Pickup 1989; Schutt 1989; Ray era!. 1991; Belliveau er al., 1991). Thus, natural transformation may be a mechanism responsible for the dissemination of small plasmids in aquatic environments In this portion of the study natural plasmid transformation of marine microbial communities was demonstrated and the distribution of competence among marine bacteria was estimated Integration of these results led to a computational model by which the magnitude of horizontal plasmid transfer by natural transformation was calculated These studies represent the first report of natural plasmid transformation of any natural microbial community and suggests that natural transformation may be an important route of plasmid transfer among marine microbial populations

PAGE 141

120 Methods Preparation of transforming DNA and gene probes Purification of plasmid DNA pQSR50 plasmid DNA was purified by alkaline lysis as previously described (Griffith 1988; Frischer eta/., 1990) followed by passage through a pZ523 plasmid purification column to remove chromosomal contamination (5' --> 3 Inc, West Chester, PA) (Chapters 1 through 4) Multimerization of plasmid preparations was accomplished by cutting to completion at the unique EcoRl site of pQSR50 followed by ligation with T4 DNA ligase also as previously described Construction of p0SR50 gene probes. Three independent pQSR50 gene probes were utilized in these studies : NPTII, pM62, and pQSR50 NPTII was constructed as described in Chapter 2 and is specific for the neomycin phosphotransferase II (nptii) gene carried by Tn5 (Figure 1, Chapter 2). pM62 was generated by subcloning the 1. 7 Kb Ava! fragment of R1162 into the pGEM4Z vector (Promega Biotech, Madison, Wis) [ 35S]RNA probes NPTII and pM62 were prepared by transcription of the subcloned restriction fragments with T7 or SP6 RNA polymerase, using [35S]UTP (1 ,320 1-'Ci/ mmol ; NEN Research Products, Boston, Mass) as described by Promega (Riboprobe system or Riboprobe Gemini system. Transcription of cloned DNA. Promega Technical Bulletin 002. Promega Biotech, Madison, Wis 1988). A third probe, homologous to the full pQSR50 molecule, was prepared by random labelling of A:vai digested pQSR50

PAGE 142

121 Ava! digested pQSR50 was labelled with ess] dCTP according to manufacturer 'instructions except that 50 ng of DNA per reaction was used and the Klenow incubation was increased from one to three hours (Prime-A-Gene, Promega Technical Bulletin 049. Promega Biotech, Madison, Wis, 1989). In all cases unincorporated nucleotides were removed from random labelled probes by spun columns (Maniatis eta/., 1982). Figure 20 shows the region of homology to pQSR50 for all probes. Transformation assays Distribution of naturally transformable marine bacteria Culturable heterotrophic bac teria (12) and 18 pathogenic Vibrio isolates from diseased fish and shellfish collected from Florida waters were screened for the abil i ty to be transformed by plasmid DNA Vibrio isolates were a gift from Drs Daniel Lim and Donald McGarry (Universit y of Sou t h Florida, Department of Biology Tampa FL) Other bacteria were isolated on artifi c ial seawater agar medium supplemented with peptone (5 g il) and yeast extra c t (1 gil) [ASWJP+ PY]. Colonies were randomly selected and restreaked to assure clonal origins. All isolates were sensitive to kanamycin and streptomycin at 500 and 1 000 respectively. The ability of these isolates to be transformed was initially determined by a modified transformation filter assay that utilized a liquid enrichment step to increase assa y sens i t i vity by one to two orders of magnitude in comparison with the standard a s say and thus, eliminated the tedious plate counting procedure.

PAGE 143

Figure 20. Restriction map ofpQSR50. The locations of EcoRI, BamHI, Hindlll, Bglll, Pstl, A val, and Narl recognition sites are shown. Size of fragments are given in base pairs. The location of Tn5 is indicated by dashed lines. Regions of probe homologies fro pM62 and NPTII are indicated neor, neomycin phosphotransferase.

PAGE 144

CJ t tl 123 0 0> ,.._ CXI -.., --N N ,.._ ,.._ N oos 000 OOS6 0006 $ 8 (! N ,.._ CXI ,.._ N It) OOSB -01 0008 M 0 $ ---lHZ OOSL CXI SV69 .QO..Q.( ---(;i -----It) OOS9 It) 0009 0 0 ;: M ooss -It) 99Z ooos ,.._ .. ., N oosv ,.._ U") c ooov 69Z 1-M 9H ,.._ ,.._ N -0 ----.., N 01 "' svz -"' ---------,.._ ,.._ .., -oos =

PAGE 145

124 In this assay one ml of an overnight culture of each of the isolates was treated as in the standard assay except that following the overnight (16 to 20 hours) incubation (cells and 4 1-Lg transforming DNA) the filter was added to 100 ml of ASWJP+PY in a 500 ml flask and allowed to incubate at 29C. After a one hour recovery period kanamycin (500 1-Lg/ml) and streptomycin (1 ,000 1-Lg/ml) was added to the flask and incubation continued Following overnight incubation one ml of the culture (whether growth was apparent or not ) was transferred to 50 ml ASWJP+ PY containing kanamycin and streptomycin and incubated another 24 hours Following this second incubation plasmid minipreparat ions from cultures were prepared according to Maniatis (1982) except that 7 m1 of culture were used. Plasmid DNA was restriction digested, electrophoresed through a 1% agarose gel, transferred to a charged nylon filter, and hybridized with the NPTII probe Strains in which hybridization signal was detected were subjected to the standard transformation assay (Chapter 2; Frischer et a/., 1990; Frischer er a/., 1993) to determine transformation frequencies Transformation of indigenous marine bacterial populations To determine whether natural marine bacterial assemblages could be transformed by plasmid DNA, microbial populations from surface and deep water, and populations extracted from sediments marine sponges, holothurians, and coral surface layer (mucus) were used in transformation filter assays. Twenty to 100 liters of water, or 1 liter of mucus from the coral surface layer were collected, concentrated to approximately 35 ml by vortex flow filtration using a Membrex Benchmark Rotary Biofiltration device (Membrex Inc, Garfield NJ) equipped with a 100 Kdal ultrafilter (Jiang er al., 1992).

PAGE 146

125 For sediment samples and sediment from holothurian guts, sediment (approximately 20 cc) was extracted by vigorous vortexing in 20 ml of sterile artificial seawater. Sponge samples (approximately 10 g wet wt) were cut into 2 cm3 square pieces and blended for 5 min with 50 ml sterile ASWJP in a 230 ml beadbeater model 909-1 (Biospects Products Bartlesville OK) surrounded by an ice bath. Remaining sponge pieces were removed by gentle centrifugation (1, 100 rpm, 1 min). For each treatment 5 ml of bacterial sus pension was filtered onto a sterile 0.2 J.Lm Nuclepore filter and transferred cell-side-up to solid artificial seawater media. Multimerized pQSR50 DNA (4 J.Lg), suspended in 100 J.Ll of 4.2 mM MgC12 was spread over the cell spot, followed by incubatin g for 16 to 20 hours. Control treatments were treated identically except that calf thymus DNA was used instead of plasmid DNA. After incubation the cells were resu s p ended in 10 ml of sterile artific ial seawater supplemented with 5 g/L peptone and 1 g / L yea st extract (ASWJP+PY) and allowed to incubate at 29 C with shaking for one hour. The suspensions were then serially diluted and plated on selective (ASWJP+ PY containing 500 J.Lgl ml kanamycin and 1,000 J.Lg/ml streptomycin) and non-selective media ( ASWJP+ PY). Verification of transformation Colony hybridization Presumptive plasmid transformants, identified as antibiotic resi stant colonies, were lifted onto sterile charged nylon filters, and hybridized with the [ 35S] labeled NPTII probe as previously described (Chapter 2; Frischer et al., 1990).

PAGE 147

126 Transformation frequencies were reported as the number of hybridizing antibiotic resistant colonies/total viable count and limit of detections were calculated as 1/viable count. Restriction analysis. Colonies which resulted in positive hybridization signals were also subcultured and plasmid DNA was purified by the alkaline lysis mini-prep procedure, as previously described (Chapter 2; Maniatis et a/., 1982). Microgram quantities of plasmid DNA from confirmed transformants were prepared by large scale alkaline lysis of chloramphenicol amplified cultures (Maniatis eta/., 1982) All marine transformant strains were found to be sensitive to chloramphenicol at a concentration of (170 J.4g/ ml in ASWJP+ PY (data not shown) However, it is not known whether the chloramphenicol treatment resulted in an increased plasmid yield in the marine strains As previously described all large scale plasmid preps were purified of chromosomal contamination by passage through a pZ523 column (5' --> 3' Inc, West Chester, PA). Plasmid mini-prep samples from potential transformants were digested to completion with restriction endonucleases, including EcoRI HindJII, and Bglll, and electrophoresed through a 1% agarose gel (tranformant MG-1 was not digested with BgTII). DNA was transferred to charged nylon by Southern transfer and detected by RNA :DNA hybridization with the [35S] labeled probe NPTII as previously described (Chapter 2). Large scale plasmid purifications were digested to completion with restriction endonucleases EcoRI, Hindiii, Bg/II, Aval, and Narl. Restriction digested DNA was electrophoresed through 1% Hoechst 33258 agarose gels (DeFlaun and Paul, 1987), and

PAGE 148

127 transferred to Zeta Probe charged modified nylon membranes (BioRad, Richmond, CA) by Southern transfer After vacuum baking at 80C for 2 hours, the filters were wet in 0 4 M Tris-Cl (pH 8.0) and prehybridized in 1% sodium dodecyl sulfate (SDS), lOX Denhardt's solution (per liter: 2 g Ficoll, 2 g polyvinylpyrrolidone, 2 g bovine serum albumin [GSA; Fraction V], 1M NaCl, 10 J.tg/ml denatured salmon sperm DNA, 10 mM dithiothreitol (DTI) sealed in a plastic bag. After three hours incubation at 65C with gentle shaking the prehybridization solution was removed and replaced with freshly denatured [35S]-randomly labelled pQSR50 probe. Denaturation was accomplished by heating to 95C for 3 minutes followed by rapid chilling on ice. Incubation was continued for 6 to 12 hours at 65C. After hybridization the probe was removed and saved for later use. The filters were washed two times in 2X SSPE (IX SSPE is: 0 2 M NaCl, 10 mM NaH2P04, 1 mM EDTA [pH 8.0] 1 mM DTI for 15 minutes each at room temperature; two times in 2X SSPE 2% SDS for 45 minutes each at 65C The filters were then briefly rinsed in O.IX SSPE at room temperature, dried under an infrared heat lamp mounted on glass plates, and exposed to Kodak XAR 2 film at -40C (Eastman Kodak Co., Rochester, NY). A restriction map of pQSR50 including EcoRI, Hindlll, Bglll, Aval, and NarJ recognition sites and the location of Tn5 is shown in Figure 20 PCR amplification of the nvtll in plasmid transformants. To further verify the transfer of pQSR50, or at least the nptll gene (kanamycin resistant determinant), PCR amplification of the nptii coding region was carried out on transformants and, in the case of transformants of isolates on the isolates before exposure to pQSR50. A 739 bp

PAGE 149

128 product was produced with PCR primers 21-mer [5'-AGC GGC GAT ACC GTA AAG CAC-3'] (#1) and 21-mer [5'-TTG AAC AAG ATG GAT TGC ACG-3 ] (#3) PCR reaction mixtures contained 200 J.!M each dATP, dCTP, dGTP, and dTTP, 10 mM Tris HCl (pH 8 3) 50 mM KCl, 7 5 mM MgC12 1.0 J.!M of each primer, 1 ng template DNA, and 2.5 units of Taq polymerase (Perkin-Elmer Cetus, Norwalk, CT.) in 100 Ill. Reaction mixtures were heated to 95 C for 5 min, followed by 40 amplification cycles (1 min at 95 C, 1 min at 55C, and 1.5 min at 72 C). Samples were held at 72 C for 10 minutes following the final amplification cycle Thermal cycling was in a Perkin Elmer DNA Thermal Cycler (Perkin-Elmer corp Norwalk CT or an Ericomp twin block thermal cycler (model TCX15A, Ericomp, Inc. San Diego, CA). Re s ult s Transformation of indigenous marine bacterial populations The res u lts o f 14 transformation experiments with microbial assembl a ges con c entrated from surface and deep water, and extracted from sediments, marine sponges holothurians and coral surface layer (mucus) samples are given in Tables 17 through 19 Colony hybridizations of selective antibiotic plates from a transformation experiment in which transformation was observed (Gulf of Mexico deep basin 24 oOO'N 7 8 3 3 W) are shown in Figure 21.

PAGE 150

Table 17. Natural plasmid transformation of indigenous marine bacterial populations (water column) Location N Shore Beach St. Pete, FL Transformation detected ( + 1 -) Transformation frequency ND Limit of detection 6.1 X 10 I O Florida shelf Gulf of Mexico 27'N 85 20'VV Presence of sequences homologous to NPTII in the indigenous population Gulf of Mexico + deep basin (1500 meters) 24 00'N 78 33'VV Atlantic sewage outfall near Miami 2547'N 80'VV 3.6 X 3.3 X 10-IO ND 7.46 X 10"9 Bahamas Presence of sequences homologous to NPTII in the 25 24' 40N 78 55 .60VV indigenous population Atlantic (1500 meters) 25'N 76 51 'VV Bahamas (Joulter's Cay) + ND : Not detected ND 1.13 X 10"9 2 2 X 10"10 5. 7 X 10"10 129

PAGE 151

Table 18 Natural plasmid transformation of indigenous marine bacterial populations (sediment) Transformation Transformation Limit of Location detected (+I-) frequency detection N Shore Beach ND 3 4 X 1Q9 St. Pete, FL N. Shore Beach ND 5.4 X 1Q9 St. Pete FL Florida shelf ND 1.1 x 10 9 ND : Not detected 130 Although antibiotic resistant colonies appeared on both calf thymus (CT) and pQSR50 plates, only those from the pQSR50 treatment gave hybridization signals to the NPTII probe. However, the RS 1162/RSF 1010 probe pM62 hybridized to both the calf thymus and pQSR50 treatment plates indicating that RSF1010/R1162-like plasmids resided in these populations prior to the addition of pQSR50 (Figure 22) Transformation was observed in five out of fourteen experiments, and transformation frequencies ranged from 3 6 X to 1.1 X 10 "9 transformants per recipient when transformation was observed In two water column samples (Gulf of Mexico surface water and Bahamas surface water) homologous sequences to Tn5 were detected in the indigenous microbial flora

PAGE 152

Table 19. Natural plasmid transformation of i nd igenous marine bacterial populations (invertebrates) Transformat ion Transformation Limit of Location de tected ( + 1-) frequency detection Bahamas ND 2.3 x w-'o Joulter's Cay (reef) 25 19' .07N 78' .33W (coral surface layer) Bahama s + 8.4 x w-9 2.9 x w u Joulter's Cay (reef) 25 19' .07N 78 05' .33W (sponge) Bahamas + 4.5 X 10-8 5.o x w -9 Cat Cay 25 31 'N 79'W (sponge) Bahamas + 4 0 X 10-S 3 o x 109 Cat Cay 25 31 'N 79 20'W (sediment for seacucumber gut) ND: Not detected 131 In these experiments both control (calf thymus DNA treatment) and the sample that was exposed to pQSR50 contained bacteria that hybridized to the NPTII probe Therefore we were not able to determine whether transformation had occurred in these samples.

