Low
DEFB4
Copy Number and High Systemic
hBD-2 and IL-22 Levels Are Associated with
Dermatophytosis
Sameh W. Jaradat
1,2,3, Susana Cubillos
1, Nadine Krieg
1, Katja Lehmann
1, Bassam Issa
4, Susann Piehler
1,
Sylvi Wehner-Diab
1, Uta-Christina Hipler
1and Johannes Norgauer
1Dermatophytes initiate dermatophytosis, but susceptibility to infection is dictated by host genetic factors, although the role of some of these—such as human beta-defensin 2 (hBD-2) genomic (DEFB4) copy number (CN) variation and its induction by IL-22—remains unclear. This was investigated in this cross-sectional study in 442 unrelated Caucasian subjects, including 195 healthy controls and 247 dermatophytosis patients who were divided into five subgroups according to clinical presentation. DNA samples were evaluated for DEFB4 CN variation by relative quantification using the comparative CT method, and serum hBD-2 and IL-22 levels were determined by ELISA.
DEFB4CN in patients was significantly lower and, except in the tinea cruris subgroup, serum hBD-2 levels were higher than in controls. The positive correlation between hBD-2 levels andDEFB4CN observed in controls was not detected in patients, who also had higher serum IL-22 levels that were positively correlated with hBD-2 levels. Moreover, unlike in control subjects, the serum IL-22 level was negatively correlated with DEFB4CN in patients. Taken together, these findings suggest an association between decreasedDEFB4CN, elevated serum hBD-2 and IL-22 levels, and dermatophytosis, underscoring a gene/cytokine interaction in the occurrence of this infection. Journal of Investigative Dermatology(2015)135,750–758; doi:10.1038/jid.2014.369; published online 25 September 2014
INTRODUCTION
Dermatophytosis (Tinea) is a dermatophytic fungal infection of keratinized tissues. Trichophyton rubrum (T. rubrum) is the major etiologic agent for superficial dermatophytosis world-wide (Havlickova et al., 2008), and it can adhere to and invade superficial keratinized tissue via carbohydrate-specific adhesins and by secreting several kinds of keratinolytic proteases (Esquenazi et al., 2004; Grumbt et al., 2013). However, not all individuals are equally susceptible to dermatophytic infections (Faergemannet al., 2005), which is dependent on the interplay of dermatophytes with environmental and host factors. Epidemiological studies have identified many predisposing factors for infection, such as diabetes mellitus, psoriasis vulgaris, advanced age, and previous injury to the nails (Nenoff et al., 2014). Family history is also believed to be a risk factor (Ghannoum et al.,
2013), and pedigree analysis has shown autosomal dominant inheritance of T. rubrum susceptibility (Zaias et al., 1996). Genetic association studies have shown that susceptibility to T. rubrumonychomycosis is reduced by the presence of HLA DR53 and DR6 alleles (Zaitz et al., 1996; Asz-Sigall et al., 2010) but increased by HLA-DR8 (Garcı´a-Romero et al., 2012). In addition, a 44C/G polymorphism in thedefensin beta (DEFB) 1 gene is correlated with increased fungal carriage (Jurevicet al., 2003).
Human beta-defensins (hBDs) are a major subfamily of natural antimicrobial peptides with a net positive charge encoded by the DEFBgene cluster on chromosome 8p23.1. One member, hBD-2, is expressed by several epithelial cell types with well-documented microbicidal activities in innate and adaptive immunity (Yanget al., 1999; Semple and Dorin, 2012), including chemoattraction of memory T cells and immature dendritic cells, as well as proinflammatory effects on immune cells and human keratinocytes. The expression of hBD-2 (encoded by DEFB4) is regulated by interactions between microbial factors, cytokines, cell differentiation status and type, and genetic variation among individuals (Groth et al., 2010; Weinberget al., 2012). TheDEFBlocus is subject to extensive variation in copy number (CN), which can range from 2 to 12 (Harder et al., 2004; Hollox, 2010), and a strong correlation exists between the DEFB4 CN and mRNA level (Groth et al., 2010). A high CN underlies susceptibility to some inflammatory disorders such as psoriasis (Holloxet al., 2008), systemic lupus erythematosus,
1Department of Dermatology, Jena University Hospital, Jena, Germany; 2Department of Conservative Dentistry, Jena University Hospital, Jena, Germany; 3Jena School for Microbial Communication, Friedrich Schiller University of Jena,
Jena, Germany and4Private Dental Practices, Orland Park, Illinois, USA Correspondence: Sameh W. Jaradat, Department of Dermatology, Jena University Hospital, Erfurter Strasse 35, 07743 Jena, Germany. E-mail: [email protected]
Received 23 April 2014; revised 21 July 2014; accepted 12 August 2014; accepted article preview online 1 September 2014; published online 25 September 2014
Abbreviations: CN, copy number; DEFB, defensin beta genes; DEFB4, defensin beta4 gene; hBD-2, human beta-defensin-2; hBDs, human beta-defensins; OR, odds ratio; T. rubrum, Trichophyton rubrum
and anti-neutrophil cytoplasmic antibody–associated small vasculitis (Zhou et al., 2012), whereas a lower CN is a risk factor for severe chronic periodontitis (Jaradatet al., 2013).
IL-22 is a glycoprotein belonging to the IL-10 cytokine family (Dumoutier et al., 2000) whose major function is to protect skin keratinocytes against microbes by enhancing the expression of antimicrobial peptides, including hBD-2, as well as metalloproteinases that eliminate invading pathogens (Wolk et al., 2006). IL-22 synergizes with other cytokines such as tumor necrosis factor-ato induce an innate immune response in primary human keratinocytes against Candida albicans infection (Eyerich et al., 2009). Several T-cell populations and innate lymphoid cells in humans express IL-22 (Sonnenberg et al., 2010), which regulates cells at outer-body barriers via a transmembrane receptor complex that is restricted to nonhematopoietic cells such as those of the skin and tissues of the digestive and respiratory systems (Eyerich et al., 2010). IL-22 has been implicated in a variety of human diseases, such as inflammatory skin disorders and bowel disease, as well as chronic infections (Nenoffet al., 2014).
