Abstract Lysyl hydroxylase (LH) catalyzes the post-translational formation of hydroxylysines in collagens and collagenous proteins. Three lysyl hydroxylase isoforms, LH1, LH2 and LH3, have been identified from different species. In addition, LH2 has two alternatively spliced forms, LH2a and LH2b. The hydroxylysines have an important role in the formation of the intermolecular collagen crosslinks that stabilize the collagen fibrils. Some of the hydroxylysine residues are further glycosylated. In this thesis the substrate amino acid sequence specificities of the LH isoforms were analyzed using synthetic peptide substrates. The data did not indicate strict amino acid sequence specificity for the LH isoforms. However, there seemed to be a preference for some sequences to be bound and hydroxylated by a certain isoform. Galactosylhydroxylysyl glucosyltransferase (GGT) catalyzes the formation of glucosylgalactosylhydroxylysine. In this study, LH3 was shown to be a multifunctional enzyme, possessing LH and GGT activities. The DXD-like motif, characteristic of many glycosyltransferase families, and the conserved cysteine and leucine residues in the N-terminal part of the LH3 molecule were critical for the GGT activity, but not for the LH activity of the molecule. The GGT/LH3 protein level was found to be decreased in skin fibroblasts and in the culture media of cells collected from members of a Finnish epidermolysis bullosa simplex (EBS) family, which was earlier reported to have a deficiency of GGT activity. In this study, we showed that the reduction of enzyme activity is not due to a mutation or lower expression of the LH3 gene. Our data indicate that the decreased GGT/LH3 activity in cells has an effect on the deposition and organization of the key extracellular matrix components, collagen types VI and I and fibronectin, and these changes are transmitted to the cytoskeletal network. These findings underline LH3 as an important extracellular regulator.

Epidermolysis bullosa simplex (EBS) and epidermolytic ichthyosis (EI) are rare skin fragility diseases characterized by intra-epidermal blistering due to autosomal dominant-negative mutations in basal (KRT5 or KRT14) and suprabasal (KRT1 or KRT10) keratin genes,  respectively. Despite vast knowledge in the disease pathogenesis, the pathomechanisms are not fully understood, and no effective remedies exist. The purpose of this work was to search for keratin gene mutations in EBS patients, to develop in vitro models for studying EBS and EI, and to investigate novel pharmacological approaches for both diseases. We identified both novel and recurrent KRT5 mutations in all studied EBS patients but one which did not show any pathogenic keratin mutations. Using cultured primary keratinocytes from EBS patients, we reproduced a correlation between clinical severity and cytoskeletal instability in vitro. Immortalized keratinocyte cell lines were established from three EBS and three EI patients with different phenotypes using HPV16-E6E7. Only cell lines derived from severely affected patients exhibited spontaneous keratin aggregates under normal culture conditions. However, heat stress significantly induced keratin aggregates in all patient cell lines. This effect was more dramatic in cells from patients with a severe phenotype. In organotypic cultures, the immortalized cells were able to differentiate and form a multilayered epidermis reminiscent of those observed in vivo. Addition of two molecular chaperones, trimethylamine N-oxide dihydrate (TMAO) and sodium 4-phenylbutyrate (4-PBA), reduced the keratin aggregates in both stressed and unstressed EBS and EI keratinocytes, respectively. The mechanism of action of TMAO and 4-PBA was shown to involve the endogenous chaperone system (Heat shock proteins e.g. Hsp70). Besides, MAPK signaling pathways also seemed to be incriminated in the pathogenesis of EBS. Furthermore, depending on which type of keratin is mutated, 4-PBA up-regulated Hsp70 and KRT4 (possibly compensating for mutated KRT1/5), and down-regulated KRT1 and KRT10, which could further assist in protecting EBS and EI cells against stress. In conclusion, novel and recurrent pathogenic keratin mutations have been identified in EBS. Immortalized EBS and EI cell lines that functionally reflect the disease phenotype were established. Two pharmacologic agents, TMAO and 4-PBA, were shown to be promising candidates as novel treatment of heritable keratinopathies in this in vitro model.