Humanized Mice

Participant-Derived Xenograft (PDX) Mouse Models

Our interest in developing and applying mouse models was initially motivated by the need for a system in which to evaluate and optimize cure-focused therapeutic strategies prior to pilot clinical trials. Existing models had limitations in their abilities to study methods to enhance HIV-specific T-cell responses. The resulting models, however, also appear to have potential for studying T-cell-based vaccine strategies – a potentially new and exciting direction for our lab.

Mouse cells do not support replication of HIV, but models of infection can be generated by engrafting human immune cells into mice that have been engineered to have no immune systems of their own (NSG mice) – including the CD4+ T-cells needed to support viral replication. A challenge, however, is that these human immune systems can recognize mouse cells as foreign and start to attack them, a process termed graft versus host disease (GvHD).

From McCann et al, J Exp Med, 2021. HST = HIV-specific T-cells, NG = IL-15 nanogels. Results show that HSTs transiently suppress HIV viral load and delay CD4 depletion, and that this is enhanced by NGs.

We found that engrafting NSG mice with only the memory subset of CD4+ T-cells (excluding naïve cells), we achieve a system that supports robust HIV replication but in which GvHD is abrogated (1). By using cells from study participants of people living with HIV, we can then isolate immune effectors (e.g. CTL or NK cells) from these same people, apply methods to enhance these cells in vitro, and then assess how they perform in vivo. We have used this approach to show that HIV-specific T-cell lines to be infused into study participants in one of our clinical trials indeed show antiviral activity in vivo, and further that this can be enhanced by targeting delivery of IL-15 using nanoparticles (1). We are expanding these efforts to test an array of therapeutic approaches designed to enhance the in vivo efficacy of natural or engineered (CAR) HIV-specific T-cells.

Moving beyond this first generation ‘CD4-PDX’ model, we have recently validated a method that allows engraftment of NSG mice with whole PBMC, while still avoiding GvHD. Initial results suggest that new HIV-specific T-cell responses are primed in infected mice, raising the possibility that we may also be elicit responses with vaccines. We are in the process of exploring this possibility using mRNA vaccine platforms, similar to those used for SARS CoV-2. The flexibility and rapidity of mRNA-based immunogen generation would pair well with these same properties of the PBMC PDX mouse model to enable testing and iterative improvements of strategies to elicit effective HIV-specific T-cell responses.

  1. McCann, C. D. et al. A participant-derived xenograft model of HIV enables long-term evaluation of autologous immunotherapies. J Exp Med 218, doi:10.1084/jem.20201908 (2021).