Although it has been suggested that host-specific genetic factors predispose individuals to dermatophytosis, the role of DEFB CN variation in the occurrence of infection remains unknown. To investigate the possibility that DEFB4 CN is associated withT. rubrum–mediated dermatophytosis,DEFB4 CN variation, as well as serum IL-22 and hBD-2 levels, was examined in unrelated Caucasian dermatophytosis patients and healthy control subjects.
RESULTS
Demographic and clinical data of participants
Characteristics of the study population are shown in Table 1. Median ages and sex distribution for the combined group of
patients and healthy control subjects were similar, and there were no significant differences in these variables between the tinea corporis, tinea pedis, tinea unguium, and lesions at multiple sites subgroups and controls. However, age and sex distributions were associated with the presence of tinea cruris (Po0.001).
Dermatophytosis patients have lowerDEFB4CN
DEFB4 CN varied between 2 and 11 per diploid genome (mean±SD, 4.53±1.25; median, 4; range, 2–11) in the
control group. In the combined group of patients, CN was between 2 and 10 (mean±SD, 3.9±1.27; median, 4; range,
2–10) (Supplementary Table S2 online). The meanDEFB4CN in the combined group of patients was lower than that in the control group (Po0.0001; Bonferroni-corrected). Similarly, the distribution of DEFB4 CN differed significantly between the combined group of patients and the controls (Po0.0001; Bonferroni-corrected), and there was a shift toward lower CN (p4 copies) in patients (Po0.0001 (Bonferroni-corrected); odds ratio (OR)¼3.11, 95% confidence interval¼1.99– 4.83) (Figure 1a). A similar trend was observed in the tinea corporis (P¼0.0018 (Bonferroni-corrected); OR¼3.25, 95% confidence interval¼1.73–6.10), tinea cruris (P¼0.0361 (Bonferroni-corrected); OR¼2.57, 95% confidence interval¼1.42–5.03), tinea pedis (P¼0.0219 (Bonferroni-cor-rected); OR¼2.67, 95% confidence interval¼1.42–5.03), and tinea unguium subgroups (P¼0.0016 (Bonferroni-cor-rected); OR¼3.22, 95% confidence interval¼1.74–5.96), as well as in the subgroup comprising patients with lesions at multiple sites (P¼0.0029 (Bonferroni-corrected); OR¼3.79, 95% confidence interval¼1.83–7.83) (Figure 1b–f), whereas no significant differences were detected between patient subgroups.
Table 1. Demographic and clinical data for the dermatophytosis study population
Clinical presentation
Variables
Controls (n¼195)
Combined group of patients (n¼247) Tinea corporis (n¼59) Tinea cruris (n¼24) Tinea pedis (n¼61) Tinea unguium (n¼64) Multiple sites1 (n¼39) Age2 44 (19–75) 42 (18–75) 39 (19–75) 27.5 (18–45) 42 (21–75) 44 (19–75) 39 (18–69) Pvalue3 0.158 0.241 o0.001 0.364 0.480 0.165 Sex4 Male 103 (52.8)5 126 (51.0)5 24 (40.7)6 21 (87.5)6 34 (55.7)6 31 (48.4)6 16 (41.0)6 Female 92 (47.2)5 121 (49.0)5 35 (59.3)6 3 (12.5)6 27 (44.3)6 33 (51.6)6 23 (59.0)6 Pvalue7 0.424 0.380 o0.001 0.243 0.860 0.482
1Lesions at multiple sites, such as tinea pedis and unguium, tinea pedis and manus/manuum, or tinea manus/manuum with infected fingernails. 2Age (in years) is given as the median and range of (minimum – maximum) values.
3Mann–Whitney test (each subgroup of patients versus controls). 4Biologically defined.
5Sex data are given asn(%); i.e., the number of male (n¼103) or female (n¼92) controls divided by the total number of controls (n¼195) or the number of
male (n¼126) or female (n¼121) cases divided by the total number of cases (n¼247).
6Sex data are given asn(%); i.e., the percentage of men or women in each patient subgroup divided by the total number in each subgroup (e.g., in the tinea
corporis subgroup, number of men (n¼24) or women (n¼35) divided by the total number in the subgroup (n¼59)).
Serum hBD-2 level is elevated in dermatophytosis patients
The serum hBD-2 level was higher in the combined group of patients than in the control group (median, 118.16 pg ml1; range, 59.51–195.62 vs. median, 104.49 pg ml1; range, 53.70–179.68; Po0.0001; Bonferroni-corrected) (Figure 2a) and was also higher in patients with lesions at multiple sites (median, 128.78 pg ml1; range, 96.33–195.62), tinea corporis (median, 123.18 pg ml1; range, 94.25–171.06), tinea pedis (median, 118.04 pg ml1; range, 59.51–156.61), or tinea unguium (median, 115.08 pg ml1; range, 74.58–166.21) than in controls (Po0.0101; Bonferroni-corrected). However, there was no significant difference in the hBD-2 level of tinea cruris patients (median, 105.69 pg ml1; range, 63.50–148.35) and
control subjects (P¼0.453). Among patient subgroups, the hBD-2 level was highest in patients with lesions at multiple sites, followed by the tinea corporis, tinea pedis, tinea unguium, and tinea cruris groups (P¼0.0002; Bonferroni-corrected).