PAGE 153

Figure 21. CT pQSRSO Transformation experiment with a concentrated mixed microbial population. Autoradiography of colony lifts of antibiotic resistant colonies hybridized with the [35S] labelled probe NPTII. Calf thymus DNA control(A) and pQSRSO plasmid DNA treatment (B). From water column populations; Gulf of Mexico, deep basin. Table 16. ........ (j.) N

PAGE 154

Figure 22 CT pQSRSO Transformation experiment with a concentrated mixed microbial population. Autoradiographs of colony lifts of antibiotic resistant colonies hybridized with the [3 5S] labelled pM62. Calf thymus DNA control (A) and pQSR50 plasmid DNA treatment (B). From water column populations; Gulf of Mexico, deep basin. Table 16. Transformation frequencies of these isolates ranged from > 4. 5 X 10"10 to 9. 3 X 10"9 transformants/recipient with pQSR50 plasmid DNA These frequencies are comparable to those observed for the Vibrio parental strain of the HFT strains (DI-9). ,__ w w

PAGE 155

134 Distribution of naturally transformable marine bacteria The results of screening 26 marine and 4 subsurface soil bacterial isolates for the ability to be transformed by pQSR50 appear in Table 20. Three of 30 (10%) of the isolates were found to be competent for heterologous pQSR50 plasmid DNA. Transformable isolates were assayed to determine transformation frequencies with pQSR50 (Table 21). Transformation frequencies ranged from >4. 5 X 10"1 0 to 9.3 X 1() 9 transformants / recipient with pQSR50 plasmid DNA. Several of the transformable isolates were identified by metabolic and fatty acid (MIDI Newark, DE) analysis and the Biolog system (Haywood, CA) and included members of the genera Vibrio, and Pseudomonas (Table 21). Analysis of plasmid DNA from transformants Restri ction analy s i s To further verify transformants, plasmid DNA from isolates bef ore and aft e r transformation and transformants from mixed environmental samples were subjected to restriction analysis The plasmid pQSR50 is linearized by EcoRI while HindU! generates a diagnostic 3 3 Kb fragment and Bgni generates a 2.8 Kb fragment which hybridizes to the NPTII probe The location of these restriction sites are shown in Figure 20 Figure 23 shows Hindiii restriction digestions o f 10 transformants from the Gulf of Mexico deep water sample (Tl -T9) as well as two kanamycin and streptomycin resistant colonies (CT-1 and CT-2) that grew on the calf thymus control

PAGE 156

Table 20 Distribution of marine bacteria which can be transformed by plasmid DNA Source or Number of Number of location isolates transformants St. Petersburg Pier, FL 7 1 N Shore Beach 1 1 St. Petersburg, FL Diseased fish and 18 0 shellfish DOE subsurface program 4 1 (Savana River site)* Total 30 3 (10%) Soil subsurface isolate 135 treatment from the same experiment. Seven of ten pQSR50 presumptive isolates hybridized to the NPTII probe, however, the expected 3.3 Kb Hindiii diagnosti c fragment was not seen in any of the transformants Interestingly, continued culture of these transformants resulted in the loss of the hybridization signals in all but one (T9) of these transformants (data not shown). Restriction analysis of other transformants indicated that this apparent loss of restriction sites in transformants was not unusual.

PAGE 157

Table 21. Strain DI-9 USFS3420 14 SJl-T GH131 Transformation of marine and subsoil isolates Identity Vibrio sp. V. parahaemolyticus Unknown Unknown Ps. cepacia Transformation frequency 3 5 X 10'9 1.9 X 10'9 9.3 X 10'10 > 4 5 X 10'10 nd Transformation detected only by liquid enrichment culture. nd: Not done. 136 Restriction profiles of all transformants from natural populations were different than the original plasmid (with the exception of the EcoRI digestion), whereas transformation of isolates in culture resulted in either maintenance or alteration of the expected restriction profile (Figure 24). The undigested plasmid profiles were faint and indicative of a higher molecular weight form than the native plasmid, which displayed supercoiled and relaxed forms However, in all cases digestion with EcoRI resulted in the expected 14.4 Kb fragment diagnostic for the linearized pQSR50 molecule.

PAGE 158

MW (Kb) 23.1-+ 9 .4-+ 4 .3-+ 2 .3-+ 2 .0-+ Figure 23 0 I{) 0:: CJ) a a. or-N 0 0 1-1-N () () (t) oct 11-11-10 Cl:) ,....
PAGE 159

Figure 24. 138 pQSRSO 14T pQSRSO S J 1 SJ1T A kb 231 ... 94 ... 6 5 4 3 -.. pQSR50 M G 1 pQSR50 BS 10 c i I i I i I J i I J D oll Autoradiographs of restriction digestions of purified pQSR50 DNA and plasmid DNA from pQSR50 transformants. Water co l umn isolate 14T (A), sediment isolate SJl and SJl transformed (SJl T; B), MG 1 (Bahamas Joulter's Cay Table 16; C), and BS-10 (Bahamas Joulter s Cay sponge Table 18; D) Undigested EcoRl Hindiii, and B gni digestions are s h own Hybridi z ed with [3 5S] labelled NPTII.

PAGE 160

139 To determine the extent of the differences in restriction digestion profiles between pQSR50 and plasmid DNA purified from transformants, large scale plasmid purifications of transformants were prepared and digested with enzymes EcoRI, Hindiii, Bgni, Ava!, and Narl. Several of the transformant strains contained multiple plasmids (notably strain MG-1; Figure 25) and thus it was necessary to hybridize plasmid digestions to the complete pQSR50 probe to highlight pQSR50-like plasmids against the background of multiple plasmid bands. Figure 26 shows autoradiographs of restriction digests of plasmid DNA purified from the marine pQSR50 transformant strains SJl-T, BS-10, MG1 T9, and 14 T. As seen when the 1. 8 Kb NPTII probe was used single bands were observed in the EcoRI, HindU!, and Bgni digestions indicating that these enzymes did not generate the expected Tn5 internal diagnostic fragments. Ava! and Narl, which recognize sites within Tn5 as well as in the R1162 portion of pQSR50 (Figure 20) generated multiple fragments which differed from those observed in pQSR50. As previously observed (Figure 24a) transformant 14T appeared to be identical to pQSR50, although additional restriction fragments were observed in the Narl digestion (Figure 26a). These variant fragments may be due to the hybridization of labelled pQSR50 to endogenous plasmid or nuclear DNA present in strain 14T, alternatively, these fragments may be attributed to incomplete digestion. Digests of transformants SJlT, BS-1 0, MG1, and T9 were all unique and differed from pQSR50 digestion patterns (Figure 26 b-e). The dominant restriction fragments of strains SJ 1-T (transformant of a Tampa Bay sediment isolate) and BS-10 (a transformant from a Bahamas sponge) were identical

PAGE 161

MW (Kb ) 23. 1 9.4 6 .5 4 3 a: 0 (.) 0 ::J w pQSR50 '6 (ij = c: 0 J: III 140 MG-1 (ij as (,) --aH CIS z :::::J l: III z Figure 25. Hoechst 33258 gels (1%) o f restriction digestions of purified pQSRSO DNA and plasmid DNA from pQSRSO transformant MG 1 (Bahamas, Joulter's Cay, Table 16

PAGE 162

A Kb 23 1 9 4 65 23 2 0 Figure 26. p Q S RSO 1 4 T pQSR5 0 pQSRSO 8 Kb 23 1 9 4 6 5 T9 141 SJ1 T Autoradiographs of restriction digestions of pQSR50 and pQSR50 transformants Water column isolate 14T (A), sediment isolate SJl-T (B), MG-1 (Bahamas, Joulter s Cay Table 16) (C), BS-10 (Bahamas Joulter's Cay sponge, Table 16) (D), and T 9 (Gulf of Mexico, Table 16) (E)

PAGE 163

142 although SJl-T also appeared to contain the expected internal diagnostic Avai and Nar! fragments that were not present in BS-10 (Figure 25b,c) This analysis indicated that at least in strains BS-10, MG-1, and T9, altered restriction sites are not only associated with the Tn5 portion of the molecule but also within the R1162 portions since R1162 internal Avai fragments (doublet 2790 2890 bp) and Nar! fragment (5764 bp) were not present i n these strains. Location of transformed DNA. Since the uncut plasmid bands seen during restrict ion analysis of the transformants resulted in faint high molecular weight bands it seemed possible that the pla smid or transposon had integrated into the host s chromosome. Furthermore, a chromosomal insertion may account for the loss of restriction sites in Tn5, although if this is the case the mechanism responsible for the loss of internal restriction sites is unclear To localize the transforming DNA a plasmid prepar a tion purified of chromosomal contamination, and a total genomic DNA preparation (Marmur, 1961) of the transformant T9 were compared by hybridization DNA was dotted on a nylon membrane at concentrations sufficient to detect a single chromosomal insertion and hybridized to the NPTII probe (Figure 27). Hybridization signals were associated with the plasmid preparation, and were virtually identical in intensity to the hybridization signal seen when purified pQSR50 DNA from E. coli RM1259 was compared Only faint signals were detected in the total genomic prep, even when 1,000 ng of DNA was dotted Faint hybridization signals associated with the total genomic fraction are consistent with the presence of plasmid DNA in these samples. Thus it appears that the hybridizing signals are extrachromosomal.

PAGE 164

Figure 27. 0> c 1.0 0> c 0 1.0 C\J 0> c 0 T"" 0> c 0 0 1.0 0> c 0 1.0 0> c 0 0 0 T"" 0> c 0 0 T"" 0> c 0 1.0 C\J pQSR50 T9 plasmid T9 chromosome Autoradiograph of a dot plot of purified pQSR50, total genomic DNA preparations of pQSR50 transformant T9 and purified plasmid DNA from T9 (Water column population; Gulf of Mexico, deep basin, Table 16). Plasmid DNA's were dotted in concentrations ranging from 5 to 250 ng per dot and chromosomal DNA was dotted at 250, 500, and 1,000 ng per dot. Hybridized with e5S] labelled NPTII. w

PAGE 165

144 PCR amplification Figure 28 shows PCR amplification of the nptii coding region of Tn5 from transformants from natural populations and isolates before and after transformation PCR primers numbers one and three were used to amplify the 739 bp product (Table 24). The expected PCR product was produced from all transformants, but was not seen in the parental Vibrio strain DI-9 or the isolate SJI previous to exposure to pQSR50 These results corroborated our previous findings which had indicated that transfer of pQSR50 (or at least Tn5) had occurred, despite the altered restriction profiles exhibited by several transformants. Discussion These studies focused on the detection of natural plasmid transformation in the marine environment independent of the HFT Vibrio strains. Transformation of mixed natural microbial communities from water and extracted from various marine invertebrates (sponges and holothurians) was shown. Interestingly no transformation was observed from sediment samples, corroborating the results of previous studies which suggested that transformation is not favored in sediment environments (Paul et al 1992; Chapter 4). Although previous studies have speculated on, defined, and demonstrated the potential for natural plasmid transformation in aquatic environments this is the first report of natural plasmid transformation of any mixed microbial population.

PAGE 166

0> I 0 1-.;t ,.... ,.... ...., (/) I-I ,.... ...., (./) 0 ,.... I (/) CD 0> I145 Figure 28 Autoradiography of PCR products of nptii coding region from purified pQSR50 and plasmid DNA from pQSR50 transformants PCR primers used were #1 and #3 (Table 23 and Fig u re 30). Hybridized with [35S] labelled RNA probe NPTII.

PAGE 167

146 The property of genetic competence has been shown to be widespread among representatives of all major taxa of prokaryotes and archaea, yet only a few representatives of each taxa have been shown to be competent (Saunders and Saunders, 1988) A systematic survey for competence among the prokaryotes has not been done but the widespread distribution of competence among prokaryotes suggest that natural transformation is an important microbial process rather than a curiosity of a few species To determine the potential importance of natural transformation as a route of gene transfer among environmental bacteria the ability of 26 marine and 4 subsurface soil bacterial isolates to be transformed by plasm i d DNA was examined Ten percent (3 of 30) were found to be transformable with heterologous plasmid DNA. In a similar study Stewart and Cyr (1987) screened 95 marine bacterial isolates for the ability to be transformed by heterologous chromosomal DNA They reported that 16% (15 of 95) could be transformed. If these estimates of the abundance of naturally transformable environmental bacteria are correct then the importance of natural transformation as a route of horizontal bacterial gene exchange is likely to have been substantially underestimated. The success of these studies is most likely due to the high sensitivity of the filter assay. Optimization of the transformation filter assay with the HFf Vibrio system (Chapters 1 and 2) and the use of state-of-art cell concentration techniques, particularly from water column environments (Jiang et a/., 1992), contributed greatly to the sensitivity of the assay It may be that if the assay sensitivity were increased further, transformation would have been observed in all samples examined

PAGE 168

147 Restriction analysis of purified plasmid DNA from all transforrnants demonstrated that the restriction profiles from transforrnants from the natural populations were different than the original plasmid, while transformation of isolates in culture resulted in either maintenance or alteration of the expected restriction profile (Figure 23, 25). Differences in restriction profiles indicated that sites in the R1162 and Tn5 regions were affected by the transformation process Thus it seems apparent that the plasmid (pQSR50) underwent some alteration during the process of transformation (probably rearrangement or host restriction modification) which altered its restriction profile. These hypothesizes are explored further in the following chapter Significance of natural plasmid transformation in the marine environment An important question in the field of environmental bacterial genetics is the contribution made by different gene transfer mechanisms to the genetic composition of microbial communities Therefore, it is important to estimate the magnitude of gene transfer rates Additionally, predicting the probability of a gene transfer event occurring is of concern to those interested in using genetically engineered microorganisms (GEM's) in the environment, and to those responsible for assessing the associated risks/benefits of this technology As discussed in the introductory chapter, the potential exists for engineered sequences to be transferred to natural microbial populations via a gene transfer event and thereby alter the genetic composition of the microbial community The introduction of new genes into natural microbial communities can be considered a

PAGE 169

148 form of "genetic pollution" (Paul et al., 1991). Genetic pollution can be defined as the introduction of new genetic material or the transfer of genes in the environment resulting from or related to anthropogenic activities. In contrast to other forms of pollution, genetic pollution, once established in a component of an ecosystem, has the capacity of self-propagation. The contribution of natural transformation to bacterial horizontal gene flow is generally considered to be trivial in comparison to conjugation. However, a quantitative estimate of gene flow by natural transformation (or transduction) has not been made. By applying the results of this study (transformation frequencies of marine isolates and our estimate of the distribution of competence among marine bacteria) to a simple mass action model of plasmid transfer by conjugation (Smets et a/ 1990; Rittmann et al 1990) an estimation of the rate of formation of transformants can be computed. There are two components to the overall rate of gene transfer : the rate of formation (T F) and the rate of loss (T J. However, these rates cannot be assumed to be constant. For example, the loss rate is likely to be influenced by the change in fitness associated with a transformation event while the formation rate may be influenced by the number of potential transformation donors. However, with respect to the experiments conducted in these studies, the rate of plasmid loss (T J is considered to be negligible since the experiments were conducted for short time periods under strong selective pressure for plasmid maintenance. Similarly, the increase in numbers of plasmid donors (transformants) during the experiment is likely to be negligible due to the short incubation periods. Thus the overall transformation rate can be approximated by a single

PAGE 170

149 equation describing the transformation formation rate (f F) Mass balance models, resembling enzyme kinetic models, have successfully been applied to conjugation studies (Smets et al. 1990; Rittmann et al., 1990). These models have defined the formation rate of transconjugates in terms of the number of donors and recipients and a kinetic transfer coefficient (equation 5 .1; Smetts et al., 1990). (5.1) Adaptation of this model for transformation can be made by substituting the number of plasmid donors for a concentration of transforming DNA. Thus, the transformation formation rate can be expressed as a function of a kinetic transfer coefficient CKt1 ) the concentrat ion of transforming DNA (D) and the number of competent recipient bacteria (R) as shown in equation 5.2 (5.2) Similar to a dissociation constant (Km) in a Michaelis-Menten expression the 1(.1 can be expressed as the slope of a transformation frequency versus DNA concentration curve (Figure 29). Kinetic coefficients may also be empirically determined by back calculation from transformation experiments with marine isolates and from successful transformation experiments with mixed marine microbial populations For example, if in a typical transformation filter assay (one ml of a stationary culture [approximately 109 cells/ml] immobilized on a filter, incubated for 24 hours in the presence of 4 J.i.g transforming DNA) a transformation frequency of 3.5 x 109 was observed, the apparent transformation formation rate would be: 3 5 transformants per ml per day or 3500 transformants per liter per day.

PAGE 171

= 0 Slope Cone. DNA 150 Figure 29. Ideal transformation curve. The Kinetic Transfer coefficient (KJ represents the slope of the transformation frequency versus the concentration of transforming DNA.

