If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Institute of Human Genetics, Medical Center–University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, GermanyFaculty of Biology, University of Freiburg, Freiburg, Germany
Trichilemmal cysts are common hair follicle–derived intradermal cysts. The trait shows an autosomal dominant mode of transmission with incomplete penetrance. Here, we describe the pathogenetic mechanism for the development of hereditary trichilemmal cysts. By whole-exome sequencing of DNA from the blood samples of 5 affected individuals and subsequent Sanger sequencing of a family cohort including 35 affected individuals, this study identified a combination of the Phospholipase C Delta 1 germline variants c.903A>G, p.(Pro301Pro) and c.1379C>T, p.(Ser460Leu) as a high-risk factor for trichilemmal cyst development. Allele-specific PCRs and cloning experiments showed that these two variants are present on the same allele. The analysis of tissue from several cysts revealed that an additional somatic Phospholipase C Delta 1 mutation on the same allele is required for cyst formation. In two different functional in vitro assays, this study showed that the protein function of the cyst-specific 1-phosphatidylinositol 4, 5-bisphosphate phosphodiesterase delta-1 protein variant is modified. This pathologic mechanism defines a monoallelic model of the two-hit mechanism proposed for tumor development and other hereditary cyst diseases.
). Trichilemmal cysts develop mostly on the scalp. In rare cases, ossification or the development of proliferating trichilemmal tumors has been observed (
), the underlying genetic mechanisms are still not understood. In 2005, the trait was localized to the TRICY1 locus on chromosome 3p21.2-24 without the detection of mutations in the candidate genes CTNNB1 and MLH1 (
). Seidenari and colleagues have proposed clinical diagnostic criteria for hereditary trichilemmal cysts and analyzed PTCH1 within the TRICY1 locus, but they could not find any disease association (
Phospholipase C delta 1 (PLCD1) lies within the TRICY1 locus and encodes 1-phosphatidylinositol 4, 5-bisphosphate phosphodiesterase delta-1 (PLCδ1), a member of the phospholipase C family. PLCδ1 plays an important role in intracellular calcium signaling via the hydrolysis of phosphatidylinositol 4, 5-bisphosphate (PIP2) into the second messengers diacylglycerol (DAG), especially 38:4 and 38:5 species (
Inositol lipid binding and membrane localization of isolated pleckstrin homology (PH) domains. Studies on the PH domains of phospholipase C delta 1 and p130.
), a catalytic center composed of the X and Y domain and a C-terminal calcium-binding C2 domain. PLCD1 is ubiquitously expressed in human tissues by two different transcript variants (NM_001130964.1 and NM_006225.3). Translation results in a 777 amino acid (NP_001124436.1)– or a 756 amino acid (NP_006216.2)–sized isoform, which differ only in the N-terminus. In this study, we refer to transcript variant 2 (NM_006225.3) and isoform 2 (NP_006216.2). Gain- and loss-of-function germline mutations in PLCD1 are associated with leukonychia totalis (MIM: #151600). This rare disease is characterized by complete whitening of the nail plate and is usually inherited in an autosomal dominant manner, but an autosomal recessive form has also been described (
Here, we identified the pathologic mechanism for the development of hereditary trichilemmal cysts. All the affected individuals analyzed carry a germline PLCD1 high-risk allele with variants c.1379C>T, p.(Ser460Leu) and c.903A>G, p.(Pro301Pro) (NM_006225.3, NP_006216.2). As we could show by two different functional assays, a second somatic PLCD1 mutation, usually c.2234C>T p.(Ser745Leu), on the same allele (in cis) is required for cyst formation.
Results
Whole-exome sequencing reveals a specific PLCD1 haplotype in individuals with hereditary trichilemmal cysts
Initially, blood-derived DNA from five individuals with hereditary trichilemmal cysts was analyzed by whole-exome sequencing. All the individuals shared heterozygous variants NC_000003.12:g.38010450T>C, NM_006225.3:c.903A>G, NP_006216.2:p.(Pro301Pro) (MAF = 0.27, rs9857730) and NC_000003.12:g.38009720G>A, NM_006225.3:c.1379C>T, NP_006216.2:p.(Ser460Leu) (MAF = 0.025, rs75495843) in PLCD1. To evaluate these variants in other affected individuals, exons 6 and 9 of PLCD1 were sequenced in a cohort of 12 Tunisian families with hereditary trichilemmal cysts (autosomal dominant mode of inheritance) and partially incomplete penetrance. In total, 35 affected and 25 unaffected individuals were analyzed. We detected the combination of the 2 PLCD1 variants in all 35 affected individuals and in 11 out of 25 (44%) unaffected individuals. It must be noted that variant c.1379C>T, p.(Ser460Leu) has always been reported in combination with c.903A>G, p.(Pro301Pro). In the 1000 Genome Project, this haplotype was detected with a frequency of 0.013. A Fisher exact test between affected and unaffected individuals demonstrated a highly significant correlation (P < 0.001) between the occurrence of the PLCD1 variants and the trichilemmal cysts.
Identification of rare cyst-specific somatic PLCD1 variants in trichilemmal cyst tissue
To evaluate an effect of the identified PLCD1 haplotype at RNA-level, cDNA from PLCD1 transcript 2 (NM_006225.3) from the tissue of five cysts was analyzed by reverse transcriptase–PCR and Sanger sequencing. We identified additional somatic PLCD1 variants in cystic tissue: variant NC_000003.12:g.38007810 G>A, NM_006225.3:c.2234C>T, NP_006216.2:p.(Ser745Leu) four times in a heterozygous state, and variants g.38008070G>A, c.2129C>T, p.(Ser710Phe) and g.38008067G>A, c.2132C>T, p.(Ser711Phe) heterozygous in one individual. These cyst-specific rare variants were detected neither in blood or buccal mucosa DNA nor in hair root cDNA (Figure 1a). In addition, we sequenced the genomic DNA extracted from the cystic tissue of 10 additional unrelated individuals and detected variant c.2234C>T, p.(Ser745Leu) in 9 individuals and c.2132C>T, p.(Ser711Phe) in 1 individual. The cyst donors were from Tunisia and Germany and of different ethnicities. The c.2234C>T, p.(Ser745Leu) variant could not be found in the available DNA samples from blood. A summary of the sequencing results is shown in Table 1. All three somatic mutations in trichilemmal cysts are not listed in the ExAC database or the 1000 Genomes Project. The somatic variants affect highly conserved serine residues, and only a duplication of serine at position 745 is described for one individual in the ExAC database. The variants are classified as potentially pathogenic by the PolyPhen-2, Mutation Taster, and Mendelian Clinically Applicable Pathogenicity scores. Because this study showed that trichilemmal cysts harbor additional somatic PLCD1 mutation(s), the next question to be addressed was whether these variants are present on the same allele as the inherited variants. To analyze the sequences of the two alleles on the RNA and cDNA level, allele-specific reverse transcriptase–PCRs of the C>T transmission at position c.2234 p.(Ser745Leu) were performed. In all five samples, the cyst-specific mutation c.2234T, p.(745Leu) occurred in cis with the inherited variants c.903G, p.(301Pro) and c.1379T, p.(460Leu), whereas the second allele showed wild-type variants at all positions (Figure 1c). The reverse transcriptase–PCR results were confirmed by the individual cloning of the two alleles from cyst-derived cDNA of patient 3. To obtain further insight into the allelic state on the cellular level, we isolated and cultivated keratinocytes from cystic material of patient 3 (process description in the Supplementary Materials and Methods online). DNA and RNA were extracted from these cells and analyzed in the same manner as the cystic tissue by DNA/cDNA sequencing and allele-specific PCR. The sequence electropherograms (Figure 1d) show no significant differences in the peak height of the two different variants at position c.2234. Allele-specific PCRs confirmed the monoallelic state of the three variants.
Figure 1Analysis of the somatic PLCD1 variant c.2234C>T, p.Ser745Leu. (a) Sequence electropherograms from buccal mucosa, hair roots, and two cysts of patient 4. The variant is present in both cysts but not in buccal mucosa and hair roots. (b) The p.Ser745 is highly conserved in different species. (c) Schematic overview of the allele-specific RT-PCRs with two different reverse primers ending at position c.2234. Sequencing of the PCR products revealed that the cyst-specific somatic variant c.2234T is located on the same allele (high-risk allele) as the germline variants c.1379T and c.903G, whereas the variants c.1379C and c.903 (wild-type allele) occur on the other allele with c.2234C (wild type). (d) Sequence electropherograms of cystic tissue and cyst-derived keratinocytes from patient 3. The electropherograms show no differences in the peak heights between wild type (C) and mutated variant (T) at position g.38007810 G>A, c.2234C>T. PLCD1, phospholipase C delta 1; RT-PCR, reverse transcriptase–PCR.
PLCδ1 protein expression in trichilemmal cyst tissue and normal human skin
To evaluate PLCδ1 protein expression, hematoxylin and eosin and immunohistological staining with paraffin-embedded sections and western blot analysis of skin and trichilemmal cyst samples (Figure 2) were performed. Hematoxylin and eosin staining of cystic tissue confirmed the trichilemmal state with the typical absence of a granular layer (Figure 2b) compared with that of normal human skin (Figure 2a). While western blot analysis revealed similar expression levels of PLCδ1 in the skin and trichilemmal cyst tissue, immunohistological staining showed that the tissue distribution of PLCδ1 differs between the skin (all epidermal layers) and trichilemmal cysts (primarily basal layer).
Figure 2PLCδ1 protein expression analyses of normal human skin and trichilemmal cysts. (a, b) Hematoxylin and eosin staining of normal human skin (a) and trichilemmal cyst of patient 3 (b). Trichilemmal cysts show an aberrant cellular structure compared with that of normal human skin. The cells differentiate from the outer basal layer (bottom of 2b) into the cyst lumen without the formation of a granular layer or stratum corneum. (c, d) Immunofluorescence staining of PLCδ1 (red, with blue DAPI nuclei counterstaining) of normal human skin (c) and the trichilemmal cyst of patient 3 (d). PLCδ1 is equally expressed in the different cellular layers of normal human skin, with the exception of the stratum corneum. Trichilemmal cysts primarily express PLCδ1 in the basal layer. (e) Western blot analysis of a commercially available human skin lysate and whole cellular protein lysate from the trichilemmal cyst of patient 3 confirmed the expression of PLCδ1 in the human skin and the trichilemmal cyst. Arrows indicate the direction of epidermal differentiation, and asterisks mark the basal cell layer. Scale bars = 50 μm. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PLCδ1, phospholipase C delta 1 protein.
