Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Apr 1.
Published in final edited form as: Thromb Haemost. 2015 Feb 5;113(4):750–758. doi: 10.1160/TH14-10-0884

Tissue Factor and Toll Like Receptor (TLR)4 in Hyperglycemia-Hyperinsulinemia: Effects in Healthy Subjects, and Type 1 and Type 2 Diabetes Mellitus

Anamika Singh 1, Guenther Boden 2, A Koneti Rao 1,3
PMCID: PMC4677063  NIHMSID: NIHMS736609  PMID: 25653143

SUMMARY

Background

Diabetes mellitus (DM) patients have increased cardiovascular events. Blood tissue factor-procoagulant activity (TF-PCA), the initiating mechanism for blood coagulation, is elevated in DM. We have shown that hyperglycemia (HG), hyperinsulinemia (HI) and combined HG+HI (induced using 24 hr infusion clamps) increases TF-PCA in healthy and T2DM subjects, but not in T1DM subjects. The mechanisms for this are unknown. DM patients have elevated plasma lipopolysaccharide (LPS), a toll-like receptor (TLR) 4 ligand. We postulated that TLR4 plays a role in modulating TF levels.

Objectives and Methods

We studied the effect of HG+HI on TLR4 and TF-PCA in vivo during 24 hr HG+HI infusion clamps in healthy subjects, and T1DM and T2DM subjects, and in vitro in blood.

Results

In vivo, in healthy subjects, 24 hr HG + HI infusion increased TLR4 6-fold, which correlated with TF-PCA (r= 0.91, p<0.0001). T2DM patients showed smaller increases in both. In T1DM subjects, TLR4 declined (50%, p<0.05) and correlated with TF-PCA (r=0.55; p<0.05). In vitro, HG (200 mg/dl added glucose) and HI (1-100 nM added insulin) increased TF-PCA in healthy subjects (~2-fold, 2-4 hr). Insulin inhibited by ~30% LPS-induced increase in TF-PCA and high glucose reversed it. TLR4 levels paralleled TF-PCA (r=0.71, p<0.0001); HG and HI increased TLR4 and insulin inhibited LPS-induced TLR4 increase.

Conclusions

This is first evidence that even in healthy subjects, HG of short duration increases TLR4 and TF-PCA, key players in inflammation and thrombosis. TLR4-TF interplay is strikingly different in non-diabetic, T1DM and T2DM subjects.

Keywords: Diabetes mellitus, Hyperglycemia, Hyperinsulinemia, Tissue factor, Toll-like receptor-4

INTRODUCTION

Patients with diabetes mellitus (DM) have a 2–8-fold higher risk of cardiovascular morbidity and mortality compared with matched controls [1]. Patients with type 2 DM (T2DM) have hyperinsulinemia (HI) and hyperglycemia (HG), both independent risk factors for mortality. Hyperglycemia in hospitalized patients, with or without diabetes, is associated with increased risk for complications and mortality [2-5]. Numerous alterations have been described in patients with T2DM and type 1 diabetes mellitus (T1DM) in blood coagulation system and platelets [6-10], which indicate they are in a prothrombotic state.

Tissue factor (TF) is the primary physiological initiator of blood coagulation and thrombosis [8, 11, 12]. In addition to the TF present in the adventitia of blood vessels and in atherosclerotic plaques, which initiates blood coagulation when vessel walls are injured or plaques fissured [13], there is a circulating pool of TF in blood that is thrombogenic [8, 11, 14-16]. We and others have shown that patients with both T2DM and T1DM have elevated levels of whole-blood TF-PCA and are in a procoagulant state [17-19]. Our studies in healthy non-diabetic subjects using glucose and insulin infusion clamps showed that HG (~200 mg/dl), HI and combined HG+HI elevate whole blood TF-procoagulant activity (TF-PCA) by two, six and nine-fold, respectively [20]. This was associated with increased plasma factor VIII and thrombin-antithrombin complexes indicating there was thrombin generation in vivo. Further, we have reported that raising blood insulin levels, and especially raising blood glucose and insulin levels (HG +HI) together to levels frequently seen in diabetic patients, increased TF-PCA and thrombin generation in T2DM patients [18]. In striking contrast, in T1DM patients [19], HG+HI did not cause any increase in TF-PCA by 24 h, rather a decline was seen with the combination of HG and HI and with selective HG. These findings indicated that the mechanisms regulating the TFPCA and the effects of HG and HI in these three subject groups are distinct, and they are unknown.

Multiple studies highlight the link between TF, toll-like receptor (TLR)4, lipopolysaccharide (LPS) (a TLR4 ligand), and the role of TLR4 in regulating TF expression. Monocytes are a major source of TF in blood and LPS induces monocyte TF surface expression and procoagulant activity [12]. Serum LPS is elevated in both T1DM [21] and T2DM patients [22, 23], and the levels have correlated with insulin and triglyceride levels [23]. LPS stimulates monocytes and other cells via the TLR4, which is an evolutionarily preserved pattern-recognition receptor, expressed on several cell types, including monocytes and platelets [24-26]. Activation of the innate immune system via TLR is implicated in the pathogenesis of insulin resistance and inflammation in DM [27-29]. TLR4 and TLR2 expression is increased in insulin resistant target tissues, skeletal muscle and adipose tissue, of T2DM subjects [27]. Moreover, nutritional free fatty acids whose levels are increased in obesity and DM activate TLR4 signaling in macrophages and adipocytes providing a link between immunity and insulin resistance [29]. Of note, TLR4 gene Asp299Gly polymorphism, which impairs inflammatory responses, is associated with a reduction in circulating C-reactive protein level and a decrease in the risk of angiographic coronary artery disease and clinical DM [30]. In vitro, TF production by LPS-stimulated monocytes is suppressed by insulin [31], similar to insulin's inhibition of platelet function [32], and monocytes from T2DM have impaired sensitivity to the inhibitory effects of insulin, resulting in enhanced TF production [31, 33].

