Tumor Necrosis Receptor Superfamily Interact with Fusion and Fission of Mitochondria of Adipose Tissue in Obese Patients without Type 2 Diabetes

Interactions between receptors and ligands of the tumor necrosis factor superfamily (TNFSF) provide costimulatory signals that control the survival, proliferation, differentiation, and effector function of immune cells. All components of the TNF superfamily are associated with NF-kB functions that are not limited to cell death and may promote survival in the face of adipose tissue inflammation in obesity. Inflammation dysfunction of mitochondria is a key factor associated with insulin resistance in obesity. The aim of the study was to analyze the relationship of soluble forms of receptors and ligands of the TNF superfamily in blood plasma with mitochondrial dynamics in adipose tissue (greater omentum (GO) and subcutaneous adipose tissue (Sat)) of obese patients with and without type 2 diabetes mellitus (T2DM). Increased plasma sTNF-R1, sTNF-R2, sTNFRSF8 receptors, and ligands TNFSF12, TNFSF13, TNFSF13B are characteristic of obese patients without T2DM. The TNF-a levels in blood plasma were associated with a decrease in MFN2 gene expression in GO and IL-10 in blood plasma. The TNFSF12 levels contributed to a decrease in glucose levels, a decrease in BMI, and an increase in IL-10 levels by influencing the MFN2 gene expression in GO, which supports mitochondrial fusion.


Introduction
Interactions between ligands of the tumor necrosis factor superfamily (TNFSF) and tumor necrosis factor receptor superfamily (TNFRSF) provide co-stimulation signals that control the survival, proliferation, differentiation, and effector function of immune cells. Receptors and ligands can be classified into three groups. The first class is death receptors (DR), which contain the death domain (DD). The second class is receptors that interact with members of the TRAF family (TNFR-associated factor). The third class consists of decoy receptors (DcR) without intracellular interacting partners that act as inhibitors of the TNFSF ligand [1]. TNFR1 belongs to the first class of receptors and is expressed in various types of cells. TNFR1 is a DR and contains a death domain DD. TNFR1 is activated by cytotoxicity signaling pathways and promotes an acute inflammatory response. TNFR1 activation induces apoptosis or necroptosis and activates nuclear factor-κB (NF-κB) [1].
TNFR2 belongs to the second class of receptors, does not contain a death domain, and does not cause cell death. TNFR2 stimulates the signaling of the non-canonical NF-κB pathway. Circulating sTNFR2 has been shown to increase in response to TNF and is negatively associated with insulin sensitivity in obese patients [2].
Thus, it is assumed that the processes of mitochondrial fusion are induced under conditions of optimization of mitochondrial bioenergetics. In contrast, the division functions are associated with the degradation of mitochondria (damage to mitochondria). Successful control of these processes can provide several therapeutic approaches, one of which is the modulation of processes associated with mitochondria: mitochondrial dynamics, apoptosis, and autophagy.
All of the above indicates the relationship between the superfamily TNF and NF-κB signaling with mitochondrial biogenesis [13,14]. A deeper understanding of the relationship between NF-κB and mitochondrial dynamics is an important area of research for regulating genes and pathways activated in chronic inflammation.
Thus, the study aimed to analyze the relationship of soluble forms of receptors and ligands of the TNF superfamily in blood plasma with mitochondrial dynamics in adipose tissue (greater omentum (GO) and subcutaneous adipose tissue (Sat)) of obese patients with and without T2DM.
Samples of biological material (biopsies from SAT and GO) from obese patients were collected during different types of bariatric surgery according to the indication: Gastric bypass or longitudinal resection. During routine laparoscopic surgeries, tissue biopsies from healthy donors were taken: Inguinal and femoral hernias (right or left), diaphragmatic and abdominal hernias, and nephroptosis. Patients with type 2 diabetes take metformin at a dose of 500-1500 mg per day. Patients with type 2 diabetes did not receive insulin therapy.
Newly diagnosed cases of T2DM were found in 18% of the patients who participated in the study. The duration of T2DM in the studied patients was 2.0 ± 1.3 years. All blood samples were collected in the morning on an empty stomach according to the protocol. The value of glycated hemoglobin (HbA1C) was 5.21 (5.1-5.4) % in the control group, 5.81 (5.6-5.9) % in the group of obese patients without T2DM, 7.59 (6.45-8.67) % in the group of obese patients with T2DM. Glucose values after the test breakfast were 5.5 ± 0.8 mmol/l in the control group, 6.2 (5.8-7.1) mmol/l in obese patients without T2DM, 11.03 ± 5.23 mmol/l in the group of obese patients with T2DM. Venous blood samples were taken 60 min after the test meal for glucose determination. Breakfast consisted of Buckwheat milk porridge without sugar (200 g), jelly without sugar (150 g). The test breakfast contained 9.1 g protein, 88.1 g carbohydrate, and 10.67 g fat. The total calorie content of the test breakfast was 466 kcal.
The presence of obesity and T2DM was established on the basis of a detailed clinical and instrumental examination in a specialized hospital, guided by the World Health Organization (1999-2013) criteria for diagnosing diabetes and other types of hyperglycemia [15]. Informed consent was signed by all patients. Verification of the diagnosis and recruitment of patients into the study groups was carried out at the Department of Reconstructive and Plastic Surgery on the basis of a regional clinical hospital in Kaliningrad.
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board (or Ethics Committee) of IKBFU (protocol code 2 and 29 November 2018).
The concentration of the TNF-α molecule in the blood plasma was measured using the ELISA method (Vector-Best kits, Russia) on a Lasurit automatic enzyme-linked immunosorbent analyzer (Dynex Technologies, Chantilly, VA, USA).

