Hyperglycemia induces Toll like receptor 4 expression and activity in mouse mesangial cells: relevance to diabetic nephropathy

Abstract

Diabetes is a proinflammatory state. The pattern recognition receptors, Toll-like receptors (TLRs), are increased in diabetic patients and have been suggested to play a role in diabetic nephropathy (DN). Progression of DN involves altered mesangial cell (MC) function with an expansion of the mesangial matrix. There is a paucity of data examining the role of TLR and its expression in MC. We hypothesize the expression of TLRs in the mesangium might be an important factor contributing to mesangium expansion and nephropathy. Thus we evaluated the effect of high glucose on TLR2 and TLR4 expression in mouse mesangial cells (MMC) in vitro. Exposure of MMC to 25 mM glucose for 24 h resulted in increased TLR4 mRNA and cell surface receptor expression compared with 5.5 mM glucose (P < 0.05). Interestingly, we were not able to detect expression of TLR2 in MMC. Furthermore, expression of a TLR4 downstream signaling cascade including myeloid differentiation factor 88 (MyD88), interferon regulatory factor 3 (IRF3), and Toll interleukin receptor domain containing adaptor inducing interferon-β (TRIF)-related adaptor molecule (TRAM) were significantly increased in cells exposed to 25 mM glucose (P < 0.05). There was also a significant increase in NF-κB activation along with increased secretion of inflammatory cytokines IL-6 and monocyte chemotactic protein-1. Levels of transforming growth factor-β were also significantly increased in the presence of 25 mM glucose (P < 0.05). Collectively, these data suggest that hyperglycemia activates TLR4 expression and activity in MC and could contribute to DN.

diabetes is a serious global disease affecting 348 million people worldwide and nearly 36 million people in the United States alone (6). Diabetic nephropathy (DN) is a leading cause of end-stage renal disease and is associated with increased mortality and morbidity (6, 28). Diabetic patients with renal disease also have an increased risk of cardiovascular disease (CVD)-related morbidity and mortality (16, 48).

Expansion of glomerular mesangium, podocyte loss, thickening of glomerular and tubular basement membrane, and tubular epithelial cell (TEC) dysfunction are key characteristics of DN (37, 51). The mesangium is composed of mesangial cells (MC) and ECM proteins secreted by MC and function as a central support for glomerular capillaries (1). There are many reports signifying the role of abnormal glomerular MC expansion and their altered function under the diabetic milieu causing occluding of glomerular capillaries (1, 51). There are also reports suggesting glomeruli have excessive accumulation of ECM protein in DN (36). Diabetic conditions also promote overproduction of ECM protein by MC, with increased formation of advanced glycation end products (AGEs), plasminogen activator inhibitor-1 (PAI-1), reactive oxygen species (ROS), activation of PKC, and production of various cytokines such as transforming growth factor (TGF-β) (19, 20, 27, 42).

Toll-like receptors (TLRs) are pattern recognition receptors and have a predominant role in the activation of innate and adaptive immune responses (26). TLRs initiate immune responses after recognition of pathogens via pathogen-associated molecular patterns and non-microbe molecules, including recognition of several endogenous ligands (damage-associated molecular patterns, ECM degradation products, fatty acids, heat shock proteins, AGEs) released during tissue damage and inflammation (26, 41, 52). Activation of TLRs results in the triggering of a signaling cascade with release of several inflammatory cytokines (3). Enhanced expression of TLR2 and TLR4 has been reported in animal models of atherosclerosis and in atherosclerotic plaque (31, 33, 39). Previous studies from our group have reported significantly increased expression of TLR2 and TLR4 and its association with an increase in levels of endogenous ligands and their activated downstream signaling cascades in monocytes from patients with type 1 diabetes mellitus (T1DM) as well as type 2 diabetes (T2DM) (9, 11–13). Increased TLR2 and TLR4 expression under a diabetic milieu was also associated with upregulation of NF-κB activity and increased release of various biomediators such as IL-6, TNF-α, monocyte chemotactic protein-1 (MCP-1), and IL-1β. Both TLR2 and TLR4 expression and activity are further enhanced in patients with diabetic microvascular disease, mainly nephropathy (14).

