Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2025 Sep 25;15:32878. doi: 10.1038/s41598-025-16862-0

Intracellular pH links energy metabolism to lymphocyte death and proliferation

Wei-ping Zeng 1,✉, Shuangshuang Yang 2, Baohua Zhou 2
PMCID: PMC12464272  PMID: 40998896

Abstract

The role of intracellular pH (pHi) of lymphocytes in the control of the magnitude of immune response is unknown. The central question addressed in this report is whether energy metabolism affects pHi, which in turn regulates the death and proliferation of the lymphocytes and hence the magnitude of the immune response. To this end, we studied lymphocytes in the in vitro model of anti-CD3 activation and the in vivo mouse model of ovalbumin sensitization and challenge. We found that low pHi induces apoptosis of proliferating lymphocytes, whereas high pHi is conducive to their survival. In the in vivo model, treating the mice with the metabolic regulators dichloroacetate or C75 that increase the influx of carbons derived from pyruvate and fatty acid to the TCA cycle, respectively, lowered pHi. Treatments with the metabolic regulators CB-839 or GSK2837808A that inhibit glutaminolysis and aerobic glycolysis, respectively, also lowered pHi. Proliferation powered by high mitochondrial membrane potentials (MMPs) in lymphocytes of low but not high pHi was accompanied by apoptosis. After antigenic challenge, lymphocytes of high pHi increased and assumed a positive relation between pHi and MMPs, while lymphocytes of low pHi and with an inverse relation between pHi and MMPs diminished. These changes were largely dependent on glutaminolysis and aerobic glycolysis. It is therefore concluded that glutaminolysis and aerobic glycolysis are important for counterbalancing the acidic effects of pyruvate and fatty acid energy metabolism to promote a favorable pHi environment for lymphocyte survival and the progression of the immune response.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-16862-0.

Subject terms: Metabolomics, Apoptosis, Cell death and immune response

Introduction

Cell death and proliferation play fundamental roles in many biological processes including immune response. A typical in vivo immune response peaks at around one week and may last for several weeks1 during which lymphocytes expand and execute their effector functions. During this dynamic process, continuous shift of balance between lymphocyte proliferation and apoptosis determines the magnitude of the immune response. An early study suggests that selective use of energy metabolic pathways may play a role in determining the magnitude of immune response. It was found that the increase of energy production via aerobic glycolysis, i.e., ATP production by breaking down glucose in the cytosol without entering the TCA cycle in the mitochondrion, strongly correlates with lymphocyte proliferation2. This phenomenon was consistent with the previously described preferential use of aerobic glycolysis over mitochondrial respiration for energy production in tumor cells and other proliferating normal cells, which has come to be known as the Warburg effect3–7. It is estimated that 70–80% of human cancers use the Warburg effect for energy production8 and the contribution of aerobic glycolysis to total energy production ranges from less than 5% to more than 50%9. The relative contribution of aerobic glycolysis to total energy production in lymphocytes is less clear.

The end product of the Warburg effect lactic acid is exported out of the cell so that it creates an acidic extracellular environment. Many studies have investigated the biological consequences of the extracellular acidosis. In tumors, the acidic extracellular microenvironment facilitates tumor cell invasion of local tissues10,11. Extracellular acidosis is also a long recognized feature of inflammation12. In general, extracellular acidosis dampens the effector functions of innate immune cells such as the neutrophils, tumor-associated macrophages, and NK cells13–16. It also suppresses T cell-mediated immunity by inhibiting cytotoxicity and blocking cell cycle progression and the expression of IL-2 and IFN-γ17.

However, the salient feature of the Warburg effect is its low efficiency of energy production as it net produces only 2 ATPs per glucose versus 34 ATPs through mitochondrial respiration18. Like glucose, glutamine is also highly utilized by proliferating cells. Glutamine is imported into the mitochondria, where it is deamidated by glutaminase and further catabolized to produce ATPs, a process referred to as glutaminolysis. Intriguingly, like the Warburg effect, glutaminolysis is also extremely inefficient for energy production. Overconsumption of glutamines by cancer cells is well documented and it contributes to cachexia, a main direct cause of death of cancer patients19. However, in the highly proliferative glioblastoma cells, up to 60% of glutamine-derived carbons are not oxidized to produce ATPs but rather secreted as lactates and to lesser extent as alanines20.

The reason why the proliferating cells that have higher demand for energy than resting cells favor the low efficient Warburg effect and glutaminolysis for energy production is not fully understood. A prevalent view is that intermediates of these two pathways are diverted to other pathways such as the pentose phosphate pathway (PPP) for biomass synthesis, and excess carbons are dumped in favor of the production of NADPH for anabolic biochemical reactions21. However, aerobic glycolysis and glutaminolysis are not the only pathways that can supply substrates for biomass synthesis. Anabolic pathways are self regulated by feedback mechanisms. For example, in highly proliferating cells reactions in the non-oxidative phase of PPP are reversed so that ribose-5-phosphate can be produced for nucleotide synthesis22. Further, even the intermediates of the TCA cycle, particularly citrate, malate and oxaloacetate, can exit the cycle to support biomass synthesis5,23. Therefore, supporting biomass synthesis may not be an indispensible function of aerobic glycolysis and glutaminolysis.

Here, we report evidence that in addition to supporting biomass synthesis, the Warburg effect and glutaminolysis also support the survival of proliferating lymphocytes. We show that influx of carbons derived from pyruvate or fatty acids to the TCA cycle lowers intracellular pH (pHi), which in turn induces apoptosis. In contrast, glutaminolysis and the Warburg effect raise pHi to allow the proliferating lymphocytes to survive.

Results

Opposite effects of high and low pHi on lymphocyte proliferation

Normal lymph node cells from unimmunized mice were labeled with carboxyfluorescein succinimidyl ester (CFSE) and stimulated in vitro with anti-CD3 antibodies plus IL-2 or IL-2 only. Lymphocyte proliferation was detected by the progressive reduction of the CFSE signal following each cell division. The pHi of the lymphocytes was measured by staining the cells with the pH indicator pHrodo™ Red, whose fluorescence intensity (detected in PE channel) is inversely related to pH value. After the in vitro stimulation with anti-CD3 antibodies plus IL-2 or IL-2 alone, live lymphocytes (for live lymphocyte gates see Fig. S1A) were analyzed to determine the relation between pHi and proliferation. As expected, only minimum cell proliferation was detected in lymphocytes stimulated with IL-2 alone (Fig. S1B). In contrast, anti-CD3 antibodies plus IL-2 induced robust proliferation of CD4 and CD8 T cells, whereas most B cells remained quiescent (Fig. 1A). Based on their pH values, the CD8 T cells and B cells were classified into pHi-low, medium, and high populations whereas no pHi-low population was detected in CD4 T cells (Figs. 1A and S1B). Since we observed differences in cell divisions between pHlo and pHimed/hi populations in the flow cytometry plots, we decided to further investigate. While the pHilow CD8 T cells could undergo 1 to 2 divisions in the anti-CD3 plus IL-2-stimulated cultures, the more proliferative CD8, as well as CD4, T cells that had undergone 3 or more divisions were found in the pHimed/hi populations (5.59 ± 1.44% and 12.34 ± 6.39% vs. 0% and 1.13 ± 0.57% in pHilo populations of CD4 and CD8 T cells, respectively) (Fig. 1A). Likewise, divided cells (38.3% and 5.65% for CD4 and CD8 T cells, respectively) in the IL-2-only cultures were found primarily in the pHimed/hi populations (Fig. S1B). However, it should be noted that intermediate or high pHi values in the absence of anti-CD3 stimulation were not sufficient to trigger lymphocyte division as most of such lymphocytes remained quiescent (Fig. S1B). Thus, low pHi was unfavorable for whereas intermediate or high pHi was conducive to high rates (≥ 3 divisions) of lymphocyte proliferation.

Fig. 1.

Fig. 1

Open in a new tab

Relations between lymphocyte proliferation and pHi. (A) CFSE-labeled lymph node cells were stimulated in vitro with anti-CD3 antibody plus IL-2 for 2.5 days. Left panels show a representative experiment. The left top panels show the staining of CFSE and the pH indicator pHrodo™ Red (detected in PE channel) of live (Zombie-violetlow-neg) CD4 and CD8 T cells, and B cells (CD19). The pHi values of the lymphocytes were categorized as low (pHlo), medium (pHmed) or high (pHhi). Numbers in the plots are percentages of the cells. The left lower panels are histograms of CFSE staining of the same cells. Right panels show the mean percentages of cells that have undergone ≥ 1 (top) or ≥ 3 (lower) divisions in all live CD4, CD8 T cells and B cells (CD19), data are pooled from three samples in three similar experiments. (B) Overlaid histograms of pHrodo™ Green staining (detected in FITC channel) of ex vivo live CD4 and CD8 T cells, and B cells from mice one day after in vivo treatments with saline or saline plus HOAc (60 µl 175mM HOAc in saline/mouse, i.t.). Negative controls were CD4 and CD8 T cells, and B cells without pHi staining. One of two similar experiments is shown. (C) Mean folds, relative to saline treatment controls, of the numbers of live CD4 and CD8 T cells and B cells in the mediastinal lymph nodes (MLNs) of ovalbumin (OVA)-sensitized and challenged mice treated with saline plus HOAc, data are pooled from five similar experiments (5 samples). (D) Left panels are histograms of Ki-67 staining of the T and B cells in live Lym2 populations (large lymphoblasts, see Fig. S3A for Lym2 gates) of the MLNs in a representative experiment. Numbers in the histograms are the percentages of Ki-67 high (hi) or low (lo) cells. Right panels show the mean folds, relative to saline controls, of the percentages of Ki-67 high and low cells, data are pooled from three similar experiments. Error bars in all bar graphs are standard deviations. Statistical significance is determined by Student’s t test. P values of the test are shown in the graphs.

To further investigate the relationship between low pHi and lymphocyte proliferation, we sought to manipulate pHi by treatments with a mild HOAc solution (87.5 mM HOAc in saline). For in vitro treatments, lymphocytes were incubated in FBS supplemented with 1/10 vol. of the HOAc solution to better mimic in vivo condition and avoid interference by pH buffering agents in cell culture medium. On the other hand, in vivo treatments were carried out by intratracheal (i.t.) injection of the HOAc solution (60 µl 175mM HOAc in saline/mouse) (i.t. injection was used because we studied the lymph nodes that drain the lungs). Both in vivo and in vitro, HOAc treatments effectively lowered the pHi in CD4, CD8 T cells and B cells (Figs. 1B, S2A). The effects of the in vitro HOAc treatments on lymphocyte proliferation were studied by analyzing the level of Ki-67, a reliable marker for cell proliferation24. The in vitro HOAc treatment almost completely depleted the Ki-67+ but not the Ki-67− lymphocytes (Fig. S2B), demonstrating that the HOAc treatment preferentially depleted proliferating lymphocytes.

For the in vivo study, mice were sensitized and intratracheally challenged with ovalbumin (OVA). In the lung-draining lymph nodes25 (collectively referred to as mediastinal lymph nodes (MLNs) hereafter), HOAc treatments reduced the numbers of live CD4 and CD8 T cells and B cells (Fig. 1C). The MLNs contained lymphocytes of relatively small and large sizes, which we referred to as Lym1 and Lym2 populations, respectively (see Fig. S3A for gating). The Lym2 cells consisted of large percentages of highly proliferating (Ki-67hi) cells (Fig. 1D upper). HOAc treatments preferentially depleted the Ki-67hi cells, but not Ki-67lo cells (Fig. 1D). On the other hand, 32–92.7% of the Lym1 cells expressed only low levels of Ki-67. HOAc treatment showed only modest effects on Ki-67lo CD4 T cells and B cells of the Lym1 populations although it caused sizable reduction of Ki-67lo CD8 T cells (Fig. S3B). These results showed that the proliferative status of lymphocytes positively correlated with their susceptibility to depletion by HOAc treatments.

Induction of apoptosis by low pHi

We further investigated whether the low pHi-induced depletion of lymphocytes was mediated by apoptosis. Since pHi in apoptotic (dead) cells cannot be measured because of their leaky cell membranes, we instead examined early apoptosis of live lymphocytes (for live lymphocyte gates, see Fig. S4A) in the MLNs and the non-draining lymph nodes (NDLNs). In both the MLNs and NDLNs, early apoptotic (Annexin V+7-AAD−) cells were found mainly in the pHilo populations of live total lymphocytes, CD4, CD8 T cells and B cells (Fig. 2A). In separate experiments, we used the activation of caspase 3 as an alternative measurement of early apoptosis. Caspase-3 activation in live MLN lymphocytes (gating shown in Fig. S4B) confirmed the results of Annexin V staining and further showed that the low pHi-induced apoptosis was mediated by caspase activation (Fig. 2B). To corroborate these findings, the MLN cells were treated with HOAc in vitro. The treatments caused dose-dependent induction of apoptosis in the different lymphocyte populations (Fig. 2C top and bottom left panels). Again, the early apoptotic (7AAD− Annexin V+) cells were found primarily in the pHilo populations (Fig. 2C middle panels). Notably, the percentages of early apoptotic pHlo CD4 T cells and B cells in the HOAc treatment groups were lower than saline controls (Fig. 2C bottom right). This showed that the HOAc treatment not only induced the apoptosis of Annexin V− live cells but also promoted the Annexin V+ early apoptotic cells to progress to apoptosis therefore reduced their proportions. Collectively, these results demonstrated a causal relationship between low pHi and apoptosis of the lymphocytes.

Fig. 2.

