The Effects of Environmental Stressors and the Pathogen on Survival and Health of Juvenile Freshwater Turtles Francis J. Polakiewicz and Rachel M. Goodman Department of Biology, Hampden-Sydney College, Hampden-Sydney, VA 23943 By exposing wildlife to pesticides and fertilizers, humans may be indirectly increasing host susceptibility to pathogens. In order to further our understanding of the interaction between the ranavirus and anthropogenic factors, their effects need to be studied using reptiles. Observation of the effects on reptilian immune function in the presence of herbicides is important to show how wildlife populations may be affected by pathogens in the mildly polluted watersheds of central Virginia. In this study, 200 juvenile Red Eared Slider Turtles (Trachemys scripta elegans) were used to investigate the response of aquatic reptiles to ranavirus and herbicide exposure. After an experimental group of 100 turtles was exposed to a ranavirus, the change in mass of the turtles was measured over a six week period under exposure to Roundup Pro, ShoreKlear, Atrazine, and tap water in groups of 40 turtles per treatment. The study suggested that survival of sliders was not affected by exposure to the ranavirus or the herbicides. INTRODUCTION Newly discovered pathogens, those that have switched hosts or expanded in distribution, can have important impacts on both humans and wildlife. Emerging pathogens within the genus, es (in the family Iridoviridae), are carried by and can be fatal to ectotherms including amphibians, reptiles, and fish (Gray et al. 2009). Although ranaviruses have been known to impact fish in wild and domestic populations for decades, they were previously unknown to impact other wildlife populations (Whittington et al. 2010). Thus, the virus has not been researched extensively by virologists or wildlife biologists outside of fish until the past two decades (Chinchar 2002; Marschang & Miller 2012). Recently, outbreaks of have occurred in several amphibian species and may have devastating impacts on the biodiversity of these organisms (Chinchar 2002). Reptiles are another class of ectotherms that are affected by the ; the virus has been isolated from turtles, tortoises, snakes, and lizards (Marschang et al. 1999, Belzer & Seibert 2007). A ranavirus was isolated from the liver and stomach of a domestic female leaf-tailed Gecko (Uroplatus fimbriatus) by Marschang, et al. (2005). There have also been recorded cases of ranavirus in green pythons (Hyatt 2002), although the virus has not been surveyed in wild populations of snakes. These reports of infection indicate that impacts of the virus on wild populations and/or detection ability among the scientific community are on the rise. es are double-stranded DNA viruses with an icosahedral shape that is visible in the cytoplasm of host cells (Chinchar 2002). infection is not always lethal for the host species. A series of nonlethal infections on frogs performed by Gray et al. (2009) suggested that some exposed individuals were able to survive subsequent exposures of due to fortified immune systems. Because exposure to the virus does not always cause mortality, this may in part explain why some populations that are exposed to ranavirus remain asymptomatic, while others experience massive die-offs (Lesbarreres et al. 2011). This variation may be due to genetic differences in viral strains or in host susceptibility, weakened immune systems in the hosts, or other factors. Decreased immunity in some vulnerable populations has been hypothesized to be due to exposure to anthropogenic stressors (Gray et al. 2009; Chen & Robert 2011). Stressors are defined as any variables that cause excessive production of the stress hormones known as corticosteroids which help organisms survive in the short term. However, continuous release of corticosteroids over longer periods can cause a sharp decrease in immune function (Hill & Wyse 1989). By exposing wildlife to synthetic chemicals including pesticides and fertilizers, humans may be indirectly increasing host susceptibility to pathogens. Pesticides, specifically organochlorines, can directly cause fatality in reptiles (Hall 1980). Two herbicides that are extensively used in North America are Atrazine (2-chloro-4-ethylamino-6- siopropylamino-s-trazine) and glyphosate-based herbicides (De Solla et al. 2005; Giesy et al. 2000). A study comparing the effects of Roundup (glyphosate plus the surfactant polyethoxylated tallowamine, orpoea) done by Giesy and colleagues (2000) found that Roundup was relatively toxic to aquatic ectotherms; however the same glyphosate-based herbicide without surfactant had minimal toxicity. Lab and field trials by Hayes et al. (2002) linked Atrazine to the morphological feminization of male leopard frogs. Atrazine also caused an increase in thyroxine in the plasma and decreased growth of exposed Tiger salamanders http://sciencejournal.hsc.edu 1
