Bacterial Community in the Skin Microbiome of Frogs in a Coldspot of Chytridiomycosis Infection

Chytridiomycosis is a fungal disease caused by the pathogens, Batrachochytrium dendrobatidis (Bd) and B. salamandrivorans (Bsal), which has caused declines in amphibian populations worldwide. Asia is considered as a coldspot of infection, since adult frogs are less susceptible to Bd-induced mortality or morbidity. Using the next-generation sequencing approach, we assessed the cutaneous bacterial community composition and presence of anti-Bd bacteria in six frog species from India using DNA isolated from skin swabs. All the six frog species sampled were tested using nested PCR and found Bd negative. We found a total of 551 OTUs on frog skin, of which the bacterial phyla such as Proteobacteria (56.15% average relative abundance) was dominated followed by Actinobacteria (21.98% average relative abundance) and Firmicutes (13.7% average relative abundance). The contribution of Proteobacteria in the anti-Bd community was highest and represented by 175 OTUs. Overall, the anti-Bd bacterial community dominated (51.7% anti-Bd OTUs) the skin microbiome of the frogs. The study highlights the putative role of frog skin microbiome in affording resistance to Bd infections in coldspots of infection.

Over a decade, amphibians' (frogs, caecilians, and salamanders) skin microbiome has been studied extensively to understand their role in the amphibian panzootic disease, called chytridiomycosis [1][2][3]. The disease is caused by two aquatic fungal pathogens: Batrachochytrium dendrobatidis (Bd) infects frogs, salamanders, and caecilians [4], and B. salamandrivorans (Bsal) infects both frogs and salamanders [5,6]. They have caused declines in over 500 amphibian species, and possible extinctions of 90 species, worldwide [7]. In adult frogs, the skin microbiome plays an important role in the host's defense mechanism [8]. The frog's skin microbiome produces a plethora of anti-fungal metabolites to protect the host [2,8], and the bacterial community on the skin might confer resistance to pathogens [9]. Skin microbiomes of frogs have revealed some important anti-Bd bacteria, such as Janthinobacterium lividum (family Oxalobacteraceae) [2,8] and Serratia marcescens (family: Yersiniaceae) that produce anti-fungal metabolites: "Violacein" [8] and "Prodigiosin," respectively. Both violacein and prodigiosin compromise the integrity of the cell membrane [10,11], and prodigiosin also inhibits RNA and protein synthesis in bacteria [12]. Consequently, these two metabolites are known to inhibit Bd fungal growth [2]. Since anti-Bd bacteria are present on frogs that show resistance to Bd [13], investigating such skin microbiomes for potential anti-Bd bacterial isolates would benefit disease mitigation strategies.
There are over 395 frog species in India, and the list is growing, as new taxa are being discovered rapidly (see, http://amphibiaweb.org/). With high species richness and endemism, the stakes are high for understanding the role of Bd on frog populations, and the mechanisms by which frogs might be able to resist Bd infection. Frog populations in the region exhibit low Bd prevalence and high Bd haplotype diversity, which points at possible historical host-pathogen co-evolution [14]. Low prevalence and low mortality could a l s o i m p l y t ha t f r o g s s h o w s o m e r e s i s t a nc e t o chytridiomycosis, and this could be mediated by the skin microbiome. Investigating the role of frog's skin microbiome in coldspots of Bd infection could reveal new pathways that inhibit the pathogen. Therefore, we studied the bacterial community composition in six frog species from two hotspots for amphibian species richness, namely the Western Ghats and the Andaman Islands. We explored the frog skin microbiome data to infer the role of bacterial community in affording them with resistance to Bd infections. (Full details on the methods are provided in the Supplementary file S1. ) We generated a total of 18,543 good quality sequences from 18 frogs belonging to six species that had no Bd infection (Table 1). We aligned all good quality sequences in mothur and assigned them identity using SILVA v123 database. In all, we identified 1351 unique sequences (at ≥ 97% sequence similarity). These unique sequences revealed 551 operational taxonomic unit (OTU), belonging to five major bacterial phyla. Average relative abundance (ARA) of bacterial taxa in the microbiome was calculated as percentage of the total good quality bacterial sequences obtained. Proteobacteria (56.15% ARA) was the most dominant phylum, followed by Actinobacteria (21.98% ARA), Firmicutes (13.7% ARA), Bacteroidetes (7.2% ARA), and Acidobacteria (0.49% ARA). Acidobacteria was present only in B. beryet (1.99% relative abundance) and in Limnonectes doriae (0.99% relative abundance), from the Andaman Islands. In the case of aquatic Nyctibatrachus poocha, Proteobacteria (6.98% relative abundance) was found to be the least abundant, and Actinobacteria (49.12% relative abundance) was the most abundant ( Fig. 1b). Rarefaction revealed that L. doriae had the highest and Ghatixalus asterops had the least number of OTUs (Table 1). Cluster analysis revealed that the bacterial community did not form clusters based on the niche occupied by the frogs (Fig. 1a).