Fig. 2

Open in a new tab

Induction of apoptosis by low pHi in lymphocytes from the lymph nodes of OVA-sensitized and challenged mice. (A) Upper panels are ex vivo live lymphocytes from MLNs or NDLNs stained with Annexin V and the pH indicator pHrodo™ Green in a representative experiment. Numbers in the plots are the percentages of cells in each quadrant. Lower panels compare the mean percentages of early apoptotic cells (Annexin V+) between pHi high (Q1 + Q4) and low (Q2 + Q3) cells of live total lymphocytes (Lym), CD4, CD8 T cells and B cells (CD19), data are pooled from eight similar experiments. (B) Upper panels are ex vivo live MLN lymphocytes stained with the caspase-3 substrate NucView® 405 and pHrodo™ Green in a representative experiment. Lower panel compare the mean percentages of cells with caspase 3 activation (NucView® 405+) between pHi high and low cells, data are pooled from three similar experiments. (C) MLN lymphocytes were incubated in FBS plus 1/10 volume of saline or saline containing 43.75mM (0.5x) or 87.5mM (1x) HOAc. The top panels are plots of 7AAD and Annexin V staining of the indicated lymphocyte populations, and the middle panels are plots of Annexin V and pHi staining of the indicated live (7AAD−) lymphocytes in a representative experiment. The lower panels are mean percentages of apoptotic cells in the indicated lymphocytes (left) and early apoptotic cells in the indicated live pHi-low lymphoyctes (right). Error bars in all bar graphs are standard deviations. Statistical significance of differences between pHi high and low cell populations (A and B) or between saline and 0.5x HOAc, or between 0.5x and 1x HOAc treatments (C) was determined by Student’s t test, *, p < 0.05; **, p < 0.01.

Inverse relation between MMPs and pHi at the early stage of immune response

Since energy production by mitochondrial respiration generates protons, we investigated whether this process contributes to the regulation of pHi in lymphocytes under physiological conditions. To this end, NDLNs and MLNs were harvested at an early stage of 3 days after the initial OVA challenge to measure the mitochondrial membrane potentials (MMPs) and pHi of the lymphocytes. Comparisons between MLNs and NDLNs, as well as lymph nodes of naïve mice, found that the pHi values of live MLN lymphocytes (for live lymphocyte gates see Fig. S4C) were much higher than those of the live lymphocytes of NDLNs and naïve lymph nodes (Fig. 3A,B left). In Fig. 3A, most lymphocytes in the NDLNs (97.2%) and naïve lymph nodes (89%) were in the right two quadrants where cells had relatively low pHi (pHi-lo gate). In contrast, most lymphocytes in the MLNs (83.3%) were found in the left two quadrants of relatively high pHi (pHi-hi gate) although substantial number of cells remained in pHi-lo gate. The pHi of lymphocytes in the MLN-pHi-hi gate were significantly higher than those of the lymphocytes in the MLN-pHi-lo, NDLN-pHi-lo or Naïve-pHi-lo gates (Fig. S5). However, the MMPs of the MLN-pHi-hi lymphocytes were conversely lower than those of MLN-pHi-lo, NDLN-pHi-lo or Naïve-pHi-lo lymphocytes (Fig. 3B right), demonstrating that the rise of pHi in the MLN-pHi-hi lymphocytes at the early stage of immune response was associated with the decline of their MMPs.

Fig. 3.

Fig. 3

Open in a new tab

Interrelations among MMPs, pHi, apoptosis and proliferation at the early stage of immune response in the lymph nodes of OVA-sensitized and challenged mice. (A) Ex vivo live lymphocytes from NDLNs and MLNs and lymph nodes of unimmunized (naïve) mice stained with pHrodo™ Green and MitoSpy Orange (indicator of MMPs) in a representative experiment. (B) Left two panels are the overlaid histograms of the pHi of the same cells in (A) (left) and the MMPs (right) of their pHi-low (Q2 + Q3) and pHi-high (Q1 + Q4) populations as defined in (A). The right two panels compare the mean pHi values (MFI of pHrodo™ Green minus negative controls) and MMPs (MFI of MitoSpy Orange minus negative controls) of the indicated lymphocyte populations, data are pooled from three similar experiments. Error bars are standard deviations. Statistical significance of differences was determined by Student’s t tests, p values of the test are shown. (C) The left panel is a plot of live MLN lymphocytes showing the gates of R1 to R8 subpopulations for further analyses and sorting. (D) Plotting of the pHi values against the MMPs of the R1 to R8 subpopulations defined in (C) left. (E) Top left panels are overlaid histograms of Annexin V staining of the R1 to R8 subpopulations defined in (C) left, and the top right panel is a bar graph showing the MFI of Annexin V staining (minus negative controls) of the cell subpopulations. Lower left panels are overlaid histograms of Ki-67 intracellular staining of sorted R1-4 and R7 subpopulations defined in (C) right. Lower right panel is a bar graph showing the MFI of Ki-67 staining (minus negative controls) of the sorted subpopulations.

Further examination of cell distribution in Fig. 3A found that the pHi of most lymphocytes in the NDLNs uniformly displayed an inverse relation with MMPs. However, the distribution of the MLN lymphocytes was more heterogeneous. While MLN lymphocytes in the pHilo MMPhi (Q2) quadrant maintained a strong inverse relation between pHi and MMPs similarly to that of the NDLN lymphocytes, the inverse relation of lymphocytes in the pHihi quadrants (Q1 and Q4) appeared to be weak (Fig. 3A). To confirm this notion, we arbitrarily divided the MLN lymphocytes into R1 to R8 subpopulations of progressively higher pHi, where R1 and R2 were equivalent to cells in the pHilo MMPhi quadrant whereas R3, R4, R7 encompassed the majority of cells in the pHihi quadrants (Fig. 3C). Plotting the pHi values of these subpopulations against their MMP values showed that indeed cells in the pHi-lo subpopulations (R1 + R5 and R2 + R6) displayed a steep inverse relation between pHi and MMPs (slopes = 2.38, 1.52, 0.71, and 3.77 for Lym, CD4, CD8 and CD19 populations, respectively) whereas cells in the pHi-hi subpopulations (R3 + R7 and R4 + R8) showed a flatter or weaker inverse relation between pHi and MMPs (slopes = 0.82, 0.33, 0.30, and 0.29 for Lym, CD4, CD8 and CD19 populations, respectively) (Fig. 3D).

