Present address Walter E. Rhodes, US Fish and Wildlife Service/DMBM-MBS, Laurel, MD 20708, USA.
Michelle A. Johnston, Department of Environmental Health Sciences, Arnold School of Public Health, 921 Assembly Street, Columbia, SC 29208, USA. E-mail: email@example.com
Aims: To determine whether American alligators (Alligator mississippiensis) are an unrecognized poikilothermic source of faecal coliform and/or potential pathogenic bacteria in South Carolina’s coastal waters.
Methods and Results: Bacteria from the cloaca of American alligators, as well as bacteria from surface water samples from their aquatic habitat, were isolated and identified. The predominant enteric bacteria identified from alligator samples using biochemical tests included Aeromonas hydrophila, Citrobacter braakii, Edwardsiella tarda, Escherichia coli, Enterobacter cloacae, Plesiomonas shigelloides and putative Salmonella, and these were similar to bacteria isolated from the surface waters in which the alligators inhabited. Based on most-probable-number enumeration estimates from captive alligator faeces, faecal coliform bacteria numbered 8·0 × 109 g−1 (wet weight) of alligator faecal material, a much higher concentration than many other documented endothermic animal sources.
Conclusions: A prevalence of enteric bacteria, both faecal coliforms and potential pathogens, was observed in American alligators. The high faecal coliform bacterial density of alligator faeces may suggest that alligators are a potential source of bacterial contamination in South Carolina coastal waters.
Significance and Impact of the Study: These findings help to increase our understanding of faecal coliform and potential pathogenic bacteria from poikilothermic reptilian sources, as there is the potential for these sources to raise bacterial water quality levels above regulatory thresholds.
Because it is difficult and costly to detect each pathogen and parasite that may be present in a variety of surface waters, a number of culturable indicator bacteria are used to suggest the presence/absence of waterborne pathogens to standardize water quality monitoring in aquatic environments throughout the United States (USEPA 1986; Scott et al. 2002; Moore et al. 2005). In South Carolina, current indicator bacteria include culturable faecal coliforms for general primary (watershed) contact recreation and shellfish-harvesting waters, Escherichia coli for primary contact recreation in fresh water, and enterococci for primary contact recreation in some marine waters (Scheuler and Holland 2000; SCDHEC 2004a, 2007). Faecal coliform bacteria are Gram-negative, facultatively anaerobic, nonspore-forming bacilli that ferment lactose with gas production (Brown 2005). They belong to the Enterobacteriaceae family, commonly known as enteric bacteria, meaning they reside in the gastrointestinal tract and are usually shed from the body in faecal material. Faecal coliforms appear in great quantities in the gastrointestinal tracts and faeces of people and endothermic, or warm-blooded, animals (Scott et al. 2002; Simpson et al. 2002; Anderson et al. 2006; Mohapatra et al. 2007).
High concentrations of indicator faecal coliform bacteria are a chief pollutant in South Carolina, causing bacterial levels in waterbodies to rise above regulatory standards, increasing health hazards to contact recreational users and shellfish consumers (SCDNR 2004; SCDHEC 2007). This is a notable concern in the southeastern United States, where recreational waters and shellfish harvests have substantial economic value, but are restricted or closed because of the elevated levels of indicator bacteria. Often, these waters border remote coastal marsh areas with no obvious source of faecal contamination. Over 40% of the oyster shellfish grounds in South Carolina were conditionally or permanently closed to harvest during the 2004/2005 season, the majority of which were a result of water quality concerns regarding indicator faecal coliform levels where no major pollution sources were identified (SCDHEC 2007).
The concept of indicator organisms is a principal component of water quality standards and regulatory microbiology. However, the concept has major limitations because it oversimplifies the complex dynamics of microbial ecology and is unable to link organisms associated with faecal contamination to their potential sources (Stewart-Pullaro et al. 2006). For example, many states recommend the use of faecal coliforms or E. coli as a pathogen indicator, even though these bacteria have been shown to live outside of host environments as part of the natural microflora in nutrient-rich subtropical habitats (Winfield and Groisman 2003). It has also been suggested that E. coli from faecal sources are not able to survive outside of host environments as long as specific pathogens, such as Salmonella spp., questioning its validity as a reliable indicator (Rhodes and Kator 1988; Winfield and Groisman 2003). Indicators are not equivalent to pathogens, and therefore there is no guarantee that indicators and pathogens will always coexist. As a result of these concerns, the United States Environmental Protection Agency (USEPA) has recently suggested the re-evaluation of current bacterial pathogen indicators and the use of alternative indicators, or the establishment of advanced ways to detect specific pathogens themselves in surface waters (USEPA 2007).