Out of the 551 OTUs identified, 285 OTUs (51.7%) were anti-Bd bacterial OTUs, and they constituted 74.2% of the average relative abundance in each of the six species. The percentage of anti-Bd OTUs, represented as a fraction of the total number of OTUs recorded from each frog species, showed variation. The highest percentage of anti-Bd bacterial OTUs were present on G. asterops (171/216, 79.1% anti-Bd OTUs), followed by Boophis madagascariensis (ns_6), Colostethus panamensis (in-hibitory_2), Alytes obstetricans (ns_64), and Caulobacteraceae bacterium (KU738923.1), respectively, from Woodhams et al. [15] and Muletz et al. [16] database. The average relative abundance of these OTUs was 17.8%. Percentage of anti-Bd bacterial OTUs, represented as a fraction of the total anti-Bd OTUs in Globally, frog skin microbiomes are replete with Proteobacteria, Bacteroidetes, and Actinobacteria [17][18][19]. Proteobacteria is also abundant in soils [20], and thereby, associated with tropical frog skins [17,21], and majority of them (80% of the total OTUs present) are in the anti-Bd databases [15]. Bacterial family Pseudomonadaceae (Proteobacteria) has been reported from several frogs' skin microbiomes with varying abundance [19]. However, this bacterial family was not found in B .beryet, L. dorie, N. poocha, and R. chlorosomma (Fig. 1b). Family Xanthomonadaceae (Proteobacteria) was dominant in the bacterial community in all the frog species, except N. poocha. Actinobacteria was dominant (49.12% relative abundance), and Proteobacteria was the least represented (6.98% relative abundance) in N. poocha. Actinobacteria are associated with freshwater ecosystems [22], and N. poocha being a mountain streamdwelling frog shares the same habitat with the bacterial phylum. The differences in the bacterial phyla composition on N. poocha could also be due to seasonal differences, or changes in pH of water in streams [23].
In terrestrial frogs (L. doriae and R. chlorosomma), OTUs were higher than in aquatic (N. poocha) and arboreal (G. asterops and B. beryet) frogs (Table 1). A similar pattern has been documented in a global comparison of frog skin microbiomes [19]. Bacterial OTUs in terrestrial frogs could be attributed to their contact with soil, which has a rich bacterial community. Based on bacterial OTUs and their abundance, there was no consistent association between the niche of the frogs and their skin bacterial community composition (Fig. 1a). This lack of association suggests that frog skin bacterial community might not be an accurate descriptor of the niche of the frog.
Low abundance of anti-Bd bacteria on frog's skin has been linked to high prevalence of Bd in the hotspots of infection [23,28]. However, currently, there is no evidence suggesting that a high abundance of anti-Bd bacteria on frog's skin is associated with low prevalence of Bd infection in coldspots of infection. A rich anti-Bd bacterial community that thrives on the skin of frogs has been revealed through this study. It could be one of the reasons for the low prevalence of Bd in frog populations in the region. Anti-Bd bacteria provide frogs with protection from Bd infections in different ways: (i) by producing anti-Bd metabolite [29], (ii) by producing biofilm on frog skin surface [1], and (iii) by host-mediated selection of anti-Bd metabolite producing bacteria on its skin [30]. Since Fig. 1 Graph represents the cluster analysis and the relative abundance of the frog skin microbiome: a cluster analysis of frog species is calculated using Bray-Curtis method. Font color of species name indicates niche of frog. b Bacterial phylum-level relative abundance in six frog species. c bacterial Family level relative abundance in six frog species (Plots were prepared using R (v3.6.3) software, and modified using Inkscape 0.91 (https://inkscape.org/en/)) frog skin microbiome reports are scarce from Asia, future studies should focus on role of skin bacterial community in affording protection from Bd infection by attenuation of the pathogen [31]. Bacterial flora in the soil or stream might be an important source for the frog skin microbiome; therefore, we hypothesize that Bd might be experiencing strong selection pressure both in the environment, and on the frogs in coldspots of infection.
Acknowledgments We thank the State Forest Departments of Kerala and Andaman and Nicobar Islands for providing necessary permits and necessary logistic support. We thank Lilly Margaret and Sachin Anand for collecting samples from Western Ghats and Andaman Islands, respectively. A. Sreenivas assisted us in generating sequences using the Roche GS-FLX System, and we thank him for this. We thank the two anonymous reviewers of this manuscript for their valuable comments.

Compliance with Ethical Standards
Competing Interests The authors declare that they have no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.