(Ambystoma tigrinum) Larson et al. 1998). Another herbicide that is commonly used in the United States is 2,4-D which mimics a plant hormone that controls growth (Garner, 1957). Willemsen and Hailey (2000) performed a field study in which 2, 4-D and 2,4,5-T were applied to the fauna in a Testudo hermanni (tortoise) habitat annually over 20 years. Their findings showed that the tortoises exposed to herbicides had, swollen eyes, fluid discharge from the nose and immobility, and that juveniles were more at risk than older age groups in the population (Willemsin and Hailey 2000). A study by Willimgham (2003) of how endocrine-disrupting compounds affected Red Eared Slider turtles (Trachemys scripta elegans) suggested that even low doses of estrogen mimics caused a significant number of males to be feminized during embryogenesis. To date, all research on the interaction between synthetic chemical stressors and vulnerability to ranaviruses has been focused on fish and amphibians. In order to further our understanding of the interaction between the ranavirus and anthropogenic factors, their effects need to be studied using reptiles. Observation of the effects on reptilian immune function in the presence of herbicides is important to show how wildlife populations may be affected by pathogens in the mildly polluted watersheds of central Virginia. In this study, the Red Eared Slider Turtle (T. s. elegans) was used to investigate the response of aquatic reptiles to ranavirus and herbicide exposure. METHODS The experiment was conducted in a laboratory at Hampden-Sydney College from May 18-July 2, 2012. Juvenile Red Eared Sliders, ranging from 4.7-8.3 g in mass and 27.8-35.0 mm in carapace length, were shipped from a commercial supplier in Florida to the lab at Hampden Sydney College (www.turtleshack.com). Ten turtles were immediately euthanized humanely via injection of sodium pentaboarbitol. Necropsies were performed in which liver, spleen and intestine samples were taken. Samples were sent to Dr. Debra Miller at the Center for Wildlife Health and Biomedical and Diagnostic Sciences at the University of Tennessee, Knoxville. Polymerase Chain Reaction (PCR) was used to make sure the turtles were not already carrying. The PCR was performed by the use of protocol and primer sets (MCP4 and MCP5) as found in Mao et al. (1996, 1997). Approximately 500 base pair sequences for the major capsid protein gene found in the ranavirus DNA were targeted. The products of the PCR were determined using electrophoresis on a 1.0% agrose gel. The 200 juvenile turtles were placed in 5.7L Sterillite plastic tubs containing 500mL of tap water (approx. 1cm of standing water). The turtles were divided and placed five per shelf. Each shelf had one control turtle that received tap water throughout the experiment and four turtles, each exposed to one of the following herbicides Roundup Pro (glyphosate plus a surfactant), ShoreKlear (glyphosate but no surfactant), 2,4-D, and Atrazine. Of 40 shelves in total, half of the shelves and the turtles contained therein were exposed to ranavirus. To minimize the potential bias of any environmental effects in the rooms while controlling possible contamination between tubs, alternating shelves contained ranavirus-absent and ranavirus-exposed turtles (as visualized in Table 1 below). The tubs on each set of two shelves were exchanged every week. Shelf 1 Shelf 2 No Chem No Chem - Round Up Round Up - Shore Klear Shore Klear - Atrazine Atrazine - 2,4-D Table 1. Visual representation of shelf order of sliders that was kept throughout the experiment. 2,4-D Rananvirus- Throughout the experiment, any tap water used on a given day to make solutions was left in a container overnight and left open to air to allow chlorine gas to escape. Plastron and carapace length were measured in millimeters using electronic calipers. On May 21, the following chemical treatments were created using commercially available herbicides: Roundup Pro 134.8 ul/6 L water (4,000mcg a.e./l), ShoreKlear 10 ul/6 L water (400 mcg a.e./l), Atrazine 2.64 ul/6 L water (10 mcg a.e./l), and 2,4-D 300.80 ul/ 6 L water (30 mcg a.e./l). During this week, tail tips were removed using sterile scalpels and frozen for later detection of initial ranavirus infection, if necessary. On each processing day, 50 turtles were weighed in grams using an analytical balance. The tubs from the previous week were replaced with new ones containing 500ml of the respective chemical treatments, or plain water in the case of controls. Each week, tubs on shelves containing ranavirusexposed turtles were exchanged with tubs on the nearest shelves containing non-exposed turtles so as to eliminate any bias due to differing environments between shelves (for example, tubs in shelf 1 would be replaced with those in shelf 2 in Figure 1, and vice versa). Turtles were fed 10 turtle pellets of Hatchling Turtle Food (from www.turtleshack.com) every other day for the entirety of the experiment. The turtles without http://sciencejournal.hsc.edu 2
ranavirus were fed first to avoid contamination, and the food sources from the ranavirus control group and the ranavirus exposed group were kept in separate containers. The mixing of chemical treatments and handling of turtles as described above was repeated with 50 turtles on each day of May 22-24 resulting in initial setup of 200 turtles in tubs. exposures began during May 27- May 30, 2012. Turtles were processed each day as follows: 25 control turtles were processed in the morning, and 25 experimental turtles were exposed to the ranavirus and processed in the afternoon. This order of processing control groups first was used for the remainder of the experiment to avoid contaminating the control group. We disinfected all equipment and furniture at the end of each day with 1% Nolvasan (Chlorhexidine diacetate left on surface/material for five minutes). All water buckets, stirring spoons, measuring cups, beakers, feeding spoons, plastic tubs, dissection areas and instruments, balance and lab table were soaked between workdays in 1% Nolvasan. During each processing or necropsy done throughout the experiment, nitrile gloves were worn and disposed of between turtles. Additionally, turtles being weighed or measured were placed in plastic sandwich bag to avoid direct contact with balance or calipers. The turtles were weighed again, and carapace and plastron length were measured. Tubs were replaced and filled with 400 ml of water containing 1x10^3 titer of ranavirus for exposed turtles and 400 ml of plain water for non-exposed turtles. These ranavirus treatments or controls were left in place for four days to allow contact with and ingestion of virions. Turtles were not exposed to both ranavirus and chemical treatment at the same time to prevent potential interactions between the herbicides and the virions in solution. The ranavirus solutions were replaced with the original chemical treatments from May 30-June 2, 2012. Mass was measured again; however carapace and plastron length were not recorded because they had been done four days prior and caused more handling and potential stress to the turtles. For the next four weeks, we conducted the following procedure each week for all tubs, processing 50 turtles per day on a four-day cycle: We weighed and measured turtles and examined them for disease symptoms (swelling, discoloration, and possibly pathogenic spots found but not identified, as described in the Results). Tubs were replaced with 500ml of new water containing the respective chemical treatment. At the end of each processing day, waste-water from the containers was treated with approximately 600ml bleach/12liters of waste water before disposal. Beginning one week after ranavirus exposure, all turtles were checked every twelve hours for signs of death and illness. To distinguish between sleeping and dead/disabled turtles, we gently poked turtles on the head or shell with a piece of spaghetti (discarded immediately after each turtle was prodded) to test for responsiveness. Turtles that were not exposed to ranavirus were checked first to avoid contamination of the control group. Turtles began dying on May 31, 2012. Turtles that were found