Another assumption surrounding faecal indicator organisms is that current water quality testing standards and bacterial source tracking methods are based on the premise that the main sources of faecal coliform bacteria (especially E. coli) inhabit the gastrointestinal tract and faeces of endothermic animals (Harwood et al. 1999). However, molluscs and insects have been known to harbour faecal coliform bacteria and potential pathogens, as have some poikilotherms (i.e., animals whose body temperature is variable and fluctuates with that of the environment), generically known as cold-blooded animals (Harwood et al. 1999; Kloot et al. 2006). If water quality is to be properly assessed, it is useful to determine the hosts of the indicator bacteria present in surface waters and the potential pathogens associated with them to construct a source material budget of bacterial inputs, as management strategies from human sources are likely to be different than those from animal sources (Hagedorn et al. 1999; Guan et al. 2002; Kelsey et al. 2003; Cox et al. 2005). In spite of this, information regarding the contribution of faecal coliform bacteria from poikilothermic animals to coastal surface waters is limited, even though these animals, such as the American alligator (Alligator mississippiensis), are numerous on the southeastern coast of the United States.
American alligators are reptiles that inhabit fresh and brackish waters including rivers, swamps, streams, lakes, creeks, tidal marshes and ponds from North Carolina to Texas (Rhodes 1998). As a result of habitat loss and increasing population densities because of their protection as a threatened species, alligators often inhabit golf course and stormwater retention ponds, raising the potential for human interaction (Wilkinson and Rhodes 1997; Rhodes 1998). As a result of seasonal temperature variations and opportunistic eating habits, alligators have intermittent defecation rates. Therefore, little information is known about bacterial loading in waterbodies from alligator faeces, even though they are commonly found in large numbers in South Carolina coastal environments.
The objectives of this research were to (i) establish whether American alligators harbour faecal coliform bacteria, (ii) identify faecal coliform bacteria and/or potential pathogens from alligator faecal material, (iii) determine the density of faecal coliform bacteria in alligator faeces and (iv) determine whether alligator faecal bacteria could contribute to water quality degradation in the South Carolina coastal zone. Because the faeces of endothermic animals are thought to be the primary source of faecal coliform and pathogenic bacteria in coastal waters, discovering the predominant bacteria excreted from alligators will significantly improve our knowledge of the bacteria harboured by poikilothermic animals. By evaluating identification methods associated with alligator faecal bacteria, results may be helpful to managers interested in establishing water quality standards and best management practices to support both public and animal health.
Materials and methods
Study area description
Sampling took place at Palmetto Bluff in Bluffton, South Carolina, a coastal development located between Hilton Head Island, South Carolina, and Savannah, Georgia, USA (Fig. 1). Although the development has a recreational golf course, marina and homesites, much of the land remains undeveloped, bordering important marsh and shellfish habitat near the May River. Palmetto Bluff was selected as the study site, because it contains various waterbodies that serve as habitat for American alligators, such as stormwater retention ponds, freshwater lagoons and a brackish, relic rice field marsh impoundment. Various waterbodies surrounding Palmetto Bluff have been placed on the EPA’s Clean Water Act Impaired Water Body 303(d) List because of faecal coliform level violations from unknown sources (SCDNR 2004).
In adherence with the South Carolina Department of Natural Resource catch and release permitting standards, 31 American alligators were sampled from various waterbodies at Palmetto Bluff in June of 2006 (n =14) and August of 2006 (n =17). Sampling took place in the summer, because alligators are most active during the warm months of the year.
