Effects of Urbanization on Cutaneous Bacteria as Inhibitors of the Amphibian Fungal Pathogen Batrachochytrium dendrobatidis

Part of the Young Naturalist Awards Curriculum Collection.

by Rebecca, grade 12, New York - 2012 YNA Winner


An Eastern redback salamander in dirt with some wood chips and brown decaying leaf pieces.
An Eastern redback salamander

As I flipped over the board, a thrill overcame me: it was my first salamander find of the day under one of the many coverboards at the New York Botanical Garden. The delicate creature wriggled its head into the dirt, hoping that the minimal camouflage would be enough for it to go undetected. Scooping up the salamander and the wad of dirt underneath it, I used a Cool Whip container as a temporary observatory. The salamander had a vibrant red stripe down its back and clung to the walls of the container while I measured it. This was the first salamander I had ever collected, and from then on finding salamanders was always a pleasant surprise. These Eastern Redback salamanders were being surveyed at the New York Botanical Garden, and I was honored to be involved. Intrigued by these beautiful amphibians, I was driven to research the worldwide amphibian decline, which is currently affecting approximately one-fifth of amphibian species (Baillie et al., 2004).

This decline has been attributed to environmental disturbances and the global emergence of a pathogenic fungus, Batrachochytrium dendrobatidis (Davidson 2004; Hamer & McDonnell, 2008), the causal agent of chytridiomycosis (Rachowicz et al., 2006; Skerratt 2007). Chytridiomycosis damages the epidermis of amphibians through the keratinization of the skin (Nicols et al., 2001), ultimately leading to suffocation (Voyles et al., 2009). However, limited information is known about the appearance of Batrachochytrium dendrobatidis and chytridiomycosis in relation to the decline of amphibians, as the disease was identified less than 15 years ago (Berger et al., 1998). Similar strains of B. dendrobatidis have been found in environments with both declining and stable amphibian species, suggesting that the pathogenicity of the fungus may be dependent on environmental conditions (Blaustein & Kiesecker, 2002).

Environmental conditions have also been found to contribute directly to amphibian declines. Previous research has shown that amphibians are particularly sensitive to habitat fragmentation, limiting not only the genetic diversity of amphibian populations, but also the diversity of their cutaneous bacteria (McKinney 2002; Noël et al., 2007; Gibbs 2001). Other factors leading to the decline of amphibians are the declining diversity of native plants and the rapid spread of invasive species (Clements & Moore, 2005), which alter the stoichiometry of soil nutrients (Ehrenfeld 2003). The modification of habitat conditions could, in turn, alter the virulence of chytridiomycosis (Savage 2011); however, this hypothesis has yet to be proved. 

Amphibian declines may have cascading effects on native ecosystems because amphibians influence overall ecosystem dynamics (Davic & Welsh, 2004). Their role in the decomposition of leaf litter is believed to aid in the maintenance of carbon dioxide emissions into the environment (Wyman 1998). Ecosystem characteristics such as invertebrate populations, leaf litter decomposition, and soil pH vary with amphibian population densities, making salamanders effective indicators of the overall health of forest ecosystems (Welsh 2001). The relationship between the health of salamander populations and greater forest health has been attributed to their sensitive skin, which is necessary for respiration but also makes them vulnerable to environmental changes (Welsh 2001). Furthermore, since amphibians are indicators of the overall health of the ecosystem, their decline may imply that forest ecosystems are unstable. 

Eastern Redback salamanders (Plethodon cinereus) are an ideal species to study toexamine the effects of urbanization on native ecosystems. The species is strictly terrestrial, enabling it to have much broader ranges than salamander species that rely on bodies of water for part or all of their life stages (Wake 1987). Additionally, Eastern Redback salamanders are resistant to habitat fragmentation (Gibbs 1998), making them ideal for study in urban environments. It is also known that, like many amphibians, Eastern Redback salamanders have resident cutaneous bacterium with the ability to inhibit the growth of B. dendrobatidis (Harris et al., 2006; Lauer et al., 2007). In this mutualistic relationship, the salamanders provide a suitable environment for bacterial growth while the bacteria produce antifungal metabolites, thus protecting the amphibian from chytridiomycosis (Brucker et al., 2008).

