Prevalence of Malassezia species on the skin of HIV-seropositive patients

Malassezia is a genus of lipophilic yeasts residing on the skin of warm-blooded animals. The correlation between specific species and their involvement in skin diseases has been well researched. However, only very few studies have investigated the distribution of Malassezia spp. on the healthy skin of patients infected with human immunodeficiency virus (HIV). The purpose of this work was to analyze whether the composition of Malassezia spp. isolated from the skin of the HIV-infected patients differs from that of healthy individuals. The study included a total of 96 subjects, who were divided into two equally sized groups: HIV-seropositive and HIV-seronegative. The specimens were collected from the subjects by swabbing four anatomical sites (face, chest, back, and scalp). Species were identified using phenotype-based methods, and the identification of strains isolated from the HIV-seropositive patients was confirmed by PCR sequencing of the rDNA cluster. Malassezia spp. were isolated from 33 (69%) HIV-seropositive patients and 38 (79%) healthy volunteers. It was found that men were much more likely to have their heads colonized with Malassezia spp. than women. The most prevalent species on the skin of both HIV-seropositive and HIV-seronegative individuals were Malassezia sympodialis, M. globosa, and M. furfur, albeit at different proportions in the two populations. The diversity of Malassezia spp. was the highest on the face of the HIV-seropositive patients (Shannon–Weiner Index H = 1.35) and lowest on the back of the healthy volunteers (H = 0.16). The phenotype- and molecular-based identification methods were congruent at 94.9%. It was observed a tendency that the HIV-seropositive patients had higher CD4+ cell counts, indicating higher colonization with Malassezia spp.

www.nature.com/scientificreports/ cells from skin express homing receptors that promote their trafficking to the sites of inflammation in the skin. Specifically, skin-homing Th17 and Th22 cells contain chemokine receptors (CCR6+, CCR4+, CCR10+, CLA+), which direct their E selectin-dependent extravasation mediated by cutaneous lymphocyte antigen and migration along the gradients of chemokine ligands (CCL20, CCL22, and CCL27) 29 . As mentioned above, Th17 activation is a mechanism of defense against Malassezia, induced via secreted metabolites. Lack of these lymphocytes disturbs the immunological response of the skin, causing imbalance in skin microbial homeostasis, including the overgrowth of Malassezia.
The most prevalent Malassezia-associated skin infection is pityriasis versicolor, which presents as hyperpigmented or hypopigmented finely scaled macules or patches mainly in the neck, arms, and trunk. Another skin disease caused by Malassezia spp. is folliculitis, which typically manifests as pruritic, follicular papulopustular eruptions distributed on the back, chest, and upper limbs 30 . In other skin diseases including SD, atopic dermatitis (AD), and psoriasis (PS), the Malassezia yeasts exacerbate the symptoms and perpetuate the condition, probably due to the dysfunction of the skin barrier and exposure of the immune system to Malassezia antigens, which stimulate allergic inflammatory responses 31,32 . Such diseases are often found in HIV patients.
Central venous catheter-related Malassezia infections are more common in preterm neonates than adults. Systemic infections have been reported in the recipients of hematopoietic cell transplants, patients with underlying hematologic malignancies, cancer patients receiving monoclonal antibody therapy, and patients with other immunodeficiency states (e.g. solid organ transplantation, diabetes mellitus, prolonged glucocorticoid therapy, HIV). High temperature and humidity may facilitate the colonization of catheters, while lipid infusions might predispose individuals to both catheter colonization and infections. Additional risk factors of infections include low birthweight, severe comorbidities, and arterial catheterization for longer than 9 days. However, despite the presence of fungemia, disseminated fungal infection is uncommon 33 .
Most approaches focus on the role of Malassezia in skin diseases, especially severe SD, which is characteristic of HIV patients. So far, only very few studies have investigated the distribution of Malassezia spp. on the healthy skin of the HIV-infected patients [34][35][36][37][38][39] . This work analyzed whether the composition of Malassezia spp. isolated from the skin of the HIV-infected patients differs from that of healthy individuals.

Results
Demography and clinical details. The detailed demographic data of the HIV-seropositive patients and the control group are presented in Table 1. All the enrolled participants were of Caucasian origin.
Mycological investigation. Malassezia spp. were isolated from 33 (69%) out of 48 HIV-seropositive patients (74% males and 44% females) and 38 (79%) out of 48 healthy subjects (83% males and 69% females). In eight HIV-seropositive patients (21.6%) who did not report the use of antifungal agents, Malassezia spp. were not found in the cultures derived from the specimens of any studied location. Likewise, Malassezia yeasts were not identified in seven (63.6%) out of 11 patients who used or were suspected of using antifungal agents (Fisher's exact test, p = 0.0219). www.nature.com/scientificreports/ The most colonized body sites were the chest and shoulders, while the least colonized was the head. No differences were observed in the frequency of isolation depending on the HIV status in each of the studied locations (Mantel-Haenszel test, p = 0.508) ( Table 2).
Men were much more likely to have their heads colonized with Malassezia spp. compared to women (16% vs 0%, respectively). This finding was statistically significant in the case of the control group (Fisher's exact test, p = 0.022) and the two groups combined (Fisher's exact test, p = 0.010). When the results from the two groups were analyzed together, the back of men was found to be more frequently colonized with Malassezia spp. compared to women (χ 2 -test, p = 0.015; Fig. 1).
Prevalence of Malassezia spp.. The two most prevalent Malassezia species on the skin of the HIV-seropositive and HIV-seronegative individuals were M. sympodialis and M. globosa; however, their frequencies differed considerably in the two populations (45.5% vs 89.3% and 28.8% vs 6.7%, respectively). Malassezia furfur was isolated from 16.7% of the HIV-seropositive patients, and only from 1.3% of healthy volunteers. Three spp., namely Malassezia slooffiae, M. restricta, and Malassezia obtusa, were isolated only from the HIV-seropositive patients, but at a low frequency of 3% (M. slooffiae, M. restricta) or 1.5% (M. obtusa). The detailed data are shown in Table 3 and Fig. 2.
The biodiversity of Malassezia spp. was the highest on the face of the HIV-seropositive patients (H = 1.35), while it was the lowest on the back of healthy volunteers (H = 0.16; Table 3).

Discussion
Numerous studies have focused on the role of Malassezia yeasts in the pathogenesis of SD, AD, and PS 40 , but only a few have analyzed the importance of Malassezia as a part of natural microbiota on the healthy skin in the context of HIV status. There are two approaches to determining the presence of microbes in a given niche. The first is methods consisting of the isolation of microorganisms on culture media, followed by their identification (by morphological, biochemical, or by nucleic acid analysis or protein analysis methods). The second method consists in searching for the nucleic acid corresponding to taxa directly in the sample and does not require the isolation of living microorganisms. The limitation of the first one is the different survival of microorganisms outside the host organism, which means that some species may not be detected at all. Also, the proportions of species may be disturbed in favor of species growing faster on the media. In the genomics methods, the results appear to be more reliable. However, finding of fungal DNA does not necessarily equate the presence of viable cells.
As described in the Introduction, HIV patients differ from the noninfected individuals in SSL composition and immunological defense. It is therefore expected that the healthy skin of HIV patients should have different Malassezia spp. independent of skin diseases. The main finding of this study was the differences in the composition of Malassezia species between the populations examined. Generally, it was more diverse in www.nature.com/scientificreports/ the HIV-seropositive group than in healthy subjects. Regardless of the presence of Malassezia, HIV infection is regarded as a predisposing factor for SD, and plenty of studies have investigated the role of Malassezia in SD. However, because of the lack of broad lipidomic, genomic, and metabolomic studies, we are not sure if there is confusion between cause and effect.
In this study, we found no differences in the frequency of Malassezia isolation, depending on the HIV status. Håkansson et al. quantified Malassezia cultures retrieved from 12 HIV-seropositive and 12 HIV-seronegative homosexuals and showed that the mean number of Malassezia/cm 2 , the mean serum antibody titers against the yeasts, and the occurrence of cutaneous disorders did not differ significantly between the two studied groups 38 . Likewise, no differences were found between the HIV-seropositive and control groups by de Vroey and Song, who evaluated samples from the forehead and back of subjects and by Wikler et al. who compared samples from the chest and back 35,36 .
di Silverio et al. found that the abundance of Malassezia spp. was greater on the skin of HIV-seropositive than healthy individuals when they evaluated the number of fungal elements in microscopic smears from scalp, forehead, nose, and axillae. However, no differences were found when smears from the chest, groin, and back were compared between the two patient groups 37 . Pechere and Saurat compared the density of Malassezia yeasts on the forehead of 40 HIV-seronegative and 38 HIV-seropositive people with clinically normal skin and found that it was much higher in the latter group 41 .
