Advertisement
Journal of Investigative Dermatology Home

Clonal Expansion of Second-Hit Cells with Somatic Recombinations or C>T Transitions Form Porokeratosis in MVD or MVK Mutant Heterozygotes

Open AccessPublished:June 15, 2019DOI:https://doi.org/10.1016/j.jid.2019.05.020
      Patients with disseminated superficial actinic porokeratosis (DSAP) and linear porokeratosis (LP) exhibit monoallelic germline mutations in genes encoding mevalonate pathway enzymes, such as MVD or MVK. Here, we showed that each skin lesion of DSAP exhibited an individual second hit genetic change in the wild-type allele of the corresponding gene specifically in the epidermis, indicating that a postnatal second hit triggering biallelic deficiency of the gene is required for porokeratosis to develop. Most skin lesions exhibited one of two principal second hits, either somatic homologous recombinations rendering the monoallelic mutation biallelic or C>T transition mutations in the wild-type allele. The second hits differed among DSAP lesions but were identical in those of congenital LP, suggesting that DSAP is attributable to sporadic postnatal second hits and congenital LP to a single second hit in the embryonic period. In the characteristic annular skin lesions of DSAP, the central epidermis featured mostly second hit keratinocytes, and that of the annular ring featured a mixture of such cells and naïve keratinocytes, implying that each lesion reflects the clonal expansion of single second hit keratinocytes. DSAP is therefore a benign intraepidermal neoplasia, which can be included in the genetic tumor disorders explicable by Knudson’s two-hit hypothesis.

      Graphical abstract

      Abbreviations:

      DSAP (disseminated superficial actinic porokeratosis), LOH (loss of heterozygosity), LP (linear porokeratosis)

      Introduction

      In several genetic tumor syndromes, multiple proliferative tumors develop with age in patients with monoallelic germline mutations of certain causative genes when second hit genetic changes occur in the wild-type allele of that gene. This mechanism is termed the Knudson “two-hit” hypothesis (
      • Knudson A.G.
      Two genetic hits (more or less) to cancer.
      ). Such diseases include neurofibromatosis (OMIM 162200) and nevoid basal cell carcinoma syndrome (OMIM 109400).
      Porokeratosis encompasses a group of keratinization disorders characterized by circular or annular skin lesions with a distinct hyperkeratotic rim termed the cornoid lamella (
      • Sertznig P.
      • von Felbert V.
      • Megahed M.
      Porokeratosis: present concepts.
      ). Known causative genes are MVD, MVK, PMVK, FDPS, and SLC17A9 (
      • Cui H.
      • Li L.
      • Wang W.
      • Shen J.
      • Yue Z.
      • Zheng X.
      • et al.
      Exome sequencing identifies SLC17A9 pathogenic gene in two Chinese pedigrees with disseminated superficial actinic porokeratosis.
      ,
      • Zhang S.Q.
      • Jiang T.
      • Li M.
      • Zhang X.
      • Ren Y.Q.
      • Wei S.C.
      • et al.
      Exome sequencing identifies MVK mutations in disseminated superficial actinic porokeratosis.
      ,
      • Zhang Z.
      • Li C.
      • Wu F.
      • Ma R.
      • Luan J.
      • Yang F.
      • et al.
      Genomic variations of the mevalonate pathway in porokeratosis.
      ). Clinically, porokeratosis is classified based on the lesional skin distribution. Disseminated superficial actinic porokeratosis (DSAP) develops over the entire skin of adults, particularly on the extremities. Linear porokeratosis (LP) appears during infancy or early childhood on a limited, segmental skin area, typically along Blaschko’s lines (
      • Sertznig P.
      • von Felbert V.
      • Megahed M.
      Porokeratosis: present concepts.
      ). Both sporadic and familial cases of DSAP and LP have been reported; familial cases are usually associated with autosomal-dominant inheritance (
      • Schamroth J.M.
      • Zlotogorski A.
      • Gilead L.
      Porokeratosis of Mibelli. Overview and review of the literature.
      ). Congenital heterozygous mutations in genes encoding enzymes of the mevalonate pathway, including MVD and MVK, are responsible for familial and sporadic DSAP and LP (
      • Zhang S.Q.
      • Jiang T.
      • Li M.
      • Zhang X.
      • Ren Y.Q.
      • Wei S.C.
      • et al.
      Exome sequencing identifies MVK mutations in disseminated superficial actinic porokeratosis.
      ,
      • Zhang Z.
      • Li C.
      • Wu F.
      • Ma R.
      • Luan J.
      • Yang F.
      • et al.
      Genomic variations of the mevalonate pathway in porokeratosis.
      ). The coexistence of DSAP and LP in single patients or families suggests that acquired second hit genetic changes in those with monoallelic, pathogenic mutations underlie the development of porokeratosis (
      • Commens C.A.
      • Shumack S.P.
      Linear porokeratosis in two families with disseminated superficial actinic porokeratosis.
      ,
      • Happle R.
      Somatic recombination may explain linear porokeratosis associated with disseminated superficial actinic porokeratosis.
      ); such second hit changes were recently identified in the MVD or MVK genes of three patients with LP (
      • Atzmony L.
      • Khan H.M.
      • Lim Y.H.
      • Paller A.S.
      • Levinsohn J.L.
      • Holland K.E.
      • et al.
      Second hit, postzygotic PMVK and MVD mutations in linear porokeratosis.
      ).
      Here, we showed that individual DSAP skin lesions exhibited unique second hit genetic changes in the presence of a monoallelic germline mutation in either MVK or MVD. Most skin lesions exhibited either second hit somatic homologous recombinations that rendered the monoallelic mutation biallelic or C>T transition mutations of the wild-type allele. The second hits differed among skin lesions in DSAP but were identical in those of congenital LP, indicating that such second hits occurring sporadically in the postnatal period and once in the embryonic period trigger DSAP and LP, respectively.

      Results

      Japanese probands with DSAP and LP were recruited for genetic analyses. Figure 1 and Supplementary Table S1 show the clinical courses and phenotypes. Skin lesions were distributed from birth on limited segmental body regions in the two patients with LP (LP1 and LP2, Figure 1a), but lesions developed later in life and were widely distributed, especially on the extremities, in the seven patients with DSAP (DSAP1–7, Figure 1b). Figure 1c and d shows typical annular skin lesions exhibiting rim-specific inflammation and cornoid lamella formation. We performed genetic analyses after obtaining written informed consent in accordance with the guidelines of the institutional review board of Keio University School of Medicine. Monoallelic congenital mutations were found in MVD or MVK in the peripheral blood leukocytes of all patients (Supplementary Table S1). The most common mutation was c.746T>C (rs761991070, p.F249S) of MVD (RefSeq: NM_002461), which was present in seven of the nine patients (LP2 and DSAP1–6).
      Figure thumbnail gr1
      Figure 1Clinical manifestations of early onset LP and late-onset DSAP. (a) Skin manifestations of early onset LP in patients LP1 and LP2. (b) Skin manifestations of late-onset DSAP on the lower extremities of patients DSAP1, DSAP3, and DSAP7. (c) Typical annular skin lesions in patients DSAP6 and DSAP7. (d) Hematoxylin and eosin staining of the lesional skin of patients LP2 and DSAP6 showing cornoid lamella in the stratum corneum, melanophages, and cell infiltration in the upper dermis. DSAP, disseminated superficial actinic porokeratosis; LP, linear porokeratosis. Bar = 100 μm. Participants gave their written informed consent to publish the clinical images.
      We performed multiple biopsies of both lesional and non-lesional skin of the two patients with LP (LP1 and LP2) and the seven patients with DSAP (DSAP1–7). We separated each biopsy sample into epidermis and dermis via dispase treatment (
      • Kitano Y.
      • Okada N.
      Separation of the epidermal sheet by dispase.
      ) and purified genomic DNA from these tissues.
      In patient LP1, who exhibited a monoallelic germline c.127_128delCT mutation of MVD, a c.683G>A (p.R228Q) mutation in MVD was specifically identified in the epidermis of all three lesional biopsies (Table 1 and Supplementary Figure S1a and b). PCR-based cloning and sequencing of genomic DNA showed that the c.683G>A mutation and the germline mutation of c.127_128delCT were located on different chromosomes, exhibiting compound heterozygosity (Supplementary Figure S2 and Supplementary Table S2). Thus, the de novo c.683G>A mutation was the disease-causative second hit in patient LP1.
      Table 1Genetic Analyses of Multiple Skin Biopsies of Linear Porokeratosis and Disseminated Superficial Actinic Porokeratosis
      Case/Germline Mutation Found in Leukocytes and Skin (1st Hit)Biopsied SampleSanger SequencingSingle Nucleotide Polymorphism Array
      Epidermis Specific Mutations (2nd hit)Dermis Specific MutationsHR at the 1st Hit Locus in the EpidermisRange of the HR (Mb)Estimated %

