Conceived and designed the experiments: JT DLK. Performed the experiments: JT JY HND W-SC MPGD CKYY. Analyzed the data: JT HND JY. Wrote the paper: JT DLK.
Drs. Jide Tian and Daniel L. Kaufman are co-inventors of a patent titled “GABA Modulates Inflammation”, but both had no patent-related activity of employment, consultancy, products in development or modified products. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials. The authors declare no other conflicts of interest.
Adipocyte and β-cell dysfunction and macrophage-related chronic inflammation are critical for the development of obesity-related insulin resistance and type 2 diabetes mellitus (T2DM), which can be negatively regulated by Tregs. Our previous studies and those of others have shown that activation of γ-aminobutyric acid
Obesity is associated with the development of type 2 diabetes mellitus (T2DM) and stems from the imbalance between calorie intake and expenditure, as well as genetic factors, leading to the accumulation of excess fat in the body. T2DM is characterized by impaired glucose tolerance, insulin resistance and insufficient insulin production by the pancreatic islet β-cells
Previous studies have shown that adipocyte and β-cell dysfunction along with low-grade macrophage-related chronic inflammation are critical for the development of obesity-related insulin resistance and T2DM
To test this hypothesis, we employed the HFD-induced obesity and T2DM model and treated orally with GABA to test whether activation of GABA receptors could prevent the HFD-induced obesity and T2DM, and improve glucose tolerance and insulin sensitivity after the onset of T2DM. We found that oral administration of GABA did not affect the amount of water and food consumption by animals, but reduced the HFD-induced gain in body weight and epididymal fat mass, accompanied by reducing the numbers of infiltrated macrophages in adipose tissues. Furthermore, treatment with GABA improved glucose tolerance and insulin sensitivity in mice, even after the onset of obesity and T2DM. Finally, we observed that treatment with GABA significantly increased the frequency of peripheral Tregs in mice, which are known to negatively regulate inflammation.
All experiments were approved by the Animal Research Committee of University of California, Los Angeles (protocol number 1993-211). Male C57BL/6 mice at 4 weeks of age were from Jackson Laboratories (Bar Harbor, ME, USA) and were housed in a specific pathogen free facility. The mice were fed with HFD beginning at 5 weeks of age (60% fat of caloric intake at 5.32 kcal/g, Research Diets, New Brunswick, USA) and provided with plain water (control) or water containing 2 mg/ml of GABA (Sigma, St. Louis, USA). The water bottles were changed weekly with fresh material. Their food intake and water consumption were measured weekly and their body weights, fasting blood glucose, glucose tolerance, and insulin sensitivity were measured longitudinally.
In addition, we tested whether treatment with GABA could modulate glucose intolerance and insulin resistance in mice after the onset of obesity and T2DM. C57BL/6 mice were fed with HFD and plain water for 20 weeks and tested for fasting blood glucose, glucose tolerance, and insulin sensitivity. Individual mice with obesity (body weights > 48 g) and T2DM (fasting blood glucose >145 mg/dL; blood glucose at 2 hours post-intraperitoneal glucose challenge >220 mg/dL; blood glucose at 30 min post insulin injection < 28% before insulin injection) were randomly divided and fed with HFD and plain water or water containing 2 mg of GABA for another 12 weeks, respectively. Their food intake and water consumption were measured weekly and their body weights, fasting blood glucose, glucose tolerance, and insulin sensitivity were measured longitudinally.
The mice were fasted for 16 hours and the concentrations of blood glucose in individual mice were measured using an OneTouch ultra blood glucose meter (LifeScan, Milpitas, USA). Glucose tolerance in individual mice was injected intraperitoneally with 2 g/kg glucose (Sigma) and their blood glucose concentrations were measured at 15, 30, 60, 90, 120 and 180 min post challenge and the areas of under the curve (AUC) of blood glucose levels were calculated. For insulin tolerance tests, non-fasted individual mice were measured for the concentrations of blood glucose as 0 time point and injected intraperitoneally with 0.75 units/kg of human regular insulin (Eli Lilly, Indianapolis, USA). Subsequently, the concentrations of blood glucose were measured at 15, 30, 60, 90, 120 and 150 min post insulin injection. The changes in the percentage of blood glucose relative to the values at 0 time point for individual mice were calculated.
