A novel anti-HIV immunotherapy to cure HIV

Combined antiretroviral therapy (ART) has saved millions of lives but does not ‘cure’ HIV infection. The virus persists latently as proviral DNA integrated in long-lived CD4+ T cells in immunologically and pharmacologically privileged viral sanctuaries. This latent ‘viral reservoir’ is the major hurdle in curing HIV-infected individuals [1], as it recrudesces after cessation of ART. It has been estimated that ART alone will not deplete the viral reservoir in infected individuals in their life time [2]; therefore, life-long ART is required to prevent virus reactivation and development of AIDS. As the virus continues to replicate at undetectable levels, and there are toxic side effects of ART, these patients are predisposed to accelerated aging, development of several non-AIDS related cancers and comorbidities such as cardiovascular disease. The report of a functionally cured, HIV-infected person, the so-called Berlin patient, in 2009 was a seismic event in the history of HIV [3,4]. The patient suffered from acute myeloid leukemia and received heterologous stem cell transplants from a donor carrying a homozygous delta-32 CCR5 gene, rendering his T cells resistant to R5-tropic HIV infection. He discontinued ART after the transplant and has remained free from detectable viremia and any symptom of AIDS. Not surprisingly, this report set off massive efforts to develop a cure for the infection. It was also realized that ‘true cure’ (complete eradication of the virus from the body) may not be a realistic goal. Instead, a ‘functional cure’ (prevention of the disease in the absence of ART) would be sufficient. One approach to such a cure is dendritic cell–based immunotherapy [5,6], in which injections of ex vivo developed, HIV antigen-expressing autologous dendritic cells can both activate the latent HIV reservoir [7] and induce antiviral cytotoxic T lymphocytes (CTL) that decrease viral reservoirs [8]. However, this effect is transient, as the virus eventually rebounds several weeks after cessation of ART [9]. Although these dendritic cell–based HIV therapies are well tolerated and safe, a recent mega-analysis indicates an overall success rate of only 38% [10]. Therefore, the approach needs novel improvements. To improve on this effectiveness, Guardo et al.[11] have proposed, in this issue of AIDS, a novel strategy for targeting dendritic cells directly in vivo by combining two previously described approaches – a TRIMIX adjuvant and HTI, an HIV T cell immunogen. TRIMIX is three mRNAs encoding CD40L, CD70 and a constitutively activated Toll-like receptor (TLR)-4 [12]. CD40L serves as a T helper cell surrogate by inducing IL-12 that ‘licenses’ dendritic cells to prime naïve T cells [13]. By priming HIV-specific naïve CD8+ T cells, CD40L-stimulated dendritic cells induce de novo antiviral CTL responses, which are more effective than preexisting virus-specific memory T cells in killing HIV-infected CD4+ T cells in chronically infected persons on ART [14]. Recovery of naïve CD8+ T cells, although reduced in numbers, occurs in chronic HIV-infected patients on ART [15]. Stimulation of dendritic cells with CD40L also induces extensive nanotube formation, wherein these highly reticulated dendritic cells communicate and transfer signals to other dendritic cells and potentially T cells during immunotherapy [16]. TLR4 activates and cause maturation of dendritic cells, as previously shown with melanoma patients [12]. The approach adopted by Guardo et al.[11] is, indeed, just one example of how cancer immunotherapy has largely set the direction of HIV immunotherapies [17]. Finally, CD70 (a member of the TNF-family) binds to costimulatory CD27 on naïve T cells [18]. However, the CD70/CD27 axis also plays a negative role in chronic viral infections by activating programmed cell death protein 1 (PD-1) and other immune checkpoints [19]. Other dendritic cell activating molecules (e.g. TLRs) also induce these checkpoints. It is therefore essential to assess whether mRNA therapy activates PD-1 and other immune checkpoint molecules such as cytotoxic T lymphocyte-associated antigen 4, lymphocyte-activation gene 3, T cell immunoreceptor with Ig and ITIM domains, T cell immunoglobulin mucin 3, signaling lymphocyte activation molecule family receptor 2B4, and glycosylphosphatidylinositol-anchored protein CD160. If so, this approach could be combined with simultaneous use of one or more checkpoint inhibitors. The use of these inhibitors has shown promise in reducing HIV replication in humans and animal models of infection [20]. A desirable characteristic of CTL generated in HIV immunotherapy is expression of CXCR5. This chemokine receptor is essential for signaling CD8+ T cells to traffic to B cell follicles, which are a key sanctuary for latent HIV [21]. Moreover, studies are also needed to determine whether the mRNA antigen preparations induce CCL19 or CCL21, and not CCL22 production, which respectively attract naïve T cells and Tregs toward activated dendritic cells [22,23]. Tregs can suppress effector function of virus-specific T cells [24]. HTI mRNA codes for 16 antigenic fragments in Gag, Pol, Vif and Nef [25]. The fragments were selected on the basis of screening three large cohorts of HIV-infected individuals for the highest, in-vitro CD8+ and CD4+ T cell reactivity (IFN-γ and granzyme B production). These HIV antigens are relatively conserved and are predominantly targeted by individuals with reduced viral loads. Guardo et al.