PAGE 172

151 From this apparent TF can be derived as shown in equation 5.3 and calculated in 5.4. = Tp/DR. (5.3) = 3500 transformants/L*day (5.4) [4000 ng DNA/0 .001L]*[109 cells/0.001L] = 8. 75 x 10'16 transformants L I ng DNA cells day Based on an average of transformation formation rates of marine isolates in this study (1700 transformants per liter per day) an average value of 4.25 x IQ-16 is calculated. Based on this transfer coefficient and average DNA and bacterial concentrations found in estuarine environments the rate of transformant formation ranges from 1 to 5 x 104 transformants per liter per day. Transformation rates (TF) were calculated based on the distribution of competence among isolates determined in this study (10%; Table 19), average extracellular DNA concentration, and bacterial direct counts found in Tampa Bay, Florida were taken to be 4 to 14 flg DNA/L and 2 to 4 x 109 cells / L, respectively (DeFlaun et al., 1987). Similarly, transfer coefficients can be calculated from the transformation frequencies observed when mixed microbial populations were used as transformation recipients. In this case transformation coefficients ranged from less than 1.33 x 1
PAGE 173

Table 22. Estimates of natural transformation and conjugation kinetic transfer coefficients and formation rates from filter assays Kinetic Transfer Transformation Estimated Transformation Transformation Coefficient (K11) 1 Formation Rate (T r)2 Rates per Year in Recipients Tampa Bay, FI Natural Transformation 4.2 x 10 6 1 to 5 x 1Q-4 1 to 6 x to Competent marine isolates > 1.3 X 10-16 to > 5.3 X 10-.'i to 6 x 1010 to Mixed marine t.o x 10-2 1.5 2 x 1015 heterotrophic populations t.4 x 10-11 5 .5(3.9) x 10-.s 6 X 1010 HFf marine Vibrio strain WJT -1 C Conjugation 104 to 10-9 Not Available Not Applicable Bacteriodes ruminicola, E. coli Antibiotic resistant coliforms 1 Kinetic Transformation Coefficient (Kn) expressed as Transformants*LI(ng DNA *Recipient*Day). 2 Transformation Formation Rate (Tr) expressed as Transformants/(L *Day). Reference This Study This Study This Study Smets et al., (1990) ....... VI N

PAGE 174

153 transfer coefficients obtained in this manner tend to be lower than those observed in liquid assays (Smetts et al 1990). Thus, our estimation of the transformation coefficient may have lead to an underestimation of the transformant formation rate in the marine water column Alternatively, this calculation assumed that all dissolved DNA has the ability to transform all competent cells which is unlikely. Moreover, our estimate of was obtained under laboratory conditions which may not be representative of in situ conditions Conjugation transfer kinetic coefficients have been reported to be in the range of 105 to 10-9 in similar filter assays (Table 22) and comparison of plasmid transformation rates to conjugation rates indicate that the rate of plasmid transfer by natural transformation is three to nine orders of magnitude smaller than conjugal transfer rates. However, when transformation rates are extrapolated to the ecosystem scale (e g an estuary such as Tampa Bay), transformation rates of 6 x 1010 to 2 x 1015 per year are calculated. Considered at this scale transformation has the capability to alter the genetic structure of marine microbial communities, particularly if the transfer involves plasmids which confer significant selective advantages to a recipient. These studies represent the first report of natural plasmid transformation of any natural microbial community and suggests that natural transformation may be an important route of plasmid transfer among marine microbial populations Furthermore, these studies report an initial estimate of the distribution of natural competence among marine microbial flora which have been used to estimate the magnitude of plasmid transfer rates by natural transformation in a marine environment.

PAGE 175

154 CHAPTER 6: NATURAL TRANSFORMATION AS A POSSIBLE MECHANISM OF PLASMID EVOLUTION Introduction Plasmids represent a group of accessory extrachromosomal genetic elements consisting of circular covalently closed double stranded DNA. Plasmids are primarily a prokaryotic feature but have been observed in eukaryotes as well. However, eukaryotic plasmids are not believed to be as mobile as thei r prokaryotic counterparts (Esser et al., 1986; Hildebrand et a/., 1991). Plasmids have two essential characteristics ; they are self-replicating and they control their copy number in a host cell. Because of their small size, ability to self-replicate, mobility, and their non-essential nature, plasmids are considerably more flexible (genetically) than the bacterial chromosome. As long as the replication functions remain, additional genes can be added, deleted, or shuffled with, in most cases, only minor effects on the plasmid or the bacterial host. Indeed, these characteristics have facilitated their exploitation as vectors in laboratory gene cloning protocols As in the laboratory, plasmids are considered to be the major vehicle for the transfer of genes between bacteria in nature, and the transfer of plasmids in the environment plays a vital role in the evolution of bacterial populations {Thomas, 1989).

PAGE 176

155 In theory, any gene can be transferred via a plasmid although some genes seem to be more readily transferred than others (e g antibiotic resistance versus 16S rRNA genes). Transposable elements play a crucial, but not necessarily exclusive, role in this process. Genes that are transferred may be only transiently plasmid associated or may become a permanent component of the plasmid genome. The genes, that have come to reside on plasmids confer a variety of phenotypes on their bacterial hosts, ranging from bacterial resistance and toxin production to symbiotic and metabolic determinants As previously discussed (Chapter 1), plasmids are commonly found in many natural and diverse microbial assemblages Similar, but non identical, plasmids are frequently isolated from different organisms and environments, suggesting a close genetic relation between such plasmids It has been speculated that such plasmids arose as the result of gene transfers between plasmids (Cohen, 1976; Datta and Hughes 1983 ; Helinski eta!., 198 5) In the previous chapter evidence was presented that demonstrated the transfer of the plasmid encoded neomycin phosphotransferase gene (nptii) to marine bacteria by natural transformation. However, in several instances we were not able to recover a plasmid identical to the transforming plasmid (as judged by restriction endonuclease profiling). In these studies two hypotheses that could explain these observations were pursued: host-dependant restriction modification, or the occurrence of genetic changes in the transformed plasmid

PAGE 177

156 Methods Bacterial strains The strains and plasmids used in this study are listed in Table 23 All strains, excluding the E coli strain RM1259 (pQSR50), are marine bacteria that have been transformed by pQSR50 (Chapter 5) Preparation of DNA and gene probes Purification of plasmid DNA Microgram quantities of p l asmid DNA from transformant strains and E coli RM1259(pQSR50) were prepared by large scale alkaline l y sis of chloramphenicol (170 1-'g/ml) amplified cultures as previously described (Chapter 5). Plasmid preps were further purified by passage through a pZ523 column to remove chromosomal contamination (5' --> 3' Inc West Chester PA) Construction of pOSR50 gene probes. In addition to probes NPTII and pM62, described in the previous chapters (Chapters 2 and 5), probes homologous to the inverted repeat of Tn5 (IRJ and the entire pQSR50 plasmid were prepared. Both of the probes were constructed by the random priming method (Feinberg and Vogelstein, 1983; Feinberg and Vogelstein, 1984) IRL was produced by labelling of a PCR amplified fragment spanning the left inverted repeat region of Tn5 in pQSR50 (Figure 30 and Table 23) PCR primers 21-mer #4 [5'-TCT AGA CGT GCA ATC CAT CIT GTT CAA-3']

PAGE 178

157 Table 23. Bacterial strains Strain Description Source or reference SJ1-T pQSR50 transforrnant of This study (Chapter 5) unidentified Tampa Bay sediment isolate. BS-10 pQSR50 transforrnant of This study (Chapter 5) unidentified bacterium from a Bahamas sponge MG-1 pQSR50 transformant of This study (Chapter 5) unidentified bacterium from surface water collected in a mangrove environment, Bahamas T9 pQSR50 transformant of This study (Chapter 5) unidentified bacterium from deep water collected in the Gulf of Mexico 14-T pQSR50 transformant of This study (Chapter 5) unidentified Tampa Bay water column isolate. E. coli RM1259 F thi lacY thr leuB Meyer era!. (1982) (pQSR50) SupE44 trpE5 and 18-mer #11 [5'-GGT ACC GAT CCT CGC CGT ACT GCC-3'] were used to amplify purified pQSR50 DNA. Primers 4 and 5 contain Xbal and Kpnl restriction sites at their 5 ends, respectively. PCR amplification was accomplished as previously

PAGE 179

158 described (Chapter 5) The amplified product was labelled b y the random priming method with digoxigenin (DIG) dUTP follow i ng the instructions provided by the Genius non radioactive random prime DNA labelling kit (Boehringer Mannheim, Indianapolis, IN) Avai digested pQSR50 plasmid DNA purified from E coli RM1259 was labelled with [35S]-dCTP by the random priming method according to manufacturer instructions except that 50 ng of DNA was used per reaction and the Klenow incubation was increased from one to three hours (Promega Biotech Technical Bulletin number 049, 1989) In all cases unincorporated nucleotides were removed from randomly labelled probes by spun columns (Maniatis et al 1982) The region of homology to pQSR50 for all probes is shown in F i gure 30. Deletion mapping To identif y and map possible deletions responsible for the apparent loss of the Hindiii and B g lii restriction sites in transformants 11 PCR primers were designed to ampli f y the nptii gene and left inverted repeat region of Tn5. PCR reaction conditions were as described in Chapter 5. Primer sequences are listed in Table 24 and located on the pQSR50 map (Figure 30)

PAGE 180

Figure 30. Restriction maps of pQSR50, Tn5, and Left Inverted Repeat region of Tn5 (IRL) with the neomycin phosphotransferase gene. Regions of probe homology Arrows indicate the location and direction of PCR primers used in these studies. Abbreviations: Km, kanamycin; Sm, streptomycin, neoR, neomycin phosphotransferase.

PAGE 181

pQSR50 (14.4 Kb) Tn 5 (5.7 Kb) IRL + neoR Tn 5 R1162 Pst I Hind Ill Pst I Bam H I Bg/11 Pst 1 EcoR I fccR I 8gf II Sail Hind 111 P.tt I P.fr I .............. Km ****** IRL Pstl NPT II Hlrd Ill Bgtll -------pM62 Pstl5811 BsmHI neoR Sm BgtiiHind Ill Rsal Psrl Hindlll Bglll (11) (10) (9) (8) (7) _.. (3) --(4) (6) (5) (2) (1) --

PAGE 182

161 Table 24. PCR primers Primer# + I-* Primer sequence (5'-->3') bp Location 1 AGC GGC GAT ACC GTA 21 3420-3440 AAG CAC 2 GGT CTA GAT CGC CCA 29 2964 2984 ATA GCA GCC AGT CC 3 + TTG AAC AAG ATG GAT 21 2704-2724 TGC ACG 4 TCT AGA CGT GCA ATC 27 2704-2724 CAT CIT GTT CAA 5 + GGG TCG ACG ACA GGA 29 2678-2698 TGA GGA TCG TIT CG 6 + TCT AGA CAG GGG ATC 24 2654 2671 AAG ATC TGA 7 + GGT ACC GCT CAG AGA 24 2336-2353 AAG CIT CAC 8 + CGC CCA CTG CGC AGG 21 6059 6079 CTC AAG 9 + GGG TCG ACG CGT CAT 28 2181-2201 CGA CAT ITA TAC C 10 TCT AGA GTC CTG CAG 24 1862 1879 ATA AGC ATG 11 + GGT ACC GAT CCT CGC 24 1351-1368 CGT ACT GCC + 1 refers to primer orientation. Plus(+) 5'->3' as drawn in figure 20 (->). Minus() 3'-->5' as drawn in figure 29 (<-).

PAGE 183

162 Results Restriction modification versus genetic rearrangement or deletions Methyl modification. Evidence for differences in DNA methyl modification between pQSR50 plasmid DNA prepared from E. coli RM12459 and plasmid DNA from marine transformants were revealed by comparison of restriction digestion profiles of the isoschizomers Dpnl and Mbol. Both enzymes recognize the sequence 5' GATC 3' but Dpnl cuts this sequence only when the adenosine base is methylated in the six position (GM6ATC) while Mbol will cut only the non-methylated sequence. Figure 31 shows an autoradiograph of a Southern transfer of pQSR50 and the transformants BS-10, SJl-T, T9, and 14-T cut with Dpnl and Mbol. The filters were hybridized to the [35S] labelled NPTII probe. Two diagnostic restriction fragments are identified by this probe if cutting occurred (309 and 223 bp) pQSR50 purified from E. coli appeared to be methylated, as it was cut by Dpnl and not digested by Mbol, while all the transformants were digested by Mbol, indicating that they did not contain 6-methyl-adenine in the recognition site. These observations indicate that a methyl modification difference between E coli purified pQSR50 and plasmid DNA from marine transformants exists Furthermore, the diagnostic fragments generated by Dpnl digestion of pQSR50 were observed in the Mbol digestion of transformant DNA suggesting that there is not a sequence difference between pQSR50 and plasmid DNA from the transformants. Restriction fragments in BS-10, SJ 1-T, and T9 lanes are slightly retarded and blurred compared to the pQSR50

PAGE 184

MW Kb 23.1 -9 .4-6.5 -2.3f:8 =: 1 .0.93-163 pQSRSO BS-10 SJ1-T T9 14-T abc abc abc abc abc Figure 31. Autoradiograph of undigested (a), Dpnl-digested (b), and Mbol-digested (c) pQSR50 (purified from E. coli) and plasmid DNA from pQSR50 transformant strains BS10, SJl-T, T9, and 14-T HindiTI-digested lambda DNA and BstNI-dige sted pBR322 DNA molecular weight markers are indicated. Hybridized with e5S] labelled NPTII.

PAGE 185

164 and 14-T lanes This is likely due to the overloading of these lanes which was required to visualize the hybridization bands. Mboi digestion of PCR amplification products using primers #1 and #5 from pQSR50, SJl-T, and BS-10 yielded identical fragments confirming sequence identity between pQSR50 and the plasmids found in the marine transformants in this region (Figure 32). Deletion of Hindiii and BgiTI restriction sites The loss of the diagnostic Hindiii and Bglii fragments from transformants suggested that these sites had been lost, possibly due to a deletion. To determine if the HindiiJ and BgiTI sites had been deleted two approaches were taken. First, the probe pM62, homologous to the R1162 region immediately flanking Tn5 in pQSR50, IRL, homologous to the inverted repeat regions of Tn5, and the NPTII probes were used to determine if a major deletion had occurred in the transformants exhibiting altered restriction profiles All three of these probes hybridized to plasmid preparations of pQSR50 (purified from E coli RM1259), MG-1, BS-1 0, T9 and SJI T (Figure 33). Thus, there does not seem to be a deletion, at least a large one, in the region of lost restriction sites of these transformants To determine whether a small deletion existed in the inverted repeat region of Tn5, PCR primers were synthesized based on the sequence of Tn5 (Auerswald et al., 1981 [GenBank accession #V00617]; Johnson and Reznikoff, 1981 [GenBank accession #J01836 M10580]; Rothstein and Reznikoff, 1981 [GenBank accession #V00614]; Rothstein and Reznikoff, 1981a [GenBank accession #V00616]; Beck et al 1982 [GenBank accession #V00618]; Fuller et al., 1984 [GenBank accession #V00615]; Mazodier et al., 1985 [GenBank accession #X01702];

PAGE 186

1 2 pQSRSO a b SJ1-T a b BS-10 a b 165 Figure 32 Undigested (a) and Mbol-digestions (b) of the 762 bp PCR product of the nptll coding region from pQSR50 and transformant strains SJlT and BS10 BstNI-digested pBR322 DNA and a 1 Kb ladder (BRL) molecular weight standards are indicated Hoechst 33258 stained agarose gel (1% w/v).

PAGE 187

Figure 33 pQSR50 MG-1 c( z BS-10 c T9 SJ1-T 166 PROBE . -;. \ ....... ,/,. .. -;...t ,;. .. . .... . . :t-. 1 . ....... ... : . 1.... /. I I .... Autoradiographs of dot blots of pQSR50 transformants Hybridized w ith [ 3 5S] labelled NPTII and pM62 and DIG labelled IR L Location s of probes are shown in Figure 30.

PAGE 188

167 Yin and Reznikoff, 1988 [GenBankaccession #M21318]; Balzques et al., 1991 [GenBank accession #X57709]) to several sites in the left inverted repeat region of Tn5 and were used in tandem with primers located in the nptll coding region that had been shown to work in all transformant strains Primer sequences are given in Table 24 and are shown localized on Tn5 (Figure 29). PCR reactions were carried out as described in Chapter 5. All primer pairs amplified the predicted size product with template DNA from pQSR50 purified from E. coli RM1259 and from plasmid preparations of the pQSR50 transformant MF-1 (DI-9 transformed with pQSR50; Table 25). The same primers did not amplify template DNA from transformants SJl, MG-1, BS-10, or T9 suggesting that a sequence change in this region may have occurred in these strains. A summary of these PCR results are shown in Table 25. Dis c ussion In the previous chapter the transfer of the plasmid encoded nptii gene to several marine bacteria by natural transformation was demonstrated. Restriction profiles of all transformants from natural populations were different than the original plasmid used as transforming DNA (pQSR50 purified from E. coli), whereas transformation of isolates in culture resulted in either the maintenance or alteration of the expected restriction profile (Figures 24, 26). In these studies two alternative hypothesizes that could explain these findings were explored

PAGE 189

Table 25 PCR amplification of Tn5 from pQSR50 and pQSR50 transformants DNA AmQlification ( +/-) Primer Expected frag patr size (bp) pQSR50 DI-9 MF1 SJJ SJl-T MG-1 BS-10 T9 1, 3 739 + + + + + + 4, 11 1375 + + 4, 7 312 + + nd nd 4, 8 621 + + nd nd 1, 6 814 + + nd nd 10, 11 527 + + nd nd 1' 7 1108 + + nd nd 2, 5 297 + + nd nd nd + + 1, 5 762 + + nd + + + 2, 9 919 + + nd nd nd nd : Not done See Figure 30 for location of primers Each amplification was attempted a minimum of five times.

