Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 May 15;10(5):e1004333.
doi: 10.1371/journal.pgen.1004333. eCollection 2014.

Mutations in the Cholesterol Transporter Gene ABCA5 Are Associated With Excessive Hair Overgrowth

Affiliations
Free PMC article

Mutations in the Cholesterol Transporter Gene ABCA5 Are Associated With Excessive Hair Overgrowth

Gina M DeStefano et al. PLoS Genet. .
Free PMC article

Abstract

Inherited hypertrichoses are rare syndromes characterized by excessive hair growth that does not result from androgen stimulation, and are often associated with additional congenital abnormalities. In this study, we investigated the genetic defect in a case of autosomal recessive congenital generalized hypertrichosis terminalis (CGHT) (OMIM135400) using whole-exome sequencing. We identified a single base pair substitution in the 5' donor splice site of intron 32 in the ABC lipid transporter gene ABCA5 that leads to aberrant splicing of the transcript and a decrease in protein levels throughout patient hair follicles. The homozygous recessive disruption of ABCA5 leads to reduced lysosome function, which results in an accumulation of autophagosomes, autophagosomal cargos as well as increased endolysosomal cholesterol in CGHT keratinocytes. In an unrelated sporadic case of CGHT, we identified a 1.3 Mb cryptic deletion of chr17q24.2-q24.3 encompassing ABCA5 and found that ABCA5 levels are dramatically reduced throughout patient hair follicles. Collectively, our findings support ABCA5 as a gene underlying the CGHT phenotype and suggest a novel, previously unrecognized role for this gene in regulating hair growth.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. Whole-exome sequencing in a case of congenital generalized hypertrichosis terminalis (CGHT) revealed a splice site mutation in ABCA5.
(A) Clinical photos of proband that illustrate the excessive hair overgrowth on the back (upper photo) and leg (lower photo). The proband is two years of age in this photo. (B) Pedigree of the family from a consanguineous union (indicated by double lines), where the proband (arrow) is the only affected member. Asterisks indicate the individuals on whom whole-exome sequencing was performed. (C) Quantification of the length (mm) of patient versus control hair follicles derived from the forearm revealed that patient hair follicles are 87% longer, with an average length of 29.6 mm (±0.9 mm) compared to 15.9 mm (±0.6 mm) for control follicles (***P<0.0001). (D) Whole-exome sequencing filtering strategy, including the number of alterations and genes obtained at each level. One alteration (c.4320+1G>C) was identified in the ABCA5 gene, and cosegregation analysis in the family members revealed that both parents are carriers of this mutation, whereas the two unaffected sisters do not carry this mutation. (E) View of ABCA5 (red) and the surrounding genes (black boxes) on chr17q24.2-q24.3, where arrows indicate direction of transcription. A magnified view of the c.4320+1G>C mutation in exon 32 of ABCA5 with the exon 32-intron 32 boundary reference sequence annotated in colors corresponding to sequencing peaks. The G>C substitution is indicated below the control sequence. (F) Sequencing of control, carrier, and affected genomic DNA confirmed heterozygosity for the c.4320+1G>C mutation (boxed) in the father (I-1) and homozygosity in the proband (II-1). All coordinates reference the UCSC Genome Browser human reference genome hg19.
Figure 2
Figure 2. The ABCA5 c.4320+1G>C mutation leads to aberrant splicing and nonsense mediated decay, and ABCA5 is abundantly expressed in the skin and hair follicle.
(A) Depiction of normal splicing between exons 31 and 32 (black lines) and aberrant skipping of exon 32 as a result of the c.4320+1G>C mutation (red lines), where the adjoining of exon 31 to 33 leads to premature termination during translation. The out-of-frame exon 33 sequence (red box) is indicated by the hatched lines, followed by a stop codon. Sequencing of the mutant transcript (MT) confirmed the aberrant splicing event, where the amino acid sequence is listed above and the stop codon is indicated by an asterisk. RT-PCR on RNA from whole skin followed by gel electrophoresis of the exon 31–33 amplicon demonstrates complete loss of the wild-type (WT) transcript in the patient (P), and very low levels of the mutant (MT) transcript in the carrier (C). (B) qRT-PCR of ABCA5 transcript levels in the carrier and patient revealed a significant reduction of levels in patient whole skin, keratinocytes and fibroblasts. A Student unpaired t test was performed with a cutoff P value of 0.05 for statistical significance. (C) ABCA5 is expressed at strong levels throughout the outer root sheath (ORS) and dermal sheath (arrow) of the human hair follicle by in situ hybridization using an antisense probe, whereas the sense probe produced minimal background signal. (D) ABCA5 localizes to the perifollicular dermis, dermal sheath, as well as the hair follicle ORS and inner root sheath (IRS) by immunohistochemistry on paraffin-embedded sections. (E) The anti-rabbit IgG primary antibody control produced no signal. (F) Immunohistochemistry of ABCA5 on hair follicle end bulbs revealed strong expression in the dermal sheath and perifolliclar dermis but no expression was observed in the dermal papilla. (G) ABCA5 is endogenously expressed in plucked scalp hair follicles, microdissected ORS, as well as perifollicular dermis (including dermal sheath), determined by RT-PCR. Expression was normalized to GAPDH levels. M  =  marker.
