Single-cell mapping of metabolic dependences using ATP as a biomarker

doi:10.21203/rs.3.rs-4836421/v1

Download PDF

Article

Single-cell mapping of metabolic dependences using ATP as a biomarker

https://doi.org/10.21203/rs.3.rs-4836421/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Metabolic dependences highlight a cell's reliance on specific pathways to meet its bioenergetic needs, with these pathways being interrogated using chemical inhibitors to assess their significance. While surrogate markers of bioenergetics (e.g., oxygen consumption) have yielded important insights, we asked whether metabolic dependences could be defined using ATP as a biomarker. To address this gap, we developed Mitochondrial/Energy Flow Cytometry (MitE-Flo), a method that evaluates the contributions of glycolysis, fatty acid oxidation (FAO), and oxidative phosphorylation (OXPHOS) to cellular ATP content. In models of mitochondrial disease due to complex I or complex IV deficiency, we identified impaired OXPHOS with a compensatory shift to glycolysis. To define the utility of ATP monitoring in immunometabolism research, we analyzed previously inaccessible cell populations: light zone (LZ) and dark zone (DZ) germinal center (GC) B cells. Highly proliferative DZ B cells exhibited elevated ATP levels and a preference for FAO and OXPHOS over glycolysis, with uniform increased activity across ETC complexes. In contrast, less proliferative LZ B cells showed lower ATP levels and an equal reliance on glycolysis and OXPHOS. Using ATP as a biomarker to define metabolic dependences provides valuable insights into disease states and elusive immune cell subtypes, thereby enhancing the metabolism research toolkit.

Biological sciences/Biochemistry

Biological sciences/Immunology

Biological sciences/Physiology

Metabolic dependences underscore the reliance on specific pathways used by cells to fulfill their bioenergetic requirements. Interrogation of metabolic dependences involves employing chemical inhibitors to evaluate the importance of select pathways. Over the past decade, technological advancements in high-resolution respirometry (i.e., Oroboros), extracellular flux analysis (i.e., Seahorse), and Single Cell ENergetIc metabolism by profilIng Translation inhibition (SCENITH) ¹, have enhanced our capabilities for profiling glycolysis, fatty acid oxidation (FAO) and oxidative phosphorylation (OXPHOS) ². Although these technologies provide an informative surrogate metric of bioenergetics, they oftentimes do not directly correlate with ATP production ³. Advancing the field of metabolism necessitates a platform that determines metabolic dependences using ATP as a biomarker.

Based on this need, we asked whether ATP could be used as an indicator for determining metabolic dependences. To address this question, we developed Mitochondrial/Energy Flow Cytometry (MitE-Flo), a method that leverages a fluorescent ATP sensor with targeted inhibitors and the precision of flow cytometry. MitE-Flo enables bioenergetic profiling of disease states, such as mitochondrial disease, and previously inaccessible cell types, such as light and dark zone germinal center B cells, providing insights into metabolic dependences. By quantifying ATP directly and profiling cellular identities and activation states, MitE-Flo offers a complementary tool for the comprehensive evaluation of bioenergetics.

Cells with distinct bioenergetic capacities display varying ATP levels

Cells that are metabolically active generally exhibit higher levels of ATP due to their increased bioenergetic demands. The terminal phosphate group of ATP carries a highly negative charge which can be mobilized via various chemical reactions within the cell, providing energy that allows cellular work to be performed ⁴. To profile ATP production in cells with varying bioenergetic capacities, we utilized ATP-Red, a fluorescent probe that specifically targets ATP, shows a strong positive correlation with ATP concentration, and does not cross-react with other metabolites ⁵. In the absence of ATP, the probe forms a closed ring structure. When ATP is present, the covalent bonds between boron and ribose are broken by the negative charge, opening the ring, producing fluorescence (Fig. 1A). Based on these properties, we hypothesized that profiling changes in cellular ATP content could help determine metabolic dependences by flow cytometry. We began by evaluating the optimal concentration of ATP-Red for flow cytometry by serial dilution (1.25 µM to 1.10 µM) in Madin-Darby canine kidney cells (MDCK cells, 0.2x10⁶ cells/well). A dose response curve was observed where higher concentrations of ATP-Red (5 and 10 µM) correlated with stronger fluorescence intensities (Fig. 1B). Cell viability was unaffected by ATP-Red concentration (Figure S1A). Next, we sought to determine the optimal timing for incubation. We observed that all timepoints studied displayed significantly robust fluorescence intensities relative to control (Fig. 1C); however, timepoints greater than 20 minutes affected cell viability (Figure S1B). Based on our results, we chose the following conditions for subsequent experiments: ATP-Red (10 µM) incubated for 20 minutes.

We next asked whether our experimental conditions could detect differences in cellular ATP content between cell lines with differing metabolic activity or substrate availability. Compared to unlabeled cells, we observed a significant shift in ATP-Red fluorescence in MDCK cells (canine kidney cells, (Figure S1C)), unstimulated splenic CD4⁺ and CD8⁺ T cells (murine, Figures S1D-E) and lymphoblastoid cell lines human, (Figure S1F) to varying degrees. For ATP-Red to be useful in profiling metabolic dependences in bioenergetics, it must be able to respond to changes in cellular metabolism, i.e., fluctuations in ATP concentration. To answer this question, we first applied glucose at increasing concentrations 30 minutes before staining with ATP-Red. We observed that glucose supplementation (100 mM and 250 mM) augmented ATP production (Fig. 1D) with normal viability (Figure S1G).

