Veterinary Research Communications https://doi.org/10.1007/s11259-022-09951-2 ORIGINAL ARTICLE Study of the variation of the Malassezia load in the interdigital fold of dogs with pododermatitis Leyna Díaz1 · Gemma Castellá1 · M. Rosa Bragulat1 · Andreu Paytuví‑Gallart2 · Walter Sanseverino2 · F. Javier Cabañes1 Received: 17 November 2021 / Accepted: 4 June 2022 © The Author(s) 2022 Abstract The yeast Malassezia pachydermatis is a common inhabitant of the skin and mucosae of dogs. However, under certain circumstances this yeast can overgrow and act as an opportunistic pathogen causing otitis and dermatitis in dogs. Canine pododermatitis is a common disorder in dogs in which M. pachydermatis acts as an opportunistic pathogen. In the present study, the presence of Malassezia yeasts was assessed and quantified in samples collected from the interdigital space of dogs with pododermatitis before and after treatment, and from healthy dogs. The samples were subjected to two different cytological examinations, culture on Sabouraud glucose agar and modified Dixon’s agar and a quantitative PCR targeting the internal transcribed spacer (ITS) genomic region. A selection of samples was analyzed by next generation sequencing (NGS) using the D1D2 domain of the large subunit of the ribosomal DNA as target. The pododermatitis samples before treatment showed higher cell counts, colony-forming units and ITS copies than the rest of samples. The NGS analysis revealed that Ascomycota was the main phylum in the healthy and post-treatment samples. However, Basidiomycota and M. pachyder- matis was more abundant in the pododermatitis samples before treatment. These results support M. pachydermatis as an opportunistic agent in canine pododermatitis by a variety of methods, and demonstrate the correlation between cytologic and molecular methods for quantification. Keywords Interdigital fold · Malassezia · M. pachydermatis · Mycobiome · Pododermatitis Introduction Canine pododermatitis is a common disorder in the gen- eral veterinary practice. Malassezia pachydermatis com- The yeast Malassezia pachydermatis is part of the normal monly acts as an opportunistic pathogen that produces podo- microbiota of the skin and mucosae of healthy dogs (Guillot dermatitis in dogs, especially in atopic animals or animals and Bond 1999; Cabañes 2021). Despite being part of the diagnosed with endocrine disease. It may affect the inter- normal microbiota, under certain circumstances, the popula- digital spaces, nail folds, and nails, and usually more than tion of M. pachydermatis can overgrow and this yeast may one paw is affected (Duclos 2013; Bajwa 2016). In dogs with act as an opportunistic pathogen. Thus M. pachydermatis Malassezia pododermatitis, pruritus is a major and constant is considered one of the most frequent aetiological agents sign often associated with erythema and/or greasy exudate. responsible for otitis and dermatitis in dogs (Bond et al. In chronic cases hyperpigmentation, alopecia and lichenifi- 2020). cation may be present (Miller et al. 2000). Its diagnosis is based on history, physical examination, clinical signs, cytological examination, and response to anti- fungal therapy (Miller et al. 2000; Bond et al. 2010). Detec- * Gemma Castellá tion of Malassezia yeasts by cytology in compatible skin Gemma.Castella@uab.cat lesions is a key for the diagnosis of Malassezia pododerma- titis (Duclos 2013). Cultural techniques are primarily used 1 Veterinary Mycology Group, Department of Animal in research rather than routine veterinary practice. However, Health and Anatomy, Universitat Autònoma de Barcelona, Bellaterra, Catalonia, Spain they have been suggested to have a diagnostic value because 2 Sequentia Biotech SL, Barcelona, Catalonia, Spain 13 Vol.:(0123456789) Veterinary Research Communications they have higher sensitivity than cytology (Bond et al. 1994, Two samples were taken from each interdigital fold by 2010). simultaneously rotating fully two sterile swabs (Deltalab, To overcome limitations of culture-based techniques, Spain) for 30 s. Also, a cell tape (Scotch, Spain) for cyto- several molecular methods have been developed. To detect logical examination was obtained from each interdigital and quantify Malassezia yeast in samples, quantitative PCR fold. Tape strip was applied to the skin and then stained (qPCR) is a fast, sensitive, and precise technique (Sugita with a Diff-Quick staining (Microptic, Spain) of which the et al. 2006). The internal transcribed spacer (ITS) region first fixative step with methanol was omitted. One swab was and β-tubulin gene have been used as target to study the used for Malassezia culture and cytological examination. population of Malassezia in samples from dogs by qPCR This swab was dipped for 10 s into an Eppendorf (Deltalab, (Puig et al. 2019; Meason-Smith et al. 2020). Spain) containing 500 µL of a solution of distilled water and On the other hand, next generation sequencing (NGS) is 0.04% Tween 80 (ICN Biomedical, USA) to reduce clump a high sensitivity tool to investigate the microbial diversity formation. This suspension was used to perform cytological within samples at a lower cost without the need of a culture examinations and culture. The other swab was maintained at (Forde and O’Toole 2013; Forbes et al. 2017). Although -20 ºC and used for qPCR and NGS. only few studies have been conducted using NGS in dogs, Tang et al. (2020) highlighted the adequacy of NGS-based Culture and cytological examination methods as a diagnostic tool to diagnose canine skin and ear infections. Aliquots of 0.1 mL of the suspension were inoculated on The aim of this work was to study the presence of M. Sabouraud glucose agar (SGA; Oxoid, Spain) and modi- pachydermatis from samples of dogs with pododermatitis, fied Dixon agar (mDA) (Guého et al. 1996) plates using healthy dogs and dogs after treatment using different tech- the surface-spread method. Both media were supplemented niques including cytologic examination, culture, qPCR and with 0.05% of chloramphenicol (Sigma, Spain) and 0.05% NGS, and to determine if there is a correlation between com- of cycloheximide (Sigma, Spain) (Guého-Kellerman et al. mon cytological techniques used in veterinary practice and 2010). The plates were incubated at 35 ºC to a maximum of molecular methods. Besides, using NGS the mycobiome was 7 days and colony forming units (CFU) were counted. compared among a selection of the three different kind of A first cytological examination was performed using the samples studied. tape strip. A second cytological examination was performed from the suspension. An aliquot of 10 µL of the suspension was heat-fixed on a slide and stained with the Diff-Quick stain. The presence of M. pachydermatis cells was quantified Materials and methods by counting cells in 10 different random fields at 100 × with immersion oil. Animals and sample collection DNA extraction A total of 13 dogs including five females and seven males were enrolled in this study (Table 1). Eleven DNA was extracted from swabs using the DNeasy PowerSoil dogs were included in the pododermatitis group. All Kit (Qiagen, Germany) according to manufacturer’s instruc- were diagnosed with pruritus in at least one extremity, tions with some modifications previously described (Díaz erythema in the affected interdigital fold, and a positive et al. 2021). To control cross-contamination, a sterile swab cytology for the presence of Malassezia yeasts. A total was processed under the same conditions. The DNA was of 15 samples were collected from this group. Three stored at -20 ºC until used as template for qPCR and NGS. dogs initially diagnosed with Malassezia pododermatitis were also sampled after treatment with antifungal Malassezia pachydermatis quantification by qPCR agents (Table 1). These dogs were included in the post- treatment group and four samples were obtained. Two All samples were selected for M. pachydermatis quantifica- dogs without erythema, pruritus and skin lesions were tion. All qPCRs were performed using an Applied Biosys- assigned to the healthy group, obtaining three samples. tems 7500 real-time system, and SYBR Green chemistry A total of 22 samples were obtained following the (PowerUp SYBR Green master mix, Applied Biosystems, procedures on Animal and Human Experimentation USA) with the absolute quantification method. from UAB and Generalitat de Catalunya approved by The multicopy ITS region was used as target gene accord- Ethics Committee (Approval number, 4600; Approval ing to Meason-Smith et al. (2020) with some modifications. date, 22Feb2019). Also, the written consent of owners Primers ITS-ANA-F and PachyR (Vuran et al. 2014) were was obtained before sampling the animals. used at a final concentration of 900 nM each. The runs were 13 Veterinary Research Communications Table 1  Samples included in the study Sample Dog Breed Sex Origin Clinical signs Treatment Samples from dogs with pododermatitis 1 Dog 1 Labrador Retriever FN Left thoracic limb Pruritus, erythema, alopecia, and _ surface debris 2 Dog 2 Yorkshire F Left thoracic limb Pruritus, erythema, alopecia, and _ surface debris 3 Dog 3 Labrador Retriever ND Right pelvic limb Pruritus, erythema, and greasy _ exudate 4 Dog 4 Labrador Retriever M Right pelvic limb Pruritus, erythema, and alopecia _ 5 Dog 5 Bernese Mountain Dog M Left thoracic limb Pruritus, erythema, alopecia, and _ a small nodule 6 Dog 5 Bernese Mountain Dog M Right thoracic limb Pruritus, erythema, and alopecia _ 7 Dog 6 Mixed breed M Right thoracic limb Pruritus, erythema, and greasy _ exudate 8 Dog 7 American Staffordshire FN Right pelvic limb Pruritus, erythema, and alopecia _ 9 Dog 8 Mixed breed MN Left thoracic limb Pruritus and erythema _ 10 Dog 8 Mixed breed MN Right thoracic limb Pruritus and erythema _ 11 Dog 9 West highland white terrier M Left thoracic limb Pruritus, erythema, greasy exu- _ date, and surface debris 12 Dog 9 West highland white terrier M Right thoracic limb Pruritus, erythema, greasy exu- _ date, and surface debris 13 Dog 10 Labrador Retriever M Left thoracic limb Pruritus, erythema, alopecia, and _ surface debris 17 Dog 13 Labrador Retriever FN Right thoracic limb Pruritus and erythema _ 18 Dog 13 Labrador Retriever FN Left thoracic limb Pruritus and erythema _ Samples from healthy dogs 14 Dog 11 Labrador Retriever MN Right pelvic limb None _ 15 Dog 11 Labrador Retriever MN Left pelvic limb None _ 16 Dog 12 Labrador Retriever FN Right pelvic limb None _ Samples after antifungal treatment 19 Dog 3 Labrador retriever ND Right pelvic limb None Topical clotrimazole twice daily, 4 weeks 20 Dog 7 American Staffordshire FN Right pelvic limb None Topical miconazole, 3 weeks 21 Dog 8 Mixed breed MN Left thoracic limb None Oral itraconazole once daily,3 weeks 22 Dog 8 Mixed breed MN Right thoracic limb None Oral itraconazole once daily, 3 weeks a In bold, samples analysed by NGS. F = female, M = male, FN = neutered female, MN = neutered male, ND = no data available performed following these thermal conditions: 1 cycle of the length of the template. All samples were run in duplicate, 50 ºC for 2 min, 1 cycle of 98 ºC for 3 min, followed by 40 including the standards and negative control. The negative cycles of 95 ºC for 30 s, and 60 ºC for 1 min. Finally, a melt- control contained all the elements of the reaction mixture ing curve analysis with a gradual increase