Despite the importance of cutaneous bacteria and the sensitivity of salamanders to environmental changes, studies have not examined whether certain features of the environment influence the presence of the inhibitory bacterium. Therefore, the effect of urbanization on the presence of bacteria with the ability to inhibit B. dendrobatidis remains unknown. Research on the cutaneous bacteria of Eastern Redback salamanders could reveal the environmental conditions that are contributing to the decline of some species of amphibians. It is not known why some amphibian species are near extinction while others, including Eastern Redback salamanders, remain stable. Further understanding the decline of amphibians could provide an opportunity to understand the effect of urbanization on ecosystems and suggest steps toward the conservation of at-risk species.


The goal of this research was twofold: (a) to determine if urbanization results in a decreased ability to protect salamanders from B. dendrobatidis; and (b) to identify changes in the diversity of cutaneous bacteria on Eastern Redback salamanders.


It was predicted that: (a) urbanization would result in fewer cutaneous bacteria with the ability to inhibit the growth of the fungus; and (b) cutaneous bacterial strains would largely be unique to that location.


Defining Urbanized Land:
All locations were chosen within the New York City area. The distance from Central Park to the study location, as well as presence of invasive plant species, was used to determine whether the sites would be considered urban, suburban or rural. 

Study Sites:
My study sites included the New York Botanical Garden (urban), Rockefeller State Park (suburban), Fahnestock State Park (rural) and the White Memorial Conservation Center (rural). All of the sample locations were mixed deciduous forests. The coverboards at the New York Botanical Garden and White Memorial were pre-established; the coverboards at Rockefeller and Fahnestock were made of untreated pine, and I established them several weeks prior to salamander sampling. Five Eastern Redback salamanders were sampled at each location, beginning in the late spring.

Left: Map of study sites. Top Right: Rockerfeller State Park. Bottom Right: Fahnestock State Park
Using a spherical densitometer

Habitat data was collected to better identify the variables between the sample locations. Canopy cover was measured using a spherical densitometer at each array. Soil temperature, pH, and moisture were also taken at each array; these measurements were collected using a soil thermometer and a Kelway Soil Acidity and Moisture Tester, respectively.

Salamander Sampling:
Permits for handling salamanders and placing coverboards were obtained for all locations through the New York Department of Environmental Conservation, the New York State Office of Parks, and the Connecticut Department of Environmental Protection. Any salamanders present were captured and placed into a shallow container. Each salamander was rinsed with distilled water to remove transient bacteria before swabbing; thus, only resident microbes remained on the salamanders’ skin for sampling.

Salamander being swabbed. Photo credit Elizabeth Policello

A non-invasive sampling method was used for bacterial swabbing (Boyle 2004). Salamanders were swabbed on their left, right, and ventral sides with sterile cotton swabs (Harris 2006). Each cotton swab was streaked onto plates prepared with 1% tryptone agar and immediately sealed with Parafilm to prevent contamination. Plates were kept sealed until immediately before bacterial isolations in the lab. After the swabbing was complete, the salamander was returned to the coverboard from which it was collected. 

Challenge Assays:
Bacterial samples were grown out until individual colonies were visible. Isolations were then performed to obtain a pure culture. From each location, visually distinct bacterial isolates were noted. Appearances were used to give each species a visual identification letter (example: Species A: matte white, wispy, covers the entire plate). This system of separating unique bacterial isolates was used until proper identification can be conducted through a planned sequence analysis. Batrachochytrium dendrobatidis cultures, strain JEL 404, obtained from Dr. Joyce Longcore at the University of Maine, were cultured in a 1% tryptone broth until colonies were visible. Plates prepared with a 10% tryptone agar were then inoculated with 2 ml of the liquid B. dendrobatidis culture, and left under a laminar flow hood until dry in order to create an even distribution of the fungus. Each bacterial isolate was then streaked into a single line on the plates containing the fungus, Bd. Plates were inverted and given at least two weeks to grow out before analyzed for antifungal abilities. All challenges were conducted in a lab setting, making sure to autoclave or bleach all materials that came in contact with Batrachochytrium dendrobatidis.

