Hemidesmosomes are multiprotein complexes perfectly designed to tether intermediate filaments to the cell membrane and stabilize adhesion of epithelial cells to the underlying basement membrane in tissues exposed to mechanical stress, such as skin, intestine, bladder and cornea 1. In stratified and pseudostratified epithelia, the structural components of hemidesmosomes belong to four protein families – plakins (the bullous pemphigoid antigen 1 (BPAG1) and plectin), integrins (α6β4), collagens (collagen XVII) and tetraspanins (CD151) 2. Numerous aspects of the structure and function of hemidesmosomes have been illuminated during the past 2–3 decades (reviewed in 3). The study of hemidesmosome organization in cultured keratinocytes with two- and three-colour super-resolution microscopy revealed that collagen XVII (syn. BPAG2, BP180) and BPAG1e are characteristically arranged within hemidesmosomes with collagen XVII surrounding a central core of BPAG1e molecules, while plectin interacts simultaneously and asymmetrically with β4 integrin and keratin. In skin cross sections, hemidesmosomes of variable sizes could be distinguished with BPAG1 and plectin occupying a position in between β4 integrin and collagen XVII, and the intermediate filament system 4. Apparently static structures, hemidesmosomes undergo regulated assembly and disassembly, crosstalk with β1 integrin adhesions (designated as focal adhesions in cell culture), and interactions with the keratin intermediate filaments, microtubules and the actin cytoskeleton, and growth factor receptors 5. In addition, ultraviolet exposure modulates hemidesmosome plasticity, contributing to long-term pigmentation in human skin. Hemidesmosomes are particularly dynamic during homoeostatic epidermal turnover and wound healing, and in cancers and metastasis. Hemidesmosomal assembly occurs after initial integrin α6β4–laminin-332 binding, through the interaction of plectin (isoform a) with integrin β4. The collagen XVII–plectin interaction, although not required for induction of hemidesmosome formation, strengthens the ternary integrin α6β4–plectin-collagen XVII complex and acts as a platform for the incorporation of BPAG1e. The next recruit was found to be BPAG1e that associates via its plakin domain with collagen XVII and integrin β4 6. The keratin 5/14 intermediate network linkage provided by the C-terminal domains of both plectin isoform a, and BPAG1e is crucial for the ability of hemidesmosomes to withstand mechanical stress. The absence of keratins causes loss of plectin-β4-integrin interaction and elevated β4-integrin phosphorylation. Disassembly of hemidesmosomes occurs transiently when keratinocytes become migratory to repopulate a skin wound or during carcinoma invasion. Transient disassembly is achieved by post-translational mechanisms including phosphorylation-dependent disruption of protein–protein interactions, whereas permanent dissolution of hemidesmosomes during terminal differentiation also involves repression of gene transcription and proteolytic cleavage of hemidesmosome components. The complexity of the hemidesmosomes as integrated microsystems is further increased by the existence of a number of associated proteins with regulatory roles. For example, SOXF transcription factors play a role in regulating formation of cytoplasmic plaque protein assembly and disrupted SOXF function results in skin blistering phenotypes. Another example is flightless I, a highly conserved member of the gelsolin family of actin-remodelling proteins which affects hemidesmosome formation. The medical relevance of hemidesmosomes and their molecular components is high because of associated genetic and autoimmune disorders (reviewed in 2). These in turn represent models for research because they reflect in a mechanistic way the functions of those proteins. In particular, the structural and functional diversity of hemidesmosomal plakins remains surprising. In basal epidermal keratinocytes, BPAG1e and plectin are intra-cellular scaffolding constituents of inner hemidesmosomal plaques and share the presence of a plakin domain, along with a rod coiled-coil and plakin-repeat domains. Besides, several isoforms of BPAG1 and plectin exist 2. These differ in size and domains and have tissue- and subtissue-specific expression patterns, suggesting adaptation to the environmental context and function. In humans mutations in the dystonin gene (DST), encoding distinct BPAG1 isoforms lead to two autosomal recessive disorders without any clinical overlap: neuropathy, hereditary sensory and autonomic, type VI (MIM#614653) and epidermolysis bullosa simplex, autosomal recessive 2 (MIM#615425). In affected members of a large consanguineous family with hereditary sensory and autonomic neuropathy type VI, manifesting with progressive limb contractures, dystonia, dysautonomia and early postnatal death, a homozygous 1-bp deletion in the DST gene, resulting in a frameshift and a premature termination codon (PTC) was identified. The mutation affected only the neuronal- and muscle-specific DST isoforms (dystonin-a1, -a2 and -a3) 7. In individuals with lifelong history of trauma-induced spontaneous blisters and erosions particularly affecting ankles and feet, homozygous mutations of the DST gene were found resulting in PTC, p.Q1124X and p.R1249X, in the coiled-coil domain 8. The coiled-coil domain is exclusively expressed in the BPAG1e and BPAG1n isoforms, which are expressed in the skin and nervous system, respectively. Electron microscopic analysis of a skin biopsy showed discrete abnormalities of hemidesmosomes, with poorly formed inner plaques. Immunofluorescence staining showed absence of BPAG1e at the dermal-epidermal junction and decreased immunoreactivity for integrin β4, plectin and collagen XVII. The clinical examination of my first patient lacking BPAG1e in the skin revealed plantar keratoderma, but was otherwise unremarkable. It was an altercation in which the young man was involved that lead to skin blistering and unravelled the minor skin fragility. Recently, a genome sequencing programme for novel undiagnosed diseases disclosed DST missense mutations in a case with skin and neurological disorder, but functional studies are required to validate the disease-causing role (9). Further cases, mutations and long-term observation are required to understand the natural history of these disorders and to decode what BPAG1 is good for, in brain and skin. The biology of the transcripts expressed in the brain, BPAG1a and b seems to be more complex than just assuring anchorage and adhesion: interaction with microtubuli, vesicle transport, Golgi positioning and regulation of the organization and dynamics of the actin network 2. In skin, in vivo lack of BPAG1e does not seem to crucially disrupt hemidesmosomal and epidermal homoeostasis, being probably buffered by other plakins. The precise underlying molecular mechanisms of how blistering and keratinization anomalies overlap remains to be established. Targeting BPAG1e by genetic or autoimmune mechanisms causes different degrees of skin fragility in humans suggesting distinct mechanisms of interference with hemidesmosomes. Challenges for further research are to elucidate the dynamics and signalling events linked to hemidesmosomes in the three-dimensional environment skin in situ, making use of advanced imaging technologies. The majority of the studies carried out so far made use of in vitro 2-D culture systems with a single epithelial cell type, in which no real hemidesmosomes but hemidesmosome-enriched protein complexes are formed. In addition, systems biology tools are required to integrate hemidesmosomal structural and regulatory molecules, as well as cytoskeletal interactions into the model. The reason for the puzzling number of isoforms of plakins and their roles in tissue homoeostasis remains to be elucidated. This will be facilitated by the identification of further human mutations and the illumination of their consequences and by mouse models. Finally, genetic disorders associated with other plakins, including microtubule-actin cross-linking factor 1, envoplakin, periplakin and epiplakin, remain to be discovered. The author has declared no conflicting interests. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.