Grantee Research Project Results
Final Report: Prevalence and Survival of Microorganisms in Shoreline Interstitial Waters: A Search for Indicators of Health Risks
EPA Grant Number: R828830Title: Prevalence and Survival of Microorganisms in Shoreline Interstitial Waters: A Search for Indicators of Health Risks
Investigators: Rogerson, Andrew , McCorquodale, Don , Estiobu, Nwadiuto
Institution: Nova Southeastern University , Florida Atlantic University - Boca Raton
EPA Project Officer: Hahn, Intaek
Project Period: August 1, 2001 through July 31, 2003
Project Amount: $312,570
RFA: Recreational Water Quality: Indicators and Interstitial Zones (2000) RFA Text | Recipients Lists
Research Category: Aquatic Ecosystems , Water , Ecological Indicators/Assessment/Restoration
Objective:
The overall objective of this research project was to determine whether alternative beach management action should be considered or whether the health risk of indicator organisms in sand is negligible. The specific objectives of this research project are to: (1) document the numbers of fecal organisms (Escherichia coli, enterococci, and fecal coliforms) in beach sand and determine whether they are attached or free in interstitial water; (2) document the number of potentially novel indicator organisms (Staphylococcus aureus, Pseudomonas aeruginosa, Clostridium perfringens, Vibrio, coliphage), and examine the occurrence of eukaryotic microbes with possible health importance; (3) compare the survival of indicator organisms in water versus sand and account for differences; and (4) look for evidence of increased health risk to individuals (particularly children) exposed to beach sand rich in indicator organisms (beach questionnaire).
There are clues in the literature that, when taken together, suggest that the microbiological quality of beach sand may constitute a health risk to bathers, particularly children who spend time in the “swash zone” and in wet sand. Current beach monitoring practices only monitor organisms in the water, yet sand has the potential to accumulate bacteria through filtration and favor the survival of indicator organisms through growth in protected microhabitats. As a first step, we examined the temporal variation of a range of microbial indicators in sand (wet and dry) and water. The study examined traditional fecal indicators (enterococci, E. coli, and fecal coliforms) as well as nontraditional indicators (coliphage, P. aeruginosa, C. perfringens, Vibrio sp., and Staphylococcus aureus).
The parameters affecting the abundance of fecal (or indicator) bacteria in sand are summarized in Table 1. They were examined in an attempt to understand the significance of any accumulations of bacteria in beach sand. Three contrasting beaches were selected to study the relative numbers of fecal and nontraditional indicators in water and sand. Three beaches were sampled between August 2001 and 2003 in South Florida: Hobie Beach in Miami (25°44’ 22.5”N, 80°10’ 18.7”W), Hollywood Beach (26°02’ 02.56”N, 80°06’ 50.36”W), and Fort Lauderdale Beach (26°07’ 17.35”N, 80°06’ 14.24”W). These beaches differed in terms of the number of users frequenting the beach and in their proximity to obvious sources of sewage pollution (e.g., ocean
Table 1. Summary of Main Parameters Influencing the Numbers of Indicator Bacteria Present in Both Wet and Dry Beach Sand. Factors in bold were specifically addressed in this study.
outfalls and contaminated rivers). Sediment characteristics and seawater properties also differed among the beaches.
Project Activities (August 2001-August 2003)
(1) Enumerated traditional beach fecal organisms, E. coli, enterococcus, and fecal coliforms.
(2) Enumerated beach S. aureus, P. aeruginosa, Vibrio, and C. perfringens.
(3) Enumerated coliphage.
(4) Compared Enterolert and Colilert for counting enterococci and E. coli in recreational waters and beaches.
(5) Determined survival of enterococci, E. coli, S. aureus, and C. perfringens in sand and water over time under various environmental conditions (mesocosm experiments).
(6) Determined if bacteria are free or attached on a sandy beach.
(7) Determined microspatial and macrospacial variation of beach microbes.
(8) Determined if air acts as a vector for moving fecal organisms from water to beach sand (aerobiology).
(9) Quantified the filtering capacity of sand to accumulate fecal organisms.
(10) Determined the scale of washout of bacteria from sand into the ”swash zone.”
(11) Investigated the palatability of fecal organisms to micrograzers.
(12) Enumerated enterococci in surface offshore waters (0-3,000 m offshore).
(13) Identified enterococci from sand and surrounding habitats.
