C57BL/6JJcl mice (CLEA Japan, Tokyo, Japan) and B6.CB17-Prkdcscid/SzJ (SCID) mice (The Jackson Laboratory, Bar Harbor, ME) were maintained with free access to chow and water. The mice were fed standard chow diet (STD) (MF, Oriental Yeast, Japan) or high fat-high sucrose diet (HFS: Protein: 16.4% kcal, Fat: 58% kcal, Carbohydrate: 25.5% kcal, Energy Density: 5.56 kcal/g) (D12331, Research Diets, New Brunswick, NJ).
For tumor growth assay, 4-week-old C57BL/6JJcl and B6.CB17-Prkdcscid/SzJ male mice were fed with STD or HFS for 6 weeks and injected with 4 × 105 B6 OVA-gene introduced B16 melanoma MO5 cells. Seven days after injection of MO5 cells, they were divided into 4 groups fed with (1) STD with water (STD), (2) STD with 5 mg/dl metformin (WAKO, Tokyo, Japan) water (STD + Met), (3) HFS with water (HFS), and (4) HFS with 5 mg/dl metformin water (HFS + Met) for 4 weeks. Mice that received metformin at 5 mg/ml achieved 1.70 μg/mL, which was consistent with plasma concentrations in patients (0.5–2 μg/mL)39. Tumor infiltrating lymphocytes (TILs) were isolated at 48 h after the administration of metformin and subjected to cytokine assay.
4-week-old C57BL/6JJcl mice were fed with STD and HFS for 12 weeks and they were divided into STD, STD + Met, HFS, and HFS + Met for 2 weeks. Then, splenocytes were isolated and subjected to cytokine assay and Extracellular Flux Analyzer.
Tumor growth assay
The mice were intradermally inoculated with 4 × 105 B6 OVA-gene introduced B16 melanoma MO5 cells in 100 μl of RPMI on the back with a 27-gauge needle after the hair shaved. The tumor diameter was measured with Vernier calipers in every 3 days until 25 days after the injection, and the volume of the tumor was calculated using a formula V = W2 × L)/2 (V = volume, W = Minor axis, L = Major axis).
Oral glucose tolerance test (OGTT)
The 4 groups of 16-week-old C57BL/6JJcl mice under STD, STD + Met, HFS, and HFS + Met for 12 weeks were fasted for 16 h and subjected to Oral Glucose Tolerance Test (OGTT). A glucose (2 g/kg) solution was given and blood samples were taken from tail veins at basal, 30, 60, and 120 min after the administration of the glucose solution.
Isolated mouse splenocytes without red blood cells or tumor infiltrating lymphocytes (TILs) were stimulated in the presence of 50 ng/ml Phorbol 12-Myristate 13-acetate (PMA) (Sigma-Aldrich, St. Louis, MO) and 2 µM Ionomycin (Sigma-Aldrich), and treated with BD Golgi Stop (BD Biosciences, San Jose, CA) for 6 h at 37 °C in a humidified atmosphere of 5% CO213,40. After the culture, the cells were stained with fluorescent dye and monoclonal antibodies (mAbs): Zombie Aqua, CD3 (BV521), and CD8 (APC/Fire) mAbs (BioLegend, San Diego, CA). Next, the cells were resuspended in BD Cytofix/Cytoperm solution (BD Biosciences). The cells were further stained in BD Perm/Wash buffer with IFN-γ (FITC), TNF-α (BV421), and IL-2 (PE/Cy7) mAbs. Finally, these cells were analyzed by FACS Canto II flow cytometer (BD Biosciences).
Human peripheral blood mononuclear cells (PBMCs) were stimulated in the presence of 50 ng/ml PMA and 2 µM Ionomycin, and then treated with BD Golgi Stop for 6 h at 37 °C in a humidified atmosphere of 5% CO2. After the culture, the cells were stained with fluorescent dye and monoclonal antibodies (mAbs): Zombie Aqua (BV521), CD8 (APC/Fire), PD-1 (PE/Cy7) mAbs (BioLegend, San Diego, CA). Next, the cells were resuspended in BD Cytofix/Cytoperm solution (BD Biosciences). The cells were further stained in BD Perm/Wash buffer with IFN-γ (FITC), TNF-α (BV421), and IL-2 (APC) mAbs. Finally, these cells were analyzed by FACS Canto II flow cytometer (BD Biosciences).
Extracellular acidification rate (ECAR) and oxygen consumption rate (OCR)
CD8 + T cells were purified from mouse splenocytes as well as human peripheral blood mononuclear cells (PBMCs) from the patients with T2D (DM) and subjects with normal glucose tolerance (NGT) using MACS (Magnetic-activated cell sorting). The CD8 + T cells were suspended in XF medium (non-buffered RPMI 1640 containing 10 mM glucose, 2 mM L-glutamine, and 1 mM sodium pyruvate) and 7 × 105 cells seeded in each well. Then, ECAR and OCR were measured with XFe96 Extracellular Flux Analyzer (Agilent Technologies, Santa Clara, CA) by serial injections of following regents at final concentrations of 1 μM oligomycin, 8 μM carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP), 1 μM Rotenone/Antimycin, and 100 mM 2-deoxy-D-glucose (2-DG).
In clinical randomized control trial, CD8 + T cells were isolated from human PBMCs of the patients with T2DM treated with metformin and without metformin in the following clinical study. The CD8 + T cells were suspended in XF medium (non-buffered RPMI 1,640 containing 10 mM glucose, 2 mM L-glutamine, and 1 mM sodium pyruvate) and 7 × 105 cells seeded in each well. Then, ECAR and OCR were measured with XFe96 Extracellular Flux Analyzer.
Metformin and immune exhaustion in type 2 diabetes; randomized control trial (METRO)
To investigate whether the production of multiple cytokines from CD8 + cells is impaired under high glucose conditions in the patients with T2D and whether metformin improves the functional abnormalities of CD8 + cells, clinical trial “Metformin and immune exhaustion in type 2 diabetes; Randomized Control Trial (METRO)” was performed (UMIN000018637). METRO study was approved by Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences and Okayama University Hospital, Ethics Committee (m05004). We recruited the hospital in-patients with T2D, (1) who understood the purpose, methods and side effects of metformin and gave written informed consents, (2) who were able to claim the unwanted symptoms and signs, (3) who were not less than 20 years old and not more than 75 years old, and (4) who did not receive metformin. The patients with lactic acidosis, moderate renal dysfunction, hemodialysis, severe liver dysfunction, shock, heart failure, cardiac infarction pulmonary embolism, respiratory failure, hypoxia, alcohol intake, dehydration, diarrhea with dehydration, vomiting, severe ketosis, diabetic coma and precoma, type 1 diabetes, severe infection, pre- and post-operation, severe trauma, malnutrition, famine, panhypopituitarism, adrenal failure, pregnancy, and allergic to biguanides were excluded.
We randomly assigned the patients with T2D to metformin group, Met(+) (n = 9), non-metformin group, Met(−) (n = 10) (online supplementary Figure S6). In the metformin group, the initial dose was 500 mg/day (250 mg twice daily), and the dose was increased up to 1,500 mg/day (500 mg thrice daily) over 2 weeks by confirming the absence of adverse effects. Any antidiabetic drugs other than metformin were allowed to be used in both groups to achieve the blood glucose control. Blood samples were collected at baseline, 1 week at hospital and 2 months at out-patient clinic, to evaluate the function of CD8 + T cells. Blood samples from subjects with normal glucose tolerance (n = 15) were used as control.
Separation and storage of blood cells
About 20 ml blood samples were collected from the patients with T2D and the subjects with normal glucose tolerance. PBMCs were isolated by density gradient centrifugation using a Ficoll-Paque (GE Healthcare Bio-Sciences, Piscataway, NJ). The 2 × 106 PBMCs were suspended in 500 µl of BAMBANKER (LYMPHOTEC, Tokyo, Japan) media, and the cells are placed in cryotubes for freezing and preservation. Then, the cells are frozen and preserved at -180 °C for further use.
Cytomegalovirus (CMV) assay
Human PBMCs were stimulated with 1 μg/ml HLA-A*24:02 CMV pp65 peptide in AIM-V media (Thermo Fisher Scientific, Waltham, MA) supplemented with 5% pooled serum and 10 U/ml IL-2) on 24-well plates and incubated for 2 weeks at 37 °C in a humidified atmosphere of 5% CO2. Then, the half volume of medium was changed with freshly prepared AIM-V media containing 5% pooled serum and 10 U/ml IL-2 in every 2 days and cultured for additional 1 week (cultured PBMCs). After the removal of CD8 + T cells by CD8 MicroBeads (MiltenyiBiotec), 1.0 × 106 PBMCs were incubated in AIM-V containing 5% pool serum and IL-2 10U/ml in the presence (CMV +) and absence (CMV−) of 1 μg/ml HLA-A*24:02 CMV pp65 peptide for 1 h. The CMV + and CMV− were mixed with equal cell numbers of “cultured PBMCs” and treated with BD Golgi Stop for 6 h, respectively. After the culture, the mixed cells were stained with Zombie (BV521), CD8 (APC/Fire), PD-1 (PE/Cy7), Tim-3 (PE) mAbs. Next, they were resuspended in BD Cytofix/Cytoperm solution. There after cells were stained in BD Perm/Wash solution with IFN-γ (FITC), TNF-α (BV421), IL-2 (APC) mAbs and analyzed by FACS Canto II flow cytometer (BD Biosciences).
All results are expressed as means ± standard deviation (SD). Computations assumed that all groups were samples from populations with the same scatter. The significance was determined by Mann–Whitney U test and two-way repeated measure ANOVA with Bonferroni method or Tukey–Kramer, and analysis of covariance (ancova) for comparison of 2 regression lines using SPSS software. A P value of < 0.05 was considered significant.
The animal experiments were approved by the Animal Care and Use Committee of the Department of Animal Resources, Advanced Science Research Center, Okayama University (OKU-2016295, OKU-2017310, and OKU-2019350). All animal experiments were performed in accordance with relevant guideline and regulations.
For human research, METRO study was approved by Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences and Okayama University Hospital, Ethics Committee (m05004) and performed in accordance with relevant guidelines/regulations.
The informed consent was obtained from all participants.