Epidermolytic palmoplantar keratoderma (EPPK) is caused by mutations in KRT9 and less often, KRT1. All known mutations in KRT9 have been found in regions of the gene encoding the conserved central α-helix rod domain. In the present study, we investigated the molecular basis of EPPK in a patient of Ashkenazi Jewish origin. The patient was found to carry a novel missense mutation in KRT9, resulting in the substitution of a poorly conserved leucine for valine at position 11 of the amino acid sequence. Despite its unusual location, the mutation was shown to be pathogenic through activation of a cryptic donor splice site, resulting in the deletion of 162 amino acids. The present data indicate the need to screen keratin genes in their entirety, as mutations altering domains of lesser functional importance may exert their deleterious effect at the transcriptional level.
Epidermolytic palmoplantar keratoderma (EPPK; OMIM #144200) is an autosomal dominant genodermatosis that develops within the first months after birth, and is characterized clinically by diffuse yellowish thickening of the skin on the palms and soles, with erythematous borders. Histologically, the disease features include epidermolytic hyperkeratosis, consisting of perinuclear vacuolization of the keratinocytes and large irregularly shaped keratohyaline granules located in the granular layers of the epidermis. EPPK is caused by mutations in the keratin 9 gene, KRT9, and in a minority of cases, in KRT1.
Keratin 9 is exclusively expressed in the suprabasal layers of palmoplantar epidermis and, like all other keratins, is composed of three major domains. The first 153 amino acids correspond to the keratin 9 head domain; amino acids 154–485 form the central α-helix rod domain, which is composed of four helical subsegments (1A, 1B, 2A and 2B) that are interrupted by three nonhelical linker domains (L1, L12 and L2); and the tail domain is comprised of amino acids 466–623.
All 27 deleterious mutations identified to date in KRT9 alter the 1A or 2B segments of the α-helix rod domain, which are highly conserved in all keratin proteins and are essential for the formation of the keratin heterodimer (http://www.interfil.org). In the present report, we describe a mutation located outside of these regions, causing a typical disease phenotype through activation of a cryptic donor splice site.
A 19-year-old man presented with congenital palmoplantar keratoderma. The patient was of Ashkenazi Jewish origin, and had been born in Russia. His medical and family histories were unremarkable, and he denied taking any medication.
On examination, diffuse yellowish thickening of the palmoplantar skin was seen, with no other dermatological findings (Figs 1a,b). The patient denied hyperhidrosis, and he had no nail or hair abnormalities.
On histological examination of a skin biopsy, epidermolytic changes were seen in the spinous layers of the epidermis (Fig. 2).
The patient and family were approached for genetic studies. The protocol was approved by our institutional review board and by the Israel National Committee for Human Genetic Studies, in accordance with the principles of the Declaration of Helsinki. Informed consent was obtained from all participants before enrolment.
Peripheral blood was obtained from the patient and his mother, and DNA was extracted. The entire coding regions of KRT9 and KRT1 were scrutinized for pathogenic alterations by direct sequencing (BigDye Terminator Cycle Sequencing Kit; Applied Biosystems, Foster City, CA, USA) on an automated sequencer (ABI 3100 Genetic Analyzer; PE Applied Biosystems) The oligonucleotide sequences and PCR settings are available upon request. No pathogenic sequence alteration was found in KRT1, but a heterozygous T>G transversion was seen at KRT9 cDNA position 31 (c.31T>G). This mutation is predicted to result in the substitution of a valine for a leucine residue at position 11 of the keratin 9 amino acid sequence (p.Leu11Val) (Fig. 3a). This mutation is located within a gene region coding for the head domain of the protein. The mutation was not found in a DNA sample obtained from the patient's mother. Other members of the family were unavailable for testing.
We then verified the c.31T>G mutation using a PCR restriction fragment length polymorphism assay. A PCR fragment 218 bp long was amplified from genomic DNA (primers shown in Table 1) The resulting amplicons were digested overnight with the DNA endonuclease BsaAI (New England Biolabs, Ipswich, MA, USA), which showed that the mutation was not present in a panel of 100 healthy population-matched controls (Fig. 3b).
