Functional Studies of CCAAT/Enhancer Binding Protein Site Located Downstream of the Transcriptional Start Site

Cell culture and cell treatments

The U-937 human monocytic cell line (American Type Culture Collection. ATCC, CRL-1593.2) was grown in Roswell Park Memorial Institute medium (RPMI)-1640 media (Cellgro). Media was supplemented with 10% heat-inactivated fetal bovine serum (FBS; Hyclone), antibiotics, (penicillin, 100 U/mL, and streptomycin, 100 µg/mL; Cellgro), glucose (4.5 g/L, Cellgro), sodium pyruvate (1 mM, Cellgro), and HEPES (10 mM, Cellgro). 293T cells were maintained in Dulbecco’s Modified Eagle Medium (ATCC) supplemented with FBS (10%), glucose solution (10%), sodium bicarbonate (2%), and antibiotics (penicillin and streptomycin at 40 µg/mL each). The cells were maintained at 37°C with 5% CO₂.

Cloning and site-directed mutagenesis

The LTR-containing DNA fragment (approximately 640 bp) was derived from the LAI molecular clone of HIV-1. The HIV-1 LAI-LTR was PCR amplified using the forward primer: 5′-GGGGTACCTGGAAGGGCTAATTCACTCC-3′ and reverse primer: 5′-TCCCCCGGGTG TAGAGATTTTCCACA-3′ (Integrated DNA Technologies). The italicized nucleotides indicate the restriction endonuclease binding sites used for cloning. The amplified product was digested with KpnI and Sma I (Promega, Madison, WI) and ligated into a modified pGL3-Basic vector, which contains the firefly luciferase (Luc) gene (Promega), to construct the parental LAI-LTR-Luc expression construct. The parental construct was used as a template for site-directed mutagenesis using the QuikChange mutagenesis procedure as described by the manufacturer (Stratagene) to construct the mutant construct LAI-DS3-9C-Luc. The following primers were used for site-directed mutagenesis, and the nucleotide that was mutated is underlined TAGTCAGTGTGCAAAATCTCTAGC (Integrated DNA Technologies). The LAI-DS3-9C-Luc has been shown to contain the DS3 element with a G-to-C bp change at position 9 of the binding site of the subtype B consensus sequence (a sequence alteration specifically shown to completely abrogate C/EBPβ binding). ⁵ All plasmids used in these studies were sequenced to verify the sequence configurations. Sequences were analyzed using Lasergene software (DNASTAR, Inc.).

The C/EBPβ-2 expression construct was generated by PCR amplification from human C/EBPβ cDNA (Open Biosystems, Human Verified Full-length cDNA Clones, MHS 1011) utilizing forward primer: 5′-CACC ATGGAAGTGGCCAACTTCTACTA-3′ and the reverse primer: 5′-CTAGCAGTGGCCGGAGGAGGCGAG-3′ (Integrated DNA Technologies, Coralville, IA). The italicized nucleotides in the forward primers correspond to sequence necessary for directional cloning into the pcDNA3.1 TOPO vector (Invitrogen), while the underlined portion corresponds to the respective start site of translation. The amplified C/EBPβ PCR product was ligated into the pcDNA3.1 TOPO vector as described by the manufacturer (Invitrogen). The plasmid was verified by sequencing. To confirm proper protein expression, 30 µg of the C/EBPβ construct was transiently transfected into 3.0 × 10⁷ 293 F cells using 293fectin as described by the manufacturer (Invitrogen), and the cell nuclear extract was harvested for further analyses 48 hours posttransfection. Western immunoblot analysis was performed with the nuclear extract using antibody specific for C/EBPβ (C/EBPβ, sc-150, Santa Cruz Biotechnology Inc.) for detection of a 45-kD protein. Electrophoretic mobility shift supershift analyses were performed to identify the specific proteins involved in DNA–protein complex formation (data not shown).

Transient transfection analyses

Exponentially growing U-937 cells were seeded at 1 × 10⁶ cells in 2 mL of growth medium in 6-well plates on the day of transfection. Fugene6 transfection reagent was utilized in the transient transfection as described by the manufacturer (Rocher). Briefly, 1 µg LAI-LTR-luciferase (LAI-LTR-Luc reporter construct or LAI-LTR containing the DS3-9C variant (LTR-DS3-9C-Luc) and 50 ng pRL-TK Renilla luciferase internal control (Promega) were transfected together or cotransfected with other expression vectors: pcDNA3.1-C/EBPβ-2 and/or pcDNA3-Tat86. pcDNA3.1-C/EBPβ-2 is described above. pcDNA3-Tat86 expression vector was provided by Dr. Kamel Khalili (Temple University, Philadelphia, PA). The pcDNA3.1 vector without an insert was used to give each transfection an equal amount of total DNA. Cells were harvested 24 hours posttransfection, and cell lysates were prepared using 50 µL 1× passive lysis buffer (Promega). Luciferase activity was assayed using the dual luciferase assay system as described by the manufacturer (Promega). Normalization to an internal control plasmid was not performed in the experiments with cotransfection expression vectors because previous studies and our results have demonstrated the responsiveness of widely used internal control vectors to cotransfected transcriptional regulators.^45–48 Each value represents the average of triplicate transfection reactions and is representative of at least three independent experiments. The error bars shown in each figure indicate the standard deviation.

Molecular clones and infection experiment

An infectious molecular clone corresponding to the LAI strain of HIV-1 (pLAI.2) was obtained as a glycerol stock from the NIH AIDS Research and Reference Reagent Program (Catalog number 2532, NIH, MD). Escherichia coli containing the molecular clone were grown in Luria-broth (MILLER) supplemented with ampicillin (100 mg/mL) at 30°C, 200 RPM overnight. DNA was isolated using an EndoFree Maxiprep procedure as described by the manufacturer (USB). The 3′ LTR was digested from the molecular clone using AatII and BamHI (NEBiolabs) and ligated into pUC19 (NEBiolabs). The LTR was subjected to site-directed mutagenesis to incorporate 9C mutations into C/EBP-DS3. Mutagenesis primers were the same as utilized in site-directed mutagenesis for constructing LAI-LTR-DS3-9C-Luc described above. The mutated LTR was digested from pUC19 and ligated back into the parental molecular clone. The parental and mutant molecular clones were sequenced completely to confirm the presence of DS3 mutant and the absence of any other mutations in the HIV-1 genome subsequent to the mutagenesis process.

Molecular clone DNA (10 µg) was transfected into 293T cells in 10 cm dishes using the ProFection mammalian transfection system (E1200; Promega). Forty-eight hours after transfection, cell supernatants were collected and assayed for p24 using Enzyme-linked immunosorbent assay (ELISA) as directed by the manufacturer (Perkin Elmer). U-937 cells were seeded at a density of 6 × 10⁴ cells/well in a 96-well v-bottom plate. Cells were then incubated for two hours with 25 ng/mL p24 of molecular clone-derived HIV-1 LAI parental or LAI 9C strains complexed with Transfectam (Promega). Virus–Transfectam complexes were prepared by mixing 25 ng/mL p24 of virus with 5 mg/mL of Transfectam in a total of 0.5 mL of serum-free RPMI. After incubating at 37°C for 45 minutes, the medium volume was increased to 3 mL with RPMI containing antibiotics and FBS to bring the serum concentration to 10%. Following the two-hour incubation with virus, cells were washed and subsequently cultured. The supernatant was collected, and the cells were washed, supplied with new media, and split at 3-day intervals for a total of 12 days. The supernatant from days 3, 6, 9, and 12 was subsequently assayed for HIV-1 production by determining the level of p24 core antigen in the supernatant using an HIV-1 p24 antigen ELISA assay (ZeptoMetrix Corp.). Infectivity was expressed relative to mock-infected cells.

Statistical analysis

The results were statistically analyzed by Student’s t-test. Differences between groups were considered significant if P < 0.05 was obtained.

C/EBP-DS3 affects HIV-1 LTR transcription activated by C/EBPβ in U-937 cells

Although U-937 cells represent a promonocytic cell line, undifferentiated U-937 cells are exclusively susceptible to infection by CXCR4-utilizing (X4) HIV-1 strains and have been utilized in a number of HIV-1 replication studies to examine selected aspects of the viral life cycle. ⁴⁹ In particular, U-937 cells and X4 HIV-1 strains have been previously utilized for identifying upstream C/EBP binding sites required for HIV-1 replication in cells of the monocyte–macrophage lineage.^6,7 Moreover, electrophoretic mobility shift analyses have shown that HIV-1 C/EBP-DS3 is able to form a DNA–protein complex containing C/EBPβ protein from U-937 cell nuclear extract. ⁵ Therefore, the role of C/EBP-DS3 in HIV-1 transcription was first investigated in U-937 promonocytic cells.

U-937 promonocytic cells were transfected with 1 µg parental HIV-1 LAI-LTR-Luc or the LAI-DS3-9C-Luc in the absence or presence of increasing amounts of C/EBPβ-2 (activator) expression plasmid (Figure 1). Western immunoblot assays were performed to demonstrate that the protein expression levels of C/EBPβ were increased in a dose-dependent manner when cotransfected with parental LTR or LTR-DS3-9C (Figure 1C). Expression of increasing amounts of C/EBPβ-2 (from 150, 300, 600 to 1200 ng per reaction) led to a dose-dependent increase in HIV-1 LTR activity in both the parental and DS3-9C LTR, with the parental LTR demonstrating a 2- to 5-fold higher level of activity (Figure 1). Specifically, in the presence of 1200 ng of C/EBPβ expression vector, when comparing to fold over the parental LTR basal activity, the maximal transactivation levels of parental LTR and LTR-DS3-9C were 5.5- and 1.9-fold, respectively (Figure 1A). When the results were analyzed as fold over their own basal activity, the transactivation levels were 5.5- and 3.3-fold, respectively (Figure 1B). These results indicated that, similar to the upstream C/EBP binding sites, C/EBP-DS3 was able to affect HIV-1 transcription through interaction with C/EBPβ. Specifically, the reduced occupancy of DS3 resulted in reduced ability to be transactivated by C/EBPβ.

OPEN IN VIEWER

Figure 1.

DS3-9C exhibited reduced ability to be transactivated by C/EBPβ. U-937 cells were transiently transfected with 1 µg LAI-LTR-Luc or LAI-DS3-9C-Luc in the absence or presence of increasing amounts of C/EBPβ, as indicated in the figures. Twenty-four hours posttransfection, cell lysates were harvested and luciferase activity was measured. (A) Results were analyzed by comparing the fold over the parental basal LTR activity level. (B) Results were analyzed by comparing the fold over its own basal LTR activity level. Asterisk indiscates that P value < 0.001. (C) The protein expression levels of C/EBPβ as demonstrated by Western blot were increased with the increasing amounts of C/EBPβ expression vectors cotransfected with LAI-LTR-Luc (upper) or LTR-DS3-9C-Luc (lower) in U-937 cells.

Tat activates the LTR-DS3-9C variant in U-937 monocytic cells

Previous studies have shown that some transcription factor binding sites located downstream of the HIV-1 transcriptional start site play an important role in the Tat-mediated transactivation of the HIV-1 LTR.^41,50 To determine whether C/EBP-DS3 could also affect Tat transactivation of the LTR, U-937 cells were transfected with 1 µg parental LAI-LTR-Luc or LTR-DS3-9C-Luc reporter vectors in the absence or presence of increasing amounts of Tat. As shown in Figure 2, Tat expression increased the activity of both the parental and DS3-9C-containing LTRs. In the presence of 50 ng of Tat expression vector, the activity of the HIV-1 LTR was similar between the parental and DS3-9C-containing configurations, with elevated activities of 13.8- and 14.2-fold, respectively. In the presence of increasing amounts of Tat expression vectors from 150, 300 to 600 ng, both LTRs exhibited increased activity in a dose-dependent manner. Specifically, the parental LTR transactivation levels were increased up to 20-, 29.6-, and 35.7-fold, respectively, which was higher than the levels of LTR-DS3-9C, 16.7-, 19.7-, and 24.7-fold, respectively (Figure 2A). However, when 1200 ng Tat expression vector was added to the cells, the transactivation level of LTR-DS3-9C (38.9-fold) was similar to that of parental level (36.4-fold). These results indicated that the LTR-DS3-9C variant needed much higher concentrations of Tat to be activated to levels similar to parental LTRs when there were low levels of C/EBPβ present (similar to those detected in U-937 cells). When analyzing the results obtained with the parental and LTR-DS3-9C variant by comparing each LTR to its own basal transcription level, increasing Tat expression increased the transcription levels of DS3-9C-containing LTRs by 26.0-, 31.9-, 38.0-, 47.5-, and 74.3-fold, respectively, which were actually higher than the levels of the parental LTR, 13.8-, 20.0-, 29.6-, 35.7-, and 36.4-fold, respectively (Figure 2B). These results suggest that Tat can help overcome the loss of C/EBP-mediated LTR activation, at least under selected physiological conditions.

OPEN IN VIEWER

Figure 2.

DS3-9C impacts Tat transactivation of the HIV-1 LTR. U-937 cells were transiently transfected with 1 µg LAI-LTR-Luc or LAI-DS3-9C-Luc in the absence or presence of increasing amounts of Tat expression vectors as indicated. Cell lysates were collected and processed for luciferase activity 24 hours posttransfection. (A) Fold over parental basal activity. (B) Fold over their own basal activity. Asterisk indicates that P value < 0.05.

HIV-1 LTR activities are further elevated by C/EBPβ and Tat together

Since the LTR-DS3-9C variant exhibited a lower transactivation capability in the presence of C/EBPβ while achieving normal or even a higher level in the presence of Tat, and both factors have been shown to be important for HIV-1 gene expression in cells of monocyte–macrophage lineage, further investigations to determine whether there was a specific interaction between C/EBPβ, Tat, and the HIV-1 LTR-DS3 variant were performed. U-937 cells were transfected with 1 µg parental or LTR-DS3-9C in the absence or presence of 50 ng Tat expression vector and increasing amounts of C/EBPβ expression vector. As shown in Figure 3, the cooperative interaction between Tat and C/EBPβ-2 has been observed in both LTR configurations. In particular, when analyzed as fold over the parental basal LTR activity level (Figure 3A), increasing amounts of C/EBPβ-2 (100–1200 ng) increased the parental LTR from 26-, 26.3-, 43.0- to 41.3-fold, respectively, while the LTR activity of the LTR-DS3-9C variant were increased by 16.8-, 33.2-, 30.6-, and 16.0-fold, respectively. Unlike the parental LTR, with greater amounts of C/EBPβ-2 (600 and 1200 ng), the transcription level of LTR-DS3-9C was decreased. When analyzed as fold was over their own basal transcription activity (Figure 3B), the maximal transcription level of LTR-DS3-9C-driven transcription was obtained with 50 ng Tat and 300 ng C/EBPβ-2 (55.9-fold), which was higher than the highest level achieved with the parental LTR (43.0-fold). These results indicated that the LTR-DS3-9C variant exhibited a greater ability to be induced by Tat and C/EBPβ-2 together, once LTR-DS3-9C basal level was enhanced to the level of the parental LTR.

OPEN IN VIEWER

Figure 3.

Tat and C/EBPβ were able to cooperatively increase HIV-1 parental LTR and LTR-DS3-9C. (A) and (B) U-937 cells were transiently transfected with 1 µg LAI-LTR-Luc or LTR-DS3-9C-Luc in the absence or presence of 50 ng Tat and increasing amounts of C/EBPβ (100, 300, 600, to 1200 ng) together. (C) and (D) U-937 cells were transiently transfected with 1 µg LAI-LTR-Luc or LAI-DS3-9C-Luc in the absence or presence of 100 ng C/EBPβ and increasing amounts of Tat (100, 300, 600, to 1200 ng) together. Twenty-four hours posttransfection, cell lysates were collected and luciferase activity was measured. The results shown in (A) and (C) indicate the fold over parental LTR basal activity, and the results shown in (B) and (D) indicate the fold over their own basal transcription level.

In contrast, when U-937 cells were transfected with 1 µg parental LAI-LTR-Luc or LAI-DS3-9C-Luc in the absence or presence of 100 ng C/EBPβ expression vector and a series of increasing amounts of Tat expression vectors, LTR activity was increased in both parental and DS3-9C-containing LTRs. The maximum activity of the parental LTR and the LTR-DS3-9C variant occurred with 100 ng C/EBPβ and 600 ng Tat expression vector, 37.7- and 50.2-fold, respectively. In addition, in all Tat levels examined, LTRs containing the DS3-9C configuration resulted in increased LTR activity over the parental LTR demonstrating that loss of C/EBP binding to DS3 can be overcome by increased expression of C/EBPβ (which can be found in the cells of the activated monocyte–macrophage lineage) and increased expression of Tat, leading one to conclude that DS3-9C may be involved in controlling HIV-1 replication in activated monocyte–macrophage environments.

DS3-9C decreases HIV-1 LAI replication in U-937 monocytic cells

To determine the effects of DS3 on HIV-1 replication, the DS3-9C variation was incorporated into an infectious molecular clone of the LAI strain of HIV-1. Molecular clone-derived viral particles were used to infect U-937 cell lines (Figure 4). Levels of p24 in the media were measured at 3, 6, 9, and 12 days postinfection. In U-937 cells, when compared with the parental LAI viral strain, the HIV-1 DS3-9C-containing variant viruses resulted in lower replication levels at each time point examined. However, the maximal p24 level of parental virus was 213 ng/mL at 3 days postinfection, after which time lower levels of virus were observed. However, the replication levels of HIV-1 DS3-9C variant viruses continuously increased from day 3 (p24 level was 53 ng/mL) to day 12 (p24 level was 112 ng/mL) postinfection (Figure 4). Although the HIV-1 LAI strain, a CXCR4-utilizing virus, was examined in this study, it was able to effectively infect U-937 cells, which has also been demonstrated in other studies.^51,52 Therefore, the comparison of the replication levels of these two viruses indicated that the DS3-9C variant exhibited a decreased replication ability at early time points but clearly has the ability to reach parental levels at later time points.

OPEN IN VIEWER

Figure 4.

HIV-1 LAI molecular clones containing DS3-9C demonstrated an altered replication phenotype. Molecular clone-derived viral particles were used to infect U-937 monocytic cells as described in the “Materials and methods” section. Levels of p24 in the media were measured at 3, 6, 9, and 12 days postinfection. The gray line corresponds to the HIV-1 LAI parental virus. The black line corresponds to the LAI virus containing the DS3-9C variant configuration.