Elevated expression of the toll like receptors 2 and 4 in obese individuals: its significance for obesity-induced inflammation

43 individuals were recruited from the local clinics. Written informed consent was obtained from all participants for inclusion in the study and the study protocol was approved by the institutional ethics committee (Ethical Review Committee of Dasman Diabetes Institute). Lean, overweight and obese subjects are asymptomatic or free of disease. Diabetic subjects were on medication of glucophage alone or in combination with crestor or zestril or januvia. Blood samples were collected after overnight fasting for isolation of PBMCs and for the determination of blood-derived factors including blood glucose, total cholesterol, high density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and hemoglobin A1c (HbA1c). Participants’ age, height, weight, blood pressure and diabetes status were assessed at the time of the blood draw. All biochemical tests were performed by using standard kits. The characteristics of the participants are described in Table 1. Adipose tissue samples were obtained from 21 individuals with different BMI.

For PBMCs isolation, fresh blood samples were collected from participants in EDTA-tubes. PBMCs were isolated by using Ficoll-Hypaque density gradient centrifugation [15].

Human adipose tissue samples were collected via abdominal subcutaneous fat pad biopsy lateral to the umbilicus using standard surgical techniques. In brief, the region was sterilized using alcohol, and locally anesthetized with 2 ml 2% lidocaine. Through a small superficial 0.5 cm skin incision, fat tissue was collected via punch biopsy. Tissue samples were stored in formalin or snap-frozen in liquid nitrogen and stored at −80°C. Samples were then used for immunohistochemical staining.

Frozen sections (4 μm) of adipose tissue were cut. Slides were deparaffinized in xylene and rehydrated in pure ethanol to water. Antigen retrieval was done by placing the slides in target retrieval solution pH6.0 (Dako) in the pressure cooker boiling for 8 minutes and cooling down for 15 minutes. After washing in phosphate buffer saline (PBS), endogenous peroxidase was blocked with 3% H₂O₂ for 30 minutes. The slides were blocked with 5% milk for 1 hour followed by the 1 hour incubation with 1%BSA solution. The slides were incubated in primary antibody overnight at room temperature in 1:300 dilution of rabbit polyclonal TLR-2 antibody (ProSci Incorporated), 1:400 dilution of rabbit polyclonal TLR-4 antibody (ProSci Incorporated), 1:200 dilutions of rabbit polyclonal antibody (ProSci Incorporated), 1:200 dilution of TRAF6 antibody (ProSci Incorporated), 1:200 dilution of MyD88 (ProSci Incorporated) and 1:200 dilution of IRAK antibody (ProSci Incorporated). After washing with PBS–0.05% Tween, slides were incubated for 1 hour with the respective secondary antibodies. Bound antibody was detected with HRP EnVision Plus Kit. Color was developed in 3, 3’-diaminobenzidine chromogen substrate. The sections were then washed in running tap water, lightly counterstained with Gill’s hematoxylin, dehydrated through ascending graded alcohols, cleared in xylene, and mounted in DPX. Two different observers who were blinded to the source of the tissues quantified the expression of each antigen semiquantitatively on a 3-point scale, where 0 = no staining; 1 = mild expression, limited areas stained; and 3 = strong expression, strong overall staining. Immunohistochemical staining of the section of human spleen tissue (HST) was used as positive control for TLR2. Immunohistochemical staining of the section of breast cancer tissue was used as positive control for TLR4 [16].

Total RNA was extracted using RNeasy kit (Qiagen). The cDNA was synthesized with 0.5ug of total RNA using high capacity cDNA reverse transcription kit (Invitrogen). Polymerase chain reaction (PCR) was performed using the One Step PCR kit (Promega). The primer pairs used were as follows: TLR2 Fwd (5′-ATTGTGCCCATTGCTCTTTC-3′) and TLR2 Rev (5′-TTCTTCCTTGGAGAGGCTGA −3′); TLR4 Fwd (5′-AATCCCCTGAGGCATTTAGG-3′) and TLR4 Rev (5′-CCCCATCTTCAATTGTCTGG −3′); MyD88 Fwd (5′-GGATGGTGGTGGTTGTCTCT-3′) and MyD88 Rev (5′-AGGATGCTGGGGAACTCTTT-3′); IRAK1 Fwd (5′- AGCCCCTTCTTCTACCAAGC-3′) and IRAK1 Rev (5′-AGGAAGCTCTGCTTCACTGC-3′) TRAF6 Fwd (5′-CTGCAAAGCCTGCATCATAA-3′) and TRAF6 Rev (5′- GGGGACAATCCATAAGAGCA −3′); TNF-α Fwd (5′-CAGAGGGCCTGTACCTCATC-3′) and TNF-α Rev (5′- GGAAGACCCCTCCCAGATAG −3′); IL-6 Fwd (5′- CAGGGGTGGTTATTGCATCT-3′) and IL-6 Rev (5′-AAAGAGGCACTGGCAGAAAA-3′) and GAPDH Fwd (5′-ATCGTGGAAGGACTCATGACCACA-3′) and GAPDH Rev 5′-TAGAGGCAGGGATGATGTTCTGGA-3′. The RNA was reverse-transcribed at 50°C for 30 minutes and reverse transcriptase was inactivated at 95°C for 15 minutes. PCR was run at: 30 cycles of 94°C for 1 minute, 50°C for 1 minute, 72°C for 1 minute, and a final extension step at 72°C for 10 minutes. The PCR product (10 μl) was analyzed on 1.4% agarose gel to detect TLR2, TLR4, MyD88, IRAK1, TRAF6, IL-6, TNF-α and GAPDH cDNA amplification. Relative band quantification was performed by using Gel Doc™ XR + imaging system (Bio Rad,USA). Density of the bands was expressed in arbitrary units. Jurkat cells were used as a –ve control for TLR2 and TLR4 [17]. THP1 cells were used for + ve control for TLR2 and TLR4 [18].

