Author's personal copy

Biological Conservation 159 (2013) 413–421

Contents lists available at SciVerse ScienceDirect

Biological Conservation journal homepage: www.elsevier.com/locate/biocon

High prevalence of infection in tadpoles increases vulnerability to fungal pathogen in high-Andean amphibians Alessandro Catenazzi a,⇑, Rudolf von May b, Vance T. Vredenburg a a b

Department of Biology, San Francisco State University, San Francisco, CA 94132, USA Museum of Vertebrate Zoology, University of California, Berkeley, CA 94720, USA

a r t i c l e

i n f o

Article history: Received 18 September 2012 Received in revised form 16 November 2012 Accepted 19 November 2012

Keywords: Batrachochytrium dendrobatidis Chytridiomycosis Telmatobius jelskii Peru Andes Streams Infection prevalence Epizootic Larval stage

a b s t r a c t The amphibian pathogen Batrachochytrium dendrobatidis (Bd) is causing population declines and species extirpations worldwide. Montane amphibians in tropical and temperate regions are especially vulnerable to chytridiomycosis. High-elevation amphibian assemblages typically include few species, so that epizootics should be limited once high frog mortality hinders transmission of the pathogen. We hypothesized that tadpoles of a high-elevation frog in the Peruvian Andes, Telmatobius jelskii, could function as reservoir for Bd in Andean streams. We postulated that, for tadpoles to function as an efficient reservoir of Bd, they should inhabit streams over extended periods of time, and have high prevalence of Bd. We surveyed streams between 2400 and 4850 m in the wet and dry seasons of 2010, where we captured, swabbed and determined the developmental stage of 458 tadpoles. We found that cohorts of tadpoles overlap continuously in these streams, as a consequence of multiple breeding events throughout the year. Prevalence of Bd among tadpoles averaged 53.1% (95% confidence interval: 49.8–56.3%); 8 out of 13 streams inhabited by T. jelskii had a prevalence greater than 50%. Prevalence of Bd was also higher during the dry season and increased with the age of the tadpoles. Our results support the hypothesis that the year-long presence of infected tadpoles in streams makes high-Andean Telmatobius frogs especially vulnerable to chytridiomycosis. The genus is already extirpated in Ecuador, and has been observed to decline rapidly in Peru, Bolivia and Argentina. Conservation strategies to mitigate the impact of Bd on populations of Telmatobius should consider aquatic life-stages. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction High elevation amphibians are threatened by chytridiomycosis, an emergent and highly virulent disease caused by the fungus Batrachochytrium dendrobatidis (Bd; Berger et al., 1998; Kilpatrick et al., 2010; Vredenburg et al., 2010). Bd epizootic events, which cause population declines and extirpations, have been documented in montane areas in California (Vredenburg et al., 2010), the Andes in Peru (Catenazzi et al., 2011; Seimon et al., 2007), and the Pyrenees and similar regions of the Iberian peninsula (Walker et al., 2010). The devastating impact of chytridiomycosis on montane amphibians is puzzling for a number of reasons. The optimal growth for Bd in culture is reportedly 15–25 °C (Piotrowski et al., 2004; Woodhams et al., 2008). These temperatures seldom occur at high elevation, even in tropical regions. The persistence of Bd in populations of highly vulnerable hosts is especially intriguing, because amphibian species richness at high-elevations is low, which limits the number ⇑ Corresponding author. Address: Department of Biology, San Francisco State University, 1600 Holloway Ave., San Francisco, CA 94132, USA. Tel.: +1 (305) 396 2626. E-mail address: [email protected] (A. Catenazzi). 0006-3207/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biocon.2012.11.023

of alternative hosts. During a chytridiomycosis outbreak, the high mortality of infected host frogs should quickly reduce transmission, and therefore reduce the risks of future outbreaks and epizootics (Anderson and May, 1979). Therefore, in order to persist Bd requires an alternative host or an environmental reservoir. In contrast to adults, chytridiomycosis is not lethal to tadpoles of most species (Blaustein et al., 2005). Bd infects the keratinized mouthparts of tadpoles (Fellers et al., 2001), and infected tadpoles can be recognized by their abnormal and depigmented oral disks (Knapp and Morgan, 2006; Vredenburg and Summers, 2001). The larval stages of montane amphibians often spend extended periods of time in water before completing metamorphosis, and frequently over-winter in temperate regions (Bosch and Martinez-Solano, 2006; Bosch et al., 2001; Briggs et al., 2010). Understanding the link between amphibian larval development and Bd epidemiology is crucial for designing strategies to reduce the impact of this disease on potentially susceptible species (Bosch and Martinez-Solano, 2006; Briggs et al., 2010; Conradie et al., 2011). Here we explore the hypothesis that long-lived tadpoles function as reservoirs for Bd in populations of high-Andean frogs. We investigated whether tadpoles of Telmatobius jelskii continuously occupy aquatic habitats throughout the year, and denoted their Bd infec-

