Kathrin A. Giehl1,2, Thomas Herzinger2, Hans Wolff1,2, Miklós Sárdy1,2, Tanja von Braunmühl2, Valérie Dekeuleneer3, Yves Sznajer4, Dominique Tennstedt3, Pascaline Boes4, Stefan Rapprich5, Nicola Wagner5, Regina C. Betz6, Markus Braun-Falco2, Tim Strom7, Thomas Ruzicka2 and Gertrud N. Eckstein7
1Center for Rare and Genetic Skin Diseases, Department of Dermatology, 2Department of Dermatology, Ludwig-Maximilian University, Munich, Germany, 3Department of Dermatology, 4Center for Human Genetics, Cliniques Universitaires St Luc, UCL, Brussels, Belgium, 5Department of Dermatology, Klinikum Darmstadt, Darmstadt, 6Institute of Human Genetics, University of Bonn, Bonn, and 7Institute of Human Genetics, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany
Punctate palmoplantar keratoderma (PPKP1; Buschke-Fischer-Brauer) is a rare autosomal dominant inherited skin disease characterized by multiple hyperkeratotic papules involving the palms and soles. Mutations have been found at 2 loci, on chromosomes 15q22–15q24 and 8q24.13–8q24.21. We recently identified mutations in 3 families, in the AAGAB gene on 15q, which encodes the alpha- and gamma-adaptin-binding protein p34. The current study examined 14 additional families, comprising a total of 26 affected individuals and identified 8 novel mutations in 9 families. In one family a mutation representing a known SNP that was present only in the affected individuals was found, and in 4 other families, previously reported mutations were found (1, 2). These results confirm the role of AAGAB in PPKP1. Our findings suggest that there is no correlation with age, but with mechanical factors. No additional obvious genotype–phenotype correlation was observed, even when comparing different types of mutations. Rather, identical genotypes presented a very broad interfamilial and intrafamilial variability of phenotypes. Key words: punctate palmoplantar keratoderma; PPKP1; AAGAB; phenotype-genotype correlation.
Accepted Nov 24, 2015; Epub ahead of print Nov 26, 2015
Acta Derm Venereol 2016; 96: XX–XX.
Kathrin Giehl, Center for Rare and Genetic Skin Diseases, Department of Dermatology, Ludwig-Maximilian-University Munich, Frauenlobstr. 9-11, DE-80337 Munich, Germany. E-mail: firstname.lastname@example.org
Punctate palmoplantar keratoderma type I (PPKP1; OMIM #148600), also called keratosis punctata palmoplantaris Buschke-Fischer-Brauer, was first described in 1910 by Buschke & Fischer (3) and Brauer (4). It is mostly an autosomal dominant inherited disease, but sporadic cases have also been described (5). It is characterized by multiple hyperkeratotic papules with central indentation irregularly distributed on the palms and soles, as opposed to the mostly diffuse hyperkeratinization observed in other palmoplantar keratodermas (6, 7). In mechanically irritated areas, confluent plaques can be found. The lesions usually develop in early adolescence, but can also occur later in life (5). The incidence of PPKP1 has not been extensively evaluated, but has been estimated at 1.17 per 100,000 inhabitants in Croatia (8). There have been reports in which punctate palmoplantar keratoderma was associated with the development of both early- and late-onset malignancies, including squamous cell carcinomas (9–11). Possible treatments include keratolytic ointments, mechanical reduction of the hyperkeratoses, and systemic retinoids.
Recently, the identification of pathogenic mutations in AAGAB, which encodes the alpha- and gamma-adaptin-binding protein p34, was described for the first time by the authors of the current paper and others (1, 12), after previous reports had mapped the critical region to chromosomes 15 and 8 (13–18).
The current study examined 14 additional families for mutations in AAGAB, and studied a possible genotype–phenotype relationship and the phenotypic variability within and between the families.
MATERIALS AND METHODS
All family members gave informed consent prior to participation. The study was approved by the institutional review board of the Ludwig-Maximilian University of Munich, and adhered to the principles of the Declaration of Helsinki. Blood samples were taken from all 33 investigated individuals. To establish the phenotype, the palms and soles of all individuals participating in this study were clinically assessed (mild: < 10 papules on hands and feet; moderate: < 50 papules on hands and feet; severe: ≥ 50 papules on hands and feet, elevated lesions, confluent hyperkeratosis). All patients were asked to report a subjective severity score of 0–10 describing the pain caused by the skin lesions.