Peripheral blood mononuclear cells (PBMCs) are often considered for investigating many aspects of immunological responses. In previous studies, it has been shown that TLR2 and 4 are highly expressed in PBMCs during course of inflammatory disorders. However, since expression of TLR2 and 4 in PBMCs in obesity has yet not been well defined, we measured levels of mRNA for TLR2 and 4 in PBMCs from obese, overweight and lean individuals. As shown in Figure 1A, TLR2 mRNA showed the highest expression, followed by TLR4 in obese subjects and there was significant difference from lean subjects (P < 0.05). Significant differences in the expression of TLRs were also found between overweight and lean subjects (P < 0.05) (Figure 1B, C). Expression levels of TLRs increased significantly with increasing BMI (TLR2: r = 0.91; TLR4: r = 0.88, P < 0.0001) (Figure 1D, E).

TLRs’ activation requires the signal transduction molecules myeloid differentiation protein (MyD88), IL-1 receptor-associated kinase 1(IRAK1), and tumor necrosis factor (TNF) receptor-associated kinase 6 (TRAF6) which led to the hypothesis that these TLR’s adaptor proteins (MyD88, IRAK1 and TRAF6) could be increased in obese individuals. Therefore, we also measured the expression of TLRs’ adaptor proteins MyD88, IRAK1 and TRAF6 in PBMCs used in the previous experiment. We found that expression levels of MyD88 and IRAK1 were significantly increased in PBMCs from overweight and obese subjects (P < 0.05) (Figure 2A, B, C). However we noticed a strong constitutive expression of TRAF6 in PBMCs from all groups of subjects (Figure 2A, D). Linear regression analysis revealed a significantly positive correlation between MyD88/IRAK1 expression and BMI (MyD88: r = 0.95, P < 0.0001; IRAK1: r = 0.78, p < 0.002) (Figure 2E, F). No association was observed between TRAF6 expression and BMI (Figure 2G).

To identify TLR2 and TLR4 expression in adipose tissues, we performed immunohistochemical analysis of the sections from 21 subjects (lean 6; overweight 7; obese 8) that were stained with the respective antibodies. Staining intensity of TLR2 and TLR4 expression was greater (P < 0.05) in obese subjects compared to lean or overweight subjects (Figure 3A, B, D and E). It was noticed that staining intensities of TLR2 and TLR4 were observed in the areas of the adipose tissue infiltrated with inflammatory cells. TLR2 and TLR4 positive cells were more abundant in adipose tissue section from obese as compared to lean individuals (Figure 3A and D). Linear regression analysis also revealed a significantly positive correlation between TLR2/TLR4 expression levels and BMI (TLR2: r = 0.68, P = 0.0007; TLR4: r = 0.43, P = 0.052) (Figure 3C and F). We also measured the expression of TLRs’ adaptor proteins MyD88, IRAK1 and TRAF6 in adipose tissues used in the previous experiment. We found that only MyD88 levels were increased in adipose tissues from obese individuals (Figure 4A, B). Linear regression analysis revealed a significantly positive correlation between MyD88 protein expression and BMI (r = 0.58, P = 0.005) (Figure 4C).

It has been known that elevated circulating levels of TNF-α and IL-6 in obese subjects play an important role in the development of insulin resistance [19‐22]. However, less information is available about the simultaneous expression of TNF-α and IL-6 with TLRs in PBMCs from same individuals. Therefore, we investigated the expression of these cytokines in the PBMCs and their association with the expression of TLRs and their adaptor proteins. Transcript levels of TNF-α and IL-6 were significantly higher in PBMCs from overweight and obese subjects compared to lean subjects (P < 0 · 05) (Figure 5A, B, C). Linear regression analysis demonstrated an association between the transcript expression levels of cytokines and TLRs expression (TLR2 vs TNF-α: r = 0.92, P < 0.0001; TLR2 vs IL-6: r = 0.91, P < 0.0001; TLR4 vs TNF-α: r = 0.92, P < 0.0001; TLR4 vs IL-6: r = 0.81, P < 0.001) (Figure 5D, E). The increased expression of TNF-α and IL-6 in PBMCs from obese subjects correlated with increased expression of TLRs’ adaptor proteins (MyD88 vs TNF-α: r = 0.9, P < 0.0001; MyD88 vs IL-6: r = 0.91, P < 0.0001; IRAK1 vs TNF-α: r = 0.86, P < 0.0002; IRAK1 vs IL-6: r = 0.77, P < 0.002) (Figure 5F, G).

Since our data showed that TLRs expression was high in obese individuals and was significantly correlated with inflammatory cytokines which causes insulin resistance. Therefore, we measured TLRs’ expression in PBMCs from 10 diabetic subjects with different BMI. Our results showed that the subjects with type-2 diabetes had significantly elevated mRNA levels of TLR2 and TLR4 as compared with non-diabetic obese subjects (Figure 6A, B and Figure 1A). The basal levels of TNF-α and IL-6 were high in all groups of diabetic subjects and there was no significance difference (data not shown).

Spearman’s rank correlation was performed to assess the association of TLRs’ expression with FBG and HbAIc. Correlation analysis showed that elevated levels of TLR2 in obesity was significantly correlated with FBG (r = 0.61; P < 0.0025) and HbAIc (r = 0.44; P < 0.03) (Table 2). Similarly, TLR4 positively correlated with FBG (r = 0.52; P < 0.01) and HbAIc (r = 0.48; P < 0.03) (Table 2).