Figure 3
Figure 3. ABCA5 protein levels are significantly reduced in CGHT patient keratinocytes and hair follicles.
(A) Immunofluorescence staining on cultured keratinocytes revealed a dramatic decrease in ABCA5 localization in CGHT keratinocytes compared to control (B) Immunoblotting on protein extracted from control and patient keratinocytes demonstrated loss of a 215 kDa band, which corresponds to the glycosylated form of the full-length ABCA5 protein, as well as a 187 kDa band, which is the unmodified protein in the patient compared to control. (C) Loss of ABCA5 localization to the outer root sheath (ORS) and inner root sheath (IRS) within patient hair follicles by immunofluorescence staining. Affected hair follicles in anagen and catagen were obtained from forearm skin biopsies, and control catagen hairs were obtained from forearm skin and anagen hairs from the occipital scalp. The anti-rabbit IgG primary antibody control produced no signal. HS  =  hair shaft; DP  =  dermal papilla; DS  =  dermal sheath.
Figure 4
Figure 4. Homozygous loss-of-function of ABCA5 perturbs lysosome function, resulting in an overall accumulation of autophagosomes and intracellular cholesterol levels in CGHT keratinocytes.
(A) Control and CGHT keratinocytes were cultured in normal media in the presence or absence of 50 nM bafilomycin, fixed and immunostained. Confocal analysis of LC3 (green), Lamp1 (red) and p62 (blue) revealed an increased number of LC3-positive particles as well strong accumulation of p62 particles in affected keratinocytes. BAF-treated control cells possessed a two-fold increase in the number of LC3 puncta, whereas no significant difference was observed in BAF-treated affected cells (quantified in Figure S5). (B) Control and CGHT keratinocytes were fixed and immunostained for Filipin and Lamp1 (red), which revealed an accumulation and redistribution of free cholesterol to Lamp1-positive organelles in affected keratinocytes.
Figure 5
Figure 5. Cytogenetic analyses, breakpoint mapping, and copy number variant analysis in a sporadic case of CGH revealed a t3;17 translocation that leads to a cryptic 1.3 Mb deletion of chr17q24.2-24.3, and SOX9 expression is reduced in patient keratinocytes.
(A) Chromosomal paint was performed in a sporadic case of CGHT with commercially available whole chromosome probes for both chromosomes 3 and 17. Note: the probes do not bind to any of the other chromosomes except 3 and 17, and the derivative chromosomes suggesting that this is an isolated rearrangement between these two chromosomes. DAPI staining was used to visualize the entire chromosomal set (blue). (B) Copy number variant analysis using the Affymetrix Cytogenetics Whole Genome 2.7M array validated the cryptic deletion that spans 1.3 Mb (red box). RefSeq genes are depicted as black boxes. (C) Region of chr17q24.2-q24.3, including the genes and 1.3 Mb deletion identified by FISH and CNV analysis. The quantitative PCR (qPCR) amplicons are indicated as vertical lines, with three amplicons residing within and two flanking the deletion. qPCR was performed to confirm the deletion on control (C) and affected (A) genomic DNA using relative quantification and the primers listed in the Materials and Methods. (D) qRT-PCR of SOX9 on RNA from control and affected keratinocytes revealed a 2.5-fold decrease (P>0.001) in expression in patient keratinocytes. A Student t test (unpaired) was performed with a cutoff P value of 0.05 for statistical significance. All coordinates reference the UCSC Genome Browser human reference genome hg19.
Figure 6
Figure 6. ABCA5 levels are markedly reduced in patient hair follicles of the sporadic CGHT case.
(A) qRT-PCR for ABCA5 on control and affected keratinocytes and fibroblasts cultured from whole skin biopsies revealed a significant reduction in the patient cells compared to controls. A Student unpaired t test was performed with a cutoff P value of 0.05 for statistical significance. (B) Immunoblotting on protein extracted from carrier and patient fibroblasts in the presence or absence of the enzyme, PNGaseF revealed loss of a ∼100 kDa band that is the glycosylated form of the protein in patient fibroblasts relative to controls. (C) Immunofluorescence staining on control and patient hair follicles in the catagen and anagen stages demonstrated reduced ABCA5 localization throughout the outer root sheath (ORS) and inner root sheath (IRS) of patient hair follicles (compare to control catagen hair follicle in Figure 3C). HS  =  hair shaft; HC  =  hair canal.

Similar articles

See all similar articles

Cited by 9 articles

See all "Cited by" articles

References

    1. Beighton P (1970) Congenital hypertrichosis lanuginosa. Arch Dermatol 101: 669–672. - PubMed
    1. Wendelin DS, Pope DN, Mallory SB (2003) Hypertrichosis. J Am Acad Dermatol 48: 161–179 quiz 180–161. - PubMed
    1. Garcia-Cruz D, Figuera LE, Cantu JM (2002) Inherited hypertrichoses. Clin Genet 61: 321–329. - PubMed
    1. Fantauzzo KA, Kurban M, Levy B, Christiano AM (2012) Trps1 and its Target Gene Sox9 Regulate Epithelial Proliferation in the Developing Hair Follicle and are Associated with Hypertrichosis. PLoS Genet 8: e1003002. - PMC - PubMed
    1. Fantauzzo KA, Tadin-Strapps M, You Y, Mentzer SE, Baumeister FA, et al. (2008) A position effect on TRPS1 is associated with Ambras syndrome in humans and the Koala phenotype in mice. Hum Mol Genet 17: 3539–3551. - PMC - PubMed

Publication types

MeSH terms

Supplementary concepts

Feedback