Abstract
Following an encounter with a pathogen or vaccination, a population of memory CD8+ T cells specific for the encountered antigen remain to prevent or reduce a secondary infection. An increase in the number or function of these memory CD8+ T cells could improve the efficacy of vaccines. An elongated mitochondrial morphology creates a spare respiratory capacity (SRC) which allows memory CD8+ T cells to rapidly respond to secondary antigen encounters. Therefore, we hypothesized that elongation of mitochondria by deletion of the mitochondrial fission protein, DRP1, could generate a large pool of memory CD8+ T cells to form a rapid secondary response. However, DRP1-deficient CD8+ T cells are compromised in number and function in a CD8 intrinsic manner during a primary response. Nonetheless, a disproportionately large pool of memory CD8+ T cells does form. We did not identify markers of differentiation towards memory precursors nor decreased cellular death in DRP1-deficient CD8+ T cells. In a dLckCre x Drp1flfl mouse model, the CD8+ T cell secondary response is not augmented as hypothesized, but also is not reduced or compromised as seen in the primary response. However, DRP1-deficient cells are not able to compete with DRP1-replete secondary T cells when placed in the same environment. Together, these findings indicate that altering mitochondrial dynamics could augment the memory CD8+ T cell pool, but the timing and conditions of the alteration would be critical.
Additionally, tumor infiltrating lymphocytes (TIL) demonstrate poor glycolytic and OXPHOS metabolism and may benefit from augmenting T cell metabolism. Similar to memory CD8+ T cells, elongation of mitochondrial morphology may support the metabolic need for additional OXPHOS capacity. Using super-resolution confocal microscopy, we visualized B16cOVA mouse melanoma TIL and observed extremely small, rounded mitochondria as the dominant mitochondrial morphology, in contrast to ex vivo effector CD8+ T cells, which have a combination of mostly small mitochondria and some more elongated or interconnected mitochondria. We hypothesized that mitochondrial elongation by loss of DRP1 could increase the capacity for OXPHOS in TIL. The dLckCre x Drp1flfl CD8+ T cells contained an elongated mitochondrial network under conditions that produced fragmented mitochondria in wild type CD8+ T cells. Therefore, we assessed the mitochondrial and metabolic function of DRP1-deficient CD8+ T cells. While the naïve CD8+ T cells from dLckCre x Drp1flfl mice have elevated glycolytic metabolism, OXPHOS metabolism, and significant SRC, this increased metabolism did not hold for primary effector, memory, or secondary effector CD8+ T cells. Further, the mitochondrial membrane potential of DRP1-deficient CD8+ T cells is decreased compared to DRP1-competent CD8+ T cells, which would indicate poor mitochondrial function in long-lived cells. B16cOVA mouse melanoma tumors grown in dLckCre x Drp1flfl mice do not have a decrease in tumor outgrowth or increase in TIL function. Therefore, pharmacological methods of mitochondrial alteration were assessed. The combination of a mitochondrial fusion enforcer, M1, and a mitochondrial fission inhibitor, Mdivi, results in increased mitochondrial spare respiratory capacity and mitochondrial mass in effector CD8+ T cells, however, the kinetics of division are delayed similar to DRP1-deficient effector CD8+ T cells. However, M1 and Mdivi combination treatment reduced the mitochondrial membrane potential of TIL treated in vitro. Ultimately, more studies on methods and timing of mitochondrial elongation are needed to determine if this is a viable method of enhancing TIL function or tumor control.
TIL function could also be improved by enhancing glycolysis. Previous studies in the Bullock lab had determined that murine and human melanoma TIL have poor ENOLASE activity compared to effector CD8+ T cells. In collaboration with the lab of Ken Hsu, we developed a fluorescent small molecule reporter of ENOLASE activity in order to assess characteristics of single cells. This ENOLASE reporter also identified reduced ENOLASE activity in TIL, but further clarified that while the majority of TIL have low ENOLASE activity, a small fraction maintains higher ENOLASE activity. We will further refer to the ENOLASE reporter as “Repo”. A greater fraction of TIL with higher ENOLASE activity maintains cytokine function and degranulation capabilities compared to TIL with lower ENOLASE activity. Compared to effector CD8+ T cells, TIL have very few post-translational modifications of ENOLASE. Effector CD8+ T cells have many sites of acetylated lysine of ENOLASE, which are potentially deacetylated in TIL by SIRT2. Inhibition of SIRT2 results in increased ENOLASE activity and acetylation in effector CD8+ T cells but more studies are necessary to determine if this holds true for TIL.
