Evaluation of the Bacteroides fragilis phage assay as an alternative indicator of sewage pollution

Evaluation of the Bacteroides fragilis phage assay as an alternative indicator of sewage pollution

Material Information

Evaluation of the Bacteroides fragilis phage assay as an alternative indicator of sewage pollution
McLaughlin, Molly Rose
Place of Publication:
Tampa, Florida
University of South Florida
Publication Date:
Physical Description:
vi, 72 leaves : ill. ; 29 cm.


Subjects / Keywords:
Bacteroides fragilis ( lcsh )
Sewage -- Microbiology ( lcsh )
Water -- Pollution -- Environmental aspects ( lcsh )
Dissertations, Academic -- Marine science -- Masters -- USF ( FTS )


General Note:
Thesis (M.S.)--University of South Florida, 2000. Includes bibliographical references (leaves 67-72).

Record Information

Source Institution:
University of South Florida
Holding Location:
Universtity of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
027451283 ( ALEPH )
45801130 ( OCLC )
F51-00152 ( USFLDC DOI )
f51-152 ( USFLDC Handle )

Postcard Information



This item is only available as the following downloads:

Full Text


EVALUATION OF THE BACTEROIDES FRAGILIS PHAGE ASSAY AS AN ALTERNATIVE INDICATOR OF SEWAGE POLLUTION by MOLLY ROSE MCLAUGHLIN A thesis submitted in partial fulfillment of the requirement for the degree of Masters of Science Department ofMarine Science University of South Florida May 2000 Major Professor: Joan Bray Rose, Ph D


Graduate School University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Master s Thesis This is to certify that the Master's Thesis of MOLLY ROSE MCLAUGHLIN with a major in Marine Science has been approved by the Examining Committee on December 17 1999 as satisfactory for the thesis requirement for the Master's of Arts degree Examining Committee: Professor : Joan Br;y Rose, Ph D Merli'ber : John H Ph D. Member : V\lter ie J Harwood Ph D.


Acknowledgments I wish to extend my heartfelt thanks to the members of my committee Dr. Joan Rose Dr. John Paul and Dr. Jody Harwood for their patience and guidance throughout this process This project was funded in part through the Tampa Bay Healthy Beaches Project in association with the Pinellas County Department ofHealth, the Florida Environmental Protection Agency the Southwest Florida Water Management District and the Tampa Bay Estuary Program A special thanks to the Hillsborough County Environmental Protection Commission for providing monthly control samples and to Javier Mindez and Dr. Juan Jofre at the University of Barcelona for providing strain organisms and technical support throughout the project. Thanks to all the members of the Water Pollution Lab for your assistance and support


Table of Contents List ofTable ......... ... ... . ... . .............. ................... ..... ......... ..... .... . .... .......... ....... ..... .... ii List of Figures .... .................. .... ......... .............................. ..................... ..... ................. iii Abstract ... ..... ............ .......... ........ ........ .... .................................... ................................ iv Introduction ... .... ..... ... .... ........ . ............................. ... ........ .. ... .... .... . .... ............. ... ..... 1 Historical Background ...... .... .................. ... ... .............. .............. .... ............... ..... . 1 Public Health Concerns ................... ................. . .. . ......... .. ................ 2 Contamination of Recreational water ............ ........ ... .... .............. .... . ........ ......... ... 6 Indicator Concept ... ..... .... ........ ...... ........ ... ........... ... ... ...... ......... . ................ ........... 9 Survival of Microbial Pathogens and Indicators in the Envirorunent ...... ........ ...... .l3 Recreational Water Standards ... ........... .... ........ .... ... ........... ................. ..... ... .......... 14 Indicator Levels and Illness ............... ........................................................... ... . .... 16 Bacteroides fragilis, an Alternative Indicator. .......... . ...... ............... ....................... 17 Objectives ofThis Study ....... ........ ......... . ... ... ... .... ............... ... ............................... 20 Material and Methods ...... ......... ............... ... .... .... ... .... .... ....... ......... ............ .. ... ............. 21 Site Selection ........... ........ .... ... ... ........... ...................... .. ....... .... ...... ... .... . . ..... .... ... 21 Bacteria and Phage strains ...... . .. .... ............................................................ .... . ... ... 21 Quality Control Organisms ............... ... ......... ............. ............................................... 21 Sample Collection ............. .......................... ................. ....................... ....... ....... ..... 22 B. fragilis BPRMA Media and Reagents ... ............... .... ... . ....... . .... ...... ......... . ...... 22 Fecal Indicators ... .......... .... ............. ... ..... .... .... ... .............. ... .. .... ............ ........ ..... 22 B fragilis Overlay Protocol .... ................................. .... ....... ........ ........................... . 23 B fragilis Bacteriophage Enrichment Assay ........ ............ ..... ............. .... ........... .... 25 Statistical Analysis for Environmental Samples . .... .. . ... ................................... .... ... 25 Transmission Electron Microscopy .... .... ... ... ............... ....... .... ......... ... . .... ... ..... ..... 26 Phage Survival Study ..... .. ... ......... ... ..... ... ........ .... ............................ ... ......... ......... 26 Statistical Analysis of Survival Curves ......... .... ... ... ... ... .......... ..... .... ........... . . . ... .. .. .26 Domestic Wastewater Survey ............................................. ... ...... . .... .... . ...... .. ... ... 27 Results ..... ........... ........ . ....... ... ... ........ .... ..... ... ............. ....... .. ....... ............... .... ............ 28 Environmental Survey .. ... ..... ............. ........... ..... . .... ...... ...... .......... ............ .... ... . ... 28 Survival in Seawater ......... . ... . ................................. ... ... ........ .... .................. ........ . 48 Treatment Plant Indicator Levels ................ ............. ..... .............. .. ..... .... . ... ........... 52 Discus s ion and Conclusions . . . ........... ... ... ........... ................. .. . .... .... ........... ... ...... . 57 References .... ................ ........ .............. ... ................................... ................ ..... .............. . 6 7


List of Tables Table 1 Waterborne human pathogens .... ................. ...... ......... ........ .................. ...... 3 Table 2 Environmental samp le sites ............ ........... ................. .... ... ...................... ... 30 Table 3 Geometric Monthly Mean oflndicators ........................... .... .... ..... ........ ... .32 Table 4 B fragilis phage B40-8 (host ATCC 51477) ....................... ......... ........ .. ...... 36 Table 5 B fragilis phage B56-3 (host RYC2056) ....... ...... ... ... ....... .......... ........ . ..... 37 Table 6 Prokaryote numerical ranking of sites based on indicator levels ...... ...... ..... .42 Table 7 Binary Logistic Regression results ..... ........ . ............... ............ . .... .... .... ..... .45 Table 8 Decay rates ofbacteriophages ...... .... ......................................... ................... 49 Table 9 Indicator survey of wastewater treatment plants ........... .... ............ ................ 56 Table 10 Summary of B fragilis Phage Environmental Studies .. ................................ 59 Table 11 Summary of Sewage Indicator Studies .......................................................... 61 Table 12 Average indicator levels from treatment plant survey ..................... .... .......... 64 11


List of Figures Figure 1 Fecal/Oral Transmission ofPathogens ..... ... .................. .... ......... .... .... ... ..... 7 Figure 2 Proposed US EPA standards ... ........................................... ......... ................ .15 Figure 3 Map depicting sample sites TB 1 through TB22 .............. . .. ......... ............. .29 Figure 4 Summary oflndicator Levels for sampling sites .... ........ ......... ........ ... ... ... 33 Figure 5 Summary of C. perfring e ns levels for sample sites ...... .............................. .33 Figure 6 Example of seasonal variation of indicator levels at 5 sites ................ ..... .... . ..... ............ ..... .... .............. ........ ..... ........... .... ... ...... 34 Figure 7 TEM Micrograph ofB563 B.fragilis bacteriophage (host RYC2056) from stock culture ........ ... ............. ... .................................. 39 Figure 8 TEM Micrograph of B40-8 B fragilis bacteriophage (host ATCC 51477) from stock culture .................... ..... ............. ........ ....... 39 Figure 9 TEM Micrograph : TB 10 Little Manatee River-phage isolated using B. fragilis host RYC2056 ... ........ . ... ..... ...... ........... .............. .40 Figure 1 0 TB2 Alafia River phage isolated using B. fragilis host ATCC 51477 ............. ......... ... ... ...... .. .... ............. ............................. 40 Figure 11 Cluster o bservation showing % similarities of sample site based on geometric monthly averages of the indicator organisms .... ..... ..... ....... ... ....... ..... .... ... ... .. ..... ........................ 43 Figure 12 Scatterplot of B fragilis phage B563 and coliphage ....................... . . ...... .46 Figure 13 Scatterplot of B fragilis phage B56-3 and fecal Coliforms ..... ....... ..... . .... ..... ........ ........ .. . .................................. ............... 47 Figure 14 Survival curve at l0C ........ ..... ................................. .................... ......... ... 50 Figure 15 Survival curve at 20C ................... ............. ..... .... .... .... .............. . .... ..... ... 50 Figure 16 Survival curve at 30C ......... ............ .... ............ ............................. ... .......... 51 Ill


EVALUATION OF THE BACTEROIDES FRAGILIS PHAGE ASSAY AS AN ALTERNATIVE INDICATOR OF SEWAGE POLLUTION by MOLLY ROSE MCLAUGHLIN An Abstract Of a thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts Department of Marine Science University of South Florida May 2000 Major Professor: Joan Bray Rose Ph.D. IV


Traditional fecal coliform bacteria have been found to be inadequate for determination of the significance and sources of feca l contamination in ambient waters. In this study, the Bacteroides.fragilis bacteriophage strains B56-3 (host RYC2056) found in both animals and humans and strain B40-8 (host ATCC 51477 HSP40) found only in humans were evaluated as possible alternative indicators of fecal contamination. Water samples were assayed in the drainage basins flowing into Tampa Bay, Florida. Phage strain B56-3 was detected at 14 ofthe 22 sites, and strain B40-8 was detected at 6 of22 sampling sites. Sewage influent and effluent from area wastewater treatment plants were also sampled to determine a possible environmental source for the B. fragilis bacteriophage Phage B56-3 ranged from 1 19 x 104 to 1.11 x 105 PFU per 100 ml, while phage B40-8 was from 66.7 to 350 PFU per 100 ml in 100% of the sewage influent sampled. Of the 14 chl orinated effluent samples tested 3 tested positive for the presence of th e animal/human phage strain, B56-3. Levels in the effluent were < 10 PFU/100ml no B40-8 phage were detected. A survival study compared the survival ofB56-3, B40-8 and MS2 coliphage in seawater. MS2 coliphage decay rates were 0.1774 log10 day -1 at 10 C, but increased to 0 .926 4log10 dai1 at 20C and 3.0107log1o dai1 at 30C .. Bacteroides fragilis phage per sis ted much lon ger in the seawater compared to the coliphage and at l0C no appreciab le die-off occurred during the 30 days of the study. Phage B40-8 decay rate s were 0.1697 log10 dai1 at 20 C and 0.0846 log10 dai1 at 30 C B56-3 phage decay rates were 0.512log10 dai1 at 20C and 0.1687log10 day-1 at 30C. In conclusion, the Bacteroides fragilis phage found in sewage is susceptible to chlorine, but very per s istent in warm marine waters compared to traditional coliphage v


virus indicators. Phages B40-8 and B56-3 we r e found in 27% and 63% respectively, of the sites in Tampa Bay, indicating fecal imputs. Abstract Approved : ___ vi Maj r Professor: Joan Bray Rose, Ph.D. ofessor, Department of Marine Science Date Approved: __


Introduction Historical Background The connection between the quality of water we drink and the occurrence of illness has been recognized as far back in human history as the age of Hippocrates (460-354 B. C). Hippocrates in his writings spoke of health and water, suggesting water be boiled and strained before consumption as a preventative health measure .(!) (2) Ancient Egyptian wall paintings dating from the 15th century depict a filtering device used for water and beer. Descriptions of basic water treatment method s have also been found in ancient Sanskrit medical writings. (2) Although the exact relationship b etween water and disease was not fully understood until centuries later early civilizations understood the need for a clean water supply. The first epidemiological study showing a scie ntific link between water and disease came in 1854 with Dr. John Snow of London The famo us Broad Street Pump", and the outbreaks of cholera associated with using this particular water supply by the community lead Dr Snow to propose the theory that infectious disease could be caused by ingesting water contaminated with sewage (which had been fo und to be contaminating the well from a nearby draining cesspool (2)). This theory was later proven feasible by the discovery of the dise ase-germ theory, which proposed that microscopic organisms could cause disease in humans.(l-3) In 1884 Koch was able to isolate Vibrio cholerae the bacteria that causes cholera and in 1892 his studies in Hamburg and Altona showed a link between the quality of the water source and the incidence of illn ess associated with water contaminated by human 1


sewage. Cholera was not a major problem in the United States at this time, however, dysentery and typhoid fever (Salmonella typhi) outbreaks were a serious concern. During the Civil War ( 1860-65) the practice of disposing of human waste up stream of the camp's drinking water supply led to major outbreaks of dysentery among the troops (2) In 1887 Sedgwick, at Massachusetts Institute of Technology, explored methods of treatment for drinking water and wastewater. With the utilization of basic water treatment methods such as filtration, and with the advent of chlorination in 1907, typhoid cases in the United States dropped from 36 to 5 cases per 100,000 people between 1900 and 1928 .(4) Public Health Concerns Microbial pathogens are microscopic organisms that can infect and cause illness in the host organism. Pathogens can infect human s by ingestion, inhalation and body contact. (5) Diarrhea caused by human patho genic microorganisms in water is one of the leading causes of death in d eveloping countries today Children in these countries may experience 10 to 12 episodes of diarrhea per year.(4) There are four main types of human pathogens that can be found in water and wastewater ; bacteria viruses protozoan parasites and helminths. (See Table 1) 2