In the present study, the most prevalent species, with abundant growth, was found to be M. sympodialis which was observed in 89.3% of healthy volunteers and 45.4% of the HIV-seropositive patients. This species was mainly recovered from the chest and back of healthy subjects, with an isolation rate of 52-54%. The next two most common species were M. globosa and M. furfur, which were found at a higher frequency in the HIVseropositive individuals (i.e. 28.8% vs 16.7% and 6.7% vs 1.5%, respectively). A previous study on the prevalence of Malassezia spp. in the Polish population, showed the dominance of M. sympodialis with single M. globosa, M. slooffiae, and M. restricta. However, the results of that study, as well as ours, may have been influenced by the use of the culture method 11 . In the same temperate climate and similar latitude, a Canadian study was performed by Gupta and Kohli using the contact plate method. They found a very congruent distribution of Malassezia spp. In Europe, M. sympodialis was found as the predominant species on healthy skin, especially on the trunk, in Bosnian and Herzegovinian populations 42 . This species was also most commonly found on the skin of healthy people or the healthy skin of dermatological patients in Spain 43 and Sweden 13 . In a Portuguese study which was based on the contact plate method and used adhesive tape for sample collection on the forearm and matched site, M. sympodialis was shown to be strongly predominant, followed by single M. restricta and M. globosa 44 . In the study of Aspiroz et al. (1999), the most abundantly isolated species in healthy Spanish people was M. globosa, yet M. sympodialis was predominant on the back 44 .
All the abovementioned studies are based on culture methods. In one study based on direct microbiome analysis, M. restricta (47%) was dominant on the side of the nose of healthy Swiss individuals, and M. sympodialis (25%) was the second most frequent species 45 . Furthermore, in the study on Bosnian and Herzegovinan populations, M. restricta was the most prevalent species on healthy scalp, while M. furfur and M. sympodialis were more frequent in the cultures derived from healthy trunk skin 42 .
Malassezia colonization in Asians is very distinct, but two species are dominant: M. restricta and M. globosa. It was found that M. sympodialis was less frequent even in the culture-dependent studies. In a Japanese study conducted using both culture and nonculture methods, M. restricta was the predominant species on the face of men, M. globosa and M. dermatis on the upper trunk of men, and M. globosa and M. sympodialis on the upper trunk of women, but there were seasonal variations observed in the results 9 . Another large Japanese study based on direct PCR method conducted on 770 healthy subjects showed that the predominant species was M. restricta, followed by M. globosa, and the findings varied depending on age and sex 46 Similar results were found in a study from South Korea, where the dominant species were M. restricta (55%) and M. globosa (22.5%), while M. sympodialis was isolated only in 10% of patients, mainly from the chest 47  Analysis of data on the prevalence of Malassezia spp. in different ethnic groups is hard not only due to discrepancies in the sampling methodologies and approaches to obtain information about species (culture vs omics methodology) but also due to numerous additional factors. The primary factor is the difference in the sebum  (13)  www.nature.com/scientificreports/ composition of individuals with different origins. Six fatty acids were identified that were significantly different in quantity between African and Caucasian Americans-synthesized naturally in the skin, as the 14:0, 16:1, n10, and 18:1, n9 fatty acids; acquired exclusively from the diet, as the iso-18:2 fatty acid; and others may be a product of bacterial metabolism, as the 15:0 or 17:1 fatty acids-which would indicate microbiota differences on the surface of the skin between these ethnic groups 7 . Further influencers of Malassezia factors are the place of living (urban or rural area), hygiene habits (washing/oiling the body and hair), use of cosmetics and their composition, climatic differences even within a given country (dry in the mountain area or humid on the coast), clothing composition (natural or artificial), etc. 46,48,50 . Moreover, the spectrum and frequency of Malassezia spp. in healthy individuals have been shown to differ according to sex, location on the skin, and sampling season; species diversity and isolation rate have been found to be the highest in samples collected from the upper trunk of men during summer 9,51 . For this reason, although all the study participants were Caucasians, we tried to make our groups very similar in terms of sex, age, place of living, etc. Differences in the isolated Malassezia spp. were observed among HIV-seropositive patients, with respect to their CD4+ counts. In general, the lower was the CD4+ count, the lower was the recovery rate of Malassezia spp. Although the observed correlation was not statistically significant, some studies have described a similar observation. In a study from Indonesia, the density of Malassezia cells showed no significant relationship with the CD4+ lymphocyte count, albeit the numbers supported our observations, with lower Malassezia CFU counts seen in patients with low CD4+ levels 52 . The most recent study of Moreno-Coutino et al. showed very similar results (rate of Malassezia isolation from patients with CD4+ counts > 500, 200-499, and < 200 cells/mm −3 was 60%, 27%, and 13%, respectively). Interestingly, patients with SD did not show this correlation in their study 39 .  www.nature.com/scientificreports/ Another explanation is antifungal prophylaxis in immunosuppressed patients. In this study, only one patient with a low CD4+ count (i.e. < 200 cells/cm −3 ) declared using fluconazole. In addition, the use of antiseptic cosmetics and personal care products could potentially affect the results.