diseased or near death during the 12 hour checks were immediately necropsied or kept in ice water for up to 12 hours to prevent decay before necropsy. First, the mass, carapace length, and plastron length were recorded along with any disease symptoms described above. Sterile scalpels, scissors and forceps were used to remove a sample of liver, spleen, intestines and the left kidney from each turtle, which were placed in vials and frozen. PCR was performed (as previously described) at the Center for Wildlife Health and Biomedical and Diagnostic Sciences at the University of Tennessee, Knoxville. The remaining bodies were stored in formalin for preservation and sent to the same lab for further analysis using histopathology. Turtles that did not expire during the study were humanely euthanized four weeks after ranavirus infection via injection of sodium pentaboarbitol by Dr. Mark French of The Ridge Animal Hospital in Farmville VA. The necropsy procedure described above was repeated for all turtles. On each day, turtles in the control group were necropsied before the turtles in the ranavirusexposed group in order to prevent any contamination of the tissue taken. Dissecting tools were scrubbed with antibacterial soap to remove any film from necropsies and disinfected in 1% Nolvasan to between uses on different turtles. RESULTS Thirty-five turtles died before the euthanasia at the conclusion of the experiment. The first turtle expired May 31, 2012. While examining the turtles during processing, white circular fuzzy spots were found on the limbs of some turtles. We suspect they were caused by fungal or bacterial infection; forthcoming results of histopathology will shed light on this probable second pathogen in our system. Also, swelling of the neck and axillary regions of the turtles was commonly found. This swelling has been reported as a symptom of ranavirus, but is probably also a generalized immune response to other stressors, since we observed it in turtles not exposure to the virus. http://sciencejournal.hsc.edu 3
Response Variable Factor DF F P Change in Exposure 1,193 0.00 0.995 Mass (g) Treatment 4,193 0.68 0.610 Change in Carapace Length (mm) Change in Plastron Length (mm) Maximum # White Spots Table 2. Results of Analyses of Variance (ANOVA) on responses of juvenile turtles to ranavirus and chemical treatments (four herbicides) in a laboratory experiment. We found that neither ravavirus exposure nor chemical treatment had any effect on turtles mass, plastron length, or carapace length (see Table 2). The maximum number of white spots found on turtles at any time during the experiment was significantly impacted by the chemical treatments (Table 2). Turtles exposed to tap water (No chem) were found to have more spots on average than any other chemical treatment group. There was a trend (thought not statistically significant with p=0.061) for ranavirus exposure to effect the number of spots found on turtles (Table 2). Exposure to the four herbicides did not have a significant effect on whether turtles cleared ranavirus by the end of the experiment (among those exposed to ranavirus; see Table 3). Neither ranavirus exposure nor chemical treatment affected the survival rate of turtles during the experiment (Table 3). Similarly, neither factor had any impact on swelling in the turtles (Table 3). Response Variable Test Result Swelling Survival Independent Variable X 2 P DF Treatment 1.04 1.000 9 Exposure 16.4 0.003 3 Treatment 2.72 0.987 9 Exposure 1.70 0.791 3 Treatment 1.71 0.998 9 Table 3. Results of Chi Squared tests on categorical responses of juvenile turtles to exposures of ranavirus and four herbicide treatments in a laboratory experiment. DISCUSSION Exposure 1,193 0.01 0.929 Treatment 4,193 0.41 0.803 Exposure 1,193 0.07 0.793 Treatment 4,193 1.98 0.099 Exposure 1,190 3.54 0.061 Treatment 4,190 4.98 0.001 The current study did not find any changes in plastron length, carapace length, or mass of juvenile turtles due to exposure to ranavirus or the four herbicides Round Up Pro, ShoreKlear, 2,4- D, Atrazine. We also did not find any overall growth during the course of the experiment, which suggests that our laboratory conditions may not have been well-suited to housing this species. Avery et al. (1993) found that sliders (Trachemys scripta) housed in water ranging from 22-30 C and consuming 25-40% protein showed