Furthermore, it has recently been shown that compared with wild-type diabetic mice, both TLR2 and TLR4 knockout (KO) mice have less progression of nephropathy (16, 32). However, neither of the studies documented the role of mesangial expansion. There has been a paucity of data on the expression of TLRs in MC and their contribution toward an increased inflammatory response in the diabetic milieu in mesangium. There is a wealth of data supporting the role of TLR expression and activation of their signaling pathway in hyperglycemia-induced cell damage (9, 11–13, 23). Hence there is a possibility of increased expression of TLRs in mesangium contributing to increased biomediator release, ECM production, and it is important to determine its role in DN. The present study evaluated the expression of Toll-like receptors TLR2 and TLR4 and activation of its downstream signaling cascade in mouse mesangial cells (MMC) under hyperglycemic conditions.

MATERIALS AND METHODS

Reagents

Mouse SV40 MES-13 glomerular mesangial cell lines were obtained from American Type Culture Collection (Manassas, VA). Anti-mouse TLR4 antibodies and isotype-matched IgG controls were purchased from eBioscience (San Diego, CA). Glucose and mannitol were purchased from Sigma-Aldrich (St. Louis, MO). Western blot antibodies were purchased from Cell Signaling Technology (Danvers, MA) and Imgenex (San Diego CA). Fetal bovine serum was purchased from Axenia Biologix (Dixon, CA). Trypsin EDTA used was purchased from Lonza (Allendale, NJ).

Cell Culture

MMC were cultured in growth media composed of endotoxin-free RPMI media (GIBCO, Grand Island, NY) supplemented with 5.5 mM glucose, 200 mM l-glutamine (Irvine Scientific, Santa Ana, CA), 1 M HEPES (Irvine Scientific), antibiotics and antimycotics, and 10% FBS. The endotoxin levels in the culture media and reagents were evaluated using a limulus amoebocyte lysate (LAL) assay (Lonza), and average endotoxin levels were <100 EU/ml consistently in all the experiments. The media was replaced every 72 h. For all experiments MMC grown to passage 5 were used.

Treatments

Cells were plated at a density of 1 × 10⁶ in each well of six-well tissue culture-treated plates and were cultured in growth media for 24 h. After achieving 80–90% confluence in six-well plates, cells were placed in depleting media (endotoxin-free RPMI supplemented with 200 mM l-glutamine, 1 M HEPES) without FBS for 24 h before any further treatments. The MMC were exposed to 5.5 mM glucose, 25 mM glucose, and as an osmotic control 19.5 mM mannitol was added along with 5.5 mM glucose for 24 and 48 h. Cell viability was determined by trypan blue exclusion and was 95%. Cell supernatants, lysates, and RNA were collected and used for ELISA, Western blotting/flow cytometry, and RT-PCR, respectively.

RNA Extraction and RT-PCR

RNA was isolated from cells using TRIzol (Invitrogen, Carlsbad, CA). RT-PCR was performed using mouse TLR2 and TLR4 primers purchased from Invivogen (San Diego, CA) following the manufacturer's cycling parameters. Amplification of GAPDH (Diagenode, Denville, NJ) was used as a loading control. Band intensities were determined using Image Quant Software (GE Healthcare Biosciences, Piscataway, NJ) as described previously (9). Data are presented as fold-induction of TLR4 transcripts normalized to GAPDH in MMC.

ELISA

IL-6, TNF-α, MCP-1, and TGF-β were measured by ELISA using reagents from R&D Systems (Minneapolis, MN) as reported previously (10). Release of various inflammatory cytokines was measured in the supernatant at 24 h of various treatments. The intra- and interassay coefficients of variation (CVs) were determined to be <10%.

Fluorescence-Activated Cell Sorting

TLR4 surface expression was determined by flow cytometry as described previously (9, 10, 12). Briefly, cells were incubated with anti-mouse TLR4 antibodies (eBioscience San Diego, CA) or isotype controls, and surface expression of TLR4 was analyzed using a BD fluorescence-activated cell sorting (FACS) Array Bioanalyzer (Franklin Lakes, NJ). Results are expressed as mean fluorescence intensity (MFI)/10,000 cells. The intra- and interassay coefficients of variation (CVs) were determined to be <10%.

Western Blotting

Western blotting was performed to determine the TLR4 downstream signaling events as described previously (10). Briefly, 20 μg of the total protein was resolved, transferred, and probed with antibodies for myeloid differentiation factor 88 (MyD88), interferon regulatory factor 3 (IRF3), TRIF-related adaptor molecule (TRAM), followed by washing and incubation with horseradish peroxidase-conjugated secondary antibodies. The membranes were developed with ECL (Thermo Fisher Scientific, Pittsburgh, PA) as described previously (9, 10). The stripped membranes were further probed with β-actin antibody (Sigma-Aldrich St Louis, MO) as a loading control. Data are presented as fold-induction normalized to β-actin.

NF-κb p65 Assay

Nuclear extracts were prepared from the cells using a reagent from Thermo Fisher Scientific. NF-κB p65 activity was determined at 24 h using a nonradioactive TransAM transcription factor assay (Active Motif, Carlsbad, CA) of the nuclear extracts of MMC following different treatments as described previously (9, 10). The intra-assay CVs were determined to be <14%.

Statistical Analysis

All the experiments were repeated three times in duplicate. Results of experimental studies are reported as the means ± SD of at least three representative independent experiments. Differences were analyzed by Student's t-test, and a probability value of P < 0.05 was considered significant. All statistical analyses were performed using Graphpad Prism software.

RESULTS

MMC were exposed to 5.5 mM (normal glucose) and 25 mM glucose for 24 h. The mRNA expression levels of TLR2 and TLR4 were determined. Figure 1A depicts significantly increased mRNA expression of TLR4 with 25 mM glucose compared with 5.5 mM glucose (P < 0.05) at 24 h. However, we did not observe any further increase at 48 h (data not shown), so we focused on the 24-h time point for all further experiments. Additionally, an osmotic control consisting of 19.5 mM mannitol was added to normal glucose for each set of experiments, and its addition did not result in a significant increase in TLR4 expression, hence suggesting a glucose-induced increase in TLR4 expression was not an osmotic effect. Interestingly, we failed to observe any increase in expression of TLR2 in MMC after repeating five independent, different sets of experiments.

Fig. 1.

A: Hyperglycemia induces Toll like receptor 4 (TLR4) mRNA expression in mouse mesangial cells (MMC). Representative RT-PCR gel shows significantly increased TLR4 mRNA expression on exposing MMC to 25 mM glucose compared with 5.5 mM at 24 h. Densitometric values are normalized to GAPDH and are expressed as means ± SD. *P < 0.05 compared with 5.5 mM glucose or mannitol control. B: surface expression of TLR4 in MMC. TLR 4 surface expression was assessed in the MMC after glucose challenge by flow cytometry at 24 h. Values are expressed as MFI/10,000 events (means ± SD). *P < 0.05 compared with 5.5 mM glucose. #P < 0.05 compared with mannitol control.

Furthermore, flow cytometry revealed significantly increased mean fluorescence intensity of TLR4 expression in the presence of 25 mM glucose compared with normal glucose (P < 0.05) or mannitol control (P < 0.01) (Fig. 1B). Since TLR4 signals via both MyD88 and non-MyD88 pathways, we further confirmed the downstream signaling mediators of the TLR4 pathway by Western blotting for both the MyD88 and non-MyD88 pathways (IRF3 and TRAM). There was significantly increased expression of MyD88, TRAM, and IRF3 in the presence of 25 mM glucose compared with normal glucose conditions as shown in Fig. 2 (P < 0.05), suggesting that both pathways were activated under hyperglycemia.

