Epidermal galactose spurs chytrid virulence and predicts amphibian colonization
Batrachochytrium salamandrivorans (B. salamandrivorans) culture conditions and zoospore isolationB. salamandrivorans type strain (AMFP 13/01)8 was grown in tryptone-gelatin hydrolysate-lactose (TGhL) broth and incubated for 5−7 days at 15 °C. Zoospores were harvested by replacing the TGhL broth with distilled water. The collected water was filtered through a sterile mesh filter with pore size 10 µm (Pluristrainer, PluriSelect) to remove sporangia. Zoospore viability and mobility were confirmed using light microscopy.Salamander skin lysate binding assayBinding of B. salamandrivorans spores to the protein or carbohydrate fractions from fire salamander (Salamandra salamandra) skin was tested by treating fire salamander sloughed skin lysates enzymatically with glycoside hydrolases, followed by protein precipitation. An overview of the skin lysate binding assay is shown in Supplementary Fig. 3.To collect the sloughed skin, ten captive-bred adult fire salamanders were housed at 15 ± 1 °C on moist tissue. The sloughed skin samples were ground with liquid nitrogen into a fine powder and then homogenized, using 3 ml RadioImmunoprecipitation assay (RIPA) buffer (Sigma-Aldrich) per gram of tissue. Samples were incubated for 1 h at 4 °C, centrifuged at 27.000 × g for 10 min and the supernatant was subsequently collected. Protein concentration was determined using the PierceTM BCA Protein Assay Kit (Thermo Fisher Scientific). The obtained skin lysate was equally divided, one part was treated with Protein Deglycosylation Mix II and two parts were kept as crude skin lysates. Protein Deglycosylation Mix II (New England BioLabs) was used to remove N-linked and O-linked glycans from glycoproteins. According to the manufacturer’s instructions, 5 µl 10× Deglycosylation Mix Buffer I and 5 µl Protein Deglycosylation Mix II were added to 40 µl skin lysate. The mixture was incubated at 37 °C for 16 h. Protein precipitation was conducted on the redundant Protein Deglycosylation Mix II treated and crude skin lysates. The precipitation was performed by slowly adding saturated ammonium sulfate solution to the skin lysates to achieve a final concentration of 75%. Samples were then centrifuged at 21.130 × g for 30 min to separate the precipitated proteins from the supernatant. The precipitated protein pellets were resuspended in 300 µl of 0.05 M carbonate−bicarbonate coating buffer (3.7 g NaHCO3, 0.64 g Na2CO3, 1 L distilled water, pH 9.6). Each skin lysate solution was adjusted to the volume of 300 µl by adding a coating buffer. One hundred µl of each skin lysate solution was coated in each well of 96-well polystyrene microtiter plates (MaxiSorpTM plate, Thermo Fisher Scientific) in three technical replicates. As controls, coating buffer (negative control) and 75% ammonium sulfate solution were also coated on the 96-well plates. After incubation at 4 °C for 24 h the coated plates were washed three times with washing buffer (0.01 M PBS-Tween 20, pH 7.4) and blocked with 1% BSA overnight at 4 °C. Plates were then again washed three times with washing buffer and three times with distilled water. One hundred µl of B. salamandrivorans zoospore suspension (1 × 107 zoospores per ml) were added in each well. Plates were incubated for 20 min at 15 °C and washed five times with distilled water to remove the unbound zoospores. Digital photographs were taken through via an inverted light microscope at 100 × magnification. Five pictures were taken for each well and zoospores in each photograph were counted in a blind fashion. Three independent repeats of the experiment were conducted (biological replicates).Carbohydrate binding assayTo further determine which carbohydrates expressed on fire salamander sloughed skin can mediate the binding of B. salamandrivorans zoospores, B. salamandrivorans binding against four carbohydrates; N-acetylglucosamine (GlcNAc), N-acetylgalactosamine (GalNAc), mannose, and lactose was tested. The three monosaccharides and the disaccharide (Sigma-Aldrich) were dissolved and