Chytrid Treatments and Their Compatibility with Amphibian Tissue


Batrachochytrium dendrobatidis (Bd) is an aquatic amphibian pathogen (Berger et al. 1998). It was first found and associated with amphibian population declines in 1998; however, cases of infection have been found in specimens from as early as 1968 (Richards-Hrdlicka 2012). Bd has infected amphibians all over the world and in extreme cases has caused extinction (Stuart et al. 2004). Although many strains have developed that are specific to their regions, the pathogen is thought to have originated in Brazil (Lips 2014). It is not known how the fungus spread, but there are theories that amphibian trade, migratory birds, and reservoir species caused the spread of the disease (Garmyn et al. 2012; McMahon et al. 2012).

Bd is an aquatic chytrid fungus and the first chytrid found to have a vertebrate host. It “feeds” off of the keratinized skin tissue of amphibians. The fungus dwells in the epidermis, causing thickening of the stratum corneum (the superficial layer of the skin), mostly in the skin of the “ventral abdomen, hind limbs, and feet” (Longcore, Pessier, and Nichols 1999). Swelling of the “drinking patch” found in the hindquarters is a common symptom. This thickening disrupts the osmoregulatory function of the skin, causing dehydration (from a decline in electrolytes), weight loss, and eventually cardiac arrest (Voyles et al. 2012). Lesions can also occur in the internal organs of the host (Berger and Speare 1998).

The fungus is unicellular, starting as a mobile zoospore with a single flagellum. This allows the fungus to infect its host in the water. After the zoospore is lodged into the epidermis, it grows and becomes a stationary thallus. This thallus uses asexual reproduction to cleave its cytoplasm to form two new zoospores (Berger et al. 2005). This life cycle allows infection to both worsen in the host and spread to other individuals.

Antimicrobial peptides serve as a possible treatment against infection. Temporin peptides, produced in dermal granular glands, defend against general skin infections in amphibians and other animals (Simmaco et al. 1996). These temporins have shown to be effective against gram-positive bacteria because of the peptide’s alpha-helix structure. This alpha-helix allows the peptide to attach to Bd’s membrane and causes lysing (Rollins-Smith 2003). Other peptides, such as Magainin II and I, are effective at inhibiting zoospore growth (Ramsey et al.), hinting that peptides could help amphibians defend against infection.

Another viable treatment involves probiotic bacteria. Bacteria can be grown on the skin of amphibians to defend against Bd (Antwis 2014). Bacteria found in the genus Bacillus has shown to defend against Bd. This genus can is also commonly found on amphibian skin. This would allow it to be a viable treatment for the disease (Harris et al. 2006). The bacteria Janthinobacterium lividum also significantly decreases the infection in amphibians exposed to Bd. Amphibians exposed to Bd after treatment with the bacteria have normal growth compared to amphibians that are healthy. This bacterium also prevents zoospore growth in the skin of amphibians (Harris et al. 2009).

A novel approach to probiotic treatments can also be taken with bacteria found in sourdough bread. The bacteria Lactobacillus hammesii releases an acid that exhibits antifungal activity, preventing mold growth in bread (Black et al. 2013). Because of this antifungal activity, this bacterium should be able to defend against Bd. It should also be safe as a treatment for amphibians since it is a non-toxic bacteria.

The facility in which experimentation was conducted is a BSL-1 laboratory. Because of the restrictions for BSL-1 labs, Bd could not be used since it is a pathogen to amphibians. For sake of experimentation, a model species of the fungus had to be used. Homolaphlyctis polyhriza (Hp) is a non-pathogenic fungus in the same order of Bd (Rhizydietes) (Encyclopedia of Life 2013). In fact, Hp is the most closely related species to Bd (Longcore, Letchr, and James 2011), with only a 10% genetic difference between the two species. Hp is an ideal model species for experimentation.