To obtain a cloacal swab from each alligator, the sound of a juvenile alligator distress call was simulated on a pond bank. As the targeted alligator instinctually followed the sound and swam towards the edge of the water, a weighted treble hook on a fishing rod was cast, or a snare pole was looped around the neck (depending on the alligator’s proximity to the bank), and the alligator was pulled onto the bank of the pond where its mouth was secured with electrical tape. A towel was draped over the eyes to reduce the stress level on the animal. Once secure, the alligator was turned over on its side, the area around the cloacal opening (i.e., orifice for reproduction and excretion) was rinsed with deionized water, and a sterile cotton-tipped transport swab (Fisher Band; Fisher Scientific) was inserted into the cloaca. The swab was then placed into its provided holding tube and immediately stored on ice until laboratory analysis (within 24 h). The alligator was sexed (male, female or undetermined hatchling), measured for snout-to-tail length (cm) and marked with a dot of spray paint on the neck to prevent resampling. In preparation for release, the towel and tape were removed, and the alligator was released back into the pond from which it was detained. The maximum handling time for any individual alligator never exceeded 10 min.
The majority of the alligators sampled were males (52%), while the remainder included females (35%) and hatchlings (13%) of indeterminate sex because of their small size. The alligators sampled ranged in size from 0·5 to 2·7 m, with an average length of 1·8 m. The majority of the alligators sampled were located in a relic rice field impoundment, while the habitat from which the second most number of alligators sampled was the golf course retention ponds (Table 1). Water samples were also obtained from the seven waterbodies containing the selected alligators for this study in accordance with the surface water sampling procedures of the American Public Health Association Standard Methods for the Examination of Water and Wastewater (APHA 1998).
Table 1. Location of sampled alligators and their corresponding aquatic habitat. The measured faecal coliform densities in each waterbody are listed as faecal coliform CFU 100 ml−1
Faecal coliform CFU 100 ml−1
*Exceeds standards for shellfish harvesting (<14 CFU 100 ml−1).
†Exceeds standards for primary contact recreation (<200 CFU 100 ml−1).
Rice field pond
Golf course pond 1
Golf course pond 2
Golf course pond 3
Stormwater pond 1
Stormwater pond 2
The 31 cloacal swabs were streaked for isolation onto MacConkey agar plates (MAC; Difco) to select for lactose-utilizing Gram-negative bacteria and onto Hektoen Enteric agar plates (HE; Difco) to select for faecal coliform enteric bacilli. After 18–24 h of incubation at 37°C, well-isolated, bacterial colonies of varying morphology from each streaked swab sample were archived in 20% glycerol/80% tryptic soy broth (TSB; Difco) at −80°C for future examination. A total of 310 cloacal isolates were collected from the 31 alligators.
For the water samples from each sampled waterbody, two replicate aliquots of 50 ml were filtered through separate 0·45-μm pore-size, 47-mm diameter membrane filters (Fisher Brand; Fisher Scientific) (APHA 1998). The filters were placed on MAC and HE agar plates and incubated at 37°C for 18–24 h. After incubation, representative bacterial colonies from the water samples were picked from the filters and restreaked to MAC and HE agar plates for isolation and incubated at 37°C for 18–24 h. After incubation, well-isolated, bacterial colonies of varying morphology from each water sample were archived in 20% glycerol/80% TSB at −80°C for future examination. A total of 100 isolates were collected from water samples.
Each of the 410 archived bacterial isolates was streaked from its storage tube onto tryptic soy agar plates (TSA; Difco) and incubated at 37°C for 18–24 h. Following incubation, colonies were identified to the species level using Analytical Profile Index (API) 20E biochemical test strips for Enterobacteriaceae and other non-fastidious Gram-negative rods in accordance with the manufacturer’s directions (bioMérieux 2006). Results were compared to bioMérieux’s API Online Profile Recognition System database to identify the cloacal and water sample isolates. Based on the API 20E protocol, isolates with a percent identification <80% are not reliable; therefore, a percent identification above 80% was considered a reliable identification as calculated by the Online Recognition System.
Faecal coliform bacteria from the water samples were enumerated using standard membrane filtration techniques with mFC agar plates (USGS 1987; APHA 1998). The samples were filtered at volumes of 5, 10, 25 and 50 ml through separate 0·45-μm pore-size, 47-mm diameter membrane filters (Fisher Brand; Fisher Scientific). The filters were placed on mFC agar plates (mFC; Difco), sealed in water-tight plastic bags and submerged in a 44·5°C water bath for 24 ± 2 h using standard techniques (APHA 1998). After incubation, all blue colonies (signifying lactose utilization by faecal coliforms) were counted and recorded as faecal coliform colony forming units per 100 ml for each sample (CFU 100 ml−1).
