- Email: [email protected]
Abnormal keratin expression in circumscribed palmar hypokeratosis
Abnormal keratin expression in circumscribed palmar hypokeratosis Akira Ishiko, MD, PhD,a Itaru Dekio, MD, PhD,a,c Atsushi Fujimoto, MD,a Kaori Kameyama, MD, PhD,b Mitsuo Sakamoto, PhD,c Yoshimi Benno, DVM, PhD,c Masayuki Amagai, MD, PhD,a and Takeji Nishikawa, MD, PhDa Tokyo and Saitama, Japan Background: Circumscribed palmar or plantar hypokeratosis (CPH) is a rare skin disorder only recently described. Objective: To determine the diagnostic features and to provide insight into the pathogenesis of CPH, with analysis of two new Japanese cases. Methods: Dermoscopy, immunohistochemistry, electron microscopy, polymerase chain reaction amplification for human papillomavirus (HPV) DNA and 16S microbial rRNA gene profiling were conducted. Results: Dermoscopy showed characteristic features using both dry and jelly immersion observation; steplike desquamation and a homogeneous erythema with regularly distributed whitish spots. Immunohistochemistry revealed strong staining with anti-pankeratin antibody (AE11AE3) and anti-keratin 16 antibody, and decreased expression of keratin 2e. EM revealed a breakage of the corneocytes within their cytoplasm, but structures for cell attachment were intact. HPV and lesion-specific bacteria were not detected. Limitations: The number of cases analyzed was two. Conclusion: Hyperproliferative epidermal state along with enhanced corneocyte fragility may account for the unique features in CPH. ( J Am Acad Dermatol 2007;57:285-91.)
C
ircumscribed palmar or plantar hypokeratosis (CPH) was first described by Perez et al1 in 2002 as a disease showing asymptomatic, well-circumscribed, and depressed erythema present for many years on the palms or soles. Its main pathologic feature shows a characteristic sharp change resembling a depressed step-off or pit from the thick horny layer to the thin horny layer in the
From the Departments of Dermatologya and Pathology,b Keio University School of Medicine, Tokyo, and Microbe Division, Japan Collection of Microorganisms, RIKEN BioResource Center, Saitama.c Supported in part by Grants-in Aid for Scientific Research (C) (18591258) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Conflicts of interest: None declared. Accepted for publication February 18, 2007. Reprint requests: Akira Ishiko, MD, PhD, Department of Dermatology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku, Tokyo 160-8582, Japan. E-mail: [email protected]. Published online June 6, 2007. 0190-9622/$32.00 ª 2007 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2007.02.035
Abbreviations used: CPH:
circumscribed palmar or plantar hypokeratosis EM: electron microscopy HPV: human papillomavirus PCR: polymerase chain reaction rRNA: ribosomal ribonucleic acid
involved skin. Several reports have been published with increasing frequency since the first report,2-12 and the disease has now become recognized worldwide as a distinct disease entity. The pathogenesis of this disease, however, has not been definitely established. The original paper suggested that CPH is a benign clonal epidermal malformation1; however, the tendency for progressive enlargement2 suggests a dynamic proliferative disorder, although without any known tendency toward oncogenic transformation. Other potential etiologies include a primary keratinizing disorder7 localized to the granular and horny layers, trauma,10 and human papillomavirus (HPV) infection of keratinocytes.8 Herein we report 285
286 Ishiko et al
J AM ACAD DERMATOL AUGUST 2007
Fig 1. Case 1. Clinical and histologic features. A, A 17- 3 12-mm, well-demarcated, slightly depressed erythema with irregular border was seen on thenar site of left palm of a 58-year-old Japanese woman. B, Histology demonstrated thick horny layer demarcated by depressed thin horny layer of the lesional area. Lesional epidermis showed slight acanthosis. C, Corneocytes became eosinophilic and frayed at the edge and horny layer sharply changed the thickness. D, Lymphoid cell infiltration was seen around dilated capillaries in superficial dermis.
new Japanese cases, which were analyzed by dermoscopy, immunohistochemistry, electron microscopy (EM), HPV polymerase chain reaction and 16S microbial ribosomal RNA (rRNA) gene profiling, to search for the diagnostic features and to provide further insight into the pathogenesis of CPH.