There are many possible treatments for Chytridiomycosis; however, not much has been done to compare these treatments for both effectiveness and compatibility with amphibian tissue. This experiment compared treatments that have already been proven effective, along with a novel treatment using the bacteria Lactobacillus hammesii. It is expected that the Magainin II peptide will be most effective against the fungus and the most compatible with amphibian tissue. It is also expected that L. hammesii will be effective and compatible as a novel treatment for Bd infections. This is assuming that the model species, Hp, reacts to the treatments as Bd would and that the amphibian cells react the way that cells would react on an amphibian in vivo.



Hp cells

Homolaphlyctis polyrhiza thali (mature)

In this experiment there were two phases. The first determined the effectiveness of each treatment for Chytridiomycosis, and the second determined the compatibility of those treatments with amphibian tissue. The treatments used against the fungus were the bacteria Lactobacillus hammesii, Bacillus subtilis, and Janthinobacterium lividum and the Magainin II peptide. B. subtilis, J. lividum, and Magainin II were chosen as treatments because they have shown activity against B. dendrobatidis (Harris et al. 2006; Harris et al. 2009; Ramsey et al. 2010). L. hammesii was chosen as a novel treatment because it has shown antifungal activity in bread (Black et al. 2013) and should be a safe treatment for amphibians.

Hp liquid

Hp liquid media with cultures 

The fungal species Homolaphlyctis polyrhiza was used as the model species for B. dendrobatidis. A non-pathogenic model had to be used for Bd because of the BSL-1 laboratory. H. polyrhiza is non-pathogenic yet is both genetically and morphologically similar to Bd as it is the closest relative to the pathogen (Longcore, Letcher, and James 2011). This made it perfect for experimentation.

The cultures for the Phase I bacterial treatments (L. hammesii, B. sibtilis, and J. lividum) were grown in test tubes. The cultures were then streaked onto PmTG agar plates (Harris et al. 2006). There were four plates for each treatment, representing a trial each. The fungus was introduced to each plate by adding 1 mL of PmTG media containing Hp.

For the Maiganin II peptide treatment, a solution of the peptide was made. To apply the treatment, 1 mL of the solution was added to four plates each. Afterwards, 1 mL of the PmTG media with Hp was added to each plate. There was also a control treatment in which the Hp was added to PmTG agar plates without any treatment.

All plates were placed in an incubator at 23⁰C and were allowed to grow for 3-4 days, or until there was substantial growth. After this growth period, pictures of all 25 plates were taken. These photos were then analyzed using the imaging software Image J. This program allowed the area of the fungal colonies to be taken. This area was then taken as a percentage as the whole area of the petri dish.

kidney cells

Xenopus laevis kidney cells

In Phase 2, kidney cells from Xenopus laevis (African clawed frogs) in the American Type Culture Collection (ATCC) were used as the amphibian tissue model. Kidney cells were used because they were available from the ATCC; there were no skin cell lines available for use. The cells were cultured in well plates in a sterile glass jar with 5% CO2 levels at room temperature.

applying treatment

Applying treatments to kidney cells

Before adding treatments, observations were made of the media (looking for cloudiness or discoloration that would indicate contamination). All of the treatments were added to fresh media, which was then added to the cultures. There were four trials for each treatment (Magainin II, L. hammesii, B. subtilis, and J. lividum). The control received media without additives. These treatments were left with the cell cultures for three hours. Afterwards, 500 µL of the media was removed and replaced with 500 µL of Lactate Dehydrogenase (LDH) Cytotoxicity Assay.

After 30 minutes, 1 mL of the mixture was pipetted into a sterile, lint-free cuvette. Each trial for each treatment was placed in a Spectro Vis at 492 nm and the absorbance for each treatment was taken.

LDH Cytotoxicity Assay measures the LDH released into media by damaged cells. Higher absorbance means that more LDH was released and more amphibian cells were killed or damaged.