      of HR Mosaicism
      Other HRs and Their Range (Mb) and % Mosaicism
      LP1/ Heterozygote of MVD c.127_128delCT mutationLS #1c.683G>A in MVD ex7 (p.R228Q)NDNP
      LS #2c.683G>A in MVD ex7 (p.R228Q)NDNP
      LS #3c.683G>A in MVD ex7 (p.R228Q)NDNP
      Non-LS #1NDNDNP
      LP2/ Heterozygote of MVD c.746T>C (p.F249S) mutationLS #1LOH of the 1st hitNDChr16 HR76.36–ter58.6 %ND
      LS #2LOH of the 1st hitNDChr16 HR76.36–ter62.8 %ND
      LS #3LOH of the 1st hitNDChr16 HR76.36–ter70.3 %ND
      Non-LS #1NDNDNo HR in Chr16ND
      Non-LS #2NDNDNP
      DSAP1/ Heterozygote of MVD c.746T>C (p.F249S) mutationLS #1LOH of the 1st hitNDChr16 HRentire 16q16.1 %ND
      LS #2LOH of the 1st hitNDChr16 HRentire 16q47.1 %ND
      LS #3c.682delC in MVD ex7 (p.R228fs)NDNo HR in Chr16ND
      LS #4LOH of the 1st hitNDChr16 HR56.61–ter49.7 %ND
      LS #5c.366_367GG>AA in MVD ex4 (p.G123S)NDNP
      LS #6c.403G>A and c.403+1G>A in MVD ex4 and int4NDNP
      LS #7NDNDNo HR in Chr16Chr15 HR (entire 15q), 18.0 %
      LS #8LOH of the 1st hitNDChr16 HR78.69–ter67.5 %ND
      DSAP2/ Heterozygote of MVD c.746T>C (p.F249S) mutationLS #1LOH of the 1st hitNDChr16 HR64.19–ter75.5 %ND
      LS #2LOH of the 1st hitNDChr16 HRentire 16q76.2 %ND
      LS #3LOH of the 1st hitNDChr16 HRentire 16q61.8 %ND
      LS #4LOH of the 1st hitNDChr16 HR68.70–ter69.7 %ND
      LS #5LOH of the 1st hitNDChr16 HR52.89–tercenter: 82.9 %

      periphery: 36.0 %
      Chr17 HR
      (29.42–ter),
      center: 18.5 %
      periphery: 14.9 %
      Non-LS #1NDNDNo HR in Chr16ND
      DSAP3/ Heterozygote of MVD c.746T>C (p.F249S) mutationLS #1c.377C>T in MVD ex4 (p.S126F)NDNP
      LS #2LOH of the 1st hitChr16 HR59.54–ter86.4 %ND
      LS #3c.113C>T in MVD ex2 (p.S38F)NDNP
      LS #4LOH of the 1st hitNDChr16 HRentire 16q50.0 %ND
      LS #5LOH of the 1st hitNDChr16 HRentire 16q65.4 %ND
      Non-LS #1NDNDNP
      DSAP4/ Heterozygote of MVD c.746T>C (p.F249S) mutationLS #1LOH of the 1st hitLOH of the 1st hitChr16 HRentire 16q9.6 %ND
      LS #2NDNDNo HR in Chr16ND
      LS #3LOH of the 1st hitNDChr16 HRentire 16q7.4 %ND
      LS #4LOH of the 1st hitNDChr16 HRentire 16q9.9 %ND
      LS #5NDNDNo HR in Chr16ND
      LS #6LOH of the 1st hitNDChr16 HRentire 16q29.2 %ND
      DSAP5/ Heterozygote of MVD c.746T>C (p.F249S) mutationLS #1LOH of the 1st hitNDChr16 HR68.60–ter57.6 %ND
      LS #2LOH of the 1st hitNDChr16 HRentire 16q35.0 %ND
      LS #3NDNDNP
      LS #4LOH of the 1st hitNDChr16 HRentire 16q52.4 %ND
      LS #5LOH of the 1st hitNDChr16 HRentire 16q82.6 %ND
      DSAP6/ Heterozygote of MVD c.746T>C (p.F249S) mutationLS #1LOH of the 1st hitNDChr16 HRentire 16q60.6 %Chr19 HR (29.84-ter), 15.7 %
      LS #2LOH of the 1st hitNDChr16 HRentire 16qcenter: 87.4 %