CASE REPORTS Case 1 In 2000, a right-handed 58-year-old Japanese woman presented with an erythematous eruption on the left palm that had existed for 3 years. The lesion was solitary, showed asymptomatic erythema, which had gradually enlarged. Physical examination revealed a 17- 3 12-mm, well-demarcated, and slightly depressed erythema with an irregular border on the thenar aspect of her left hand. The initial clinical diagnosis was acute eczema, but various topical therapies including steroid, 20% urea, maxacalcitol, or vitamin A were ineffective during the 5-year follow-up period before performing a skin biopsy. Culture from the scale revealed two colonies of Staphylococcus epidermidis. When a biopsy
specimen was taken from the edge of the erythema, the lesion had slightly enlarged to 17 3 14 mm (Fig 1, A). Case 2 In 2001, a 73-year-old Japanese woman presented with multiple erythematous eruptions on her left palm, which had persisted for more than 10 years. The erythema was asymptomatic, but had enlarged very slowly and the number of affected areas had increased. There were two nail-sized depressed areas of erythema on the thenar part of the palm of her left hand. In addition, two similar, but smaller areas of erythema were seen around them (Fig 2, A). Dermoscopic views without using jelly showed stairlike desquamation at the edge of the lesion and this may be characteristic of CPH (Fig 2, B). With jelly immersion, homogeneous, structureless erythema was seen with regularly distributed whitish spots, which were suggestive of acrosyringeum (Fig 2, C). Treatment with vitamin A ointment, calcipotriol, maxacalcitol, 10% urea, zinc sulfate, steroid, tacrolimus, and suprofen were all disappointing. Culture from the scale was negative.
J AM ACAD DERMATOL
Ishiko et al 287
VOLUME 57, NUMBER 2
Fig 2. Case 2. Clinical and histologic features. A, Two depressed areas of erythema on thenar site of left palm of a 73-year-old woman. B, Dermoscopic view without using jelly showed stairlike desquamation at the edge of the lesion. C, Dermoscopic views using jelly immersion showed homogeneous, structureless erythema with regularly distributed whitish spots, which were suggestive of acrosyringeum. D, Lesional epidermis showed psoriasiform acanthosis.
Histopathology There was an abrupt decrement in the thickness of the stratum corneum, although in the context of both a preserved epidermal thickness and a granular cell layer that was only slightly diminished (Figs 1, B, and 2, D). A curious and distinctive feature was the frayed quality of the stratum corneum adjacent to the zone of hypokeratosis, most reminiscent of a staircase (Fig 1, C). There was a morphologic change in the corneocytes that defined the lateral margins of the hypokeratotic zone. In particular, the corneocytes more superficially manifested a basophilic and vacuolated appearance, while those in the lower portion exhibited hypereosinophilia. A faint perivascular lymphocytic infiltration was seen around dilated capillary in the superficial dermis (Fig 1, D). These features were shared in both cases. Immunohistochemistry Using commercially available polyclonal antibodies against HPV (Dakocytomation, Glostrup, Denmark), we failed to find any specific labeling in the epidermal keratinocyte in either case. To determine whether the thin horny layer is due to hypoproliferation of epidermis, Ki-67 (MIB-1,
Dakocytomation) was stained. The results showed that Ki-67epositive basal cells were more frequently seen in the lesional epidermis than in the nonlesional epidermis (Fig 3, A) in both CPH cases. To characterize the lesional keratinocytes, several antikeratin antibodies were used, including 34-bE12 (Enzo Diagnostics, Farmingdale, NY), AE11AE3 (Dakocytomation), CAM 5.2 (Becton Dickinson, San Jose, Calif), cytokeratin 2e (clone: Ks2.342.7.1, Progen Biotechnik, Heidelberg, Germany), cytokeratin 7 (clone:OV-TL-12/30, Dakocytomation), cytokeratin 9 (clone Ks9.701Ks9.216, Progen Biotechnik), cytokeratin 16 (clone SS025, Novocastra Laboratories Ltd, Newcastle, UK), and cytokeratin 20 (clone:Ks20.8, Dakocytomation). AE11AE3 (Fig 3, C) and cytokeratin 16 (Fig 3, E) showed prominent suprabasilar staining of lesional keratinocytes in contrast to its virtual absence in nonlesional skin. Conversely, cytokeratin 2e manifested a diminution in staining in the lesional zone relative to nonlesional skin (Fig 3, F ). There was no difference in the extent of positive staining with cytokeratin 9 between lesional and nonlesional skin. Both lesional and nonlesional skin failed to show reactivity with CAM 5.2, cytokeratin 7 and 20.
288 Ishiko et al
J AM ACAD DERMATOL AUGUST 2007
Fig 3. Case 2. Imunohistochemstry. A, Ki-67epositive basal cells were seen more frequently beneath lesional epidermis than in nonlesional epidermis. B, Anti-pankeratin 34bE12 showed no difference between lesional and nonlesional epidermis. C, Anti-pankeratin AE11AE3 showed strong positive staining at suprabasal keratinocytes restricted to lesional epidermis; this contrasted with the significantly weaker staining in perilesional skin. D, Acrosyringeal epithelium was spared from overexpression of AE11AE3. E, Cytokeratin16 was positive at suprabasal keratinocytes only in the lesion. F, Cytokeratin 2e expression was decreased in lesional suprabasal keratinocytes.