of temperature and water instead of DNA template spiked. from 60 ºC to 95 ºC was performed. To achieve absolute quantification of the samples, the standard curve method NGS and data analysis was used. A standard curve of genomic DNA extracted from the neotype strain of M. pachydermatis CBS1879 was Due to the high technical cost of the analysis, five sam- constructed, including 7 ten-fold dilutions, from 18 ng to ples were selected for the metagenomics NGS analysis of 0.018 pg of DNA. The quantity and purity of the gDNA the fungal 26S rRNA gene. These samples were selected were determined by spectrometry (NanoDrop 2000; Thermo based on dog’s breed and the results of the cytological Scientific, Spain). The number of copies of the template was examination and culture. The samples belonged to four calculated considering the amount of a template present and different Labrador Retriever dogs. One sample from the 13 Veterinary Research Communications healthy group (sample 16) with negative cytology and no out using Bray–Curtis dissimilarities; cluster significance growth in the culture media used was included. Three dif- was assessed with PERMANOVA and ANOSIM (R pack- ferent samples from pododermatitis group (samples 1, 3 age vegan). and 4) with a positive cytological examination and growth on both culture media used were selected. Also, a sample Statistical analysis from the post-treatment group (sample 19) was included. Quality control was performed at IGA Technology. Other statistical analyses were conducted by Minitab 17 sta- DNA concentration was evaluated by using a Qubit 2.0 tistical software (Minitab). The Ryan-Joiner normality test Fluorometer (Invitrogen, USA). Amplicon-seq librar- was applied to determine whether data followed a normal ies of D1/D2 regions of the fungal 26S rRNA gene were distribution. Differences in both cytologic examinations, obtained from each sample by following 16S Metagen- CFU in SGA and mDA, and qPCR quantification values omic Sequencing Library Preparation protocol with minor between samples were tested by Kruskal–Wallis test. Stu- modifications. Briefly, the forward primer (NL1) and the dent’s t test was applied in values of CFU in both culture reverse primer (NL4) (Reynolds and Taylor 1993) contain- media. A multiple regression analysis was used to study the ing Illumina overhang sequences necessary for the compat- relationship between the qPCR quantification values and val- ibility with Illumina index and sequencing adapters were ues obtained in the rest of variables. The significance level used for the first PCR amplification under the following for all statistical analyses was set at P ≤ 0.05. conditions: 95 °C for 3 min; 28 cycles of: 95 °C for 30 s, 55 °C for 30 s, 72 °C for 30 s; 72 °C for 5 min; hold at 4 °C. Upon the clean-up, the second PCR was performed Results under the following conditions: 95 °C for 3 min; 9 cycles of: 95 °C for 30 s, 55 °C for 30 s, 72 °C for 30 s; 72 °C for Culture and cytological examination 5 min; hold at 4 °C. Relevant flow-cell binding domains and unique indices (NexteraXT Index Kit, Illumina, USA) The results of the two cytological examinations performed were integrated to the amplicon target. Libraries were nor- (tape strip and heat-fixed slide) and the results obtained malized by Qubit 2.0 Fluorometer, pooled and sequenced in both SGA and mDA culture media are summarized in on MiSeq using paired 300‐bp reads, and MiSeq v3 Table 2. reagents (Illumina). Sequence reads were analyzed in The 15 samples from the pododermatitis group had a the cloud bioinformatics platform GAIA (https://​m etag​ positive cytological evaluation for Malassezia yeast cells, enomi​cs.​seque​ntiab​iotech.​com) (Giampaoli et al. 2020). with a range of 0.1–45.9 cells/field (mean = 7.37) on the tape Sequencing quality was assessed using FastQC (bioinfor- strip. All samples had growth on SGA plates ranging from 1 matics.babraham.ac.uk/projects/fastqc/) and BBduk (jgi. to 468 CFU/plate (mean = 83.8) (Tables 2 and 3). doe.gov/data-and-tools/bbtools/), setting a minim length The three samples from the healthy group had a nega- of 35 bp and a minimum Phred-quality score of 25. tive cytological evaluation for the presence of Malassezia The resulting high-quality reads were mapped against a yeasts and no growth was observed in any of the culture custom-made cured database of 26S fungal sequences from media used. NCBI (Díaz et al. 2021). For the taxonomic classification, In the post-treatment group, no growth of M. pachyder- the mapping-based approach against the database with the matis was obtained. All samples had a negative cytological BWA mapper (Li and Durbin 2010) was followed by an in evaluation except for sample 19 which showed a low cell house Lowest Common Ancestor (LCA) algorithm. Follow- count in the first cytological examination. ing the limits proposed by GAIA software, the minimum Although in almost all samples the SGA medium gave identity thresholds applied to classify the reads into dif- higher CFU counts than mDA, no significant differences ferent taxonomic levels were species (99%), genus (98%), (P = 0.382) were obtained. Statistically significant differ- family (96%), order (94%), class (92%), and phylum (90%). ences in tape-strip, heat-fixed slide, plate counts in SGA and Those taxa with abundance below 0.01% considering its mDA media values were found between samples from dogs mean across the different samples within an experimen- with pododermatitis signs and healthy and post-treatment tal group were filtered out for further analysis. DESeq2 dogs (Table 3). (v1.26) (Love et al. 2014) was used to carry out differen- tial abundance analyses. GAIA also assesses the