Tracking Outcomes of Snake Fungal Disease in Free-ranging Pygmy Rattlesnakes (Sistrurus miliarius)

Snake fungal disease (SFD) is caused by the fungus Ophidiomyces ophiodiicola (Allender et al. 2015; Lorch et al. 2015) and is a threat to snake populations in the US (Lorch et al. 2016). The disease has now been confirmed in at least 18 states in the eastern US (Lorch et al. 2016) and has been linked to the decline of two rattlesnake (Viperidae) populations (Allender et al. 2011; Clark et al. 2011) and a population of a threatened colubrid snake (Lorch et al. 2016). Little is known regarding the individual outcomes of infection in free-ranging individuals. Previous studies indicate high mortality in laboratory settings (Allender et al. 2011, 2015), and effective treatment interventions in the laboratory have been elusive. Infection is often associated with winter hibernation (Lorch et al. 2016), and infection severity peaks in cool winter months and is negatively related to population energetic status in populations where winter hibernation does not occur (McCoy et al. 2017). No estimate of mortality in field-active snakes has been published, and no studies have reported the individual outcomes of SFD in unmanipulated, free-ranging individuals. To describe the outcomes of SFD in individual snakes in the field, we monitored the presentation of clinical signs of SFD in a population of pygmy rattlesnakes (Sistrurus miliarius) for 2 yr. Pygmy rattlesnakes are active and accessible year round in Central Florida, and a winter-high and summer-low pattern of SFD severity has been described in previous investigations (McCoy et al. 2017). We sought to establish whether snakes identified as severely infected, based on the presentation of clinical signs, died or if they survived and cleared clinical signs of infection.

We studied snakes at Lake Woodruff National Wildlife Refuge in central Florida (29°05′N, 81°26′W). The presence of SFD had been documented at this site (Cheatwood et al. 2003; Lorch et al. 2016; McCoy et al. 2017). Surveys were conducted on a single, 10-ha mesic hammock surrounded by freshwater marsh (May et al. 1996). From 18 January 2015 to 11 December 2016, the severity of clinical signs of SFD was visually scored 535 times on a total of 222 individual adult pygmy rattlesnakes using methods described by McCoy et al. (2017). The severity of infection was scored on a zero- to three-point scale with zero describing individual snakes with no clinical signs of infection, one describing snakes with at least one visible lesion consistent with SFD, two describing snakes with five or more visible lesions, and three describing snakes with severe lesions on the head and vent in addition to body lesions. Snakes that presented clinical signs intermediate between two levels were assigned half scores. The visual scoring method that we used (McCoy et al. 2017) was validated by PCR in a previous study on the Lake Woodruff population (Lind et al. 2018). In that study, 29 snakes with varying degrees of infection were sampled using skin swabs, and real-time PCR was used to identify the presence and abundance of O. ophiodiicola in samples. The PCR cycle threshold (inversely related to the abundance of the target organism) was strongly correlated with SFD scores and all snakes identified as having clinical signs consistent with SFD (with an infection severity score ≥1) were PCR positive. Four out of the 12 snakes assigned a score of 0 in the study were PCR positive, indicating that snakes without clinical signs can carry the fungus. At its initial capture, each snake was marked by intraperitoneal injection of a passive integrated transponder (PIT) tag (Avid®, Norco, California, USA), a technique that does not appear to cause negative impacts in this species (Jemison et al. 1995). A blood sample was also drawn from the caudal vein using a 1-mL syringe and a 27-ga needle as part of another study. Individuals were processed in the field and released at the point of capture. Recaptures were opportunistic and snakes were marked dorsally with a nail polish color code to ensure that snakes were not sampled more than twice per month. On three occasions in the winter of 2015, individuals were encountered with severe clinical signs of infection and were also severely emaciated. Field processing, including PIT tagging, was not performed in such cases, and these individuals were removed from the recapture analysis. When it was assumed that these individuals were not recaptured, statistical differences in recapture probabilities were unchanged.

To determine if mortality associated with SFD decreased the likelihood of recapturing an individual, recapture data on the 86 individuals sampled in January, February, and March of 2015 and 2016 were analyzed. Snakes were divided into three groups based on their severity of infection scores (low=0–0.5, medium=1–1.5, severe=2–3). A Fisher's exact test was then used to determine whether the probability of recapture was different among the three groups within each year. Individuals sampled in the winter of 2015 had a higher probability of recapture, as individuals could be recaptured for the entire 21 mo remaining in the study compared to only 9 mo for snakes sampled in the winter of 2016. The individual outcomes of infection in the population were described for the 12 snakes (female=4; male=8) that satisfied the following criteria: 1) they were repeatedly sampled at least four times, 2) they exhibited severe signs of infection at a minimum of one point in the study, and 3) we had complete mass and length data (used to calculate body condition index; BCI) for at least four observations. Individual energetic status was estimated by a BCI calculated from the residual of the linear regression of log-transformed mass on log-transformed snout to vent length.

The severity of clinical signs of infection was not related to the probability of recapture later in the study (Table 1). Six out of the 12 individuals (50%) with severe clinical signs of infection in the winter of 2015 were recaptured later in the study with no clinical signs of infection. Individual snakes with extensive recapture histories demonstrated that individuals could clear the visible signs of infection and, in some cases, could fluctuate between severe and no or low clinical signs of infection over a 2-yr period (Fig. 1).

We described long-term temporal patterns of SFD severity in free-ranging individuals and demonstrated that the outcomes of SFD in field-active pygmy rattlesnakes in central Florida are variable. Mortality rates in the field are likely not as high as reported in laboratory studies conducted on other pit vipers (Allender et al. 2011, 2015). Snake fungal disease appears to cause high rates of mortality in snake populations where it has recently emerged and in small, vulnerable populations that are already subject to habitat loss and or other interacting stressors. The disease in our study population has been endemic for at least 20 yr (Cheatwood et al. 2003). Disease severity may be higher in populations that have been only recently exposed to a pathogen (Daszak et al. 1999), which may explain the lack of a relationship between infection severity and the probability of recapture observed in our study population. However, the mechanism underlying interpopulation variation in effects on population dynamics (if present) will require direct investigation.

Description of the individual outcomes of infection clearly showed that recovery occurred in many cases, even from severe infections. The clearing of clinical signs of infection did not necessarily indicate the clearing of the infection. Ophidiomyces DNA can be found on the skin of snakes exhibiting no visual signs of infection (Hileman et al. 2017; Lind et al. 2018), and several individuals in this study (e.g., M-894 and M-781, Fig. 1) appeared to clear clinical signs of infection in the summer of 2015 but exhibited severe clinical signs again in the winter of 2016, suggesting that the infection had never cleared or that there was chronic exposure of snakes to the fungus in the environment. A small study on a larger-bodied pit viper, the timber rattlesnake (Crotalus horridus), demonstrated low mortality in eight wild adult snakes with severe facial lesions that were then held in captivity without treatment (McBride et al. 2015). Laboratory estimates of mortality in pygmy rattlesnakes have not been established. However, given that individuals can and do clear clinical signs, no intervention or interventions done in the wild may be favored over interventions that involve bringing individuals into captivity for treatment.

Our results did not suggest that host survival or population dynamics were not affected by SFD, as our sampling was not designed to estimate these variables. Our results also did not rule out the potential for sublethal effects on individual fitness. The population-wide negative relationship between the severity of clinical signs and BCI as described by McCoy et al. (2017) appeared to be evident in our description of individual outcomes. Body condition also covaries with components of the hematologic response to SFD in the eastern massasauga (Sistrurus catenatus; Allender et al. 2016). In many individuals in our study, BCI clearly declined as SFD score rose (e.g., snakes F-374, F-305, F-052, and F-122). That this relationship was evident suggested a relationship between postparturient energetic status in the fall (parturition occurs in late summer in this population) and the susceptibility to SFD. Many individual snakes in our study population were recaptured multiple times and never demonstrated clinical signs of SFD, even in winter (Lind et al. 2018). The factors that underlie within and among individual variation in the severity of infection (e.g., energetic, reproductive, or immune status), as well as the fitness outcomes of infection, warrant further investigation. Elucidation of such mechanisms may prove critical to effective conservation and management strategies.

We thank the many undergraduates at Stetson University who assisted with field sampling. We especially thank Ethan Royal for assistance in the field. We thank Candice Stevenson for access to the Lake Woodruff National Wildlife Refuge. All animal care practices and experimental procedures were approved and overseen by the IACUC committee at Stetson University (SU-1001).