Although other ion channels contribute to MMP, the MMP is mainly determined positively by mitochondrial respiration that creates the proton gradient across the mitochondrial inner membrane and negatively by ATP synthesis that is powered by the proton gradient. Therefore, the lower MMPs in P (pH-hi) lymphocytes could be mainly due to lower mitochondrial respiration or dysfunctional ATP synthase. To investigate these alternatives, the MLN lymphocytes were incubated in FBS plus pyruvate and L-glutamine with increasing concentrations of the protonophore carbonyl cyanide-4- trifluoromethoxy phenylhydrazone (FCCP) or oligomycin A to assess the basal levels of MMP and the functionality of ATP synthase, respectively. As expected, FCCP decreased MMPs in both the N  (pH-lo) and P lymphocytes of live total lymphocytes, CD4, CD8 T cells, and B cells. In all lymphocyte populations, P lymphocytes showed lower MMPs than their N lymphocyte counterparts (Fig. S6 upper panels). Conversely, oligomycin A treatments increased MMPs because it inhibits ATP synthase. Importantly, the increases of MMPs by oligomycin A were observed in both the N and P lymphocytes, and the MMPs in P lymphocytes remained lower than those of their N lymphocytes counterparts (Fig. S6 lower panels). These results showed that the ATP synthase in both the N and P lymphocytes was functional and negatively contributed to MMPs.

We further analyzed the differences of early apoptosis among the subpopulations. We found that early apoptosis progressively decreased (indicated by the decrease of MFI of Annexin V staining) as the pHi increased (Fig. 3E upper panels). This observation is consistent with the notion that low pHi induces apoptosis in Fig. 2. We also determined the proliferative status of the lymphocytes by intracellular staining of Ki-67 in the sorted R1, R2; and R3, R4, R7 subpopulations, which represented the majority of cells in the pHi-lo and pHi-hi subpopulations, respectively (see sorting gates in Fig. 3C right). Cells expressing high levels of Ki-67 were found in the pHi-lo subpopulations with the R1 subpopulations expressed the highest levels of Ki-67, whereas pHi-hi subpopulations expressed relatively low levels of Ki-67 (Fig. 3E lower panels). Thus, the pattern of Ki-67 staining resembles that of the Annexin V staining. Collectively, Fig. 3 shows that at the early stage of immune response high rates of lymphocyte proliferation powered by high MMPs were accompanied by high rates of apoptosis and low pHi. This explains why the R1 cells, despite their high rates of proliferation, constituted only small percentages of the live lymphocytes. This explanation is consistent with the in vitro findings that lymphocytes of high proliferation rates failed to accumulate in the pHi low populations (Fig. 1A). The small numbers of highly proliferative lymphocytes in Fig. 3 were likely newly activated lymphocytes that replenished the lymph nodes.

Dynamic changes of pHi and its relations with MMPs

We extended the study of MMPs and pHi of lymphocytes in the MLNs to later time points of day 4, 5 and 7. Strikingly, at day 7 the lymphocytes showed a “v” shape of distribution. We therefore divided the lymphocytes at the cusp of the v shape distribution into pHi low and high populations, referred to as the N and P lymphocytes, respectively (Fig. 4). As the immune response progressed from day 4 to day 7, the percentages of the P lymphocytes increased with corresponding decreases of the percentages of the N lymphocytes. At all the time points, the N lymphocytes maintained the inverse relation between pHi and MMPs. However, at day 4, the P lymphocytes of the CD4 and CD8 T cells started to show a positive relation between pHi and MMPs. This positive relation became more pronounced as the immune response further progressed to day 5 and day 7. The appearance of such positive relation in the P lymphocytes of the B cells was slower, but also became apparent at day 7 (Fig. 4).

Fig. 4.

Fig. 4

Open in a new tab

Dynamic changes of pHi and MMPs of lymphocytes in the MLNs of OVA-sensitized and challenged mice. Left panels show ex vivo live total lymphocytes, CD4, CD8 T cells and B cells stained with MitoSpy Orange and pHrodo™ Green. The lymphocytes are divided into populations of relatively high (P) and low (N) pHi. MLNs were harvested on day 4, 5 or 7 after the initial OVA challenge. Numbers in the plots are percentages of the N and P lymphocytes. Right panels show the mean percentages of P and N lymphocytes at the indicated time points, data are pooled from three similar experiments. Error bars are standard deviations. Statistical significance of differences was determined by Student’s t tests, p values of the test are shown.

Regulation of pHi and N and P lymphocyte populations by energy metabolic pathways

The relations between MMPs and pHi confirmed a role of energy metabolism in the regulation of pHi, and consequently also suggested its role in controlling the relative proportions of the N and P lymphocytes. We therefore used pharmacological compounds to test this hypothesis and pinpoint the specific energy metabolic pathways for such regulations. Compound GSK2837808A (LDH inh for short) is an inhibitor of lactate dehydrogenase26. Lactate dehydrogenase prevents pyruvate from entering the TCA cycle by converting the pyruvate to lactate. Therefore LDH inh is an inhibitor of aerobic glycolysis and the Warburg effect. CB-839 is an inhibitor of glutaminase 1 hence an inhibitor of glutaminolysis27 because it inhibits deamidation of glutamine, the first and essential step of glutaminolysis; dichloroacetate (DCA) is an inhibitor of pyruvate dehydrogenase kinase therefore promotes the influx of carbons derived from pyruvate to the TCA cycle (referred to hereafter as influx of pyruvate)28; and C75 is an activator of carnitine palmitoyltransferase 1 A therefore promotes the influx of carbons derived from fatty acids to the TCA cycle (referred to hereafter as influx of fatty acids)29–31.

OVA-sensitized and challenged mice were treated by i.p. injection of each of the compounds. The results showed that inhibiting glutaminolysis or the Warburg effect, or alternatively increasing the influx of pyruvate or fatty acids, lowered the pHi of CD4, CD8 T cells and B cells (Fig. 5A,B). Consistent with this finding, inhibition of glutaminolysis or the Warburg effect decreased the percentages of the P lymphocytes (Fig. 5C,D) and their absolute numbers (Fig. 5E). Although the percentages of the N lymphocytes increased due to the reduction of the P lymphocytes, inhibition of either pathway did not increase the absolute numbers of the N lymphocytes, but rather significantly decreased the absolute numbers of the N lymphocytes of the CD8 T cells and B cells (Fig. S7). Increase of the influx of pyruvate or fatty acids did not decrease the percentages of the P lymphocytes of the CD4 and CD8 T cells but reduced the percentages of the P lymphocytes of the B cells (Fig. 5C,D). However, regardless of the changes in the percentages of the P and N lymphocytes, the increase of influx of pyruvate or fatty acids reduced the absolute numbers of both the P and N lymphocytes of T and B cells (Figs. 5E and S7).

Fig. 5.

Fig. 5

Open in a new tab

Regulation of pHi and the proportions of the P and N lymphocytes by energy metabolic pathways. OVA-sensitized and challenged mice received i.p. injections of vehicle control or the indicated energy metabolic regulators as described in the Methods. Ex vivo live lymphocytes of the MLNs from the mice were analyzed. (A) Overlaid histograms of pHrodo™ Green staining of the T and B cells. (B) Means of changes of pHi (MFI of pHrodo™ Green) of lymphocytes treated with the metabolic regulators as compared with vehicle controls, data are pooled from 4 similar experiments. (C) Ex vivo live total lymphocytes (Lym), CD4, CD8 T cells and B cells from mice received the indicated treatments. Numbers in the plots are the percentages of the P and N lymphocyte populations. (D,E) Means of changes of the percentages (D) and numbers (folds) of P lymphocytes as compared with vehicle controls after the treatments with the metabolic regulators, data are pooled from 3 similar experiments. Error bars in all bar graphs are standard deviations. Statistical significance of differences from the vehicle controls was determined by Student’s t test, *, p < 0.05; **, p < 0.01.