Table 1 Waterborne human pathogens Group Pathogen Disease or Symptom Viruses Enteroviruses (polio, echo coxsackie Meningitis,paralysis,rash fever myocarditis, respiratory disease,diarrhea Hepatitis A and E Infectious hepatitis Norwalk virus Gastroenteritis Rota virus Gastroenteritis Astrovirus Gastroenteritis Calicivirus Gastroenteritis Adenovirus Gastroenteritis,eye infections respiratory disease Reovirus Respiratory disease Bacteria Salmonella Typhoid, gastroenteritis Shigella Gastroenteritis,dysentery Campylobacter Gastroenteritis Vibrio cholerae Gastroenteritis, Cholera Yersinia enteroco l itica Gastroenteritis Escherichia coli (pathogenic) Gastroenteritis hemolytic uremic syndrome Legion ella Pneumonia Protozoa Nagleria Meningoencephalitis Entamoeba histolytica Amoebic dysentery Giardia lamblia Gastroenteritis Cryptosporidium Gastroenteritis Helminths Ascaris lumbricoides Ascariasis Trichuris trichiora Whipworm Necuter americanus Hookworm Adapted from Pollution Science, 1997 ( 4) Some of the bacterial pathogens include Salmonella typhi (typhoid fever) Shigella (acute gastroenteritis and fever), Vibrio cholerae (cholera), Campylobacter jejuni, Yersina enterocolytica, and pathogenic E. coli 0157 : H7. 3


Viruses the second type of microbial pathogens are obligate intracellular parasites measuring 20-200 nm and consisting of genetic material (RNA or DNA) inside a protein capsid. As the definition states they cannot replicate outside of a host cell, but are very stable in their environmental form and may persist for long periods outside of their host.( 4) Viruses that replicate in the intestinal tract of animals are referred to as enteric viruses These viruses are transmitted by a fecal-oral route that is they are excreted in the feces of infected individuals and ingested by the next host.(6) Enteric viruses can not only infect the cells of the intestine, but may migrate from the intestinal tract and replicate in other organs of the body such as the liver, heart, eye, skin and nerve tissue. An example of virus migration is Hepatitis A virus moving from the intestinal tract and infecting the liver.(4, 7) Most of these viruses are host specific; they can only replicate inside the human body and do not cause illness in other mammals Enteric viruses can include Hepatitis A and E rotavirus, norwalk and adenovirus. (4) Enteroviruses are a subgroup of enteric virus and include poliovirus coxsackie A and B virus and echovirus. Clinical symptoms can cover a broad range, including skin rash fever respiratory infections, eye and ear infections and acute gastroenteritis. ( 4 8) The sy mptoms of gastroenteritis can include nausea, vomiting, abdominal pain and diarrhea (9-11) Rotavirus is a major cause of pediatric diarrhea worldwide, and the leading cause of pediatric death in Africa, Asia and Latin America Norwalk virus is a major cause of acute gastroenteritis in the United States.( 4) Although only 50% of infected individuals will manifest clinical symptoms, some illnesses caused by these viruses can be quite severe. Coxsackie A and B viruses have been linked to heart disease 4


meningitis paralysis, insulin dependent diabetes and complications during pregnancy (4 8, 12) Wastewater may contain up to 140 different types of enteric viruses ( 4 13) Infected individuals are always present in a community, however, a significant percentage may carry viruses without manifesting any clinical symptoms Regardless of any outward symptoms, infected individuals may shed up to 106 to 1012 viruses per gram of feces. The concentrations of enteric viruses in wastewater averages 103 per liter in the United States, while Africa and Asia the level may reach 105 per liter.(7))(4) The third type of mircrobial pathogens are protozoan or single celled parasites which infect and replicate in the intestinal tract of humans .. Two examples are Giardia Iamblia and Crytosporidium.(4 12) These organisms have environmental stages termed cysts or oocysts which are excreted by infected individuals and are very resistant to environmental stress Helminths the fourth type of pathogen are worms that can also colonize the intestinal tract and shed eggs into the feces of infected individual s .(4) Both protozoan s and helminths can cause acute gastroenteritis and abdominal pain In cases of severe infections, helminths can actually cause obstruction in the small intestines. For normal healthy adults, exposure to microbial pathogens may not pose much of a health risk Mortality among adults can be 0 1% or less. However, the mortality rate can increa s e 10 to 100 time s in geriatric nursing home patients and may reach 50% in immunocompromised patients.(4, 12) Immunocompromised refers to an individual with an impaired immune system, and can include HIV patients, cancer patients and organ donor recipients. Also at increased risk are young children, the elderly and pregnant women. The majority of deaths in the United States linked to diarrhea occur in elderly 5


patients Gerba et al (1996) estimated that the irnmunocompromised children, the elderly and pregnant women make up almost 20% of the total population of the United States and this percentage will continue to increase In all infected persons, outcome from exposure depends on host immunity, nutrition, age of host and the type and strain of microorganism ( 12) Contamination of Recreational Water There are four main transport pathways where by microbial pathogens from wastewater may enter the environment and eventually infect humans These include exposure to contaminated drinking water, recreational water, shellfish and irrigation water and crops.(14) (See figure 1) The risk is related to the numbers of pathogens excreted, their survival through natural environmental processes and their ability to be tran spo rted throu g h the environment. 6


Figure 1 Fecal / Oral Transmission of Pathogens Oceans and Estuaries Shellfi s h Human and Animal Waste Rivers and Lake s Adapted from Wastewater Microbiology, 1994 (8) 7 Ground Water C rop s and Aero s ol s


Every day in third world countries, billions of gallons of poorly treated or untreated wastewater effluent are dumped into waterways and along coastlines, often near public recreational water areas.(8) Contamination of marine water, fresh water, and ground water can also occur from non-point sources, including runoff from land by storm waters, landfill leaching, agricultural runoff, wildlife and septic tanks. Septic tanks may be the most significant source of sewage contamination to ground water in the United States (7) Contamination to marine recreational waters can occur through sewage disposal into estuaries, contaminated river flow into coastal areas and marine sewage outfalls, whose distances can vary from 1 to 4 miles offshore.(&) It is estimated that around 4000 closings and advisories are issued each year in the United States for all types of recreational waters. The most common cause is microbial water pollution (15) Since 1953, epidemiology surveys have b een undertaken to study the effects on human health of utilizing contaminated recreational water areas.(13, 16, 17) Bathers may ingest between 1 0 and 15 ml of seawater or freshwater each time they bathe. Seawater especially is irritating to the mucous membranes and skin, which may interfere with the barrier effect of these organs in preventing infection.(16) Throughout these studies, the most common symptom associated with swimming and water quality has been gastroenteritis.(18-20) Respiratory problems reported include sore throat, cough and whee z ing Infections can also occur in the eye ear and wounds (9-11, 16) The standard incubation time is between 3 to 5 days for symptoms to manifest.(!?) Bacterial pathogens causing respiratory illnesses in swimmers have included Pseudomonas, Staphylococ c us aureus, Aeromonas, and Legionel/a. Ear and wound infections can result from exposure to Pseudomonas aeruginosa Vibrio parahaemolyticus and Vibrio 8


vulnificus Pseudomonas can come from wastewater, but all three of these organisms have environmental sources. Skin infections can be caused by Staphylococcus aureus, Aeromonas, Vibrio spp. and Mycobacterium.(8) Other bacterial pathogens, viruses and protozoan parasites previously described are also are of concern in contaminated recreational waters In addition to the effects of the illness, economic effects can result, including absence from work, medical treatment and depending on the severity of the illness hospital costs.(11) Fattal et al in 1991 (17) suggested that bathers themselves may be contributing to infectious pathogens found in heavily populated recreational areas, or those areas with a low flow, or flushing rate He also suggested using Staphylococcus aureus levels as an indication of bather density. In most of the studies, outcome from exposure depended on immunity and health of the individual, and the length and type of exposure.(! I) Epidemiology studies are extremely difficult to interpret, earlier studies have been found to have major design flaws. The exact level of exposure, accuracy in reporting illnesses, and the exclusion of children in most studies due to ethical considerations may effect the results of these studies.(16) In one ofthe few studies to include children, Alexander et al. (1992), did show an increased risk of gastroenteritis among children swimming in contaminated waters. (19, 20) Indicator Concept In 1891, scientists began to link two theories together, one; that water could cause disease, and two; that infected individuals excrete pathogens in their feces The search began for a detectable microorganism whose presence would indicate wastewater contamination of water. In 1892, Schardinger suggested using Bacillus coli (now termed 9


Escherichia coli), as an indicator of human sewage contamination. This bacteria had been isolated in 1885 from the human intestinal tract. When the American Public Health Association issued its first "Standard Methods of Water Analysis" in 1905, Bacillus coli was officially recommended as an indicator organism of water contamination. At the time, B. coli could not be distinguished from other bacterial normal intestinal flora such as Klebsiella, Enterobacter or Citrobacter in basic laboratory testing, so the term "Total Coliforms" was used to describe these organisms. The definition of coliforms was broadened in 1937 to include all aerobic and faculative anaerobic gram negative non spore forming bacilli that ferment lactose with gas production (3) This definition has continued to evolve and now includes "with 48 hours incubation at 37C". (8) Eventually, methods were developed to isolate Escherichia coli and Klebsiella pneumoniae, as well as other organisms, from the rest of the total coliform group. "Fecal Coliforms", defined as coliforms that ferment lactose at 44.5C, were believed to be a more specific indicator of wastewater contamination.(8) Fecal coliforms have been found to include mainly E. coli with percentages ranging from 60 to 90% depending on the source. (21) The group of bacteria known as fecal streptococci has been suggested as a possible alternative indicator organism. Fecal streptococci includeS. bovis, S. equinus, and S. avium E. faecalis, and E faecium. The streptococci were identified by Thiercelin in 1899 and further divided into serotypes, or Lancefield groups, in 1933 (22) Lancefield groups are based on the antigenic differences in the C carbohydrate of the cell wall of the bacterium.(3) A subgroup of fecal strep is the enterococci and include Enterococcus.faeca/is (Lancefield Group D Strep) and E. faecium. 10


E. faecium is only of fecal origin, but E. faecalis can be found in the environment. (23, 24) Enterococci do not reproduce in the environment, and survive better in seawater than fecal coliforms.(3, 8) The concentration in feces has been reported to be around 106 per gram and 10-100 times less than fecal coli forms in sewage They also are more resistant to environmental stress than fecal coliforms, and mimic viruses in regards to survival in seawater (24, 25) Clostridium perfringens is also found in wastewater at concentrations of 103 to 1 04 per 1 00 ml and has been evaluated as a possible bacterial indicator. Clostridium is an anaerobic gram positive spore forming rod that is very resistant to disinfection .(3, 6 8) These organisms can survive for extended periods of time due to these endospores, and therefore useful as a tool in determining past pollution events. They have also been utilized as a biological wastewater tracer, and in detecting sewage contamination of marine waters.(3, 8) The Bacteriophages, or viruses infecting bacteria, have also been used as indicators, particularly for indicating the presence of pathogenic viruses. These "phages are similar in morphology, structure, size and chemical composition to enteric viruses and are easier to detect with simple laboratory methods (8, 19, 26-28) Bacteriophages pose no human health risk, are found in greater numbers than enteric viruses in wastewater, and cannot replicate without a host.(26) They are more resistant to disinfection than the bacterial indicators, and give a better indication of the fate of enteric viruses in the environment. D'Herrelle first isolated bacteriophages from human fecal material and Guelin in 1948 was the first researcher to suggest bacteriophages as an alternative indicator of fecal pollution.(28, 29) 11


