Elevation of FMR1 mRNA levels occurs for unmethylated alleles both within the premutation range and extending into the full-mutation range (Tassone et al. 2000a-c; Salat et al. 2000; Kenneson et al. 2001). The concomitant deficit in FMRP was originally suggested to be the stimulus for increased FMR1 mRNA production, in the absence of any increase in mRNA stability, essentially as a feedback response to lowered protein (FMRP) levels (Tassone et al. 2000a). Recently, increased levels of run-on transcription in a premutation cell line (compared with a normal control) have been observed, providing direct evidence of transcriptional activation for expanded (premutation) alleles (Tassone et al., unpublished results). Moreover, using both quantitative RT-PCR and RNA in situ hybridization experiments, Tassone et al. (unpublished results) demonstrated that higher FMR1 mRNA levels are not due to nuclear sequestration. In particular, FMR1 mRNA is not retained in the nucleus, but is mainly localized in the cytoplasm of lymphocytes carrying either normal or premutation alleles.
As noted already, although FMR1 mRNA levels are increased in the premutation range, FMRP expression is decreased. The FMRP deficit is r(CGG)-dependent and is due to decreased translational efficiency (Primerano et al. 2002). Reduced translational efficiency was observed both in cell lines and in transient transfection experiments using expanded alleles spanning the entire premutation range (Primerano et al. 2002; Chen et al. 2003). Particularly for premutation alleles, a smaller fraction of FMR1 mRNA was found to be associated with polysomes, while the majority of the expanded-repeat mRNA was associated with inactive ribonucleoprotein particles. These findings, namely, increased FMR1 mRNA expression levels and deficit in translation efficiency in premutation alleles, have also been confirmed by in vivo translation experiments using a reporter (luciferase) mRNA with the 5'-UTR of the FMR1 gene, the latter harboring varying numbers of r(CGG) repeats. Interestingly, the decreased translation efficiency, evident in the premutation range, was also observed for an allele near the gray zone (45-54 CGG repeats). Translation efficiency gradually decreased with an increasing r(CGG) repeat number (Chen et al. 2003).
The precise mechanism by which the expanded r(CGG) repeat impedes translation is not understood at present. What is surprising is that translation occurs at all for larger premutation alleles, since the predicted free energies of stabilization of the r(CGG) repeat element would be expected to completely block translation. In this regard, an internal ribosome entry site (IRES) was identified near the 5' end of the 5'-UTR, upstream of the r(CGG) repeat (Chiang et al. 2001). FMR1 IRES activity was found to be of moderate strength compared with that of other known IRESs (Chiang et al. 2001); its role in the regulation of FMRP expression is not known at present. Interestingly, cellular IRESs have been shown to increase the translational efficiency of several den-dritically localized mRNAs, including the microtubule-associated protein 2 (MAP2), the a-subunit of the Ca2+/calmodulin-dependent protein kinase II (a-CaMKII), cytoskeleton-associated protein, arc, dendrin, and neurogranin (RC3) (Pinkstaff et al. 2001). IRESs that can mediate cap-independent translation could be used for a rapid and local synthesis of proteins in dendrites. Although translation at dendrites occurs by both cap-dependent and cap-independent mechanisms, the translation mediated by IRES in the RC3 gene is relatively more efficient in dendrites than in the cell body (Pinkstaff et al. 2001). The finding that five different neuronal mRNAs are translated in den-drites by an IRES-mediated mechanism suggests that IRES sequences may control translation in specific neuronal regions.
Was this article helpful?