SCFAs are the main metabolites produced by gut microbiota through bacterial anaerobic fermentation of non-digestible carbohydrates, such as dietary fiber (62). They are fatty acids with fewer than six carbon atoms, the most abundant of which are acetate (two carbons) and propionate (three), which are mainly produced by members of the Bacteroidetes phylum, as well as butyrate (four), predominantly produced by members of the Firmicutes phylum. Acetate derived from colonic bacterial fermentation can flow into the blood compartment and together with endogenous acetate can exert systemic activity upon the immune system. Propionate is a precursor of glucogenesis and has a direct effect on T cell activity locally. Butyrate effects are mainly restricted to the gut, where it is used as energy sourced by the colonocytes and, together with propionate, is processed in the liver (63). The effect of SCFAs on the metabolism of T-cells is yet to be understood but there is abundant data that suggest that SCFAs regulate the adaptive immune system through the modulation of mTOR activity, glucose metabolism, histone acetylation and cytokine gene expression (64). Multiple studies have confirmed that T cells are susceptible to SCFA exposure in vitro, promoting or inhibiting a specific T cell phenotype, and therefore supporting their use in ex vivo T cells expansion.

While research on the effect of microbial metabolites on T cells has primarily been focused on SCFAs, there are also other metabolites that have been investigated. For instance, the inosine, a purine metabolite that can be produced by Bifidobacterium pseudolongum, has been shown to boost the CD4+ Th1 differentiation in the presence of IFN-γ through adenosine A2A receptor (56). Moreover, the bacterial transformation of host bile acids has been demonstrated to directly modulate balance of Th17 and Treg cells (65, 66). Furthermore, tryptophan metabolites are reported to be essential for intestinal immunity and perform their effect on T cells through aryl hydrocarbon receptor signaling (67). Although these bacterial-derived metabolites are promising alternatives to SCFAs, in this manuscript, we will focus on the new and fast developing research field of the T and CAR-T cell ex vivo manufacturing optimization process through the SCFA bacterial derived metabolites. Results summarized in this review, might appear difficult to interpret as the effects of SCFAs are clearly dependent on immunological context and the studies compared often used a wide range of different conditions, from concentrations to exposure time. This is expected, as the field is only just starting to investigate the effect of gut microbial-derived metabolites on immune populations ex vivo. This lack of clarity and consistency reflects the urgent need for more rigorous and systematic protocols to better understand and learn from the microbiome-T cell crosstalk.

The effect of SCFAs in regulating T cell differentiation into effector and IL-10+ regulatory T cells was studied by Park et al. In this study T cells were activated in vitro for 5-6 days in the presence or absence of the SCFAs at different concentrations. The results showed that naïve CD4+ T cells, isolated from murine spleens and lymph nodes (LN), treated with acetate and propionate in vitro, led to differentiation into Th17 cells, which was observed through the increase in IL-17+ and IL-10+ cells as well as the increase in the transcription of the associated genes IL-17A, IL-17F, Rorc, RORα, T-bet, and IFN-g (25) Tables 1 , 2 .

Studies over the last 10 years have shown the ability of SCFAs to modulate and alter polarized CD4+ T cells via epigenetic and metabolic processes. Under tolerogenic conditions, naïve CD4+ T cells supplemented with low or physiogical concentrations of SCFAs seem to play a role in facilitating the activity of T regulatory cells, promoting the production of IL-10 and FoxP3+ Treg polarization (22, 39). Furusawa et al. isolated naïve CD4+ T cells from murine spleen and LNs. After three days of expansion cells were differentiated into Tregs in the presence or absence of acetate, propionate or butyrate, for an additional 2 days. Butyrate significantly increased the concentration of Foxp3+ cells; propionate did so moderately, while acetate showed no effect. The butyrate effect was in part mediated by histone H3 acetylation of the Foxp3 locus (22). Another study provided additional evidence indicating an increase of Foxp3+ CD4+ T cells for lower concentrations of propionate but not for higher ones (23). A third independent study, partially confirmed the enhanced Foxp3 expression in purified murine CD4+ T cells using a suboptimal concentrations of TGF-β1 and low concentrations of butyrate. However, the effect was not observed in the absence or under optimal concentrations of TGF-β1. In any case, higher concetrations of Butyrate did not enhance the expression of Foxp3 and instead, induced the expression of T-bet and IFN-γ via histone acetylation (HDAC inhibition), which is associated with an inhibition of Treg differentiation (24).

