Expression of the c-myc Proto-oncogene Is Essential for HIV-1 Infection in Activated T Cells

PS-ODNs used in this study were synthesized by Oligo Etc. Sequences used were as previously described (23): c-myc antisense, AACGTTGAGGGGCAT, located in exon 2 of initiation site of translation; sense c-myc, ATGCCCCTCAACGTT; non-sense, AGTGGCGGAGACTCT; and scrambled, AAGCATACGGGGTGT containing a GGGG motif (32). The oligonucleotides were dissolved in 30 mM Hepes (pH 7.0). Purified mAbs to human CD8 (G10-1, IgG2a), CD16 (FC-2, IgG2b), CD20 (1F5, IgG2a), and HLA-DR (HB10a, IgG2a) were produced in our lab and used to purify human primary CD4⁺ T cells as previously described (14). Goat anti–mouse IgG conjugated to magnetic microbeads was purchased from Miltenyi Biotec. mAbs to human CD3 (64.1, IgG2a) and CD28 (9.3, IgG2a) were used to activate CD4⁺ T cells as previously described (14). Phospho–c-Myc (Thr58/Ser62) polyclonal antibody was purchased from New England Biolabs. Anti–human c-Myc mAb (9E10, IgG1), rabbit polyclonal antibody specific to NH₂-terminal region 1–262 amino acids of c-Myc (N-262), and rabbit polyclonal anti-ERK1 (c-16) antiserum were obtained from Santa Cruz Biotechnology. PE-conjugated anti–HIV-1 p24 protein mAb was purchased from Coulter Corp. TUNEL (TdT-mediated dUTP nick-end labeling) detection kits were obtained from Boehringer Mannheim.

Enriched preparations of human CD4⁺ T cells were isolated from peripheral blood samples from healthy, HIV-seronegative donors as follows: PBLs were obtained by centrifugation over Ficoll-Hypaque, and then E-rosette–positive (Er⁺) cells were isolated as previously described (33). CD4⁺ T cells were obtained by negative selection of Er⁺ cells depleting CD8⁺, CD16⁺, CD20⁺, and HLA-DR⁺ cells with mAb-coated beads. The purity of isolated CD4⁺ cells was >97% as monitored by flow cytometry. Cells were cultured in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum, 2 mM glutamine, 10 U/ml penicillin, 10 mg/ml streptomycin, 1 mM pyruvate, and nonessential amino acids.

HIV-1 strain Lai was prepared as previously described (14). Cells were infected with HIV-1 at a multiplicity of infection of 0.01 per cell.

DNA was extracted from HIV-1– and heat-inactivated HIV-1–infected cells as previously described (14). PCR was performed as described (14) with some modifications including: 50 ng of DNA/sample for amplification of β-globin, 100 ng for LTR/LTR products, 250 ng for LTR/ gag products, and 750 ng for LTR/circle products. PCR mixtures contained 1 μM of each primer, 200 μM each of the four deoxynucleoside triphosphates, 1.5 mM MgCl₂, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, and 0.2 U Tag DNA polymerase (GIBCO BRL). The final volume was 50 μl. The reaction was subjected to 32 cycles (30 cycles for β-globin) of denaturation for 45 s at 94°C, annealing for 1 min at 60°C, and elongation for 2 min at 70°C. PCR products were subjected to 2% agarose gel containing 0.01 μg/ml ethidium bromide and were visualized by UV light. Primers used in this study have been described (14).

After various treatments, 5 × 10⁶ primary CD4⁺ T cells were lysed in 500 μl lysis buffer (2% NP-40, 0.5% sodium deoxycholate, 0.2% SDS, 25 mM Tris-HCl, 50 mM NaCl, 1 mM PMSF, 1 mM Na₃VO₄, 10 μM E-64 [trans-epoxysuccinylt-l-leucylamido (4-guanidino)-butane], 1 μg/ml pepstatin, 10 μg/ml leupeptin, and 0.1% aprotinin). After incubation on ice for 30 min the cells were sonicated. The cell lysates (equivalent to 10⁶ primary CD4⁺ T cells) were mixed with 2× SDS loading buffer (125 mM Tris-HCl [pH 6.8], 4% SDS, 20% glycerol, 83 mM dithiothreitol, and 0.02% bromophenol blue), incubated at 100°C for 5 min, electrophoresed by 8% SDS-polyacrylamide gel, and then transferred to nitrocellulose membranes (Schreicher & Schuell). The membranes were blocked with 5% nonfat milk-TBST (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 0.1% Tween 20) at 4°C overnight, followed by incubation with primary antibodies (in 5% BSA-TBST) at 4°C overnight. After washing, membranes were incubated with horseradish peroxidase–conjugated second antibodies at room temperature for 1 h. Bands on the blotted membranes were detected by incubation with enhanced chemiluminescence reagent (ECL) (Amersham) for 1 min and exposure to Kodak X-Omat film (Eastman-Kodak Co.).

