ApoB RNA editing demonstrates stringent requirements in the nucleotide sequences that flank the targeted base (Fig. 2). These sequence requirements have been mapped by several groups, the strategy for which was based on the observation that C to U RNA editing of apoB could be recapitulated using synthetic templates containing 30 nucleotides flanking the edited base (25-29). Emerging from the detailed characterization of the cis-acting elements required for C to U RNA editing of apoB (findings described in greater detail below), was the question of whether other transcripts might also undergo similar modification. Extensive database searches of transcripts containing similar cis-acting elements yielded a number of candidates demonstrating homology to the apoB-editing site. Based upon such a homology search, Skuse and coworkers identified the NF1 transcript as containing a potentially edited site, a cytidine base at nucleotide 3916 in the cDNA, positioned in a sequence context with ~ 50% identity to human apoB (Fig. 2) (30,31). C to U RNA editing of NF1 results in modification of a CGA (arginine) codon to a UGA, translational stop codon. Tumors from patients with NF1 manifested C to U RNA editing of the NF1 transcript, and follow-up studies suggested that the extent of RNA editing appeared to correlate with the malignancy grade of the tumor, more malignant tumors demonstrating higher levels of RNA editing (32). What might these findings imply for the biology of NF1 and how does this example fit with what is known about C to U RNA editing?
NF1 is a common genetic disease, affecting ~ 1 in 3000 individuals worldwide. It is also a cancer predisposition syndrome, which results from a combination of germline and somatic mutations in the neurofibromatosis tumor-suppressor gene product, neurofibromin (33). The tumor-suppressor function of neurofibromin is believed to reside in a region that contains intrinsic Ras-GTPase activating protein (GAP) activity and which is homologous to other Ras-GAP proteins conserved from yeast through mammals (reviewed in (33)). Loss of GAP activity leads to increased Ras activity, which in turn promotes cell proliferation and imparts a growth advantage (34). C to U RNA editing of NF1 is predicted to result in translational termination of the edited transcript, with elimination of the GAP domain in neurofibromin, thereby potentially contributing to the growth advantage demonstrated in these tumors. Thus, NF1 RNA editing could potentially represent a novel mechanism whereby the clinical manifestations of this disease may be conditionally modified. This becomes a relevant consideration in the context of understanding the clinical heterogeneity of NF1, which often yields marked differences in disease manifestations even within family members carrying identical germline mutations (33). Based upon these considerations, therefore, a testable hypothesis is that C to U RNA editing of NF1 may function as a conditional genetic modifier of the neurofibromatosis tumor-suppressor gene by inactivating the remaining wild-type allele in subjects, who have inherited a germline mutation in the first allele. The mechanism of such conditional inactivation would reflect translational termination of neurofibromin in tumors that demonstrate C to U RNA editing of NF1. This said, the crucial piece of information yet to be provided is formal proof that the edited NF1 transcript encodes a truncated protein. An alternative possibility is that the edited mRNA, by virtue of the in-frame translational termination codon, becomes unstable and subject to nonsense-mediated decay (35). In either instance, the net effect would be to reduce production of the full-length neurofibromin tumor-suppressor gene product.
What is the relationship of this form of C to U RNA editing to that described for apoB? We have recently revisited the question of C to U RNA editing of NF1 in tumors from patients with neurofibromatosis. These latest findings suggest that C to U RNA editing occurs in only a subset of peripheral nerve sheath tumors and was undetectable in tissues from unaffected subjects. In exploring the features that distinguish tumors, in which C to U RNA editing was demonstrated, two defining characteristics emerged. First, tumors that demonstrated C to U RNA editing were all found to express apobec-1 mRNA, the catalytic deaminase of the apoB RNA-editing holoenzyme (31). This feature is considered particularly relevant since the underlying mechanism of NF1 RNA editing and its relationship, if any, to apoB RNA editing was previously unresolved. It bears emphasis that only those tumors, in which C to U RNA editing of NF1 was found, demonstrated the presence of apobec-1 mRNA. Secondly, alignment of the RNA sequence of human apoB and NF1 over a 40 nucleotide stretch encompassing their respective edited sites reveals 50% identity, but more significantly revealed only 6 of 11 matches in a critical region (described in detail below), referred to as the ''mooring sequence'' (Fig. 2) (26,28). This degree of mismatch in the context of apoB mRNA would eliminate C to U RNA editing (26-28). Accordingly, we reasoned that there might be compensatory alterations in either the primary sequence of the NF1 transcript or in the secondary structure of the relevant region that permits presentation of the targeted cytidine in an appropriate configuration relative to the active site of the deaminase, apobec-1. These studies revealed the presence of a consensus binding site for apobec-1, UUUN[A/U]U, located immediately adjacent and 5' to the edited base in NF1 RNA (Fig. 2) (31,36). This consensus binding site in apoB RNA is located 3' of the edited base and spans the 5' end of the mooring sequence (36). In addition to the consensus binding site for apobec-1, we demonstrated that C to U editing of NF1 RNA was confined to transcripts in which an alternatively spliced downstream exon was included in the template (31). Specifically, a 63-nucleotide exon, 23A, located 3' of the edited base (23-1), was required for editing of cytidine 3916 in NF1 cDNA (31). Analysis of multiple tumors derived from different patients demonstrated a significant correlation between the extent of C to U RNA editing and the relative abundance of the alternatively spliced exon 23A (31). Modeling predictions further supported the hypothesis that the presence of the alternatively spliced downstream exon 23A in NF1 RNA allows the transcript to fold in a manner that resembles the stem-loop/bulge appearance of apoB RNA. This folding is predicted, in turn, to allow presentation of the targeted cytidine in a favorable configuration relative to apobec-1. Thus, the implications arising from these latest studies point toward NF1 RNA editing, representing an opportunistic outcome of several discreet events, including aberrant expression of apobec-1 and differential alternative splicing of the NF1 transcript (31). Even under these conditions, however, the extent of C to U editing of this transcript, generally less than 20%, is considerably less efficient than observed with the canonical substrate, apoB, which in the small intestine is typically ~ 90% (30,31,37). This difference in editing efficiency may be accounted for by sequence divergence in the immediate vicinity of the targeted base in NF1 compared to the conserved sequence in apoB RNA.
Was this article helpful?
This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.