Author's personal copy

414

A. Catenazzi et al. / Biological Conservation 159 (2013) 413–421

tion status. We also quantified Bd infection intensity and prevalence in both tadpoles and post-metamorphic stages of T. jelskii in order to evaluate the threat of chytridiomycosis in the Puna ecosystem of the Peruvian central Andes. The genus Telmatobius, consisting of approximately 60 species is mainly distributed in the tropical Andes, with most species distributed in Peru and Bolivia (Lehr, 2005). Species in this genus have declined dramatically over the past three decades (Merino-Viteri et al., 2005). For example, the three Telmatobius species known from Ecuador were extirpated in the late 1980s and early 1990s and are now thought to be extinct (Merino-Viteri et al., 2005). The last individuals of Telmatobius found in Ecuador showed symptoms of chytridiomycosis (Merino-Viteri et al., 2005). Moreover, population declines of Telmatobius marmoratus, Telmatobius mendelsoni and Telmatobius timens in Peru (Catenazzi et al., 2011; Seimon et al., 2007; von May et al., 2008) and of two species of Telmatobius in Argentina (Barrionuevo and Mangione, 2006) have been associated with outbreaks of Bd. Additional threats for species of Telmatobius are habitat loss, agricultural expansion, trout farming, contamination from mining and agriculture, and human harvesting and consumption (Angulo, 2008; Catenazzi et al., 2010; De la Riva and Lavilla, 2008; von May et al., 2008). T. jelskii inhabits bogs, streams and rivers with slow-moving waters in the high Andes of central Peru, and are restricted to the regions of Junín, Huancavelica, and Ayacucho (Sinsch, 1986, 1990; Sinsch et al., 1995; Vellard, 1955). Adults of T. jelskii reproduce all year long under favorable environmental and habitat conditions (Sinsch, 1990). Tadpoles of T. jelskii are thought to spend several months in the cold water of high-elevation streams before completing metamorphosis (>3 months according to Sinsch (1986, 1990). Therefore, if tadpoles carry Bd, they may function as reservoirs for chytridiomycosis over long periods of time. This may keep Bd in the ecosystem when adults are not present or if adults have been extirpated during an outbreak of chytridiomycosis (Briggs et al., 2010, 2005). In this study, we assessed whether tadpoles of T. jelskii fulfill two important prerequisites for functioning as reservoir: (1) tadpoles are infected with Bd throughout the year, and (2) tadpoles inhabit aquatic breeding sites all year long.

2. Methods We studied 22 streams in the Ayacucho region of central Peru (Fig. 1, online kml file), as part of a long-term program that is monitoring the impact of a trans-Andean natural gas pipeline. We sampled 9 streams in both the wet (20 March–6 April 2010) and dry (8–18 July 2010) seasons, 4 streams only in the wet season and 9 streams only in the dry season. These streams were located at elevations between 2400 and 4854 m. Tadpoles and frogs were captured along linear transects of 200 m of stream and with a search effort of six person-hours per transect. We searched for post-metamorphic stages by visually inspecting pools and riparian areas, as well as manually, by feeling for frogs moving under rocks, mud and along the edges of riffles, runs and pools. We determined developmental stage of tadpoles using standard techniques (Gosner, 1960). Tadpoles hatch around Gosner stages 18–20, become active swimmers and algal grazers at stage 25, interrupt feeding and start reabsorbing tail and undergoing metamorphosis at stage 42, and complete metamorphosis at stage 46. At each stream, we measured water temperature in pools occupied by tadpoles between 10:00 and 13:00. We averaged temperature measurements across all occupied pools within a stream. To measure Bd infection load, each animal was swabbed with a synthetic dry swab (Medical Wire and Equipment Co. Ltd.). In postmetamorphic stages, swabs were stroked across the skin a total of 30 times: 5 strokes on each side of the abdominal midline, 5