Gene Expression Research
Total RNA from homogenized biopsies of AT was isolated using an ExtractRNA kit (Evrogen, Moscow, Russia). The resulting RNA was dissolved in 30 µL of nucleasefree water. The purity and concentration of the isolated RNA were determined using a spectrophotometer Nanovue Plus (GE Healthcare Bio-Sciences, Uppsala, Sweden). The quality of total RNA was determined by the RIN index (the RNA integrity number). The reverse transcription procedure was performed using (dT) 23 (Beagle, Moscow, Russia) and M-MLV reverse transcription (Evrogen, Moscow, Russia). To determine relative gene expression levels, qPCR was performed using qPCRmix-HS reagents (Evrogen, Moscow, Russia). As a template, we used 4 µL of cDNA, as a reference gene-RPLPO (a large ribosomal protein). Sequences of primers and probes for PCR are listed in Table 1. Table 1. Sequences of primers and probes for PCR.

Western Blot Analysis
Protein lysates were obtained using RIPA buffer (n = 8) (BioRad, Hercules, CA, USA) and normalized by the Bradford method (reagent Bradford BioRad buffer, USA) with separation by polyacrylamide gel electrophoresis (BioRad, Hercules, CA, USA). The transfer of proteins was carried out on a PVDF membrane at +4 C for 2 h (BioRad, Hercules, CA, USA). Hybridization of the target proteins was carried out with the corresponding primary antibodies and secondary antibodies labeled with horseradish peroxidase (Thermo Fisher Scientific, Waltham, MA, USA) using a substrate for visualization of Western blots ECL Plus (Thermo Fisher Scientific, Waltham, MA, USA). The analysis of the staining intensity of the bands was carried out in the Image J program with normalizations for the GAPDH protein.