Cyst-specific PLCδ1 variant affects in vitro TRPC4 channel activation and intracellular DAG concentrations
Finally, this study ascertained whether the pathologic mechanism of the cyst-specific PLCδ1 protein p.(Pro301Pro, p.(Ser460Leu), p.(Ser745Leu) is due to an altered physiological function. In heterologous expression systems, TRPC4 channels are activated by Gi coupled GPCRs via a PLCδ1-dependent mechanism (
). Therefore, Chinese hamster ovary cells transiently expressing TRPC4, GabaB (1a/2), and one of the PLCδ1 protein variants—wild-type, inherited high-risk allele (p.[Pro301Pro], p.[Ser460Leu]), cyst-specific variant (p.[Pro301Pro], p.[Ser460Leu], p.[Ser745Leu]) or a naturally nonexisting (artificial) variant with the somatic mutation p.(Ser745Leu)—were analyzed (Figure 3a). In contrast to the prominent TRPC4 activation in wild-type PLCδ1-expressing cells, channel activation in cells overexpressing the cyst-specific PLCδ1 (P < 0.01) or the artificial variant p.Ser745Leu (P < 0.01) was not observed. Similarly, a lipase-deficient PLCδ1 NP_006216.2:p.(His311Ala) (
). TRPC4 activation was only slightly reduced in cells overexpressing the PLCδ1 high-risk allele variant p.(Pro301Pro), p.(Ser460Leu) compared with that in the wild type. These results suggest an impaired function of the trichilemmal PLCδ1 protein variant. An alternative explanation could be a rather unspecific reduction of the basal cellular processes, that is, protein expression/trafficking in the cells that overexpress the trichilemmal PLCδ1 mutant. To evaluate this hypothesis, Kir3.1/3.2 channel activation was analyzed in an additional experiment. We expressed Kir3 channels, which are independently activated from PLCδ1 by Gi coupled GPCRs via Gby, instead of TRPC4 in Chinese hamster ovary cells. The results were similar to those observed in the TRPC4 experiment; a robust activation of Kir3 channels in the PLCδ1 wild-type–expressing cells were detected, but no significant channel activation was detected in cells that express the cyst-specific PLCδ1 variant (see Supplementary Figure S2 online). Based on these results, we cannot exclude the possibility that the observed effect of the cyst-specific PLCδ1 variant is independent of its enzymatic function. For that reason, we performed another, more direct assay by the measurement of intracellular concentrations of DAG in HEK293T cells that overexpress the PLCδ1 protein variants as described in the TRPC4 assay (Figure 3b). DAG is one of the reaction products of PIP2 hydrolysis generated by the enzymatic function of PLCδ1. Using this experimental setup, we identified a significant reduction of the DAG species that are primarily generated by PLCδ1-mediated PIP2 hydrolysis, DAG 38:4 and 38:5 (
), in cells overexpressing the cyst-specific PLCδ1 variant in comparison to the wild-type PLCδ1 (P < 0.05) expressing cells. The DAG concentrations in cells with the high-risk allele protein variant were not significantly reduced compared with those in cells with wild-type PLCδ1 protein. In contrast to the TRPC4 assay, we identified a significant difference between the cells that express the cyst-specific protein variant and the lipase-deficient variant. As an additional negative control, we used HEK293T cells that overexpress beta-galactosidase (LacZ). These cells showed similar DAG levels as cells overexpressing the lipase-deficient PLCδ1 variant. The overexpression of wild-type PLCδ1 or mutant variants in HEK293T cells did not affect the levels of the other DAG species (see Supplementary Figure S2C). Transfection controls for both functional assays are shown in Supplementary Figure S2. To investigate the impact of point mutations on the intracellular localization of PLCδ1 protein, we overexpressed wild-type and cyst-specific variant p.Pro301Pro, p.Ser460Leu, p.Ser745Leu in HEK293T cells and performed immunofluorescence costaining of FLAG-tagged PLCδ1, but we did not detect a significant difference in the intracellular localization (see Supplementary Figure S3 online).
Figure 3Functional analyses of cells overexpressing different PLCδ1 protein variants. (a) PLCδ1-dependent TRPC4 channel activation in CHO cells. Cells overexpressing the cyst-specific protein variant p.(Pro301Pro), p.(Ser460Leu), p.(Ser745Leu) and the hypothetic protein variant p.(Ser745Leu) showed a similar reduced activation of the TRPC4 channel as cells overexpressing a known lipase-deficient protein variant p.(His311Ala) compared with cells overexpressing the wild-type protein (**P ≤ 0.01, ns = not significant). Values (n ≥ 10) represent the means and standard errors of three independent experiments. (b) Concentration of DAG species 38:4 and 38:5 in HEK293T cells. Cells overexpressing the trichilemmal or the hypothetic protein variant showed significantly reduced DAG concentrations compared with cells overexpressing the wild-type protein (*P ≤ 0.05, **P ≤ 0.01, ns = not significant) but significantly higher DAG concentrations compared with cells overexpressing the lipase-deficient protein variant (#P ≤ 0.01). Values represent the means and standard deviations (n = 3) and are representative for three independent experiments. Both assays were analyzed via one-way ANOVA tests with subsequent post-hoc Bonferroni tests. ANOVA, analysis of variance; CHO, Chinese hamster ovary; DAG, diacylglycerol; HEK, human embryonic kidney; PLCδ1, phospholipase C delta 1 protein; TRPC4, transient receptor potential channel 4.