strokes on the inner thighs of each hind leg, and 5 strokes on the foot webbing of each hind leg. Tadpoles were swabbed with 10 strokes on the mouthparts. We used a real-time Polymerase Chain Reaction (PCR) assay on material collected on swabs to detect Bd and quantify the level of infection (Boyle et al., 2004). This assay compares the sample to a set of standards and calculates a genomic equivalent for each sample (i.e., the number of copies of DNA of Bd on each swab, or Zswab). We followed DNA extraction and real-time PCR methods of Hyatt et al. (2007) and Boyle et al. (2004), except that we analyzed single-swab extracts once instead of 3 times (Kriger et al., 2006; Vredenburg et al., 2010). The real-time PCR technique for Bd uses DNA extracts from swabs that are diluted 80fold during extraction and PCR. Thus, to estimate Zswab, we multiplied the genomic equivalent values generated during the realtime PCR by 80. Confidence intervals for prevalence data follow the Wilson procedure (Newcombe, 1998). We compared Bd prevalence, Zswab and ontogenetic structure of tadpole populations between wet and dry season surveys. These comparisons were made at three levels: (1) within stream comparisons for sites 6, 7 and 13 (Fig. 1); (2) within watershed comparisons, by pooling data from all streams in the Apacheta and Huamanga watersheds (9 streams); and (3) within elevational range comparisons, by pooling data from all streams in the 3800–3899 (2 streams), 3900–3999 (1 stream), and 4000–4099 m ranges (3 streams). We used this approach instead of a general model because we were constrained in our sampling locations. Our permits restricted us to visiting streams near the buried natural gas pipeline where the company had agreements with local communities. As a result, our dataset had important gaps with respect to elevation, number of sites per watershed, and number of streams visited per season. We additionally evaluated if, when controlling for developmental stage, body size differed between infected and uninfected tadpoles. If Bd infection did not affect tadpole growth and development, curves of body size versus developmental stage would have similar slopes. To test this prediction, we treated developmental stage as a covariate in an analysis of covariance (ANCOVA) comparing the body size of infected and uninfected tadpoles. For this analysis we only considered tadpoles between development stages 25 (presence of keratinized mouthparts) and 40 (when tadpoles stop growing in size and enter metamorphosis). We separated the frequency distribution of tadpole developmental stages into different components using finite mixture distribution models. We assumed that developmental stages followed a gamma distribution corresponding to distinct breeding and egg laying events. We labeled these components as cohorts of tadpoles. The term cohort refers here to tadpoles sharing the same developmental stage at the time of our surveys. We first fitted mixtures of unconstrained gamma distributions, and then produced separate constrained analyses. We chose the unconstrained model only if it was a better fit to the frequency distribution than the constrained model. We used the R package mixdist to fit the finite mixture distribution models (http://cran.r-project.org/web/packages/ mixdist/index.html). This package contains functions for fitting finite mixture distribution models to grouped data and conditional data by the method of maximum likelihood using a combination of a Newton-type algorithm and an expectation–maximization algorithm. The package also reports results of a chi-square test for the fit between an empirical histogram and the histogram calculated from the sum of the mixture distributions. 3. Results 3.1. Prevalence of Bd and infection intensity in tadpoles of T. jelskii Infection with Bd was widespread at the three spatial scales (Figs. 2–4, Table A1). Prevalence of Bd across all tadpoles was

Author's personal copy

A. Catenazzi et al. / Biological Conservation 159 (2013) 413–421

415

Fig. 1. Location of streams studied in the Peruvian Andes. An open square with a cross mark indicates streams where Telmatobius jelskii were not found. A star indicates presence of individuals of T. jelskii infected with Batrachochytrium dendrobatidis. A closed square indicates presence of T. jelskii but no Bd infection. A dashed line indicates the course of the buried gas pipeline. Inset depicts location of Peru in South America; the rectangle indicated the location of the study region. A kml file with the precise location of each sampling site is available online.

53.1% (95% confidence interval: 49.8–56.3%; n = 458 tadpoles in 13 streams). Prevalence averaged 55.1 ± 10.0% per stream (range 0– 100.0%; n = 13 streams; 8 streams had prevalence > 50%). Prevalence varied seasonally (v2 = 13.41, df = 1, p < 0.01) and averaged 61.2% (95% confidence interval: 54.9–67.1%) in the dry season and 42.2% (95% confidence interval: 35.6–49.0%) in the wet season across all tadpoles. Similarly, intensity of infection (measured as Zswab) was higher during the dry season (Zswab = 289.2 ± 51.3 zoospore equivalents; n = 245 tadpoles) than it was during the wet season (Zswab = 181.3 ± 42.9 zoospore equivalents; n = 213; Welch’s t test on log-transformed Zswab, t = 3.01, p < 0.01). Prevalence of Bd in tadpoles increased with elevation (logistic regression with binomial errors, p < 0.01; n = 12 streams with P12 tadpoles), but elevation explained little of the variation in intensity of infection (linear regression on log-transformed Zswab, p < 0.01, R2 = 0.09). There was no effect of water temperature (logistic regression with binomial errors, p = 0.30; n = 12 streams) on prevalence of Bd. Prevalence of Bd and intensity of infection varied with ontogeny and body size. Prevalence increased with both developmental stage (logistic regression with binomial errors, p < 0.01) and body length (p < 0.01; Fig. 5). Similarly, intensity of infection increased with developmental stage (linear regression on log-transformed Zswab, p < 0.01, but note R2 = 0.33) and body length (p < 0.01, R2 = 0.26). Uninfected tadpoles were larger and heavier than infected tadpoles at Gosner stages above 33 (Fig. 6). When controlling for developmental stage, there was a significant interaction between infection status and the covariate developmental stage (linear models, body size- and mass-developmental stage interactions significant at p < 0.0001). 3.2. Prevalence of Bd and infection intensity in juveniles and adults of T. jelskii Prevalence across all frogs was 27.3% (n = 77 frogs in 11 streams). The intensity of infection averaged 8.3 ± 4.6 zoospore equivalents (range 0.04–92.0, n = 21 infected frogs). Prevalence

