HSP47: Drug Discovery
In the course of the last years, several diseases have been linked to Hsp47 and its interaction partners, such as collagen-related and neurodegenerative diseases as well as cancer. Hsp47 is an ER-resident procollagen-specific molecular chaperone which represents a crucial factor for the biosynthesis of collagen. Hsp47 expression levels tightly correlate with that of collagens in numerous tissues and cell types, and Hsp47 deficiency directly affects the function of these cells. Recent evidence suggests that the expression of Hsp47 is down-regulated by malignant transformation of fibroblastic cells with Rous sarcoma virus 230 while its expression is significantly up-regulated during the differentiation of the mouse teratocarcinoma cell line F9 exposed to dibutyryl cyclic AMP and retinoic acid, respectively 177. Notably, the expression of Hsp47 is consistently up- or down-regulated together with alterations in the expression of distinct types of collagens. The modulation of Hsp47 levels thus has therapeutic potential in treating diseases with an altered expression of Hsp47. The in vivo function of Hsp47 has been elucidated in SERPINH1-knockout mice where gene disruption was found to cause severe abnormality in the disposal of mature type I collagen and fibril structures of type I collagen in mesenchymal tissues, resulting in embryonic lethality 231. Moreover, the secretion rate of type I and IV collagens was significantly reduced in cells from SERPINH1-knockout mice, leading to accumulation of procollagen in the ER and its dilation 232. The accumulation of procollagen in the ER of SERPINH1−/− embryos resulted in ER stress-induced apoptosis, as evidenced by the induction of CHOP and the splicing of XBP1 mRNA 233. Misfolded procollagens that aggregate in the ER of SERPINH1−/− cells are not removed by ERAD, but are eliminated by autophagy 233. Treatment with inhibitors of autophagy or antisense RNAs targeting autophagy-related genes significantly increases the formation of procollagen trimer aggregates in the ER and concomitant cell death rates 234, 235. In contrast, activation of autophagy by rapamycin reduces the amount of aggregates 235. These findings reveal the impact of autophagy in blocking cytotoxicity due to agglomeration of misfolded proteins that are not discharged by ERAD.
In order to study the function of Hsp47 in collagen biosynthesis at later stages of the development, a Cre-loxP system has been used to delete the SERPINH1 gene in mouse chondrocytes. Herein, SERPINH1 deficiency was found to cause accumulation and partial aggregation of type I procollagen molecules in the ER, leading to ER stress-mediated apoptosis 236.
A variety of fibrotic diseases including lung fibrosis and those affecting other organs show an up-regulated expression of Hsp47 13, 150-155. These disorders are characterized by arteficial deposition of collagens in the ECM, which disassembles the typical structure and integrity and weakens the physiological functions of any organ 210. In this context, elevated levels of Hsp47 and collagen have been reported to coincide with the onset of liver fibrosis 213, intestinal fibrosis 214, and glomerulonephritis 215. In an experimental glomerulosclerosis model induced by anti-thymocyte antibodies, knockdown of Hsp47 using antisense-RNA against SERPINH1 significantly suppressed the enhanced production of collagen in mouse kidneys 237. With respect to the lungs, previous studies clearly demonstrated an up-regulation of human Hsp47 in fibrotic lesions of idiopathic pulmonary fibrosis (IPF), idiopathic nonspecific interstitial pneumonia (NSIP), and diffuse alveolar damage (DAD) 152, 211, 212. Vitamin A-storing hepatic stellate cells (HSCs) were recently identified as the key cell type in liver fibrogenesis as they differentiate into collagen-producing myofibroblasts upon exposure to inflammatory conditions 216. In an attempt to efficiently target HSCs, vitamin A-coupled liposomes containing siRNA against Hsp47 were constructed and delivered to HSCs in animal models of human liver cirrhosis 216. In this study, Hsp47 siRNA was found to induce apoptosis in HSCs and to considerably improve liver fibrosis in rats. Owing to these observations, knockdown of Hsp47 could improve liver fibrosis, on the one hand, by decreasing the number of HSCs in the liver and, on the other hand, by inhibiting procollagen secretion from HSCs. It should be pointed out that deletion of SERPINH1 in HSCs led to accumulation and partial aggregation of immature type I procollagen in the ER 236. The data also clearly demonstrated that the inhibition of autophagy culminates in an increase in procollagen accumulation and subsequent induction of ER stress-mediated apoptosis, as evidenced by an up-regulation of Grp78/BiP, CHOP and Grp94. These findings reveal that apoptosis induction in HSCs might be attributed to a combinatorial action of SERPINH1 knockout and autophagy blockage 236. Repression of Hsp47 or pharmacological inhibition of Hsp47 function thus represents a promising therapeutic strategy for treating various fibroses. However, blockage of Hsp47-procollagen interaction in the ER should also be taken into consideration for treating fibrotic diseases. Promising candidates were recently identified by different approaches yielding compounds that may provide the basis for the development of novel antifibrotic therapeutics 238, 239. Notwithstanding the fact that a large number of individuals contemporarily suffer from fibrotic diseases, only single drugs such as pirfenidone (Esbriet®) and nintedanib (Ofev®) have been approved for the treatment of idiopathic pulmonary fibrosis (IPF) in many countries, including the European Union (EU), the USA and Japan.