After demonstrating ATP concentrations in supplemented states, we next assessed the dynamics of ATP production in activated cells. T cells undergo metabolic reprogramming following activation which includes a marked increase in glycolysis, OXPHOS, and ATP production ⁶. To examine changes in ATP content due to cellular activation, we stimulated mouse splenic T cells with anti-CD3/CD28 for 72 hours. Flow cytometric analysis with ATP-Red displayed increases in ATP concentration of CD4⁺ (Fig. 1E) and CD8⁺ (Fig. 1G) T cells along with markers of activation (Figs. 1F and 1I).

Differential glycolytic dependence revealed through ATP

As a major energy generating pathway in intermediary metabolism, glycolysis has evolved in nearly all organisms, with glucose being the most common source of cellular energy and a substrate for many biochemical processes ^{7, 8}. By inhibiting hexokinase and glucose-6-phosphate isomerase, ultimately impairing glycolytic flux, 2-deoxyglucose (2-DG) is an important tool for studying substrate level phosphorylation and glycolytic flux (Fig. 2A) ⁹. To determine the utility of ATP-Red in defining glycolytic dependence, we evaluated ATP production in the presence of 2-DG in MDCK cells, Jurkat cells, and lymphoblastoid cell lines (LCLs). These cells were chosen due to their distinct origins, roles, and metabolic demands. To ascertain the optimal concentration of 2-DG for our experiments, we evaluated cell viability across a gradient of doses. The concentration of 100 mM was identified as the ideal dosage, striking a balance between effective inhibition and maintaining cell viability (Figures S2A-S2F). Following 2-DG treatment, MDCK cells (0.2x10⁶ cells/well) showed a notable decrease in ATP fluorescence, indicative of a glycolytic dependence of 68% to total ATP synthesis (Fig. 2B). Similar to MDCK cells, ATP fluorescence in LCLs (0.2x10⁶ cells/well) and Jurkat cells (0.2x10⁶ cells/well) was also decreased (Fig. 2C and 2D), albeit to lesser extent, indicating glycolytic dependences of 28% and 27%, respectively. Our findings that in MDCK cells glycolysis accounts for a larger proportion of cellular ATP production when contrasted with LCL and Jurkat cells.

To corroborate glycolytic suppression by 2-DG, we performed extracellular flux analysis on MDCK cells (0.4x10⁴ cells/well), where we replicated our flow cytometry protocol timing: 250 mM glucose was injected into the well at 20 minutes followed by 100 mM 2-DG 20 minutes later. With this method, MDCK cells experienced a 51% decrease in extracellular acidification (Fig. 2E), consistent with our observations using ATP. Additionally, we determined the ATP concentration with a colorimetric method in MDCK cells treated with 100 mM 2-DG. ATP concentration was reduced by 61% compared to the untreated control (Fig. 2F) substantiating our results obtained with ATP-Red (Figs. 2B-2D).

Differential FAO dependence revealed through ATP

Beta-oxidation of mitochondrial fatty acids (i.e., FAO) leads to the production of acetyl-CoA for the tricarboxylic acid (TCA) cycle and reducing equivalents for the respiratory chain. Both can lead to the production of ATP through OXPHOS (Fig. 3A). Carnitine palmitoyl transferase 1 (CPT1) is an essential component of the carnitine shuttle, the mechanism by which long chain fatty acids, a primary fuel, enter the mitochondria for FAO. Etomoxir is a CPT1 inhibitor, used to assess the role of fatty acid oxidation in cellular metabolism (Fig. 3A). In MDCK, LCLs and Jurkat cells, the optimal dose for FAO inhibition was determined by viability as above, yielding an optimal dose of 100 µM (Figures S3A-S3F). To understand the contributions of FAO to ATP production, we evaluated the effect of etomoxir on ATP fluorescence in MDCK cells (0.2x10⁶ cells per condition). The observed decrease in fluorescence following etomoxir treatment (Fig. 3B) suggested an FAO dependence of 39% for ATP production. To compare FAO dependence in cell types with distinctive metabolic profiles, we incubated LCLs and Jurkat cells with etomoxir as above. LCL cells also showed a reduction in ATP fluorescence in response to etomoxir (Fig. 3C), indicating and FAO dependence of 32%. Similar results were found for Jurkat cells (0.2x10⁶ cells/well), albeit to a lesser degree: FAO dependence of 24% (Fig. 3D). To corroborate the changes with etomoxir treatment, we studied palmitate oxidation by extracellular flux analysis. MDCK cells (0.4x10⁴/well) were treated with palmitate or BSA (mock control) for 20 minutes followed by etomoxir (100 µM) injection 20 minutes later. In MDCK cells, etomoxir resulted in a reduction in OCR (Fig. 3E), analogous to our ATP fluorescence readings in MDCK cells. Additionally, we determined the ATP concentration with a colorimetric method in MDCK cells treated with etomoxir (100 µM). Using this method, an FAO dependence of 38% was observed, nearly matching our ATP-Red studies (Fig. 3F).

Differential OXPHOS dependence revealed through ATP

Mitochondria are ubiquitous organelles within eukaryotic cells due to their essential function of supplying cellular energy in the form of ATP ¹⁰. Originally identified as an antibiotic produced by fungi, oligomycin is a potent inhibitor of mitochondrial ATP synthase (Fig. 4A), and useful in assessing OXPHOS dependence ¹¹. To profile OXPHOS dependence with ATP-Red, we treated various cell lines with oligomycin. Optimal dosing of oligomycin (10 µM) was determined as above (Figures S4A - S4F). We first evaluated the effect of oligomycin on ATP production in MDCK cells (0.2x10⁶ cells per condition). With oligomycin, we observed a reduction in ATP production (Fig. 4B), suggesting and OXPHOS dependence of 39%. In LCL cells (0.2x10⁶ cells/well) and Jurkat cells (0.2x10⁶ cells/well), oligomycin (10 µM) also decreased ATP fluorescence (Figs. 4C and 4D), indicating OXPHOS dependences of 35% and 43%, respectively. To corroborate our findings, we measured the OCR by extracellular flux analysis in MDCK cells (0.4x10⁴ cells/well). Following oligomycin treatment, OCR fell by 52% (Fig. 4E). Additionally, we determined the ATP concentration with a colorimetric method in MDCK cells treated with oligomycin (10 µM). Similar to our ATP-Red results, we observed a reduction in ATP concentration, indicating an OXHOS dependence of 46% (Fig. 4F).

