HIV-1 structural gene expression requires binding of the rev trans-activator to its RNA target sequence

The human endogenous retrovirus K Rev response element coincides with a predicted RNA folding region

“…The biological activity of the orthologs of K-Rev observed in exogenous retroviruses, such as H-Rev and HTLV-I Rex, requires the recruitment of these viral regulatory proteins to a discrete primary RNA-binding site that is presented in the context of a larger region of tightly folded RNA termed a "response element" (Hanly et al+, 1989;Malim et al+, 1989Malim et al+, , 1990Zapp & Green, 1989;Bartel et al+, 1991;Heaphy et al+, 1991;Bogerd et al+, 1992)+ Although the primary binding site is clearly essential for biological activity, it is not sufficient and the H-Rev-or Rex-induced export of a target viral RNA therefore requires an almost full-length RNA response element located in cis (Malim et al+, 1989;Huang et al+, 1991)+ Because the biological activity of Rev or Rex response elements is dependent on their folding, these RNA elements can be effectively identified, and their structure predicted, using computer analysis+ Indeed, structures for several retroviral response elements, including both the ;234-nt HIV-1 RRE and the ;254-nt HTLV-I, were first predicted by computer analysis and only subsequently validated by mutational analysis (Hanly et al+, 1989;Malim et al+, 1989;Le et al+, 1990)+ The K-RRE coincides with a predicted unusual RNA folding region Previously, we and others have used functional assays to roughly map the extent of the HERV-K Rev response element (K-RRE) to between residues 8719 and 9152 according to the prototypic HERV-K10 proviral sequence reported by Ono et al+ (1986) (Magin et al+, 1999;Yang et al+, 1999)+ This 434-nt RNA sequence, which is entirely located in the U3 region of the HERV-K LTR, is considerably larger than either the H-RRE or the HTLV-I RxRE, and it therefore appeared likely that the actual extent of the K-RRE would be smaller than this initial estimate+ To gain insight into the true size of the K-RRE, and to determine whether the K-RRE indeed coincides with an unusual folding region (UFR), we used the same computational approach that previously allowed the definition of not only the H-RRE and the HTLV-I RxRE but also the HIV-2 RRE and the Visna-Maedi Virus RRE (V-RRE) (Hanly et al+, 1989;Malim et al+, 1989;Le et al+, 1990;)+ Initially, we focused our attention on sequences extending from 8504 to 9250 within the HERV-K108 proviral clone (Barbulescu et al+, 1999), a sequence that fully includes, but extends significantly beyond, the functionally defined 8719-9152 K-RRE sequence+ Monte Carlo simulations were used to search for any UFR within this HERV-K sequence by calculation of the statistical significance and thermod...…”

“…Although retroviruses normally infect somatic cells, infection of germ line cells can also occur+ If the infected germ cell then participates in the formation of progeny, the resultant integrated provirus will be vertically inherited and can then become a fixed part of the host, or even the species, genome+ Although comparatively rare, the difficulty of reversing such proviral integrations has led to their accumulation over time and endogenous proviruses now constitute ;0+1% of the entire human genome (reviewed by Patience et al+, 1997)+ Vertical, as opposed to horizontal, transmission of retroviruses does not require virally encoded gene products and the selective pressure to maintain these in an intact form is therefore lost+ Indeed, retroviral proteins can be deleterious to the cell and their functional expression may therefore be actively selected against over time+ In fact, the large majority of endogenous retroviruses bear obvious frame-shift or deletion mutations and are clearly defective+ Although this applies also to the HERV-K family of endogenous retroviruses, several individual HERV-K proviruses appear to be only minimally defective (Barbulescu et al+, 1999) and at least some proviruses encode functional forms of the enzymes critical for the retroviral life cycle, such as reverse transcriptase (Berkhout et al+, 1999)+ Nevertheless, no HERV-K endogenous provirus has been shown to be replication competent and it is probable that HERV-K proviruses that are not obviously defective have nevertheless accumulated deleterious point mutations over time+ The lack of a replication-competent HERV-K means that it is not possible to demonstrate that the K-Rev protein or K-RRE target sequence used in the present analysis are fully biologically active by, for example, demonstrating their ability to support viral replication in culture, as would be the case if an exogenous retrovirus was being analyzed+ Although this is a significant concern, it is nevertheless clear that K-Rev does interact with the K-RRE in a manner that is similar, albeit not identical, to what is seen for the equivalent nuclear RNA export factors encoded by exogenous retroviruses such as HIV-1 and HTLV-I+ Specifically, the K-RRE, like the H-RRE and the RxRE (Hanly et al+, 1989;Malim et al+, 1989), clearly coincides with a region of the HERV-K genome that is very likely to fold into an extensive RNA stem-loop structure (Fig+ 1; Table 1)+ Computational analysis of the secondary structure likely to be adopted by the ;416-nt K-RRE predicts a highly stable RNA structure that, like the H-RRE (Malim et al+, 1989(Malim et al+, , 1990, can be subdivided into a set of stem-loop subdomains (Fig+ 2)+ Efforts to identify the specific binding site for K-Rev on the K-RRE, using a series of in vitro and in vivo assays , revealed the existence of two specific binding sites for K-Rev in the K-RRE, coincident with predicted stem-loop subdomains SLIIb and SLIIf+ Interestingly, a single specific binding site for H-Rev coincident with the predicted SLIII subdomain of the K-RRE (Fig+ 2) was also observed+ Importantly, each of these RNA sequences proved able to specifically recruit either the K-Rev or the H-Rev protein when assayed in the human nucleus, the milieu where this interaction would normally be predicted to occur, and specific binding of K-Rev to the SLIIb and SLIIf subdomains of the K-RRE was also readily detectable in vitro and in the yeast three-hybrid assay+ Together, these data convincingly identify sequences within the K-RRE that have the potential to recruit either the K-Rev...…”