increases of approximately 5% in carapace length, 4% in plastron length, and 21% in mass during a five week period. The complete lack of growth we found in a similar time period (nearly five weeks) could be attributed to differences in our housing temperature or quality of food pellets. We do not have the nutritional composition of our food source, but it is formulated for juveniles and used routinely in a major commercial breeding operation and pet shop reptile provider. Therefore, we suspect the lack of growth was due to our relatively cold lab temperatures, which averaged 18-22 C during the experiment. Surprisingly, we found that turtles not exposed to any chemicals or ranavirus had a larger number of spots (possibly pathogenic) than turtles that were exposed to ranavirus or any of the four herbicides. We suggest that the herbicides may have a directly negative effect on the putative pathogen that caused the spots. Additionally, perhaps the presence of ranavirus could have a negative effect on spot number through competition with the putative pathogen in the host turtles. We will be able to make more informed interpretation of this issue when the causative agent of the white spots is identified by Dr. Miller via histopathology. Survival of sliders was not affected by exposure to the ranavirus or the herbicides. Either the amount of time the turtles were exposed to the ranavirus (approximately 4 days) or the concentration of virons (1x10^3 titer) in the exposure solution could have been too low to ensure uptake of the virus. The prevalence of the virus after death or euthanasia was 16%. This percentage of prevailing virus in the turtles indicates that 84% of the ranavirus exposed group either cleared the virus or did not pick the virus up during the exposure period. Another possibility is that concentrations of the herbicides in this experiment may not have been high enough to affect the survival of turtles exposed to ranavirus. Willemsen and Hailey (2001) found that juvenile tortoises (Testudo hermanni) living in an environment with 2,4-D sprayed on surrounding plant life had population decreases of 50% each year. This study used a greater concentration and longer exposure to 2,4-D than our experiment. That study also had a two year timeline to observe the changes in the population, whereas our study was more limited in time. A study by Forson and Storfer http://sciencejournal.hsc.edu 4
(2006) found that long-toed salamanders (Ambystoma macrodactylum) exposed to 184ug/L of Atrazine had a significantly lower mass at metamorphosis than the salamanders exposed to 18.4ug/L, 1.84ug/L of Atrazine and the control group. The concentrations used in the laboratory were set to duplicate those found regularly in natural environments (Solomon &Thompson 2003, Gilliom et al. 2007, Relyea 2007, US EPA 2005). Future research could be done to investigate the effects of increased concentration on the Red Eared sliders (T. s. elegans). The swelling found in the axillary region and the neck of the turtles was not affected by the presence of chemical exposure, but was more common in turtles by ranavirus exposure. This finding supports swelling as a common symptom of ranavirus infection. However, swelling was found in sliders that were not exposed to ranavirus as well, perhaps as a response to the possible pathogen associated with white spots or a to the herbicide treatments. Our current findings suggest that the low concentrations of herbicides used during the experiment did not have a significant impact on growth or survival of juvenile sliders in the presence of ranavirus. However, we are hesitant to make broader conclusions and recommendations based on this study because of our lack of growth data and the probable presence of a second pathogen that was unintentionally introduced to the system. Ideally, this study would be repeated in a warmer environment suited to this species, and with a population of turtles not carrying large numbers of additional pathogens. Additionally, further experiments in which sliders are exposed to varying concentrations of each herbicide may yield more insight into specific concentrations at which growth, survivability, and susceptibility to ranavirus are affected. REFERENCES 1. Avery, H. W., Spotilat, J. R., Flscher, R. U. Jr., Standora, E. A. and Avery, S. B. (1993). Roles of Diet Protein and Temperature in the Growth and Nutritional Energetics of Juvenile Slider Turtles, Trachemys Scripta. Physiological Zoology 66: 902-25. 