Abstract
CCNs are structurally related matricellular proteins that are highly expressed in many embryonic and adult tissues, including the skeletal system and tumors, where canonical cap-dependent translation is suppressed under hypoxic environments. CCNs are encoded by mRNAs containing long G/C rich 5'-untranslated regions (5'-UTRs). Given that they are expressed under conditions of cellular stress, it has been suggested that the long G/C-rich regions contain internal ribosomal entry sites (IRES) that allow these mRNAS to be translated under conditions where cap-dependent translation is suppressed. Previously published work supported this possibility. However, recent studies have shown that a number of previously reported cellular IRES elements do not in fact possess IRES activity. Here we aimed to reveal whether the 5'UTRs of CCNs harbor IRES activities. The 5'UTRs of CCN1, 2, and 4 were tested in this study. Our results showed that the 5'UTRs of these genes do not contain IRES elements, but instead appear to contain cryptic promoters. Both promoterless and hairpin-containing dicistronic tests showed that transcription was initiated by cryptic promoter elements in 5'UTRs of CCN1, 2, and 4. When dicistronic mRNAs were translated in vitro or in vivo, no IRES activities were detected in the 5'UTRs of CCN1, 2, and 4. Furthermore, these cryptic promoter activities from 5'UTRs of CCN1, 2, and 4 could be detected in various cell types, including chondrocytes, osteoblasts, and endothelial cells, where the cryptic promoter permitted varying degrees of activation. In addition, the core promoter element of the CCN2 5'UTR was identified. CCNs are expressed under conditions of cellular stress, and it has been suggested that some CCN family members utilize IRES-mediated translation initiation to facilitate this expression. We found no evidence for IRES activity, but rather found that the unusually long 5'UTRs of CCNs 1, 2, and 4 harbor cryptic promoters that showed varying degrees of activity in different cell types. These results suggest that these promoters may contribute to the regulation of CCN genes in vivo.