diversity Quantification of M. pachydermatis by qPCR within (alpha-diversity) and between (beta-diversity) sam- ples. Alpha and beta diversities were calculated using the The standard curve generated was linear over 7 ten-fold dilu- R package phyloseq (McMurdie and Holmes 2013). A tions of M. pachydermatis DNA, from 1.2 × ­1012 to 1.2 × ­105 Principal Coordinates Analysis (PCoA) was also carried copies. The qPCR amplifications of each standard showed 13 Veterinary Research Communications Table 2  Results obtained in Sample Dog Cytological examination (cells/ Culture CFU qPCR samples, including the two field) cytological examinations performed presented in counted Tape strip Heat-fixed SGA mDA ITS nº of copies cells per field, the colony slide forming units (CFU) per plate in both Sabouraud’s glucose Samples from dogs with pododermatitis agar (SGA) and modified 1 Dog 1 1.3 1.1 79 66 2.6 × ­107 Dixon agar (mDA), and qPCR 2 Dog 2 0.2 0 1 0 1.4 × ­106 quantification values 3 Dog 3 12.6 0.4 218 1 1.2 × ­107 4 Dog 4 1.4 0.5 76 14 7.5 × ­106 5 Dog 5 6.2 2.2 2 9 2.1 × ­108 6 Dog 5 11.4 0.6 120 311 8.1 × ­107 7 Dog 6 8.4 0.2 4 0 1.7 × ­107 8 Dog 7 45.9 0.7 468 153 2.7 × ­108 9 Dog 8 6.8 0.3 172 91 1.3 × ­106 10 Dog 8 1.3 0.3 10 17 1.6 × ­106 11 Dog 9 4.7 0.2 38 15 1.6 × ­108 12 Dog 9 2.3 0 3 1 1.9 × ­107 13 Dog 10 4.7 1.3 59 24 2.8 × ­108 17 Dog 13 3.2 0 2 1 2.2 × ­106 18 Dog 13 0.1 0 5 1 6.5 × ­106 Samples from healthy dogs 14 Dog 11 0 0 0 0 4.9 × ­106 15 Dog 11 0 0 0 0 1.1 × ­105 16 Dog 12 0 0 0 0 1.2 × ­106 Samples after antifungal treatment 19 Dog 3 0.2 0 0 0 1.3 × ­107 20 Dog 7 0 0 0 0 3.7 × ­105 21 Dog 8 0 0 0 0 6.5 × ­105 22 Dog 8 0 0 0 0 1.7 × ­105 a In bold, samples analysed by NGS Table 3  Mean values, standard deviation (SD), and range of values quantification values between samples from dogs with sings of podo- for plate counts (CFU) in Sabouraud’s glucose agar (SGA) and modi- dermatitis and samples from healthy and post-treatment dogs fied Dixon agar (mDA), cytologic examination (cells/field) and qPCR Dogs with pododermatitis (n = 15) Healthy dogs (n = 3) Dogs after treatment (n = 4) P value mean value ± SD range mean value ± SD range mean value ± SD range CFU in SGA 83.8 ± 126.0 1.0–468.0 0±0 0–0 0±0 0–0 0.001 CFU in mDA 46.9 ± 85.0 0–311.0 0±0 0–0 0±0 0–0 0.004 Tape-strip 7.37 ± 11.34 0.1–45.9 0±0 0–0 0.05 ± 0.10 0–0.2 0.001 Heat-fixed slide 0.52 ± 0.61 0–2.2 0±0 0–0 0±0 0–0 0.015 qPCR 7.2 × ­107 ± 1.0 × ­108 1.3 × ­106–2.8108 2.0 × ­106 ± 2.5 × ­106 1.1 × ­105–4.9 × ­106 3.6 × ­106 ± 6.4 × ­106 1.7 × ­105–1.3 × ­107 0.014 amplification plots corresponding to mean Cq values of obtained from the healthy group. Samples from the post- 14.36 to 34.33. The standard curve yielded r2 values > 0.99, treatment group showed quantifications values of 1.7 × ­105 to and slope values of − 3.44. Amplification was obtained in 1.3 × ­107 copies. Statistically significant differences in qPCR all samples with a Cq value lower than 35. The Cq values of values were found between samples from the pododermati- samples are detailed in Supplementary Table 1. tis group and samples from the healthy and after treatment As shown in Table 2, in the pododermatitis group, quanti- groups (Table 3). fication values of 1.3 × ­106 to 2.8 × ­108 copies were obtained Multiple regression analysis showed that the relation- whereas values from 1.1 × ­105 to 4.9 × ­106 copies were ship (r2 = 70.7%) between qPCR quantification and both 13 Veterinary Research Communications cytological examinations in the model was statistically sig- Although in the PCoA plots, the samples clustered in two nificant (P < 0.001). different groups, one including the healthy and post-treat- ment samples and one including all three pododermatitis NGS data analysis samples, there was no significant difference between the two different categories based on PERMANOVA ( F = 18.9, The five samples were correctly sequenced, and the gener- P = 0.1) and ANOSIM (R = 1, P = 0.1) tests probably due to ated fastq files reported an average value of 62,178 LSU the small sample size. reads passing filter. The number of generated sequences The taxonomic composition of the samples was investi- reads of each sample is described in Table 4. The raw gated at various taxonomic levels. Within the pododerma- sequencing data is available at the NCBI database (SRA titis group, Basidiomycota with a median of abundance of accession number: PRJNA742914). 22.18% was the main phylum (Fig. 1). At the level of class The Shannon diversity index was selected to measure (Fig. 2), Malasseziomycetes showed the highest abundance alpha diversity. The sample of the healthy dog was the most (16.57%) being Malasseziales (16.33%) the main order diverse with an index of 3.32 (Table 4). Samples from the (Fig. 3). With a median of abundance of 15.94%, Malassezi- pododermatitis group showed less diversity, on average, than aceae was the main family (Fig. 4). Malassezia was the main the healthy and post-treatment samples. genus with a median of abundance of 14.26% (Fig. 5) and, The Bray Curtis index was used to measure beta diversity. M. pachydermatis was the predominant species (10.64%). To Samples from the healthy and post-treatment group with an a lesser extent, other Malassezia species were identified in index of 0.323 were more similar compared to the podo- this group of samples like M. restricta (0.04%), M. globosa dermatitis samples (average of 0.716 between the healthy (0.02%), and M. sympodialis (0.01%). An average of 6.29% sample and average of 0.496 between the post-treatment of the Malassezia sequences could not be identified to the sample). Also, pododermatitis samples were more similar species level (Fig. 6). This