Table 1. Primer sequences used in the study
cDNA, complementary DNA; RFLP, restriction fragment length polymorphism.
Conservation analysis showed that the Leu11 amino acid is poorly conserved across species (Fig. 3d). and the ConSurf server showed that the head domain of KRT9 has no homologue among other keratins. However, using the Berkeley Drosophila Genome Project, we found that the mutation is likely to activate a cryptic donor splice site. To confirm this hypothesis, total RNA was extracted from a skin biopsy (RNeasy Extraction Kit; Qiagen Inc., Valencia, CA, USA). cDNA was synthesized (Thermo Scientific Verso cDNA Synthesis Kit, ABgene, Surrey, UK) and amplified by PCR with exon-crossing primers (Table 1) located in exons 1 and 3 of the KRT9 gene. PCR amplification was performed with Taq polymerase Q (Qiagen Inc.). The reverse transcriptase PCR product was extracted from the gel, purified (QIAquick Gel Extraction Kit; Qiagen Inc.) and sequenced directly as described above. We identified an in-frame deletion 486 bp long located within exon 1 of KRT9 (c.31–516del486) (Fig. 3c). The deletion was found to result from the use of donor and acceptor splice sites located at positions 30 and 516 of the cDNA sequence, respectively. The mutation is predicted to result in the deletion of 162 amino acids (p.Leu11_Gln172del). Thus, we identified a novel EPPK-causing mutation located in a region of the KRT9 gene, encoding a poorly conserved region of the protein that is not routinely screened for diagnostic purposes. The mutation leads to a deletion 162 bp long, which includes part of the highly conserved 1A helical segment of the α-helix rod domain, where 20 of the 27 mutations known to cause EPPK are located (http://www.interfil.org).
At present, most of the human mutation databases contain data primarily derived from genomic DNA analysis. Because RNA is not routinely sequenced, most splice-site mutations are usually identified by prediction within classic consensus splice sites. As previously shown in several keratinopathies, disruption of native splice sites can lead to activation of cryptic splice sites.[4, 5] All other point mutations in gene coding sequences are usually scored as missense, nonsense or silent mutations. However, exonic mutations altering mRNA splicing have been described in several genetic disorders including neurofibromatosis, type 1 (OMIM #162200), cystic fibrosis (OMIM #219700), ataxia telangiectasia (OMIM #208900), familial breast–ovarian cancer (OMIM #604370) and Netherton syndrome (OMIM #256500). The present data further underline the importance of RNA analysis and full gene screening in the molecular analysis of typical phenotypes for which no mutations are found in routinely screened regions.
In conclusion, we report here the first example of a splice-site activating exonic mutation in KRT9, leading to EPPK.
EPPK is an autosomal dominant disorder associated with mutations in the KRT9 gene and less often, in the KRT1 gene.
All known mutations in KRT9 are located in gene regions encoding the 1A or 2B segments of the conserved α-helix rod domain of the protein.
We report here the first example of a splice-site activating exonic mutation in KRT9 leading to EPPK.
These results show the importance of RNA analysis and full gene screening in the molecular analysis of typical phenotypes for which no mutations are found in routinely screened regions.
We are very grateful to the family members for their participation in this study.
Most mutations in KRT9 have been found to affect:
The head domain of the protein
The central helical rod domain of the protein
The tail domain of the protein
Both the head and the tail domain of the protein
Both the head and the central helical rod domain of the protein
A 5- year old patient presents in your clinic with palmoplantar keratoderma. On histology, epidermolytic changes are evident. In which gene(s) are the mutations most likely to be found ?
KRT6a and KRT1
KRT6a and KRT16
Which type of palmoplantar keratoderma is likely to be seen in this patient ?
To identify the consequence of a splice-site mutation, what is the experimental approach most often used ?
Epidermolytic palmoplantar keratoderma has usually its onset during the: