The spectrum of cutaneous lymphomas in Japan: a study of 62 cases based on the World Health Organization Classification.

BACKGROUND: The relative incidence of malignant lymphoma subtypes differs according to geographic location. This study investigated the epidemiology of cutaneous lymphoma subtypes in Japan and compared it with other countries. METHODS: Sixty-two patients with cutaneous lymphoma attending the Department of Dermatology, National Hospital Organization Hokkaido Cancer Center were reviewed. The World Health Organization classification of hematopoietic and lymphoid malignancies was adopted. RESULTS: Of the 62 patients, 31 had primary cutaneous lymphoma (PCL) and 31 had secondary cutaneous lymphoma (SCL). T- and natural killer (NK)-cell lymphoma accounted for 80% of PCL, of which, mycosis fungoides accounted for almost 35%. Of the 31 patients with secondary cutaneous lymphoma, 17 patients (54%) had T- and NK-cell lymphoma, including nine adult T-cell leukemia/lymphoma patients, and 14 patients (46%) had B-cell lymphoma, including 11 diffuse large B-cell lymphoma patients. The majority of patients with SCL and NK-cell lymphoma with primary or secondary skin lesions had a poor outcome. CONCLUSIONS: PCL in this study showed a similar incidence to that of other institutions in Japan, while also demonstrating different frequencies from that of other countries, suggesting that the relative frequency of different PCL subtypes differ according to geographical location, similar to previous reports of systemic malignant lymphoma.

Erratum to: Genetic studies of 20 Japanese families of dystrophic epidermolysis bullosa.

In Table 1 and in Result and discussion section, C2875F should read C2876F.

Genetic studies of 20 Japanese families of dystrophic epidermolysis bullosa.

Dystrophic EB (DEB) is clinically characterized by mucocutaneous blistering in response to minor trauma, followed by scarring and nail dystrophy, and is caused by mutations in the COL7A1 gene encoding type VII collagen. DEB is inherited in either an autosomal dominant (DDEB) or recessive (RDEB) fashion. DDEB basically results from a glycine substitution mutation within the collagenous domain on one COL7A1 allele, while a combination of mutations such as premature stop codon, missense, and splice-site mutations on both alleles causes RDEB. In this study, mutation analysis was performed in 20 distinct Japanese DEB families (16 RDEB and four DDEB). The result demonstrated 30 pathogenic COL7A1 mutations with 16 novel mutations, which included four missense, five nonsense, one deletion, two insertion, one indel, and three splice-site mutations. We confirmed that Japanese COL7A1 mutations were basically family specific, although three mutations, 5818delC, 6573 + 1G > C, and E2857X, were recurrent based on previous reports. Furthermore, the Q2827X mutation found in two unrelated families would be regarded as a candidate recurrent Japanese COL7A1 mutation. The study furthers our understanding of both the clinical and allelic heterogeneity displayed in Japanese DEB patients.

Identification of COL7A1 alternative splicing inserting 9 amino acid residues into the fibronectin type III linker domain.

Type VII collagen is the major component of anchoring fibrils within the cutaneous basement membrane zone. The large amino-terminal noncollagenous domain of type VII collagen interacts with various extracellular matrix proteins and contributes to the dermal-epidermal attachment. The purpose of this study was to detect alternative splicing of COL7A1 transcript encoding the noncollagenous 1 domain. The alternative splicing in this region may affect interactions of the noncollagenous 1 domain with extracellular matrix proteins and also dermal-epidermal adhesion. Thus we examined expression of the alternative splicing in situations relating to wound healing and skin remodeling that required dermal-epidermal binding and detachment. Amplification of overlapping cDNA from keratinocytes using reverse transcription-polymerase chain reaction identified alternative splicing, which was generated by a different exon 18 acceptor site 27 bp upstream from the common acceptor site. Expression of this alternatively spliced transcript differed among several cell types. The nine amino acid residues GPLTLPLSP from the 27 bp nucleotides were inserted into the linker of fibronectin type III domains. This insertion was suggested to contribute to flexibility of the linker of fibronectin type III domains and may affect the interactions between the noncollagenous 1 domain and extracellular matrix proteins. Treatment with transforming growth factor-beta 1, which is known to promote wound healing and skin remodeling, enhanced the expression of this 27 bp transcript. Furthermore, keratinocyte biopsies from the wound edge of patients with epithelizing skin ulcers showed a significant increase in the 27 bp transcript expression compared with normal keratinocytes from steady-state body sites. These results suggest that amino acid variation of this alternative splicing may have some role in dermal-epidermal adhesion, wound healing, and skin remodeling. To the best of our knowledge, this is the first evidence of alternative splice insertion of a small peptide into the linker region of the fibronectin type III domains, a common motif within modular proteins.

