Validation of [³H]thymidine incorporation and its application : to detecting natural transformation in the marine environment

Validation of [³H]thymidine incorporation and its application : to detecting natural transformation in the marine environment

Material Information

Validation of [³H]thymidine incorporation and its application : to detecting natural transformation in the marine environment
Jeffrey, Wade H.
Place of Publication:
Tampa, Florida
University of South Florida
Publication Date:
Physical Description:
xiv, 180 leaves : ill. ; 29 cm


Subjects / Keywords:
Heterotrophic bacteria ( lcsh )
Marine bacteria ( lcsh )
Thymidine ( lcsh )
Dissertations, Academic -- Marine Science -- Doctoral -- USF ( FTS )


General Note:
Thesis (Ph. D.)--University of South Florida, 1989. Includes bibliographical references (leaves 162-180).

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:
025638329 ( ALEPH )
25937340 ( OCLC )
F51-00175 ( USFLDC DOI )
f51.175 ( USFLDC Handle )

Postcard Information



This item is only available as the following downloads:

Full Text


VALIDATION OF [3H]THYMIDINE INCORPORATION AND ITS APPLICATION TO DETECTING NATURAL TRANSFORMATION IN THE MARINE ENVIRONMENT by Wade H. Jeffrey A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Marine Science in the University of South Florida April, 1989 Major Professor: Dr. John H. Paul


Graduate Council University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Ph.D. Dissertation This is to certify that the Ph.D. dissertation of Wade H. Jeffrey with a major in Marine Science has been approved by the examining committee on January 12, 1989 as satisfactory for the dissertation requirement for the Ph.D. degree Examining Committee: .. Dr. /e'Cf/goty J Ste;.rrt Dr. Ectwa'?'CtvaA ffi""e't < jt)r hose# To}res


DEDICATION To the Big People without their love, support, help, guidance, parental discretion, and genes (both dominant and recessive), this would never have been possible ii


ACKNOWLEDGEMFNTS The completion of this degree and my graduate career is due in part to the help, assistance, companionship, and hatred of many people who must be acknowledged for their contributions. First and foremost, I would like to thank Dr. John H. Paul for all his excellent guidance, direction, and words of wisdom. He put up with me for a long tire. I d like to thank Dr. Greg Stewart for his expertise and input, especially into the last two chapters of this beast. I would also like to thank the other members of my committee, Drs. Ted Van Vleet, Jose Torres, and Vargo for their advice and contributions to the overall scope of this project. You've all done me many favors over the years which are all greatly appreciated. M3.y all your graduate students be geniuses and may your research coffers overflow. A very special thanks to Dr. Bob Potter for coming to my rescue. Much thanks to Dr. M3.ry J):!Flaun for countless favors, discussions, real food, arguments, (sh eesh sounds like we're married) and for showing me that it is possible to get out of this place. Thanks to Lisa Cazares for technical assistance par excellence . wipe tests, pipets, buffers, orders, and for being the best probe maker I know. Thanks to Cheri Harpole for all the clean dishes and the smiles, Andy David for sharing in the dilemma of not having a last name, for all the survival tips, and for saving my notebooks from the dripping ceiling, Dave Williams for all the graphics assistance, Peter Betzer for numerous letters, and iii


Denise help above and beyond the call of. duty are countless other people who contributed to my life as a student .. some I mentioned in my thesis and may not acknowledge again. Rest assured that your help was greatly appreciated. But, there are still a few, thanks to Jack Hickman for dynes of the world, grouper sandwiches, chicken 'wangs' and especially for that house on the beach that year. It may be the only reason I made it. To Holli Jo for all the personal space (it's yours now), to Kevin for real science by real men, notarized documents, and for the baseball games in the To my brothers Roe and Dean for music news and views, to the Office and Trap for financial assistance, cheap thrills, and moral lessons, the Wall of Shame spiritual enlightenment, to Harv and all the yuppie power drinkers for lunar landings, to Joel for all the Buds, to John and Joline for legal aid, to Roger for fine to Tom for AK speak, to Pam for water skiing, to all the contributors to the Wall of Shame for improved office decoration, to Diet Coke for caffeine, and to Willy for still Thanks to Marie and her butthead buddy Bryan for not engaging their brains before they acted and to to my ulcer. A very special thanks to Becky Vanneman sanctuary, sustinence, and a feeling of 'family'. Finally, my greatest thanks goes to my (ya, the big people) for all my ambitions and keeping me going when I might not have otherwise And to my fiance Virginia, for making me see beyond the petri dish, for always being there when I needed a new perspective, and for Your reward ... you have to marry me. iv


TABLE OF CONTENTS LIST OF TABLES viii LIST O F FIGURES X ABSTRACT xii CHAPTER 1: INTRODUCTION Isotope Dilution 11 Non-Specific Macromolecular Labelling 15 Ability of Bacteria to Take Up and Incorporate Thymid ine 1 6 A Direct Comparison Between [3H]Thymidine Incorporation and DNA Synthesis 20 Detection of Natural Transfor-p1ation 21 CHAPTER 2: ISOTOPE DIWTION 33 Introduction 33 Methods and Materials 34 Sampling sites 34 Materials 34 Isotope dilution assays 34 Results 36 Discussion 41 CHAPTER 3: MACROMOLECULAR LABELLING 44 Introduction 44 Methods and Materials 44 Sampling sites 44 Materials 45 Macromolecular fractionation 45 Seasonal patterns of macromolecular labelling 46 Effect of exogenous thymidine concentration on macromolecular labelling patterns in v


diffet'ent size fr'actions 46 Cor'r'elations of labelling patter'ns with envir'onmental par'ameter's 47 Linear'ity of labelling patter'ns 47 Spatial vat'iation of labelling patter'ns 47 Seasonal patter'ns 49 Labelling patter'ns as a function of exogenous thymidine concentration and size fr'action 50 COr'r'elations of labelling patter'ns with micr'obial biomass and activity pat'ameter's 53 Effect of FdU on non-specific labelling of mact'omolecules 58 Compar'ison non-specific labe3ling dur'ing [methylH]thymidine and [6-H]thymidine incor'por'ation 61 Discussion 64 CHAPTER 4: DIRECT COMPARISON OF THYMIDINE INCORPORATION WITH DNA SYNTHESIS 74 Intr'oduction 74 Methods and M:l.ter'ials 74 Cultur'e Studies 74 Field Studies 75 Results 77 Discussion 80 CHAPTER 5: THYMIDINE KINASE ACTIVITY, THYMIDINE TRANSPORT AND INCORPORATION INTO DNA 85 Intr'oduction 85 Methods and Mlter'ials 85 Thymidine incorpor'ation in mar'ine isolates 85 Thymidine uptake vs. incor'por'ation 87 Thymidine kinase enzyme assays 88 Molecular' pr'obing for' thymidine kinase genes 89 M13 single str'anded pr'obes 89 Hybr'idization conditions for' M13 pr'obes 90 Constr'uction of Ribopr'obe RNA pr'obes 91 Hybr'idization conditions for' RNA pr'obes 93 Str'ingency effects on Ribopr'obe hybr'idization 93 Sensitivity and specificity of pr'obing 94 Chr'omosomal DNA pr'eparations 94 Pr'epar'ation of samples for' pr'obing 95 Results 96 Ability of mat'ine bacter'ia to incot'por'ate thymidine 96 vi


Thymidine uptake and and thymidine kinase enzyme activity 96 thymidine kinase genes: M13 single stranded DNA with HSV tdk 98 Probing for thymidine kinase genes: Riboprobe RNA 100 Discuss ion 11 0 CHAPTER 6: DETECTION OF NATURAL TRANSFORMATION 118 Introduction 118 and t-Bterials 118 Transforming DNA 118 Generation of plasmid multimers 120 Natural transformation assays in environmental samples 120 Transformation in environmental sediments 122 Filter transformation assays of marine isolates 125 Gene probing of potential 126 Sediment transformation assays with Vibrio parahaemolyticus 1 Z7 Chromosonal DNA trans format ion in V. parahaemolyticus 128 Amplification of detectable transformants in a mixed population 129 Results 129 of multimers for use in transformation assays 129 in environmental samples 131 Transformation in isolates 138 Chromosomal DNA transformation of V. parahaemolyticus 143 Amplification and selection for transformants in a mixed population 143 Thymidine to detect natural in v. parahaemolyticus 145 Discuss ion 146 CHAPTER 7: SUMMARY AND CONCWSIONS 156 LITERATURE CITED 162 vii


LIST OF TABLES 1. of cell division conversion factors derived from [ H]thymidine incorporation values. 10 2 Summary of isotope dilution values. 37 3. [ 3H]Thymidine incorporation into DNA before and after correction for isotope dilution. 41 4. Spatial in percent of radiolabel incorporated into DNA during [ H]thymidine incorporation. 49 5 Effect of exogenrus thymidine concentrations on total rates of incorporation in different size fractions. 50 6 Macromolecular labelling patterns as a function of exogenous thymidine concentration and size fraction. 52 1. of percent of radiolabel into DNA during [ H]thymidine incorporation assays with environmental parameters. 55 8. Effect of FdU on [3H]thymidine incorporation rates into DNA and percent radiolabel into DNA in vat'irus environments. 60 9. Comparison of thymidine incorporation with DNA synthesis for V. proteolyticus. 79 10. Comparison of thymidine incorporation with DNA synthesis for environmental bacteria. 80 11. Bacterial strains and plasmids. 86 12. Thymidine incorporation rates of marine bacterial isolates. 97 13. Thymidine uptake and incorporation and thymidine kinase enzyme activity in selected bacteria. 98 14. Transformation of marine populations in Bayboro Harbor ( 3 /19-21/87) 133 15. Transformation by plasmid DNA in the Medard Reset'voir 134 16. Tt'ansformation by plasmid DNA in offshore water's ( 5 / 13-15 / 87). 135 viii


17. Transformation in a coral surface microlayer of the Dry Tortugas ( 5 / 14-16-87). 136 18. Transformation in the coral surface microlayer of Cay Sal Bank ( 7/23 -25/88 ) 137 19. Transformation in the Medard Reservoir by p l asmid DNA (3/7-1 0/88) 138 20. Thymidine incorporation rates f o r filter transformation samples from t h e Medard Reservoir ( 3 / 7-10 /87) 138 21. Transformation in the Medard Reservoir with plasmid DNA (9/30-10 / 2/87). 139 22. Tampa Bay sediment transformation experiment ( 6 / 29-7/1 /88). 141 23. Sediment transformation, Cay Sal Bank ( 7 / 25-27/88) 141 2 4. Summary o f environmental transformation data. 142 2 5. Transformati o n frequencies f o r V parahaemolyticus 143 26. Thymidine incorporation to detect natural transformation in v. parahaemolyticus: amplification and selection experiment. 146 ix


1. 2. 3. 4. 5. 6 7 8. 9. 10 11. 12 13. 14. 15 LIST OF FIGURES Summary of pyrimidine nucleotide biosynthesis. Reactions involved in the de novo synthesis of dTMP. Thymidine kinase reaction and its relation to pyrimidine nucleotide salvage. Location of sampling sites along the Alafia and Ccystal Rivers. Isotope dilution analysis of Bayboro Harbor and Charlotte Harbor. Inhibition of intracellular isotope dilution by FdU in Bayboro Harbor. L3nearity of macromolecular labelling during [ H]thymidine incorporation in Bayboro Harbor waters Seasonal pattern of labelling in Bayboro Harbor. Diurnal pattern of [3H]thymidine incorporation and percent of radioactivity incorporated into DNA for the Medard Reservoir and the Crystal River Station 2. Effect of FdU on non-specific macromolecular labelling at higher concentrations of [3H]thymidine in Bayboro Harbor Effect of amethopterin and trimethoprim on macromolecular labelling patterns in Bayboro Harbor Comparison of incorporation3in Bayboro Harbor by using [methylH]thymidine and [6-H]thymidine. Direct comparison of thymidine incorporation, DNA synthesis, and cells produced for v. proteolyticus. Summary of the procedure used to make RNA probes using the riboprobe system. Probing sensitivity of M13 single stranded DNA probes with HSV tdk. X 7 12 18 35 38 39 48 5 1 57 59 62 63 78 92 99


16. Colony lifts probed with M13 single stranded DNA probes with HSV tdk. 17. Colony lifts of control organisms probed for thymidine kinase genes with Riboprobe RNA probes 18. Colony lifts of marine isolates probed with f coli tdk gene probe. 19. Stringency effects on standard curves probed with _E;_. coli tdk gene probes. 20. Stringency effects o n chromosomal DNA dot blots probed with f coli tdk gene probes. 21. Maps of transforming plasmids used in this study. 22. Schematic representation of sampling procedures for environmental transformation assays. 23. Schematic representation of liquid selection/ amplification procedure for isolating transformants. 24. Construction of multimeric forms of plasmid DNA. 25. Autoradiogram of Southern transfer of parahaemolyticus transfonnant. 26. Autoradiogram of Southern transfer of miniprep DNA from the liquid selection/amplification transformation experiment with the mixed bacterial population using pKT230. xi 101 102 104 105 108 119 121 130 132 140 144


VALIDATION OF [3HJTHYMIDINE INCORPORATION AND ITS APPLICATION T O DETECTING NATURAL TRANSFORMATION IN THE MARINE ENVIRONMENT by Wade H. Jeffrey An Abstract Of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Marine Science in the University of SOuth Florida April, 1989 Major Professor: Dr. John H Paul xii


The [ 3 H]thymidine method to estimate DNA synthesis and production was examined by investigating the four major factors and assumptions associated with the technique. When compared to fluorometrically determined of DNA synthesis, [ 3 H]thy midine consistently DNA synthesis by 6 to 8-fold, indicating the inability of [ 3 H]thymidine and isotope dilution assays to accurately determine the amount of thymine bases incorporated into DNA. Non-specific labelling of macromolecules other than DNA was ubiquitous and the percentage of radioactivity incorporated into DNA was inversely related to total rates of thymidine incorporation but independent of any other parameter examined. The use of specific and a of [methyl-3H]thymidine with [6-3 H]thymidine indicated that non-specific labelling was not the result of a demethylation reaction but the result of [ 3H]thymine catabolism Four of the 41 marine bacterial isolates examined were incapable of incorporating thymidine into DNA and lacked thymidine and thymidine kinase activity. of an coli tdk into one of these in high levels of thymidine kinase activity but no capacity to or transport thymidine by this The inability to incorporate thymidine into DNA by these isolates was due to the absence of a thymidine transport mechanism, thymidine kinase, or both. Natural in aquatic was examined by adding antibiotic encoding DNA and selecting for growth of transformants in the xiii


of nutrients and antibiotics. Rates of [3H]thymidine incorporation were higher in treatments receiving transforming DNA than in controls in 21 of 31 treatments implying that transformation may have occurred. A Vibrio parahaemolyt i cus isolate was naturally transformed with the broad host range plasmids pKT230 and pGQ3. This organ ism also transformed in sediment microcosms in the presence of another antibiotic resistant marine organism. Liquid media selection allowed detection of transformation at frequencies too low to be detected by standard plate counting methods. [ 3 H]Thymidine incorporation could not be used to detect transformation of V parahaemolyticus because this organism could not transport thy midine into cells. Abstract approved : xiv Dr. Jdhn H Paul Major Professor Associate Professor Dept of Marine Science Date of tAppro{Tal


CHAPTER 1 : INTRODUCTION Bacteria are an integral part of biogeochemical cycling in the world's oceans. For example, marine bacteria are known to use up to fifty percent of the carbon fixed by primary production ( Pomeroy, 1974; Azam et al., 1983). The microbial loop theory states that bacteria in the marine environment use dissolved organic matter generated by phytoplankton as an energy source. The bacteria, in turn, are grazed upon by flagellates. Microzooplankton then graze upon the flagellates. Thus, energy released as DOM by phytoplankton is returned to the main food chain (albeit inefficiently) in a microbial loop of bacteria flagellates microzooplankton (Azam et al., 1983) Although this theory has recently been questioned (Ducklow et al., 1986; 1987; Sherr et al., 1987), the above discussion indicates the need to accurately quantify bacterial heterotrophic production. This has proved difficult, however, as researchers continue to search for the optimal means to measure growth rates and productivity. There are several basic factors that must be considered when measuring bacterial activity. Incubation times must be short enough to minimize the accumulation of inhibitory products and the depletion of the substrate. Containers of low surface area to volume ratios should be used to minimize interactions with surfaces. Radio labelled substrates must have high enough specific activity so that their addition does not substantially increase the concentration of the


substrate within the natural sample Additionally, since isotopic equilibrium of internal and external pools is not always immediate, measurements should not be taken before equilibration occurs. Measurements taken over a time course allow determination of optimal incubation times before substrate depletion and bottle effects become significant (Stanley and Kanopka, 1985) Activity measurements, which are often the rate of one particular reaction or process, must be converted into units of carbon production to estimate heterotrophic productivity. Since any conversion 2 introduces uncertainty to the productivity value, it is important that the cell component or process on which the productivity estimate is based be directly linked to cell growth or carbon. Perhaps the best example is DNA synthesis. Once it is initiated, DNA synthesis proceeds to completion triggering a cycle of cell division (Lar k 1 969) DNA synthesis is regulated primarily at the site of initiation. Faster growth rates are the result of multiple replication forks proceeding along the chromosome since the the rate at which replication forks travel is constant ( Lark, 1969). Intracellular turnover of DNA is limited to excision and repair processes. The synthesis of other cellular components such as RNA and protein i s only correlated with cell division when cell growth is balanced (i. e cell properties such as DNA, RNA, protein, and cell number increase by the same factor during that time; Campbell, 1957). The relationship of RNA synthesis to growth is much more co mplex than DNA synthesis. While cells in balanced growth show a direct relationship between RNA synthesis and growth rate, this phenomenon does not occur in slowly growing cells. Different t ypes o f RNA have different rates of synthesis. While


approximately 97% of total RNA composition is ribosomal and transfer RNA in rapidly growing cells, these two types of RNA account for only 50% of the RNA synthesis. Messenger RNA, which is unstable and quickly turns over, comprises the rest (Nierlich, 1974) Several methods are currently used to estimate bacterial production including 14co 2 uptake (Romanenko et al., 1972; Overbeck, 1979; Riemann and Sondergaard, 1984), 35so4 incorporation into protein (Monheimer, 1978; Jordan and Petersen, 1978; Jordan and Likens, 1980; Pedros-Alio and Brock, 1982), and frequency _of dividing cells (FDC; Hagstrom et al., 1979; Newell and Christian, 1981; Riemann and 3 Sondergaard, 1984). Each of these techniques is limited in accuracy or ability to distinguish bacterial productivity from eukaryotic productivity. The use of radiolabelled nucleic acid precursors to estimate nucleic acid synthesis is the most commonly used means to estimate heterotrophic productivity. Several requirements must be met in order to use nucleic acid synthesis to esimate growth rates. The specific activity of the precursor immediately before incorporation into the macromolecule must be known. The radioactivity at the end of the experiment must be in the macromolecule under consideration. Finally, the incorporation of the added radioactive precursor should be by only one biosynthetic pathway This avoids complications due to differing kinetics (Moriarty, 1986). The commonly used radiolabelled precursors include the of [3H]uridine (LaRock et al. 1979), 32Po4 (Karl and Bossard, 1985a,b), C3H]adenine ( Karl, 1979; 1981; 1982; Karl et al., 1981a,b; Winn and Karl, 1984), and [3H]thymidine (Fuhrman and Azam, 1980; 1982; Moriarty and Pollard, 1981 ; 1982; Riemann and


Sondergaard, 1984; Pollard and Moriarty, 1984; Hollibaugh, 1988). Of these four techniques, [3H]adenine incorporation into RNA and DNA and [ 3 H]thymidine incorporation into DNA are the most commonly used, yet each has its own set of limitations. [ 3H]Adenine is believed to be uniformly assimilated by all heterotrophic and autotrophic microoroganisms. This technique is used 4 to measure total "microbial" nucleic acid synthesis. The results allow prediction of total microbial carbon production and mean specific growth rates (Winn and Karl, 1984). In addition, the accuracy of adenine incorporation depends upon; ( i) the reliability of the measurement of isotope dilution that occurs before and during incorporation; and (ii) the assumption that all cells take up adenine equally (Karl and Bossard, 1985a). This second assumption has been recently questioned by Fuhrman et al. ( 198 6a b). [3H ]Aden ine incorporation requires the determination of the specific activity of microbial [ 3 HJATP and [ 3H]dATP pools. This is determined from total microbial ATP, both bacterial and microeukaryotic. Fuhrman et al. ( 1986a,b) reported, however that in non..:oligotrophic environments, adenine added in nanomolar quanti ties is primarily taken up by bacteria. Conversely, rrost of t he ATP is associated with larger organisms. Bacterial ATP represents only 0 .5..:17% of the particulate ATP in freshwater and marine euphotic zones (Stanley and Kanopka, 1985). Additionally, DNA synthesis in eukaryotes is not continuous. It relies on continuous protein synthesis. If protein synthesis stops, DNA synthesis stops (Moriarty, 1986). Measured DNA and RNA synthesis values have agreed closely with predicted values based on [ 3 H]adenine incorporation in batch and


continuous cultures of a marine bacterium and several algae (Winn and Karl, 1984). Rates determined just after [3H]adenine was added to the samples were inaccurate, presumably due to compartrentalization of intracellular ATP. This indicates the need for long incubations for [ 3H]adenine incorporation assays (at least 10% of the doubling time of the organisms) These long incubations increase the chances of deleterious bottle effects described previously. 5 In addition to those limitations, the accuracy o f [3H]adenine incorporation remains unknown. As yet, there has not been a measure of DNA synthesis by some absolute non-isotopic means with which to co mpare [ 3H]adenine derived DNA synthesis values. Winn and Karl (1984) found good agreement between [3H]adenine derived and 32Po4 derived values. Unfortunately, they only measured DNA directly for their algae samples, not for their bacterial samples. The use of [3H]thymidine (thymine deoxyribose; TdR) incorporation into DNA has gained much wider use. Only dividing bacteria synthesize DNA and the rate of synthesis is proportional to the rate o f biomass formation during bacterial growth (Maaloe and Kjelgaard, 1966) Since cellular DNA content i s fairly constant and does not turnover, DNA synthesis is perhaps the best indicator of cell growth (Van Es and 1982) [3H]Thymidine is an effective means to pulse label DNA in bacteria (Moriarty, 1986) It is rapidly, efficiently, and stably taken up by bacteria. Thymidine is quickly converted to nucleotides and DNA labelling is believed to occur with little dilution by intracellular pools (Kornburg, 1980) Bacterial nucleotide synthesis occurs by two major routes. The first is de novo synthesis in which nucleotides are synthesized from


basic cellular components The second is the salvage pathway which serves to incorporate free bases and nucleosides that arise from catabolic activities (Kornbur-g, 1980). Figure 1 presents a general outline of the two pathways as they pertain to the incorporation of thymine bases into DNA. 6 [3H]Thymidine incorporation, as it is commonly used, is a function of the salvage pathway and is dependent on the activity of thymidine kinase (E.C. This enzyme phosphorylates thymidine to deoxythymidine monophosphate (dTMP). It is then further phosphorylated to dTDP and dTTP by the action of d'IMP kinase and nucleotide diphosphate kinase, respectively. dTTP is then incorporated into DNA by the action of DNA polymerase (Kornburg, 1980). De novo synthesis functions primarily through the action of thymidylate synthetase (E.C. 2 .1.1.45; Kornburg, 1908) This enzyme catalyzes the methylation of dUMP to form dTMP. The importance of this reaction lies in the convergence of the two pathways in the synthesis of d'IMP. As stated previously, the accuracy of [ 3 H]thymidine incorporation to measure DNA synthesis requires the determination of the specific activity of the final precursor to DNA, dTTP. Ideally, all dTTP would be derived from added [ 3 H]thymidine and thus the specific activity of the dTTP would be the same as the added [ 3 H]thymidine. However, de novo synthesis of dTMP results in dilution of the final precursor and complicates the determination of its specific activity. Further dilution may occur by the de novo synthesis of thymidine nucleotides v i a dUTP and cytidine nucleotides (Figure 1; Moriarty, 1986) coli there are two functionally separate pools o f dTTP. The


"' ); ., ,u URt.CIL 1 URACIL 5-CARBOXYLATE AS-FORMYL URACIL AS-HYOROXYMETIIYL URAC I L n HIYMJNE COz 0 2 COz Oz COz 0 2 C O z Oz] deoA + + + + + + + + sue KG sue KG sue KG sue KG TIIYMI 01 NE <-1 tdk DNA dTTP dTDP 3 dTMP I thyA dCOP ndk dCTP dcd dUTP dut dUMP I ndk dUDP l nrd CMP cmk COP ndk CTP p yrG UTP ndk UMP s ( )T:SIIIE _2___.. URACIL \ I a (ulcJ: -RNA l 7 10 --------------CYTIDINE -URI DINE URIOINE Figure 1. Summary of pyrimidine nucleotide biosynthesis. The enzymes involved include cmk, CMP (dCMP) kinase; dcd, 'dCTP deaminase; dut, dUTPase; ndk, nucleoside diphosphokinase; nrd, ribonucleoside diphosphate reductase; pyrG, CTP synthetase; pyrH, UMP kinase; thyA, thymidylate synthase; tdk, thymidine kinase; deoA, thymidine phosphorylase; 1, uracil 5-carboxylic acid decarboxylase; 2, thymine 7-hydroxylase; 3, dTMP kinase; 4, ribonucleotide glycosylase; 5, cytidine kinase; 6, uracil phosphoribosyl transferase; 7, uridine phosphorylase; 8, uridine kinase; 9, cytosine deaminase; 10, cytidine deaminase. Figure derived from Munch-Petersen (1983).


smaller dTTP pool is rapidly turned over for DNA synthesis in the DNA polymerase complex (Matthews et al., 1979) while a larger dTTP pool, derived from the DNA polymerase complex, is used for control. dTTP is an important regulator in pyrimidine nucleotide biosynthesis and as such the dTTP pools must be closely regulated ( Moriarty, 1986 ) The two routes of de novo synthesis of thymidine nucleotides, via cytidine nucleotides and uridine nucleotides (Figure 1) are both regulated by dTTP which inhibits ribonucleoside reductase and deoxycytidylate deaminase An increase in the larger dTTP pool derived from the DNA polymerase complex is quickly followed by an inhibition of de novo synthesis, resulting in a rapid depleti on of the smaller biosynthetic pool. The larger pool cannot penetrate the multienzyme complex to serve as a precursor for DNA synthesis (Matthews et al. 1979). If exogenous thymidine is not supplied in great enough quanti ties to maintain dTTP at concentrations necessary for DNA synthesis, de novo synthesis supplies the remaining dTMP required to keep dTTP at the optimal level and dilution of the isotope occurs (Pollard and Moriarty, 1984). It has been proposed that large amounts of exogenous thymidine nM for aquatic samples) will therefore provide sufficient dTTP to inhibit de novo synthesis (Pollard and Moriarty, 1984) Many investigators have derived conversion factors to estimate bacterial heterotrophic productivity based on [3H]thymidine incorporation measurements. Farly productivity values derived from thymidin e incorporation wer e determined using average DNA content per cell values (4 fg/cell; Gilles et al., 1970; Wallace and Morowitz, 1973; o r 2.5 fg/cell; Fuhrman and Azam, 1982) and an assumed A-T base pair content of 50%. Using this information, Moriarty and Pollard 8


(1981 ;1982) derived values of 1 3 x 1018 cells grown per mole thymidine incorporated. Fuhrman and Azam (1982) derived a value of 1.7 2.4 x 10 18 cell/mole thymidine. As a result, most early reports used a common conversion factor of 2 x 1018 cells per mole thymidine incorporated (Moriarty, 1986). Many subsequent investigators have determined conversion factors based o n measured changes in cell number in samples in which simultaneous rates of thymidine incorporation were determined. These experiments usually involve removal of bacterivores by selective filtration, specific eukaryotic inhibitors, and or the dilution of 9 samples in filtered water. In general, most conversion factors are approximately 2 0 x 1018 cells/mole thy midine incorporated. There have been, however, many exceptions Table 1 surrmarizes the conversion factors reported. Recent evidence suggests that cell volume changes may be closely related to production estimates from thymidine incorporation. Riemann et al. (1987) reported significant changes in cell volumes during their experiments but very consistent production values based on cell number. Coveney and Wetzel ( 1988), however, reported variable ce l l production values but consistent thymidine incorporation per cell volume. Bacterial heterotrophic productivity is often reported in terms of bacterial cell growth. However, to compare heterotrophic productivity with primary productivity, further conversion factors are required. An easy way to derive heterotrophic productivity values in terms of carbon produced is by multiplying the derived value of cells/mole thymidine incorporated by an average cell ca rbon content. This gives results in terms of g rams carbo n per liter per hour. An alternative would be to


Table 1 of cell division conversion factors derived from [ H]thymidine incorporation va l ues 1018 cells/mol Sample Type Reference 0 .2-1 .3 1.3 1.7 2.4 1 9-68 1 .9-2.2 0 .9-7.0 2 0--7.2 5 8 8.7 2 .2 2.0-6.0 1 0-1 2 1.1 0 44 2 .15 1 1 coastal waters marine sediment Nearshore waters Offshore waters Dilute seawater Lake water cult Freshwater Freshwater Freshwater Freshwater Oligotrophi c Dilute seawater Dilute seawater Lake sediments Fuhrman and Azam, 1980 Moriarty and Pollard, 1981; 1982 Fuhrman and Azam, 198 2 Fuhrman and Azam, 1982 Kirchman et al. 1982 Bell et al. 1983 Riemann, 1984 Riemann, 1 985 Murray and Hodson, 1985 Lovell and Kanopka, 1985 Ducklow and Hill, 1985 Bell, 1986 Riemann et al., 1987 Bell and Ahlgren, 1987 Lake water Scavia and Laird, 1987 Diluted freshwater Smits and Riemann, 1988 Olig Lake cult Coveney and Wetzel, 1988 use an average bacterial cell volume (Krambeck et al. 1981; Fry and Davies, 1985; A W David and J. H. Paul, submitted to J Microbiol Meth.) and use published estimates of g rams ca rbon/cubic micron (Bratback 1985; Bjornsen, 1986; Lee and Fuhrman, 1987) 10 Introducing additional conversio n factors increases the potential error into the productivity estimate. However, there are no other current means to estimate productivity from [3H]th ymidine incorporation data. Considering all of the above information, the validity of thymidine incorporation as a measure of bacterial gro wth to estimate productivity is contingent upon four factors and assumptions:


(i) the need to assess and correct for isotope dilution (ii ) the assumption that all bacteria ubiquitously label with [ 3H]thymidine (i.e. contain thymidine kinase) (iii) the need to determine the amount of radioactivity incorporated specifically into DNA (iv) the empirical relationship between thymidine incorporation and DNA synthesis. Each of these will be examined in greater detail below Isotope Dilution 11 An essential requirement for accurate [3H]thymidine incorporation into DNA measurements is the need to know the specific activity of the final precursor to DNA, dTTP (Moriarty, 1986). Due to the presence of more than one dTTP pool within the cell (see above) it is impossible to measure the specific activity of the dTTP pool used for DNA synthesis. Werner (1971) demonstrated that the rate of labelling of DNA by thymine and thymidine was greater than the labelling rate of the extractable dTTP pool. Isotope dilution is most commonly the result of de novo synthesis of thymine nucleotides. The most prevalent reaction is that of thymidylate synthetase (Figure 2) which catalyzes the methyl-transfer from 5:10-methylenetetrahydrofolate to the 5-position of dUMP. There is a simultaneous two_:electron reduction of the c1 carbon from methylene to a methyl group. The net result is the methylation of dUMP to form dTMP and the production of dihydrofolate (Pastore and Friedkin, 1962). This reaction increases intracellular pools of dTMP and effectively dilutes the d'IMP formed via the phosphorylation of


5,10-methylene tetrahydrofolate tetrahydrofolate dihydrofolate NADPH + W trimethoprlm amethopterin 12 Figure 2. Reactions involved in the de novo synthesis of dTMP. thyA, thyrnidylate synthase; folA, dihydrofolate FdU, of thymidylate synthase. Trimethoprim and of dihydrofolate


exogenously added [ 3 H]thymidine. Once this pool is diluted, it is no longer possible to assume that the specific activity of the dTTP pool is the same as the exogenously added [3H]thymidine (Moriarty, 1986). Another possible route of de novo synthesis of thymidine nucleotides is via cytidine nucleotides (Figure 1) Evident ly, this second pathway is considered minor, as it has received little if any discussion in the literature. 13 Isotope dilution is regulated by dTTP, which serves as an effective inhibitor of thymidylate synthase, nucleoside reductase, and deoxycytidylate deaminase (Moriarty 1986) Increases in dTTP will slow down or inhibit the supply of dTMP from de novo synthesis. It has been proposed that large concentrations of exogenous l y added thymidine will serve to provide large enough cellular dTTP pools such that isotope dilution will be inhibited. Moriarty and Pollard (198 1 ; 1982) have proposed the use of isotope dilution analysis to determine the specific activity of the dTTP pool Their technique is derived from the earlier work of Forsdyke (1968; 1971 ) and coworkers (Sjostrom and Forsdyke 1974; Scott and Forsdyke, 1976; 1980) whose original technique was developed using mammalian tissue cells. The technique calls for the addition of a fixed amount of radioactive thymidine and various arrounts of unlabelled thymidine to a series of replicate samples containing growing bacteria. The total concentration of added thymidine is plotted against the reciprocal of the radioactivity in the extracted DNA. If no isotope dilution has occurred from sources other than the unlabelled thymidine, the plot should pass through the origin. A negative intercept on the ordinate (total concentration of added thymidine) is representative of isotope dilution by other sources of


thymine bases in DNA (Pollar-d and M:>r-iar-ty, 1984). This value r-epr-esents the sum of all pools that dilute the [3H]thymidine pr-iorto incor-poretion into DNA (M:>riarty, 1986). A necessar-y condition for isotope dilution analysis to wor-k is that the rate limiting step forincor-por-ation of thymidine be the final DNA polymerase step. This condition is met in bacter-ia with normal r-egulatory mechanisms (Pollard and Mor-iar-ty, 1984) but may not be found in bacter-ia gr-owing in oligotr-ophic envir-onments (Mor-iarty, 1986). 14 It is impor-tant to note that DNA synthesis r-ates ar-e independent of the amount of added thymidine because of the double contr-ibution of de novo and salvage pathways which ensur-e sufficient amounts of dTTP necessar-y forDNA synthesis. Lar-ge exogenous pools of thymidine will allow the mor-e ener-gy efficient salvage pathway to supply the major-ity of dTTP forDNA synthesis. If an insufficient amount is provided by salvage pathways, then de novo synthesis contr-ibutes the r-emainder-. Regar-dless of which pathway co ntr-ibutes dTTP, the DNA synthesis r-ate is constant (M:>r-iarty, 1986). Isotope dilution may be caused by extr-acellularsour-ces of nucleobases, nucleosides, ornucleotides, orany compound which effectively competes forthe enzymes involved in thymidine salvage. It is commonly assumed that ambient concentr-ations of thymidine in seawater-ar-e negligible. This may pr-ove to be a false assumption as the components of dissolved o r-ganic matterin seawater-ar-e further c har-acter-ized DNA has been found to be a ubiquitous component of the DOMin mar-ine and fr-eshwaterenvir-onments (DeFlaun et a l ,1986; 1987) It seems likely that the components of DNA might also exist in sufficient amounts in seawaterdue to the pr-esence of


15 deoxyribonucleases (Maeda and Taga, 1973; 1974; Paul et al., 1987). The isotope dilution technique described above only estimates total dilution. It is unable to separate intracellular from extracellular sources of isotope dilution. Non-specific Macromolecular Labelling When was first proposed as a means to estimate bacterial DNA synthesis, one of the advantages was thought to be that it would specifically label DNA. Because thymine bases are only found in DNA (Kornburg, 1980), placing the radiolabel on the methyl group of thymidine would only label DNA since the uracil formed by the demethylation of thymidine would not be radiolabelled and radioactivity would not be incorporated into RNA. The released [ 3HJ could enter other biosynthetic reactions but it was believed that during short term incubations the amount of [ 3HJ incorporated into other macromolecules would be minimal (Moriarty, 1986) and preliminary reports supported this assumption (Fuhrman and Azam, 1980 ; 1982) While this apparently holds true for bacteria in culture Ccarmody and Herriott, 1970; Karl, 1982; Jeffrey and Paul, 1986a), there is overwhelming evidence indicating that significant percentages (often greater than 50%) of the radioactivity from [ 3 H-methyl]thymidine is incorporated into RNA, protein, and lipids (Find l ey et al., 1985; Jeffrey and Paul, 1986b; Robarts et al., 1986; Hollibaugh, 1988; Thorn and Ventullo, 1988; K. carmen et al., Lirnnol. Oceanogr in press). The biochemicial pathway by which non-specific labelling occurs in nature remains unknown and may reflect an absence of the process for bacteria in culture.


