A METHOD OF PROFILING THE ENERGETIC METABOLISM OF A POPULATION OF CELLS

Abstract

The present invention relates to a method of profiling the energetic metabolism of a single cells. The inventors designed a method that rapidly and efficiently measures the protein synthesis level in single cells upon inhibition of the different energy producing pathways. The method developed by the inventors permits to acquire energetic metabolism profiles with single cell resolution in non-abundant cells ex-vivo and permits to decrease to a minimum manipulation time, incubations and cost of sample preparation. The method is particularly suitable for determining activation state of a cell, diagnosing inflammatory diseases, or predicting the response of a subject to immunotherapy treatment. In particular, the present invention relates to a method of profiling the energetic metabolism in single cells comprising measuring the protein synthesis level of the cells and contacting the cell with different inhibitors of metabolic pathways.

Claims

1. A method of profiling the energetic metabolism profile of a population of cells comprising: i) providing four samples [S1], [S2], [S3] and [S4] of said population of cells ii) measuring the protein synthesis level [LCo] in sample [S1] in the absence of any inhibitor; iii) contacting the sample [S2] with an inhibitor [A] of energy production resulting from glycolysis and oxidative phosphorylation of glucose-derived pyruvate and measuring the protein synthesis level [LA] in said sample; iv) contacting the sample [S3] with an inhibitor [B] of energy production resulting from TCA cycle and oxidative phosphorylation comprising pyruvate oxidation, oxidation of fatty acids and oxidation of amino acids and measuring the protein synthesis level [LB] in said sample [S3]; v) contacting the sample [S4] cells with both inhibitors [A] and [B] and measuring the protein synthesis level [L(A+B)] in said sample [S4]; vi) assessing the glucose dependency of the population of cells; vii) assessing the mitochondrial dependency of the population of cells; viii) assessing the glycolytic capacity of the population of cells; ix) assessing the capacity for the oxidation of fatty acids and the oxidation of amino acids of the population of cells and then x) determining the energetic metabolism profile of the population of cells based on assessments made in steps vi), vii), viii) and ix).

2. The method of claim 1 wherein the population of cells consists of a homogeneous population of cells or a heterogeneous population of cells.

3. The method of claim 1 wherein the population of cells is from a biological sample obtained from a subject.

4. The method of claim 1 wherein the population of cells comprises immune cells; natural killer cells; myeloid cells, neutrophils, eosinophils, mast cells, basophils, and/or granulocytes.

5. The method of claim 1 wherein the inhibitor [A] is selected from the group consisting of 2-Deoxy-Glucose, 2-[N-(7-Nitrobenz-2-oxa-1,3-diaxol-4-yl)amino]-2-deoxyglucose/2-NBDG, Phloretin, 3-Bromophyruvic acid, Iodoacetate, Fluoride and 6-Aminonicotinamide.

6. The method of claim 1 wherein step iii) is performed in the presence of pyruvate or acetate.

7. The method of claim 1 wherein the inhibitor [B] is selected from the group consisting of Oligomycin (A/B/C/D/E//F and derivates), Rotenone, Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone/FCCP, Trimetazidine/TMZ, 2[6(4-chlorophenoxy)hexyl]oxirane-2-carboxylate/Etamoxir, Bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl sulfide/BPTES, and enasidenib.

8. The method of claim 1 wherein the inhibitor [B] is an inhibitor of wild type or mutant enzymes of mitochondrial metabolism pathways.

9. The method of claim 1 wherein the glucose dependency of the cells is assessed by calculating formula (I): $\begin{array}{cc}\mathrm{Glucose}\mathrm{dependency}=\left(\left[LCo\right]-\left[LA\right]\right)/\left(\left[LCo\right]-\left[L\left(A+B\right)\right]\right)\times 100.& \left(I\right)\end{array}$

10. The method of claim 1 wherein the mitochondrial dependency of the cells is assessed by calculating formula (II): $\begin{array}{cc}\mathrm{Mitochondrial}\mathrm{dependency}=\left(\left[LCo\right]-\left[LB\right]\right)/\left(\left[LCo\right]-\left[L\left(A+B\right)\right]\right)\times {100}_{.}& \left(\mathrm{II}\right)\end{array}$

11. The method of claim 1 wherein the oxidation of fatty acids and oxidation of amino acids capacity is assessed by calculating formula (IV): $\begin{array}{cc}\mathrm{Oxidation}\mathrm{of}\mathrm{fatty}\mathrm{acids}\&\mathrm{oxidaion}\mathrm{of}\mathrm{amino}\mathrm{acids}\mathrm{capacity}=\left(1-\left(\left[LCo\right]-\left[LA\right]\right)/\left(\left[\mathrm{LCo}\right]-\left[L\left(A+B\right)\right]\right)\right)\times 100.& \left(\mathrm{IV}\right)\end{array}$

12. The method of claim 1 which further comprises the steps of: providing a further sample [S5] of the population of cells, contacting said sample [S5] with an inhibitor [C] of energy production resulting from oxidation of fatty acids and measuring the protein synthesis level [LC] in said sample and, assessing the dependency of oxidation of fatty acids of the population cells.

13. The method of claim 12 wherein, the inhibitor [C] is Trimetazidine/TMZ or 2[6(4-chlorophenoxy)hexyl]oxirane-2-carboxylate/Etamoxir.

14. The method of claim 12 wherein the dependency of oxidation of fatty acids is assessed by calculating formula (V): $\begin{array}{cc}\mathrm{Dependency}\mathrm{of}\mathrm{oxidation}\mathrm{of}\mathrm{fatty}\mathrm{acids}=\left(\left[\mathrm{LCo}\right]-\left[\mathrm{LC}\right]\right)/\left(\left[\mathrm{LCo}\right]-\left[L\left(A+B\right)\right]\right)\times 100.& \left(V\right)\end{array}$

15. The method of claim 1 which further comprises the steps of providing a further sample [S6] of the population of cells, contacting the sample [S6] with an inhibitor [D] of the production of energy resulting from oxidation of amino acids and measuring the protein synthesis level [LD] in the sample, and assessing the dependency of oxidation of amino acids of the population cells.

16. The method of claim 15, wherein the inhibitor [D] is selected from the group consisting of Bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl sulfide/BPTES, Aminooxyacetic acid/ADA and epigallocatechin-3-gallate/EGCG.

17. The method of claim 15 wherein the dependency of oxidation of amino acids is assessed by calculating formula (VI): $\begin{array}{cc}\mathrm{Dependency}\mathrm{of}\mathrm{oxidation}\mathrm{of}\mathrm{acids}=\left(\left[\mathrm{LCo}\right]-\left[\mathrm{LD}\right]\right)/\left(\left[\mathrm{LCo}\right]-\left[L\left(A+B\right)\right]\right)\times 100.& \left(\mathrm{VI}\right)\end{array}$

18. The method of claim 1 wherein the protein synthesis levels [LCo], [LA], [LB] and [L(A+B)] are determined by contacting the samples [S1], [S2], [S3] and [S4], respectively, with puromycin and then with monoclonal antibodies specific for puromycin, wherein the monoclonal antibodies are conjugated with a detectable label.

19. The method of claim 18 wherein the detectable label is a heavy metal, a fluorescent label, a chemiluminescent label, an enzyme label, a bioluminescent label, a colloidal gold, or a DNA-barcode oligonucleotide.

20. The method of claim 18 wherein the protein synthesis levels are assessed by cytometry, cytof or Cite-seq.

21. The method of claim 1 which further comprises identifying a particular cell type in said population of cells, using of a panel of binding partners specific for one or more cell surface markers of interest.

22. The method of claim 1 wherein determining the energetic metabolism profile of the population of cells according step x) indicates whether said population of cells has a respiratory profile or a glycolysis profile.

23-24. (canceled)

25. Use of the method of claim 1 for predicting whether a subject suffering from cancer will be eligible to a therapy, in particular chemotherapy or immunotherapy.

26. A kit for performing the method of claim 1 comprising: an amount of inhibitor [A], an amount of inhibitor [B], an amount of puromycin, an amount of monoclonal antibodies specific to puromycin, optionally an amount of inhibitor [C], optionally an amount of inhibitor [D], optionally an amount of pyruvate, optionally an amount of acetate, optionally a panel of antibodies for cell sorting, and optionally a software package for calculating the different formulas (I-VI) suitable for assessing the metabolic profile.

27. The method of claim 1, wherein the glycolytic capacity of the cells is assessed by calculating the formula (III): $\begin{array}{cc}\mathrm{Glycolytic}\mathrm{capacity}=\left(1-\left(\left[LCo\right]-\left[LB\right]\right)/\left(\left[LCo\right]-\left[L\left(A+B\right)\right]\right)\right)\times 100.& \left(\mathrm{III}\right)\end{array}$

Description

FIGURES

[0086] FIG. 1. Rational and principle of ZENITH. A) Different cells use different source of energy to produce ATP and GTP. These sources are mostly glucose, aminoacids or fatty acids. Cells using these different molecules display different energetic metabolism profiles that are depicted. Protein synthesis consumes around 30% of the cellular energy (ATP/GTP) available and is thus rapidly impacted by nutrients starvation. B) To determine the EM profile using ZENITH, cell samples are subdivided and incubated with or without energetic pathways inhibitors (e.g. No inhibitor, inhibitor A, inhibitor B and inhibitors A+B). Cells produce ATP using glycolysis by oxidative phosphorylation of the glycolysis products (GlycOxPhos) or by degrading Fatty acids and aminoacids (FAO&Glnlysis). Inhibitor A blocks Glycolysis and GlycOxPhos, while inhibitor B blocks the production of energy from GlycOxphos, FAO and amino acids, and the combination of Inhibitors A and B (and or C) blocks the production of ATP from all sources. The bulk of treated cells (Control, A, B and A+B) are processed for flow cytometry with same combination of monoclonal antibodies to identify single cells and protein synthesis level after a short puromycin pulse. The impact of each treatment (energetic pathway inhibitors) on protein synthesis levels is quantified by multiparametric flow cytometry and processed with appropriate analysis sofwares. C) EM profiles are directly inferred from puromycin incorporation levels (MFI euro) in control cells (PuroCo), substracted form the levels found in the cells treated with the different inhibitors as described in the formulas.

[0087] FIG. 2. ATP levels and protein synthesis. 4×104 MEFs were seeded in 96 well plates and treated with 2-DG+Oligo for different periods of time. A) Effect of protein synthesis and RNA synthesis inhibition on the pool of ATP. Total ATP levels in the absence (Co) or presence of translation inhibitor (CHX), transcription inhibitor (ActD) or both CHX+ActD (5 minutes of treatment). B) Levels of ATP as a function of time after 2-DG+Oligo treatment. C) Levels of translation (Puromycin MFI) as a function of time after 2-DG+Oligo treatment measured in single cells by FACS. D) Correlation between the level of ATP and level of translation (Puro WI). Two tailed t-test (A) on at least 3 independent experiments (*p<0.05; **p<0.001). N=3 independent experiments (*p<0.05; **p<0.001).

[0088] FIG. 3. EM profile using ATP levels or protein synthesis activity in MEFs and changes in EM profile in human and mouse T cells upon activation. A) ATP levels in MEFs treated for 20-30 minutes with the metabolic inhibitors. B) EM profile established with the ATP levels shown in (A). C) EM profile established with ATP levels at different timepoints after adition of metabolic inhibitors. D) Level of protein synthesis in MEFs treated for 20 minutes with the metabolic inhibitors. E) EM profile calculated with the Puro incorporation levels shown in (D). G) Mouse splenic CD8 T cells where analyzed using ZENITH. Total Puro MFI is indicated in Control cells (non-activated) or PMA/Ionomycin-treated. H) EM profiles established with values shown in (G). I) EM profile of human blood central memory CD4+ T cells from two different subjects (P1 and P3) was generated in control (non-stimulated) or activated (aCD.sup.3/aCD28 beads) using the ZENITH method. A combination of antibodies that allow for the identification of central memory CD4 T (CD3+CD4+CD45RO+CCR7+) cells was used to estimate the metabolic profile in the absence or presence of T cells stimulation. N=3, ANOVA on at least 2 independent experiments (*p<0.05; **p<0.005; ***p<0.0005).

[0089] FIG. 4. EM profile based on ZENITH in WT and in PERK KO bmDCs differentiated in vitro with Flt3L. A) Flt3L-bmDC cultures from WT and CD11c-Cre PERKflox/flox mice were analyzed using ZENITH to determine if PERK and ER-stress are implicated in the translation loss induced by 2-DG treatment. Distinct DC subsets displaying key markers (see methods) are indicated as DC1 (XCR1+), DC2(CD11B+)/DC3 and DC6 (pDC). B) Puromycin incorporation levels measured by flow cytometry in DC subsets from the same in vitro culture. DCs were incubated with Control, 2-DG, Oligo and 2-DG+Oligo prior puromycin incorporation. C) Glucose and mitochondria Dependency or Glycolytic and FAO&Glnlysis (FAO) Capacity, established from the raw values in B) show that even in the absence of PERK, cells undergo translation inhibition. Data shown as mean±SD of three independent mice and are representative of two independent experiments; *p<0.05, **p<0.001.

[0090] FIG. 5. EM profile based on ZENITH of bone Marrow derived Dendritic cells subsets (BMDCs). A) Example of protein synthesis profile of BMDCs from three mice (biological replicates) in response to 2DG, Control treatment, Oligomycin and 2-DG+Oligomycin. The area under the curve of protein synthesis levels represent the capacity of the cells to sustain metabolic functions (i.e. Translation) in the presence of each inhibitor, allowing to obtain the metabolic profile (0, 1, 2, 3, 4, 5). B) ZENITH metabolic profile of FLT3L derived BMDCs from three (I, II, III) WT mice non-stimulated (Left) or stimulated during 4 hours with LPS (Right). Dendritic cells subpopulations were analysed and their metabolic profile is shown. C) The metabolic profile of each replicate (n=3) from each mice (n=3), obtained in the different DCs subpopulations can be used to cluster cells subsets with similar metabolic profiles. D) Statistical analysis of the metabolism profile obtained in the different DC subsets, reveals that DC2 are the subset that performs a stronger metabolic switch upon LPS, while pDCs do not change their metabolism.

[0091] FIG. 6. Parallel Seahorse® and ZENITH metabolic anlaysis of resting and activated T cells. A) Correlation between the changes in glycolytic capacity of steady state and activated (aCD3/CD28) T cells from three different subjects measured by Seahorse® and ZENITH (P1, P2, P3, Pearson r=0.92; R.sup.2=0.85; p<0.01). B) Basal Oxygen Consumption Rate (OCR) in non-activated (non-Act) and activated T cells (aCD3/CD28). Each bar represents the mean of P1, P2 and P3 (in triplicates). C) Basal translation levels (anti-Puro, Geometric Mean Fluorescence, intensity) in non-activated (non-Act) and activated T cells (aCD3/CD28). Each bar represents the mean of P1, P2, and P3 (in triplicates).

EXAMPLE

[0092] Material & Methods

[0093] Cells and Cell Culture

[0094] Mouse splenocytes from WT C57BL/6J (Jackson) or PERK KO C57BL/6J background mice (Zhang et al., 2002) were cultured in DMEM containing 5% of Fetal Calf Serum (FCS) and 50 μM of 2-Mercaptoethanol (Mouse cells culture media, MCCM) at 37° C. 5% of CO2 GM-CSF BM-derived dendritic cells (GM-bmDCs) were differentiated in vitro from the bone marrow of 6-8-week-old male mice, using GM-CSF, produced by J558L cells. Bone marrow progenitors were plated at 0.8×106 cells/ml, 5 ml/well in 6-well plates, and cultivated with RPMI (GIBCO), 5% FCS (Sigma-Aldrich), 20 μg/ml gentamycin (Sigma-Aldrich), 50 μM β-mercaptoethanol (VWR), and GM-CSF. The medium was replaced every 2 days; BM-derived DCs were used for experiments at day 6. Similarly, FLT3L BM-derived dendritic cells (FLT3L-bmDCs) were differentiated by adding FLT3L to RPMI, 10% of Fetal Calf Serum (FCS) and 50 μM of 2-Mercaptoethanol (Mouse cells culture media, MCCM) during 6 days at 37° C. 5% of CO2. To obtain splenocytes, eight weeks old wild type C57BL/6J mice were sacrificed by cervical dislocation and splenectomized. Single cells suspentions from the spleens were generated and cultured in MCCM as previously described. Mononuclear cell enriched from blood of healthy donors was submitted to Ficoll-paque plus (PBL Biomedical Laboratories). PBMCs and Whole blood were cultured in the absence (non stimulated) or in the presence of for indicated time. Immune cell stimulations were performed in the absence (Control) or presence of 0.1 μg/ml of extrapure Lipopolysacharide (Invivogen LPS, Cat. tlrl-3pelps), 10 μg/mlPoly I:C (Invivogen, Cat. No. tlrl-pic), CpG-A ODN 2216 (Invivogen, Cat. No. tlrl-2216) or PMA (5 ng/ml; Sigma, Cat. no. P-8139) and ionomycin (500 ng/ml; Sigma, cat. no. 1-0634) over night for T cell stimulations and 4 hours for Dendritic cells.

[0095] ATP Measurement

[0096] 20×104 MEFs were seeded in 100u1 of 5% FCS DMEM culture media ON in opaque 96 well plates. Cells were incubated with the inhibitors for the times indicated in the figures. After, 100 ul of Cell titer-Glo luminiscence ATP reconstituted buffer and substrate (Promega, Cat. No. G7570) was added to each well and Luminiscence was measured after 10 minutes following manufacturer instructions. A standard curve with ATP was performed using the same kit and following manufacturer instructions.

[0097] Metabolic Flux Analysis (Seahorse®)

[0098] OCR and ECAR were measured with the XF24 Extracellular Flux Analyzer (Seahorse Bioscience). 4.105 cells with αCD3/αCD28 beads or not, were placed in triplicates in XF medium (nonbuffered Dulbecco's modified Eagle's medium containing 2.5 mM glucose, 2 mM L-glutamine, and 1 mM sodium pyruvate) and monitored 25 min under basal conditions and in response to 10 mM Glucose, 1 μM oligomycin, 100 mM 2-Deoxy-Glucose. Glycolytic capacity was measured by the difference between ECAR level after add oligomycin and before add glucose. OCR, ECAR and SRC parameters was analyzed and extract from Agilent Seahorse Wave Desktop software. Glycolytic capacity was obtained by the difference between ECAR level after add Oligomycin and before add Glucose.

[0099] ZENITH

[0100] Cells were plated at 1-10×106 cells/ml, 0.5 ml/well in 48-well plates. Experimental duplicates were performed in all conditions. After differentiation, activation or harvesting of human of cells, wells were treated during 45 minutes with Control, 2-DeoxyGlucose (2-DG, final concentration 100 mM; Sigma-Aldrich Cat. No. D6134-5G), Oligomycin (Oligo, final concentration 1 μM; Sigma-Aldrich Cat. No. 75351), BPTES (final concentration 1 μM; Cat. No. SML0601), Trimetazidine (TMZ, final concentration 1 μM; Sigma-Aldrich Cat. No. 653322) or a combination of the drugs at the final concentrations before mentioned. As negative control, the translation initiation inhibitor Harringtonine was added 15 minutes before addition of Puromycin (Harringtonine, 2 μg/ml; Abcam, cat. ab141941). Puromycin (Puro, final concentration 10 μg/ml; Sigma-Aldrich, Cat. No. P7255) is added during the last 15 minutes of the metabolic inhibitors treatment. After puro treatment, cells were washed in cold PBS and stained with a combination of fluorochrome cell viability marker and conjugated antibodies against different surface markers during 30 minutes at 4° C. in PBS 5% FCS, 2 mM EDTA (FACS wash buffer). After washing with FACS wash buffer, cells were fixed and permeabilized using BD Cytofix/Cytoperm™ (Catalog No. 554714) following manufacturer instructions. Intracellular staining of Puro using fluorescently labeled anti-Puro monoclonal antibodies was performed by incubating cells during one hour (1:1000, Clone 12D10, Merk, Catalog No. MABE343) at 4° C. diluted in Permwash.

[0101] Flow Cytometry

[0102] Flow cytometry was conducted using BD LSR II and BD LSR Fortessa X-20 machine (BD Biosciences™) and data were analyzed with FlowJo (Tree Star™) and/or Cytobank. The antibodies used to stain mouse splenocytes were anti-Puro-AF488 (Merk, Catalog No. MABE343) or anti-Puro-Clone R4743L-E8, rat IgG2A in house produced and conjugated with Alexa Fluor 647 or Alexa-Fluor 488, anti-Phospho-S6-PE (Cell Signaling Technology, Catalog No. #5316), anti-Ki67 PE-eFluor 610 (eBioscience™, Catalog No. 61-5698-82) CD4-APC-eF780 (eBioscience™, Catalog No. 47-0042-82), CD8-APC (eBioscience™, Catalog No. 17-0081-83), CD8O-PercPCy5.5 (Biolegend™, Catalog No. 104722), anti-B220-BV421 (Biolegend™, Catalog No. 103251), anti-MHC-II-AF700 (eBioscience™, Catalog No. 56-5321-82), LIVE/DEAD™ Fixable Aqua Dead Cell Stain (Invitrogen™, Catalog No. L34957). The following anti-Human antigens antibodies were used for staining whole blood and PBMCs upon ZENITH protocol application. Alexa Fluor-488 Mouse Anti-Human Axl (Clone 108724, R&D Biosystems, Cat. No. FAB154G), Alexa Fluor-647 Mouse Anti-Puromycin (clone 12D10, Millipore, Cat. No. MABE343-AF647), BUV395 Mouse Anti-Human CD11c (Clone B-1y6, BD Bioscience, Cat. No. 563787), BUV737 Mouse Anti-Human CD86 (Clone FUN-1, BD Bioscience, Cat. No. 564428), BV510 Mouse Anti-Human CD19 (Clone HIB19, BD Bioscience, Cat. No. 740164), BV510 Mouse Anti-Human CD3 (Clone HIT3a, BD Bioscience, Cat. No. 564713), BV510 Mouse Anti-Human CD56 (Clone B159, BD Bioscience, Cat. No. 740171), BV605 Anti-Human HLA-DR (Clone L243, BioLegend, Cat. No. BLE307640), BV650 Mouse Anti-Human CD16 (Clone 3G8, BD Bioscience, Cat. No. 563692), BV711 Mouse Anti-Human CD14 (Clone M5E2, BD bioscience, Cat. No. 740773), BV785 Mouse Anti-Human CD45RA (Clone HI100, BioLegend, Cat. No. BLE304140), Live Dead Fixable Aqua Dead Cell Stain Kit (Life Technologies, Cat. No. L34957), PE Rabbit anti-Phospho-S6 Ribosomal Protein (Ser235/236) monoclonal (Clone D57.2.2E, Cell signaling, Cat. No. 5316S), PE Rat Anti-Human Clec9A/CD370 (Clone 3A4, BD Bioscience, Cat. No. 563488), PE-Cy7 Mouse Anti-Human CD22 (Clone HIB22, BD Bioscience, Cat. No. 563941), AF488 Mouse Anti-Human CD38 (Clone HIT-2, BioLegend, Cat. No. BLE303512).

[0103] Animal Studies

[0104] Wild type C57BL/6 mice were purchased from Charles River and maintained in the animal facility of CIML under specific pathogen-free conditions. C57Bl/6 embryos at gestation embryonic day 12.5 (E12.5) were depleted of all organs and brain, and dissociated in Liberase Tm (0.5 mg/ml, Roche), DNasel (0.2 mg/ml, Roche) in PBS for 15 min at 37° C. while stirring constantly. Cell suspensions were washed with RPMI (Thermo scientific), supplemented with 2% heat-inactivated FCS, 100U/ml penicillin, and 100 mg/ml streptomycin prior Immuno-staining and flow cytometry analysis.

[0105] This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals the French Ministry of Agriculture and of the European Union. Animals were housed in the CIML animal facilities accredited by the French Ministry of Agriculture to perform experiments on alive mice. All animal experiments were approved by Direction Départementale des Services Vétérinaires des Bouches du Rhône (Approval number A13-543). All efforts were made to minimize animal suffering.

[0106] Statistical Analysis

[0107] Statistical analysis was performed with GraphPad Prism software. When several conditions were to compare, we performed a one-way ANOVA, followed by Tukey range test to assess the significance among pairs of conditions. When only two conditions were to test, we performed Student's t-test or Welch t-test, according the validity of homoscedasticity hypothesis (*P<0.05, **P<0.01, ***P<0.005).

[0108] Results

[0109] Given the hurdles find in other metabolism analytical techniques, our objective was to develop a method responding to three main imperatives: A) To profile EM in non-abundant cells ex-vivo; B) To decrease to a minimum manipulation time, incubations and cost of sample preparation; C) To acquire EM profiles with single cell resolution. Changes in the EM profile require the activation of different signaling pathways and also genetic reprograming that enable cells to change their catabolism. ZENITH is based on this concept and integrates that most of the energy that the cell obtain from degrading glucose, aminoacids or lipids is constantly consumed by the protein synthesis machinery (Buttgereit and Brand, 1995; Lindqvist et al., 2017; Schimmel, 1993). Conceptualizing that protein synthesis intensity directly reflects ATP consumption, we designed a method that rapidly and efficiently measure by flow cytometry protein synthesis activity in single cells upon inhibition of the different energy producing pathways (FIG. 1). The two inhibitors that are shown in the illustrations and used in the experiments are 2-Deoxy-D-Glucose (2-DG, Inhibitor A) and Olygomycin A (Oligo, inhibitor B). Glucose can be used in two ways to produce ATP, Glycolysis alone (FIG. 1B) or Glycolysis followed by Oxidative Phosphorylation (GlycOxPhos). As shown in FIG. 1B, the presence of inhibitor A, blocks ATP/GTP production from glucose degradation, and only allow cells to produce ATP/GTP from acetyl-CoA derived from oxidation of fatty acids (FAO) or oxidation of amino acids (ie Glnlysis). In contrast, the inhibitor B, impairs the ATP production from mitochondrial respiration (OxPhos), also resulting in the inhibition of the TCA cycle, and consequently inhibiting energy production from acetyl-CoA, FAO and Glnlysis, and thus cells can only produce ATP to sustain protein synthesis by performing Glycolysis. Cells that have a glycolytic EM profile, express a set of enzymes with nucleoside di-phosphate kinase activity (NDPK/NME) that rapidly exchange ATP to GTP and are able to sustain translation when mitochondrial respiration is inhibited. In contrast, cells that have a respiratory EM profile, that are forced to switch to glycolysis (i.e. by blocking mitochondrial respiration) can potentially produce ATP to survive, but they are not adapted to generate GTP and do not sustain protein synthesis efficiently.

[0110] Classically, measurements of the rate of protein synthesis required the use of radioactive amino acids. Puromycin (puro) is an antibiotic, that due to its tRNA-AA mimetic molecular structure is very efficiently incorporated into the nascent polypeptide chains during the process of mRNA translation by the Ribosomes. We have previously developed and patented fluorescent monoclonal antibodies anti-puromycin and demonstrated that the level of anti-puromycin staining is a very precise measure of the level of protein synthesis (Schmidt et al., 2009), patent FR08 56499). To further optimize the signal to noise ratio of puromycin intracellular detection during ZENITH, we screened and developed a novel monoclonal anti-puromycin antibody specifically adapted to intracellular flow cytometry (data not shown). The metabolic profile of each cell is obtained by the combination of a panel of monoclonal antibodies that enable us to identify single cells from different types, and a directly coupled to fluorochrome anti-Puro monoclonal antibody to monitor the level of protein synthesis by the use of multiparametric flow cytometry. Taking into account the mechanism of action of each of the inhibitors, the direct interactions between the EM pathways and global protein synthesis intensity, we apply a mathematic formula to integrate the results obtained from the different measurements into one simple graph that illustrates the metabolic profile of each cell subsets (FIG. 1C). The capacity of certain cells to compensate and change their metabolism upon treatment with inhibitors, can be misinterpreted as a defect of our method. ZENITH measure the glycolytic capacity of cells when mitochondrial respiration is inhibited, and the glucose dependency in the presence of 2-DG. The FAO and Glnlysis dependency can also be calculated in the presence of TMZ (FAO inhibitor) and BPTES (Glutaminolysis inhibitor). Moreover, the low dependency in any of the sources can be used to identify the cells with high degree of EM plasticity. In other words, we quantify the degree of synergy in the presence of inhibitors of several pathways, as compared to the sum of effects of the inhibitors separately. This measure allow us to identify cells with a more plastic and rapidly adaptable EM metabolism, something that is informative and was only observed in certain cells.

[0111] Changes in ATP Levels Tightly Correlate with Protein Synthesis Activity.

[0112] Our methodology was validated by correlating total ATP levels with protein synthesis measured by flow cytometry. Using non-transfrormed mouse embryonic fibroblasts (MEFs), we tested whether translation or transcription inhibition (with Cycloheximide/CHX, or ActinomycinD/ActD, respectively) impacted ATP levels, when the different energetic pathways are inhibited (Oligo+2-DG). As shown in FIG. 2A, only when CHX was added, the pool of ATP was found to be significantly higher compared to Oligo+2-DG treatment (Co) (FIG. 2A). This result, is in accordance with previous results showing that translation has a higher energetic cost as compared to transcription, and thus is a more sensitive marker of the metabolic status of the cells. After, to determine if they show similar pattern we tested how the levels of ATP and the level of protein synthesis change as function of time upon blockade of ATP synthesis. We incubated MEFs during different times with Oligo+2-DG, and ATP levels were quantified by luminescence using Cell titer-Glo® (FIG. 2B), and protein synthesis after puromycin incorporation by flow cytometry (FIG. 2C). We observed a statistically significant correlation between the levels of ATP and the level of total protein synthesis (FIG. 2D, Pearson r=0.985, R squared=0.9703, p<0.0001) These results demonstrate that variation in intracellular levels of ATP and of protein synthesis activity can be directly correlated and that translation intensity can be used as a read out to monitor variations in ATP availibility and synthesis and perform EM profiling.

[0113] Protein Synthesis Levels in Response to Metabolic Inhibitors Describe Better the EM Activity than Bulk ATP Levels.

[0114] Human and mouse embryonic fibroblasts have been shown to display high Oxygen Consumption Rate (OCR) and low ExtraCellular Acidification Rate (ECAR) (Suganya et al., 2015; Zhang et al., 2012). This important ratio of OCR/ECAR is characteristic of non-transformed cells that have high mitochondrial activity, relative to a low glycolytic activity (Suganya et al., 2015; Zhang et al., 2012). As a consequence MEFs do not acidify their culture media, even when cultured at high density. This phenotype is also observed in resting T cells, but is rapidly lost when T cells are activated by APCs and engage in a rapid switch to glycolysis (MacIver et al., 2013).

[0115] Based on these observations and high mitochondrial dependency of fibroblast and T cells, we compared EM profiles obtained from quantifying ATP levels (bulk ATP levels) to profiles obtained from quantifying protein synthesis by flow. The formulas used for calculating the EM profiles with ATP is developed as such:

[00007]$\mathrm{Glucose}\mathrm{dependency}=\frac{{\mathrm{ATP}}_{\mathrm{Co}}-{\mathrm{ATP}}_{2\mathrm{DG}}}{{\mathrm{ATP}}_{\mathrm{Co}}-{\mathrm{ATP}}_{2\mathrm{DG}+\mathrm{Oligo}}}\times 100$$\mathrm{Mitochondrial}\mathrm{dependency}=\frac{{\mathrm{ATP}}_{\mathrm{Co}}-{\mathrm{ATP}}_{\mathrm{Oligo}}}{{\mathrm{ATP}}_{\mathrm{Co}}-{\mathrm{ATP}}_{2\mathrm{DG}+\mathrm{Oligo}}}\times 100$$\mathrm{Glycolytic}\mathrm{Capacity}=\left(1-\frac{{\mathrm{ATP}}_{\mathrm{Co}}-{\mathrm{ATP}}_{\mathrm{Oligo}}}{{\mathrm{ATP}}_{\mathrm{Co}}-{\mathrm{ATP}}_{2\mathrm{DG}+\mathrm{Oligo}}}\right)\times 100$$\mathrm{FAO}\&\mathrm{Glnlysis}\mathrm{Capacity}=\left(1-\frac{{\mathrm{ATP}}_{\mathrm{Co}}-{\mathrm{ATP}}_{2\mathrm{DG}}}{{\mathrm{ATP}}_{\mathrm{Co}}-{\mathrm{ATP}}_{2\mathrm{DG}+\mathrm{Oligo}}}\right)\times 100$

[0116] In MEFs the maximum glycolytic capacity calculated by using ATP levels (FIG. 3A) is almost 100% and is significantly higher than the maximun FAO&Glnlysis capacity (FIG. 3B, p<0.001 legend). By performing kinetic studies, we could observe that the total EM profile remain unchanged, despite being calculated at different times after treatment with the inhibitors (FIG. 3C), as exemplified by the levels of ATP that remained unchanged in the presence of Oligomycin (FIGS. 3A and 3B). These results demonstrate that in highly respiratory cells, like MEFs, global ATP levels after treatment with metabolic inhibitors are not able to reflect energetic mitochondrial dependency. Oligomycin inhibition on respiration could only be observed when 2-DG+Oligo was also added to the cells. These results, evidenced that a measure of ATP levels upon inhibition of the different energy producing pathways does not permit to obtain an accurate EM profile of the cells, due to the induction of probable compensation mechanisms. In contrast, when ZENITH was used to profile the same MEFs, a completely different, but coherent, picture emerged (FIGS. 3D, 3E and 3F). ZENITH revealed that MEFs have higher mitochondrial dependency than glucose dependency, and also a higher FAO&Glnlysis capacity than glycolytic capacity (FIG. 3E, p<0.05 legend), confirming previous published observation on the EM of these cells (Suganya et al., 2015; Zhang et al., 2012). To further demonstrate the validity of ZENITH, we analyzed mouse splenic cells (n=5l) and CD8+ T cells activated or not for 6 h with PMA/Ionomycin. CD8+ T cells activation leads to a dramatic switch in EM profile, increasing metabolic activity and switching from respiration to aerobic glycolysis (MacIver et al., 2013). Resting splenic CD8 T cells displayed significantly lower metabolic activity than their activated counterparts (FIG. 3G, p<0,0001). Determination of the EM transition profile of non-activated to activated CD8 T cells, indicated a strong reduction in mitochondrial energetic dependency (from 55% to less than 15%, FIG. 3H, p<0,0001) and an increase in glycolytic capacity, reaching almost 90% (FIG. 3H, p<0,0001). These results were confirmed in human central memory CD4+ T cells isolated from blood and after activation with beads coated with the agonistic anti-CD3 and anti-CD28 antibodies (FIG. 3I). Altogether these results, demonstrate that ZENITH by analyzing protein synthesis in single cells exposed to metabolic inhibitors provides an accurate measure of the respective contributions of EM pathways activity in individual cells present in heterogenous populations.

[0117] Protein Synthesis Inhibition Upon Short Exposure to 2-DG is not Due to ER-Stress Induction.

[0118] It has been previously shown that treatment of cells with 2-DG during several hours can induce endoplasmic reticulum (ER) stress due to a defect in glycosylation and folding of recently translated proteins (Liu et al., 2016; Marquez et al., 2017; Zhang et al., 2015). ER-stress induce the activation of PERK, a kinase that phosphorylates the alpha subunit of eukaryotic translation initiation factor 2 (eIF2a) that acts as a dominant negative inhibitor of translation initiation. To address whether ER stress could interfere with protein synthesis upon 2-DG treatment, we took advantage of the CD11c-CRE PERKflox/flox mice that delete PERK in all DC subsets (CD11c+ cells, data not shown). 2-DG effect was compared both in WT PERKflox/flox (control) and in CD11c-CRE PERKflox/flox (PERK KO) DCs derived from bone marrow precursors with Flt3L. Irrespective of their genetic background, the different DC subsets had the same capacity to decrease protein synthesis in response to 2-DG and displayed the same EM characteristics, suggesting that PERK is not implicated in the loss of translation observed upon the short 2-DG treatment required to perform ZENITH.

[0119] Metabolic Profile of In-Vitro Generated and Ex-Vivo Mouse Dendritic Cells

[0120] Most studies about the metabolism of mouse DCs, have been performed using in-vitro differentiation cultures of bone marrow that enable to obtain large amounts of this cells. DC differentiation in response to FLT3L and GM-CSF, require in-vitro incubation of the hematopoietic precursors during at least one week in fully supplemented media and FCS, prior potential flow based sorting or magnetic purification. Both of these processes may lead to metabolic alterations and major differences with their ex-vivo counterparts. Although the validity of GM-CSF bmDC, as a truly relevant DC model is a matter of debate (Guilliams and Malissen, 2015; Helft et al., 2015; Helft et al., 2016; Lutz et al., 2016), most metabolic studies have been performed using this system (Everts et al., 2012; Kelly and O'Neill, 2015; Wolf et al., 2016). GM-CSF derived bone marrow cultures generate an heterogeneous population of macrophages (GM-Macs), Dendritic cells (GM bmDCs) and immature DCs (Helft et al., 2015; Lutz et al., 2017). The single cell resolution of ZENITH allows the characterization of the metabolic profile of cell populations present in low proportion in the sample, without the need for further manipulation and isolation. We confirmed that the bulk of immature GM-CSF-bmDC show high dependency on glucose and switch towards increased glycolytic capacity and decreased respiration upon activation with LPS (data not shown), as previously described (Everts et al., 2012). Upon LPS treatment, not all cells follow exactly in the same kinetics of activation, something that is revealed by their heterogenous levels of maturation markers. While some already start to show a mature phenotype, they also start to show higher dependency on mitochondrial respiration (data not shown, LPS immatureDC 5% vs Mature DC-A 45%, P<0.001). Interestingly, this data is in accordance with previous studies suggesting that immature GM-bmDCs have an EM profile that is equivalent to the highly glycolytic human blood monocytes (data not shown) (Cheng et al., 2014; Oren et al., 1963).

[0121] Moreover, we could identify in the culture, a subpopulation of steady state “mature” DCs charcterized by high levels of surface and CD86 (Mature DC-B, data not shown), that displayed a distinct metabolic profile and performed different metabolic change after LPS stimulation (data not shown). Although the origin and function of this population remain to be established, it could not have been identified using existing technologies, further highlighting the power of ZENITH to establish EM profiles in an unbiased fashion and concomitantly on different cell subsets. ZENITH allows therefore a functional cell population classification based on the clustering of individual cell metabolic profiles, independently of their abundance in the studied sample.

[0122] The EM profile of FLT3L-derived and splenic DC ex-vivo was also established. In contrast to GM-CSF-derived, FLT3L-bmDC encompass several DC subsets including, DC1 (XCR1+ cDC), DC2 (SIRPA+CD1 lb+cDC) and DC6 (CD123+ SiglecH+ pDCs) (Merad et al., 2013). FLT3L-derived bmDC were compared to freshly isolated splenic DCs and correlation of EM profiles with the different DC subsets and their activation state was carried-out. Flt3L bmDCs and splenic DCs are highly dependent on mitochondrial respiration, with lower glycolytic capacity in steady state condidions, with DC6/pDC being the most dependent (data not shown). In splenic DC, the LPS dependent switch to glycolysis was only partially observed in DC1 and not in DC2 nor DC6. This situation was inverted in Flt3L-bmDC, with DC2 implementing glycolysis almost exclusively, while DC6 remained unaffected. These results could be also dependent on the capacity of the different DC subsets to be activated directly by LPS, with TLR4 being mostly expressed on DC2. Altogether, our results indicate that while the metabolic profile of GM-CSF and FLT3L-bmDCs, do not entirely recapitulate the EM profile of explanted splenic DCs. Upon LPS-activation, a switch to glycolysis can be observed although in different individual subsets according to the origin of the DCs under examination. Nevertheless, EM profiles of mouse DCs strongly correlated with the level of immune activation, but not maturation, and may represent a surrogant marker of inflammatory status.

[0123] ZENITH Recapitulates Seahorse® EM Profiling of Steady State and Activated T Cells.

[0124] The metabolic switch of T cells to aerobic glycolysis upon activation was originally documented in the 1970s and more recently confirmed using the Seahorse® method. To benchmark our method, we monitored the variations in EM observed in isolated bulk human blood T cells at steady state or upon activation in parallel by Seahorse® and ZENITH®. Upon activation, an increase in the glycolytic capacity of T cells was observed by both Seahorse® and ZENITH® (Data not shown). Measures obtained with the two methods were in excellent agreement (correlation Spearman r squared 0.85, P<0.01) (FIG. 6A). We observed a significant decrease in the spare respiratory capacity in bulk of T cells upon activation with Seahorse (Data not shown). Interestingly, an increase in OCR by Seahorse, was paralleled with a strong increase in the global level of PS measured by ZENITH® although at a larger extent (FIGS. 6B and 6C, respectively). Overall, the EM profiles of T cells upon activation obtained by Seahorse® and by ZENITH® were therefore very consistent, during which the level of translation (FIG. 6C) correlated with the global metabolic activity of the cells and changes in the response to inhibitors confirmed the metabolic switch towards aerobic glycolysis that occurs upon T cell activation. However, ZENITH® showed two main advantages over Seahorse® measurements. First, the magnitude of the signal and deviation of measurements with ZENITH® were superior (FIG. 6B vs 6C). Second, ZENITH analysis was performed with 10 fold less T cells (1.2.Math.10.sup.5 in triplicates vs. 1.2.Math.10.sup.6 cells, respectively). Moreover, ZENITH® could incorporate a full FCS spectrum of T cell markers in the analysis allowing to study in parallel different CD3+ T cells subpopulations present into the bulk sample (FIG. 6), encompassing naïve, memory and effector CD4+ or CD8+ T cells subsets.

[0125] Metabolic Deconvolution of Blood T Cell Subsets by ZENITH® Identifies a Memory CD8+ T Cells Subset Constitutively Displaying High Glycolytic Capacity.

[0126] To expand upon the ability to deconvolve T cell subpopulations, we next applied ZENITH® to mixed populations that previously were inaccessible to single-cell analyses. For this, we took advantage of staining for CD45RA, IL7RA (CD127), CCR7, CD45RO, CD57, PD1, and Perforin all of which allow subset analyses of naïve and memory CD4+ and CD8+ cells from total human blood preparations. Application of antibodies to these nine markers yielded six phenotypically distinct clusters/subpopulations with different abundances (Data not shown). The metabolic profiles of non-activated naïve T cells (CD8 or CD4), as well as memory (EM and CM) CD4 and highly differentiated CD8 (HDE) showed a medium-high degree of mitochondrial dependence (Data not shown), consistenly to what was previously reported on their metabolic activity.sup.4. In contrast, the less abundant cell subsets such as CD8 early effector memory (EEM) and Natural Killer (NK) cells (that co-purfied with T cells and represented just 5% of the cells) showed higher glycolytic capacity. To determine if similar metabolic trends are observed in other species and preparations, we performed ZENITH® on resting and activated mouse splenic T cells and human blood central memory CD4 T cell subsets (Data not shown). As a result we could also determine a highly consistent switch of mouse and human T cells upon activation towards high glycolytic capacity and high glucose dependence (Data not shown).

[0127] We note that during bulk T cells analysis, naïve cells represented 42%, (the majority out of 78%) and thus likely drives the EM monitoring performed for unsegregated population by Seahorse® (FIG. 6B). Thus, Seahorse® measurements indicate a rather low “mean” glycolytic rate/capacity and high “mean” oxygen consumption rate (Data not shown). Seahorse analysis of resting T cells is in accordance with the metabolism of naïve T cells determined by ZENITH® (Data not shown). However, Seahorse results completely masked the presence of CD8+ EEM that represents no more that 5% of the cells (representing 500 cells in each of the treatments, resulting in 2000 cells) and presented, at steady state high glycolytic capacity.

[0128] Another important feature of the multiparametric ZENITH® resolutive power is the possibility to feature single cell behaviors according to their sensitivity to metabolic inhibitors independently of their phenotype. This allows to identify functional metabolic heterogeneity first, and to determine the phenotype or sort, different cells afterwards. As a proof of concept, resting purified T cells were treated with Oligomycin to inhibit mitochondrial respiration prior translation monitoring. As result, the histogram plotting translation levels showed two T cell subpopulations, one with high and one with low levels of translation (Data not shown). The population that showed high level of translation upon mitochondrial inhibition were labeled as “Glycolytic” and the cells that blocked translation as “Mitochondrial dependent” (Data not shown). As shown in the T-SNE, the phenotype of Glycolytic and Respiratory T cells, recapitulated our previous results (Data not shown) and showed that the expression of CD45RA, mostly present in naïve T and NK cells, correlates well “Mitochondrial dependence” (Data not shown). In conclusion, we found that ZENITH® allow for both the measurement of the EM profile of known non-abundant cell subsets of interest, but also to sort and identify “unknown” cells with interesting metabolic profile present within a heterogenous sample.

[0129] Profiling the Metabolic State of Human Tumor-Associated Myeloid Cells.

[0130] Immunotherapies are a game changer in oncology yet only a fraction of patients show complete immune-mediated rejection of the tumor. The variations observed in patients responses to treatment have created a strong need for understanding the functional state of tumor-associated immune cells (immunoprofiling).sup.6. We thus tested whether ZENITH® could be used for paralleled phenotypic and metabolic profiling of human tumor samples and what this would reveal about the heterogeneity of immune cell subsets comparing tumors of diverse origins, notably comparing a tumor with tumor-free adjacent tissue. We thus performed ZENITH® using PMBCs from healthy donors, using two cancers from the same tissue (explanted meningioma, brain metastasis (originated from a breast cancer)), and comparing renal carcinoma tumors and renal juxtatumoral tissue. In the case of renal carcinoma and juxta-tumoral tissue, both ZENITH® and single cell RNA seq analysis were performed in parallel on the same sample. While we focused on the tissue settings, in our results heatmap, we included the EM profile of when the same immune cells where identified in the blood from healthy donors.

[0131] We observed 8 different myeloid populations in meningioma and 6 different subset in renal carcinoma (Data not shown), that were all profiled by ZENITH® (Data not shown). Upon clustering of the different cell subsets based on EM profiling, two groups emerged, a “Glycolytic cluster” and a “Respiratory cluster” (Data not shown). Mono1 and Neutrophils displayed glycolytic metabolism profiles in all blood samples and tumors tested (Data not shown). In contrast, Mono2, DC1 and DC2 showed relatively high glycolytic capacity when isolated from kidney tumor and juxtatumoral tissues, while these subsets showed high respiratory metabolism profile in the two brain tumors. Conversely, tumor-associated macrophages (TAM), showed high mitochondrial dependence, while juxta-tumoral macrophages displayed high glycolytic capacity (Data not shown), suggesting that tumor microenvironment modifies TAM EM. The decrease of glycolytic capacity in TAM as compared to juxta-tumoral macrophages was previously associated with increased immunosuppression in the tumor environment, tumor progression via both nutritional and immunological cues, and poor patient survival.sup.32. These results demonstrate again the analytical capacity and descriminative power of ZENITH® and suggest that in addition to the tumor type, it's anatomical origin could influence the metabolism of immune subsets introducing and additional layer of heterogeneity in the tumor environment. This sensitivity would be unknown using currently available bulk measures.

[0132] Linking scRNA-Seq and Functional Energetic Metabolism Profile in Tumor-Associated Myeloid Cells.

[0133] As these results were not previously observed and the ZENITH® method is new, we also sought to extend and validate the findings, by processing in parallel the same sample using single-cell RNA-seq. We therefore aimed to compare in each population, the functional EM profile obtained by ZENITH® and metabolic gene expression profile obtained by mRNA sequencing. To do this, we first identified specific glyolytic and respiratory gene signature that correlate in with functional metabolism in different myeloid cells in the blood (Data not shown). Then we aimed to tested the expression (mRNA levels) of these glycolytic and respiratory metabolic gene signatures in the different myeloid populations of the tumor. To do so, sorted myeloid cells (CD45+Lin-HLA-DR+) from the renal carcinoma and its juxtatumoral tissue and performed single cell RNA-seq using the 10× Genomics Chromium platform paired with deep sequencing (Data not shown). Analysis of 12,801 cells for the tumor and 2,080 for the juxta tumoral tissue yielded 6 and 5 high quality population clusters respectively. To rigourously identify the myeloid populations we checked the expression of characteristic signatures of these populations33 to establish cellular identities in the tSNE representations. This process allowed us to identify both Mono1, Mono2, and DC clusters (Data not shown). We focused on 5 monocytes and macrophages (clusters expressing MAFB and/or CSF1R) that we monitored by flow cytometry and were present both in tumor and juxta-tumoral tissue (Data not shown). By checking the expression of classical markers (i.e FCGR3A/CD16 and CD14 (Data not shown) we confirmed that clusters 0 and 1 represent CD14+CD16− classical monocytes. Cluster 2 represents CD14−CD16+ non classical monocytes (Mono2), while co-expression of CD14 and CD16 for the clusters 3 and 4, suggest macrophage-like phenotype. Altogether, those results indicate that tumor micro environment specifically modify macrophages metabolism profile functionally and at the transcriptional level. Therefore, single cell RNA sequencing analysis confirmed results obtained by performing ZENITH® on all the different myeloid cell subsets identified (Data not shown). Moreover, in strong agreement with our ZENITH® data by FACS, monocytes clusters (0, 1, 2) presented an enrichment in glycolytic signature both in tumor and juxta tumoral tissue. However, and still in agreement with ZENITH® functional data, macrophages (cluster 3) showed high expression of the respiratory signature in the tumor while this was not detectable in juxta tumoral tissue (Data not shown). As observed for the monocytes, dendritic cells presented an enrichment in glycolytic signature both in tumor and juxta tumoral tissue (Data not shown). Moreover, performing ZENITH® allowed us to identify a functional gene signature identified on myeloid cells sorted from PBMC (Data not shown) that can be extended to myeloid cell sorted from tissue and tumors. Therefore, combinig ZENITH® profiling and single cell RNA sequencing can be used to profile energetic metabolism of a variety of cell type and tissues

DISCUSSION

[0134] ZENITH represents as a novel and rapid technique to evaluate the energetic metabolism profile of cells by flow cytometry at single cell resolution. We showed that this simple profiling method allows a direct integration of energetic metabolism measure with transcriptomic data, in turn permitting immune cell population clustering, analysis of immunostimulatory activity and immunomonitoring in cancer patients. We could show that while human blood monocytes and neutrophiles exhibit a very strong glucose dependency and glycolytic capacity at steady state, dendritic cells have low glycolytic capacity and higher mitochondrial respiratory activity. Upon LPS activation, DCs subsets, undergo a strong switch towards glycolysis and as Dendritic cells mature an increase in their mitochondrial dependency and decrease their glycolytic capacity is observed.

[0135] Altogether, the different glycolytic EM profile observed among blood cells might reflect functional requirements, like for migrating neutrophils, monocytes (Mono1), and moDCs that need to access peripheral tissues that are poorly oxygenized, as compared to the blood stream. EM activity is clearly dependent on a gene expression program that also dictates functional cellular differentiation and is implemented in the cells prior reaching the tissue where they display their effector functions. EM profiling can therefore be used to predict the immune status in patients or organs. Steady state human blood and tonsil DCs or spleen derived DCs migrate to highly irrigated secondary lymphoid tissues, a situation that correlates with their high mitochondrial dependency. Once in secondary lymphoid organs, mature DCs can activate naïve and central memory T cells that circulate and remain most of their life between the blood and lymph stream. We have confirmed here that, while naïve and central memory T cells show at steady state high mitochondrial dependency (more than 50%) and low glycolytic capacity (less than 45%), their EM profile changes dramatically upon activation, showing a decrease in mitochondrial dependency (to less than 20%) and an increase in glycolytic capacity (to more than 80%). When T cells are activated in the lymph nodes, they proliferate, differentiate in highly glycolytic effector T cells and re-enter in the blood circulation with the ability to find inflammed postcapillary venules and extravasate. However, the first metabolic switch occurs even 6 hours after activation, and thus suggest that even in secondary lymphoid organs, such as the spleen there are zones or situations where high glycolytic potential is required or associated with efficient immune response.

[0136] ZENITH will enable researchers to harness the high throughput power of multiparametric flow cytometry that is used for diagnostics in many fields of human medicine (oncology, immunology, infectology, etc), without the need for additional specialized instruments and trained staff to be implemented. The single cell resolution of ZENITH allows for the design a screening strategy based on gene inactivation or silencing to identified genes or cells affecting Energetic matabolism. For example CAS9 based knock out screenings could be used to identify new genes involved in the regulation of the metabolic profile in dendritic cells. The capacity to sort single cells that display a particular metabolic profile and perform next generation sequencing, will for the first time allow the identification of novel genes involved in metabolism and its regulation ex-vivo in a cell type specific manner.

[0137] This method can also be used to analyze samples by Flow cytometry but also, to perform ex-vivo imaging of live tissue slices or whole organs by fluorescent microscopy. This will allow us to combine our current knowledge of anatomopathology, with information about metabolism in different areas and cellular activities in the compromised tissue. Ultimately this method has the potential to become a routine pathology study that can bring important information for clinicians with the need of choosing the best treatment against cancer and other diseases.

[0138] We plan to perform ex-vivo ZENITH to perform imaging of live tissue slices or whole embryo, organs and tumors by fluorescent microscopy. We will apply this technology to study transgenic mice that express reporter fluorescent protein in immune cells subpopulation that cannot be studied by current available techniques because their attachment to the tissue and membrane composition make it difficult to isolate as intact single cell suspensions.

[0139] The field of immunometabolism in oncology has very promising applications. This technology will enable to determine if the metabolic profile of tumor cells and immune cells infiltrating the tumor is informative to predict tumor progression. Transcriptomic data obtained from several tumor sets suggest that the expression pattern of genes involved in tumor metabolism, is a better predictor of tumor progression that the level or kind of immune cells infiltrating the tumor. Thus, ZENITH has the potential to become a method used for diagnostics and immunomonitoring in human oncology, an area of human medicine in most urgent need for novel therapies and markers.

[0140] ZENITH has the potential to be applied in many basic and translational research studies, especially in the interphase between oncology, immunology and immunometabolism. It can be used to analyze blood cells, secondary lymphoid organs and tumors and tumor-infiltrating cells and in neurobiology. This method can be used not only to analyze samples by Flow cytometry but also, to perform ex-vivo imaging of live tissue slices or whole organs by fluorescent microscopy. Ultimately, it has the potential to become a routine pathology study that can bring important information for clinicians with the need of choosing the best treatment against cancer and other diseases. Indeed flow cytometers are available in most research institutes and hospitals, ZENITH represents the most accessible method for functional metabolic profiling and presents several advantages regarding to sensitivity, accesibility, single cell resolution, stability of the readout, time required, compatibility with fixation and sorting compared to other methods. Importantly, ZENITH can establish the metabolic profile of very infrequent cells, as exemplified by the analysis of early effector T cells that represent around 5% of the total T cells isolated from blood (500 cells), thus representing a gain of sensitivity of aproximatively 800 fold compared to Seahorse® measurements (400.000 in triplicates).

[0141] Given the direct relationship between EM and the functionality of lymphoid effector cells and myeloid cells, ZENITH analysis could be used to define the ‘immune EM contexture’ and complement the establishment of an immunoscore that's define immune fitness of tumours and predict and stratify patients for tailored therapies.

REFERENCES

[0142] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure. [0143] Assmann, N., and Finlay, D. K. (2016). Metabolic regulation of immune responses: therapeutic opportunities. J Clin Invest 126, 2031-2039. [0144] Buttgereit, F., and Brand, M. D. (1995). A hierarchy of ATP-consuming processes in mammalian cells. Biochem J 312 (Pt 1), 163-167. [0145] Cheng, S. C., Quintin, J., Cramer, R. A., Shepardson, K. M., Saeed, S., Kumar, V., Giamarellos-Bourboulis, E. J., Martens, J. H., Rao, N. A., Aghajanirefah, A., et al. (2014). mTOR- and HIF-lalpha-mediated aerobic glycolysis as metabolic basis for trained immunity. Science 345, 1250684. [0146] Connolly, N. M., Dussmann, H., Anilkumar, U., Huber, H. J., and Prehn, J. H. (2014). Single-cell imaging of bioenergetic responses to neuronal excitotoxicity and oxygen and glucose deprivation. J Neurosci 34, 10192-10205. [0147] Dougan, M., and Dranoff, G. (2009). Immune therapy for cancer. Annu Rev Immunol 27, 83-117. [0148] Everts, B., Amiel, E., Huang, S. C., Smith, A. M., Chang, C. H., Lam, W. Y., Redmann, V., Freitas, T. C., Blagih, J., van der Windt, G. J., et al. (2014). TLR-driven early glycolytic reprogramming via the kinases TBK1-IKKvarepsilon supports the anabolic demands of dendritic cell activation. Nat Immunol 15, 323-332. [0149] Everts, B., Amiel, E., van der Windt, G. J., Freitas, T. C., Chott, R., Yarasheski, K. E., Pearce, E. L., and Pearce, E. J. (2012). Commitment to glycolysis sustains survival of NO-producing inflammatory dendritic cells. Blood 120, 1422-1431. [0150] Ganeshan, K., and Chawla, A. (2014). Metabolic regulation of immune responses. Annu Rev Immunol 32, 609-634. [0151] Guilliams, M., and Malissen, B. (2015). A Death Notice for In-Vitro-Generated GM-CSF Dendritic Cells? Immunity 42, 988-990. [0152] Helft, J., Bottcher, J., Chakravarty, P., Zelenay, S., Huotari, J., Schraml, B. U., Goubau, D., and Reis e Sousa, C. (2015). GM-CSF Mouse Bone Marrow Cultures Comprise a Heterogeneous Population of CD11c(+)MHCII(+) Macrophages and Dendritic Cells. Immunity 42, 1197-1211. [0153] Helft, J., Bottcher, J. P., Chakravarty, P., Zelenay, S., Huotari, J., Schraml, B. U., Goubau, D., and Reis, E. S. C. (2016). Alive but Confused: Heterogeneity of CD11c(+) MHC Class II(+) Cells in GM-CSF Mouse Bone Marrow Cultures. Immunity 44, 3-4. [0154] Jewett, M. C., Miller, M. L., Chen, Y., and Swartz, J. R. (2009). Continued protein synthesis at low [ATP] and [GTP] enables cell adaptation during energy limitation. J Bacteriol 191, 1083-1091. [0155] Kelly, B., and O'Neill, L. A. (2015). Metabolic reprogramming in macrophages and dendritic cells in innate immunity. Cell Res 25, 771-784. [0156] Lindqvist, L M., Tandoc, K., Topisirovic, I., and Furic, L. (2017). Cross-talk between protein synthesis, energy metabolism and autophagy in cancer. Curr Opin Genet Dev 48, 104-111. [0157] Liu, H., Kurtoglu, M., Leon-Annicchiarico, C. L., Munoz-Pinedo, C., Barredo, J., Leclerc, G., Merchan, J., Liu, X., and Lampidis, T. J. (2016). Combining 2-deoxy-D-glucose with fenofibrate leads to tumor cell death mediated by simultaneous induction of energy and ER stress. Oncotarget 7, 36461-36473. [0158] Lutz, M. B., Inaba, K., Schuler, G., and Romani, N. (2016). Still Alive and Kicking: In-Vitro-Generated GM-CSF Dendritic Cells! Immunity 44, 1-2. [0159] Lutz, M. B., Strobl, H., Schuler, G., and Romani, N. (2017). GM-CSF Monocyte-Derived Cells and Langerhans Cells As Part of the Dendritic Cell Family. Front Immunol 8, 1388. [0160] MacIver, N. J., Michalek, R. D., and Rathmell, J. C. (2013). Metabolic regulation of T lymphocytes. Annu Rev Immunol 31, 259-283. [0161] Marquez, S., Fernandez, J. J., Teran-Cabanillas, E., Herrero, C., Alonso, S., Azogil, A., Montero, O., Iwawaki, T., Cubillos-Ruiz, J. R., Fernandez, N., et al. (2017). Endoplasmic Reticulum Stress Sensor IRElalpha Enhances IL-23 Expression by Human Dendritic Cells. Front Immunol 8, 639. [0162] Merad, M., Sathe, P., Helft, J., Miller, J., and Mortha, A. (2013). The dendritic cell lineage: ontogeny and function of dendritic cells and their subsets in the steady state and the inflamed setting. Annu Rev Immunol 31, 563-604. [0163] Miller, A., Nagy, C., Knapp, B., Laengle, J., Ponweiser, E., Groeger, M., Starkl, P., Bergmann, M., Wagner, O., and Haschemi, A. (2017). Exploring Metabolic Configurations of Single Cells within Complex Tissue Microenvironments. Cell Metab 26, 788-800 e786. [0164] Neijssel, O. M., M. J. Teixeira de Mattos, and D. W. Tempest (1996). Growth yield and energy distribution. [0165] Newick, K., O'Brien, S., Moon, E., and Albelda, S. M. (2017). CAR T Cell Therapy for Solid Tumors. Annu Rev Med 68, 139-152. [0166] Oren, R., Farnham, A. E., Saito, K., Milofsky, E., and Karnovsky, M. L. (1963). Metabolic patterns in three types of phagocytizing cells. J Cell Biol 17, 487-501. [0167] Page, D. B Postow, M. A., Callahan, M. K., Allison, J. P., and Wolchok, J. D. (2014). Immune modulation in cancer with antibodies. Annu Rev Med 65, 185-202. [0168] Schimmel, P. (1993). GTP hydrolysis in protein synthesis: two for Tu? Science 259, 1264-1265. [0169] Schmidt, E. K., Clavarino, G., Ceppi, M., and Pierre, P. (2009). SUnSET, a nonradioactive method to monitor protein synthesis. Nat Methods 6, 275-277. [0170] Steinman, R. M. (2012). Decisions about dendritic cells: past, present, and future. Annu Rev Immunol 30, 1-22. [0171] Steinman, R. M., Hawiger, D., and Nussenzweig, M. C. (2003). Tolerogenic dendritic cells. Annu Rev Immunol 21, 685-711. [0172] Suganya, R., Chakraborty, A., Miriyala, S., Hazra, T. K., and Izumi, T. (2015). Suppression of oxidative phosphorylation in mouse embryonic fibroblast cells deficient in apurinic/apyrimidinic endonuclease. DNA Repair (Amst) 27, 40-48. [0173] Villani, A. C., Satija, R., Reynolds, G., Sarkizova, S., Shekhar, K., Fletcher, J., Griesbeck, M., Butler, A., Zheng, S., Lazo, S., et al. (2017). Single-cell RNA-seq reveals new types of human blood dendritic cells, monocytes, and progenitors. Science 356. [0174] Wallace, D. C., Fan, W., and Procaccio, V. (2010). Mitochondrial energetics and therapeutics. Annu Rev Pathol 5, 297-348. [0175] Wolf, A. J., Reyes, C. N., Liang, W., Becker, C., Shimada, K., Wheeler, M. L., Cho, H. C., Popescu, N. I., Coggeshall, K. M., Arditi, M., et al. (2016). Hexokinase Is an Innate Immune Receptor for the Detection of Bacterial Peptidoglycan. Cell 166, 624-636. [0176] Zhang, J., Nuebel, E., Wisidagama, D. R., Setoguchi, K., Hong, J. S., Van Horn, C. M., Imam, S. S., Vergnes, L., Malone, C. S., Koehler, C. M., et al. (2012). Measuring energy metabolism in cultured cells, including human pluripotent stem cells and differentiated cells. Nat Protoc 7, 1068-1085. [0177] Zhang, L., Su, J., Xie, Q., Zeng, L., Wang, Y., Yi, D., Yu, Y., Liu, S., Li, S., and Xu, Y. (2015). 2-Deoxy-d-Glucose Sensitizes Human Ovarian Cancer Cells to Cisplatin by Increasing ER Stress and Decreasing ATP Stores in Acidic Vesicles. J Biochem Mol Toxicol 29, 572-578. [0178] Zhang, P., McGrath, B., Li, S., Frank, A., Zambito, F., Reinert, J., Gannon, M., Ma, K., McNaughton, K., and Cavener, D. R. (2002). The PERK eukaryotic initiation factor 2 alpha kinase is required for the development of the skeletal system, postnatal growth, and the function and viability of the pancreas. Mol Cell Biol 22, 3864-3874.