COMMENTARY

Seeing is believing: Structure of the catalytic domain of HIV-1 integrase in complex with human LEDGF͞p75

Christina Marchetti Bradley and Robert Craigie* Laboratory of Molecular Biology, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, 5 Center Drive, Bethesda, MD 20892

etroviruses covalently insert transfer, the 3Ј hydroxyl groups are (2), eliminating the possibility that the their genome into the DNA of joined to opposite strands of the host interaction is weak, indirect, or medi- the host cell and subsequently DNA at sites separated by 5 bp in the ated by DNA or another protein. The coopt cellular machinery for case of HIV-1. Cellular enzymes repair integrase-interacting domain of RDNA replication, transcription, and pro- the resulting intermediate to complete LEDGF͞p75 was mapped independently tein expression (1). These viruses also integration. by two groups using cell-based (11) and exploit cellular proteins to assist in this Integration takes place in a more in vitro and bioinformatics (9) methods. stable insertion of their genetic material complex environment than these bio- Cherepanov et al. (2) report the crystal into the host genome, a process called chemical assays suggest. Before integra- structure of this domain and the cata- integration. In a recent issue of PNAS, tion, the viral cDNA is associated with a lytic subdomain of HIV-1 integrase. Cherepanov et al. (2) reported the first number of viral and cellular proteins in Both HIV-1 and feline immunodeficiency structure of a retroviral integrase, the a large nucleoprotein assembly called virus PICs can be immunoprecipated with viral enzyme that catalyzes integration, the preintegration complex (PIC). Many anti-LEDGF͞p75 antibodies, suggesting in complex with a host protein. This de- of the protein components of the PIC that LEDGF͞p75 is indeed a compo- finitive structural evidence for an inter- are derived from the core of the infect- nent of these PICs (12). But what is the action between HIV-1 integrase and any ing virion, but some are acquired from function of this interaction? The answer other protein adds to the significant evi- to this question is not clear, but many dence for the role of host proteins in groups are currently investigating its integration that has been accumulated This definitive possible effect on integrase activity, PIC over the past decade (ref. 3 and refer- nuclear import and trafficking, and inte- ences therein). structural study adds gration site target selection. The infectious particles of retrovi- Does LEDGF͞p75 play a role in the ruses, called virions, contain two RNA to the significant catalytic steps of breaking and joining copies of their genomes. After viral DNA? HIV-1 integrase carries out these entry and a series of poorly understood evidence for the role reaction steps in vitro without the assis- uncoating steps, the RNA is released tance of other proteins, but the reac- into the host cytoplasm. Here the viral of host proteins in tions lack the full fidelity of integration enzyme reverse transcriptase synthesizes in vivo. In particular, many reaction a double-stranded DNA copy of the integration. products result from insertion of only genome by using the RNA as a tem- one viral DNA end into one strand of plate. The next step is the hallmark of target DNA rather than concerted inte- the retroviral life cycle: the viral cDNA the cytoplasm of the infected cell. The gration of pairs of viral DNA ends. ͞ is transported to the nucleus and is in- viral proteins associated with the HIV-1 LEDGF p75 has been reported to stim- serted into the host genome (1). Inte- PICs include integrase, reverse tran- ulate integration in in vitro assays that gration is required for infection and scriptase, matrix, and Vpr (4–6). In do not distinguish single-end integration ensures the stable association of the addition, the list of host proteins that from concerted integration (7). It would have been reported to associate with be interesting to determine whether viral genome in the host cell for subse- ͞ quent generations. From its new posi- HIV-1 PICs or to interact directly with LEDGF p75 influences the fidelity of tion, the viral genome is transcribed, HIV-1 integrase continues to grow. HIV-1 DNA integration. However, leading to the synthesis of viral proteins These proteins include INI1, HMGA1, other retroviral integrases, such as RSV, and full-length transcripts of the genome BAF, EED, p300, and LEDGF͞p75, the do not interact with LEDGF (13), so a and, ultimately, to new virus particles. topic of this Commentary (reviewed in potential role in the catalytic steps of The chemistry of integration is cata- ref. 3). The role of most of these cellu- integration is not a universal paradigm. lyzed by the viral enzyme integrase, lar proteins in integration is largely Another possible role for auxiliary PIC proteins, such as LEDGF͞p75, is in many copies of which are found in the unknown. nuclear import and targeting. Retroviral virion (1). The integration reaction has Expression of HIV-1 integrase in PICs must enter the host nucleus to been successfully recapitulated in vitro human cells from a synthetic gene led to access their target, and, because lentivi- with recombinant integrase and short the discovery that integrase stably asso- ruses like HIV-1 efficiently infect nondi- DNA oligonucleotides representing the ciates with LEDGF͞p75 (lens epithelium- viding cells, these PICs must in theory viral DNA ends and the target DNA. derived growth factor͞transcription traverse intact nuclear membranes. With this approach, the biochemical coactivator p75) and that this protein Transport of particles into and out of mechanism of integration has been elu- stabilizes the association of integrase cidated. First, a pair of dinucleotides with DNA (7). The interaction between Ј ͞ at the 3 ends of the viral DNA are integrase and LEDGF p75 was con- Conflict of interest statement: No conflicts declared. cleaved, exposing the conserved CA firmed with recombinant proteins by See companion article on page 17308 in issue 48 of volume sequence that marks the boundary solution methods, such as pull-down 102. between the host DNA and the inte- assays (8–10), gel filtration chromatog- *To whom correspondence should be addressed. E-mail: grated viral cDNA. Next, during strand raphy, and analytical ultracentrifugation [email protected].

