The Mcpt5-Cre⁺R-DTA⁺ mouse line was established as described previously (20). Mast cell-deficient (Mcpt5-Cre⁺R-DTA⁺, n = 8) and -proficient (Cre-negative R-DTA⁺, n = 10) littermate male mice were subjected to a high-fat diet (HFD, 60% calories from fat) (Research Diets, New Brunswick, NJ, USA) for a time period of 21 weeks. Body weight was recorded on a weekly basis. At week 10 of the HFD feeding, metabolic performance was measured in a subgroup of mice using comprehensive laboratory animal monitoring system (CLAMS). Mice were placed in metabolic cages (PhenoMaster; TSE Systems, Bad Homburg, Germany) with free access to water and food and 12/12 h light/dark cycle for three consecutive days and nights. Volume of oxygen consumption (VO₂) and carbon dioxide production (VCO₂) were determined automatically every 20 min as well as water consumption and food intake. The respiratory exchange ratio (RER) was calculated as VCO₂/VO₂. Motility was recorded by an infrared sensor and measured every 20 min as number of events per period (20 min); averages were calculated for day and night periods. Data were normalized with respect to body weight using analysis of covariance (ANCOVA). Cholesterol and triglycerides levels were measured in blood after 16 h of starvation with AccuTrend (Roche, Mannheim, Germany). Glucose tolerance test (GTT) and insulin tolerance test (ITT) were performed as previously described (21, 22). GTT was performed after 16 h of starvation after 18 weeks on HFD by injecting glucose (1 g/kg of bodyweight) i.p. and by measuring glucose in blood with glucometer (Accu-Chek, Roche, Mannheim, Germany) at different time points, as shown in the figures. ITT was performed after 6 h of starvation after 19–20 weeks on HFD by injecting insulin 1.5 IU/kg of bodyweight i.p. and glucose was measured at different time points, as shown in the figures. After euthanizing mice, sera, subcutaneous and gonadal AT (sAT and gAT), liver, pancreas, and muscle (quadriceps) were collected and used either directly or stored at −80°C for further analysis. Animal experiments were approved by the Landesdirektion Sachsen, Germany.

Stromal vascular fraction (SVF) was isolated as described previously (23, 24). Briefly, sAT and gAT were isolated and immediately minced and digested with type I collagenase (Gibco, Darmstadt, Germany). Digested AT was filtered through 100 μm cell strainer (BD, Heidelberg, Germany), and the adipocyte fraction was removed by centrifugation. SVF cells in the pellet were washed and resuspended in FACS buffer (PBS, 0.1%BSA, 0.1%NaN₃) and analyzed for the content of inflammatory and immune cells by using flow cytometry (FACS Canto II, BD Bioscience). Mφs were detected as CD11b⁺ F4/80⁺; M1 Mφs were detected as CD11b⁺ F4/80⁺ CD11c⁺ cells. Total T cells (CD3⁺) were further divided into T helper (CD3⁺ CD4⁺) and cytotoxic T cells (CD3⁺ CD8⁺). Fc receptors on leukocytes were blocked by anti-CD16/CD32 antibody (BD, Heidelberg, Germany). The following fluorescently labeled antibodies were used: CD11b-APC, CD11c-PE (BD, Heidelberg, Germany), F4/80-Alexa-fluor 488 (eBioscience, Frankfurt, Germany), CD8a-APC, CD4-FITC, CD3e-PE (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany).

Plasma was obtained from heparinized blood from mice that were starved for 16 h. Insulin was measured in plasma by using insulin immunoassay (Crystal Chem, Cologne, Germany). After killing mice, the serum was collected and analyzed for adiponectin by using an immunoassay (R&D Systems, Wiesbaden-Nordenstadt, Germany).