Underpinning cellular energy production, the chemiosmotic hypothesis describes how ETC components generate a proton gradient to drive ATP synthesis ¹². To elucidate the specific metabolic dependences of individual ETC components, MDCK cells underwent targeted inhibition. Optimal concentrations of rotenone (RO, targeting NADH-ubiquinone oxidoreductase, CI), 3-nitropropionic acid (3-NP, targeting succinate-ubiquinone oxidoreductase, CII), antimycin A (AA, targeting ubiquinol–cytochrome c oxidoreductase, CIII), and sodium azide (AZ, targeting cytochrome c oxidase, CIV) were determined as described above (Figures S4G – S4O). Initially, the effect of RO (2 µM) on CI was assessed by measuring ATP fluorescence in MDCK cells (2 × 10⁵ cells per condition), revealing a CI dependence of 18.1% (Fig. 4G). Confirmation of CI inhibition was achieved by evaluating the OCR using extracellular flux analysis post-treatment (Figs. 4H and S4P). Subsequently, CII was investigated using 3-NP (400 µM), revealing a CII dependence of 16.6% (Fig. 4I). Similar to rotenone, treatment with 400 µM 3-NP led to a substantial reduction in OCR (Figs. 4J and S4Q). Assessment of CIII involved treating cells with antimycin A (2 µM), resulting in a CIII dependence of 24% (Fig. 4K). The associated decrease in OCR post-treatment paralleled the effects observed with the other inhibitors (Figs. 4L and S4R). Lastly, CIV was evaluated using sodium azide (6 mM), revealing a CIV dependence of 32% (Fig. 4M). OCR assessment confirmed a significant decrease, consistent with the observed impairment in ATP production (Figs. 4N and S4S). Collectively, our ETC complex dependence data underscore its integral role in maintaining mitochondrial ATP production, with complex-specific inhibitors inducing distinct reductions in cellular respiration and energy output.

Mathematical integration of metabolic dependences for comprehensive bioenergetic profiling (MitE-Flo)

Utilizing selective inhibitors targeting critical junctures in glycolysis, FAO, and OXPHOS (Fig. 5A), we incorporated our findings to establish bioenergetic profiles. The inhibitors' efficacy in reducing ATP production manifests as fluorescence shifts (i.e., diminished ATP fluorescence), offering insights into each pathway for ATP generation (Fig. 5B). By quantifying these shifts via flow cytometry, we ascertained the metabolic pathway dependences by calculating the geometric mean fluorescence intensity (MFI) (Fig. 5C). An integrated mathematical model (Fig. 5D) enabled us to quantify the metabolic dependence of each pathway for cellular ATP production. In MDCK cells, our analysis revealed glycolytic, FAO, and OXPHOS dependences of 42.8%, 25.9%, and 32%, respectively.

To incorporate ETC dependences into our understanding of overall OXPHOS functionality, we applied our MitE-Flo system in a manner akin to our earlier analysis of central metabolic pathways (Fig. 5E). The extent of inhibition caused by targeted inhibitors, leading to a fluorescence shift, informs us about the ETC complex dependences on ATP production within mitochondria (Fig. 5F). Based on these calculations (Fig. 5G), we determined the contributions of complex I (8.7%), II (6.9%), III (10.2%) and IV(10.4%), normalized to OXPHOS, to ATP production in MDCK cells.

MitE-Flo delineates metabolic dependences in mitochondrial disease

Cells affected by mitochondrial dysfunction exhibit metabolic adaptations to cope with impaired OXPHOS. For example, in a mouse model of mitochondrial encephalopathy, increased flux through glycolysis helps compensate for impaired OXPHOS ¹³. To determine the utility of MitE-Flo to determine the metabolic phenotype of a disease state marked by compromised bioenergetics, cells deficient in the nuclear genes Surfeit 1 (SURF1) or NADH:ubiquinone oxidoreductase subunit S4 (NDUFS4) were studied. SURF1 is an integral component of ETC complex IV (cytochrome c oxidase, COX) biogenesis. NDUFS4 is a nuclear-encoded accessory subunit of the mitochondrial membrane respiratory chain NADH dehydrogenase (complex I). Deficiencies in either of these genes produce OXPHOS dysfunction. We chose to edit MDCK cells specifically, based on the robust bioenergetic profiles displayed above. SURF1 KO and NDUFS4 KO cells demonstrated lower levels of ATP-Red consistent with OXPHOS deficiency (Fig. 6A). Glycolytic dependence was markedly elevated, exceeding that of wild-type cells; an allostatic adaptation facilitating ATP production (Figs. 6B and 6E). OXPHOS (Figs. 6C and 6E) and FAO (Figs. 6D and 6E) dependences were decreased as expected for mitochondrial disease. We further validated our findings by creating a bioenergetic plot (Fig. 6F), integrating glycolysis (ECAR) and OXPHOS (OCR) measurements from extracellular flux analysis. Echoing our MitE-Flo results, the SURF1 KO and NDUFS4 KO cells exhibited impaired OXPHOS with an increase in glycolysis. This glycolytic upsurge persisted after normalizing for ATP content (Fig. 6G).

