Skip to Main Content


  1. Eukaryotic cytoskeletons are composed of three filament networks. Actin microfilaments (6 nm) and microtubules (30 nm) are composed of proteins that are highly conserved in evolution from yeast to humans. Intermediate filaments (IFs, 10 nm), on the other hand, first appeared in evolution among the brachiopods and seem to have evolved to serve the specialized architectural requirements of higher eukaryotic cells. They constitute a superfamily of more than 40 proteins, which are differentially expressed in most, if not all, cells of higher organisms.

  2. Despite their diversity in amino acid sequence, all IF proteins have a common secondary structure, consisting of a central, α-helical rod domain and flanking nonhelical head (N-terminal) and tail (C-terminal) segments. IF proteins form parallel, coiled-coil dimers, and more than 10,000 of these are needed to make each 10-nm filament. Dimers associate in a head-to-tail fashion to make linear arrays, which then pack laterally as two antiparallel chains arranged to form apolar protofilaments (2–3 nm). Two protofilaments intertwine to form protofibrils (4.5 nm), and approximately four of these then constitute the overall 10-nm diameter of the IF.

  3. IF proteins have been subdivided into five distinct subtypes. Type I and II IF proteins are the keratins, which form obligatory heteropolymers in vitro in the absence of other auxiliary proteins or factors. Approximately 30 keratins of two distinct types are coexpressed as pairs in epithelial cells at various stages of differentiation and development. In the epidermis, keratins K5 and K14 are the major structural proteins of basal cells. As keratinocytes commit to terminal differentiation, they switch off expression of this pair and switch on expression of keratins K1 and K10, which constitute approximately 85 percent of total protein in the fully differentiated squame.

  4. The first disorder of keratin to be genetically defined was epidermolysis bullosa simplex of the Dowling-Meara type (D-M EBS) (MIM Nos. 148066 and 131760), a blistering skin disease where the innermost basal cells of the epidermis rupture and degenerate upon mechanical stress. This form of EBS is typified by clumps or aggregates of keratin material in basal epidermal cells. The elucidation of the genetic basis of EBS as a disorder of keratins K5 (MIM No. 148040) and K14 (MIM No. 148066) represents a novel example of the use of reverse genetics, beginning with the cloning of the basal cell keratin cDNAs and genes, defining key residues important for K5 and K14 filament assembly, engineering dominant negative-acting K14 mutants, discovering that these mutant genes cause EBS in transgenic mice, and finally focusing on human EBS and demonstrating that patients with this disorder have mutations in key residues critical for filament formation. Classic genetics substantiate the reverse genetic findings: Human families with any of four different subtypes of EBS-Dowling-Meara, Koebner (MIM No. 131900), Weber-Cockayne (MIM No. 131800), and mottled pigmentation (MIM No. 131960) have genetic defects that map to either chromosome 12 or chromosome 17, at locations where the respective genes for K5 and K14 reside.

  5. Many point mutations in the genes encoding K5 and K14 genes have now been found in patients with EBS, and the location of the mutation in the keratin polypeptides correlates well with earlier predictions based on random mutagenesis studies. In humans and in mice, severity in phenotype correlates with the degree to which a particular mutation affects 10-nm filament assembly, with the most severely disrupting mutations giving rise to defects in filament elongation and the milder mutations affecting lateral interactions within the filament. EBS with mottled pigmentation differs in that the location of the mutation resides in a region of K5 that is involved in connecting keratin IFs to desmosomes and possibly to other proteins/organelles within the basal epidermal cell.

  6. The advantage in a reverse genetic approach is that once the paradigm for a keratin disorder had been determined, a list could be made of additional disorders likely to involve defects in one or more of the 30 different keratin genes within the human genome. From a study of EBS, it was predicted that keratin disorders should have the following in common: cytolysis of a subset of epithelial cells sharing a common pattern of keratin expression, cytolysis as a consequence of mild physical stress, clumps or aggregates of keratin protein in the cells undergoing degeneration, and little or no involvement of epithelial cells that do not express the keratin(s) of interest. Since all epithelial cells express a type I keratin that maintains a structurally important arginine that is also a genetic hotspot, even a residue likely to be mutated could be predicted.

  7. The first keratin disorder shown to fit the paradigm set by EBS is epidermolytic hyperkeratosis (EH), a skin disorder showing normal basal cells but degenerating suprabasal cells leading to crusting of the skin over mechanically stressed regions such as elbows and knees. Since epidermal cells make a switch in keratin expression as they commit to terminal differentiation and move outward toward the skin surface, it was correctly predicted and subsequently demonstrated that EH is a genetic disorder involving mutation in the genes for K1 and K10. The most frequent mutation found in these patients is the highly conserved arginine residue near the N-terminal end of the rod domain of K10. A milder form of EH affecting only the uppermost suprabasal layers has been shown to involve both mild mutations in the genes for K1 or K10 and severe mutations in the gene for K2e, a keratin expressed later during differentiation. A rare form of the clinically mosaic disorder, referred to as epidermal nevus with EH involvement, was demonstrated to be a genetically mosaic disorder involving K10 and/or K1 mutations. Other diseases that fit the paradigm for a genetic disorder of keratin and that have been shown to involve mutations in keratins include white sponge nevus (WSN), an esophageal and oral epithelial disorder involving K4 and K13 mutations; pachyonychia congenita (PC), a hair and nail disorder involving K6 and K16 mutations; monilethrix, a brittle hair disorder involving defects in the Ha and Hb hair-specific keratins; and Meesman's corneal dystrophy, a degenerative disorder of the cornea involving mutations in the genes for K12 and K3. Some correlations have been found between cirrhosis of the liver and genetic differences in K8 and K18, but as yet, functional and genetic evidence is lacking to ascertain whether these differences are causative.

  8. Since keratin mutations in humans give rise to epithelial cells displaying perturbations in IF architecture and mechanical fragility, it seems likely that defects in other IF gene family members will generate similar problems in other cell types. In this regard, it was shown recently that mice that are null for the gene encoding desmin, the IF protein of muscle cells, exhibit generalized muscle weakness and degeneration. Furthermore, desmin mutations recently were discovered in several patients suffering from generalized myopathies and who have disorganization within their muscle Z-bands, the residence of desmin IFs.

  9. While studies involving genetic disorders of keratin are relatively new, the field has opened up exciting new prospects for elucidating the genetic bases of other human diseases that are typified by alterations in IF networks and cell degeneration. It seems likely that many of these diseases will either involve IF gene defects or defects in proteins or organelles that associate with 10-nm filaments. In this regard, it has been discovered that functional loss of the bullous pemphigoid antigen 1 (BPAG1) gene is responsible for EBS and rapid degeneration of the sensory neurons in mice, a condition referred to as dystonia musculorum (dt/dt). A similar disorder in humans, EBS with muscular dystrophy, has been shown to arise from functional loss of the plectin gene, sharing a high degree of sequence identity with the BPAG1 gene. The proteins encoded by these genes are part of a growing group of IF linker proteins that function to connect the IF network to different cellular junctions and/or organelles. Taken together, the study of IF disorders and disorders of their associated proteins overtly illustrates the power of combining classic genetic approaches with reverse approaches (involving cell biology and transgenic mouse technology) in understanding the genetic bases of human diseases.

Pop-up div Successfully Displayed

This div only appears when the trigger link is hovered over. Otherwise it is hidden from view.