Prevention and/or treatment of inflammatory skin disease

Abstract

Disclosed is a substance which down-regulates the activity of a MAST gene, or the activity of a transcription or translation product of a MAST gene, for use in the prevent and/or treatment of an inflammatory skin condition in a mammalian subject.

Claims

1. A method of treating an inflammatory skin condition in a mammalian subject that has a bacterial skin infection, the method comprising administering to the mammalian subject an effective dose of a short interfering RNA molecule that down-regulates the expression of a MAST4 gene.

2. The method of claim 1, wherein the skin condition comprises dandruff.

3. The method of claim 1, wherein the short interfering RNA molecule is essentially identical in sequence to a portion of the MAST4 gene transcript.

4. The method of claim 3, wherein the short interfering RNA molecule comprises a 20 to 30 nucleotide sequence identical to a portion of the MAST4 gene transcript.

5. The method of claim 1, wherein a formulation comprising the effective dose of the short interfering RNA molecule in admixture with a dermatologically acceptable bulking agent, diluent, carrier or excipient is administered to the mammalian subject, and wherein the mammalian subject is a human subject.

6. The method of claim 5, wherein the formulation comprises a skin penetration enhancer.

7. The method of claim 5, wherein the formulation is in the form of liquid, cream, gel, lotion, paste or patch.

8. The method of claim 5, wherein the formulation is in the form of a shampoo and/or a conditioner.

9. The method of claim 1, wherein the inflammatory skin condition is selected from the group consisting of atopic dermatitis (AD), psoriasis (PSO), seborrheic dermatitis, dandruff, rosacea, and acne.

10. A method of preventing an inflammatory skin condition in a mammalian subject that has a bacterial skin infection, the method comprising administering to the mammalian subject an effective dose of a short interfering RNA molecule that down-regulates the expression of a MAST4 gene.

11. The method of claim 10, wherein the skin condition comprises dandruff.

12. The method of claim 10, wherein the short interfering RNA molecule is essentially identical in sequence to a portion of the MAST4 gene transcript.

13. The method of claim 10, wherein the short interfering RNA molecule comprises a 20 to 30 nucleotide sequence identical to a portion of the MAST4 gene transcript.

14. The method of claim 10, wherein a formulation comprising the effective dose of the short interfering RNA molecule in admixture with a dermatologically acceptable bulking agent, diluent, carrier or excipient is administered to the mammalian subject, and wherein the mammalian subject is a human subject.

15. The method of claim 14, wherein the formulation comprises a skin penetration enhancer.

16. The method of claim 14, wherein the formulation is in the form of liquid, cream, gel, lotion, paste or patch.

17. The method of claim 14, wherein the formulation is in the form of a shampoo and/or a conditioner.

18. The method of claim 10, wherein the inflammatory skin condition is selected from the group consisting of atopic dermatitis (AD), psoriasis (PSO), seborrheic dermatitis, dandruff, rosacea, and acne.

19. A method of treating a microbial-induced inflammatory skin condition in a mammalian subject in need thereof, the method comprising administering to the mammalian subject an effective dose of a substance that down-regulates the expression of a MAST4 gene, wherein (a) the substance acts at the nucleic acid level, down-regulating the transcription of the MAST4 gene and/or the translation of the MAST4 gene transcript; (b) the substance is a short interfering RNA molecule or a CRISPR-Cas component; and (c) the inflammatory skin condition is selected from the group consisting of atopic dermatitis (AD), psoriasis (PSO), seborrheic dermatitis, dandruff, rosacea, and acne.

Description

(1) The invention will now be further described by way of illustrative example and with reference to the accompanying drawings, in which:

(2) FIG. 1 shows SEQ ID NO: 1 setting forth the nucleotide sequence of part of the human MAST4 gene around a single nucleotide polymorphism identified by the inventors as associated with susceptibility to seborrheic dermatitis;

(3) FIG. 2 shows the general chemical structure of quinoxaline and derivatives thereof;

(4) FIG. 3 illustrates the structure of certain quinoxaline derivatives which are preferred for use in the present invention;

EXAMPLES

Example 1—Candidate Gene Association Study (CGAS) and Genome-Wide Association Study (GWAS)

(5) In this example, the inventors conducted a candidate gene association study (CGAS) to investigate whether genetic variants previously associated with AD and PSO are also associated with an increased risk for SD, and conducted the first genome-wide association study (GWAS) to identify novel genetic variants associated with SD.

(6) Methods

(7) Study Population.

(8) The Rotterdam Study (RS) is an ongoing prospective cohort study of chronic diseases, including skin diseases, in the elderly. A detailed description of the study can be found elsewhere (Hofman et al. The Rotterdam Study: 2016 objectives and design update. Eur J Epidemiol. 2015; 30(8):661-708). In brief, the Rotterdam Study consists of a major cohort (RS-I) and two extensions (RS-II and RS-III). RS-I started in 1990 and initially included 7,983 participants living in the Ommoord district in Rotterdam, The Netherlands. RS-II started in 2000 and includes 3,011 participants. RS-III is a further extension of the cohort, started in 2006, and includes 3,932 participants. In total, the Rotterdam Study consists of 14,926 subjects aged 45 years or over. The cohort consists predominantly (90%) of participants of North-European ancestry.

(9) The Rotterdam Study was approved by the Medical Ethics Committee of the Erasmus MC and by the Ministry of Health, Welfare and Sport of the Netherlands. All participants provided written informed consent to participate in the study and to obtain information from their treating physicians.

(10) Case Definition

(11) A total of 4,454 participants from the Rotterdam Study with available genotype data underwent a full body skin examination (FBSE) by trained physicians to assess whether a participant had SD. The clinical diagnosis was based on the presence of greasy scaling, erythema and a characteristic distribution of the scalp, face and/or chest. Participants who did not have any sign of active disease and without a history of ketoconazole use (≥3 prescriptions using pharmacy linkage data) were included as controls.

(12) Selection of Candidate Loci

(13) PubMed was searched for well-powered European GWAS that showed an association between single mucleotide polymorphisms (SNPs) and PSO or AD. The inventors selected SNPs with a significance level of <5.0×10.sup.−8. The reported (lead) SNP was used for the SNP-based association analysis and was used to select a locus region for a gene-based candidate gene association study.

(14) For lead SNPs located within a gene, a 30 kb region (plus 15 kb downstream and 15 kb upstream of the gene) was added to include regulatory elements close to the gene. If the lead SNP was intergenic and located more than 15 kb away from a gene, the inventors also added 15 kb downstream and 15 kb upstream from the base-pair position of the lead SNP. The NCBI Variation Viewer (http://www.ncbi.nlm.nih.gov/variation/view/) with build CRCh37 was used to retrieve the gene genomic coordinates. These coordinates were used to retrieve SNPs in the corresponding regions of the Rotterdam Study cohorts.

(15) Genotyping and Imputation

(16) DNA extraction from whole blood genotyping, imputation and quality control were carried out following standard protocols, as described earlier (Hofman et al. Eur J Epidemiol. 2015; 30(8):661-708; Verkouteren et al., J Invest Dermatol. 2015; 135(8):2135-8). The Illumina Infinium II HumanHap550 BeadChips were used to genotype the RS-I and RS-II cohorts while Illumina Human610-Quad BeadChips were used to genotype the RS-III cohort.

(17) Quality control on the single-nucleotide polymorphism (SNP) included removing SNPs with Hardy-Weinberg equilibrium deviations (p-value<5×10.sup.−6), genotype call rate <97%, gender mismatch and a high mean autosomal heterozygosity. First-degree relatives were removed and SNPs were filtered out if they had a minor allele frequency of less than 1% or an imputation quality (Rsq) of less than 0.3.

(18) Candidate Gene Study

(19) The candidate gene association studies (CGAS) were performed in two steps. First, a SNP-based association study of previously associated PSO and AD SNPs and second, a gene-based association analysis was conducted to screen for additional variants within the regions of interest.

(20) The SNP-based CGAS of SD cases and controls were performed using the imputed dosage data of each Rotterdam Study cohort using a logistic regression with an additive model. The output per SNP for each Rotterdam Study cohort was meta-analyzed using p-values derived from a likelihood ratio test as summary statistics (Willer et al., METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics. 2010; 26(17):2190-1; Aulchenko et al., ProbABEL package for genome-wide association analysis of imputed data. BMC Bioinformatics. 2010; 11:134).

(21) Best-guessed genotypes were used for the gene-based CGAS. The genotypes of the three Rotterdam Study cohorts were estimated from the imputed dosage data using the GCTA software with default parameters (Yang et al., Am J Hum Genet. 2011; 88(1):76-82).

(22) The gene-based logistic regression was conducted using the set-based test implemented in PLINK v.190 (paramters: p-value=0.05, maximum number of SNPs=15, r.sup.2=0.5, permutations=10000) (Purcell et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007; 81(3):559-75). Both SNP-based and gene-based associations were adjusted for age, sex and four principal components (PCs).

(23) Genome-Wide Association Study

(24) A GWAS of SD cases and controls was performed using the imputed dosage data of each Rotterdam Study cohort using a logistic regression with an additive model. The model was adjusted for age, sex and four PCs. The GWAS analyses were implemented in the ProbABEL package (Aulchenko et al., cited above). The inflation factor λ was estimated close to 1.0 and was not further considered.

(25) Next, the GWAS summary statistics per Rotterdam Study cohort were meta-analyzed using p-values derived from a likelihood ratio test. The meta-analysis was carried-out using METAL (Willer et al., cited previously). The threshold for the p-value of genome wide significance was set at a p-value≤5×10.sup.−8.

(26) Results

(27) Study Population

(28) 4,050 of the 4,454 participants were included for the CGAS and GWAS of SD. Of these, 609 (15%) were diagnosed during a FBSE by a research physician as having SD. Males were more likely to have SD than were females (60.4% of cases vs 41.8% of controls were males; p-value<0.001) and SD patients were older compared to those without SD (68.94 vs 67.97, p-value: 0.018). SD cases were more likely to have the diagnosis of psoriasis than controls (4.8% vs 3.0%; p=0.035).

(29) Candidate Genes Approach

(30) The CGAS did not yield any significant locus for SD after correcting for multiple testing.

(31) Genome Wide Association Study

(32) To discover new loci for SD, the inventors performed a GWAS on SD cases against controls and over 7 million loci and found two genome-wide significant associations (Table 1). The most significant SNP was r58331610 that mapped to an intronic region of the MAST4 gene at chromosome 5 (p-value: 1.75×10.sup.−8). The second hit was the rs16944244 SNP that mapped to an intergenic region at chromosome 17p12, between the genes PIRT and SHISA6 (p-value: 2.10×10.sup.−8). In addition, 68 SNPs in seven different loci were found to be significantly associated with SD with suggestive genome-wide associations (p-value<5×10.sup.−6) in this relatively small sample of SD patients. Several SNPs mapped to protein-coding genes including GRM3 (rs6978155, 5.24×10.sup.−7), KIAA1324L (rs73382367, 6.51×10.sup.−7) and SENP2 (rs13081203, 3.64×10.sup.−6), while others were intergenic. None of these loci have been previously associated with PSO or AD.

(33) TABLE-US-00001 TABLE 1 Gene SNP Chr Location Allele Freq Pvalue MAST4 rs58331610 5 66020457 a g 0.8615 .sup. 1.75 × 10.sup.−8* Intergenic rs16944244 17 11026693 c g 0.9796 .sup. 2.10 × 10.sup.−8* Intergenic rs78160483 17 11026297 t g 0.9818 8.38 × 10.sup.−8 Intergenic rs16944241 17 11026414 a g 0.9818 8.38 × 10.sup.−8 Intergenic rs6546997 2 75658780 t c 0.1894 9.53 × 10.sup.−8 Intergenic rs8075550 17 11027059 a g 0.9594 1.04 × 10.sup.−7 Intergenic rs11870758 17 11015902 a t 0.9722 1.21 × 10.sup.−7 MAST4 rs12188593 5 66030069 t c 0.1359 1.27 × 10.sup.−7 Intergenic rs182131544 17 11023964 a g 0.9788 1.32 × 10.sup.−7 Intergenic rs111448323 17 11022815 a g 0.9786 1.50 × 10.sup.−7 Intergenic rs4791451 17 11027570 c g 0.9808 1.54 × 10.sup.−7 Intergenic rs77699632 17 11020937 a t 0.0214 1.73 × 10.sup.−7 Intergenic rs111341744 17 11021220 a g 0.9784 1.79 × 10.sup.−7 Intergenic rs75205996 17 11021311 t c 0.9784 1.79 × 10.sup.−7 Intergenic rs58692658 17 11019423 t c 0.022 2.32 × 10.sup.−7 Intergenic rs77592872 17 11018841 a t 0.9781 2.34 × 10.sup.−7 Intergenic rs138424360 17 11018089 a g 0.0219 2.37 × 10.sup.−7 Intergenic rs80298958 17 11016351 c g 0.9784 3.39 × 10.sup.−7 Intergenic rs76730906 17 11019589 t g 0.9754 3.85 × 10.sup.−7 GRM3 rs6978155 7 86477894 a t 0.0543 5.24 × 10.sup.−7 GRM3 rs6957842 7 86477870 a t 0.9457 5.28 × 10.sup.−7 GRM3 7:86475369:I 7 86475369 d i 0.9459 5.63 × 10.sup.−7 GRM3 rs6960053 7 86473063 a g 0.0541 5.68 × 10.sup.−7 GRM3 rs6974507 7 86472032 t g 0.9459 5.75 × 10.sup.−7 GRM3 rs7801589 7 86470943 t c 0.9459 5.76 × 10.sup.−7 GRM3 rs6465087 7 86470488 t c 0.9459 5.78 × 10.sup.−7 GRM3 rs6954573 7 86472161 a t 0.9455 5.86 × 10.sup.−7 KIAA1324L rs73382367 7 86525065 a g 0.9436 6.51 × 10.sup.−7 KIAA1324L rs6465089 7 86564795 t c 0.0563 7.17 × 10.sup.−7 GRM3 7:86478749:I 7 86478749 d i 0.9453 7.44 × 10.sup.−7 GRM3 rs6947778 7 86476124 c g 0.0545 8.11 × 10.sup.−7 GRM3 rs6967992 7 86476311 t c 0.9455 8.11 × 10.sup.−7 GRM3 rs6955565 7 86472484 a g 0.0545 8.12 × 10.sup.−7 GRM3 rs6955452 7 86472584 c g 0.9455 8.15 × 10.sup.−7 GRM3 rs6955917 7 86472763 a g 0.0545 8.15 × 10.sup.−7 KIAA1324L rs6958716 7 86594383 t c 0.943 1.18 × 10.sup.−6 *statistically significant

(34) Discussion

(35) In this study the inventors did not find statistically significant associations between previously associated PSO/AD SNPs and SD. However in the GWAS for SD, two SNPs with genome-wide significant associations were observed. In addition, genome-wide suggestive associations were found in other genes that may be relevant to the SD pathogenesis that need to be confirmed in other cohorts.

(36) In the pilot GWAS for SD, several SNPs were significantly associated (p-value≤5.0×10.sup.−8) with SD, including variants located at the MAST4 gene. MAST4 belongs to the protein kinase (PK) group, which plays a critical role in intracellular signal transmission cascades. The precise function of MAST4 is unknown, but the gene is expressed in different skin cells (Garza et al. J Clin Invest. 2011; 121(2):613-22). Only 12 MB upstream of MAST4 lies the gene CD180, which directly interacts with the TLR4 signaling complex (Divanovic et al. Nat Immunol. 2005; 6(6):571-8). TLR4 plays an important role in the innate immunity of the skin. Expression studies have shown a disturbance in of expression of TLR4 in AD, PSO and contact dermatitis skin and a decreased expression of TLR4 is also found in treatment of SD with lithium gluconate (Elewa et al., J Dermatol. 2015; 52(5):467-76; Ballenger et al., Arch Dermatol Res. 2008; 300(5): 215-23). MAST3, a paralog of MAST4, is also known to influence TLR4 (Labbe et al. Genes Immun. 2008; 9(7):602-12).

(37) The MAST4 gene has not previously been associated with any skin condition, such as PSO or AD.

(38) The SNP rs58331610 maps to a 9.6kb region of chromosome 5 in the first intron of the MAST4 gene. The genomic DNA nucleotide sequence immediately around the SNP is set forth in SEQ ID NO: 1 and is shown in FIG. 1. In that Figure, the location of the polymorphism is indicated by R (purine, i.e. A or G).

(39) At time of writing, the most recent human genome assembly (GRCh 38/hg 38) provides the co-ordinates for SEQ ID NO: 1, shown in FIG. 1, as: chr5:66,724,129 to 66,725,129 (1001 bp). This region of the genome is located within the MAST4 gene, intron 1 (which has co-ordinates 66,597,019 to 66,759,708).

(40) The frequency of the G allele in the Rotterdam Study group (which is largely of European Caucasian ancestry) is 0.139. According to the findings of the Genome Wide Association Study the G allele is associated with ‘protection’ from developing seborrheic dermatitis.

(41) Interestingly, the data from the 1000 genome project shows that the frequency of this protective rs58331610 ‘G allele’ varies substantially between different ancestral groups around the globe, ranging from <10% in some African groups up to 50% in some East Asian and Central and South American ancestral populations.

(42) The SNP is strongly associated with SD, so it is possible that levels of expression of the MAST4 gene are associated with a predisposition to developing the condition. It follows that modulating the activity of the MAST4 gene, either up or down, may have a beneficial effect on prevention and/or treatment of the disease.

(43) Not only that, but there are three other MAST genes known in humans (MAST 1, 2 and 3), all of which are expressed in the skin. It is possible that up- or down-regulation of one or more of MAST 1-3 may also have a beneficial effect on prevention and/or treatment of SD. However, MAST4 appears to be the MAST gene which is most highly expressed in human skin, and is therefore the prime candidate for prophylactic and/or therapeutic intervention in SD.

(44) Using the MAPPER tool (genome.ufl.edu/mapper), the inventors found evidence that the rs58331610 SNP is present within a predicted C/EBP β transcription factor binding site.

(45) Activity of C/EBP β is important in the regulation of genes involved in immune and inflammatory responses. C/EBP-β transcription factor regulates IL-17 responsive genes and is expressed preferentially in differentiated keratinocytes. Expression of C/EBPβ is up-regulated in psoriatic compared to healthy skin. C/EBP-β binding sites are particularly found in regulatory sequences of genes that are associated with the inflammatory response (europepmc.org/backend/ptpmcrender.fcgi?accid=PMC4324884&blobtype=pdf). CEBPβ was first identified on the basis of its ability to regulate gene transcription in response to IL-1 and IL-6. (ncbi.nlm.nih.gov/pmc/articles/PMC1222736/pdf/12006103.pdf).

(46) Alternative names include NF-IL6, TCF5, IL-6DBP, LAP, CRP2, NF-M, AGP/EBP, ApC/EBP.

(47) Accordingly, one hypothesis is that genetic variation at the SNP locus could affect the C/EBPβ binding site and alter the binding affinity for the transcription factor and hence modulate transcription and expression of the MAST gene.

Example 2—Demonstration that MAST4 Kinase is Required for Inflammatory Response to Microbial Challenge and that Inhibition of MAST4 Kinase can Reduce Inflammatory Response in Normal Human Epidermal Keratinocytes

(48) Method:

(49) Normal Human Epidermal Keratinocytes were seeded at 40,000 cells per well in 24 well plates and grown for 16 hours in media (EpiLife® medium MEP1500CA)+HKGS supplement (S-001-5) at 37° C., 5% CO.sub.2. The cells were transfected using RNAiMAX with 20 nM of either a MAST4 target specific stealth siRNA (Thermofisher HSS180104) or a medium GC specific control siRNA. After 48 hours the medium was removed and cells were challenged with 10 μg/ml peptidoglycan from Staphylococcus aureus (Sigma 77140) for 24 hours. The medium was harvested from the cells and total RNA was extracted from each cell sample using Qiagen.

(50) RNeasy mini kit and quantitated on a NanoDrop. MAST4 gene expression was measured using standard gene expression analysis conditions and TaqMan® expression assay Hs00389519_m1. Duplicate 100 μ1 samples from each experimental well were analysed for IL8 using R&D systems IL8 duo-set ELISA.

(51) Table 2 below shows the relative quantitation (RQ) of MAST4 gene expression in the cell samples compared to vehicle control and normalized to “housekeeper” (Actin) gene expression. The MAST4 siRNA knockdown successfully reduced MAST4 transcript levels by >80%.

(52) Peptidoglycan challenge induced MAST4 transcript expression, increasing it by >25% in the cells transfected with control siRNA. This demonstrates that MAST4 gene expression is induced by peptidoglycan, which is a bacterial cell wall component, supporting a role for MAST4 in epidermal keratinocyte inflammatory immune response (Wang et al 2001 Infect. Immun. 69(4), 2270-2276). Peptidoglycan is known as a toll-like receptor 2 (TLR2) activator (Dziarski and Gupta 2005 Infect. Immun. 73(8), 5212-5216) and therefore this data supports a role for MAST4 in TLR2 mediated signaling in Normal Human Epidermal Keratinocytes.

(53) TABLE-US-00002 TABLE 2 Relative IL-8 induction by Quantitation of peptidoglycan siRNA Treatment MAST4 expression IL-8 (pg/ml) (pg/ml) Control Vehicle 1 26.73 +/− 9.33 NA MAST4 Vehicle 0.166  32.94 +/− 10.96 NA Control Peptidoglycan 1.257 245.94 +/− 49.86 219.21 MAST4 Peptidoglycan 0.18 175.30 +/− 24.16 142.36

(54) Table 2 also shows the production of inflammatory cytokine, IL-8 in response to the peptidoglycan challenge in the different cell populations. In control siRNA transfected keratinocytes, peptidoglycan challenge results in >9-fold increase in IL-8 production. In contrast, cells transfected with the MAST4 kinase siRNA duplex show a more modest 5-fold induction of IL-8 in response to the peptidoglycan challenge. This data shows that an 80% knockdown of MAST4 at the gene expression level is accompanied by a 35% reduction at the functional level, in the IL-8 response to peptidoglycan challenge. This data supports a key role for MAST4 in epidermal keratinocyte inflammatory signalling in response to a TLR2 agonist and demonstrates for the first time that inhibition of the MAST4 kinase can reduce inflammatory response to a microbial challenge in an in vitro epidermal model of skin inflammation.

Example 3—Epigenetic Study

(55) In this example, the inventors conducted a genome-wide study of DNA methylation in human scalp biopsies taken from healthy and dandruff volunteers, to investigate whether there are epigenetic differences associated with dandruff.

(56) Subjects

(57) Subjects were female, UK residents of a northern European ancestry (Caucasian) and aged 23 to 37. Dandruff suffers were defined as current anti-dandruff users, self-reported to have had dandruff for at least two years, and as had signs of dandruff during a visual assessment (Harding et al., Arch Dermatol Res (2002) 294:221-230) after 5 weeks use of a non-dandruff beauty shampoo. Healthy subjects were defined as not currently using anti-dandruff shampoo, did not report having dandruff for 2 years, and had no signs of dandruff after 5 weeks use of a non-dandruff beauty shampoo.

(58) Samples and Analysis

(59) 4 mm scalp biopsies were taken from healthy (n=25) and dandruff (n=23) scalps and snap frozen on dry ice. DNA was extracted using a standard Phenol Chloroform method (Molecular Cloning A laboratory manual, 3.sup.rd edition. Sambrook and Russell. Cold Spring Harbor Laboratory Press, 2001). DNA methylation was measured by first using the Zymo bi-sulphite conversion method (EPIC Methylation arrays: www.illumina.com) followed by interrogation on the Infinium methylation EPIC beadchip 850K arrays (Illumina). Statistical analysis of the chip data involved conversion of beta values to m-values (Du et al., BMC Bioinformatics 2010, 11:587). Raw m-value data were normalised using quantile normalisation and linear statistical models were used to identify probes that were differentially methylated regions between dandruff and healthy control subjects. After statistical filtering, beta-values were used to filter for a 5 percent difference between dandruff and healthy controls.

(60) Findings

(61) Differentially methylated sites were identified with statistical cut-offs of FDR adjusted p-value<0.05 and delta-beta>=5% (Du et al. 2010). At this level, 701 methylation sites were differentially methylated in dandruff relative to healthy samples. One of these differentially methylated sites was cg13811092 which is located at nucleotide 66010095 on chromosome 5 and another is cg03133881 which is located at nucleotide 46467354 on chromosome 1 in human genome assembly GRCh37/hg37. This former site is located within intron 1 of the MAST4 gene and is significantly hypo-methylated in dandruff compared with healthy scalps; this site lies in a region of ‘open chromatin’ close to an enhancer element (so there are likely to be transcription factors binding to this site in scalp) and 10 kb upstream of rs58331610, the genetic variant identified in the genome wide association study of seborrheic dermatitis. The latter CpG site is significantly hypo-methylated in dandruff compared with health scalps. The site is located within intron 6 of the MAST2 gene, and is also within a region of ‘open chromatin’.

(62) In the literature, it is well documented that hypo-methylation within a regulatory region of a gene can be associated with altered expression of the gene and, as a result, its function (Bonder et al. Nat Genet. 2017 January; 49(1):131-138). Therefore, the presence of differentially methylated sites in the MAST4 and MAST2 genes is supportive evidence that these genes play a role in dandruff etiology, particularly when they coincide with a regulatory region such as a site of open chromatin.