Because of the difficulty in collecting alligator faeces and the unknown length of time after defecation if found in the environment, enumeration of alligator faecal material could be highly variable. To address this issue, a fresh faecal sample was obtained from an aquarium enclosure containing three captive alligators at the Cypress Gardens Aquarium and Reptile Complex, in Moncks Corner, South Carolina. Bacteria from the alligator faecal sample was enumerated using multiple dilution most-probable-number (MPN) analysis to estimate alligator faecal material density following standard methods (APHA 1998). A 1-g portion (wet weight) of alligator faeces was diluted in 99 ml of 0·1% sterile Peptone water (Peptone; Fisher Scientific) to make a 1 : 100 dilution of the faecal material. Volumes were dispensed through a 10−9 dilution series in A-1 single-strength and double-strength medium (A-1; Difco) for detecting faecal coliforms. After incubation at 35°C for 3 h, then 44·5°C for the remaining 21 ± 2 h, culture tubes were recorded as positive for faecal coliforms when gas and turbidity were present. Based on the number of tubes positive for faecal coliforms, a MPN of faecal coliform bacteria was computed from a probability table (APHA 1998).
To compare bacterial identifications from alligators and surface waters, similarity coefficients were calculated using the Jaccard index to compare the similarity between sample sets (Real and Vargas 1996). Jaccard’s index of similarity was calculated by dividing the number of common bacterial species in the alligator cloacal sample and the water sample sets by the number of bacteria that occurred in both sample sets. The result was multiplied by 100 to give the percent of bacterial species found in both the alligator and the aquatic habitat from which it was captured.
The API 20E biochemical tests were able to successfully identify 83% of alligator cloacal isolates and 76% of water isolates. A 100% identification rate was not achieved because of the fact that identifications below 80% were not used in statistical calculations of data. Within the alligator isolates, 19 species of enteric bacteria were identified. The most frequently isolated bacteria from the alligator cloacal samples included Plesiomonas shigelloides, Edwardsiella tarda and Aeromonas hydrophila. The bacteria E. coli, Citrobacter braakii, Enterobacter cloacae, putative Salmonella and Klebsiella pneumonia were also isolated, but at lower prevalence rates (Fig. 2a).
Within the water sample isolates, 13 species of enteric bacteria were identified. The most frequently isolated bacteria from the water samples included Aer. hydrophila, Ent. cloacae and Cit. braakii. The bacteria Kl. pneumonia, Ple. shigelloides, Cit. freundii, putative Salmonella and E. coli were also isolated at lower prevalence rates (Fig. 2b).
There were ten bacterial species common to the alligator and water isolates out of the 22 total species of enteric bacteria identified, resulting in an overall similarity coefficient of 45% for the pooled isolates. Of these, the most common species identified, all of which are faecal coliforms, were Aer. hydrophila, Cit. braakii, Ent. cloacae, E. coli, Ple. shigelloides and putative Salmonella (Fig. 2). Individually, the percent similarity of bacterial isolates from alligators compared to the water isolates also found in their corresponding aquatic habitat were 40% for the isolates from the 16 alligators in the rice field pond, 22–44% for the isolates from the eight alligators in the three golf course ponds, 36% for the isolates from the three alligators in the lagoon pond and 25–60% for the isolates from the four alligators in the two stormwater ponds.
Based on the MPN estimate from the captive alligator faecal sample, faecal coliform bacteria numbered 8·0 × 109 g−1 of faecal material, a much higher concentration than many other documented endothermic animal sources (Table 2).
Table 2. Density of faecal coliform bacteria in animal faeces
Faecal coliform [density per gram (wet wt) faeces]
All the water samples collected, with the exception of one of the golf course ponds, did not exceed the South Carolina Department of Health and Environmental Control’s primary contact recreational criteria for faecal coliform levels (<200 CFU 100 ml−1) (Table 1).