At the end of the experiments, mice were sacrificed and their visceral adipose tissue (VAT) were dissected and weighed. One portion of the VAT was fixed with 4% paraformaldehyde for 24 hours and the VAT sections at 5 µm were stained with Toluidine blue O. A total 5 sections from individual mice of each group (n = 8) were examined in a blinded fashion using the Lplab image analysis software.
Another portion of the VAT was fixed with the Bouin buffer for 48 hours and then paraffin-embedded. Their sections at 5 µm of 100 µm intervals were rehydrated, treated with 3% of H202 in methanol, and subjected to antigen retrieval with 0.01 M sodium citrate buffer (pH 6.0) in a high pressure-steamer. Subsequently, these sections were blocked sequentially with a biotin blocking system (Dako) and 20% of FBS in PBS. The sections were probed with biotinylated anti-F4/80 (PharMigen, San Diego, USA) overnight at 4°C. After washing, the immunocomplex was visualized using HRP-striptoavidin (Dako) and the peroxidase substrate diaminobenzidine (DAB, Dako), followed by counterstaining with hematoxylin and examination under a light microscope. A total of 400 nuclei from 5 fields of 5 sections of each mouse were counted in a blinded manner and the percentage of macrophage infiltrates was calculated.
Groups of C57BL/6 mice were fed with plain water or water containing 2 mg/ml of GABA for 4 weeks. Their splenic mononuclear cells were prepared and the frequency of CD4+Foxp3+ Tregs was characterized by flow cytometry analysis. Briefly, splenic mononuclear cells (106/tube) were treated in duplicate with anti-CD16/32 on ice for 20 minutes. After washing, the cells were stained with FITC-anti-CD4 (BD PharMigen, San Diego, USA) for 25 minutes, fixed, permeabilized and stained with PE-anti-Foxp3 or isotype control (Biolegend, San Diego. USA), respectively. After washing, the cells were characterized by flow cytometry analysis and the percentage of CD4+Foxp3+ cells in total CD4+ T cells was calculated.
Data are expressed as mean or mean ± SD of each group of mice. The difference between groups was analyzed by Students' t-test and X2 test. The potential correlation between measures was analyzed by Sperman's correlation analysis. A p value of less than 0.05 was considered statistically significant.
GABAA-R subunits are expressed by immunocompetent cells, including T cells and macrophages and are also expressed by adipose tissues
Groups of C57BL/6 mice were fed with HFD and provided with plain water or water containing 2 mg/ml of GABA for 20 weeks. The amounts of water and food consumed by individual mice and their body weights were measured longitudinally. Data are expressed as mean values of water and food consumed by individual mice per day and mean values of body weights of each group of mice (n = 20 per group) at the indicated time points post-HFD. Intragroup variations were less than 10% for the amounts of water and food consumption and less than 15% for the values of body weights. (A) The amounts of water consumed; (B) The amounts of food consumed; and (C) The body weights. *p< 0.05; ** p<0.01 vs. the GABA-fed mice.
During the study period, we measured the concentrations of fasting blood glucose, glucose tolerance and insulin sensitivity in the water-fed control and GABA-fed mice longitudinally. We found that the concentrations of fasting blood glucose gradually increased in control mice and some mice at 20 weeks post HFD developed T2DM. In contrast, the concentrations of fasting blood glucose in mice that received GABA remained at a similar level throughout the observation period (
Groups of C57BL/6 mice were fed with HFD and given plain water or water containing 2 mg/ml of GABA for 20 weeks. The concentrations of fasting blood glucose, intraperitoneal glucose tolerance (IPGT), and insulin sensitivity of individual mice were tested longitudinally, as described in the
To test whether treatment with GABA could modulate glucose tolerance and insulin sensitivity in mice after the onset of obesity and T2DM, C57BL/6 mice were fed with HFD for 20 weeks and their fasting glucose, glucose tolerance, and insulin sensitivity were measured. Individual mice with obesity (body weight > 48 g) and T2DM (abnormal fasting blood glucose and impaired glucose tolerance and insulin sensitivity, see
C57BL/6 mice were fed with HFD for 20 weeks and their fasting blood glucose and IPTG were measured. Individual mice with body weight >48g, fasting glucose level of >145 mg/dL and abnormal IPTG (blood glucose >220 mg/dL at 2 hours post-IPGT) were considered to have obesity and T2DM, and randomly treated orally with plain water (control) or water containing 2 mg/ml of GABA for another 12 weeks. Their body weights, fasting glucose concentrations, IPGT and insulin sensitivity were measured longitudinally. (A) The body weights; (B) The concentrations of fasting blood glucose; (C) The dynamic changes following glucose challenge; and (D) The percentages of blood glucose concentrations following insulin challenge. Data are expressed as mean values of each group (n = 4 per group) at the indicated time points and the concentrations of blood glucose before insulin challenge in individual mice were used as 100%. *p< 0.05; ** p<0.01 vs. the GABA-fed mice.
Previous studies have shown that macrophage-related chronic inflammation in adipose tissues is crucial for the development of insulin resistance, obesity, and T2DM
Groups of C57BL/6 mice were fed with HFD and provided with plain water or water containing 2 mg/ml of GABA for 20 weeks. The mice were sacrificed and the amounts of VAT in individual mice were measured. One portion of the VAT was fixed with 4% paraformaldehyde for 24 hours and the VAT sections at 5 µm were stained with Toluidine blue O, followed by examination under a light microscope. The sizes of adipocytes in 5 sections of individual mice from each group (n = 8) were examined in a blinded fashion. Another portion of VAT was fixed with the Bouin buffer for 48 hours and the VAT sections were subjected to immunohistochemistry analysis of infiltrated macrophages using anti-F4/80 antibodies and DAB substrate (brown), and the percentages of macrophages in 400 nuclear cells from 5 sections of each mouse in individual groups of mice were quantified in a blinded manner. Data are representative images of the adipocytes, macrophages stained and expressed as mean ± SEM from each group of mice (n = 8 for immunohistological examination and n = 12 for measuring the amounts of VAT per group). *p< 0.05; ** p<0.01 vs. the GABA-fed mice.
The Foxp3+ Tregs are negative regulators of inflammatory responses and recent studies have indicated that Tregs can inhibit obesity-related inflammation and insulin resistance in mice
C57BL/6 mice were fed with plain water or water containing GABA (2mg/ml) for four weeks. Their splenic mononuclear cells were prepared and treated with anti-CD16/32. Subsequently, splenic mononuclear cells (106/tube) were stained in duplicate with FITC-anti-CD4 and after washing, the cells were fixed, permeabilized and stained with PE-anti-Foxp3, followed by flow cytometry analysis. The cells were stained with single fluorescent-labeled antibodies or with isotype-matched IgG as controls. Data are representative charts and expressed as mean ± SEM of the percentage of CD4+Foxp3+ Tregs in splenic CD4+ T cells of different groups of mice (n = 3–4 mice per group) from two separate experiments. *p<0.01 vs. the water-fed mice.
Macrophage-related chronic inflammation in adipose tissues is associated with the development of insulin resistance and glucose intolerance, leading to the development of obesity and T2DM. Our previous studies, and those of others, have shown that GABA, through the activation of GABAA-R, inhibits inflammation in mouse models of autoimmune diseases
Furthermore, while most of the HFD-fed control mice displayed a significant increase in body weights, the mice that had been fed with HFD and GABA had significantly less gain in body weights over the observation period. Over-consumption of calories can cause obesity and lead to the development of insulin resistance and T2DM
Recent studies have shown that adipose tissue-related chronic inflammation contributes to the development of insulin resistance, a key component of metabolic syndrome, and leads to the development of obesity and T2DM as well as other metabolic disorder-related diseases
Notably, Tregs are potent inhibitors for macrophage activation and function
In summary, our data demonstrated that oral treatment with GABA inhibited the HFD-induced obesity and improved glucose intolerance and insulin sensitivity, even after the establishment of obesity and T2DM in mice. Furthermore, oral treatment with GABA reduced the HFD-induced adipocyte hypertrophy and adipose tissue mass, accompanied by significantly reduced macrophage infiltrates in the adipose tissues. Furthermore, we found that GABA treatment increased the frequency of splenic Tregs in mice. Apparently, GABA, through its GABAA-Rs on adipocytes, macrophages and T cells, inhibited chronic inflammation in adipose tissues, leading to the improvement of glucose tolerance and insulin sensitivity in HFD-fed mice. Given that GABA mainly acts on the peripheral GABA receptors and is safe for human consumption, GABA and other GABAA-R agonists may be valuable for the prevention and treatment of obesity and T2DM in the clinic.