      periphery: 48.6 %
      ND
      LS #3LOH of the 1st hitNDChr16 HRentire 16q75.0 %ND
      LS #4LOH of the 1st hitNDChr16 HRentire 16q45.6 %ND
      LS #5c.3G>A in MVD ex1 (p.M1I)NDNP
      Non-LS #1NDNDNP
      Non-LS #2NDNDNo HR in Chr16ND
      DSAP7/ Heterozygote of MVK c.1073A>C (p.Q358P) mutationLS #1c.575G>A (p.G192E) and c.602C>T (p.S201F) in MVKNDNo HR in Chr12ND
      LS #2LOH of the 1st hitNDChr12 HR82.49–ter87.4 %ND
      LS #3c.815_816CC>TT (p.S272F) in MVKNDNo HR in Chr12ND
      LS #4LOH of the 1st hitNDChr12 HRentire 1293.6 %ND
      LS #5c.1126G>A (p.G376S) in MVKNDNo HR in Chr12ND
      LS #6c.814T>C (p.S272P) in MVKNDNo HR in Chr12ND
      Abbreviations: Chr, chromosome; DSAP, disseminated superficial actinic porokeratosis; ex, exon; HR, somatic homologous recombination; int, intron; LOH, loss of heterogeneity; LS, lesional skin; LP, linear porokeratosis; Mb, mega base; ND, not detected; Non-LS, non-lesional skin; NP, not performed; ter, terminus of the long arm of chromosome.
      In patient LP2, who had a monoallelic germline c.746T>C mutation of MVD, Sanger sequencing revealed loss of heterozygosity (LOH) of the germline c.746T>C mutation specifically in the epidermis of all three lesional biopsies (Table 1 and Supplementary Figure S1c–e). Whole-exome sequencing revealed no de novo mutation in any genes in the lesional epidermis (data not shown). Single nucleotide polymorphism array analyses revealed that the LOH of the MVD mutant was caused by an epidermis-specific LOH of the terminal ∼14-Mb region of chromosome 16q, including the MVD locus, and was thus not attributable to a change in gene dosage (Supplementary Figure S1e). These results indicate that postzygotic somatic homologous recombination in chromosome 16q rendered the monoallelic c.746T>C mutation of MVD biallelic (Supplementary Figure S1e) and was the disease-causative second hit in patient LP2.
      We thus confirmed the recent report of
      • Atzmony L.
      • Khan H.M.
      • Lim Y.H.
      • Paller A.S.
      • Levinsohn J.L.
      • Holland K.E.
      • et al.
      Second hit, postzygotic PMVK and MVD mutations in linear porokeratosis.
      showing second hit genetic changes in the skin lesions of patients with LP with monoallelic germline mutations in MVD or MVK. Next, we explored whether second hit genetic changes were evident in the skin lesions of patients with DSAP. We performed multiple biopsies of skin lesions of patients with DSAP 1–7 (Table 1). Of the 40 lesions biopsied, 26 lesional epidermal samples exhibited LOH of the congenital monoallelic mutation attributable to somatic homologous recombination of the chromosome in which the congenital mutation was located (chromosome 16 for MVD and chromosome 12 for MVK; Figure 2a–c and Supplementary Figure S3 for DSAP1; summarized in Figure 2d and Table 1). The extent of homologous recombination differed among lesions; recombination of the entire long chromosome arm was the most common event (Figure 2d). Ten lesional epidermal samples exhibited acquired mutations, principally C>T and CC>TT transitions or the complementary G>A and GG>AA transitions (8 out of 10 acquired mutations; summarized in Figure 2e and Table 1). PCR-based cloning and sequencing of genomic DNA demonstrated that the acquired and congenital mutations were located on different chromosomes, exhibiting compound heterozygosity (Supplementary Figure S2 and Supplementary Table S2). The remaining four samples contained no detectable acquired mutations or LOH, suggesting that other mechanisms were involved in permanent inactivation of the wild-type allele (e.g., microdeletion of several exons or mutations in transcriptional regulatory regions). Thus, each DSAP lesion exhibited a unique second hit change. The changes were principally homologous recombinations and C>T transitions; a postnatal second hit triggering biallelic deficiency of MVD or MVK was required for porokeratosis to develop.
      Figure thumbnail gr2
      Figure 2Different second hits in patients with LP and DSAP. (a) LOH in the lesional skin epidermis of patient DSAP1. (b) Log R ratio and B-allele frequency of Chr16 in the lesional epidermis. Pink: homologous recombination region; gray: segments lacking detectable single nucleotide polymorphisms. (c) Schematic of homologous recombination. Red and yellow bars: chromosome containing the congenital pathogenic mutation (black circle) and the wild-type gene, respectively. (d, e) Summary of homologous recombinations (d) and somatic mutations (e) in epidermal samples of lesional skin from the indicated subjects. Each bar in (d) shows a recombinant chromosome in a skin lesion. Red lines in (e) show the allele containing the congenital mutation (black circle) of each subject; green lines in (e) show the allele containing the acquired mutation. BAF, B-allele frequency; Chr, chromosome; DSAP, disseminated superficial actinic porokeratosis; LOH, loss of heterozygosity; LP, linear porokeratosis; LRR, log R ratio.
      Clinically, DSAP skin lesions first present as tiny erythematous papules that gradually expand to form annular lesions with hyperkeratotic rims. Our results suggest that each DSAP lesion is an independent colony originating from a single second hit stem cell. Next, we determined the ratios of naïve and second hit cells in the centers and peripheral rings of annular DSAP skin lesions. The characteristic lesions of patients DSAP2, DSAP6, and DSAP7 were biopsied, and each sample was separated into a central cylinder of normal-looking skin and the annular, hyperkeratotic hyperpigmented rim (Figure 3a and b). Each sample was further separated into epidermis and dermis and analyzed (Figure 3c–e).
      Figure thumbnail gr3
      Figure 3Genetic characterization of annular skin lesions of disseminated superficial actinic porokeratosis. (a) Sample processing. (b) Annular lesion before biopsy (left) and after dissection (right). Dotted lines indicate dissection margin; *, Dermis protruding laterally beneath the epidermis after dissection. (c, d) Genetic analyses of the indicated samples. Sanger sequencing chromatograms reveal the congenital heterozygous mutations (black arrowheads); loss of heterozygosity (open arrowheads in (c) and, in (d), mosaicism of acquired mutations (black arrows, wild-type nucleotides; open arrowheads, mutant nucleotides). B-allele frequency of Chr16 reveals homologous recombination (pink region in c). Right: Estimated ratio of naïve (green) and second hit (purple) cells as determined by digital PCR analyses. (e) Estimated percentage of second hit cells in the epidermis of three annular lesions as revealed by digital PCR analyses. (f) Development of porokeratosis skin lesions. BAF, B-allele frequency; Chr, chromosome; LS, lesional skin.
      In lesional skin (LS) samples DSAP2 LS #5 and DSAP6 LS #2, the second hit genetic change was homologous recombination in the long arm of chromosome 16 (Figure 3c and Supplementary Table S2). Sanger sequencing and single nucleotide polymorphism array analyses revealed that the wild-type allele of MVD mostly was absent in the epidermis of the central normal-looking area, but was present in the epidermis of the peripheral ring (Figure 3c). The ratio of naïve cells (cells with the monoallelic MVD c.746T>C mutation) to second hit cells (with the biallelic MVD c.746T>C mutation), calculated via digital PCR analyses, revealed a higher proportion of second hit cells in the central area than in the peripheral ring (Figure 3c and Supplementary Figures S4 and S5).
      In a lesional skin sample DSAP7 lesional skin #1, two acquired mutations of MVK, c.575G>A and c.602C>T, constituted the second hit genetic changes. Sanger sequencing revealed that the ratios of the two second hit mutant alleles to the wild-type allele were constant (Figure 3d), suggesting that the two second hits occurred nearly simultaneously. The ratio of naïve cells (cells heterozygous for the MVK c.1073A>C mutation) to second hit cells (compound heterozygotes for the MVK c.1073A>C mutation and the two acquired mutations), calculated via digital PCR analyses, revealed a higher proportion of second hit cells in the central area than in the peripheral ring (Figure 3d and Supplementary Figure S6).
      Multiple comparisons of the three skin lesions showed that the estimated ratios of second hit to naïve cells differed significantly, being lowest in the surrounding normal epidermis, highest in the epidermis of the central lesion, and nearly equal in the epidermis of the peripheral ring (Figure 3e), indicating that clonal expansion of a single second hit stem cell had caused the development of each DSAP skin lesion (Figure 3f).