percentage includes sequences among them with an index ranging from 0.112 to 0.196. marked as unknown (6.04%) because a match with enough Table 4  NGS reads after filter Sample Number reads after % Reads classified Shannon species Number of and biodiversity data obtained quality processing to genus diversity species identi- from metagenomics analysis fied 16 (healthy) 52,488 52.19 3.32 220 1 (pododermatitis) 73,732 26.15 2.37 247 3 (pododermatitis) 60,227 21.58 1.46 201 4 (pododermatitis) 62,178 26.23 2.35 219 19 (post-treatment) 106,095 45.07 2.48 248 Fig. 1  Relative abundance of fungal phyla across different interdigital fold samples 13 Veterinary Research Communications Fig. 2  Relative abundance of fungal classes across the differ- ent interdigital fold samples Fig. 3  Relative abundance of fungal orders across the differ- ent interdigital fold samples coverage and identity was not found in our database. Also, Within the healthy and post-treatment samples, Asco- the percentage includes ambiguous sequences (0.21%) mycota was the predominant phylum with an abundance because they matched with two different sequences from the of 72.42% and 56.97% respectively (Fig. 1), and Doth- database but could not discriminate between them. Malasse- ideomycetes was the main class (Fig. 2). Capnodiales zia pachydermatis and M. sympodialis showed a higher with an abundance of 44.54% and 34.64% respectively, abundance in the pododermatitis group (P < 0.05) while no was the predominant order (Fig. 3). At the family level differences were observed in the abundance of M. restricta (Fig. 4), Cladosporiaceae showed the highest abundance (P = 0.024), and M. globosa (P = 0.56) between both groups. 13 Veterinary Research Communications Fig. 4  Relative abundance of fungal families across the differ- ent interdigital fold samples Fig. 5  Relative abundance of fungal genera across the differ- ent interdigital fold samples in both samples (36.02 and 33.81% respectively) and Discussion Cladosporium was the main genus with an abundance of 32.40% and 32.15% respectively (Fig. 5). Malasse- Samples of the pododermatitis group showed the highest zia pachydermatis (Fig. 6) was also detected but with number of yeast cells in the two cytological examinations low abundance (0.06% and 0.03% respectively). Other and the highest number of CFU in the plates. Whereas Malassezia species were identified in these samples. samples from the healthy and after treatment groups Malassezia restricta (0.02%) was identified in both sam- showed much less counted cells in both cytological exami- ples, and M. globosa (0.04%) was only present in the post- nations and no growth was observed in the plates. Clinical treatment sample. The percentage of Malassezia sequences signs are associated to a hypersensitivity reaction against that could not be identified to the species level in healthy the yeast. The proliferation of Malassezia within the stra- samples decreased to 0.03% whereas in post-treatment tum corneum produces numerous antigens and allergens samples was 0.08%. that activate an immune response resulting in epidermal 13 Veterinary Research Communications Fig. 6  Relative abundance of M. pachydermatis and unidentified taxa across the different samples. (A) Healthy and post-treatment samples. (B) Pododermatitis samples damage and inflammation (Bond et al. 2020). Thus, low after effective treatment. Due to the higher sensitivity of the yeast counts can produce severe clinical sings and any qPCR method, DNA of M. pachydermatis was detected in number of the yeast may be considered significant (Bond all samples. As cytological examination is considered a use- et al. 2020). However, in the veterinary practice more than ful technique for diagnosing Malassezia pododermatitis, the 3–5 cells or more per high power field in the skin or a results of both cytological examinations were compared to number higher than 70 CFU could be indicative of infec- those obtained by qPCR. A good correlation was observed tion by Malassezia spp. (Cafarchia et al. 2005b; Bajwa between both diagnostic methods. These results showed that 2017). this qPCR method could be a fast useful technique to reliably All samples with a positive cytological examination had detect Malassezia pachydermatis pododermatitis in dogs. growth, except the sample 19 that showed a low cell count The fungal diversity studied using NGS was reduced in on the tape strip but had no growth. All the samples with a samples of the pododermatitis group compared with the negative cytology showed no growth in any of the culture healthy and post-treatment samples. These results are in media used. Thus, cytological examination is a useful tech- accordance with other studies where a reduction of fungal nique for diagnosing Malassezia pododermatitis (Cafarchia species diversity and richness was observed in samples from et al. 2005a; Duclos 2013). Modified Dixon’s agar has been dogs with otitis when compared to healthy dogs (Korbelik suggested as the preferred media to culture M. pachyderma- et al. 2018) and after treatment (Puigdemont et al. 2021). tis due to the rapid growth of the colonies and its potential Also, atopic dogs usually present lower diversity in their to support the lipid-dependent isolates that exist (Puig et al. microbiome when compared to healthy animals (Meason- 2017; Bond et al. 2020). However, in our study no significant Smith et al. 2015). differences were obtained in terms of CFU between SGA The results obtained in the metagenomic analysis were and mDA. different between the pododermatitis samples and the In our study, two cytological examinations were done healthy and post-treatment samples. Basidiomycota was the using