Mosaic Manifestation of Autosomal Dominant Skin Disorders

Abstract

A mosaic arrangement of skin lesions has been noted in numerous autosomal dominant traits including hereditary multiple skin tumors, disorders of keratinization, diseases of the connective tissue or bones, and vascular disorders. Hereditary multiple skin tumors originate from allelic loss and thus represent mosaics. Multiple lesions may sometimes involve only a segmental area, in the form of simple segmental mosaicism. Moreover, a pronounced segmental involvement may be overlaid on the ordinary, nonsegmental trait, in the form of superimposed mosaicism. Both simple segmental and superimposed mosaic manifestations have been documented in many autosomal dominant traits such as trichoepitheliomatosis, hereditary basaloid follicular hamartoma, multiple syringomas, multiple spiradenomas, glomangiomatosis, blue rubber bleb angiomatosis, neurofibromatosis 1, neurofibromatosis 2, leiomyomatosis, Gorlin syndrome, and hereditary nonsyndromic basal cell carcinoma. Examples of simple segmental involvement were reported in multiple trichodiscomas, cylindromatosis, schwannomatosis, and mastocytosis. Cases suggesting superimposed mosaicism were reported in Hornstein-Knickenberg syndrome (illegitimately named “Birt-Hogg-Dubé syndrome”), Legius syndrome, and PTEN hamartoma syndrome. Some hereditary tumor syndromes such as leiomyomatosis, neurofibromatosis 1, glomangiomatosis, and—most likely—blue rubber bleb angiomatosis appear to be especially prone to develop a superimposed mosaic involvement. Molecular proof of superimposed mosaicism has so far been provided in Gorlin syndrome, neurofibromatosis 1, Legius syndrome, and PTEN hamartoma syndrome. In the group of keratinization disorders, superimposed mosaicism has been documented in patients with keratinopathic ichthyosis of Brocq, bullous ichthyosis of Siemens, Darier disease, Hailey-Hailey disease, KID syndrome, and the various forms of porokeratosis. Remarkably, a woman with plaque-like porokeratosis of Mibelli gave birth to two sons who both had linear lesions in the form of superimposed mosaic involvement. Hence the underlying gene locus may be a hotspot for postzygotic recombination, as also assumed in disseminated superficial actinic porokeratosis. Molecular proof of simple segmental mosaicism has so far been provided in epidermolytic ichthyosis of Brocq, Darier disease, and KID syndrome. Molecular evidence of superimposed mosaicism was presented in cases of Darier disease, Hailey-Hailey disease, and various types of porokeratosis. Cases of simple segmental mosaicism with molecular proof were reported in pachyonychia congenita of the Jadassohn-Lewandowsky type and in the Galli-Galli variant of Dowling-Degos disease, and clinical examples without molecular data were reported in Costello syndrome and acrokeratoelastoidosis. Moreover, cases suggesting superimposed mosaicism have been documented, without molecular data, in autosomal dominant acanthosis nigricans, autosomal dominant dyskeratosis congenita, and porokeratosis palmaris, plantaris et disseminata. In disorders of connective tissue or bones, both types of mosaicism were documented in tuberous sclerosis and in autosomal dominant forms of Ehlers-Danlos syndromes. In tuberous sclerosis, molecular proof of simple segmental involvement was provided, whereas clinical features suggesting a superimposed mosaic manifestation are frequently noted and described under various names such as cobblestone nevus, forehead patch, or fibrous hamartoma of infancy. A simple segmental involvement in the form of pigmentary disturbances following Blaschko’s lines was documented in a girl with Brachman de Lange syndrome, a multisystem birth defect characterized by short stature, peculiar facial appearance, bone anomalies, and mental deficiency. Cases suggesting superimposed mosaicism were reported, under the name “juvenile elastoma,” in patients with Buschke-Ollendorff syndrome. Remarkably, such pronounced segmental lesions are rather often noted in several members of a family, indicating that the underlying LEMD3 locus may be a hotspot for postzygotic recombination. Moreover, a particular bone disease called melorheostosis may occur as a superimposed mosaic manifestation of Buschke-Ollendorff syndrome. Cases suggesting superimposed mosaicism have also been described in Marfan syndrome and Ehlers-Danlos syndrome type 3. In Albright’s hereditary osteodystrophy, many persuasive examples of superimposed mosaic involvement were found in the literature. In hereditary osteomatosis cutis, cases of superimposed mosaicism are so impressive that they have even got, erroneously, its own OMIM number 166350 under the name “progressive osseous heteroplasia.” A girl with Zimmermann-Laband syndrome, a phenotype characterized by thick lips, bulbous nose, hypoplasia or absence of nails or terminal phalanges, increased joint mobility, hepatosplenomegaly, and mental deficiency, showed hemihyperplasia with enlargement of her external genitalia and ipsilateral segmental hypertrichosis and hyperpigmentation. Superimposed mosaicism is possible but not certain. In hereditary hemorrhagic telangiectasia (Osler-Rendu-Weber syndrome), one case of simple segmental involvement of the face was reported. On the other hand, a case suggesting superimposed mosaicism in the form of a pronounced linear lesion involving the forehead has likewise been documented. Molecular studies are so far lacking. Rhodoid nevus syndrome (aka “capillary malformation-arteriovenous malformation”) is characterized by a familial occurrence of multiple disseminated rhodoid nevi. These circular or oval lesions are often surrounded by an anemic halo. They are caused by RASA1 or EPHB4 mutations. A superimposed mosaic manifestation occurs rather frequently in the form of arteriovenous malformation with a concomitant large, segmentally arranged form of rhodoid nevus. For this reason, the phenotype had been described under the name “capillary malformation-arteriovenous malformation,” but it is now clear that the presence of an arteriovenous malformation is by no means a prerequisite for the diagnosis of the syndrome. Within the group of epidermolysis bullosa, revertant mosaicism is frequently noted, whereas reports on forward mosaicism are virtually absent. Two exceptional cases were found in the literature. A self-limited form of dystrophic epidermolysis bullosa (“transient bullous dermolysis of the newborn”) is caused by COL7A1 mutations. It tends to heal spontaneously. A female neonate had linear lesions involving one of her legs, giving rise to scarring. Analysis of blood DNA revealed compound heterozygosity at COL7A1, without presence of mosaicism. Skin samples were not available for analysis, which is why the cause of this particular form of cutaneous mosaicism remains unknown. Another peculiar mosaic phenotype in the form of superficial acantholysis arranged along Blaschko’s lines was reported by Jokiaho et al. (1989) under the term “miliaria rubra-like lesions” in a newborn girl with dysmorphism and a parietal meningocele. The linear acantholytic disorder resolved spontaneously within 1 week.