      Discussion

      Our observations indicated that DSAP is a benign intraepidermal neoplasia, which can be included in the genetic tumor disorders explicable by Knudson's “two-hit” hypothesis (
      • Knudson A.G.
      Two genetic hits (more or less) to cancer.
      ) and develops in the presence of a monoallelic germline mutation of either MVK or MVD. The principal second hit genetic changes are postzygotic C>T transitions and somatic homologous recombinations that render the monoallelic mutation biallelic, indicating that a postnatal second hit is required for porokeratosis to develop. Embryonic and postnatal second hits trigger LP and DSAP, respectively, as shown in Figure 4. DSAP probably reflects widespread postnatal genetic changes, including somatic homologous recombinations and C>T transitions, in aged skin.
      Figure thumbnail gr4
      Figure 4Schematics of the pathogenesis of linear porokeratosis and DSAP. The early-onset segmental form (linear porokeratosis, left panels) and late-onset sporadic form (DSAP, right panels) of porokeratosis are caused by second hit genetic changes (shown in red) that occur de novo. Red circles, skin lesions; "mut" in black, allele with the congenital pathogenic mutation; "mut" in red, allele with acquired genetic changes. DSAP, disseminated superficial actinic porokeratosis; WT, wild-type allele of the causative gene.
      Seven of the nine Japanese patients analyzed in this study (LP2 and DSAP1–6) had a germline c.746T>C (p.F249S) mutation in MVD. In a Japanese cohort database search, the c.746T>C mutation was identified in nine out of 7,104 alleles (minor allele frequency = 0.0013, Japanese Multi Omics Reference Panel), indicating that approximately one per 400 Japanese individuals is estimated to have a pathogenic mutation in MVD. The c.746T>C mutation of MVD is probably the major cause of LP and DSAP in Japan.
      C>T and CC>TT transitions are induced in keratinocytes by UV light (
      • Brash D.E.
      UV signature mutations.
      ,
      • Pfeifer G.P.
      • You Y.H.
      • Besaratinia A.
      Mutations induced by ultraviolet light.
      ). In contrast, somatic homologous recombination is induced by double-stranded DNA breaks and subsequent repair of the breaks using the homologous chromosome as a template (
      • Mehta A.
      • Haber J.E.
      Sources of DNA double-strand breaks and models of recombinational DNA repair.
      ); the cause of such breaks remains unknown (
      • Vilenchik M.M.
      • Knudson A.G.
      Endogenous DNA double-strand breaks: production, fidelity of repair, and induction of cancer.
      ,
      • White R.R.
      • Vijg J.
      Do DNA double-strand breaks drive aging?.
      ). Although DSAP has been attributed to chronic sun exposure, the face is affected in only ∼15% of patients (
      • Sertznig P.
      • von Felbert V.
      • Megahed M.
      Porokeratosis: present concepts.
      ), suggesting that factors other than UV light induce somatic homologous recombinations in the epidermis and thus the development of DSAP.
      Recent advances in analytical methods have revealed the progressive accumulation of cells with genomic damage, including homologous recombination, in humans (
      • Abyzov A.
      • Mariani J.
      • Palejev D.
      • Zhang Y.
      • Haney M.S.
      • Tomasini L.
      • et al.
      Somatic copy number mosaicism in human skin revealed by induced pluripotent stem cells.
      ,
      • Abyzov A.
      • Tomasini L.
      • Zhou B.
      • Vasmatzis N.
      • Coppola G.
      • Amenduni M.
      • et al.
      One thousand somatic SNVs per skin fibroblast cell set baseline of mosaic mutational load with patterns that suggest proliferative origin.
      ,
      • Forsberg L.A.
      • Gisselsson D.
      • Dumanski J.P.
      Mosaicism in health and disease - clones picking up speed.
      ,
      • Laurie C.C.
      • Laurie C.A.
      • Rice K.
      • Doheny K.F.
      • Zelnick L.R.
      • McHugh C.P.
      • et al.
      Detectable clonal mosaicism from birth to old age and its relationship to cancer.
      ,
      • Wang Y.
      • Masaki T.
      • Khan S.G.
      • Tamura D.
      • Kuschal C.
      • Rogers M.
      • et al.
      Four-dimensional, dynamic mosaicism is a hallmark of normal human skin that permits mapping of the organization and patterning of human epidermis during terminal differentiation.
      ). Disseminated epidermal homologous recombinations have been found in patients with ichthyosis with confetti (OMIM 609165) and loricrin keratoderma (OMIM 604117); revertant cells that have lost the disease-causing mutation via homologous recombination form normal-looking patches on affected skin (
      • Choate K.A.
      • Lu Y.
      • Zhou J.
      • Choi M.
      • Elias P.M.
      • Farhi A.
      • et al.
      Mitotic recombination in patients with ichthyosis causes reversion of dominant mutations in KRT10.
      ,
      • Choate K.A.
      • Lu Y.
      • Zhou J.
      • Elias P.M.
      • Zaidi S.
      • Paller A.S.
      • et al.
      Frequent somatic reversion of KRT1 mutations in ichthyosis with confetti.
      ,
      • Suzuki S.
      • Nomura T.
      • Miyauchi T.
      • Takeda M.
      • Fujita Y.
      • Nishie W.
      • et al.
      Somatic recombination underlies frequent revertant mosaicism in loricrin keratoderma.
      ). DSAP is almost a mirror image of such diseases. In ichthyosis with confetti, revertant cell colonies usually appear during the teenage years (
      • Choate K.A.
      • Lu Y.
      • Zhou J.
      • Choi M.
      • Elias P.M.
      • Farhi A.
      • et al.
      Mitotic recombination in patients with ichthyosis causes reversion of dominant mutations in KRT10.
      ), much earlier than in DSAP (
      • Sertznig P.
      • von Felbert V.
      • Megahed M.
      Porokeratosis: present concepts.
      ). Probably this is because aberrant keratin filaments formed in the nuclei of keratinocytes of patients with ichthyosis with confetti (
      • Choate K.A.
      • Lu Y.
      • Zhou J.
      • Choi M.
      • Elias P.M.
      • Farhi A.
      • et al.
      Mitotic recombination in patients with ichthyosis causes reversion of dominant mutations in KRT10.
      ) mechanically accelerate double-stranded DNA breakage and somatic recombination.
      DSAP lesions commonly develop later in life. We found that postnatal homologous recombination occurs sporadically in the epidermis. Notably, probable disease-neutral homologous recombinations were identified in three of the 40 DSAP samples (Table 1), suggesting that somatic homologous recombination is not a rare event in the epidermis. Previous chronological observations of DSAP concluded that several extrinsic factors, including aging and immunosuppression, induced the development of multiple skin lesions (
      • Sertznig P.
      • von Felbert V.
      • Megahed M.
      Porokeratosis: present concepts.
      ). However, how these factors affect the incidence of somatic recombination caused by double-stranded DNA breaks remains to be determined, as does the survival rate of recombinant cells that expand clonally to form skin lesions.
      Circular or annular lesions are characteristic of porokeratosis. We found that each lesion formed via clonal expansion of second hit cells. Dense infiltrations of mononuclear immune cells and cornoid lamella formation were evident at the rim but not in the center. Future studies should explore how a biallelic deficiency in genes encoding mevalonate pathway enzymes triggers clonal expansion of such cells, how naïve and second hit cells mix to form cornoid lamellae, and how inflammatory reactions are induced at the rim. The findings will contribute to the development of novel therapeutic strategies for porokeratosis.

      Materials and Methods

       Sample preparation

      Sample collection was approved by the Ethics Committee of Keio University School of Medicine and performed after obtaining written informed consent. Participants gave their written informed consent to publish the clinical images. The skin was biopsied under local anesthesia to obtain full-thickness skin specimens. The epidermis was separated from the dermis by dispase treatment, as described previously (
      • Kitano Y.
      • Okada N.
      Separation of the epidermal sheet by dispase.
      ).
      The annular skin lesions were dissected into the peripheral ring and central cylinder, as shown in Figure 3a. The full-thickness annular skin lesion was punched out with a punch biopsy with a diameter identical to that of the skin lesion. The central normal-looking area was punched out ex vivo from the lesional specimens with a punch biopsy instrument with a diameter identical to that of the central area.