Higher rates of proliferation but no apoptosis of MMP-high P lymphocytes

MLN lymphocytes at the late stage of immune response (day 7) were studied to determine the inter-relations among MMPs, pHi, apoptosis and proliferation in the P lymphocytes, and compared them with the N lymphocytes. Similarly to the early stage of immune response in Fig. 3, distinct populations of early apoptotic cells were detected in the N lymphocytes with either high or low MMPs. In contrast, no distinct populations of early apoptotic cells were detected in the P lymphocytes regardless of their MMP levels. (Fig. 6A,B). However, both the P and N lymphocytes with higher MMPs had higher rates of proliferation as judged by their higher Ki-67 levels than their counterparts with lower MMPs (Fig. 6C,D). Thus, unlike the N lymphocytes, P lymphocytes could proliferate, even at high rates, without triggering apoptosis. Inhibition of the Warburg effect and glutaminolysis reduced the percentage of the P lymphocytes and conversely increased the percentage of the N lymphocytes (Fig. 6E middle). Increase of the influx of pyruvate and fatty acids also increased the percentage of the N lymphocytes but to lesser degree, and the treatment elevated the MMP levels of both the N and P lymphocytes so that the percentages of the MMP-high lymphocytes were increased (Fig. 6E right). However, the pHi values of the different P and N lymphocytes in the metabolic treatment groups were lower than those of their counterparts of the vehicle controls (Fig. S8), in agreement with the roles of the energy metabolic regulators in modulating pHi demonstrated in Fig. 5. Like in Fig. 6B, early apoptotic populations were detected in the N but not P lymphocytes of the different treatment groups (Fig. 6F). Moreover, the proliferative statuses of the pHi-lo or hi lymphocytes in the metabolic treatment groups were similar to those of their counterparts in the vehicle control group (Fig. 6G), confirming the notion that it is the pHi and MMP profiles that correlate with the proliferative statuses of the lymphocytes.

Fig. 6.

Fig. 6

Open in a new tab

Interrelations among MMPs, pHi, apoptosis and proliferation of P (left two quadrants) lymphocytes and comparisons with N (right two qudrants) lymphocytes. (A) The P and N populations of T and B cells of MLNs are divided into subpopulations of low (lo) or high (hi) MMPs. (B) Overlaid histograms of Annexin V-BV650 staining of the subpopulations defined in (A). (C) Overlaid histograms of Ki-67 staining of the sorted MMP-high or –low subpopulations of the P and N lymphocytes. (D) Means of folds of Ki-67 MFI (minus negative controls) of the sorted subpopulations relative to the average MFI (minus negative controls) of the live total lymphocytes, CD4, CD8 T cells or B cells in all samples, data are pooled from four similar experiments. (E) MMP and pHi staining of live MLN lymphocytes of mice that received the indicated treatments, numbers are percentages of the cells. (F) Overlaid histograms of Annexin V-BV650 staining of the subpopulations defined in (E). (G) Means of folds of Ki-67 MFI of the sorted subpopulations as defined in (E) relative to the average MFI of the indicated lymphocyte populations of all samples, data are pooled from three similar experiments. Error bars in all bar graphs are standard deviations. Statistical significance of differences determined by Student’s t test, level of the significance (* p < 0.05; ** p < 0.01) or the p values are shown.

Finally, in this study, the MMP is measured by staining the lymphocytes with MitoSpy Orange. MitoSpy Orange is a tetramethylrosamine (TMRM)-based dye32. It is well established that the accumulation of TMRM-based dyes in mitochondria is dependent on mitochondrial membrane potential as the accumulation is abolished by treatment with the protonophore FCCP that depolarizes the mitochondrial inner membrane33. However, it is possible that both the MMPs of individual mitochondria and the total mass of mitochondria may contribute to the total MitoSpy signal of a cell. To investigate the possibility, we stained the day 7 MLN cells with Mito Tracker Green, which measures the total mass of mitochondria. We found that in most lymphocytes, mitochondrial mass slightly increases as pHi decreases, but there are small numbers of pHi-low lymphocytes that have low mitochondrial masses (Fig. S9). On the other hand, mitochondrial masses were low in pHi-high lymphocytes. Therefore, the high MMPs in the pHi-high lymphocytes was not due to high mitochondrial mass whereas the high MMPs in most pHi-low lymphocytes (Fig. S9 upper) was partially attributable to high mitochondrial masses.

Discussion

While previous studies have investigated the biological consequences of extracellular acidosis, in this study, we discovered a critical role of intracellular pH (pHi) in regulating the death and proliferation of lymphocytes therefore the magnitude of immune response. The in vitro and in vivo evidence in the present study established a causal relationship between low pHi and apoptosis. First, we found a clear association between low pHi and apoptosis. We further showed that treating lymphocytes with HOAc lowered pHi and induced apoptosis of lymphocytes in a dose dependent manner. Arguably, HOAc treatment can alter both intracellular and extracellular pH. However, caspase activation is an intracellular process, therefore it is reasonable to argue that the apoptosis was caused by low intracellular rather than extracellular pH. It is also noteworthy that in the in vivo experiment, only minuscule amount of HOAc (0.6–60 µl of 175mM HOAc in saline) diluted in saline was administered to an individual mouse. The HOAc treatments caused little, if any, extra stress to the mice, which is reminiscent of the fact that humans suffer no serious pathophysiological consequences from transient severe acidosis as the result of grand mal seizure or strenuous physical excises34,35. Therefore, it could be speculated that potential low pHi-based therapies for diseases attributed to dysregulated cell death and proliferation might be well tolerated by human subjects.

We further demonstrated that under normal physiological condition, energy metabolism is a major mechanism for regulating pHi. Specifically, influx of carbons derived from pyruvate or fatty acids to the TCA cycle lowers pHi, whereas glutaminolysis and aerobic glycolysis raise pHi. Importantly, in lymphocytes of low pHi, high rates of proliferation powered by high MMPs were accompanied by apoptosis. This result was consistent with the in vitro data showing that highly proliferating lymphocytes failed to accumulate in the pHi-low populations. Therefore, it appears that proliferating lymphocytes use glutaminolysis and the Warburg effect for energy production to avoid low pHi-induced apoptosis. Although our study focuses on lymphocytes, glutaminolysis and the Warburg effect are found in various types of normal and malignant cells21,36. In malignant cells, it is believed that the excretion of the end product of the Warburg effect lactate creates an acidic extracellular tumor microenvironment that facilitates tumor cell invasion of local tissue10,11. Our data suggest that the acid excretion may also be necessary for the tumor cells to avoid low pHi-induced apoptosis. Thus, while the Warburg effect is generally associated with cell proliferation, our study demonstrated a clear link between the Warburg effect and cell survival. Furthermore, it must be pointed out that the Warburg effect and the dependence on glutamines for cell proliferation are evolutionarily conserved37–40. Therefore, the same survival strategy may apply to other cell types across species, particularly cells that are highly active in energy production.