PCR primers were used as described earlier (1). The coding region and the exon-intron junctions were sequenced by the Sanger method using Big Dye Terminator cycle sequencing kit v3.1 (Applied Biosystems/Life Technologies, Grand Island, NY, USA) and analysed on an ABI 3730 Genetic Analyzer (Applied Biosystems). Sequences were analysed for mutations with the Staden Package 2.0 (open source sequence analysis tool; http://staden.sourceforge.net). Segregation analysis was performed on all available family members to confirm co-segregation with the phenotype in all pedigrees.
Splice site variants were analysed in silico with CRYP-SKIP 9 (http://cryp-skip.img.cas.cz/) to predict the probability of exon skipping or cryptic splice site activation of the mutated allele. The other changes were studied using different programs to predict whether variants are deleterious. PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/bgi.shtml) classifies an amino acid substitution as probably damaging, possibly damaging, benign, or of unknown significance. SIFT (http://sift.jcvi.org/) predicts whether a change is tolerated or damaging. For mutations affecting the start codon, the ORF Finder (Open Reading Frame Finder; http://www.ncbi.nlm.nih.gov/projects/gorf/) from the US National Center for Biotechnology Information (NCBI) was used to check if there was an alternative open reading frame with sequence similarity to the AAGAB open reading frame. This tool identifies all open reading frames using the standard or alternative genetic codes.
Mutation analysis of AAGAB in 26 affected individuals with palmoplantar keratoderma from 14 families revealed 12 different mutations, 8 of which are novel. These changes include 3 mutations affecting the translation initiation site, 1 missense mutation, 3 frameshift mutations, 2 splice site mutations, and 3 nonsense mutations (Table SI1).
Segregation analysis was performed in all cases when related individuals joined the study, and co-segregation of the mutation with the disease was observed in all available families (Fig. S11).
Keratoses on the soles were more severe than those on the palms, especially at points of pressure. Phenotypic variability could be observed between, as well as within, families (Fig. 1). Of the 26 affected individuals, 17 were women and 9 were men. The mean onset of disease was 18.2 years. A detailed description of the phenotypes is given in Table SII1.
Fig. 1. Clinical images of phenotypic variation within one family and influence of water exposure and mechanical irritation. Moderate disseminated plane papules on (a) palm and (b) soles of 83-year-old index patient (65325, family 1) with punctate palmoplantar keratoderma (PPKP1). The patient’s palm (a) before and (c) after 5 min of water exposure shows an increased prominence of papules. (d) The nephew (72382) has a very mild phenotype, which was more obvious at (e) the tips of the fingers, probably as a result of mechanical irritation associated with playing guitar for many years.
The following genotype variants are described in Appendix S11:
This study examined 14 families with PPKP1 and identified 8 previously unknown mutations in 9 families, 3 known mutations in 4 families, and a mutation, representing a known SNP in one family. In addition, using molecular genetic analysis, the aim of this study was to establish a genotype–phenotype correlation and examine for inter- and intra-familiar phenotypic variability.
To date, a total of 38 mutations have been reported by several authors in families with PPKP1 Buschke-Fischer-Brauer (2, 12, 19–27), as summarized in Fig. 2. The mutations occurred in a variety of different populations, including individuals of European, Asian, African, Jewish, Mexican and Canadian origin. Among these mutations, 17 were frameshift mutations, 10 nonsense mutations, 6 mutations affecting the start codon, 4 splice site alterations, and 1 missense mutation. Except for exon 3, where only frameshift mutations have been reported, the other mutation types occurred in almost every coding exon of AAGAB, with the exception of exons 9 and 10 (Fig. 2). This broad range of mutation types and distribution reflects the high variability of clinical phenotypes among families, but does not explain the intra-familiar differences. Although we compared the clinical phenotypes of all patients with different types of mutations, no correlation between the severity of the phenotype and the type or location of the mutation could be identified.
Fig. 2. Schematic representation of AAGAB showing mutation sites associated with punctate palmoplantar keratoderma (PPKP1). (A) Mutations in PPKP1 cases identified so far. (B) Genomic structure of AAGAB. Ten exons are indicated as black bars. The non-coding parts of exons 1 and 10 are indicated with a smaller bar (gene annotation from RefSeq NM_024666.3; chromosomal location 15q22.33–15q23). (C) Rab-like GTPase domain (P-loop_NTPase) and adaptin-binding domain in the protein sequence NP_078942 (length = 315 amino acids). (D) Newly identified mutations in AAGAB in 10 families with PPKP1. The different mutations types are indicated with coloured boxes (blue = frameshift, green = affecting start codon, red = nonsense, purple = splice site, yellow = missense).
This phenotypic variability within families leads to the assumption that one or more modifier genes interact with AAGAB or with the gamma-adaptin and alpha-adaptin subunits involved in clathrin-coated vesicle trafficking, thus influencing the final phenotype.