       Exome sequencing

      Exome sequencing and data analyses were performed as described previously (
      • Kubo A.
      • Shiohama A.
      • Sasaki T.
      • Nakabayashi K.
      • Kawasaki H.
      • Atsugi T.
      • et al.
      Mutations in SERPINB7, encoding a member of the serine protease inhibitor superfamily, cause Nagashima-type palmoplantar keratosis.
      ,
      • Takenouchi T.
      • Yamaguchi Y.
      • Tanikawa A.
      • Kosaki R.
      • Okano H.
      • Kosaki K.
      Novel overgrowth syndrome phenotype due to recurrent de novo PDGFRB mutation.
      ,
      • Takenouchi T.
      • Yoshihashi H.
      • Sakaguchi Y.
      • Uehara T.
      • Honda M.
      • Takahashi T.
      • et al.
      Hirschsprung disease as a yet undescribed phenotype in a patient with ARID1B mutation.
      ). Exome sequencing produced approximately 6 billion paired reads per sample that mapped to the human genome sequence assembly (hs37d5). The average coverage of the targeted exonic regions was ×115, and 99.0% of the targeted regions were covered with over ×10 reads on average.

       Single nucleotide polymorphism BeadChip analysis

      DNA amplification, labeling, and hybridization were performed according to the manufacturer’s instructions using the HumanCytoSNP-12 v2.1 DNA Analysis BeadChip Kit (WG-320-2103; Illumina, San Diego, CA). The array slides were scanned on an iScan system (Illumina), and log R ratios and B-allele frequencies were calculated and visualized using KaryoStudio data-analysis software (ver. 1.4; Illumina). The percent mosaicism was calculated from the difference between the average B-allele frequency value in the uniparental disomy region and the average B-allele frequency value in the biparental disomy region.

       Sanger sequencing and subcloning of the genomic DNA

      Genomic DNA was extracted from peripheral blood leukocytes and skin specimens using a Maxwell RSC Instrument and Maxwell RSC Blood DNA Kit (Promega, Madison, WI). Amplification of the genomic DNA and subsequent direct Sanger sequencing of MVD and MVK were performed using the primer pairs MVD_ex"n"F and R (n = 1–10) and MVK_ex"n"F and R (n = 2–11, Supplementary Table S3). To determine whether the congenital and acquired mutations were located on the same or on different chromosomes (Supplementary Figure S2), the genomic DNA, including both the congenital and the acquired mutations, was amplified by PCR using KOD-FX Taq (TOYOBO, Tokyo, Japan) with the primer pairs shown in Supplementary Table S4. The amplified DNA was subcloned into the pCR-Blunt vector (Invitrogen, Carlsbad, CA), and each clone was sequenced with the primers shown in Supplementary Table S4.

       PCR-based detection of the allele-specific mutations

      Genomic DNA was amplified via PCR with primers designed to include both the congenital and the acquired mutations in the amplicons (Supplementary Figure S2). The expected PCR amplicons and the amplicons aberrantly produced by PCR chimera are shown in Supplementary Figure S2. If the acquired and congenital mutations were located on different alleles, PCR amplicons possessing both mutations would rarely be detected because such amplicons would be the result of PCR chimera (Supplementary Figure S2). If the acquired and congenital mutations were located on the same allele, PCR amplicons possessing only the acquired mutation would rarely be detected because such amplicons would also be the result of PCR chimera (Supplementary Figure S2). Supplementary Table S2 lists the results.

       Digital PCR

      To estimate the percentage of second hit cells and naïve cells in the epidermis of each skin lesion, the rates of wild-type and mutant alleles were quantified using the QuantStudio 3D Digital PCR System comprising a ProFlex Base PCR system, including a chip adapter kit, an automatic chip loader, and the QuantStudio 3D Instrument (Thermo Fisher Scientific, Carlsbad, CA). The wild-type and mutant alleles of the c.746T>C mutation of MVD were detected using the primer pairs and TaqMan probes listed in Supplementary Table S5. The wild-type and mutant alleles of the c.575G>A and c.602C>T mutations of MVK were detected using the primer pairs and TaqMan probes shown in Supplementary Table S5. We prepared reaction mixtures containing 18 μl of 2× QuantStudio 3D Digital PCR Master Mix (Thermo Fisher Scientific), 0.9 μl of 40× TaqMan primer-probe mixture, as shown in Supplementary Table S5, and 1.8 μl diluted genomic DNA (5–10 ng/ml). We loaded 14.5 μl of each reaction mixture onto a QuantStudio 3D Digital PCR 20K v2 Chip (Thermo Fisher Scientific) according to the manufacturer’s instructions. Gene amplification was performed under the following conditions: 96 °C for 10 minutes, 39 cycles at 56 °C for 2 minutes and at 98 °C for 30 seconds, and a final 2-minute extension at 60 °C. After amplification, the chips were imaged on the QuantStudio 3D Instrument and analyzed with QuantStudio 3D AnalysisSuite cloud software (Thermo Fisher Scientific).

       Statistical analysis

      The percentage mosaicism was calculated from the ratio between the numbers of wild-type and mutant alleles. The percentage mosaicism was statistically compared among the epidermis of the central cylinder lesion, peripheral ring lesion, and normal skin adjacent to the lesion by one-way analysis of variance with multiple comparison tests using PRISM 6 software for Mac OS X (GraphPad Software, San Diego, CA).

       Web resources

      The URLs used to present the data herein are as follows:
      Japanese Multi Omics Reference Panel, https://jmorp.megabank.tohoku.ac.jp

      ORCIDs

      Conflict of Interest

      The authors state no conflict of interest.

      Acknowledgments

      We are grateful to all of the individuals who made invaluable contributions to this research. We thank Hiromi Kamura for the technical support and our colleagues in the Department of Dermatology, Keio University School of Medicine for the clinical support. This study was supported by a research grant from Maruho Co, Ltd , to the "Keio-Maruho Laboratory of Skin Barriology" and by research grants from the Japan Agency for Medical Research and Development under "Practical Research Project for Rare/Intractable Diseases Program" (Grant Number 18ek0109301s0201) and "Program for an Integrated Database of Clinical and Genomic Information" (Grant Number JP18kk0205002).