We next studied the contributions of individual ETC components. Treatment with the drugs resulted in a slight viability reduction in SURF1 KO and NDUFS4 KO cells (Fig. 6H), likely due to their inherent mitochondrial vulnerability. Nonetheless, the viability levels were sufficient for subsequent ATP-Red concentration assays in live cells. Comparative analysis of the ETC dependence using the calculations above indicated a marked decrease in the activities of complexes C-I to C-IV in SURF1 KO and NDUFS4 KO cells (Figs. 6I − 6L). The greater depression in SURF1 KO was not surprising since COX is the final and rate limiting step of the respiratory chain ¹⁴. After adjusting for the overall OXPHOS dependence, the depression of CI to CIV activity in SURF1 KO and NDUFS4 KO became more evident (Fig. 6M). Extracellular flux analysis of MDCK cells also confirmed this overall depression of OXPHOS (Fig. 6N).

MitE-Flo delineates metabolic dependences in dark zone and light zone germinal center B cells

While MitE-Flo had proven effective in defining bioenergetic phenotypes in a disease state like mitochondrial disease, our next goal was to investigate small, previously inaccessible cell populations to characterize normal physiology. In the highly adaptive milieu of germinal centers (GC) where affinity maturation transpires, activated B cells undergo brisk proliferation with somatic hypermutation (SHM) in the dark zone (DZ, CXCR4⁺), followed by selection and class switch recombination within the light zone (LZ, CD23⁺) ^{15, 16}. Gene expression analyses and extracellular flux analysis have partially characterized the metabolic phenotypes of GC B cells, indicating a predilection for oxidative OXPHOS and FAO generally ^{16, 17, 18}. However, LZ and DZ GC B cells may exhibit distinct bioenergetic profiles, a phenomenon not yet documented in the biomedical literature. To define the utility of MitE-Flo in immunometabolic investigations, we sought to elucidate the metabolic dependences of DZ and LZ GC B cells; cells previously inaccessible to direct bioenergetic profiling. To promote GC formation in the spleen, C57BL/6J mice were immunized with sheep red blood cells (Fig. 7A). LZ and DZ GC B cells were identified utilizing a focused gating strategy (Figure S5A). The viability of the sorted cells was comparable (Fig. 7B), but DZ GC B cells prevailed in number (Fig. 7C), mirroring their proliferative vigor. Post-sorting of GC B cell fractions (Figure S5B), LZ GC B cells manifested amplified HIF-1α and HK2 mRNA levels (Figs. 7D and 7E) compared to DZ, suggesting a shift towards glycolysis ^{15, 17, 19}. Elevated HIF-1α protein levels in LZ GC B cells, as assessed in permeabilized cells by flow cytometry, further supported a glycolytic phenotype (Fig. 7F). ATP-Red analysis disclosed an enhanced ATP concentration within DZ GC B cells, indicative of increased metabolic demands (Fig. 7G). MitE-Flo, complemented with cell-specific markers, was then applied to delineate metabolic and ETC dependences for both DZ and LZ GC B cells. Corroborating our PCR findings, LZ GC B cells displayed equal dependences on glycolysis and OXPHOS whereas DZ GC B cells prioritized FAO and OXPHOS, after ATP level normalization (Figs. 7H − 7J and 7O). To conclude our examination of LZ and DZ cell populations, we assessed the energetic performance of mitochondrial complexes C-I to C-IV, determining their contributions to ATP production relative to their overall OXPHOS dependence (as depicted in Fig. 7H). By employing our equations, we determined the different ETC complex dependences. Our analysis revealed consistently higher ETC dependences across all complexes in DZ cells (Figs. 7K-N). After normalizing for relative ATP levels, we demonstrated an increase in FAO and OXPHOS dependences in DZ GC B cells (Fig. 7O). ETC dependence analysis, normalized to OXPHOS, uncovered nearly equal CI - CIV dependences in DZ GC B cells, suggesting an overall boost in ETC complexes (Fig. 7P). Taken as a whole, we have shown MitE-Flo to be an indispensable instrument for decoding the bioenergetic landscapes of GC B cells, a complex and unique cell population.

In this contribution, we have established that ATP can be used as a biomarker to determine metabolic dependences using MitE-Flo, a flow cytometry-based bioenergetics platform. Applying MitE-Flo to normal physiology as well as pathology, we probed the metabolic nuances of models of mitochondrial disease and germinal centers. NDUFS4 KO and SURF1 KO cells demonstrated allostatic adaptations in glycolytic dependence secondary to OXPHOS dysfunction. Concerning normal physiology in small populations, MitE-Flo illuminated the differential metabolic dependences of dark zone (DZ) and light zone (LZ) GC B cells, a finding previously limited to RNA sequencing and bulk analysis. Our results elucidate distinct metabolic strategies: LZ cells depend on glycolytic/OXPHOS dependence, while DZ cells depend on FAO/OXPHOS, underscoring the metabolic plasticity essential for GC B cell function.

The detailed analysis of ATP, a crucial indicator of cellular energy status, is integral for unraveling metabolic processes ²⁰. Employing ATP-Red for its specificity for ATP fluctuations allows for precise bioenergetic profiling ⁵. MitE-Flo leverages this specificity, enabling the direct measurement of intracellular ATP via flow cytometry, and thus facilitates the identification of cellular metabolic phenotypes. This is particularly revealing of the metabolic pathways, glycolysis, FAO, and OXPHOS, that are engaged to meet the bioenergetic demands, especially during the metabolic shifts from OXPHOS to glycolysis upon cell activation [30].

Insights into ATP dynamics are foundational for elucidating the concept of metabolic allostasis ²¹, a critical cellular process for maintaining energy balance that is necessary for cellular viability and stress response adaptation. It is in this pursuit that the concept of metabolic dependence is highlighted, where cells exhibit a preference for certain pathways to generate ATP. Utilizing metabolic inhibitors, we can discern these metabolic preferences, providing insight into how cells strive to maintain energy balance. In the context of our NDUFS4 and SURF1 KO cells, an increase in glycolytic dependence in response to OXPHOS deficiency indicates a shift within metabolic allostasis, revealing an adaptive strategy rather than a state of metabolic inflexibility. This observation hints at broader systemic changes that, under prolonged or excessive mitochondrial stress, could potentially lead to either a resilient or an enduringly altered metabolic phenotype ^{21, 22, 23}.

The glycolytic, FAO, and OXPHOS pathways contribute variably to cellular ATP production, shaped by the cell’s energy demands and specific context. Enhanced energy requirements typically prioritize glycolysis for its swift ATP yield, while FAO is more integral during stable energy conditions ²⁴. Despite the role of OXPHOS as a primary ATP source, certain diseases and activation states in cells, such as T cells, can shift the preference towards glycolytic dependence, even when oxygen is plentiful ^{20, 25}. The interplay among these metabolic pathways reflects a cell’s adaptability (i.e., allostasis), influenced by factors like energy needs, oxygen availability, and cell status; attributes that are also reflected in the metabolic nuances of GC B cells.

In the GC, B cells increase glucose uptake and mitochondrial biogenesis, indicative of an enhanced metabolic state ¹⁶. Previously, the metabolic nuances between LZ and DZ GC B cells were obscured when analyzed in bulk, which prevented detailed characterization of their distinct metabolic phenotypes. Utilizing MitE-Flo has not only allowed for the direct demonstration of the metabolic differences between these GC B cell subtypes, but also revealed unexpected aspects of metabolic diversity, expanding our understanding of immune cell metabolism. Traditionally, it has been held that proliferating cells primarily exhibit a dependence on glycolysis for energy. Contrary to this view, DZ GC B cells, though highly proliferative, rely more on FAO and OXPHOS. In comparison, LZ GC B cells, which proliferate to a lesser extent, show a glycolytic dependence. This behavior of GC B cells calls into question the established belief that glycolysis is the dominant metabolic pathway in all rapidly dividing lymphocytes.

The development of MitE-Flo for metabolic phenotyping represents a significant advancement in the field of cellular metabolism, providing a refined lens for examining the nuances of cellular bioenergetics. Its capacity to assess glycolytic, FAO and OXPHOS/ETC dependences offers a powerful method for characterizing metabolic states, which could be transformative in disease staging. In inflammatory diseases or cancer, where metabolic shifts often correlate with disease activity, MitE-Flo could be utilized to monitor these changes, potentially serving as a biomarker for disease progression or response to therapy ^{26, 27, 28, 29, 30}. The precise data gathered by MitE-Flo on mitochondrial health and function could also aid in diagnostics for a range of diseases characterized by altered bioenergetics, including mitochondrial diseases. Further research and validation are required to confirm its efficacy and reliability in these contexts, which could pave the way for more targeted and effective treatments, informed by detailed bioeneregtic profiles.

In summary, MitE-Flo combines bioenergetic profiling with the granularity of flow cytometry. This platform is robust and allows for the study of disease states and rare cell populations with minimal disturbance. The protocol is quick, inexpensive, and widely accessible due to the availability of reagents and ubiquitous nature of flow cytometry at most research institutions. Most importantly, MitE-Flo selectively and directly measures ATP, the primary energy currency of the cell and biomarker of cellular bioenergetic status.

Cell culture: Madin Darby canine kidney cells (MDCK)

Cells were routinely subcultured every 3 days (72 hours ± 6 hours) at a seeding density of 1x10⁵ cells/cm² or 5x10⁴ cells/cm² (DMEM + 10% FBS + Penicillin-Streptomycin (100 U/mL) in 75 cm² vent-cap T-flasks. The total volume of media was 20 mL per flask. Cultures were maintained at 37° C ± 1 C in a humidified incubator with 5% CO₂. Cell counts were determined by standard counting technique using an automatic cell counter (1:2 dilution in 0.04% trypan blue). All cultures were maintained using aseptic technique. At the end of the culture, the cells were detached with 2.0 to 3.0 mL of trypsin-EDTA 0.05% solution and incubated for 10 minutes at 37 °C ± 1 °C in a humidified incubator with 5% CO₂. The cell suspension was washed with 1X PBS and centrifuged at 1250 RPM for 5 minutes. The cell pellet was resuspended in Optimem media where they were counted using an automatic cell counter (1:2 dilution in 0.04% trypan blue). A concentration of 0.2x10⁶ cells/tube was used for the respective analyses and drug treatments.

Cell culture: Immortalized line of human T lymphocyte cells (Jurkat)

Cells were routinely subcultured every 2 and 3 days (48–72 hours). Cultures were maintained at a cell concentration between 1x10⁵ and 1x10⁶ viable cells/mL (RPMI-1640 Medium + 10% FBS + Penicillin-Streptomycin (100 U/mL)) in 25 cm² or 75 cm² culture flasks. Cultures were maintained at 37 °C ± 1 °C in a humidified incubator with 5% CO₂. Cell counts were determined by standard counting technique using an automatic cell counter (1:2 dilution in 0.04% trypan blue). All cultures were maintained using aseptic technique. At the end of the culture, the cells were collected, and the cell suspension was washed with 1X PBS and centrifuged at 1250 RPM for 5 minutes, the cell pellet was resuspended in Optimem media where they were counted using an automatic cell counter (1:2 dilution in 0.04% trypan blue). A concentration of 0.2x10⁶ cells/tube was used for the respective analyses.