completely different techniques. In all cases, the tape main phylum identified on pododermatitis samples while strip method showed the highest abundance of cells, in Ascomycota was the main one on healthy and post-treatment agreement with other studies (Bond et al. 2020). A study on samples. Within the pododermatitis group, Malassezia was dogs with Malassezia pododermatitis showed that the num- the majority genus identified in all three samples, and M. ber of yeasts recovered with swabs were significantly lower pachydermatis the main species. The genus Malassezia was than those obtained by using other techniques as tape strip- also present in the healthy and post-treatment samples in ping, superficial scraping, and direct impressions (Miller a low abundance as it is a common member of the skin of et al. 2000). dogs. Within the healthy and post-treatment samples, Clad- Although a low number of samples has been analyzed, osporium was the main genus followed by Alternaria. These ITS copies of M. pachydermatis were increased on most of genera are considered transient mycobiota as they are com- the samples of the pododermatitis group (73.3%) compared monly isolated from environmental samples. to that of healthy and post-treatment groups. Besides, the Our results agree with those obtained in studies of the ear qPCR values dramatically decreased in samples from dogs and skin mycobiota of healthy dogs and dogs diagnosed with 13 Veterinary Research Communications otitis or skin pathologies. In all these studies the main phylum results obtained in this study. Also, further studies includ- identified in healthy dogs was Ascomycota and the main genus ing more breeds would be of interest to study the genetic was part of the transient mycobiota. The second main genus predisposition of some breeds to Malassezia pododermatitis. identified in healthy dogs was Malassezia. In dogs diagnosed The NGS analysis allowed a better understanding of with otitis or dermatitis, Basidiomycota was the main phy- dynamics of M. pachydemartis pododermatitis. There is a lum and Malassezia the main genus identified. Malassezia difference in the results obtained in terms of fungal spe- pachydermatis was the main species identified in dogs with cies identified and richness of fungal populations between otitis and atopic dermatitis (Korbelik et al. 2018; Bradley et al. healthy and post-treatment samples and pododermatitis sam- 2020; Puigdemont et al. 2021). A study conducted by Tang ples. In the pododermatitis group Malassezia is predomi- et al. (2020) identified M. pachydermatis as the only yeast nant and in the healthy and post-treatment samples transient species present in the samples of both healthy dogs and dogs mycobiota predominates. Also, statistically significant dif- with otitis/dermatitis. These results differ from those obtained ferences were observed between dogs with pododermati- by Meason-Smith et al. (2020) where M. restricta was the tis before and after treatment and healthy dogs in values main species detected in healthy and allergic canine skin. of cytological examination, CFU and qPCR. The results However, in this paper only molecular techniques were used, obtained showed a good correlation between cytologic and so this species was not isolated from those animals. molecular methods and its value as a diagnostic tool. Focusing only on Malassezia, M. pachydermatis was the main species detected in all the samples, but M. restricta, Supplementary Information The online version contains supplemen- tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 11259-0​ 22-0​ 9951-2. M. globosa, M. sympodialis were also detected. Malasse- zia restricta and M. globosa have never been isolated from Acknowledgements The authors thanks Mar Bardagí for the samples dogs (Bond et al. 2010). Although M. sympodialis has been kindly provided for our work. We also thank Carolina Gómez from the previously cited from dogs with dermatitis and/or otitis, its Veterinary Mycology Group of Universitat Autònoma de Barcelona identification was based solely on phenotypic characteris- (UAB) for valuable technical assistance. tics (Raabe et al. 1998; Nardoni et al. 2004; Cafarchia et al. Authors’ contributions Gemma Castellá, M. Rosa Bragulat, and F. 2005a). Thus, the presence of small amounts of these three Javier Cabañes contributed to the study conception and design. Mate- species in our samples could be a result of cross-contami- rial preparation, data collection and analysis were performed by Leyna nation. These species are commonly found on the skin of Díaz, Gemma Castellá, M. Rosa Bragulat, and F. Javier Cabañes. Leyna healthy humans (Sugita et al. 2010) and thus, the contami- Díaz, Gemma Castellá, Walter Sanseverino, and Andreu Paytuví-Gal- lart performed bioinformatics analyses. The first draft of the manuscript nation may have occurred during the sample extraction or was written by Leyna Díaz and all authors commented and contrib- contamination of the kit reagents, or they may be on the uted to the preparation of the final manuscript. All authors read and dog’s skin as a result from interaction with humans or with approved the final manuscript. the indoor environment (Frau et al. 2019; Díaz et al. 2021). Metagenomic studies on dog skin selected the ITS region Funding Open Access Funding provided by Universitat Autònoma de Barcelona. Financial support came from Servei Veterinari de Bacte- as target gene. Although the ITS region is considered the riologia i Micologia from the UAB. universal barcode for fungi (Schoch et al. 2012), its variabil- ity among fungal species could lead to an incorrect estima- Data availability The raw sequencing data is available at the NCBI tion of fungal populations (Scorzetti et al. 2002; De Filippis database (SRA accession number: PRJNA742914). et al. 2017). The LSU selected in this study as target gene has shown greater biodiversity in biological samples and Declarations works better for species discrimination in Basidiomycota and yeasts, especially within Malassezia genus (Hoggard et al. Ethics approval All samples were obtained following the procedures on Animal and Human Experimentation from UAB and Generalitat de 2018; Mota-Gutiérrez et al. 2019). The use of this region Catalunya approved by Ethics Committee (Approval number, 4600; allowed us the description of fungal diversity and the taxo- Approval date, 22Feb2019). Also, the written consent of owners was nomic classification of several Malassezia species. There obtained before sampling the animals. was a percentage of unidentified Malassezia species within Consent to participate Written informed consent was obtained from the pododermatitis (6.29%) and the healthy/post-treatment the dog-owners. (0.05%) group. However, all the sequences belonged to the class Malasseziomycetes and could represent new taxa that Consent for publication Dog-owners signed informed consent regard- have not been yet described. ing publishing their dogs’ data. The major limitation of this study was the sample size. Due to the high cost of the method, a total of five samples Conflict of interest Walter Sanseverino and Andreu Paytuví-Gallart were employed by company Sequentia Biotech S.L. All other authors were selected for metagenomic analysis. Future studies with declare no competing interests. a larger sample size would be needed to clarify some of the 13 Veterinary Research Communications Open Access This article is licensed under a Creative Commons Attri- extraction and amplicon production strategies deeply inf luence bution 4.0 International License, which permits use, sharing, adapta- the outcome of gut mycobiome studies. Sci Rep 9:9328. https://​ tion, distribution and reproduction in any medium or format, as long doi.​org/​10.​1038/​s41598-​019-​44974-x as you give appropriate credit to the original author(s) and the source, Giampaoli S, De Vittori E, Frajese GV, Paytuví A, Sanseverino W, provide a link to the Creative Commons licence, and indicate if changes Anselmo A, Barni F, Berti A (2020) A semi-automated protocol were made. The images or other third party material in this article are for NGS metabarcoding and fungal analysis in forensic. Foren- included in the article's Creative Commons licence, unless indicated sic Sci Int 306:110052. https://​doi.​org/​10.​1016/j.​forsc​iint.​2019.​ otherwise in a credit line to the material. If material is not included in 110052 the article's Creative Commons licence and your intended use is not Guého E, Midgley G, Guillot J (1996) The genus Malassezia with permitted by statutory regulation or exceeds the permitted use, you will description of four new species. Antonie Van Leeuw 69:337–355. need to obtain permission directly from the copyright holder. To view a https://​doi.​org/​10.​1007/​BF003​99623 copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. Guého-Kellerman E, Boekhout T, Begerow D (2010) Biodiversity, phylogeny and ultrastructure. Malassezia and The Skin, 1st edn. Springer, Berlin, pp 17–63 Guillot J, Bond R (1999) Malassezia pachydermatis: a review. Med Mycol 37:295–306. https://​doi.​org/​10.​1046/j.​1365-​280x.​1999.​ References 00237.x Hoggard M, Vesty A, Wong G, Montgomery JM, Fourie C, Douglas Bajwa J (2016) Canine pododermatitis. Can Vet J 57:991–993 RG, Biswas K, Taylor MW (2018) Characterizing the Human Bajwa J (2017) Cutaneous cytology and the dermatology patient. Can Mycobiota: A Comparison of Small Subunit rRNA, ITS1, ITS2, Vet J 58:625–627 and Large Subunit rRNA Genomic Targets. Front Microbiol Bond R, Collin NS, Lloyd DH (1994) Use of contact plates for the 9:2208. https://​doi.​org/​10.​3389/​fmicb.​2018.​02208 quantitative culture of Malassezia pachydermatis from canine Korbelik J, Singh A, Rosseau J, Weese S (2018) Analysis of the otic skin. J Small Anim Prac 35:68–72 mycobiota in dogs with otitis externa compared to healthy indi- Bond R, Guillot J, Cabañes FJ (2010) Malassezia yeasts in animal viduals. Vet Dermatol 29:417e138. https://​doi.​org/​10.​1111/​vde.​ disease. Malassezia and The Skin, 1st edn. Springer, Berlin, pp 12665. 271–299 Li H, Durbin R (2010) Fast and accurate long-read alignment with Bond R, Morris DO, Guillot J, Bensignor EJ, Robson D, Mason KV, Burrows-Wheeler transform. Bioinformatics 26:589–595. https://​ Kano R, Hill PB (2020) Biology, diagnosis and treatment of doi.​org/​10.​1093/​bioin​forma​tics/​btp698 Malassezia dermatitis in dogs and cats Clinical Consensus Guide- Love MI, Huber W, Anders S (2014) Moderated estimation of fold lines of the World Association for Veterinary Dermatology. Vet change and dispersion for RNA-seq data with DESeq2. Genome Dermatol 31:28–74. https://​doi.​org/​10.​1111/​vde.​12809 Biol 15:550. https://​doi.​org/​10.​1186/​s13059-​014-​0550-8 Bradley CW, Lee FF, Rankin SC, Kalan LR, Horwinski J, Morris DO, McMurdie PJ, Holmes S (2013) phyloseq: An R Package for repro- Grice EA, Cain CL (2020) The otic microbiota and mycobiota in a ducible interactive analysis and graphics of microbiome census referral population of dogs in eastern USA with otitis externa. Vet data. PLoS ONE 8:e61217. https://d​ oi.​org/​10.​1371/​journa​ l.​pone.​ Dermatol 31:225–249. https://​doi.​org/​10.​1111/​vde.​12826 00612​17 Cabañes FJ (2021) Diagnosis of Malassezia dermatitis and otitis in Meason-Smith C, Olivry T, Lawhon SD, Hoffmann AR (2020) dogs and cats, is it just a matter of counting? Rev Iberoam Micol Malassezia species dysbiosis in natural and allergen-induced 38:3–4. https://​doi.