A short-term screening protocol, using fibrillin-1 as a reporter molecule, for photoaging repair agents.

Photoaged skin is characterized by coarse and fine wrinkles. The mechanisms of wrinkle formation are undetermined, but appear to be due to changes within the matrix of the dermis and at the dermal-epidermal junction. Previous studies have identified marked reductions in procollagens I and III, collagen VII, and the fibrillin-rich microfibrillar apparatus in this area. Topically applied all-trans retinoic acid can repair photoaged dermal matrix, but this takes at least 6 mo of treatment. In this study, we have examined the abundance and distribution of fibrillin-1 prior to, and following, 192 wk of all-trans retinoic acid treatment. We have further developed a short-term protocol to determine the utility of potential repair agents, using fibrillin-1 as the marker for outcome. Individuals with clinically assessed severe photoaging were recruited to the study (n = 8). 0.025% all-trans retinoic acid, 5% sodium lauryl sulfate (irritant control), or vehicle were applied under occlusion to photoaged extensor forearm. A fourth control area was also occluded. After 96 h, punch biopsies were taken under local anesthesia and processed for either transmission electron microscopy or snap frozen. Frozen sections were prepared for immunohistochemistry and in situ hybridization immunohistochemistry. Electron microscopy revealed aberrant elastic fibers in the papillary dermis of photoaged forearm skin, with sparse microfibrillar apparatus and interstitial collagen. After application of 0.025% all-trans retinoic acid, there was increased deposition of both these dermal matrix components, with the aberrant elastic fibers no longer apparent. Significant increases (p < 0.05) were observed at the protein and mRNA levels for fibrillin-1 following all-trans retinoic acid and sodium lauryl sulfate treatments, with all-trans retinoic acid having a significantly greater effect than irritant control (p < 0.001); however, neither application had significant effect on the abundance of collagen VII or its mRNA. Investigation of collagen I synthesis revealed no difference following treatments. To ascertain the clinical relevance of using fibrillin-1 as a marker for photoaging, facial skin was biopsied at baseline and after long-term (192 wk) topical all-trans retinoic acid treatment (n = 5). Biopsies were wax-embedded and sections prepared for immunohistochemistry for fibrillin-1. Significant increases in the abundance of the microfibrillar apparatus was observed proximal to the dermal- epidermal junction (p < 0.001) following long-term all-trans retinoic acid application. This study indicates that all-trans retinoic acid can significantly affect fibrillin-1 content in photoaged skin. Furthermore, fibrillin-1 can be used as a "reporter" molecule in short-term protocols for testing the utility of topical agents in the repair of photoaged skin.

Fibrillin assembly: dimer formation mediated by amino-terminal sequences.

We have investigated recombinant fibrillin-1 (profib-1) and fibrillin-2 (glyfib-2) molecules encoding the proline- or glycine-rich regions with flanking domains (exons 9-11), in order to establish whether these sequences might mediate specific molecular recognition events important in fibrillin assembly. Our data demonstrate that both recombinant molecules can form extracellular dimers, but highlight subtle differences in the stability of these dimers. Following expression in COS-1 cells, SDS-PAGE analysis showed that glyfib-2 was present intracellularly as monomers, and extracellularly as monomers and disulphide-bonded dimers. Size fractionation in native non-reducing conditions prior to SDS-PAGE analysis highlighted that glyfib-2 also formed non-covalent associations. In contrast, profib-1 appeared monomeric in cells and medium. Using an in vitro translation system supplemented with semipermeabilised HT1080 cells together with chemical crosslinking, dimers of the fibrillin-1 and fibrillin-2 molecules were detected. Dimerisation was not cell-dependent since molecules translated in the absence of cells dimerised, and was not an intracellular event as judged by proteinase K digestions. A crosslinking and coimmunoprecipitation strategy provided a means of investigating whether molecular chaperones might be involved in preventing dimerisation of translocated molecules. Proteinase K-resistant recombinant molecules associated rapidly with BiP, and thereafter with protein disulphide isomerase and calreticulin. Differences between the two fibrillin isoforms in ability to form stable dimers prompted investigation of the proline- and glycine-rich sequences. Differences in solubility and pI were apparent that may contribute to reduced stability of proline-rich region interactions. These studies suggest that extracellular dimer formation mediated by interactions of the proline- and glycine-rich regions may be a crucial early step in the extracellular assembly of fibrillin into microfibrils.

Fibrillin degradation by matrix metalloproteinases: identification of amino- and carboxy-terminal cleavage sites.

Fibrillin molecules form the structural framework of elastic fibrillin-rich microfibrils of the extracellular matrix. We have investigated the proteolysis of recombinant fibrillin molecules by five matrix metalloproteinases. Cleavage sites were defined at the carboxy-terminal end of the fibrillin-1 proline-rich region and the corresponding fibrillin-2 glycine-rich region (exon 10), and within exon 49 towards the carboxy-terminus of fibrillin-1. Cleavage at these sites is predicted to disrupt the structure and function of the fibrillin-rich microfibrils.