The proportion of Malassezia spp. among HIV patients with a high level of CD4+ lymphocytes was not similar to the proportion among healthy controls. This fact suggests that immunological response has no direct influence on Malassezia diversity.
According to Gupta and Kohli 10 , Malassezia spp. colonized the skin of 80-90% of healthy volunteers who were aged 15 years or older. In their study, the rate of Malassezia colonization (growth from at least one anatomical site) was 79% among healthy volunteers and 69% among the HIV-seropositive patients. In our study, we found a relatively low rate of colonization in comparison to others. In particular, we did not isolate any Malassezia strain from the scalp of women. This could be explained by the wide use of antidandruff shampoos with antimicrobial compounds in Poland. We did not control this factor, and the data collected from subjects were difficult to interpret because many patients did not pay attention to the type of shampoo they used. However, we noticed that 40% of participants declared using antidandruff shampoo, but there was no difference in the occurrence of Malassezia among groups (data not shown).
The underrepresentation of M. restricta could be explained by the slower growth and formation of small colonies on Leeming Notman agar, Oil-Potato Dextrose agar, and mDA compared to other species 53 . Studies based on molecular methods in different populations frequently have shown M. restricta as the most abundant Malassezia species on healthy and diseased human skin. Some other factors, including those related to population, may also play a role in this relationship.
The results of phenotypic and molecular identification were discordant in approximately 5% of cases. The reason for this might be misidentification with the use of phenotypic tests. Molecular-based methods serve as a more accurate and reliable means of species identification, since phenotypic tests might reveal only important diversity within Malassezia spp. 54,55 .
A limitation of the study was that only isolates recovered from the HIV-seropositive patients were subjected to confirmatory speciation using the molecular typing method. This was because isolates from healthy subjects were not preserved and were unavailable for subsequent molecular studies. However, since we observed a high concordance of results between the phenotype-based and molecular-based methods used for species identification among the HIV-positive patients (95%), and the species composition determined only by conventional methods, as in the case of the HIV-seronegative group, seems to be highly accurate.
Another limitation of the study was that the prevalence of other fungi and bacteria was not investigated in the evaluated skin samples. Indeed, the exploration of the entire skin mycobiome of the study subjects, which is achievable using metagenomic sequencing, would add more information for a better understanding of any potential shifts in the skin microflora associated with chronic inflammatory conditions, such as HIV infection.
In conclusion, the examined HIV-seropositive and HIV-seronegative individuals showed the same three core Malassezia spp. and rank-frequency distribution. In the former group, however, the species spectrum was wider to some extent. To better recognize this relationship, further studies with more patients are needed.

Materials and methods
Patient recruitment procedure. The study analyzed a total of 96 individuals. This included 48 HIVseropositive patients, who were treated at the Clinic of Infectious Diseases, Department of Gastroenterology, Hepatology and Infectious Diseases, Jagiellonian University in Kraków, and 48 healthy (HIV-seronegative) volunteers (control group).
Written informed consent statements obtained from all the HIV-infected patients and healthy volunteers, of both sexes and over the age of 18 years, who were included in the study. Patients with conditions that would not allow collecting material or who did not give the written consent for participation were excluded. The specimens were collected from 16.03.2011 to 27.04.2012. The participants were also surveyed during sample collection.