Electron microscopy Ultrastructural examination showed signs of corneocyte splitting through the cytoplasm with intact cell-cell attachment at the edge of the lesion where the horny layer sharply changes thickness (Fig 4, A and B). Vacuolar alteration of corneocytes involving the lower edges of the zone of hypokeratosis was seen (Fig 4, C). HPV PCR Since a recent report suggested an involvement of HPV type 4 in CPH skin, we performed PCR for HPVspecific DNA using paraffin-embedded tissue. DNA was extracted from 10-m thick paraffin sections and direct sequencing with general primers GP5 and GP6
were performed as described previously.8 As a result, we could not detect any HPV DNA in either case. 16S microbial rRNA analysis (terminal restriction fragment length polymorphism) It is reported that several uncultured bacterial species exist in human skin surfaces,13 and such unknown bacteria may cause the loss of the horny layer. Therefore we employed a culture-independent molecular analysis to search for causative microorganisms. A sample was taken by scrubbing the lesional or nonlesional skin with sterile swabs from case 2. 16S ribosomal gene, which was present in all known bacteria, was amplified with universal primers, FAM-27F (fluorescent-labeled) and 1492R
Ishiko et al 289
J AM ACAD DERMATOL VOLUME 57, NUMBER 2
Fig 5. 16S microbial rRNA gene profile (terminal RFLP) from the skin surface swabs. 16S ribosomal gene, which was present in all known bacteria, was amplified with universal primers. PCR products, which should contain most bacterial genes from the skin surface, were compared between (A) lesional and (B) nonlesional skin using restriction enzyme cutting length. Several bands were detected including band derived from P acnes. There was no difference between lesional skin and nonlesional skin. Fig 4. Case 2. EM of edge of lesion. A, At edge of lesion where the horny layer showed a sharp change in thickness, corneocytes showed intracytoplasmic splitting. B, Cell-cell detachment was not apparent and corneodesmosomes, cornified cell envelope, and keratin patterns were normal. C, At bottom of depressed area, outer layer of corneocytes contained many vacuoles of variable size. Bars = 1 m.
(nonlabeled). 59-Fluorescent-labeled PCR products, which should contain most bacterial genes on the skin surface, were compared between lesional and nonlesional skin using fluorescent-labeled restriction enzyme cutting length.13 As a result, several bands were detected, including a band derived from Propionibacterium acnes (Fig 5). Other bands indicated the presence of unidentified bacteria that cannot be detected by culture. There was no difference between lesional band patterns and nonlesional band patterns, that is, there were no lesionspecific bacteria.
DISCUSSION Herein we report two new Japanese cases with CPH, one of which had multiple lesions. To date, 28 cases have been reported in the literature and the
Table I. Twenty-eight reported cases of CPH Age (y) of first visit Male/female Predilection site
Right/left Solitary/multiple Suspected etiologic factors
35-89 (mean, 63.5) 5/23 Thenar (palm) (17/27) Hypothenar (palm) (7/27) Thumb (4/27) Sole (2/27) 16/14 24/4 Clonal epidermal proliferation, trauma, bacteria, HPV-4
clinical characteristics were summarized in Table I. CPH occurs mainly in elderly women and the predilection site was the thenar palm, which is consistent with our cases. No apparent deviation between the right and left hand was seen, and most cases occur in a solitary fashion. The diagnostic features of CPH are summarized in Table II. Using dermoscopy with non-jelly immersion technique, stair-like desquamation at the edge of the lesion was seen that was compatible with the characteristic pathological findings. With jelly immersion, homogeneous
290 Ishiko et al
J AM ACAD DERMATOL AUGUST 2007
Table II. Diagnostic findings of CPH Clinical features Dermoscopy Histopathology Immunohistochemistry Electron microscopy
Well-demarcated, depressed erythema persisting for years on palms or soles Without jelly; erythema with stair-like desquamation with jelly; structureless erythema with regularly distributed whitish spots Abrupt decrement in thickness of stratum corneum with frayed edge Ki-67", CK16", AE11AE3", CK2e# Breakage of corneocytes, vacuolization of corneocyte
structureless erythema was seen with regularly distributed whitish spots. The erythema likely represents mild inflammation with dilated capillaries in the superficial dermis, and the whitish spots may represent sparing of the acrosyringeum from excoriation in the thin horny layer. Thus dermoscopic observation faithfully reflects the histologic CPH characteristics and may have some diagnostic value in CPH. By EM, corneocytes showed cytoplasmic breakage or cleavage, and there was no significant detachment between the cells. This may indicate that the loss of the horny layer may be due to the fragility of the lesional corneocyte and does not arise from a cell adhesion disturbance. Despite the thinning of the horny layer, the malpighian layer in the lesion was hypertrophic. MIB-1 proliferation marker Ki-67 antibody