PAGE 190

169 The most direct approach to differentiate between sequence and non-sequence modifications would have been the direct sequencing of the regions of interest. However DNA sequencing of the region containing the suspect Hindiii and Bgni sites (inverted repeat of Tn5) from the marine transfonnant strains was not possible Restriction fragments from these regions were not available for cloning into sequencing vectors nor could this region be amplified by the polymerase chain reaction, despite the construction of 11 independent PCR primers for this region. Furthermore, attempts to artificially transform or to mobilize plasmid DNA from transformants into E coli or the Vibrio HFT strains were unsuccessful (data not shown). Therefore it was necessary to address the question of genetic versus non-genetic differences indirectly Non-sequence versus sequence modifications Comparison of Dpni and Mboi digestions indicated that plasmid DNA purified from E coli was methylated while plasmid DNA purified from the transformant strains was not. Addi tionally there was no evidence that suggest the loss of Dpni / Mboi sites since the same fragments were observed in all strains These results suggest that differences in restriction profiles might be due to methylation differences, however, it must be stressed that these results apply only to the coding region of nptii whereas the Hindlii and B gni sites occur outside of this region in the inverted repeat regions of Tn5. Thus, from these data, it is not possible to eliminate the possibility of the existence of genet i c differences outside the nprii coding region.

PAGE 191

170 However, a difference in methylation patterns between transforming pQSR50 DNA and DNA purified from the transformants cannot fully explain the observed loss of restriction sites, particularly with regards to the 2.3 Kb Bgni fragment. The Bgni recognition sequence contains an internal Dpnl!Mbol recognition sequence 5' AGATCT 3'. Since Bgni in not sensitive to (dam methylation) it is unlikely that the loss of the Bglll site in transformant strains SJI-T, BS-10, MG-1, and T9 can be explained by methylation The probes NPTII IRL, and pM62 hybridized to all transformant strains indicating that no major deletion in the region of the "lost" Hindlll and Bgni restriction sites. However, it was not possible to amplify this region by PCR suggesting that smaller deletions may have occurred. Alternatively, these results are consistent with a deletion of one of the inverted repeats of Tn5 (presumably the left inverted repeat) that would have eliminated the diagnostic Hindiii and Bg!II restriction fragments, prevented PCR amplification with all primer sets tested yet still allowed for positive hybridization by NPTII, IRL, and pM62 if the other inverted repeat was intact. Further investigation would be required to test this possibility. The generation of new plasmids by natural transformation The evolutionary origin of plasmids is uncertain. Some may be descendants of the nucleic acid molecules present in the cells before the evolution of the chromosome (e.g. Dyson, 1985), while others are probably descended from viruses (Broda, 1979)

PAGE 192

171 Perhaps the most important aspect of the evolution of plasmids is that they sometimes merge partially or completely with chromosomes allowing the movement of bacterial genes between chromosomes and plasmids (Rowbury, 1977; Hartl and Dykhuizen, 1984), from one plasmid to another (Godwin and Slater, 1979; Broda, 1979), or from one bacterial species to another (Campbell, 1981). Gene flow between plasmids and chromosomes is thought to be particularly common for genes on transposons, that insert themselves in both plasmids and chromosomes, and in bacterial species like E coli that harbor plasmids that are able to incorporate themselves temporarily in chromosome However, other mechanisms may also contribute to the genetic versatility of plasmids In these studies evidence was presented suggesting that during the process of natural transformation the plasmid pQSR50 underwent genetic changes in addition to non sequence modifications These changes may have been due to an association of the transforming plasmid with homologous DNA present in the recipient. Indeed in at least one example, recipient bacterial populations were shown to contain DNA homologous to R1162 prior to exposure to pQSR50 DNA (Figure 22, Chapter 5). Alternatively these changes may be an artifact of transformation by multimeric plasmid molecules. The consistent recovery of the correct sized EcoRI restriction fragment and the apparent high molecular weight of undigested plasmid DNA in most of the transformants suggests the presence of multimerized molecules in these transformants Plasmids contained by the transformants could be the products of transformation by multiple identical plasmids, recombination events resulting from the presence of self-homology in multimers, or some illegitimate recombination event. However, it is d i fficult to visualize the loss of internal

PAGE 193

172 restriction fragments by these mechanisms. Three model scenarios depicting the interaction of a transforming plasmid with regions of homology in the recipient resulting in the formation of a new plasmids are presented in Figure 34. In the first scenario (A) single stranded transforming plasmid DNA is transported into a recipient carrying a plasmid that contains regions of homology to the transforming plasmid The transforming DNA is transiently stabilized by association with the endogenous plasmid and then is reestablished as an independent replicon. This model suggests that the resulting plasmids may be unstable since it is likely that both belong to identical incompatibility groups. Interestingly, plasmid instability was observed in several transformation experiments (Chapter 5) In the second scenario (B) a transient association of transforming plasmid DNA with a region of chromosomal homology is depicted. As in the first scenario single stranded transforming DNA, incorporated into the recipient, transiently associates with a region of homology followed by the reestablishment of an independent replicon. In this case the transformed plasmid may acquire a portion of the recipient chromosome. The third scenario presented (C) suggests a permanent association of transforming DNA with a resident plasmid resulting in the formation of a modified plasmid. A fourth possibility is the permanent integration of a transforming plasmid into a homologous region of the host chromosome. However, in our studies this does not seem to have occurred since all hybridization signals from transformants were associated with an extrachromosomal DNA fraction.

PAGE 194

173 A c _0 ;(a ) A c60 co B c c _0 Ei ) c A eOs ) Figure 34. Proposed mechanisms for the generation of new plasmids during natural plasmid transformation Transient association of transforming plasmi d with endogenous plasmid (panel A). Transient association of transforming plasmid with recipient chromosome (panel B). Permanent association of transforming plasmid with an endogenous plasmid (panel C).

PAGE 195

174 In these studies evidence for both non-genetic (methylation) and genetic modifications were observed to have been a result of the natural transformation of marine bacteria. However, non-sequence modifications were not sufficient to explain all observed differences in restriction profiles of the transforrnants Three mechanistic models were proposed that outlined potential recombination pathways that would lead to the generation of new plasmid molecules during the process of natural plasmid transformation The relationship between these models and the Vibrio HFT transformation system are unclear since in the Vibrio system the expected plasmid molecule was consistently generated after transformation. Moreover since multimerization of the transforming plasmid DNA resulted in increased transformation efficiencies this suggested that the expression of the transformed plasmid was a result of recircularization with itself and not by an involvement with endogenous regions of homology (Chapter 2). Although continued investigation into the mechanisms of transformation in marine bacteria are required (particularly at the level of recombination), these studies demonstrate the plausibility of the hypothesis that the process of natural plasmid transformation promotes the formation of new plasmids among marine bacteria.

PAGE 196

CHAPTER 7: NATURAL PLASMID TRANSFORMATION IN THE MARINE ENVIRONMENT : A DISCUSSION AND SUMMARY 175 The aim of th i s chapter is to summarize the results of this study and to place them into the general context of bacterial gene transfer in marine environments It is well established that genes, both chromosomally and plasmid encoded can be and have been transferred between microbial populations in numerous and d i verse environments (Chapter 1 ; Levy and Miller 1989 ; Fry and Day, 1990) Especially important seems to be the transfer of plasmids The promiscuity of some naturally occurring plasmids, especially those which are conjugative, have been well documented (Chapter 1 ; Thomas 1989 ; Fry and Day 1990) Other studies have indicated the potential for transduction (Miller et al., 1992; McHenney and Baltz, 1988 ; Saye et al ., 1987; Keynan et al., 1974) and natural transformation of plasmids in nature (Chamier er al. 1993 ; Rochelle et al. 1988 ; Paul et al. 1991a). It i s even possible that there are yet undiscovered mechanisms by which genetic information can be shared between similar and diverse bacteria. The studies described in this work focused on the transfer of plasmid encoded genes in marine bacteria by natural transformation. The goals of the project were twofold First, to develop a model system that could be used to investigate natural transformation in marine environments (Chapter 2) Emphasis was placed on

PAGE 197

176 determining whether transformation in a naturally transformable marine organism was analogous to the transformation systems of other well studied organisms. Development of a marine transformation system also enabled us to assess the impact of several environmental parameters on the transformation process (Chapter 3) The second goal of these studies was to demonstrate the transfer of plasmid encoded genes to indigenous marine microbial populations by natural transformation (Chapter 5) Follow up on in i tial work in our laboratory, that reported the first example of natural plasmid transformation of a marine organism (Vibrio DI-9 [Jeffrey et al 1990]) led to the discovery of a high frequency of transformation variant of this strain These strains which we termed HFT strains, were transformed at significantly higher frequencies than was the wild type strain with a variety of IncQ/P4 plasmids (Frischer er a! ., 1990) This high frequency of transformation phenotype extended both to plasmid and chromosomal DNA indicating that it was not a condition associated with efficient plasmid transformation ( i. e. plasmid uptake or recircularization) but represents a generalized cond i tion The HFT phenotype was stable and heritable, suggesting that it was due to a genetic change rather than a phys i ological phenomenon Furthermore, another Vibrio strain, Vibrio parahaemolyticus (USFS3420) could also be transformed and an HFT derivative of this strain was isolated (MF-4C) This suggested that transformation was a general property of Vibrios. However, 18 other Vibrio strains were examined for the ability to be transformed by heterologous plasmid DNA and none of these strains were found to be transformable (Chapter 5) The nature of the genetic difference between the wildtype low frequency of

PAGE 198

177 transformation strain (LFT) and the HFT strains remains unclear. Preliminary evidence suggests that the difference in transforming activity may involve DNA binding and uptake since a difference in colony morphologies was seen between the LFT and HFT strains; suggestive of a cell wall modification. In the future it will be interesting to identify functional and genetic differences between these strains. Approaches which could be utilized may include transposon mutagenesis to artificially create HFT mutations or complementation assays utilizing restriction fragments of wildtype strains to restore the wildtype phenotype Functional differences could be ascertained using as a point of departure the identification of protein differences between the strains. Regardless of the exact nature of the difference(s) between the LFT and HFT strains, the observation that a spontaneous mutation can give rise to a highly competent strain of what is normally only marginally competent bacterium, is significant. Although competence among bacteria seems to be a property of a relatively few species, if a mutation can lead to competence in normally non-transformable organisms, or organisms that transform poorly transformation may be more important as a route of gene transfer than was previously believed. Thus, the finding that a high frequency of transformation mutation can occur in a marine bacterium suggests that natural transformation by these mutants could be a means of plasmid dissemination in marine and aquatic environments. In the third chapter laboratory experiments using a filter transformation assay were described. These studies were designed to evaluate, under controlled conditions, whether conditions in the marine environment were conducive to transformation of the

PAGE 199

178 HFT marine Vibrio strains. These studies indicated that natural conditions would facilitate transformation in an organism similar to the marine Vibrio strains It was observed that the amount of DNA necessary for transformation ( > 0 1 fg/cell) was well within the concentration range of dissolved DNA concentrations found in most marine environments and that the time required for transformation to occur (one to four hours) was rapid enough to compete with the digestion of dissolved DNA by extracellular nucleases. Furthermore, the transformation efficiency was not significantly affected over a wide range of nutrient, salinity, and temperature conditions. The type of plasmid DNA (different plasmids as well as different topological plasmid forms) was also examined with respect to effect on transformation efficiency. Plasmid multimers exhibited the highest transforming activity, consistent with the hypothesi s that, as in other organisms, plasmid establishment in Vibrio involves homologou s recombination Plasmid preparations enriched in monomeric and linear molecules also exhibited transforming activity. However, analysis of dose dependence suggested that two-hit kinetics (i.e two plasmid molecules were required per transformation event) were required for linear and monomeric molecules while tran s formation by multimeric molecules could be best explained by a first order model. However, further investigation s utilizing pure preparations of different topological plasmid molecules is required to absolutely determine the transformation activity of these plasmid forms. The development and maintenance of competence in the HFT marine Vibrio strains was also examined and determined to be unique when compared to competence

PAGE 200

179 development exhibited by other transformable organisms. Competence in most other systems studied exhibit only a short (20 min to 30 hour) window of competence. In the Vibrio HFf strains competence was observed throughout the growth phase, although highest transformation frequencies were observed when cultures reached late log phase. Competence was maintained at relatively high levels for at least ten days in both nutrient and starvation media. In the future it will be interesting to determine whether genetic competence in the marine Vibrio strains are under the control of a specific regulon involving specialized gene products, as seems to be the case in Bacillus, H. injlenzae, S pneumoniae, and N. gonorrhoeae, or whether competence in these strains is the result of a defect in an unrelated cellular function (e.g. membrane permeability, macromolecular transport nuclease activities, or any DNA processing function) which under normal situations protect a cell from the invasion of exotic genetic information Regardless it appears unlikely that it is possible to extrapolate information about the transformation systems of marine organisms based on non-marine organisms. Likewise, it is unknown if the transformation system of the Vibrio HFT strain is representative of transformation systems in other environmental bacteria. The integration of these results suggests that the marine environment could readily support natural transformation Dissolved DNA appears to be sufficiently abundant, the sequestering of dissolved DNA by a competent marine organism occurred on a time scale at least as rapid as the turnover rate for dissolved DNA measured in a variety of marine environments, physical and chemical conditions of most marine environments are conducive to transformation, and competence is maintained for sufficient times to favor

PAGE 201

180 an encounter between transforming DNA and a competent organism Concurrent with these lab studies marine microcosm (water and sediment) were conducted (Paul eta!. 1991; Paul er a!. 1992). Although these studies were not conducted as part of the studies included in this dissertation their relevance to these studies warrants discuss ion. In microcosm studies the potential for transformation of the HFT Vibrio strains in somewhat more realist i c environmental conditions (compared with transformation filter assays) was demonstrated As might have been expected lower transformation frequencies were seen in marine microcosms than were observed in the optimized filter assay In sterile microcosms transformation efficiencies were reduced by one to three orders of magnitude compared to transformation frequencies observed in the filter assay. When the ambient population was present, transfer was observed only in the water column microcosms and was reduced by one to three orders of magnitude in comparison to sterile water microcosms (Paul er a!., 1991a). The presence of nonsterile sediment inhibited the transformation process in all assays to below the detection limits of the system. No transfer to the ambient population was observed in these experiments These studies further demonstrate the potential for plasmid transfer to occur by natural transformation in the marine environment or at least in the water column. Using the HFT Vibrio system inter-as well as intraspecific plasmid transfer was demonstrated (Paul et al. 1992). These observations are especially relevant since they indicate that contact-mediated intergeneric plasmid exchange can occur in the absence of detectable viable donor cells and that small non-conjugative plasmids can be spread

PAGE 202

181 through heterogenous microbial communities by natural transformation. The plasmid pQSR50 could be transferred to the Vibrio HFf strains from Vibrio donor cells as well as from E. coli cells containing the plasmid. Transforming DNA could be presented either in a purified extracellular form or contained in donor cells. Viability of intact plasmid bearing donor strains was not required as heat-inactivated plasmid containing cells could act as donors (albeit at reduced efficiencies; [Paul et al 1992]). Nucleic acid synthesis inhibitors (nalidixic acid and rifampicin) inhibited transfer. There was, however, a requirement for cell contact since plasmid transfer was completely inhibited when donor and recipient cells were separated by a 0.2 J.Lm membrane and no transfer was observed when spent medium alone was used as a source of transforming DNA (Paul eta! 1992). These investigation demonstrate the potential for contact-mediated plasmid transformation to occur between widely differing genera and a marine bacterium. These studies may be particularly relevant in the assessment of genetic risk to the environment, particularly from wastewater treatment systems where plasmid containing enterics may be released into marine environments In Chapter 4 preliminary in situ transformation experiments were described. The objectives of these studies were to explore the potential for natural transformation in somewhat more realistic environments than were used in the microcosm studies As observed in microcosm studies, transformation of the Vibrio strains was observed in sterile incubations Purified plasmid DNA or intact plasmid donor cells were both capable of producing transformants. In all incubations the presence of the ambient microbial community reduced the transformation frequencies below detection limits.