thereafter diluted in coating buffer to achieve a concentration of 5% (w/v). Then they were coated in triplicate wells by incubating at 4 °C for 24 h42. Plates were rinsed three times with washing buffer and blocked with 1% BSA overnight at 4 °C.Hundred μl of B. salamandrivorans zoospore suspension (1 × 107 zoospores per ml) was added in each well and incubated for 20 min at 15 °C. After washing the wells five times with distilled water to remove unbound zoospores, the plates were evaluated using a light microscope. Digital photographs were taken at 100 × magnification. Five pictures were taken for each well and zoospores in each photograph were counted in a blind fashion. Three independent repeats of the experiment were conducted (biological replicates).In this experiment the highest level of B. salamandrivorans spores binding to lactose was observed. Lactose is a dissacharide consisting of glucose and galactose. Therefore, in the following experiments galactose, glucose and their derivatives will be tested separately.Carbohydrate chemotaxis testChemotaxis of B. salamandrivorans toward free carbohydrates was tested as previously explained (Supplementary Fig. 4)12. The sugars D-Glucose (Sigma-Aldrich), D-mannose (Sigma-Aldrich), Lactose (Sigma-Aldrich), and D-galactose (Sigma-Aldrich) were tested as attractant for B. salamandrivorans. The monosaccharides instead of the amide derivatives were used in this experiment to exclude any chemotactic signalling activity of the amides. Sugars were dissolved in distilled water, filter sterilized, and tested at a 0.1 M concentration. Hematocrit capillaries (75 mm length; Hirschmann laborgeräte, Eberstadt, Germany) were filled with 60 µl carbohydrate solution, vehicle control capillaries with 60 µl sterile distilled water. To prevent leakage, the capillaries were sealed with wax plugs (Hirschmann laborgeräte, Eberstadt, Germany) at one side. Each capillary was swiped on the outside with lens paper (Kimtech Science, Kimberley Clark, Roswell, GA, USA) to remove possible attractant spillover. Capillaries were incubated in 400 µl inoculum containing 106 B. salamandrivorans zoospores in water and placed in a holder inclined about 65° upwards. The assay was incubated for 90 min at 15 °C, after which the capillaries were removed and swiped again at the outside to remove B. salamandrivorans zoospores possibly adhering on the outside. Inocula were checked for motility of the zoospores using an inverted microscope (Olympus CKX 41, Hamburg, Germany). Contents of the capillaries were collected and centrifuged for 2 min at 16.000 × g. The supernatant was removed as much as possible. The pellet was suspended in 100 µl Prepman Ultra Sample Preparation reagent (Applied Biosystems, Life Technologies Europe, Ghent, Belgium) and DNA was extracted according to the manufacturer’s guidelines. For each sample, the number of B. salamandrivorans zoospores was quantified using quantitative real-time PCR (qPCR)41, and data were analyzed using the Bio-Rad CFX manager 3.1. The primers and probe can be found in Supplementary Table 11. Within each assay, all carbohydrates and negative controls were tested at least in triplicate (technical replicates) and three independent repeats of the assay were performed (biological replicates).Carbohydrate transcriptome testRNA preparation: total RNA was isolated from B. salamandrivorans zoospores treated with different carbohydrates. Therefore, newly released zoospores (less than 2 h after induction of spore release by adding water) were harvested from 175 cm2 cell culture flasks by replacing the TGhL broth with distilled water, which was filtered using a sterile mesh filter with pore size 10 µm (Pluristrainer, PluriSelect). Six-biological replicates containing 4 × 107 zoospores were obtained. Each biological replicate consisted of a pool of spores harvested from three cell culture flasks. Per biological replicate, the spores were divided into 4 eppendorfs (107 zoospores/eppendorf) which were treated for 1 h at 15 °C with H2O (control), 50 mM (D-galactose), 50 mM (D-glucose), or 50 mM (D-mannose) (Supplementary