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0509078102 PNAS ͉ December 6, 2005 ͉ vol. 102 ͉ no. 49 ͉ 17543–17544 Downloaded by guest on September 23, 2021 the nucleus is controlled by the pore interaction with LEDGD͞p75, fails to in the distribution of HIV-1 integration complexes found in the nuclear enve- abolish nuclear import of HIV-1 inte- sites. lope, which permit passive diffusion of grase, although it blocks replication of the The work of Cherepanov et al. (2) particles Ͻ9 nm. The HIV-1 PIC, which virus at the integration step (17). Further- represents a major advance in the field is significantly larger, must therefore be more, depletion of LEDGF͞p75 by small of retroviral integration because of the actively transported into the nucleus and inhibitory RNA had no effect on viral unambiguous nature of the interaction harbor a nuclear localization signal replication or nuclear import of PICs, sug- and the unprecedented crystal structure (NLS) (14, 15). Where is the NLS in the gesting that if LEDGF͞p75 is indeed in- of a complex between integrase and any HIV-1 PIC? Attempts to define an NLS volved in this process it is by a redundant other protein. Like all breakthrough results, this structure and the other that is required for nuclear import of mechanism (12). experimental work present many HIV-1 PICs have yielded inconclusive The final role that PIC proteins can unresolved questions. Beyond the re- results. Although some individual com- play in integration, target-site selection, maining functional questions outlined ponents of the PIC, such as integrase, has received considerable interest of late. above, some structural issues need to be matrix, and Vpr, carry potential NLS Integration confers the potential for long addressed. It would be interesting to signals, none of these has been shown to persistence of the viral genome after in- know the structural details of the inter- be essential for the nuclear import of fection. Whether the integrated provirus action between LEDGF͞p75 and the HIV-1 PICs. LEDGF͞p75 is an alterna- is a harmless or deadly addition to the N-terminal domain of integrase, particu- tive candidate for directing nuclear genome depends on the site of integration larly in light of the fact that the struc- import of HIV-1 PICs, although the and the potential for transcription. The ture of full-length integrase has yet to data are far from unambiguous. mechanisms that regulate transcriptional be determined. The interaction of inte- When HIV-1 integrase is expressed in activity of proviral DNA are not well un- grase and LEDGF͞p75 may represent the absence of other viral proteins, it re- derstood but are likely to involve not only the best hope for obtaining structural sides in the nucleus (7), where it forms a positional effects due to the chromatin information on full-length integrase. complex with LEDGF͞p75. The nuclear environment at the site of integration but Finally, beyond the intellectually com- distribution pattern of integrase mutants also epigenetic mechanisms. Recent find- pelling aspects of this problem is the fact during interphase and of the integrase ings that HIV-1, murine leukemia virus, that integrase, as one of three virally en- mutants binding to condensed chromo- and avian sarcoma virus differ in their coded enzymes, is an important target for developing antiretroviral compounds. In- somes in mitosis correlates with their co- preference for sites of integration (18–22) tegrase inhibitors that specifically block localization with LEDGF͞p75 (8). RNA have led to considerable interest in the ͞ strand transfer are in advanced stages of interference knockdown of LEDGF p75 underlying mechanisms of targeting and development (23). A new frontier for inte- leads to the diffuse redistribution of inte- the possible roles of cellular proteins. grase-targeting drugs would be ones that ͞ grase in cells (8). These data suggest that LEDGF p75 possesses an additional func- inhibit specific interactions in the PIC LEDGF͞p75 could be the factor responsi- tion that may be relevant to target selec- necessary for integration in vivo. Muta- ble for the nuclear import of integrase. To tion: it tethers integrase to DNA and tions of the LEDGF͞p75 binding site on further address this question, Maertens et chromatin (7, 12, 17). In fact, the IN integrase have been shown to block viral al. (16) used deletion and missence mu- Q168A integrase mutation that disrupts replication, suggesting that modulation of tagenesis to isolate the NLS in LEDGF͞ the interaction with LEDGF͞p75 and this binding site by a small molecule or p75; they found a canonical simian virus blocks viral replication at the integration peptide could have serious effects on in- 40-like NLS that transferred nuclear im- step prevents chromosomal tethering of fection, regardless of the function of port activity to an otherwise diffuse ␤- integrase (17). LEDGF͞p75 does not LEDGF͞p75. galactosidase enzyme. A single amino acid interact with integrases from Moloney change in this NLS was sufficient to murine leukemia virus or RSV (13), a This research was supported in part by the exclude mutant LEDGF͞p75 from the feature that may or may not be related to Intramural Research Program of the National Institutes of Health, National Institute of Di- nucleus and to abolish nuclear import of a potential role in target site selection. It abetes and Digestive and Kidney Diseases. HIV-1 integrase (16). However, Q168A, would be interesting to know whether de- C.M.B. is a Paul Sigler͞Argoron Fellow of an integrase mutation that disrupts the pletion of LEDGF͞p75 leads to a change the Helen Hay Whitney Foundation.

