In mice with specific Drp1 ablation during T cell development, the number of mature T cells in the thymus was reportedly decreased, whilst normal organelle function was retained and all mature T cell subsets were correctly represented (16). Therefore, knockout of Drp1 only appears to affect the number of mature T cells and has no impact on their function or differentiation into subtypes. Furthermore, the reduced proliferation of Drp1-knockout mature T cells was also observed after stimulation with antigen (16). Interestingly, this reduction in the clonal expansion rate could be reversed by activating Drp1-S616E overexpression (16). Although the underlying mechanism remains unclear, the impairment of T cell migration function may partly explain this phenomenon. T cells migrate from the cortex to the medulla in the thymus and undergo strict positive and negative selection (17). Knockout of Drp1 inhibits T cell migration, thereby leading to the accumulation of T cells in the thymic cortex (16). This, in turn, affects the screening of T cells and leads to excessive T cell clearance. It is unclear whether Drp1 is involved in regulating positive and negative selection, although knockout of Drp1 does not lead to an increase in autoreactive T cells.

The migration of T cells depends on their unique cytoskeleton, which converts chemical signals into mechanical energy and promotes migration (18, 19). During this process, Drp1 is phosphorylated at serine 616 under the action of the mitogen-activated protein kinase pathway (14), then the mitochondria are fragmented and relocated to the uropod along the microtubules ( Figure 1 ), providing ATP for the movement of myosin (20). Interestingly, the lack of Drp1 leads to the fusion of mitochondria, which is not conducive to microtubule-mediated transport, while overexpression of Drp1 can promote mitochondrial relocation and accelerate T cell migration (16). This may be due to the inability of microtubules to transport excessively large organelles, or the disturbance of mitochondrial dynamics by changes in the morphology of mitochondria (20). Impaired migration not only affects the maturation of T cells in the thymus (reducing the survival rate of thymocytes during positive selection), but also hinders the migration of T cells to the blood and secondary lymphoid organs, which play a role in immune surveillance (16).

As a second messenger, the calcium current is indispensable for T cell activation. Mitochondria of the immune synapse ensure the calcium release-activated calcium channels remain open by absorbing calcium, keeping the calcium current at a low level most suitable for T cell activation (21, 22) ( Figure 1 ). The calcium current regulates the AMP-activated protein kinase pathway (23) and the mechanistic target of rapamycin (mTOR) pathway (24) to control the metabolic reprogramming of T cells to meet the metabolic needs of the activated state (25, 26). The complete activation of Drp1 depends on calcium-dependent calcineurin-regulated serine 637 dephosphorylation (27) and activation of phosphorylated serine 616 by the mitogen-activated protein kinase pathway (16). Activated Drp1 mediates mitochondrial division and migration to immune synapses (28) ( Figure 1 ). The calcium influx of Drp1-deficient T cells increases during activation, which further leads to the over-activation of AMP-activated protein kinase and a decrease in mTOR-cMyc (16). Since cMyc is necessary for transcription when metabolic genes are activated in T cells (29), its reduction may affect their metabolic transcription. In addition, the electron transport chain located in the IMM produces reactive oxygen species (ROS), which are intrinsically linked to the stability of mitochondrial dynamics, because high levels of ROS can lead to cell damage (30, 31).