Serum IL-22 level is elevated in dermatophytosis patients
IL-22 was detectable (42.6 pg ml1) in 191 of 247 (77.3%) patients and in 101 of 195 (51.8%) controls (Po0.001) (Figure 2b). The median serum IL-22 level in the combined group of patients was about 3-fold higher than that in the control group (median, 46.45 pg ml1; range, 2.60–71.18 vs.
median, 16.05 pg ml1; range, 2.60–61.03;Po0.0001;
(Bon-ferroni-corrected)), and IL-22 levels in the tinea corporis
F requency (%) F requency (%) Controls Controls Tinea cruris Controls Controls Tinea pedis Tinea corporis Controls Tinea ungulum Combined group of patients
Controls
Tinea lesions at multiple sites
20.6 13.9 4.6 34.9 30.4 25.1 OR = 3.11, CI = 1.99–4.93 P < 0.0001 41.2 34.9 30.4 28.3 46.6 18.5 40 50 30 20 10 <4 4 >4 13.8 12.5 13.9 4.6 25.0 34.9 20.8 25.1 20.8 4.6 18.8 13.9 23.4 29.7 25.1 34.9 13.3 4.2 8.2 16.7 13.3 6.5 8.2 8.1 20.6 27.1 13.9 4.6 16.4 13.9 21.3 34.9 36.1 25.1 9.8 13.3 13.1 8.2 8.2 7.7 2.6 13.3 15.4 25.1 28.2 34.9 20.5 13.9 3.3 15.3 34.9 30.5 25.1 15.3 13.3 3.3 8.2 8.5 4.6 30 40 35 20 25 10 15 5 DEFB4 CN variation DEFB4 CN variation F requency (%) OR = 3.25, CI = 1.73–6.10 P < 0.0001 42.4 34.930.5 27.1 46.6 18.5 40 50 30 20 10 <4 4 >4 DEFB4 CN variation F requency (%) OR = 2.67, CI = 1.42–5.03 P = 0.0219 37.7 34.936.1 26.2 46.6 18.5 40 50 30 20 10 <4 4 >4 DEFB4 CN variation F requency (%) OR = 2.57, CI = 1.73–6.10 P = 0.0361 37.5 34.9 20.8 41.7 46.6 18.5 40 50 30 20 10 <4 4 >4 DEFB4 CN variation 12.4 13.3 6.3 8.2 9.4 F requency (%) OR = 3.22, CI = 1.74–5.96 P = 0.0016 42.2 34.9 29.7 28.1 46.6 18.5 40 50 30 20 10 <4 4 >4 DEFB4 CN variation F requency (%) OR = 3.79, CI = 1.83–7.86 P = 0.0029 46.1 34.9 28.2 25.6 46.6 18.5 40 50 30 20 10 <4 4 >4 DEFB4 CN variation 2 3 4 5 6 .7 DEFB4 CN variation 2 3 4 5 6 .7 DEFB4 CN variation 2 3 4 5 6 .7 DEFB4 CN variation 2 3 4 5 6 .7 DEFB4 CN variation 2 3 4 5 6 .7 DEFB4 CN variation 2 3 4 5 6 .7 F requency (%) F requency (%) 30 40 35 20 25 10 15 5 30 40 35 20 25 10 15 5 F requency (%) 30 40 35 20 25 10 15 5 F requency (%) 30 40 35 20 25 10 15 5 F requency (%) 30 40 35 20 25 10 15 5 4.6 25.6
Figure 1. Frequency distribution ofbeta defensin 2(DEFB4) copy number (CN) variation.Values are for controls (n¼195) and (a) the combined group of
dermatophytosis patients (n¼247), and patients with tinea (b) corporis (n¼59), (c) cruris (n¼24), (d) pedis (n¼61), and (e) unguium (n¼64), as well as (f) with multiple lesions (n¼39). CN was rounded to the nearest integer assigned to bins ofo4, equal to 4, and44, and the final value was one that was the same in two assays. Odds ratios (ORs) and confidence intervals (CIs) were calculated for cases with44 CN versusPcases withp4 CN.Po0.05 was significant (Bonferroni-correctedPvalues).
(median, 46.45 pg ml1; range, 2.60–69.59), tinea cruris (median, 48.55 pg ml1; range, 2.60–66.25), tinea pedis (median, 47.30 pg ml1; range, 2.60–65.73), and tinea unguium (median, 36.22 pg ml1; range, 2.60–71.18) sub-groups, as well as in the subgroup comprising patients with lesions at multiple sites (median, 54.07 pg ml1; range, 2.60– 60.54), were higher than those in controls (Po0.0024; Bonferroni-corrected), whereas no significant differences were found among the patient subgroups (P¼0.603).
DEFB4CN is negatively correlated with serum IL-22 but not with
hBD-2 levels in dermatophytosis patients
Serum hBD-2 level was positively correlated with DEFB4 CN in healthy controls (r¼0.370; 95% confidence interval for r (Fisher’s z transformed)¼0.530 to 0.246; Po0.0001 (Bonferroni-corrected)); however, this was not observed in the combined group of patients nor in any of the patient subgroups (Figure 3a).
There was a negative correlation between DEFB4 CN and IL-22 level in the combined group of patients (r¼ 0.480; 95% confidence interval for r (Fisher’s z transformed)¼
0.649 to 0.397; Po0.0001 (Bonferroni-corrected)), and in the tinea corporis (r¼ 0.392; 95% confidence interval for r (Fisher’s z transformed)¼ 0.680 to 0.148; P¼0.024 (Bonferroni-corrected)), tinea cruris (r¼ 0.528; 95% confi-dence interval for r (Fisher’s z transformed)¼ 1.00 to 0.174; P¼0.0386 (Bonferroni-corrected)), tinea pedis (r¼ 0.517; 95% confidence interval for r (Fisher’s z transformed)¼ 0.829 to 0.315; P¼0.0022 (Bonferroni-corrected)), and tinea unguium (r¼ 0.584; 95% confidence interval for r (Fisher’s z transformed)¼ 0.920 to 0.418; Po0.0001 (Bonferroni-corrected)) subgroups, as well as in the subgroup comprising patients with multiple lesions (r¼ 0.483; 95% confidence interval for r (Fisher’s z transformed)¼ 0.853 to 0.201;P¼0.0342 (Bonferroni-corrected)) (Figure 3b). When the combined group of patients was divided into two categories of low (p4) and high (44) CN, the serum IL-22 level for low CN was higher than in the high CN category (median, 53.04 pg ml1; range, 2.6071.18 vs.
median, 25.21 pg ml1; range, 2.6068.66; Po0.0001 (Bonferroni-corrected)). There was no significant correlation between DEFB4 CN and serum IL-22 level in the control group (r¼ 0.009; 95% confidence interval for r (Fisher’s z transformed)¼ 0.150 to 0.132;P¼0.263).