To our knowledge, this is the first study designed to determine whether American alligators are an unrecognized poikilothermic source of faecal coliform and/or pathogenic bacteria in surface waters. Despite the fact that American alligators are a federally threatened species ‘due to similarity of appearance to other crocodilians’ under the Endangered Species Act, they are numerous in southeastern coastal environments. There are currently at least 100 000 alligators in the state of South Carolina (Rhodes 1998; Waters et al. 2007). The results confirm a prevalence of enteric bacteria, including both faecal coliform and potential pathogens, observed in alligators. This finding may suggest that the presence of alligator faecal material could have an impact on water quality, especially in coastal areas near recreational and shellfish waters, where alligator populations are densely concentrated.
The present water quality testing standard is based on the principle that faecal coliforms, specifically E. coli, are the inhabitants of the gastrointestinal tracts and faeces of primarily endothermic animals. Ectothermic and poikilothermic animals are seldom considered to be a probable source of these bacteria in surface waters, even though they are frequent residents of coastal habitats. The results of this study demonstrate that enteric bacteria, including several faecal coliforms such as E. coli, Enterobacter spp. and Citrobacter spp. and potential pathogens, such as Kl.pneumonia and putative Salmonella, can be isolated from American alligator faecal material.
Water samples were collected from seven waterbodies (some of which are used for fishing and recreation within Palmetto Bluff) serving as habitat for the sampled alligators. All the water samples collected, with one exception, satisfied the South Carolina Department of Health and Environmental Control’s primary contact recreational criteria (geometric mean of faecal coliforms <200 CFU 100 ml−1) (SCDHEC 2004b) (Table 1). Of the water samples collected, only golf course pond 2 exceeded the allowable amount of faecal coliform bacteria for contact recreational waters by containing 2600 CFU 100 ml−1. Other locations, such as the relic rice field pond, had elevated faecal coliform levels, but not high enough to exceed the contact recreational criteria. While these locations are not shellfish-harvesting waters, four of the seven areas sampled exceeded the geometric mean shellfish standard for the allowable amount of faecal coliform bacteria (<14 CFU 100 ml−1) (SCDHEC 2004b, 2007) (Table 1). This may be significant if run-off from these waters flow into nearby marsh areas where filter-feeding shellfish grow.
The most common enteric bacteria identified from both the alligator and the water samples were Aer. hydrophila, Cit. braakii, Ent. cloacae, E. coli, Ple. shigelloides and putative Salmonella (Fig. 2). These are faecal coliforms or potential pathogens that may cause fever, chills, abdominal pain, nausea, diarrhoea and/or vomiting especially in young children, the elderly or immunocompromised individuals (Corrente et al. 2004; Moore et al. 2005). For example, various strains of Salmonella cause gastroenteritis, while Aer. hydrophila is a pathogen that has been directly linked with massive fish kills in Florida (Gorden et al. 1979; Corrente et al. 2004). The co-occurrence of the same bacterial species in alligators and adjoining surface waters may suggest that the alligators represent a source of bacterial pollution and contribute to water quality degradation; however, further research using molecular bacterial source tracking methods is needed to definitively investigate the contribution of faecal coliform bacteria from poikilothermic animals as bacterial isolates from water samples may stem from other resident wildlife such as wading birds, raccoons, etc.
The estimated high bacterial density of the captive alligator faecal sample may suggest that wild alligators could potentially contribute high concentrations of faecal coliforms to South Carolina coastal waters. Based on the resulting MPN estimates, the faecal coliform bacteria numbered 8·0 × 109 g−1 (wet weight) of alligator faeces, which is a much higher concentration than many other documented endothermic animal sources, including humans, ducks and dogs (Scheuler and Holland 2000; Cox et al. 2005) (Table 2). Since wild alligators do not defaecate regularly because of diet and seasonal behaviours, it is difficult to estimate a daily load of faecal bacteria generated from alligators. However, as captive alligators are thought to defecate once per week (K. Smith, Cypress Gardens, SC, personal communication), wild alligators may contribute a weekly load of c. 8·0 × 109 faecal coliform bacteria per gram of faeces. This calculation should be considered an estimate for wild alligators, as there is little information regarding defecation rates and faecal coliform concentrations from noncaptive alligators because of variations in diet, feeding frequency, etc. Unfortunately, this calculation could not be duplicated as only one captive alligator faecal sample was permitted for this study.