      Author Contributions

      Conceptualization: AK; Data Curation: AK, TS, KN; Formal Analysis: AK, TS, HS, KN; Funding Acquisition: AK, KH, YN, KK, MA; Investigation: AS, SA, SS, HF, TK, KN; Project Administration: AK; Resources: AK, NO, NUA, DY; Supervision: MA; Validation: AK, MA; Visualization: AK; Writing - Original Draft Preparation: AK; Writing - Review and Editing: AK, KN, MA

      Supplementary Material

      Figure thumbnail fx2
      Supplementary Figure S1Genetic changes in patients with linear porokeratosis LP1 and LP2. Genetic analyses of peripheral blood leukocytes (blood) and the epidermis and dermis of LS and non-LS of patients LP1 (a, b) and LP2 (c–e). (a) Sequencing reads of the c.127_128delCT and c.683G>A mutations of MVD via exome sequencing of the indicated samples of patient LP1 (see ). (b) Sanger sequencing chromatograms of the congenital c.127_128delCT mutation and acquired c.683G>A mutation of MVD in the indicated samples of patient LP1. Arrowheads, heterozygous c.127_128delCT mutation of MVD; arrows, wild-type c.683G of MVD; open arrowheads, mosaicism of c.683G>A mutation of MVD. (c) Sequencing reads of the congenital heterozygous c.746T>C mutation of MVD via next-generation sequencing of the indicated samples of patient LP2 (). (d) The heterozygous c.746T>C mutation of MVD (black arrowheads) and loss of heterozygosity (open arrowhead) in the indicated samples in patient LP2. (e) LRR and BAF of Chr16 in the lesional epidermis and dermis of lesional skin sample LS#1 of patient LP2. The somatic homologous recombination region is shown in pink in the BAF plot. Chr16 containing the congenital MVD mutation (black circle) and wild-type MVD are indicated by the red and yellow bars in the schematics, respectively. Gray bars, segments where no SNP was detected. BAF, B-allele frequency; Chr, chromosome; LP, linear porokeratosis; LRR, Log R ratio; LS, lesional skin; non-LS, non-lesional skin; SNP, single nucleotide polymorphism.
      Figure thumbnail fx3
      Supplementary Figure S2Schematic of PCR-based detection of allele-specific mutations. Green bar, chromosome with the wild-type allele; pink bar, chromosome with a congenital mutant allele; black oval, congenital mutation; red oval, de novo mutation; arrows indicate the locations and directions of primers used for subcloning.
      Figure thumbnail fx4
      Supplementary Figure S3Genetic analyses of the skin lesions of late-onset DSAP in patient DSAP1. (a) Sanger sequencing chromatograms of the congenital heterozygous c.746T>C mutation of MVD (arrowheads) in the indicated samples from patient DSAP1 (). (b) LOH of the c.746T>C mutation of MVD (open arrowheads) and the LRR and BAF of Chr16 in the indicated samples from patient DSAP1. The region of homologous recombination is indicated in pink in the BAF. The schematics show Chr16 with a homologous recombination (double-headed arrows) in a second hit cell (yellow bar, chromosome with the wild-type allele; red bar, chromosome with the allele of the congenital mutation [black circle]). (c) Second hit mutations in the indicated samples from patient DSAP1. Black arrowheads, transition mutations; black arrow, 1-bp deletion mutation; open arrows, small peaks of 1-bp frame-shifted nucleotides attributable to the c.682delC mutation; open arrowhead, c.746T>C mutation with a small peak of frame-shifted c.747C. BAF, B-allele frequency; bp, base pair; Chr, chromosome; DSAP, disseminated superficial actinic porokeratosis; LOH, loss of heterozygosity; LRR, log R ratio.
      Figure thumbnail fx5
      Supplementary Figure S4Digital PCR analyses of the loss of heterogeneity in a single annular lesion of porokeratosis in patient DSAP2. Quantification of the reference and mutant alleles via digital PCR for the c.746T>C mutation of MVD in lesional (LS #5) and non-lesional (Non-LS #1) skin of patient DSAP2. Representative plot of the digital PCR (left panels) and the number of copies/μl (right panel, mean ± SEM, n = 3). The data points are color-coded according to the following calls: FAM (blue), VIC (red), FAM + VIC (green), and NOT AMPLIFIED (yellow). DSAP, disseminated superficial actinic porokeratosis; LS, lesional skin; non-LS, non-lesional skin; SEM, standard error of the mean.
      Figure thumbnail fx6
      Supplementary Figure S5Digital PCR analyses of the loss of heterogeneity in a single annular lesion of porokeratosis in patient DSAP6. Quantification of reference and mutant alleles via digital PCR of the c.746T>C mutation of MVD in lesional (LS #2) and non-lesional (Non-LS #1) skin of patient DSAP6. Representative plot of the digital PCR (left panels) and number of copies/μl (right panel, mean ± SEM, n = 3). The data points are color-coded according to the following calls: FAM (blue), VIC (red), FAM + VIC (green), and NOT AMPLIFIED (yellow). DSAP, disseminated superficial actinic porokeratosis; LS, lesional skin; non-LS, non-lesional skin; SEM, standard error of the mean.
      Figure thumbnail fx7
      Supplementary Figure S6Digital PCR analyses of the two second hit mutations in a single annular lesion of porokeratosis in patient DSAP7. Quantification of the reference and mutant alleles via digital PCR for the (a) c.575G>A and (b) c.602C>T mutations of MVK in lesional and non-lesional skin of patient DSAP7. A representative plot of the digital PCR (upper panels) and the number of copies/μl (bottom panels, mean ± SEM, n = 3). The data points are color-coded according to the following calls: FAM (blue), VIC (red), FAM + VIC (green), and NOT AMPLIFIED (yellow). DSAP, disseminated superficial actinic porokeratosis; SEM, standard error of the mean.
      Supplementary Table S1Clinical Characteristics of the Subjects with Linear Porokeratosis (LP1 and LP2) and Disseminated Superficial Actinic Porokeratosis (DSAP1–7)
      SubjectAge (years), sexAge (years) at OnsetAge (years) at Biopsies for Genetic AnalysesDistribution of Skin LesionsAge (years) at Initial SCC OnsetFamilial/SporadicGene Mutation Detected in Peripheral Blood Leukocytes
      LP154, maleBirth50 and 53Unilateral linear on buttock, thigh, and inguinal region48SporadicHeterozygous MVD c.127_128delCT mutation
      LP224, femaleBirth20 and 24Unilateral linear on abdomen, back, buttock, thigh, and inguinal regionNDMother had disseminated lesionsHeterozygous MVD c.746T>C (p.F249S) mutation
      DSAP176, male6475Disseminated on the entire body, except the face; more prevalent on the extremitiesNDOne of his four siblings had disseminated lesionsHeterozygous MVD c.746T>C (p.F249S) mutation
      DSAP275, female6974Disseminated on the entire body except the faceNDSporadicHeterozygous MVD c.746T>C (p.F249S) mutation
      DSAP377, male30s76Disseminated on the entire body, except the face; more prevalent on the extremitiesNDFather and one of his siblings had disseminated lesionsHeterozygous MVD c.746T>C (p.F249S) mutation
      DSAP443, male20s42Disseminated predominantly on the extremities with some on the trunk and none on the faceNDSporadicHeterozygous MVD c.746T>C (p.F249S) mutation
      DSAP567, female6467Disseminated on the entire body except the faceNDSporadicHeterozygous MVD c.746T>C (p.F249S) mutation
      DSAP675, male1775Disseminated on the entire body except the face60Father and one of father's four siblings had disseminated lesionsHeterozygous MVD c.746T>C (p.F249S) mutation
      DSAP776, male6376Disseminated on the entire body including the faceNDSporadicHeterozygous MVK c.1073A>C (p.Q358P) mutation
      Abbreviations: DSAP, disseminated superficial actinic porokeratosis; LP, linear porokeratosis; ND, not detected; SCC, squamous cell carcinoma.
      Supplementary Table S2Analyses of the Congenital and Acquired Mutations in the Subcloned Genomic DNA of the Lesional Epidermis
      Skin LesionsNumber of Clones
      Including Neither of the MutationsIncluding Only the Congenital MutationIncluding Only the Acquired MutationIncluding Both MutationsUnreadable
      LP1 LS #11418601
      LP1 LS #31215302
      DSAP1 LS #32119401
      DSAP1 LS #5812901
      DSAP1 LS #6215903
      DSAP3 LS #15211301
      DSAP3 LS #37111411
      DSAP6 LS #50251401
      DSAP7 LS #13161300
      DSAP7 LS #31161500
      DSAP7 LS #50251601
      DSAP7 LS #60211100
      Abbreviations: DSAP, disseminated superficial actinic porokeratosis; LP, linear porokeratosis; LS, lesional skin.
      Supplementary Table S3Primers Used in Cloning and Sanger Sequencing
      Primer NameLocationDirectionPrimer Sequence
      MVD_g10415Fupstream of exon 1ForwardGACCTACGTCAGCAACCCATC
      MVD_ex1Fupstream of exon 1ForwardCGTCCATTGGCTGAGAGGTA
      MVD_ex1Rintron 1ReverseAGCTTGTCACGCGAAGGAG
      MVD_g14717Fintron 1ForwardCTGGAGTAGAAGCGATGAAGGAAT
      MVD_ex2Fintron 1ForwardACAGGTTGTGCGGTGAGAG
      MVD_ex2Rintron 2ReverseTGGTTCACTGTGGTCTCCAA
      MVD_ex3Fintron 2ForwardAGCTGAGACTGGCCTTTCC
      MVD_ex3Rintron 3ReverseCCAGCTGGTCATTGAGGTG
      MVD_ex4Fintron 3ForwardTCCCGGGTAAGAAACTCTCC
      MVD_ex4Rintron 4ReverseCAGCTGCAACTGGAAGATGA
      MVD_ex5Fintron 4ForwardGAAGCCCTGTCATCTGGAAA
      MVD_ex5Rintron 5ReverseCAACCACGCTTCTGTTCCTT
      MVD_ex6Fintron 5ForwardAAGGAACAGAAGCGTGGTTG
      MVD_ex6Rintron 6ReverseCATCTCTGCATTTGGCTCCT
      MVD_ex7Fintron 6ForwardAGGAGCCAAATGCAGAGATG
      MVD_ex7Rintron 7ReverseAGGGTGGAGCTTTCAGACAC
      MVD_ex8Fintron 7ForwardGGCAGACGTGTCAAGTACCA
      MVD_ex8Rintron 8ReverseGTGAGCGGCCTCTCTTTCTT
      MVD_ex9Fintron 8ForwardATAATCCTGCGGTGTTGAGG
      MVD_ex9Rintron 9ReverseGAGCCTGTGTCAGACTCACC
      MVD_ex10Fintron 9ForwardGTGGGTGTCACCACTCAGG
      MVD_ex10Rdownstream of exon 10ReverseGGGACCTCTCCTGACACCT
      MVD_ex10_1310Rdownstream of exon 10ReverseACCCACATGTCCCAGGAGT
      MVK_ex2Fintron 1ForwardTGGTACGTAGCAAGTGCTGG
      MVK_ex2Rintron 2ReverseGTGCCTCAGGGTGTCCTTTT
      MVK_ex3Fintron 2ForwardCTGTTCCCAGGCTTAGTGGG
      MVK_ex3Rintron 3ReverseAACACCCGGGAAATAGCTGG
      MVK_ex4Fintron 3ForwardCCATGTTCCAATTCCAGTGA
      MVK_ex4Rintron 4ReverseCATGATGAGGACAGCCAATG
      MVK_ex5Fintron 4ForwardTGATTCAGCATGAGTTCCTGA
      MVK_ex5Rintron 5ReverseATCCCACTCTGCCAGCACTA
      MVK_ex6_35Fintron 5ForwardGGCACGCTCCGTAGCTGGAGAGGTTCAGAGTGGAC
      MVK_ex6Fintron 5ForwardTCCGTAGCTGGAGAGGTTCA
      MVK_ex6Rintron 6ReverseAATCCATTGATTCGATTCCTC
      MVK_ex7Fintron 6ForwardAACTGAAGCTGGGCTGACTC
      MVK_ex7Rintron 7ReverseGCCACATTGGGGGATGAAGA
      MVK_ex8Fintron 7ForwardTCCAGTGTCACCTCTTGTGC
      MVK_ex8Rintron 8ReverseCCTATAACCCATCAGCCAGAG
      MVK_ex9_36Fintron 8ForwardGGCTCAGAAGAGAGGTGGTTTCCCCAGAGGATGGGC
      MVK_ex9Fintron 8ForwardGTGGTTTCCCCAGAGGATGG
      MVK_ex9Rintron 9ReverseTAGCTTCCGGGGGATTCTGA
      MVK_ex10Fintron 9ForwardGAAATGCTGCAGGGCAGAAG
      MVK_ex10Rintron 10ReverseTCCAGGTGGACCCCTCTTAG
      MVK_ex11Fintron 10ForwardGCAGGTAACCTTGGGCTTTA
      MVK_ex11_34Rdownstream of exon 11ReverseGGATGCAGAGGGCAGACCATGCCTCCCTAGGTCC
      MVK_ex11Rdownstream of exon 11ReverseAGACCATGCCTCCCTAGGTC
      Supplementary Table S4Primers Used for Subcloning of Genomic DNA and Subsequent Sanger Sequencing to Evaluate the Alleles of the 1st and 2nd Hits
      Skin Lesions

