Resistance to RevM10 inhibition reflects a conformational switch in the HIV-1 Rev response element

Abstract

Nuclear export of certain HIV-1 mRNAs requires an interaction between the viral Rev protein and the Rev response element (RRE), a structured element located in the Env region of its RNA genome. This interaction is an attractive target for both drug design and gene therapy, exemplified by RevM10, a transdominant negative protein that, when introduced into host cells, disrupts viral mRNA export. However, two silent G->A mutations in the RRE (RRE61) confer RevM10 resistance, which prompted us to examine RRE structure using a novel chemical probing strategy. Variations in region III/IV/V of mutant RNAs suggest a stepwise rearrangement to RevM10 resistance. Mass spectrometry was used to directly assess Rev ''loading'' onto RRE and its variants, indicating that this is unaffected by RNA structural changes. Similarity in chemical footprints with mutant protein implicates additional host factors in RevM10 resistance.chemical footprinting ͉ HIV RNA evolution ͉ mass spectrometry ͉ Rev/RRE interaction E xport of a subset of HIV-1 mRNAs is a crucial process in the life cycle mediated by the Rev protein. Although completely spliced viral mRNA is transported to the cytoplasm without a need for Rev, export of unspliced or partially spliced transcripts mandates binding of Rev to the Rev response element (RRE), a structural feature present in these RNA species. The Rev-RRE complex subsequently interacts with cellular factors, such as CRM1 (1, 2); Ran-GTP (3); and, perhaps, eIF-5A (4), to facilitate export.The Rev nuclear localization signal (NLS), located near the N terminus, is an arginine-rich domain (5) that mediates import (6) and RRE binding (7,8). The NLS is also flanked by regions implicated in Rev multimerization (9, 10). The nuclear export signal (NES), located near the C terminus, is a leucine-rich domain (11, 12) that interacts with the nuclear export factor CRM1 (2). Because Rev function is vital for viral replication, it is an attractive target for drug design and gene therapy. In one study, a Rev variant (RevM10), carrying two mutations in the NES, induced a transdominant negative phenotype and, when introduced into human T-cells, disrupted export of Revdependent viral mRNAs (13). A number of gene therapy vectors (14-16) have been devised to deliver RevM10, and a phase 1 clinical trial was been performed in ref.14.The precise mechanism by which RevM10 inhibits Rev function is unknown. Selection for viral resistance in infected cell cultures, using virus derived from the NL4-3 molecular clone, demonstrated, surprisingly, that silent mutations at nucleotides (nts) 164 (G:A164) and 245 (G:A254) in the RRE (designated RRE61) induced resistance to RevM10 (17). Because the Env sequence is unaffected by these mutations, this implies structural differences in the mutant RNA. We therefore used selective 2Ј-hydroxyl acylation analyzed by primer extension (SHAPE) (18) to derive secondary structures of WT and mutant RREs at single nucleotide resolution.Our work illustrates that the two silent G:A mutations create an RRE with s...

Analysis of the Relationship between the Structure and Function of the HIV-1 Rev Response Element (RRE)

“…Although the mechanism of resistance mediated by single nucleotide changes remains somewhat mysterious, our recent studies have unexpectedly shown that the changes cause a rearrangement of stem loops III, IV and V of the RRE into an alternative stable secondary structure that is more similar to the 5 stem-loop structure than the 4 stem-loop structure (137). Together these studies again highlight the point that regions in the RRE outside of the primary Rev binding site are important for Rev function.…”

“…Interestingly, the secondary structure determination of an in-vitro transcribed and folded 232-nt short NL4-3 RRE by the same SHAPE technology gave a different result (137). This study reported that the RRE forms an alternative 4 stem-loop structure, which structurally differs from the 5 stem-loop structure only in the rearrangement of stem-loops III and IV ( Figure 5B).…”

“…Past studies (32,57,128,137,155,238) on the secondary structure of the subtype B RRE have either supported a 5 stem-loop structure or a 4 stem-loop structure. Early studies on the secondary structure of the RRE were done on the HXB2 short RRE, which supported a 5 stem-loop structure.…”

“…Other studies (32,137,155) have reported that the RRE forms an alternative 4 stem-loop structure (Figure 5B), which structurally differs from the 5 stem-loop structure only in the rearrangement of stem-loops III and IV. In the 5 stem-loop structure, the stem-loop III and IV form separate stemloop structure whereas in the alternative structure, stem-loops III and IV combine to form a single stem-loop III/IV.…”

“…Although there is evidence (77,137,224) that stem loops III, IV, V of the RRE contribute to Rev-RRE function, the precise functional role of these regions is not well understood. In this thesis, we have re-examined the secondary structure of the RRE and compared the functional importance of the 5 stem-loop and the 4 stem-loop RRE structures.…”