[11] show that monocyte-derived dendritic cells electroporated with TRIMIX/HTI mRNA express activation markers and induce antigen-specific responses in vitro as determined by T-cell proliferation and production of IFN-γ, intranodal injection of the mRNA preparation of mice induces antigen-specific CTL responses against multiple epitope and human lymph node explants exposed to the mixture activate dendritic cells and induce production of several proinflammatory cytokines and chemokines. However, information is needed for IL-12p70 production, which is essential for activating CD4+ helper T-cell responses that are required for priming naïve CD8+ T cells to become broadly reactive CTL [26]. As mentioned above, the induction of de novo antiviral cellular responses from naïve T cells is likely to be more effective in controlling viral replication. TRIMIX/HTI mRNA induces relatively weak proliferative responses in CD4+ T cells [11]. This may be due to a relative lack/insufficiency of CD4+ T cell–targeted epitopes in the HTI mixture. The number of such epitopes has not been revealed. The low responses could also be due to their poor presentation by major histocompatibility complex (MHC) class II molecules. As mRNA is translated in the cytoplasm, this mode of antigen expression favors its presentation via MHC class I molecules. Exogenous antigens, on the other hand, are presented by MHC class II on dendritic cells (a requirement for antigen presentation to CD4+ T cells), although they are also presented via MHC class I molecules through cross-presentation. Overall, dendritic cells poorly present endogenously produced antigens via MHC class II molecules [27]. This approach, therefore, may consider adding sequences to their CD4+ T cell–targeted epitopes that would direct them to lysosomes and MHC class II loading compartments. Finally, therapeutic mRNAs should be of high purity and free from any double-stranded RNA. As the mRNA is delivered to the cytosol, in which it is translated, the presence of double-stranded RNA in the mRNA preparation could induce type 1 interferon via activating TLR-3. The interferon could inhibit translation of the therapeutic mRNA and reduce its therapeutic effect [28]. The novel therapy designed by Guardo et al.[11] is meant to be delivered directly to dendritic cells in vivo by intranodal injections. This route is preferable over injections of antigen/mRNA-pulsed/electroporated dendritic cells. The main reason is that the half-life of the injected dendritic cells is short. However, the caveat is that naked mRNAs, of which TRIMIX/HTI comprises, are prone to degradation by ubiquitously present RNAses. For this reason, it would be important either to make it more resistant to RNAses by incorporating modified nucleotides like pseudouridines and/or deliver it as complexes with protamine, cationic lipids or in nanoparticles [29,30]. The novel TRIMIX/HTI preparation, like other such mRNA, offers several advantages over other immunotherapies. Single-stranded mRNA per se binds TLR 7 and 8 and acts as an adjuvant [28]. Therefore, mRNA therapies theoretically require no adjuvant, although in this particular formulation, TRIMIX provides the adjuvant effect. Unlike DNA, mRNA does not pose a risk of integration into host genome. TRIMIX/HTI comprises well defined molecules that cost less compared to protein and peptide based therapies. Moreover, it comprises epitopes that are relatively conserved, present in diverse circulating viruses, and are targeted by the virus-specific T cells. Escaping these conserved epitopes could cost the virus replication fitness. The authors adopted this strategy to counter viral mutability as well as its diversity. It, therefore, may be used as a global anti-HIV therapeutic. In brief, the novel mRNA and dendritic cell–based formulation designed by Guardo et al.[11] is a major advancement toward developing an immunotherapy that is effective and scalable for HIV-infected patients. Such mRNA-based immunotherapies have shown recent promise in treating cancers [31]. The real test of this approach, however, will be whether it can control viral recrudescence in HIV-infected individuals upon cessation of ART, and if so, for how long. Acknowledgements The work of C.R.R. is supported by grants from the National Institutes of Health. The research in A.A.'s laboratory has been supported by the Canadian Institutes of Health Research. We thank Dr Robbie Mailliard for critical review of the manuscript. Conflicts of interest There are no conflicts of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.