After activation, T cells differentiate into i) effector T cells (T_E), which participate in immune reactions to foreign antigens, and ii) memory T cells (T_M), which differentiate rapidly into T_E to participate in the immune response on subsequent exposure to the same antigen. The metabolic patterns of different T cell subsets differ greatly (32, 33). Mitochondria, as metabolic centers, affect the differentiation of T cells. During the differentiation of naïve T cells into T_E, the metabolic mode changes from oxidative phosphorylation and fatty acid oxidation to glycolysis (34). This process depends on the precise regulation of the calcium current at immune synapses by Drp1, which promotes the transcription of glycolysis genes by maintaining the activation of mTOR/cMyc (16) ( Figure 1 ). Fragmented mitochondria were predominantly observed in T_E, and this phenotype depended on the phosphorylation of Drp1 at Ser616 to mediate mitochondrial division (35). Knockout of Drp1 in vitro promotes the transition of T cells to a memory-like phenotype due to the inability of mitochondria to divide (16). The same results were obtained when the glycolysis of T cells was inhibited (36). It should be noted that although knockout of Drp1 appears to promote the conversion of T_E to T_M by inhibiting glycolysis (37), another study has shown that promoting glycolysis can also increase the formation of memory T cells (38), and the primary metabolism of T_M remains controversial (4). Interestingly, the mitochondrial fusion protein Opa1 is not only implicated in effector and memory fate decisions (35), but also affects T cell development in the thymus (39, 40). We believe that both mitochondrial morphologies play a role in T cell development, from maturation to apoptosis, after participating in the immune response. It is crucial to clarify the specific changes in mitochondria and the alteration of pathways during these processes. Whether it is an autoimmune disease that suppresses the immune response, or an anti-tumor treatment that needs to enhance the immune response, more research is needed to clarify the relationship between mitochondrial metabolism and T cell differentiation.

To avoid autoimmunity, effector T cells die following the immune response. This process is called activation-induced cell death and is regulated by mitochondria. Drp1-mediated mitochondrial fission generates ROS that regulate the transcription and expression of CD95L (FasL), which is essential for CD95 (Fas)-dependent apoptosis (41) ( Figure 1 ). In addition, mitochondria also play other roles in apoptosis. The activation of protein kinase A downstream of the T cell receptor (TCR) signal can inhibit autophagy induced by activation-induced cell death, leading to the accumulation and division of damaged mitochondria, thereby releasing cytochrome C and driving apoptosis (42).

Some studies have shown that the mitochondrial function of tumor-infiltrating lymphocytes is abnormal (50, 52, 53). Persistent antigen stimulation in the TME leads to the damage of mitochondrial function ( Figure 2 ) and further affects the ability of T cells to proliferate (54).

TCR can increase the level of glycolysis by co-stimulation with CD28 receptor (55). In vitro, glycolysis increased significantly in T cells after long-term stimulation with antigen (54) ( Figure 2 ), which was consistent with transformation of the metabolic mode during the differentiation of effector T cells. However, in this case, the tricarboxylic acid cycle and the synthesis of nucleotide triphosphate were affected, indicating that this continuous stimulation impaired the ability of mitochondrial β-oxidation in T cells (54) ( Figure 2 ). The dysfunction of mitochondria leads to the deficiency of ATP synthesis in T cells. It can be speculated that oxidative phosphorylation is impaired in T cells stimulated by antigen, which further aggravates the dependence of T cells on glycolysis ( Figure 2 ). In addition, continuous antigen stimulation leads to the accumulation of high levels of ROS in T cells, which directly interferes with ATP synthesis (54) ( Figure 2 ). The proliferation of T cells requires the synthesis of DNA and protein, which is inseparable from the production of ATP (56–58). Therefore, although long-term antigen stimulation maintains a high level of glycolysis, the decrease in mitochondrial oxidative phosphorylation affects the ability for T cell proliferation. Long-term stimulation also increases the level of calcium in T cells and further stimulates the increase in mitochondrial ROS (59, 60) and nuclear factor of activated T cells (NFAT) (61, 62), which can directly activate exhaustion-related transcription factors (63) ( Figure 2 ). Excessive calcium ions in T cells can also block the interaction between kinesin-1 and microtubules and prevent mitochondria from moving along microtubules in T cells (64, 65). In addition, hypoxia, as another feature of the TME, cooperates with continuous stimulation to impair the respiratory ability of T cells and leads to the expression of co-inhibitory molecules (66). Unfortunately, no research has focused on Drp1 changes in this chronic process.

Interestingly, after long-term stimulation of TCR, mitochondria in T cells exhibited a disrupted membrane structure, decreased mitochondrial cristae length and a decreased number of cristae, which resulted from the accumulation of damaged mitochondria that could not be cleared in a timely manner (67). Damaged mitochondria prevent T cells from being fully utilized and from performing their normal functions (see section Drp1 and Mitophagy for more detail). The same result was observed in T cells treated with the Drp1 inhibitor Mdvi-1, that is, mitochondria showed a damaged morphology (67). Previously, we discussed the important role of Drp1-mediated mitochondrial division in T cell energy metabolism and the regulation of ROS production, and emphasized the necessity of mitochondrial relocation in the immune synapse for regulation of the calcium current. Therefore, there is good reason to further explore the changes in Drp1 during this process.

