HIV-1 and autophagy: Gag, Nef, and Tat

Human Immunodeficiency Virus (HIV) is a lentivirus and the causative agent of Acquired Immunodeficiency Syndrome (AIDS), infecting cells of the human immune system including CD4+ T lymphocytes, dendritic cells and macrophages as well as cells of the nervous system, in particular microglia. The infection of these cells can lead apoptosis or to latently infected cells, the latter constituting a viral reservoir.

HIV genome (linearised)

HIV-1 Gag, Nef and autophagy: inhibition of autophagy by Beclin-1 binding?

Patients infected with a virus that harbours a deletion of the nef gene either do not progress to full blown AIDS or -if so- with a considerable delay compared to patients infected with “standard” HIV as well as lower virus loads, suggesting that the expression of Nef is necessary for maintaining high viral loads and AIDS pathogenicity. In addition, deletions of nef within the genome of the simian equivalent of HIV, SIV, have similar effects in Rhesus monkeys.  In general, Nef is believed to contribute to the pathogenesis of AIDS by interfering with the antiviral immune response on multiple levels, including the downregulation of cell surface molecules (MHC-Class I, Tetraspanins, and CD4) thus preventing the recognition of infected cells by CD8⁺ lymphocytes and promoting the dissemination of infectious virions respectively, thus partially overlapping with the functions of other accessory proteins such as Vpu, Vif, and Vpr/Vpx (in HIV-2).   In addition, Nef from non-pathogenic SIV has been shown to downregulate the expression of TCR-CD3 thus preventing the activation of bystander T lymphocytes - similar to HIV-1 Δ Nef but not wt HIV-1. In all instances however are indirect effects of Nef expression rather than direct functions.

HIV-1 Nef: domains (top) and membrane topology (bottom)

As part of the innate antiviral response against HIV-1 in macrophages, HIV-1 is degraded via autophagy in a Nef dependent manner as cells infected with a HIV-1 Δ Nef exhibit increased viral titers, suggesting that Nef inhibits either the formation of mature autophagosomes or alternatively the degradation of mature autophagosomes by inhibiting the fusion with the lysosome, akin to the coronaviral PLP2 or nsp-6. Experimental evidence suggests that Nef does bind Beclin-1 either directly or indirectly thus blocking the maturation of the autophagosome but interestingly not the induction of autophagy. Indeed, Nef co-localises with both Atg7 and Atg12 in addition to Beclin-1 and 2xFYVE-GFP ( a marker for membranes containing phosphatidylinositol 3-phosphate (PI3P) ) probably by binding Beclin-1, indicating that the formation of phagosomes is not impaired. So how is the fusion with lysosomes inhibited? As mentioned in a previous post, Beclin-1 is also required for the fusion of the lysosome with the autophagosome via a complex anchored at the lysosomal membrane, a process which can be inhibited by Rubicon binding to Beclin-1. In the case of HIV-1 Nef, binding of Nef to the complex therefore might either prevent UVRAG from binding Beclin-1 or the Beclin-1/UVRAG complex from being formed at the lysosome (personal opinion). Given that Nef not interferes with the formation of autophagic vesicles, the binding of Nef per se seems not sufficient to impair the function of Beclin-1 during the formation of the phagosome suggesting that Nef either does not bind to ER localised Beclin-1 or is prevented from doing so by a different HIV-1 derived protein. Unfortunately, the author of this post is not aware of any study expressing Nef by itself in the context of investigating autophagy, but past results obtained by the author of this post indicate that in SupT1 cells Nef does exhibit a punctate distribution pattern in the cytoplasm.