Fig. 5). After 1 h, the zoospores were centrifuged for 5 min at 4.000 × g at 15 °C to remove the supernatant, after which RNA was extracted using the RNeasy mini kit (Qiagen)18. The RNA was treated with Turbo™ DNase (Ambion), following the manufacturer’s instructions. RNA degradation and contamination were monitored on 1% agarose gels. The RNA purity was checked using the NanoPhotometer® spectrophotometer (IMPLEN, CA, USA). Finally, the RNA integrity and quantitation were assessed using the RNA Nano 6000 assay kit of the Bioanalyzer 2100 system (Agilent Technologie, CA, USA).Library preparation for transcriptome sequencing: Whole-transcriptome sequencing libraries were constructed and sequenced on the Illumina HiSeq platform (Novogen, China). A total amount of 1 μg RNA per sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext® UltraTM RNA Library Prep Kit for Illumina® (NEB, USA) following the manufacturer’s recommendations and index codes were added to attribute sequences to each sample. Briefly, mRNA was purified from total RNA using poly-T oligo-attached magnetic beads. Fragmentation was carried out using divalent cations under elevated temperature in NEBNext First Strand Synthesis Reaction Buffer (5X). First-strand cDNA was synthesized using random hexamer primer and M-MuLV Reverse Transcriptase (RNase H-). Second strand cDNA synthesis was subsequently performed using DNA Polymerase I and RNase H. Remaining overhangs were converted into blunt ends via exonuclease/polymerase activities. After adenylation of 3′ ends of DNA fragments, NEBNext Adaptor with hairpin loop structure was ligated to prepare for hybridization. In order to select cDNA fragments of preferentially 150−200 bp in length, the library fragments were purified with AMPure XP system (Beckman Coulter, Beverly, USA). Then 3 μl USER Enzyme (NEB, USA) was used with size-selected, adaptor-ligated cDNA at 37 °C for 15 min followed by 5 min at 95 °C before PCR. Then PCR was performed with Phusion High-Fidelity DNA polymerase, Universal PCR primers, and Index (X) Primer. At last, PCR products were purified (AMPure XP system) and library quality was assessed on the Agilent Bioanalyzer 2100 system.Clustering and sequencing: The clustering of the index-coded samples was performed on a cBot Cluster Generation System using PE Cluster Kit cBot-HS (Illumina) according to the manufacturer’s instructions. After cluster generation, the library preparations were sequenced on an Illumina platform and paired-end reads were generated.Quality analysis, mapping, and assembly: Raw data (raw reads) of FASTQ format were first processed through fastp (version 0.20.0). In this step, clean data (clean reads) were obtained by removing reads containing adapter and poly-N sequences and reads with low quality from raw data. At the same time, Q20, Q30, and GC content of the clean data were calculated (Supplementary Table 12). All the downstream analyses were based on the clean data with high quality. Reference genome and gene model annotation files were downloaded from genome website browser (NCBI/UCSC/Ensembl) directly. Paired-end clean reads were mapped to the B. salamandrivorans reference genome using HISAT2 (version 2.0.5) software18. Featurecounts (version 1.5.0-p3) were used to count the read numbers mapped to each gene, including known and novel genes (Supplementary Table 13). And then RPKM (reads per kilobase per million) of each gene was calculated based on the length of the gene and reads count mapped to this gene.Gene expression, differential expression, enrichment, and coexpression- analysis: Differential expression analysis was performed using the DESeq2 R package43. The resulting P-values were adjusted using the Benjamini and Hochberg’s approach for controlling the false discovery rate (FDR). Genes with an adjusted P-value < 0.05 found by DESeq2 were assigned as differentially expressed. Protein domains were annotated with PFAM version 27 and 33 and KEGG domains, Gene Ontology (GO) enrichment analysis of differentially expressed genes was implemented by the clusterProfiler R package44 