Programmed cell death protein-1 (PD-1) is a co-inhibitory receptor expressed on the surface of T cells (68), which mediates immunosuppression (69). Upregulation of PD-1 expression is one of the characteristics of exhausted T cells (49). Anti-PD-1 therapy can effectively improve the activity of tumor-infiltrating T cells (70, 71). The effect of PD-1 on tumor-infiltrating T cells is at least partly achieved by inhibiting Drp1.

PD-1 regulates T cell metabolism. First, PD-1 enhances T cell fatty acid β-oxidation and inhibits the T cell glycolysis pathway and amino acid metabolism by changing the length and number of mitochondrial cristae (72, 73). Second, PD-1 inhibits the expression of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α, the most important gene that drives mitochondrial biogenesis), polarizing mitochondria and resulting in the production of high levels of ROS, leading to a metabolic imbalance (74, 75) ( Figure 2 ). Metabolic stress not only affects the differentiation of T cells, but also weakens the anti-tumor ability of T cells.

Previous studies have shown that PD-1 can inhibit the proliferation and migration of T cells (76, 77). Simula and colleagues used a mouse tumor model to explore whether PD-1 regulates these processes by affecting Drp1 (78). Their study demonstrated that the PD-1 signal inhibits the division of mitochondria in T cells by inhibiting the phosphorylation of Ser616 sites, and this inhibition is achieved through the ERK and mTOR pathways (76, 78, 79) ( Figure 2 ). Compared with thymocyte Drp1-knockout mice, PD-1 inhibitors significantly enhanced the anti-tumor effect of wild-type mice, and this difference was related to the phosphorylation level of Drp1 in T cells (78). Interestingly, consistent with the effect of Drp1 on T cell migration described earlier, Simula and colleagues found that the density of tumor-infiltrating T cells in wild-type mice was higher, which was related to Drp1-mediated polarization and rearrangement of the mitochondria located in the uropod of T cells (78).

These results suggest that in the case of persistent metabolic disorders, PD-1 gradually induces T cells and causes their dysfunction by regulating Drp1. As a key factor of PD-1, Drp1 has great potential as a target for synergistic anti-PD-1 therapy.

Regardless of the cause of metabolic stress (i.e., hypoxia, or continuous antigen stimulation and activation of the PD-1 signal pathway), mitochondrial damage ensues. As the main functional organelle in cells, mitochondria are constantly undergoing self-renewal, and damaged mitochondria are removed in a timely manner. Although the proteasome system can degrade mitochondria that undergo reversible damage, the clearance of mitochondria during an immune response mainly occurs through mitophagy (80).

Yu and colleagues showed that antigen stimulation combined with activation of the PD-1 signal impaired the autophagy activity of mitochondria, resulting in the accumulation of a large number of mitochondria in tumor-infiltrating T cells (67). These accumulated mitochondria are characterized by the destruction of the mitochondrial membrane and the structure of the cristae (67). The accumulated mitochondrial mass increases and the membrane potential decreases, indicating that these mitochondria are dysfunctional and depolarized (67). The inability to clear these dysfunctional mitochondria leads to a decrease in mitochondrial adaptation, resulting in the expression of more co-suppressor molecules, such as PD-1, on T cells (67). In addition, further analysis showed that the chromatin accessibility of T cells accumulating a large number of damaged mitochondria changed with DNA methylation, which was related to exhaustion (67).

Drp1-mediated mitochondrial fission plays a role in mitochondrial autophagy (81). Significant inhibition of mitochondrial autophagy was observed in Drp1-negative B cells (82). In addition, the importance of mitochondrial division for mitochondrial autophagy has been observed in a variety of cells (83). It can be speculated that damaged mitochondria are marked and separated by division, and mitochondrial autophagy will target and eliminate depolarized mitochondria (82).