(14) Conducted rapid detection of S. aureus, P. aeruginosa, and total bacteria using peptide nucleic acid probes.
(15) Documented free-living and pathogenic beach protozoa.
(16) Enumerated sand yeasts.
(17) Performed complementary molecular methods for the detection of fecal organisms.
(18) Distributed beach questionnaires to identify possible health risks.
Summary/Accomplishments (Outputs/Outcomes):
Regardless of the indicator bacteria in question, densities were always higher in dry sand (above the high tide level) relative to wet sand. Moreover, counts were always higher in sand than in water. Hobie Beach had the highest counts of the three beaches. When the sand data were normalized for water content of the sand (i.e., the counts per g were converted to counts per mL based on the amount of water film trapped in the sand), the densities increased dramatically. Generally, numbers of bacteria in water were concentrated some 20-fold in the wet sand and some 160-fold in the dry sand (data expressed as per g sand). When normalized for water content, however, concentration factors were substantially increased (see Table 2). In the case of enterococci, bacteria were 240 times more abundant in wet sand relative to water and 5,500 times more abundant in dry sand (data expressed as per mL water). Because bacteria reside in the water fraction, it is the amount of water in the sand, rather than the “weight” of the sand particles, that has the most relevance in terms of absolute concentrations of bacteria in the sand environment.
It also should be noted that there was poor correlation between beach counts of enterococci and E. coli when traditional membrane filtration methods were compared with the Enterolert and Colilert methods (correlation coefficients of 0.26 and 0.19). This was partly because of the fact that at higher salinities, counts increased when using the rapid method (especially when cultures were seeded with native bacteria) and the percentage of false positives increased. Overall, bacterial densities by the Colilert and Enterolert methods were higher than values obtained by traditional methods.
The results of the experiments on the mesocosms, microspatial distribution, sand filtration, consumption, “wash out” from sand, and offshore counts all provided information that helped to explain the high indicator bacterial counts in sand and the consequences of these counts for water managers.
Table 2. Mean Number of Bacteria (Data Pooled From All Three Beaches) in the Water, Wet Sand, and Dry Sand. Densities as per 100 mL (i.e., sand counts normalized for water content).
The mesocosm experiments clearly showed that all bacteria (fecal indicators and nonfecal indicators) had improved survival in sand (wet and dry) relative to seawater. In most cases, populations of bacteria increased in the first few days in sand. Surprisingly, growth of fecal bacteria in “dry” sand was not limited by the low moisture content of the sand, which was one-half that of wet sand. The only treatment that promoted longer survival of bacteria in seawater was the addition of nutrients, although the addition of nutrients to sand had no effect. This suggests that there are nutrient microhabitats (pockets) in sand that enhance survival. Consequently, “starving” bacteria in seawater were less tolerant of environmental fluctuations such as temperature and salinity changes.
Mesocosm experiments with grazing protozoa suggested that predation was a major factor influencing the numbers of indicator bacteria found in beach sand. The palatability study showed that micrograzers consumed, and grew on, fecal bacteria. Predation also may account for why fecal bacteria were more abundant in dry sand compared to wet sand. It is likely that many predators were inactive in dry sand because of the limited water film. This agrees with direct observations that showed fewer and smaller predators in dry sand compared to wet sand. Thus, because of reduced predation in the natural dry sand mesocosm experiment (nonsterilized with indigenous bacteria and micrograzers), enterococci and E. coli did not decline to below 5 percent survival. In natural wet sand, however, the levels of enterococci and E. coli dropped to below 5 percent survival in 5 and 2 days, respectively.
In situ mesocosms showed that indicator bacteria were filtered from the water and accumulated in sand. Moreover, the washout of sand indicator bacteria, and the fact that they should be considered “environmental” rather than true indicators of sewage contamination, suggest that their presence in the swash zone could influence the outcome of routine beach testing. Traditionally, the water quality of recreational beaches relies on testing the numbers of indicator bacteria in the water a few meters offshore. The present study has shown, however, that there are considerably more indicator organisms in the zone 0 to 5 m from shore than in the water extending from 5 to 3,000 m offshore. This strongly suggests that the washout of bacteria from sand influences the bacterial counts in the water. Experimentally, we have confirmed that wave action can remove almost all of the indicator bacteria from sand in one or two tidal cycles. In short, rain events, storms, and tidal wave action are probably washing fecal indicator bacteria from the sand to give misleadingly high counts in near-shore waters. This may be resulting in unwarranted beach closures. It must be confirmed that the fecal indicator counts in sand represent harmless “environmental” bacteria rather than significant levels of sewage contamination. This was addressed both directly and indirectly through a beach questionnaire and preliminary molecular work (see conclusions).