Statistical Analysis
The normal distribution of quantitative indicators was checked using the Kolmogorov-Smirnov test and the Shapiro-Wilk test. With a normal distribution, the hypothesis of equality of the mean sample values was tested using Student's t-test. The nonparametric Kruskal-Wallace test applied the significance of differences between independent quantitative samples not having a normal distribution law. In the case of statistically significant differences between the groups, analysis was performed using the Mann-Whitney test. Differences were considered significant at a significance level of p < 0.05.
To determine relative gene expression levels, qPCR was performed using qPCRmix-HS reagents (Evrogen, Moscow, Russia). As a template, we used 4 µL of cDNA, as a reference gene-RPLPO (a large ribosomal protein). Gene expression levels were calculated using the delta-delta Ct method, also known as 2-∆∆Ct. The formula was used to calculate the relative frequency of gene expression in the samples [16].
The analysis of the band staining intensity was carried out using the Image Lab Software (Bio-Rad, Hercules, CA, USA) with normalizations for the glyceraldehyde-3phosphate dehydrogenase (GAPDH) protein (#ZG003, Thermo Fisher Scientific, Waltham, MA, USA).
The presence of a relationship between the studied parameters was carried out using Spearman correlation. Correlations were calculated in each group and for all patients (general correlations). For the analysis of an adequate linear regression model, the following regression residues were considered: the lack of autocorrelation of residues (Durbin-Watson test, p-values > 0.05) and its normal distribution and the consistency of the dispersion residues (heteroscedasticity test, p-values < 0.05 were considered significant). Multi-linear regression Least squares Gaussian distribution of residuals was visualized using GraphPad Prism 9.00 (GraphPad Software Inc, San Diego, CA, USA). Statistical analyzes and graphs were generated in GraphPad Prism 8.01 from the output data (GraphPad Software Inc, San Diego, CA, USA).

Biochemical Parameters of Obese Patients
The results of blood biochemical parameters are presented in Figure

The Concentration of Plasma Cytokines of the TNF Receptor Superfamily in Obese Patients
The concentration of plasma cytokines of the TNF receptor superfamily (receptors-sTNF-R1, sTNFRSF8, and ligands-TNFSF12, TNFSF13, TNFSF13B) in obese patients with T2DM was lower than in patients without T2DM and the control group (p < 0.05) ( Table 1). The levels of receptors sTNF-R1, sTNF-R2, sTNFRSF8, and ligands TNFSF12, TNFSF13, TNFSF13B in obese patients without T2DM were higher in comparison with the group of patients with T2DM and the control group (p < 0.05) ( Table 2).

The Concentration of Plasma Cytokines of the TNF Receptor Superfamily in Obese Patients
The concentration of plasma cytokines of the TNF receptor superfamily (receptors-sTNF-R1, sTNFRSF8, and ligands-TNFSF12, TNFSF13, TNFSF13B) in obese patients with T2DM was lower than in patients without T2DM and the control group (p < 0.05) ( Table 1). The levels of receptors sTNF-R1, sTNF-R2, sTNFRSF8, and ligands TNFSF12, TNFSF13, TN-FSF13B in obese patients without T2DM were higher in comparison with the group of patients with T2DM and the control group (p < 0.05) ( Table 2). Plasma levels of receptors and TNF ligand decreased with increasing BMI. A negative relationship was found between the levels of sTNF-R1, sTNF-R2, sTNFRSF8, TNFSF12,   sTNF-R1, sTNF-R2, sTNFRSF8, TNFSF12, TNFSF13, TNFSF13B levels were po tively correlated with each other in obese patients ( Figure 3). A regression analysis was performed to prove that an increase in TNF class 2 rece tors and ligands in blood plasma was not randomly associated with an improvement carbohydrate and lipid metabolism parameters in the serum of obese patients. We an lyzed whether the receptors affect the biochemical parameters of obese patients (Figure (Table 3). Using multivariate regression analysis, we showed that an increase in sTNF-R TNFSF12, and TNFSF13B levels decreased BMI and plasma glucose levels. Multivari A regression analysis was performed to prove that an increase in TNF class 2 receptors and ligands in blood plasma was not randomly associated with an improvement in carbohydrate and lipid metabolism parameters in the serum of obese patients. We analyzed whether the receptors affect the biochemical parameters of obese patients ( Figure 4) ( Table 3). Using multivariate regression analysis, we showed that an increase in sTNF-R1, TNFSF12, and TNFSF13B levels decreased BMI and plasma glucose levels. Multivariate regression analysis showed that an increase in TNFSF12 and TNFSF13B levels was associated with a decrease in blood glucose concentration, a reduction in BMI, and an increase in plasma IL10 levels.
regression analysis showed that an increase in TNFSF12 and TNFSF13B levels was associated with a decrease in blood glucose concentration, a reduction in BMI, and an increase in plasma IL10 levels.   Thus, increases in TNFSF12, TNFSF13B, and sTNF-R1 levels are associated with decreased glucose concentration and decreased BMI in obese patients.

The NF-kB Levels in Obese Patients
NF-kB1 gene expression in GO and SAT was increased in obese patients without T2DM compared with the control group (p < 0.05) ( Figure 5A). NF-kB1 gene expression in GO was higher in patients with T2DM than in the control group and was not different from that of obese patients without T2DM (p < 0.05). NF-kB1 gene expression in SAT was lower in patients with T2DM than in obese patients without T2DM (p < 0.05). NF-kB (RelA) protein production in GO and SAT did not change in obese patients without T2DM compared to the control group (p < 0.05) ( Figure 5B,C).

Mitochondrial Dynamics-Division and Fusion
DNM1L gene expression in GO and SAT was higher in obese patients without T2DM than in the control group ( Figure 6). MFN2 gene expression in GO and SAT was higher in obese patients without T2DM than in the control group (Figure 7). TFAM gene expression in GO and SAT was higher in obese patients without T2DM than in the control group ( Figure 8). DNM1L, MFN2, and TFAM gene expression were positively correlated with

Mitochondrial Dynamics-Division and Fusion
DNM1L gene expression in GO and SAT was higher in obese patients without T2DM than in the control group ( Figure 6). MFN2 gene expression in GO and SAT was higher in obese patients without T2DM than in the control group (Figure 7). TFAM gene expression in GO and SAT was higher in obese patients without T2DM than in the control group ( Figure 8). DNM1L, MFN2, and TFAM gene expression were positively correlated with each other in GO. NF-kB1, DNM1L, MFN2, and TFAM gene expression correlated positively with each other in SAT (Figure 9).
The production of dinamine-1-like protein in SAT was higher in obese patients with T2DM than in obese patients without T2DM ( Figure 6). The production of transcription factor-A protein in SAT was lower in obese patients with T2DM than in the control group ( Figure 5).
Thus, increased activity of mitochondrial dynamics genes is characteristic of obese patients without T2DM. In obese patients with T2DM, the production of dinamine-1-like protein was increased. In contrast, the production of transcription factor A protein was decreased compared to the other groups of patients. each other in GO. NF-kB1, DNM1L, MFN2, and TFAM gene expression correlated positively with each other in SAT (Figure 9). The production of dinamine-1-like protein in SAT was higher in obese patients with T2DM than in obese patients without T2DM ( Figure 6). The production of transcription factor-A protein in SAT was lower in obese patients with T2DM than in the control group ( Figure 5). Thus, increased activity of mitochondrial dynamics genes is characteristic of obese patients without T2DM. In obese patients with T2DM, the production of dinamine-1-like protein was increased. In contrast, the production of transcription factor A protein was decreased compared to the other groups of patients.

TNF Receptors and Ligands Levels Have Been Associated with Components of Mitochondrial Dynamics
It was found that the levels of sTNF-R1, sTNF-R2, sTNFRSF8, and TNFSF12 were positively correlated with MFN2 gene expression in GO (Figures 10 and 11). The levels of TNFSF12, TNFSF13, and TNFSF13B were positively correlated with TFAM (GO) ( Figures  10 and 11). TNFSF13 level was positively correlated with DNM1L gene expression in GO (Figures 10 and 11).

TNF Receptors and Ligands Levels Have Been Associated with Components of Mitochondrial Dynamics
It was found that the levels of sTNF-R1, sTNF-R2, sTNFRSF8, and TNFSF12 were positively correlated with MFN2 gene expression in GO (Figures 10 and 11). The levels of TNFSF12, TNFSF13, and TNFSF13B were positively correlated with TFAM (GO) (Figures 10 and 11). TNFSF13 level was positively correlated with DNM1L gene expression in GO (Figures 10 and 11).

