Characterization of the HIV-1 integrase chromatin- and LEDGF/p75-binding abilities by mutagenic analysis within the catalytic core domain of integrase

Abstract

Background: During the early stage of HIV-1 replication, integrase (IN) plays important roles at several steps, including reverse transcription, viral DNA nuclear import, targeting viral DNA to host chromatin and integration. Previous studies have demonstrated that HIV-1 IN interacts with a cellular Lens epithelium-derived growth factor (LEDGF/p75) and that this viral/cellular interaction plays an important role for tethering HIV-1 preintegration complexes (PICs) to transcriptionally active units of host chromatin. Meanwhile, other studies have revealed that the efficient knockdown and/or knockout of LEDGF/p75 could not abolish HIV infection, suggesting a LEDGF/p75-independent action of IN for viral DNA chromatin targeting and integration, even though the underlying mechanism(s) is not fully understood.

Results: In this study, we performed site-directed mutagenic analysis at the C-terminal region of the IN catalytic core domain responsible for IN/chromatin binding and IN/LEDGF/p75 interaction. The results showed that the IN mutations H171A, L172A and EH170,1AA, located in the loop region 170EHLK173 between the alpha4 and alpha5 helices of IN, severely impaired the interaction with LEDGF/p75 but were still able to bind chromatin. In addition, our combined knockdown approach for LEDGF/p75 also failed to dissociate IN from chromatin. This suggests that IN has a LEDGF/p75-independent determinant for host chromatin binding. Furthermore, a single-round HIV-1 replication assay showed that the viruses harboring IN mutants capable of LEDGF/p75-independent chromatin binding still sustained a low level of infection, while the chromatin-binding defective mutant was non-infectious.

Conclusions: All of these data indicate that, even though the presence of LEDGF/p75 is important for a productive HIV-1 replication, IN has the ability to bind chromatin in a LEDGF/p75-independent manner and sustains a low level of HIV-1 infection. Hence, it is interesting to define the mechanism(s) underlying IN-mediated LEDGF/p75-independent chromatin targeting, and further studies in this regard will help for a better understanding of the molecular mechanism of chromatin targeting by IN during HIV-1 infection.

Figure 1

Identification of chromatin binding sites within IN CCD. A). 293T cells were transfected with different CMV-YFP-IN expressors (including the wild type IN and different mutants, as indicated). At 48 h post-transfection, cells were fractionated into chromatin-bound (lower panel) and non-chromatin-bound (upper panel) fractions as described in Materials and methods. YFP-IN in each fraction was analyzed by IP and WB with anti-GFP antibody. B). The intensity of both the chromatin-bound and non-chromatin-bound YFP-IN was densitometrically determined. The data are presented as the percentage of chromatin-bound YFP-IN to total input. Results are representative of two independent experiments.

Figure 2

Identification of LEDGF/p75-binding sites within IN CCD. A). The CMV-YFP-INwt/mut or CMV-YFP plasmid was co-transfected with SVCMVin-T7-LEDGF expressor in 293T cells. After 48 h of transfection, 90% cells were lysed and subjected to co-IP assay. The IN bound T7-LEDGF/p75 was precipitated by using rabbit anti-GFP antibody and detected by WB using mouse anti-T7 antibody (upper panel). 10% cells were lysed with 0.5% NP-40, directly loaded on 10% SDS-PAGE gel and probed with anti-T7 or anti-GFP antibody to detect T7-LEDGF or YFP-IN expression (middle or lower panel). B). The intensity of protein bands was densitometrically determined. Results were expressed as the ratio of bound T7-LEDGF/p75 expression (mutants/wild-type) which was normalized by total input. Binding affinity to LEDGF/p75 of YFP-IN wild type was arbitrarily set as 100%. Results are representative of two independent experiments.