within streams averaged 15.7 ± 4.2% (0–39.4%; n = 11 streams). Only two streams had sufficient sample sizes (n > 10) of both tadpoles and frogs for comparison. In Site 7, prevalence of infection for dry and wet season combined was 33.7% among tadpoles and 39.4% among frogs, whereas in Site 13 prevalence was 53.0% among tadpoles and 14.3% among frogs. 3.3. Structure of tadpole populations Our exploratory analyses revealed between 3 and 5 cohorts of tadpoles in the wet and dry seasons (Figs. 2–4; Table 1). The relative proportions of tadpole cohorts identified on the basis of developmental stage differed between the wet and dry seasons. The first cohort (generally corresponding to a mode of distribution around Gosner stage 25) was numerically dominant in the wet season (P50% of all tadpoles within streams, watersheds and elevational ranges; Table 1). During the dry season, the oldest cohort (generally corresponding to a mode of distribution around Gosner stages 35–38) was often as large as the first cohort (in 2 out of 3 streams, 1 out of 2 watersheds, and 2 out of 3 elevational ranges; Table 1). 4. Discussion Our data show that Bd prevalence and infection levels are high in tadpoles of Telmatobius, a genus that has experienced population declines and species extinction in the tropical Andes (Barrionuevo and Ponssa, 2008; Catenazzi et al., 2011; De la Riva and Lavilla, 2008; Merino-Viteri et al., 2005). Our results support the hypothesis that tadpoles are an important reservoir for Bd in populations of T. jelskii, similar to results found for other stream-breeding amphibians (Hero et al., 2005) or amphibians with extended development times as aquatic larvae (Bosch and Martinez-Solano, 2006; Bosch et al., 2001). Bd was present in tadpoles at all sites, except one stream, and prevalence of infection was high, exceeding 50% in most streams. This result confirms previous findings of high Bd prevalence in tadpoles of Telmatobius frogs. For example, prevalence was 63.6% among tadpoles of a population of T. marmoratus in the Andes of Cusco in southern Peru (Catenazzi et al., 2011). At

Author's personal copy

416

A. Catenazzi et al. / Biological Conservation 159 (2013) 413–421

Fig. 2. Frequency distribution of developmental stages for tadpoles in Site 6 (top row), Site 7 (middle row), and Site 13 (bottom row), based on visual transects from March and April (wet season) and July 2010 (dry season). The fitted mixture of distributions is shown in green, whereas the red triangles refer to the modes of each distribution (mean stages of Table 1; see Section 2). Pie charts indicate prevalence of infection with Batrachochytrium dendrobatidis: white = no infection; very light gray = 0 < Zswab < 1; light gray = 1 6 Zswab < 10; gray = 10 6 Zswab < 100; dark gray = 100 6 Zswab < 1000; black: 1000 6 Zswab < 10,000). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

our study sites in central Peru, prevalence among tadpoles varied seasonally, and the highest prevalence was observed during the dry season. During the dry season both tadpoles and frogs are more

likely to congregate in pools (pers. obervation), a situation that facilitates transmission of Bd within and between life stages of T. jelskii. The proportion of tadpoles with zoospore

Author's personal copy

A. Catenazzi et al. / Biological Conservation 159 (2013) 413–421

417

Fig. 3. Frequency distribution of developmental stages for tadpoles in the 3800–3899 m (top row), 3900–3999 m (middle row), and 4000–4099 m (bottom row) elevational ranges, based on visual transects from March and April (wet season) and July 2010 (dry season). See Fig. 1 for details.

equivalents > 1000 is higher in the dry season than in the wet season, further increasing the opportunities for transmission of Bd through zoospores shed by the mouthparts of tadpoles infecting the skin of pool-dwelling juveniles and adult frogs. The high prevalence of Bd in these streams might be due to long persistence times of free-living zoospores in pools during the dry

season, especially at high elevations. Mitchell et al. (2008) predicted that the longer Bd is able to persist in water, the greater the impact the fungus will have on host populations. In fast-flowing streams such as the ones inhabited by T. jelskii, it could be difficult for the zoospores to remain in pools during periodic floods in the wet season. The longer persistence of free-living zoospores in

Author's personal copy

418

A. Catenazzi et al. / Biological Conservation 159 (2013) 413–421

Fig. 4. Frequency distribution of developmental stages for tadpoles in the Huamanga watershed, 3800–4222 m (top row) and the Apacheta watershed, 3910–4548 m (bottom row), based on visual transects from March and April (wet season) and July 2010 (dry season). See Fig. 1 for details.

pools during the dry season, associated with low water levels and slow flow, could further increase transmission of Bd within and between life stages of T. jelskii during the dry season. In support of this hypothesis, the increase in prevalence of Bd was noticeable in high elevation streams (Fig. 3), where water flow is greatly reduced during the dry season. Similar results of higher prevalence of Bd during periods of minimum flows have been observed in other stream and riverine systems (Conradie et al., 2011). The increase in prevalence of infection and infection intensity (Zswab) with developmental stage and body size of tadpoles is consistent with the observation that tadpoles spend considerable amounts of time in water prior to development (Figs. 2–4; Sinsch, 1986, 1990). This extended aquatic period increases the chances that tadpoles will be infected by Bd, a finding supported by similar trends that have been reported elsewhere (Conradie et al., 2011; Smith et al., 2007). In addition to prolonged exposure to free-living zoospores, the higher prevalence and infection intensity in cohorts of older and larger tadpoles could also indicate enhanced transmission from numerically abundant cohorts of younger and smaller tadpoles. Overall, these findings further support the idea that tadpoles are important reservoirs of Bd for these aquatic frogs. Results from the ANCOVA suggest that observed differences in tadpole body size may be the result of changes in both growth and development associated with infection by Bd. Our data show