stained increased cell numbers in the lesional basal layer. Anti-pankeratin antibody AE11AE3 showed strong immunostaining in the lesional suprabasal keratinocytes, and this may be useful for the diagnosis of CPH. This pattern of staining is also seen in epidermis overlying scars in case with epidermolysis bullosa14 and lichen sclerosus15 and regarded as proliferative changes, which might be associated with minor trauma. By testing the cytokeratins recognized by AE11AE3, but not by 34bE12 antibody, we found hyperexpression of keratin 16 in the lesion; this may be responsible for the strong staining of AE11AE3. Keratin 16 is expressed in the proliferative epidermis; therefore hyperexpression of keratin 16 is consistent with increased MIB-1epositive basal cells and epidermal hypertrophy. This proliferative phenotype may be a reactive one against some sort of irritation resulted from the early loss of the horny layer or may in itself be pathogenic by leading altered keratinization. In fact, keratin 2e expression was decreased in the lesion and this might be a reason of corneocyte fragility, although further biological investigation will be required. Neither of our cases showed any evidence of HPV infection using immunohistochemistry and HPV DNA amplification techniques, indicating that HPV is, at least, not the major cause of CPH. Like pitted keratolysis, some bacterial infection may
cause the loss of the horny layer. 16S Microbial rRNA gene is present in all known bacteria, and its conserved region is suitable for amplification. The length of the restriction enzymeedigested fragments may be used to identify specific bacteria and this method enables us to detect bacteria that cannot be cultured. Although we have detected several kinds of bacterial gene using this method, there was no difference of microbial components between samples from the lesional skin and perilesional normal skin. This indicated there were no disease sitee specific pathogen candidates. In conclusion, CPH still remains an enigma with regard to pathogenesis. It is certainly well defined from a clinical and light microscopic perspective. Our study suggests that a hyperproliferative epidermal state along with enhanced corneocyte fragility may account for its unique features. Specific triggers, such as trauma, microbes and HPV, cannot be implicated in every case. We thank Dr Yuuichiro Yamasaki for referring a patient, Dr Almut Bo¨er for performing HPV PCR, and Dr James R. McMillan for proofreading this manuscript.
REFERENCES 1. Perez A, Ru¨tten A, Gold R, Urbina F, Misad C, Izquierdo MJ, et al. Circumscribed palmar or plantar hypokeratosis: a distinctive epidermal malformation of the palms or soles. J Am Acad Dermatol 2002;47:21-7. 2. Obermoser G, Zelger B. ‘‘Multifocal’’ circumscribed palmar hypokeratosis: malformation or not? J Am Acad Dermatol 2003;49:1197-8. 3. Burnett JW, Harvey VM. Circumscribed palmar or plantar hypokeratosis: report of two additional cases. J Am Acad Dermatol 2004;51:843. 4. DeBloom JR, Ting W, Stone MS, Arpey CJ. Circumscribed palmar hypokeratosis. J Am Acad Dermatol 2004;51:319-21. 5. Ru¨tten A, Wecker-Brosi H, Gruhlke G, Kutzner H, Cirne de Castro L, Requena L. [Circumscribed acral hypokeratosis]. Hautarzt 2004;55:1060-3. German. 6. Mensing C, Schleusner V, Sander C, Mensing H. Circumscribed palmar or plantar hypokeratosis: two cases of a recently described entity of unknown origin. Am J Dermatopathol 2005;27:247-9. 7. Urbina F, Misad C, Gonzalez S. Circumscribed palmar hypokeratosis: clinical evolution and ultrastructural study after prolonged treatment with topical calcipotriol. J Eur Acad Dermatol Venereol 2005;19:491-4.
J AM ACAD DERMATOL
Ishiko et al 291
VOLUME 57, NUMBER 2
8. Bo¨er A, Falk T. Circumscribed palmar hypokeratosis induced by papilloma virus type 4. J Am Acad Dermatol 2006;54:908-9. 9. Lee SE, Kim YC, Kim SC. Circumscribed palmar or plantar hypokeratosis: report of a Korean case and published work review. J Dermatol 2006;33:403-5. 10. Resnik K, DiLeonardo M. Circumscribed palmar hypokeratosis: new observations. Am J Dermatopathol 2006;28:112-6. 11. Shirai A, Seki M, Hino H. Circumscribed palmar or plantar hypokeratosis. Practical Dermatology (jpn) 2005;27:71-4. 12. Murata M, Saitou Y, Mitsuto I. A case of circumscribed palmar hypokeratosis. Jpn J Clin Dermatol (jpn) 2006;60:27-9. 13. Dekio I, Hayashi H, Sakamoto M, Kitahara M, Nishikawa T, Suematsu M, et al. Detection of potentially novel bacterial
components of the human skin microbiota using cultureindependent molecular profiling. J Med Microbiol 2005;54: 1231-8. 14. Smoller BA, McNutt NS, Carter M, Gottlieb AB, Hsu A, Krueger J. Recessive dystrophic epidermolysis bullosa skin displays a chronic growth-activated immunophenotype. Implications for carcinogenesis. Arch Dermatol 1990;126:78-83. 15. Carlson JA, Ambros R, Malfetano J, Ross J, Grabowski R, Lamb P, et al. Vulvar lichen sclerosus and squamous cell carcinoma: a cohort, case control, and investigational study with historical perspective; implications for chronic inflammation and sclerosis in the development of neoplasia. Hum Pathol 1998;29: 932-48.