The present study identified low pHi as a natural and cell intrinsic trigger of apoptosis that is intricately linked to cellular energy metabolism. This discovery is distinct from the earlier observation of cytosolic acidification in cultured cells as a result of the redistribution of mitochondrial contents caused by external agents that compromise the mitochondrial membrane integrity41–44. In contrast, the low pHi observed in the lymphocytes in this study was largely the result of normal mitochondrial energy production fueled by carbons derived from pyruvate and fatty acids.

On the other hand, we also observed a bi-phasic increase of pHi in lymphocytes during the immune response. At the early stage of the immune response, the increase of pHi is associated with the reduction of MMPs. As the immune response progresses, the increase of pHi is largely dependent on glutaminolysis and aerobic glycolysis, and the lymphocytes of high pHi show a positive relation between pHi and MMPs. The exact chemical modes of action for glumtaminolysis and aerobic glycolysis to increase pHi are beyond the scope of this study. Nonetheless, it is tempting to speculate that ammonia derived from glutaminolysis may lower the overall proton content in the cells. In contrast, aerobic glycolysis followed by lactate excretion by itself is pH neutral. However, it lessens the burden of energy production from mitochondrial respiration, therefore could indirectly reduce the production of protons from mitochondrial carbon oxidation thereby also lowers the overall proton content in the cells.

The dynamic nature of immune response stresses the importance of understanding and medically interfering with the molecular events in lymphocytes in accordance with the kinetics of immune response. However, unlike in the laboratories, in clinical settings the time of initial antigen exposure is usually uncertain therefore it is difficult to determine the stage in the natural time course of an immune response at the time of medical treatment. This study shows that the stage of an immune response can be determined by examining the pHi of the lymphocytes and its relations with MMPs without having to know the time of the initial antigen exposure. Specifically, the early stage of immune response can be determined by the dominance of low pHi lymphocytes with an inverse relation between pHi and MMPs. As the immune response progresses, high pHi lymphocytes increase and assume a positive relation between pHi and MMPs. At the late stage, the immune response is dominated by high pHi lymphocytes with a positive relation between pHi and MMPs. Applying this immune stage characterization system to clinical cases could potentially improve the precision and efficacy of therapies that target immune responses. However, we must point out that further studies are needed to analyze the antigenic specificity of the lymphocytes at the different stages of immune response.

Nonetheless, based on the evidence in the present study, we conclude that glutaminolysis and aerobic glycolysis in proliferating lymphocytes play a critical role in counterbalancing the acidic effect of pyruvate and fatty acid energy metabolism to provide a favorable pHi environment for the survival of the lymphocytes and progression of the immune response.

Methods

Mice and OVA immunization model

Balb/c mice were purchased from Jackson Laboratory (Bar Harbor, ME) and housed in the animal facility of Charles River Accelerator and Development Lab (CRADL) (Cambridge, MA). Additional mice were maintained at Therazwimm animal facility. Animal studies were performed according to the protocols approved by the CRADL and Therazwimm Institutional Animal Care and Use Committees. All studies were performed in compliance with the Animal Welfare Act and Regulations45. OVA sensitization and challenge were performed as previously described46,47. Briefly, Balb/c mice were sensitized by i.p. injection of 20 µg OVA (Sigma-Aldrich, St. Louis, MO) in total volume of 100 µl PBS and Alum adjuvant (33.4% in volume) (Thermo Scientific, Rockford, IL). Again, two to three weeks later, the sensitization is repeated with 500 µg OVA mixed in Alum adjuvant. Two to three weeks later, mice were challenged with 100 µg OVA in 60 µl saline by i.t. instillation. The study is reported in accordance with ARRIVE guidelines48.

In vitro stimulation of lymphocytes

Lymph node cells were washed 3 times with plain PBS, then incubated (5 × 106/ml) in 1µM Carboxyfluorescein succinimidyl ester (CFSE) (Fluka/Sigma-Aldrich, Burlington, VT) in plain PBS at room temperature for 7 min. After the incubation, 1/4 volume of FBS was added to stop the labeling, and cells were washed 4 times with PBS plus 1% FBS. The labeled cells were cultured in complete RPMI-1640 medium (RPMI-1640 supplemented with 1x glutamax, 1x pen/strep and 5% FBS) plus IL-2 (20 units/ml) or IL-2 and anti-CD3 antibody (1 µg/ml) (BD Pharmingen, San Diego, CA) for 2.5 days.

Measuring pHi, MMP and early apoptosis

MMP and pHi were determined by staining lymphocytes with the MMP indicator MitoSpy Orange (Biolegend, San Francisco, CA) and the pH indicator pHrodo™ Green/Red AM (ThermoFisher Scientific, Waltham, MA), respectively. Intensity of MitoSpy staining is MMP-dependent, and intensity of pHrodo™ staining is inversely correlated with pHi value. Apoptosis and early apoptosis were detected by staining the lymphocytes with 7AAD and Annexin V (Biolegend) or NucView® Caspase-3 Substrate (Biotium, Fremont, CA). For staining with MitoSpy, pHrodo™ and Annexin V, cells were washed once with Live Cell Image Solution (LCIS) (Life Technology/ThermoFisher Scientific, Grand Island, NY). The cells (1–2 × 107/ml) were re-suspended in LCIS supplemented with 0.7mM CaCl2, plus 1µΜ pHrodo™, 100nM MitoSpy Orange, Annexin V and/or the Caspase-3 Substrate, and Zombie viability dye (Biolegend) at dilutions recommended by the manufacturers individually or in combination. The cell suspension was incubated in 37 °C water bath for 20 min and in ice water for 15 min, then washed in LCIS supplemented with 0.7mM CaCl2 and 1% FBS. The cells were re-suspended in LCIS supplemented with 0.7mM CaCl2, 7-AAD was added to the cells prior to flow cytometric analysis. MMP, pHi and early apoptosis were determined in gated live (7AAD− or Zombie−/lo) lymphocytes. In flow cytometric data analysis, the ranges of the axes of MMPs and pHi were adjusted (for example, by biexponential transformation) to include all live lymphocytes within the boundaries and the ranges were also shortened to better visualize the heterogeneity of the lymphocytes.