The study followed the Declaration of Helsinki guidelines (2008) and was approved by the Bioethics Committee of the Jagiellonian University (No. KBET/33/B/2011, dated 28.04.2011).
Malassezia isolation. Specimens were obtained from four anatomical sites: face (nasolabial wrinkles and forehead), chest (breastbone), back (interscapular region), and scalp. They were collected by rubbing 10-15 times with viscose swabs moistened with sterile saline. The samples were delivered to the laboratory within 2 h, where they were immediately inoculated onto modified Dixon's agar (mDA) medium (malt extract 36 g, mycological peptone 10 g, ox bile 20 g, Tween 40 10 cm −3 , glycerol 2 cm −3 , oleic acid 2 cm −3 , chloramphenicol 0.5 g, and agar 15 g/1000 cm −3 distilled water) and cultured in a humid chamber at 32 °C for 2 weeks and checked at 7 and 14 days. Species identification. Biochemical methods. Species were first identified using phenotypic methods, which involved the examination of macro-and micromorphologies and determination of biochemical profiles based on the organisms' requirement for exogenous lipid for their growth, catalase production, esculin hydrolysis, and assimilation of Cremophor EL and Tween. All the phenotypic tests were carried out and interpreted essentially as described elsewhere 56 (Fig. 4).
Strains isolated from the HIV-seropositive patients, after biochemical identification, were inoculated into cryo-tube with modified Dixon broth which was supplemented with glycerol and gradually cooled: firstly in a refrigerator to 4 °C, then in a freezer to − 22 °C, and finally stored at − 80 °C for further analysis 57  www.nature.com/scientificreports/ Molecular methods. For strains isolated from the HIV-seropositive patients, the identification of phenotypebased species was confirmed by molecular typing. Briefly, DNA was extracted from yeast cultures on mDA medium, using Genomic Mini AX Yeast kit (A&A Biotechnology, Poland) following the manufacturer's instructions. Molecular identification was performed by sequence analysis of a part of the rDNA cluster, comprising the internal transcribed spacer (ITS) regions ITS1 and ITS2, a complete 5.8S rRNA gene, and partial regions of the 18S and 26S rRNA genes. For this purpose, primers V9 (5′-TGC GTT GAT TAC GTC CCT GC-3′) and RLR3R (5′-GGT CCG TGT TTC AAGAC-3′) were used. PCR amplification, sequencing, sequence alignment, and mapping were carried out as described previously 11 . In short, the PCR mixtures with a total volume of 25 µl, containing approximately 20 ng of genomic DNA each, were prepared using TopTaq Master Mix kit (Qiagen, Germany) following the vendor's protocol. After initial denaturation at 94 °C for 5 min, the reaction mixtures were run through 30 cycles at 94 °C for 45 s, 56 °C for 30 s, and 72 °C for 1.5 min, and a final step at 72 °C for 10 min. Amplicons were sequenced either directly with the same primers as those used for the amplification, or when difficult-toread sequences were obtained, after cloning into the pGEM-T Easy Vector system (Promega, USA). Consensus sequences were assembled with ChromasPro ver. 1.7.1 (Technelysium, Australia) and searched against the Gen-Bank database of the National Center for Biotechnology Information (NCBI) using the BLASTN algorithm (https ://blast .ncbi.nlm.nih.gov/). For species identification, distance scores of up to 2.00% (98% match) were used as a proxy, and the species showing the closest match was considered as correctly identified.
All the determined nucleotide sequences were deposited in the GenBank database under the accession numbers listed in Supplementary Table. CD4+ level. The level of CD4+ lymphocytes in HIV patients was calculated during routine diagnostics at the Clinic of Infectious Diseases using flow cytometry methods with FACS Canto II (Becton Dickinson). Three milliliters of standardized cell suspensions were prepared by adding an actively growing colony to sterile water and adjusted to 1 on the McFarland scale. Next, the suspensions were added to the liquefied Sabouraud Glucose Agar medium and cooled to approximately 50 °C to solidify (1). In the solid medium, five holes of 2 mm were prepared (2) and filled with lipid compounds (Tween and Cremophor EL) (3). The prepared plates were incubated for 7-10 days at 32-34 °C and read visually (4).

Scientific Reports
| (2020) 10:17779 | https://doi.org/10.1038/s41598-020-74133-6 www.nature.com/scientificreports/ 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://creat iveco mmons .org/licen ses/by/4.0/.