Chimeric antigen receptors against HIV: Bispecific CD4-based designs for enhanced breadth/potency and minimal immunogenicity

Abstract

Abstract Chimeric antigen receptor (CAR) technology has potential to achieve durable (life-long?) control of HIV in the absence of antiretroviral drugs, i.e. a “functional cure”. Thus in targeting the HIV-1 Env glycoprotein expressed on the surface of virus-producing cells, CAR design must strive not only for high potency and HIV specificity, but also to approach the ideal qualities of inescapablilty and non-immunogenicity. Our strategy involves bispecific targeting domains consisting of an extracellular region of human CD4, the primary HIV receptor, attached to a second motif that recognizes a distinct highly conserved region on the otherwise highly variable Env. The second motif is intended both to enhance CAR potency, and to prevent the CD4 from acting as an HIV entry receptor on CAR-expressing CD8+ T cells. We have focused on second motifs predicted to lack the immunogenicity associated with antibody-based CARs. A particularly favorable construct contains CD4 linked to the carbohydrate recognition domain of human mannose-binding lectin (MBL), which recognizes the highly conserved oligomannose patch on gp120 (Ghanem et al., Cytotherapy). In another variant, CD4 is linked to the N-terminal portion of the CCR5 co-receptor. We found that the bispecific CARs broadly suppressed infection by genetically diverse HIV-1 isolates, with significantly greater potency than a monospecifc CD4 CAR. Similar results were obtained using rhesus macaque PBMCs and a rhesus version of the CD4-MBL CAR. These results set the stage for testing the activities of the CD4-based bispecific CARs in relevant animal models including rhesus macaques and humanized mice.