TNF Receptors and Ligands Levels Have Been Associated with Components of Mitochondrial Dynamics
It was found that the levels of sTNF-R1, sTNF-R2, sTNFRSF8, and TNFSF12 were positively correlated with MFN2 gene expression in GO (Figures 10 and 11). The levels of TNFSF12, TNFSF13, and TNFSF13B were positively correlated with TFAM (GO) ( Figures  10 and 11). TNFSF13 level was positively correlated with DNM1L gene expression in GO (Figures 10 and 11).   We checked if the multiple regression model included DNM1L, MFN2, and gene expression in GO and SAT. By sequential fitting, it was found that only the TN dependent multiple regression model was statistically significant. Moreover, w conditional parameter MFN2 gene expression, the regression coefficient and sign increased (Table 4).
Thus, we have demonstrated an association between sTNF-R1, sTNF-R2, sTN TNFSF12, TNFSF13, TNFSF13B, and mitochondrial dynamics in obese patients. Th significant contribution was made by TNFSF12 level and contributed to an incr MFN2 gene expression in GO, an increase in IL -10 level, a decrease in glucose lev a reduction in BMI in obese patients. We checked if the multiple regression model included DNM1L, MFN2, and TFAM gene expression in GO and SAT. By sequential fitting, it was found that only the TNFSF12 dependent multiple regression model was statistically significant. Moreover, with the conditional parameter MFN2 gene expression, the regression coefficient and significance increased (Table 4). Table 4. Models of multiple regression analysis of the relationship between the sTNF-R1. TNFSF12 and TNFSF12 levels and glucose level, IL10 level, BMI, DNM1L gene expression in GO, MFN2 gene expression in GO, and TFAM gene expression in GO. Thus, we have demonstrated an association between sTNF-R1, sTNF-R2, sTNFRSF8, TNFSF12, TNFSF13, TNFSF13B, and mitochondrial dynamics in obese patients. The most significant contribution was made by TNFSF12 level and contributed to an increase in MFN2 gene expression in GO, an increase in IL -10 level, a decrease in glucose level, and a reduction in BMI in obese patients.

TNF-a Level Was Associated with MFN2 Gene Expression in GO
The multiple regression model showed that an increase in TNF-a level influenced a decrease in IL10 level and MFN2 gene expression in GO (Table 5). Table 5. Multiple regression analysis models of the relationship between TNF-a level and Il-10 level and MFN2 gene expression in GO.