In this study, we present a mechanism for trichilemmal cyst formation induced by PLCD1 mutations. We detected PLCD1 variants c.903A>G, p.(Pro301Pro) and c.1379C>T, p.(Ser460Leu) in DNA from the blood of 35 affected individuals from 12 Tunisian families and 14 unrelated affected individuals with North African and Caucasian ethnicity, respectively. Variant c.1379C>T, p.(Ser460Leu) was only found in combination with c.903A>G, p.(Pro301Pro) in all the individuals analyzed and in the 1000 Genomes Project. In addition, this haplotype was significantly (P < 0.001) more often detected in affected than in unaffected individuals. Because the development of hereditary trichilemmal cysts involves incomplete penetrance and may also appear in persons older than 60 years, we consider this haplotype to be a PLCD1 high-risk allele for hereditary trichilemmal cysts. Based on the findings of cyst-specific somatic PLCD1 mutations in all the cysts analyzed (15) and that these mutations occur in cis, we assume that this somatic mutational event triggers cyst formation. Specifically, variant c.2234C>T, p.(Ser745Leu) appears to be a genetic driver for cyst development and a mutational hotspot in hair follicle because 13 out of 15 cyst samples analyzed from individuals with different ethnic backgrounds carried this variant. Moreover, two of these cysts developed on different areas on the scalp of patient 4 at the same time. Because c.2234C>T p.(Ser745Leu) was not detected in hair root–derived cDNA of this patient, we conclude that this mutation evolved independently twice in different hair follicle cells. We have no simple explanation why the somatic PLCD1 variants occur specifically on the high-risk allele, but a similar effect has been observed for myeloproliferative neoplasms in the JAK2 gene (
). Two different hypotheses (hypermutability and fertile ground hypothesis) have been proposed to explain this circumstance in the JAK2 gene, but it is still controversial (
The three somatic PLCD1 mutations detected (c.2234C>T, p.[Ser745Leu], c.2129C>T, p.[Ser710Phe], and c.2132C>T, p.[Ser711Phe]) concern highly conserved serine residues (Figure 1b) located in the C2 domain of PLCδ1. The C2 domain is important for the enzymatic activation of PLCδ1 through Ca2+-dependent binding to membrane-bound phosphatidylserine (
). Heterologous PLCδ1 overexpression experiments in mammalian cells led to reduced TRPC4 activation, as well as reduced intracellular DAG concentration levels, in cells overexpressing the cyst-specific protein variants p.Pro301Pro, p.Ser460Leu, and p.Ser745Leu compared with cells expressing the wild-type protein or the lipase-inactive variant. Based on these results, we suggest that the enzymatic activity of the cyst-specific PLCδ1 variant is impaired but not completely abolished. Speculating about the pathogenic effect, we suggest either a change in protein structure or the loss of an essential phosphorylation at serine residue 745. In mice, the evolutionary conserved serine residue at position 745 is partially phosphorylated (
). An effect of this phosphorylation onto protein function has not been tested, but, for example, the phosphorylation of serine at position 1164 in the C2 domain of PLCγ2 results in reduced Ca2+ signaling under specific conditions (
). It should be noted that the naturally nonexisting protein variant with the cyst-specific mutation c.2234C>T, p.(Ser745Leu) showed the same reduced activity as the cyst-specific protein, but this mutation never occurred independently of the high-risk allele. Because the mutation of one PLCD1 allele seems to be crucial for the development of trichilemmal cyst, we assume that the mutated protein has a dominant-negative effect on the functional wild-type protein and/or a dosage-dependent effect on PLCδ1 activity.
In Plcd1 knockout mice, it has been shown that the gene plays an important role in hair formation and skin development. As a result, knockout mice are nude because of hair loss and develop hair follicle–derived intradermal cysts. In contrast to trichilemmal cysts, these cysts show a skin-like differentiation (
). This issue confirms well-known differences in skin development/regulation between mice and humans.
In the autosomal recessive form of leukonychia totalis, protein-truncating PLCD1 mutations are described leading to the loss of function or reduced enzyme activity, whereas in the autosomal dominant disease form, the mutation NP_006216.2:p.Ala574Thr (ENST00000463876.5) results in an increased activity compared with the wild-type protein (
). The genetic mechanism of FLOTCH syndrome has not yet been clarified. However, the fact that leukonychia totalis and trichilemmal cyst arise together in affected individuals implies that PLCD1 mutations are probably involved in the development of both disorders in FLOTCH syndrome. Moreover, PLCD1 plays an important role in different tumors as a tumor-suppressor gene. It is epigenetically silenced through hypermethylation in breast cancer, gastric cancer, and chronic myeolid leukemia (
Phospholipase C delta 1 is a novel 3p22.3 tumor suppressor involved in cytoskeleton organization, with its epigenetic silencing correlated with high-stage gastric cancer.
In this study, we identified the pathologic mechanism of trichilemmal cyst development, which shows similarities to the two-hit hypothesis proposed by Knudson (
). According to Knudson, some types of cancer arise because of accumulation of mutations on both alleles of a tumor-suppressor gene. The first mutation is normally inherited (first-hit), and the second mutation occurs somatically on the other allele of the affected gene (second-hit). Alternatively, trichilemmal cyst development depends on an inherited PLCD1 high-risk allele and a somatic mutation in cis (second-hit) on the high-risk allele. Therefore, we propose a monoallelic model of the two-hit mechanism for trichilemmal cyst formation (Figure 4). We cannot completely exclude that there is a loss-of-heterozygosity event in PLCD1 on the single-cell level because of a mixture of normal and cystic cells in the analyzed cystic tissue, but sequencing of DNA and cDNA from cyst-derived keratinocytes showed no imbalances in the heterozygous state of PLCD1.
