DARPin-targeted Chimeric Antigen Receptor T cells:CD4 as a cellular target shows potential to evade HIV latency reservoir

Abstract

Specific targeting of HIV infected cells has been a challenge since the discovery of the virus. To date, antiretroviral therapy is capable of inhibition of viral spread and controlling the viral load at levels below detection limit. However, the latent viral reservoir persists even during intensive treatment, and produces new infectious viral particles as soon as medication is discontinued. But antiretroviral therapy is cost-intensive, causes major side effects and involves close monitoring of the patient. Relapse due to viral mutation may also require multiple changes in the medication strategy. It is therefore essential for a sustainable cure, to also fully eliminate the latent viral reservoir. Chimeric Antigen Receptor (CAR)-T cell therapy has been very suc-cessful in the treatment of hematologic malignancies. Autologous T cells of patients are genetically modified to express a chimeric receptor, which enables them to specifically bind their tar-get antigen and deplete the target cell without the need of MHC-presentation. Transferring this CAR-T cell technology from targeting of malignant B cells to HIV-infected T cells could potentially be the way to a sterilising cure. However, the development of anti-HIV CAR-T cells has been dormant in the past years due to lack of efficiency, and because targeting of viral proteins on the surface of infected cells did not deplete the latent viral reservoir, hence did not show any improvement compared to conventional antiretroviral therapy. Fast progress in recent development of next generation CARs led to significant enhancement of efficacy and specificity. In addition to that, targeting of cellular rather than viral proteins could potentially overcome persistence of the latent reservoir. Therefore, CD4 as the exclusive HIV entry receptor into the host cell could be a suitable cellular target for anti-HIV CAR-T cells, because all infected cells express CD4, including the latent reservoir. One aim of this thesis was to investigate whether CAR-T cells can specifically and efficiently deplete their CD4-positive target cell population. Therefore, a second-generation CAR was utilised, bearing CD3zeta and CD28 co-stimulatory domains, as successfully used in clinical trials. A CD4-specific Designed Ankyrin Repeat Protein (DARPin) was used as a binding domain, since this molecule was shown to be small, well expressed on the cell surface, specific and not immunogenic. Generation of an expression plasmid, encoding all CAR domains in a cassette system, provided a platform for flexible and individual CAR design. Transduction of primary human T cells with gamma-retroviral particles led to CAR expression on the cell surface. Co-culture resulted in dose-dependent CAR-mediated T cell activation and cytotoxicity towards target but not towards non-target cells, verifying specificity and potency of anti-CD4-DARPin CAR-T cells. High efficiency was shown during co-culture with non-target cells including de-creasing amounts of target cells. CD4 specific cytotoxicity was also observed against very low target cell levels. Human CD4 was selected as a target because it is the entry receptor for HIV into the host cell, and therefore expressed on all HIV-positive cells, irrespectively of their activation state. Another question of this thesis was therefore, if CD4-specific CAR-T cells have the potential to deplete the CD4-positive T cell compartment and thereby also the latent HIV reservoir. Anti-CD4-DARPin CAR-T cells showed efficient and specific cytotoxicity against an HIV-latency model - towards activated, as well as latently infected cells with low levels of target expression. Furthermore, CD4 T cells were efficiently depleted in an autologous setup, without major impact on the CD4-negative T cell population. Taken together, anti-CD4-DARPin CAR-T cells have shown specificity, efficiency and potency, the main properties, required for success-ful CAR-T cell therapy. With regard to further investigation of this approach in an immunocompetent animal model, anti-CD4-DARPin CARs were not only expressed on human, but also non-human primate T cells. Moderate depletion of autologous CD4 target cells was observed using T cells of rhesus macaques, providing the groundwork to test this approach in a well-established immunocompetent animal model. In summary, the results of this work provide evidence, that anti-CD4-DARPin CAR-T cells are a promising approach to specifically deplete potential HIV host cells, along with elimination of latently infected cells. This was achieved by combination of a potent 2nd generation CAR, driving efficient T cell activation with a specific and well expressed DARPin binding domain, target-ing a cellular protein to capture all potentially HIV-positive cells, including those in a latent stage. These data provide a basis for further investigation of anti-CD4-DARPin CAR-T cells for HIV therapy.