Overall structure of the Hsp47/procollagen complex.
Hsp47 is a procollagen-specific molecular chaperone, crucially involved in the biogenesis and structural assembly of collagen 2, 3. Hsp47, also termed Serpin H1, is a non-inhibitory serpin belonging to clade H of the SERPIN superfamily of proteinase inhibitors. The human SERPIN superfamily comprises 35 putatively functional protein-coding genes and six pseudogenes, while in mice 60 protein-coding genes and 12 pseudoegenes have been described 34. The corresponding gene products differ from each other by expression level, subcellular location and amino acid constitution (see also section “Family Members“ and Table 2). Serpins are classified into clades due to their sequence similarity and phylogenetic relationship. 16 Clades have been defined (termed A–P), with 9 clades (A–I) covering human serpins (Table 2). The two largest clades of serpins identified so far are the extracellular ‘clade A’ members and the intracellular ‘clade B’ serpins. Hsp47/Serpin H1 is the ER-resident representative of the huge SERPIN superfamily and one of the predominant ER polypeptides.
In-depth investigations have been done to identifying Hsp47/serpin isoforms and their relation to physiology and pathophysiology. Three isoforms as the result of alternative splicing have been described for Serpin A1 and Serpin A3. However, no experimental confirmation is available for the shorter isoforms apart from the canonical sequence. Interestingly, seven isoforms have been reported for Serpin A9 with isoform 9 bearing an additional amino acid stretch of 18 amino acid at the N-terminus compared to the canonical sequence. Multiple isoforms have also been characterized for Serpin B1, -B3, -B5, -B7, -B11, ‑13 as well as Serpin E1-3, Serpin F2, and Serpin G1. For instance, isoform 3 of Serpin G1 harbors a GFLEPQ insertion at position 17 yielding a putative protein 505 amino acid residues in length. Isoform 2 of Serpin F2 bears an LAL → VQP substitution at position 120 – 122 and lacks the amino acid stretch 56 – 119 of the canonical sequence. Further serpins with amino acid insertions comprise isoform 3 of Serpin E2, and isoform 2 of Serpin B12, respectively. Deletions yielding shorter amino acid stretches can be found, e.g. in isoform 2 of Serpin B1, -B3, ‑B5, ‑B7, ‑B11 and ‑13 as well as in isoform 3 of Serpin B11. As mentioned earlier, no data are available on the biological or pathological significance of these serpin isoforms. For more detailed information on serpin isoforms please refer to the UniProt data base. Noteworthy, a variety of human diseases have been attributed to SERPIN gene and/or protein variations. Mutations in SERPINA1 have been linked to emphysema, hepatitis, and hepatocellular carcinoma 97-99 whereas mutations in SERPINA10 and SERPINC1 have been found as being associated with pregnancy complications 100 and familial venous thromboembolic disease 101, respectively. Mutations in SERPINH1 and SERPINF1 have been linked to osteogenesis imperfecta 102, 103. A single nucleotide exchange in the SERPING1 gene has been identified to cause type II hereditary angioneurotic edema 103. The expression of modified serpins has been described in various pathologies, including renal and cardiovascular injury 104, chronic obstructive pulmonary disease 105, autoimmune disease 58, insulin resistance 106, 58, neurodegeneration 20, 58, and cancer 94, 95, 107-109. In this context, down-regulation and/or intracellular location of Serpin B5 (maspin) has been linked to tumor progression and prognosis, indicating its putative function as a tumor suppressor110. A single amino acid substitution in Serpin I1 has been reported to cause serpin polymerization leading to familial dementia 93, 111.