C
ircumscribed palmar or plantar hypokeratosis (CPH) was first described by Perez et al1 in 2002 as a disease showing asymptomatic, well-circumscribed, and depressed erythema present for many years on the palms or soles. Its main pathologic feature shows a characteristic sharp change resembling a depressed step-off or pit from the thick horny layer to the thin horny layer in the
From the Departments of Dermatologya and Pathology,b Keio University School of Medicine, Tokyo, and Microbe Division, Japan Collection of Microorganisms, RIKEN BioResource Center, Saitama.c Supported in part by Grants-in Aid for Scientific Research (C) (18591258) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Conflicts of interest: None declared. Accepted for publication February 18, 2007. Reprint requests: Akira Ishiko, MD, PhD, Department of Dermatology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku, Tokyo 160-8582, Japan. E-mail: [email protected]. Published online June 6, 2007. 0190-9622/$32.00 ª 2007 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2007.02.035
Abbreviations used: CPH:
circumscribed palmar or plantar hypokeratosis EM: electron microscopy HPV: human papillomavirus PCR: polymerase chain reaction rRNA: ribosomal ribonucleic acid
involved skin. Several reports have been published with increasing frequency since the first report,2-12 and the disease has now become recognized worldwide as a distinct disease entity. The pathogenesis of this disease, however, has not been definitely established. The original paper suggested that CPH is a benign clonal epidermal malformation1; however, the tendency for progressive enlargement2 suggests a dynamic proliferative disorder, although without any known tendency toward oncogenic transformation. Other potential etiologies include a primary keratinizing disorder7 localized to the granular and horny layers, trauma,10 and human papillomavirus (HPV) infection of keratinocytes.8 Herein we report 285
286 Ishiko et al
J AM ACAD DERMATOL AUGUST 2007
Fig 1. Case 1. Clinical and histologic features. A, A 17- 3 12-mm, well-demarcated, slightly depressed erythema with irregular border was seen on thenar site of left palm of a 58-year-old Japanese woman. B, Histology demonstrated thick horny layer demarcated by depressed thin horny layer of the lesional area. Lesional epidermis showed slight acanthosis. C, Corneocytes became eosinophilic and frayed at the edge and horny layer sharply changed the thickness. D, Lymphoid cell infiltration was seen around dilated capillaries in superficial dermis.
new Japanese cases, which were analyzed by dermoscopy, immunohistochemistry, electron microscopy (EM), HPV polymerase chain reaction and 16S microbial ribosomal RNA (rRNA) gene profiling, to search for the diagnostic features and to provide further insight into the pathogenesis of CPH.
CASE REPORTS Case 1 In 2000, a right-handed 58-year-old Japanese woman presented with an erythematous eruption on the left palm that had existed for 3 years. The lesion was solitary, showed asymptomatic erythema, which had gradually enlarged. Physical examination revealed a 17- 3 12-mm, well-demarcated, and slightly depressed erythema with an irregular border on the thenar aspect of her left hand. The initial clinical diagnosis was acute eczema, but various topical therapies including steroid, 20% urea, maxacalcitol, or vitamin A were ineffective during the 5-year follow-up period before performing a skin biopsy. Culture from the scale revealed two colonies of Staphylococcus epidermidis. When a biopsy
specimen was taken from the edge of the erythema, the lesion had slightly enlarged to 17 3 14 mm (Fig 1, A). Case 2 In 2001, a 73-year-old Japanese woman presented with multiple erythematous eruptions on her left palm, which had persisted for more than 10 years. The erythema was asymptomatic, but had enlarged very slowly and the number of affected areas had increased. There were two nail-sized depressed areas of erythema on the thenar part of the palm of her left hand. In addition, two similar, but smaller areas of erythema were seen around them (Fig 2, A). Dermoscopic views without using jelly showed stairlike desquamation at the edge of the lesion and this may be characteristic of CPH (Fig 2, B). With jelly immersion, homogeneous, structureless erythema was seen with regularly distributed whitish spots, which were suggestive of acrosyringeum (Fig 2, C). Treatment with vitamin A ointment, calcipotriol, maxacalcitol, 10% urea, zinc sulfate, steroid, tacrolimus, and suprofen were all disappointing. Culture from the scale was negative.