      (Exon Numbers for the Location of the 1st and 2nd Hits)
      Primers Used for CloningPrimers Used for Sequencing
      LP1-LS #1 and #3

      (exon 2 and 7)
      MVD_g14717F

      MVD_ex8R
      MVD_g14717F

      MVD_ex7R
      DSAP1-LS #3

      (exon 7 and 7)
      MVD_ex5F

      MVD_ex8R
      MVD_ex7R
      DSAP1-LS #5

      (exon 7 and 4)
      MVD_ex4F

      MVD_ex8R
      MVD_ex4F

      MVD_ex7R
      DSAP1-LS #6

      (exon 7 and 4)
      MVD_ex4F

      MVD_ex8R
      MVD_ex4F

      MVD_ex7R
      DSAP3-LS #1

      (exon 7 and 4)
      MVD_ex4F

      MVD_ex8R
      MVD_ex4F

      MVD_ex7R
      DSAP3-LS #3

      (exon 7 and 2)
      MVD_g14717F

      MVD_ex8R
      MVD_ex2F

      MVD_ex7R
      DSAP6-LS #5

      (exon 7 and 1)
      MVD_g10415F

      MVD_ex10_1310R
      MVD_ex1F

      MVD_ex7R
      DSAP7-LS #1

      (exon 11 and 6)
      MVK_ex6_35F

      MVK_ex11_34R
      MVK_ex11F MVK_ex6R
      DSAP7-LS #3

      (exon 11 and 9)
      MVK_ex9_36F

      MVK_ex11_34R
      MVK_ex11F MVK_ex9R
      DSAP7-LS #5

      (exon 11 and 11)
      MVK_ex9_36F

      MVK_ex11_34R
      MVK_ex11F MVK_ex9R
      DSAP7-LS #6

      (exon 11 and 9)
      MVK_ex9_36F

      MVK_ex11_34R
      MVK_ex11F MVK_ex9R
      Abbreviations: DSAP, disseminated superficial actinic porokeratosis; LP, linear porokeratosis; LS, lesional skin.
      Supplementary Table S5Primers and TaqMan Probes Used for Digital PCR Analyses
      AssayPrimers and TaqMan ProbesDirectionSequence
      MVD c.746T>C
      Primers
      MVD746_FForwardAGATGGCCCGCTGCAT
      MVD746_RReverseCTGGTTGCTGTCCTTCATGGT
      TaqMan probes
      MVD746_T_VICForwardVIC-AGCGAGACTTCCCCAGCT-MGB
      MVD746_C_FAMForwardFAM-AGCGAGACTCCCCCAGCT-MGB
      MVK c.575G>A
      Primers
      MVK575_FForwardACCTACCCCAGGTGCTGACA
      MVK575_RReverseGACCAAGGAGGATTTGGAGCTAA
      TaqMan probes
      MVK575_G_VICReverseVIC-TTCTCTCCCCTTGGAAG-MGB
      MVK575_A_FAMReverseFAM-CATTCTCTCCTCTTGGAAG-MGB
      MVK c.602C>T
      Primers
      MVK602_FForwardACCTACCCCAGGTGCTGACA
      MVK602_RReverseGACCAAGGAGGATTTGGAGCTAA
      TaqMan probes
      MVK602_C_VICForwardVIC-AACCCCTCCGGAGTGG-MGB
      MVK602_T_FAMForwardFAM-AACCCCTTCGGAGTGG-MGB
      Abbreviations: FAM, FAM fluorescent dye; MGB, Minor Groove Binder; VIC, VIC fluorescent dye.