Serum IL-22 and hBD-2 levels are positively correlated in dermatophytosis patients
IL-22 and hBD-2 levels were positively correlated in the combined group of patients (r¼0.552; 95% confidence interval for r (Fisher’s z transformed)¼0.495 to 0.747; Po0.0001 (Bonferroni-corrected)), as well as in the tinea corporis (r¼0.484; 95% confidence interval forr(Fisher’s z transformed)¼0.266 to 0.790; P¼0.001 (Bonferroni-cor-rected)), tinea pedis (r¼0.614; 95% confidence interval forr (Fisher’s z transformed)¼0.458 to 0.972; Po 0.0001(Bonfer-roni-corrected)), tinea unguium (r¼0.640; 95% confidence interval for r (Fisher’s z transformed)¼0.490 to 0.992; Po0.0001 (Bonferroni-corrected)), and patients with multiple lesions (r¼0.611; 95% confidence interval for r (Fisher’s z transformed)¼0.384 to 1.00; Po0.0001 (Bonferroni-cor-rected)) (Figure 3c), but not in the tinea cruris subgroup (r¼0.371; 95% confidence interval for r (Fisher’s z transformed)¼ 0.038 to 0.818;P¼0.073) or in the controls (r¼0.0692; 95% confidence interval for r (Fisher’s z transformed)¼ 0.073 to 0.211;P¼0.091).
DISCUSSION
The current study investigated the association between the occurrence of superficial dermatophytosis resulting from T. rubruminfection andDEFB4CN variation and serum levels of hBD-2 and IL-22. TheDEFB4CN distribution observed here in controls (Figure 1) was consistent with earlier findings in European populations (Holloxet al., 2008). The CN was lower in dermatophytosis patients regardless of the clinical presentation, which has not been previously reported, but it is consistent with a general role of human genetic variation in antifungal immunity (Zaitzet al., 1996; Asz-Sigallet al., 2010; Garcı´a-Romeroet al., 2012). The high CN and elevated hBD-2
80 60 40 20 30 50 70 75 65 55 45 35 25 15 5 10 P < 0.0001 P < 0.0001 P = 0.453 P = 0.0101 P < 0.0001 P < 0.0001 P < 0.0001 P < 0.0001 P < 0.0001P < 0.0001P < 0.0001P < 0.0001 200 180 160 140 120 100 80 60 40 20 Controls Tinea unguium Tinea pedis Tinea crur is Tinea cor por is Combinedpatients
Multiplesites Controls Tinea unguium Tinea pedis Tinea crur is Tinea cor por is Combinedpatients Multiplesites hBD-2 ser um concentr ation (pg ml –1 ) IL-22 ser um concentr ation (pg ml –1 )
Figure 2. Serum levels of human beta defensin (hBD)-2 and interleukin (IL)-22 in dermatophytosis patients (n¼247) and healthy controls (n¼195).(a) Human
beta defensin (hBD)-2; (b) interleukin (IL)-22. Patient subgroups are represented by different colors. The medians are indicated by horizontal bars. Comparisons between patient subgroups and the control group were performed with the Mann–Whitney test, and comparisons between subgroups were performed with the Kruskal–Wallis test.Po0.05 was significant (Bonferroni-correctedPvalues).
expression in psoriasis patients may also explain the relatively low prevalence of bacterial and viral infections among those patients (Jansenet al., 2009).
Except for those in the tinea cruris subgroup, serum hBD-2 level was higher in all patients than in controls (Figure 2). Unexpectedly, there was no correlation betweenDEFB4 CN variation and serum hBD-2 levels in patients, although a positive correlation was found in controls (Figure 3). These findings suggest that in the absence of infection, genetic composition is the main factor controlling hBD-2 production, as evidenced by the finding that CN variation accounts for 50% of the total variation in DEFB4 mRNA expression in healthy subjects (Hollox et al., 2003). Given that hBD-2 peptide is an inducible rather than a constitutively expressed peptide (Bissellet al., 2004), it is possible that the higher level in patients is owing to the upregulation of expression by T. rubrum rather than a variation in CN; i.e., regulation of DEFB4 is largely governed by T. rubrum infection. These
results are in agreement with recent studies showing increased hBD-2 expression in the superficial epidermal layers of patients with tinea corporis (Brasch et al., 2014) or tinea pedis (Kawaiet al., 2006). In contrast, hBD-2 expression was less evident in the stratum corneum under a high fungal hyphae load, indicating that hBD-2 expression was suppressed by infection (Jensen et al., 2007).
Another possible explanation for the high hBD-2 level in patients is thatT. rubruminfection introduces immunomodu-latory factors—mainly cytokines—into the infected area and that these factors induce hBD-2 production (Campos et al., 2006; Pereira et al., 2013). Indeed, the level of IL-22
cytokine—which enhances antimicrobial peptide
production—was positively correlated with that of hBD-2 in patients (Figure 3). Moreover, IL-22 was detected in a greater number of patients and the serum level was about 3-fold higher than that in controls (Figure 2). In CD3/28-stimulated T cells, hBD-2 increases IL-22 production, which, in turn,
200 180 160 140 120 100 80 60 40 20 200 180 160 140 120 100 80 60 40 20 10 40 8010 40 80 10 40 80 10 40 80 10 40 80 10 40 8010 40 80 80 60 40 20 30 50 70 10 2 3 4 5 6.7 2 3 4 5 6.7 2 3 4 5 6.7 2 3 4 5 6.7 2 3 4 5 6.7 DEFB4 CN variation DEFB4 CN variation 2 3 4 5 6.7 Controls Tinea unguium Tinea pedis Tinea cruris Tinea corporis Combined patients Multiple sites 2 3 4 5 6.7 Controls 2 3 4 5 6.7 2 3 4 5 6.7 2 3 4 5 6.7 2 3 4 5 6.7 2 3 4 5 6.7 Tinea unguium Tinea pedis Tinea cruris Tinea corporis Combined patients 2 3 4 5 6.7 Multiple sites Controls Tinea unguium Tinea pedis Tinea cruris Tinea corporis Combined patients Multiple sites 2 3 4 5 6.7 r=0.370 P<0.0001 r=–0.049 P=0.719 r=0.002 P=0.818 r=–0.068 P=0.799 r=–0.018 P=0.904 r=–0.023 P=0.711 r=–0.080 P=0.741 r=–0.009 P=0.263 r=–0.480 P<0.001 r=–0.392 P=0.024 r=–0.528 P=0.038 r=–0.517 P=0.002 r=–0.584 P<0.001 r=–0.4831 P=0.0342 r=0.0692 P=0.091 r=0.552 P<0.0001 r=0.484 P=0.001 r=0.371 P=0.073 r=0.614 P<0.0001 r=0.640 P<0.0001 r=0.699 P<0.0001 hBD-2 ser um le v el (pg ml –1 ) hBD-2 ser um le v el (pg ml –1 ) IL-22 ser um le v el (pg ml –1 )
IL-22 serum level (pg ml–1)
Figure 3. Correlation betweenbeta defensin 2(DEFB4) copy number (CN) variation and serum human beta defensin (hBD)-2 and interleukin (IL)-22 levels.