that infection with Bd is associated with small body size at advanced developmental stages. This small body size could be caused by either slower growth or faster development in infected tadpoles. This result stands in contrast to previous studies that focused on African tadpoles, which, when controlling for developmental stage, found no difference in body size between infected and uninfected tadpoles (Smith et al., 2007). Given that body size is correlated with body mass in T. jelskii, our results imply that infected tadpoles and froglets weigh less than uninfected tadpoles at metamorphosis, as shown elsewhere (Garner et al., 2009). Our results also suggest that infected tadpoles have a higher body mass and size than the uninfected earlier in development, although sample size for infected tadpoles is small. Controlled experiments where individual tadpoles are followed throughout their development are needed to verify these patterns and to test the hypothesis that infection with Bd influences development. Our results of high prevalence of Bd in Andean streams above 4000 m contradict the hypothesis that low temperatures limit the impact of Bd on amphibians (Puschendorf et al., 2009). Tadpoles at these high elevation streams live at temperatures below the thermal optimum for Bd growth, which is between 15 and 25 °C based on laboratory studies of Bd cultures (Piotrowski et al., 2004; Woodhams et al., 2008). Despite the low temperatures recorded during our study, Bd is widespread geographically and in-

Author's personal copy

A. Catenazzi et al. / Biological Conservation 159 (2013) 413–421

419

Fig. 6. Relationship between (A) body size or (B) mass and developmental stage in infected and uninfected tadpoles of T. jelskii. Fig. 5. Prevalence of infection (proportion of sampled tadpoles infected) with Batrachochytrium dendrobatidis as a function of (A) developmental stage and (B) body size in tadpoles of Telmatobius jelskii.

fects the majority of tadpoles. Knapp et al. (2011) similarly concluded that cold waters did not protect high elevation Rana sierrae in the Sierra Nevada of California from infection with Bd. Moreover, in the Andes Bd has been found in ponds recently formed from deglaciation above 5200 m (Seimon et al., 2007), where freezing temperatures occur throughout the year. Therefore, low temperatures do not offer a refuge from Bd for larval and adult amphibians at high elevations. In support of this finding, we did not find any relationship between water temperature and prevalence of Bd in our streams (note, however, that our temperature measurements were limited to the time of the two visits in the wet and dry season). Although we did not observe mass die-offs of tadpoles or adults of T. jelskii, the high prevalence of Bd we observed could cause outbreaks of chytridiomycosis in the future. It is possible that tadpoles tolerate Bd infection throughout their larval development and experience low mortality rates, though we did not experimentally test the effect of Bd infection levels on tadpole mortality. We currently lack an understanding of levels of prevalence and infection intensity in tadpoles that could elicit local and regional outbreaks, similar to those described for adult R. sierrae and R. muscosa in California (Vredenburg et al., 2010). However, tadpoles infected with Bd have been shown to grow slower and to be more likely to die once they reach metamorphic stages than non-infected tadpoles (Garner et al., 2009). In the adults of T. jelskii that we swabbed, the number of zoospore equivalents is much lower than the 10,000 zoospore equivalents associated with symptoms of chytridiomycosis and epizootic outbreaks in other species of frogs (Vredenburg et al., 2010). On the other hand, only half the streams

we surveyed had populations of T. jelskii. We do not know whether these streams supported populations in the past, but the species was abundant according to historic accounts (Vellard, 1955; see sites for specimens collected between 1950 and 1975 in the online kml file). Our findings have important implications for conservation of the endangered Telmatobius frogs. Any in situ disease mitigation strategy for populations of Telmatobius should consider all life stages. Reintroduction of the species in streams where it has disappeared by transferring larvae from other populations should include antifungal treatment of tadpoles with keratinized mouthparts (Gosner stages between 25 and 42) to reduce or eliminate Bd infection. These reintroductions may have occurred over the past decade because farmers in southern Peru, where Telmatobius frogs are praised for their putative medicinal properties (Angulo, 2008), have noticed the extirpation of frog populations over large areas. During surveys in Coline (Quispicanchis, Cusco), in March 2008 local inhabitants reported that populations of T. marmoratus had been reintroduced after all frogs had died some years earlier (pers. comm. to A. Catenazzi). Moreover, the live trade of Telmatobius frogs in the tropical Andes further promotes local reintroductions and could facilitate the spread of Bd (Catenazzi et al., 2010). Managing Bd infection in wild populations is challenging (Woodhams et al., 2011). Few approaches have been experimented specifically for stream-breeding amphibians, where the interaction between Bd infection, flow rate and other environmental stressors may further complicate conservation efforts. The problem calls for creative solutions, and our study suggests that for T. jelskii and other high-elevation stream-breeding frogs, these solutions should consider reducing pathogen transmission within and between life stages.