and dcGOR R package45. GO terms with corrected P-value less than 0.05 were considered significantly enriched by differential expressed genes. ClusterProfiler R package44 was also used to test the statistical enrichment of differentially expressed genes in KEGG pathways.Detection of protease activityThe influence of carboydrate exposure on protease activity of B. salamandrivorans zoospores was assessed. Therefore, zoospores were harvested from 175 cm2 cell culture flasks by replacing the TGhL broth with distilled water, which was filtered using a sterile mesh filter with pore size 10 µm (Pluristrainer, PluriSelect). A pool containing approximately 5 × 107 zoospores/ ml was obtained. 200 µl of the spore suspension (107 spores) was added to eppendorfs containing 200 µl H2O (H2O; n = 3), 200 µl 100 mM D-Glucose (Glc; n = 3), 200 µl 100 mM D-mannose (Man; n = 3), 200 µl 100 mM D-galactose (Gal; n = 3), or as a control, 200 µl H2O containing protease inhibitor mix (P8215, Sigma-Aldrich) (PI; n = 3). After 1.5 h at 15 °C, the zoospores were centrifuged for 5 min at 4.000 × g at 15 °C and the supernatant was collected. Protease activity in the supernatant was analyzed using the Pierce Fluorescent Protease Assay Kit (Thermo Fisher Scientific), according to the manufacturer’s instructions. Three independent repeats of the experiment were performed (biological replicates).Identification of B. salamandrivorans lectin genesPotential candidates of carbohydrate-binding molecules (CBMs) were identified in the B. salamandrivorans (AMFP) genome listed in the NCBI database (Bioproject PRJNA311566).B. salamandrivorans (AMFP 13/01) coding regions from the single annotated genome present on NCBI database (Bioproject PRJNA311566) were used to single out potential lectin genes of interest that could serve as genes of carbohydrate-binding proteins. The lectin candidates were identified with BLASTp (BLAST + 2.9.0) over the FungiDB database (constituting 199 candidates, database accessed 1st March 2018) using the stringent e-value cutoff of 1e−50 to avoid spurious hits46,47.From these, five candidates that referred to lectins and carbohydrate-binding were manually selected using the NCBI CDD (v3.16) conserved domain software with default settings48.Expression of two of these genes (BSLG_00833 and BSLG_02674) was confirmed by a previous mRNA expression analysis (Bioproject PRJNA311566)18.AnimalsThe animal experiments were performed following the European law and with the approval of the ethical committee of the Faculty of Veterinary Medicine (Ghent University EC) (EC2015/86). Only captive bred animals were used. Fire salamander larvae belonging to different life stages49 were used in a B. salamandrivorans infection trial.For lectin-histochemical staining, skin samples were collected from amphibian species Salamandra salamandra (n = 10), Ichthyosaura alpestris (n = 12), Lissotriton helveticus (n = 13), Pleurodeles waltl (n = 11), Lissotriton boscai (n = 3), Alytes obstetricans (n = 10), Cynops pyrrhogaster (n = 3), Triturus anatolicus (n = 3), Triturus marmoratus (n = 3), Calotriton asper (n = 10), Bombina variegata (n = 5), Rana temporaria (n = 10), Epidalea calamita (n = 5), Pelobates fuscus (n = 5) and Salamandra lanzai (n = 3). Tail or toe clips, ventral and dorsal skin samples were collected from animals that were euthanized with natrium pentobarbital 20% (KELA). The collected samples were immediately fixed in Bouin’s solution for 24 h.Mucosome samples were collected by bathing animals in HPLC-grade water for 1 h from 21 amphibian species (different animals as the ones used for the tissueclips), namely Lissotriton helveticus (n = 3), Pleurodeles waltl (n = 3), Lissotriton boscai (n = 3), Triturus anatolicus (n = 3), Triturus marmoratus (n = 3), Cynops pyrrhogaster (n = 3), Ichthyosaura alpestris (n = 3), Salamandra salamandra (n = 3), Lyciasalamandera helverseni (n = 3), Speleomantes strinatii (n = 2), Paramesotriton hongkongensis (n = 2), Plethodon glutinosus (n = 2), Chioglossa lusitanica (n = 3), Pachyhynobius shangchengensis (n = 3), Calotriton asper (n = 