Total RNA was isolated from AT and liver by using phenol/chlorophorm extraction (TRI Reagent, MRC, Cincinnati, OH, USA). The expression of inflammatory (IL-1β, IL-6, IL-10, TNF, MCP-1, and F4/80) and metabolic (PPARγ, Srebp1c, LPK, GK, G6P, FATP2, CD36, MTTP and Glut-2) genes (for detailed description see Mast Cell Deficiency Does Not Affect Obesity-Related Liver Steatosis and Inflammation section in Results) was analyzed using SsoFast EvaGreen Supermix (BioRad) on real-time PCR cycler CFX384 (BioRad, Munich, Germany). The expression was normalized to Cre- (mast cell proficient) group by using 18S RNA as a reference and quantified with ΔΔCt method (25). Primers used in the study: mTNF-a_F – AGCCCCCAGTCTGTATCCTTCT, mTNF-a_R – AAGCCCATTTGAGTCCTTGATG, mIL-10_F – TAAGGCTGGCCACACTTGAGA, mIL-10_R – AGCTGCTGCAGGAATGATCA, mF4/80_F – TCAAGGCCATTGCCCAGAT, mF4/80_R – TCCCGTACCTGACGGTTGAG, mIL-1β_F – ATCCCAAGCAATACCCAAAG, mIL-1β_R – GTGCTGATGTACCAGTTGGG, mMCP-1_F – GCATCTGCCCTAAGGTCTTC, mMCP-1_R – AAGTGCTTGAGGTGGTTGTG, mPPARg_F – GAGTGTGACGACAAGATTTG, mPPARg_R – GGTGGGCCAGAATGGCATCT, Srebp1c_F – GATCAAAGAGGAGCCAGTGC, Srebp1c_R – TAGATGGTGGCTGCTGAGTG, mGK_F – TGCGGAGATGCTCTTTGACT, mGK_R – TCTCGGAGAAGTCCCACGAT, mLPK_F – CTTGCTCTACCGTGAGCCTC, mLPK_R – ACCACAATCACCAGATCACC, mG6Pase_F – TGGAGTCTTGTCAGGCATTG, mG6Pase_R – TCCAAAGTCCACAGGAGGTC, mFATP2_F – CTACGCATCCACTGAAGGCA, mFATP2_R – AGTCCAACCTCACCTTTGGG, mCD36_F – AGGTCTATCTACGCTGTGTTC, mCD36_R – ATGGTTGTCTGGATTCTGGAG, Mttp_F – CACTCAGGCAATTCGAGACA, Mttp_R – TCTGGCTGAGGTGGGAATAC, mGlut2_F – ATTCGCCTGGATGAGTTACG, mGlut2_R – CCAGCGAAGAGGAAGAACAC, m18S rRNA_F – GTTCCGACCATAAACGATGCC, and m18S rRNA_R – TGGTGGTGCCCTTCCGTCAAT.

Paraffin stocks were prepared from gAT and liver, 5 μm thick cuts were transferred onto slides and were stained with Mayer’s hematoxylin (SAV, Flintsbach a. Inn, Germany) and counterstained with 1% eosin (Seipt, Germany) for the analysis of adipocyte size and liver steatosis. Mast cell accumulation in gAT was detected by using Giemsa staining (Sigma-Aldrich, Munich, Germany). Steatosis, lobular inflammation, and hepatocellular ballooning were evaluated according to previously published criteria, following the NASH-CRN Committee scoring system (26). Steatosis and inflammation were scored using a 0–3 scale, ballooning by using a 0–2 scale. The NAFLD activity score (NAS) was defined as the sum of steatosis, lobular inflammation, and ballooning, thus ranging from 0 to 8. A microscope coupled to a computerized system (Zeiss, Oberkochen, Germany) and equipped with the AxioVision Rel. 4.8 software (Carl-Zeiss MicroImaging GmbH, Jena, Germany) was used.

Student’s t-test was used for statistical analysis of most experiments, non-parametrical Mann–Whitney U test was used for quantitative Real-Time PCR (qPCR) evaluation and ANCOVA, with respect to mouse bodyweight, was used for analysis of data from metabolic cages. All data are expressed as means ± SEM; the level of significance was set at p < 0.05.