Data Analysis

kidney cells with treatment

Kidney cells with treatments

The amount of Lactate Dehydrogenase (LDH) was measured from the amphibian tissue after application of the treatments. This was performed by running an assay and then measuring the absorbance. A higher absorbance is directly related to the amount of LDH released from the cells. Damaged or dead cells would release more LDH.

Most of the treatments have a similar absorbance reading, at about 0.12 nm. Standard error bars are shown to give standard deviation. The treatment MII had a large standard deviation because of a huge outlier. The other three data points for Magainin II were about the same, so the outlier was removed to gain more conclusive results. The higher absorbance in the outlier could have been a result of using a dirty cuvette or running the absorbance test on the wrong side of the cuvette.  

graph 1

Graph 1 represents the altered data (removal of the outlier). The average and standard deviation of the MII treatment are significantly decreased. However, the treatment containing this peptide has the largest absorbance. This shows that more cells were damaged with this particular treatment.

Numerically, the most compatible treatment with amphibian tissue is the Bacillus subtilis bacteria. However, the standard deviations between the treatments overlap, meaning there is no statistically significant difference between their effects on amphibian tissue.

To further analyze the data, Analysis of Variance (ANOVA) was taken, with a confidence level of 95% (p ≤ 0.05). The resulting values from the original set of data (with the outlier) was p = 0.321 and F =1 .28. P = 0.611 and F = 0.69 from the second set of data (Graph 1). Both of these p values are greater than 0.05, and the first F value is close to 1. This means that the difference between the treatments’ effects on the tissue is not statistically significant. Since all treatments have a similar effect on the tissues compared to the control, they are considered safe for amphibian tissues. These treatments do not damage or kill amphibian tissues based on the data from observing their effects on the amphibian kidney cells.

graph 2

Graph 2, shown above, presents the efficiency of each treatment at killing Hp. The area of the fungal colonies was taken using imaging software. The percent of the fungus’ area relative to the petri dish was then calculated (the area of each petri dish was about 5,674 mm2).

A lower percentage of the fungal colonies corresponds with a higher efficiency. The treatments with the lowest averages were Bacillus subtilis and Lactobacillus hammesii. These treatments were also significantly more efficient than the other two treatments. Numerically, the Lactobacillus hammesii is the most efficient, but since the standard deviation overlaps with that of the Bacillus subtilis, they are statistically similar. Compared to the control, Janthinobacterium lividum and Magainin II were the same and had little to no effect on the fungus.

ANOVA with a confidence level of 95.0% was taken. The resulting p value was greater than 0.001 and F = 56.51. A low p value along with a large F value shows that the difference between the treatments is statistically significant.



Treatments previously proven to be effective at controlling Barachochytrium dendrobatidis were compared in both their effectiveness and their compatibility with amphibian cells. A novel treatment, Lactobacillus hammesii, was also tested in these aspects.

This novel bacterial treatment proved to be the most effective in controlling Homolaphlycits polyrhiza growth. On one of the plates (Trial 2) there was some Hp growth, but the fungus looked “filmy” and unhealthy (see Appendix B). The bacterial treatment, Bacillus subtilis, was also very effective. This bacterial treatment showed a clear zone of inhibition between the bacteria and the fungus (Trial 4).

The other treatments that were expected to be effective had little to no effect on the model fungus. Janthinobacterium lividum and Magainin II had similar results as the control. On some plates, it appeared that the Hp was growing over the Janthinobacterium lividum (Trials 2 and 4).

Data from the compatibility test showed that all treatments had no effect on the amphibian kidney cells. Using this model, no treatments are expected to be harmful to amphibian tissue. Other research has shown that Magainin II peptides can inhibit zoospore growth (Ramsey et al. 2010); this experiment supports this research.

The hypothesis that Magainin II would be the most compatible was partially correct. Maganin II did not harm the cells, and neither did the other treatments. The hypothesis that Lactobacillus hammesii would be an effective treatment was supported. It was the most effective treatment at controlling the population of the fungus. The goal of finding a viable treatment for Hp was achieved, and the goal of finding a treatment compatible with amphibian tissue was partially achieved.