​org/​10.​1016/j.​riam.​2020.​03.​001 atopic dermatitis in dogs. Med Mycol 58:756–765. https://​doi.​ Cafarchia C, Gallo S, Capelli G, Otranto D (2005a) Occurrence and org/​10.​1093/​mmy/​myz118 population size of Malassezia spp. in the external ear canal of Meason-Smith C, Diesel A, Patterson AP, Older CE, Mansell JM, dogs and cats both healthy and with otitis. Mycopathologia Suchodolski JS, Rodrigues Hoffmann A (2015) What is living on 160:143–149. https://​doi.​org/​10.​1007/​s11046-​005-​0151-x your dog’s skin? Characterization of the canine cutaneous myco- Cafarchia C, Gallo S, Romito D, Capelli G, Chermette R, Guillot J, biota and fungal dysbiosis in canine allergic dermatitis. FEMS Otranto D (2005b) Frequency, body distribution, and population Microbiol Ecol 91:fiv139. https://​doi.​org/​10.​1093/​femsec/​fiv139 size of Malassezia species in healthy dogs and in dogs with local- Miller WH, Griffin CE, Campbell KL (2000) Fungal skin diseases. ized cutaneous lesions. J Vet Diagn Invest 17:316–322. https://d​ oi.​ Muller & Kirk’s Small Animal Dermatology, 6th edn. Saunders, org/​10.​1177/​10406​38705​01700​403 Philadelphia, pp 336–422 De Filippis F, Laiola M, Blaiotta G, Ercolini D (2017) Different ampli- Mota-Gutiérrez J, Ferrocino I, Rantsiou K, Cocolin L (2019) Metatax- con targets for fungal sequencing-based studies of fungal diversity. omic comparison between internal transcribed spacer and 26S Appl Environ Microbiol 83:e00905-e917. https://d​ oi.o​ rg/1​ 0.1​ 128/​ ribosomal large subunit (LSU) rDNA gene. Int J Food Microbiol AEM.​00905-​17 290:132–140. https://​doi.​org/​10.​1016/j.​ijfoo​dmicro.​2018.​10.​010 Díaz L, Castellá G, Bragulat MR, Martorell J, Paytuví-Gallart A, San- Nardoni S, Mancianti F, Corazza M, Rum A (2004) Occurrence of severino W, Cabañes FJ (2021) External ear canal mycobiome of Malassezia species in healthy and dermatologically diseased dogs. some rabbit breeds. Med Mycol 59:683–693. https://​doi.​org/​10.​ Mycopathologia 157:383–388. https://​doi.​org/​10.​1023/b:​myco.​ 1093/​mmy/​myaa0​97 00000​30416.​36743.​dd Duclos D (2013) Canine pododermatitis. Vet Clin North Am Small Puig L, Bragulat MR, Castellá G, Cabañes FJ (2017) Characteriza- Anim Pract 43:69–70. https://d​ oi.o​ rg/1​ 0.1​ 016/j.c​ vsm.2​ 012.0​ 9.0​ 12 tion of the species Malassezia pachydermatis and re-evaluation Forbes JD, Knox NC, Ronholm J, Pagotto F, Reimer A (2017) of its lipid dependence using a synthetic agar medium. PLosONE Metagenomics: the next culture-independent game changer. Front 12:e0179148. https://​doi.​org/​10.​1371/​journ​al.​pone.​01791​48 Microbiol 8:1069. https://​doi.​org/​10.​3389/​fmicb.​2017.​01069 Puig L, Castellá G, Cabañes FJ (2019) Quantification of Malassezia Forde BM, O’Toole PW (2013) Next-generation sequencing technolo- pachydermatis by real-time PCR in swabs from the external ear gies and their impact on microbial genomics. Brief Funct Genom- canal of dogs. J Vet Diagn Invest 31:440–447. https://​doi.​org/​10.​ ics 12:440. https://​doi.​org/​10.​1093/​bfgp/​els062 1177/​10406​38719​840686 Frau A, Kenny JG, Lenzi L, Campbell BJ, Ijaz UZ, Duckworth CA, Puigdemont A, D’Andreano S, Ramió-Lluch L, Cuscó A, Francino O, Burkitt MD, Hall N, Anson J, Darby AC et al (2019) DNA Brazis P (2021) Effect of an anti-inflammatory pomegranate otic 13 Veterinary Research Communications treatment on the clinical evolution and microbiota profile of dogs patients using real-time PCR. Microbiol Immunol 50:549–552. with otitis externa. Vet Dermatol 32:158-e37. https://​doi.​org/​10.​ https://​doi.​org/​10.​1111/j.​1348-​0421.​2006.​tb038​25.x 1111/​vde.​12930 Sugita T, Boekhout T, Velegraki A, Guillot J, Hađina S, Cabañes Raabe P, Mayser P, Weiss R (1998) Demonstration of Malassezia fur- FJ (2010) Epidemiology of Malassezia-related skin diseases. fur and M. sympodialis together with M. pachydermatis in veteri- Malassezia and The Skin, 1st edn. Springer, Berlin, pp 65–119 nary specimens. Mycoses 41:493–500. https://​doi.​org/​10.​1111/j.​ Tang S, Prem A, Tjokrosurjo J, Sary M, Van Bel MA, Rodrigues- 1439-​0507.​1998.​tb007​12.x Hoffmann A, Kavanagh M, Wu G, Van Eden ME, Krumbeck JA Reynolds DR, Taylor JW (1993) The Fungal Holomorph: Mitotic, Mei- (2020) The canine skin and ear microbiome: A comprehensive otic and Pleomorphic Speciation in Fungal Systematics, 1st edn, survey of pathogens implicated in canine skin and ear infections CAB international, Surrey, UK using a novel next-generation-sequencing-based assay. Vet Micro- Schoch CL, Seifert KA, Huhndorf S, Robert V, Spouge JL, Levesque biol 247:108764. https://​doi.​org/​10.​1016/j.​vetmic.​2020.​108764 CA, Chen W et al (2012) Nuclear ribosomal internal transcribed Vuran E, Karaarslan A, Karasartova D, Turegun B, Sahin F (2014) spacer (ITS) region as a universal DNA barcode marker for Fungi. Identification of Malassezia species from pityriasis versicolor Proc Natl Acad Sci U S A 109:6241–6246. https://​doi.​org/​10.​ lesions with a new multiplex PCR method. Mycopathologia 1073/​pnas.​11170​18109 177:41–49. https://​doi.​org/​10.​1007/​s11046-​013-​9704-6 Scorzetti G, Fell JW, Fonseca A, Statzell-Tallman A (2002) Systematics of basiodiomycetous yeasts: a comparison of large subunit D1/D2 Publisher's note Springer Nature remains neutral with regard to and internal transcribed spacer rDNA regions. FEMS Yeast Res jurisdictional claims in published maps and institutional affiliations. 2:495–517. https://​doi.​org/​10.​1111/j.​1567-​1364.​2002.​tb001​17.x Sugita T, Tajima M, Tsubuku H, Tsuboi R, Nishikawa A (2006) Quan- titative analysis of cutaneous Malassezia in atopic dermatitis 13