“…Like the K-RRE (Fig+ 2), the H-RRE is predicted to consist of a long basal RNA stem topped by an RNA bulge from which extends a series of smaller RNA stemloop structures termed SLII, III, IV, and V (Malim et al+, 1990)+ In the case of the H-RRE, there is a single primary binding site for H-Rev that maps to SLII+ The precise, individual deletion of stem-loop III, IV, or V does not markedly inhibit H-RRE function, although larger deletions, which in principle also leave SLII intact, do block H-RRE activity (Malim et al+, 1989(Malim et al+, , 1990)+ By analogy, we predicted that precise deletion of SLIIb and/or SLIIf in the K-RRE might block the ability of the K-RRE to respond to K-Rev, whereas the exact deletion of other predicted stem-loop subdomains, such as SLIId, might leave K-RRE function unimpaired+ We therefore precisely and individually deleted each of the predicted SLII stem-loop subdomains indicated in Figure 2, as well as SLIII, and asked whether the resultant K-RRE mutants would respond to K-Rev expression in vivo using the pDM128 assay (Fig+ 7)+ Again, H-Rev and HTLV-I Rex, which presumably bind to a different target than K-Rev in the K-RRE, were used as controls FIGURE 6. Both H-Rev and K-Rev bind to specific subdomains of the K-RRE in the human cell nucleus+ This experiment measures the ability of fusion proteins consisting of H-Tat fused to either H-Rev or K-Rev to activate HIV-1 LTR-based indicator constructs bearing precise substitutions of the indicated RNA element for the H-TAR element+ The parental pBC12/CMV plasmid served as a Negative control (Neg)+ 293T cells were transfected as previously described (Blair et al+, 1998) and induced CAT enzyme levels measured ;48 h after transfection+ The indicated data represent the average of four independent experiments and were normalized to the CAT activity observed in cells cotransfected with pBC12/H-TAR/CAT and pH-Tat/ H-Rev, which was arbitrarily set at 100+ to test whether individual deletions would result in destabilization and/or complete misfolding of RNAs bearing the K-RRE, which should abrogate the response to all three retroviral export factors+ We also hoped that this analysis might help to further define the binding site(s) for either H-Rev or Rex on the K-RRE, if this indeed coincided with a single predicted stem-loop subdomain+ Unfortunately, the results obtained from this quite extensive analysis were only moderately informative, as FIGURE 7.…”

Nucleolar trafficking is essential for nuclear export of intronless herpesvirus mRNA

“…of the unspliced 9-kb and singly spliced 4-kb HIV-1 RNAs out of the nucleus (36)(37)(38)(39), and, as such, Rev is functionally similar to the ORF57 protein from herpesviruses. Although the actual localization of Rev itself to the nucleolus has yet to be shown to be essential for its role in HIV-1 replication, it appears that the nucleolus has a fundamental role to play in HIV-1 replication.…”

Functional mapping of the rev-responsive element of human immunodeficiency virus type 2 (HIV-2): influence of HIV-2 envelope-encoding sequences on HIV-1 gp120 expression in the presence or absence of Rev

“…The RRE has a complex secondary structure consisting of five stem-loops around a central stem that appear to be necessary for the Rev-RNA interaction (Dayton et al, 1989;Olsen et al, 1990). Direct binding of the recombinant HIV-1 Rev protein to the RRE in vitro has been demonstrated and the Rev-binding site was mapped to a stem-loop structure within the 5' portion of the RRE (Daly et al, 1989;Heaphy et al, 1990;Malim et al, 1990;Zapp & Green, 1989). …”