PAGE 203

182 These studies further demonstrated the potential for natural plasmid transformation in marine environments and illustrate the potential for the HFT strains to be used in situ Some of the deficits of the system were also highlighted, namely low assay sensitivity In future studies it will be desirable to develop systems which can be used outside of containments and which are better suited for survival in natural environments. Chapters five and six describe studies that documented the transfer of plasmid encoded genes to indigenous marine bacteria via natural transformation This has been a long term goal in our laboratory initiating in 1986 when it was observed that the addition of plasmid DNA containing antibiotic resistance genes (kanamycin and streptomycin) to natural water populations resulted in increased heterotrophic activity, measured compared to controls, when populations were exposed to antibiotics (Jeffrey, Ph.D. thesis, 1989). However, despite the reproducibility of these observations, transformation was never verified in these experiments by the isolation of a transformant. The development and optimization of the transformation filter assay (chapters two and three) along with the development of efficient means to concentrate natural marine microbial communities (Jiang er a!. 1992) allowed us to effectively screen marine microbial populations for the ability to transform Two approaches were taken, first, marine sediment and water isolates were screened for transformation. The distribution of competence among isolates was determined and found to be approximately 10%. Second mixed microbial communities were concentrated and exposed to transforming plasmid DNA directly. In these experiments 5 of 14 microbial assemblages that were examined contained transformable members. Transformation frequencies of isolates and

PAGE 204

183 of concentrated mixed populations were observed to be in range of 1()9 transformants per recipient, similar to the transformation frequency observed in the wildtype Vibrio transforming strains Few studies have conducted systematic searches for the distribution of competence among diverse bacteria. Furthermore, these studies have been biased towards selection of organisms that exhibit very high transformation efficiencies. Although high frequencies are required if the goal is to develop a system as a tool of molecular biology, gene transfer frequencies that are low in comparison (e g. 10"9 versus 102 per recipient) may have significance in nature particularly if the genes involved confer selective advantages as often is the case for plasmid-encoded genes The magnitude of gene transfer by natural transformation was estimated by the use of a second order kinetics model which assumed the rate of formation of transformants is proportional to the concentration of available transforming DNA the concentrat i on of competent bacteria and is the result of the interaction of these two parameters Based on known concentrations of dissolved DNA, the estimate of number of transforming bac teria determined in these studies, and the empirical determination of kinet i c transfer coefficients
PAGE 205

184 Interestingly, in the majority of examples where transformation of marine bacteria was observed the plasmid DNA contained by the transformants appeared to be changed with respect to its restriction profiles as compared to the original transforming plasmid. Two principal hypotheses were explored in chapter 6 to explain these results: restriction modification or genetic rearrangement. Differences in methylation were found between transforming DNA (purified from E coli) and plasmid DNA from the transformants, suggesting that variant host restriction mod i fication systems may have resulted in differences seen in restriction digestion patterns. However, the observed non-sequence modification d i fferences were not sufficient to explain all the observed differences. Therefore it is likely that in addition to host dependent DNA modification, the process of natural plasmid transformation resulted in the genetic alteration of the transformed pla s mid. Perhaps this process may be referred to as "natural genetic engineering" Although bacterial gene transfer mechanisms were initially discovered m laboratory investigations bac t erial gene transfer is not limited to such environments Geneti c exchange between bacteria is not simply a curiosity of the laboratory invented by humans for the benefit of scientific research and biotechnological application Rather the inherent genetic abilities of bacteria have been harnessed, and in some cases creatively enhanced, for laboratory use. It should be kept in mind that the exchange of genetic information between bacteria is a normal process which occurs in natural environments The full implications, ecological or otherwise, of gene flow among microbial populations is not fully understood In the future it will be important to concentrate on establishing links between the occurrence of gene transfer events and

PAGE 206

185 ecological and evolutionary consequences of such events. In this project the transfer of plasmid encoded genes by natural transformation was investigated as a means by which small molecular weight plasmids can be transferred among marine microbial populations. These studies have indicated the potential for natural plasmid transfer, described a unique marine transformation model system, and documented the transfer of plasmid encoded genes to marine microbial assemblages by natural transformation. In general terms these studies strongly suggest that natural plasmid transformation could have a profound impact on the genetic constitution of marine microbial populations.

PAGE 207

186 LITERATURE CITED Aardema, B W. M G Lorenz, and W. E Krumbein. (1983) Protection of sediment adsorbed transforming DNA against enzymatic inactivation. Appl. Environ. Microbial. 46: 417-420 Ahlquist, E.F., C.A. Fewson, D A Ritchie, J Podmore, and V Rowell. (1980) competence for genetic transformation in Acinetobacter calcoaceticus NCIB8250. FEMS Microbial. Lett. 7: 107-109. Ahlquist E F., C A. Fewson, J Podmore and V. Rowell. (1980). Competence for genetic transformation in Acinetobacter calcoaceticus NCIB8250 FEMS Microbial. Lett. 7:107-109. Akiba, T. L. Koyama, Y Ishiki, S. Kimura, and T. Rukushima (1960). On the mechanism of the development of multiple-drug-resistant clones of Shigella. Jap. J. Microbial. 4 :2 19-227 Albano, M J Hahn and D Dubnau (1987) Expression of competence genes in Bacillus subtilis. J Bacterial. 169:3110-3117. Albritton W L., J W. Bendler, and J K Setlow. ( 1981). Plasmid transformation in Haemophilus injluenzae. J. Bacteriol. 145: 1099-1101. Altherr M.R. and K L. Kasweck (1982). In situ studies with membrane diffusion chambers of antibiotic resistance in Escherichia coli. Appl. Environ. Microbial. 44:838-843 Amabile Cuevas, C.F. and M E Chicurel. (1993). Horizontal gene transfer American Scientist 81 : 332-341. Amabile-Cuevas, C.F. and M E. Chicurel. (1992) Bacterial plasmids and gene flux. Cell 70:189-199 A min M K and M.J. Day. (1988). Donor and recipient effects on transduction frequency in situ. REG EM 1, abstract no 2

PAGE 208

187 Auerswald: E. A. G. Ludwig, and Schaller, H. (1981). Structural analysis of Tn5. Cold Spnng Harb Symp Quant. Bioi. 45 : 107-113. Aviles, M., J.C. Codina, A Perez-Garcia, F. Cazorla, R. Romero, and A. de Vicente. (1993). Occurrence of resistance to antibiotics and metals and of plasmids in bacterial strains isolated from marine environments. Wat. Sci. Tech 27:475-478. Bagdasarian, M., R.L.B. Ruckert, F C.H. Franklin, M M. Bagdasarian, and J. Frey (1981). Specific purpose plasmid cloning vectors. IT. Broad host range, high copy number RSF1010 derived vectors, and a host-vector system for gene cloning in Psuedomonas. Gene 16:237-247. Bale, M J., J C. Fry, and M J. Day. (1987). Plasmid transfer between strains of Pseudomonas aeruginosa on membrane filters attached to river stones. J. Gen Microbiol. 133: 3099-3107. Bale, M .J., J.C. Fry, and M.J. Day (1988) Transfer and occurrence of large mercury resistance palsmids in river epilithon Appl. Environ. Microbiol. 54:972-978 Balganesh, M., and J.K. Setlow. (1985) Differential behavior of plasmids containing chromosomal DNA insertions of various sizes during transformation and conjugation in Haemophilus injluenzae. J. Bacteriol. 161 : 141-146. Balzquez, J., J. Davies, and F Moreno. (1991). Mutations in the aphA 2 gene of transposon Tn5 mapping within the regions highly conserved in aminoglycoside phosphotransferases strongly reduce aminoglycoside resistance. Mol. Microbiol. 5:1511-1518. Bannister, J V. and M W Parker (1985). The presence of a copper/zinc superoxide dismutase in the bacterium Photobacterium leiognathi: a likely case of gene transfer from eukaryotes to prokaryotes. Proc. Natl. Acad Sci. USA 82 : 149-152 Barany, F., and M E. Kahn. (1985) Comparison of transformation mechanisms of Haemophilus parainjluenzae and Haemophilus injluenzae. J Bacteriol. 161 : 72-79. Barany F., and A. Tomasz. (1980) Genetic transformation of Streptococcus pneumoniae by heterologous plasmid deoxyribonucleic acid. J. Bacteriol. 144 : 698-709 Baross, J.A., J. Liston and R Y. Morita (1973). Some implications of genetic exchange among marine Vzbrios, including Vibrio parahaemolyticus, naturally occuring in the Pacific oyster. In: International symposium on Vibrio parahaemolyticus, pp. 129-137 Saikon Publishing Co Tokyo.

PAGE 209

188 Baumann, P., and R.H.W. Schubert 1984 Family II. Vibrionaceae Veron 1965, 5245AL, p In: N R Krieg and J G Holt (ed.), Bergey's manual of systematic bactenology, vol. 1. The Williams & Wilkins Co Baltmore Baya, A.M., P.R. Grayton V L. Brown, D.J. Grimes, E. Russek-Choen, and R R. Colwell. (1986). Coincident plasmids and antimicrobial resistance in marine bacteria isolated from polluted and unpolluted atlantic ocean samples Appl. Environ Microbial. 51:1285-1292 Bazalyan, V L. and T.A Ayzatullin (1979) Kinetics of enzymatic hydrolysis of DNA in seawater Oceanology 19:30-33. Beck, E., G Ludwig, E A Auerswald, B. Reiss and H Schaller. (1982) Nucleotide sequence and exact localization of the neomycin phosphotransferase gene from transposon Tn5. Gene 19 : 327-336 Beebee, T .J. (1991 ). Analysis, purification and quantification of extracellular DNA from aquatic environments. Freshwater Bio. 25 : 525-532. Behnke C. (1981) Plasmid transformation of Streptococcus sanguis (Challis) occurs by circular and linear molecules. Mol. Gen Genet. 183 : 490-497. Belliveau, B.H. M.E. Starodub, and J T Trevors. (1991). Occurrence of antibiotic and metal resistance and plasmids in Bacillus strains isolated from marine sediment. Can J Microbial. 37 : 513-520. Bensi G A. Iglesias, U Canosi, and T.A Trautner. (1981) Plasmid transformation in Bacillus subrilis. The significance of partial homology between plasmid and recipient cell DNAs Mol. Gen. Genet. 184 : 400-404. Bergh, D.E. (1989). Transposable elements in prokaryotes, p. 99-137. In S B Levy and R. V.Miller (ed ), Gene transfer in the environment, McGraw-Hill Book Co ., New York. Binns, A .N., and M.F. Thomashow (1988). Cell biology of Agrobacterium infection and transformation of plants Ann. Rev Microbial. 42:575-606 Biswas G.D., T Sox, E Blackman, and P.F. Sparling (1977) Factors affecting genetic transformation of Neisseria gonorrhoeae. J. Bacterial. 129:983-992. Biswas, G .D., J Graves, R. Schwalbe, and P.F. Sparling. (1986). Construction of isogenic gonococcal strains varying in the presence of a 4 2-K.ilobase cryptic plasmid J. Bacterial. 167:685-694.

PAGE 210

189 Bolinches J., M L. Lemos, B. Fouz, M Cambra, J.L. Larsen, and A E. Roranzo. (1990). Serological relationships among Vibrio anguillarum stranis. J. Aquat. Animal Health 2:21-29. Bork, P., and R.F. Doolittle. (1992). Proposed acquisition of an animal protein domain by bacteria Proc. Natl Acad. Sci. USA. 89:8990-8994 Bradley, D.E., J Whelan (1985) Conjugation systems of IncT plasmids J Gen Microbiol. 131 : 2665-2671. Bratbak, G M Heldal, S. Norland, and T F. Thingstad (1990). Viruses as partners in spring bloom microbial trophodynamics Appl. Environ. Microbiol. 56 : 1400 1405. Bratbak, G., M. Heldal T.F. Thingstad B Riemann, and O.H. Haslund (1992). Incorporation of viruses into the budget of microbial C-transfer. A first approach Mar Ecol. Prog. Ser. 83:273-280 Broda, P (1979). Plasmids Freeman, San Francisco. Brokamp, A and F R.J Schmidt. (1991) Survival of Alcaligenes xylosoxidans degrading 2 2-dichloropropionate and horizontal transfer of its halidohydrolase gene in a soil microcosm Curr Microbiol. 22 : 299-306. Buluwela, L., A Forster T. Boehm, and T H Rabb i tts (1989). A rapid procedure for colony screening using nylon filters Nucleic Acids Res 17:452 Burton, N.F., M J Day and A.T Bull. (1982). Distribution of bacterial plasmids in clean and polluted sites in a south Wales river. Appl. Environ Microbiol. 44:1026-1029. Buzby J S R.D. Porter, and S.E. Stevens, Jr. (1985). Expression of the Escherichia coli lacZ gene on a plasmid vector in a cyanobacterium. Science 230:805 807. Campbell, A (1981) Evolutionary significance of accessary DNA elements in bacteria Annu Rev. Microbiol. 35 : 55-83 Canosi U G Morelli and T A. Trautner. (1978) The relationship between molecular structure and transformation efficiency of some S aureus plasmids isolated from B subrilis. Mol. Gen. Genet. 166: 259-267. Canosi, U., A Iglesias, and T A. Trautner. (1981) Plasmid trasnforrnation in Bacillus subrilis: effects of insertion of Bacillus subrilis DNA into plasmid pC194. Mol. Gen Genet. 181:434-440.

PAGE 211

190 Carlson, C.A., L .S. Pierson, J.J. Rosen, and J L. Ingraham. (1983). Pseudomonas stutzeri and related species undergo natural transformation J. Bacteriol. 153:93 99. Carlson, C.A., S.M. Steenbergen, and J L. Ingraham (1984). Natural transformation of Pseudomonas stutzeri by plasmids that contain cloned fragments of chromosomal deoxyribonucleic acid. Arch. Microbiol. 140:134-138 Carlson, C .A., G.J. Stewart, and J.L. Ingraham. (1985). Thymidine salvage in Pseudomonas stutzeri and Pseudomonas aeroginosa provided by heterologous expression of Escherichia coli thymidine kinase gene. J. Bacteriol. 163 :291 295 Chamier, B., M .G. Lorenz, and W. Wackemagel. (1993). Natural transformation of Acinetobacter calcoaceticus by plasmid DNA adsorbed on sand and groundwater aquifer material Appl. Environ. Microbiol. 59 : 1662-1667 Chandler, M. (1991). New shuttle vectors for Haemophilus influenzae and Escherichia coli: P15A-derived plasmids replicate in H. injluenzae Rd. Plasmid 25:221-224. Chauvat, F. C Astier, F. Vedel, and F. Joset-Espardellier. (1983). Transformation in the cyanobacterium Synechococcus R2: Improvement of efficiency; role of pUH24 plasmid. Mol. Gen. Genet. 191: 39-45 Chen, J D ., and D A. Morrison (1987). Modulation of competence for genetic transformation in Streptococcus pneumoniae. J. Gen Microbiol 133: 1959-1967 Cohan F M M.S. Roberts and E.C. King. (1991). The potential for genetic exchange by transformation within a natural population of Bacillus subtilis. Evolution 45: 1383-1421. Cohen, S (1976). Transposable genetic elements and plasmid evolution. Nature 263:731 738 Christopher, F H Franklin, and R. Spooner (1989). Broad-host-range cloning vectors, p. 248-267 In C.M. Thomas (ed.), Promiscuous plasmids of Gram-negative bacteria. Academic Press, Inc (London), Ltd., London. Church, G.M. and W. Gilbert Genomic sequencing. Proc Natl. Acad. Sci. USA 81:1991-1995. Clark, A.J. (1973). Recombination deficient mutants of Escherichia coli and other bacteria Annu. Rev Genet. 7:67-86.

PAGE 212

191 Clewell, .. B E.E. Ehrenfeld, F. An, R.E. Kessler, R. Wirth, M. Mori, C. Kitada, M I?e, and A Suzuki (1987). Sex phennones and plasmid-related COnjugatiOn m Streptococcus faecalis. In: J F. Ferretti and R. Curtiss m (eds ). Genetics pp. 2-8 American Society for Microbiology, Washington, Clewell, D.B., and C. Gawron-Burke (1986). Conjugative transposons and the dissemination of antibiotic resistance in streptococci. Annu. Rev. Microbial. 40: 635-659. Conchas, R F M L. Lemos, J.L. Barja, and A E. Toranzo (1991). Distribution of plasmidand chromosome-mediated iron uptake systems in Vibrio anguillarwn strains of different origins Appl. Environ. Microbial. 57 : 2956-2962 Contente, S. and D Dubnau (1979) Characterization of plasmid transformation in Bacillus subtilis: kinetic properties and the effect of DNA conformation Mol. Gen Genet. 167 : 252 258. Coughter J .P., and Stewart, G J. (1989). Genetic exhange in the environment. Antonie van Leeuwenhoek 55 : 15-22. Crosa, J.H. (1980). A plasmid associated with virulence in the marine fish pathogen Vibrio anguillarwn specifies an iron sequestering system Nature 284:566-568 Cruze J.A. J T Singer and W R. Finnerty (1979) conditions for quantitative transformation in Acin.erobacter calcoaceticus. Curr. Microbial. 3:129-132. Danner, D B., R. A Deich, K. L. Sisco and H 0 Smith (1980) An eleven-base pair sequence determines the specificity of DNA uptake in Haenwphilus transformation Gene 11: 311-318. Danner D .B., H.O. Smith, and S.A. Narang. (1982) Construction of DNA recognition sites active in Haemophilus transformation Proc. Natl Acad Sci. USA 79:2393 2397 Datta, N., and V.M. Huges (1983) Plasmids of the same Inc groups in Enterobacteria before and after the medical use of antibiotics. Nature. 306:616-617. DeFlaun M F J.H. Paul, and D Davis (1986) Simplified method for dissolved DNA determination in aquatic environments. Appl. Environ. Microbial. 52 : 654-659 DeFlaun, M .F., J.H. Paul, and W H. Jeffrey. (1987). Distribution and molecular weight of dissolved DNA in subtropical estuarine and oceanic environments Mar. Ecol. Prog Ser. 38 : 65-73.