L. hammesii could be a very effective and safe treatment for amphibians infected with Bd. It is effective at controlling Hp growth and is compatible with amphibian cells. This treatment could be further tested by observing its effects on the pathogen Batrachochytrium dendrobatidis and on the whole amphibian organism. Since it has proven effective with the models, it is expected to be an effective treatment against Chytridiomycosis.

While processing the photos with image J, the total area of fungal growth may not have been completely accounted for, but the results are only slightly altered, if at all. Using the pathogen rather than a model fungus would give more conclusive results. But since these two species are so closely related, similar results are expected. Eventually, developments for the application of L. hammesii as a treatment should be explored.

Finding an effective treatment for Chytridiomycosis is a large step in the process of developing a cure. This research is critical because it gives the amphibian population a chance to thrive. Many species have suffered from this disease, and some populations have gone extinct. Using effective treatments can help amphibians worldwide overcome infection.



This research would not be possible without the support of many individuals. Dr. Joyce Longcore, of Maine University, donated the fungus for experimentation. Dr. Louis Rollins-Smith, of Vanderbilt University, provided information via email that was helpful for my experimentation. Dr. Michael Gänzel, of the University of Alberta, was able to donate the bacteria Lactobacillus hammessi.



Antwis, R.E., et al. “Ex situ Diet Influences the Bacterial Community Associated with the Skin of the Red-Eyed Tree Frogs (Agalychnis callidryas).” PLoS ONE 9.1 (2014). 

Berger, L., and R. Speare. “Chytridiomycosis: A new disease of wild and captive amphibians.” ANZCCART Newsletter 11.4 (1998): 1-3.

Berger, L., et al. “Chytridiomycosis causes amphibian mortality associated with population  declines in the rain forests of Australia and Central America. PNAS 95 (1998): 9031-6.

Berger, L., A.D. Hyatt, R. Speare, and J.E. Longcore. “Life cycle stages of the amphibian chytrid Batrochochytrium dendrobatidis.” Diseases of Aquatic Organisms 68 (2005): 51-63.

Black, B.A., E. Zanini, J.M. Curtis and M.G. Ganzle. “Antifungal Hydroxy Fatty Acid Produced during Sourdough Fermentation: Microbial and Enzymatic Pathways and Antifungal Activity in Bread.” Department of Agricultural, Food and Nutritional Science 79.6 (2013): 1866-1873.

Garmyn, A., et al. “Waterfowl: Potential Environmental Reservoirs of the Chytrid Fungus Batrachochytrium dendrobatidis.” PlosONE 7.4 (13 April 2012).

Garner, T.W., et al. “The Emerging Amphibian Pathogen Batrachochytrium dendrobatidis Globally Infects Introduced Populations of North American Bullfrog, Rana catesbeinana.” Biol Lett 2.3 (22 Sept. 2006): 455-9.

Harris, R.N., et al. “Skin microbes on frogs prevent morbidity and mortality caused by a lethal skin fungus.” International Society for Microbial Ecology 3 (2009): 818-824.

Harris, R.N., T.Y. James, A. Lauer, M.A. Simon, and A. Patel. “Amphibian Pathogen Batrachochytrium dendrobatidis is inhibited by the Cutaneous Bacteria of Amphibian Species.” Eco. Health 3 (2006): 53-56.

Lips, K. “A Tale of Two Lineages: Unexpected Long-Term Persistence of the Amphibian-Killing Fungus in Brazil.” Molecular Ecology 23 (2014): 747-749.

Longcore, J.E., P.M. Letcher, and T.Y. James. “Homolaphyectis polyrhize gen. et sp. Nov, a species in the Rhizophydiales (Chitridiomycetes) with multiple rhizoidal axes.” Mycotaxon 118 (2011): 433-440.