Cell culture: Lymphoblastoid cell lines (LCL)

Cells were routinely subcultured every 3 days (72 hours). Cultures were maintained at a cell concentration between 1x10⁵ and 1x10⁶ viable cells/mL (RPMI-1640 Medium + 10% FBS + sodium pyruvate 1 mM + Glutamine 2mM + MEM NEAA 1x + Penicillin-Streptomycin (100 U/mL)) in 25 cm² or a 75 cm² culture flasks. Cultures were maintained at 37 °C ± 1 °C in a humidified incubator with 5% CO₂. Cell counts were determined by standard counting technique using an automatic cell counter (1:2 dilution in 0.04% trypan blue). All cultures were maintained using aseptic technique. At the end of culture, the cells were collected, and the cell suspension was washed with 1X PBS and centrifuged at 1000 RPM for 5 minutes, the cell pellet was resuspended in Optimem where they were counted using an automatic cell counter (1:2 dilution in 0.04% trypan blue). A concentration of 0.2x10⁶ cells/tube was used for the respective analyses.

Measurement of cell respiration with Seahorse flux analyzer

Extracellular flux analysis (Seahorse, Agilent, Santa Clara, CA.) can determine the rate of mitochondrial oxidative phosphorylation, through measurement of the oxygen consumption rate (OCR) measured in (pmol/min/mg/protein), in a 96 microplate-based assay platform. For this methodology, basal OCR was performed on MDCK cells (4x10⁵ cells/well) and obtained by the OCR of the last reading before adding the Oligomycin 10 µM. The assay was performed in XF Assay Modified DMEM. Three consecutive measurements were performed under basal conditions and after the sequential addition of the following ATP synthase inhibitor.

Measurement of extra cellular acidification rate (ECAR) with Seahorse flux analyzer

The extracellular acidification rate (ECAR) was determined in MDCK cells (4x10⁵ cells/well) suing extracellular flux analysis as above and measured in (mpH/min/mg/protein). This was obtained by taking the ECAR of the last reading before adding 250 mM glucose as an inducer of ATP production and increase in extracellular acidification. ECAR was then taken in the last measurement after the addition of 2-DG 100 mM to obtain the inhibitory effect of the drug. The assay was performed in XF Assay Modified DMEM. Three consecutive measurements were performed under basal conditions and after the sequential addition of the following electron transport chain inhibitors.

β-oxidation of palmitate

Palmitate oxidation was performed on MDCK cells measuring the OCR. MDCK cells (4x10⁵ cells/well) were incubated overnight. OCR were determined while using the XF Cell Palmitate-BSA FAO Substrate according to the manufacturer’s instructions.

Mitochondrial complex activity assay

The mitochondrial complex activity assay was conducted in MDCK cells. Briefly, MDCK WT and SURF1 KO cells (4x105 cells/well) were incubated overnight. Oxygen consumption rates were assessed following the protocol outlined by ³¹.

Flow cytometer and cell sorting

Flow cytometric analysis was conducted using the CytoFLEX LX flow cytometer (Beckman Coulter) equipped with suitable laser configurations. Cells were labeled with a panel of fluorescent conjugated antibodies targeting specific mouse antigens, including anti-mouse CD4, anti-mouse CD8, anti-mouse CD23, anti-mouse CXCR4, anti-mouse B220/CD45, anti-mouse HIF-1α, and anti-mouse CD19. Additionally, a LIVE/DEAD™ Fixable Near IR (876) Viability Kit (Thermofisher #L34981) was utilized to discriminate between viable and dead cells and ATP sensor for ATP measurement. Fluorescence data were acquired and analyzed using appropriate software FlowJo™ version 10, enabling the characterization of distinct cell populations based on their antigen expression profiles. The BD FACSDiva 8.0.2 flow cytometer was utilized for cell sorting purposes. Cells of interest were prepared and labeled with appropriate reagents, including anti-CD23 and anti-CXCR4 antibodies, to selectively target the light and dark zone populations, respectively. Subsequently, fluorescence data were acquired and analyzed using BD FACSDiva 8.0.2 software, with gating strategies (Figure S5B) applied to identify and separate the target cellular subpopulations based on their expression of CD23 and CXCR4. Sorted cells were then collected for downstream analyses. This approach facilitated precise characterization of cellular subpopulations and proved effective for flow cytometry studies. For mouse HIF-1α protein labeling, permeabilization was performed with 100% cold methanol for 5 minutes. Subsequently, the cells were washed with 1X PBS, FBS and a spin for 5 minutes at 1250 rpm to subsequently carry out the labeling with anti-mouse HIF-1α. mention that due to the limitation of the MitE-Flo protocol in terms of membrane permeabilization, both the mouse HIF-1α protein labeling and the metabolic analysis by MitE-Flo were performed separately.

Realtime PCR for metabolic genes

RNA extraction was performed in splenic DZ GC B (CXCR4⁺) and LZ GC B (CD23⁺) cells. Total RNA was extracted from cultured cells using the RNeasy Mini Kit (Qiagen) according to the manufacturer's instructions. RNA concentration and purity were assessed using a Synergy Biotek spectrophotometer (Agilent). First-strand cDNA synthesis was performed using the iScript™ cDNA Synthesis Kit (Biorad) according to the manufacturer's protocol. Briefly, 1 µg of total RNA was reverse transcribed in a final 20 µL reaction mixture. The reaction was incubated at 25° C for 5 min, followed by 46° C for 20 min and 95° C for 1 min to inactivate the reverse transcriptase enzyme. Gene expression levels were quantified by real-time PCR using the CFX96 Touch Real-Time PCR Detection System (Biorad). TaqMan gene expression assays, mouse HK2 Mm00443385_m1 and mouse HIF-1α Mm00468869_m1 (Thermofisher) were used to quantify the expression of the target genes, HIF-1α and HK2 and the endogenous control gene mouse β-actin Mm01205647_g1 (Thermofisher). Each reaction mixture contained 10 µL of TaqMan Universal PCR Master Mix, 1 µL of TaqMan Gene Expression Assay (20×), 4 µL of cDNA template, and nuclease-free water to a final volume of 20 µL. Thermal cycling conditions consisted of an initial denaturation step at 95° C for 10 min, followed by 40 cycles of denaturation at 95° C for 15 s and annealing/extension at 60° C for 1 min. Relative gene expression levels were calculated using the comparative Ct (2^−ΔΔCt) method. Data were analyzed using CFX Maestro Software 2.3 for Windows (Biorad).

Colorimetric ATP

MDCK cells were divided into 4 experimental conditions and treated with the different inhibitory drugs oligomycin (10 µM), etomoxir (100 µM), 100 mM 2-DG and untreated control cells, to which the extraction was performed. of ATP using the recommendations of the kit and the enzymatic reaction was carried out according to the instructions of the kit (ATP kit MAK190 SIGMA), where we later measured the absorbance at a wavelength of 570 nm. We calculated ATP concentrations using a standard curve provided by the manufacturer.

Drugs stock

The drugs (Oligomycin 5 mM sigma), (Etomoxir 20 mM sigma), (rotenone 5 mM sigma), and (antimycin A 3 mM sigma), were resuspended in 100% ethanol, (2-deoxy-D-glucose, 1 M sigma), ( 3-NP 1 M sigma) and (sodium azide 5 M sigma) in Optimem media, for stock generation and were stored in aliquots at -20°C until use according to the manufacturer’s instructions.

Metabolic profile by flow cytometry (MitE-Flo)

For the cellular metabolic profile analysis, we used MDCK, LCL, Jurkat, and primary mouse cells at a concentration of 0.2x10⁶ cells/well in Optimem media without fetal bovine serum. The cells were divided into 5 different wells at this same concentration and to which inhibitory drugs were added individually (10 µM oligomycin, 100 mM 2-DG, 100 µM etomoxir) panel A and (2 µM rotenone, 400 µM 3-NP, 2 µM antimycin A and 6 mM sodium azide) panel B, plus one well where the cells received the addition of all drugs together at the same final concentration. The cells were incubated at 37 °C for 30 minutes, then a final 10 µM ATP-Red concentration was added plus an LIVE/DEAD™ Fixable aqua or LIVE/DEAD™ Fixable Near IR (876) Viability Kit probe in a ratio of 1/1000. In the same step cells were labeled with a panel of fluorescent conjugated antibodies targeting specific mouse antigens, including anti-mouse CD4, anti-mouse CD8, anti-mouse CD23, anti-mouse CXCR4, anti-mouse B220/CD45, anti-mouse HIF-1α, and anti-mouse CD19. in samples from mouse spleens. In the presence of drug treatments, the cells were incubated at 37 °C for 20 minutes. After incubation the cells were washed with 1X PBS and centrifuged at 1250 RPM for 5 minutes and resuspended in 1% PFA and subsequently analyzed by flow cytometry.

Data analysis

All values are presented as mean and standard error of the mean (SEM) of a number of independent experiments (> 3). As most of our datasets did not display normality, non-parametric tests were applied Mann-Whitney. Comparisons between more than two paired data sets were made using the ordinary one-way ANOVA with a Multiple Comparison Test. For all statistical comparisons we used a Prism 10.2 software, where a p value < 0.05 was considered significant. All illustrations were created with BioRender.com.

Acknowledgments: The authors would like to acknowledge the NHGRI animal facility and the Intramural Research Program of the NIH.

Funding: National Institutes of Health Intramural Research Program HG200381-03

Author contributions:

Conceptualization: JM, PJM

Methodology: JM, PJM

Investigation: JM, ALF, TNT, MK, SA, EMGL, SK, PJM

Supervision: PJM

Writing – original draft: JM, PJM

Writing – review & editing: JM, ALF, TNT, EW, EMGL, APW, PJM

Competing interests: Authors declare that they have no competing interests.

Data and materials availability: All data are available in the main text, supplementary materials, or by reasonable request from the authors.