Crossing of Mcpt5-Cre transgenic mice (19) to the R-DTA line (27) results in profound deficiency for connective tissue mast cells, the subset of mast cells populating most tissues, including AT, due to selective suicidal expression of diphtheria toxin A in Mcpt5-Cre⁺R-DTA⁺ animals. Lack of connective tissue mast cells is reflected by absence of IgE-mediated anaphylaxis, whereas the numbers of other major immune cell types are not affected (28). We assessed the involvement of mast cells in diet-induced obesity-related metabolic dysregulation. First, a group of mast cell-deficient and mast cell-proficient littermate control mice was followed on standard diet for >15 weeks. Under these conditions, mast cell-deficient mice displayed no differences with regards to body weight, AT and liver weight, glucose tolerance, and further metabolic parameters, e.g., blood cholesterol, blood triglycerides, or blood insulin, as compared to controls (data not shown). We, then, performed a detailed analysis of mice in the course of HFD-induced obesity. In contrast to Kit^W-sh/W-sh mice (12), but similarly to the Cpa3^Cre+ mice (16), Mcpt5-Cre⁺R-DTA⁺ mast cell-deficient and -proficient (Cre-negative R-DTA⁺) mice that were fed a HFD did not display any significant differences in body weight (Figure 1A), body weight gain (Figure 1B), glucose tolerance (Figure 1C), and insulin resistance (Figure 1D). Thus, mast cell-deficiency does not affect systemic metabolic dysregulation and development of insulin resistance due to obesity. In accordance with the body weight data, the weights of the sAT, gAT, liver, muscle (quadriceps), and pancreas were not affected by mast cell-deficiency in diet-induced obesity (Figure 2A). Moreover, no differences in fasting blood cholesterol (Figure 2B) and triglyceride levels (Figure 2C) or fasting plasma insulin levels (Figure 2D) were found between obese mast cell-deficient and -proficient mice. Furthermore, mast cell deficiency did not affect adiponectin levels (Figure 2E). We subjected obese mast cell-deficient and -proficient mice to CLAMS analysis and observed no alterations in RER (Figure 2F), calories consumption (Figure 2G), water and food intake (Figures 2H,I), and motility (Figure 2J) between Cre-negative and Cre⁺ mice. Additionally, histological analysis of the AT did not display any differences between obese mast cell-deficient and -proficient mice (Figure 2K), with regards to AT morphology or adipocyte size. Together, mast cells are dispensable in obesity-related metabolic dysregulation and insulin resistance development.

As mast cells have been previously proposed to contribute to a variety of immune responses (8–10) and especially to obesity-related AT inflammation (12), we next evaluated inflammatory cell accumulation in the AT. Consistent with the findings in mast cell-deficient Cpa3^Cre+ mice (16), flow cytometric analysis did not reveal any differences in the accumulation of total Mφs or pro-inflammatory M1-polarized CD11c⁺ Mφs in the sAT or the gAT between obese Mcpt5-Cre⁺R-DTA⁺ mast cell-deficient and obese Cre-negative R-DTA⁺ control mice (Figures 3A–D). No difference in Mφ accumulation was observed when data were expressed as the percentage of SVF (Figures 3A,B) or when data were expressed as cell numbers per gram of tissue (Figures 3C,D). Similarly, mast cell deficiency did not affect accumulation of total T cell numbers (CD3⁺ cells), CD4⁺ T helper cell numbers, or CD8⁺ cytotoxic T cell numbers in the obese sAT or gAT (Figures 3E–H). Histological analysis of Giemsa-stained sections confirmed the virtual absence of mast cells in gAT of Mcpt5-Cre⁺R-DTA⁺ mice (Figures 3I,J). Quantitative PCR analysis of the gAT of obese mast cell-deficient and -proficient mice did not reveal any difference in the expression of cytokines IL-10, IL-1β, IL-6, and TNF and of the chemokine MCP-1 and thus in the inflammatory environment of the AT (Figure 3K). In conclusion, mast cells do not regulate AT inflammation in the course of diet-induced obesity.

We also performed histological analysis of the livers of obese mast cell-deficient and -proficient mice. Hepatic steatosis (Figures 4A,B), as well as lobular inflammation (Figure 4C), hepatocellular ballooning (Figure 4D), and overall NAS (Figure 4E) did not differ between the two groups. Moreover, quantitative PCR did not display any differences in the expression of a series of factors involved in hepatic metabolism (Figure 4F), including lipogenesis [peroxisome proliferator-activated receptor gamma (PPARγ), sterol regulatory element-binding protein 1C (Srebp1c)], glycolysis [liver-type pyruvate kinase (LPK), glucokinase (GK)], the transport of fatty acids [fatty acid transport protein 2 (FATP2)], and triglycerides [CD36, microsomal triglyceride transfer protein (MTTP)] and glucose uptake by cells [glucose transporter 2 (Glut-2)], due to mast cell deficiency. The only significant difference between the two groups was in the expression of the gluconeogenic gene glucose-6-phosphatase (G6P, Figure 4F), which was slightly increased in mast cell-deficient mice, as compared to the control group. Further, we analyzed the inflammatory milieu of the liver of obese mast cell-deficient and -proficient mice and found no differences in the expression of F4/80 (as a surrogate marker for the presence of Mφs and Kupffer cells) or of the cytokines IL-1β, IL-6, and TNF and of the chemokine MCP-1 (Figure 4G). Therefore, mast cell deficiency does not contribute to obesity-associated hepatic steatosis and metabolic dysregulation.