Basal shuttle of NF-kappaB/I kappaB alpha in resting T lymphocytes regulates HIV-1 LTR dependent expression

Abstract

Background: In HIV-infected T lymphocytes, NF-kappaB/Rel transcription factors are major elements involved in the activation of LTR-dependent transcription from latency. Most NF-kappaB heterodimer p65/p50 is sequestered as an inactive form in the cytoplasm of resting T lymphocytes via its interaction with I kappaB inhibitors. In these cells, both absolute HIV latency and low level ongoing HIV replication have been described. These situations could be related to differences in the balance between NF-kappaB and I kappaB alpha ratio. Actually, control of I kappaB alpha by cellular factors such as Murr-1 plays a critical role in maintaining HIV latency in unstimulated T lymphocytes. Formerly, our group demonstrated the presence of nuclear I kappaB alpha in T cells after PMA activation. Now we attempt to determine the dynamics of NF-kappaB/I kappaB alpha nucleocytosolic transport in absence of activation as a mechanism to explain both the maintenance of latency and the existence of low level ongoing HIV replication in resting CD4+ T lymphocytes.

Results and conclusion: We show that the inhibition of the nuclear export by leptomycin B in resting CD4+ T cells resulted in nuclear accumulation of both I kappaB alpha and p65/RelA, as well as formation of NF-kappaB/I kappaB alpha complexes. This proves the existence of a rapid shuttling of I kappaB alpha between nucleus and cytosol even in absence of cellular activation. The nuclear accumulation of I kappaB alpha in resting CD4+ T lymphocytes results in inhibition of HIV-LTR dependent transcription as well as restrains HIV replication in CD4+ T lymphocytes. On the other hand, basal NF-kappaB activity detected in resting CD4+ T lymphocytes was related to low level HIV replication in these cells.

Figure 1

Subcellular localization of IκBα and p65/RelA in CD₄⁺ T lymphocytes. Cells were treated or not with 20 nM LMB and then fixed, permeabilized and stained with specific antibodies against IκBα and p65/RelA. A secondary antibody conjugated with Texas Red (Molecular Probes) was used. Images were taken by confocal microscopy.

Figure 2

Subcellular localization of EYFP-IκBα and EYFP-p65 fusion proteins in CD₄⁺ T lymphocytes. (a) Cells were transiently transfected with 1 μg of either EYFP-IκBα or EYFP-p65 expression vectors per million of cells. LMB was added immediately after transfection. After 18–24 hours of incubation, cells were analyzed by confocal microscopy. pEYFP-C1 vector was used as control of unspecific distribution. (b) Resting purified CD₄⁺ T lymphocytes were transiently transfected with the control plasmid pcDNA3.1 by using an Amaxa nucleofector and a classical electroporator (Equibio). As occurs in untransfected resting T cells (lane 1), NF-κB was not induced in resting CD₄⁺ T lymphocytes after electroporation (lanes 3 and 4). As a positive control, NF-κB (p50/p65) binding was induced in these cells by PMA activation (lane 2).

Figure 3

Kinetic analysis of nuclear IκBα translocation. One CD₄⁺ T lymphocyte transfected with EYFP-IκBα vector was photographed before and after treatment with LMB up to 30 minutes. Photographs were taken in vivo by confocal microscopy every minute after adding LMB.

Figure 4

Analysis of nuclear NF-κB/IκBα complexes in CD₄⁺ T cells and IκBα pool dependence on de novo protein synthesis. (a) Analysis of subcellular distribution of p65/RelA and IκBα in CD₄⁺ T cells and presence of NF-κB/IκBα complexes in the nucleus after treatment with LMB or PMA. Ten micrograms of cytosolic and nuclear extracts from CD₄⁺ T cells treated with either PMA or LMB during 4 and 6 hours respectively were analyzed by Western Blot using antibodies against p65/RelA and IκBα. Immunoprecipitation assays were performed using 100 μg of these cytosolic and nuclear extracts, which were incubated with 5 μg of an antibody against p65/RelA conjugated with agarose. IκBα and p65/RelA complexes were characterized by immunoblotting. (b) Contamination with cytosolic proteins during nuclear protein extraction or accumulation of cytosolic proteins in the nucleus after treatment with LMB was assessed by Western Blot using an antibody against both p105 and p50/NF-κB1 proteins. (c) Analysis of NF-κB DNA-binding activity in CD₄ ⁺T cells treated with either PMA or LMB. Three micrograms of nuclear extract were incubated with an oligonucleotide containing the double consensus motif κB present in the HIV LTR labeled with [α-³²P]-dCTP. Protein extracts were obtained from CD₄⁺ T cells after treatment with either LMB or PMA for 6 and 4 hours respectively. (d) Analysis of IκBα pool dependence on de novo protein synthesis. Ten micrograms of nuclear extracts from CD₄⁺ T cells incubated with 20 nM LMB for 4 hours and 10 μg/ml CHX and/or 25 ng/ml PMA for 4 hours,30 min and 2 hours, respectively, were analyzed by Western Blot.

Figure 5

Influence of IκBα over-expression on HIV-LTR transactivation. Resting CD₄⁺ T cells were transfected with LTR-LUC vector together with (a) pcDNA3.1 and/or CMV-Tat expression vectors or (b) pcDNA3.1 and/or CMV-Tat and/or CMV-IκBα expression vectors, as indicated. Cells were treated with LMB immediately after transfection and/or with PMA two hours after transfection, as indicated. Luciferase activity was measured 18 hours after transfection. Numbers on the top of the bars represent fold transcriptional activity relative to unstimulated T cells transfected with pcDNA3.1.

Figure 6

HIV replication in resting or activated CD₄⁺ T cells transfected with an infectious molecular HIV-1 clone. Highly purified CD₄⁺ CD₂₅ ^- CD₆₉ ^- DR^- T cells were transfected with the NL4.3 infectious molecular HIV-1 clone together with CMV-IκBα or pcDNA3.1 as negative control, and then activated with anti-CD₃ and IL-2, PHA and IL-2, or maintained in the absence of activation. Viral replication was determined by quantification of HIV p24-gag antigen in culture supernatants (a) after 5 days of transfection or (b) after 7 days of transfection. Numbers on the top of the bars represent fold HIV-replication relative to unstimulated T cells transfected with pcDNA3.1. Differences in p24-gag production were significant for resting and anti-CD₃-activated T cells (p < 0.05) and a trend towards statistical significance was found in PHA-activated T cells (p = 0.081).