A three-level scale was used to differentiate the antifungal properties of the bacteria. “No antifungal ability” was characterized as a plate covered with the fungus with the initial bacterial streak no longer visible (Figure 1). In the cases of “weak inhibition,” the initial streak was visible, but fungal growth was still prominent (Figure 2). Challenges that exhibited “strong antifungal properties” were classified either by prominent bacterial growth within the streak, or a noticeable zone of inhibition (Figure 3). Zones of inhibition were light hazy areas surrounding the streak, indicating that an antifungal metabolite had been excreted and the fungus was receding to the outside of the plate.

Left to right, three petri dishes showing: no antifungal ability, weak inhibition, and, finally, strong antifungal properties.
L to R: Figures 1, 2 and 3

Statistical Analysis

One-way ANOVAs were run between the locations for each habitat variable, and T-tests were conducted in Minitab 15 in order to identify specific differences between each location.


Figure 4: Antifungal properties of each bacterial isolate, classified by strength and location

The antifungal abilities of the bacterial isolates were compared by location. Surprisingly, the New York Botanical Garden, Rockefeller Park, and Fahnestock State Park displayed similar results, with one to three isolates in each category (Figure 4). At White Memorial, the majority of the isolates exhibited weak signs of inhibition. None of the bacterial isolates at White Memorial exhibited strong antifungal properties; thus, amphibians at White Memorial may not be adequately protected from the pathogenic fungus.

Bacterial richness and diversity can be seen from the number of visually distinct bacterial isolates obtained from swab samples. Each location had its own unique combination of bacterial isolates; however, some isolates were commonly found in multiple locations, such as Isolate A (Table 1).

Table 1: Summary of isolates found at each location

Site Total bacterial isolates Distinct bacterial isolates Reoccuring isolates Unique isolates
NYBG 8 4 A, D, E, P H, I, J, K
Rockerfeller 7 4 A, R, W S, V, Y, Z
Fahnestock 8 6 D, E B, C, F, L, M, N
White Memorial 9 5 A, P, R, W G, Q, T, U, X

Significant differences were seen between all of the sample locations for soil temperature (Table 2; p= 0.0003), with the strongest variance seen between White Memorial Conservation Center and Fahnestock State Park (p= 0.002); however, this excluded the temperature change from Fahnestock to Rockefeller (p >0.05). The three alternate habitat variables showed no significant variation between the sites (p > 0.05).

Table 2: Mean Habitat Conditions Measured at Coverboard Sites    sd= 2.05

Location Distance to Central Park pH Soil Moisture Canopy Cover Soil Temperature
NYBG (urban) 10.3 miles 5.58 56.40% 82.99% 15.44 (a) C.
Rockefeller (suburban) 31.3 miles 6.30 48.00% 86.74% 18.34 (b) C.
Fahnestock (rural) 56.5 miles 5.86 38.00% 92.56% 17.22 (c) C.
White Memorial (rural) 96.3 miles 5.82 60.00% 91.73% 14.04 (d) C.
    NS NS NS p= 0.0003


This study set out to determine the percentage of bacterial isolates in each location with the ability to inhibit the pathogenic fungus Batrachochytrium dendrobatidis. Contrary to what was expected, bacteria isolates from Fahnestock, Rockefeller, and the New York Botanical Garden all exhibited bacteria isolates with inhibitory characteristics. Only the rural location, White Memorial, had a mix without inhibitory bacterial isolates. These results suggest that location does have an impact on both bacterial richness and diversity on salamanders. 

I had predicted that urbanization would result in a decreased presence of bacteria with the ability to inhibit the growth of the fungus. However, bacteria from the New York Botanical Garden, Rockefeller, and Fahnestock State Park all exhibited bacterial isolates in the inhibition category (Figure 4). I observed similar levels of human disturbance at these three locations, in contrast to White Memorial Park, where the coverboards were set away from frequently used trails, thus limiting human disturbance. Munawar and Weisse (1989) suggested that the function of microfloral communities changes based on ecosystem stressors. Thus, the level of human disturbance in three of the chosen locations may be putting pressure on the cutaneous bacteria to inhibit the growth of the fungus. Furthermore, the results may suggest that the salamanders at White Memorial have not been exposed to the pathogenic fungus, while the bacteria at the other three locations were adapting in order to survive in its presence.

Figure 5: Obtaining a water sample

My prediction that bacterial diversity would be greater in the rural location than the urban locations was supported by the data. The two more urbanized locations only had four isolates unique to that location, versus six and five isolates at Fahnestock and White Memorial, respectively (Table 1). The diversity of the bacterial communities (an example can be seen in Figure 5) has been found to be controlled by soil pH and plant diversity (Fierer & Jackson, 2005). When comparing variables between the habitats, I found that variations in soil pH were not statistically significant (Table 2), suggesting that plant diversity may have played a greater role in altering the diversity of the cutaneous bacteria. Additionally, all of the parks investigated in this study had different proportions of native and non-native plant species present, further supporting the idea that plant diversity impacted why bacterial diversity varied so greatly between locations.

Soil temperatures differed significantly between the locations, except when Fahnestock was compared to Rockerfeller. The pH values of the soil did not vary significantly between the sites, but this result was expected. The rock content of the soil is the major contributor to its acidity, with anthropogenic disturbances contributing to an alteration of the pH (Ulrich 1986). My sample locations had similar types of bedrock due to their relatively close proximity (U.S. Geological Service 2011), which explains the similarity in their soil pH. Testing a location with a different rock content and slope might provide a larger pH variation in future studies.

This research is the first to identify a relationship between the environment and the effectiveness of cutaneous bacteria in inhibiting the growth of B. dendrobatidis. In previous literature (Berger et. al, 2004), the influence of temperature is suggested to be key to the diversity of cutaneous bacteria. Future research should consider habitat variables such as temperature.

A limitation in my study was developing a reliable urban-to-rural gradient. Urban-to-rural gradients are very subjective, as there is no set way to define them. Gradients are more complex than simply the distance from a densely populated area. Hence I chose a variety of factors to create a gradient for this study (McDonnell et. al, 1997). Future research should incorporate a GPS mapping component to better classify locations as urban, suburban, or rural.

This study was the first to examine and compare the cutaneous bacteria on salamanders along an urban-to-rural gradient. This study is also the first to document differences in the bacteria’s inhibition ability between locations, thus showing the influence of the environment on salamanders’ defense abilities. In future research, I would take a more comprehensive approach and include environmental variables such as water chemistry and environmental toxins to see their effect on the microbial communities on salamanders.

Additional amphibian species could also be considered for future research, especially those in decline. Finally, this study suggests that future conservation efforts for amphibians might focus on increasing the presence of inhibitory bacteria rather than trying to eradicate the disease, which may only exacerbate the problem.


This study is the first to document how urbanization impacts cutaneous bacteria on salamanders that inhibit the pathogenic fungus Batrachochytrium dendrobatidis. Habitat disturbance appears to influence the function of the salamanders’ microfloral community, and urbanization may actually condition an amphibian against the pathogenic fungus. It appears that soil temperature and plant diversity contribute to a change in the diversity of the salamanders’ microfloral community, and thus their susceptibility to infection. This is an exploratory study, and further research is needed to correlate specific habitat conditions to their effects on the salamanders’ microfloral community. This research could be applied to help create a targeted conservation effort that focuses on expanding the amphibians’ microbial community. This study might provide a strategy to prevent worldwide amphibian declines by helping amphibians to raise natural defenses against Batrachochytrium dendrobatidis.


Baillie, J.E.M., C. Hilton-Taylor, S.N. Stuart (eds.) The IUCN Red List of Threatened Species: A Global Species Assessment. Gland, Switzerland and Cambridge, U.K.: International Union for the Conservation of Nature and Natural Resources, 2004.

Becker, Matthew, and Reid Harris. “Cutaneous bacteria on the Eastern Redback salamander prevent morbidity associated with a lethal disease.” PLoS One 5.6 (2010).

Berger, L., et al. “Effect of Season and Temperature on Mortality in Amphibians due to Chytridiomycosis.” Australian Veterinary Journal 82.7 (2004): 31-36.

Berger, Lee, et al. “Chytridiomycosis causes amphibian mortality associated with population declines in the rain forests of Australia and Central America.” Proceedings of the National Academy of Sciences 95.15 (1998): 9031-9036.

Boyle, D.G., et al. “Rapid quantitative detection of chytridiomycosis (Batrachochytrium dendrobatidis) in amphibian samples using real-time Taqman PCR assay.” Disease of Aquatic Organisms 60 (2004): 141-148.

Brucker, R.M., et al. “Amphibian chemical defense: antifungal metabolites of the microsymbiont Janthinobacterium lividumon the salamander Plethodon cinereus.” Journal of Chemical Ecology 34 (2008): 1422-1429.

Clemants, Steven, and Gerry Moore. “ The changing flora of the New York metropolitan region.”Urban Habitats 3.1 (2005): 192-210.

Davic, R.D., and H.H. Welsh Jr. “On the ecological roles of salamanders.” Annual Review of Ecology, Evolution, and Systematics 35 (2004): 405-434.

Davidson, Carlos. “Declining downward: amphibian population declines in California and historical pesticide use.” Ecological Applications 14.6 (2004): 1892-1902.

Ehrenfeld, J.G. “Effects of exotic plant invasions on soil nutrient cycling processes.” Ecosystems 6 (2003), 503-523.

Firer, Noah, and Robert Jackson. “The Diversity and Biogeography of Soil Bacterial Communities.” Proceedings of the National Academy of Sciences 103.3 (2006): 626-631.

Gibbs, James. “Demography versus habitat fragmentation as determinants of genetic variation in wild populations.” Biological Conservation 100.1 (2001): 15-20.

Gibbs, James. “Distribution of woodland amphibians along a forest fragmentation gradient.” Landscape Ecology13 (1998): 263-68.

Hamer, Andrew, and Mark McDonnell. “Amphibian ecology and conservation in the urbanizing world: A review.” Biological Conservation 141 (2008): n. page.

Harris, R.N., et al. “Amphibian pathogen Batrachochytrium dendrobatidisis inhibited by the cutaneous bacteria of amphibian species.” EcoHealth 3 (2006): 53-56.

Hyde, Erin, and Theodore Simons. “Sampling Plethodontid Salamanders: Source of Variability.” Journal of Wildlife Management 65.4 (2001): 624-32.

Lauer, A., et al. “Common cutaneous bacteria from the Eastern Redback salamander can inhibit pathogenic fungi.” American Society of Ichthyologists and Herpetologists 3 (2007): 630-640.

Lauer, Antje, et al. “Diversity of cutaneous bacteria with antifungal activity isolated from female four-toed salamanders.” International Society for Microbial Ecology 2 (2008): 145-157.

Marco, A., C. Quilchano, and A.R. Blaustein. “Sensitivity to nitrate and nitrite in pond-breeding amphibians from the Pacific Northwest, U.S.” Environmental Toxicology and Chemistry 18.12 (1999): 2836-2839.

McDonnell, Mark, et al. “Ecosystem Processes Along an Urban-to-Rural Gradient.” Urban Ecosystem 1 (1997): 21-36.

McKinney, Michael. “Urbanization, biodiversity, and conservation.” BioScience 52.10 (Oct 2002): 883-890.

Morehouse, E.A., et al. “Multilocus sequence typing suggests the chytrid pathogen of amphibians is a recently emerged clone.” Molecular Ecology 12 (2003): 395-403.

Munawar, M., and T. Weisse. “Is the ‘Microbial Loop’ an Early Warning Indicator of Anthropogenic Stress?” Hydrobiologia 188.1 (1989): 163-74.

Nichols, D.K., E.W. Lamirande, A.P. Pessier, and J.E. Longcore. “Experimental Transmission of Cutaneous Chytridiomycosis in Dendrobatid Frogs.” Journal of Wildlife Diseases 37.1 (2001): 1-11.

Noël, Sarah, Martin Oulet, Patrick Galois, and Francoise-Joseph Lapoint. “Impact of urban fragmentation on the general structure of the Eastern Redback salamander.”  Conservation Genetics 5.10 (2007): 2077.

Rachowicz, L.J., et al. “Emerging infectious disease as a proximate cause of amphibian mass mortality.” Ecology 87.7 (2006): 1671-1683.

Rousk, Johannes, et al. “Soil bacterial and fungal communities across a pH gradient in an arable soil.” International Society for Microbial Ecology Journal 4 (6 May 2010): 1340-1351.

Savage, Anna, Michael Sredl, and Kelly Zamudio. “Disease dynamics vary spatially and temporally in a North American amphibian.” Biological Conservation 144.6  (June 2011); 1910-1915.

Schweitzer, Peter. “New York Geology.” U.S. Geological Survey, U.S. Department of the Interior, 4 Oct 2011. Retrieved on 14 Nov 2011 from http://mrdata.usgs.gov/sgmc/ny.html.

Skerratt, Lee Francis, et al. “Spread of chytridiomycosis has caused the rapid global decline and extinction of frogs.” EcoHealth 4 (2007): 125-134.

Ulrich, Bernhard. “Natural and Anthropogenic Components of Soil Acidification.” Journal of Nutrition and Soil Science 149.6 (1986): 702-717.

Voyles, J., et al. “Pathogenesis of chytridiomycosis, a cause of catastrophic amphibian declines.” Science 326.5952 (2009): 582-585.

Wake, David. “Adaptive Radiation of Salamanders in Middle American Cloud Forests.” Annals of the Missouri Botanical Garden 74. 2 (1987): 242-264.

Welsh, Hartwell H. and Sam Droege. “A case for using Plethodontid salamanders for monitoring the biodiversity and ecosystem integrity of North American forests.” Conservation Biology 15.3 (2001): 558-569.

Wyman, Richard L. “Experimental assessment of salamanders as predators of detrital food web effects on invertebrates, decomposition and the carbon cycle.” Biodiversity and Conservation 7.5 (1998): 641-650.


I am very thankful for my mentors, Dr. Jim Lewis at Fordham University and Ms. Jessica Arcate Schuler at the New York Botanical Garden, for their aid in developing techniques and providing locations at which I could conduct my research. I would also like to thank Dr. Joyce Longcore at University of Maine, Orono, for graciously providing access to Batrachochytrium dendrobatidis cultures. I am grateful for the access to all of the participating parks: the New York Botanical Garden, Rockefeller State Park, Clarence Fahnestock State Park, and White Memorial Conservation Center. I would also like to thank the New York Department of Environmental Conservation, the Connecticut Department of Environmental Protection, and the New York State Parks Department for their assistance in obtaining permits. Finally, I would like to thank my teachers, Mr. Angelo Piccirillo, and Ms. Valerie Holmes, for their vital support and guidance, as well as my parents, for their undying encouragement and support.