Discussion
This study investigated the role of TNF receptors and ligands in plasma of obese patients in relation to the presence or absence of T2DM. It was found that obese patients with T2DM had impaired lipid metabolism and high glucose levels. In obese patients without T2DM, levels of glucose, cholesterol, high-density lipoproteins (HDL), low-density lipoproteins (LDL), and triglycerides were predominantly within reference values (World Health Organization (1999-2013)).
Increased levels of the receptors sTNF-R1, sTNF-R2, sCD30/TNFRSF8, and the ligands TVVEAK/TNFSF12, APRIL/TNFSF13, BAFF/TNFSF13B are characteristic of obese patients without T2DM. STNFR1 and sTNFR2 can suppress the TNF-mediated inflammatory response in blood plasma [17]. This fact indicates compensatory mechanisms associated with the action of TNF receptors and ligands in obese patients without T2DM. Receptors interacting with type 2 and type 3 TNF family members may protect against metabolic complications [1]. Correlation and regression analysis of TNFSF12 and TNFSF13B levels showed that increased levels were associated with lower BMI, lower glucose levels, and increased IL10 (Figure 4 and Figure S1).
High levels of sTNF-R1, TNFSF12, TNFSF13, and TNFSF13B contribute to the maintenance of normal lipid metabolism in obese patients (p < 0.05). A decrease in HDL and an increase in LDL are associated with obesity. Plasma levels of sTNF-R1, TNFSF12, TNFSF13, and TNFSF13B correlated positively with HDL levels and negatively with LDL levels [18].
A positive correlation was found between the levels of sTNF-R1, sTNF-R2, sTN-FRSF8, TNFSF12, TNFSF13, and TNFSF13B in blood plasma with the levels of the antiinflammatory cytokine IL-10 and a negative correlation with TNF. IL-10, known as a factor that inhibits cytokine synthesis, is considered an essential immunoregulatory cytokine. IL-10 is compensatorily synthesized in response to the action of TNF and other proinflammatory cytokines and can inhibit inflammation under various pathophysiological conditions [19]. Data indicate the anti-inflammatory role of high concentrations of sTNF-R1, sTNF-R2, sTNFRSF8, TNFSF13, and TNFSF13B in obesity.
The levels of sTNF-R1, sTNF-R2, sTNFRSF8, TNFSF12, TNFSF13, and TNFSF13B decreased with increasing BMI in obese patients. This fact might indicate depletion of compensatory potential in obesity.
Thus, an increase in sTNFRSF8 and TNFSF13 is associated with an increase in the production of IL-10, HDL and a decrease in plasma BMI, LDL, and TNF in obese patients. Moreover, an increase in plasma levels of sTNFRSF8, TNFSF12, and TNFSF13 was associated with a decrease in serum glucose concentration in obese patients. Increased levels of sTNF-R1, sTNF-R2, sTNFRSF8, TNFSF12, TNFSF13, and TNFSF13B were found to be protective in relation to the development of T2DM in obese patients (p < 0.05). Multivariate regression analysis revealed that an increase in TNFSF12 and TNFSF13B levels was associated with a decrease in blood glucose concentration, a decrease in BMI, and an increase in plasma IL10 levels. Multivariate regression analysis revealed that an increase in sTNF-R1 levels was associated with a decrease in blood glucose concentration and BMI.
Adipose tissue of different localization regulates metabolism in obesity [20]. VAT is the most metabolically active fat depot [21]. Metabolic complications are thought to be associated with metabolic abnormalities in GO [20,21]. NF-kB1 gene expression in the GO was increased in obese patients with and without T2DM compared to the control group.
SAT is considered to be a protective fat depot [22]. Anti-inflammatory cytokines and adipokines are thought to be produced in SAT [20,23]. High NF-kB1 gene expression is unexpected in obese patients without metabolic complications. The role of NF-kB needs to be further investigated to understand why NF-kB1 gene expression is elevated in visceral and subcutaneous adipose tissue in obese patients without T2DM, and NF-kB1 gene expression is not as significant in patients with T2DM.
Activation of the non-canonical pathway begins with the stabilization of NF-κBinducing kinase (NIK). NIK phosphorizes and activates IKKα. The homodimeric IKKα complex phosphorylates and triggers the processing of p100. In addition, the p100 subunit is processed to p52, which is a distinctive feature of this pathway. Interestingly, untreated p100 inhibits DNA binding and nuclear localization of the NF-κB heterodimer [8]. p52 and its hetero-or homodimeric partner are released to bind to DNA in the nucleus and affect transcription [7,8].
The non-canonical activation of NF-κB is stimulated by family members of specific TNF receptors belonging to the second and third classes [8]. The TNFSF12, TNFSF13, and TNFSF13B levels were correlated with NF-kB1 gene expression in GO ( Figure S2). The stimulation of NF-κB gene expression in SAT and GO in obese patients without T2DM might be due to the activation of the non-canonical NF-κB pathway. In contrast, in obese patients with T2DM, the canonical NF-κB pathway is activated.
We found that gene expression of NF-kB1 correlated positively with components of mitochondrial dynamics at GO and SAT (Figure 9). NF-kB can trigger genes for cell death and genes that contribute to survival under oxidative stress [8]. Recent studies have docu-mented the importance of mitochondrial dynamics in the pathogenesis of T2DM [12,[24][25][26][27]. This study shows that obese patients without T2DM have increased mitochondrial fission, fusion, and mitochondrial transcription: DNM1L, MFN2, and TFAM gene expression in GO and SAT was higher than in the control group. Judging from the change in DNM1L and MFN2 gene expression, mitochondrial division and fusion are in equilibrium in obese patients without T2DM. In this study, NF-kB1, DNM1L, MFN2, and TFAM gene expression was unidirectional in GO and SAT.