The amount of non-specific labelling of macromolecules appears to vary temporally and spatially and has been reported to positively correlate with primary production, bacterial activity, and seasonality for a lake in South Africa ( Robarts et al., 1986) Hollibaugh (1988) reported various levels of macromolecular labelling for two bays on the California coast. The percentage of label into DNA was found to vary between different sampling locations, the length of incubation, the amount of added thymidine, and whether the radiolabel was on the methyl group or the 6 position of thymidine. Carmen et al. (in press) reported that in coastal sediments a significant amount of radiolabel incorporated into DNA was due to recycling of catabolic products from thymidine. At this writing, the only general consensus regarding non..:.specific macromolecular labelling is that it occurs extensively in most environments and that the radioactivity incorporated into DNA needs to be separated from the other macronolecules. Ability of Bacteria to Take Up and Incorporate Thymidine The validity of the C 3H]thymidine technique also relies on the assumption that all of the active bacteria within the experimental sample are cap able of incorporating exogenous thymidine into DNA. Thymidine must be rapidly and efficiently transported into the cell before it can be incorporated into DNA and little is known about nucleotide transport systems in bacteria. The pyrimidines thymidine, uridine, cytidine, and deoxycytidine as well as the purines adenosine and guanosine h ave been shown to be transported intact into E. coli in an energy dependent process (Munch-Petersen et al., 1979) There is 16


17 evidence suggesting shared uptake mechanisms (Mygind and Munch-Petersen, 1975) and thymidine has been shown to compete with uridine for transport (Roy-Burman and Visser, 1981) and incorporation ( MJriarty, 1986). These non-specific transport systems will effectively allow the presence of other nucleosides to affect thymidine incorporation rates. The extent of these effects should be apparent from isotope dilution analysis (see discussion above) After thymidine is transported into the cell it must be phosphorylated to dTMP by the action of thymidine kinase (Figure 3). It is believed that the majority of environmental bacteria possess this enzyme (Fuhrma n and Azam, 1982) although studies exami ning the presence of thymidine kinase are limited. The fungi Neorospora crassa, Aspergillus nidulans, and Saccharomyces cerevisiae, the algae Euglena gracilus, Synechoccus, Thalassiosira, Isochrysis,and P latymonas, and the cyanobacteria Anacystis and Synechocystis all have been reported to lack thymidine kinase as have the nuclei of the eukaryotic algae Bryopsis, Chlamydomonas, Dictyota, Euglena, Padina, and Spirogyra (Moriarty, 1986). The lack of thymidine kinase in microeukaryotes results in little incorporatio n of [3H]thymidine by these organisms. The incorporated radioactivity is usually the result of recycling of catabolic products (Moriarty 1986). Rivkin ( 1986) and Rivkin and Voytek (1986), however, demonstrated [3H]thymidine i ncorporat i on by axenic laboratory cultures of marine diatoms and dinoflagellates during a 2-6 hr period prior to cytokinesis, but not during other times in the cell cycle. These investigators make note that their incubations were long and that they used very high concentrations of exogenous thymidine Their techniques would not preclude the use of


Phosphorylation ADP L Mg2+ T Thym idine k inase Thymidine SALVAGE PATHWAY Thymidi ne Monophosphate dTMP .l dTOP + dTTP J, DNA Figure 3. Thymidine kinase reaction and its relation to pyrimidine nucleotide salvage. 18


[ 3 H]thymidine to label primarily bacterial cells during short term incubation with the more commonly used lower thymidine concentrations. 19 The lack of thymidine kinase in eukaryotes and the preponderance of the enzyrre in bacteria makes [ 3H]thymidine incorporation an attractive means to measure bacterial activity. While thy midine kinase is assumed to be widespread in bacteria, studies examining the presence of the enzyrre in bacteria have been limited and some exceptions to the assumed presence of the enzyrre have been made. The most indepth studies have been performed by Saito and coworkers (Saito and Tomioka, 1984; Saito et al., 1985). In general, the actinomycetes and related organism s (e.g. mycobacteria, Rhodococcus, Streptomyces, Corynebacterium) have been found to lack thymidine kinase activity (Saito and Tomioka, 1984). In a subsequent report, Saito et al. (1985) examined 131 strains of 73 species and found that while t he majority had thymidine kinase activity, approximately had little if any enzyrre activity. The most notable groups lacking thymidine kinase activity were the pseudomonads Vibrio spp Candida, and the actinomycetes species. These four groups accounted for approximately half of the bacteria which lacked thymidine kinase activity. Other bacteria which have been demonstrated to lack thymidine kinase activity include Neisseria meningitidis (Jyssum and Jyssum, 1970), Acinetobacter calcoaceticus, Moraxella osloensis (Jyssum and Bovre, 1974), and Brevibacterium ammoniagenes (Auling et al., 1982) As yet, there has not been an in depth examination of marine bacteria for thymidine kinase Fuhrman and Azam ( 1982) used microautoradiography and co ncluded that only a very small portion of the active bacteria in their samples were unable to incorporate


thymidine Pseudomonas fluorescens and another pseudomonad isolated from a freshwater lake were unable to incorporate [ 3 H]thyrnidine ( Ramsay, 1 97 4 ) Pollard and Moriarty ( 1984) found that a few pseudomonad species were also unable to incorporate thymidine. Owing to the preponderance of Vibrio and Pseudomonas spp. in the marine env ironment, a significant percentage of the bacteria in a particular sample may be incapable of thymidine incorporation. A Direct Comparison Between f3H]Thymdine Incorporation and DNA Synthesis While [ 3H]thymidine incorporation remains the most commonly used means by which bacterial activity or productivity is estimated, there has yet to be a direct comparison between [ 3 H]thymidine incorporation and a nonisotopic measurement of DNA synthesis. Winn and Karl ( 1984) performed a direct comparison be tween [3H]adenine incorporation derived rates of RNA and DNA synthesis and compared them to known synthesis rates for the marine bacterium 20 Serratia marinorubra and three marine algae, Pavlova lutheri, Chlorella cordata, and Cylindrotheca sp. They used 32Po4 incorporation to get DNA and RNA synthesis rates for the bacterium and chemical determination of DNA and RNA for the algae. They reported close agreement between [ 3 H]adenine derived values and other nucleic acid synthesis rates. On average, in a heterogeneous environmental population, one must assume that 25% of the bases in DNA are thymine Therefore, for every dTTP incorporated into DNA by DNA polymerase, one dATP, dGTP, and dCTP should also be incorporated. Using an average molecular weight of a


21 DNA nucleotide base pair of 624 g/mol (Lewin 1987) four nucleotides would have a molecular weight of 1248 g/mol. One mole of thymidine incorporated would result in 1248 grams of DNA being synthesized. Since thymidine incorporation rates are most commonly given in terms of moles thymidine incorporated per liter per hour, DNA synthesis may be calculated by multiplying the thymidine incorporation rate by 1248. Once the thymidine incorporation rate has been corrected for isotope dilution and non...:.specific labelling of macromolecules, thymidine incorporation derived values and directly rreasured amounts of DNA synthesized should be in very close agreement If they are not, then some or all of the above assumptions need to be questioned or other factors not yet known may invalidate thymidine incorporation as a means by which to estimate DNA synthesis in bacteria. Detection of Natural Transformation There are three general types of bacterial genetic exchange. Conjugation is a type of mating between cells that is plasmid mediated and requires cell to cell contact resulting in the transfer of the conjugative plasmid and sometimes a single stranded copy of some or all of the donor chromosome to the recipient. Transduction originates from erroneous bacteriophage replication resulting in the transfer of small fragments of chromosomal DNA (Stewart and carlson, 1986). Natural transformation is the process by Which bacterial cells bind and take up extracellular DNA from their surrounding environment and incorporate it into one of the cell's replicons (Stewart and Carlson, 1986) Transformation was the first mechanism of genetic exchange reported for bacteria. The early work involved the exchange of a "transforming


factor" which changed morphology and virulence in Streptococcus pneumoniae (Griffith, 1928) Several years later, the work of Avery et al. (1944) identified the transforming factor as DNA. 22 It is important to note the difference between natural transformation and artificial transformation, a process commonly used with coli in molecular cloning techniques In artificial transformation, cells become competent (the cell state which allows the ability to take up DNA) as a direct result of physical, chemical, or enzynBtic manipulation. In contrast, the functions for natural transformation are encoded on the cell's chrorrosome. Consequently, competence is a heritable trait and is part of the normal physiology of the bacterium (Stewart and Carlson, 1986) There are five general steps involved in the process of natural transformation. In order, they are; (i) the deve lopment of competence; (ii) DNA binding to the cell, (iii) taking up of the DNA; (iv) the formation of a pre integration complex; and ( v) incorporation and integration of DNA into one of the recipient cell's replicons (Smith et al., 1981) In general, Gram..:.positive and Gram-negative bacteria appear to follow the same set of five steps during transformation. There may be, however, some differences in the actual mechanisms involved during each step for different types of organisms (Smith et al., 1981) The three genera of bacteria which have been studied extensively for their ability to naturally transform are Streptococcus, Haemophilus, and Bacillus although transformable species have also been identified in Streptomyces, Pseudomonas, Acinetobacter, Moraxella, Neisseria, Achromobacter, Methylobacterium organophilum, Azotobacter


23 vinelandii, Micrococcus radiodurans, Mycobacterium (Stewart and Carlson, 1986) and the cyanobacteria Synechoccus (Chauvat et al., 1983) and Anacystis nidulans (Golden and Sherman, 1984). The recent review of natural transformation by Stewart and Carlson ( 1986) and the earlier review by Smith et al. (1981) contain full accounting on the mechanisms involved in natural transformation. For a complete description of the natural transformation process, the reader is referred to these reports. There have been many studies examining genetic exchange by conjugation (Mach and Grimes, 1982; Gealt et al., 1985; McPherson and Gealt, 1986; Bale et al., 1987; Trevors and Starodub, 1987; O 'Morchoe et al., 1988; Genther et al., 1988) and transduction (Saye et al., 1987; Germida and Khachatourians, 1988) in environmental bacteria. However, the detection of natural transformation in the environment has never been documented. In contrast to conjugation and transduction, natural transformation involves the incorporation of extracellulr DNA and is thus susceptible to DNase activity. In addition, competence is not a constitutive process of the vast majority of transformable bacteria. Therefore, natural transformation in the environment is directly correlated with the stability of extracellular DNA and environmental factors which induce competence (Coughter and Stewart, in press). The ubiquitous presence of extracellular DNA in aquatic environments is now well documented (DeFlaun et al., 1986; 1987) and turnover times for dissolved DNA may range from 6 to 15 hr (Paul et al., 1987; Paul et al., unpublished results). It now becomes important to determine whether the potential exists for the dissolved DNA pool to serve as a source of gene sequences for natural transformation in the


environment. Deter:mining the ability of environmental bacteria to naturally transform will yield valuable information in two areas of microbial research. It will give insight into the natural flux of microbial genes through the environment and genetic adaptation by bacteria to stress and selective pressures brought on by environmental changes. Of a more timely importance, detection of natural 24 transformation in the environment could be very important in determining risk assessment for the release of genetically engineered microorganisms (GEM's) into the environment. The debate over the release of GEM's is quite intense and well documented (Tangley, 1985; Stotzky and Babich, 1986; Sharples, 1987; Davis, 1987; Baskin, 1987; Fox, 1988; Marx, 1987; Crawford, 1987; Williams, 1987; US Congress OTA, 1 988). The concerns about the safety of released GEM's are well publicized and discussed in many publicationns (e.g. General Accounting Office, 1988; National Academy of Sciences 1987). Release has been limited to a very few field tests (Lindow et al., 1988; Lindow and Panopoulos, 1988; Drahos et al., 1988; Bishop et al., 1988) and the debate over the safety of such releases continues Once a GEM is released, whether intentionally or accidentally, it will eventually die and a part of its recombinant DNA sequences will become part of the extracellular DNA pool. It is imperative to determine whether these recombinant DNA sequences may transform the ambient microbial populations and thus propagate the recombinant sequence in the environment A limited number of studies have examined transformation by bacterial isolates in sediment microcosms DNA, whether it is extracellular (Maeda and Taga, 1974; Lorenz et al., 1981; Aardema et


al., 1983; Lorenz and Wacker.nagel, 1987) or associated with dead cells (Novitzky, 1986) is known to be resistant to DNase degradation in sediment environments. Graham and Istock (1978; 1979) demonstrated transformation by Bacillus subtilis in sterile soils. After 8 days, 79% of the isolated organisms exhibited altered genotypes. Transformation frequencies for subtilis associated with sterile sea sand were 25 to 50 times higher than in liquid culture transformations (Lorenz et al., 1988). However, it is unknown whether these sterile microcosms accurately reflect interactions in natural microbial communi ties. 25 Plasmids are one type of extrachromosomal genetic element found in bacteria. They are circular DNA molecules which replicate autonomously to the host chromosome (Lewin, 1987). They occur frequently in both Gram-positive and Gram-negative bacteria (Willets, 1985) and they often carry genes coding for extremely important functions such as antibiotic resistance (Levy et al., 1981; Obrien et al., 1985; Baya et al., 1986; Frederickson et al., 1988) or the abilty to degrade certain compounds and thus allow growth in a wider range of environments However, plasmids are often recognized as dispensable because there is little evidence to suggest that plasmids in nature carry genes with essential functions s uch as respiratory genes (Beringer and Hirsch, 1984). The flux of antibiotic resistance via plasmids is well documented (Levy et al., 1981; Beringer and Hirsch, 1984; Obrien et al., 1985; Davies, 1986; Baya et al., 1986; Levy and Marshall, 1988) In addition, plasmids are perhaps the most popular vector for genetic manipulation in bacteria. These manipulations often include the insertion of antibiotic resistance genes which allow for selection to detect DNA


26 transfer between microbes plants, and animals (Davies 1986). Previous studies on plasmid mediated natural transformation have all been under laboratory conditions and have concentrated on five species, Bacillus subtilis, Streptococcus pneumoniae, Haemophilus influenzae, Azotobacter vinelandii, and Neisseria gonorrheae There are several similarities as well as differences between these organisms in their mechanisms for plasmid transformation. Plasmid transformation of B. subtilis occurs at a much lower frequency than does transformation with chromosomal DNA. Chromosomal transformation requires uptake of approximately one molecule of transforming DNA per transformant While plasmid transformation requires uptake of 103 to 104 molecules per transformant ( Contente and Dubnau, 1979). Only multimeric forms of plasmids may transform B. subtilis (Canosi et al., 1978; Mottes et al., 1979) These may be natural forms or those made by linearization and ligation of monomeric plasmid molecules. B subtilis transforms by first binding DNA, nicking it, then taking up one strand as the other strand is degraded ( de Vos et al., 1981). Immediately after uptake, monomer DNA is always found as single stranded (ssDNA) with a shorter length than the parent plasmid while multimeric plasmid DNA is found as ssDNA, dsDNA, and partially dsDNA (de Vos et al. 1981) The double strand is the result of annealing and synthesis of the second strand. Multimers are believed 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 recE4-independent


reannealing event. DNA synthesis converts them into double stranded DNA and superhelical turns are introduced to result in a covalently closed circular monomeric plasmid (de Vos et al., 1981) Monomer plasmid will transform if the recipient cell contains a plasmid with homologous DNA ( Contente and Cubnau, 1979) or if the transforming plasmid contains a sequence hO!IDlogous to recipient chromosomal DNA (Bens i et al., 1981; Canosi et al., 1981) Transformation with homologous DNA is recE4 dependent (Canosi et al., 1981 ) Transformation will also repair deficiencies in homologous DNA in the recipient (Iglesisas et al., 1981) In addition, the entire plasmid may enter a recipient's chromosome This process may interupt a gene function which returns if the plasmid is excised from the chromosome forming an intact autonorrous plasmid within the cell (Hofmeister et al., 1983). In contrast to B subtilis, S. pneumoniae may be transformed with monomeric plasmid DNA. However, this process occurs with two hit kinetics. Two rronomer i c strands enter the cell separately and associate to form a duplex The remaining gaps are regenerated resulting in an intact plasmid (Saunders and Guild, 1981a). The process is not perfect as regenerated plasmids often contain deletions (Saunders and Guild, 1981a). Multimeric plasmids transform with first order kinetics (Saunders and Guild, 1981b). Open circular and linear forms of plasmids also transform, but at 35-fold lower frequencies than circular forms (Saunders and Guild, 1 981b). All plasmid transformations are very inefficient, requiring the uptake of 10 to 103 donor strands per 27


recipient. After DNA is nicked during binding to the cell surface, it is cleaved by a subsequent break in the complementary strand. Continuous processive membrane nuclease activity degrades the co mplimentary strand as the original nicked strand enters the cell (Lacks, 1979). Interactions between homologous sequences in the recipient and transforming DNA in S. pneumoniae are similar to those observed for B. subtilis. Transformation in the presence of homologous DNA requires the uptake of only one plasmid molecule (Lopez et al., 1982) Monomeric plasmid containing homologous chromosomal DNA will transform a pproximately ten times more than plasmid without the chromosomal insertion. Recombination and exchange between homologous sequences in recipients and transforming DNA is similar to that observed in subtilis (Lopez et al., 1982) The amount of exchange is proportional to the length of t he homologous sequence While transformation with plasmid DNA for B subtilis and S. pneumoniae is generally similar in f unction to transformation with chromoso mal DNA, there is apparently some difference between transformation with the two types of DNA in H influenzae. Transformation occurs at much greater frequencies with chromosomal DNA. 28 Monomeric plasmids will transform although multimeric forms are favored (Notani et al., 1981). High concentrations of divalent cations stimulate circular or linear plasmid uptake but have no effect on chromosomal DNA uptake (Gromkova and Goodgal, 1981). I t is important to note that cations effect uptake in cells which are a l ready c ompetent cations do not create competence as they do for E coli (Gromkova and Goodgal, 1 981). H. influenzae will only transform with


homologous DNA and recognizes an eleven base pair sequence for DNA binding (Danner et al. 1980) and the receptor sites are the same for chromosomal and plasmid DNA (Gromkova and Goodgal, 1981) Donor DNA enters Hemophilus in specialized membrane extensions called transformasomes that serve to facilitate transport of the DNA across the cell wall. Linear DNA then undergoes degradation of one strand as it exits the transfonnasome (Barany et al., 1983) before it recircularizes and the second strand is synthesized. 29 Plasmid transformation will increase greatly if the recipient cell contains a homologous DNA sequence (Albritton et al., 1981). However, if a chromosomal insert is contained on the plasmid, the marker is usually transferred to the recipient chromosome and the plasmid is lost (Setlow et al., 1 981). Establishment of the plasmid under these circumstances is two orders of magnitude less than marker transfer to the chromosome. In limited cases, chromosomal sequences have been inserted into the plasmid during recombination events (Setlow et al., 1981). Stuy and Walter (1986) have reported, however, that homology facilitated transformation in H. influenzae was similar to that of B subtilis and S pneumoniae Intact plasmids were recovered after transformation. Like H. influenzae, Azotobacter vinelandii is transformed at much lower frequencies by plasmid DNA (Doran et al., 1987). The binding and uptake mechanisms are similar for both forms of DNA. Linearized plasmid DNA has a 2 to 3 times greater transformation efficiency than does closed circular plasmid DNA (Dora n et al., 1987) No deletions were reported in plasmids isolated from transformants Transformation is increased 10 to 50 fold if a chromosomal insert is included on the


30 transforming plasmid. However, the marker is transferred to the chromosome through a recomb in at ion event and the plasmid is not maintained in the recipient (Doran et al., 1987) Other studies have derronstrated that plasmid DNA mai ntenance in a transformant is physiologically demanding and may disrupt normal cell functions such as nitrogen fixation and cell size (Glick et al., 1986) Like the other organisms examined, chrorrosomal and plasmid DNA are taken up with the same efficiencies by N. gonorrheae (Biswas et al., 1986). However, the transformation frequency for plasmid DNA is orders of magnitude lower Transforming plasmid DNA enters gonorrheae in linear double stranded form (Biswas and Sparling, 1981; Biswas et al., 1986) although the rare entrance of intact circular rrolecules has not been ruled out (Biswas et al., 1986). Transformation in this organism is also facilitated by the presence of homologous plasmids in the recipient cell (Biswas et al. 1982). When no homologous sequences are present in the recipient, 10 to 1 OO% of the transformants contain plasmids with significant deletions (Biswas et al.; 1982). Deletions are random and are not a function of the plasmid DNA since the same plasmids are recovered intact from transformed E. coli (Biswas et al. 1982) Several generalities on plasmid transformation may be drawn from the preceeding discussion. Plasmid transformation occurs at very low frequencies compared to chromosomal DNA mediated t ransformation. Plasmid monomers have either no or very low transforming capacity. Multimeric forms transform more frequently or in some organisms, two hit kinetics has resulte d in transformation by monomers. Plasmids isolated from transformants often have sequence deletions.


Transformation is greatly enhanced by the presence of horoologous sequences within the recipient cell that may be located on either the chromosome or resident plasmids within the recipient cell. Due to the low frequencies at which plasmid transformation occurs, detection of this process may prove difficult and time consuming Isolation of transformants by plating on selective media is difficult due to the large ambient antibiotic resistant bacterial populations (Baya et al., 1986; Alvero, 1987). Detection of gene sequences in transformants is a promising technique (Ogram and Saylor, 1988) However, limits in sensitivity may hinder its use when low 31 transformation frequencies occur and when natural antibiotic resistance is high. In addition, both of these techniques may be time consuming Results may not be available for 2 to 7 days. [3H ]Thymidine incorporation as a means to measure bacterial activity and growt h may provide an alternative. This technique i s very sensitive in measuring small changes in activity (Jeffrey and Paul, 1986a) [3H]Thymidine incorporation could be measured and compared between samples that received transforming DNA and samples that did not. In the presence of selective pressures for transformants (i.e. antibiotics coinciding with the resistance genes carried on t he tranforming DNA), higher [ 3H]thymidine incorporation rates in the transforming DNA amended sample would indicate expression of the antibiotic resistance gene in that bacterial population. I n this way, a quantitative measure of growth could be compared between the sample receiving transforming DNA and growth in the ambient antibiotic resistant population. The difference in rates would be due to growth facilitated by the incorporation and expression of the transforming resistance gene


32 The objectives of this study were to examine each of the four factors which effect the accuracy of thymidine incorporation (isotope dilution, non-specific labelling, the ability of the bacteria to incorporate thymidine, and the empirical relationship between thymidine incorporation and DNA synthesis) to determine the significance of each during routine thymidine incorporation assays. Once the limitations of the technique were determ ined, [ 3H]thymidine i ncorporation was evaluated as a methcxi to detect natural genetic transformation in aquatic bacteria.


CHAPTER 2: ISOTOPE DILUTION Introduction The validity of the [3H]thymidine incorporation technique has come into question owing to the uncertainty in estimating intracellular and extracellular thymine nucleoside and nucleotide pools (termed isotope dilution; Bell, 1986; Pollard and Moriarty, 1986). Intracellular isotope dilution is caused by de novo synthesis of pyrimidines in bacteria. The enzyrre thymidylate synthetase (EC 2.1. 1.45) catal yzes the conversion of dUMP to dTMP (Kornberg, 1980). dTMP synthesized by this pathway dilutes the radiolabelled dTMP which has been formed from exogenous [ 3H]TdR (Pollard and Moriarty, 1984) Extracellular isotope dilution may be caused by ambient concentrations of thymidine or compounds which compete for the same enzymes as thymidine during salvage of exogenous nucleosides. Extracellular isotope dilution is usually considered to be insignificant, although hydrolysis of extracellular DNA (DeFlaun et al., 1986) by de o xyribonucleases (Maeda and Taga, 1973; 1974; Paul et al., 1987) may provide a significant source of dissolved nucleotides, nucleosides, and nucleobases in aquatic environments In this part of the study the extent of isotope dilution was examined in a variety of environments. Additionally, the effect of 5-fluoro 2' deoxyuridine (FdU), an irreversible inhibitor of thymidylate synthetase (Danenburg and Lockshin, 1981), on isotope dilution was examined in an effort to 33


inhibit intracellular isotope dilution and thus allow the separation of intracellular from extracellular dilution. Methods and Materials 34 Sampling sites. The sampling sites included Bayboro Harbor, a eutrophic embayment o f Tampa Bay St. Petersburg, FL., and the Alafia River (Figure 4) which drains approximately 100 km2 of farm l and and phosphate mining and processing operations (Paul et al., submitted to Limnol. Oceanogr.) with an estimated discharge into Tampa Bay east of Tampa, FL of 13,000 L/sec. The Medard Reservoir, which drains into the Alafia River, was also sampled. The Crystal River (Figure 4) Crystal River, FL, is a spring fed river with a 40 krn2 watershed and a discharge rate of 27,600 Lisee (Paul et al., submitted ) An offshore station in the Gulf of Mexico and Charlotte Harbor, an estuary located in Southwest Florida, and Lake Maggiore, St. Petersburg, FL were also sampled. Materials. 5-Fluoro and thymidine were obtained from Sigma Chemical Co., St. Louis, MO. [Methy l 3 H]Thymidine (60 -80 Ci/mmole) was obtained from ICN Pharmaceuticals Inc., Irvine, CA. (15.3 Ci/mmole) was obtained from New England Nuclear, Boston, MA. Isotope Dilution Assays. Isotope dilution assays were conducted at several different locations by the method of Moriarty and Pollard (1981, 1982) with thymidine concentrations ranging from 2.5 to 40 nM. In some instances, a parallel series of flasks was amended with 12. 5 nM FdU. Incubation times were minutes. In the presence of FdU, the of an isotope dilution plot equals the arrount of


A Tampa Bay 8 Gulf of Mexico N t W(DAAO IUJE,.WO.,. ) / Figure 4 Location of sampling sites along the Alafia and Crystal Rivers. 35 3


36 extracellular isotope dilution, assuming thymidylate synthetase has been inhibited. High concentrations of exogenous thymidine have been reported to inhibit intracellular isotope dilution by causing feedback inhibition of thymidylate synthetase (Pollard and Moriarty, 1984). FdU was examined for the ability to inhibit intracellular isotope dilution to the same extent as high thymidine concentrations by measuring thymidine incorporation at 2.5 nM C3H]thymidine in the presence and absence of 12.5 nM FdU and comparing with thymidine incorporation at 35 nM [3H]thymidine. In a separate experiment, intracellular isotope dilution was estimated by noting the differen ce in rates of 2.5 nM [ 3 H]thymidine incorporation into DNA in the presence and absence of 12.5 nM FdU. Using the difference in incorporated DPM 's and the known specific activity of the added [3H]thymidine, the amount of intracellular isotope dilution was calculated. Total isotope dilution (Pollard and Moriarty, 1984) was determined and extracellular dilution was calculated by subtraction. Results Table 2 summarizes the isotope dilution values determined for a variety of environments. Only four samples had significant amounts of dilution. In two of the three experiments where intracellular and extracellular dilution were separated, 94% and 89% of the isotope dilution could be attributed to extracellular sources for Bayboro Harbor and Charlotte Harbor, respectively.


Table 2. Summary of isotope dilution values Location Total Dilutiona Bayboro Harbor 7 /85 8.36 + 1.55 Bayboro Harbor 7 /86 2 .07 + 0 72 Bayboro Harbor 1/87 3 .33 + 0 .52 Bayboro Harbor 6/87 0. 73 + 0.96 Bayboro Harbor 8/87 3.16 + 2 .85 Bayboro Harbor 10/87 1.00 + 2.23 Crystal River #1 4 /87 1.34 + 0 .69 Crystal River #4 4/87 2 .05 + 1 .20 Crystal River #2 7 /87 0 97 + 2 .80 River 117 4 /87 Alafia 1.54 + 2 .19 Medard Reservoir 4/87 0. 08 + 0. 71 Medard Reservoir 9 /87 2.78 + 1.90 Offshore 5/85 0.99 + 0.76 Charlotte Harbor 8/85 4.07 + 1.75 Lake M:l.ggiore 11 /85 . 0 73 + 1.63 a Values presented as equivalent nM concentrations of thymidine the 95% con fidence interval Not significantly different from zero ( P > 0 95; Student's t test; Zar, 1984) The results of two isotope dilution assays conducted in the presence and absence of FdU are shown in Figure 5 From the 37 Y-intercepts, total isotope dilution was the equivalent of 8.36 1.55 nM thymidine (95% confidence interval) and 4.07 1.75 nM thymid ine for Bayboro Harbor and Charlotte Harbor, respectively. The Y-intercepts from the FdU treatments yield extracellular isotope dilution equivalent to 7 .89 0.93 nM thymidine and 3 .64 1.49 nM thymidine for Bayboro Harbor and Charlotte Harbor, respectively.


Q) c: : E >.._ 20 :!! 10 c: Bayboro Harbor 20 10 0 6 0.9 104 /DPM Charlotte Harbor 1 2 3 104 /DPM Figure 5. Isotope dilution analysis of Bayboro Harbor and Charlotte Harbor. (A) Data for Bayboro Harbor. The total dilution was found to be 8.36 + 1.55 nM thymidine, the extracellular dilution determined from the y-intercept of-the FdU treatments was equivalent to 7.89 + 0.93 nM, and the intracellular isotope dilution was equivalent tc 0.47 nM thymidine. TB) Data for Charlotte Harbor. The total dilution was found to be 4.07 + 1.75 nM thymidine, the extracellular dilution was equivalent to 3.64 + 1 .49 nM thymidine, and the intracellular isotope dilution was equivalent to 0.43 thymidine. w co


39 600 A B 2 5 nM 2 5 nM TdR 12 5 nM FdU 200 !..I ci. 90 ... MIN 0 c (J .51000 34. 7 nM TdR a: 'C ... 0 G) 0 E 600 a. 200 90 MIN Figure 6. Inhibition of isotope dilution by FdU in Harbor. [ H]Thymidine incorporation is shown at 2 5 nM [ (A), in the presence of 12.5 nM FdU (B) and at 35 nM [ H]thymidine (C). Rates of incorporation into DNA were 88.0, 163.2, and 167.5 pmol/L hr for panels A, B, and C, respectively. Equivalent rates of incorporation into DNA fo r panels B and C co nfirm the inhibiti o n of intracelluar isotope dilution.


Intracellular isotope dilution was determined to be negligible, 0.47 nM for Bayboro Harbor and 0.43 nM for Charlotte Harbor FdU inhibited intracellular isotope dilution to the same degree as did high concentrations (35 nM) of thymidine (Figure 6). [3H]Thymidine added at 2.5 nM was a sufficiently low concentration to enable intracellular isotope dilution to occur. The calculated rate of thymidine incorporation into DNA was 88.1 pmoles thymidine/Lh (Figure 6a). The addition of 12.5 nM FdU inhibited intracellular isotope dilution and resulted in a rate of thymidine incorporation into DNA o f 163.2 pmoles/Lh (Figure 6b). The addition of 35 nM [3H]thymidine (Figure 6c) resulted in a thymidine incorporation rate of 167.5 pmoles/Lh. In another Bayboro Harbor experiment, total isotope dilution was determined to be 3. 33 nM. Intracellular and extracellular isotope dilution were calculated to be 1.90 nM and 1.43 nM, respectively. Using this information, the data in Table 3 were derived. The rates corrected for isotope dilutio n are in good agreement over the concentration range of thymidine employed. 40


Table 3. C3H]Thymidine incorporation into DNA before and after correction for isotope dilution [ 3H]TdR incorpora3iona at following cone. of added [ H]TdR (nM) 2 .5 35 50 COntrol c Corrected 159.4b 206.7 229;2 226.6 244.4 a Values expressed as pmol/L hr into DNA b Treatment in the presence of 12.5 nM FdU c Values are corrected for extracellular isotope dilution only. FdU and high thymidine concentrations presumably inhibit intracellular isotope dilution Discussion In the limited examples where isotope dilution was significant, FdU has been shown to inhibit intracellular isotope dilution and has allowed intracellular and extracellular isotope dilution to be separated and ind icated that intracellular isotope dilution was minimal. Addition of high [3H]thymidine concentrations (e.g. 35 nM) either rendered the amount of intracellular isotope dilution negligible, or prevented it by feedback inhibition of thymidylate synthase (Pollard and Moriarty, 1984). Extracellular dilution was found to be significant in only two of fifteen cases examined (e. g. Bayboro H3.rbor and Charlotte Harbor, Figure 5 ; Table 2) The nM thymidine equivalent values presented for extracellular isotope dilution do not necessarily represent ambient thymidine concentrations in seawater Extracellular isotope dilution 41 could be caused by the presence of any nucleic acid component which may compete with thymidine for the same enzyrJEs in pyrimidine salvage


pathways. Broad specificity ribonucleoside and deoxyribonucleoside transport systems (Munch-Petersen and Mygind, 1983) and non-specific pyrimidine ribonucleoside kinases (Newhard, 1983) have been identified for bacteria in culture. Reduced rates of [3H]thymidine uptake in the presence of thymine ( J. Paul, unpublished results) and reduced [3H]thymidine incorporation in the presence of uridine (Moriarty, 1986; data not shown) and deoxyadenosine (data not shown) imply the presence of non-specific salvage pathway enzymes in marine bacteria. Although the exact nature of extracellular diluting compounds is 42 unknown, ambient concentrations of thymidine in Bayboro Harbor of 5 nM (1.2 do not seem unreasonable. An average extracellular DNA value for these waters is 10 DNAIL (DeFlaun et al., 1987). Using an average molecular weight of DNA of 1 248 g/mole (Lewin, 1987) and considering the presence of extracellular nucleases (Paul et al., 1987), it is conceivable that a compound Which is approximately 19% of the weight of DNA might be present in seawater at levels of 1 2 % of the weight of DNA present, although turnover times for thymidine would be much faster than those for DNA. Although high isotope dilution values obtained in Bayboro Harbor and Olarlotte H3.rbor are conceivable, they appear to be exceptions. Isotope dilution measurements from a wide variety of environments indicate that dilution is insignificant in most instances (Table 2). The results indicate that, with the exception of three eutrophic locations (Bayboro H3.rbor, the Medard Reservoir, and Charlotte Harbor), isotope dilution is a negligible factor and its significance would be greatly reduced especially if C3 H]thymidine were ad ded at 35 nM. When significant dilution does occur, FdU has proven as effective


in inhibiting intracellular isotope dilution as high concentration of ekrvnorcX thymidine. Under these circumstances, intracellular and extracellular dilution may be separated. However, the Moriarty and Pollard (1981, 1982) method cannot determine whether all dilution has been accounted for. There is a need to verify the accuracy of this technique. As will be discussed in Chapter 4, the isotope dilution technique grossly underestimates the number of thymine bases incorporated into DNA. 43


44 CHAPTER 3: MACROMOLECULAR LABELLING Introduction Incorporation of radioactivity from [3H]thymidine into macromolecules other than DNA (Fallon and Newell, 1986; Hollibaugh, 1988; Robarts et al., 1986) also raises questions about the validity of the method. Macromolecular labelling patterns have been found to vary both temporally and spatially ( Robarts et al., 1986; Hollibaugh, 1988) yet factors which affect labelling patterns unknown. The biochemical pathways involved in if ic labelling are also unknown, mst likely because bacteria in culture incorporate [ 3H]thymidine solely into DNA (carmody and Herriott, 1970; Karl, 1982; Jeffrey and Paul, 1986a). In this part of the study, environmental factors which might affect macrorrolecular labelling patterns were examined. In addition, specific metabolic inhibitors were used in an effort to determine by what pathways labelling occurs. Methods and Materials Sampling sites. The sampling sites included Bayboro Harbor, the Medard Reservoir, and the Alafia and Crystal Rivers (Figure 4). Both rivers were sampled six times between February 1987 and April 1988 between 8 and 10 am. All samples were processed immediately (Paul et al., submitted). Additionally, a coral surface microlayer (CSM) from the Dry Tortugas (Paul et al., 1986) an offshore station in the Gulf of


Mexico located at 25' N, 82'W, and Charlotte Harbor, an estuary located in Southwest Florida, were also sampled. Materials. 5-Fluoro 2'-deoxyuridine thymidine, and trimethoprim, were all obtained from Sigma Chemical Co., St. Louis, MO. [Methyl..:.3 HJThymidine (6o..:.8o Cilmmole) was obtained from ICN Pharmaceuticals Inc., Irvine, CA. [6..:.3 H]Thymidine (15. 3 Ci/mmole) was obtained from New England Nuclear, Boston, MA. Macromolecular Fractionation. Acid/base hydrolysis of 45 macromolecules was performed by the method of Riemann and Sondergaard (1984). Total macromolecules (here defined as DNA, RNA, and protein) were collected by precipitation with cold 5% trichloroacetic acid (TCA). RNA was hydrolyzed in a subsample which was treated with 1 N NaOH (final concentration) for 1 hr at 60C, acidified with TCA, and chilled. DNA and RNA were hydrolyzed in a third subsample by treatment in 20% TCA for 30 min at 95C. The precipitates from each treatment were collected by filtration and radioactivity was determined by liquid scintillation counting. The majority of samples were collected on 0 2 llffi pore diameter Nuclepore (Pleasanton Ca.) filters which were prepared for scintillation counting by the method of Kobayashi and Harris (1978) In a few experiments, total macromolecules and the warm NaOH treated sub sample were collected on Millipore GS (0. 22 llffi pore diameter; Millipore Corp., Bedford, Mass.) filters. These filters were treated with 0.5 ml HCl at 95 for 15 min. After the filters had cooled 1 ml of ethyl acetate was added to dissolve the filters followed by 10 ml Aquasol 2 (New England Nuclear). Failure to first treat Millipore filters with 0.5 N HCl resulted in significantly lower recovery of radioactivity and subsequently reduced rates of


[3H]thymidine incorporation even though efficiencies, as determined by the sample channels ratio method, were unchanged (data not shown) Millipore filters were not used to collect protein fractions (hot acid treatments) as radioactivity retention was approximately double that of Nuclepore filters (data not s hown). It was assumed that this was due to the more reactive nature of the mixed ester construction of Millipore filters which bound radioactive macromolecular hydrolysates. Seasonal patterns of macromolecular labelling. Surface water samples were taken at least once per month from Bayboro Harbor from November 1 985 through December 1987. Subsamples were taken for bacteria l direct counts (Porter and Feig, 1980). In mst samples, temperature was measured and recorded and salinity was determined by refrac tometry for samples taken during the second year. Water samples were amended with 2.5 nM [3 H]thymidine and radioactivity incorporated into the different macromlecular fraction determined during 90 min incubations. Effect of exogenous thymidine concentration on mac romolecular labelling patterns in different size fractions. The effect of 46 thymidine concentrations on macromolecular labelling patterns was determined in water samples from Bayboro Harbor. Repli cate samples were amended with thymidine a t final co ncen trations ranging from 2.5 to 500 nM. At each sampling time, one subsample from each treatment was subjected to acid-base hydrolysis. A second subsample was first passed through a 1 .0 urn pore size Nuclepore filter and the filtrate treated to hydrolysis.


Correlations of labelling patterns with environmental parameters. The timing of this project coincided with a National Science Foundation project entitled "Extracellular DNA in the rivers of southwest Florida." As part of this project, six sampling trips were made to the Crystal and Alafia Rivers during a 14 rronth period between February 1987 and April 1988. Additionally, 52 hr diel studies were m ade of station 2 of the Crystal River and The Medard Reservoir. These samplings included measurements for [ 3H]thymidine incorporation, primary production, bacterial counts, chlorophyll particulate DNA, dissolved DNA, dissolved organic ca rbon, temperat ure, salinity, and nutients (N and P). The methods have been described in Paul et al. ( submitted ) These data were made available for compariso n with the percentage of radiolabel incorporated into DNA during [3H]thymidine incorporation measurements in the study. Correlations between the 47 arcsine transformed percent (Zar, 1984) of radiolabel incorporated into DNA were determined using the Regress II statistics package (Human Systems Dynamics) for the Apple lie computer. Results Linearity of labelling patterns. Repeated experiments hav e demonstrated that total [ 3 H]thymidine incorporation is linear over time (data not shown). Figure 7 indicates that the percentage of radioactive label incorporated into each o f the three macromolecular fractions was constant with time. This fact was used in later experiments allowing only beginning and endpoints to be sampled. Spatial variation of labelling patterns. Table 4 provides a summary of the percentage of radiolabel incorporated into DNA (%DNA)


600 c. ... 0 400 (.) c: -a: tn Q) RNA 0 E c. 200 Protein 20 40 60 Time in Min Figure 7. Linearity of macromolecular labelling during [3H]thymidine incorporation in Bayboro Harbor waters. Lines plotted were determined by linear regression t-nalysis of the data for each macromolecular fraction. R values were 0 999, 0.990, and 0.920 fo r the t otal, warm NaOH, and hot acid treatment s respectivel y 48


49 Table 4. Spatial in percent of radiolabel incorporated into DNA during [ H]thymidine incorporation % Radiolabel into DNA Location Minimum M3.ximum M3an S D N Bayboro 6 1 95.5 55.8 2 3.6 26 Crystal R 2 9 4 57.2 45.5 8. 1 16 C.R. st 1 26 88 52. 8 26. 5 5 C R st 2 29.4 57.2 44. 3 9 5 6 C .R. st 3 31.4 56.8 47.9 9 5 6 . C R st 4 39.8 50.8 44. 7 5 0 6 C.R. diel 29.4 64 46. 0 10. 5 13 Alafia 28. 8 78. 6 51.8 11. 2 48 A R st 1 42. 2 70. 9 54.7 12.4 6 A.R. st 3 53. 5 61.6 57.0 3.4 6 . A.R. st 5 40.3 68. 8 53. 5 10. 4 6 A .R. st 6 39. 7 78.6 57. 3 13.5 6 A.R. st 7 3 7 0 59. 4 50. 2 9.4 6 A.R. st 8 28. 8 52.9 45.7 9 5 6 Medard 32. 6 59.6 40. 2 10. 3 6 Med diel 50. 2 62. 5 55. 7 3 6 13 Both Riv 28. 8 78. 6 50. 2 10. 8 64 All samp 6 1 95. 5 51.6 14.4 121 in the environments sampled. As is evident, non.::.specific labelling varies over a wide range between different locations. The greatest range was recorded dur ing the seasonal study of Bayboro Harbor while the diel sampling of the Medard Reservoir had the least variability. However, t he averages of all groups of samples ranged between 40 to 60% with an a verage for all data of 51. 6 14. 4 percent. Seasonal patterns. Seasonality of non-specific labelling is shown in Figure 8. The percentage of label incorporated into DNA


50 increased markedly during the winter months of 1986. However, the onset of spring reversed this trend dramatically and non -specific labelling increased through the sunmer months. During the fall of 1986, the trend again reversed itself but no consistent pattern was demonstrated from that point through all of 1987. The amount of label incorporated into RNA was inversely proportional to the amount o f label incorporated into DNA. In many respects, the RNA curve (F i gure 88) is the inverse image of the DNA curve (Figure 8A). A noticeable difference occurs in the second half of 1987 where the percentage of radioactivity incorporated into RNA steadily decreases from mid summer through late fall. The percentage of radiolabel in DNA during that time was much more irregular. The pattern of the amount of label into protein (Figure 8C) also follows that of RNA until the end of 1 987. Labelling patterns as a function of exogenous thymidine concentration and size fraction. The effect of exogenous thymidine concentration on rates of incorporation between different size fractions is presented in Table 5. Rates of incorporation increased as thymidine concentrations increased up to 125 nM, most likely due to isotope dilution. As thymidine concentrations increased, the Table Effect of exogenous thymidine concentrations on total rates of incorporation in different size fractions pmoles/L hr (percent of whole sample) TdR cone. whole < 1 wn > 1 lJITl 2.5 149.4 + 7.0 (100) 96. 0 + 8 6 ( 64) 53. 4 ( 35) 35 187.4 + 7.9 (100) 11 4 7-+ 7. 0 ( 61 ) (39) 125 229.4 + (100) 170;1 + 15.7 ( 74) 59;3 (26) 500 242. 4 + 28. 3 (100) + 15.7 (86) 32.-8 (14)


100 80 60 40 c w !;( 20 a: 0 Q. a: 0 (,) 100 z > 80 1> 1-60 Q 40 0 < a: 20 ..I 0 1LL. 100 0 1-80 z w (,) a: w Q. 20 DNA : RNA : PROTEIN N D J D 1986 FMAMJJASOND 1987 Figure 8 Seasonal pattern of macromolecular labelling in Bayboro Harbor 5 1


Table 6. Macromolecular patterns as a function of exogenous thymidine concentration and size fraction 2.5 nM 35 nM 125 nM 500 nM -Totb d <1 ).illlc >1 )..1ITl tot <1)..1ITI >h.1m tot <1)..1ITI >1 ).1lll tot DNA 67 70 63 61 82 Z7 64 64 65 72 65 100 RNA 2 1 21 21 25 11 47 24 26 17 23 17 0 Prot 11 9 15 14 8 25 12 10 18 5 18 0 a values presented are percent of the radiolabel incorporated into the different fractions b unfractionated c passed through a 1 pore size filter, then subjected to acid-base hydrolysis d determined by subtracting incorporation rates for each macromolecular fraction in the< fraction from the unfractionated sample then determining percentage of radiolabel in each group. V1 1\)


53 percentage of the total incorporated thymidine into the< fraction increased. Conversely, the percentage of the total thymidine incorporated into the > fraction decreased The effect of exogenous thymidine concentrations on macromolecular labelling patterns is shown in Table 6. In whole water, the perce ntage of radiolabel incorporated into DNA did not change significantly from 2 5 to 500 nM thymidine. All values ranged between 61 to 72%. The amount of radiolabel incorporated into RNA for whole samples was even more consistent, varying from only 2 1 to 25%. The amount of radiolabel in the protein fraction also varied little, ranging from 5 to 14%. The effect of thymidine concentration on macromolecular labelling in the different size fractions was more complex. In general, size fraction had no influence on labelling patterns when 2 5 or 125 nM exogenous thymidine were added ( Table 6). At 35 nM, however, there was a dramatic drop in the amount of label into RNA in the < 1um fraction and an inrease in the percentage of label incorporated into DNA. Because rates of incorporation into the > 1 ]..1IIl fraction are determined by subtraction of the < 1]..1IIl fraction from t he total unfractioned sample, there were even greater changes in the > 1]..1IIl fraction. In that case, label into DNA dropped to 27% while label into RNA and protein both increased to 47 and 25%, respectively. The results for 500 nM added thymidine indicate little difference between the unfractionated and < lllffi sample. However, the > 1 j.llil fraction had essentially all of its incorporated label in DNA. Correlations of labelling patterns with microbial biomass and activity parameters. In Bayboro Harbor, the percent of radiolabelincorporated into DNA (%DNA) was found to have a significant


54 inverse correlation (R = ...,0.62; (P < 0.001) with total rates of thymidine incorporation. However, no correlation was found between %DNA and cellular rates of incorporation (R = cell density (R or water temperature (R = ...,0.12). The results from samples taken in the Alafia and Crystal Rivers are presented in Table 7. COrrelations were calculated for each individual station as well as grouping the stations by rivers and environmental similarities. Significant trends within each station were either limited or were mirrored in correlations for grouped samples (data not shown) The dominant trend in these data is that the a.rrount of radiolabel incorporated into DNA is inversely correlated with rates of thymidine incorporation parameters (total rates of incorporation, the log transformed rates of incorporation) which agrees with the data from Bayboro Harbor. While this does not hold true for all samples, this relationship prevails more than any other. Primary production also negatively correlated with %DNA in the Alafia River stations. Fluctuations in nutrient co ncentrations presented some interesting correlations with %DNA. The data base for Po4 in the Crystal River was limited because P04 concentrations were often too low to be accurately measured. However, for the samples where Po4 concentrations were available (n = 10), there was a significant correlation (R = 0.80; P < 0.05) between Po4 levels and the percentage of label incorporated into DNA. In the Alafia River and the Medard Reservoir, there are instances where labelling patterns correlated with nitrogen conce ntrations. However, these correlations are inconsistent and again are often limited by the availability of specific nutrient parameters.


Table 7. percent radiolabel into DNA during [ H]thymldlne 1ncorporat1on assays with environmental parameters Parameter All Sta. Cryst R AlaR Medard C.R. Diel Med. Diel n = 64 n = 16 n = 42 n = 6 n = 13 n = 13 Temp .03 .23

The diel experiments conducted at Station 2 of the Crystal River and the Medard Reservoir allowed a more in depth study of these two environments and the results are presented in Figure 9 and Table 7 There was only slight variation in the percentage of label incorporated into DNA for the M3dard Reservoir (F i gure 9A). The rate of incorporation into DNA paralleled the total rate of incorporation. In contrast, the total rate of incorporation, the rate of incorporation into DNA, and the percentage of radiolabel incorporated into DNA all fluctuated greatly at the Crystal River station. In Figure 9 B, it is apparent that there was a strong inverse relationship between total rate of incorporation and percent of label incorporated into DNA. The arrount of label incorporated into DNA decreased as the dark hours progressed, then increased during daylight. 56 Table 7 presents numerical correlations for the data collected at the two diels studies. As indicated by Figure 9, there was little fluctuation in the labelling patterns at the Medard Reservoir during this study. Consequently, little correlation with other parameters occured. There were significant correlations with primary productivity and phosphate levels. However, these may have been coincidental as there was little indication from other data (Table 7) that labelling patterns correlated with either of these two parameters in the Alafia River or the Medard Reservoir. There was no correlation with N02 levels as was found in the Medard samples taken during t h e seasonal studies. Again, this indicates a coincidental relationship between label incorporated into DNA and nutrient levels for the Medard Reservoir. In contrast, at the Crystal River diel sampling, there were


A 12 24 12 Time of day 100 24 12 Figure 9. Diurnal pattern of [3H]thymidine incorporation and percent of radioactivity incorporated into DNA for the Medard Reservoir and Crystal River Statif,n 2 Solid lines represent total incorporation of [ H]thymidine and dashed lines represent percent of radioactivity incorporated into DNA. 57


several parameters which significantly correlated with the percent of label incorporated into DNA. The mst notable were thymidine incorporation rates and primary productivity (Table 7). However, only the thymidine incorporation rates were consistent with other data (Table 7). The primary productivity value was significant (P < 0 .05 ) and positive whereas the value for the Medard diel correlation was negative. Unfortunately, Po4 levels were again not measurable for these samples so no further examination of the correlation of P04 with labelling patterns reported for the rest o f the river (Table 7) could be made. The significant correlation with temperature is a further reflection of the diurnal nature of the labelling pattern for these samples. The water temperature varied three degrees during the sampling period. 58 Effect of FdU on non-specific Labelling of Macromolecules. Figure 6 and Table 8 demonstrate the ability of 12.5 nM FdU to completely inhibit labelling of macromlecules during [ 3 H]thymidine incorporation at low levels (2 .5 nM) of exogenous thymidine in Bayboro Harbor (Figure 6a,b). However, in the presence of 35 nM C3H]thymidine, 12.5 nM FdU had little inhibitory effect on non.:.:specific labelling (data not shown). Increasing the FdU concentration in proportion to the higher [ 3H]thymidine concentration had only a slight inhibitory effect on labelling (Figure 10), but in no instance as great as the effect of 2.5 nM thymidine and 12.5 nM FdU. Table 8 summarizes the effect FdU had on labelling patterns in a variety of environments. Results similar to those found for Bayboro


400 A B c. 0 (J c:::: -a: 200 0 400 Q) 0 E c. 200 Figure 10. c D 35 nM TdR 75 nM FdU DNA MIN 90 35 nM 35nM DNA 35 nM TdR 175 nM FdU DNA ,,,,, j!/;'/////// MIN 90 Effect of FdU on macromolecular labelling at higher co ncentrations o f [ H]thymidine in Bayboro Harbor 59


Harbor were obtained when a coral surface microlayer from the Dry Tortugas was investigated (Table 8). At low [ 3H]thymidine concentrations, 12.5 nM FdU effectively inhibited non.:..specific labelling. In an offshore site in the Gulf of Mexico, there was an increase in the percent of non-specific labelling as well as a decrease in the rate of C 3H]thymidine incorporation into DNA in the presenc of FdU (Table 8). Table 8 Effect of FdU on [ 3 H]thymidine incorporation rates into DNA and percent radiolabel into DNA in various environments Control + 12.5 nM FdU Location pm::>les/L hr %DNA pm::>les/L hr %DNA Bayboro 194.3 48 213. 8 93 Bayboro 182. 1 48 147.8 98 Dry Tort CSM 2 9 34 8 1 81 Offshore 0.8 55 0 5 36 Alafia St. 1 69 80 49. 9 80 Alafia St. 8 107. 7 46 103.1 84 Ccyst Riv St. 23. 5 59 22.0 77 Medard winter 44.3 71 25. 9 86 Medard summer 181 9 46 234. 0 91 60 At the ups tream Station 1 of the Alafia River, there was a minimal amount of non.:..specific labell ing of macromolecules (Table 8) and FdU caused a noticeable decrease in the rate of C 3H]t h ymidine incorporation into DNA similar to that observed in oligotrophic offshore waters and a sample collected from the Medard Reservoir in winter (Table 8). However, at the more eutrophic estuarine Alafia River Station 8, FdU


again caused a significant inhibition of non-specific labelling of macromolecules (Table 8). There was no discernible difference in rates of [ 3 H]thymidine incorporation into DNA betwen the two treatments at Station 8 and in the Crystal River sample, most likely owing to the absence of intracellular isotope dilution in these samples. To determine the mechanism by which FdU inhibited non...:.specific labelling, other metabolic inhibitors were examined. In laboratory cultures amethopterin and trirnethoprim have been shown to inhibit dihydrofolate reductase (DHFR; M:>llgaard and Newhard, 1983). This reaction regenerates the tetrahydrofolate which donates the methyl 61 group to dUMP during the thymidylate synthetase reaction (Mollgaard and Newhard, 1983) It is assurned that environmental bacteria possess similar pathways. Inhibition of DHFR would yield similar results as obtained with FdU should reversal of thymidylate synthetase be the cause of non-specific labelling. However, these two inhibitors had little effect on labelling even at concentrations which began to reduce total cellular activity (Figure 11) Comparison of non...:.specific labelling during [methyl...:.3 H]thyrnidine and incorporation. The effect of FdU on non:specific macromolecular labelling by and [6...:.3 H]thymidine appears in Figure 12. [6...:.3H]Thymidine incorporation resulted in non:specific labelling of macromolecules (Figure 12C), although to a lesser extent than did [methyl...:.3 H]thymidine. FdU at 12.5 nM completely inhibited labelling when either form of [ 3 H]thymidine was employed (Figure 1 28, D) There was no difference in calculated rates of thymidine incorporation into DNA between the two forms of [3H]thymidine. Higher rates of [ 3 H]thymidine incorporation into DNA


A 4oo Control .,.. 200 I ...1 ci. ... 0 (.) c tJ) .! 400 0 E a. 200 c Trimeth. MIN 90 B A met h D Ameth + Trimeth. MIN 90 11. Effect of and on labelling patterns in In each tn=atment, the were added at 100 62


... 0 (,) c: 0 Q) -0 E Q. A 300 300 / / 1 //I Pr: I 90 MIN B [Methyi-3HJ TdR FdU DNA DNA MIN 90 Figure 12 Comparison of Bayboro Harbor by using H]thymidine and [6-H]thymidine. [methyl-H]thymidine in thEJ a b sence (A) and presence (B) of 12. 5 nM FdU and [6-H]thymidine in the absence (C) and presence (D) of 12. 5 nM FdU shown. 63


in the presence of FdU (Figure 12B,D) are most probably due to inhibition of intracellular isotope dilution by FdU. Discuss ion The methodology used throughout this study was the acid-base hydrolysis protocol of Riemann and Sondergaard (1984). Recent evidence suggests that this may not be an ideal means by which to separate nacromolecules labelled during [ 3 H]thymidine incorporation. Various other methods have been proposed for the separation of labelled macromolecules including specific enzymatic digestion (Robarts et al., 64 1986; Servais et al., 1987), the use of hydroxylapatite columns (Witzel and Graf, 1984) and dialysis (Pollard, 1987). Robarts et al. (1986) reported a 64% recovery of c32P]DNA using hydrolysis and in this study, 60 to 100% of added [ 3H]DNA was recovered depending on its source (see below). t-bre interesting, several investigators have reported that the fraction most commonly considered to be RNA is more likely to contain radiolabelled lipids (Robarts et al., 1986; Carman et al., in press). RNase digestion determined that only a minor portio n (-7%) of the incorporated label was in RNA (Riemann 1984). This may limit the amount of information which may be gathered about the other molecules other than DNA which may be labelled. However, this study has focused on the amount of label in DNA and not o n whic h other molecules are labelled. Recent investigations ( Robarts et al., 1986; Servais et al., 1987; Hollibaugh, 1988) have reported temporal and spatial fluctuations in macromolecular labelling patterns during [ 3 H]thymidin e incorporation assays in several environments. The results presented here are


consistent with those findings. The average amount of radiolabel incorporated into DNA (%DNA) varied widely between locations although the averages of all locations were quite close (Table 4) and the average for all samples was 51.6 14.4 percent with a range of 6.1 to 95 .5%. The data in Table 4 have not been corrected for percent 65 recovery of radiolabelled DNA since recovery varied depending on how it was determined. Using V. proteolyticus cells radiolabelled with [ 3H]thymidine, all of the radioactivity was recovered as DNA. In contrast, subjecting [ 3H]>.DNA to acid.::.base hydrolysis resulted in 60-85% recovery of DNA. Since recovery varied and it was not determined for all samples, no universal correction factor was applied to the data. The linearity of labelling patterns reported for Bayboro Harbor agreed with reports from South San Francisco Bay (Hollibaugh, 1988). Robarts et al. ( 1986), however, reported inconsistent labelling patterns over time for a hypertrophic lake. M:>riarty (1986) stated that in short term incubations, the vast majority of radiolabel will be incorporated into DNA. However, the results presented here and in South San Francisco Bay (Hollibaugh, 1988) indicate that at least in eutrophic estuaries, non...:.specific labelling begins immediately and continues at a constant rate. During the first year, the seasonal pattern of macronoleular labelling in Bayboro Harbor (Figure 8) was very similar to that reported by Robarts et al. (1986). In a hypertrophic lake, they reported a peak in non-specific labelling in the summer and a minimum amount in the winter (Robarts et al., 1986). In Bayboro Harbor, the greatest percentages of radiolabel incorporated into DNA occurred in


66 the winter ITDnths with minima in the sumner. The absence of any consistent pattern in radiolabelling during the second year may have been related to differences in weather patterns between the two years. There was below average rainfall for St. PetersbUr:'g in 1986 while the sumner of 1987 was marked by frequent severe rainstorms. Bayboro Harbor receives major inputs from rainwater runoff. Consequently, during 1986 when little rain fell, the microbial environment was perhaps allowed to proceed normally between seasons. During the second year When frequent rains contributed large amounts of r:'Unoff to the harbor, these periodic disturbances prevented normal seasonal succession in the microbial environment. The effect of exogenous thymidine concentrations indicated that labelling patterns in Bayboro Harbor were essentially independent of thymidine concentration in the size fractions examined (Table 6). Since variations were small and displayed no consistent pattern, the differences noted between treatments may have been in part due to the artifacts created by filtration to separate size fractions. Bacteria often respond erratically to filtration during thymidine incorporation as indicated by occasionally finding the rates in the < fraction to be greater than those in the unfiltered sample (unpublished obser'Vations). Removal of grazers cannot be the cause of these results since filtration occurs at the end of the incubation with [3H]thymidine. The relatively consistent rates of incorporation and macroiTOlecular labelling patterns indicated that even at very high concentrations of thymidine, bacteria are the organisms primarily responsible for incorporation. Other seemingly significant variations between size fractions were in part due to the subtraction method used


to determine the> data. Additionally, the specific activity of the 500 nM sample was low enough that minor changes in incorporated radioactivity resulted in significant differences in thymidine incorporation rates. It might be expected that at high concentrations, microalgae may begin to incorporate significant amounts of thymidine (Rivkin, 1986; Rivkin and Voytek, 1986). However, the increase of incorporation by the < 1 lJIIl fraction (primarily bacteria) at higher thymidine concentrations do not support that hypothesis. 67 Exogenous thymidine concentrations had no effect on macromolecular labelling patterns in unfiltered water in Bayboro Harbor. However, this effect was not consistent with previous reports. Karl (1982) reported increasing incorporation into DNA with increasing thymidine concentrations from 1 .25 to 125 nM. However, a later experiment provided opposite results with decreasing incorporation into DNA with increasing thymi dine concentrations. Hollibaugh ( 1988) reported a decrease in radiolabel incorporated into DNA with increasing thymidine concentrations for South San Francisco Bay waters. Differences between results reported from these studies may in fact be attributable to differences in environment Just as percent of radiolabel incorporated into DNA varies spatially, so too might effects of exogenous thymidine concentrations. In previous investigations, the percentage of label incorporated into DNA has been reported to negatively correlate with total thymidine incorporation rates and primary productivity and positively correlate with oxygen concentrations (Robarts et al. 1986). Lovell and Konopka ( 1985a) reported only 28% of the radio label in DNA in anaerobic hypolirnnetic samples and 80% in epilirnnetic and metalimnetic samples.


Hanson and Lowry (1983) reported greater non -specific labelling with samples taken at 500 to 2000 m depths. The significant negative correlations of percent labelled DNA with total rates of thymidine incorporation for Bayboro Harbor and other samples (Table 7) is consistent with results reported by Robarts et al. (1986) In essence, the faster bacteria cells grow, the greater the amount of non-specific labelling. Gr-eater non..:.specific labelling in productive waters is supported by the negative correlation of percent labelled DNA with primary production in the Alafia River and Medard samples. The exception to this trend is the positive correlation reported for the diel study at Station 2 of the Crystal River No explanation can be made for this anorraly. It has been suggested that under organic nutrient limiting conditions, thymidine is more likely to be catabolized to a carbon and energy source than to be incorporated as a precurser into DNA (Fallon and Newell, 1986; Servais et al., 1987; Carman et al., in press). There was no correlation between %DNA and DOC. However, the technique used cannot distinguish between usable carbon and refractory forms of carbon. When examining the inorganic nutrient data, the Po4 concentrations in the Crystal River were often too low to measure and it is possible that these waters were phosphate limited. However, where data was available, the most significant correlation was that between %DNA and P04 indicating that the more available Po4 the more label was incorporated into DNA implying that under nutrient stressed conditions, catabolic enzymes were more likely to degrade thymidine. The data presented in Table 7 and Figure 9 indicated that little diurnal fluctuation in %DNA occurred during the diel study at the 68


69 Medard Reservoir. However, the Crysta l River sampling demonstrated wide fluctuations in most of the parameters measured and very strong correlations between %DNA and many parameters Thymidine incorporation rates fluctuated on a diurnal pattern most similar to those recorded in Bayboro Harbor during the wet season (Paul et al., 1988) with maximum rates at night and minimum rates in the afternoon. It is possible that measured fluctuations were related to tidal flow. The increases in %DNA c losely followed the observed lower t ides such that the diurnal variation may be due to changes in population between the up river freshwaters and those waters brought in from the Gulf of Mexico with the incomi ng tide. If this were true, however, it might be expected that %DNA would decrease from station 1 over the springs at Kings Bay down the river to Station 4 located at the mouth of the river. However, examination of the data from these stations indicates that this was not the case as no consistent relationship was reco rded between stations. The effect of terrestrial input attibutable to building developments along the shore cannot be easily ascertained and its possible that fluctuations in measured parameters were due to changes in these inputs carried by tidal flow Perhaps the best independent indicator of tidal influence is salinity. During the six seasonal samplings of the Crystal River, average salinity at Station 2 was 1.6 0.9 ppt indicating that the high flow rate downriver caused by the freshwater spring sources significantly dilutes any input from the Gulf of Mexico which tidal flow might bring. Since salinity is reduced approximately 15-fold from Gulf of Mexico water, it seems unlikely that the large differences in measured parameters would be due to diluted input from the Gulf of Mexico. Unfortunately, there are no


70 other data other than that from the Medard Reservoir with which to compare the Crystal River diel data. Many investigators have conducted diel studies using [3H]thymidine incorporation (Riemann et al., 1982; Moriarty and Pollard, 1982; Paul et al., 1988) but the focus of these reports was not how labelling patterns changed Therefore, if they did collect these data, it was not reported. Overall, non-specific labelling seems to most often correlate with bacterial activity. Faster growing cells incorporate less of the radiolabel into DNA. In some instances, %DNA correlated with nutrient levels and primary productivity. The inconsistency with which these parameters correlate with each other, however, indicates that labelling patterns are dependent on no single factor but perhaps a combination of several parameters. It is possible that labelling patterns in bacteria are less influenced by the environment than they are a function of the particular bacterial population Which inhabits that environment. It is essential, therefore, that when estimates of DNA synthesis or heterotrophic productivity are made from thymidine incorporation values, determination of specific radioactivity in DNA must be determined. The studies using specific inhibitors yielded much more conclusive information concerning the pathways involved with non...:.specific mac rom lecular labelling. Since FdU is a known inhibitor of thymidylate synthetase, this enzyDE may be the active site for the inhibition of non.::.specific labelling. It has been proposed that demethylation of [3H]dTMP by a reversal of thymidylate synthase would produce radio labelled 5, lO...:.methylene tetrahydrofolate which could donate tritiated methyl groups to several reactions in RNA and protein


71 synthesis (Hollibaugh, 1988). While this reaction has been shown to be reversable using halogenated analogs, it has been shown to be irreversible with regards to d'IMP demethylation under laboratory conditions (Commerford and Joel, 1979). Additionally, the inability of the dihydrofolate reductase inhibitors, amethopterin and trimethoprim, to reduce non-specific labelling (Figure 11) suggests that thymidylate synthase is not involved in the non.::specific labelling process. Radiolabelling of RNA and protein fractions observed during [6 -3H]thymidine incorporation (Figure 12C) indicates that thymidylate synthase is not involved since the label is not associated with the methyl group of thymidine. Since FdU completely inhibits labelling of [6-3 H]thymidine under the same conditions as it does for [methyl.::3H]thymidine, the inhibition by FdU of non-specific labelling must be due to enzymatic reactions not associated with the methyl grou p of thymidine. A more likely pathway for non.::specifi c labelling from [3H]thymidine incorporation involves the conversion of thymidine to thymine via the thymidine phosphorylase (EC 2 4 .2.4) reaction. The further degradation of thymine to uracil is not a single dernethylation reaction but rather stepwise dehydrogenation and decarboxylation reactions by thymine 7.::hydroxylase (EC and uracil.::5.::carboxylic acid decarboxylase (Newhard, 1983). These reactions could account for tritium from the methyl group entering into a series of catabolic or anabolic pathways, and may account for the production of 3H2o observed by others (Karl 1982; Hollibaugh, 1988) Razzell and casshyap (196 4) reported on the ability of FdU to inhibit thymidine phosphorylase in Escherichia coli. Attempts to block


72 thymidine phosphorylase with other known inhibitors, uridine and deoxyadenosine (Budman and Pardee, 1967) were inconclusive. Both of these substrates significantly reduced total thymidine incorporation (data not shown) rrost likely due to competition for transport or incorporation enzymes. It seems likely that this first reaction in the catabolism of thymidine may be the target for FdU in estuarine bacteria where non-specific labelling was inhibited. This would prevent the further degradation of thymidine and the subsequent release of tritium. At first glance, the effect of FdU on non-specific labelling appears to be sampling site dependent. However, the ability of FdU to inhibit labelling was significantly correlated with total thymidine incorporation per litre per hour (R=0.84; P

(1986), may be related to production rates. In oligotrophic fresh and marine waters, FdU caused a decrease in the rate of [ 3 H]thymidine incorporation into DNA. In these waters where isotope dilution was negligible (Chapter 2) and heterotrophic productivity was low, FdU may have a different effect on bacterial DNA synthesis. One of the reasons fluorinated pyrimidines are comm:mly used as anti-cancer agents is that they are incorporated into nucleic acids. Fluorinated dUTP may be incorporated into DNA which causes nicks in the DNA strands (Martinet al., 1980). In those environments in which thymidylate synthetase activity is low and therefore does not bind FdUMP, the inhibitor may have a greater likelihood of being incorporated into DNA. In conclusion, macromoleular labelling is ubiquitous 73 during thymidine incorporation assays with environmental samples The results presented here indicate that determination of specific radioactivity incorporated into DNA is essential when [ 3 H]thymidin e incorporation is used to estimate DNA synthesis or heterotrophic activity. FdU may completely inhibit non.:.specific labelling at low [3H]thymidine concentrations. Caution must be exercised, however, to determine that extracellular isotope dilution is insignificant when low exogenous [ 3 H]thymidine concentrations are added. Results have also indicated that non.:.specific labelling is not caused by a reversal of thymidylate synthetase activity nor some other uncharacterized thymidine demethylation reaction.


CHAPTER 4: DIRECT Ca1PARISON OF THYMIDINE INCORPORATION WITH DNA SYNTHESIS Intmduction The third factor which effects the accuracy of [ 3 H]thymidine incorporation is the empirical relationship between thymidine incorporation and DNA synthesis (Kar l and Bossard, 1986) There has not yet been an absolute standard non-isotopic determination of bacterial DNA synthesis under laboratory and environmental conditions with which results from C3 H]thymidine incorporation could be compared. The relationship between [3H]thymidine incorporation and DNA synthesis has always been assumed to be direct. However, the exact nature of this relationship has never been examined. This chapter presents the first comparison of bacterial DNA synthesis as estimated by [3H]thymidine incorporation and as measured directly using a non-isotopic means. Methods and Materials Culture Studies. The culture organism used was Vibrio proteolyticus which has been previously shown to incorporate [3H]thymidine into DNA (Jeffrey and Paul, l986a). Overnight cultures were diluted 1:50 into 300 ml of fresh growth media (ASWJP+PY; Paul, 1982) and grown at room temperature for 3 hr on a gyratory shaker at 100 rpm. Six 25 ml samples were removed to sterile 125 m l polycarbonate flasks for isotope dilution analysis (Pollard and 74


75 Moriarty, 1984). To all flasks, (ICN Pharmaceuticals, Inc.) was added at 2.0 Unlabelled thymidine was added such that total thymidine concentrations ranged from 1 to 30 To the remaining cell solution, thymidine was added at 2.0 and a final concentration of 10 Subsamples were taken immediately and at 20 min intervals up to a total of 2 hr for measurements of thymidine incorporation, direct counts, and cellular DNA. Triplicate 2.0 ml samples for thymidine incorporation were added to test tubes containing 0.2 ml 100% trichloroacetic acid (TCA), mixed, and placed on ice. After a minimum of 1 hr, the contents were filtered onto 0.2 pore size polycarbonate filters (Nuclepore Corp., Pleasanton, Calif) and each washed with 5 ml ice cold 5% TCA. Radioactivity was determined by liquid scintillation counting. Acid/base hydrolysis of macrorolecules was not required to determine non.:..specific labelling since this organism incorporates [ 3 H]thymidine only into DNA (Jeffrey and Paul, 1986a; data not shown). Subsamples for direct counts were fixed and diluted in 0.2 filtered ASWJP containing 1.8% formalin. Direct counts were later determined by epifluorescence microscopy (Porter and Feig, 1980). Cellular DNA was determined fluorometrically using Hoechst dye 33258 by the method of Paul and Meyers (1982). This technique involves extracting DNA by sonication, removing cell debris and filter particles by centrifugation, and reacting the clarified extract with Hoechst 33258. DNA concentration was determined against known standards. Field Studies. For calibration of thymidine incorporation with DNA synthesis in natural waters, samples were taken from Bayboro Harbor and oligotrophic offshore waters of the southeast Gulf of Mexico


Water samples were gently filtered (<0.2 atm vacuum) through a 1 pore size Nuclepore filter to remove the majority of bacterivores (Anderse n and Fenchel, 1985). The sample was then incubated at room temperature for 2-4 hr on a gyrotatory shaker at 100 rpm. Isotope dilution (Pollard and Moriarty, 1984) was then determined in a series 76 of 25 ml subsamples with thymidine added at a range of concentrations from 2.5-30 nM. To the remaining sample, [ 3 H]thymid ine was added at 35 nM and subsamples immediately taken fo r thymidine incorporation, direct counts, and cellular DNA. Replicate subsamples were subjected to acid/base hydrolysis as described previously (Riemann and Sondergaard 1984) to determine the amount o f non-specific macromolecular labelling during [ 3 H]thymidine incorporation. Sampling intervals were extended to 2 h (Bayboro Harbor) or 4 hr (offshore waters) to ensure significant changes in the measured parameters. Subsamples for DNA were filtered onto 0.22 pore size membrane filters (GS type, Millipore Corp., Bedford, MA) and stored frozen in 3 ml sec (0 .15 M NaCl, 0 .015 M sodium acetate; Paul et al., 1986) for later analysis (Paul and 1982). Direct counts of subsamples were fixed in 1.8% formalin for later epifluorescence microscopy determinations (Porter and Feig, 1980). The efficiency of acid/base hydrolysis of DNA was determined by either subjecting [ 3 H]thymidine labelled proteolyticus cells (which label only DNA) to acid-base hydrolysis or adding 0.1 J-1Ci/ml high specific activity [3H]end-labelled ADNA (Maniatis et al., 1982) to water samples and immediately subjecting 2 ml subsarnples to acid/ base hydrolysis (Riemann and Sondergaard 1984). Total DNA was determined


by adding a 2 ml sample to 0.2 ml 100% TCA, chilling, and collecting the precipitate by filtration. This value was taken as 100% of the added radioactivity. Percent recovery was calculated after determining the amount of radioactivity lost during warm base hydrolysis and the amount of radioactivity remaining after hot acid extraction. Results 77 The results from the DNA calibration experiment using V. proteolyticus are presented graphically in Figure 13 and numerically in Table 9 Figure 13 demonstrates the exponential i ncrease in each of the measured parameters during the duration of the experiment. In Table 9, the DNA values are the amount of DNA increase during the twenty min time interval. The calculated DNA va l ue is determined by using 624 g/mol as an average molecular weight base pair (Le win, 1987) and a 50% A-T base pair composition. The value is corrected for isotope dilution by determining the amount of total dilution, Which in the case of the exponentially growing V. proteolyticus in ASWJP+PY, was equivalent to 5 6 ).IM. Thymidine incorporation rates were then calculated using 15.6 ).1M as the total thymidine concentrations. The corrected moles of thymidine incorporated was then multiplied by 1248 g of DNA per mole (2X base pair mol. wt) to get estimated DNA synthesis. The ratios of the measured to the calculated DNA values indicated a significant underestimation of DNA synthesis as determined by thymidine incorporation during the two hour duration of the experiment The values range from 17. 1 to 4. 6 fold and average an eight fold underestimation of DNA synthesis as determine d by thymidine incorporation (Table 9).


78 0 """" 10!. 3 40 80 120 Minutes F i gu r e 1 3 Direct co mparison of thy midine incorporation, DNA synthesis, and cells p r oduced fo r proteolyticus.


Table 9. Comparison of thymidine incorporation with DNA synthesis for V. proteolyticus 6DNA 79 Time interval Measuredb Calculatedc Measured/d 10 17 (min) (ng/ml) (ng/ml) Calculated TdR Incorp. 0-20 650 38 17. 1 2.24 20-40 294 46 6.4 2;83 40-60 218 45 4.8 4.29 60-80 368 79 4;6 2;23 80-100 702 76 9.2 5.67 100-'-120 482 80 6;0 5; 10 a Average of all values 3.73 + 1 5 X 1017 b Determined by direct fluorometric measurements of DNA c Calculated from thymidine incorporation by multiplying moles thymidine incorporated by 1248 g DNA per mole. Corrected for isotope dilution and assumes 50% A-T base pair composition d Average of all values = 8 .0 + 4.7. Excluding 0-20 min outlier average is equal to 6 2 1:-8 Similar results were o bserved during experiments comparing DNA synthesis with [3H]thymidine incorporation using environmental samples (Table 10). In Bayboro Harbor waters, [ 3 HJthymidine incorporation underestimated DNA synthesis by an average factor of 5.8. In offshore waters, DNA synthesis was underestimated by approximately 8 fold (Table 10) The values were corrected for isotope dilution (less than or equivalent to 3 nM thymidine for these samples), non-specific macromolecular l abelling, and effic iency of acid/base hydrolysis. Determined extraction efficiency values have ranged from 60-100% but the majority were between 80 to 85%. Therefore, a conservative


80 extraction efficiency of 80% was used. Table 10. Comparison of thymidine incorporation with DNA synthesis for environmental bacteria 6DNA DNA I cell Measa ca l c t1:!as/b t1:!as ca lc. 1018 cells/mol (ng/ml) ( ng/ml) Calc (fg) ( fg ) TdR Incorp. BBH 8 .00 1.35 5.9 14.0 0.51 0.6 1:62 5;8 8;7 0;8 0:9 Offshore 0.765 0.096 8.0 5.3 0 .64 4 3 a Determined during a 2 hr time interval for Bayboro Harbor and 4 hr interval for offshore waters b Average value= 6.6 + 1.2. Values corrected for isotope dilution, non-specific macromolecular labelling, and efficiency of acid/base hydrolysis Discussion The data presented here indicate that thymidine incorporation significantly and consistently underestimates DNA synthesis in three widely different aquatic environments. The amount by which [3H]thymidine incorporation underesti mates DNA synthesis is not significantly different regard less of whether an exponentially growing culture (Table 9), a eutrophic estuarine sample (Bayboro Harbor, Tabl e 10) or oligotrophic offshore waters (Table 10) are examined. Further evidence of the underestimations is indicated by values for DNA content per cell (Table 10). The measured DNA/cell values presented here are s lightly larger than those reported by others (Fuhrman and Azam, 1982; Paul and Carlson, 1984; Paul et al. 1985), but they may be due to the


higher growth rate of these cells caused by stimulation from filtration and incubation. Actively growing cells will have more than one copy of the genome since DNA is synthesized before cell division (Freifelder, 1983) In contrast, the DNA per cell values derived from [3H]thymidine incorporation and cell counts were extremely low ( Table 10), approximately seven to nine times lower than published values (Fuhrman and Azam, 1982; Paul and Carlson, 1984; Paul et al., 1985) 81 It is unlikely that the assumptions and corrections used to calculate DNA synthesis are the reasons for the large discrepancy between measured values and those derived from [ 3H]thymidine incorporation. Using an average molecular weight of a base pair of 624 g/mol (Lewin, 1987) is appropriate on the basis of DNA structure and is commonly used to estimate DNA content in a variety of procedures (Lewin, 1987). The percent efficiency of DNA hydrolysis is conservative. Using 100% efficiency would actually increase the ratio of measured to predicted DNA. Assuming 50% A-T base pair composition also seems logical for a heterogeneous population. Any variations would be slight and certainly could not account for the differences in measured vs. calculated values. The data presented here are independent of the percentage of active cells in each sample. The increase in DNA pe r volume of liquid over time was measured which was presumably attributable to the active cells, the same active cells which were incorporating [ 3 H]thymidine Inactive cells were not synthesizing DNA and did not contribute to or detract from the results. It is assumed that the vast majority of active bacteria in these samples were incorporating thymidine The limited available information (Fuhrman and Azam, 1982; see Chapter 5)


82 appears to support this assumption. Values for cells synthesized per mole of thymidine incorporated are presented (Table 10) as this has become the most common means by which thymidine incorporation is correlated with cell growth (Bell, 1986; Riemann et al., 1987; Coveney and Wetzel, 1988). As is apparent by the data in Tables 9 and 10, there is a wide range in values. The values for V. proteolyticus and Bayboro Harbor are quite similar, yet much lower than the value for the offshore waters. The reason for this is not immediately clear, although the faster growth rates and larger cellular DNA contents for these samples contribute to these results. Greater numbers of cells synthesized per mole thymidine incorporated for the offshore sample indicates a slower growth rate and lower DNA content per cell. However, all these values fall within the range reported by other researchers (Bell, 1986; Riemann et al., 1987; Covenay and Wetzel, 1988). The large variance in values presented here indicates that this type of conversion factor for thymidine incorporation to productivity may not be consistent in all environments or growth rates. In contrast, the ratio of measured to calculated DNA synthesized appears constant. Presumably, thymidine bases incorporated into DNA may arise from two biosynthetic pathways. The salvage pathway incorporates exogenous thymidine to dTMP, which is twice more phosphorylated to dTTP before incorporation into DNA (Kornberg, 1980). The de novo synthesis pathway uses intracellular constituents to eventually synthesize dTMP from dUMP via the thymidylate synthetase reaction (Kornburg, 1980). By determining the amount of label incorporated into DNA and isotope dilution, all of the thymine bases should be accounted for. The


dramatic and consistent underestimation of DNA synthesis by [3H]thymidine incorporation in three widely different environments indicates that there are other significant sources of thymine bases incorporated into DNA for which there c an be no accounting. The apparent explanat ion for these results is that isotope dilution analysis is not appropriate for use with in situ [3H]thymidine incorporation assays. The isotope dilution method (t-briarty and Pollard, 1981; 1982; Pollard and Moriarty, 1984) is based on a technique originally designed for use with mammalian cell cultures (Sjostrom and F orsdyke, 1973 ; Forsdyke, 1968). Several studies have examined its use in environmental samples and the method appears to 83 function much as it does when applied to mammalian tissue cells. These previou s studies have demonstrated that isotope dilution analysis accounts for some intra and extracellular dilution of the isotope, although the amount for environmental samples was rarely more than the equivalent of a few nM thymidine (Chapter 2). In previous studies, there was not means to determine whether all dilution was measured. By directly measuring the amount of DNA synthesized in these samples using the Ibechst 33258 technique (Paul and Meyers, 1982) the number of thymine bases incorporated into DNA was calculated. Results consistently indicated substantial sources of thymine bases that could not be accounted for by isotope dilution analysis. These results agree very closely with those of Fuhrman and Azarn (1982) who elected to determine dilution by comparing [ 3 H]thymidine incorporation with labelled phosphorous incorporation into DNA. They reported a 3 -7 fold underestimation of DNA synthesis on the basis of [ 3 H]thymidine incorporation compared with phosphorus incorporation.


84 Although C3 H]thymidine incorporation has its limitations, it is still the simplest and most rapid means by Which bacterial activity can be examined in most environments. caution is suggested when using this technique as to how the results obtained may be interpreted and applied. Bacterial activity, DNA synthesis, and heterotrophic production are three different processes. Although [3H]thymidine incorporation may be used to estimate all three of these processes, the assumptions required to use [ 3 H]thymidine for each are different. It is the responsibility of the researcher to show that the thymidine method is valid for a particular application. Relative bacterial activity between similar samples may easily be compared using gross [3H]thymidine incorporation values. In this case, non-specific labelling and isotope dilution becomes less important. However, when attempts are made to extrapolate [ 3 H]thymidine incorporation to DNA synthesis values, labelling, isotope dilution, and other factors become very important. The data presented here and that of Fuhrman and Azam (1982) indicate that estimates of DNA synthesis may be obtained by multiplying [ 3H]thymidine predicted values by 6-8. Heterotrophic productivity values may be derived from empirical or measured values for cells produced per mole of thymidine incorporated without first going through DNA synthesis estimates. However, the increasing body of literature indicates that no single conversion factor holds for all environments at all times. Most recent evidence indicates that cell volume may be more closely related to [ 3 H]thymidine incorporation derived productivity estimates than to cell numbers (Coveney and Wetzel, 1988).

PAGE 100

CHAPTER 5: THYMIDINE KINASE ACTIVITY, THYMIDINE TRANSPORT AND INCORPORATION INTO DNA Introduction 85 cne of the essential assumptions for the validity of the [ 3H]thymidine incorporation method to estimate bacterial productivity is that all of the bacteria in the sample are capable of incorporating thymidine into DNA. For exogenous thymidine to be incorporated into DNA, it must first be transported into the cell then phosphorylated. Thymidine kinase is the enzyme responsible for the initial phosphorylation of thymidine to dTMP and is therefore believed to be essential for thymidine incorporation. Previous investigations have assumed the ubiquity of thymidine incorporation but there has never been an in depth study of thymidine kinase activity in marine bacteria. In this part of the study, a collection of marine isolates were examined for their ability to incorporate thymidine and attempts were made to determine the presence of thymidine kinase in these organisms by molecular probing for the thymidine kinase gene and an enzyme assay for thymidine kinase activity. Methods and Materials Thymidine incorporation in marine isolates: Bacterial cultures were isolated from surface water samples of Bayboro Harbor and from a coral surface microlayer of the Dry Tortugas. The majority of bacteria were isolated on ASWJP+PY plates (Paul, 1982) A few were isolated on

PAGE 101

86 ASWJP agar plates containing 10 mM glucose or 10 mM thymidine as sole carbon sources. All isolates, however, were rraintained on ASWJP+PY as well as the initial isolating medium. In addition to the bacteria isolated specifically for this study, nine biofouling organisms isolated from the Chesapeake Bay (Paul and Loeb, 1983) were examined. Vibrio alginolyticus was the gift of Dr. Andrew Gordon of Old Dominion University. Pseudomonas altantica was donated by Dr. Chris Low of the University of Tennessee and Vibrio parahemolyticus was given by Dr. Gregory Stewart of the University of South Florida. A list of othe r isolates and plasmids used in this study is presented in Table 11. Table 11. Bacterial strains and plasmids. Strain E. coli HB101 E. coli HB101 P stutzeri JM300 E. coli C600 JL4000 E. coli HB101 JL4062 P. aeruginosa PA01 E coli C600 E coli JM1 01 Plasmid pSBTI<2.0 pSB-NI wild type pKT230 pell3 pR02317 wild type wild type Pb,C c c P,C c c Antibiotic Source AMP AMP kan kan AMP P Berg, M. Diekmann Stanford University, Stanford, ca. G. Stewart, University South Florida, Tampa G. Stewart G. Stewart G. Stewart M. DeFlaun, Tufts Univ Boston, Ma. Bethesda Research labs Gaithersburg, Md a Thymidine kinase genes on chromosome (C) or plasmid ( P ) b Herpes Simplex Virus thymidine kinase gene

PAGE 102

87 To detennine whether these organisms could incorporate thymidine, each was inoculated from an agar plate into 25 ml ASWJP+PY and grown over night at room temperature on a gyrotatory shaker at 140 rpm. The cells were then diluted 1:25 in fresh growth media and grown until A600 = 0 .25. They were diluted 1:2 into fresh growth media (approximate cell density = 5 x 107 cells/ml) and thymidine added at 1 and a final concentration of 1 Radioactivity incorporated into the cold TCA precipitable fraction was determined after 1 hr. Thymidine uptake vs. incorporation. A series of organisms were examined for the ability to take up (transport) thymidine and then the ability to incorporate thymidine into cold TCA insoluble material. Of the organisms examined, three were known to be capable o f thymidine incorporation and four were known to be unable to incorporate thymidine. In addition, v. parahaemolyticus which had been transformed with the pOQ3 plasmid (see Chapter 6) was also examined Overnight cultures were diluted 1:10 into fresh growth media and allowed to grow for 90 min and subsampled for bacterial direct counts (Porter and Feig, 1980). [3H]Thymidine was added to the remaining cell solution at a final concentration of 100 nM. Samples were taken immediately and again after 30 min of Uptake was detennined by adding subsamples to tubes containing 1 ml of 5 mM unlabelled thymidine in ASWJP. Samples were then filtered onto 0.2 Nuclepore filters and washed with 5 ml of the 5 mM tnlabelled thymidine solution. Incorporation was determined by adding replicate subsamples to tubes containing 1 ml ice cold 10%TCA. After 30 min on ice, the samples were filtered onto 0.2 pore size Nuclepore filters and the filters were washed with 5 ml ice cold 5% TCA. Formalin killed controls were

PAGE 103

examined to determine abiotic adsorption of radioactivity to the filters. Radioactivity in all samples was determined by liquid scintillation counting and rates were determined by linear regression analysis after correcting for formalin controls. 88 Thymidine kinase enzyme assays. Nine isolates were examined for thymidine kinase activity in cell free extracts using the protoco l outlined by Beck et al. (1972) and Chen and Prusoff (1978). Batch cultures (250 ml) were grown overnight and harvested by centrifugation for 10 min at 10,000 x g coli C600 and Pseudomonas stutzeri JM300 were washed in 0.9% NaCl and the marine isolates were washed in ASWJP. The cells were collected by centifugation and resuspended in 5 ml assay buffer (0.1 M Tris-HCl, pH 7.8; mM MgC12 ; 2 mM The cell solution was transferred to an acid washed snap cap vial and sonicated on ice for three 20 sec intervals at 70 w. The cell extract was centifuged for 15 min at 17,000 x gat 3 C and the supernatant was collected and 1/5 volume 10% streptomycin sulfate was added to precipitate nucleic acids. The sample was mixed and stored on ice for 30 min, clarified by centrifugation for 15 min at 17,000 x g and 3 C, and dialyzed against enzyme buffer for 2 hr' at 4. Enzyme reactions (50 consisted of 42 of cell extract, 2 50 mM ATP (final concentration 25 mM), 2 25 mM KCl (final concentration 1 mM), 2 of mM thymidine (final concentration 10 and 2 of [3H]thymidine in deionized water. The reactants were mixed and 5 samples were taken immediately and after 15 min incubation at 37 C. Each sample was spotted onto a DEAE cellulose filter disk (Whatrnan) placed on a filtration manifold with the vacuum on and allowed to dey for 30 sec then washed with 10 ml methanol.

PAGE 104

89 Radioactivity in C3H]dTMP bound to the was determined by liquid scintillation counting in the extracts was by the method of et al. (1951). Molecular thymidine kinase genes. Single stranded DNA a Herpes simplex (HSV) thymid i ne kinase gene (tdk) M13 templates (Bethesda Bethesda, Md) by the method of Holben et al. (1988) Single stranded RNA probes HSV tdk and an coli tdk us ing pGEM 3Z and 4Z plasmids Biotech, Madison, WI). M13 single Plasmid pSBTK2. 0 possesses a 2 0 Kb containing a thymidine kinase gene HSV and was donated by Paul and Diekmann of Stanford Uni ty. The 2 0 Kb Hind III -Barn HI was subcloned into the double Ml3 mpl8 replicate (RF) using techniques (Bethesda Md; Maniatis et al., 1982). Single strand Ml3 containing the 2.0 Kb was isolated and infected coli JMlOl host cells by polyethylene glycol/NaCl of infected supernatants (Bethesda The DNA was in TES buffer (20 rnM pH 7 5 10 rnM NaCl, 0.1 rnM EDTA) then phenol: chloroform and ethanol at -:20C. Purified single DNA was collected by and in TES. This DNA was used to single specific DNA by the method of Holben et al., (1988). A primer was bound to the DNA and the extended using the Klenow of DNA I in the of one to [32 P]deoxynucleotide (NEN, Boston, MA) and unlabelled

PAGE 105

90 triphosphates. The mixture was then restriction endonuclease enzyme digested at a site immediately downstream of the cloned fragment. The DNA was combined with 150 of loading buffer (89 mM Tris pH 8 .0; 89 rnM boric acid, 2 mM EDTA, 0.1% xylene cyanole, 0.1% bromophenol blue, 80% formamide) and denatured by immersion in boiling wate r for 10 min. The sample was electrophoresed on a 4% polyacrylamide gel containing 50% (w/v) urea for 4-6 hr at 400V. The radiolabelled probe fragment was located by autoradiography (30 to 60 sec exposure) and excised. The gel fragments containing the probe were extraction for 6 hr to overnight at 37 in elution buffer (0.3M Liel, 10 mM Tris pH 7.5, 0 .05 % SDS, 0.1 mM EDTA). The solution was passed through a QUIK-SEP column (Isolabs, Inc., Akron, Chio) to remove gel fragments to yield the purified probe. Hybridization conditions for Ml3 probes. All membranes were charged modified nylon (Zeta-Probe; Bio Rad, Richmond eA; MSI, Westborough, MA). Filters were floated on 6X sse until thoroughly wetted, immersed for five min at room temperature, then pre-washed for 3 hr at 42e in a solution containing 50 mM (pH 8 0) 1M Nael, 1 mM EDTA, and 0.1% SDS. Filters were prehybridized for an additional 5 1/2 hr at 42 in a solution containing 50% formamide, 5X Denhardts solution, 0.1% SDS, 0.1 mg/ml denatured salmon sperm DNA, and 5X SSPE (Maniatis et al., 1982). After 5.5 hr, the purified probe was added directly to the prehybridization solution and hybridization was allowed to proceed for 36 hr at 42. The filters were then washed four times for 10 min each in 2X sse containing 0.1% SDS at room temperature followed by 3 washes for 90 min each at 68 in 1X sse containing 0.1% SDS. The filters were dried and hybridized radioactivity was

PAGE 106

determined by autoradiography using Kodak XAR film (Eastman Kodak Co., Rochester, New York). 91 Construction of Riboprobe RNA probes. The second probing strategy involved cloning the 2.0 Kb HSV tdk and a 1 .4 Kb Bam HI fragment from pGQ3 (carlson et al., 1986) containing coli tdk into the pGEM 3Z and 4Z plasmids of the Riboprobe vector system (Prornega Biotech, Madison, WI). These plasmids contain promoters for SP6 and T7 RNA polyrnerases flanking the multiple cloning site which allowed the generation of sequence specific single stranded RNA probes. The methodology used is essentially that of Church and Gilbert ( 1984) and that provided by Promega Biotech (Bulletin # 002, 033, 037) and the reactions for making probes are summarized in Figure 14. The 2.0 Kb Hind III -BAM HI fragment containing the HSV tdk was cloned into the pGEM4Z vecter and the 1. 4 Kb Bam HI fragment containing _. coli tdk was cloned into the pGEM3Z vector using standard techniques (Maniatis et al., 1982). Each plasmid was transformed into JM109 (Maniatis et al. 1982) and large scale plasmid purification was performed by the methods of Littman (personal communication) and Zervos et al. (1988) Purified plasmid was digested with a restriction endonuclease immediately down stream to the cloned fragment from the promoter for the RNA polymerase which would be used (Figure 14). After digestion was complete, the DNA was purified by phenol:chloroform and chlorofonn:isoamyl alcohol extraction. Sodium acetate was added to 0.3 M and the DNA precipitated in 2 volumes of ethanol overnight at -20 c The DNA was collected by centrifugation, washed once in 70% ethanol, then resuspended in TE (10 rnM Tris-HCl, 1 rnM EDTA, pH 8 .0) to a final concentration of approximately 1 [35s]UTP was incorporated into

PAGE 107

Synthesis Of Specific RNA Probes T7 Promoter t Restriction Digest t Add RNA Polymerase Add ( 32 P] UTP + NTP's SP6 Promoter Add DNase + Phenol chloroform extraction Et-OH ppt i Spun column l ssRNA probe 92 Figure 14. Summary of the procedure used to make RNA probes using the Riboprobe system

PAGE 108

RNA by the action of the desired RNA polymerase After RNA synthesis, an RNase-free DNase (RQ1) was added to remove the DNA template and diethylpyrocarbonate was added to denature all enzymes. The RNA was 93 purified by phenol: ch loroform and chloroform:isomyl alcohol extraction then precipitated in 2 volumes of ethanol at -80 C for 1 hr. The RNA was collected by centrifugation, washed once with 70% ethanol, and resuspended in STE (10 mM Tris-HCl, 100 mM NaCl, 1 mM EDTA, pH 8.0) containing 10 mM dithiothreitol (DTT). Unincorporated nucleotides were removed by the spun column procedure of Maniatis et al. (1982) Hybridization conditions for RNA probes Filters were wetted and washed for 5 min at room temperature in 0.4 M Tris (pH 8.0). Prehybridization and hybridization solutions consisted of 0.25 M NaHP04 (pH 7.2), 0.25 M NaCl, 1 mM EDTA, 7% SDS, 10 mM DTT, and 0.1 mg/ml denatured salmon sperm DNA. Prehybridization was for 1 5 min at 42 C. The prehybridization solution was withdrawn and replaced with fresh solution containing the probe and hybridization was allowed to proceed overnight at 42 C. After hybridization, the filters were washed for 5 min at room temperature in 2X SSC containing 10 mM DTT followed by three 1 hr washes at 65 C in PSE (0.25 M NaHP04 2% SDS, 1 mM EDTA) and three 30 min washes in PES (0 M NaHP04 SDS, and 1 mM EDTA) at 65. Filters were dried and radioactivity was detected by autoradiography. Stringency effects on Riboprobe hybridization. Three hybridization and three washing conditions were exrrained for their effects on hybridization. The three hybridization solutions were as described previously except that NaCl concentrations were 0.25, 0 .5, or 0 .75 M Filter washing was manipulated by performing variations on the

PAGE 109

above described procedure. After an in it ia 1 sse wash ( see above) all three treatments received three 60 minute washes in PSE at room temperature. Variations occurred in the final wash. Treatment I received no further washes, treatment II received two 15 min washes in PES at 37, and treatment III received two washes in PES at 50. Sensitivity and specificity of probing. To establish the 94 sensitivity of the probing technique, standard curves were constructed by dot blotting purified DNA preparations. For Riboprobe RNA probes, plasmids pWJl, pWJTK, PGQ3, PS8TK2.0, and calf thymus DNA were dotted in concentrations ranging from 100 ng to 1 pg of DNA. For Ml3 single stranded DNA probes, Ml3 RF, Ml3 clone containing the 2.0 Kb HSV tdk insert, pSBTK 2.0, pOQ3, and calf thymus DNA were spotted. Dot blots were prepared for probing as described by M:miatis et al. (1982). Probing sensitivity and specificity were determined by autoradiography. Chromosomal DNA preparations. 100 ml of phase cells were harvested for 10 min at 10,000 x g, washed once in TES, harvested again, and resuspended in 25 ml of a solution containing 25% sucrose, 50 mM Tris pH 8.0, and 1 mg/ml lysozyme (Sigma Chemical Co.). After 30 min on ice, 1.25 ml 0.5 M EDTA was added. After an additional 15 min on ice, 2.8 ml 10% Sarcosyl (IBI, New Haven, CT) in TES and 10 of 5 mg/ml heat treated RNase were added. The solution was mixed and stored at room temperature for 30 min. Two mg proteinase K (Bethesda Research Laboratories, Gaithersburg, Md) was added and the solution incubated at 37 for 30 min. The mixture was transferred to a 50 ml conical centrifuge tube and extracted twice with an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1) and twice with an equal volume of chloroform:isoamyl alcohol (24:1). Two volumes of ethanol

PAGE 110

95 were added, mixed, and the solution placed in ice for 30 min. DNA was twirled up on a glass rod and dialyzed overnight against TES. Subsamples of the viscous DNA solutions were quantified via the Hoechst 33258 method (Paul and Meyers, 1982) to approximate DNA concentrations in each extract. Preparation of samples for probing. Dot blots of chromosomal extracts and plasmid preparations of different organisms were spotted on charge modified nylon (Biorad). After drying under a heat lamp, the filters were placed on 0 .5N NaOH saturated blot block (Biorad) for 15 min, followed by two similar treatments for 3 min on 1M Tris (pH 7 .5 ) saturated blot block. Finally, each filter was placed on blot block saturated with 0.5 M Tris (pH 7.5) and 1.5 M NaCl for 5 min. The filters were then blotted dry and baked at 80C under vacuum for 2 hr (Maniatis et al., 1982). Filters were stored in sealed zip-lock bags in a freezer until probing. Colony lifts using charged modified nylon (MSI) were prepared for probing by the method of M:miatis et al. (1982). Five ml of the appropriate growth media was inoculated and the cells grown overnight. Five of the grown culture was spotted on a nutrient agar plate and the colony allowed to develop for 24 to 48 hr. After lifting off the colony with the filter, the filter was treated by placing it cell side up on blot block saturated with the following solutions; 10% SDS for 3 min, 0.5 N NaOH for 15 min, twice on 1 M Tris pH (7.5) for 3 min, and 0.5 M (pH 7.5) and 1.5 M NaCl for 5 min. Samples were blotted dry then placed in a vacuum oven for 2 hr at 80 Dried filters were stored in sealed zip-lock bags in a frost-free freezer.

PAGE 111

96 Results Ability of marine bacteria to incorporate thymidine. Thymidine incorporation rates under standard conditions for the marine isolates are presented in Table 12. Thirty seven of forty-one or 90% of the organisms examined were capable of incorporating thymidine into cold TCA insoluble material. The four organislins which did not incorporate thymidine were organisms number 3, 8, 27, and V. parahaemolyticus. Organism 1 was considered capable of thymidine incorporation owing to the fact that it grew at much slower rates under the standard conditions applied than did the other organisms. Thymidine uptake and incorporation and thymidine kinase enzyme activity. All of the organisms which were capable of thymidine incorporation derronstrated thymidine uptake as well as significant thymidine kinase enzyme activity in cell-free extracts (Table 13). Those organisms which could not incorporate thymidine often demonstrated small amounts of cell associated radioactivity after 30 min. Although this cell associated radioactivity was significantly greater than that associated with formalin killed controls (P < 0.05), it was usually two orders of magnitude less than that associated with those organisms capable of thymidine incorporation. Organisms incapable of thymidine incorporation also lacked thymidine kinase activity in cell-free extracts (Table 13). V. parahaemolyticus containing p003 had little thymidine transport and no thymidine incorporation, however, it did have a large amount of thymidine kinase activity indicating active expression of the transformed gene. The fact that this organism could not incorporate thymidine even with an active thymidine kinase indicates that the inability to incorporate

PAGE 112

Tabl e 12. Thymidine inco rporation rates of marine bacterial isolates Organism nmol/ L hr Organism nmol /L hr 9 0 + 0 3 22 71.7 + 5.0 2 7 1. 0 + 1.2 23 215.2 + 6.4 3 1 .5 + 0.8 24 194. 0 + 5 8 4 126. 0 + 8. 1 25 496.8 + 34.0 -5 26. 7 + 3 2 26 375. 8 + 2.7 6 1 9 1 .8 + 6.9 27 0.0 7 91 8 + 2. 5 28 106. 0 + 9.6 8 0.0 29 4 40.0 + 8.6 -9 129.2 + 11 5 30 94. 3 + 1 7.3 10 90. 8 + 0.5 31 94.3 + 2 4 11 1 29.7 + 0.4 A 237.3 1. 1 1 2 7 4.3..: 0. 4 B 81. 3 + 6 9 13 N.D. c 93.3..:: 1. 5 14 1 91.0 + 1.4 D 82. 3 ..: 0.4 1 5 139.7..: 2.2 E 9 4. 8 + 1 1 1 6 500. 0 + 19.2 F 61 3 + 3.3 17 597. 8 + 24.1 G 8 9.3 ..: 1. 1 1 8 90. 7 + 4. 1 v. proteolyt 100. 0 + 3 1 19 92. 3 9.5 A citrea 226.7 + 7.6 20 194. 3 ..: 3 3 P atlantica 114.8 + 0.41 21 N.D. v. alginolyt 127.5 + 6.2 v. parahae m 0.0 Organisms numbered were isolated from Bayboro Harbor or the Dry Tortu gas O r ganisms V. proteo lyticus, and A. citrea -wer e biofouling bacteria isolated from the 01esapeake Bay. Cell den sities were normalized to A600=0.25 before the measurements were made. N.D. signifies not determ1ned. 97

PAGE 113

98 thymidine is due to the absence of a functioning transport system. Table 13 Thymidine uptake and incorporation and thymidine kinase enzyme activity in selected bacteria Organism Uptakea Incorporationb Thymidine Kinase Activityc _. coli C600 26.80 31 .63 282 V. proteolyticus 12.46 40.72 63 alginolyticus N.D. N.D. 530 P. atlantica N .D. #23 56.93 #27 1 .17 P. stutzeri JM300 0.46 #8 2.45 V. parahaem 1 .31 V. parahaem(pGQ3) 0.23 N.D. 77.17 0.94 0 .29 0.02 0.03 0.01 a Rates given in moles taken up/cell hr b Rates given in moles incorporated/cell hr c Enzyme activity given as nmol/min ug protein N.D. not determined 12 37 2 2 0 852 Probing for thymidine kinase genes: M13 single stranded DNA probes with HSV tdk. The sensitivity and specificity of the M13 probing system using the thymidine kinase gene from HSV is presented in Figure 15. The limit of sensitivity was approximately 30 pg (one third the weight of the spotted M13 clone DNA). There was no significant hybridization with either M13 RF or calf thymus DNA and there was only slight hybridization with purified pGQ3 which contains the coli tdk. This implies little homology between the HSV tdk and the coli tdk

PAGE 114

M13RF: M13TK ; pSBTK2.0 pGQ3 t CT 99 Figure 15. Probing sensitivity of M13 single stranded DNA probes with HSV tdk. DNA concentrations spotted were (A) 100 ng; (B) 10 ng; (C) 1 ng; (D) 100 pg; (E) 10 pg. M13TK was RF containing HSV tdk and CT was calf thymus DNA.

PAGE 115

that was cormborated when colony lifts of several ._. coli and Pseudomonas spp were probed (Figure 16) There was a very strong 100 hybridization between the probe and the coli containing the pSSTK2. 0 parent plasmid. There was limited hybridization with the coli containing the (the same plasmid as pS8TK2. 0 without the 2 0 Kb tdk insert) indicating limited vector hybridization. There was no hybridization of the probe to either of the two tdk Pseudomonas spp examined, only slight hybridization of the probe to the ._. coli containing pKT230, and no hybridization to coli JL4062 containing the pOQ3 plasmid. These results indicate that the HSV tdk probe is specific for itself and will not detect E coli thymidine kinase genes. When the HSV tdk probe was used to probe co lony lifts of the 42 marine isolates, there was no hybridization of the probe to any o f the organisms (data not shown). Probing for thymidine kinase genes: Riboprobe RNA probes. The results from the M13 system developed probes were unsatisfactory because the specific activity of the probes was low, the methodology was complex and time consuming, and this system required the undesirable use of 3 2 P-labelled nucleotides. As a consequence, the 1 6 Kb coli tdk fragment from pOQ3 and the 2 0 Kb HSV tdk were cloned into Riboprobe vectors pGEM4Z and pGEM3Z, respectively, resulting in plasmids pWJTK containing the 2 .0 Kb HSV tdk fragment and pWJ1 containing the 1 .6 Kb E coli fragment. Each of the probes was hybridized to a series of col ony lifts of ... coli and Pseudomonas spp (Figure 17). The.. co l i tdk probe hybridized with all four of the._ coli strains and did not hybridize to either of the tdkPseudomonas spp ( Figure 17A). The tdk from HSV gave similar results as those

PAGE 116

101 pSBTK2.0 pSB-NI JM300 PA01 pKT230 Figure 16. Colony lifts probed with M13 single stranded DNA probes with HSV tdk. The E. coli organisms C600(pKT230), JL4062, and HB101 (pSB-NI) all had functional E. coli thymidine kinase genes HB1 01(pSBTK2 .0) contained the HSV tdk and neither of the two Pseudomonas sp. (JM300 or PA01) contained thymidine kinase genes.-

PAGE 117

tdk from E. coli tdk from HSV PA01 pSBTK2.0 c.. 3: w 0 0 pSB-NI PA01 pSBTK2.0 Figure 17. Colony lifts of control organisms probed for thymidine kinase genes w ith Riboprobe RNA probes. E coli organisms (pKT230, JL4062, pSSTK2 0 and pSS-NI) all had functional thymidine kinase genes either on chromosomes or plasmids. The two Pseudomonas sp. ( JM300 and PA01 ) did not have thymidine kinase c.. 3: CN 0 0 _.. 0 1\)

PAGE 118

103 obtained using the M13 probe (Figure 178). There was hybridization to the parent plasmid containing strain and to the coli containing pKT230, however, there was no hybridization of the probe to the other E. coli strains or to either of the Pseudomonas spp In light of the fact that the HSV tdk probes could not detect thymidine kinase ge ne in the E. coli, all r-emaining probing was done with the coli tdk probe contained in pWJ1 Colony lifts of the marine isolates were probed but no significant hybridization occu red (Figur-e 18). Ther-e was slight hybridization with organism 27, yet this was an organism which did not incorporate thymidine (Table 12). Ther-efore, the pr-obe was most likely hybr-idizing to some other homologous sequence. Since these organisms were capable of thymidine incorporation, yet they did not hybridize to the probe, several stringencies of hybridization and washing conditions were examined. Standard curves which were hybridized under three different stringencies and washed with two washing regimes are presented in Figur-e 19. Increasing salt concentration in the hybridization solution resulted in minimal increases in hybridization. Similarly, the two washing regimes had only a slight effect on hybridization of the pr-obe to DNA on standard curves Sensitivity of the probe to pWJ1 was to approximately 1 pg DNA or-0.3 pg of the tdk fragment while sensitivity to the parent plasmid pGQ3 was approximately 10 pg or 3 pg of the tdk fr-agment and hybridization to the two plasmids containing the HSV tdk was two orders of magnitude lower. Liquid scintillation counting of standard curves indicated that there was only 1% of the hybridization to the HSV tdk DNA as there was to E. coli tdk (data not shown). Under the

PAGE 119

A 1''Z 28 '7rr; ' \ ". ". 12 . 2 ' 18 '\ 1 'L 11'1 l c 29 13 '" .. _16 10 l 26 .<.'J-l-:l 9 '15 5 31 i 25 8 I f "l! 7 6 ' 24 21 23 22 . >"', *" ; .. . . . .. ___ .!.. --" ,:; "' ')'. F .. v a c D e F G Vp Ac Pa Va 18. Colony lifts of isolates with E coli tdk gene designations the same as those in Table 13. _. 0 .:=

PAGE 120

Figur-e 19. Str-ingency effects on standar-d curves pr-obed with E. coli tdk gene pr-obes. DNA concentrations on each spot wer-e-cAT 100 ng; (B) 10 ng; (C) 1 ng; (D) 100 pg; (E) 1g pg; and (F) 1 pg. Hybr-idizations wer-e all over-night at 42 C at thr-ee str-ingencies depicted by Ar-abic numer-als. (1) 0.75 M NaCl; (2) 0 5 M NaCl; (3) 0.25 M NaCl. Differ-ent washing r-egimes ar-e r-epr-esented by Roman numer-als. Both r-eceived a washing for5 min at r-oom temper-atur-e in 2X SSC containing 10 mM OTT and thr-ee washings in PSE for-60 min each at r-oom temper-atur-e. (II) r-ecei6ed an additional two washings for15 min each in PES at 50 C

PAGE 121

1 0 6 ABC D E F pWJ1 pWJTK pGQ3 1 pSBTK2.0 CT v 2 c c 0 3 II

PAGE 122

107 least stringent conditions, there was some non-specific hybridization of the probe to calf thymus DNA Which was detectable by autoradiography at 100 ng DNA (Figure 19) In an effort to reduce hybridization interference which may arise when probing crude colony lifts, chromosomal DNA preparations dot blotted on charged modified nylon. A indicated that salt concentrati on in the hybridization solution had little effect on hybridization of the probe to chromosomal DNA dot blots from several coli, Pseudomonas spp., and bacteria (data not shown). in a second experiment, approximately 1 ug of chromosomal and plasmid DNA from several organisms was dot blot ted on charge modified nylon. Hybridization was conducted at the middle salt concentration (0.5 M NaCl) and the effect of three washing regimes examined 20). As expected, the chromosomal dots from E. coli JM101 and JL4062 were strongly hybridized with the was little hybridization to! coli C600 or HB101. It was believed that this may have been due to a minimal amount of DNA having been dotted for these samples due to the highly viscous of DNA preparations. Colony lifts of these two organisms strongly hybridized to the probe (Figure 20) indicating the ability of this probe to detect thymidine kinase genes in all coli strains examined. The did not hybridize to of the two tdk Pseudomonas spp. or to any of the environmental organisms examined except number 27. These results were identical to those reported for co lony lifts (Figure 17, 18). Stringency had little effect on hybridization results. At the lowest stringency washing, there was an equal or greater amount of hybridization of the probe to calf thymus DNA as there was to any of

PAGE 123

Figure 20. Stringency effects on chromosomal DNA dot blots probed with E. coli tdk gene probes. Each dot was spotted with approximately 1 DNA. Samples probed (L-R) were (A) pGQ3, pSeTK2.0, E. coli JM101, E. coli C600, E. coli HB101, E. coli JL4062, calf thymus DNA:-""[BJP. stutzeri JM300, P. aerUgillosa PA01, #8, lfZ{, lf3; and "l/:6. (C) V. -proteolyticus, V. alginolyticus, P. atlantica, and 1111. Lack of hybridization to E:coli c6oo and HB101 was believed to be due to insufficient DNA being dotted on the filter. Colony lifts were probed and included to demonstrate that the E. coli probe did hybridize to these strains. All three filters were hybridized at 0.5 M NaCl at 42 C. Different washing regimes are designated by Roman numerals. All three were washed for 5 min at room temperature in 2X SSC containing 10 mM DTT and three times in PSE for one hr at room temperature. II was washed twice in PES at 3 7 for 30 min and III was washed twice in PES at 50 for 30 min.

PAGE 124

_C600 HB101 A & B c A 8 c \ A I I I 8 Ill c 109

PAGE 125

the environmental organisms except number 27 implying that any slight binding of the probe to environmental organisms was probably due to non-specific hybridization. Discussion The results presented herein support the assumption that the vast majority of bacteria in aquatic environments are capable of incorporating thymidine into DNA. Of the 41 organisms examined, only four were not able to incorporate [3H]thymidine into TCA precipitable material (Table 12). All of the organisms which could not incorporate thymidine also did not demonstrate significant thymidine kinase activity in cell-free extracts (Table 13). tvblecular probing for 110 thymidine kinase genes using both coli and HSV tdk probes proved unsuccessful. Only one of the environmental organisms demonstrated significant hybridization to the coli tdk probe, yet this organism was incapable of thymidine incorporation implying limited conservation of nucleotide sequences among bacterial thymidine kinase genes, the presence of nucleotide kinases, or unique pathways for nucleoside phosphorylation. It must be noted that the enzyrrE assay used in this study and the incorporation of [3H]thymidine into TCA precipitable material cannot distinguish between the activity of thymidine kinase and a non-specific nucleoside kinase which may use thymidine as a substrate. While it is assumed that thymidine kinase was involved with thymidine incorporation by the marine isolates, other uncharacterized enzymes may have been present. The organisms which were used in this study were isolated from a

PAGE 126

111 wide variety of envirorurents and 90% were capable of thymidine incorporation. However, all of these organisms were maintained on nutrient rich agar, and as such, must be classified as part of the minority of environmental bacteria which are culturable (Colwell et al., 1985; 1988; Rollins and Colwell, 1986). The data presented herein do not allow speculation on the proportion of viable but bacteria which might be capable of thymidine incorporation. Using microautoradiography, Fuhrman and Azam (1982) demonstrated that the majority of active bacteria in coastal waters were also capable of taking up [3H]thymidine. The technique they used, however, cannot distinguish between uptake and incorporation of thymidine One of the organisms which did incorporate thymidine and demonstrated thymidine kinase activity was Pseudomonas atlantica. Previous investigations have reported that f stutzeri, aeruginosa, mendocina, alcaligenes, P. pseudoalcaligenes (carlson et al., 1985), pseudomallei (Saito et al., 1985), bathycetes, and P. marine (Pollard and Moriarty, 1984) all do not incorporate thymidine. However, nautica and an unidentified freshwater Pseudomonas species have been shown to incorporate low levels of thymidine (Pollard and Moriarty, 1984) while P. douderoffi, P. nigrafaciens (Pollard and -Moriarty, 1984), maltophilia, fluorescens (Saito et al., 1985) have demonstrated high levels of thymidine incorporation. This study is the first report of thymidine kinase activity and thymidine incorporation in f atlantica. As more orga nisms have been examined, the assumption that Pseudomonas species are incapable of thymidine incorporation has proved invalid. Lack of thymidine incorporation by V. parahaemolyticus agrees with

PAGE 127

112 results reported by Saito et al. (1985). They examined two strains of V. parahaemolyticus as well as V. fluvialis and reported no thymidine kinase activity in any of them. However, the high thymidine incorporation rates and thymidine kinase activity reported here for v. proteolyticus and V. alginolyticus indicated that lack of thymidine kinase activity is not indicative of Vibrio species. The inability of V. parahaemolyticus to incorporate thymidine was due to both the absence of an active transport system as well as thymidine kinase. By transforming V. parahaemolyticus with pGQ3 containing the E. coli thymidine kinase gene (see Chapter 6), it was hoped that thymidine incorporation could be provided to this organism as has been demonstrated in several Pseudomonas species and a thymidine kinase mutant of Salmonella typhimurium (carlson et al., 1985). The three transformants examined, however, demonstrated no incorporation of thymidine even after growth in the presence of 1 00 nM thymidine for 4 hr (data not shown). Enzyme assays indicated that the E. coli tdk was functional and expressing in transformants (Table 13) but thymidine incorporation was prohibited by the absence of a functioning thymidine transport system in this organism (Table 13). The ability to incorporate thymidine does not appear to be a consistent characteristic even within members of the same genus as exemplified by Vibrio and Pseudomonas species. Saito et al. (1985) examined 131 strains of 73 species and found a mixture of results. The majority of Bacillus, Staphylococcus, Clostridium, Salmonella, Escherichia, Klebsiella, Serratia, Proteus, and Bacteroides species examined demonstrated significant thymidine kinase activity, although there were exeptions within most genera The Lactobacillus strains

PAGE 128

examined could be divided into two groups, those with high thymidine kinase activity, and those with little if any enzyme activity. The distribution of the enzyme could not be linked to patterns of carbohydrate fermentation. In contrast, virtually all of the actinomycetes and related organisms as well as Candida species did not demonstrate thymidine kinase activity. Saito et al. (1985) compared thymidine kinase activity with location on the 5S rRNA phylogenie tree but could report no consistent relationships. 113 The organisms that were incapabl e of thymidine incorporation did demonstrate limited thymi dine transport or uptake. Presumably the uptake of [3H]thymidine was an active process since washing with unlabelled thymidine did not remove it and formalin killed controls did not show this characteristic. However, the amount of thymidine taken up by these cells was usually two orders of magnitude less than that taken up by cells capable of thymidine incorporation. Whether the thymidine taken up by cells that do not incorporate thymidine would be capable of being incorporated in the presence of an active thymidine kinase is unknown, although the absence of incorporation by V. parahaemolyticus(pGQ3) implies that the limited cell associated thymidine is not available for incorporation. Without an active thymidine kinase, thymidine taken up by cells would be limited to catabolism by thymidine phosphorylase. carlson et al. (1985) reported on the absence of thymidine phosphorylase in organisms that also lack thymidine kinase It is unknown whether the marine bacteria lacking thymidine kinase contain thymidine phosphorylase, although presence of this enzyme might explain the limited capacity for thymidine transport that these organisms demonstrated.

PAGE 129

114 Thymidine kinase deficiency is an interesting phenotype. The majority of the organisms examined grew quickly in ASWJP+PY but absence of a functional salvage pathway would be disadvantageous to fast g rowing cells (Saito and Tomioka, 1984) since de novo synthesis of nucleotides is energetically rore expensive than is the salvage of preformed nucleosides (Okazaki and Kornberg, 1964; Saito and Tomioka, 1984). The evolution of thymidine kinase deficiency must allow some other advantage not known at this time. While thymidine incorporation and enzyme assays provided consistent results concerning the presence or absence of thymidine kinase, molecular probing using E. coli and HSV tdk probes proved less beneficial. Neither of the two probes hybridized with colony lifts of chromosomal DNA dot blots of any of the marine isolates which contained thymidine kinase. The HSV tdk probes did not hybridize to all of the coli strains (Figure 16; 17B) and derocmstr.ated only limited hybridization to purified coli tdk conta ined on pGQ3 (Figure 15). These results indicate only limited sequence homology between HSV and E. coli thymidine kinase genes The E. coli tdk probe proved able to detect thymidine kinase genes in all of coli strains examined (JM101, C600, HB101, JL4062; Figure 17A; 20) indicating some conservation of the gene sequence within E. coli strains. However, probes failed to detect thymidine kinase genes in any of the marine isolates even under the least stringent conditions applied that would allow hybridization to occur between sequences with less than 70% homology (Wahl et al., 1987) indicating limited homology and conservation of thymidine kinase DNA sequences between different bacterial species.

PAGE 130

115 The rrajority of molecular analysis of thymidine kinase has involved viral genes because they are regulated during virus growth and they serve as a ver y useful selective marker in the ge netic transfection and transformation of many thymidine kinase lacking mammalian cell lines (Garapin et al., 1981). Nucleotide sequences for thymidine kinase genes have been determined for (Wagner et al., 1981), (Kit et al., 1983), marmoset herpes virus (Otsuka and K i t 1984), shope fibroma virus (Upton and McFadden, 1986), Epstein-Barr virus (EBV; Littler et al., 1986), as well as vaccinia virus (Weir and Moss, 1983), variola virus, and rnonkeypox virus (Upton and McFadden, 1986). Other eukaryotic thymidine kinase genes that have been sequenced include those from human, mouse, hamster, and chicken cells (Upton and McDadden, 1986). Several studies have examined the homology between sequences of different organisms' thymidine kinase genes. HSV-1 and HSV2 contain 91% sequence homology in their promoters for thymidine kinase and approximately 89% homology in their structural genes (Kit et al., 1983). The thymidine kinase of EBV produces a protein with antigenic cross reactivity with the HSV thymidine kinase. Analysis of predicted amino acid sequence between EBV and indicated 27% identical amino acids and further 17% conservative amino acid changes (Littler et al., 1986). In contrast, the nucleotide sequence of the marmoset herpes virus thymidine kinase gene does not contain appreciable sequence homology to either HSV-1 or HSV-2 (Otsuka and Kit, 1984) Upton and McFadden (1986) compared the nucleotide sequences of thymidine kinases from shope fibroma virus, vaccinia virus, variola virus, and rnonkeypox virus, as well as those from human, mouse, hamster, and chicken cells.

PAGE 131

116 Significant homology was reported between shope fibroma virus, variola, vaccinia, and monkeypox viruses as well as the human, mouse, hamster, and chicken thymidine kinases. However, there was no obvious homology between these genes and the thymidine kinase genes from HSV (Upton and McFadden, 1986). One reason for these results may be that both EBV and HSV thymidine kinases are substantially larger than those from vaccinia and the eukaryotes and they serve as deoxynucleoside kinases with broad substrate specificity whereas those from the other organisms are exclusively thymidine kinases (Littler et al., 1986). Phosphate acceptors for E. coli thymidine kinase include thymidine, deoxyuridine, and the 5-halogenated analogs 5 -fluoro-, 5-chloro-, 5-brorro-, and 5...:.iododeoxyuridine (Okazaki and Kornberg, 1964). Phosphate may be donated by most nucleotide triphosphates except dTTP with the preferred donors being ATP, dGTP, dATP, dCTP, GTP, and ITP (Okazaki and Kornberg, 1964). While no information is known about the substrate specificity of the thymidine kinase enzymes of the organisms examined in this study, they did not hybridize to probes constructed from either the broad substrate specificity HSV tdk or the limited substrate specificity E. coli tdk signifying that lack of homology may be related to some other characteristic of the enzymes. The apparent lack of homology between E. coli thymidine kinase genes and those for the environmental organisms may provide a valuable lesson for other researchers. The use of molecular techniques to examine environmental research questions has become a popular approach in microbial ecology. As more genetic markers are determined, molecular probing of samples will increase in attempts to genetically characterize different samples and populations. Caution must be used,

PAGE 132

117 however, when using heterologous gene probing in mixed populations. The efficiency and accuracy of probing environmental samples will be dependent on conservation of gene sequences between the probe and target sequences and results could be open to interpretation until gene conservation is known.

PAGE 133

118 CHAPTER 6: DETECTION OF NATURAL TRANSFORMATION Introduction The detection of natural genetic transformation in environmental samples is of importance for determining risk assessment for the release of genetically engineered microorganisms and for learning more about genetic adaptation to environmental stress and change. In this part of the study, natural transformation with plasmid DNA was studied in both bacterial isolates and in environmental samples The ultimate goa l was to determine if growth rates, as measured by [ 3 H]thymidine incorporation in the presence of oolective agents, could be used as an indicator of natural transformation. Methods and Materials Transforming DNA. The broad host range plasmid pKT230 (Bagdasarian et al., 1981) was used throughout this study. It was originally obtained by replacing the small Pst I generated fragment of RSF1010 with the PACYC177 plasmid resulting in an 11.9 Kb chimera of the two plasmids (Figure 21) encoding for resistance to the two protein synthesis inhibitors kanamycin and streptomycin. In subsequent experiments, a derivative of pKT230, pGQ3, was also used. This plasmid was formed by the addition of a 1 .6 Kb fragment containing an coli thymidine kinase gene into the unique Bam HI site of pKT230 (Figure 21; Carlson et al., 1985). Isolation of the plasmids from their E coli

PAGE 134

aJ Q) 3 ::J: .. \ ""'11ct s, If: '11ct .t ''" ,..---....... ,,, "'lit ( ss\\ Xho 1 Xhol pKT 230 I I pGQ3 11. 1 Kb 1 3.5 Kb Figure 21. Maps of the transforming DNA plasmids used in this study. pOQ3 was formed by inserting the E. coli tdk into pKT230 at the Bam HI restriction site. ----__. 1.0

PAGE 135

hosts was performed by large scale alkaline lysis procedures (G. Littman personal communication) followed by purification with pZ523 columns (5 _..:.> 3', Paoli, Pa.; Zervos et al., 1988) Generation of plasmid multimers : pKT230 and pOQ3 were digested with the restriction endonuclease Xho I (IBI, New Haven, CT) which resulted in linear plasmid molecules. The restriction enzyme was denatured by heating the sample to 70 C for 20 min and after cooling, T4 DNA ligase (Bethesda Research Laboratories, Gaithersburg, Md. ) was 120 added to anneal the linearized plasmid molecules into multimeric forms. The detailed reactions were as follows: 10 pKT230 or pGQ3 in 30 TE 4.5 pasteurized 1% BSA 4 5 10 X restriction enzyme buffer (IBI) 5 Xho I (25 The mixture was gently vortexed and incubated at 37 for 2 hr then the reaction was terminated by heating at 70 C for 20 min. After cooling, the ligation reaction was performed as follows: 45 DNA solution 15 5 X ligation buffer (BRL) 20 sterile deionized water 16 T4 ligase The mixture was incubated overnight at 15 C and the reaction terminated by the addition of EDTA to a final concentration of 25 mM. The success of the procedure was determined by agarose gel electrophoresis (DeFlaun and Paul, 1986) using a 0 .4% agarose gel. . Natural transformation assays i n environmental samples A schematic representation of the procedure for detecting natural transformation in aquatic environments is presented in Figure 22. Water samples from Bayboro Harbor, the Medard Reservoir, the coral surface microlayers of the Dry Tortugas and Cay Sal Bank, Bahamas, and

PAGE 136

1 Liter Sample 8 100 ng/ml DNA (CJ: or pKT 230) 20 hrlt-------.-Filter Subsamples Add Peptone Yeast / "Probe No Ab Ab1 Ab2 Ab3 Ab Plates 24 hr Probe 8 hr l I I I 3H Thymidine lncorp. 1 -direct count & plate 16 hrl I l I 3HThymldlne lncorp. 2 -direct count & plate Figure 22. Schematic representation of sampling procedures for environmental transformation assays. 121

PAGE 137

122 an oligotrophic offshore sample from the southeastern Gulf of Mexico were amended with either purified pKT230 or calf thymus DNA to a final concentration of 100 ng/ml. The samples were incubated at room temperature or 28 C with gentle shaking for 20 hr then subsamples were taken for direct detection of transformants by molecular probing and for plating in attempts to isolate transformants. To the remaining sample, sterile concentrated and yeast extract (Difco, Detroit, Mich.) was added as a nutrient source to a final concentration of m5 g/ml peptone and 1 mg/ml yeast extract except where noted The sample was then subdivided into smaller groups which were given increasing amounts of kanamycin and streptomycin and incubated again for 4 to 8 hr then sub sampled for [ 3 H]thymidine incorporation and a second [ 3H] thymidine incorporation assa y was performed 8 to 16 hr later. [ 3H]Thymidine incorporation assays were conducted using 35 nM [3H]thymidine (approx. 2 to 2.5 Radioactivity incorporated into cold TCA precipitable macromolecules was determined as described previously. Rates were calculated using linear regression and compared between samples amended with pKT230 and calf thyrrus DNA for the different antibiotic concentrations. Transformation in environmental sediments Bacterial populations in surface sediments from two locations were tested for the ability to naturally transform. Fine sand sedirents from shallow water were collected off of North Shore Park, St. Petersburg, FL using a small piston coring device constructed by cutting off the end of a sterile 60 cc syringe (Becton Dickenson, Rutherford, N. J.). The top 1 em of sediment was repeatedly sampled until the syringe contained 60 cc of

PAGE 138

123 sediment. This sediment was transferred to a sterile plastic beaker and mixed thoroughly with a sterile spatula. Sediment microcosms were constructed in vertically mounted sterile 60 cc syringes. The syringe was plugged with sterile silanized glass wool ( Applied Sciences Laboratories, St. College, Pa.) and 25 cc of sediment was loaded into the column The colunn was packed with three 25 ml washes of autoclaved and filter sterilized ASWJP. Seventy..:.five ).Jg of DNA (monomer pKT230, multimer pKT230, or calf thymus DNA) was added to each column in 25 ml autoclaved and sterile filtered ASWJP by allowing it to drain through the column until the liquid meniscus touched the sediment surface. The flow was stopped by imbedding the 18 gauge needle on the end of the column in a neoprene stopper and the columns allowed to incubate overnight at room temperature. The columns were washed by passing 25 ml 1 M NaCl through them and sacrificed by removing the ends with a sterile scalpel, the cotton plug was discarded, and the sediments mixed with a sterile spatula. Using a miniature coring device made from a sterile 10 cc syringe, eight 2 cc subsamples were placed in sterile 15 ml conical tubes. Two ml of sterile ASWJP+PY containing 250 ).Jg/ml kanamycin and 1000 vg/ml streptomycin was added to each tube and vortexed for 15 sec. The tubes were then placed at a 60 angle on a gyratory shaker and incubated with shaking for later [3H]thymidine incorporation assays (see below) A 1 cc sediment subsample was taken and vortexed with 1.7 ml sterile ASWJP for 2 min. The supernatant was sampled for plating on ASWJP+PY 1 .5% agar plates amended with 250 vg/ml kanamycin and 1000 ).Jg/ml streptomycin in attempts to isolate transformants. [3H]Thymidine incorporation assays were performed by the method of

PAGE 139

124 carmen et al. (in press) at 5 and 25 hr after the initial subsampling from the columns. Four of the eight 2 cc samples from each treatment were used for each C3H]thymidine assay. At the beginning of the assay, 0.5 ml Formalin and 0.5 ml 10 mM unlabelled thymidine were added to two tubes and the samples vortexed thoroughly This sample was the zero time point blank used to account for abiotic binding of the isotope. [ 3H]Thymidine was then added to all four tubes to a final concentration of 125 nM. The two assay tubes were placed back on the gyratory shaker and incubated with rigorous shaking for 3 hr. The two control tubes were centrifuged at maximum velocity for 10 min in a clinical centrifuge kept in a small refrigerator. The supernatant was discarded and 10 ml ice cold 10% TCA was added to each sample, vortexed, and stored on ice for 90 min with intermit tent vortexing at 30 min intervals. The samples were centrifuged as described above for 10 min and the supernatant discarded. The pellet was washed once with ice cold 10% TCA and twice in ice cold absolute ethanol and briefly dried in a 60 sand bath before 5 ml 1 N NaOH was added and the sample extracted at 37 for 1 hr with intermittent vortexing every 20 min. The sample was centrifuged and 1 rnl aliquots of the supernatant added to liquid scintillation vials. One hundred u l of concentrated HCl was added to neutralize the sample followed by 10 ml Aquasol (New England Nuclear) and radioactivity in RNA and DNA was determined by liquid scintillation counting After the 3 hr incubation, [ 3H]thymidine incorporation in the remaining two samples was terminated with the addition of Formalin and 10rnM unlabelled thymidine and the samples processd as described above. A second sediment sample was collected in a seagrass flat in Cay

PAGE 140

125 Sal Bank, Baha.IIE.s. The sediments in this location were a course cabonate sand The top 1 em of sediment was collected and colunns were prepared and treated as previously described with the addition of two additional treatments consisting of rifampin resistant chromosomal DNA isolated from several marine Vibrio spp. rifampin resistant mutants (G. Stewart, personal communication) and a calf thymus control for selection in the presence of rifampin (50 ug/ml) [3H]Thymidine incorporation assays in carbonate sediments required a modification of the extraction procedure. After the addition of formalin and unlabelled thymidine, the sample was transferred to a 100 ml plastic beaker and approximately 25 ml of 2 N HCl was slowly added to dissolve the carbonate sediments The contents of the beaker were then transferred to a 50 ml centrifuge tube and chilled for 90 min on ice. The precipitate was collected by centifugation in a clinical centrifuge for 10 min and the pellet was washed twice with ice cold absolute ethanol then dried before 3 ml 1 N NaOH was added The sample was vortexed and incubated at 37 for 1 hr and aliquots counted by liquid scintillation counting Filter transformation assays of marine isolates. Thirty marine isolates were screened for the ability to naturally transform with plasmid DNA us ing filter transformation assays. Cultures were inoculated from plates into 25 ml ASWJP+PY and grown overnight at room temperature on a gyratory shaker. They were then diluted 1 :25 into fresh media and grown for 2 hr at room temperature at 140 rpm to ensure that cells were in log phase growth One ml of cells was filtered onto a sterile 0.2 pore size N uclepore filter keeping the area covered by cells on the filter to a minimum (approximately 2 em in diameter).

PAGE 141

126 When the filter was almost dry it was placed cell side up on a 1 .5% agar ASWJP+PY plate and one of pKT230 DNA in 20 TE was added to the cells. Controls received no DNA and the cells were incubated overnight at room temperature. The filter was then aseptically removed and transferred to 10 ml ASWJP+PY in a 125 ml flask and shaken for 1 hr at 150 rpm to rerrove cells from the filter. T ransformation was determined by spread plating on ASWJP+PY plates amended with 100 kanamycin and 250 ml streptomycin. Total viable cells were determined by plating on ASWJP+PY without antibiotics and transformation frequency was determined as number of transformants per number of viable cells. Gene probing of potential transformants. Potential transformants were isolated and maintained on ASWJP+PY agar plates conta i ning 100 kanamycin and 250 streptomycin Seven ml of media were inoculated with a potential transformant or the wild type and grown overnight. The cells were harvested at 8000 rpm for 10 min at 15 C, the pellets resupended in STE, and harvested for 5 min at 12,000 rpm in a microfuge. The supernatant was discarded and the plasmid DNA extracted by the alkaline lysis procedure of Maniatis et al. After ethanol precipitation, the DNA pellet was resuspended in 20 sterile deionized water. One fourth of the extract was subjected to single restriction endonuclease digestion with Xho I and one fourth was double digested with Xho I and Bam HI. The samples were electrophoresed on a 1% agarose gel (DeFlaun and Paul, 1986) alongside similarly digested samples of pKT230 and the DNA transferred to charged modified nylon by the method of Southern (1975; Maniatis et al., 1982). The filter was dried and stored as described above.

PAGE 142

127 The 1.4 Kb Xho I-Bam HI fr-agrrent of pKT230 was cloned into the Ribopr-obe vectorsystem Pr-obes wer-e gener-ated using the sarre techniques as desribed forthe thymidine kinase pr-obes (Chapter5). Hybridization conditions wer-e identical using 0 .25 M NaCl in all hybridizations. Filterwashing consisted of one wash in 2X sse with 10 mM OTT forfive min at r-oom temper-atur-e followed by thr-ee 60 min washes at 65 C in PSE and thr-ee 30 min washes in PES at 65 c. Filter-s wer-e dried and hybr-idization determined by autoradiogr-aphy. Sediment micr-ocosm tr-ansformation assays with Vibr-io par-ahaemolyticus. Fine sand sur-face sediments wer-e collected fr-om shallow water-s of Tampa Bay, St. Peter-sbur-g, FL. Upon returning to the labor-atory, sediments were washed repeatedly by vortexing in ASWJP and sterilized by two consecutive autoclavings for 1 hr each. A series of sediment columns was then constructed by clamping sterile 10 cc syringes (Becton Dickenson) to a ring stand and plugging the bottom of each with ster-ile silanized glass wool. Three cm3 of sterilized sediment was added to each column and packed by three washes of 3 ml autoclaved/filter sterilized ASWJP. Thr-ee ml of autoclaved/filtered sterilized ASWJP containing 5 vg/ml of the appropriate DNA was allowed to drain through the column. The three types of DNA used were monomers of pKT230, multimers of pKT230, and calf thymus DNA (Type I, Sigma). The DNA solutions were held on the columns for 1 hr followed by a seco ndary 3 ml containing 109 cells/ml of parahaemolyticus Cells were grown to A600 = and harvested at 8000 rpm for 10 min at room temperature, resuspended in their original volume, then added and allowed to drain through the columns until the meniscus of liquid just touched the sediment surface. The columns were stoppered and

PAGE 143

incubated at room temperature overnight then sacrificed aseptically with a sterile scalpel by removing the end and the glass wool plug The 3 cm3 of sediment was then passed into a sterile 15 ml conical tube, five ml sterile ASWJP was added, and the contents vortexed thoroughly for 2 min to remove cells from the sedime n t The sediment 128 was allowed to briefly settle and the supernatant sampled for plating. The remaining supernatant was added to 35 ml ASWJP+PY amended with 100 )lg/ml kanamycin and 250 )lg/ml streptomycin and incubated at room temperature at 1 00 rpm on a gyratory shaker and sub sampled for [3H]thymidine incorporation assays. At each sampling time, 10 ml was rerroved from the original flask and placed in a sterile 25 ml polymethyl pentene flask (Nalg ene) [ 3H]Thymidine was added at 35 nM (2 1 )lCi/ml) and total TCA precipitable fractions were collected by filtration onto 0 2 )lm pore..:.size Nuclepore filters at 0 and 60 min. Each filter was washed with 5 ml 5% ice cold TCA and incorporated radioactivity was determined by liquid scintillation counting (Kobayashi and Harris, 1978). Chromosomal DNA transformation in V parahaemolyticus. Transformation of V. parahaemolyticus by rifampin resistance encoding v parahaemolyticus DNA was performed using sediment columns. Each 3 cm3 column was loaded with 15 )lg RIF DNA as previously described. Cells were grown to A600 = and then added to columns by one of three methods. In the first, cells were loaded onto the column without first harvesting them. For a second column, cells were first harvested by centrifugation then resuspended in an equal volume of sterile ASWJP, and finally, cells were first harvested then resuspended in one-tenth the original volume of sterile ASWJP then loaded onto the third column.

PAGE 144

129 Amplification of detectable transformants in a mixed population. Sediment column transformation with pKT230 monomers, rnultimers, and calf thymus DNA was performed as previously described using V. parahaemolyticus and an unidentified kanamycin and streptomycin organism hereafter referred to as UNKS1. These two organisms were easiliy distinguishable as V. parahaemolyticus produced creme colored colonies while those of UNKS1 were bright yellow. A schematic representation of the sampling regime i s presented in Figure 23. Twelve hr after the columns were sacrificed and inoculated into 35 ml antibiotic amended ASWJP+PY, 2 ml subsamples were diluted 1 :10 into fresh antibiotic ASWJP+PY. Twelve hr later, the original flask was subsarnpled again for a 1 :10 dilution. All flasks were allowed to grow for an additional 48 hr, by which time the cells had reached stationary phase. Each of the nine flasks was again subsarnpled for 1:10 dilutions and [3H]thymidine incorporation was determined on 10 ml aliquots of each final dilution after cells reached exponential growth 4 hr later. The remaining cell solution was allowed to grow to stationary phase. Seven ml subsarnples were collected from each flask and plasmid DNA extracted by the alkaline lysis procedure (Maniatis et al., 1982). Each DNA extract was dissolved in 20 deionized water. Five of extract was electrophoresed on a 1% agarose gel (DeFlaun and Paul, 1986) and transferred to charge modified nylon (Southern, 1975; Maniatis et al., 1982) The filter was subsequently probed with the 1 4 Kb Xho I -Barn HI fragment of pKT230 using the Riboprobe technique. Results Construction of multimers fo r use in transformation assays. The

PAGE 145

Sediment Microcosm -Vortex 2 i'nin & A Figure 23. Schematic representation of liquid selection/amplification procedure for isolating trans formants Sediment columns were sacrificed and vortexed for 2 min in ASWJP. 4 ml of the supernatant was added to flas k (A). Arrows represent 1 :10 dilutions into fresh media containing antibiotics a t after the designated time interval. w 0

PAGE 146

presence of multimers in the DNA preparations used for transformation assays with V. parahaemolyticus was confirmed by the reduced migration distance of the multimer DNA preparation when analyzed by gel elecrophoresis (Figure 24). The reactions used to generate multimers were limited to small scale. Ten plasmid DNA could be co nverted to multimers very efficiently yet attempts to scale up the reactions to larger amounts were unsuccessful (data not shown). Transformation in environmental samples. The results of [3H]thymidine incorporation assays conducted during a transformation experiment in Bayboro Harbor are pre se n ted in Table 14. Kanamycin was effective in inhibiting microbial growth (i.e. [ 3 H]thymidine 131 incorporation) in calf thymus DNA controls. As kanamycin was added in increasing concentrations, thymidine incorporation rates dropped from 2268 pmoles/L hr (no antibiotic) antibiotic to 14.9 pmoles/L hr (100 kanamycin). The transforming DNA (pKT230) did not serve to stimulate growth in the absence of antibiotic selection. High rates of [3H]thymidine incorporation for both the calf thymus control and the pKT230 samples after 27 hr with 100 }.lg/ml kanamycin demonstrated the presence of antibiotic resistant bacteria in these waters. After 27 hr, the enhanced growth due to the presence of the transformants was overshadowed by the growth of the even greater population of ambient antibiotic resistant organisms. However, the [ 3 H]thymidine incorporation rates for the samples in the presence of 50 }.lg/ml kanamycin demonstrate increased growth owing to the presence of pKT230 in both the 8 and 27 hr samples In addition, there was enhanced growth in the treatments containing 100 }.lg/ml kanamycin after 8 hr. Samples amended with rifampin resistance encoding chromosomal DNA

PAGE 147

23.1-9.4-6.6-4.4-2.3_ 2.0-pKT230 ABC D pGQ3 ABCDE Figure 24. Construction of multimeric forms of plasmid DNA. 132 I. (A) Hind III A molecular weight standards. Molecular weights given in Kb; (B) undigested pKT230; (C) Xho I d igested pKT230; (D) Xho I digested pKT230 after overnight treatment ligase to produce multimers. II. (A) Hind III A molecular weight standards; (B) undigested pOQ3; (C) Xho I digested p0Q3; (D) and (E) replicate samples of Xho I digested pGQ3 after overn ight treatment with

PAGE 148

Table 14. Transformation of marine populations in Baybor Harbor (3/19-21 /87) 8hr 27hr 133 prrol/L hr % CT DNA piJDl/L hr % CT DNA CT DNA No kan 2268 + 130 677 + 121 pKT230 No kan 1552 439 68 1108 + 70 164 CT DNA 50 kan 45.9 + 4 5 205 + 109 pKT230 50 kan 1129 + 57 2460 2112 +50 1030 CT DNA 100 kan 14.9 + 6.0 5395 + 143 pKT230 100 kan 24.7+1.1 166 5200 + 256 96 CT DNA 10 rif 5185 + 411 1783 + 55 RIF Chrom 10 rif 6866 + 110 118 1480 + 75 83 CT DNA 50 rif 2.8 + 2.6 0.0 -:-; RIF Chrom 50 rif 32.2 11.3 1167 28. 1 + 10.8 -2800 had greater [ 3 H]thyrnidine incorporation rates in the presence of 50 rifampin than their calf thymus controls at both samplings. After 8 hr there was an 11...:.fold increase in growth in the RIF DNA treated sample. After 27 hr, there was no growth in the calf thymus control, therefore, there was at least a increase in growth in the RIF DNA treated sample (Table 14). The results from an experiment conducted with water from the Medard Reservoir are presented in Table 15. After 8 hr, there was no discernable difference between the treatments and controls. However, after 25 hr, in the presence of 100 kanamycin, the pKT230 monomer amended sample had a 12..:.fold higher growth rate than did the calf thymus DNA control. In samples treated with 250 kanamycin, the pKT230 monomer amended sample had a 6.::.fold greater thymidine incorporation rate. In oligotrophic waters, the presence of the pKT230 plasmid and 100

PAGE 149

Table 15. Transformation by plasmid DNA in the Medard Reservoir ( 4 /23 -25 /87) 8hr' 25 hr' 134 prrol/L hr' % CT DNA prrol/L hr' % CT DNA CT DNAa 2 1 + 10 pKT230 mono 233 + 20 pKT230 mult 485 + 4 CT DNAb 0 pKT230 rrono 0 pKT230 mult 0 CT DNAc 0 pKT230 mono 0 pKT230 mult 1 CT DNAd ND pKT230 mono ND pKT230 mult ND a No antibiotic b Kanamycin added at 50 11g/ml c Kanamycin added at 100 11g/ml d Kanamycin added at 250 11g/ml ND not determined 1547 + 138 1100 1915 + 114 2310 2673 + 330 24 + 21 0 9 + 21 19 + 23 236 + 8 1 3 12 + 18 73 + 18 1 + 4 124 173 37 1242 3 608 3 11g/ml kanamycin resulted in an 825% increase in growth over the calf thymus control (Table 16) after 9 hr' of incubation, In each of the other treatments, except those with 50 11g/ml kanamycin, growth was also stimulated in the pKT230 samples, but not to as high a degree as for the 9 hr sample. Results from a cora l surface microlayer sample from the Dry Tortugas also yielded similar results (Table 17). Using double selection for both kanamycin and streptomycin, the plasmid amended treatment had a greater growth rate after 5 hr' and a 27-fold increase in growth after 15 hr.

PAGE 150

135 Table 16 Transformation by plasmid DNA in offshore waters (5/13-15/87) 9 hr 2 4 hr Treatment piiOles/L hr % CT DNA piiOl/L hr % CT DNA CT DNAa 3076 + 104 pKT230 3375 + 82 110 CT DNAb 3685 + 22 pKT230 543 + 3 15 CT DNAc 28. 9 + 3. 9 pKT230 + 825 CT DNAd 25.3 + 0.4 pKT230 38;9 + 154 a No added antibiotic b Antibiotic concentration was 50 kanamycin c Antibiotic concentration was 100 kanamycin ND ND ND ND 3760 + 263 5454 + 143 2695 + 55 3460 + 156 d Antibiotic concentration was 100 kanamycin and 25 streptomycin ND not determined ,145 :i128 [3H]Thymidine incorporation rates determined for a transformation assay in a coral surface microlayer sample from cay Sal Bank are presented in Table 18. For each of the two types of DNA, thymidine incorporation rates indicated that transformation may have occurred. At 5 hr there was no significant difference in the rifampin treated samples However, the pKT230 sample had a 359% greater thyroid ine incorporation rate than did its corresponding calf thyrrus control. After 15 hr of incubation, there was no significant difference between the pKT230 treatment and its calf thymus control. There was a 6-fold increase in the RIF DNA treated sample over its calf thymus control. In a transformation experiment conducted with water from the

PAGE 151

Table 17. Transformation in a coral surface microlayer of the Dry Tortugas (5/14-16 /87) 5hr 15 hr 136 pnol/L hr % CT DNA pnol/L hr % CT DNA CT DNAa 3332 + 189 ND pKT230 4151 + 382 125 ND CT DNAb 9310 + 800 ND pKT230 5191 + 407 56 ND CT DNAc 11547 + 1026 866 + 120 pKT230 0 :j1682 + 76 CT DNAd 118 + 15 221 + 317 pKT230 9437 260 7900 6038 + 1086 a No added antibiotic b Antibiotic concentration was 100 kanamycin c Antibiotic concentration was 250 ).lg/ml kanamycin d Antibiotic concentration was 250 ).lg/ml kanamycin and 100 ).lg/ml streptomycin ND not determined 194 2700 Medard Reservoir in March, 1988, an additional treatment was added. Filter transformations were conducted similarly to those done with culture organisms. After the cells were removed from the filter by washing in nutrient media for 1 hr, a 2 ml subsample was inoculated into 25 ml nutrient media containing antibiotics. After 12 hr, [3H]thymidine incorporation was determined. While there was no significant difference in thymidine incorporation rates for the liquid transformation assay (Table 19), the filter transformation assay provided different results (Table 20). The pKT230 amended sample had almost three times greater gro wth than did the calf thymus DNA control. In another water colunn experiment from the Medard Reservoir in the

PAGE 152

fall of 1987, three concentrations of antibiotic and three nutrient concentrations were examined for their effect on growth of transformants. In general, there was no strong evidence in any of the treatments that transformation may have occurred (Table 21) Table 18. Transformation in the coral surface microlayer of Cay Sal Bank ( 7/23-25 /88) 5hr 15 hr 137 prrol / L hr % CT DNA prrol/L hr % CT DNA CT DNAa RIF chrorro CT DNAb pKT230 0.0 0:0 16 + 10 56+ 21 359 a Antibiotic concentration was 50 rifampin 1193 + 38 7153 + 572 340 + 29 275 + 25 b Antibiotic concentration was 250 kanamycin and 1000 streptomycin 600 81 The results from the sediment transformation assay in Tampa Bay sediments are presented in Table 22. There was no significant difference between calf thymus controls or samples amended with pKT230 monomers or multimers. In the sediments of Cay Sal Bank, however, both RIF chrorrosomal DNA and pKT230 rronomers and multimers resulted in increases in growth over the calf thymus controls ( Table 23). A summary of all of the environmental transformation experiments appears in Table 24. Based on C3 H]thymidine incorporation data and using at least a 150% increase in growth as a positive indicator of transformation, 21 of 31 treatments (68%) showed higher growth rates in the presence of the transforming DNA. Transformation was indicated in approximately two-thirds o f the samples from each of the three environmental types examined.

PAGE 153

Table 19. Transformation in the Medard Reservoir by plasmid DNA (3/7-10/88) 9hr 29 hr 138 piTDl/L hr % CT DNA prrol / L hr % CT DNA CT DNAa 0 0 pKT230 7 + 10 >700 0 CT DNAb 2 + 3 0 pKT230 4 + 4 200 0 a Antibiotic concentration was 100 kanamycin b Antibiotic concentration was 200 kanamycin and 200 streptomycin Table 20. Thymidine incorporation rates for filter transformation samples from the Medard Reservoir (3/7-10/88) pmol/L hr % CT DNA CT DNAa 87. 1 + 25.8 "'-pKT230 226:4 + 14:6 260 CT DNAb 48.1 + 3.6 pKT230 27:4 + 2:5 57 a Antibiotic concentration was 100 kanamycin and 100 streptomycin b Antibiotic concentration was 200 kanamycin and 200 streptomycin Transformation in marine isolates. Of the thirty isolates examined for the ability to genetically transform with plasmid DNA, only one organism, identified as Vibrio parahaemolyticus (G. Stewart, personal communication), successfully incorporated and maintained the plasmid. The presence of the plasmid in a potential transformant was confirmed by Southern transfer of restriction digests of plasmid rninipreps and probing with the 1.4 Kb Xho I-Bam HI fragment of pKT230

PAGE 154

139 Table 21. Transfo rmati on in the Medaro Reservoir with plasmid DNA (9/30-10/2/87) 8hr 26 hr puol/L hr % CT DNA puol/L hr % CT DNAa,d 537 + 13 547 + 44 pKT230 603 + 37 112 1173"+179 CT DNAa,e 123 + 7 54 + 18 pKT230 43 + 3 33 12 + 7 CT DNAa, f 46 + 10 36 + 9 pKT230 56+ 20 122 12 + 10 CT DNAb, d 1978 + 25 388 + 353 pKT230 1797 + 556 91 758 + 558 CT DNAb,e 13 + 2 0 pKT230 2"+5 1 5 27 + 5 CT DNAb' f 8 + 3 14 + 8 pKT230 2 4 + 1 0 300 4 + 3 CT DNAc,f ND 116 + 31 pKT230 ND 41 + 12 a no added nutrients b Nutrients added to 0 .25 g/L peptone and 0 .05 g/L yeast extract c Nutrients added to 5 g/L peptone and 1 g/L yeast extract d No antibiotics added e Antibiotic concentrat i on was 100 kanamycin f Antibiotic concentration was 100 kanamycin and 100 streptomycin ND not determined CT DNA 214 22 33 195 >2700 29 35

PAGE 155

140 1234567 --Figure 25. Autoradiogram of Southern transfer of V. parahaemolyticus transformant. pKT230 transformants probed with Rib oprobe RNA probe made from the 1.4 Kb Xho I /Bam HI fragment of pKT230. (1) undigested pKT230; pKT230; (3) Xho I/Bam HI double digestion o f pKT230; ( 4 ) undigested DNA from aplasmid miniprep of a V. p arahaemolyticus transformant; (5) Bam HI digested t ransformant DNA; (6) Xho I/Bam HI double digestion of transformant DNA; ( 7) V parahaemolyticus wild type DNA.

PAGE 156

141 Table 22. Tampa Bay sediment transformation experimenta (6/29-7/1/88) 8hr pmol/gdw hr % CT DNA 25 hr pmol/gdw hr % CT DNA CT DNA pKT230 rronomer pKT230 mult imer 73. 7 + 5.7 + 15.2 80:6 + 15.5 81 109 65. 8 + 28.8 62;7 + 26 37.0 + 95 56 a Antibiotic concentration for all samples was 250 vg/ml kanamycin and 1000 vg/ml streptomycin Table 23. Sediment transformation, Cay Sal Bank (7125-ZT /88) pmoles /gr dry wt hr % CT DNA C T DNAa 790 + 58 :;;RIF chrorro 1268 + 336 160 CT DNAb 3868 + 4045 pKT230 rrono 5818 + 383 150 pKT230 mult 6904 + 1164 178 a Concentration of rifampin = 50 vg/ml b Antibiotic concentrations were 250 vg/ml kanamycin and 1000 vg/ml s trep tomyc in using a Riboprobe RNA probe (Figure 25) The restriction pattern of the transformant and the purified plasmid are identical. The slight difference between restriction patterns between pKT230 and the transformant appearing in Figure 25 was due to incomplete digestion of the pKT230 DNA by the restriction endonucleases There was no hybridization of the probe to the wild type strain. The plasmid was incorporated into the cell intact as there is no apparent difference in size between the parent plasmid and that isolated from the transformant. Several protocols were examined in attempt to increase the

PAGE 157

142 Table 24. Summary of environmental transformation data Sample a DNAb Number of Transformation c Location Treatments Indicated by TdR Bayboro (3/87) we PL 2 2 Bayboro ( 3/87) we CHR 2 1 Madard ( 4/87) we PL 3 2 Medard (4/87) we PL-MLT 3 1 Medard (9/87) we PL 4 3 Madard (3/88) we PL 2 2 Medar-d (3/88) Filter-PL 2 1 Offshore (5/87) we PL 3 2 Dr-y Tort (5/87) CSM PL 3 2 Cay Sa 1 ( 7 I 88) CSM PL 1 1 Cay Sal (7/88) CS1 CHR 1 1 Cay Sal (7/88) SED PL 1 1 cay Sal (7/88) SED PL-MLT 1 1 Cay Sal (7/88) SED CHR 1 1 Tampa Bay (6/88) SED PL 1 0 Tampa Bay (6/88) SED PL-:MLT 1 0 Total 31 21 a we = watercolumn ; CSM = cor-al surface micr-olayer-; SED = sediment b Type of transforming DNA used. PL = plasmid monomers; PL-MLT = plasmid multimer-s; CHR = chromosomal c Tr-ansformation indicated by [3H]thymidine incor-poration if greater than 150% of the ca l f thymus contr-o l in 21 of 31 tr-eatments tr-ansfor-mation fr-equency of this or-ganism with plasmid DNA (Table 25) The or-iginal tr-ansformants were isolated fr-om a filter-tr-ansformation assay. Tr-ansformation with RIF chromosomal DNA has been incr-eased by 1 to 2 or-ders of magnitude when transformation is conducted in sediment microcosms (G. Stewar-t personal communication). However, no incr-ease in plasmid transfor-mation frequency resulted when sediment columns were used with plasmid monomers ormultimers (Table 25) The majority of tr-ansfomation assays wer-e conducted using selection on 1 00 lJg/ml kanamycin and 250 lJg/ml str-eptomycin. However, this organism has no toler-ance to kanamycin at concentrations as low as 5 lJg/ml (data not

PAGE 158

Table 25. Transformation frequencies parahaemolyticus DNA Type pKT230 mono Filter Trans ND Sed. Trans 3. 1 X 1 0-8 ND not determined pKT230 mult 0 .8..:.2.4 X 10..:.8 0 7 X 108 RIF chromo ND 1 6-2 o x 1 o-7 shown). Reduction of antibiotic by 50% failed to increase transformation frequencies in sediment transformation assays (data not s h own). Chromosomal DNA Transformation of V parahaemolyticus 143 Transformation with RIF chromosomal DNA was approximately 2 orders of magnitude greater than transformation by plasmid DNA (Table 25) The methods by which cells were loaded onto the columns did not significantly affect transformation frequency Cells which were not harvested prior to loading on the column transformed with a frequency of 2.0 x Cells whi ch were harvested and resuspended in nutrient-free media transformed at a frequency of 1 6 x regardless of whether they were resuspended in an equal arrount of media or concentrated 10fold (Table 25). Amplification and selection for transformants in a mixed population After sacrificing the sediment column, the inoculum from the sample Which received multimer DNA contained a mixed population of wil d type cells, transformants, and UNKS1 cells at concentrations of 1 5 x109, 10, and 230 cells/ml, repectively, based on plate counts. The results from the Southern blot hybridized with pKT230 is shown in Figure 26. There was no hybridization to any of the treatments which received calf thymus DNA o r plasmid monomers. However, in each of the

PAGE 159

144 1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Figure 26. Autoradiogram of Southern transfer of miniprep DNA from the liquid se lection/amplif icat ion transforrna tion experiment with the mixed bacterial population using pKT230. Southern transfer was probed with Riboprobe RNA pro be made from the 1 4 Kb Xho I/Bam HI fragment of pKT230. CT =calf thymus receiving samples; MONO = plasmid nxmomer receiving samples; MULT = plasmid multimer receiving samples A-F depict sample based on Fig 23. (1) undigested pKT230; (2) CT A; (3) CT B; (4) CT C ; {5) MONO A; (6) MONO B ; (7) MONO C; (8) MULT A; (9) MULT B ; ( 1 0) MULT C; (11) CT D; (12) CT E ; ( 1 3) CT F; ( 1 4 ) MONO F; ( 1 5 ) MUL T D ; ( 1 6 ) MUL T E; ( 1 7 ) MULT F; ( 1 8) undigested pKT230.

PAGE 160

145 treatments that received multimer DNA, there was hybridization of the probe to the sample indicating that transformation had occurred. Amplification of the transforming DNA was indicated by the differences in hybridization intensity. In the original 0, 12, and 24 hr samples, (I,II, and III, respectively) hybridization occurred at increasing levels between successive dilutions. Similarly, the dilutions made after 48 hr (IV, V, VI) showed much greater hybridization than did the flasks from which they were sampled. These results indicate that successive dilutions resulted in an increasing selection for transformants and amplification for the plasmid over time within the mixed population samples. Thymidine incorporation to detect natural transformation in V. parahaemolyticus. In two experiments with a mixed population of V parahaemolyticus and UNKS1 [ 3 H]thymidine was used in attempts to indicate transformation (these experiments occurred before it was determined that '!._. parahaemolyticus was incapable of thymidine incorporation, see Chapter 5). During the selection and amplification of transformants using pKT230, thymidine incorporation was measured in exponentially growing cells in each of the final dilutions (Table 26). There was no apparent difference between either the calf thymus or monomer treatments while those receiving multimer DNA were slightly lower on average. After v. parahaemolyticus was transformed with pGQ3, the columns were sacrificed, washed with sterile ASWJP, and inoculated into antibiotic amended media. Plate counts indicated that the inocula for the calf thymus control and the plasmid amended samples had initial populations of 1250 and 545 UNKS1 cells/ ml, respectively. The number of transformants was too l o w to detect. Fifty-four hr after

PAGE 161

146 Table 26. Thymidine incorporation to detect natural transformation in V. parahaemolyticus: amplification and selection experiment pmoles thymidine incorp/L hr a Sample D E F Average C.T. DNA Monomer 5198 + 166 5133 111 7000 + 178 5814 + 103 5720 + 336 5873 + 326 5973 + 9zr 5607 + 411 a Samples are as those described in Figure 23 Multimer 4497 + 248 1895 + 28 -3924 + 82 3439 + 1367 inoculation, the calf thymus control sample had a thymidine incorporation rate of 46. 3 prnoles/L hr while the plasmid amended sample had a rate of 3553.3 120.7 pmoles/L hr. After another 10 hr, the plasmid amended sample displayed visible cell growth as indicated by increased turbidity within the flask. Within an additional 24 hr, both flasks appeared saturated with bacterial cells. Further analysis of transformants indicated that increased thymidine incorporation was not due to transformants which were still incapable of thymidine incorporation (Chapter 5) but rather to different growth rates of the UNKS1 cells in the two flasks. Discussion Natural transfonnation has been examined in both environmental samples and bacterial isolates. Although no transformants were isolated from environmental samples, greater [3H]thymidine incorporation rates were recorded in samples that received transforming DNA when grown under selective pressure in approximately two-thirds of

PAGE 162

147 the different treatm:mts implying that transformation rray have occurred. Strongest results appeared in samples from water column and coral surface microlayer samples. Although sediment samples also demonstrated enhanced growth when receiving transforming DNA, the amount of increased growth was generally lower t han that observed in other environments These results may have been partially due to the increased difficulty in working with sedirent samples compared to liquid samples. Failure to isolate transformants from these samples may have been due to the very low frequency at Which transformation It is possible that transformants may not have been stable or that plasmids were taken up, transcribed, and translated, but not replicated. Another indication of transformation was the presence in several plasmid a.mended water colurm samples of bacterial cells which were elongated several fold over their normal size (data not shown). These cells were not observed in any calf thymus DNA control treatment or in samples which received plasmid DNA without antibiotic selection. Apparently, a phenotypic change caused by the plasmid and selected for by the antibiotics was seen in these samples. In general, increased [3H]thymidine incorporation rates in plasmid arrended samples were not due t o the transforming DNA serving as a growth factor or nutrient source. In the absence of antibiotics, thymidine incorporation rates between calf thymus controls and samples receiving transforming DNA were usually equ ivalent. From the data presented in the tables it is apparent that there were large ambient antibiotic resistant bacterial populations in the rrajority of location sampled. Antibiotics were often added at concentrations an orde r of

PAGE 163

magnitude higher than that recommended for experiments with bacterial cultures (Mmiatis et al., 1982). To inhibit growth of bacterial populations from Tampa Bay sediments, kanamycin and streptomycin must be added at final concentrations of at least 0.5 and 1 0 mg/ml (data not shown). Even with the high concentrations of antibiotics employed in the transformation experiments, extensive growth occurred in the majority of calf-thymus controls after 24 hr. 148 Thirty heterotrophic marine bacterial isolates were examined for the ability to naturally transform with plasmid DNA. Only one, identified as Vibrio parahaemolyticus, appeared transformable using standard filter transformation assays. parahaemolyticus is a common marine pathogen which causes gastroenteritis usually associated with the ingestion of improperly handled or prepared seafood (Joseph et al., 1982). It is the leading source of food poisoning in Japan (Kaneko and Colwell, 1973) and the second leading cause of gastroenteritis in Calcutta (Sarkar et al., 1985). This organism is found throughout the world in estuaries and environments but rarely in offshore waters (Joseph et al., 1982). It is found associated with a variety of higher o r ganisms in the marine environment including plankton, fish, and shellfish (Joseph et al., 1982). v parahaemolyticus plays an important ecological role in estuaries such as the Chesapeake Bay where it digests chitin and mineralizes organic matter (Colwell et al., 1977). The distribution of this organism is independent of fecal coliform and enterococc i cell numbers and not related to proximity of sew age outfall (Shiaris et al., 1987). Although considered halophilic, it has been isolated infrequently from freshwate r water column and sediment samples The majority of organisms isolated from freshwater

PAGE 164

149 environments were associated with biological hosts, particularly the gastrointestinal tract of fishes (Sarkar et al., 1985). The exact pathogenic mechanism of V. parahaemolyticus is unclear. The rrost studied virulence factor has been the therrrostable direct hemolysin (TDH) responsible for the Kanagawa phenomenon (KP). TDH has been established as the cause of virulence in KP+ strains and almost all clinical isolates of V parahaemolyticus have been KP+. In contrast, less than 1% of the environmental isolates of this organism + h ave been KP (Jo seph et al., 1982; Sarkar et al., 1987). Although KP strains do not produc e TDH (Takeda, 1983) they have been reported to cause acute gastroenteritis and to produce 30 to 40% mouse lethality by virulence factors which are still unknown (Sarka r et al., 1987) A strong preference for transformatio n by rnultirneric plasmid forms was observed In an isolated experiment, transforrnants were detected from cells treated with an unfractioned plasmid preparation. In all other experiments, however, transformation was detected only in samples treated with rnultimers. This implies that the transforrnants resulting from unfractioned DNA may have arisen from naturally occurring multimers or concatarners in the plasmid DNA preparation. The absence of detectable transformants from monomers in the amplified samples (Figure 26) implied that rnultimeric forms were required. Southern transfer and hybridization studies indicated that plasmids were incorporated without apparent deletions ( Figure 25) as has bee n reported for another transformable bacterium, S pneumoniae (Saunders and
PAGE 165

150 frequencies for '!._. parahaemolyticus. No significant difference in transformation frequencies resulted with harvested and washed cells implying that this organism does not produce a competence factor as has been reported for Bacillus and Streptococcus ( Q:>odgal, 1982). The physiology of competence developrrent of '!._. parahaemolyticus remains unknown. Competence is often induced by changes in cell physiology associated with the shift from exponential to stationary growth phase (Stewart and Carlson, 1986). Apparently, transformation in V. parahaemolyticus is not effected by gradual or sudden (as caused by harvesting and resuspension in media ) onset of stationary phase. Transformation frequencies for V. parahaemolyticus with chrorrosomal DNA have been reported to increase 1 to 2 orders of magnitude when transformation was conducted in sedirrent microcosms and compared to filter transformations (G. Stewart, personal communication). This is similar to increased transformation of B. subtilis attached to sand grains reported by Lorenz et al. (1988). However, plasmid transformation frequencies for V. parahaemolyticus did not increase in sediment microcosms. In their review of V. parahaemolyticus, Joseph et al. (1982) reported a general sensitivity to chloramphenicol, gentamycin, kanamycin, nitrofurantoin, tetracycline, doxycycline, and streptomycin, but resistance to ampicillin, carbenicillin, clindamycin, colistin, erythromycin, and penicillin. t-blitoris et al. (1985) examined antibiotic resistance in 199 isolates of V parahaemolyticus from Indonesia and reported 92 antibiotic resistance patterns of which approximately half were represented by a single isolate. In general,

PAGE 166

151 this organism was found to be susceptible to chloramphenicol and resistant to lincomycin, nafcillin, and penicillin. Neither of the studies mentioned above reported on the incidence of plasmids and associated antibiotic resistance. Early reports found cryptic plasmids in several y_. parahaemolyticus strains (Guerry and Colwell, 1977; Twedt et al., 1981; Arei et al., 1983). The ability of Y._. parahaemolyticus to acquire plasmid encoded antibiotic resistance through conjugation from E. coli was reported by Guerry and Colwell ( 1977) Hayashi et al. (1982) reported the conjugal transfer and stable maintenance of a multiple antibiotic resistance encoding plasmid from y_. anguillarurn to V. parahaemolyticus. Arai et al. ( 1985 ) isolated a multiple antibiotic resistance encod ing conjugal plasmid from y_. parahaemolyticus which could be conjugally transferred to and remain stable in other strains of V parahaemolyticus, V alginolyticus, NAG vibrio, and E. coli. The natural transformation of v. parahaemolyticus with plasmid DNA reported here is important for several reasons. Although natural transformation by plasmid DNA of freshwater isolates of Acinetobacter calcoaceticus has been recently demonstrated (Rochelle et al., 1988), this study reports the first incidence of natural transformation by plasmid DNA reported of a marine isolate. Stewart and co workers (Zuklic et al., 1987; Stewart at al. 1988; Coughter and Stewart, in press) have reported natural transformation by homologous ch romosomal DNA in 14% of randomly selected culturable marine heterotrophs. In this study, only 1 of 30 (3%) rrarine isolates examined was transformable with plasmid DNA. This indicates, however, that natural transformation may be one source of horizontal gene transfer in the

PAGE 167

152 marine environment not previously recognized. Plasmid transformation by V. parahaemolyticus is also important because this organism is a pathogen. Antibiotic resistance is often encoded on plasmid DNA (Levy et al., 1981; Obrien et al., 1985; Baya et al., 1986; Frederickson et al., 1988) and the flux of antibiotic resistance in other systems is well documented (Levy et al., 1981; Beringer and Hirsch, 1984; Obrien et al., 1985; Davies, 1986; Baya et al., 1986; Levy and Marshall, 1988). This study deiJX)nstrates the ability of an organism which is one of the main causes of seafood poisoning in certain parts of the world to acquire antibiotic resistance through natural transformation which may create future difficulty in treating patients who have acquired gastroenteritis from this organism. The ability of V. parahaemolyticus to transform in sediment microcosms is important as this organism has been isolated from sediment samples in many locations (Joseph et al., 1982). In the Chesapeake Bay, interaction with sediment in the winter is essential for its survival (Kaneko and Colwell, 1973). The selection for transformants and amplification of transforming DNA has potential applications for further studies o f natural transformation. Sediment ecosystems may hold the greatest possibility for detecting natural transformation due to the large bacterial populations and the resistance of extracellular DNA to degradation (Maeda and Taga, 1974; Lorenz et al., 1981 ; Aardema et al., 1983; Lorenz and Wackernagal, 1987). Transformation has been reported to be enhanced by the presence of surfaces (Ste wart et al. 1983) and attachment to sand (Lorenz et al., 1988). However, selection fo r

PAGE 168

153 growth of transformants within sediment is difficult. Antibiotics may become bound to sediments and effectively reduce the concentration needed for the selection of transformants. Removing the cells from the sediment by vortexing reduces this problem. Successive dilution series continuously select for antibiotic resistant subpopulations, and eventually, transformants should make up a significant part of one subpopulation. The experiment with V. parahaemolyticus indicates that transformants which occur at frequencies as low as 10-g, may be selected for over time in the presence of other organisms. Theoretically, this technique would allow the detection of a single transformant if it were allowed to grow and propagate under selective pressure until it became a significant part of a sample s population. Liquid selection may increase the ability to detect transformation where standard methodologies might not and it is conceivable that organisms considered to be nontransformable may be capable of transformation, but at frequencies too low to detect by previous techniques. In the mixed population samples, only a single transformant colony was detected by plating during transformation with pKT230 and no transformants were detected by plating during transformation with pOQ3. However, successive selection resulted in strong amplification of the pKT230 and pGQ3 signals for hybridization studies. This typifies the problem faced when trying to detect natural transformation in the environment. Plate count methodologies also require that transformants be culturable. Since the majority of environmental bacteria are believed to be non-culturable, however, sensitivity of detecting transformants in the environment by plating may be greatly reduced

PAGE 169

An initial goal of this study was to determine whether C3 H]thymidine incorporation could be used as an indicator of natural transformation. In 21 of 31 transformation treatments conducted in 154 various environments, thymidine incorporation was significantly greater in treatments which received transforming DNA (Table 24) implying that transformation may have occurred Unfortunately, no conclusive proof (i.e. isolation of transformants and detection of specific gene sequences of the transforming DNA with molecular probing) that transformation occurred was obtained from any of the samples where [3H]thymidine incorporation indicated that transformation may have occurred. Although parahaemolyticus is a member of the minority of bacteria that cannot incorporate thymidine (Chapter 5), the data from transformation with this organism indicates that [ 3 H]thymidine incorporation data may be misleading. In one case, slightly lower thymidine incorporation rates in samples receiving plasmid DNA were recorded (Table 26) yet molecular probing verified that transformation had occurred When parahaemolyticus was transformed with pOQ3 in the presence of the unidentified antibiotic resistant organism UNKS1, intitial [3H]thymidine incorporation rates in the plasmid arrended sample were almost two orders of magnitude greater than the calf thymus control. These results were even more misleading since plate counts indicated that the calf thymus treatment originally had twice as many UNKS1 cells as did the plasmid amended sample, It might be expected, therefore, that the calf thymus control would produce higher thymidine incorporation rates due to its larger initial cell densisty. Molecular probing again verified that transformation had occurred but examination of isolated transformants indicated that even though they contained

PAGE 170

155 active thymidine kinase, they were still incapable of thymidine incorpo ration (Chapter 5). The different rates of incorporation were independent of transformation but were due to different growth rates of the UNKS1 cells in presumably replicate flasks with identical growth conditions. The use of thymidine incorporation as a means to detect natural transformation is subject to some of the same restrictions which apply when using the technique to estimate bacterial DNA synthesis or heterotrophic productivity. In the case of V. parahaemolyt icus, inability to incorporate thymidine due to the absence of a functional transport system prevented the use of the technique. In addition, variable growth rates between seemingly similar flasks also precluded accurate results. For reliable use of thymidine incorporation as a means to detect natural transformation, all of the organisms must be capable of thymidine incorporation and differences in growth rates between treatment and control samples must be due solely to pressures brought about by the selective agents (i.e. antibiotics) used to select for growth of transformants.

PAGE 171

CHAPTER 7: SUMMARY AND CONCLUSIONS The four major factors and assumptions associated with the use of [ 3HJthymidine incorporation were examined. Using standard i sotope dilution assays, the amount of dilution was found to be insignificant in eleven of fifteen samples taken in various l ocations in south west 156 Florida and the southeastern Gulf o f Mexico. In estuarine environments where dilution was significant, the thymidylate synthase inhibitor (FdU), inhibited intracellular isotope dilution to the sa.roo degree as did the addition of high concentrations of exogenous thymidine. This allowed the separation of intracellular and extracellular dilution and it was found that the majority of dilution was extracellular. In three different environments, an exponentially growing bacterial culture, eutrophic estuarine waters, and oligotrophic offshore waters, [3H]thymidine incorporation was found to consistently underestimate DNA synthesis by a factor of This underestimation was attributed to the failure of standard isotope dilution assays to account for all thymine bases incorporated into DNA indicating that isotope dilution assays are not applicable to in situ [3H]thymidine incorporation measurements. macromolecular labelling was found to be significant in all env ironrnents examined and t o fluctuate both temporally and spatially, although averages for all samples and locations were approximately 50%. The greatest range was observed in Bayboro Harbor

PAGE 172

157 and a seasonal pattern was observed during the first year of a two year study with the percentage of radiolabel incorporatd into DNA being greatest in the winter and lowest in the summer. Non-specific labelling began inunediately upon addition of [ 3 H]thymidine and the percentage of radioactive label incorporated into each of the three macromolecular fractions was constant with time. Labelling patterns did not fluctuate as a function of added thymidine concentrations or cell size fractions. The percent of radioactivity incorporated into DNA was most often inversely related to total thymidine incorporation rates indicating that rapidly growing cells were less likely to incorporate radioactivity into DNA. No other consistent correlation was observed between radioactivity incorporation and a variety of microbial biomass, activity, and nutrient parameters. FdU was shown to inhibit non-specific labelling under certain conditions. The use of two dihydrofolate reductase inhibitors, amethopterin and trimethoprim, and a comparison of and [6-3H]thymidine indicated that non-specific labelling does not occur by a reversal of thymidylate synthase or a demethylation reaction but via catabolism of thymidine beginning with thymidine phosphorylase. Approximately 90% of the marine heterotrophic bacterial isolates surveyed were capable of thymidine incorporation. The organisms that could not incorporate thymidine also transported thymidine poorly and lacked thymidine kinase activity. One organism that did not incorporate thymidine was later transformed with a plasmid containing an E. coli tdk. Although enzyme assays revealed high thymidine kinase activity in the transformants, these cells still failed to incorporate thymidine into DNA or transport thymidine into cells. These results

PAGE 173

indicated that while the vast major-ity of cultur-able narine heter-otr-ophic bacter-ia wer-e capable of thymidine incor-poration, the inability to incor-por-ate thymidine in some or-ganisms was due to the absence of thymidine tr-ansport, thymidine kinase, orboth. Natural tr-ansformation was investigated in sever-al envir-onments using both chr-orrosomal and plasmid DNA. In two-thir-ds of the 158 experimental tr-eatments, [ 3 H]thymidine incor-por-ation r-ates wer-e gr-eater' in samples r-eceiving tr-ansforming DNA than in controls when bacter-ia wer-e gr-own in the pr-esence of antibiotics, resistance for which was encoded by the tr-ansforming DNA. These r-esults indicate that tr-ansfor-mation may have occur-r-ed, although tr-ansformants wer-e never isolated fr-om any of these samples. In an effor-t to develop a culture system with which the envir-onmental data could be compar-ed, a strain of V par-ahaemolyticus was transfor-med with the br-oad host range plasmid pKT230 and its der-ivative, This was the fir-st known instance of natur-al tr-ansformation of a mar-ine bacter-ial isolate by plasmid DNA. This or-ganism also tr-ansformed in the presence of anotherantibiotic r-esistant mar-ine or-ganism and techniques wer-e developed using selection in liquid media which would detect transfor-mation with greater sensitivity than could standar-d plate counting assays. [ 3 H]Thymidine incor-poration data fr-om envir-onmental tr-ansfor-mation exper-iments could not be compar-ed with results using V. parahaemolyticus since this organism was unable to incor-por-ate thymidine even when tr-ansformed with a thymidine kinase gene. The r-esults fr-om tr-ansformation studies of V. par-ahae molyticus in a mixed population also derronstr-ated that [3H]thymidine incor-por-ation r-esults could be misleading. In one case,

PAGE 174

159 thymidine in samples transfornation In another instance, the sample DNA had a of [ 3 H]thymidine two of magnitude than the calf thymus control. analysis that could not thymidine and that the different independent of but due to the other antibiotic present. In conclusion, the and assumptions may significantly affect [ 3 H]thymidine assays, depending on the sampled and how the data is to be The assumption that all within a sample capable of thymidine acceptable. may be however, which dominated by species unable to thymidine (e.g. Pseudomonas) that would the validity of this assumption The of [ 3 H]thymidine into than DNA is ubiquitous in aquatic samples. While estimates of DNA synthesis the of specific into DNA, comparisons of activity and can be made using total rates of [3H]Thymidine is to DNA synthesis in yet it DNA synthesis b y a consistent factor of The this is that isotope dilution analysis is not applicable to samples. The isotope dilution has most ambiguous. assays indicate that in the majority of isotope dilution is negligible and would become insignificant if high concentrations of exogenous thymidine were

PAGE 175

160 added. The direct comparison of DNA synthesis and thymidine incorporation indicated, however, that there are significant sources of thymine bases incorporated into DNA which cannot be accounted for by isotope dilution assays. The results presented in this study indicate that thymidine incorporation has many valuable applications. Caution must be advised, however, as to how results from [3H]thymidine incorporation may be interpreted. The importance of each of the four factors examined varies with how the technique is to be applied. This project has helped to clarify when these factors apply, yet conditions vary between environments and it is the responsibility of each individual researcher to insure that [ 3 H]thymidine incorporation is valid for the particular application. Detection of natural transformation is an example of an application of thymidine incorporation that may not be valid. This application would require that all of the organisms be capable of thymidine incorporation and that differences in incorporation rates must be due solely to the presence or absence of transformants growing under the selective pressures applied. As evidenced in the study of transformation of V parahaemolyticus, these requirements are not always met and [ 3 H]thymidine incorporation results were not valid indicators of natural transformation. These results may be one reason why no transformants were isolated from any of the environmental samples. Although [3H]thymidine incorporation could not be used for transformation assays with parahaemolyticus, the transformation of this organism by plasmid DNA raises many questions. This is the first

PAGE 176

161 report of natural transformation by plasmid DNA of a marine isolate. Little is known about the development of competence, optimal conditions for transformation, or the transformat i on mechanism of this organism Although transformation by rnultimeric plasmids was preferred, it is unknown whether rnultimeric forms are required or whether marker rescue would facilitate plasmid transformation of V parahaemolyticus. This organism is a pathogen and the results presented here may lead to further studies of horizontal gene transfer and antibiotic resistance acquisition by natural transformation of pathogenic bacteria. The sediment microcosm experiments have provided the basis for further studies of natural transformation in environmental sediments Extensions of these experiments would be to determine if V. parahaemolyticus is transformable in en vironmental sediments and whether ambient sediment bacterial populatio n s are transformable. Techniques developed in the microcosm studies allow the detection of natural transformation at frequencies that are too low to detect by plating and may provide the best means to detect transformation in the environrrent

PAGE 177

162 LITERATURE CITED Aardema, B. W., M G Lorenz, and W. E. Krumbein. 1983. Protection of against enzymatic inactivation. Appl. Microbial. 46: 417-420. W. L., J W. Bendler, and J. K. Setlow 1981. Plasmid in Haemophilus influenzae. 145 : 1099-1101. C. C. 1987. Antibiotic of bacterial of lakes. System Appl. 9 : 169-172. Anderson, P and T Fenchel 1985. flagellates inseawater samples. by Linnol. 30: 198-201 Aria, T T. Ando, A Kusakabe, and A. Md. Uaalh 1983. Plasmids in Vibrio parahaemolyticus isolated in Japan and Bangladesh with special to different distributions. Microbial. Irnrnunol. 27: 1021 1029. Aria, T H Harnashirna, and H Hasegawa. 1985. Isolation of a new plasmid a strain of Vibrio Immunol. 29: 103-112 Auling, G., H. Prelle, and H Dickman. 1982. of into DNA of and the of kinase Eur. J. Biochem. 121: 365..;370. Azam, F T. Fenchel, Field, J.S. Gray, L .A. Meyer-Reil, and F Thi ngstad 1983. The ecological role of water-column microbesin the sea. Ecol. Progress Ser 1Q: 257-263 Bagdasa rian, M R Lurz, B Ruckert, F C H Frankl i n, M M J Frey, and K N Timmis. 1981. Specific purpose plasmid cloning vectors rr. Broad host high copy number, RSF 1 010 derived and a host-vector system gene cloning in Pseudomonas. Gene 16: 237-247. Bale, M J J C Frey, and M J Day. 1987. Plasmid transfer between strains of Pseudomonas aeruginosa on membrane filters attached to stones. J. Gen. Microbial. 133: 3099-3107 Barany, F., M. E Kahn, and H. 0 Smith 1983. and integration of DNA in Haemophilus influenzae

PAGE 178

transformation. Proc. Nat. Acad. Sci. USA 80: 7274-7278. Baskin, Y. 1987. Testing engineered microbes in the field. ASM News 53: 6 11617: Baya, A. M., P. R. Brayton D. J. Grimes, E Russeck-Cohen, and R R. Colwell. 1986. Coincident plasmids and antimicrobial resistance in marine bacteria isolated from polluted Atlantic Ocean samples Appl. Environ Microbial. 51: 1285-1292. Beck, C.F. J.C. Ingraham, J Neuhard, and E Thomassen. 1972. 163 Matabolism of pyr imi dines and pyr imidine nucleosides by Salmon e lla typhurium. J. Bacterial. 1 10: 2 1 9 228. Bell, R. T. 1986. Further verification of the isotope dilution aproa'Ch -for estimating the degree of participation of [ H]thymidine in DNA synthesis in studies of aquatic bacterial production. Appl. Environ. Microbial. 52: 1212-1214. Bell, R. T., and I. Ahlgren. 1987. Thymidine incorporation and m i crob ial respiration in the surface sediment of a hypertroph i c l ake. Limnol. Oceanogr. 32: 476-482. Bell, R. T., G M Ahlgren, and I. Ahlgren. 19833 Estimating bacterioplankton production by measuring [ H]thymidine incorporation in a eutrophic Swedish lake. Appl. Environ Microbial. 45: 17091721. Bensi G;, A Iglesias, u canosi, and T A. Trautner. 1981. Plasmid transformation in Bacillus subtilis The significance of partial homology between plasmid and recipient cell D NAs. 1'-bl. Gen. Genet 184: 400-404. Beringe r J E., and P. R. Hirsch. 1984. Genetic adaptation to t h e environment. Therole of plasmids in microbial ecology. In: M o Klug and c. A Reddy (ed .), Current perspectives in microbial ecology American Society for Microbiology, Washington, D.C. Bethesda Research Laboratories. 1985. BRL M13 cloning/di deoxy sequencing instruction manual. Bethesda Research Laboratories, Gaithersburg Md. Bishop, D. P F Entwistle, I R. cameron, C. J. Allen, and R D Possee. 1988. Fie l d trials of genetically engineered Baculovirus insectisi des, p. In M Sussman, C H. Collins, F A Skinner, and D. E; Stewart-TUll {eds.), The release of genetically engineered microorganisms. Academic-Press, New York. Biswas G; D., K L Burnstein, and P F. Sparling. 1986. Linearization of donor DNA during pl asmid transformation in Neisseria gonorrhoeae. J Bacterial. 168: 756-7 6 1. Biswas, G. D J F Graves, T. E. Sox, F. C Tenover, and P. F. Sparling. 1982; Marker rescue by a homolo gous recipient plasmid

PAGE 179

during transformation of gonococci by a hybrid Per plasmid. J Bacteriol. 151: 77 82. Biswas, G. D., and P. F. Sparling. 1981. Entry of double stranded deoxyribonucleic acid during transformation of Neisseria gonorrhoeae. J. Bacteriol. 145: 638-640. Bjornsen, P. K. 1986. Automatic determination of bacterioplankton biomass by image analysis. Appl. Environ. Microb ial. 5 1 : 1199-1204. 164 Bratbak, G. 1985 Bacterial biovolume and biomass estimations. Appl. Environ. Microb io l. 4 9: 1 488-1 4 93. Budrnan, D R., and A. B Pardee. 1967. Thymidine and t h ymine incorporation into deoxyribonucleic acid: inhibition and repression of thymidylate phosphorylase of Escherichia coli. J Bacterial. 94: 1 546-1550 --Gampell, A 1957. Synchronization of cell division. Bacteriol. Rev. 21 : 263 272 . Canosi, U., A Iglesias, and T. A Trautner. 1981. Plasmid transformation in Bacillus subtilis: effects of insertion of Bacillus subtilis DNA into plasmid pC194. Mol. Gen Genet. 181: 434-440. Ganosi, U., G. Morelli, and T. A. Trautner. 1978. The relationship between molecular structure and transformation efficiency of some S. aureus plasmids isolated from B. subtilis. Mol. Gen. Genet 166: 259 267. -Garlson, 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. -Carm:xiy J. M., and R M. Herriott. 1970. Thymine and thym idine uptake by Haemophilus inflaenzae and the labelling of deoxyribonucleic acid. J. Bacterial. 101: 5255 30. Chauvat, F., Astier, F. Vedel, and F. Joset-Espardellier. 1983. Trans format ion in the cyanobactettum Synechococcus R2: Improvement of efficiency; role of pUH24 plasmid. Mol. Gen. Genet. 191: 39-45 Chen, M. s., and W. H. Prussoff. 1978. Thymidine kinase of Escherichia coli. Meth. Enzynol. 5l: 354-360 Church, G M., and W. Gilbert. 198 4. Genomic sequencing. Proc. Nat Acad; Sci. USA 81: 1991-1995. Colwell, R R., P. Brayton, D J. Grimes, D. Roszak, S Huq, and L Palmer. 1985. Viable but non-culturable Vibrio cholerae and

PAGE 180

165 related pathogens in the environment: implication for release of genetically engineered microorganisms. Bio/Technology l: 817-820. Colwell, R. R J Kaper, and S. W. Joseph. 1977 V ibrio cholerae, Vibrio parahaemolyticus, and other vibrios: occurrence and distribution in Chesapeake Bay. Science 198: 394-396 Colwell, R. R C. Somerville, I. Knight, and W Straube. 1988. Detection and monitoring of genetically engineered micro-organisms, p 47-60. In M Sussman, C. H. Collins, F. A Skinner, and D. E. Stewart-Tu l l (-eds.), The release of genetically engineered microorganisms. academi c -Press, New York. Commerford, S. L., and D D. Joel. 1979. Iododeoxyuridine administered to mice is de-iodinated and incorporated into DNA primarily as thymidylate Biochem. Biophys. Res. Comm. 86: 112-118 Contente, S., and D. Dubnau. 1979a Marker rescue transformation by linear plasmid DNA in Bacillus subtilis. Plasmid 2: 555-571 Contente S., and D. Dubnau. 1979b 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 G J Stewart. 1988 Genetic exchange in the environment. Antonie van Leeuwenhoek 54: in press. Coveney, M F and R. G Wetze l 1 988 Experimental evaluation of conversion factors for the [ 3 H]thymidine incorporation assay of bacterial secondary production. Appl. Environ. Microbial. 54: 201 8-2026. Crawford, M. 1987. californai field test goes forward Science 2 36: 511. Danenburg, P. V., and A Lockshin 1981. Fluorinated pyrimidines as tight-binding inhibitors of thymidiylate synthetase. Pharmac Ther 13: 69-90 Danner, D B., R A Deich K. L Sisco, and H 0 Smith 1980 An sequence determines the specificityofDNA uptake i n Haemophilus transformation. Gene ll: 311-318. Davies J. 1986. Mi crob io l. Rev. A new look at antibiotic resistance. 39 : 363-371. FEMS Davis, B D. 1987. Bacterial domestication: underlying assumptions Science 235: 1329-1335 DeFlaun, M F and J H. Paul. 1986. Hoechst 33258 staining of DNA in agarose gel electrophoresis. J. Microbial. Meth. 2 : 265-270 DeFlaun, M F J.H. Paul, and D. Davis 1986. Simplified meth o d for

PAGE 181

dissolved DNA determination in aquatic environments Environ. Microbial 52: 654-659 Appl. 166 DeFlaun, M. F., J. H. Paul, and W. H Jeffrey. 1987. Distribution and rrolecular weight of dissolved DNA in subtropical estuarine and oceanic environments. Mar. Ecol. Progr Ser. 38: 65-73. de Vos, W. M., G. Venema, U. Canosi, and T. A. Trautner. 1981. Plasmid transformation in Bacillus subtilis: fate of plasmid DNA. Mol. Gen. Genet 181 : 424-433 Doran, J. L., W H Bingle, K. L. Roy, K. Hiratsuka, and W. J Page. 1987. Plasmid tran s formati on of Azotobacter vinelandii OP. J Gen. Microbial. 133: 2059-2072. Drahos, D. J., G. F Barry, B C Hemmng, E J. Brandt, H D Skipper, E. L.Kline, D. A. Kluepfel, T. A. Hughes, and D. T. G:>oden. 1988.Pre-release testing procedures: US field test of a lacZY-engineered soil bacterium, p. In M Sussman, C H Collins, F. A. Skinner, and D. E Stewart -full-reds. ) The release of genetically engineered microorganisms. Academic Press, New York. Ducklow, H.W., and S.M. Hill. 1985. Tritiated thymidine incorporation and the growth of heterotrophic bacteria in warm core rings. Limnol Oceanogr 30: Ducklow, H.W. D.A. Purdie, P.J. LeB. Wiliams, and J.M. Davies 1986. Bacterioplankton: A sink for carbon in a coastal marine plankton community. Science 232: 865-867 Ducklow, H.W. D A Purdie, P.J. LeB. Williams, and J.M Davies. 1987. Bacteria: Link or sink? Science 235: 88. Fallon, R. D., and S Y. Newell 1986. Thymidine incorporation by the microbial community of standing dead Spartina alterniflora. Appl Environ Microbial. 52: Findley, S. E G., J. L. Meyer, and R. T. Edwards. 19843 Measuring bacterial production via rate of incorporation of H-thymid ine into DNA. J Microb. Meth. 2 : 57-72 Forsdyke, D.R. 1968. Studies of the incorporation of [5-3H]uridine during -activation and transformation of lymphocytes induced by phytohaerrogglutination. Biochem J 107: Forsdyke D. R 1971. Application of the isotope dilution principle the analysis factors affecting the incorporation of [ H]uridine and [ H]cytidine into cultured lymphocytes Biochem J 125: 721-732 Fox, J L 1988. Biotechnology alfresco. Bioscience 38 : 533-537 Frederickson, J. K., R. J. Hicks, S W. Li, and F J Brockman. 1988

PAGE 182

167 Plasmid incidence in bacteria from deep subsurface sedirrents. Appl. Environ. Microbiol. 54: 2926-2923. Freifelder, D. 1983. M:>lecular:Biology. Jones and Bar-tlett Publisher:s Inc ,Boston. Fry, J. C., and A. R. Davies. 1985. An assessment of methods for measur-ing volumes of planktonic bacter:-ia with particular reference to television image analysis. J. Appl. Bacteriol. 58: 105-112. Fuhrman, J .A., and F. Azam. 1980. Bacterioplankton secondar-y production estimates for coastal waters of British Columbia, Antar-ctica, and California. Appl. Envir-o n Micr:-obiol. 39: 1085-1095. Fuhr-man, J .A., and F. Azam. 1982. Thymidine incor-por-ation as a measure of heter-otrophic bacter:-ioplankton pr-oduction in marine surface water-s: Evaluation and field results. Marine Biology 66: 109-120. Fuhr:'man, J.A., H.W. Ducklow, D.L. Kir-chman, and G.B. McManus. 1986a Does adenine incor-por-ation into nucleic acids measure total microbial pr-oduction? Limnol. Oceanogr. 31: 627636. Fuhrman, J.A., H.W. Ducklow, D.L. Kirchman, and G.B. McManus. 1986b. Adenine and total microbial pr-oduction: A reply. Limnol. Oceanogr. ]l: 1395-1400. Gar:-apin, A. c., F. Colber:-e-.:Garapin, M. Cohen-Solal, F. Hor:-odniceanu, and P; Kourilsky. 1981. Expr:-ession of herpes simplex virus type I thymidine kinase gene in Escherichia coli. Proc. Nat Acad. Sci. USA 78: 815-819. --Gealt, M. A., M. D. Chai, K B. Alper-t, and J. C. Boyer-. 1985. Transfer of plasmids pBR322 and pBR325 in wastewater fr-om labor-atory strains of Escherichia coli to bacter-ia indigenous to a waste disposal system. Appl. Environ. Microbiol. i2: General Accounting Office. 1988. Biotechnology. Managing the risks of field testing genetica1ly engineer-ed organisms GAO/RCED-88-.:27. Gener-al Accounting Office, Washington; D .C. Genther, F. J., P. 01atter:-jee, T. Bar-key, and A. Bourqum. 1988. Capacity of aquatic bacter-ia to act as recipients of plasmid DNA. Appl. Envir. Micr:-obiol. 54: 115-117. Ger:-mida, J. J., and G. G. Khachatourians. 1988. Tr-ansduction of Escher-ichia coli in soil. Can. J. MicrobioL 3 4: 190-193 Gillis, M., J. De Ley, and M. De Cleene. 1970. The deter-mination of molecularweight of bacterial genome DNA fr-om restriction r-ates. Eur. J. Biochem. 143-153. Glick, B. R., H. E. Br-ooks, and J.J. Pasternak. 1986 Physiological

PAGE 183

effects of plasmid DNA transformation on Azotobacter vinelandii. Can. J Microbiol. 32: 145-148. Golden, S., and L. Sherman. 1 984. Optima l conditions for genetic transformation of the cyanobacterium Anacystis nydulans R2. J Bacteriol. 158: 36-42. Goodgal, S H 1982. DNA uptake in Haemophilus transformation. Ann. Rev. Genet. 16! 169-192. Graham, J B., and C A Istock. 1978. Genetic exchange in Bacillus subtilis in soil. Mol ec Gen. Genet. 166: 287-290 J B., and C A Istock. 1979. Gene exchange and natural selection cause Bacillus subtilis to evolve in soil culture. Science 204: 637-639. Griffith, F. 1 928 The significance of pneumococcal types J Hyg. 27:113-159 . Gromkova, R., and s Goodgal. 1981. Uptake of plasmid deoxyribonucleic acid by Haemophilus. J Bacteriol. 146: 79-84. Guerry, P., and R R. Colwell 1977. Isolation of cryptic plasmid deoxyribonucleic acid from Kanagawa-positive strains of Vibrio parahaemolyticus. Infect. Irnmun ...!...: 328..:.334. 168 Hagstrom, A., U. Larsson P. Horstedt, and S. Normak. 1979. Frequency of dividing cells, a n e w approach to the determination of bacterial growth rates in aquatic environments Appl. Environ Microbiol. 30 : 805-8 1 2 Hanson, R. B., and H. K. Lowery. 1983. Nucleic acid synthesis in oceanic microplankton from the Drake Passage, Antarctica: Evaluation of steady..:.state gro wth. Mar. Biol. 73: 79-89 Hayashi F., K. Harada S. Mitsuhashi, and M Inoue 1982. Conjugat i o n of drug-reststance plasmids from Vibrio anguillarum to Vibrio parahaemolyticus Microbiol. Immunol. 26: 479-485 Hofmeister, J., M Israeli..:.Reches, and D Dubnau. 1983. Integration of plasmid pE194 at multiple sites qn the Bacillus subtilis ch romosome. Mol. Gen. Genet 189: 58-68. Holben, W. E., J. K Jansson B K. Chelm, and J M Tiedje. 1 988 DNA probe method for the detection of speciftc microorganisms in the soil bacterial community. Appl. Environ Microbiol. 54: 703-.711. Hollibaugh, J T 1988. Limitat i ons of the [3H]thymidine method for estimating bacterial p roductivity due to thymidine metabolism Mar. Ecol Progr. Ser. 19-30. Iglesias, A., G Bensi U canosi and T A Trautner. 1981. Plasmid

PAGE 184

169 transformation in Bacillus subtilis. Alterations introduced into the recipient-hom::>logous DNA of hybrid plasmids can be corrected in transformation. Mol. Gen. Genet. 184: 405-409 Jeffrey, W.H., and J H Paul. 1986a. Activ ity of an attached and Vibrio sp. as measured by thymidine incorporation, p-iodonitrotetrazolium reduction, and ATP/DNA ratios. Appl. Environ Microbiol 51: 150-156 Jeffrey, W.H., and J H Paul. 1986b. Activity measurements of planktonic and microfouling communities in a eutrophic estuary. Appl Environ. Microbiol. 51: 157-162. Jordon, M., and G E Likens. 1980. Measurment of planktonic bacterial production in an oligotrophic lake. Lirmol. Oceanogr. 25: 719-:732. Jordon, M., and B .J. Peterson. 1978. Sulfate uptake as a measure of bacterial production Limnol. Oceanogr 23: 146-150. Joseph, S W., R R. Colwell, and J B Kaper. 1982. Vibrio parahaemolyticus and related halophilic vibrios. CRC Crit. Rev. Microbiol 10: Jyssum, S., and K Bovre 1974. Search fo r thymidine phosphorylase, nucleoside deoxyribotransferase and thymidine kinase in Moraxella Acinetobac ter, and allied bacteria. Acta ?athol. Microbiol. Scand. Sec. B. 82: 57-66. Jyssum, S., and K Jyssum and dTMP by Neisseria of labelled material. 683-691. 1970. Uti lization of thymine thymidine, meningitidis 1 Growth response and uptake Acta ?athol. Microbiol. Scand. Sec B 78: Kaneko, T., and R. R. Colwell. 1973. Ecol o gy o f Vibrio parahaemo lyticas in Chesapeake Bay. J. Bacteriol. Karl, D M 1979. Measuremen t of microbial activity and growth in the ocean by rates of stable ribonucleic acid synthesis. Appl. Environ. Microbiol 38: 850-860. Karl D .M. 1981. Simul taneous rates of ribonucleic acid and deoxyribonucleic acid synthesis for estimating growth and cell division of aquatic microbial cormnunities. Appl. Environ Microbiol. 40: 802-810 Karl, D. M. 1982. Selected nucleic acid precursers in studies of aquatic microbial ecology. Appl. Environ Microbiol 44: 891-902 Karl, D.M., and P : Bossard . 1?85a. mic rob3a1 nucleic acidsynthesrs and spec1f1c growth rate by POu and [ H]adenine : Fie l d comparison Appl. Environ Microbiol. 50: 706-709 Karl, D.M., and P. Bossard 1985b. Measuremen t and significance of

PAGE 185

ATP and adenine nucleotide pool turnover in microbial cells and environmental samples. J. Microbial. Methods 1: 125-139. Karl, D M., C.D. Winn, and D.C.L. Wong. 1981a. RNA synthesis as a measure of microbial growth in aquatic environments. I. 170 Evaluation, verification, and optimization of methods. Mar. Biol. 64: 1-1 2. Karl, D.M., C.D. Winn, and D.C.L. Wong. 1981b. RNA synthesis as a measure of microbial growth in aquatic environments. II. Field applications. Mar. Biol. 64: 13-21. Kirchn'an, D., H. Ducklow, and R Mitchell. 1982. Estimates of bacterial growth from changes in uptake rates and biomass. Appl. Environ. Microbial. 44: Kit, S., M Kit, H. Qavi D. Trkula, and H. Otsika 1983. Nucleotide sequence of the herpes type 2 (HSV-2) thymidine kinase gene and predicted amino acid sequence of thymidine kinase polypeptide and its comparison with the HSV-1 thymidine kinase Biochlin. Biophys. Acta 741: 158-170. Kobayashi, Y., and w. G. Harris. 1978. LSC applications notes #1-30 New England Nuclear Corp., Kornberg, A. 1980. DNA replication. W.H. Freeman, San Francisco. Krambeck, C., H.-J. Krambeck, and J. Overbeck 1981. Microcomputer assisted biomass determination of plankton bacteria on scanning electron micrographs. Appl. Environ. Microbial. 42: 142-149. Lacks, S. 1979. Uptake of circular deoxyribonucleic acid and mechanism of deoxyribonucleic acid transport in genetic transformation of Streptococcus pneumoniae. J. Bacterial. 138: --. -Lark, K. G. 1969. Initiation and control of DNA synthesis. Ann. Rev. Biochem. 38: 569-604. LaRock, P.A., R.D. Lauer, J.R. Schwarz, K.K. Watanabe, and D.A. Wiesenburg. 1979. Microbial biomass and activity distribution in an anoxic, hypersaline basin. Appl. Environ. Microbial. 37: --Lee, s., and J A. Fuhrman. 1987. Relationships between biovolume and biomass of naturally derived marine bacterioplankton. Appl. Environ. Microbial 53: Levy, s. B., R. c. Clowes, and E. L. Koenig. 1981. Molecular biology, pathogenicity, and ecology of bacterial plasmids. Plenum Press, New York. Levy, s. B., and B. M. Marshall. 1988. Genetic transfer in the natura l environment, p. 61-76. In M. Sussman, C H. Collins, F.

PAGE 186

A. Skinner, and D. E. The release of genetically engineered microorganisms. Academtc Press, New York. Lewin, B 1987. Genes III. John Wiley and Sons, New York. Lindow, S E., G. R Knudsen, R. J. Seidler, M. V. Walter, V. w. 171 Lambou, P s Amy, D. Schmedding, V. Prince, and S. Hern. 1988. Aerial dispersal and epiphytic survival of Pseudomonas syringae during a pretest for the release of genetically engineered strains into the environment. Appl. Environ. Microbiol. 54: 1557-1563. Lindow, E., and N. J. Panopoulos. 1988. Field tests of recombinant ice Pseudomonas syringae for biological frost control in potato, p M Sussman, C. H. Collins, F A. Skinner, and D E. Ceds.), The release of genettcally engineered microorganisms. ADademic Press, New York. Littler, E., J. Zeuthen, A McBride, E Sorensen, K. Powell, J. Walsh-Arrand, and J. Arrand. 1986. Indication of epstein-barr virus-coded thymidine kinase. EMBO J. 5: Lopez, P M. Espinosa, D. L. Stassi, and S A Lacks. 1982. Facilitation of plasmid transfer in Streptococcus pneumoniae by chromosomal homology. J. Bacterial. 150: 692-701. Lorenz, M G., B Aardema, and W. Kramben. 1981. Interaction of rrarine sediments with DNA and DNA avai1abili ty to nucleases. Mar. Biol. 64: 225230. Lorenz, M. G., B. Aardema, and W Wackernagel. 1988. Highly efficient genetic transformation in Bacillus subttlis attached to sand grains. J Gen. Microbiol. 134: 107-112. Lorenz, M. G., and W. Wackernagel. 1987. Adsorption of DNA to sand and variable degradation rates of adsorbed DNA. Appl. Environ Microbial. 53: 2948-2952. Lowry, 0. H., N J Rosenbrough, A. L. Farr, and R J Randall. 1951. Protein measurement with the folin phenol reagent J Biol. Chern. 193: 265-275 Lovell, c R and A. Konopka. 1985a. Primary and bacterial production in two dimictic Indian lakes. Appl. Environ. Microbial. 49: 485-491. Lovell, c. R., and A. Konopka. 1985b. Seasonal bacterial production in a dimictic lake as measured by increases in cell number and thymidine incorporation Appl. Environ. Microbial. 49: 492-500. Maaloe, 0 and N.O. Kjeldgaard. 1966. Control of macromolecular synthesis. W.A". Benjamin, Inc., New York. Mach, P. A., and D J Gr'irres. wastewater treatment plant. 1982. RPlasmid transfer in a Appl. Environ. Microbiol. 44:

PAGE 187

1395-1403 Maeda, M and N. Taga. 1973. Deoxyribonuclease activity in seawater and sedimants. Mar. Biol. 20: 58-63. Maeda, M., and N. Taga. 1974. Occurrance and distribution of deoxyribonucleic acid hydrolyzing bacteria in seawater. J. Exp. Mar. Biol. Ecol. 14: 157-169. 172 Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. t-t>lecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N. Y. Martin, D. S., R. L. Stolfi, R. C. Sawyer, R. N. Nayaki, S. Spiegelman, C. W: Young; and T 1980. An overview of thymidine cancer 45: 1117-1128. Marx, J. L. 1987. Assessing the risks of microbial release. Science. 237: 1413-1417. Mathews, C K., T.W. North, and G.P. Veer Reddy. 1979. Multienzyme complexes in DNA precursor biosynthesis, p; 133-156. In: G. Weker (ed.), Advances in enzyme regulation. Pergamon Press,()xford. McPherson, P., and M. A. Gealt. 1986. Isolation of indegenous wastewater bacterial strains capable of mobilizing plasmid pBR325. Appl. Envir. Microbial. 51: 904-909. t-nlitoris, E., S. w Joseph, M. I. Krichevsky, W. Sindhuhardja, and R. R. Colwell. 1985. Characterization and dtstribution of Vibrio parahaemolyticus isolated in Indonesia. Appl Environ. Microbial. 50: 1388-:1394. Mollgaard, H., and J. Neuhard. 1983. Biosynthesis of deoxythymidine triphosphate, p. 149-201. In : A. Munch-Petersen (ed.), Metabolism of nucleotides, and nucleobases in microorganisms. Academic Press, N Y. Monheimer, R.H. 1978. Difficulties in interpretation of microbial heterotrophy from sulfate uptake data: laboratory studies. Lirnnol. Oceanogr. 23: 150-154 Moriarty, D J W. 1986. Measurement of bacterial growth rates in aquatic systems from rates of nucleic acid synthesis. Adv. Microb. Ecol. 9: 245-292. Moriarty, D.J .W., and P.C. Pollard. 1981. DNA synthesis as a measure of bacterial growth rates in seagrass sedirrents. Mar. Ecol. Progr. Ser. 2= 151-156. Moriarty, D.J.W., and P .C. Pollard. 1982. Diel variation of bacterial productivity in seagrass (Zostera capricorni) beds measured by rate of thymidine incorporation into DNA. Mar. Biol. 72: 165-173.

PAGE 188

Mottes M., G Grandi V. Sgaramella, U. Canosi, G Morelli, and T A 1979. Different specitic activities of the monomeric and oligomeric fo rms od plasmid DNA in transformation of B subtilis and E col i Mol. Gen. Genet 174: 281-286. A 1983. Metabolism of nucleotides, nucleosides, and nucleobases in microorganisms Academic Press, N Y Munch-Petersen, A and B. Mygind. 1983. Transport of nucleic acid precursers, p 259-305. In : A. Munch-Petersen (ed ) Metabolism of nucletides; nucleosideS, and nucleobases in microorganisms. Academic Press, N. Y. Munch-Petersen, A., B. Mygind, A. Nicolarsen, and N J Pihl. 1979. Nucleoside transport in cells and membrane vescicles from Escherichia coli K12. J. Biol. Chern. 254: 3730-3737 Murray, R E., and R. E Hodson. 1985. Annual cycle of bacterial secondary production in five aquatic habitats o f the Okefenokee Swamp ecosystem. Appl. Environ. Microbial. 49: 650-655. Mygind, B., and A. Munch-Petersen. 1975. Transport of pyrimidine nucleosides in ce lls of Escherichia coli K12. Eur J Biochem. 59: 365-372. National Academy of Sci ences. 1987. Introduction of recombinant engineered organisms in the environment: key issues. National Academy Press, Washington, D .C. 173 Neuhard J 1983. n ucleosides, p of nucleotides; Academic Press Utilization of preformed pyrimidine bases and 95-148 I n: A Munch-Petersen (ed .), Metabolism nucleosides and nucleobases in microorganisms. N Y Newell, S Y., and R.R. Christian. 1981 Frequency of dividing cells as an estimator of bacterial productivity. Appl. Environ Microbial 42: 23-31. Nierlich D P 1974. Regulatio n of bacterial g r o wth. Sc i ence 184: 104 3-,1050: Notani N.H. J K Setlow, D. Mccarthy, and Clayton 1981. Transformation of Haemophilus influenzae byplasmid RSF0885. J Bacterial. 148: 812-816. Novitzky, J. A 1986. Degradation of dead microbial biomass in marine sediment Appl. Environ Microbial 52: 504-509. Obrien, T. F., K H Mayer, H Kishi, E. Gillence, M Syvanen, and J D Hopkins 1985. Intercontinental spread of a new antibiotic resistance gene on an endemic plasmid. Science 230 : 87-88. Ogram, A v and G s Sayler. 1988. The use of gene probes in the

PAGE 189

rapid analysis of natural microbial cormnunities J. Industr. Microbial. ].: 281 Okazaki, R., and A. Kornberg. 1964. Deoxythyrnidine kinase of Escher"ichia coli I. Purification and some properties of the enzyme. J Biol. Chern. 239: 269-274 174 O'Morchoe S. B., 0 Ogunseitan, G. S Sayler, and R. v. Miller. 1988. Conjugal transfer of R68.45 and FP5 between Pseudomonas aeruginosa strains in a freshwater environment Appl. Environ. Microbial. 54: 1923-1929. otsuka, H., and S. Kit 1984. Nucleotide sequence of the marmoset herpesvirus thymidine kinase gene and predicted amino acid sequence of thymidine kinase polypeptide. Virol. 135: 316-330. Overbeck, J. 1979. Dark CO uptake-biochemical background and its relevance to in situ bacterial production. Arch. Hydrobiol. Beih. Erebn. Lirrnol-. 12: 38-47 Pastore, E. J., and M. Friedkin. 1962. The enzymatic synthesis of thymidylate II. Transfer of tritium from tetrahydrofolate to the methyl group of thymidine. J. Biol. Chern. 237: 3802-3810. Paul, J H. 1982. The use of Hoechst dyes 33258 and 33342 for enumeration of attached and planktonic bacteria. Appl. Environ. Microbial. 43: 939-944. Paul J. H., and D. J. Carlson. 1984. Genetic material in the marine environment: implication for bacterial DNA. Lirnnol. Oceanogr 29: 10911097. -Paul, J. H., M. F. DeFlaun, and H. Jeffrey. 1986. Elevated levels of microbial activity in thecoral surface microlayer. Mar. Ecol. Progr. Ser. 33: 29-40 Paul, J. H., M. F. DeFlaun, W. H. Jeffrey, and A. W. David. 1988. Seasonal and diel variability in dissolved DNA and in microbial biomass and activity in a subtropical estuary. Appl. Environ. Microbial. 54: Paul, J. H., w. H. Jeffrey, and M. F. DeFlaun. 1985. Particulate DNA in subtropical oceanic and estuarine planktoni"c environments. Mar. Biol. 90: Paul, J.H., W.H. Jeffrey, and M.F. DeFlaun. 1987. The dynamics of extracellular DNA in the marine environment; Appl. Environ. Microbial. 53: 170-179. Paul, J. H., and G. I. Loeb. 1983. Improved microfouling assay employ ing a DNA-specific fluorochrome and polystyrene as substratum. Appl. Environ. Microbial. 338-343 Paul, J.H., and B. Meyers. 1982. Fluorometric determination of DNA in

PAGE 190

175 aquatic microorganisms by use of Hoechst 33258. Appl. Environ. Microbial. 1393-1399. C., and T.D. Brock. 1982. Assessing biomass and production of bact"er"ia in eutrophic Lake Wisconsin. Appl. Environ. Microbial. 44: 203-218. Pollard, P.C., and D.J.W. Moriarty. 1984. Validity of the tritiated thymidine method for esirnating bacterial growth rates: measurements of isotope dilution during DNA synthesis. Appl. Environ. Microbial. 48: 1076-1083. Pomeroy, L.R. 1974. The oceans food web, a changing paradigm. Bioscience 24: 499-504. Porter, K. G., andY. S Feig. 1980. and counting aquatic The use of DAPI for identifying Limnol. Oceanogr 25: 943-948. Prornega Biotech. 1987. Technical Bulletins 002, 033, and 037. Promega Biotech, Madison, WI. Promega Biotech 1987. Prornega Notes 7. Prornega Biotech, Madison, WI. Ramsay, A. J. 1974. The use of autoradiography to determine the proportion of bacteria metabolizing in an aquatic habitat. J. Gen. Microbial. 80: 363-373. Razzell, W. E., and P. cassyhap. 1964. Substrate specificity and induction of thymidine phosphorylase in Escherichia coli. J Biol. Chern. 239: Riemann, B. 1984. Determining of natural assemblages of freshwater bacteria by means of H-thymidine incorporation into DNA: comments on methodology Arch. Hydrobiol. Beih. Ergebn. Lirnnol. 19: 67-80. Riemann, B. 1985. Potential importance of fish predation and zooplankton grazing on natural populations of freshwater bacteria. Appl. Environ. Microbial. 50: 187-193. Riemann, B., P. K. Bjornsen, S Newell, and calculation of cell of coastal measured incorporation of [ H]thymidine. 471 ,.476. R. Fallon. 1987. oar ine bacteria based on Li1111ol. Oceanogr. 32: Riemann, B., J. A. Fuhrman, and F. Azarn. 3982. Bacterial secondary production in freshwater measured by H-thymidine incorporation. Microb. Ecol. 8: 101-:114. Riemann, B., and M. Sondergaard. 1984. of diel rates of bacterial secondary production in aquatic environments. Appl. Environ. Microbial. 47: 632-638.

PAGE 191

176 Rivkin, R. B. 1986. Incorporation of tritiated thymidine by eukaryotic microalgae. J. Phycol. 22: 1937198. Rivkin, R. B., and M A. Voytek. 1986. Cell division rates of eukaryptic algae measured by tritiated thymidine incorporation into DNA: coincident measur'ements of photosynthesis and cell division of individual species of phytoplankton isolates fr'Om natural populations. J. Phycol. 22: 199-205. Robar'ts, R.D., R.J. Wicks, and L.M. Sephton. 1986. Spatial and tempora1 3 variations in bacterial macrorrolecular' labelling with [methyl-H]thymidine in a hyper'saline lake. Appl. Environ. Microbial. 52: 1368-1373. Rochelle, P. A., M. J. Day, and J. C. Fry. 1988. Occurrence, transfer, and mobilization in epilithic strains of Acinetobacter of plasmids capable of transformation. J. Gen. Microbial. 134: 2933-2941. Rollins, D. M., and R. R. Colwell. stageof Campylobactel:' jejuni natural aquatic env ir'onment. 531""'538. 1986. Viable but non-cultu!:'able and its !:'ole in survival in the Appl. Environ. Microbial. 52: Romanenko, V.I., J. Overbeck, and Y.I. Sorokin. Estimation of pr'Oduction of heterotrophic bacteria using C, p. 82-85. In: Y.I. Sorokin and H. Kadota (ed.), Techniques for the assessment of microbial production and decomposition in freshwaters. International Biological Programme Handbook no. 23. Blackwell Scientific Publishers, Oxford, England. S., and D. W. Visser. 1981. Uridine and uracil transport in Escherichia coli and transportin deficient mutants. Biochim. Biophys. Acta 646: 309""'319. Saito, H., and H. Tomioka. 1984. Thymidine kinase of bacteria: acttvi ty of the enzyme in act inomycetes and related organisms. J. Gen. Microbial. 130: 1863-1870. Saito, H., H. Tomioka, and S. Ohl
PAGE 192

177 transforms Streptococcus pneumoniae. Mol. Gen. Genet. 181: Saunders, C. W., and W R. Guild. 1981b. Pathway of plasmid trans format ion in pneurrococcus: open circular and 1 inear molecules are active. J. Bacterial. 146: 517-526. Saye, D. J., 0. Ogunseitan G. S Sayler, and R. V Mil ler. 1987. Potential for transduction of plasmids in a natural freshwater environment: effect of plasmid donor concentration and a natural microbial community on transduction in Pseudomonas aeruginosa. Appl. Environ. Microbial. 53: 987-995. saylor, G.S., M.S. Shields, E.T. Tedford, A. Breen, S W Hooper, K .M. Sirotkin, and J. W. Davis. 1985. Application of DNA-:DNA co lony hybridization to the detection of catabolic genotypes in environmental samples. Appl. Environ. Microbiol. 49: 1295-1303. Scavia D., and G A. Laird. 1987. Bacterioplankton in Lake Michi gan : dynamics, controls, and significance to carbon flux. Linnol. Oceanogr 32: 1017-1033. Scott, F. W., and D. Forsdyke. 1976. Isotope dilution studies of the effects of 5-'fluoro deoxyurid ine and hydroxyurea on the incorporation of deoxycytidine and thymidine by cultured thymus cells. can. J. Biochem. 5 4: 238-248 Scott, F. W., and D R. Forsdyke. 1980. Isotope dilution analysis of the effects of deoxyguanosine and deoxyadenosine on the incorporation of the thymidine and deoxycytodine by hydroxyurea-treated thymus cells. Biochem J. 190: 721-730. Servais, P., J. G Billen, and J. Vi ves..:.Rego. 1987. Determining [ H]thymidine incorporation into bacterioplankton DNA: improvement of the method by DNase treatrnent Appl. Environ. Microbiol. 53: 1977-:1979. Setlow, J. K., N. K. Notani, D. McCarthy, and N.-L. Clayton 1981. Transformati"on of Haemophilus influenzae by plasmid RSF0885 containing a cloned segment of chromosomal deoxyribonucleic acid. J. Bacteriol. 148: 804-811. Sharples F E 1987. Regulatio n of prod ucts from biotechnology. Science 235: Sherr, E.B. B .F. Sherr, and L.J Albright. 1987. Bacteria: Link or sink? Science 235: 88. Shiaris, M. P., A. C Rex, G. W. Pettibone, K Keay, P. McManus, M A Rex, J Ebersole, and E. Gallagher 1 987. Distribution of indicator bacteria and Vibrio parahaemolyticus in sewage polluted intertidal sedirnents. Appl. Environ. Microbiol 53: 1756-1761 Sjostrom, D.A., and D.R. Forsdyke. 1974. Isotope dilution analysis of rate limiting steps and pools affecting the incorporation of

PAGE 193

thymidine and deoxycytidine into cultured thymus cells. Biochem. J. 13 8 : 253-262. Smith, H. 0., D. B. Danner, and R. A. Deich. 1981. Genetic transformation. Ann. Rev. Biochem: 50 : Smits, J. D3 and B. Riemann. 1988 Calculation o f cell production from [ H]thymidine incorporation with freshwater bacteria. Appl Environ. Microbial. 54: 2213-2219. Stanley, J. T., and A. Kanopka. 1985. Maasurements of i n situ activities of nonphot osy n thetic microorganisms in aq ua t i c and terrestrial habitats. Ann. Rev. Microbial. 39: 321-346 Stewart, G.J., and C.A. carlson. 1986. transformation. Ann. Rev. Microbial. The biology of natural 40: 211-235 Stewart, G. J., C. A. Carlson, and J. L. Ingrham. 1983. Evidence for an active role of donor cells in natural transformi n g Pseudomonas stutzeri. J. Bacterial. 156: 30-35. Stewart, G. J., D. Cyr, and S Singalliano. 1988 The capacity for horizontal gene transfer by natural transformation in marine bacteria. First International Confere nce o n the Release of Genetically Engineered Microorganisms (REGEM 1) Abstract It 85A. 178 Stotzky, G., and H. Babi ch. 1986. Survivl of, and genetic t ransfer by, genet ically engineered bacteria in natural environments. Adv. Appl. Microbial. l]_: 93-138 Stuy, J H., a n d R. B Walker transfer in H. 1 986. Homology facilitated p l asmid Mol. Gen. Genet. 203 : 288 295. Southern, E. M. 1975. Detection of specific sequences aiiDng DNA fragments separat ed by ge l electrophoresis. J M::>l. Biol. 98: 503 5 17. Takeda, Y. 1 983 Therm:::>stabl e direct heiiDlysin of Vibrio parahaemolyticus Pharmacal. & Ther. 12: 123-146. Tangley, L. 1985. Releasing engineered organism s in the environment Bioscience 35: 470-473. Thorn, P. M., and R. M Ventullo. 1988. Measurement of bacterial growth rates in subsurface sed irrents using incorporation of tritiated thymidine into DNA. Microb Ecol. 16: 3-16 Trevors, J T., a nd M. E. Starodub. 1 987 R P l asmid transfer in non...:.sterile agricultural soil. Syst. Appl. Microbial. .2_: 3 1 2-315. Twedt, R M., D F Brown, and D. L Zink. 1981. Comparison of plasm i d deoxyribonucleic ac i d content, cu lture characteristics, and indices of pathogenicity aiiDng selected strains of Vibrio parahaemolyticus. Infect. Immun. 33: 322-325.

PAGE 194

179 Upton, C., and G. t-bFadden. 1986. Identification and n ucleotide sequence of the thymidine kinase gene of shope fibroma virus. J Virol. 60: 920-927. US Congress, Office of Technology Assessment 1988. New developments in biotechnology --field testing engineered organisms: genetic and ecological issues, OTA-BA,350. US Government Printing Office, Washington, DC. Van Es, F .B., and L.A. Meyer-Reil. 1982. Biomass and metabolic activity of heterotrophic marine bacteria. Adv. Microb. Ecol 6 : 111-170. Wagner, M J., J. Sharp, and W. Summers. 1981. Nucleotide sequence of the thymidine kinase gene of herpes simplex virus type I Proc. Nat Acad. Sci. USA 78: 1441-1445. Wahl, G. M., S L. Berger, and A. R Kimmel. 1987. Molecular hybridization of immobilized nucleic acids: theoretical concepts and practical considerations. Meth. Enzymol. 152: 399-407 Wallace, D C., and H. J Morowitz. 1973. Genome size and evolution. Chromosoma 40: 121-126. Weir, J P., and B Moss 1983. Nucleotide sequence of vaccinia virus thymidine kinase gene and the nature of spontaneous frameshift mutations. J. Virol. 46: 530-537. Werner, R 1971. Nature of DNA precursors. Nat. New Biol. 233: 99:--1 03. Willets, N 1985. Plasmids. In: J. Scaife, D Leach, and A Galizzi (ed.); Genetics of bacteria. Academic Press, New Y ork. Williams, D. Germany. 1987. Political problems of genetic engineering in West ASMNews 53 : Winn, C D., and D M Karl. 1984. Laboratory calibrations of the technique for measuring rates of RNA and DNA synthesis in marine microoganisms Appl. Environ Micobiol 4 7 : 835-842. Witzel, K P., and G Graf 1 984. On the use of differen t nucleic acid precursors for the measurement of microbial nucleic acid turnover. Arch. Hydrobiol 22: 59-65 Zar, J H. 1984. Biostatistical analysis. Prentice Hall, Inc., Englewood Cliffs. Zervos, P H., L. M Morris, and R J. Hellwig 1988. A novel method for rapid isolation of plasmid DNA. Biotechniques 238-242. Zuklic F W., T Kraxberger, and G J Stewart 1987 Detection of natural transformation in a coral-associated marine bacterium

PAGE 195

Abstract N-47, American Society for Microbiology 87th Annual t-t:!eting 180


Download Options

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


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.


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.


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.


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.