Characterization of Skin Microbiome in Tinea Pedis

Abstract

Tinea pedis has been associated with Trichophyton rubrum infection. However, it’s not clear whether other microbes were implicated in disease pathogenesis. The composition of microbial communities of patients with tinea pedis and healthy controls were analyzed to identify the characteristics of tinea pedis and differences associated with clinical patterns. We found that microbial community structures were different in patients with tinea pedis compared with healthy controls. Moreover, skin microbiome varied in different forms of tinea pedis. Healthy controls exhibited greater fungal diversity than patients with tinea pedis. In patients with tinea pedis, the dominant bacterial and fungal genera were Staphylococcus and Trichophyton. Compared with healthy controls, Corynebacterium tuberculostearicum was decreased and T. rubrum was increased. C. tuberculostearicum was more abundant in vesicular tinea pedis than in hyperkeratotic and interdigital tinea pedis. Interdigital tinea pedis had a higher detection rate of Corynebacterium minutissimum and T. rubrum than the other forms. These results indicated that bacterial microbes may take part in the development of tinea pedis.

Electronic supplementary material

The online version of this article (10.1007/s12088-019-00816-y) contains supplementary material, which is available to authorized users.

Introduction

Normal skin is colonized by a lot of microorganisms that live harmlessly as commensals on its surface [1]. Environmental characteristics play a strong role in shaping skin microbial communities [2]. Each region of the skin has a distinct ecological niche. The skin of the foot represents a unique environment, with a large number of sweat glands but lack of sebaceous glands. Sebaceous secretions are thought to have antimicrobial properties [3]. Furthermore, wearing occlusive shoes may create a high relative humidity and warm conditions. The fungi and gram-negative bacterial thrive in warm and humid conditions [4]. The excessive growth of normal microorganisms could result in disease.

Tinea pedis is the most common dermatophytoses in the post-pubescent period, with an incidence rate of nearly 20% [5, 6]. Tinea pedis has a high infectivity and recurrence rate, which affects the quality of life of patients. Moreover, the toxic effects of the long course of treatment, including topical and systemic antifungal therapies, lead to the need for alternative strategies.

In recent years, many dermatologists have focused their attention on the role of skin microbiome in tinea pedis. Keisha found that fungal diversity was increased while bacterial diversity was decreased on skin of the foot compared with the other sites [7]. Most authors consider dermatophytes as the most frequent etiological agents, followed by yeasts and non-dermatophyte moulds [8, 9]. Djeridane reported that the main microbes in tinea pedis were Candida parapsilosis and Trichophyton rubrum [10]. According to a 10-year retrospective study, the predominant pathogen was T. rubrum in tinea pedis, which accounts for 71.19% [9]. The isolation rate of T. rubrum is up to 98.1% in tinea pedis and tinea unguium in Tunisia, followed by C. parapsilosis [11]. At present, most studies have focused on the role of dermatophytes in the pathogenesis of tinea pedis. However, the role of other microbes in tinea pedis has been neglected. Understanding the structure of microbial communities and their function in tinea pedis may provide some new approaches for the treatment of tinea pedis. The purpose of this work was to evaluate the microbial composition of tinea pedis and attempted to investigate the relationship between skin microbes in tinea pedis.

Methods

Subjects and Sampling

Between August–September 2016 and March 2017, a total of 26 patients with tinea pedis and 10 healthy controls were recruited (Supplementary Material-Table 1). Patients were confirmed through the direct microscopy in potassium hydroxide (KOH). Specimens for microbial analysis from the lesions were collected if fungus elements were positive (KOH-positive). None of the subjects had received topical anti-fungal medications for 4 weeks or oral anti-fungal medications for 6 months prior to the study. Subjects younger than 18 years and older than 60 years old and those with diabetes mellitus, peripheral vascular disease or immunosuppression were excluded from the study. The protocol of this work had been approved by the Scientific and Ethical Committee of the Shanghai Dermatology Hospital. Informed consents were obtained from all the participants.

The skin samples were obtained by repeatedly rubbed 4 times over the object region using iCleanhcy flocked swabs (Huachenyang Corporation, Shenzhen, China). Lesional specimens were collected directly from the plantar or interdigital area of the foot. Control samples were obtained from the healthy skin of a healthy person on a similar site of the foot. Samples were stored at − 80 °C prior to further processing.

DNA Extraction

DNA was extracted from samples by using the QIAamp DNA Stool Mini Kit (QIAGEN) according to the operation instructions. The concentration and purity of DNA were checked by running the samples on 1.0% agarose gels.

Pyrosequencing

The 16S rRNA V1–V2 region was amplified with the primers 27-F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 338-R (5′-TGCTGCCTCCCGTAGGAGT -3′). And the ITS region was amplified with the primers ITS1-F (5′-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2 (5′-GCTGCGTTCTTCATCGATGC -3′). Cycling conditions were 94 °C for 2 min, followed by 25 cycles of 94 °C for 30 s, 56 °C for 30 s, and 72 °C for 30 s, at last a final step at 72 °C for 5 min. All the samples were sequenced via IlluminaMiSeq platform by TinyGene Bio-Tech Co., Ltd (Shanghai, China). Raw data have been submitted to the NCBI Sequence Read Archive (SRP167680).

Statistical Analysis

The 16S and ITS sequences were analyzed by software mothur version 1.33.3, UPARSE (usearch version v8.1.1756) and R (version 3.2.3). Sequences were clustered into operational taxonomic units (OTUs) at the 97% similarity threshold following the UPARSE pipeline (http://drive5.com/usearch/manual/uparsecmds.html). The taxonomy of the OTU representative sequences was analyzed against Silva 119 database or unite database using an 80% confidence level by the classify.seqs command in mothur (version 1.33.3). Standard Normality and Equality of Variance Tests were performed to determine whether data are parametric and/or non-parametric. For statistical analysis, Independent T-test and ANOVA(LSD/Dunnett’s comparison) and Wilcoxon Test (for non-parametric data) were performed, The level of significance was p-values < 0.05.

Results

Metagenomics Analysis of Skin Microbiota in Patients with Tinea Pedis and Healthy Controls

In the skin microbiota of patients with tinea pedis, we identified 26 bacterial phyla, 462 genera and 488 species; 8 fungal phyla, 372 genera and 581 species. In healthy controls, we identified 12 bacterial phyla, 187 genera and 128 species; 5 fungal phyla, 162 genera and 206 species. The most prevalent bacteria phyla in tinea pedis were Firmicutes, Actinobacteria, Proteobacteria, and Bacteroidetes. Firmicutes and Actinobacteria comprised more than 80% of the total sequences. Staphylococcus constituted more than 30% of the bacterial genera. The most abundant species in patients with tinea pedis were Corynebacterium tuberculostearicum and Staphylococcus pettenkoferi. Patients with tinea pedis showed a decrease in C. tuberculostearicum (6.49%) as compared to healthy controls (22.54%) (p < 0.05) (Fig. 1). Ascomycota and Basidiomycota were the most prevalent fungal phyla, which accounted for 87.37% of the total sequences. Furthermore, patients with tinea pedis showed higher abundance of Ascomycota (75.92%) and lower abundance of Basidiomycota (11.45%) as compared to healthy controls (p < 0.05). The major fungal genera were Trichophyton, Aspergillus, and Candida in patients with tinea pedis. An increase in Trichophyton (27.26%) was observed in patients with tinea pedis as compared to healthy controls (0.065%) (p < 0.05). The most abundant fungal species included of Trichophyton rubrum (26.75%), Cladosporium halotolerans (5.45%), and Candida parapsilosis (4.79%). Trichophyton mentagrophytes var. interdigitale accounted for only approximately 0.95% of the total sequences. Patients with tinea pedis showed an increase in T. rubrum (26.75%) as compared to healthy controls (0.01%) (p < 0.05) (Fig. 2).

Fig. 1.

Distribution of the bacterial species in patients with tinea pedis and healthy controls. a Relative abundances of bacterial species in the skin of healthy controls, b Relative abundances of bacterial species on lesional skin samples of tinea pedis

Fig. 2.

Distribution of the fungal species in patients with tinea pedis and healthy controls. a Relative abundances of fungal species in the skin of healthy controls, b Relative abundances of fungal species on lesional skin samples of tinea pedis

Correlations Between Bacteria and Fungi

To determine the correlation between bacteria and fungi, we studied the relationship between T. rubrum and bacterial species. T. rubrum was positively correlated to Corynebacterium minutissimum (p < 0.05), while it was negatively correlated to C. tuberculostearicum, S. pettenkoferi, and Paracoccus sphaerophysae (Supplementary Material-Table 2).