PAGE 213

192 DeFlaun M.F., and J. H. Paul. (1987) Hoechst 33258 staining of DNA in agarose gel electrophoresis J. Microb. Methods 5:265-270. M .F., and J.H. Paul. (1989). Detection of exogenous gene sequences in dissolved DNA from aquatic environments Microb. :Ecol. 18:21-28 De Graaff J., L.P. Elwell and S Falkow. (1976). Molecular nature of two betalactamase-specifying plasmids isolated from Haemophilus injluenzae type B. J. Bacteriol. 126 : 439-446 DeVos, W.M., and G. Venema. (1981) Fate of plasmid DNA in transformation of Bacillus subtilis protoplasts. Mol. Gen Genet. 182:39-43. Deich, R.A. H.O. Smith (1980) Mechanism of homospecific DNA uptake in Hae mophilus injlue n zae transformation. Mol. Gen. Genet. 177 : 369 374. Doran, J L. W H Bingle, K.L. Roy, K. Hirtsuka, and W J Page. (1987) Plasmid transformation of Azot obacter vinelandii OP J Gen Microbiol. 133:2059 2072 Dowson, C G A Hutchison, J A Brannigan, R C. George, D Hansman, J. Linares A. Tomasz J M Smith, and B.G Spratt. (1989) Horizontal transfer of penicill i n-binding protein genes in pencillin-resistant clinical isolates of Streptococcus pneumoniae. Proc. Natl. Acad. Sci. USA 86:8842-8846 Dowson, C .G., A. Hutchison, N. Woodford A.P. Johnson, B.C. George, and B. Spratt. (1990) Penicillin-resistant viridans streptococci have obtained altered penicillin binding protein genes from penicillin-resistant strains of Streptococcus pneumoniae. Proc Natl. Acad Sci USA 87:5858 5862 Dubnau, D ( 1991) The regulation of genetic competence in Bacillus subtilis. Mol. Microbiol. S: 11-18 Duncan, K E C.A. !stock, J.B. Graham and N Ferguson. (1989) Genetic exchage between Bacillus subtilis and Bacillus licheniformis: variable hybrid stability and the nature of bacterial species. Evolution 43: 15851 609. Dyson, F. (1985). Origins of Life Cambridge Univ. Press Cambridge. Ehrlich, S.D. (1977) Replication and expression ofplasmids from Staphylococcus aureus in B a cillus subtilis Proc. Natl. Acad Sci. USA 74 : 1680-1682 Elwell L.P. M Roberts, L.W. Mayer and S Falkow (1977) Plasmid-mediated betalactamase production in Neisseria g onorrhoeae. Antimicrob. Agents Chemother. 11: 528-533.

PAGE 214

193 Esser, K U Kuck, C. Lang-Hinrichs, P. Lemke, H.O. Osiewacz, U. Stahl, and P. Tudzynsi. (1986). Plasmid ofEukaryotes, pp. 1-214. Springer Verlag, New York. Essich, E., E Jr. Stevens, and R.D Porter (1990). Chromosomal transformation in the cyanobacterium Agmenellwn qUlldruplicatum. J. Bacteriol. 172:1916-1922 Feinberg, A P and B Vogelstein (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132 : 6-13 Feinberg A.P. and B. Vogelstein (1984) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem 137:266-267 Felsenstein, K M (1988) A modified protocol for the pZ523 columns: a legitimate alternative for plasmid DNA Biotechniques 6:847-848. Foster, T.J. (1983) Plasmid determined resistance to antimicrobial drugs and toxic metal ions in bacteria. Microbiol. Rev. 47 : 361-409. Fox M.S. and Hotchkiss, R.D. (1957) Initiation of bacterial transformation Nature 179 : 1322-1325. Fox M .S., and Allen, M .K. (1964). On the mechanism of deoxyribonucleate integration in pneumococcal transformation Proc. Natl. Acad Sci USA 52:412-419. Frey, J., and M. Bagdasarian (1989) The molecular biology of IncQ plasmids. In: C M. Thomas (ed.) Promiscuous plasmids of Gram negative bacteria, pp 79-94 Academic Press London. Frischer, M.E., J M. Thurmond, and J H Paul. (1990). Natural plasmid transformation in a high-frequency-of-transformation marine Vibrio strain Appl. Environ. Microbiol. 56:3439-3444 Frischer, M.E., J.M. Thurmond, and J.H. Paul. (1993) Factors affecting competence in a high frequency of transformation marine Vibrio J Gen. Microbiol. 139:753761. Fry, J .C., and M J Day (1990). Plasmid transfer in the epilithon. In: J C. Fry and M.J. Day (eds.) Bacterial genetics in natural environments p 55-80 Chapman and Hall, London Fuller, R.S B. E Funnell, and A. Kornberg (1984) The dnaA protein complex with the E coli chromosomal replication origin (oriC) and other DNA sites Cell 38: 889-900

PAGE 215

194 Fulthorpe R .R., and Wyndham, R.C (1991) Transfer and expression of the catabolic plasmid pBRC60 in wild bacterial recipients in a freshwater ecosystem Appl. Env i ron Microbiol. 57:1546-1553 Garcia, E ., P Lopez, M T Perez Urena and M. Espinosa (1978). Early stages in Bacillus subtilis transfonnation : association between homologous DNA and surface structures. J Bacteriol. 135 : 731-740. Gauthier, M.J., F Cauvin, and J -P Breittmayer (1985) Influence of salts and temperature on the transfer of mercury resistance from a marine pseudomonad to Escherichia coli Appl. Environ. Microbiol. 50 : 38-40. Gauthier M J., and V A Breittma y er. (1990 ) Gene transfer in marine environments In: J C Fry and M J Day (eds .) Bacterial genetics i n natural environments Chapman and Hall, London. Gealt M. A. M D Chai K. B Alpert and J. C Boyer (1985) Transfer of plasmids pBR322 and pBR325 in wasterwater from laboratory strains of Escherichia coli to bacteria indigenous to a waste disposal system. Appl. Environ Microb i ol. 49 : 836-841. Genthner, F ., P. Chatterjee, T. Varkey and A Bourquin. (1988) Capacity of aquatic ba c teria to act as recipients of plasmd DNA. Appl. Environ. Microb i ol. 54 : 115117 Gheyse n, G., P. Dhaese M. van Montagu and J S c hell (1985). DNA flux across genetic barriers: The crown gall phenomenon. In: B Hohn and E.S. Dennis ( eds .). Plant gene research G e net i c flux in plants pp 11-47 Springer-Verlag Wien New York Gibbs C P. B.Y Reimann, E Schultz A Kaufmann, R. Haas, and T F Meyer (1989) Reassortment of pilin genes in Neisseria gonorrhoeae occurs by two distinct mechanisms Nature 338 : 651-652 Glassman D .L., and L.A McNicol. (1981). Plasmid frequency in natural populations of estuarine microorgan i sms. Plasmid 5 : 231 Gl i ck B R. H E Brooks J J Pasternak (1985). Transfonnation of Azotobacter vinelandii with plasmid DNA J Bacteriol. 162 : 276-279 Golden S S., and Sherman L.A (1984) Optimal conditions genetic transfonnation of the cyanobacterium Anacystis n idulans R2. J Bactenol. 158 : 36 42

PAGE 216

195 Goodgal S H. (1982). DNA uptake in Haemophilus transformation Ann Rev. Genet. 16: 169-192. Goodgal S.H., and R.M. Herriott (1961) Studies on transformations of Haemophilus injluenzae. I. competence J. Gen. Physiol. 44:1201-1227 Goodman, S.D. and J J Scocca (1988) Identification and arrangement of the DNA sequence recognized in specific transformation of Neisseria gonorrlweae. Proc Natl. Acad. Sci. USA 85 : 6982-6986 Goodman, S.D., and J J Scocca. (1991). Factors influencing the specific interaction of Neisseria gonorrhoeae with transforming DNA J Bacteriol. 173 : 5921-5923 Goodwin, D ., and J.H. Slater (1979). The influence of growth envirnment on the stability of a drug resistance plasmid in Escherischia coli K12 J. Gen Microbiol. 111:201-210. Graham J.B., and C A. !stock. (1978). Genetic exchange in Bacillus subtillis in soil. Mol. Gen. Genet. 166:287-290. Graham J .B., and C.A. !stock. (1979) Gene exchage and natural selection in Bacillus subrilis in soil. Molec. Gen. Genet. 166 : 287-290. Graham B.J ., and C .A. !stock (1981) Parasexuality and microevolution in experimental populations of Bacillus subrilis. Evolution 35:954 963 Griffith, F (1928). The significance of pneumococcal types. J Hyg 27 : 113-159. Griffith O M (1988). Large-scale isolation of plasmid DNA using high speed centrifugation methods. BioTechniques 6 : 725-727. Gromkova, R and S H. Goodgal. (1979) Transformation by plasmid and chromosomal DNAs in H influenza. Biochem Biophys Res. Commun 88 : 1428-1434. Gromkova, R. and S H Goodgal (1981). Uptake of plasmid DNA by Haemophilus. J Bacteriol. 146:79-84. Gryczan, T.J. s Contente and D Dubnau. (1978) Characterization of Staphylococcus aureus plasmids introduced by transformation into Bacillus subtilis. J. Bacteriol. 134 : 318-329 Gryczan, T ., s. Contente and D. Dubnau (1980) cloning of heterologous chromosomal DNA by recombination between a plasmid vector and a homologous resident plasmid in Bacillus subrilis. Mol. Gen Genet. 177 : 459-467.

PAGE 217

196 Guerry, P 1 Van Embden and S Faalkow (1974) Molecular nature of two non. conjugative plasmids carrying drug res i stance genes J Bacteriol. 117 : 619-630 Hartl, D W ., and D E Dykhuizen. (1984 ). The population genetics of Escherischia coli. Annu Rev Genetics 18 : 31-68 Heinemann J.A and G.F. Sprague (1989) Bacterial conjugative plasmids mobilize DNA transfer between bacteria and yeast. Nature 340 : 205-209 Helinski, D S Cohen D Clewell, D Jackson and A. Hollaender (eds.) (1985 ). Plasmids in Bacteria Plenum Press New York H e ns c h k e, R B and F R.J. S c hmidt. (1990 ) Plasmid mobilization from genetically en g ineered bacteria t o members of the indigenous so i l microflora in situ Cur. Mi c robio l. 20 : 105-110. Hepburn A G. (1982) The biolo g y of the crown gal l -a plant tumour induced by A grobacrerium tume fac iens. In: M E Rhodes-Robertson and F A Skinner (eds ). Ba c teri a and plants. pp 101-113 Academic Pres s New York Herriott R M ., E M Mey er, and M Vogt. (1970) Refined nongrowth media for stage II de v elopme n t of c ompetence in Hae m ophilus injluenzae. J Bacteriol. 101:517 5 2 4 H i ldebrand M., D .K. Corey J.R Ludwig A K. Teng-Yung Feng and B E Volcani (1991 ). Plasm i ds in diatom spec i es J Bacteriol. 173 : 5924 5927 H ill, K E., A J Weightman, and J C Fry (1992) Isolation and screening of plasmids from the ep i lithon whi c h mobilize recombinant plasmid pD10 Appl. Env i ron M ic robi o l. 58 : 1292-1300 Hille, J., A. Hoekema P Hooykaa s and R A Schilperoort (1984) Gene organization of the Ti pla s mid. In: D P S Verma and T. Hohn (eds ). Plant gene research Gene s involved in microbe plant interactions pp 287-309. Springer-Verlag New York Hoo y kaas P .J.J., and R A S c hilperoort (1984). The molecular g e netics of crown gall tumorigenesis Annu Rev Genet. 22:209-283 Hopwood D A D J Lydiate F Malpartida, and H.M. Wright. (1984) Conjugative sex plasmids of Streptomyces In: D R. Helinski, S N Cohen, D B. Clewell D.A. Jackson and H Hollaender (eds ) pp 615-634 Plasmids in bacteria Plenum New York

PAGE 218

197 Houck, M .A., J.B. Clark, K.R. Peterson, and M.G. Kidwell (1991). Possible horizontal transfer of Drosophila genes by the mite Proctolaelaps reg a/is. Science 253: 1125 1129 Hui, F M and D.A. Morrison (1991) Genetic transformation in Streptococcus pneumoniae: nucleotide sequence analysis shows comA, a gene required for competence induction, to be a member of the bacterial ATP-dependent transport family. J. Bacteriol. 173:372-381. Ippen-Ihler, K. (1989) Bacterial conjugation In: S B. Levy and R.V. Miller (eds ) Gene transfer in the environment. McGraw-Hill Publishing Company, New York. Ippen Ihler, K.A. and E G. Minkley Jr. (1986) The conjugation system ofF, the fertility factor of Escherichia coli. Annu. Rev. Genet. 20:593-624. Jeffrey, W H (1989) Validation of [3H] thymidine incorporation and its application to detecting natural transformation in the marine environment. Doctoral disseration, University of South Florida. Jiang S .C., J.M. Thurmond, S L. Pichard and J H Paul. (1992). Concentration of microbial populations from aquatic environments by vortex flow filtration Mar. Ecol. Prog. Ser. 80:101-107 Jeffrey, W.H., J H Paul, and G.J Stewart. (1990) Natural transformation of a marine Vibrio species by plasmid DNA. Microbial Ecol. 19:259-269 Juni, E. and A. Janik (1969) Transformation of Acinerobacter calcoaceticus (Bacterium anitratum). J Bacteriol. 98 : 281-288 Juni E (1972). Interspecies transformation of Acinetobacter: genetic evidence for a ubiquitous genus. J Bacteriol. 112:917-931. Juni, E., and Heym, G.A. (1980). Transformation assay for identification of psychrottrophic Achromobacters. Appl. Environ Microbiol. 40: 1106-1114 Kahn, M .E., M Concino, R Gromkova, and S H Goodgal (1979). DNA binding activity of vesicles produced by competence-deficient mutatns of Haemophilus. Biochem Biophys Res. Commun. 87: 764-772 Kahn, M.E., F. Barany, and H O Smith. (1983) Transformasomes: specialized membraneous structures which protect DNA during Haemophilus transformation. Proc. Natl. Acad. Sci USA 80: 6927-6931.