Flow cytometry

Fluorochrome-conjugated antibodies against mouse CD4, CD8, CD19, TCRβ and Ki-67, 7-AAD, Zombie-Green/Violet fixable viability dyes, fluorochrome or biotin-conjugated Annexin V, fluorochrome-conjugated streptavidin, and Foxp3 buffer set were purchased from Biolegend (San Diego, CA). Lymphocytes were stained for cell surface markers or with Annexin V. Live and dead cells were distinguished by staining the cells with the Zombie dyes or 7-AAD. For Ki-67 staining, cells were fixed with CytoFix/CytoPerm buffer, washed with PBS plus 1% FBS and 1x CytoPerm buffer, and incubated with antibodies against mouse Ki-67. For lymphocyte cell count by flow cytometry, lymph node cells of different mice were prepared as single cell suspensions in the same volume of buffer. An aliquot of the same volume of cell suspension were stained with 7AAD with or without staining with other dyes and antibodies. The stained cells were re-suspended in the same volume of buffer and analyzed by flow cytometry with the same acquisition volume, flow speed and acquisition time. All flow cytometry experiments were carried out using FACSCanto (BD. Bioscience, Franklin Lakes, NJ) or Attune™ cytometer (Invitrogen, Carlsbad, CA). Typically more than 100,000 events are collected. Data analyses were carried out using FlowJo.

Treatments with FCCP or oligomycin A

FCCP and oligomycin A were purchased from MedChemExpress (Monmouth Junction, NJ). MLN cells were suspended in FBS supplemented with 1mM sodium pyruvate, 2mM L-glutamine, 100 units/ml penicillin and 100 µg/ml streptomycin at 1.5 × 106 cells/ml. Vehicle (5% DMSO in saline), FCCP or oligomycin were added to the cell suspensions at the final concentrations indicated in each experiment. The cells were incubated at 37 °C, 5% CO2 for 1 h. After the incubation, cells were washed with plain PBS, then stained with pHrodo™ Green and MitoSpy Orange to measure pHi and MMPs.

Treatments with pH modifiers or energy metabolic regulators

Lymphocytes were treated with saline or HOAc to determine the effects of the pH modifiers on pHi, cell proliferation and apoptosis. For in vitro treatments, lymphocytes were incubated in FBS supplemented with 10% volume of saline or saline plus 87.5 mM (1x) or 43.75 mM (0.5x) HOAc in 37 °C water bath for 5 h. For in vivo treatments with HOAc, OVA-sensitized mice were challenged on day 0 and day 1, mice receive i.t. instillation of 60 µl saline or saline plus 175 mM HOAc on day 2, the mice were sacrificed on day 3. For in vivo treatments with energy metabolic regulators, OVA-sensitized mice were challenged on day 0 and day 1, and on day 2 and day 3 received i.p. injection of 200 µl vehicle (saline plus 5% DMSO) or vehicle plus 48nM GSK2837808A, 10 µM CB-839, 10mM dichloroacetate, or 1 mM C75, mice were sacrificed 1 day (on day 5) after the second treatments. All energy metabolic regulators and Mito Tracker Green were purchased from MedChemExpress (Monmouth Junction, NJ).

Statistical analyses

Data of the numbers or percentages of cell populations, or mean fluorescence intensities (MFI) are presented as Mean ± standard deviation. The Student’s t-test was used to determine the statistical significance of the differences between experimental groups. A p value < 0.05 is considered statistically significant. The p values or significance levels (*: p < 0.05; **: p < 0.01) are shown in the figures.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1 (2.2MB, docx)

Acknowledgements

WZ thanks Doug Green for reading the manuscript and constructive comments.

Author contributions

WZ conceived, designed and carried out the study, and wrote the manuscript. SY and BZ assisted with flow cytometry data acquisition.

Funding

No external fund was used.

Data availability

All data are presented in the main text and supplemental materials.