Figure 3

Differential effects of IN mutations within ₁₇₀EHLK₁₇₃ region on chromatin- and LEDGF-binding. A). Diagram of amino acids sequence and introduced mutations in HIV-1 IN ₁₇₀EHLK₁₇₃ domain. B). Chromatin binding profiles of IN double mutants within ₁₇₀EHLK₁₇₃. 293T cells were mock-transfected or transfected with equal amount of CMV-YFP-IN wild type or double mutants EH170,1AA, EK 170,3AA, HL171,2AA and HK171,3AA. At 48 h post-transfection, cells were fractionated into chromatin-bound and non-chromatin-bound fractions as described in Materials and methods. YFP-IN in each fraction was analyzed by IP and WB with anti-GFP antibody. Chromatin binding affinity was quantified by laser densitometry and results are shown as the percentage of chromatin-bound to total input of YFP-IN (lower panel). C) LEDGF-binding affinity within IN ₁₇₀EHLK₁₇₃ by co-IP assay. 293T cells were co-transfected with the SVCMVin-T7-LEDGF/p75 expressor and CMV-YFP-INwt/mut plasmid as indicated. After 48 h of transfection, 90% of cells were lysed and subjected to co-IP assay as described before. The upper panel showed the bound T7-LEDGF/p75 in each sample. 10% of cell lysates were used to detect the expression of YFP-INwt/mut and T7-LEDGF/p75 by WB using anti-GFP and anti-LEDGF antibodies respectively (middle panel and lower panel). D). LEDGF-binding affinity within IN ₁₇₀EHLK₁₇₃ detected by chemiluminescent co-IP assay. AcGFP1-INwt/mut or AcGFP1-C and ProLabel-LEDGF fusion proteins were coexpressed in 293T cells. After 48 h of transfection, cells were lysed and immunoprecipitated with anti-GFP antibody and the chemiluminescent signals from ProLabel-LEDGF present in the complexes were measured by using ProLabel Detection Kit II and valued as relative luminescence units (RLU). Results are representative of two independent experiments.

Figure 4

Subcellular localization of IN ₁₇₀EHLK₁₇₃ mutants in COS-7 cells. COS-7 cells were transfected with different CMV-YFP-IN fusion protein expressors as indicated for 48 h. After fixation and permeabilization, cells were incubated with primary rabbit anti-GFP antibody followed by secondary FITC-conjugated anti-rabbit antibodies, and the nuclei were stained with DAPI. Cells were visualized by a Carl Zeiss microscopy (Axiovert 200) with a 63× oil immersion objective.

Figure 5

LEDGF/p75 is not required for chromatin binding of IN. A). Transient knockdown of LEDGF/p75 by siRNA had no effect on IN nuclear localization. 293T cells were transfected with either 20 nM negative control (NC) siRNA or 20 nM si-LEDGF PSIP1HSS146003 for 24 h before transfection with CMV-YFP-IN wild type. At 48 h post-transfection, cells were fixed, permeabilized and detected for YFP-IN and LEDGF/p75 expression by using anti-GFP or anti-LEDGF antibodies. The nuclei were stained with DAPI. B). Analysis of chromatin binding affinity of IN on LEDGF/p75 knockdown cells. The lentiviral shRNA-mediated LEDGF/p75 stable knockdown 293T cells were transfected with 20 nM si-LEDGF for 48 h and further transfected with YFP-IN wild type or mutant V165A and were analyzed for its chromatin binding affinity. In parallel, cells were either mock-transfected or transfected with negative control siRNA to study chromatin binding of YFP-IN wild type. The chromatin bound and non-chromatin-bound fractions of YFP-IN wild type or V165A were showed as indicated. The LEDGF/p75 expression level in each sample was verified by WB with anti-LEDGF antibody. Endogenous beta-actin was used for normalization of sample loading.

Figure 6

A) The differential replication profiles of IN mutant viruses within the loop ₁₇₀EHLK₁₇₃ on HIV-1 single-cycle replication. 293T cells were co-transfected with an RT/IN/Env-deleted HIV-1 provirus NLlucΔBglΔRI, each CMV-Vpr-RT-IN (wt/mut) expressor and a VSV-G expressor to generate single round replication competent virus. To test the effect of different IN mutant viruses on HIV-1 infection, C8166 T cells were infected with equal amount of VSV-G pseudotyped IN mutant viruses (adjust by p24 level) for 48 h. 1 × 10⁶ cells were collected and cell-associated luciferase activity was measured by luciferase assay at 48 h post-infection. B) Followed by 12 h infection with single cycle replicating viruses on dividing C8166 T cells, total DNA was extracted and amplified for total viral DNA and human β-globin gene using corresponding primers by Real time PCR. Total HIV-1 DNA levels were expressed as copy numbers per cell, with DNA template normalized by the amplification of the β-globin gene. NC: negative control or 70°C inactivated wt virus.