Figure 4Monoallelic two-hit model for trichilemmal cyst formation. (a) Comparison between the Knudson two-hit model for tumorigenesis and the proposed monoallelic two-hit model for trichilemmal cyst formation. As an alternative explanation, loss of heterozygosity could occur on the cellular level and induce cyst formation. X symbolizes mutations. (b) Detailed description of the trichilemmal cyst development mechanism. Affected individuals carry a hereditary PLCD1 “high-risk” allele as a first-hit. For cyst formation, a second somatic mutation (primarily c.2234C>T, p.Ser745Leu) has to occur on the same allele. PLCD1, phospholipase C delta 1.
Although the molecular mechanism is not yet completely elucidated, our data show that a combination of a PLCD1 high-risk allele and a somatic mutation on the same allele is the genetic trigger for trichilemmal cyst development.
Materials and Methods
Sample collection
A total of 60 DNA samples from Tunisian families with hereditary form of trichilemmal cysts, including 35 affected individuals, were initially collected. Additional blood and cyst samples (fresh and paraffin-embedded) were collected in Sfax and Freiburg for further analysis. DNA was extracted using standard phenol-chloroform procedure or Maxwell 16 extraction robot (Maxwell RSC DNA FFPE Kit, Promega, Mannheim, Germany) according to the manufacturer’s instructions.
Exome sequencing
Whole-exome sequencing of blood-derived DNA from five affected Tunisian individuals was performed using an AmpliSeq Exome Kit (Thermo Fisher Scientific, Waltham, MA) for library preparation and sequencing on the Ion Proton platform (Thermo Fisher). The whole process was conducted according to the manufacturer’s instructions. Base calling and sequence alignment were performed using Ion Torrent Suite software version 4.4.0.6 (Invitrogen Life Technologies, Gaithersburg, MD). Interpretation of single nucleotide variants was performed with the following online prediction tools: Mutation Taster (
Genomic DNA and allele-specific reverse transcriptase–PCRs were performed with Taq polymerase (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The primers and annealing temperatures are listed in Supplementary Table S1 online. The sequencing reactions were performed using a BigDye Terminator v1.1 Cycle Sequencing Kit (Thermo Fisher). Capillary electrophoresis was run on either a 3130 or 3500xL Genetic Analyzer (Thermo Fisher).
RNA extraction, cDNA synthesis, and PLCD1 transcript analysis
The RNA from cyst material and hair, including the roots, was extracted with a RotiQuick Kit (Carl Roth, Karlsruhe, Germany) according to the manufacturer’s instructions. The RNA quality was analyzed using a NanoDrop spectrophotometer (Thermo Fisher). At least 100 ng of RNA was used for reverse transcription with ProtoScript II Reverse Transcriptase (New England Biolabs, Beverly, MA) according to the manufacturer’s instructions, and a 1:1 mixture of random hexamer and Oligo dT15 primers. PLCD1 transcript analysis of cyst cDNA was performed by PCRs using Phusion Polymerase (Thermo Fisher) according to manufacturer’s instructions and subsequent Sanger sequencing. Primers and annealing temperatures are listed in Supplementary Table S1.
Histological analysis
The cyst material from patient 9 and normal human breast skin were fixed in 4% formaldehyde, gradually dehydrated with ethanol and embedded in paraffin (Paraplast Plus, McCormick Scientific, St. Louis, MO). Seven-micron sections of the biopsy samples were used for histological staining. The sections were gradually dewaxed and stained with hematoxylin and eosin Schnellfärbekit (Roth). These sections were dehydrated and mounted in Roti-Histokit media (Carl Roth). The images were recorded with a Carl Zeiss Axioskop 40 light microscope (Carl Zeiss, Jena, Germany) equipped with an AxioCam MRc5 camera and analyzed with AxioVision Rel. 4.6 software.
Immunofluorescence and western blot analysis
Immunofluorescence staining of the paraffin sections and western blot were performed as described by
with the following modifications: antigen retrieval of the staining was performed with Tris-EDTA buffer (pH 9) at 95 °C for 15 minutes. Primary anti-PLCδ1 antibody (ab154610, Abcam, Cambridge, England) was diluted 1:100 in phosphate buffered saline with Tween, and secondary antibody α-rabbit IgG AF 594 (ab150080, Abcam) was diluted 1:200 in phosphate buffered saline with Tween. Images were processed with ImageJ2 (Fiji/ImageJ2, Madison, WI) and Inkscape v.0.48 (inkscape.org).
For western blot, trichilemmal cyst tissue was homogenized with a cooled Douncer and dissolved in RIPA buffer. The applied whole-skin lysate was commercially available (Novus Biologicals, Littleton, CO). The primary anti-PLCδ1 antibody (sc-374329, Santa Cruz Biotechnology, Dallas, TX) was used in a 1:100 dilution, and the loading control anti-GAPDH (sc-47724, Santa Cruz) was used in a 1:200 dilution. For both primary antibodies, α-rabbit IgG (HRP) PI-1000 (Vector Laboratories, Burlingham, CA) was used as the secondary antibody in a 1:10,000 dilution.
Cloning of PLCD1 in pCMV6-AN-3DDK and pcDNA4-HisMax C
Cloning of the PLCD1 transcript (ENST00000334661) cDNA from cyst-derived material was performed using the Zero Blunt TOPO PCR Cloning Kit for Sequencing (Thermo Fisher) and chemically competent DH5α E. coli cells. PLCD1 wild-type and cyst-specific protein variant sequences were subcloned into mammalian pCMV6-AN-3DDK (OriGene, Rockville, MD), pcDNA4-His Max C (Thermo Fisher), and pET302/NT-His (Thermo Fisher) vectors (primer pairs listed in Supplementary Table S1). The ligated vectors were transformed into DH5α E. coli cells (Subcloning Efficiency DH5α Competent Cells, Thermo Fisher) and selected on Luria-Bertani agar plates with 100 μg/ml ampicillin. PLCD1 high-risk allele variants c.903A>G, p.(Pro301Pro), c.1379C>T, p.(Ser460Leu), c.2234C>T, p.(Ser745Leu) and variant c.931C>G, c.932A>T, p.(His311Ala) were generated by site-directed mutagenesis of wild-type or cyst-specific PLCD1 variant with the primers listed in Supplementary Table S1.
Electrophysiological assay TRPC4 and Kir 3.1/3.2 channel activation
Electrophysiological recordings from Chinese hamster ovary cells transiently transfected with either TRPC4 or Kir3.1/3.2 channels, GabaB (1a/2) combined with one of the cloned PLCD1 variants were performed in whole-cell-mode at room temperature (22 °C–24 °C). Detailed information about the procedure is provided in the Supplementary Materials and Methods online.
DAG cell concentration measurements with overexpressed PLCδ1 variants
HEK293T cells were transfected with pcDNA4-HisMax C plasmids encoding PLCδ1 variants according to the manufacturer’s instructions. Cell pellets were extracted as described by
using an ACQUITY-UPLC system (Waters Corporation, Milford, MA). DAG species were analyzed by selected reaction monitoring ([MNH4]+ to [RCOO+58]+) of the respective esterified fatty acid. Data acquisition was performed by MS Workstation (Bruker, Billerica, MA). Detailed information about the procedure is provided in the Supplementary Materials and Methods.
Statistics
GraphPad Prism version 7 (GraphPad Software, San Diego, CA) was used for all the statistical analyses. The correlation between the high-risk allele and the occurrence of trichilemmal cysts was tested using Fisher exact test. The functional assays were analyzed via one-way analysis of variance tests with subsequent post-hoc Bonferroni tests.
Study approval
The Tunisian family cohort DNA samples were collected in 2004. At that time, all the individuals had an explanatory discussion with a physician and gave oral informed consent for the genetic studies of hereditary trichilemmal cysts. Blood and cyst samples from patients (fresh and paraffin-embedded) were collected with the written informed consent of patients. Approval was provided by the Ethics Committee of the University of Freiburg (statement 161/16, April 26, 2016) and the Ethics Committee of Sfax, Tunisia (statement 0057/2017). The written consent is part of the respective ethics vote and is available in Sfax, Tunisia (Tunisian individuals) or in Freiburg, Germany (German individuals) with consent for the publication of anonymized case data. The research was conducted according to the guidelines of the Declaration of Helsinki.
Data availability statement
Datasets related to this article can be found at https://doi.org/10.17632/68bxdnx84t.1 hosted at an open-source online data repository hosted at Mendeley Data.
We would like to thank Bernd Rösler, Gabriele Grüninger, and Juna Leppert for their technical support; Kristin Technau-Hafsi for providing paraffin-embedded trichilemmal cyst samples; Marc-Alexander Rauschendorf for providing advice with respect to different methods and discussion of results; Bernd Fakler and Rudolf Zechner for providing advice and help regarding the functional assays; and Susan Cure for manuscript proofreading. We are very grateful to all the participants of the study who provided DNA samples and/or trichilemmal cyst material. The study was partially supported by a grant from the German Research Foundation, DFG (FI1767/3-1) and funding from the Austrian Science Fund (FWF Project P25944 to FPWR).
Whole PLCD1 transcript (ENST00000334661) cDNA from the cyst-derived material of patient 3 was amplified with the primer pair ATGGACTCGGGCCGGGACT and CTAGTCCTGGAGGGAGATCTTCACA using Phusion Polymerase (Thermo Fisher Scientifc, Waltham, MA). The blunt-end PCR fragments were ligated into a pCR 4Blunt-TOPO cloning vector using the Zero Blunt TOPO PCR Cloning Kit for Sequencing (Thermo Fisher) and chemically competent DH5α E. coli cells. The transformed cells were incubated overnight on Luria-Bertani agar plates (100 μg/ml ampicillin) at 37 °C. Ten clones were picked, cultivated overnight in liquid Luria-Bertani media containing ampicillin (12 μg/ml), and the plasmid DNA was extracted using QIAprep Miniprep Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. One microliter of each sample was directly used for Sanger sequencing as described for the PCR products.
PLCD1 wild-type and cyst-specific protein variant c.903A>G, p.(Pro301Pro), c.1379C>T, p.(Ser460Leu), c.2234C>T, p.(Ser745Leu) sequences from the pCR4-4Blunt-TOPO cloning vector were subcloned into mammalian pCMV6-AN-3DDK (OriGene, Rockville, MD), pcDNA4-His Max C (Thermo Fisher), and pET302/NT-His (Thermo Fisher) vectors. Sequences were amplified by PCR with Hot Start Q5 polymerase (New England Biolabs, Beverly, MA) and the primer pairs listed in Supplementary Table S1 online (annealing temperature: 65 °C, 35 PCR cycles, 80 second extension). PCR products were visualized on 1% agarose gels and purified using a Monarch DNA Gel Extraction Kit (New England Biolabs). The PCR products and vectors were subsequently digested for 1 hour at 37 °C with XhoI and AsiSI (pCMV6-AN-3DDK) or BamHI (pcDNA4-HisMax C) restriction enzymes (New England Biolabs) and purified with Monarch PCR & DNA Cleanup Kit (New England Biolabs). Ligation was performed with 150 ng vector DNA, 180 ng PCR product, and 5 units of T4 DNA ligase (Thermo Fisher). The ligated vectors were transformed into DH5α E.coli cells (Subcloning Efficiency DH5α Competent Cells, Thermo Fisher), and positive clones were selected on Luria-Bertani agar plates with 100 μg/ml ampicillin. The correct integration of the PCR products into the vectors was tested by sequencing of the whole insert. The PLCD1 high-risk allele variants c.903A>G, p.(Pro301Pro), c.1379C>T, p.(Ser460Leu), an artificial variant with the somatic mutation c.2234C>T, p.(Ser745Leu), and the lipase-deficient protein variants c.931C>G, c.932A>T, p.(His311Ala) were generated by site-directed mutagenesis PCR (annealing temperature: 60 °C, 18 PCR cycles, 9-minute extension) of the wild-type or the cyst-specific PLCD1 variant with the primers listed in Supplementary Table S1.
Electrophysiological assay TRPC4 and Kir 3.1/3.2 channel activation
Electrophysiological recordings from Chinese hamster ovary cells transiently transfected with the indicated constructs were performed in the whole-cell-mode at room temperature (22 °C–24 °C). Currents were recorded with an EPC9 amplifier, low-pass filtered at 3 kHz and sampled at 5–20 kHz. Recording pipettes were made from quartz glass and had resistances of 2–3 MOhm. The intracellular solution used to record the TRPC4 channels was composed of (in mM) 140 CsCl, 1 MgCl2, 0.05 EGTA, 10 HEPES, pH 7.2. The intracellular solution used to record the Kir3.1/3.2 channels was composed of (in mM) 32.5 KCl, 107.5 K gluconate, 4 MgATP, 0.6 NaGTP, 10 mM trisphosphocreatine, 5 EGTA, 10 HEPES, pH 7.2.The extracellular solution contained (in mM) 144 NaCl, 5.8 KCl, 0.9 MgCl2, 1.3 CaCl2, 0.1 NaH2PO4, 5.6 D-glucose, and 10 HEPES (pH 7.4). TRPC4 channel mediated currents because of the following voltage ramp protocol were continuously recorded: (i) 100 mV for 100 milliseconds, (ii) ramp from +100 mV to –100 mV within 300 milliseconds, and (iii) –100 mV for 100 milliseconds, interval 6 seconds. Kir3.1/3.2 channel mediated currents due to the following voltage ramp protocol were continuously recorded: (i) –80 mV for 100 milliseconds, (ii) ramp from –100 mV to +100 mV within 500 milliseconds, and (iii) –100 mV for 100 milliseconds, interval 3 seconds. Rapid application/removal of Baclofen (100 μM, dissolved in extracellular solution) was performed using the RCS-200 system (Biologic Science instruments, Claix, France).
DAG cell concentration measurements with overexpressed PLCδ1 variants
HEK293T cells were transfected with pcDNA4-HisMax C plasmids encoding PLCδ1 variants according to the manufacturer’s instructions. After cultivation of the cells in DMEM supplemented with 10% fetal bovine serum for 24 hours at 37 °C and 7% CO2 in a humidified atmosphere, the cell pellets were collected, extensively washed with ice-cooled phosphate buffered saline (PBS), and extracted as described by Matyash et al. (
). In brief, the samples were homogenized using two 6-mm steel beads on a Mixer Mill (Retsch, Haan, Germany); 2 × 15 seconds, frequency 30/s) in 700 μl methyl-tert-butyl ether/methanol (3/1, v/v) containing 500 pmol butylated hydroxytoluene, 1% acetic acid, and 100 pmol of internal standards (ISTD, 14:0-14:0 DAG; Avanti Polar Lipids, Alabaster, AL) per sample. Total lipid extraction was performed under constant shaking for 30 minutes at room temperature. After the addition of 140 μl dH2O and additional incubation for 30 minutes at room temperature, the samples were centrifuged at 1,000 × g for 15 minutes to obtain phase separation. Five hundred microliters of the upper, organic phase were collected and dried under a stream of nitrogen. The lipids were dissolved in 150 μl 2-propanol/methanol/dH2O (7/2.5/1, v/v/v) for UPLC-MS analysis. Chromatographic separation was modified according to
using an ACQUITY-UPLC system (Waters Corporation, Milford, MA), equipped with a Luna omega C18 column (2.1 × 50 mm, 1.6 μm; Phenomenex, Torrance, CA) starting a 20-minute linear gradient with 80% solvent A (methanol/dH2O, 1/1, v/v; 10 mM ammonium acetate, 0.1% formic acid, 8 μM phosphoric acid). The column compartment was maintained at 50 °C. An EVOQ Elite triple quadrupole mass spectrometer (Bruker, Billerica, MA) equipped with an ESI source was used for detection. The DAG species were analyzed by selected reaction monitoring ([MNH4]+ to [RCOO+58]+ of the respective esterified fatty acid; 15 eV, 50 milliseconds, 0.7 resolution Q1/Q3). Data acquisition was performed by an MS Workstation (Bruker). Data were normalized for recovery as well as extraction and ionization efficacy by calculating the analyte/ISTD ratios (AU). The DAG content was then normalized to the cellular protein content. Therefore, the cell pellets were solubilized in 0.3 N NaOH/0.1% SDS after lipid extraction at 65 °C overnight, and the protein content was determined using Pierce BCA reagent (Thermo Fisher Scientific) and BSA as standard.
Isolation and cultivation of cyst-derived keratinocytes
After surgery, the trichilemmal cyst was stored in DMEM media with high glucose, 25 mM HEPES+ 1% Antibiotic & Antimycotic without fetal calf serum (Thermo Fisher Scientific). For cell isolation, a trichilemmal cyst was opened with a sterile scalpel, and noncystic materials (blood vessels, dermal tissue, and cystic content) were removed. Subsequently, the remaining cyst material was washed five times with PBS buffer (8% Antibiotic & Antimycotic Solution + 2% ciprofloxacin) and cut into small pieces. These small pieces were transferred into a sterile Petri dish and incubated in 5 ml trypsin/EDTA (0.05 %/0.02%)-PBS solution for 1 hour. Trypsin was inactivated by the addition of 10% fetal calf serum solution. Afterward, the cystic pieces and the supernatant were centrifuged for 5 minutes at 300g. The supernatant was removed; the cystic pieces were washed with 10 ml PBS and centrifuged for 5 minutes at 300 × g. The supernatant was finally removed, and the cystic pieces were stored in 7 ml serum-free media (Thermo Fischer Scientific) with 1% penicillin or streptomycin, 0.5% ciprofloxacin at 37 °C, 5% CO2. The next day, the cystic pieces were washed as described before and cultivated in 7 ml serum-free media (Thermo Fischer Scientific) with 1% penicillin or streptomycin, 0.5% ciprofloxacin at 37 °C, 5% CO2. After 2 days, the cell clusters started to form and to proliferate.
Overexpression of PLCδ1 in HEK293T cells and subsequent Immunocytochemistry analysis
HEK293T cells were seeded on 2-well Lab-Tek II Chamber Slides (Thermo Fisher) and transfected with either pCMV6-AN-3DDK-PLCD1 wild type or pCMV6-AN-3DDK-PLCD1-c.903A>G-c.1379C>T-c.2234C>T. For cell transfection, 2.5 μg vector DNA and 6 μl Lipofectamine LTX Reagent (Invitrogen, Carlsbad, CA) were used according to the manufacturer’s instructions for Transfecting Plasmid DNA into HEK 293 Cells Using Lipofectamine LTX Reagent for 6-well plates. The cells were incubated at 37 °C, 5% CO2 for 21 hours. Afterward, the cells were fixed in 4% paraformaldehyde for immunocytochemistry analysis. The cell membranes were permeabilized in PBS with 0.5% Triton X 100. Because of interfering endogenous PLCδ1, overexpressed FLAG-tagged PLCδ1 proteins were incubated with a primary FLAG-tag antibody (F1804, Sigma-Aldrich, St. Louis, MO, 1:200 dilution) and secondary antibody α-mouse IgG Fab2 Alexa Fluor 488 Conjugate (Cell Signaling Technology, Boston, MA) 1:200 in phosphate buffered saline with Tween. Cell nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI, 1:1,000, Sigma-Aldrich). Images were taken with a Carl Zeiss Axio Imager 2 microscope (Carl Zeiss, Jena, Germany) equipped with an AxioCam MRc5 camera and the ZEN 2012 software and were processed with the freely available software ImageJ2 (Fiji/Image2, Madison, WI) and Inkscape v.0.48 (inkscape.org).
Supplementary Figure S1Pedigrees of Family 4 and Family 5. Both families show an autosomal dominant inheritance for trichilemmal cysts.
Supplementary Figure S2Additional data of the functional assays. (a) Transfection controls of overexpressed PLCδ1 protein variants for the two different assays. (b) Results of the electrophysiological Kir3.1/3.2 assay. The cyst-specific PLCδ1 variant showed a reduced activation of the Kir3.1/3.2 channels compared with the wild-type protein. (c) Concentrations of all analyzed DAG species in the HEK cells. DAG, diacylglycerol; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HEK, human embryonic kidney; PLCδ1, phospholipase C delta 1 protein; TRPC4, transient receptor potential channel 4.
Supplementary Figure S3Analysis of the intracellular PLCδ1 localization. (a–d) FLAG-tagged wild-type PLCδ1 (a, b) and cyst-specific variant p.Pro301Pro, p.Ser460Leu, p.Ser745Leu (c, d) overexpressed in HEK293T cells. Immunofluorescence costaining of FLAG-tag PLCδ1 (green) and cell nuclei (DAPI, blue) showed no differences in the intracellular localization between the wild-type and cyst-specific protein variant. Both variants are primarily localized in the cytoplasm and slightly membrane associated. HEK, human embryonic kidney; PLCδ1, phospholipase C delta 1 protein.
Phospholipase C delta 1 is a novel 3p22.3 tumor suppressor involved in cytoskeleton organization, with its epigenetic silencing correlated with high-stage gastric cancer.
Inositol lipid binding and membrane localization of isolated pleckstrin homology (PH) domains. Studies on the PH domains of phospholipase C delta 1 and p130.
Trichilemmal or “pilar” cysts are commonly found on the scalp and are derived from the outer root sheath of the hair follicle. Multiple trichilemmal cysts present in an autosomal dominant pattern of inheritance, yet the genetic mechanism has remained elusive. In this issue, Hörer et al. (2019) highlight predisposing variants in PLCD1 in such families and propose a monoallelic mutational mechanism that drives cyst formation.
31554-4&id=gr1.jpg){kind=link}
31554-4&id=gr2.jpg){kind=link}
31554-4&id=gr3.jpg){kind=link}
31554-4&id=gr4.jpg){kind=link}
31554-4&id=fx1.jpg){kind=link}
31554-4&id=fx2.jpg){kind=link}
31554-4&id=fx3.jpg){kind=link}