Declarations

Competing interests

The corresponding author declares financial and non-financial interests. Other authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Malhotra, D., Burrack, K. S., Jenkins, M. K. & Frosch, A. E. Antigen-Specific CD4(+) T cells exhibit distinct kinetic and phenotypic patterns during primary and secondary responses to infection. Front. Immunol.11, 2125 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Wang, T., Marquardt, C. & Foker, J. Aerobic Glycolysis during lymphocyte proliferation. Nature261, 702–705 (1976). [DOI] [PubMed] [Google Scholar]
  • 3.Bissell, M. J., Hatie, C. & Rubin, H. Patterns of glucose metabolism in normal and virus-transformed chick cells in tissue culture. J. Natl. Cancer Inst.49, 555–565 (1972). [PubMed] [Google Scholar]
  • 4.Munyon, W. H. & Merchant, D. J. The relation between glucose utilization, lactic acid production and utilization and the growth cycle of L strain fibroblasts. Exp. Cell. Res.17, 490–498 (1959). [DOI] [PubMed] [Google Scholar]
  • 5.Vander Heiden, M. G., Cantley, L. C. & Thompson, C. B. Understanding the Warburg effect: The metabolic requirements of cell proliferation. Science324, 1029–1033 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Warburg, O. Versuche an überlebendem carcinomgewebe. Biochem. Zschr. 142, 317–333 (1923). [Google Scholar]
  • 7.Warburg, O. On the origin of cancer cells. Science123, 309–314 (1956). [DOI] [PubMed] [Google Scholar]
  • 8.Vaupel, P. & Multhoff, G. Revisiting the Warburg effect: Historical dogma versus current Understanding. J. Physiol.599, 1745–1757 (2021). [DOI] [PubMed] [Google Scholar]
  • 9.Zu, X. L. & Guppy, M. Cancer metabolism: Facts, fantasy, and fiction. Biochem. Biophys. Res. Commun.313, 459–465 (2004). [DOI] [PubMed] [Google Scholar]
  • 10.Corbet, C. & Feron, O. Tumour acidosis: From the passenger to the driver’s seat. Nat. Rev. Cancer. 17, 577–593 (2017). [DOI] [PubMed] [Google Scholar]
  • 11.Estrella, V. et al. Acidity generated by the tumor microenvironment drives local invasion. Cancer Res.73, 1524–1535 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Dubos, R. J. The micro-environment of inflammation or Metchnikoff revisited. Lancet269, 1–5 (1955). [DOI] [PubMed] [Google Scholar]
  • 13.Behnen, M., Moller, S., Brozek, A., Klinger, M. & Laskay, T. Extracellular acidification inhibits the ROS-Dependent formation of neutrophil extracellular traps. Front. Immunol.8, 184 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Colegio, O. R. et al. Functional polarization of tumour-associated macrophages by tumour-derived lactic acid. Nature513, 559–563 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Fischer, B., Muller, B., Fisch, P. & Kreutz, W. An acidic microenvironment inhibits antitumoral non-major histocompatibility complex-restricted cytotoxicity: Implications for cancer immunotherapy. J. Immunother. 23, 196–207 (2000). [DOI] [PubMed] [Google Scholar]
  • 16.Gabig, T. G., Bearman, S. I. & Babior, B. M. Effects of oxygen tension and pH on the respiratory burst of human neutrophils. Blood53, 1133–1139 (1979). [PubMed] [Google Scholar]
  • 17.Bosticardo, M. et al. Biased activation of human T lymphocytes due to low extracellular pH is antagonized by B7/CD28 costimulation. Eur. J. Immunol.31, 2829–2838 (2001). [DOI] [PubMed] [Google Scholar]
  • 18.Mookerjee, S. A., Gerencser, A. A., Nicholls, D. G. & Brand, M. D. Quantifying intracellular rates of glycolytic and oxidative ATP production and consumption using extracellular flux measurements. J. Biol. Chem.292, 7189–7207 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Braha, A. et al. Nutritional interventions to improve cachexia outcomes in cancer-a systematic review. Medicina (Kaunas)58, 966 (2022). [DOI] [PMC free article] [PubMed]
  • 20.DeBerardinis, R. J. et al. Beyond aerobic glycolysis: Transformed cells can engage in glutamine metabolism that exceeds the requirement for protein and nucleotide synthesis. Proc. Natl. Acad. Sci. U S A. 104, 19345–19350 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Pavlova, N. N. & Thompson, C. B. The emerging hallmarks of cancer metabolism. Cell. Metab.23, 27–47 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Ghanem, N. et al. The Pentose phosphate pathway in cancer: Regulation and therapeutic opportunities. Chemotherapy66, 179–191 (2021). [DOI] [PubMed] [Google Scholar]
  • 23.Bailis, W. et al. Distinct modes of mitochondrial metabolism uncouple T cell differentiation and function. Nature571, 403–407 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Darzynkiewicz, Z. et al. Initiation and termination of DNA replication during S phase in relation to cyclins D1, E and A, p21WAF1, Cdt1 and the p12 subunit of DNA polymerase delta revealed in individual cells by cytometry. Oncotarget6, 11735–11750 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Takahashi, S. & Patrick, G. Patterns of lymphatic drainage to individual thoracic and cervical lymph nodes in the rat. Lab. Anim.21, 31–34 (1987). [DOI] [PubMed] [Google Scholar]
  • 26.Billiard, J. et al. Quinoline 3-sulfonamides inhibit lactate dehydrogenase A and reverse aerobic Glycolysis in cancer cells. Cancer Metab.1, 19 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Gross, M. I. et al. Antitumor activity of the glutaminase inhibitor CB-839 in triple-negative breast cancer. Mol. Cancer Ther.13, 890–901 (2014). [DOI] [PubMed] [Google Scholar]
  • 28.Stakisaitis, D. et al. The importance of gender-related anticancer research on mitochondrial regulator sodium dichloroacetate in preclinical studies in vivo. Cancers (Basel)11, 1210 (2019). [DOI] [PMC free article] [PubMed]
  • 29.Idrovo, J. P., Yang, W. L., Nicastro, J., Coppa, G. F. & Wang, P. Stimulation of carnitine palmitoyltransferase 1 improves renal function and attenuates tissue damage after ischemia/reperfusion. J. Surg. Res.177, 157–164 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Su, W. et al. Restoration of CPT1A-mediated fatty acid oxidation in mesothelial cells protects against peritoneal fibrosis. Theranostics13, 4482–4496 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Xue, Y. et al. Inhibition of carnitine palmitoyltransferase 1A aggravates fatty liver graft injury via promoting mitochondrial permeability transition. Transplantation105, 550–560 (2021). [DOI] [PubMed] [Google Scholar]
  • 32.Biolegend www.biolegend.com/ja-jp/products/mitospy-orange-cmtmros-12370
  • 33.Desai, S. et al. Performance of TMRM and mitotrackers in mitochondrial morphofunctional analysis of primary human skin fibroblasts. Biochim. Biophys. Acta Bioenerg.1865, 149027 (2024). [DOI] [PubMed] [Google Scholar]
  • 34.Bangsbo, J., Johansen, L., Graham, T. & Saltin, B. Lactate and H + effluxes from human skeletal muscles during intense, dynamic exercise. J. Physiol.462, 115–133 (1993). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Orringer, C. E., Eustace, J. C., Wunsch, C. D. & Gardner, L. B. Natural history of lactic acidosis after grand-mal seizures. A model for the study of an anion-gap acidosis not associated with hyperkalemia. N Engl. J. Med.297, 796–799 (1977). [DOI] [PubMed] [Google Scholar]
  • 36.DeBerardinis, R. J. & Cheng, T. Q’s next: The diverse functions of glutamine in metabolism, cell biology and cancer. Oncogene29, 313–324 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Parveen, S. et al. Glutamine metabolism Inhibition has dual Immunomodulatory and antibacterial activities against Mycobacterium tuberculosis. Nat. Commun.14, 7427 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Pasquali, C. C. et al. The origin and evolution of human glutaminases and their atypical C-terminal Ankyrin repeats. J. Biol. Chem.292, 11572–11585 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Pfeiffer, T. & Morley, A. An evolutionary perspective on the Crabtree effect. Front. Mol. Biosci.1, 17 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Vemuri, G. N., Altman, E., Sangurdekar, D. P., Khodursky, A. B. & Eiteman, M. A. Overflow metabolism in Escherichia coli during steady-state growth: Transcriptional regulation and effect of the redox ratio. Appl. Environ. Microbiol.72, 3653–3661 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Barry, M. A. & Eastman, A. Endonuclease activation during apoptosis: The role of cytosolic Ca2 + and pH. Biochem. Biophys. Res. Commun.186, 782–789 (1992). [DOI] [PubMed] [Google Scholar]
  • 42.Gottlieb, R. A., Nordberg, J., Skowronski, E. & Babior, B. M. Apoptosis induced in Jurkat cells by several agents is preceded by intracellular acidification. Proc. Natl. Acad. Sci. U S A. 93, 654–658 (1996). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Liu, D. et al. Caspase-8-mediated intracellular acidification precedes mitochondrial dysfunction in somatostatin-induced apoptosis. J. Biol. Chem.275, 9244–9250 (2000). [DOI] [PubMed] [Google Scholar]
  • 44.Matsuyama, S., Llopis, J., Deveraux, Q. L., Tsien, R. Y. & Reed, J. C. Changes in intramitochondrial and cytosolic pH: Early events that modulate caspase activation during apoptosis. Nat. Cell. Biol.2, 318–325 (2000). [DOI] [PubMed] [Google Scholar]
  • 45.National Research Council. Guide for the care and use of laboratory animals, 8th ed. (The national academies press, Washington, D.C., 2011).
  • 46.Rangasamy, T. et al. Disruption of Nrf2 enhances susceptibility to severe airway inflammation and asthma in mice. J. Exp. Med.202, 47–59 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Zeng, W. P. et al. Staphylococcal enterotoxin A-activated regulatory T cells promote allergen-specific TH2 response to intratracheal allergen inoculation. J. Allergy Clin. Immunol.139, 508–518e504 (2017). [DOI] [PubMed] [Google Scholar]
  • 48.du Percie, N. et al. The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. PLoS Biol.18, e3000410 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1 (2.2MB, docx)

Data Availability Statement

All data are presented in the main text and supplemental materials.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES