Mutations in the c-Secretase Genes NCSTN, PSENEN, and PSEN1 Underlie Rare Forms of Hidradenitis Suppurativa (Acne Inversa)

Thursday, 24 May, 2012
  • Andrew E. Pink,
  • Michael A. Simpson,
  • Nemesha Desai,
  • Dimitra Dafou,
  • Alison Hills,
  • Peter S Mortimer,
  • Catherine H Smith,
  • Richard C. Trembath,
  • Jonathan N.W. Barker


Mutations in the c-Secretase Genes NCSTN, PSENEN, and PSEN1 Underlie Rare Forms of Hidradenitis Suppurativa (Acne Inversa)
Journal of Investigative Dermatology (2012) 132, 2459–2461; doi:10.1038/jid.2012.162; published online 24 May 2012
Hidradenitis suppurativa (HS; OMIM142690) is a chronic inflammatory skin disease that presents with nodules, cysts, and abscesses in apocrine gland–bearing sites. It affects 1% of Europeans and is primarily thought to be a disease of follicular occlusion (Alikhan et al.,2009). Associated factors include smoking (B90%), obesity (475%), and infection (Alikhan et al., 2009). HS may segregate as an autosomal dominant trait, and heterozygous mutations in the g-secretase genes NCSTN, PSENEN, and PSEN1 have recently been reported in a small number of kindreds (Wang et al., 2010; Li et al., 2011; Liu et al., 2011; Pink et al., 2011). We previously identified mutations in NCSTN and PSENEN in two of seven multiplex HS pedigrees. In contrast to previous studies, which have predominantly focussed on the identification of mutations in multiplex kindreds, the aim of this study was to determine the prevalence of mutations in each of these genes among a large cohort of subjects sequentially recruited from a tertiary referral HS clinic. We performed mutational analysis of NCSTN,
PSENEN, and PSEN1 by direct sequencing of all coding regions and assessment of large-scale deletions and duplications by multiplex ligation phosphorylation assay (MLPA). In all, 48 individuals with HS (diagnostic criteria from Von der Werth et al., 2000) were sequentially consented and recruited from our tertiary referral clinic. The study was conducted in accordance with the Declaration of Helsinki Principles and approved by the East London REC2 (09/H0704/50). Population data are shown in Supplementary Table S1 online. The cohort was of mixed ethnicity, the average body mass index (BMI) was 31.1 (clinically obese), 69% reported a smoking history, and the average Sartorius score (Sartorius et al., 2003) was 49.9. A total of 20 (42%) patients reported a family history of HS, 19 consistent with autosomal dominant inheritance. The DNA was extracted from venous blood or saliva. All coding regions and associated splice sites of NCSTN, PSEN1, and PSENEN were amplified by PCR, using exon flanking intronic primers and Sanger sequenced in all 48 patients. In three subjects we identified heterozygous DNA changes in NCSTN (Supplementary Table S2 online), a missense substitution (NCSTN c.553G4A, p.Asp185Asn), and two single-base substitutions located within 10 bp of donor splice junctions (NCSTN c.996þ7G4A and NCSTN c.1101þ10A4G). To our knowledge these variants have not previously been reported and none were
observed in dbSNP, 1,000 genomes, or in 400 European controls. All 48 patients were assessed for whole gene/exon deletions and duplications in NCSTN, PSENEN, and PSEN1 using MLPA (Supplementary Methods online; Hills et al., 2010). No genomic deletions or duplications were detected. The heterozygous missense variant in exon 5 of NCSTN (c.553G4A, p.Asp185Asn) was identified in a 45-yearold female (patient HS-01; Supplementary Table S2 online, Figure 1). The missense variant is predicted to lead to the substitution of an evolutionary conserved aspartic acid residue with an asparagine residue. A heterozygous substitution of the conserved seventh base of intron 8 of NCSTN (c.996þ 7G4A) was identified in a 38-year-old female (patient HS-02 Supplementary Table S2 online, Figure 1). Computational splice site analysis (https://splice. predicts this substitution to have a detrimental effect on splicing. NCSTN mRNA expression levels were significantly lower than wild-type controls, suggesting that the mutant transcript is subject to nonsense-mediated
decay (Supplementary Methods and Supplementary Figure S1 online). A third heterozygous variant was detected in the donor splice site of NCSTN exon 9 (c.1101þ10A4G) in a 24-year-old male (patient HS-03; Supplementary Table S2 online, Figure 1). Computational splice site analysis suggests that this mutation is unlikely to have a significant effect on splicing. Indeed, no aberrant transcripts were detected (via reverse transcriptase–PCR and sequencing of the full-length NCSTN transcript, Supplementary Figures S2 and S3 online) and NCSTN mRNA expression level was within the range of wild-type controls (data not shown). The potential pathogenicity of this variant is therefore unclear and it may represent a rare variant unlinked to HS in this individual. The two individuals with likely pathogenic variants in NCSTN had chronic, severe disease (Sartorius scores of 56 and 160, respectively) comprising painful nodules, cysts, abscesses, sinus tracts, and scarring, involving at least the axillae, groin, and buttocks (see Supplementary Table S2 online). Their & 2012 The Society for Investigative Dermatology.

Abbreviations: BMI, body mass index; HS, hidradenitis suppurativa; MLPA, multiplex ligation
phosphorylation assay disease started at the age of 13 and 35 years, respectively. They had a raised BMI (38 and 37) and were both current smokers. Both reported a history of type II diabetes, neither had acne vulgaris, and there was no history of inflammatory bowel disease or other neutrophilic dermatoses. The only effective treatment option was surgical resection (see Supplementary Table S2 online). Interestingly, neither of these individuals reported a family history of HS or any associated conditions. g-Secretase is an endoprotease complex involved in the intramembranous cleavage of numerous type-1 transmembrane proteins. To date, mutations have been reported in three of the six genes encoding proteins integral to the complex (eight in NCSTN, three in PSENEN, and one in PSEN1, Figure 2) (De Strooper et al., 1999; Wang et al., 2010; Li et al., 2011; Liu et al., 2011; Pink et al., 2011). All but one of these mutations were observed in multiplex kindreds, whereas the HS cohort presented here is representative of patients recruited from a tertiary referral clinic (only 42% of patients reported a family history). Our data suggest that NCSTN, PSEN1, and PSENEN are only responsible for a small proportion of HS cases (o7%), and this figure may be lower in the general disease population given the phenotypic severity of the cohort studied. Of the 14 potential g-secretase mutations now reported in HS, 2 are nonsense mutations, 6 result in frameshifts, 4 in altered splicing, and 2 are missense mutations. The pattern of mutations suggests that loss-of-function of components of the g-secretase complex underlies the disease. Interestingly, the NCSTN missense mutation reported here (p.Asp185Asn) and the previously reported p.Pro211Arg (Li et al., 2011) are both located within the ectodomain of NCSTN (Figure 2). These observations support a key functional role for this domain within the g-secretase complex and a critical role in HS pathogenesis. It is currently difficult to draw robust genotype–phenotype correlations; however, the clinical phenotype of all reported mutation-positive cases is severe and extensive.


Where can I read this paper?

Full article on Journal of Investigative Dermatology.

Published: Journal of Investigative Dermatology (2012) 132, 2459–2461; doi:10.1038/jid.2012.162; published online 24 May 2012