Localisation of Nef in SupT1 cells stably transfected with a inducible plasmid treated with 4-HT (middle and bottom) 

or untreated (top) and co-transfected with HIV-1 Vpr (top and bottom) Arrows indicate punctae in Nef expressing cells 

One possibility is that in infected cells another viral protein is inducing autophagy which is subsequently inhibited by Nef. Indeed, the same study, which showed that Nef inhibits autophagy, also demonstrated that the viral Gag/p17 protein co-localises and co-purifies with LC3-II in extracts of monocyte derived macrophages infected with a pseudotyped VSV -HIV construct. Interestingly, the accumulation of autophagic markers is dependent on Nef -but is the formation of the phagophore? If so, then the question remains how does Nef impair the degradation of the autophagosome? The answer might be, that Nef interacts with a (newly identified) negative regulator of autophagy, Golgi-associated plant pathogenesis-related protein 1 (GAPR-1). Transfection of HeLa cells with a siRNA targeting GAPR-1 increases the number of Beclin-1/WIPI2 positive punctae, and thus early autophagosomes, whereas in control cells Beclin-1 is mainly localised to the Golgi. GAPR-1 therefore tethers Beclin-1 to the Golgi and renders Beclin-1 inactive. Binding of Golgi resident Beclin-1 -but not ER resident Beclin-1- by Nef might lead to the formation of Beclin-1/Nef/GAPR-1 complex, thus preventing Golgi resident Beclin-1 from being forming a complex with UVRAG. I should note that this represents a model which needs to be tested, but may be supported by findings that the expression of a D174A/D175A mutant of Nef which does not bind Beclin-1 still inhibits autophagy despite that a synthetic peptide which represents amino acids 257–337 of Beclin-1 -representing  the region within the beclin 1 evolutionarily conserved domain (ECD) that binds Nef, counteracts Nef mediated inhibition of autophagy and decreases viral replication. The presence of both Nef and Gag at the site of the formation of the phagophore can be explained by the interaction of Nef with Gag. 

Domains of Beclin-1 (top) and potential interaction with Nef (bottom) 

Consequently, the formation of the phagophore may or may not dependent on Gag. It is interesting though that the expression of Nef increases the levels of Gag, which makes sense if Nef prevents the degradation of Gag by autophagy and allows the targeting of Gag-Pol to the site of viral assembly (the plasma membrane) instead.  Alternatively, the Nef-Gag complex located at the ER might be directed to the site of viral assembly and the inhibition of the degradation of this complex by autophagy be a mechanism to increase traffic of the complex to the plasma membrane. Indeed, the transport of ATP from the amphisome to the plasma membrane in vesicles in a VAMP7 dependent manner has been reported.
Furthermore, in cells infected with a pseudotyped VSV- Δ Nef-Gag virus, Gag accumulates at the ER instead the plasma membrane, but the effect of chemicals inhibiting autophagy has not been determined (yet). 

Based on results obtained by the author of this post in the past, in the presence of HIV Vpr, Nef positive punctae are largely absent if Vpr is expressed. HIV Vpr has been show to induce mitochondrial damage and thus potentially not only induce apoptosis but also mitophagy. If this is the case, then Nef might facilitate and not block the engulfment of mitochondria in LC3 positive vesicles whilst preventing degradation of damaged mitochondria in this specific case. Since Nef itself induce the oxidative response and depolarisation of mitochondria under hypoxia indicated by results published in 2011, the accumulation of depolarised mitochondria in Nef expressing cells might lead to apoptosis and incomplete mitophagy itself. 
In SupT1 cells stably expressing an inducible Nef plasmid, the expression of both Nef and Vpr damages mitochondria as indicated by loss of mitochondrial potential and induce mitochondria dependent apoptosis. Indeed both the expression of Vpr and Nef (as well as the combined expression) induce apoptosis, although it is not clear if this is due to inhibition of the clearance of damaged mitochondria by mitophagy via inhibiting mitophagy or not.

Nef and Vpr in SupT1 cells stably expressing inducible Nef plasmid: Mitochondrial depolarisation
(CCCP: positive control)

Expression of Nef in the context of a proviral construct does not prevent apoptosis

                                                                                             HIV-1 Tat

HIV-1 not only infects cells of the immune system such as macrophages or T lymphocytes but also enters the brain early during the infection, infecting- an albeit limited number of- microglia and astrocytes. Long-term survival of these cells has been implicated in serving as a viral reservoir and be one of the obstacles in treating HIV-1 positive patients.  As outlined in previous posts, one of the mechanisms preventing apoptosis of cells infected with viruses in general is the expression of antiapoptotic proteins while autophagy represent a second mechanism to counteract apoptosis. As outlined above, the Nef protein of HIV inhibits autophagy and the inhibition of autophagy by Influenza Virus M2 protein has been implicated in inducing apoptosis.  Regarding HIV-1 Nef, preliminary data obtained by the author of this blog indicate that the expression of the proviral pNL43-Δ Nef nor pNL43- Nef* does prevent apoptosis in transfected SupT1 cells (whether these constructs inhibit autophagy or not was not tested) suggesting that under these conditions Nef does not contribute to the induction of apoptosis. It should be noted however, that the expression of Nef by itself induces caspase dependent apoptosis in various cell lines. 

Among the proteins that regulate the interplay between autophagy and apoptosis, it has been reported that the expression of BAG3 enhances autophagy whilst inhibiting apoptosis as well as degradation of proteins via the proteasomal pathway. Indeed, BAG3 expression is upregulated in both glial and T lymphocytes infected with HIV-1. Following infection of the host cell with HIV the viral genome is imported into the nucleus where it is integrated into the host genome. In the absence of any stimulation, the integrated genome is not expressed (latent phase). In order to stimulate gene expression, the viral enhancer and promoter elements contained in the viral long term element (LTR) located at the end of integrated provirus need to be activated by HIV Tat (trans-activator of transcription), one of the earliest genes being expressed. Tat interacts with cellular histone acetyltransferases (HATs) in particular p300/CBP, that are recruited to the viral LTR, thereby acetylating nucleosomes within the promoter region.

In addition to activating the expression of viral genes, Tat also activates the expression of cellular genes. Following the transfection of a plasmid allowing the expression of HIV-1 Tat into normal human astrocytes as well as U87MG cells expressing GFP-LC3, the formation of mature autophagosomes can be observed as evidenced by an increase in GFP-LC3 positive punctae as well as LC3-II. Additionally, these punctae co-localise with lysosomal markers (although no control experiment using Chloroquine nor E-64d has been conducted). Since the levels of BAG3 are increased following the transfection with Tat and cells transfected with siRNA targeting BAG3 do not show an increase in mature autophagosomes it has been proposed that Tat either increases BAG3 levels independent of inducting BAG3 expression (as evidenced by qRT-PCR). Furthermore, the increase in BAG3 protects cells from apoptosis thus linking the induction of autophagy by Tat to the inhibition of apoptosis. Although the mechanism of how Tat stabilises BAG3 is not known, it might be possible that Tat expression facilities the accumulation of misfolded proteins in the ER and that BAG3 is stabilised as part of the ER stress response. Indeed, Tat has been reported to induce the ER stress response under hypoxia.

Nef, Gag, and Tat induce autophagy whilst Nef also inhibits
the fusion of the autophagosome with the lysosome

In summary, the inhibition of Nef of the autophagy pathway might explain the accumulation of endosome like structures and vacuoles in infected cells. Short peptides composed of the Beclin-1 interaction of Nef but lacking the ability of Nef to localise to the Golgi might stimulate autophagy and decrease viral titers in patients infected with HIV-1 Δ Nef. Vice versa, peptides derived from the domain of Beclin-1 interacting with Nef have been shown to inhibit the replication of West Nile Virus (WNV) and Chikungunya Virus (CHIKV) in mice as well as HIV-1 in vitro. In these cases the peptide tested, Tat-Beclin1, binds viral proteins that otherwise bind cellular Beclin-1 and thus inhibit viral proteins that inhibit autophagy. In other words, it increases autophagy. Personally I would like to test this peptide in cells expressing the coronaviral nsp-6 and/or PLP2.

In the case of Tat, Tat might induce the accumulation of (misfolded) proteins in the ER under hypoxia conditions and maybe under normal atmospheric conditions and thus induce BAG3 mediated autophagy. Unfortunately, a paper describing the inhibition of ER stress apoptosis by HIV-1 Nef had to be retracted.

Further reading

Kyei GB, Dinkins C, Davis AS, Roberts E, Singh SB, Dong C, Wu L, Kominami E, Ueno T, Yamamoto A, Federico M, Panganiban A, Vergne I, & Deretic V (2009). Autophagy pathway intersects with HIV-1 biosynthesis and regulates viral yields in macrophages. The Journal of cell biology, 186 (2), 255-68 PMID: 19635843 

Haller C, Müller B, Fritz JV, Lamas-Murua M, Stolp B, Pujol F, Keppler OT, & Fackler OT (2019). HIV-1 Nef and Vpu are Functionally Redundant Broad-Spectrum Modulators of Cell Surface Receptors Including Tetraspanins. Journal of virology PMID: 25275127

Shoji-Kawata S, Sumpter R, Leveno M, Campbell GR, Zou Z, Kinch L, Wilkins AD, Sun Q, Pallauf K, MacDuff D, Huerta C, Virgin HW, Helms JB, Eerland R, Tooze SA, Xavier R, Lenschow DJ, Yamamoto A, King D, Lichtarge O, Grishin NV, Spector SA, Kaloyanova DV, & Levine B (2019). Identification of a candidate therapeutic autophagy-inducing peptide. Nature, 494 (7436), 201-6 PMID: 23364696 

Geist MM, Pan X, Bender S, Bartenschlager R, Nickel W, & Fackler OT (2019). Heterologous Src homology 4 domains support membrane anchoring and biological activity of HIV-1 Nef. The Journal of biological chemistry, 289 (20), 14030-44 PMID: 24706755 

Dinkins, C., Pilli, M., & Kehrl, J. (2019). Roles of autophagy in HIV infection Immunology and Cell Biology DOI: 10.1038/icb.2019.88 

Dinkins, C., Arko-Mensah, J., & Deretic, V. (2010). Autophagy and HIV Seminars in Cell & Developmental Biology, 21 (7), 712-718 DOI: 10.1016/j.semcdb.2010.04.004 Yang YP, Hu LF, Zheng HF, Mao CJ, Hu WD, Xiong KP, Wang F, & Liu CF (2019). Application and interpretation of current autophagy inhibitors and activators. Acta pharmacologica Sinica, 34 (5), 625-35 PMID: 23524572 

Kang R, Zeh HJ, Lotze MT, & Tang D (2011). The Beclin 1 network regulates autophagy and apoptosis. Cell death and differentiation, 18 (4), 571-80 PMID: 21311563 

Zhong Y, Wang QJ, Li X, Yan Y, Backer JM, Chait BT, Heintz N, & Yue Z (2009). Distinct regulation of autophagic activity by Atg14L and Rubicon associated with Beclin 1-phosphatidylinositol-3-kinase complex. Nature cell biology, 11 (4), 468-76 PMID: 19270693 

Minoia M, Boncoraglio A, Vinet J, Morelli FF, Brunsting JF, Poletti A, Krom S, Reits E, Kampinga HH, & Carra S (2019). BAG3 induces the sequestration of proteasomal clients into cytoplasmic puncta: implications for a proteasome-to-autophagy switch. Autophagy, 10 (9), 1603-21 PMID: 25046115

Gamerdinger M, Kaya AM, Wolfrum U, Clement AM, & Behl C (2011). BAG3 mediates chaperone-based aggresome-targeting and selective autophagy of misfolded proteins. EMBO reports, 12 (2), 149-56 PMID: 21252941

Carra S, Seguin SJ, & Landry J (2008). HspB8 and Bag3: a new chaperone complex targeting misfolded proteins to macroautophagy. Autophagy, 4 (2), 237-9 PMID: 18094623 

Bruno, A., De Simone, F., Iorio, V., De Marco, M., Khalili, K., Sariyer, I., Capunzo, M., Nori, S., & Rosati, A. (2019). HIV-1 Tat protein induces glial cell autophagy through enhancement of BAG3 protein levels Cell Cycle DOI: 10.4161/15384101.2019.952959 

 Romani B, Engelbrecht S, & Glashoff RH (2010). Functions of Tat: the versatile protein of human immunodeficiency virus type 1. The Journal of general virology, 91 (Pt 1), 1-12 PMID: 19812265

Tiede LM, Cook EA, Morsey B, & Fox HS (2011). Oxygen matters: tissue culture oxygen levels affect mitochondrial function and structure as well as responses to HIV viroproteins. Cell death & disease, 2 PMID: 22190005