Coliphage are bacteriophage that infect E. coli and have been extensively studied as a possible indicator.(26, 30-35) However, it has been suggested that environmental replication and long term survival in tropical soils can occur (26) Two types of coliphage have been evaluated somatic and F specific. Somat i c coliphages absorb to and infect their host through the bacterial cell wall, while F-specific coliphages absorb to and infect their host through the pili.(28) F-specific coliphages are host-specific for E. coli only but only 1-3% ofhumans carry this phage. (3) There are several inherent disadvantages to using total and fecal coli forms as indicators of wastewater contamination Not all of the microorganisms included in the total coliform group are of fecal origin so their presence does not always indicate the presence of fecal contamination.(4, 21, 35, 36) Both total and fecal coliforms are found in the intestinal flora of all warm-blooded animals other than humans and may in fact, replicate in tropical environments.(4, 8 37) Fujioka and Byappanahalli(38) reported evidence of environmental replication of E. coli in tropical soils in Hawaii and a study by River et al(39) detected E. coli from pristine vegetation in a rain forest, providing evidence for environmental sources and long term survival outside of human hosts In addition to questions about the source of fecal coliforms, they also are less resistant to environmental stress, water treatment and disinfection than some pathogens .(8, 19) In general, the bacterial indicators previously mentioned give little indication of the viral load in water They are less resistant to environmental stress than viruses and protozoa do not give an accurate picture of their fate and transport, and their presence in water may not indicate human sewage contamination but may be from environmental, livestock or wildlife sources. Enteric viruses have been found in waters deemed 12


acceptable by bacterial standards.( 40) It is unlikely that an ideal indicator can be found, however, it may prove useful to employ combinations of indicators based on the type and use of the water in question.( 19) Survival of Microbial Pathogens and Indicators in the Environment Factors controlling survival of pathogens and indicator organisms in seawater include temperature, sunlight, moisture, pH, salts organic material and adsorption to particulates, species of virus and predation by protozoans, as well as other biotic factors.(4 8 ) Most enteric bacteria survive only a few days outside of the human body Viruses and protozoans persist much longer. ( 4) Total and fecal coliforms have shorter survival times in seawater than enteric viruses and are less resistant to inactivation by environmental factors ( 4, 26) E. coli and Salmonella were found to have great variation in survival rates depending on sunlight and predation ( 41) Therefore they are poor indicators of virus and protozoan fate in the environment.( 4) It is estimated that the amount of time for 90% (T90) ofthe total coliform population to die is 1-6 hours (42) The survival time of enteric viruses has been found to be highly variable, from 2 to 130 days in sea or estuary water, 2 to 188 days in rivers 5 to 168 days in tap water and 8 to 436 days in marine sediments.(4) Bacteriophages show similar survival patterns to enteric viruses in the environment.(43) Berry and Norton in 1976 (44) found that salinity and temperature did not have a significant effect on the survival of the T2 phage. Grazing by predators did have a significant effect on the phage's survival and this was also the conclusion of studies done by Borrego and Romero using coliphage.( 4 3) Metcalf and Stilles in 1967 found that enteric viruses persist up to 30 days in summer and up to 55 days in the winter 13


in environmental waters. When sewage contamination exists, the surviva l time increases to 14-3 5 days, and increases substantially in oysters ( 45) Recreational Water Standards Regulations for bathing waters were first considered in 1924 by the American Public Health Association. In 1936, total coliform limits of 1000 colony forming units per 100 ml was suggested. In the 40's and SO's, the United States Public Health Service started to study the relationship between illness and swimming and their work led to the 1968 recommendations from the Department of the Interior of fecal coliform limits of 200 cfu/1 OOml average density or that no more than 10% of samples collected within 30 days to exceed 400cfu/100ml.(15) These same criteria were adopted by the Environmental Protection Agency in 1976, and remained unchanged in the 1986 update. (5) (See figure 2) Recreational water guidance l e vels published by the Environmental Protection Agency in 1 986 based on the Cabelli epidemio lo gical study (1982) did not change the suggested coliform le vels, but did address alternative indicator levels. Limits of enterococci l evels for marine waters was recommended to be no more than 35 colony forming units per 100 ml, based on the geometric mean of 5 samples taken within a 30 day period The confidence limit depends on the type of exposure ( 46) 14


Figure 2 Proposed US EPA Standards FRESH-WATER enterococci E coli MARINE-WATER enterococci Criteria for indicator for bacteriological densities Single Sample Maximum Allowable Density ACCEPTABLE STEADY DESIGNATED MODERATE LIGHTLY SWIMMING STATE BEACH AREA FULL BODY USED FULL ASSOCIATED GEOMENTRIC (UPPER 75% CONTACT BODY GASTROMEAN C L ) RECREATION CONTACT ENTER! TIC INDICATOR (UPPER82% RECREATION RATE PER DENSITY C L ) (UPPER90% 1000 C L.) SWIMMERS 8 33 61 89 108 8 126 235 298 406 19 35 104 158 276 EPA Cri teri a for b at h i ng (F ull b o d y con ta ct) recrea ti on al waters Freshwater INFREQUENT LYUSED FULL BODY CONTACT RECREATION (UPPER95% C L ) 151 576 500 Based on statistically sufficient number of samples (generally not less than 5 samples equally spaced over a 30 day period), the geometric mean of the indicated bacterial densities should not exceed one or the other of the following: E. coli 126 per 1 00 ml or: Enterococci 33 per 100 rnl Marine water Based on a statistically sufficient number of samples (generally not less than 5 samples equally spaced over a 30 day period), the geometic mean of the enterococci densities should not exceed 35 per 100 ml. Adapted from Godfree et al (1990)(13) 15


Indicator Levels and Illness The exact relationship between indicator levels, enterovirus levels and illness in bathers is difficult to quantify. In the New York City study (1973-1975) by Cabelli et al, (1982), enterococci exhibited the best correlation with swimming associated illness, followed by E. coli. Fecal coliforms showed very poor correlation with illness. These findings were supported by Ferley et al 1989.(3, 21) E. coli showed a good correlation with gastroenteritis, but only in fresh water (24) According to Fleisher in 1991, the original EPA study (Cabelli, 1986) had design flaws in methods of chosing participants and interpreting the data, and should not be used to evaluate enterococcus levels Fleisher stated that the risk in the Cabelli study based on enterococci has been overestimated ( 46) He also states in 1996 that in compiling the 11 previous epidemiology studies, the lack of complete data is a primary concern Most of these studies only included acute gastroenteritis and may not have included the use of all alternative indicators. Only acute gastroenteritis has been show to have a mathematical model of dose-response, and Fleisher argues against using a single indicator. In 1996 he evaluated five indicator organisms and found only enterococci correlated with gastroenteritis among the bathers. The suggested upper limit according to this study is 32 cfu/1 OOml. (18) Most studies argue against using only one indicator organism to evaluate the risk associated with a particular recreational area,(18, 21) and most agree that current standards do not adequately assess risk of exposure to recreational waters.( 4 7) Many of the areas in question are multiple use areas, that is they are not only used for recreational purposes, but may in fact by utilized as source water for drinking, fishing and shellfish 16


harvesting, and receiving areas for treated wastewater effluent. The use of environmental m o deling to calculate risk of a recreational area may be more useful than microbial testing alone Bacteroides fragilis, an Alternative Indicator In recent years Bacteriodes fragilis bacteriophage has been evaluated as a possible alternative indicator of sewage pollution The host, Bacteriodes fragilis is an anaerobic bacterium found in the normal intestinal flora of humans and mammals It is not considered pathogenic, however it is an opportunistic pathogen and is the most common anaerobic isolate in human wound infections. There are 114 different species of the bacteria B fragilis is the most common isolate found among the Bacteroides spp., but may only comprise 0 5% of the total culturable human fecal flora, although it may be in greater numbers than E. coli (49,56,162) Allsop et al, 1985 found 1 78 x 1 03 to 1 99 x 105 per 100 ml of B fragilis bacteria in polluted waters in the United Kingdom, 8.1 x 101 to 4 24 x 104 per 100 ml i n lakes and reservo i rs. ( 48) The morphology of Bacteriodes fragilis bacteria is non-motile gram negative rods with rounded ends measuring 0 5 to 0.8 microns in diameter to 1 5 to 9 microns long They are pleomorphic, often exhibiting vacoules While they are non encapsulated when colonizing the gut, greater than 80% will become encapsulated when transferred to the blood stream or to wounds. This response may occur during the inflammatory process ( 49) The colony appears gra y and opaque when grown on nutrient agar Their growth is enhanced by the presence of bile they are resistant to kanamycin and penicillin and are producers (49 50) Recent clinical isolates have s hown resistance to clindamycin cephalosporin and chloramphenicol. (49 51) The 17


bacteria's decay rates in the environment are higher than E. coli or S. faecalis, according to Fiksdal et al (1985). (52) The phage of B .fragilis belongs to the bacteriophage group Siphoviridae and contains double-stranded DNA. These somatic phages have icosahedral heads exhibit flexible tails, and are often observed in star-shaped clusters when viewed under an electron microscope. (25, 53, 54) According to Lasobras et al (1997), bacteriophage with flexible tails are often the most resistant to environmental stress.( 53) B .fragilis phage is only found in 10-13 % of human fece s studied, so their numbers tend to be low in the environment.(25, 28) The bacte riophage and their host bacterium are strict anaerobes and do not replicate in the environment. B .fragilis bacteriophage are more resistant chlorination that S. faecalis, E. coli and some enteroviruses.(8, 26, 37) They also display a positive correlation with the levels of enteroviruses and rotaviruses (Jofre et al, 1989). Grabow and Jofre 1995 in South Africa found the B40-8, or HSP40 (host ATCC 51477) B .fragilis phage strain in 13% of humans tested, but not in animal s or birds. Somatic coliphage were found in 54% of humans and in 50-60% of animals and F-specific coliphages in 26% of humans and 60-90% in animals (26, 27, 55) B40-8 B .fragilis phage was found to be highly specific for human sewage, but natural variations exist in the gastrointestinal flora due to diet, stress and environment. (27, 56-58) S hoop et al, 1990,(59) found l evels of B .fragilis bacteria to be 5 to 9.8 x 104 colony forming units per ml in untreated sewage influent, while Allsop et al 1985 ,( 48) found le ve l s of 2.5 x 104 to 2.53 x 106 cfu per 100 ml in sewage. Researchers in France found 4.4 x 103 MPN per 100 ml in treated sewage and 4.4 x 104 MPN per 100 ml in untreated sewage.(60) Tartera et al 1989 found 5.3 x 103 cfu per 100 ml in sewage samples tested 18


compared to 1.2 x 106 per 1 00 ml average coliphage levels in sewage.( 61, 62) B. fragilis phage are not as abundant as coliphage in polluted waters.(55, 63) Environmental numbers can be highly varied depending on the source water Polluted river water levels of B.fragilis were 6.7 per 100 ml and 9.06 x 102 per 100 ml for coliphage levels (62) Lucena et al (1996)(64) found 7 perlOO ml to 5.3 x 103 per 100 ml in polluted river water, 1.2 x 102 per 100 grams in sediment and 2-3 per 100 ml in ground water Armon and Kott (1995)(65) found 1.97 to 10. 7% of drinking water samples tested were positive for the presence of B fragilis phage. Araujo et al ( 1997)( 66) found waters receiving recent sewage imput showed levels of 2.3 x 103 per 100 ml, waters receiving intermediate sewage imput showed levels of2.3 x 101 per 100 ml, and waters with persistently low levels of pollution averaged 1 67 per 100 ml and was only detected in 21. 9% of the samples tested. B fragilis phage showed no replication in the environment and demonstrated decay rates similar to human enteric viruses, coliphages and poliovirus. (25, 67) Their persistence in seawater is similar to Hepatitis A virus.(8, 68) B .fragilis phage was also show to be highly resistant to disinfection A study by Bosch et al in 1989 showed that in tapwater carrying 2-3 mg/1 residual chlorine levels with 15 minutes exposure time, B. fragilis phage were more resistant than poliovirus type 1, coliphage f2, E coli enterococci and simian rotavirus. The phage also showed higher resistance in sewage containing 20-24 mg/1 of residual chlorine with a contact time of 15 minutes F2 coliphage showed the best resistance to UV radiation, with B fragilis phage showing similar decay rates to poliovirus and rotavirus. (28) Chung and Sobsey (1993) showed that B.fragilis survived better than enterovirus in lab conditions at 5 and 25 C. (68) Jofre 19


et al (1995) found B .fragilis phage to be more resistant to treatment processes than either somatic or F-specific coliphages, or other bacterial indicators. (69, 70) Objectives of This Study In order to evaluate the potential for the B .fragilis phage assay for use as an indicator of fecal contamination in sub-tropical waters, this research focused on three main areas. I) The study of a variety of environmental waters for the presence of the B..fragilis phage. This included marine and freshwater, polluted and unpolluted areas, compared to traditional and alternative indicators. (fecal coliforms enterococci, Clostridium, and coliphage) 2) The survival and persistence of B..fragilis phage and coliphage in seawater at various temperatures 3) Domestic wastewater survey to determine normal background levels of B .fragilis phage compared to fecal coliforms, enterococci and coliphage 20


Materials and Methods Site Selection Twenty two sites were chosen as part of the Tampa Bay Healthy Beaches Project (West Central Florida) and used as sites for the envirorunental survey of the Bacteroides fragilis phage assay evaluation. The sites were chosen to represent the major drainage basins leading into Tampa Bay, and to provide several different water types, including marine, fresh, urban and rural impacted, and low intermediate and high pollution areas Three wastewater treatment plants were chosen in Hillsborough and Pinellas counties in the Tampa Bay region of West Central Florida to evaluate B fragilis phage levels in domestic sewage for our geographic region of the United States. Bacteria and Phage Strains The B fragilis bacteriophage B56-3 (ho st: RYC2056) and B40-8, or HSP40 (host: ATCC51477) and their corresponding host bacteria were kindly provided by Javier Mindez and Dr. Juan Jofr e at the University of Barcelona. Quality Control Organisms Enterococcus faecalis (ATCC 19433) was used as a positive control for the enterococci indicator assay. C lostridium perfringens (ATCC 1312 4) was used as a positive control for Clostridium indicator assay. E. coli (ATCC 15597) was used both as a positive control for E. coli and as the host bacterium. 21


Sample Collection Grab samples were collected in sterile 1 L plastic bottles and placed on ice for transportation to the lab. Samples were processed within 8 hours of collection B. fragilis BPRMA Media and Reagents All reagents and media used in the detection of B fragilis were made according to the International Standards ISO protocol CD 10705-4 (ISO/TC 147 / SC 4WG 11 N36), 1998. Fecal Indicators For each bacterial indicator assayed, volumes of the water sample were filtered through a 0.45Jlm pore size membrane filter (Osmonics) using a 47mm Gelman filter funnel fitted to a vacuum manifold. Sample volumes were determined by the fecal contamination level at each site. The filter s were then placed on the appropriate media as described below for each individual bacterial indicator assay. Fecal coliforms were enumerated according to the Standard Methods for Examination of Water and Wastewater APHA, 1989 (71) Water samples were filtered as described above and placed on mFC agar plates (Difco ) Plates were then incubated for 18 to 24 hours at 44. 5 C in a water bath. The dark blue colonies were counted as fecal coliforms. Enterococci were enumerated using Method 1600, USEPA (72) Water samples were filtered as described above. The filters were placed on mEl agar plates (Difco) and incubated for 18 to 24 hours at 41 C Those colonies exhibiting a blue halo wer e counted as enterococci. 22