Apoptotic cell death followed by HIV-1 infection was detected by TUNEL according to the manufacturer's protocol (Boehringer Mannheim). After TUNEL staining, cells were resuspended in 100 μl PBS containing 1% BSA; PE-conjugated anti–HIV-1 p24 mAb was added and incubated at 4°C for 20 min. The cells were washed with cold PBS, suspended in 1% paraformaldehyde (Sigma Chemical Co.), and kept at 4°C in the dark until flow cytometry analysis by means of FACScan^® (Becton Dickinson). Cell cycle stages were determined by measuring DNA content with propidium iodide as previously described (34).

Cell proliferation was estimated by [³H]thymidine incorporation: 10⁵ primary CD4⁺ T cells in the presence or absence of oligodeoxynucleotides were stimulated with CD3 (10 μg/ml) and CD28 (20 μg/ml) mAbs in triplicate in 96-well plates. The cells were incubated at 37°C in a 5% CO₂ incubator for 3 d. Each well was pulsed for 16 h with 0.5 μCi [³H]thymidine, and then the incorporation of [³H]thymidine radioactivity was monitored by a beta counter.

In our previous study we found that HIV-1 nuclear import required a CSA-sensitive pathway, and that both TCR and CD28 ligation are essential for this process (14). Similarly, the expression of c-Myc in primary CD4⁺ T cells required costimulation with CD3 and CD28 mAbs (Fig. 1 A); neither CD3 nor CD28 ligation alone induced c-Myc expression. Time course experiments showed that c-Myc expression increased by 4 h, peaked at 24 h after costimulation, and was sustained for 48 h. Moreover, CSA inhibited c-Myc expression (Fig. 1 A, bottom). Because of this correlation, we tested whether c-Myc might be a key regulator of HIV-1 DNA nuclear import in primary T cells. Since no c-Myc–specific inhibitor is yet available, we used a c-myc antisense PS-ODN to inhibit c-Myc function. By competitively inhibiting HIV-1 reverse transcriptase binding to the virus genome–cellular primer complex, PS-ODNs have an inhibitory effect on the initiation of HIV-1 reverse transcription in a sequence-independent manner (35, 36). However, sequence-independent PS-ODNs do not exhibit any anti–HIV-1 activity once initiation of virus reverse transcription has begun (35–38). To avoid nonspecific anti-HIV activity of sequence-independent PS-ODNs, we first infected activated CD4⁺ T cells with HIV-1 for 24 h and then administrated graded doses of c-myc antisense, sense, or non-sense PS-ODNs to the infected cells. As shown in Fig. 1 B, initiation of reverse transcription (LTR/LTR product) and full-length viral DNA synthesis (LTR/gag product) were not affected by the c-myc antisense PS-ODN.

However, nuclear import of HIV-1 DNA (LTR circles) was blocked by c-myc antisense PS-ODN even at doses as low as 1 μM. Doses below to 0.2 μM were less efficient at inhibiting LTR circle formation (data not shown). Neither c-myc sense nor non-sense PS-ODN had any effect on viral DNA nucleus translocation up to 8 μM (Fig. 1 B). Consequently, HIV-1–infected cells treated with c-myc antisense PS-ODN did not produce p24 gag protein or undergo apoptosis (Fig. 1 C). Under conditions in which HIV-1 had already entered the nucleus (e.g., at 48 h), c-myc antisense PS-ODN did block viral p24 expression (data not shown). Lack of an effect by c-myc antisense PS-ODN on full-length viral DNA synthesis was not simply because the oligonucleotides were added too late to the cultures (after 24 h infection), as full-length viral DNA was not detectable until at least 40 h after HIV infection in activated CD4⁺ T cells (reference 14 and data not shown). Thus, c-myc antisense PS-ODN apparently selectively acts on the stage of HIV-1 DNA nuclear import.