Luu and colleagues polarized CD4+ T cells towards Th17 for three days. Supplementation with pentanoate, another SCFA with 5 carbons, during the three-day process, impeded the polarization and inhibited the production of IL-17A through HDAC-inhibitory activity. In this case, and contrary to the results for the shorter SCFAs used in the study from Park et al., pentanoate led to a reduction in the transcription of the Th17 associated genes IL-17A, Rorc, Il21, Stat3, and Tgfb3. The treatment did, however, lead to the increase in the expression of IL-10 in Th17 cells. This increase seems to be due to pentanoate acting as a precursor of acetyl-CoA and activating the mTOR pathway, similarly to the mechanism for acetate. Results showed that Th17 cells treated ex vivo with pentanoate increased the extracellular acidification rate and enhanced their glycolytic activity leading to metabolic and epigenetic reprogramming and the loss of the pathogenic phenotype of Th17 cells in autoimmune disease (26).

Finally, under Th1 CD4+ polarizing conditions in the presence of IL-12, acetate and propionate supplementation potentiated the differentiation into Th1 cells, in a concentration dependent manner (25) Table 1 .

Activated CD8+ T cells cultivated in vitro using a SCFA-enriched medium have an increased production of IFN-γ and Granzyme B. SCFAs with shorter chain lengths such as acetate (two carbon chain) require greater concentrations for similar effects to ones with longer chain lengths such as pentanoate (five carbon chain) (28–31). For instance, murine CD8+ T cells polarized to CTLs in culture for three days, were supplemented with acetate for an additional three days at a concentration 25 mM. This led to a 125% increase in IFN-γ cells in comparison to untreated cells. Propionate, which contains a three-carbon chain, seems to require a concentration of approximately 5 mM to achieve similar effects (28). To obtain a comparable effect using butyrate, a SCFA with a four-carbon chain, a concentration of only 1 mM was required (29). In line with this hypothesis, murine CD8+ T expansion under CTL-inducing conditions supplemented with low concentration of acetate showed no effect over TNF-α or IFN-γ expression (27). SCFAs of shorter chain lengths are found in the body at greater concentrations (68). This might explain why T cells may could have evolved to become less sensitive to SCFAs of shorter chain lengths through a variety of mechanisms. In the example above, while acetate caused the increase in IFN-γ via the mTOR pathway (rapamycin, an mTOR inhibitor, led to an abrogation of the effect), butyrate caused it via the inhibition of HDAC activity (29). CD8+ T expansion under CTL-inducing conditions and supplemented with pentanoate, butyrate and, to a lesser extent, propionate showed an increase in the frequencies of TNF-α+ IFN-γ+ CD8+ T cells and secretion of TNF-α by CTLs. The effectiveness of butyrate and pentanoate treated T cells was further tested in a B16-OVA melanoma murine model and showed a better melanoma tumor control. Moreover, pentanoate-treated CD8+ T cells, showed an increase in in vivo persistence in a rag1-deficient mouse model, which was associated to an increase in CD25 expression and continuous IL-2 secretion (27). Qiu et al. tested CD8+ T cells under prolonged glucose restriction in vitro and showed that supplementation of acetate increased the expression of IFN-γ, although Granzyme B levels were not altered. These effects were mediated through histone acylation (31).

Differences in SCFAs concentration in medium lead to differences the expression of cytokines, however, the effect on the length of exposure seems to be less clear. While Luu and colleagues exposed murine activated CD8+ T cells to 1 mM Butyrate for three days (29), He and collaborators exposed activated CD8+ T cells to butyrate at the same concentration for just 12 hours, with the inclusion of IL-12 (28). They both obtained similar results for the proportion of IFN-γ cells, indicating a doubling in relation to untreated cells.

There are other factors to consider beyond concentrations or the duration of treatment, like whether the effect is long-lasting and leads to stable phenotypic change or their memory potential. The first one was explored by treating CTLs with butyrate for two days and then allowing the cells to grow without it for an additional three days. An increase in IFN-γ was still observed after treatment withdrawal compared to the control group, in vitro and in vivo , indicating a phenotypic stable change (29). The relationship between SCFA T cell ex vivo treatment and their memory potential were studied with transgenic gBT-I CD8+T cells treated with butyrate for three days after antigen-specific activation. While butyrate treatment reduced proliferation, it also led to greater responsiveness to IL-15 and a higher expression of transcription factor FoxO1 (30). Expression of FoxO1 is involved in the formation of memory T cells (69–71) while IL-15 promotes memory CD8+ T-cell survival and proliferation (72, 73), suggesting that butyrate may lead to a greater memory potential of the activated CD8+ T cells. This was supported by in vivo experiments that showed that butyrate treated gBT-I CD8+ T cells had a greater expansion upon antigen encounter. The suggested mechanism behind butyrate enhancement of the memory potential of CD8+ T cells might be its impact over glycolytic metabolism, uncoupling the TCA cycle from glycolitic input and favouring oxidative phosphorylation (30) Table 1.

All the studies mentioned above prove that SCFAs can modulate T cell polarization and memory formation ex vivo. There are few reports that show better therapy outcomes polarizing CAR-T cells in vitro before infusion. Interestingly, Th17-polarized human mesothelin-CAR T cells exhibited superior immunity against mesothelioma compared to Th1-polarized mesothelin -CAR T cells (74). In the same line of results, Guedan and colleagues showed that mesothelin-specific CAR Th17/Tc17 cells (that maintained Th17 function with a Th1 bias) had long-lived persistence in vivo and eradicated tumors (75). A third report studying the Th17 polarization effects on CAR-T cells, showed that a Th/Tc17 CAR-T targeting the proto-oncogene Neu, exhibited an increase antitumor immunity and improved early tumor control. The combination of the Th/Tc17 Neu-CAR-T and a STING agonist increased the trafficking, persistence, and tumor control in a murine model of breast cancer (76). Supporting the use of Th17 polarizing conditions for CAR-T cells, Fraietta and colleagues showed that CAR-T cells from chronic lymphocytic leukemia complete responders to CD19-CAR T therapy, had an enhanced transcriptomic profile of STAT3/IL-6 signaling, producing a type-17 signature compared with non-responders (77). Finally, a human mesothelin-CAR polarized towards Th9 cells was able to eliminate advanced human ovarian cancer patient-derived xenograft in humanized NSG mice. Regular expanded CAR-T or high doses of Th1+Tc1 polarized CAR-T cells could not achieve the same results (78). The differentiation status of CART cells plays a crucial role for therapeutic success as well. It is now well stablished that a less differentiated profile, like naïve or central memory T cells which have the capacity to persist and proliferate long-term in vivo lead to a better clinical outcome (77, 79). All the studies mentioned above highlight the potential of SCFAs over the process of expanding tumor-specific lymphocytes and CAR-T cell manufacturing for cancer immunotherapies. In this line, two studies recently published have explored the use of SCFAs on TILs and CAR-T cells.

Murine tumor-infiltrating lymphocytes (TILs) obtained from B16 melanoma and treated ex vivo with acetate showed a significant increase in IFN-γ secretion. This indicates that acetate can promote responsiveness in T cells isolated directly from the tumor microenvironment (31). The effect of butyrate and pentanoate supplementation during CAR-T cell ex vivo expansion was investigated under CTL-inducing conditions. Murine and human CD8+ T cells were used to generate a CAR that recognize receptor tyrosine kinase-like orphan receptor 1 (ROR1). Murine CAR-T cells treated with butyrate and pentanoate enhanced expression of CD25, as well as TNF-α and IFN-γ production. Pentanoate treated cells were further tested in a pancreatic tumor model and showed that tumor volume and weight were significantly reduced in comparison to non-treated cells. Later, the authors developed CAR-T cells from healthy human donor CD8+ T cells using a ROR1-specific CAR. After a two-step 17-day stimulation process, CAR-T cells were treated for four days with pentanoate at different concentrations. CAR-T cells pretreated with pentanoate showed upregulation of CD25 expression, IL-2 secretion, and stronger proliferation, as well as greater cytolytic activity when encountering their target antigen ROR1, together with an increase in IFN-γ and TNF-α production (27). Table 1 .