      References

        • Abyzov A.
        • Mariani J.
        • Palejev D.
        • Zhang Y.
        • Haney M.S.
        • Tomasini L.
        • et al.
        Somatic copy number mosaicism in human skin revealed by induced pluripotent stem cells.
        Nature. 2012; 492: 438-442
        • Abyzov A.
        • Tomasini L.
        • Zhou B.
        • Vasmatzis N.
        • Coppola G.
        • Amenduni M.
        • et al.
        One thousand somatic SNVs per skin fibroblast cell set baseline of mosaic mutational load with patterns that suggest proliferative origin.
        Genome Res. 2017; 27: 512-523
        • Atzmony L.
        • Khan H.M.
        • Lim Y.H.
        • Paller A.S.
        • Levinsohn J.L.
        • Holland K.E.
        • et al.
        Second hit, postzygotic PMVK and MVD mutations in linear porokeratosis.
        JAMA Dermatol. 2019; 155
        • Brash D.E.
        UV signature mutations.
        Photochem Photobiol. 2015; 91: 15-26
        • Choate K.A.
        • Lu Y.
        • Zhou J.
        • Choi M.
        • Elias P.M.
        • Farhi A.
        • et al.
        Mitotic recombination in patients with ichthyosis causes reversion of dominant mutations in KRT10.
        Science. 2010; 330: 94-97
        • Choate K.A.
        • Lu Y.
        • Zhou J.
        • Elias P.M.
        • Zaidi S.
        • Paller A.S.
        • et al.
        Frequent somatic reversion of KRT1 mutations in ichthyosis with confetti.
        J Clin Invest. 2015; 125: 1703-1707
        • Commens C.A.
        • Shumack S.P.
        Linear porokeratosis in two families with disseminated superficial actinic porokeratosis.
        Pediatr Dermatol. 1987; 4: 209-214
        • Cui H.
        • Li L.
        • Wang W.
        • Shen J.
        • Yue Z.
        • Zheng X.
        • et al.
        Exome sequencing identifies SLC17A9 pathogenic gene in two Chinese pedigrees with disseminated superficial actinic porokeratosis.
        J Med Genet. 2014; 51: 699-704
        • Forsberg L.A.
        • Gisselsson D.
        • Dumanski J.P.
        Mosaicism in health and disease - clones picking up speed.
        Nat Rev Genet. 2017; 18: 128-142
        • Happle R.
        Somatic recombination may explain linear porokeratosis associated with disseminated superficial actinic porokeratosis.
        Am J Med Genet. 1991; 39: 237
        • Kitano Y.
        • Okada N.
        Separation of the epidermal sheet by dispase.
        Br J Dermatol. 1983; 108: 555-560
        • Knudson A.G.
        Two genetic hits (more or less) to cancer.
        Nat Rev Cancer. 2001; 1: 157-162
        • Kubo A.
        • Shiohama A.
        • Sasaki T.
        • Nakabayashi K.
        • Kawasaki H.
        • Atsugi T.
        • et al.
        Mutations in SERPINB7, encoding a member of the serine protease inhibitor superfamily, cause Nagashima-type palmoplantar keratosis.
        Am J Hum Genet. 2013; 93: 945-956
        • Laurie C.C.
        • Laurie C.A.
        • Rice K.
        • Doheny K.F.
        • Zelnick L.R.
        • McHugh C.P.
        • et al.
        Detectable clonal mosaicism from birth to old age and its relationship to cancer.
        Nat Genet. 2012; 44: 642-650
        • Mehta A.
        • Haber J.E.
        Sources of DNA double-strand breaks and models of recombinational DNA repair.
        Cold Spring Harb Perspect Biol. 2014; 6: a016428
        • Pfeifer G.P.
        • You Y.H.
        • Besaratinia A.
        Mutations induced by ultraviolet light.
        Mutat Res. 2005; 571: 19-31
        • Schamroth J.M.
        • Zlotogorski A.
        • Gilead L.
        Porokeratosis of Mibelli. Overview and review of the literature.
        Acta Derm Venereol. 1997; 77: 207-213
        • Sertznig P.
        • von Felbert V.
        • Megahed M.
        Porokeratosis: present concepts.
        J Eur Acad Dermatol Venereol. 2012; 26: 404-412
        • Suzuki S.
        • Nomura T.
        • Miyauchi T.
        • Takeda M.
        • Fujita Y.
        • Nishie W.
        • et al.
        Somatic recombination underlies frequent revertant mosaicism in loricrin keratoderma.
        Life Sci Alliance. 2019; 2
        • Takenouchi T.
        • Yamaguchi Y.
        • Tanikawa A.
        • Kosaki R.
        • Okano H.
        • Kosaki K.
        Novel overgrowth syndrome phenotype due to recurrent de novo PDGFRB mutation.
        J Pediatr. 2015; 166: 483-486
        • Takenouchi T.
        • Yoshihashi H.
        • Sakaguchi Y.
        • Uehara T.
        • Honda M.
        • Takahashi T.
        • et al.
        Hirschsprung disease as a yet undescribed phenotype in a patient with ARID1B mutation.
        Am J Med Genet A. 2016; 170: 3249-3252
        • Vilenchik M.M.
        • Knudson A.G.
        Endogenous DNA double-strand breaks: production, fidelity of repair, and induction of cancer.
        Proc Natl Acad Sci USA. 2003; 100: 12871-12876
        • Wang Y.
        • Masaki T.
        • Khan S.G.
        • Tamura D.
        • Kuschal C.
        • Rogers M.
        • et al.
        Four-dimensional, dynamic mosaicism is a hallmark of normal human skin that permits mapping of the organization and patterning of human epidermis during terminal differentiation.
        PLOS ONE. 2018; 13e0198011
        • White R.R.
        • Vijg J.
        Do DNA double-strand breaks drive aging?.
        Mol Cell. 2016; 63: 729-738
        • Zhang S.Q.
        • Jiang T.
        • Li M.
        • Zhang X.
        • Ren Y.Q.
        • Wei S.C.
        • et al.
        Exome sequencing identifies MVK mutations in disseminated superficial actinic porokeratosis.
        Nat Genet. 2012; 44: 1156-1160
        • Zhang Z.
        • Li C.
        • Wu F.
        • Ma R.
        • Luan J.
        • Yang F.
        • et al.
        Genomic variations of the mevalonate pathway in porokeratosis.
        ELife. 2015; 4: e06322