T Cells Engineered For Resistance To Mycophenolate Mofetil Demonstrate Enhanced Expansion and Improved Tumour Control In Immunosuppressed Hosts

Adoptive transfer of therapeutic T cells targeting viral infection or tumour following allogeneic solid organ or hematopoietic stem cell transplantation is hindered by the requirement for immunosuppression. Recent reports have suggested the strategy of using genes encoding cyclophilin or FK506 mutants to confer resistance of T cells to calcineurin inhibitors (Brewin et al. Blood 2009;114:4792 & De Angelis et al. Blood 2009;114:4784). Mycophenolate mofetil (MMF) is another widely used immunosuppressant that acts as a non-competitive inhibitor of inosine-5’-monophosphate dehydrogenase 2 (IMPDH2), an inducible enzyme that generates guanine nucleotides for DNA and RNA synthesis in T cells. In this study, we used T cells transduced with mutated IMPDH2 that confers >2000-fold resistance to MMF (IMPDH2^R; T333I, S351Y).

Genes encoding human IMPDH2^R and IMPDH2^CS(containing a catalytic site mutation, C331A) were cloned into SFG retroviral vectors as fusions to eGFP reporter sequences. T cells were transduced with either vector, transferred to recipient mice and evaluated in the presence or absence of MMF.

IMPDH2^R-transduced (td) murine and human T cells demonstrated a selective advantage to MMF in vitro by overcoming drug-induced G₁-S phase cell cycle blockade (p=0.02, 55.0% IMPDH2^R-td vs 25.6% IMPDH2^CS-td cells in S phase) and preventing apoptosis (p=0.01, 22.5% vs. 39.3% Annexin V positive). Overexpression of the wild type IMPDH2 gene conferred intermediate protection against vector alone controls.

To test for selection of IMPDH2^R-td cells in vivo, sub-lethally irradiated (1.5-2Gy) B6.PL (Thy1.1) mice were injected with a 1:1 mix of OT-I TCR transgenic CD8 T cells transduced with IMPDH2^R (CD45.1, Thy1.2) or IMPDH2^CS (CD45.2, Thy1.2). Host mice received cognate peptide (SIINFEKL) in adjuvant or adjuvant alone, and no drug or MMF (200 μg/g/day) by daily ip injection. Compared to mice given adjuvant alone (spleen ratio IMPDH2^R:IMPDH2^CS= 1.1), selection of IMPDH2^R-td T cells occurred with antigenic stimulation even in the absence of drug, indicating a moderate increase in overall proliferative/survival potential of these cells (ratio IMPDH2^R:IMPDH2^CS = 2.5, p=0.04). As predicted, in vivo selection of IMPDH2^R-td T cells was significantly greater in the presence of MMF (ratio=7.8, drug v no drug, p=0.01). Unexpectedly, however, absolute numbers of IMPDH2^R-td T cells in the spleen were actually greater in the presence of MMF rather than its absence indicating that drug resistance per se was not the sole explanation underlying the capacity of gene-modified cells to expand in vivo (27.3 x 10⁵ MMF vs. 19.4 x 10⁵ no MMF, p=0.01).

Similarly, not only did IMPDH2^R-td OT-I cells control EL4.OVA tumor in vivo better than IMPDH2^CS-td OT-I cells in the presence of MMF (log rank p<0.0001), but they also improved tumour control to a greater extent in the presence of drug than in its absence (log rank p<0.0001). Improved tumor control was linked to higher frequencies and absolute numbers of IMPDH2^R-td T cells in the spleen of MMF-treated versus non-treated recipients, and was independent of any direct effects of MMF upon tumor growth. Higher IMPDH2^R-td T cell numbers in the presence rather than the absence of drug occurred in the context of strong MMF-mediated suppression of endogenous T and B cell compartments.

Synergy between clinically relevant doses of MMF and transfer of IMPDH2^R-td T cells is a potential means for enhancing in vivo expansion and performance of therapeutic T cells beyond that observed in the absence of drug. Future studies will evaluate the potential for this strategy used alone or in combination with other approaches designed to overcome sensitivity of therapeutic T cells to immunosuppressive drugs.

No relevant conflicts of interest to declare.