Mutation in LEMD3 (Man1) Associated with Osteopoikilosis and Late-Onset Generalized Morphea: A New Buschke-Ollendorf Syndrome Variant

Introduction. Buschke-Ollendorf syndrome (BOS) is an uncommon syndrome characterized by osteopoikilosis and other bone abnormalities, accompanied by skin lesions, most frequently connective tissue nevi. BOS is caused by mutations in the LEMD3 gene, which encodes the inner nuclear membrane protein Man1. We describe a unique case of osteopoikilosis associated with late-onset localized scleroderma and familial LEMD3 mutations. Case Report. A 72-year-old woman presented with adult-onset diffuse morphea and bullous skin lesions. Evaluation revealed multiple hyperostotic lesions (osteopoikilosis) suggestive of BOS. DNA sequencing identified a previously undescribed nonsense mutation (Trp621X) in the LEMD3 gene encoding Man1. Two additional family members were found to have osteopoikilosis and carry the same LEMD3 mutation. Conclusions and Relevance. We report a unique familial LEMD3 mutation in an individual with osteopoikilosis and late-onset morphea. We propose that this constellation represents a novel syndromic variant of BOS.


Introduction
Osteopoikilosis is a rare autosomal dominant skeletal dysplasia characterized by multiple hyperostotic lesions. The bone lesions are generally symmetric but distributed irregularly and are typically detected as incidental radiographic findings [1]. Osteopoikilosis can be an isolated skeletal abnormality or may occur in association with diverse cutaneous manifestations as a component of Buschke-Ollendorf syndrome (BOS) (OMIM166700) [2,3]. The cutaneous manifestations of BOS, commonly manifesting in childhood, include connective tissue nevi and less frequently elastomas, collagenomas, and dermatofibrosis lenticularis (also called hypertrophic scar disseminata) [4][5][6]. The genetic basis for BOS was identified in 2004 by genome-wide linkage studies. These studies uncovered a mutation in LEMD3 (LEM domain containing 3) gene [2]. The LEMD3 gene encodes the 60 kD inner nuclear membrane protein Man1. Mutations in LEMD3 are also linked to skeletal abnormalities other than BOS. These include isolated (nonsyndromic) osteopoikilosis [2] and melorheostosis, a hyperostotic anomaly characterized by radiolucent "dripping wax" appearance in the cortex of long bones [7,8]. Of note, melorheostosis itself may be an isolated radiological finding or occur in association with abnormalities in adjacent soft tissue, including linear scleroderma [9][10][11][12].
Morphea is a localized form of scleroderma characterized by skin induration in localized areas. Morphea has the highest incidence in childhood and young adults. Lateonset morphea is considerably less common. In contrast to systemic sclerosis, morphea is confined to the skin and is not associated with extracutaneous manifestations. The spectrum of morphea disorders includes linear scleroderma, plaque morphea, and diffuse morphea, which in rare cases may be extensive (pansclerotic morphea). Morphea lesions commonly occur on the extremities and the face and less 2 Case Reports in Dermatological Medicine frequently on the trunk. The etiology of morphea is unknown and its pathogenesis remains poorly understood. Transforming growth factor-beta (TGF-) is a multifunctional cytokine implicated in fibrosis in multiple organs [13]. The profibrotic responses elicited by TGF-involve both Smad-dependent canonical, as well as Smad-independent noncanonical intracellular signaling pathways [14,15]. Alterations in TGF-expression or function and in its downstream signaling mediators are implicated in the pathogenesis of localized scleroderma and systemic sclerosis [16]. Man1, the protein encoded by LEMD3, is intricately linked to TGFbiology and has complex effects on modulating TGFresponses. On one hand, Man1 interacts directly with TGFsuperfamily ligands, including bone morphogenic proteins (BMPs) and activin [17]. On the other hand, Man1 binds, via its C-terminal domain, directly to Smad [17,18]. Importantly, Man1 negatively regulates Smad-mediated TGF-signaling in a variety of cell types [2,[17][18][19][20][21][22][23]. Despite these recent molecular insights, the full spectrum of LEMD3 mutations and their impact on TGF-biology and their functional role in the phenotypic expression of BOS remain poorly understood.

Case Report
A previously healthy 72-year-old Caucasian woman presented with six months' progressive skin tightening and discoloration affecting her arms, shoulders, chest, and lower legs. Subsequently, painful erythematous patches appeared on her back, breasts, and belt line. She had no family history of scleroderma or other autoimmune disease. Physical examination demonstrated firmly indurated and hyperpigmented lesions on the arms, shoulders, chest, belt line, and lower legs and scaly erythematous and partially bullous patches over both breasts (Figures 1(a)-1(c)). She had no sclerodactyly, nailfold microvascular abnormalities or other manifestations of systemic sclerosis, and serologic tests for antinuclear, anti-Scl-70, and anti-centromere antibodies were negative. Radiographs of the hands, feet, and knees revealed numerous well-demarcated bone densities (osteopoikilosis) bilaterally ( Figure 3). Based on the presence of osteopoikilosis and skin lesions, the diagnosis of BOS was made, and genomic DNA sequencing was undertaken (see below). Further investigation identified three family members (school-aged nieces and nephews on the paternal side) who had asymptomatic osteopoikilosis, but no skin lesions. Treatment of the index case included psoralens and ultraviolet light A, oral calcitriol hydroxychloroquine, and mycophenolate mofetil, as well as topical calcipotriene, betamethasone dipropropionate, and pimecrolimus. She showed slow partial resolution of skin lesions. Subsequent course was complicated by recurrent episodes of hemorrhagic olecranon bursitis and hemorrhagic bullae over the chest, abdomen, and back.

Literature Survey and Discussion
First described in 1928, BOS is an uncommon familial syndrome characterized by osteopoikilosis associated with skin manifestations [32,33]. In children with BOS, osteopoikilosis has been reported to be accompanied by fibrotic skin lesions, including linear scleroderma, part of the morphea spectrum disorders [5,6,[34][35][36]. We are unaware of a previous description of late-onset generalized morphea associated with osteopoikilosis. The present case might represent the coexistence of two distinct disorders affecting the skin and bone. We consider this unlikely however. As osteopoikilosis has an estimated prevalence of 2/100,000 and morphea of 0.02-0.04/100,000 [37], the extreme rarity of these two conditions makes their occurrence in the same individual by chance highly unlikely. A favored alternative explanation is that late-onset generalized morphea associated with osteopoikilosis seen in the present case is in fact syndromic and represents a novel BOS variant that falls within the phenotypic continuum linked with LEMD3 mutations.
A review of over 100 published cases of BOS showed that connective tissue nevi (dermatofibrosis lenticularis disseminata) were the most frequent cutaneous manifestation. The diagnosis of BOS was characteristically made before the age of 16. A survey of cases of LEMD3-associated skin and bony lesions revealed 28 cases of melorheostosis associated with linear scleroderma, typically affecting skin adjacent to the bone lesions, with a majority of these individuals developing linear (localized) scleroderma in childhood (Table 2). However, melorheostosis frequently occurs in the absence of LEMD3 mutations [8], and thus far none of the LEMD3 mutation-proven cases of melorheostosis (Table 1) have coincided with linear scleroderma. One report of osteopoikilosis associated with scleroderma described a patient with sclerodactyly and Raynaud phenomenon, suggesting coexistent systemic sclerosis and isolated osteopoikilosis rather than syndromic BOS [6].
LEMD3 mutations show variable penetrance. There is extreme variability in the associated phenotypes, even among individuals harboring identical mutations [2]. Given such a high degree of heterogeneity and incomplete penetrance, the causal role of any particular LEMD3 mutation in a specific phenotype is difficult to discern. Although the TGF-/Smad signaling pathway plays a pivotal role in both skin and bone homeostasis, it remains unclear how Man1-Smad interactions are affected by the BOS mutations, and whether they contribute to clinical features. While the novel LEMD3 mutation described in this report is predicted to alter the C-terminal domain of Man1 required for R-Smad interactions [23], our functional studies failed to demonstrate consistent alterations in TGF-/Smad signaling in the BOS skin fibroblasts.
The coexistence of morphea and lichen sclerosus et atrophicus (LSA) changes is also of note. While this combination has been previously reported as a cause of bullous changes [40][41][42][43] in morphea, the association is relatively common in adults. A recent retrospective study confirmed the coexistence of these two entities in 26 of 91 (28.5%) of adult morphea patients compared to only 1 of 381 children with morphea [44]. Bullous LSA changes are primarily inflammatory [45] and some have suggested that LSA may represent subepithelial morphea in this context [46]. Therefore, whether the LSA changes are related to the LEMD3 Case Reports in Dermatological Medicine

Summary
In summary, we describe a case of osteopoikilosis associated with late-onset generalized morphea and associated LSA changes in an elderly individual carrying a previously undescribed familial mutation in LEMD3. We propose that in this case morphea and osteopoikilosis are linked, representing a novel BOS variant that is on the continuum of LEMD3associated skin and bone manifestations. In light of the known involvement of Man1 in modulating canonical TGFsignaling, we hypothesize that the skin and bone abnormalities associated with LEMD3 mutations might be related to altered TGF-signaling. Future studies will characterize the functional consequences of LEMD3 mutations and their role in the clinical manifestations of the syndrome. Given the diverse phenotypes associated with such mutations and poorly understood mechanisms of how Man1 protein changes contribute to the phenotypic manifestations of BOS, such studies may reveal new roles for this diverse molecule in mesenchymal cell biology.