Fig. 2.

TLR4 signaling pathways in MMC. Western blotting was performed for different adaptor proteins in the cell lysates after exposure of MMC to glucose challenge for 24 h. Densitometric ratios represent adaptor protein-to-β-actin ratios. MyD88, myeloid differentiation factor 88; IRF3, interferon regulatory factor 3; TRAM, Toll interleukin receptor domain-containing, adaptor protein-inducing interferon (TRIF)-related adaptor molecule. *P < 0.05 compared with 5.5 mM glucose.

In addition, hyperglycemia (25 mM) also induced activation of nuclear NF-κB compared with normal glucose (63.3 ± 16 vs. 24.5 ± 6 pg/μg nuclear protein, P < 0.05) and mannitol control (Fig. 3). Because activation of NF-κB p65 by TLR4 leads to transcription of various proinflammatory genes (38), we compared the levels of various inflammatory cytokines such as IL-6, TNF-α, and MCP-1 in the supernatants at 24 h. There were significant increases in levels of IL-6 and MCP-1 in the presence of 25 mM glucose compared with normal glucose (Fig. 4, P < 0.05) but not TNF-α.

Fig. 3.

NF-κB (p65) activation in MMC. ELISA was performed to study activation of transcription factor NF-κB (P65) in the nuclear extracts from mesangial cells at 24 h. *P < 0.05 compared with 5.5 mM glucose or mannitol control.

Fig. 4.

Hyperglycemia-induced circulating expression of proinflammatory cytokines and transforming growth factor (TGF)-β by MMC at 24 h. ELISA was performed to estimate the levels of released cytokines/mg protein in the supernatant. IL-6, monocyte chemotactic protein-1 (MCP-1), AND TNF-α are expressed as pg/mg cell protein. TGF-β is expressed as ng/mg cell protein. *P < 0.05 vs. 5.5 mM glucose.

Previous studies have reported the pivotal role of TGF-β overexpression in mediating MC dysfunction in DN, contributing toward deposition of ECM components in the kidney (44). We confirmed the significantly increased expression of TGF-β in MMC in the presence of 25 mM glucose compared with normal glucose (1.45 ± 0.29 vs. 0.86 ± 0.2 ng/mg protein, P < 0.05, Fig. 4).

DISCUSSION

DN, a microvascular complication, appears to have an important inflammatory component (18, 47, 54). Various reports suggest MC are more prone to oxidative stress and inflammation under high-glucose conditions (5, 40). In addition, abnormal MC expansion and glomerular basement membrane thickening are characteristic features of DN (4, 8).

The present study demonstrates that there is significantly increased TLR4 expression as well as downstream signaling of TLR4 (MyD88 and non-MyD88) pathways in glomerular MMC under high-glucose conditions; however, there was no increase in mRNA TLR2 expression. Also, there was a significant increase in expression of p65 NF-κB and secretion of inflammatory biomediators (IL-6, MCP-1). This suggests the possibility of a TLR4-mediated pathway in promoting abnormal mesangial function and inflammation in DN. Previous studies from our group have demonstrated increased expression and activity of both TLR2 and TLR4 in monocytes and its associated inflammatory mediators in patients with T1DM and T2DM (9, 12, 14). Michelsen et al. (39) found decreased lesion size, lipid content, and macrophage infiltration in the plaques of TLR4 KO mice compared with wild-type mice, hence providing evidence of a role of TLR4 in atherosclerosis. Reyna et al. (46) also found the association of increased TLR4 expression causing activation of NF-κB signaling with the release of inflammatory biomediators such as IL-6 and TNF-α in muscle biopsies of obese and T2DM patients. Recently, Lin et al. (32) found hyperglycemia-induced TLR4 expression and not TLR2 in human proximal TECs and its association with upregulation of inflammatory cytokines and chemokines via NF-κB activation, hence signifying the role of a TLR4-mediated pathway in causing tubulointerstitial inflammation in DN. Thus our novel findings of increased TLR4 in MC and those of Lin et al. in TECs underscore a role of TLR4 in DN.

Devaraj et al. (15) also reported TLR4 KO diabetic mice have significantly reduced levels of MyD88, IRAK-1 protein phosphorylation, TRIF, IRF3, and decreased NF-κB activity and release of biomediators (IL-1 β, IL-6, IL-8, IP-10, MCP-1, IFN-β, and TNF-α) compared with wild-type diabetic mice. While we previously showed that the TLR2KO-streptozotocin (STZ) mice had decreased podocyte loss and albuminuria compared with wild-type STZ mice, we failed to show any change in mesangial expansion by both periodic acid-Schiff (PAS) and silver methenamine staining at 14 wk in these mice (16). In the present study, we failed to demonstrate TLR2 expression in MC, lending plausibility to our previous negative findings suggesting the ameliorating effects of TLR2KO are largely via effects on macrophages and podocytes. While Lin et al. (32) showed a decrease in the progression of DN in the TLR4 KO STZ mice compared with wild-type diabetic mice, they ascribed the benefit to a major effect on TEC. However, they did not report on podocyte function or mesangial expansion. In contrast, Li et al. (30) reported increased TLR2 expression in renal TEC in human renal biopsy DN samples as well as in kidneys of rat models of DN. However, various other reports have also provided evidence of TLR4 association in glomerular injury in DN (2, 29, 34). It has also been shown that lipopolysaccharide activity (the ligand for TLR4) predicts the progression of DN in humans with T1DM, thus further underscoring the role of TLRs in DN (43).

While we did not study the mechanisms of the upregulation in TLR4 in MMC, in previous reports from our group we have shown that hyperglycemia-induced increased TLR4 expression is associated with increased NADPH oxidase activity via PKC (10, 24). This was also confirmed with knockout of the p47^phox subunit of NADPH oxidase (7). NADPH oxidase composed of p22^phox, p47^phox, and p67^phox plays a crucial role in the generation of ROS and hence serves a potential role in providing a host defense. It has also been suggested that formation of advanced glycation end-products (AGEs) can induce TLR activity (24). Reports have also shown AGE-modified LDL (low-density lipoprotein) induced TLR4 expression (22). Thus both of these mechanisms could account for the increase in TLR4 and will be studied in the future. Collectively, the above reports supported a role of TLR4 in DN.

Previous reports have also suggested the pivotal role of TGF-β isoforms and the overproduction of their receptors in diabetic kidney disease, which in turn is associated with hypertrophic and fibrotic/sclerotic manifestations (21, 25, 45, 49, 50, 53). The increased TGF-β also contributes to stimulation of ECM production and suppression of its degradation in DN. We also confirmed the same by showing the hyperglycemic condition (25 mM) contributes to a significantly increased release of TGF-β by MMC compared with normal glucose. Manabe et al. (35) reported 25 mM glucose-induced expression of ROS, production of TGF-β1, MCP-1, and activation of NF-κB and activator protein-1 in an in vitro normal human MC model compared with the 5 mM concentration of glucose.

Conclusion

Thus the present study makes the novel observation that TLR4 expression and activity are increased under hyperglycemia in MC and could contribute to DN. Future studies are thus needed to explore the role of hyperglycemia-induced TLR4 expression in MC using in vivo animal models.

GRANTS

This study was supported by National Heart, Lung, and Blood Institute Grants ROI-HL074360.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: H.K. and A.C. performed experiments; H.K., A.C., and I.J. analyzed data; H.K. and I.J. interpreted results of experiments; H.K. prepared figures; H.K. and I.J. drafted manuscript; H.K., A.C., and I.J. edited and revised manuscript; I.J. provided conception and design of research; I.J. approved final version of manuscript.