2. Belzer, William R., and Susan Seibert. (2007) A Natural History of in an Eastern Box Turtle Population. Turtle and Tortoise Newsletter 15:18-25 3. Chen G, Robert J. 2011. Antiviral Immunity in Amphibians. Viruses 3:2065-2086. 4. Chinchar, V. G. 2002. es (family Iridoviridae): Emerging cold-blooded killers. Archives of Virology 147: 447-70. 5. De Solla, Shane R., Pamela A. Martin, Kimberly J. Fernie, Drad J. Park, and Gregory Mayne. (2005) "Effects of Environmentally Relevant Concentrations of Atrazine on Gonadal Development of Snapping Turtles (Chelydra serpentina). Environmental Toxicology and Chemistry 25.2:520-26. 6. Forson, Diane, and Andrew Storfer. "Effects of Atrazine and Iridovirus on Survival and Life History Traits of Long-Toed Salamanders (Ambystoma Macrodactylum)."Environmental Toxicology and Chemistry 25.1 (2006): 168-73. Print. 7. Garner, R.J., 1957. Veterinary Toxicology. Bailliere, Tindall and Cox, London. 8. Giesy, J.ohn P., Stuart Dobson S, and Keith R. Solomon K.R.. "Ecotoxicological Risk Assessment for Roundup Herbicide." Rev Environ Contam Toxicol 167 (2000): 35-120. Print. 9. Gilliom RJ, Barbash JE, Crawford CG, et al. 2007. Pesticides in the nation s streams and ground water, 1992-2001. Circular 1291. U.S. Department of the Interior, U.S. Geological Survey. 10. Gray, Matthew J., Debra L. Miller, and Jason T. Hoverman. Ecology and Pathology of Amphibian es. Diseases of Aquatic Organisms 87 (2009): 243-66. 11. Hall RJ. 1980. Effects of environmental contaminants on reptiles: a review. Washington D.C. U.S. Fish and Wildlife Service. Species Science Report Wildlife No. 128:. P 1-12. 12. Hayes T, Haston K, Tsue M, Hoang A, Haeffele C, Vonk A. 2002. Feminization of male frogs in the wild. Nature 419:895-896. 13. Hill Rw, Wyse GA (1989) Animal physiology. Harper and Row, New York. 14. Hyatt, A. D. (2002) "First Identification of a From Green Pythons." Journal of Wildlife Diseases 38 (2002): 239-52. 15. Larson DL, McDonald S. Fivizzani AJ. Newton WE. Hamilton SJ. 1998. Effects of the herbicide atrazine on Ambystoma tigrinum metamorphosis: Duration. Larval growth. And hormonal response. Physiological Zoology 71:671-679 http://sciencejournal.hsc.edu 5
16. Lesbarreres D, Brunner J, Chinchar VG, Duffus A, Kerby J, Miller DL, Robert J, Schock DM, Waltzek T, Gray MJ. 2011. : Past, Present and Future. Biology Letters, DOI: 10.1098: 1-3. 17. Mao J., T.N. Tham, G.A. Gentry, A. Aubertin, and V.G. chinchar. 1996. Cloning, sequence analysis, and expression of the major capsid protein of the iridovirus frog virus 3. Virology 216:431-436. 18. Mao, J., R.P. Hedrick, and V.G. chinchar. 1997. Molecular characterization, sequence analysis, and taxonomic position of newly isolated fish iridoviruses. Virology 229: 212-220. 19. Marschang RE, Miller D. 2012. 2011 International Symposium. Journal of Herpetological Medicine and Surgery 21(1): 1-2. 20. Marschang, R. E., P. Becher, H. Posthaus, P. Wild, H.-J. Thiel, U. Muller-Doblies, E. F. Kaleta, and L. N. Bacciarini. (1999) "Isolation and Characterization of an Iridovirus from Hermann's Tortoises (Testudo hermanni)." Archives of Virology 144 (1999): 1909-922. Print. 22. Relyea RA. 2005. The impact of insecticides and herbicides on the biodiversity and productivity of aquatic communities. Ecological Applications 15:618-627. 23. Solomon KR, Thompson DG. 2003. Ecological risk assessment for aquatic organisms from overwater uses of glyphosate. Journal of Toxicology and Environmental Health, Part B 6:289 324. 24. U.S. Environmental Protection Agency. 2005. Reregistration eligibility decision 2,4-D; EPA 738-R- 05-002. Office of Prevention, Pesticides and Toxic Substances, Office of Pesticide Programs, Washington, D.C. 25. Whittington RJ, Becker JA, Dennis MM. 2010. Iridovirus infections in finfish critical review with emphasis on ranaviruses. Journal of Fish Diseases 33:95-122. 26. Willemsen RE, Hailey A. 2001. Effects of spraying the herbicides 2,4-D and 2,4,5-T on a population of the tortoise Testudo hermanni in southern Greece. Environmental Pollution 113:71-78. 21. Marschang, Rachel E., Sadine Braun S, and Paul Becher P. (2005) " Isolation of a from a Gecko." Journal of Zoo and Wildlife Medicine 36.2 (2005): n. pag. Print. http://sciencejournal.hsc.edu 6