In our study more women were affected than men (17 women, 9 men) although PPKP1 is inherited in an autosomal dominant manner. One explanation could be that more women are concerned about their skin and therefore more often seek help from a doctor.
The spectrum of the clinical phenotypes between families, within families, and among affected individuals varied extensively. The phenotypes ranged from very mild forms that were recognized only after another family member with a more severe phenotype was diagnosed, to very severe forms in which the entire sole was affected with such thick hyperkeratosis that the patient required surgery.
We found that the same mutation could lead to different phenotypes (Fig. 1) even in the same family (families 1, 4, 6). An example of clinical variability is shown in family 1 in Fig. 1. In family 4, the index patient had a moderate phenotype, and according to the patient’s history, the parents were not affected. For segregation analysis, the genotype of the parents was also examined, which revealed that the father also had the same genotype. A thorough skin examination uncovered only one lesion, which would otherwise have been interpreted as a small wart, but was now interpreted as a minimal affected phenotype. In family 6, the phenotype varied from severe to very mild and clinically almost absent.
The missense mutation in family 6, which is a known SNP with a minor allele frequency of 0.001 (dbSNP; http://www.ncbi.nlm.nih.gov/SNP/), is more likely a causal mutation than a rare polymorphism. Due to the low incidence and the possible late onset of the disease, it is very likely that this mutation leads to PPKP1 and that this disease is more common than reported. The segregation analysis also supports this assumption, as all affected members exhibit this mutation. Although this mutation was excluded in unaffected relatives of family 6, it will be necessary to detect this mutation in other independent patients to define its pathogenicity.
Four of the 26 affected individuals (66351 [family 3], 71005 [family 4], 76427 [family 6], 72608 [family 9]), all of whom are from different families, would not have been recognized as PPKP1 without the typical index patients. Mild forms of PPKP1 might often be overlooked or under-reported, and therefore the prevalence of this disease may be underestimated.
The phenotype showing increased prominence of papules upon exposure to water (Fig. 1) has so far been diagnosed as keratosis palmoplantaris maculosa seu papulosa, as in the case of our patient (65325, family 1) and in the literature (28), hereditary chronic translucent papules of the palms and soles (29) or papulotranslucent acrokeratoderma (30, 31). In these cases genetic investigation of the AAGAB gene will be necessary to confirm that these are variants of PPKP1.
Also, no correlation was found between the severity of the clinical phenotype and the subjective severity score (Table SII1). For example, all 4 affected individuals of family 6 reported relatively high severity scores, although 2 of them had only a mild or moderate phenotype. This demonstrates impressively how even seemingly mild phenotypic expression of PPKP1 can have a profound impact on the patients’ quality of life.
In 4 of 26 patients, there was no family history of palmoplantar keratoderma. Because of the phenotypic variability, it is still possible that these sporadic cases have a familial background. In our cases, it was not possible to elucidate this, as the other family members could not be examined clinically or by molecular genetics.
These observations suggest that environmental factors, such as manual labour, hobbies, walking longer distances, as well as the frequency of mechanical treatment, may influence the phenotypic differences within a family.
The results of the analysis of 26 individuals from 14 families reported in this study, and those of other analyses reported by others, indicate that no clear phenotype–genotype correlation exists in PPKP1. Our findings suggest that there are no correlations with age, but with mechanical factors, and that there is an underestimation of the disease frequency. Differential diagnosis includes calluses due to pressure points, viral warts and porokeratosis punctate palmaris et plantaris and keratoses of toxic or paraneoplastic origin. As some entities of palmoplantar keratoderma are not distinct, they might often be misdiagnosed before the molecular era. The penetrance of PPKP1 seems to be highly variable. Regardless of the genotype, phenotypes vary considerably in PPKP1 and can even mimic other clinical entities of palmoplantar keratoderma.
PPKP1 Buschke-Fischer-Brauer is an example of how genetic characterization can lead to the regrouping of clinical entities and to the definition of new entities.
The authors would like to thank the patients and their families for their participation in this study. The authors also thank R. Feldmann for technical assistance in mutation analysis, and Diana Lingk for technical assistance with the photographic images. RCB is the recipient of a Heisenberg Professorship from the German Research Counsil (DFG).
Electronic database information. The URLs of databases and predictive software used herein are as follows: Online Mendelian Inheritance in Man (OMIM): http://www.omim.org. UCSC Human Genome Browser: http://genome.cse.ucsc.edu/. SIFT: http://sift.jcvi.org/. PolyPhen-2: http://genetics.bwh.harvard.edu/pph2/bgi.shtml. CRYP-SKIP: http://cryp-skip.img.cas.cz/. dbSNP: http://www.ncbi.nlm.nih.gov/SNP/.