It can be argued that even though alligator faecal coliform concentrations are very high, they may not contribute as much bacteria to a waterbody as, for example, a dog that defecates everyday. However, a rain event may be required to transport land-based faeces to surface waters, where as alligators may defecate directly into the water. The average monthly precipitation for Bluffton, South Carolina, from May to September is c. 16 cm; therefore, if a significant rain event creating pathogen-laden stormwater run-off takes place periodically, the output of faecal coliforms from endothermic animals into surface waters may be somewhat similar to the amount from alligators that may defecate as often as once per week. During periods of drought, this would not be the case as alligator inputs would be continuous where as endothermic animal inputs would not occur until the next rain event. Other factors to consider besides intermittent faecal shedding are the faecal degradation rates that may vary significantly throughout the year.
The survival rate of the alligator isolates in coastal waters was not calculated, as this study was specifically attempting to determine the presence or absence of these bacteria as inhabitants of the gastrointestinal tracts of alligators. However, it has been established that the survival of faecal coliform bacteria, such as the indicator E. coli and other coliform bacteria, is dependent upon various factors such as temperature and salinity (Anderson et al. 2005). Under favourable conditions, E. coli has been demonstrated to survive outside of host environments as part of the natural microflora, especially in warm, subtropical habitats similar to the South Carolina coast (Solo-Gabriele et al. 2000; Winfield and Groisman 2003). More importantly, it has also been suggested that E. coli from faecal sources are not able to survive outside of nonhost environments as long as specific pathogens, such as Salmonella spp. (Rhodes and Kator 1988; Winfield and Groisman 2003). If potential pathogens are able to outlast indicator organisms in unfavourable environmental conditions, then the reliability of current water quality indicators should be questioned. The identification of E. coli, Klebsiella and putative Salmonella associated with the alligator gastrointestinal tract is a potential public health concern and should be further investigated, as more research is needed to improve our understanding of the fate and transport of faecal coliforms and potential pathogens in the environment.
Because it is challenging to acquire faecal samples from aquatic reptiles, this study relied mainly on the cloacal swabbing method of sampling. There are apparent disadvantages with this sampling method, as the cloaca excretes urine, faeces and eggs, which creates variability depending on the recent activities of the sampled animal. However, recent oviposition from the females sampled was unlikely, as sampling took place after mating season. Faecal coliforms were detected on all the collected swabs from the hatchlings of undetermined sex, female and male alligators sampled.
To conclude, this study demonstrated that American alligators found within the South Carolina coastal zone harbour enteric bacteria including indicator faecal coliforms and potential pathogens. The fact that reptiles excrete faecal coliform bacteria is significant, because alligators live in habitats where their contribution to the pool of faecal coliform bacteria may impact humans. Large alligator populations with high faecal coliform densities occupying aquatic habitats may suggest that these poikilothermic animals are an important contributor of faecal bacteria to coastal areas, contrary to common water quality testing paradigms.
This publication represents manuscript contribution no. 1583 of the Belle W. Baruch Institute for Marine and Coastal Sciences. The research described within was supported, in part, by a grant from the National Oceanic and Atmospheric Administration (grant no. NA05NOS4261154) awarded to the University of South Carolina and the South Carolina Sea Grant Consortium and NOAA Center for Coastal Health and Biomolecular Research Student Fellowship Program.
The authors acknowledge the National Oceanic and Atmospheric Administration Center for Coastal Environmental Health and Biomolecular Research, the South Carolina Department of Natural Resources, the Palmetto Bluff Conservancy, Critter Management Inc., the Cypress Gardens Aquarium and Reptile Center, the University of South Carolina Molecular Microbial Ecology Laboratory and the University of South Carolina Belle W. Baruch Institute Geographic Information Processing Laboratory for their significant contributions. This research complied with South Carolina Department of Natural Resource guidelines for alligator handling.
The National Ocean Service (NOS) and the University of South Carolina (USC) do not approve, recommend or endorse any proprietary product or material mentioned in this publication. No reference shall be made to NOS or USC, or to this publication furnished by NOS or USC, in any advertising or sales promotion which would indicate or imply that NOS or USC approves, recommends or endorses any proprietary product or proprietary material mentioned herein or which has as its purpose any intent to cause directly or indirectly the advertised product to be used or purchased because of NOS or USC publication.