The fusion of mitochondria restores their function [27,28]. Increased MFN2 gene expression in GO and SAT may be a protective mechanism for maintaining normal mitochondrial dynamics in adipose tissue in obese patients without T2DM. We used multiple regression models and tested models for the effect of TNF receptors and ligands on MFN2 gene expression in GO and SAT. TNFSF12 influenced the dependent variables-MFN2 gene expression in GO, glucose level, IL10 level and BMI (r2 = 0.62, p < 0.05). Moreover, this multiple regression model was the strongest among all studied (Table 4). We hypothesize that TNFSF12 is involved in the maintenance of mitochondrial fusion in GO obese patients without T2DM. Plasma TNFSF12 levels contribute to the maintenance of normal carbohydrate metabolism in obese patients without T2DM.
TNF receptors and ligands in blood plasma and the expression level of mitochondrial dynamics genes in SAT and GO in obese patients with T2DM did not differ from control values. However, biochemical parameters were higher in obese patients with T2DM than in the control group. This suggests that TNF receptors and ligands and associated mitochondrial dynamics do not compensate for the development of carbohydrate and lipid metabolism complications in obese patients with T2DM.
Fission processes are associated with mitochondrial degradation and are therefore induced under conditions of mitochondrial damage [25,26]. The balance of mitochondrial dynamics is essential for metabolic health [12]. The predominance of this or that process has a pathological significance [25]. The biological value of an increase in mitochondrial fission is not completely clear. It has been reported that high glucose levels can activate mitochondrial fission 1 protein (FIS1), leading to increased mitochondrial fragmentation and overproduction of ROS [25]. The DRP1 protein production in SAT was increased in obese patients with T2DM compared to the control group. The TNFSF13 was associated with mitochondrial fission-correlation and regression analysis showed a relationship between TNFSF13 level and DNM1L gene expression in GO. However, TNFSF13 was not included in the multiple regression models. We hypothesize that TNFSF13 independently affects DNM1L gene expression, and this relationship needs further investigation.
In obese patients with T2DM, mitochondrial function is impaired under oxidative stress conditions and increased ROS production [12]. The TFAM gene plays an essential role in mitochondrial biogenesis, and cell physiology maintains mtDNA replication and transcription and regulates mtDNA copy number [28]. TFAM protein production is decreased in SAT obese patients with T2DM compared to the control group. Our previous work has provided compelling evidence for the effect of high TNF levels on high mtDNA copy numbers in adipose tissue. Moreover, in patients with morbid obesity (BMI 40 kg/m 2 ), mtDNA copy number was reduced in adipose tissue compared with patients with obesity grades I and II [27,28]. Therefore, the increase in TFAM gene expression may indicate an increase in mtDNA transcription [25]. High TFAM gene expression in muscle reduces fat mass, increases energy expenditure, and decreases oxidative stress [24]. It is suggested that TFAM activation may be a therapeutic strategy for treating peripheral neuropathy [29]. In obese patients without T2DM, TFAM gene expression was increased in GO compared with the control group. In obese patients, correlation and regression analyzes showed a relationship between TNFSF13B and TFAM gene expression in GO.
The influence of TNF receptors and ligands on protection against metabolic complications in obesity, especially T2DM, is also due to blocking the effects of TNF-a. TNF-a levels correlated negatively with soluble forms of TNF receptors and ligands. Thus, a multiple regression model showed that an increase in TNF-a levels influenced a decrease in IL10 levels and MFN2 gene expression in GO.
Thus, TNF family receptors and ligands affect the components of mitochondrial dynamics only in GO but not in SAT. TNFSF12 could promote mitochondrial fusion in GO, keep energy metabolism in balance and protect against metabolic disorders.

2.
The TNFSF12 and TNFSF13B levels were associated with NFkB1 gene expression in GO in obese patients, and NFkB1 gene expression increased in GO and SAT in obese patients without T2DM compared with the control group. The NFkB1 gene expression was associated with components of mitochondrial dynamic-DNM1L, MFN2, and TFAM gene expression in GO and SAT in all obese patients. 3.
The DNM1L, MFN2, and TFAM gene expression levels in GO and SAT responsible for regulating mitochondrial dynamics were increased in obese patients without T2DM, and DNM1L and TFAM proteins production were unbalanced in patients with obesity and T2DM. 4.
The TNFSF12 levels contributed to an increase in MFN2 gene expression in GO; TNFSF13 contributed to an increase in DNM1L gene expression in GO; TNFSF13B contributed to an increase in TFAM gene expression in GO.

5.
The TNF-a levels in blood plasma were associated with a decrease in MFN2 gene expression in GO and IL-10 in blood plasma. 6.
The TNFSF12 levels contributed to a decrease in glucose levels, a decrease in BMI, and an increase in IL-10 levels by influencing the MFN2 gene expression in GO, which supports mitochondrial fusion.