The Characteristics of Microbiota in Different Clinical Patterns of Tinea Pedis

Interdigital tinea pedis showed a higher abundance of Actinobacteria compared to hyperkeratotic and vesicular tinea pedis (58.19% vs. 37.25%; p = 0.05 and 58.19% vs. 22.67%; p < 0.05, respectively). Firmicutes were more abundant in hyperkeratotic tinea pedis than in vesicular and interdigital tinea pedis (50.13% vs. 44.19% and 50.13% vs. 38.22%; p > 0.05, respectively). Similarly, Proteobacteria were more abundant in hyperkeratotic tinea pedis than in vesicular and interdigital tinea pedis (21.31% vs. 13.02% and 21.31% vs. 1.56%; p < 0.05, respectively). At the genera level, Staphylococcus was more abundant in hyperkeratotic tinea pedis than in vesicular and interdigital tinea pedis (41.89% vs. 37.96% and 41.89% vs. 31.73%; p > 0.05, respectively). Furthermore, no difference was observed for Streptococcus, Brevundimonas, or Propionibacterium. At the species level, C. tuberculostearicum was more abundant in vesicular tinea pedis than in hyperkeratotic and interdigital tinea pedis (10.07% vs. 2.84%; p < 0.05 and 10.07% vs. 5.37%; p > 0.05, respectively). S. pettenkoferi was more abundant in hyperkeratotic tinea pedis than in vesicular and interdigital tinea pedis (6.93% vs. 0.97% and 6.93% vs. 2.41%; p > 0.05, respectively). C. minutissimum was more abundant in interdigital tinea pedis than in hyperkeratotic and vesicular tinea pedis (4.61% vs. 0.19%; p < 0.05 and 4.61% vs. 0.35%; p > 0.05, respectively). No difference was observed for P. sphaerophysae (Supplementary Material-Figs. S1).

With regard to fungi, the most prevalent phyla detected in patients with tinea pedis included Ascomycota (75.92%) and Basidiomycota (11.45%). At the genera level, Trichophyton was more abundant in interdigital tinea pedis than in hyperkeratotic and vesicular tinea pedis (59.79% vs. 15.34%; p < 0.05 and 59.79% vs. 20.31%; p > 0.05, respectively). At the species level, T. rubrum was more abundant in interdigital tinea pedis than in hyperkeratotic and vesicular tinea pedis (59.79% vs. 15.34% and 59.79% vs. 18.08%; p < 0.05, respectively). No difference was observed in other fungal species (Supplementary Material-Figs. S2).

Microbial Diversity

We analyzed the difference in the Shannon diversity index in different sites which were calculated to determine the diversity in different groups. The data showed that healthy controls exhibited greater fungal diversity than patients with tinea pedis, but a similar pattern was not observed for bacteria (Fig. 3).

Fig. 3.

a Comparison of the Shannon diversity estimates among different groups for bacterial species composition, b Comparison of the Shannon diversity estimates among different groups for fungal species composition. Differences were considered statistically significant at p-value < 0.05. (Healthy control: IH-interdigital, PH-plantar; Tinea pedis: IT- interdigital; PT- plantar)

Discussion

In the present study, the characterization of skin microbiome in patients with tinea pedis was characterized. The major bacterial genera in patients with tinea pedis were Staphylococcus, Streptococcus, Brevundimonas, and Propionibacterium, while in healthy controls, the major bacterial genera were Staphylococcus, Pseudoclavibacter, Micrococcus, and Neisseria. There was no significant difference in bacterial genera composition between patients with tinea pedis and healthy controls. However, these results were not entirely consistent with previous research, especially for the skin bacteria. Marshall found that aerobic coryneform bacteria, Staphylococci, and Micrococci constituted the major microbial group in the feet of healthy persons [12]. The reason might be that these skin microbes would live without sebaceous glands or their products. The predominant organisms on moist skin were Staphylococcus and Corynebacterium [13]. Our study showed that the most common species were C. tuberculostearicum and S. pettenkoferi in patients with tinea pedis and healthy controls. Marshall’s study showed that a great number of Staphylococci and aerobic coryneform bacteria were detected on foot of all subjects [12].