Dermatophytosis patient subgroups are represented by different colors. (a)DEFB4CN was positively correlated with hBD-2 level in controls (n¼195) but not in patients (n¼247), and (b) negatively correlated with IL-22 level in patients but not in controls. (c) Serum hBD-2 and IL-22 levels were positively correlated in patients but not in controls. Spearman’s rank correlation test was used to assess correlation.Po0.05 was considered significant (Bonferroni-correctedPvalues).
induces hBD-2 expression through the Jak signal transducer and activator of transcription (Jak-STAT) signaling pathway (Aujlaet al., 2008; Kanda and Watanabe, 2012). Furthermore, IL-22 activates STAT3 in keratinocytes, whereas loss of STAT3 expression abrogated the inductive effect of IL-22 on genes that regulate keratinocyte differentiation (Wolket al., 2009). It was suggested that STAT3 expression enhances the sensitivity of keratinocytes to subsequent IL-22 exposure, thus providing regulatory control through positive feedback.
IL-22 expression is reported to be induced by NF-kB, CCAAT/enhancer-binding protein (C/EBP), or specificity pro-tein 1 (Sp1); hBD-2-induced extracellular signal–regulated kinase stimulates IL-22 production by activating Sp1 and C/EBP expression and/or inducing their phosphorylation (Kanda and Watanabe, 2012). It was therefore speculated that altered serum IL-22 level is associated with DEFB4 CN variation, which was examined in this study. Interestingly, IL-22 level was higher in patients with low CN (p4) compared with those with high (44) CN, whereas no relationship was found in controls. One interpretation of these results is that IL-22 has a protective role duringT. rubruminfection, such that the level of the antimicrobial hBD-2 is upregulated in patients who have low DEFB4 CN and in whom the hBD-2 level is insufficient for an immune response againstT. rubrum, which
increases their susceptibility to infection. This is supported by a study showing that hBD-2 inhibits T. rubrumgrowth in vitro (Fritzet al., 2012). Consequently,T. rubrummay induce IL-22 production either directly or indirectly by stimulating other factors known to enhance IL-22, such as tumor necrosis factor-a and IL-1b (Sonnenberg et al., 2010), leading to the upregulation of DEFB4 gene expression and increased production of hBD-2. This could promote clearance of the infection via antimicrobial and immunomodulatory activities, and enhance IL-22 level through reciprocal regulation, thereby eliminating the infection and restoring homeostasis. Indeed, IL-22 has been shown to contribute to antifungal resistance at epithelial surfaces by increasing the expression of other antimicrobial genes that potentiate the function of IL-22 in a positive feedback loop (Sabat et al., 2014). Putative STAT3-binding sites were identified in an analysis of a 1000-base-pair region of the DEFB4promoter (Wolket al., 2004), suggesting that it is potentially regulated by IL-22 and thatDEFB4CN has a significant role in modulating expression. That is, a higher CN would provide more STAT3-responsive promoters that are targeted by IL-22, thereby leading to higher hBD-2 peptide production. Specific transcriptional repressors may also participate in cytokine-genomic responses and suppress hBD-2 synthesis by blocking DEFB4 transcription. To
1,273 Participants were screened for eligibility (2009–2013)
Included
639 Participants met the inclusion criteria
Participants were distributed according to physicians’ diagnosis
425 Patients suspected to have dermatophytosis
35 Patients were further excluded due to poor DNA quality and difficulty in DEFB4
CN variations, hBD-2 and IL-22 analysis 247 Patients were subcategorized according
to the clinical presentation
61 Tinea pedis 39 Lesions at multiple sites 64 Tinea unguium
24 Tinea cruris 59 Tinea corporis
Microbiological tests (T. rubrum) 282 Patients confirmed with
T. rubrum infection 214 Healthy controls
19 Were further excluded due to poor DNA quality
and difficulty in DEFB4
CN variations, hBD-2 and IL-22 analysis
195 Healthy controls
Unrelated German Caucasians .18 years. They had no history or current manifestation of systemic diseases such as diabetes mellitus; no atopic or skin disorders other than the suspected fungal skin infection; no HIV infection; were not pregnant and had not used antibiot-ics or immunosuppressants within the previous 3 months; had no history of or were not currently undergoing chemotherapy or treatment with antifungal medication; had no trauma at the site of infection; were non-smokers; and had good body hygiene, with feet being washed on a daily basis.
investigate this possibility, it would be of interest to examine the expression of multiprotein regulatory complexes at promoter-proximal genomic response elements in individuals who have high DEFB4 CN and low serum IL-22 levels; the observed negative correlation between them raises several questions regarding the role ofDEFB4CN in controlling gene expression and systemic levels of other cytokines.
The negative association between DEFB4 CN and derma-tophytosis was not observed in subgroups withX7 CN, with
the exception of the tinea pedis subgroup (Figure 1). One explanation for this is that although higher CN can confer immunity to dermatophytosis and, therefore, is a selective advantage, genes with CN variation contain higher-than-average frequencies of nonsynonymous mutations (Nguyen et al., 2006). Thus, individuals with CNX7 may be
predisposed to mutations that could negatively affect the activity of the DEFB4 gene against dermatophytosis. However, given that CNX7 did protect against tinea pedis,
it would be of interest to investigate whether sites of infection can be distinguished on a molecular basis, which would lead to improved classification of dermatophytosis and therapeutic management.
In this study, patients were categorized according to the clinical presentation of dermatophytosis, and there were no significant differences between these groups in terms of CN distribution and systemic IL-22 level. Every effort was made to match patients and controls by age and sex; however, the majority of tinea cruris patients were relatively young male individuals, consistent with the demographic data on the prevalence of this form of dermatophytosis (Degreef, 2008). Nonetheless, tinea cruris patients had a lower level of systemic hBD-2 than those with other forms of tinea (Figure 2). This may be because hBD-2 is variably expressed in different regions of the human skin, with the scalp and plantar surfaces having the highest expression, and the perineum, abdomen, and chest expressing a low level of the peptide (Ali et al., 2001). Sex differences in the tinea cruris subgroup, combined group of patients, and controls was unrelated to serum hBD-2 levels (Supplementary Table S3 online), consistent with another report showing that hBD-2 expression is not sex-related (Wittersheim et al., 2013). Furthermore, age-related changes in hBD-2 expression have been reported, with higher expression being more frequently detected in older individuals (Wittersheim et al., 2013), indicating that the lower hBD-2 level in the tinea cruris subgroup may have been a result of the younger mean age of these patients.
In conclusion, an association was established between DEFB4 CN variation and the occurrence of T. rubrum–induced superficial dermatophytosis. The results suggest that low DEFB4 CN is a risk factor for dermatophy-tosis, with elevated IL-22 levels implicated in its pathogenesis. Dermatophytosis candidate genes other than DEFB4, such as Fc receptor gamma, which is used by the pattern recognition receptor dectin-2 to induce innate immune responses against T. rubrum (Sato et al., 2006; Nossent et al., 2012), have also shown variable CN, and may simi-larly contribute to the pathogenesis of dermatophyte infection. A causal relationship between these factors should be verified
through prospective cohort studies, which should also be undertaken in patient populations from different geogra-phical regions. Identifying the genetic causes underlying dermatophytosis will inform preventive treatments for persi-stent forms such as tinea unguium that undermine the quality of life of patients. The data for serum IL-22 levels and the known contribution of IL-22 to protective immunity against yeast infection raise the possibility of using IL-22 as a therapeutic agent for dermatophytosis, which merits further exploration.
MATERIALS AND METHODS
Study population
This cross-sectional study was approved by the Ethical Committee of Jena University (no. 1940-01/07) and the protocol was in accordance with the Declaration of Helsinki Principles, with all individuals providing signed, informed consent before participation. Subjects were enrolled at the Department of Dermatology from November 2009 to January 2014 and were interviewed and examined by two experienced dermatologists. All efforts were made to reduce pheno-typic heterogeneity by applying stringent criteria to the patient selection process. All participants were unrelated German Cauca-siansX18 years of age. They had no history or current manifestation
of systemic diseases such as diabetes mellitus; no atopic or skin disorders other than the suspected fungal skin infection (cases with T. rubruminfection); no HIV infection; were not pregnant and had not used antibiotics or immunosuppressants within the previous 3 months; had no history of or were not currently undergoing chemotherapy or radiotherapy or treatment with antifungal medica-tion; had no trauma at the site of infection (for cases); were non-smokers; and had good body hygiene, with feet being washed on a daily basis.
Patients were diagnosed on the basis of the presence of clinical signs of dermatophytosis and on mycologic evidence defined as a fungal culture positive forT. rubrum. Clinical signs of dermatophy-tosis were confirmed by direct microscopic examination of samples with 10–20% potassium hydroxide and by culture on Sabouraud’s dextrose agar with chloramphenicol (0.05%) and cycloheximide (0.5%) or Sabouraud’s dextrose agar with chloramphenicol but without cycloheximide. Diagnosis was confirmed only when both clinical and laboratory criteria were met. Microbiological tests were performed at Fungus Laboratory, Department of Dermatology, Jena University Hospital. Patients were divided into subgroups according to clinical presentation (Wagner and Sohnle, 1995): tinea corporis (ringworm on the trunk); tinea cruris (ringworm of the groin); tinea pedis (athlete’s foot); tinea unguium (onychomycosis and nail infections); and patients with lesions at multiple sites, such as tinea pedis and unguium, tinea pedis and manus/manuum, or tinea manus/ manuum with infected fingernails. Pooling subgroups together yielded the combined group of patients. Controls and cases were frequency matched for age and sex (Supplementary Table S1 online). Controls were recruited after cases were enrolled, as the distribution of matching factors—i.e., age and sex—was not known in advance. Controls were examined physically to exclude clinical signs of dermatophytosis (Figure 4). A total of 442 unrelated German Caucasian participants comprising 247 dermatophytosis patients (121 female and 126 male patients) and 195 controls (92 female and 103 male patients) were included in the study.
Diplotyping ofDEFB4CN variation
Genomic DNA isolation, quantification, and analysis ofDEFB4CN variation (GenBank: AF040153.1) were carried out as previously described (Abe et al., 2013; Jaradat et al., 2013). Briefly, genomic DNA was isolated from peripheral blood using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany). A Nanodrop-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE) was used to determine DNA concentration and quality; samples with optical density 260/280 and 260/230 values ofB1.8 and 2.0–2.2, respectively, were used for analyses, which were performed by individuals who were blinded to the clinical information. CN variation was determined using duplex real-time PCR with TaqMan CN target and RNase P reference assays (assay ID: 4403328)—i.e., by relative quantification using the comparative CT (DDCT) method. PCR was performed in 96-well plates on an ABI 7500 system (Applied Biosystems, Foster City, CA). All assays were performed in quadruplicate and included calibrator and negative control samples. Reaction conditions were set at 951C for 10 minutes, followed by 40
cycles of 951C for 15 seconds and 601C for 1 minute.