Air was not an important vector for the transport of fecal organisms to the beach environment. Fecal bacteria in the upper beach (dry sand) most likely come from storm surges that send waves to the upper reaches of the beach. Other sources may include runoff from urban areas/agricultural lands and from the feces of birds and other animals that can be found on the beach; once fecal bacteria are deposited in dry sand, they can grow and multiply. The multisource nature of fecal indicator bacteria was, to some degree, confirmed by Biolog System identification of enterococci.
In terms of free-living amoebae, 24 morphotypes were isolated on freshwater media, despite the salinity levels of seawater, wet sand, and dry sand. Both wet sand and dry sand are high salinity sites averaging 31 and 23 ppt salt, respectively. Acanthamoeba sp. (some isolates cause amoebic keratitis) was found in one-third of all sand samples examined. The abrasive nature of sand particles may cause damage to the cornea, providing an entry route for acanthamoebae to initiate infection. The possibility that beach sand provides a reservoir for potentially pathogenic, free-living amoebae has never before been considered, probably because it has been assumed that the high salinity water is too harsh an environment for these “freshwater” protozoa. The data, however, must be kept in perspective. Acanthamoeba keratitis is a rare condition with an estimated incidence rate of only 0.33 cases per 10,000 contact lens wearers per year, the group deemed most at risk. But it is interesting that the most common observed genotype from the beach (T4) is the same genotype that has been found in the vast majority of amoebic keratitis (AK) infections.
The possible presence of Cryptosporidium, Giardia, and Entamoeba was investigated. Unfortunately, conventional microscopic methods used to scan cysts of these pathogens could not be used for sand because of the fine particulates in the samples. The detection of these pathogens on the beach by an enzyme-linked immunosorbent assay (ELISA) test (developed for clinical trials) is interesting, but does not prove the existence of intact cysts. Mere fragments of cyst wall material would produce a positive result. Moreover, the failure to culture any Entamoeba histolytica, despite the relatively high percentage of positive ELISA results, suggests that intact cysts are not present in sand, and that these pathogens do not pose a health risk to beach users.
During the 2-year study, some 10,000 surveys were handed out, but only 892 experimental forms and 609 control forms were returned and evaluated (the control subjects had not visited a beach within the previous 9 days). Because of the low return of questionnaires (around 10 percent), data from the three beaches were pooled and illnesses were grouped (gastrointestinal [GI], constitutional, dermatological, upper respiratory). At this time, the Public Health Department of the Health Profession Division of Nova Southeastern University is statistically analyzing the data. The raw data, however, show that of those beach users reporting symptoms, 22 percent were GI, 9 percent were constitutional, 18 percent were dermatological, and 51 percent were upper respiratory. The control group reported similar frequency of illnesses: 26 percent, 10 percent, 11 percent, and 53 percent, respectively. Although these data do not consider beach activity (swimming, play in the sand, etc.) or the age of the respondents, it is evident that overall there are no clear signs of symptoms in the recreational population compared with the control (nonbeach) population.
Conclusions:
(1) Densities of all indicators (fecal and nonfecal) were higher in dry sand (above the high tide line) relative to wet sand (intertidal); both were considerably higher than in seawater.
(2) Hobie Beach (most populated) had higher counts than Fort Lauderdale Beach and Hollywood Beach (least populated).
(3) The presence of people on a beach increased the number of traditional indicators (enterococci and E. coli) in the sand. Although this may have been because of the shedding of organics (skin, etc.) stimulating growth, it also is likely that bacterial growth in the sand was stimulated by the physical disturbance of sand (walking, play, etc.). Beach raking (as practiced at Hollywood and Fort Lauderdale Beaches) may be exacerbating the problem of high sand counts.
(4) Nontraditional indicators (P. aeruginosa, Vibrio, S. aureus, C. perfringens, S- and F- coliphage) gave the same numerical trends as the traditional fecal indicators (E. coli, fecal coliforms, enterococci). For example, S. aureus throughout the year averaged 4.67 x 104 in water, 1.28 x 107 in wet sand, and 1.30 x 109 in dry sand (all counts as per 100 ml water).