J AM ACAD DERMATOL
Ishiko et al 287
VOLUME 57, NUMBER 2
Fig 2. Case 2. Clinical and histologic features. A, Two depressed areas of erythema on thenar site of left palm of a 73-year-old woman. B, Dermoscopic view without using jelly showed stairlike desquamation at the edge of the lesion. C, Dermoscopic views using jelly immersion showed homogeneous, structureless erythema with regularly distributed whitish spots, which were suggestive of acrosyringeum. D, Lesional epidermis showed psoriasiform acanthosis.
Histopathology There was an abrupt decrement in the thickness of the stratum corneum, although in the context of both a preserved epidermal thickness and a granular cell layer that was only slightly diminished (Figs 1, B, and 2, D). A curious and distinctive feature was the frayed quality of the stratum corneum adjacent to the zone of hypokeratosis, most reminiscent of a staircase (Fig 1, C). There was a morphologic change in the corneocytes that defined the lateral margins of the hypokeratotic zone. In particular, the corneocytes more superficially manifested a basophilic and vacuolated appearance, while those in the lower portion exhibited hypereosinophilia. A faint perivascular lymphocytic infiltration was seen around dilated capillary in the superficial dermis (Fig 1, D). These features were shared in both cases. Immunohistochemistry Using commercially available polyclonal antibodies against HPV (Dakocytomation, Glostrup, Denmark), we failed to find any specific labeling in the epidermal keratinocyte in either case. To determine whether the thin horny layer is due to hypoproliferation of epidermis, Ki-67 (MIB-1,
Dakocytomation) was stained. The results showed that Ki-67epositive basal cells were more frequently seen in the lesional epidermis than in the nonlesional epidermis (Fig 3, A) in both CPH cases. To characterize the lesional keratinocytes, several antikeratin antibodies were used, including 34-bE12 (Enzo Diagnostics, Farmingdale, NY), AE11AE3 (Dakocytomation), CAM 5.2 (Becton Dickinson, San Jose, Calif), cytokeratin 2e (clone: Ks2.342.7.1, Progen Biotechnik, Heidelberg, Germany), cytokeratin 7 (clone:OV-TL-12/30, Dakocytomation), cytokeratin 9 (clone Ks9.701Ks9.216, Progen Biotechnik), cytokeratin 16 (clone SS025, Novocastra Laboratories Ltd, Newcastle, UK), and cytokeratin 20 (clone:Ks20.8, Dakocytomation). AE11AE3 (Fig 3, C) and cytokeratin 16 (Fig 3, E) showed prominent suprabasilar staining of lesional keratinocytes in contrast to its virtual absence in nonlesional skin. Conversely, cytokeratin 2e manifested a diminution in staining in the lesional zone relative to nonlesional skin (Fig 3, F ). There was no difference in the extent of positive staining with cytokeratin 9 between lesional and nonlesional skin. Both lesional and nonlesional skin failed to show reactivity with CAM 5.2, cytokeratin 7 and 20.
288 Ishiko et al
J AM ACAD DERMATOL AUGUST 2007
Fig 3. Case 2. Imunohistochemstry. A, Ki-67epositive basal cells were seen more frequently beneath lesional epidermis than in nonlesional epidermis. B, Anti-pankeratin 34bE12 showed no difference between lesional and nonlesional epidermis. C, Anti-pankeratin AE11AE3 showed strong positive staining at suprabasal keratinocytes restricted to lesional epidermis; this contrasted with the significantly weaker staining in perilesional skin. D, Acrosyringeal epithelium was spared from overexpression of AE11AE3. E, Cytokeratin16 was positive at suprabasal keratinocytes only in the lesion. F, Cytokeratin 2e expression was decreased in lesional suprabasal keratinocytes.