E. coli were enumerated by taking those plates positive for fecal coliforms, transferring the membrane filter to EC with MUG media (Difco ), and incubating for an additional 24 hours at 3 7C. Colonies that fluoresced under UV light were counted as E. coli. C. perfringens_were enumerated by filtering water samples as described above. The filter was then placed on mCP agar plates (acumediaBaltimore Maryland) and incubated anaerobically in GasPakjars (BBL GasPak, Becton Dickinson) for 18 to 24 hours at 45C. Yellow colonies that turned pink or red when exposed to ammonium hydroxide fumes were counted as C. perfringens. Coliphage were enumerated according to the Standard Methods for Examination of Water and Wastewater, APHA, 1989. A 1ml aliquot of the water sample was added to a 1 ml aliquot of a log phase E. coli host bacterial culture in a tube of melted soft TSA agar and overlayed onto a TSA plate. The agar was allowed to solidify, and the plate was incubated for 18 to 24 hours at 37C Each sample was assayed using 10 replicate plates. Phage concentration of the sample was calculated by using the number of plaques that appeared on the bacterial lawn of each plate. B. fragilis Overlay Protocol The overlay protocol used in this study was as stated in the ISO protocol CD 1 0705-4 (ISO/TC 147/SC 4WG 11 N36), 1998 The media u sed to grow the Bacteroidesfragilis host contained peptone, tryptone, yeast extract, sodium chloride L-cystein, glucose, magnesium sulfate and calcium chloride in the base broth Before use, hemin, disodium carbonate and the antibiotics kanamycin monosulfate and nalidixic acid were added to the base broth and the broth was brought to room temperature. The pH of the media was 23


adjusted to the range of 6.3 to 7.3 with hydrochloric acid. (Instructions for media preparation are found in the ISO protocol listed above) The B fragilis host was grown to log phase (equivalent to 108 host cells per ml of broth) using 1 0 ml screw -top sterile glass culture tubes filled to the top with broth to provide an anaerobic environment and incubated at 3 7C. The time required for the cells to reach log phase was 4 to 5 hours. To a melted tube containing 2.5 ml of prepared B fragilis soft agar, 1 ml of a log phase host culture and 1 m1 of the water sample at room temperature was added. After gently mixing to avoid bubble formation, the sample was poured onto the surface of a BPRMA plate and swirled to distribute the agar equally. The agar was allowed to solidify and then the plate was inverted and incubated for 18 to 24 hours at 3 7C anaerobically in a GasPak jar. Plaques appearing on the bacterial lawn were considered B fragilis bacteriophage. If replicates of 1 0 are used for each sample, the limit of detection for this assay is 1 0 PFU/ 1 OOml. Araujo et al, 1993 demonstrated that background flora and suspended solids have the potential to interfere with plaque determination in the overlay method. (73) Lucena et al, 1995, noted interference with the presence/absence procedure due to background flora and turbidity. (63) Tartera et al, 1992, compared the B fragilis phage recovery of several different type of filters and filter protocols. The s tudy s recommendation for samples containing a high background flora included pre-filtering the water sample before assaying using a 0.45)lm PVDF filter (Millipore). Pretreatment of the filter with 3% beef extract was not found to significantly increase the recovery of the phage with PVDF filters. Recovery rates using the suggested protocol averaged 63.11 + / -15%. (74) 24


B.fragilis Bacteriophage Enrichment Assay One hundred ml of pre filtered seawater (0.45 J.Lm Millipore PVDF filter) was added to 100 ml of double-strength BPRMA broth in a sterile 250 ml glass flask with a screwtop lid. Both were prewarmed to at least room temperature. Thirty ml of a log-phase culture of either RYC2056 or ATCC 51477 (HSP40) B fragilis host bacterium was added and the flask was filled to the top with regular strength BPRMA broth to provide an anaerobic environment. A positive and negative control flask were used for each host. Both controls used 100 ml of a NaCl and peptone dilution buffer as the sample, and 1 J..Ll of a phage stock culture dilution was added to the positive control flask. All flasks were incubated at 3 7C for 48 hours. A 1 ml aliquot was taken from each flask and placed in a 1.5 ml sterile J..L-fuge tube with 0.4 ml of chlorofom1 and vortexed. The tubes were spun in the micro-centrifuge at 3000 rpm for 5 minutes To a tube containing 2 5 ml of melted BPRMA soft agar, 1 ml of log phase host culture was added and overlayed on BPRMA agar plates For each sample, 1 J.Ll was dotted in triplicate on the surface of the appropriate host lawn without disturbing the surface top agar The plates were allowed to solidify and were inverted and incubated overnight at 37C for 24 hours in a GasPak anaerobic jar. Samples were considered positive for the presence of the bacteriophage if clearing was noted on the lawn. All samples positive for the enrichment assay were repeated using 10 ml of pre-filtered sample and the standard overlay procedure Statistical Analysis for Environmental Samples All indicator data was log transformed and the averages used to calculate monthly geometric means The cluster observation analysis of the level of indicators at the 25


sampling sites was obtained using the MINIT AB statistical software program (Mini tab, Inc ) Cluster observations are used to classify observations into groups Binary logistic regressions were performed using MINT AB to compare the indicator levels with the presence of the B. fragilis phage. Binary logistic regressions are used to perform logistic regressions on data sets involving presence/absence with one or more predictors. Transmission Electron Microscopy One samples of a phage dilution were dotted onto formvar-coated copper EM grids and negatively stained with 2% uranyl sulfate. Phage Survival Study One hundred ml of filtered-sterilized (Millipore PVDF filter) seawater from the Gulf Pier at Ft. DeSoto St. Petersburg was placed in sterile 250 ml flasks (Salinity:35 ppt, water temp 26C) and pre-incubated at the appropriate temperature for 24 hours This allowed the seawater to be at the appropriate experimental temperature before the addition of the bacteriophages The temperatures used in the experiment were 1 0 C,20 C and 30 C Dilutions of B56-3 and B40-8 were added to the flasks at each temperature to achieve a final phage concentration of 1 06 per ml. Flasks were incubated in the dark at the appropriate temperatures Each flask was sampled once a day for 12 days, then at 18 and 24 days. Aliquots were diluted and assayed using the standard overlay protocol. Each count was done in duplicate Statistical Analysis of Survival Curves The survival curves were analyzed according to Yates et al, 1985(75). Phage concentrations at each sampling day were log transformed and plotted against the corresponding day of the experiment. A best fit linear regression line was applied to the 26


curve to achieve the R2 value and the slope ofthe line. The slope was used to calculate the decay rate of each phage using the formula -[(Log10PFU)dai1]. Domestic Wastewater Survey Influent and effluent samples were taken from three different domestic wastewater treatment plants in the Tampa Bay area of West Central Florida. Each treatment plant was sampled on two or three different days Both influent and effluent samples were pre filtered through a 0.45 micron Millipore PVDF filter to remove background flora. Influent samples were assayed using the standard overlay method in triplicate for both B fragilis phages and MS2 coliphage. Effluent samples were also assayed using the standard overlay method for B.fragilis phages and MS2 coliphage (10 replicates), as well as the enrichment procedure for both strains of B fragilis phage. In addition, all pre filtered samples were assayed for fecal coliforms, enterococci, and Clostridium perfringens using the methods previously described 27


Results Environmental Survey One of the objectives of this project was to determine if the Bacteroides fragilis phage assay would be applicable as an alternative indicator in our geogra phic location. Marine water samp les were analyzed for the presence of two B. fragilis phages, the human strain B40-8 (host ATCC 51477) and the animal/human strain B56 3 (host RYC2056). The sampling sites were chosen as part ofthe Tampa Bay Healthy Beaches Project. Thi s project is focused on a microbiological pollut ion indicator survey using the major drainage basins of the Tampa Bay region in West Ce ntr a l Florida. The sites represented several different water types, including marine, brackish and fresh water, urban and rurally impacted areas, and hi gh, intermediate and low pollution areas. (See Figure 3 and Table 2 for map of sites and site descriptions) Sites TB 1 through TB 11 were in a predominately rural area with several cattle ranches and small corrimunities with septic tank disposal systems. TB3,4,6,7 and 8 were vario u s locations along Bullfrog Creek in South Hillsborough county. This area has a history of high levels of agricultural pollution. TB 12 through TB21 were in the highly populated Pinellas county TB13, 16, 19 and 20 were beach si tes and TB22 was a control site from the middle of Tampa Bay. 28


Figure 3 Map depicting sample s ite s TB 1 through TB22 c N ..... 0 10 29


Table 2 Environmental sample sites Location Site Salinity Range Impacts Delaney Creek at US 41 TB1 0-5 ppt agriculture and industry Alafia River at US 41 TB2 8-24 ppt fertilizer plant,indu stry Bullfrog Creek at US 41 TB3 0-21 ppt recreational vehicle/trailer park Bullfrog Creek at Symmes Rd TB4 0-5 ppt Interstate I-75,residential area Alafia River at 30 l TB5 0-5 ppt residential area,agriculture Bullfrog Creek at 1-75 TB6 0-5 ppt residential area, cattle pasture Little Bullfrog headwaters TB7 0-2 ppt wildlife,agriculture Big Bullfrog headwaters TB8 0-5 ppt wildlife,agriculture Little Manatee River at 30 l TB9 0 ppt wildlife,agriculture Little Manatee River at US 41 TB10 10-19 ppt mobile home park,business area Manatee River at US 41 TB11 15-29 ppt high flow,county park,business Hillsborough River at Kennedy TB12 0-15 ppt downtown Tampa,small boats Courtney Campbell Causeway TB13 20-28 ppt public beach, dogs, birds Sweetwater Creek at Memorial Hwy TB14 0-8 ppt highly developed residential area with canals Tarpon Lake Cana l at Tampa Rd. TB15 0-2 ppt urban/business area, construction zo nes Honeymoon Island Beach TB16 30-33 ppt no swimming pri s tine beach with lar ge bird population Allen's Creek at US 19 TB17 18-21 ppt business/residential, highly developed Joe's Creek/Cross Bayou at Park St. TB18 2-14 ppt wastewater treatment plant, wildlife preserve John's Pa ss Beach TB19 31-36 ppt heavy boat traffic, tourist area hotel district North Beach, Ft. DeSoto TB20 32-36 ppt lar ge public beach, barrier island re stricts flow Salt Creek at 4'" St. S. TB21 18-24 ppt large residential and business area Control site, midd le of bay TB22 26-31 ppt EPC si te # 16, mid water sample All 22 sites were sampled monthly for 5 months to determine the levels of traditional and alternative indicators, including fecal coliforms, enterococci, Clos tridium perfringens and coliphage. Table 3 shows the geometric monthly mean of the indicators at all sites. The geometric monthly average of fecal coliforms ranged from 0.495 colony forming units (CFU) per 100 ml at the control site TB22 to 3611.8 CFU/100 ml at TB14 Sweetwater Creek. Enterococci ranged from 0.1 CFU/100ml at the control site to 3653.8 CFU/1 OOml at TB4 along Bullfrog Creek. C lo s tridium perfringen levels were below the 30


detection limit for TB 16,19,20 and 22 and were as high as 18.16 CFU/1 OOml at Sweetwater Creek. Coliphage levels ranged from below the detection limit at TB 16, 20 and 22 to 1407 2 plaque forming units (PFU) per 100ml at TB4 along Bullfrog Creek. Figures 4 and 5 are summary graphs of the geometric monthly means for all sampling sites. Sites TB4 Bullfrog Creek and TB 14 Sweetwater Creek show the highest indicator levels when compared to the remaining sites. The additional sites along Bullfrog Creek, TB3, 6 and 7, also show high levels of fecal indicators. Sites TB2, TB10 through TB13, TB 15, TB 16, and TB 18 through TB22 show overall low geometric means for the fecal indicators. Clostridium per.fringens levels, however, do not often correlate with the other fecal indicators. Site TB 14 Sweetwater Creek is the only site that shows an agreement with the traditional and alternative indicator data Figure 6 shows an example of the seasonal variation of the indicators for 5 sites in June through October of 1999. Of the five months sampled, August and September showed peaks for TB 17 Allen's Creek and TB20 North Beach. There were no obvious peaks for TB12 Hillsborough River and TB 14 Sweetwater Creek Site TB4 along Bullfrog Creek exhibited a sharp increase in the October fecal indicator levels Fifteen of the 22 sites exceeded the Florida standard of200 CFU/lOOml geometric monthly mean for fecal coliforms in ambient waters Seventeen of the 22 sites exceeded the EPA s uggest ed guidelines of 33-35 CFU/1 OOml for enterococci in ambient waters. Fujioka et al in 1985 suggested a guideline level of 50 CFU/1 OOml for C.per.fringen for waters in Hawaii (76) While some sites exceeded this level at individual samplings, the geometric means for the sampling sites all fell below the suggested guideline There are currently no suggested EPA standards for coliphage levels in ambient waters. 31


Table 3 Geometric Monthly Mean of Indicators Reported in CFU or PFU/100ml Location Site Fecal Colifonns Enterococci C. perfringens Coliphage Delaney Creek TB1 545.6* 584.2** 3 .10 Alafia River TB2 104 8 40 2** 2 06 Bullfrog Creek TB3 1440 .6* 321.2** 6 .24 Bullfrog Creek TB4 2930 .8* 3653 8** 2 63 Alafia River TB5 868.4* 1537 1** 2 30 Bullfrog Creek TB6 2680.5* 868.3** 2.41 Bullfrog Creek TB7 1490.3* 2027 9** 3 .51 Bullfrog Creek TB8 163.1 472.3** 1.73 Little Manatee River TB9 728 .8* 1052.3** 1.60 Little Manatee River TB10 238.5* 86.2** 0 25 Manatee River TB11 64 6 24 5 0 78 Hillsborough River TB12 319 3* 212.0** 2 56 Courtney Campbell Cswy TB13 287.8* 61.8** 2 75 Sweetwater Creek TB14 3611.8* 1492.3** 18.16 Tarpon Lake Canal TB15 277.4* 100.9** 5 .24 Honeymoon Island Beach TB16 31.6 13.3 0 Allen s Cre e k TB17 927 3* 255 8** 9 .20 Jo e's Creek/ C ross Bayou TB18 2 76 1 51.1** 7 28 John's P ass Beach TB19 109 2 3 0 0 North Beach Ft. DeSoto TB20 121.1 4 8 0 Salt Cre ek TB21 328 .9* 147.5** 0.80 Tampa Bay Control s ite TB22 0.495 0 1 0 *exceeds EPA gmdehnes for fecal cohforms for ambient waters (200 CFU/1 00 ml geometric monthly mean) 583 7 13.1 81.5 1407.2 69. 8 742.4 173 1 44. 7 511.6 54 8 6 7 58 3 0 6 367 8 8 5 0 18. 1 40. 2 0 6 0 15. 2 0 **exceeds EPA guidelines for enterococci for ambient waters (33 CFU/100ml fresh water, 35 CFU/1 OOml marine water geometric monthly mean) Indicator numbers are the geometric mean of results from June through October of 1999 32


Figure 4 Summary of Indicator levels for sampling sites Geometric Monthy Mean Indicators 4000 3500 oFecal Collforms Enterococci 3000 coliphage e 2500 C) C) .... ::;) u.. 2000 a.. ... 0 ::;) 1500 u.. 0 1000 1-500 0 0 -, 1 -;I n. L I r\a. ra. rt. lrL_ lfln n f1._ .. Sample site Figure 5 Summary of C.per.fringens levels for sampling sites Geometric Monthly Mean C.perlrlngens 20 18 16 H o C .perlrlngens I14 e 12 0 0 10 .... :5 u.. 8 0 6 4 2 0 Dll-1ii 1-N Ill 1-.., Ill 1-nn n "' Ill 1-"' Ill 1n n = o .D-n-0 .-N M 'lilt m m m m co 11-1-1-!-Sample sites 33 1 --"' "' ... 1ii 1ii 1ii 1-1-1--CD "' 1ii 1ii 1-10 &1 1n N Ill 1-N N Ill 1-


w Figure 6 Example of seaso n al var i a ti on of in dicat or l eve l s at 5 si t es E 0 0 ...... ::::> u. ll. I ... 0 ::::> u. u 't:J Q) E ... .E In c ... Cl 0 ...J Indicator Levels sFecal C oliforms TB4 Bullfrog Creek TB12 Hillsborough TB14 Sweetwater Creek TB17 Allen' s Creek 1EJ En terococci C perfringen s R iver IE) Coliphage 5 _ TB20 North Beach 4 1 1 1 1-11 I ltl m I 3 2 Q !lUll m ill! lll W I n W I IJJIIIft. W I fiWc l I[IWI I PIJII 11 1 p t W l l Q l J! I:WI 1 1"3. 1 I I I 5 I i I J u ne J uly A u g Se p t Oct Ju n e July Aug Sept Oct J une J u ly Aug Sept Oct J une July Aug Sept Oct June July Au g Sep t Oct Site/Month


During the first 2 months of the sampling schedule, all sites were analyzed for the Bacteroides fragilis human strain B40-8 (host ATCC 51477 (HSP40)) and the animal/human strain B56-3 (host RYC2056) using the standard overlay procedure (see Material and Methods section). Phage levels were consistently below the detection limit of 1 0 PFU per 1 OOml for all sampling sites Each site was then evaluate d for the remaining 3 months of the survey using the presence/absence B fragilis enrichment assay (see Material and Methods section). Tables 4 and 5 shows the results of the overlays and enrichments for both strains of the B fragilis bacteriophage for all sampling sites. For the B fragilis human strain B40-8 56 total enrichments were performed during the months of August, September and October Seven samp le s, or 12.5% of the total tested positive for the presence of the phage strain. Only one site TB17 Allen's Creek, tested positive for two of the three sampling dates The B fragilis animal/human phage strain B56-3 was used for a total of 57 enrichments during the month s of August, September and October. Twenty eight samples or 49% of the total, tested positive for the presence of the phage strain Three sites, TB4 and 6 along Bullfrog Creek and TB 14 Sweetwater Creek tested positive during all three months of the study. All positive enrichment samples were repeated using the standard overlay procedure (See Material and Methods sec tion) Results were consistently below the detection limit ofthe assay. Positive enrichment results can be considered to be <10 PFU/100ml for all sample sites. 35


Table 4 B.fragilis phage B40-8 (host ATCC 51477) Site Jun July Aug Sept Oct TB1 <20 pfu/1 OOml <20 pfu/1 OOml <20 pfu/1 OOml Positive Negative TB2 <20 pfu/1 OOml <20 pfu/100ml <20 pfu/1 OOml N/A Negative TB3 <20 pfu/1 OOml < 20 pfu/100ml Negative Negative Negative TB4 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Positive TB5 <20 pfu/1 OOml <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative TB6 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB7 <20 pfu/1 OOml <20 pfu/1 OOml Negative Positive Negative TB8 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB9 <20 pfu/1 OOml <20 pfu/1 OOml <20 pfu/1 OOml N/A Negative TB10 < 20 pfu/1 OOml < 20 pfu/1 OOml <20 pfu/1 OOml N/A Negative TBll < 20 pfu/100ml < 20 pfu/1 OOml <20 pfu/1 OOml NIA Negative TB12 <20 pfu/1 OOml < 20 pfu/1 OOml Positive Negative Negative TB13 <20 pfu/100ml < 20 pfu/1 OOml Negative Negative Negative TB14 <20 pfu/100ml <20 pfu/1 OOml Negative Positive Negative TB15 <20 pfu/100ml <20 pfu/100ml Negative Negative Negative TB16 <20 pfu/1 OOml < 20 pfu/lOOml Negative Negative Negative TB17 < 20 pfu/100ml <20 pfu/1 OOml Positive Positive Negative TB18 < 20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB19 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB20 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB21 <20 pfu/1 OOml <20 pfu/100ml Negative Negative Negative TB22 <20 pfu/lOOml <20 pfu/1 OOml Negative Negative Negative . Results of <20 pfu/1 OOml are from the standard overlay procedure Positive and Negative resu.lts were obtained using the enrichment assay 36


Table 5 B fragilis phage B56-3 (host RYC2056) Site Jun July Aug Sept Oct TBl <20 pfu/1 OOml <20 pfu/1 OOml <20 pfu/100ml Positive Negative TB2 <20 pfu/1 OOml <20 pfu/1 OOml 20 Qfu/1 OOml N/A Positive TB3 <20 pfu/1 OOml <20 pfu/1 OOml Negative Positive Positive TB4 <20 pfu/1 OOml <20 pfu/1 OOml Positive Positive Positive TB5 <20 pfu/1 OOml <20 pfu/100ml <20 pfu/1 OOml Positive Positive TB6 <20 pfu/1 OOml <20 pfu/1 OOml Positive Positive Positive TB7 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB8 <20 pfu/1 OOml <20 pfu/1 OOml Negative Positive Positive TB9 <20 pfu/100ml <20 pfu/1 OOml <20 pfu/1 OOml NIA Positive TB10 <2 0 pfu/100ml < 20 pfu/1 OOml < 20 pfu/100ml NIA Positive TBll <20 pfu/1 OOml <20 pfu/1 OOml < 20 pfu/100ml NIA Negative TB12 <20 pfu/100ml <20 pfu/1 OOml Positive Positive Negative TB13 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB14 <2 0 pfu/1 OOml <20 pfu/100ml Positive Positive Positive TB15 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB16 <20 pfu/100ml <20 pfu/1 OOml Negative Negative Negative TB17 <20 pfu/1 OOml <20 pfu/100ml Positive Positive Negative TB18 <20 pfu/1 OOml <20 pfu/100ml Positive Positive Negative TB19 <20 pfu/1 OOml <20 pfu/1 OOml Negative Negative Negative TB20 <20 pfu/1 OOml <20 pfu/lOOml Negative Negative Negative TB21 <20 pfu/100ml <20 pfu/1 OOml Positive Positive Negative TB22 <20 pfu/1 OOml <20 pfu/100ml Negative Negative Negative . Results of <20 pfu/1 OOml are from the standard overlay procedure, Positive and Negative results were obtained using the enrichment assay. 37


In order to confirm that the bacteriophage forming plaques on the B fragilis phage host in the enrichment procedure was morphologically consistent with the B fragilis bacteriophage, transmission electron micrographs were taken of the control bacteriophages, and of phages isolated from plaques from the environmental samples All phages isolated from the enviromental samples exhibited the consistent morphology of icosahedral heads with long flexible segmented tails. This morphological type suggest s the virus family Siphoviridae which is the family identified by Ackermann and DuBow (54) as including the B. fragilis bacteriophages Figures 7 and 8 are transmission electron micrographs showing the B40-8 and B56-3 bacteriophages from the stock culture. Figures 9 and 1 0 are the respected bacteriophages from environmental samples. 3 8


Figure 7 TEM Micrograph ofB56-3 B fragilis bacteriophage (host RYC2056) from stock culture Figure 8 TEM Micrograph ofB40-8 B.fragilis bacteriophage (host ATCC 51477 (HSP40)) from stock culture 39


Figure 9 TEM Micrograph:TB I 0 Little Manatee River-phage isolated using B.fragilis host R YC2056 Figure I 0 TB2 Alafia River phage isolated using B. fragilis host A TCC 514 77 40


Using the indicator data all sites were evaluated to detennine the degree of fecal contamination The sites were ranked numerically using 1 for the lowest geometric monthly mean of an indicator organism to 22 for the highest. The final ranking score for each site was based on the total score of all indicator organisms. (77) Table 5 shows the ranking results for each individual indicator organism and the total for each site. Sites TB14 Sweetwater Creek, TB3,4,6 and 7 along Bullfrog Creek and TB1 Delaney Creek all exhibited high total ratings in regard to indicator levels The lowest ranking sites included TB11 Manatee River, TB19 John's Pass, TB20 North Beach, TB16 Honeymoon Island Beach and the TB22 control site in the middle of Tampa Bay. Intennediate ranking was observed for the remaining sites. In addition to the numerical site ranking, a cluster observation analysis was perfonned using the geometric monthly averages of the indicators to evaluate the degree of pollution at each site (See figure 11) The results were consistent with the numerical ranking results. Sites TB4 Bullfrog Creek and TB 14 Sweetwater Creek show similarities in regard to the high level of fecal indicators present. The sites lowest in indicator levels TBll Manatee River, TB16 Honeymoon Island, TB19 North Beach TB20 John's Pass and TB22 Control site also grouped together at the other end of the cluster observation. 41


Table 6 Numerical ranking of sites based on indicator levels Location Site Fecal Enterococci c. Coliphag e Total Coli forms perfringens Sweetwater Creek TB14 22 19 22 18 Bullfrog Creek TB4 21 22 14 22 Little Bullfrog headwater s TB7 19 21 17 17 Bullfrog Creek TB6 20 17 12 21 Bullfrog Creek TB3 18 14 19 16 Delaney Creek TB1 14 16 16 20 Alafia River TB5 16 20 11 15 Allen's Creek TB17 17 13 21 10 Little Manatee River TB9 15 18 8 19 Hillsborough River TB12 12 12 13 14 Joe s Creek/Cro s s Bayou TB18 9 7 20 II Tarpon Lake Canal TBI5 10 10 18 7 Big Bullfrog headwaters TB8 7 15 9 12 Salt Creek TB21 13 II 7 9 Courtney Campbell Cswy TB13 11 8 15 5 Little Manatee River TB10 8 9 5 13 Alafia TB2 4 6 10 8 Manatee TB11 3 5 6 6 John s Pa s s TB19 5 2 3 4 Honeymoon Island Beach TB16 2 4 4 3 North Beach Ft. DeSoto TB20 6 3 2 2 Middle of Tampa Bay TB22 I 1 I 1 R a nkmg destgnated by asstgnmg number to s tte, 1 for lowest mdtcator number, 2 2 for htghest for each indicator and then ranking site overall by total 42 81 79 74 70 67 66 62 61 60 51 47 45 43 40 39 35 28 20 14 13 13 4


Figure 11 Cluster observation showing % similarities of sample site based on geometric monthly averages of the indicator organisms Similarity I 24.86 49 90 -74.95 -J_ l I I n I l l l I I I I I I I I I I I I I 100 .00 4 14 6 5 7 9 3 15 18 17 B 2 12 13 10 21 11 16 22 19 20 Sampl e sites 43


Statisical analysis of the Bacteroides fragilis phage results from the environmental survey was performed to determine if the presence of the bacteriophage was linked to the levels of traditional and alternative indicator organisms. Because of the differing nature of the two data sets (indicator data as numerical concentrations and the B. fragilis phage as presence / absence) binary logistic regressions were performed to compare the presence of B. fragilis phage strains B40-8 and B56-3 to the levels of fecal coliforms enterococci, C. perfringens and coliphage (78). Binary logistic regression will first determine if the slope ofthe regression line is different from zero, then goodness of fit tests will evaluate how well the data fits the modeled expected frequencies. Results of the regression will be presented as the percentage of concordant pairs (signifying that the model predicted the outcome of both presence and absence), percentage of discordant pairs (model did not successfully predict the outcome), and percentage of tied pairs (model predicted only one of the parameters) Table 7 shows the results of the binary logistic regression. When the animal/human strain of B fragilis phage (B56-3) was used as the outcome or response organism the concordant percentage of all the indicators combined was 92.9%. The concordant percentage when using the human strain B40-8 as the response or outcome organism was only 72.7%. There was no significant difference in the concordant percentage results between using B56-3 phage strain as the response organism, and when using both phage strains combined This indicates that there is not a strong correlation between the presence of the human B fragilis strain and the levels of the indicator organisms. In addition fecal coliforms and enterococci alone do not seem to be good predictors of the presence ofthe Bacteroides fragilis phage strain B56-3. The concordant percentage is 85. 1% when using these 2 44


indicators as the model organisms, compared with 92 9% when all the indicators are entered into the equation, or 91.4% when only C. perfringens and coliphage are used as the model organisms. Binary logistic regressions were also performed to determine if a relationship between the salinity of the sampling sites and the presence of the B fragilis phage existed The regressions did not show any significant correlation for the environmental samples. Table 7 Binary Logistic Regression results Response organism Model organism Concordant % Discordant % Tie% 856-3 (RYC2056) Fecal Coliforms 75.6 24.1 0.2 Enterococci 85.1 14 7 0.2 C.perfringens 67.0 16 9 16. 1 Coliphage 89.8 8 9 1.4 FC + Enterococci 85.0 14.5 0 5 C.perfringen + coliphage 91.4 8 6 0 0 All indicators 92.9 6 9 0 2 8408 (A TCC 51577) Fecal Coliforms 67. 2 31.8 1.0 Enterococci 66.4 32. 3 1.3 C.perfringens 55.2 29.7 15. 1 Coliphage 69.0 28.6 2 3 FC + Enterococci 66 7 32 0 1.3 C perfringen+coliphage 70. 1 28.1 1.8 All indicator s 72. 7 27.3 0 0 840-8+856-3 Fecal Coliforms 75. 1 24.7 0.2 Enterococci 85 7 14. 1 0 2 C.perfringens 64.3 18 .3 17.4 Coliphage 90.5 8 0 1.5 FC + Enterococci 85.4 14.4 0 I C.perfringen + coliphage 91.0 8.5 0.5 All indicators 92.3 7.5 0 1 45


The relationship between the presence or absence of the Bact eroide s fragilis phage and the other feca l i nd icat o rs can a l so b e depicted on a scatterplot. In Figure 12 the presence of the B fragilis phage strain is represented by a 1, and the absence of the phage strain is represented by -1. Comparing co l iphage and B fragilis phage strain B56-3 show that a majority of the positive B fragilis phage resu lt s occurred as the coliphage l evels increased However some sites with high coliphage levels were negative for the presence of the B fragilis phage Figure 13 show s a scatterplot comparing the presence of the B.fragilis phage s train B56 3 to the levels of fecal coliforms found at the sampling sites The plot shows a very poor relationship between the two indicator organisms, with equal number of positive and negative B56-3 phage results occuring as the level of feca l coliforms rise Figure 12 Scatterplot of B fragilis phage B56-3 and coliphage Bactero i des fragilis phage strain 856-3 and coliphage 1 5 1 ...... .. .... ... ...... -T """"'"' 0 .5 -.., .;, 0 II) Ill 0 5 1 1 5 2 2 .5 3 3 5 4 4 5 -0.5 -1 .... ...... ...... T .... .... .... -1. 5 C oliphage -Log transformed 46


Figure 13 Scatterplot of B. fragilis phage B56-3 and fecal coliforms Bacteroides fragilis phage strain 856-3 and fecal coliforms 1 5 1 ..... .. . .. .... Ill c 0 5 Qj u Ill .a Qj u 0 c Qj Ill e 1 2 3 4 5 D.. t? -0 5 CD II) al -1 ..... ....... ..... ..-........... ..... .... ......... -1. 5 Fecal Colifo rms Log transformed 47


Survival in Seawater The subtropical climate of Florida makes it a unique ecosystem when compared to the remainder of North America The warm marine and estuarine waters make it difficult to apply indicator assays that may be used with great success in temperate zones and in fresh water systems Survival of pathogenic microorganisms and indicator microorganisms may also differ in regards to the warm climate and saline waters. In order to determine the effects of temperature and salinity on the persistence of the Bacteroides fragilis bacteriophage B56-3 and B40-8 in seawater compared to coliphage an experiment was performed using flasks of sterile filtered natural seawater incubated in the dark at 10 20 and 30 degrees C. Dilution of phage stock cultures were added to the flasks, and each flask was sampled daily for 12 days, then at 18 and 24 days. Phage concentrations were compared among the three virus strains at each temperature. (See Material and Methods section) The results were analyzed as suggested by Yates et al, 1985 (7 5). Phage concentrations at each sampling were log transformed and plotted against the corresponding day of the experiment. Best fit linear regression l i nes were applied to the curves, and the slope of the line was used to determine the decay rate of the virus by using the following equat ion : Decay rate = [ (Log10 PFU) day I] A summary of the decay rates are presented in Table 8. The bacteriophage MS2 shows the highest decay rate for each of the three temperatures compar e d to the two B frag ilis 48


phage strains At 1 0C temperature appears to have little effect on the survival of phage strains B56-3 and B40-8 The decay rate of phage B56-3 at 20C is 0.512log10 day-1 Phage strain B40-8 does not show as strong a correlation between temperature and phage survival as strain B56-3 does at this temperature Both B .fragilis phage strains exhibit a lower decay rate at 30 C than at 20C. Bacteriophage MS2 has the highest decay rate at 30 C surviving in the seawater flask less than 36 hours. Figures 14 through 16 show the linear regression curves for the three phages at each of the experimental temperatures plotting survival against time to achieve the decay rate. Table 8 Decay rates of bacteriophage Temperature Bacteriopha g e Decay rate* Value IOUC MS2 0 1774 R" = 0.9606 B56-3 0.0085 RL0 1316 B40-8 0.0016 R L 0 0073 20uC MS2 0 9 2 64 0 9977 B56-3 0 512 R l = 0 9606 B40-8 0 1697 Rl= 0 675 5 3 0 u C MS2 3 0107 R L 1.0000 B56-3 0 168 7 R L 0 8739 B40-8 0 0846 0 7508 *reported in log10 dai1 49


Figure 14 Survival curve at 1 0C 1 0 d egrees C survival 8 y = -0 0016x + 6 6963 R1 = 0 0073 7 .. 6 y = -0 0085x + 6 53 4 R' = 0 1 316 6 5 :::> u.. 0.. 4 0 ..--1:1) .3 3 2 0 -y = -0 1 774x MS2 R = 0 9606 ATCC 6 RYC2056 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Days Figure 15 Survival curve a t 20C 20 degrees C survival 10 y = -0. 1 697x 6 5763 5 R2 = 0 6755 0 : u.. 0.. -5 0 ...... 1:1) 0 ...I -10 MS2 .ATCC -15 6 RYC2056 y = -0 926 4 x + 7 667 4 R2 = 0 9977 -2 0 Days 50


Figure 16 Survival curve a t 30C 30 degrees C survival 20 y = .0.0846x 6 6761 R1 s 0 .7508 10 0 -10 'E -20 ::) u. ll. 0 ..... -3 0 Cl 0 -J -40 -50 -60 ""' 1 2 7 8 9 1 0 11 12 13 1 4 15 16 17 18 19 20 21 22 23 24 25 y = .0.1687x 6 .2268 R' = 0 .8739 :j .MS2 I 8408 A 856 3 y = -3. 0107x 9 .7204 "R' = 1 -70 5 1


Treatment Plant Indicator Levels Environmental levels of indicators found in the Tampa Bay area including the B. fragilis phages, may differ greatly compared to levels found in other geographic areas of the world Florida's subtropical location and extensive coastal zones make it a unique ecosystem in North America Indicator organisms used with great success in temperate zones may not be applicable to the warm marine and estuarine waters of West Central Florida Studies done in Spain to determine the levels of B. .fragilis bacteriophage B40-8 (host ATCC 51477)in sewage found numbers ranging from 10to 104 PFU per 100ml. (25) B .fragilis phages are commonly found in lower numbers than coli phages which can range from 103 to 106PFU per 100 ml in domestic sewage. (55, 63) In order to compare the l e vels of B .fragilis bacteriophages in domestic sewage found in the Tampa Bay area to those levels found in the European studies, samples were taken from three area wastewater treatment plants and analyzed for the levels of B .fragilis phage strains B56-3 and B40-8 The samples were also analyzed for the presence of fecal coliforms, enterococci C. perfring e ns and coliphage The summary of these results appear in Table 9 Treatment plant design included primary treatment such as sedimentation secondary treatment utilizing activated sludge basins and tertiary or advanced treatment involving sand filtration and a disinfection step using chlorine or UV Each of the facilities chosen for the study had a different treatment plant design The Howard Curren Advanced Wastewater Treatment Plant located in the Port of Tampa was the largest facility surveyed, with a flow rate capacity of 98 million gallons per day (mgd). This plant was 52


capable of processing up to 300 mgd during storm events In addition to activated sludge, this plant design also included a denitrification filter chlorination step to treat the effluent, and a dechlorination step before discharging into Tampa Bay. The contact time for the chlorine basin was 30 minutes The Albert Whitted Water Reclamation Facility was located at the mouth of Bayboro Harbor in downtown St. Petersburg, Florida, and was rated for a flow rate capacity of 12 mgd. This plant did have a chlorination step, but the water was not dechlorinated The discharge of the plant was routed to the reclaimed water distribution system. The contact time of the chlorine basin was 15 to 30 minutes The final treatment plant of the survey was the City of St. Petersburg Southwest Water Reclamation Facility located at the southwest tip of Pinellas County. This facility was rated for a 12 mgd flow rate, although it was capable of processing up to 30 mgd during storm events. Rainwater for the city of St. Petersburg was diverted to this plant. Activated sludge and a chlorination step were included in the plant design The contact time ofthe chlorine basin was 15 to 30 minutes, and the discharge from the plant was routed to injection wells There was no dechlorination step Influent samples from all three plants were taken at the intake port at the head of each treatment plant. Effluent samp l es from the Howard Curren plant were taken after dechlorination from the discharge line leading into Tampa Bay. Effluent samples from Albert Whitted were taken after the chlorination step from the line leading to the reclaimed water distribution system Effluent samples from the Southwest plant were taken after the activated sludge basin but before the chlorination process. Final effluent samples were also taken after the chlorination step from the discharge line leading to the 53


injection wells. Both chlorinated samples from Albert Whitted and the Southwest plant were dechlorinated with sodium thiosulfate in the laboratory before analysis of fecal indicators. The Howard Curren plant had flow rates from 58.6 mgd to 59.8 mgd during the sampling periods. The average daily flow rate for November was 50 to 51 mgd. Fecal coliform levels in the influent averaged around 106 per 100 ml. Enterococci levels were 1 05 C. perfringens averaged 104 and coliphage levels were 1 05 per 1 OOml. The B. fragilis phage B56-3 ranged from 1.2 x 104 to 2.4 x 104 while phage B40-8 levels were much lower, ranging from 66 7 to 167 PFU/lOOml. The effluent showed consistent results below the detection limit of the assays for all indicators with the exception of C. perfringens, which had results of 4 CFU / 1 OOml for each of the November samples The phages B40-8 and B56-3 were negative for the enrichment procedure for both November effluent samples. The Albert Whitted facility had flow rates of 8 55 and 11.5 mgd respectively during sample collection The levels of indicators in the influent were similar to those found at Howard Curren. Phage B56-3 levels ranged from 3 .83 x 104 to 1.11 x 105 phage B40-8 levels ranged from 100 to 350 PFU/lOOml. In the effluent ofthis plant, fecal coliforms were detected at 1 and 6 CFUIIOOml enterococci was below the detection limit of the assay, and phage B56-3 was detected using the enrichment procedure in both samples collected in November. The Southwest Water Reclamation Facility was sampled at the influent pipe as in the other plants, but effluent was collected prior to and after disinfection. Influent levels of the indicators were similar to the other 2 treatment plants. The phage B56-3 showed 54


levels of2.57 x 104per 100 ml at both November samplings, and phage B40-8 was also the same for both samplings at 233 PFU /lOOml. Fecal coliforms, enterococci, C. perfringens coliphage and B. fragilis phage B56-3 were present in the pre-chlorinated effluent but not present in the post-chlorinated effluent during the November 1 01 h sampling. On November 12, however, coliphage levels of30 PFU/100ml and positive enrichment for B56-3 was detected in the post-chlorinated effluent. During this particular sampling, the facility was in the process of acid flushing the plant. This did not affect the influent, but had the potential of affecting the effluent in terms of the survival of the indicator microorganisms due to the changes in pH. 55


Table 8 Indicator survey of wastewater treatment plants Howard Curren Advanced Wastewater Treatment Plant 3 /26/99 ll/10/ 99 ll/12/99 Indicator Influent Effluent Influent Effluent Influent Effluent Fecal colifonns * 1.62 X IOU < I 2 28 X IOU N / A Enterococci * 1.54xl0, < I 1.73 X IOJ < I C perftingens * 4 2 7 X 10 4 Coliphage * 1.04 X 10 < 10 2 .99 X 10' < 10 B56-3 l.l9 X 10 2.4 X 10" Negative 1.9 X 10 Negative B40-8 100 66 7 Negative 167 Negative Flow rate 59 8 mgd 58.86 mgd All m1croorgamsms reported m CFU or PFU per l 00 ml Albert Whitted Water Reclamation Facility 817199 I III 0 / 99 11112/99 Indicator Influent Effluent Influent Effluent Influent Effluent Fecal colifonns * 2.7 xl0u I 4.13 X IOU 6 Enterococci * 2.95xlOJ < I 3 .24 x 10J < I C perftingens * 2.45 X 24 4.85 X 10 8 Coliphage l.52xl0, < 33 1.23 x to 1.2 X IOJ 3 X 10' 100 B56-3 l.ll X l 0 <33 3 83 X 10" Positive"'"' 1.01 X 10' Positive""" B40-8 350 < 33 100 Negative 300 Negative Flow rate "' 8.55 mgd 11. 5 mgd All m1croorgamsms reported m CFU or PFU per l 00 ml City of St. Petersurg Southwest Water Reclamation Facility 11110 /99 ll/12/99 Indicator Influent Pre Po s t Influent Pre Post chlorination chlorination chlorination chlorination Fecal 3 0 X 10 24 < I 2 .68 X 10 65 I colifonns Enterococci 2 88 x to, < l < I 6.15 X 10 6 < I C perftingens 1.3 X 10 272 <2 2.0 X 10" 146 < I Coliphage l.l5 X 10 290 <10 2 03 x IOJ 460 30 B56-3 2.57 X 10 67 Negative 2 57 X 10 125 Positive""" B40-8 233 Negative Negative 233 Negative Negative Flow rate 9.5 mgd 16.2 mgd"'** All m1croorgamsms reported m CFU or PFU per I 00 ml *tes t not perfonned **positive enrichment results s ignify < 10 PFUIIOOml """"' increased flow due to acidification of treatment plant 56


Discussion and Conclusions Studies have been undertaken in Europe and South Africa to determine if the Bacteroides fragilis bacteriophage could be detected in ambient water and used as an alternative indicator of fecal contamination Most of the studies determining envirorunentallevels of B.fragilis phage have been conducted in Spain (previous studies for B fragilis phage levels are summarized in Table 1 0) Results for the animal/human phage strain B56-3 have generally not been included as most studies concentrated on detecting only the B40-8 human strain. These studies describe heavily impacted or polluted waters as those directly receiving untreated sewage The methods used were either the standard overlay method or the presence / absence assay in an MPN format. The results exhibit a great deal of variation depending on the pollution level of the sampling sites; phages were found to range from 101 to 104 PFU per 100 mi. Waters that were considered low impact, or those not receiving urban sewage and runoff, did consistently fall below the detection limit of the assay for the human strain B40-8 The first of this project's objectives was to determine if the human strain B40-8 and the animal/human strain B56-3 B fragilis phages could be detected in ambient waters in the Tampa Bay region. Using the standard overlay method (detection limit of < 1 0 PFU/1 OOml), the phages were not detected at any of the sampling sites All positive results were obtained using the presence /absence procedure We detected the human strain (B40-8) at 27% ofthe sampling sites ; TBl Delaney Creek, TB4 Bullfrog Creek, TB7 Bullfrog Creek, TB 12 Hillsborough River TB 14 Sweetwater Creek and TB 17 57


Allen's Creek. More than 50% of the Delaney Creek watershed is urban development. Bullfrog Creek watershed includes 50% agricultural use 12% urban and residential. The Hillsborough River has 32% agricultural use, 25% urban and industrial impacts. Sweetwater Creek has high urban development at 69% of the total watershed. (Lipp et al 1999, submitted to Hydrobiologia) Of the sites positive for the human strain, sites TB4, 14 and 7 also ranked the highest in respect to the levels of traditional and alternative fecal indicators. The remaining three sites, TB 1, 12 and 17 ranked 6th, gth and 1Oth highest in regards to the level of indicators detected. The animal/human strain B56-3 was detected at 63% of the sites. The negative sites included TB7 15, 13, 11, 19, 16,20 and 22. Ofthose sites, TB11 19, 16,20 and 22 did rank as some of the least fecally contaminated sites, but TB7 ranked as the third most polluted and TB15 and 13 were ranked as intermediate sites. Not enough human enterovirus data has been collected from the Tampa Bay sampling sites to determine if the presence of the B. fragilis phage correlates to the presence of enteroviruses The binary logistic regressions showed that there is no strong relationship between the presence of the human strain of B.fragilis phage and the levels ofthe other indicators. While the relations hip between the animal/human strain B56-3 and the other indicators did show more correlation, the indicator bacteria alone did not appear to be a good predictor of bacteriophage levels. This supports the current belief among researchers that using indicator bacteria does not provide an adequate model for viral fate and transport. In this study, coliphage showed the best individual relationship in regards to the presence ofthe B.fragilis phage Lucena et al, 1996, found that B fragilis phage, coliphage and 58


enteroviruses clustered together better than indicator bacteria when environmental data were analyzed. (64) TABLE10 Summary of B .fragilis phage environmental studies Year 1988 (a) 1989(b) 1996 (c) 1997 (d) 1 997 (e) (a)(79) (b)(62) (c)(64) (d)(60) (e)(80) Area Spain Spain Spain France Spain Water type polluted river polluted seawater groundwater intermediate water low impact water s polluted river groundwater polluted river polluted river intermediate water low impact water Reported in PFU/1 OOml Phage 8408 Other Indicator s Ave 4 8 x Ave 7 3 x lOL Not detected 0-43 coliphage fecal coliforms enterococci C.perfringens Not detected fecal coliforms enterococci C.perfringens 7-5 .3 X 1 O J 2-3 >/= 1.6 X 10 Somatic coliphage f-specific coliphage -IoSomatic coliphage f-specific coliphage < 1 O-l.5x 1 O J Somatic coliphage f-specific coliphage <10270 Results 9 06 X IO-10 -10 -10 73 2.4 X lOL -10L Not detected -10'-10 1o -1o 10 < 10 3 6x 10 <10-680 <10400 Before determining the existing background levels of B .fragilis phage in domestic sewage, it was important to consider the levels found in feces. Allsop et al in England found B .fragilis bacteria occurred in greater numbers in feces than E. coli The B .fragilis bacteria group averaged 109 CFU per gram of feces in humans 107108 CFU per 59


gram of feces for domesticated cats and dogs, 107 CFU per gram of feces in seabirds and 104 CFU per gram of feces in both chickens and pigs. B fragilis bacteria was not found in cattle feces. ( 48, 81) Bacteroidesfragilis phage strain B56-3 (host RYC2056) has been isolated from 28% of human fecal samples tested, 31% of pigs tested and 29% of poultry It has never been isolated from cattle, sheep or horse fecal samples The levels found were 6 7 x 102 PFU per gram of feces in pigs and 1.3 X 102 PFU per gram of feces in poultry. (58, 82) The human strain B40-8 (host ATCC 51477) has been isolated from 10-13% of human fecal samples tested, but not from any animal fecal sample. (domestic animals, primates and seabirds). (26, 27 55, 58) In comparison somatic coliphage was detected in 54% of human isolates tested, 56% of domestic animals, 53-57% of primates and 60% of seabirds F -specific coliphage was found in 26% of humans 90% of domestic animals, 63-76% ofprimates and 20% of seabirds fecal samples tested. (26) All ofthe fecal studies used samples obtained from Spain, Great Britain and South Africa The second objective ofthis project was to determine the local levels of the B40-8 and B56-3 phage strains in domestic sewage in order to identify possible sources to the environment. Table 11 shows summary of previous sewage studies done in Europe and South Africa Puig et al, 1999 utilized 114 strains of B fragilis host (including B40 8) to detect bacteriophage in sewage. Sixty-six of these strains detected phage in human sewage, however the numbers were highly variable between strains. Host RYC2056 (B56-3) detected the overall highest number of phage in sewage (55) Jofre et al, 1989, found that the ratio of B. fragilis phage to enterovirus in sediments was similar to the ratio found in sewage, showing that the fate of both may be similar in the environment. 60


(83) Levels ofB40-8 were highly variable and ranged from 7 PFU to 105 PFU per 100 ml. TABLE 11 Summary of sewage indicator studies Reported in PFU/1 OOml Year Area Sample type Phage Phage Indicator Results B56-3 B40-8 1987(a) Spain sewage influent 7-l.lx 10' 1988 (b) Spain sewage influent 8.9 x to 1989 (c) Spain sewage influent 5.3 X 10' coliphage 1.2 X 10 fecal colifonns -10 enterococci -10 C. perfringens -10, 1996 (d) France sewage influent > =4.4x to 1999(e) Spain* sewage influent 32-190 82-440 E. coli 1400 7600 fecal colifonn s -10' -10' s laughter house 0-7.5 waste 2 .9-2.4xl02 E coli fecal colifonns to-10o average lev e ls for samples from Netherlands, Ireland Gennany Austr1a Portugal Gennany, Sweden, France and South Africa (a)(27) (d)(60) (b)(79) (e)(55) (c)(62) The level of B. fragilis phage in domestic sewage was much lower in this region compared to the European studies. Phage strain B40-8 detected in Tampa Bay wastewater treatment plants ranged from 66.7 to 350 PFU per 100 ml, and was found in 100% ofthe sewage influent samples tested. Phage B56-3 levels in the 1999 study by Puig et al ranged from 82 to 440 in sewage and 2.9 to 240 in slaughter house waste Our 61


study found much higher levels in sewage samples, with results ranging from 1 19 x 1 04 to 1.11 x 105 PFU per 100 ml, and was detected in 1 00% of sewage samples tested. Samples from all 3 plants gave consistent coliphage numbers of around 105 PFU per 1 00 ml, fecal coliforms numbers of 106 CFU per 100 ml, enterococci levels of 105 CFU per 1 00 ml and C. per.fringens levels of 1 04 CFU per 1 00 ml. Each indicator consistently averaged one log lower than studies done in Spain. The low numbers of the human strain B40-8 in Florida sewage may explain the low numbers in ambient water in the geographic location studied. The B56-3 phage strain found in domestic sewage should only be coming from human waste, however, the environmental sources from pigs and poultry would make it difficult to trace the source when it is found in the environment. Seabirds may be a consideration in Tampa Bay as a possible environmental source, however no studies were found determining the presence of phage B56-3 in seabird fecal samples. Positive enrichment for the B56-3 B .fragilis phage occurred in 2 of the 3 treatment plant effluent samples on 3 occasions During the fust event, fecal coliforms were 1 CFU per 100 ml, enterococci was negative, C.perfringens was 24 CFU per 100 ml and coliphage was 1 03 PFU per 100 ml. The second event showed fecal coliform levels of 6 CFU per 100 ml, enterococci was negative, C. perfringens was 8 CFU per 1 OOml and coliphage was 102 PFU per 100 ml. The last event had fecal coliform levels of 1 CFU per 100 ml, both enterococci and C. perfringens were negative, and coliphage was 30 PFU per 100 ml. The human strain B40-8 was not detected in any ofthe effluent samples analyzed, but given the low numbers in the influent, that may be expected The sewage study further highlights the fact that the indicator bacteria are not suitable models for the 62


fate and transport of human viruses. Clostridium perfringens shows to be the more resistant to treatment than the other indicator bacteria. In each case of a positive enrichment result for B56-3 in the effluent, coliphage was also detected All effluent samples showing negative enrichment results for B fragilis phage were negative for coliphage Two survival studies have been done to compare the persistence of B. fragilis phage strain B40-8 with that of human enteroviruses. None of these have addressed the survival of the phage in warm seawater. Jofre et al, 1986, utilized both freshwater and seawater in laboratory and in situ experiments The temperature for the laboratory study was 22C while the in situ experiment was conducted at 15 C. Poliovirus, simian rotavirus, f2 coliphage and B40-8 phage persistence were compared for 7 days. The study found survival of B40-8 phage similar to the f2 coliphage, and slightly greater than poliovirus and rotavirus. All viruses survived for longer in freshwater than in seawater.(67) Tartera et al, 1988, studied the inactivation of B. fragilis phage B40 -8, E coli, S.faecalis, poliovirus 1, simian rotavirus II, f2 coliphage when exposed to UV and chlorine. B40-8 phage proved more resistant to chlorine than all other organisms tested. E.coli and S.faecalis had fastest inactivation rates overall. For UV, B40-8 shows similar inactivation rates to the enteroviruses tested. F2 coliphage showed slightly greater resistance to UV. Again, the two species of bacteria showed the fastest inactivation rates (84) Table 12 shows a summary of the indicator levels found in the treatment plant survey, and the average removal rate for each Coliphage and the B. fragilis phage B56-3 showed 63


similar removal rates following chlorination Removal rates for the human strain 840-8 are difficult to compare because of the low numbers found in the influent. Table 12 Average indicator levels from treatment plant survey Indicator Influent* Effluent* Removal rate Fecal Coliforms 2.7 X 10 1.6 5 to 6 log removal Enterococci 3.1 X 0 5 log removal C. perfringens 2 6 X 10'+ 6.7 3 to 4 log removal Coliphage 1.85 X 10J 190 2 to 4 log removal B56-3phage 4.46 X 10'+ 3 of 6 positive ( < 1 0) 2 to 4 log removal 8408 phage 194 Negative 2 log removal *reported m PFU or CFU per 100 ml Yates et al, 1985, studied the fate of viruses in groundwater samples, using poliovirus, echovirus and MS2 coliphage (host A TCC 15597). The survival study used in our project utilized the experiment design found in this study. She found that temperature correlated significantly with virus decay rates Enterovirus decay rates ranged from 0.060 to 0.186 log10 daf1 at 12 C and 0.188 to 0.628 log10 day -1 at 23C. MS2 decay rates ranged from 0.012 to 0 064logiO day-1 at 4 C, 0.030 to 0.162logl0 daf1 at 12 C, and 0.187 to 0.578 log10 daf1 at 23C. The calculation of decay rates in our survival experiments utili ze d the formula developed by Yates et al in the 1985 study described above. (75) MS2 coliphage decay rates found in our study were 0.1774logJO daf1 at 1 0C but increa se d to 0 .926 4 log10 daf1 at 20C and 3.0107logJO daf1 at 30C. This i s of major concern when attempting to use coliphage as an alternative indicator in subtropical environments such as West Central Florida. Summer water temperatures in the Tampa Bay region can reach 32C, 64


and temperatures up to 35C can be found in the Florida Keys. Griffin et al, 1999, found low coliphage levels in canals throughout the Florida Keys, but 79% of these sites were positive for enteroviruses. Water temperatures at these sites ranged from 29 to 33C. (77) Bacteriodes fragilis phage persisted much longer in the seawater compared to the coliphage. The lowest temperature of 1 0C had almost no effect during the 30 days of the study on both phage strains. B40-8 phage decay rates were 0 1697 at 20 C and 0.0846 at 30 C B56-3 phage decay rates were 0 512 at 20C and 0.1687 at 30C. They both fell within the range found for enteroviruses tested at 23 C as noted in the Yates study. This shows that B. fragilis phage persist well in seawater, and that they may show similar decay rates to enteroviruses. They are also better suited than coliphage to indicate the fate and transport of viruses in subtropical climates. Final conclusions : 1) Bacteroidesfragilis phage strains B40-8 and B56-3 were found in the environment in our geographic location but in very low numbers. This differs from the findings in Europe but most of the polluted sites sampled in their studies were directly receiving untreated domestic sewage. 2) The sewage survey showed the B56-3 phage strain was found in greater abundance than the human strain B40 8 Both phage strains were consistently found in domestic sewage in the Tampa Bay area. 3) The survival study shows that the B fragilis phage strains survive better in warm seawater than MS2 coliphage, and may be a better research tool in detecting fecal contamination in subtropical environments. 65


4) This still does not completely answer the "animal vs human sources for fecal contamination" question. While the animal/human strain B56-3 does have environmental sources the high levels found in the domestic sewage in the Tampa Bay region may indicate that humans are a significant source in this area. The presence of the human strain B40-8 associated with human enterovirus data will determine if this can be a useful indicator tool Future areas of interest regarding the Bacteroides fragilis phage assay in the Tampa Bay region should include investigation into septic tank systems as a possible source to the environment, and analysis of fecal samples could determine if the local seabird population is contributing to the phage B56-3 found in the ambient water 66


References 1. Lehr J. H ., Gass,Tyler E ,Pettyjohn,Wayne A DeMarre Jack in Domestic Water Treatment (National Water Well Association, Dublin,OH), pp. 2-4. 2 Drinking Water and Health, (1977ed ) Committee, S D W. National Academy of Sciences, Washington, D C., pp. 1-8 3 Edberg, S C., LeClerc, H & Robertson, J. (1997) Critical Reviews in Microbiology 23, 179-206. 4. Pepper, I. L., Gerba,C. P ,Brusseau,M. L. (1997) Pollution Science (Academic Press, San D i ego) 5 USEPA (1998) Bacterial Water Quality Standards for Recreational Waters (Freshwater and Marine Waters (Office of Water United States Environmental Protection Agency Washington D.C.) pp. 1-13. 6 Pancorbo 0 C. & Barnhart, H M. (1992) Journal of Environmental Health 54, 57-63. 7 Yates M V ., Yates, Scott R. (1988) CRC Critical Reviews in Environmental Control17, 307-344. 8 Bitton G (1994) Wastewater Microbiology (Wiley-Liss Publication New York). 9. Kueh, C S W ., Tam, T.-Y., Lee, T., Wong S. L., Lloyd, 0. L., Yu I. T S ., Wong, T W Tam, J. S. & Bassett, D. C. J (1995) Wat e r Science and Technology 31, 1 4 10. Cartwright, R. Y. (1993) Water Science and Technology 27, 195-198. 11. Fleisher J. M Kay, D., Wyer, M.D. & Godfree, A. F (1998) International Journal of Epidemiology 27, 722-726. 12. Gerba, C P. Rose Joan B.,Haas,Charles N (1996) International Journal of Food Microbiology 30, 113-123. 13. Godfree A. Jones, F. & Kay D. (1990) Marine Pollution Bulletin 21,414-422. 67


14. Helmer, R., Hespanhol, I. & Saliba, L. J. (1991) Water Science and Technology 24,35-42. 15. USEPA (1999) EPA Action Plan for Beaches and Recreational Waters (Office of Research and Development and Office of Water, United States Environmental Protection Agency, Washington,D.C), pp. 1-19. 16 Pruss, A (1998) International Journal of Epidemiology 27, 1-9. 17. Fattal, B., Peleg-Olevsky, E. & Cabelli, V J. (1991) International Journal of Environmental Health Research 1, 204-214. 18 Fleisher, J M., Kay, D., Salmon, R. L., Jones, F Wyer, M.D. & Godfree, A. F. (1996) American Journal of Public Health 86, 1228-1234. 19. Virology, I. S. G. o. W. (1983) Water Research 17, 121-132 20. Alexander, L. M., Heaven, A., Tennant, A & Morris, R. (1992) Journal of Epidemiology and Community Health 46, 340-344. 21. Cabelli V J., Dufour A P., McCabe, L. J. & Levin, M.A. (1983) Journal WPCF 55, 1306-1314. 22. Godfree, A. F., Kay,D., Wyer,M D. (1997) Journal of Applied Microbiology Symposium Supplement 83(26), 11 OS-119S. 23. Figueras, M. J. Polo, F., Inza, I. & Guarro, J. (1997) Marine Pollution Bulletin 34, 148-156. 24. USEPA. (1986) Ambient Water Quality Criteria for Bacteria Vol. EPA440/5-84002 25 : Sinton L. W., Finlay, R. K. & Hannah D. J. (1998) New Zealand Journal of Marine and Freshwater Research 32, 323-348. 26. Grabow, W. 0 K., Neubrech, T. E Holtzhausen, C. S. & Jofre J. (1995) Water Science and Technology 31, 223-230. 2 7. Tartera C. & Jofre, J. (1987) Applied and Environmental Microbiology 53, 16321637. 28 Armon, R. & Kott Y (1996) C ritical Revi e ws in Environmental Science and Tec hnology 26, 299-335 29. Dutka B. J El Shaarawi, A., Martins M. T. & Sanchez, P. S. (1987) Water Research 21, 1127-1134. 68


30. Borrego J. J. Morinigo M.A., de Vicente, A., Comax, R. & Romero P (1987) Water Research 21, 1473-1480. 31. Karst M ., Dutkiewicz R., Hahn T. & Botzenhart, K (1991) Water Science and Technology 24,245-250. 32. O'Keefe, B & Green, J. (1989) Water Research 23, 1027-1030 33. Palmateer, G. A., Dutka E. M. Meissner S M & Sakellaris, M G (1991) Water Research 25, 355-357. 34 Reynolds, K. A. Rose J. B. & Giordano, A. T. (1993) Water Scien ce and Technology 27, 115-117. 35. Baudisova, D (1997) Water Sci e nce and Technology 35, 333-336. 36. Gallagher, T P. & Spino D F. (1968) Water Research 2, 169 175. 37. Kreader C A. (1995) Applied and Environmental Microbiology 61, 1171-1179 38. Byappanahalli, M N & Fujioka, R. S. (1998) Water Science and Technology 38, 171174. 39. Rivera S C. Hazen Terry C ., Toranzos Gary A. (1988) Applied and Environmental Microbiology 54, 513-517. 40. Marzouk Y ., Goyal S M & Gerba, C P. (1980) Water Research 14, 1585-1590 41. Rhodes, M W ., Kator Howard (1988) Applied and Environmental Microbiology 54,2902-2907. 42. Ludwig, H. F. & Storrs, P N (1970) Water Research 4, 709-720 43 Borrego J. J. & Romero P. (1985) Water Research 19, 557-562 44. Berry, S A. & Noton B. G (1976) Wat e r Research 10,323-327. 45 Fleisher J M (1991) Journal of Water Pollution Control Federation Lit Review Issu e 63, 259-265. 46 Marino, F. J., Martinez-Manzanares, E., Morinigo, M.A. & Borrego J. J. (1995) Water Science and Technology 31, 27-31 47. Allsop, K & Stickler D J (1985) Journal of Applied Bacteriology 58, 95-99 48. Brook, I. (1995) Journal of M e dical Microbiology 43, 92-98 69


49. Koneman, E. W., Allen,S. D ,Janda,W. M.,Schreckenberger,P C. Winn,W. C. Jr. (1997) Color Atlas and Textbook of Diagnostic Microbiology (Lippincott, New York) 50 Brook, I. (1990) Journal of Applied Bacteriology 69, 697-700 51. Fiksdal L., Maki,James S LaCroix,Steven J., Staley,James T. (1985) Applied and Environm e ntal Microbiology 49, 148-150. 52. Lasobras, J. Muniesa M ., Frias J Lucena F & Jofre, J (1997) Water Science and Technology 35, 129-132 53. Ackermann, H .-W. DuBow,Michael S. (1936) Viruses ofProkaryotes (CRC Press Boca Raton, Florida) 54 Puig, A. Queralt N., Jofre, J. & Araujo, R. (1999) Applied and Environmental Microbiology 65, 1772-1776 55. Havelaar, A H Furuse K., Hogeboom, W M (1986) Journal of Applied Bacteriology 60, 255-262 56 Holdeman, L. V Good l. J., Moore,W E C (1976) Applied and Environmental Microbiology 31, 359-375 57 Puig, A., Jofre, J & Araujo, R. (1997) Water Science and Technology 35, 359362 58 Shoop, D. S., Myers, L. L. & LeFever, J. B. (1990) Applied and Environmental Microbiology 56, 2243-2244. 59. Sun Z. P ., Levi Y Kiene, L., Dumoutier, N. & Lucena, F. (1997) Water,Air and Soil Pollution 96, 175-183. 60 Tartera, C., Lucena,F Jofre,J. (1989) Applied and Environmental Microbiology 55, 2696-2701. 61. Tartera, C., Lucena, F. & Jofre J. (1989) Applied and Environmental Microbiology 55, 2696-2701. 62 Lucena, F ., Muniesa M Puig, A Araujo, R. & Jofre J. (1995) Journal of Virological Methods 54, 121-130 63 Lucena, F., Araujo R. & Jofre J (1996) Water Research 30,2812-2816. 64 Armon, R. & Kott, Y. (1995) Water Science and Technology 31,215-222. 70


65. Araujo, R., Lasobras, J., Puig, A., Lucena, F. & Jofre, J. (1997) Water Science and Technology 35, 125-128. 66. Jofre, J Bosch, A., Lucena, F., Girones, R. & Tartera, C. (1986) Water Science and Technology 18, 167-173. 67. Chung, H. & Sobsey, M D (1993) Water Science and Technology 27,425-428. 68. Jofre, J., Olle E ., Lucena, F. & Ribas, F. (1995) Water Science and Technology 31,6973. 69. Jofre, J ., Olle, E., Ribas, F., Vidal, A. & Lucena, F. (1995) Applied and Environmental Microbiology 61, 3227-3231. 70. Eaton, A D., Clesceri,Lenore S ,Greenburg,Arnold E. (1995) in Standard Methods, ed. American Public Health Association, A. W. W. A., Water Environmental Federation (American Public Health Association, Washington,D C ) 71. Rhodes, M. W Kator,J. ( 1998) Journal of Applied Microbiology 83, 120-126 72. Araujo, R M., Lasobras, J., Lucena, F & Jofre, J (1993) Water Science and Technology 27, 119-122. 73. Tartera, C Araujo, R ., Michel, T & Jofre, J. (1992) Applied and Environmental Microbiology 58, 2670-2673. 74 Yates, M. V., Gerba,Charles P.,Kelley,Lee M. (1985) Applied and Environmental Microbiology 49, 778-781. 75. Fujioka, R. S. Shizumura,L. K. (1985) Journal ofWater Pollution Control Federation 57, 986-992. 76. Griffin, D. W., Gibson,Charles J. Lipp,Erin K.,Riley,Kelley,Paul,John H ,Rose,Joan B (1999) Applied and Environmental Microbiology 65, 4118-4125. 77. Lipp, E. K. (1999) Ph.D Dissertation Department of Marine Science (University of South Florida, St. Petersburg), pp 173 78. Allsop, K., Stickler D. J. (1984) Journal of Applied Bacteriology 56, 15-24. 79 Tartera, C., Jofre,J ., Lucena,F. (1988) Environmental Technology Letters 9, 407410. 80. Araujo, R M Puig,A., Lasobras,J., Lucena,F., Jofre J (1997) Journal of Applied Microbiology 82, 281-286. 71


81. Puig, A., Queralt,Nuria,Jofre Juan,Araujo,Rosa (1999) Applied and Environmental Microbiology 65, 1772-1776 82 Jofre, J., Blasi M.,Bosch,A. Lucena,F. (1989) Water Science and Technology 21, 15-19. 83 Tartera C., Bosch A.,Jofre J (1988) FEMS Microbiology Letters 56, 313-316. 72


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.