(5) In a series of laboratory mesocosm experiments with enterococci, E. coli, S. aureus, P. aeruginosa, and C. perfringens, sand promoted increased survival relative to seawater.
(6) “Natural” mesocosm experiments (with indigenous micrograzers) and palatability studies on fecal bacteria showed that micropredators (heterotrophic protozoa) were important regulators of indicator bacteria in the sand.
(7) Sand was shown to be an effective filter for removing indicator bacteria from the water column.
(8) The majority of indicator bacteria in sand were attached to sand grains (not free in interstitial space), suggesting that they were metabolically active. Indicators were distributed in micropatches again, suggesting in situ growth.
(9) Air was not an important vector for moving indicator bacteria from the swash zone to the upper beach.
(10) “Washout” of indicator bacteria from sand was considerable, suggesting that the swash zone receives significant bacterial inputs from the beach. The numbers of indicator bacteria in the water column fell off markedly from 5 m to 3,000 m offshore (mean enterococci count, 1.5 per 100 mL). This has important implications for water quality testing; perhaps samples should be collected more than 5 m from the swash zone to prevent unnecessary beach closures. Alternatively, if the scale of washout is geographically related, threshold levels might need to be readdressed for subtropical and tropical waters.
(11) The most abundant opportunistically pathogenic beach protozoan was Acanthamoeba sp. (responsible for AK infections). One-third of sand samples processed had acanthamoebae, and growth experiments suggested that populations were actively growing at salinities found in sand. Molecular genotyping showed that the majority of isolates were T4, a common environmental genotype (also the one recovered from the majority of AK infections). The potential risk from eukaryotic microbes at the beach should be further explored.
(12) Detection of enteric and nonenteric pathogens at the beach by polymerase chain reaction showed promise for future risk assessment. Molecular results correlated poorly, however, with traditional indicator counts.
(13) The beach questionnaire showed no clear signs of symptoms in the recreational population compared with the control (nonbeach) population. More rigorous analysis of the data is underway because there were slightly more dermatological symptoms in beach users. Questionnaire returns, however, were low (around 10 percent), and a more comprehensive epidemiological study may be warranted.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
Other project views: | All 21 publications | 2 publications in selected types | All 2 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Booton GC, Rogerson A, Bonilla TD, Seal DV, Kelly DJ, Beattie TK, Tomlinson A, Lares-Villa F, Fuerst PA, Byers TJ. Molecular and physiological evaluation of subtropical environmental isolates of Acanthamoeba spp., causal agent of Acanthamoeba keratitis. Journal of Eukaryotic Microbiology 2004;51(2):192-200. |
R828830 (Final) |
not available |
|
Esiobu N, Mohammed R, Echverry A, Green M, Bonilla T, Hartz A, McCorquodale D, Rogerson A. The application of peptide nucleic acid probes for rapid detection and enumeration of eubacteria, Staphylococcus aureus and Pseudomonas aeruginosa in recreational beaches of S. Florida. Journal of Microbiological Methods 2004;57(2):157-162. |
R828830 (Final) |
not available |
Supplemental Keywords:
protozoan pathogens, beach contamination, water quality, aerobiology, microbiology, public good, health, water, biology, ecology and ecosystems, environmental microbiology, environmental monitoring, health risk assessment, recreational water, susceptibility, sensitive population, genetic susceptibility, Escherichia coli, E. coli, bacteria, beach contamination, children, enterococci, environmental hazard exposures, exposure, fecal coliform, human health risk, indicator organisms, microbes, microorganisms, pathogens, recreational beaches, recreational water monitoring, sand particles, shoreline interstitial water,, RFA, Health, Scientific Discipline, Water, Health Risk Assessment, Susceptibility/Sensitive Population/Genetic Susceptibility, Environmental Microbiology, Environmental Monitoring, genetic susceptability, Ecology and Ecosystems, Biology, Recreational Water, pathogens, recreational water monitoring, sensitive populations, bacteria, E. coli, recreational beaches, exposure, microbes, microorganisms, children, fecal coliform, water quality criteria, shoreline interstitial water, beach contamination, water quality, enterococci, indicator organisms, human health risk, sand particlesProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.