3), Salamandra algira (n = 3), Salamandra lanzai (n = 2), Alytes obstetricans (n = 3), Bombina variegata (n = 2), Epidalea calamita (n = 3) and Pelobates fuscus (n = 3).Exposure of fire salamander larvae and metamorphs to B. salamandrivorans Twenty-two early-stage and 26 late-stage larvae49,50 were inoculated with 1.5 × 105 B. salamandrivorans spores per ml water during 24 h. Ten days after the inoculation all the early-stage and sixteen late-stage larvae were euthanized. The two hind legs were analyzed by qPCR to detect the B. salamandrivorans GE load. A tail clip was stained with fluorescein-labelled RCA I (see below). Ten late-stage larvae were further kept until five weeks after metamorphosis.Six one-week-old fire salamander metamorphs were inoculated with 1 ml of water containing 1.5 × 105 spores for 24 h. The animals were euthanized 10 days after inoculation. The two hind legs were analyzed by qPCR to detect the B. salamandrivorans GE load. A tail clip was stained with fluorescein labelled RCA I (see below).Lectin-histochemical stainingFluorescein labelled RCA I (Ricinus communis agglutinin I) (Vector Laboratories) and Con A (Concanavalin A) has been used to detect the expression of galactose and mannose or glucose in the epidermis of amphibians38.After 24 h fixation in Bouin’s medium (Sigma-Aldrich), samples were washed first with tap water until the water ran colourless, then washed for 24 h in 70% ethanol saturated with lithium carbonate (Sigma-Aldrich) to remove picric acid. Tissues were then dehydrated in a graded ethanol series, cleared in xylene, embedded in paraffin, and sectioned in 4−6 µm slices. Before lectin staining, the sections were deparaffinized in xylene and hydrated in a series of ethyl alcohols. For better presenting the carbohydrate antigens, we performed antigen retrieval by submerging slides in citrate buffer (10 mM citric acid, pH 6.0) and heat treating in microwave (850 W for 3.5 min plus 450 W for 10 min). The slides were rinsed with PBS (0.01 M, pH 7.4) and immersed in 1% BSA (Sigma-Aldrich) for 15 min, to prevent non-specific lectin binding. Subsequently, the sections were incubated with either lectin RCA I (15 µg/ml) or lectin Con A (5 µg/ml) for 30 min. Lectins were diluted with lectin binding buffer (10 mM Hepes, 0.15 M NaCl, pH 7.5). As a negative control, lectin RCA I was mixed with 200 mM galactose, and lectin Con A was mixed with 200 mM mannose + 200 mM glucose, before incubating with skin sections to inhibit lectin binding. For positive control, a slide of fire salamander ventral skin sample for RCA staining, and midewife toad ventral skin sample for Con A staining, was included in each experiment. The slides were then washed in PBS, and cell nuclei were stained with 10 µg/ml Hoechst 33342 Solution (Invitrogen). Coverslips were mounted with ProlongTM Gold Antifade Reagent (Invitrogen). Staining results were observed using a Leica fluorescence microscope under 10× magnification, with a 450−490 nm BP excitation filter for lectin staining and a 355−425 nm BP excitation filter for Hoechst staining. Staining pictures were taken using Leica Application Suite (LAS) X software. The lectin staining intensities were classified as intense (3), strong (2), weak (1), or negative (0) staining (Supplementary Fig. 6). Experimental positive and negative controls were defined as intense (3) and negative (0) stained, respectively, and other slides were then evaluated in comparison to the set parameters. Hoechst staining results were paired with corresponding lectin staining results, making it easier to discern the tissue structure from the dark background. The fluorescent intensities were scored by three reviewers, respectively scoring the same dataset of pictures blinded three separate times, and the mean value was taken as the final result.Free galactose, mannose, and total carbohydrates in amphibian mucosomeMucus was collected from 21 amphibian species (see above). The animal body surface and volume of bathing water were calculated as follows: surface area of anuran species in cm2 = 9.9* (mass in g)^(0.56), surface area of