Quantification of serum hBD-2 and IL-22 levels
Whole blood was collected at the time of enrollment between 7:00 hours and 12:00 hours, and the serum was separated and stored as aliquots at 801C until used for analysis. Commercially available
ELISA kits were used according to the manufacturers’ recommenda-tions to measure serum levels of hBD-2 (Beta Defensin-2 Kit with a detection limit of 7.8 pg ml1; Phoenix Europe GmbH, Karlsruhe,
Germany) and IL-22 (Human IL-22 Kit, with a detection limit of 2.7 pg ml1; R&D Systems, Minneapolis, MN). Assays were performed
in duplicate. Samples that showed a value below the minimum detectable dose for IL-22 were taken as 2.6 pg ml1. An hBD-2-positive control sample (B45–140 pg ml1) was included in each
hBD-2 quantification assay. Measurements were carried out blindly with respect to participants’ clinical characteristics and genetic data.
Statistical analysis
Statistical analyses were carried out using SPSS 22.0 for Windows (SPSS, Chicago, IL). A normal probability (Q-Q) plot and the Shapiro– Wilk test complementing the graphical analysis were used to deter-mine whether data were normally distributed. Categorical variables were compared using Pearson’sw2test. To determine the strength of the association between decreasedDEFB4CN and superficial derma-tophytosis, we re-distributed the CN data according to the median (o4, 4,44). The strength of associations was determined by calculat-ing the OR and 95% confidence interval. Continuous variables were compared using the nonparametric Mann–Whitney and Kruskal– Wallis tests (for two- and multiple-group comparisons, respectively). Correlations were analyzed by Spearman’s rank correlation (r). Correlations were defined as follows: 1.00X(r)X0.650 or1.00p(r) p0.650 (strong); 0.6504(r)X0.350 or 0.350X(r)40.650 (moderate); 0.3504(r)X0.200 or 0.200X(r)40.350 (weak); and 0.2004(r)40.200 (none). A confidence interval for (r) was established with Fisher’s z transformation. When dealing with multiple comparisons, thePvalues were corrected with Bonferroni’s correction by multiplying with the number of analyses. AllPvalues were two-sided and statistical significance was defined asPo0.05. The power was estimated for different sample sizes by using the association between higherDEFBCN and the relative risk of psoriasis reported in
one study (Holloxet al., 2008), which was the first to include both Dutch and German participants. The effect sizes ranged from 0.25 to 0.4 for mean CN difference between cases or controls, whereas the standard deviation of CNs was between 1.1 and 1.3. Power curves were plotted for the unpaired ttest for power calculation, assuming equal variance in cases and controls and a significance level of 0.05 (Supplementary Figure 1 online). In the psoriasis study, the power estimate ranged between 0.58 (based on a true difference of 0.25 and standard deviation of 1.3) and 0.98 (based on a true difference of 0.4 and standard deviation of 1.1). Using the largest observed standard deviation in this study (1.27 in cases), the power was 0.76 for a true effect size of 0.25 and 0.94 for an effect size of 0.4.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ACKNOWLEDGMENTS
This study was supported by the Jena School for Microbial Communication and was funded by Deutsche Forschungsgemeinschaft (JSMC/TP31).
SUPPLEMENTARY MATERIAL
Supplementary material is linked to the online version of the paper at http:// www.nature.com/jid
REFERENCES
Abe S, Miura K, Kinoshita A et al. (2013) Copy number variation of the antimicrobial-gene, defensin beta 4, is associated with susceptibility to cervical cancer.J Hum Genet58:250–3
Ali RS, Falconer A, Ikrem Met al.(2001) Expression of the peptide antibiotics human beta defensin-1 and human beta defensin-2 in normal human skin.
J Invest Dermatol117:106–11
Asz-Sigall D, Lo´pez-Garcı´a L, Vega-Memije ME et al. (2010) HLA-DR6 association confers increased resistance to T. rubrum onychomycosis in Mexican Mestizos.Int J Dermatol49:1406–9
Aujla SJ, Chan YR, Zheng Met al.(2008) IL-22 mediates mucosal host defense against Gram-negative bacterial pneumonia.Nat Med14:275–81 Bissell J, Joly S, Johnson GK et al. (2004) Expression of beta-defensins in
gingival health and in periodontal disease.J Oral Pathol Med33:278–85 Brasch J, Mo¨rig A, Neumann B et al.(2014) Expression of antimicrobial peptides and toll-like receptors is increased in tinea and pityriasis versicolor.Mycoses57:147–52
Campos MR, Russo M, Gomes Eet al.(2006) Stimulation, inhibition and death of macrophages infected withTrichophyton rubrum.Microbes Infect8:372–9 Degreef H (2008) Clinical forms of dermatophytosis (ringworm infection).
Mycopathologia166:257–65
Dumoutier L, Louahed J, Renauld JC (2000) Cloning and characterization of IL-10-related T cell-derived inducible factor (IL-TIF), a novel cytokine structurally related to IL-10 and inducible by IL-9.J Immunol164:1814–9 Esquenazi D1, Alviano CS, de Souza Wet al.(2004) The influence of surface carbohydrates during in vitro infection of mammalian cells by the dermatophyteTrichophyton rubrum.Res Microbiol155:144–53 Eyerich S, Eyerich K, Cavani Aet al.(2010) IL-17 and IL-22: siblings, not twins.
Trends Immunol31:354–61
Eyerich S, Eyerich K, Pennino Det al.(2009) Th22 cells represent a distinct human T cell subset involved in epidermal immunity and remodeling.J Clin Invest119:3573–85
Faergemann J, Correia O, Nowicki Ret al.(2005) Genetic predisposition– understanding underlying mechanisms of onychomycosis. J Eur Acad Dermatol Venereol19(Suppl 1):17–9
Fritz P, Beck-Jendroschek V, Brasch J (2012) Inhibition of dermatophytes by the antimicrobial peptides humanb-defensin-2, ribonuclease 7 and psoriasin.