Fibrotic diseases and cancer share several characteristics; both pathologies are characterized by genetic alterations, uncontrolled cell proliferation, altered cell interactions and communication and tissue invasion. Consequently, most studies identifying Hsp47 as a putative therapeutic target for the development of appropriate treatment drugs have been performed on cancer. Cancer cells often show up-regulated levels of Hsp47 as can be found in pancreatic ductal adenocarcinoma, breast cancer, cervical squamous cell carcinoma, glioblastoma, head and neck squamous cell carcinoma, scirrhous gastric carcinoma, ulcerative colitis-associated carcinomas, clear cell renal cell carcinoma 225, esophageal squamous cell carcinoma (ESCC), and colorectal cancer, often correlating with an unfavourable prognosis 164, 225-227. In contrast, a down-regulation of Hsp47 was determined in cancerous laryngeal tissues from laryngeal squamous cell carcinoma (LSCC) compared to non-cancerous tissues and this down-regulation correlated significantly with poor prognosis in LSCC patients 228. New evidence has emerged to demonstrate the existence of a binding site for miR-29 in the 3-UTR region of the SERPINH1 gene. The amount of miR-29, associating with this region, has been found as being inversely correlated with the expression of Hsp47 in human breast cancer tissues 167. Also, silencing of the Hsp47 expression or introduction of miR-29 in breast cancer cells suppressed the malignant phenotype of the tumor cells in vitro and considerably blocked tumor growth in vivo, together with a reduced deposition of collagen and fibronectin. The authors provide further evidence for the capacity of miR-29 to suppress the expression of multiple ECM network genes in mammary epithelial cells. These data identified Hsp47 as being a down-stream target of miR-29 which promotes breast cancer progression by modulating the deposition of the ECM 167. Accordingly, a simultaneous up-regulation of Hsp47 and down-regulation of miR-29 have also been described in cervical squamous cell carcinoma, pancreatic ductal adenocarcinoma, scirrhous carcinoma of the stomach, and lung cancer 183, 187, 240, 241. In the latter case, down-regulation of miR-29a caused over-expression of SERPINH1 and LOXL2 in lung cancer and idiopathic pulmonary fibrosis (IPF), suggesting a pathophysiological role of these genes in these disorders. Owing to the above elucidation, miR-29 is obviously critically involved in the regulation of ECM genes implying miR-29 as being a potent inhibitor of cancer cell migration and invasion. In line with this notion, pharmacological inhibition or knockdown of Hsp47 was able to reduce the levels of secreted Aβ peptides in AD patients suggesting Hsp47 as being a potential target for preventing the formation and/or growth of amyloid plaques in AD 224. However, it should be kept in mind that down-regulation of Hsp47 by antisense oligonucleotides in rabbit chondrocyes and human chondrosarcoma cells induced TNFα-independent inflammatory cartilage-degrading pathways involving iNOS and MMP-9 242. Therefore, Hsp47 down-regulation is nevertheless evidently unsuitable for the treatment of patients with RA even though the determination of Hsp47 levels in RA patients might represent a reliable marker for quantifying SFs in human synovial tissue 220. Recent evidence suggests the involvement of miR-29 in the pathogenesis of liver fibrosis. Work by Roderburg and colleagues demonstrated a significant down-regulation of miR-29a in the serum of patients with advanced liver cirrhosis compared to healthy controls or patients with early fibrosis 243. The down-regulation of miR-29 in murine hepatic stellate cells (HSCs) could be identified as being induced by TGF-β and inflammatory mediators, including nuclear factor kappa B (NF-kB). Furthermore, over-expression of miR-29b in murine HSC resulted in down-regulation of Col1a1, Col4a5, and Col5a3 expression 243. Collectively, these data validate Hsp47 as a prospective novel and key therapeutic target for future treatments of a conceivably wide spectrum of disorders.