We next studied whether c-myc antisense PS-ODN specifically inhibited full-length c-Myc protein expression. Using mAb 9E10 specific to the COOH-terminal end of c-Myc (39), we consistently observed that in the presence of c-myc antisense, sense, or non-sense PS-ODNs, the two major forms of c-Myc proteins, p64 and p67, remained relatively unchanged (Fig. 2). However, c-myc antisense PS-ODN selectively induced the accumulation of 46- and 50-kD proteins, whose expression levels were higher than that of the full-length c-Myc. Neither the c-myc sense nor non-sense PS-ODN induced accumulation of these two proteins (Fig. 2, top). These data are consistent with previous studies showing that expression of c-Myc short (c-MycS) proteins in some tumor cell lines arised from two translational initiation sites downstream of the full-length c-Myc start codon (40–45). These downstream-initiated c-MycS proteins lack most of the NH₂-terminal transactivation domain; they are produced through a leaky scanning mechanism, since optimization of the traditional initiation codon for full-length c-Myc results in less synthesis of the c-MycS proteins (45). Because the c-myc antisense oligonucleotide we used corresponds to the initiation site of full-length c-Myc mRNA, and the two smaller proteins we detected are about the same size as c-MycS isoforms, it seemed likely that the 46- and 50-kD proteins are produced through the same mechanism leading to deletion of the NH₂-terminal region. This possibility was substantiated by the fact that antibodies specific to either NH₂-terminal phosphorylated Thr58/Ser62 or the whole NH₂ terminus region of c-Myc failed to recognize 46- and 50-kD proteins (Fig. 2, middle and bottom). However, both antibodies were able to recognize p64 and p67 full-length c-Myc, which did not change expression in cells treated with different PS-ODNs (Fig. 2). The same result was obtained in a CD4⁺ lymphoid cell line, CEM (data not shown). Thus, the c-myc antisense oligonucleotides, but not control PS-ODNs, selectively induce NH₂-terminally truncated c-Myc proteins that are known to act as dominant negative inhibitors by competitively suppressing full-length c-Myc functions (45–48). Blockage of HIV-1 DNA nuclear import by c-myc antisense PS-ODN most likely is mediated by these NH₂-terminally truncated c-Myc proteins.

Finally, we tested whether c-myc antisense PS-ODN could inhibit the entry of cell cycle and proliferation induced in primary CD4⁺ T cells after TCR and CD28 ligation. Treating CD4⁺ T cells with 6 μM of c-myc antisense oligonucleotide, which efficiently blocked HIV-1 LTR circle formation, could not inhibit cell cycle progression (Fig. 3 A). Similarly, c-myc antisense, sense, and non-sense PS-ODN had no effects on CD4⁺ T cell proliferation induced by CD3 plus CD28 mAbs (Fig. 3 B). These data are consistent with previous findings that NH₂-terminally truncated c-Myc proteins do not interfere with cell growth (45, 49).

The study presented here reveals a novel function of c-Myc for regulation of HIV-1 nuclear import. Blocking of HIV-1 DNA nuclear import by c-myc antisense PS-ODN appeared to be mediated through the presence of 46- and 50-kD NH₂-terminally truncated c-Myc proteins, which do not affect cell cycle progression or cell proliferation (Fig. 3, A and B). Our data imply that the mechanism by which c-Myc controls HIV-1 DNA nuclear import is distinct from those controlling cell cycle progression. However, precisely where and how c-Myc is required for HIV-1 DNA nuclear import in proliferating CD4⁺ T cells remains to be discovered. NH₂-terminal–defective c-Myc proteins are able to heterodimerize with Max, translocate to nucleus, repress gene expression, stimulate cellular proliferation, and induce cell apoptosis (49). However, c-MycS proteins are not able to activate gene transcription (49). It is probable that c-Myc regulates HIV-1 DNA nuclear import through its transactivation activity by regulation downstream gene expression. The ability of HIV-1 to infect nondividing cells, such as monocytes, terminally differentiated macrophages, mucosal dendritic cells, or γ-irradiated cells, is believed to be a unique feature since oncoretroviruses only can establish infection when the cells undergo mitosis (50–57). The ability of HIV-1 to infect nondividing cells is presumably related to the fact that its PIC can be recognized by the cell nuclear import machinery (58–61) and actively transported through nucleopores (62). Moreover, a cellular serine/threonine protein kinase, mitogen-activated protein kinase (MAPK), can associate with HIV-1 PIC to facilitate nuclear targeting of viral DNA (63–66). It is unclear whether HIV-1 DNA nuclear import in proliferating CD4 T cells is regulated through the identical pathway seen in nondividing cells. A reasonable possibility is that c-Myc affects the expression of genes encoding cellular proteins involved in nuclear transport. Further elucidation of the role of c-Myc in regulation of expression of cellular nuclear importing molecules might help us to understand how c-Myc regulates HIV-1 DNA nuclear import.

We thank Ms. M. Domenowske for preparation of figures; Dr. Aaron Marshall and Ms. Kate Elias for editorial assistance; Drs. James Mullins and Michael Katze for critical review of the manuscript; Drs. Andrew Craxton, Raymond T. Doty, and Aaron Marshall and Mr. Aimin Jing for helpful discussion; and members of the Clark laboratory for technical assistance.

This work was supported by National Institutes of Health grant RR00166.

c-MycS

c-Myc short

CSA

cyclosporin A

PS-ODN

phosphorothioate oligodeoxynucleotide

PIC

pre-integration complexes

TUNEL

TdT-mediated dUTP nick-end labeling