PAGE 219

Kahn, M.E. and H.O. Smith. (1984) Transformation in Haeophilus: a problem in membrane biology. J. Membr. Biol. 81:89-103. 198 Kelch, W.J., and J.S Lee. (1978). Antibiotic resistance patterns of Gram-negative bacteria isolated from environmental sources. Appl. Environ. Microbiol. 36 : 450456. Keynan A., K .H. Nealson, H. Sideropoulos and J W Hastings. (1974). Marine transducing bacteriophage attacking a luminous bacterium. J. Virology 14: 333 340. Kinkle, B K and E L. Schmidt. (1991). Transfer of the pea symbiotic plasmid pJB5JI in nonsterile soil. Appl. Environ Microbial. 57:3264-3269 K.lingmuller W (1991). Plasmid transfer in natural soil : a case by case study with nitrogen-fixing Enterobacter. FEMS Microbiol. Ecol. 85:107-116. Kobori, H., C.W. Sullivan, and H Shizuya (1984) Bacterial plasm i ds in Antarctic natural microbial assemblages Appl. Environ Microbiol. 48:515-518. Kolenc R J., W.E. Inniss B R Glick, C.W. Robinson, and C.I. Mayfield (1988). Transfer and expression of mesophilic plasmid-mediated degradative capacity in a psychrotrophic bacterium Appl. Environ. Microbiol. 54:638-641. Kuramitsu H .K., and C.M. Long. Plasmid-mediated transformation of Streptococcus mutans. (1982) Infect. Immun 36 : 435 436 Lacks S (1962) Molecular fate of DNA in genetic transformation of Pneumococcus. J. Mol. Bioi 5 : 119 131. Lacks S (1977) Binding and entry of DNA in pneumococcal transformation In: Portoles A., R. Lopez, and M Espinosa (eds.) Modern trends, in bacterial transformation and transfection. pp 35-44 North-Holland, Amsterdam Lacks S. and B. Greenberg. (1973 ). Competence for deoxyribonucleic acid uptake and deoxyrobonuclease action external to cells in the genetic transformation of Diplococcus pneumoniae. J Bacterial. 114 : 152-163 Larsen J.L., H B. Rasmussen, and I. Dalsgaard (1988) Study of Vibrio anguillarum strains from different sources with emphasis on ecological and pathobiological properties. Appl. Environ. Microbiol. 54 : 2264-2267. Levy S B. and R.V Miller (eds ) (1989). Gene transfer in the environment. McGraw Hill Publishing Company, New York

PAGE 220

199 LeBlanc, D.J:, and F P Hassel. Transformation of Streptococcus sanguis Challis by plasmtd deoxyribonucleic acid from Streptococcusfaecalis. (1976) J Bacteriol. 128 : 347 355. Lederburg 1 ., and E.L. Tatum. (1946). Gene recombination in E coli. Nature 158 : 558. Lemos, M .L., P Salinas, A.E. toranzo, J L. Ba.Ija, and J H Crosa (1988). Chromosome mediated iron uptake system in pathogenic strains of Vibrio anguillarum. 1 Bacteriol. 170:1920-1925. Levisohn R., J Moreland and K H. Nealson. (1987) Isolation and characterization of a generalized transducing phage for the marine luminous bacterium Vibrio fischeri MJ-1. J Gen. Bacteriol. 133 : 1577-1582. Lopez, P., M Espinosa, D.L. Stassi, and S A Lacks (1982) Facilitation of plasmid transfer in Streptococcus pneumoniae by Chromosomal homology. J. Bacteriol. 150 : 692-701 Lorenz, M G B.W. Aardema and W E Krumbein. (1981) Interaction of marine sediment with DNA and DNA availability to nucleases Mar Biol. 64 : 225-230. Lorenz, M G., and W Wackernagel. (1987). Adsorption of DNA to sand and variable degradation rates of adsorbed DNA. Appl. Environ. Microbiol. 53:2948 2952 Lorenz, M.G. and W. Wackernagel. (1988) Impact of mineral surfaces on gene transfer by transformation in natural bacterial eivironments. In W Klingmuller (ed ) Risk assessment for deliberate releases p 110-119. SpringerVerlag KG, Berlin Lorenz, M.G. B.W. Aardema, and W Wackernagel. (1988) Highly efficient genetic transformation of Bacillus subtilis attached to sand grains J. Gen Microbiol. 134 : 107-112. Lorenz M G and W Wackernagel. (1990) Natural genetic transformation of Pseudomonas stutzeri by sand-adsorbed DNA Arch Microbiol. 154:380 385. Lorenz, M.G., and W Wackernagel. (1991). High frequency of natural transformation of Pseudomonas stutzeri in soil extract supplemented with a carbon/energy and phosphorous source. Appl. Environ Micriobiol. 57:1246-1251. Lorenz, M .G., and W. Wackernagel. (1994). Bacterial gene transfer by free DNA in the environment. Microbiol. Rev (In Press) Lyon, B.R., and R. Skurray (1987). Antimicrobial resistance of Staphylococcus aureus: genetic basis Microbiol. Rev. 51: 88-134.

PAGE 221

200 Mach, P.A., and D J. Grimes (1982). R-plasmid transfer in wastewater treatment plant. Appl. Environ. Microbiol. 44: 1395-1403. Macrina, F.L., K.R. Jones, and R.A. Welch Transformation of Streptococcus sanguis with monomeric pVA736 plasmid deoxyribonucleic acid (1981). J. Bacteriol. 146:826-830. Maeda, M and N. Taga. (1974). Occurrence and distribution of deoxyribonucleic acid hydrolyzing bacteria in seawater. J. Exp Mar. Biol. Ecol. 14:157-169 Maimone, F A Coppo, C. Pazzani, S.O. Ismail, R. Guerra, and P. Procacci. (1986) Clonal spread of multiply resistant strains of Vibrio cholerae 01 in Somalia. J. Infec. Diseases 153 : 802-803 Mancini, S Fertels D Nave, and M A Gealt. (1987) Mo1bilization of plasmid pHSV 106 from Escherichia coli HB 101 in a laboratory scale waste treatment facility Appl. Environ. Microbiol. 53:665-671. Mandel M ., and A. Higa. Calcium-dependent bacteriophage DNA infection. (1970) J. Mol. Bioi. 53 : 157-162 Maniatis, T., E. F. Fritsch, and J. Sambrook. (1982). Molecular cloning: a laboratory manual Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Margolin, P (1987) Generalized transduction In : (F.C. Neidhardt (ed ) pp. 1154-1168. Escherichia coli and Salmonella ryphimurium. Cellular and molecular biology Vol. 2. Am Soc. Microbiol Washingon, D.C. Marmur J. ( 1961). A procedure for the isolation of deoxyribonucleic acid from microorganisms J. Mol. Biol. 3:208-218. Martinez, E., B Bartolome and F. de la Cruz (1988). pACYC184-derived cloning vectors containing the multiple cloning site and lacZalpha reporter gene of pUC8/9 and pUC19119 plasmids. Gene 68:159-162. Masters, M (1985) Generalized transduction In: J Scaife, D. Leach, and A. Galizzi (eds ) pp 197-216 Genetics of bacteria Academic Press, New York. Mathis, L.S., and J J. Scocca. (1982) Haemophilus injluenzae and Neisseria gonorrhoeae recognize different specificity deeterminants in the DNA uptake step of genetic transformation J. Gen. Microbiol. 128 : 1159-1161.

PAGE 222

201 Mazodier, P P. Cossart, E Giraud and F Gasser ( 1985) Completion of the nucleotide sequence of the central region of Tn5 confirms the presence of three resistance genes. Nucleic Acids Res 13: 195-205 McHenney, M.A., and R.H. Baltz (1988) Transduction of plasmid DNA in Streptomyces spp and related genera by bacteriophage FP43 J Bacteriol. 170 : 2276 2282 Mcintire, S.A. (1989) Analysis of conjugation in bacteria. In: M Levine, R Seidler, and P.H. Pritchard (eds ). Classical and molecular methods to assess environmental application of microorganisms Am Soc Microbiol, Washington D.C. McPherson P and M.A. Gealt. (1986) Isolation of indigenous wastewater bacterial strains capable of mobilizing plasmid pBR325 Appl. Environ Microbiol. 51 : 904909. Meyer R., R Laux, G. Boch M Hinds, R Bayly, and J A. Shapiro (1982). Broad host range IncP-4 plasmid R1162: effects of deletions and insertions on plasmid maintenance and host range J Bacteriol. 152 : 140-150. Michel, B., B Niaudet, and S D Ehrlich. (1983) Intermolecular recombination during transformation of Bacillus subrilis competent cells by monomeric and dimeric plasmids Plasmid 10:1 -10. Miller R V S Ripp J Replicon, O A Ogunseitan, and T.A. Kokjohn (1992) Virus mediated gene transfer in freshwater environments In: M J Gauthier (ed ) Gene transfer and the environment. Springer Verlag, New York/ Morita R.Y. (1985) Starvat i on and miniaturization of heterotrophs with special emphasis on maintenance of the starved viable state In: M M Fletcher and G.D Floodgate (eds) Bacteria in their natural environmnets, pp 111-130 London:Academic Press. Morrison, D A. M Baker, and B Mannarelli. (1979) A protein component of the pneumococcal eclipse complex. In: Glover S.W L.O. Butler (eds ) Transformation 1978 pp 43-52, Cotswold Oxford, UK Morrison, D A and M. Baker (1979) competence for genetic transformation in pneumococcus depends on synthesis of a small set of proteins Nature 282 : 215 217 Morrison D A and W R. Guild (1973). Breakage prior to entry of donor DNA in pneumococcus transformation. Biochim Biophys Acta 299: 545-556

PAGE 223

202 Morrison, W D., R.V. Miller, and G S Sayler (1978) Frequency of F116-mediated transduction of Pseudomonas aeruginosa in a freshwater environment. Appl. Env i ron Microbial. 36:724-730 Mottes M G Grandi and V Sgaramella (1979) Different specific activities of the monomeric and oligomeric forms of plasmid dNA in transformation of B subtilis and E coli. Mol. Gen Genet. 174 : 281-286 Murchison H.H., J.F. Barrett, G.A. Cardineau, and R. Curtiss III. Transformation of Streptococcus mutans with chromosomal and shuttle plasmid (p Y A629) DNAs (1986) Infect. Immun. 54 : 273-282. Myhr, E. J L. Larsen, A Lillehaug, R Gudding M Heum and T Hastein. (199 1 ) Characterizat i on of V ibrio anguillarum and closely related species isolated from farmed fish in Norwa y Applied and Environmental Microbiology 57 : 2750-2757 Nes t er E W ., and T Kosuge (1981) P l asmids specifying plant hyperplasias. Annu. Re v Microbial. 35 : 531-565. Nes ter E.W., and B.A.D Stocker. (196 3 ). Biosynthetic latency in early stages of deoxyribonucle i c acid transformation in Bacillus subti lis. J. Bacterial 85: 785-796 Norelli J.L., T J Burr, A.M. Adriana L. Cicero, M.T. Gilbert and B.H. Katz. (1991). Homologous streptomycin resistance gene present among diverse Gram-negative bacte ri a in New York State apple orchards Appl. Environ Microbial. 57 : 486 491. Notani N .K., J.K. Setlow, D McCarthy, and N L. Clayton (1981). Transformation of Haemophilus injluenzae by plasmid RSF0885 J. Bacterial. 148 : 812-816 Notani N ., and S H. Goodgal ( 1966). On the nature of recombinants formed during tran s formation in Haemophilus injluenzae. J. Gen Physiol. 49:197-209 Novick, R P. (1969) Extrachromosomal inheritance in bacteria. Bacterial Rev 33 : 210 263 Novick R.P. I. Edelman and S. Lofdahl. (1986). Small Staphylococcus aureus plasmids are transduced as linear multimers that are formed and resolved by replicative processes J. Mol. bioi. 192:209-220 Novitsky J.A. (1986) Degradation of dead microbial biomass in a marine sediment. Appl. Environ Microbial. 52 : 504 508.

PAGE 224

203 K D Maris, K. Timmis, and D F Dwyer. (1992). Expression and transfer of engineered catabolic pathways harbored by Pseudomonas spp. introduced into activated sludge microcosms. Appl. Environ. Microbiol. 58 : 3380-3386 O'Connor, M., A. Wopat, and R S Hanson. (1977). Genetic transformation in Merhylobacterium organophilum. J Gen Microbiol. 119 : 287-296 Odelson, D A., J L Rasmussen, J. Smith, and F L. Macrina. (1987) Extrachromosomal systems and gene transmission in anaerobic bacteria. Plasmid 17 : 87-109. Ogram, A., G S Sayler and T. Barkay. (1987). The extraction and purification of microbial DNA from sediments J Microbiol. Methods 7:57-66. Ogunseitan, O.A., E T. Tedford, D Pacia, K M Sirokin, and G S Sayler (1987). Distribution of plasmids in groundwater bacteria J Ind Microbiol. 1:311-317 O'Morchoe, S., 0 Ogunseitan, G S. Sayler, and R V Miller, (1988) Conjugal transfer of R68 .45 and FP5 between Psuedomonas aeruginosa in a natural freshwater environment. Appl. Environ Microbial. 54:1923-1929. Ormerod, J.G (1988), Natural genetic transformation in Chlorobium. In: Olson, B (ed ) Green photosynthetic bacteria. Plenum Press, New York. Page, W J. (1982). Optimal conditions for competence development in nitrogen-fixing Azotobacter vinelandii. Can J Microbial. 28:389-397. Pakula, R and W Walczak. (1963) On the nature of competence of transformable streptococci J. Gen Microbial. 31:125-133. Palmen R B. Vosman, R Kok J R. van der Zee, and K.J. Hellingwerf. (1992) Characterization of transformation-deficient mutants of Acinetobacter calcoacericus. Mol. Microbial. 6 : 1747-1754. Palmen, R B Vosman, P. Buijsman, C K D Breek, and K J Hellingwerf. (1993) Physiological characterization of natural transformation in Acinetobacter calcoaceticus. J Gen Microbial. 139:295-305. Paranchych, W., and L.S Frost. (1987) The physiology and biochemistry of pil. Adv Bacterial. Physiol. 29:53-114. Paul, J.H. (1982). The use ofHoechst dyes 33258 and 33342 for enumeration of attached and planktonic bacteria Appl Environ. Microbial. 43 : 939-944

PAGE 225

204 Paul, J.H. W.H. Jeffrey, and M F. DeFlaun (1987). Dynamics of extracellular DNA in the marine environment. Appl. Environ Microbiol. 53:170-179 Paul, J.H. and B Meyers (1982) Fluorometric determination of DNA in aquatic microorganisms by use of Hoechst 33258 Appl. Environ Microbiol. 43 : 1393 1399. Paul, J H., M F DeFlaun, and W.H Jeffrey (1988) Mechanisms of DNA utilization by estuarine bacterial populations. Appl. Environ Microbiol. 43 : 1393-1399 Paul, J H ., and S L. Pichard. (1989) Specificity of cellular DNA-binding sites of microbial populations in a Florida reservoir Appl. Environ Microbiol. 55 : 2798 2801. Paul J H M.E. Frischer, and J.M. Thurmond (1991). Gene transfer in marine water column and sediment microcosms by natural transformation Appl. Environ. Microbiol. 57 : 1509-1515. Paul J H. J M Thurmond, M E. Frischer and J P. Cannon (1992) Intergeneric natural plasmid transformation between E coli and a marine Vibrio species. Mol. Ecol. 1:37-46 Paul J H ., S.C. Jiang and J.B. Rose (1991). Concentration of Viruses and dissolved DNA from aquatic environments by vortex flow filtration. Appl. Environ. Microbiol. 57 : 2197-2204. Paul, J H., J B. Rose, S C Jiang, C A Kellogg and L. Dickson (1993). Distribution of viral abundance in the reef environment of Key Largo, Florida Appl. Environ Microbiol. 59:718-724 Paul J H ., and D J Carlson (1984) Genetic material in the marine environment: implication for bacterail DNA Limnol. Oceanogr. 29 :10911097. Pickup, R.W. (1989). Related plasmids found in an English lake distric stream M i crob Ecol. 18 : 211-220 Pifer, M.L. (1986) Plasmid establishment in competent Haenwphilus injluenzae occurs by illegitimate transformation J Bacteriol. 168:683-687 Pifer, M .L., and H O Smith (1985). Processing of donor DNA during Haemophilus injluenzae transformation : analysis using a model plasmid system. Proc Natl. Acad Sci. USA 82:3731-3735

PAGE 226

205 Piggot P .J., and J.A. Hoch (1985) revised genetic linkage map of Bacillus subtilis Microbial. Rev 49 : 158-179 Pretorius-Guth, I.M., A Puhler, and R Simon. (1990) Conjugal transfer of megaplasmid 2 between Rhizobium meliloiti strains in alfalfa nodules. Appl. Environ Microbial. 56 : 2354-2359 Proctor, L.M., and J .A. Fuhrman (1990). Viral mortality of marine bacteria and cyanobacteria Nature 343 : 60-62. Randen J.V., and G Venema (1984) D i rect plasmid transfer from replica-plated E coli colonies to competent B subtilis cells Identification of an E coli clone carrying the hisH and tyrA genes of B. subtilis Mol. Gen Genet. 195 : 57 61. Ray, M K. G Seshu Kumar, and S Shivaji (1991) Plasmids from the soil bacteria of Schirmacher Oas i s, Antarctica Microbios 67 : 151-157 Redfield R J (1991) sxy-1, a Haemophilus injluenzae mutat i on caus i ng greatly enhanced spontaneous competece J Bacterial. 173 : 5612-5618 Rittmann B.E., B F Smets and D A Stahl. (1990). The role of genes in biological processes part one Environ Sci Technol. 24 : 23-29 Ro berts M C ( 1989) Gene transfer in the urogenital and repiratory tract. In : S .B. Levy and R.V Miller (eds ) Gene transfer in the environment. McGraw-Hill Publishing Company, New York Rochelle P A., M J Da y, and J C Fry (1988) Occurrence, transfer and mobilization in epilith ic strains of Acineobacter of mercur res i stance plasmids capable of transformation J G e n Microbial. 134 : 2933-2941. Rochelle P A J C Fry and M J Day (1989) Factors affecting conjugal transfer of plasmids encoding mercury res i stance from pure cultrues and mixed natural suspensions of ep i lithi c bacteria. J Gen Microbial. 135 : 409-4 2 4. Romanowski G M G Lorenz and W. Wackernagel. (1991) Adsorp ti on of plasmid DNA to mineral surfaces and protection against DNasei. Appl. Environ Mi c robial. 57 : 1062-1069 Romanowski, G M .G. Lorenz, G. Sayler and W. Wackernagel. (1992) Persistence of free plasmid DNA in soil monitored by various methods including a transformation assay Appl. Environ Microbial. 58 : 3012-3019

PAGE 227

206 Rothstein, S J. and W S Reznikoff. (1981). The functional differences in the inverted repeats of Tn5 are caused by a single base pair nonhomology Cell 23:191-199. Rothstein, S J R.A. Jorgensen, J.C.-P. Yin, Z. Yong-Di, R.C Johnson, and W S. Reznikoff. (1981). Genetic organization ofTn5. Cold Spring Harb. Symp. Quant. Biol. 45:99-105. Roybury, R (1977). Bacterial plasmids with particular reference to their replication and transfer properties. Prog. Biophys Mol. Biol. 31:271-317. Sambri, B. and M .A. Lovett. (1989) Influence of several experimental parameters on interand intra-specific mating process efficiency Microbiologica 12: 323-328 Sambrook, J., E.F. Fritch and T Maniatis. (1989) Molecular Cloning, a Laboratory Manual 2nd edn. Cold Spring Harbor NY Cold Spring Harbor Laboratory. Sandt C.H., and D S Herson (1991) Mobilization of the genetically engineered plasmid pHSV 106 from Escherichia coli HB 101 (pHSV 1 06) to Enterobacter cloacae in drinking water. Appl. Environ. Microbiol. 57:194 200. Saunders C.W., and W.R. Guild. (1980). Properties and transforming activities of two plasmids in Streptococcus pneummoniae. Mol. Gen Genet. 180:573-578. Saunders C.W., and W R Guild (1981a). Monomer plasmid DNA transforms Streptococcus pneumoniae. Mol. Gen Genet. 181:57-62. Saunders C.W., and W.R. Guild (1981b) Pathway of plasmid transformation in pneumococcus : open circular and linear molecules are active. J Bacteriol. 146 : 517-526 Saunders J R., and V.A. Saunders (1988). Bacterial transformation with plasmid DNA Methods in Microbiol. 21 : 79-128 Saye, D.J. 0. Ogunseitan, G S. Sayler, and R.V Miller. (1987). Potential for transduction of plasmids in a natural environment: effect of plasmid donor concentration and a natural microbial community in Pseudomonas aeruginosa. Appl. Environ Microbiol. 53 : 987-995. Saye, D .J., O A Ogunseitan, G.S Sayler and R V Miller. (1990) Transduction of linked chromosomal genes between Pseudomonas aeruginosa strains during incubation in situ in a freswater habitat. Appl. Environ Microbiol. 56:140-145. Schaberg D R ., and M.J. Zervos (1986). Intergeneric and interspecies gene exchange in Gram-positive cocci Antimicrob Agents Chemother. 30:817-822.

PAGE 228

207 Schofield, P.R., A H Gibson, W F Dudman, and J.M. Watson (1987) Evidence for genetic exchange and recombination of Rhizobium symbiotic plasmids in a soil population Appl. Env i ron Microbiol. 53 : 2942 2947 Schutt, C (1989) Plasmids in the bacterial assemblage of a dystrophic lake: evidence for plasmid-encoded nickel resistance Microb Ecol. 17: 49-62 Schutt C (1990) Plasmids and their role in natural aquatic bacterial communities. In: J. Overbeck and R.J Chrost (eds.). Aquatic microbial ecology: biochemical and molecular approaches Springer-Verlag, New York Scocca J J (1990 ). The role of transformation in the variab i lity of the Neisseria gonorrhoeae cell surface Mol. Microbiol. 4 : 321 327. Seifert H S R.S., Anioka, C C Marchal, P F. Sparling and So. Magdalene (1988 ). DNA transformation leads to pilin antigenic variation in Neisseria gonorrhoeae. Nature London 336 : 392-395. Setlow J K N .K. Notani D McCarthy and N L. Clayton (1981b) Transformation of Haemophilus injluenzae by plasm i d RSF0885 containing cloned segments of chromosomal deoxyribonucleic acid. J. Ba c teriol. 148 : 804 811. Seto H., and A. Toma s z ( 1975 ). Protoplast formation and leakage of i ntramembrane cell components-indu c tion by competence activator substance of pneumococci. J Ba c teriol. 121:344-353 Sh i gemitsu H. K. Terauchi and I. Koike (1991) Abundance of viruses in marine waters : assessment by epifluorescence and transfmission electron microscop y. Appl. En v iron Microbiol. 57 : 2731 2734 Singer J T., W. Choe K.A. Schmidt and R A Makula (1992 ) Virulence plasmid pJM1 prevents the con j ugal entry of plasmid DNA into the marine fish pathogen Vibrio anguil/arum 775. J Gen Microbiol. 138 : 2485 2490. Sisc o K.L. and H 0. Smi th (1979) Sequence-spec i fic DNA uptake in Haemophilus transformation Proc Natl. Acad Sci. USA 76 : 972-976 Sizemore, R .K., and R R Colwell. (1977) Plasmids Carried by antibiotic resistant marine ba c teria Antimicrob Agents Chemother 12 : 373 382. Smets, B F., B .E. Rittmann and D A Stahl. (1990) The role of genes in biolog i cal processes part two Environ Sci. Technol. 24 : 162-169

PAGE 229

208 Smit, E J.D. VanElsas, J A. VanVeen, and W.M DeVos. (1991). Detection of plasmid from Psuedomonas jluorescens to indigenous bacteria in soil using bactenophage OR2f for donor counter selection. Appl. Environ. Microbiol. 57 : 3482-3488 Smith, H.O., D.B. Danner, and R.A. Deich. (1981) Genetic transformation Annu. Rev. Biochem. 50:41-68. Smith, J.M., C.J. Dowson, and B.G Spratt. (1991) Localized sex in bacteria. Nature 349 : 29-31. Sorensen, S J. (1988) Survival of auxotrophic Escherichia coli K12 in seawater and transfer of plasmid pBR322 and pBR325 from laboratory strains of E. coli K12 to bacteria indigenous to the marine env i ronment. Special report Department of Microbiology. University of Copenhagen 1988; 73pp Sox T E W. Mohammed, P.F. Sparling (1979) Transformation-derived Neisseria gonorrhoeae plasmids with alterd structure and function. J. Bacteriol. 138 : 510 518 Sparling, P.E. (1966) Genetic transformation of Neisseria gonorrhoeae to streptomycin resistance J Bacteriol. 92 : 1364-1371. Stachel, S E ., and P C Zambryski (1986) Agrobacterium rumefaciens and the susceptible p l ant cell : a novel adaptation of extracellular recognition and DNA conjugation. Cell 47:155-157 Sternberg N., and R. Hoess (1983) The molecular genetics of bacteriophage P2. Annu Rev Genet. 17 : 123-154. Stewart, G J. (1989) The mechanism of natural transformation. In: Levy, S.B., and R V Miller (eds .) Gene transfer in the environment pp 139-164 McGraw Hill New York. Stewart G.J and D.H. Cyr. (1987). distribution of natural transformation ability among marine bacteria. EOS abstract 22A-04 1435 68, 1712 Stewart G.J., and C A. Carlson. (1986) The biology of natural transformation. Annu Rev Microbial. 40:211-235 Stewart, G .J., and C. D. Sinigalliano. (1990). Detection of horizontal gene transfer by natural transformation in native and introduced species of bacteria in marine and synthetic sediments. Appl. Environ Microbial. 56 : 1818-1824.

PAGE 230

209 Stewart, G .J., C.D. Sinigalliano and K.A Garko. (1991). Binding of exogenous DNA to marine sed i ments and the effect of DNA/sediment binding on natural transformation of Pseudomonas stulzeri strain ZoBel! in sediment columns FEMS Microbiol. Ecol. 85 : 1-8. Stuy, J.H. (1979) Plasmid transfer in Haemophilus injluenzae. J. Bacteriol. 139 : 520 529. Stuy, J .H., and R.B. Walter. (1986) Addition, deletion, and substitution of long nonhomologous deoxyribonucleic acid segments by genetic transformation of Haemophilus injluenzae. J Bacteriol. 148 : 565-571. Susskind, M M., and D. Botstein (1978). Molecular genetics of bacteriophate P22 Microbiol. Rev 42 : 385-413 te Riele H P J., and G Venema (1984) Molecular fate of heterologous bacterial DNA in competent Bacillus subtilis: Further characterization of unstable association between unstable donor and recip i ent DNA and the involvement of the cellular membrane. Mol. Gen Genet. 195 : 200-208. Thiry, G M. Thiry, J Thiry-Braipson M. Faclen, M Mergeay, and L. Ledoux (1984). Function of extrachromosomal DNA in the epiphytic bacterium Erwinia uredova. In: Helsinki D.R., Cohen, S N Clewell, D.B., Jackson, P A Hollaender, A (eds.) Plasmids in bacteria Plenum Press, London Thomas, C M (ed ) (1989) Promiscuous plasmids of Gram-negative bacteria Academic Press, London Tigari S., B E B Moseley. (1980) Transformation in Micrococcus radiodurans: Measurement of various paramters and evidence for multiple independently segregating genomes per cell. J Gen Microbiol. 119:287-296 Tolmasky M.E. P.C. Salinas, L.A Actis and J.H. Crosa (1988). Increased production of the siderophore anguibactin mediated by pJM1-like plasmids in Vibrio anguillarum. Inf Immun 56:1608-1614. Tomasz, A. (1966). Model for the mechanism controlling the expression of competent state in Pneumococcus cultures. J Bacteriol. 94 : 562-570 Tomasz, A (1970) Cellular metabolism in genetic transformation in pnemococci: Requirement for protein synthesis during induction of competence J. Bacteriol. 101 : 860-871.

PAGE 231

210 Tomasz, A and R.D. Hotchkiss. (1964) Regulation of the transformability of pneumococcal cultures by macromolecular cell products Proc Natl Acad Sci. USA 51 : 480-487 Tomasz, A and J L. Mosser. (1966). On the nature of pneumococcal activator substance Proc. Natl. Acad. Sci. USA 55:58-66 Tomb, J F H. El-Hajj, and H.O. Smith. (1991). Nucleotide sequence of a cluster of genes involved in the transformation of Haemophilus injluenz.ae Rd Gene 104 : 1 10. Top, E M Mergeay, D Springael, and W. Verstraete (1990). Gene escape model : transfer of heavy metal resistance genes from Escherichia coli to Alcaligenes eurrophus on agar plates and in soil samples Appl. Environ Microbiol. 56:2471 2479. Toranzo A .E., J.L. Barja, R R. Colweel, and F M Hetrick. (1983). Characterization of plasmids in bacterial fish pathogens. In fee. Immun 39 : 184-192 Toranzo, A E. P Combarro, M L. Lemos, and J L. Barja (1984). Plasmid coding for transferable drug resistance in bacteria isolated from cultured rainbow trout. Appl. Environ Microbiol. 48:872-877. Trevors J T and K M Oddie. (1986) R-plasmid transfer in soil and water. Can J Microbiol. 32 : 610-613 Trevors J T. T Barkay, and A W Bourquin. (1987). Gene transfer among bacteria in soil and aquatic environments: a review Can J. Microbiol. 33 : 191-198 Trevors J.T. and M E. Starodub. (1987). R-Plasmid transfer in non sterile agricultural soil. Syst. Appl. Microbiol. 9:312-315. Tucker, W T and J M Pemberton. (1980) Transformation of Rhodopsuedomonas sphaerodides with deoxyribonucleic acid isolated from bacteriophage R4>6P. J Bacteriol. 143 : 43-49 Van Den Hondel, C A M J.J., S Verbeek, A van der Ende, P J Weisbeek W E. Borrias, and G A van Arkel. (1980) Proc Natl Acad Sci. USA 77:1570-1574 Van Elsas, J D M Nikkel, and L.S Van Overbeek. (1989) Detection transfer in soil and rhizosphere, and the occurrence of homology to RP4 m so1l bacteria Curr. Microbiol. 19 : 375-381

PAGE 232

211 Veal, D.A., H.W. Stokes, and G. Daggard (1992). Genetic exchange in natural microbial communities. In K.C. Marshall (ed.), Genetic exchange in natural microbial communities. Plenum Press, New York. Venema, G. (1979). Bacterial transformation. Adv Microb. Physiol. 19: 245-331. Vijayakumar, M.N., and D.A. Morrison. (1986) Localization of competence-induced proteins in Streptococcus pneumoniae. J. Bacteriol. 165 : 689-695. Wahlund, T M ., and M T Madigan. (1991) Nitrogen fixation and genetic transformation in Chlorobium tepidum. In VII international Symposium on Photosynthetic Prokaryotes Abstract A, p 138 Wang Y., and D.E. Taylor. (1990). Natural transformation in Campylobacter species. J. Bacteriol. 172:949 -9 55 Watler, M.V. L.A. Porteous, V.J. Prince, L. Ganio, and R.J. Seidler. (1991). A microcosm for measuring survival and conjugation of genetically engineered bacteria in rhizosphere environments. Curr. Microbiol. 22: 117-121. Weinrauch Y., and D Dubnau. ( 1987). Plasmid marker rescue transformation proceeds by breakage-reunion in Bacillus subrilis. J. Bacteriol. 169 : 1205-1211. Weisberg R A. (1987). Specialized transduction. In: F.C. Neidhardt (ed.) pp. 1169 1176 Escherichia coli and Salmonella typhimurium. Cellular and molecular biology, Vol. 2 Am Soc. Microbiol. Washington D.C. Wiik, R., K A. Hoff, K Andersen and F.L. Daae. (1989) Relationships between plasmids and phenotypes of presumptive strains of Vibrio anguillarum isolated from different fish species. Appl. Environ Microbio l. 55:826-831. Wilkins, B.M. (1990) Factors influencing the dissemination of DNA by bacterial conjugation. In: J C. Fry, and M.J. Day (eds.) pp. 23-30. Bacterial genetics in natural environments Chapman and Hall, London. Willetts N and R Skurray (1987). Structure and function of the F factor and mechanism of conjugation. In: J.L. Ingraham, K.B Low, B Magasanik, F.C. Neidhart, M. Schaechter, and H.E. Umbarger (eds.) pp. 1110-1133 Escherichia coli and Salmonella typhimurium: cellular and molecular biology Am. Soc Microbiol., Washington, D.C. Willetts, N ., and R. Skurray. (1980). The conjugation system ofF-likeplasmids. Annu. Rev Genet. 14:41-76.

PAGE 233

Willetts, N. and B. Wilkins (1984) Processing of plasmid DNA during bacterial conjugation. Microbiol. Rev 48:24-41. 212 Williams H G., M J. Day and J C. Fry (1992) Detecting natural transformation of Acinetobacter calcoaceticus, in situ, with i n natural epilithon of the river Taff In: Stewart-Tull, E E .S., and M Sussman (eds ) The release of genetically mod i fied microorganisms pp. 253-255 Plenum Press, New York Williams, H G. (1993) Plasmid and chromosomal gene transfer by transformation in river epilithon. Doctoral dissertation University of Wales College of Cardiff. Wommack E.K., R T Hill M Kessel E. Russek-Cohen and R.R. Colwell (1992) Distribution of viruses in the Chesapeake bay Appl. Environ Microbiol. 58:2965-2970 Yin, J.C. and W.S Reznikoff. (1988) p2 and inhibition of Tn5 transposition J Bacteriol. 170: 3008-30015. Zambrysik, P. (1988) Basic processes underlyingA grobacterium-mediated DNA transfer to plant cells Ann. Rev. Genet. 22 : 1 30. Zar, J H. (1984). Biostatistical Anal y sis, pp 162-209. New Jersey: Prentice Hall. Zeph L.R., M.A. Onaga, and G. Stotzky. (1988) Transduction of Escherichia coli by bacteriophage P1 in soil. Appl. Environ Microbiol. 54 : 1731-1737 Zervos, P.R. L.M. Morris, and R.S. Hellwig (1988) A novel method for rapid isolation of plasmid DNA. BioTechniques 6 : 238-242 Zinder N .D., and J. Lederburg (195 2 ) Geneti c exchange in Salmonella. J. Bacteriol. 64 : 679 699.


printinsert_linkshareget_appmore_horiz

Download Options

close
No images are available for this item.
Cite this item close

APA

Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.

MLA

Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.

CHICAGO

Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.

WIKIPEDIA

Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.