Arguello RJ et al (2020) A Flow Cytometry-Based Method to Functionally Profile Energy Metabolism with Single-Cell Resolution. Cell Metab 32:1063–1075e1067
Horan MP, Pichaud N, Ballard JW (2012) Review: quantifying mitochondrial dysfunction in complex diseases of aging. J Gerontol Biol Sci Med Sci 67:1022–1035
Brand MD, Nicholls DG (2011) Assessing mitochondrial dysfunction in cells. Biochem J 435:297–312
Meurer F, Do HT, Sadowski G, Held C (2017) Standard Gibbs energy of metabolic reactions: II. Glucose-6-phosphatase reaction and ATP hydrolysis. Biophys Chem 223:30–38
Wang L et al (2016) A Multisite-Binding Switchable Fluorescent Probe for Monitoring Mitochondrial ATP Level Fluctuation in Live Cells. Angew Chem Int Ed Engl 55:1773–1776
Patsoukis N et al (2015) PD-1 alters T-cell metabolic reprogramming by inhibiting glycolysis and promoting lipolysis and fatty acid oxidation. Nat Commun 6:6692
Tirone TA, Brunicardi FC (2001) Overview of glucose regulation. World J Surg 25:461–467
Akram M (2013) Mini-review on glycolysis and cancer. J Cancer Educ 28:454–457
Pajak B et al (2019) 2-Deoxy-d-Glucose and Its Analogs: From Diagnostic to Therapeutic Agents. Int J Mol Sci 21
Ghezzi D, Zeviani M (2018) Human diseases associated with defects in assembly of OXPHOS complexes. Essays Biochem 62:271–286
H A LARDY DJ, McMURRAY WC (1958) Antibiotics as tools for metabolic studies. I. A survey of toxic antibiotics in respiratory, phosphorylative and glycolytic systems. Arch Biochem Biophys Dec 78(2):587–597
Mitchell P (1961) Coupling of phosphorylation to electron and hydrogen transfer by a chemi-osmotic type of mechanism. Nature 191:144–148
Garcia S et al (2022) Enhanced glycolysis and GSK3 inactivation promote brain metabolic adaptations following neuronal mitochondrial stress. Hum Mol Genet 31:692–704
Kadenbach B, Complex (2021) IV - The regulatory center of mitochondrial oxidative phosphorylation. Mitochondrion 58:296–302
Caro-Maldonado A et al (2014) Metabolic Reprogramming Is Required for Antibody Production That Is Suppressed in Anergic but Exaggerated in Chronically BAFF-Exposed B Cells. J Immunol 192:3626–3636
Weisel FJ et al (2020) Germinal center B cells selectively oxidize fatty acids for energy while conducting minimal glycolysis. Nat Immunol 21:331–
Dufort FJ et al (2007) Cutting edge: IL-4-mediated protection of primary B lymphocytes from apoptosis via Stat6-dependent regulation of glycolytic metabolism. J Immunol 179:4953–4957
Victora GD et al (2012) vol 120, pg 2240,. Identification of human germinal center light and dark zone cells and their relationship to human B-cell lymphomas Blood 126, 1262–1262 (2015)
Doughty CA et al (2006) Antigen receptor-mediated changes in glucose metabolism in B lymphocytes: role of phosphatidylinositol 3-kinase signaling in the glycolytic control of growth. Blood 107:4458–4465
Choi SC, Morel L (2020) Immune metabolism regulation of the germinal center response. Exp Mol Med 52:348–355
Jestin M et al (2020) Mitochondrial disease disrupts hepatic allostasis and lowers the threshold for immune-mediated liver toxicity. Mol Metab 37:100981
Ghini V, Saccenti E, Tenori L, Assfalg M, Luchinat C (2015) Allostasis and Resilience of the Human Individual Metabolic Phenotype. J Proteome Res 14:2951–2962
Smith RL, Soeters MR, Wüst RCI, Houtkooper RH (2018) Metabolic Flexibility as an Adaptation to Energy Resources and Requirements in Health and Disease. Endocr Rev 39:489–517
Alfardan R, Guo C, Toth LA, Nie D (2019) Impaired Recovery from Influenza A/X-31(H3N2) Infection in Mice with 8-Lipoxygenase Deficiency. Med Sci (Basel) 7
Moltedo B et al (2009) Cutting edge: stealth influenza virus replication precedes the initiation of adaptive immunity. J Immunol 183:3569–3573
Mazzarello AN et al (2022) B cell receptor isotypes differentially associate with cell signaling, kinetics, and outcome in chronic lymphocytic leukemia. J Clin Invest 132
Kominsky DJ, Campbell EL, Colgan SP (2010) Metabolic shifts in immunity and inflammation. J Immunol 184:4062–4068
Weyand CM, Zeisbrich M, Goronzy JJ (2017) Metabolic signatures of T-cells and macrophages in rheumatoid arthritis. Curr Opin Immunol 46:112–120
Luo Y, Ma J, Lu W (2020) The Significance of Mitochondrial Dysfunction in Cancer. Int J Mol Sci 21
McGuire PJ (2019) Mitochondrial Dysfunction and the Aging Immune System. Biology (Basel) 8
Salabei JK, Gibb AA, Hill BG (2014) Comprehensive measurement of respiratory activity in permeabilized cells using extracellular flux analysis. Nat Protoc 9:421–438

There is NO Competing Interest.

Download PDF

Version 1

posted

You are reading this latest preprint version

Single-cell mapping of metabolic dependences using ATP as a biomarker

Status:

Version 1

Abstract

Figures

INTRODUCTION

RESULTS

Cells with distinct bioenergetic capacities display varying ATP levels

Differential glycolytic dependence revealed through ATP

Differential FAO dependence revealed through ATP

Differential OXPHOS dependence revealed through ATP

Mathematical integration of metabolic dependences for comprehensive bioenergetic profiling (MitE-Flo)

MitE-Flo delineates metabolic dependences in mitochondrial disease

MitE-Flo delineates metabolic dependences in dark zone and light zone germinal center B cells

DISCUSSION

MATERIALS AND METHODS

Cell culture: Madin Darby canine kidney cells (MDCK)

Cell culture: Immortalized line of human T lymphocyte cells (Jurkat)

Cell culture: Lymphoblastoid cell lines (LCL)

Measurement of cell respiration with Seahorse flux analyzer

Measurement of extra cellular acidification rate (ECAR) with Seahorse flux analyzer

β-oxidation of palmitate

Mitochondrial complex activity assay

Flow cytometer and cell sorting

Realtime PCR for metabolic genes

Colorimetric ATP

Drugs stock

Metabolic profile by flow cytometry (MitE-Flo)

Data analysis

Declarations

References

Additional Declarations

Status:

Version 1