The role of natural killer cells in functional cure strategies for human immunodeficiency virus type 1 infection

Although antiretroviral treatment (ART) effectively suppresses human immunodeficiency virus (HIV) replication, the presence of a viral reservoir impedes eradication of the virus from the body. As a consequence, life‐long adherence to ART is required since rapid viral rebound occurs almost invariably upon treatment interruption. Therefore, researchers have tried, by restoring the anti‐viral function of effector immune cells, to develop a so‐called functional cure against HIV‐1 where viral replication can be controlled without taking daily ART and without the need for complete viral eradication. Several functional cure strategies have been tested, either aimed at enhancing the immune system, e.g. using therapeutic vaccination, or at targeting the reservoir directly, e.g. by the shock and kill strategy.
We previously performed a clinical trial where we administered autologous dendritic cells (DCs) electroporated with mRNA encoding Tat, Rev and Nef as a potential therapeutic vaccine. We showed that antigen‐specific T‐cell responses were observed in most patients, but no effect was observed on clinical parameters such as CD4+ T‐cell counts and time to viral rebound. Since natural killer (NK) cells play a central role in the antiviral immune response, we hypothesized that they are important players in the search for a functional cure against HIV‐1, by reducing viral replication and/or by targeting latently infected cells directly.
In a first part, we studied the effect of DC‐based therapeutic vaccination on the phenotype and functionality of NK cells since the crosstalk between DCs and NK cells is well described. Although the frequency of total NK cells did not change, we observed a significant increase in cytotoxic NK cell numbers following immunisation, highlighting the importance of evaluating NK cell subsets. In addition, significant alterations in NK cell migration and exhaustion markers were observed together with increased NK cell‐mediated killing and (poly)functionality. Interestingly, a modest increase in the frequency of memory‐like NK cells was observed, which was associated with increased NK cellmediated killing upon vaccination. However, these changes were lost upon treatment interruption, most likely due to a rapid viral rebound. Therefore, further research is warranted to optimize the efficacy of DC‐based vaccine‐induced NK‐cell responses.
Besides therapeutic vaccination, another well‐studied strategy is the shock and kill strategy in which latency reversal agents (LRAs) reactivate the silent viral reservoir, making it visible for recognition and elimination by the host’s immune system. Although this strategy was shown to be very potent in vitro and ex vivo, so far, in vivo no significant delays in time to viral rebound have been observed. To enhance the efficiency of this strategy, it can be combined with another immune‐stimulatory treatment(s). Therefore, we aimed to study the effect of LRA treatment on NK cells and the effect of blocking the inhibitory interaction between NKG2A receptor and HLA‐E on NK cell‐mediated killing of reactivated 20 cells. We found that the proteasome inhibitor bortezomib could induce latency reversal in J‐Lat 10.6 cells but not in other variants of the J‐Lat cell model, including J‐Lat 8.4 cells. This difference in sensitivity might be due to differences in viral integration sites. In addition, HLA‐E expression on J‐Lat 8.4 cells was significantly decreased upon treatment with bortezomib, leading to a modest increase in NK cell cytotoxicity. These results suggest that inhibition on NKG2A/HLA‐E interaction might enhance the efficiency of the shock and kill strategy by improving NK cell‐mediated killing of reactivated cells. However, these results need to be confirmed in primary latency models.
Finally, we aimed to identify a viral reservoir screening tool for PLWH to facilitate pre‐screening in the context of HIV cure trials. Currently, most cure trials focus on the inclusion of participants exhibiting a small viral reservoir size. However, the latter represent only a minority of PLWH. Plasma viral load (pVL) is a measure for viral replication and is routinely determined in the clinic. We used the qualitative information for pVL measurements below the limit of quantification as a novel screening tool for viral reservoir size. In contrast to quantification of cell‐associated DNA and RNA, pVL results are readily available for all PLWH in care.
In conclusion, we showed that NK cells play a key role in two very important functional cure strategies, therapeutic vaccination and the shock and kill strategy. Whereas the main focus of functional cure strategies has so far been the improvement of T‐cell responses, our results suggest that future research should take into account NK cell‐mediated responses as well.