1. Brown, P. O. (1997) in Retroviruses, eds. Coffin, 9. Cherepanov, P., Devroe, E., Silver, P. A. & Moisant, F., Ben-Slama, L., Witrouw, M., et al. J. M., Hughes, S. H. & Varmus, H. E. (Cold Spring Engelman, A. (2004) J. Biol. Chem. 279, 48883– (2005) J. Biol. Chem. 280, 25517–25523. Harbor Lab., Plainview, NY), pp. 161–203. 48892. 18. Schroder, A. R. W., Shinn, P., Chen, H. M., Berry, 2. Cherepanov, P., Ambrosio, A. L. B., Rahman, S., 10. Cherepanov, P., Sun, Z., Rahman, S., Maertens, C., Ecker, J. R. & Bushman, F. (2002) Cell 110, Ellenberger, T. & Engelman, A. (2005) Proc. Natl. G., Wagner, G. & Engelman, A. (2005) Nat. Struct. 521–529. Acad. Sci. USA 102, 17308–17313. Mol. Biol. 12, 526–532. 19. Laufs, S., Gentner, B., Nagy, Z., Jauch, A., Ben- 3. Turlure, F., Devroe, E., Silver, P. A. & Engelman, 11. Vanegas, M., Llano, M., Delgado, S., Thompson, ner, A., Naundorf, S., Kuehlcke, K., Schiedlmeier, A. (2004) Front. Biosci. 9, 3187–3208. D., Peretz, M. & Poeschla, E. (2005) J. Cell Sci. B., Ho, A. D., Zeller, W. J. & Fruehauf, S. (2003) 4. Farnet, C. M. & Haseltine, W. A. (1991) J. Virol. 118, 1733–1743. Blood 101, 2191–2198. 65, 1910–1915. 12. Llano, M., Vanegas, M., Fregoso, O., Saenz, D., 20. Wu, X., Li, Y., Crise, B. & Burgess, S. M. (2003) 5. Bukrinsky, M. I., Sharova, N., McDonald, T. L., Chung, S., Peretz, M. & Poeschla, E. M. (2004) Science 300, 1749–1751. Pushkarskaya, T., Tarpley, W.G. & Stevenson, J. Virol. 78, 9524–9537. 21. Mitchell, R. S., Beitzel, B. F., Schroder, A. R., M. (1993) Proc. Natl. Acad. Sci. USA 90, 6125– 13. Busschots, K., Vercammen, J., Emiliani, S., Ben- Shinn, P., Chen, H., Berry, C. C., Ecker, J. R. 6129. arous, R., Engelborghs, Y., Christ, F. & Debyser, & Bushman, F. (2004) PLoS Biol. 2, 6. Miller, M. D., Farnet, C. M. & Bushman, F. D. Z. (2005) J. Biol. Chem. 280, 17841–17847. 1127–1137. (1997) J. Virol. 71, 5382–5390. 14. Fouchier, R. A. & Malim, M. H. (1999) Adv. Virus 22. Narezkina, A., Taganov, K. D., Litwin, S., Stoy- 7. Cherepanov, P., Maertens, G., Proost, P., Res. 52, 275–299. anova, R., Hayashi, J., Seeger, C., Skalka, A. M. Devreese, B., Van Beeumen, J., Engelborghs, Y., 15. Goff, S. P. (2001) J. Gene Med. 3, 517–528. & Katz, R. A. (2004) J. Virol. 78, 11656– De Clercq, E. & Debyser, Z. (2003) J. Biol. Chem. 16. Maertens, G., Cherepanov, P., Debyser, Z., En- 11663. 278, 372–381. gelborghs, Y. & Engelman, A. (2004) J. Biol. 23. Hazuda, D. J., Young, S. D., Guare, J. P, Anthony, 8. Maertens, G., Cherepanov, P., Pluymers, W., Chem. 279, 33421–33429. N. J., Gomez, R. P., Wai, J. S., Vacca, J. P., Handt, Busschots, K., De Clercq, E., Debyser, Z. & En- 17. Emiliani, S., Mousiner, A., Busschots, K., Maroun, L., Motzel, S. L., Klein, H. J., et al. (2004) Science gelborghs, Y. (2003) J. Biol. Chem. 278, 33528–33539. M., Van Maele, B., Tempe, D., Vandekerckhove, 305, 528–532.

17544 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0509078102 Bradley and Craigie Downloaded by guest on September 23, 2021