Longcore, J.E., A.P. Pessier, and D.K. Nichols. “Batrachochytrium dendrobatidis gen. sp. Nov, a chytrid pathogenic to amphibians.” Mycologia 91.2(1999):219-227.

McMahon, T.A., et al. “Chytrid fungus Batrachochytrium dendrobatidis has nonamphibian host and releases chemicals that cause pathology in the absence of infection.” PNAS 110.1 (2 Jan. 2013): 210-215.

Ramsey, J.P., L.K. Reinert, L.K. Harper, D.C. Woodhams, and L.A. Rollins-Smith. “Immune Defenses against Batrachochytrium dendrobatidis, a Fungus Linked to Global Amphibian Declines in South African Clawed Frogs, Xenopus laevis.” Infection and Immunity 79.9 (2010): 3981-3992.

Richards-Hrdlicka, K.L. “Extracting the amphibian chytrid fungus from formalin-fixed specimens.” Methods in Ecology and Evolution 3 (2012): 842-844.

Rollins-Smith, L.A., et al. “Activities of Temporin Family Peptides against the Chytrid Fungus (Batrachochytrium dendrobatidis) Associated with Global Amphibian Declines.”   Antimicrobial Agents and Chemotherapy 47.3 (2003): 1157-1160.

Simmaco, M., et al. “Temporins, antimicrobial peptides from the European red frog Rana temporaria.” Ear. J. Biochem 242 (1996): 788-792.

Stuart, S.N., et al. “Status and Trends of Amphibian Declines and Extinctions Worldwide.” Science 306.5702 (3 Dec. 2004): 1783-1786.

Voyles, J., V.T. Vredenburg, T.S. Tunstal, J.M. Parker, C.J. Briggs, and E.B. Rosenblum. “Pathophysiology in Mountain Yellow-Legged Frogs (Rana muscosa) during a Chytridiomycosis Outbreak.” PLoS ONE 7.4 (2012).

Walker, S., et al. “Environmental detection of Batrachochytrium dendrobatidis in   a temperate     climate.” Disease of Aquatic Organisms 77 (2007): 105-112.


Appendix A (Raw Data)

Percentage of Hp Post-Treatment

Treatment Mll % control % Bs % Lh % Jl %
Trail 1 3459.253 60.96 4411.4 77.74 0 0 771.802 13.60 4232.226 74.58
Trail 2 3907.843 68.86 4684.732 82.55 716.818 12.63 133.636 2.35 3556.866 62.68
Trail 3 4022.439 70.88 4691.36 82.67 1412.194 24.88 422.039 7.43 4627.083 81.54
Trail 4 3745.732 66.00 3969.27 69.94 1752.068 30.87 57.676 1.01 3569.918 62.91
Average %   66.68   78.23   17.09   6.10   70.42


Absorbance of LDH

Treatment Bs Lh Jl M ll Control
Trail 1 0.117 0.13 0.124 0.401 0.108
Trail 2 0.126 0.11 0.12 0.12 0.112
Trail 3 0.115 0.12 0.11 0.128 0.149
Trail 4 0.107 0.125 0.124 0.142 0.124
Average 0.11625 0.12125 0.1195 0.19775 0.12325


Appendix B (PmTG plates with Hp and Treatment)

Bacillus subtilis

Bacillus subtilis

Trial 1 (top left). Trial 2 (top right). Trial 3 (bottom right). Trial 4 (bottom left).


Lactobacillus hammesii

lactobacilllus hammesii

Trial 1 (top left). Trial 2 (top right). Trial 3 (bottom right). Trial 4 (bottom left).


Janthinobacterium lividum

janthinobacterium lividum

Trial 1 (top left). Trial 2 (top right). Trial 3 (bottom right). Trial 4 (bottom left).


Magainin II


Trial 1 (top left). Trial 2 (top right). Trial 3 (bottom right). Trial 4 (bottom left).




Trial 1 (top left). Trial 2 (top right). Trial 3 (bottom right). Trial 4 (bottom left).