Elucidating the genetic basis for EB simplex created an opportunity to reexamine its pathophysiology (see Table ) and investigate, on a de novo basis, the causal relationship among keratin assembly, IF network architecture, and epithelial cell integrity.
Studies carried out in relevant mouse models confirmed that frictional trauma, often in a repeated fashion, is required to expose the fragility of basal keratinocytes and elicit bullous skin lesions. Subsequent biophysical studies conducted on keratin IFs, both when reconstituted in vitro and in cultured cells ex vivo, strongly suggested that mutant keratin–expressing basal keratinocytes are mechanically softer due to defects in the architecture of the IF network (61
). Relative to filaments reconstituted in vitro from wild-type K5 and K14 proteins, those containing EB simplex–causing K14
mutations (e.g., Arg125→Cys; ref. 61
) or K5
mutations (e.g., 1649delG; ref. 57
) exhibit markedly lower elasticity under small (linear) deformation regimes and break readily when subjected to progressively larger (nonlinear) amounts of shear stress. The deficiency observed is markedly enhanced when such filaments are crosslinked into a network, a circumstance that increases the elasticity of wild-type K5/K14 heteropolymers by 500-fold (61
), depending on concentration (67
Beil et al. (68
) showed that collapse of keratin IFs around the nucleus, in a fashion characteristic of many mutant keratin–expressing cells, causes a dramatic softening of the cytoplasm in cultured epithelial cells. When conducting live imaging studies of cultured skin keratinocytes, Werner et al. (62
) found that assembly subunits containing wild-type K14 incorporate within “growing” filaments in a specific zone of the peripheral cytoplasm; in contrast, subunits containing a K14 Arg125→Cys mutant protein fail to do so (62
). The latter finding is of interest given the earlier observation that trauma-induced rupture of basal keratinocytes often occurs in a specific area of the cytoplasm, between hemidesmosomes and the nucleus, in EB simplex skin (32
) and mouse models thereof (22
). Much progress has thus been made from a handful of biophysical and live imaging studies, so that future efforts of this type could transform our understanding of how the keratin IF network supports epithelial cell integrity in vivo.
The presence of cytoplasmic aggregates containing mispolymerized mutant keratin proteins is the defining characteristic of EBS-DM (Table and Figure ) and may worsen the impact of dominant-negative mutations. Mice made to express aggregation-prone K14 mutants (30
) exhibit an earlier onset of disease and more severe blistering and die earlier than K14
-null mice (43
). Analogously, the EB simplex phenotype exhibited by individuals null for K14
is milder than that of autosomal-dominant EBS-DM (46
). The strong missense alleles characteristic of EBS-DM likely curtail the beneficial influence of any wild-type keratin (e.g., K15 in basal keratinocytes), as these also become incorporated into the defective assembly aggregates.
Additionally, when the misfolded protein response fails to resolve these aggregates, they may overwhelm the protein homeostasis apparatus of the cell (70
). This could lead to cellular stress (72
) and influence the cell and tissue phenotypes in vivo. Cytoplasmic aggregates of mispolymerized IF proteins are characteristic of many disorders caused by mutations in genes encoding non-keratin IF proteins (8
). Alexander disease, for instance, is a devastating neurological condition caused by missense mutations in the gene encoding the type III IF protein glial fibrillary acidic protein (GFAP) (74
). Mutant GFAP-containing aggregates, readily seen in histology sections, are a cardinal feature of this condition. Such astrocyte-specific aggregates may be the source of the non-cell-autonomous pathophysiology of this condition, i.e., the Alexander disease phenotype may also reflect anomalies in CNS cell types that do not express mutant GFAP (75
Does inflammation play an important role in the EB simplex phenotype? The clinical and experimental evidence in this regard is equivocal. In EBS-Migr (Table ) (76
), skin blisters show a distinct belt-like erythema, and their margins are migratory even in trauma-free skin, suggesting a role for inflammation. In addition, there are conflicting reports as to whether the Arg125→Cys mutation in K14
, which causes EBS-DM, results in increased TNF-α secretion and increased susceptibility to TNF-α–induced apoptosis in human keratinocytes (77
). Lugassy et al. (79
) reported that K14
haploinsufficiency results in increased susceptibility of keratinocytes to TNF-α–induced apoptosis and suggested that this plays a key role in the pathogenesis of Naegeli-Franceschetti-Jadassohn syndrome (NFJS) (Table ), extending previous work on EBS-DM mutations by Yoneda et al. (80
). In contrast, Lu et al. (78
) detected higher levels of IL-6, IL-1β, and chemokines, but normal levels of TNF-α, in the skin of newborn K5
-null mice. The latter mice survived only marginally longer after birth (up to 8 hours compared with less than 2 hours) when treated with the antiinflammatory agent doxycycline (78
). Related to this, sulforaphane, another small molecule with antiinflammatory properties, does not substantially prolong the life of K5
-null mice, although it reduces skin inflammation (45
). Finally, conventional apoptosis does not occur in the epidermis of individuals with the most common variants of EB simplex, or in mouse models thereof (45
). Whether proinflammatory cytokines are produced as a secondary consequence of basal cell lysis, even when occurring on a microscopic scale (Figure ), or are present before mechanical trauma and the ensuing blistering remains unclear and requires further investigation.
Further, certain clinical attributes of EB simplex still have no sound biological basis (Table ). Topping the list of such attributes are two frequently seen phenomena: enhanced frequency of skin blistering during the warm season and age-related improvement in clinical symptoms (2
). On an anecdotal basis, bathing in cold water and, paradoxically, experiencing fever episodes seem to lessen the blistering in individuals with EB simplex (4
). Although experimental evidence is still lacking, one possibility is that conditions that activate the heat shock/stress response may help to resolve the aggregates of misassembled keratins. Understanding the mechanistic basis for these phenomena, as well as others listed in Table , may lead to new insights into keratin and epithelial biology as well as innovative ideas for treatments for EB simplex.