Electron microscopy Ultrastructural examination showed signs of corneocyte splitting through the cytoplasm with intact cell-cell attachment at the edge of the lesion where the horny layer sharply changes thickness (Fig 4, A and B). Vacuolar alteration of corneocytes involving the lower edges of the zone of hypokeratosis was seen (Fig 4, C). HPV PCR Since a recent report suggested an involvement of HPV type 4 in CPH skin, we performed PCR for HPVspecific DNA using paraffin-embedded tissue. DNA was extracted from 10-m thick paraffin sections and direct sequencing with general primers GP5 and GP6
were performed as described previously.8 As a result, we could not detect any HPV DNA in either case. 16S microbial rRNA analysis (terminal restriction fragment length polymorphism) It is reported that several uncultured bacterial species exist in human skin surfaces,13 and such unknown bacteria may cause the loss of the horny layer. Therefore we employed a culture-independent molecular analysis to search for causative microorganisms. A sample was taken by scrubbing the lesional or nonlesional skin with sterile swabs from case 2. 16S ribosomal gene, which was present in all known bacteria, was amplified with universal primers, FAM-27F (fluorescent-labeled) and 1492R
Ishiko et al 289
J AM ACAD DERMATOL VOLUME 57, NUMBER 2
Fig 5. 16S microbial rRNA gene profile (terminal RFLP) from the skin surface swabs. 16S ribosomal gene, which was present in all known bacteria, was amplified with universal primers. PCR products, which should contain most bacterial genes from the skin surface, were compared between (A) lesional and (B) nonlesional skin using restriction enzyme cutting length. Several bands were detected including band derived from P acnes. There was no difference between lesional skin and nonlesional skin. Fig 4. Case 2. EM of edge of lesion. A, At edge of lesion where the horny layer showed a sharp change in thickness, corneocytes showed intracytoplasmic splitting. B, Cell-cell detachment was not apparent and corneodesmosomes, cornified cell envelope, and keratin patterns were normal. C, At bottom of depressed area, outer layer of corneocytes contained many vacuoles of variable size. Bars = 1 m.
(nonlabeled). 59-Fluorescent-labeled PCR products, which should contain most bacterial genes on the skin surface, were compared between lesional and nonlesional skin using fluorescent-labeled restriction enzyme cutting length.13 As a result, several bands were detected, including a band derived from Propionibacterium acnes (Fig 5). Other bands indicated the presence of unidentified bacteria that cannot be detected by culture. There was no difference between lesional band patterns and nonlesional band patterns, that is, there were no lesionspecific bacteria.
DISCUSSION Herein we report two new Japanese cases with CPH, one of which had multiple lesions. To date, 28 cases have been reported in the literature and the
Table I. Twenty-eight reported cases of CPH Age (y) of first visit Male/female Predilection site
Right/left Solitary/multiple Suspected etiologic factors
35-89 (mean, 63.5) 5/23 Thenar (palm) (17/27) Hypothenar (palm) (7/27) Thumb (4/27) Sole (2/27) 16/14 24/4 Clonal epidermal proliferation, trauma, bacteria, HPV-4
clinical characteristics were summarized in Table I. CPH occurs mainly in elderly women and the predilection site was the thenar palm, which is consistent with our cases. No apparent deviation between the right and left hand was seen, and most cases occur in a solitary fashion. The diagnostic features of CPH are summarized in Table II. Using dermoscopy with non-jelly immersion technique, stair-like desquamation at the edge of the lesion was seen that was compatible with the characteristic pathological findings. With jelly immersion, homogeneous
290 Ishiko et al
J AM ACAD DERMATOL AUGUST 2007
Table II. Diagnostic findings of CPH Clinical features Dermoscopy Histopathology Immunohistochemistry Electron microscopy
Well-demarcated, depressed erythema persisting for years on palms or soles Without jelly; erythema with stair-like desquamation with jelly; structureless erythema with regularly distributed whitish spots Abrupt decrement in thickness of stratum corneum with frayed edge Ki-67", CK16", AE11AE3", CK2e# Breakage of corneocytes, vacuolization of corneocyte
structureless erythema was seen with regularly distributed whitish spots. The erythema likely represents mild inflammation with dilated capillaries in the superficial dermis, and the whitish spots may represent sparing of the acrosyringeum from excoriation in the thin horny layer. Thus dermoscopic observation faithfully reflects the histologic CPH characteristics and may have some diagnostic value in CPH. By EM, corneocytes showed cytoplasmic breakage or cleavage, and there was no significant detachment between the cells. This may indicate that the loss of the horny layer may be due to the fragility of the lesional corneocyte and does not arise from a cell adhesion disturbance. Despite the thinning of the horny layer, the malpighian layer in the lesion was hypertrophic. MIB-1 proliferation marker Ki-67 antibody stained increased cell numbers in the lesional basal layer. Anti-pankeratin antibody AE11AE3 showed strong immunostaining in the lesional suprabasal keratinocytes, and this may be useful for the diagnosis of CPH. This pattern of staining is also seen in epidermis overlying scars in case with epidermolysis bullosa14 and lichen sclerosus15 and regarded as proliferative changes, which might be associated with minor trauma. By testing the cytokeratins recognized by AE11AE3, but not by 34bE12 antibody, we found hyperexpression of keratin 16 in the lesion; this may be responsible for the strong staining of AE11AE3. Keratin 16 is expressed in the proliferative epidermis; therefore hyperexpression of keratin 16 is consistent with increased MIB-1epositive basal cells and epidermal hypertrophy. This proliferative phenotype may be a reactive one against some sort of irritation resulted from the early loss of the horny layer or may in itself be pathogenic by leading altered keratinization. In fact, keratin 2e expression was decreased in the lesion and this might be a reason of corneocyte fragility, although further biological investigation will be required. Neither of our cases showed any evidence of HPV infection using immunohistochemistry and HPV DNA amplification techniques, indicating that HPV is, at least, not the major cause of CPH. Like pitted keratolysis, some bacterial infection may
cause the loss of the horny layer. 16S Microbial rRNA gene is present in all known bacteria, and its conserved region is suitable for amplification. The length of the restriction enzymeedigested fragments may be used to identify specific bacteria and this method enables us to detect bacteria that cannot be cultured. Although we have detected several kinds of bacterial gene using this method, there was no difference of microbial components between samples from the lesional skin and perilesional normal skin. This indicated there were no disease sitee specific pathogen candidates. In conclusion, CPH still remains an enigma with regard to pathogenesis. It is certainly well defined from a clinical and light microscopic perspective. Our study suggests that a hyperproliferative epidermal state along with enhanced corneocyte fragility may account for its unique features. Specific triggers, such as trauma, microbes and HPV, cannot be implicated in every case. We thank Dr Yuuichiro Yamasaki for referring a patient, Dr Almut Bo¨er for performing HPV PCR, and Dr James R. McMillan for proofreading this manuscript.
REFERENCES 1. Perez A, Ru¨tten A, Gold R, Urbina F, Misad C, Izquierdo MJ, et al. Circumscribed palmar or plantar hypokeratosis: a distinctive epidermal malformation of the palms or soles. J Am Acad Dermatol 2002;47:21-7. 2. Obermoser G, Zelger B. ‘‘Multifocal’’ circumscribed palmar hypokeratosis: malformation or not? J Am Acad Dermatol 2003;49:1197-8. 3. Burnett JW, Harvey VM. Circumscribed palmar or plantar hypokeratosis: report of two additional cases. J Am Acad Dermatol 2004;51:843. 4. DeBloom JR, Ting W, Stone MS, Arpey CJ. Circumscribed palmar hypokeratosis. J Am Acad Dermatol 2004;51:319-21. 5. Ru¨tten A, Wecker-Brosi H, Gruhlke G, Kutzner H, Cirne de Castro L, Requena L. [Circumscribed acral hypokeratosis]. Hautarzt 2004;55:1060-3. German. 6. Mensing C, Schleusner V, Sander C, Mensing H. Circumscribed palmar or plantar hypokeratosis: two cases of a recently described entity of unknown origin. Am J Dermatopathol 2005;27:247-9. 7. Urbina F, Misad C, Gonzalez S. Circumscribed palmar hypokeratosis: clinical evolution and ultrastructural study after prolonged treatment with topical calcipotriol. J Eur Acad Dermatol Venereol 2005;19:491-4.
J AM ACAD DERMATOL
Ishiko et al 291
VOLUME 57, NUMBER 2
8. Bo¨er A, Falk T. Circumscribed palmar hypokeratosis induced by papilloma virus type 4. J Am Acad Dermatol 2006;54:908-9. 9. Lee SE, Kim YC, Kim SC. Circumscribed palmar or plantar hypokeratosis: report of a Korean case and published work review. J Dermatol 2006;33:403-5. 10. Resnik K, DiLeonardo M. Circumscribed palmar hypokeratosis: new observations. Am J Dermatopathol 2006;28:112-6. 11. Shirai A, Seki M, Hino H. Circumscribed palmar or plantar hypokeratosis. Practical Dermatology (jpn) 2005;27:71-4. 12. Murata M, Saitou Y, Mitsuto I. A case of circumscribed palmar hypokeratosis. Jpn J Clin Dermatol (jpn) 2006;60:27-9. 13. Dekio I, Hayashi H, Sakamoto M, Kitahara M, Nishikawa T, Suematsu M, et al. Detection of potentially novel bacterial
components of the human skin microbiota using cultureindependent molecular profiling. J Med Microbiol 2005;54: 1231-8. 14. Smoller BA, McNutt NS, Carter M, Gottlieb AB, Hsu A, Krueger J. Recessive dystrophic epidermolysis bullosa skin displays a chronic growth-activated immunophenotype. Implications for carcinogenesis. Arch Dermatol 1990;126:78-83. 15. Carlson JA, Ambros R, Malfetano J, Ross J, Grabowski R, Lamb P, et al. Vulvar lichen sclerosus and squamous cell carcinoma: a cohort, case control, and investigational study with historical perspective; implications for chronic inflammation and sclerosis in the development of neoplasia. Hum Pathol 1998;29: 932-48.