Author's personal copy

420

A. Catenazzi et al. / Biological Conservation 159 (2013) 413–421

Table 1 Cohorts of tadpoles identified by fitting multiple modes to univariate frequency distribution data of Gosner developmental stages. Values in bold indicate significant values for chi-square for the fit between the empirical histogram and the histogram calculated from the sum of the mixture distributions (p < 0.05). Cohorts

a

Wet season

Dry season

Proportion ± SE

Mean stage ± SE

Proportion ± SE

Mean stage ± SE

1. Within streams Site 6 Cohort 1 Cohort 2 Cohort 3

n = 46 tadpoles 0.69 ± 0.07 0.20 ± 0.06 0.11 ± 0.05

25.13 ± 0.19 30.59 ± 0.49 37.09 ± 0.64

n = 110 tadpoles 0.33 ± 0.05 0.18 ± 0.04 0.49 ± 0.05

25.41 ± 0.28 29.79 ± 0.59 35.42 ± 0.28

Site 7 Cohort 1 Cohort 2 Cohort 3

n = 27 tadpolesa – – –

– – –

n = 64 tadpoles 0.67 ± 0.06 0.09 ± 0.04 0.23 ± 0.05

25.14 ± 0.14 28.65 ± 0.47 37.04 ± 0.32

Site 13 Cohort 1 Cohort 2 Cohort 3 Cohort 4

n = 113 tadpoles 0.51 ± 0.05 0.07 ± 0.03 0.42 ± 0.05 –

24.85 ± 0.15 29.60 ± 0.65 35.85 ± 0.23 –

n = 112 tadpoles 0.38 ± 0.04 0.16 ± 0.04 0.12 ± 0.44 0.33 ± 0.05

25.68 ± 0.15 28.50 ± 0.25 35.13 ± 0.50 38.88 ± 0.28

2. Within watersheds Huamanga Cohort 1 Cohort 2 Cohort 3 Cohort 4

n = 127 tadpoles (6 streams) 0.54 ± 0.05 0.13 ± 0.03 0.25 ± 0.04 0.08 ± 0.03

24.86 ± 0.14 30.14 ± 0.53 35.49 ± 0.32 42.69 ± 0.65

n = 174 tadpoles (3 streams) 0.47 ± 0.04 0.14 ± 0.03 0.39 ± 0.04 –

25.31 ± 0.16 29.67 ± 0.42 35.82 ± 0.22 –

Apacheta Cohort 1 Cohort 2 Cohort 3 Cohort 4

n = 113 tadpoles (1 stream) 0.50 ± 0.05 0.07 ± 0.02 0.28 ± 0.05 0.15 ± 0.04

24.82 ± 0.10 29.28 ± 0.34 34.61 ± 0.27 38.07 ± 0.43

n = 164 tadpoles (3 streams) 0.63 ± 0.04 0.06 ± 0.04 0.30 ± 0.05 –

26.73 ± 0.13 34.88 ± 1.22 38.49 ± 0.40 –

3. Within elevational ranges 3800–3899 m Cohort 1 Cohort 2 Cohort 3

n = 12 tadpoles (1 stream)a – – –

– – –

n = 51 tadpoles (1 stream) 0.84 ± 0.05 0.09 ± 0.04 0.06 ± 0.03

27.03 ± 0.17 35.89 ± 0.72 41 ± 1.10

3900–3999 m Cohort 1 Cohort 2 Cohort 3 Cohort 4

n = 113 tadpoles (1 stream) 0.50 ± 0.05 0.07 ± 0.02 0.28 ± 0.05 0.15 ± 0.04

24.82 ± 0.10 29.28 ± 0.34 34.61 ± 0.27 38.07 ± 0.43

n = 112 tadpoles (1 stream) 0.43 ± 0.07 0.11 ± 0.06 0.46 ± 0.05 –

26.01 ± 0.31 28.52 ± 0.67 37.90 ± 0.26 –

4000–4099 m Cohort 1 Cohort 2 Cohort 3 Cohort 4 Cohort 5

n = 73 tadpoles (2 streams) 0.65 ± 0.06 0.11 ± 0.04 0.13 ± 0.04 0.08 ± 0.03 0.03 ± 0.02

24.73 ± 0.09 27.59 ± 0.28 31.24 ± 0.23 37.16 ± 0.33 40.98 ± 0.63

n = 176 tadpoles (3 streams) 0.46 ± 0.04 0.14 ± 0.03 0.40 ± 0.04 – –

25.30 ± 0.16 29.67 ± 0.42 35.86 ± 0.22 – –

Sample size too small to fit multiple frequency distributions

Acknowledgments We thank V. Vargas García, W. Velazco Soto and O. Chipana Marca for assistance in the field, A. Pace for assistance with logistics and travel arrangements, and T. Cheng for assistance with PCR assays for detection of B. dendrobatidis. We thank S. Hayes, J. Mastanduno, E. Foreyt and J. Christian for comments on the manuscript. We thank Peru LNG and the Biodiversity and Monitoring Assessment Program at the Center for Conservation Education and Sustainability (CCES) and Smithsonian Conservation Biology Institute. RvM was supported by a National Science Foundation Postdoctoral Research Fellowship in Biology (DBI-1103087). This work was funded by National Science Foundation Grant 1120283 (nsf.gov). This is publication #12 of the CCES’s Peru Biodiversity Program. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bio-

con.2012.11.023. These data include Google maps of the most important areas described in this article. References Anderson, R., May, R., 1979. Population biology of infectious diseases: Part 1. Nature 280, 361–367. Angulo, A., 2008. Conservation needs of Batrachophrynus and Telmatobius frogs of the Andes of Peru. Conserv. Soc. 6, 328–333. Barrionuevo, J.S., Ponssa, M.L., 2008. Decline of three species of the genus Telmatobius (Anura: Leptodactylidae) from Tucuman province, Argentina. Herpetologica 64, 47–62. Barrionuevo, S., Mangione, S., 2006. Chytridiomycosis in two species of Telmatobius (Anura: Leptodactylidae) from Argentina. Dis. Aquat. Organ. 73, 171–174. Berger, L., Speare, R., Daszak, P., Green, D.E., Cunningham, A.A., Goggin, C.L., Slocombe, R., Ragan, M.A., Hyatt, A.D., McDonald, K.R., Hines, H.B., Lips, K.R., Marantelli, G., Parkes, H., 1998. Chytridiomycosis causes amphibian mortality associated with population declines in the rainforests of Australia and central America. Proc. Natl. Acad. Sci. USA 95, 9031–9036. Blaustein, A.R., Romansic, J.M., Scheessele, E.A., Han, B.A., Pessier, A.P., Longcore, J.E., 2005. Interspecific variation in susceptibility of frog tadpoles to the pathogenic fungus Batracbochytrium dendrobatidis. Conserv. Biol. 19, 1460–1468. Bosch, J., Martinez-Solano, I., 2006. Chytrid fungus infection related to unusual mortalities of Salamandra salamandra and Bufo bufo in the Penalara Natural Park, Spain. Oryx 40, 84–89.

Author's personal copy

A. Catenazzi et al. / Biological Conservation 159 (2013) 413–421 Bosch, J., Martinez-Solano, I., Garcia-Paris, M., 2001. Evidence of a chytrid fungus infection involved in the decline of the common midwife toad (Alytes obstetricans) in protected areas of central Spain. Biol. Conserv. 97, 331–337. Boyle, D.G., Boyle, D.B., Olsen, V., Morgan, J.A.T., Hyatt, A.D., 2004. Rapid quantitative detection of chytridiomycosis (Batrachochytrium dendrobatidis) in amphibian samples using real-time Taqman PCR assay. Dis. Aquat. Organ. 60, 141–148. Briggs, C.J., Knapp, R.A., Vredenburg, V.T., 2010. Enzootic and epizootic dynamics of the chytrid fungal pathogen of amphibians. Proc. Natl. Acad. Sci. USA 107, 9695–9700. Briggs, C.J., Vredenburg, V.T., Knapp, R.A., Rachowicz, L.J., 2005. Investigating the population-level effects of chytridiomycosis: an emerging infectious disease of amphibians. Ecology 86, 3149–3159. Catenazzi, A., Lehr, E., Rodriguez, L.O., Vredenburg, V.T., 2011. Batrachochytrium dendrobatidis and the collapse of anuran species richness and abundance in the upper Manu National Park, southeastern Peru. Conserv. Biol. 25, 382–391. Catenazzi, A., Vredenburg, V.T., Lehr, E., 2010. Batrachochytrium dendrobatidis in the live frog trade of Telmatobius (Anura: Ceratophryidae) in the tropical Andes. Dis. Aquat. Organ. 92, 187–191. Conradie, W., Weldon, C., Smith, K.F., du Preez, L.H., 2011. Seasonal pattern of chytridiomycosis in common river frog (Amietia angolensis) tadpoles in the South African grassland biome. Afr. Zool. 46, 95–102. De la Riva, I., Lavilla, E.O., 2008. Conservation status of the Andean frogs of the genera Telmatobius and Batrachophrynus. In: Stuart, S.N., Hoffmann, M., Chanson, J.S., Cox, N.A., Berridge, R.J., Ramani, P., Young, B.E. (Eds.), Threatened Amphibians of the World. Lynx Edicions, Barcelona, Spain, p. 101. Fellers, G.M., Green, D.E., Longcore, J.E., 2001. Oral chytridiomycosis in the mountain yellow-legged frog (Rana muscosa). Copeia 2001, 945–953. Garner, T.W.J., Walker, S., Bosch, J., Leech, S., Rowcliffe, J.M., Cunningham, A.A., Fisher, M.C., 2009. Life history tradeoffs influence mortality associated with the amphibian pathogen Batrachochytrium dendrobatidis. Oikos 118, 783–791. Gosner, K.L., 1960. A simplified table for staging anuran embryos and larvae with notes on identification. Herpetologica 16, 183–190. Hero, J.M., Williams, S.E., Magnusson, W.E., 2005. Ecological traits of declining amphibians in upland areas of eastern Australia. J. Zool. 267, 221–232. Hyatt, A.D., Boyle, D.G., Olsen, V., Boyle, D.B., Berger, L., Obendorf, D., Dalton, A., Kriger, K., Hero, M., Hines, H., Phillott, R., Campbell, R., Marantelli, G., Gleason, F., Colling, A., 2007. Diagnostic assays and sampling protocols for the detection of Batrachochytrium dendrobatidis. Dis. Aquat. Organ. 73, 175–192. Kilpatrick, A.M., Briggs, C.J., Daszak, P., 2010. The ecology and impact of chytridiomycosis: an emerging disease of amphibians. Trends Ecol. Evol. 25, 109–118. Knapp, R.A., Briggs, C.J., Smith, T.C., Maurer, J.R., 2011. Nowhere to hide: impact of a temperature-sensitive amphibian pathogen along an elevation gradient in the temperate zone. Ecosphere 2, 1–26. Knapp, R.A., Morgan, J.A.T., 2006. Tadpole mouthpart depigmentation as an accurate indicator of chytridiomycosis, an emerging disease of amphibians. Copeia, 188– 197. Kriger, K.M., Hero, J.M., Ashton, K.J., 2006. Cost efficiency in the detection of chytridiomycosis using PCR assay. Dis. Aquat. Organ. 71, 149–154. Lehr, E. (Ed.) 2005. The Telmatobius and Batrachophrynus species of Peru. Monografías de Herpetología 7, 39–64. Merino-Viteri, A., Coloma, L.A., Almendáriz, A. (Eds.), 2005. Los Telmatobius de los Andes de Ecuador y su disminución poblacional. Monografías de Herpetología 7, 9–37.

421

Mitchell, K.M., Churcher, T.S., Garner, T.W.J., Fisher, M.C., 2008. Persistence of the emerging pathogen Batrachochytrium dendrobatidis outside the amphibian host greatly increases the probability of host extinction. Proc. Roy. Soc. B: Biol. Sci. 275, 329–334. Newcombe, R.G., 1998. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat. Med. 17, 857–872. Piotrowski, J.S., Annis, S.L., Longcore, J.E., 2004. Physiology of Batrachochytrium dendrobatidis, a chytrid pathogen of amphibians. Mycologia 96, 9–15. Puschendorf, R., Carnaval, A.C., VanDerWal, J., Zumbado-Ulate, H., Chaves, G., Bolaños, F., Alford, R.A., 2009. Distribution models for the amphibian chytrid Batrachochytrium dendrobatidis in Costa Rica: proposing climatic refuges as a conservation tool. Divers. Distrib. 15, 401–408. Seimon, T.A., Seimon, A., Daszak, P., Halloy, S.R.P., Schloegel, L.M., Aguilar, C.A., Sowell, P., Hyatt, A.D., Konecky, B., Simmons, J.E., 2007. Upward range extension of Andean anurans and chytridiomycosis to extreme elevations in response to tropical deglaciation. Glob. Change Biol. 13, 288–299. Sinsch, U., 1986. Anfibios de la Sierra central del Perú – una clave de identificación para adultos y larvas. Boletín de Lima VIII, 23–33. Sinsch, U., 1990. Froschlurche (Anura) der zentral-peruanischen Anden: Artdiagnose, Taxonomie, Habitate, Verhaltensökologie. Salamandra 26, 177– 214. Sinsch, U., Salas, A.W., Canales, V., 1995. Reassessment of central Peruvian Telmatobiinae (genera Batrachophrynus and Telmatobius). I. Morphometry and classification. Alytes 13, 14–44. Smith, K.G., Weldon, C., Conradie, W., du Preez, L.H., 2007. Relationships among size, development, and Batrachochytrium dendrobatidis infection in African tadpoles. Dis. Aquat. Organ. 74, 159–164. Vellard, J., 1955. Estudios sobre batracios andinos. III. Los Telmatobius del grupo jelskii. Memorias del Museo de Historia Natural, Universidad Nacional Mayor de San Marcos 4, 1–28. von May, R., Catenazzi, A., Angulo, A., Brown, J.L., Carrillo, J., Chávez, G., Córdova, J.H., Curo, A., Delgado, A., Enciso, M.A., Gutiérrez, R., Lehr, E., Martínez, J.L., MedinaMüller, M., Miranda, A., Neira, D.R., Ochoa, J.A., Quiroz, A.J., Rodríguez, D.A., Rodríguez, L.O., Salas, A.W., Seimon, T., Seimon, A., Siu-Ting, K., Suárez, J., Torres, C., Twomey, E., 2008. Current state of conservation knowledge on threatened amphibian species in Peru. Trop. Conserv. Sci. 1, 376–396. Vredenburg, V.T., Knapp, R.A., Tunstall, T.S., Briggs, C.J., 2010. Dynamics of an emerging disease drive large-scale amphibian population extinctions. Proc. Natl. Acad. Sci. USA 107, 9689–9694. Vredenburg, V.T., Summers, A.P., 2001. Field Identification of chytridiomycosis in Rana muscosa. Herpetol. Rev. 32, 151–152. Walker, S.F., Bosch, J., Gomez, V., Garner, T.W.J., Cunningham, A.A., Schmeller, D.S., Ninyerola, M., Henk, D.A., Ginestet, C., Arthur, C.P., Fisher, M.C., 2010. Factors driving pathogenicity vs. prevalence of amphibian panzootic chytridiomycosis in Iberia. Ecol. Lett. 13, 372–382. Woodhams, D.C., Alford, R.A., Briggs, C.J., Johnson, M., Rollins-Smith, L.A., 2008. Lifehistory trade-offs influence disease in changing climates. Strategies of an amphibian pathogen. Ecology 89, 1627–1639. Woodhams, D.C., Bosch, J., Briggs, C.J., Cashins, S., Davis, L.R., Lauer, A., Muths, E., Puschendorf, R., Schmidt, B.R., Sheafor, B., Voyles, J., 2011. Mitigating amphibian disease: strategies to maintain wild populations and control chytridiomycosis. Front. Zool. 8.