The majority of keratinocytes incorporate intradermally injected plasmid DNA regardless of size but only a small proportion of cells can express the gene product.

The expression of intradermally injected DNA by keratinocytes is found mainly in the upper and middle layers of the epidermis. To investigate the mechanism of this selective expression, we observed the sequential changes in the distribution of interleukin-6-expressing keratinocytes after the introduction of the interleukin-6 gene. Transgene expression first occurred in basal keratinocytes and subsequently expanded to all epidermal layers and then remained in the upper layers. Semiquantitative analysis indicated that keratinocytes in the lower layers incorporated and lost DNA earlier than those in the upper layers. In order to examine the effect of the DNA size on the transgene expression, we constructed a plasmid containing a full-length 9 kb cDNA of type VII collagen and introduced it into keratinocytes. The expression pattern of type VII collagen in the epidermis was the same as those for smaller genes. This suggests that plasmid size has little or no effect on the expression pattern of the transfected gene. To trace the introduced plasmid, we intradermally injected a green fluorescence protein expression plasmid coupled with a rhodamine flag. Almost all keratinocytes in the injected areas showed rhodamine fluorescence. Furthermore, some cells also expressed green fluorescence protein. A lack of rhodamine fluorescence in the nucleus suggested an impairment of plasmid DNA transport from the cytoplasm to the nucleus. Collectively, our results show that the majority of keratinocytes take up the intradermally injected DNA regardless of its size, but that the transfer of DNA from the cytoplasm to the nucleus is limiting the transgene expression.

Dominant and recessive compound heterozygous mutations in epidermolysis bullosa simplex demonstrate the role of the stutter region in keratin intermediate filament assembly.

Keratin intermediate filaments are important cytoskeletal structural proteins involved in maintaining cell shape and function. Mutations in the epidermal keratin genes, keratin 5 or keratin 14 lead to the disruption of keratin filament assembly, resulting in an autosomal dominant inherited blistering skin disease, epidermolysis bullosa simplex (EBS). We investigated a large EBS kindred who exhibited a markedly heterogeneous clinical presentation and detected two distinct keratin 5 mutations in the proband, the most severely affected. One missense mutation (E170K) in the highly conserved helix initiation peptide sequence of the 1A rod domain was found in all the affected family members. In contrast, the other missense mutation (E418K) was found only in the proband. The E418K mutation was located in the stutter region, an interruption in the heptad repeat regularity, whose function as yet remains unclear. We hypothesized that this mutated stutter allele was clinically silent when combined with the wild type allele but aggravates the clinical severity of EBS caused by the E170K mutation on the other allele. To confirm this in vitro, we transfected mutant keratin 5 cDNA into cultured cells. Although only 12.7% of the cells transfected with the E170K mutation alone showed disrupted keratin filament aggregations, significantly more cells (30.0%) cotransfected with both E170K and E418K mutations demonstrated keratin aggregation (p < 0.05). These transfection assay results corresponded to the heterogeneous clinical findings of the EBS patient in this kindred. We have identified the first case of both compound heterozygous dominant (E170K) and recessive (E418K) mutations in any keratin gene and confirmed the significant involvement of the stutter region in the assembly and organization of the keratin intermediate filament network in vitro.

Exclusion of COL7A1 mutation in Kindler syndrome.

We describe a patient with Kindler syndrome with an 18-year follow-up who was initially misdiagnosed as suffering from dystrophic epidermolysis bullosa. The patient's skin showed broad reticulate labeling for collagen VII and reduplication of the lamina densa. Screening of this patient's DNA excluded any pathogenic COL7A1 mutations.