Plasmid expressing NC and plasmid expressing Tat were cotransformed to yeast strain AH109 and the interaction of the two proteins were observed (A). To confirm specific interaction between NC and Tat, HIV-1 Vpu protein was used as a negative control. Yeast cell cotransformed with NC and Vpu didn’t grow in SD media D.O. tryptophan, leucine, and histidine. Next, in vitro
pulldown assay was performed. Purified recombinant GST-fused NC or GST alone was mixed with cell lysates of BHK-21 expressing Tat. The binding of Tat to GST-fused NC, not to GST, was observed (B). GST pulldown of Tat was visualized by western blot. Coimmunoprecipitation was also used for observing the interaction between the two proteins. From the cell lysate expressing NC and Tat, the two proteins were immunoprecipitated together (C). But, NC and Vpu proteins were not observed to interact with each other. Subcellular colocalization of the two proteins was also confirmed under laser scanning confocal microscope. It has been reported that virion NC localizes to the nucleus at the early infection stage [13
]. In our experiments, NC was predominantly observed in the cytoplasm in the absence of Tat. It seemed that only part of overexpressed NC translocated to nucleus in the absence of Tat. However, in the presence of Tat, NC was shown to translocate to the nucleus and colocalize with Tat (D). Both NC and Tat are translated in the cytoplasm and the binding between the two proteins seems to recruit NC to nucleus with their specific interaction. Tat has a nuclear localization signal (NLS) in the basic domain and mainly localizes in the nucleus, inducing transcription activation. The binding between the two proteins resulted in translocation of NC to the nucleus.
Figure 1 Interaction between HIV nucleocapsid (NC) protein and Tat (A) Plasmid expressing NC and plasmid expressing Tat were transformed to AH109 and the yeast was grown on media lacking histidine. Murine p53 and SV40 large T antigens were used as a positive control. (more ...)
The outcome of interaction between the two proteins was explored. When the two proteins were co-expressed in HEK293 cells, the amount of Tat was observed to decrease through a western blot analysis (A). Co-expression of GST did not alter the amount of Tat. To determine whether this change resulted from transcriptional control, we examined the amount of mRNA expressing Tat by reverse transcription PCR analysis. The amount of mRNA expressing Tat did not change in the presence or in the absence of NC (B). Accordingly, Tat was hypothesized to degrade posttranslationally in the presence of NC. The primary degradation mechanism of proteins in eukaryotic cells involves the ubiquitin-proteasome pathway. Ubiquitin binds to unnecessary or misfolded proteins and this modification is recognized by the 26s proteasome leading to degradation of the target proteins [15
]. To confirm whether the decrease of Tat in the presence of NC was via proteasomal degradation, MG132, a proteasome inhibitor, was used. The amount of Tat was seen to decrease less in the presence of NC when the cells were treated with MG132 (lanes 2 and 5, C). As a comparison, the amount of Tat did not decrease in the presence of HCV core (lanes 3 and 6). The proteasome constitutively degrades nearly all proteins, so treatment with MG-132 is expected to increase the amount of Tat in the cell, appearing as if there was an increase in Tat expression. However, this is actually a decrease in Tat degradation. There is a decrease in Tat level with NC, but not alone or with HCV core protein, in untreated cells. This difference is alleviated when the proteasome is shut down, thus the proteasome is most likely causing the induced degradation of Tat by NC. To see whether ubiquitination occurred before the proteasomal degradation of Tat, ubiquitination assay was performed. In the presence of NC, the ubiquitination of Tat did not increase (D). Hdm2, a proto-oncoprotein, is already known to induce the ubiquitination of Tat [16
] and this was therefore used as the positive control in this experiment.
Figure 2 NC induced proteasomal degradation of Tat in an ubiquitin-independent pathway (A) V5-tagged Tat and Flag-tagged NC or GST were expressed in HEK293 cells and detected by a western blot analysis. Tat decreased in the presence of NC (middle lane), while (more ...)
Since Tat binds to the TAR RNA element and activates transcription elongation [2
], we checked whether the degradation of Tat by NC affected this process. A reporter plasmid containing the luciferase gene under the control of HIV-1 LTR promoter was used. As negative controls, HIV p24, or GST was used instead of NC. In the presence of Tat, reporter activity increased notably (A). In the presence of NC, reporter activity decreased 22% when compared to the tests in the presence of HIV p24 or GST. The decreased reporter activity by NC was effectively rescued by treatment with MG132 (B). Thus, we have confirmed that the NC-mediated degradation of Tat results in the decrease of transcriptional activation.
Figure 3 NC inhibited transactivation of Tat. (A) A luciferase reporter plasmid under the control of the HIV-1 LTR promoter was expressed with Tat in the presence of HIV NC, GST, or HIV p24; (B) Luciferase reporter activity was measured after treatment of cells (more ...)
There has been a report that HIV Tat is degraded by cellular p14ARF
in an ubiquitin-independent pathway [17
]. In the latter study the degradation resulted in a decrease of transactivation. Many other investigations have reported that proteasomal degradation occurs independently of ubiquitination [18
]. For example, tumor suppressor p53 is degraded by both the ubiquitin-dependent and ubiquitin-independent pathways [19
]. There is also a report of viral transactivator degradation by its own viral protein. The hepatitis B virus transactivator X protein is degraded by both ubiquitin-dependent and ubiquitin-independent pathways [20
]. Proteasomal degradation of HBV X protein induced by HBV core was suggested as a new mechanism of controlling virus life cycle [21
Tat upregulates transcription from proviral genome. Transcribed early genes include Tat, Rev, and Nef. Late genes include Gag, Pol, Env, Vpr, Vpu, and Vif which are expressed in a Rev-dependent manner [22
]. Although cleaved NC is generally found after proteolytic processing of Gag in released viral particles, premature processing of HIV Gag in cytoplasm is also reported [24
]. A similar premature cytoplasmic processing of Gag was also shown in an avian retrovirus [25
]. When NC is accumulated after premature processing in cytoplasm, it would have a chance to interact with Tat which was abundantly expressed at the early stage of infection. This interaction can lead to degradation of Tat via proteasomal pathway, leading to decrease in transcription at late stage of viral infection cycle. The 22% decrease of Tat-mediated transcription in the presence of NC in our experiment cannot be responsible for the entire silencing of transcription at the late stage. Since both NC and Tat were overexpressed in our experiment, Tat degradation effect would not be fully reflected. And there could be other mechanisms yet to be elucidated responsible for this transcriptional control. Nonetheless, the virus needs to minimize unnecessary transcription from the viral genome.
Based on our findings, we suggest an additional role of NC related to transcriptional control at the late stage of HIV replication.