Garcı´a-Romero MT, Granados J, Vega-Memije MEet al.(2012) Analysis of genetic polymorphism of the HLA-B and HLA-DR loci in patients with dermatophytic onychomycosis and in their first-degree relatives. Actas Dermosifiliogr103:59–62
Ghannoum MA, Mukherjee PK, Warshaw EMet al.(2013) Molecular analysis of dermatophytes suggests spread of infection among household mem-bers.Cutis9:237–45
Groth M, Wiegand C, Szafranski K et al. (2010) Both copy number and sequence variations affect expression of humanDEFB4. Genes Immun
11:458–66
Grumbt M, Monod M, Yamada T et al. (2013) Keratin degradation by dermatophytes relies on cysteine dioxygenase and a sulfite efflux pump.
J Invest Dermatol133:1550–5
Harder J, Meyer-Hoffert U, Wehkamp K et al. (2004) Differential gene induction of human beta-defensins (hBD-1, -2, -3, and -4) in keratinocytes is inhibited by retinoic acid.J Invest Dermatol123:522–9
Havlickova B, Czaika VA, Friedrich M (2008) Epidemiological trends in skin mycoses worldwide.Mycoses51(Suppl 4):2–15
Hollox EJ (2010) Beta-defensins and Crohn’s disease: confusion from counting copies.Am J Gastroenterol105:360–2
Hollox EJ, Armour JA, Barber JC (2003) Extensive normal copy number variation of a beta-defensin antimicrobial-gene cluster.Am J Hum Genet
73:591–600
Hollox EJ, Huffmeier U, Zeeuwen PLet al.(2008) Psoriasis is associated with increased beta-defensin genomic copy number.Nat Genet40:23–5 Jansen PA, Rodijk-Olthuis D, Hollox EJet al.(2009) Beta-defensin-2 protein is
a serum biomarker for disease activity in psoriasis and reaches biologi-cally relevant concentrations in lesional skin.PLoS One4:e4725 Jaradat SW, Hoder-Przyrembel C, Cubillos S et al. (2013) Beta-defensin-2
genomic copy number variation and chronic periodontitis.J Dent Res
92:1035–40
Jensen JM, Pfeiffer S, Akaki T et al. (2007) Barrier function, epidermal differentiation, and human beta- defensin 2 expression in tinea corporis.
J Invest Dermatol127:1720–7
Jurevic RJ, Bai M, Chadwick RBet al.(2003) Single-nucleotide polymorphisms (SNPs) in human beta-defensin 1: high-throughput SNP assays and association with Candida carriage in type I diabetics and nondiabetic controls.J Clin Microbiol41:90–6
Kanda N, Watanabe S (2012) Increased serum humanb-defensin-2 levels in atopic dermatitis: relationship to IL-22 and oncostatin M.Immunobiology
217:436–45
Kawai M, Yamazaki M, Tsuboi Ret al.(2006) Human beta-defensin-2, an antimicrobial peptide, is elevated in scales collected from tinea pedis patients.Int J Dermatol45:1389–90
Nenoff P, Kru¨ger C, Ginter-Hanselmayer Get al.(2014) Mycology - an update. Part 1: Dermatomycoses: Causative agents, epidemiology and pathogen-esis.J Dtsch Dermatol Ges12:188–210
Nguyen DQ, Webber C, Ponting CP (2006) Bias of selection on human copy-number variants.PLoS Genet2:e20
Nossent JC, Rischmueller M, Lester S (2012) Low copy number of the Fc-g
receptor 3B gene FCGR3B is a risk factor for primary Sjogren’s syndrome.
J Rheumatol39:2142–7
Pereira AL, Franco GC, Cortelli SC et al. (2013) Influence of periodontal status and periodontopathogens on levels of oral humanb-defensin-2 in saliva.J Periodontol84:1445–53
Sabat R, Ouyang W, Wolk K (2014) Therapeutic opportunities of the IL-22-IL-22R1 system.Nat Rev Drug Discov13:21–38
Sato K, Yang XL, Yudate T et al.(2006) Dectin-2 is a pattern recognition receptor for fungi that couples with the Fc receptor gamma chain to induce innate immune responses.J Biol Chem281:38854–66
Semple F, Dorin JR (2012)b-Defensins: multifunctional modulators of infec-tion, inflammation and more?J Innate Immun4:337–48
Sonnenberg GF, Fouser LA, Artis D (2010) Functional biology of the IL-22-IL-22R pathway in regulating immunity and inflammation at barrier surfaces.
Adv Immunol107:1–29
Wagner DK, Sohnle PG (1995) Cutaneous defenses against dermatophytes and yeasts.Clin Microbiol Rev8:317–35
Weinberg A, Jin G, Sieg Set al.(2012) The yin and yang of human Beta defensins in health and disease.Front Immunol3:294
Wittersheim M, Cordes J, Meyer-Hoffert Uet al.(2013) Differential expression andin vivosecretion of the antimicrobial peptides psoriasin (S100A7), RNase 7, human beta-defensin-2 and -3 in healthy human skin. Exp Dermatol22:364–6
Wolk K, Haugen HS, Xu Wet al.(2009) IL-22 and IL-20 are key mediators of the epidermal alterations in psoriasis while IL-17 and IFN-gamma are not.
J Mol Med (Berl)87:523–36
Wolk K, Kunz S, Witte Eet al.(2004) IL-22 increases the innate immunity of tissues.Immunity21:241–54
Wolk K, Witte E, Wallace Eet al.(2006) IL-22 regulates the expression of genes responsible for antimicrobial defense, cellular differentiation, and mobility in keratinocytes: a potential role in psoriasis.Eur J Immunol36:1309–23 Yang D, Chertov O, Bykovskaia SNet al.(1999) Beta-defensins: linking innate and adaptive immunity through dendritic and T cell CCR6. Science
286:525–8
Zaias N, Tosti A, Rebell Get al.(1996) Autosomal dominant pattern of distal subungual onychomycosis caused byTrichophyton rubrum.J Am Acad Dermatol34:302–4
Zaitz C, Campbell I, Moraes JRet al.(1996) HLA-associated susceptibility to chronic onychomycosis in Brazilian Ashkenazic Jews. Int J Dermatol
35:681–2
Zhou XJ, Cheng FJ, Lv JCet al.(2012) Higher DEFB4 genomic copy number in SLE and ANCA-associated small vasculitis. Rheumatology (Oxford)51: 992–5