urodelan species in cm2 = 8.42* (mass in g)^(0.694), and the quantity of HPLC-grade water to add to both anuran and urodelan species was determined by dividing the surface area by 4), and animals were bathed in respective amounts of HPLC-grade water for 1 h40,51. Animal washes were collected and concentrated by SpeedVac Vacuum Concentrators (Thermo Fisher Scientific) to 100 µl. The quantities of free galactose, mannose, and total carbohydrates in 100 µl of concentrated animal wash were measured using the Galactose Assay Kit (Abcam), Mannose ELISA Kit (Aviva Systems Biology), and Total Carbohydrates Assay Kit (Abcam), as per instructions. Concentrations of free galactose, mannose, and total carbohydrates in animal washes were divided by animal body surface to get the final results of sugar concentrations per square centimetre of the body surface.Statistical analysisStatistical analyses of fire salamander skin lysate binding assay, carbohydrate-binding assay, chemotaxis assay, and protease activity assay were performed using R version 4.0.3. To account for the experimental design, Generalized Linear Mixed Models (GLMM, R library lme452) were used, specifying a nested random effect whereby technical replicates are nested within biological replicates. Count data were modelled first using a Poisson distribution, but as significant overdispersion was present in the data, a negative binomial error structure was implemented. For the protease activity assay, data do not represent counts and a log transformation on the raw values were used to ensure normality of model residuals (Shapiro-Wilk W > 0.95) allowing a Gaussian error structure (i.e., a Linear Mixed Model (LMM). To test for differences between categories, the (G)LMMs were directly fed to the glht function of the R library multcomp53, setting up contrasts for Tukey’s all-pair comparisons, resulting in Bonferroni-corrected p-values adjusted for multiple testing. Statistical analyses of the larvae infection trial were performed in R version 4.0.0, with tidyverse54 version 1.3.0, MASS55 version 7.3-51.6, VGAM56 version 1.1-3, DHARMa57 version 0.3.1 and glmmTMB58 version 1.0.2.1. Infection loads of larvae and metamorphs were compared, using the Wilcoxon rank sum test, formula Chytrid GE load ~ larvae vs metamorph status, from the stats package. The correlation between larvae Ricinus communis agglutinin (RCA) scoring (1 = weak staining, 2 = strong staining, 3 = intense staining) and infection load was performed using the glm() function on log-transformed genomic equivalents with formula log10(B. salamandrivorans load in Genomic equivalents)~ RCA score, treating RCA score as an ordered factor with guassian distribution. As non-transformed chytrid loads showed zero-inflation and overdispersion, we also fit a generalized linear model with negative binomial distribution (GE load ~ RCA score) with RCA score as an ordered factor, using glmmTMB with a zero-inflation model (~ RCA score), which showed a comparable positive correlation between RCA score and GE load (conditional model coefficient = 5.67, p = 0.003, zero-inflation model coefficient = −2.21, p = 0.016). Residuals and chi-square test indicated the negative binomial model was not a significant improvement and so the simpler generalized linear model on transformed data was included. RCA scoring and larval stage prediction probabilities in Fig. 4c were generated by polr(RCA score ~ life stage) from MASS. Model fit and appropriateness was tested using Chisq test (p = 0.003), the model fit compared favourably to a more complex multinomial logit model and a model fit based on 70% of the data predicted 70−75% of remaining data (when data repeatedly sampled with different seeds, with the final model fit to all data).The regression and correlation analyses of different amphibian species were performed in SPSS (IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY, USA). Correlations of RCA scores with B. salamandrivorans infection peak loads, mortality rates, and percentage of free galactose were calculated by two-tailed Point-Biserial Correlation (p More