The analysis of fungi in patients with tinea pedis showed that the main species were T. rubrum, C. halotolerans, and C. parapsilosis. T. rubrum was the most frequently pathogen, consistent with previous researches [9, 14]. T. rubrum had a significantly higher detection rate in patients with tinea pedis than in healthy controls. This result mirrors the findings of previous works [10, 11]. The pathogenic dermatophytes invade the stratum corneum through the release of various lipases and ceramides [15]. Our findings indicate that the overgrowth of T. rubrum had driven the development of tinea pedis. C. halotolerans and C. parapsilosis were also the main fungal species. Cladosporium species are saprobic dematiaceous fungi, which are usually associated with allergic rhinitis or localize superficially or in deep lesions [16]. C. halotolerans is the most frequent species of Cladosporium. The detection rate of C. halotolerans was 5.45% in patients with tinea pedis and 4.98% in healthy controls. However, it was not clear whether C. halotolerans was involved in the disease pathogenesis. C. parapsilosis is a saprophyte yeast of the human skin. As the most isolated yeast, C. parapsilosis has been considered as the second agent responsible for foot mycoses [11].

Few studies have examined the relationship between bacteria and fungi in tinea pedis. Our study showed that T. rubrum was positively correlated with C. tuberculostearicum and Corynebacterium minutissimum. C. tuberculostearicum, a lipophilic Corynebacterium, is normally considered an innocuous member of the healthy skin microbiota. Several studies have found that C. tuberculostearicum is associated with the formation of body odors [17, 18]. Smeekens found that the antifungal and antistaphylococcal responses were unaffected by normal Corynebacterium [13]. However, it is unclear whether C. tuberculostearicum is involved in the pathogenesis of tinea pedis. C. minutissimum is also thought to be a member of the normal skin flora, which usually grows in moist and occluded areas. A study involving 121 patients with interdigital foot infections, showed that the primary pathogen were T. rubrum (33.7%) and C. minutissimum (19.8%) [19]. These findings suggest that C. minutissimum often coexists with other dermatophytes. Some scholars have investigated the relationship between bacteria and fungi in vivo. They report that T. rubrum was significantly decreased, when cultivated with Janthinobacterium lividum, and propose that J. lividum may inhibit the growth of T. rubrum [20]. The antifungal properties of J. lividum suggest new antifungal treatments. Our result showed that Trichophyton was negatively correlated with Janthinobacterium. Another in vivo experiment showed that Pseudomonas aeruginosa can inhibit the growth of T. rubrum [4]. We found that Trichophyton was negatively correlated with Pseudomonas. The relationship between bacteria and fungi in tinea pedis needs more experimental confirmations.

Tinea pedis is classified into hyperkeratotic, vesicular, interdigital and ulcerative, according to clinical features; the first three forms were noted in our study. In vesicular tinea pedis, C. tuberculostearicum was more abundant than in other forms. Interdigital type is the most common clinical pattern and showed a higher detection rate of C. minutissimum than in other forms. This may be because C. minutissimum generally prefer to grow in moist and occluded areas. T. rubrum was found to be significantly higher in interdigital tinea pedis than in other forms. This finding was consistent with a previous study, which indicated that T. rubrum and Trichophyton interdigitale were the mian pathologic dermatophytes in interdigital tinea pedis [21].

Microbial diversity may vary from different sites and different disease states. We found that healthy controls exhibited greater fungal diversity than patients with tinea pedis. This result was consistent with previous research [22]. Another work in patients with atopic dermatitis (AD) showed skin microbial diversity was drastically reduced during flares in AD, and it can be restored with common AD therapies [23]. In the future, treatment should be focused to restore the microbial diversity, rather than to eliminate a certain microorganism.

In conclusion, the composition of bacteria and fungi of patients with tinea pedis is different from that of a healthy person. Furthermore, skin microbiota composition of different clinical forms of tinea pedis was different. However, bacterial and bacterial-fungal interactions in tinea pedis remain uncertain and need more experimental confirmations. Skin microbial dysbiosis may affect the occurrence and development of the disease. Further studies on bacterial-fungal ecology in tinea pedis should be conducted, which may eventually lead to new directions for treatment.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflict of interest.