To explore the potential role of TLR4 in regulating the effects of HG+HI on TF expression we studied TLR4 levels during HG+HI infusion clamps using samples from our studies in healthy nondiabetic subjects [20] and patients with T1DM and T2DM [18, 19]. In addition, we studied the effect in vitro of high glucose, high insulin and the combination on whole blood TF-PCA and TLR4 in healthy non-diabetic subjects in the presence and absence of LPS. Our studies document for the first time evidence that HG (~200 mg/dl) of relatively short duration (6 hr) in healthy subjects leads to a proinflammatory and prothrombotic state, with elevated TLR4 and TF-PCA. They provide new insights into changes in TLR4 and the relationship between TLR4 on TF-PCA, and on the strikingly differential effects in T1DM and T2DM patients.

MATERIALS AND METHODS

Materials

Insulin, D-glucose and LPS were obtained from Sigma Aldrich (St. Louis, MO). For the TF-PCA assay factor VIIa, factor X, pooled normal human plasma, phospholipids vesicles and Hemosil recombiplastin were obtained from American Diagnostica (Stamford, CT), Haematologic Technologies Inc. (Essex Junction, VT), George king Bio-Medical Inc. (Overland Park, KS), Avanti Polar Lipids (Alabaster, AL), and Instrumentation Laboratory Company (Lexington, MA) respectively. All the other reagents used were of analytical grade.

Methods

In vivo studies in healthy non-diabetic subjects and patients with T1DM and T2DM during hyperglycemic-hyperinsulinemic infusion clamps

Whole blood TLR4 levels were measured in blood samples available from our previously published studies on TF-PCA using infusion clamps in healthy non-diabetic subjects [20], and patients with type T1DM [19] and T2DM [18]; the details regarding the subjects and the protocols have been described. The study protocols were approved by the Institutional Review Board of Temple University Hospital. Informed written consent was obtained from all subjects after explanation of the nature, purpose and potential risks.

Subjects

Healthy non-diabetic subjects

None of the subjects had a family history of DM or any other endocrine disorder and none were taking any medication. Basal TLR4 and TF-PCA were measured in 18 healthy subjects (mean age 37.3 ± 4.1 years; 10 males). Infusion clamps were performed on 9 subjects (6 males; mean age 37.6 ± 4.2 years); in addition to basal samples, TLR4 was measured in 5 subjects at 6 hr and in 4 subjects at 24 hr, based on availability.

Patients with T1DM

Infusion clamps were performed in 6 subjects [19] (42.3 ± 6.8 years; 5 males; duration of DM 18 ± 8 years). Three subjects were studied over 6 hr; 3 were for 6 hr. In these subjects basal C-peptide levels were very low (0.25 ± 0.09 ng/ml) and did not respond to intravenous glucose. None had clinical signs of renal insufficiency and proliferative retinopathy. Six T1DM patients were on a basal/bolus insulin regimen consisting of insulin glargine in doses ranging from 15-70 units taken at night. Three patients took Novolog 70/30 mix (insulin asparte-protamine/insulin asparte) twice daily in doses ranging from 45-50 units. Only basal TLR4 and TF-PCA levels were measured in 11 T1DM patients (mean age 44.6 ± 5.1; 8 males).

Patients with T2DM

HG+HI infusion clamps were performed in 5 T2DM patients (mean age 49.2 ± 3.9 years, 4 males, duration of DM 7.5 ± 4 years) [18]. Only basal TLR4 and TF-PCA were measured in 17 T2DM patients (56.8 ± 3.4 years; 10 males).

Study Design: Combined elevation of glucose and insulin (High glucose/high insulin clamps)

Healthy, non diabetic subjects

The infusion clamps studies began at ~8AM after an overnight fast, as decribed [20]. The purpose of these studies was to examine effects of raising both blood glucose and insulin levels. A 20% glucose solution was infused intravenously at variable rates adjusted to maintain plasma glucose at ~200 mg/dl (~11 mmol/l). Serum insulin rose (~1000 pmol/L) in response to the HG [20]. Small blood samples (0.25 ml) were collected every 30-60 min initially and every 1-2 hr later for measurement of glucose concentrations. Subjects were fasting but were allowed to drink water ad lib. Plasma electrolytes were monitored every 6 hr, body weight every 12 hr and fluid balances every 6 hr. Potassium (20 mg) and magnesium (400 mg) were given PO every 12 hr. Blood samples were obtained at baseline and at 6 and 24 hr.

Type 1 diabetes mellitus patients

Study participants came to the CRC after an overnight fast. They had their last meal and insulin injection at ~8 PM. At ~8AM the next day, blood samples were collected and a 20% glucose solution was infused IV at variable rates, which were adjusted to maintain plasma glucose ~200-300 mg/dl (~11-16 mmol/l) [19]. Insulin was infused at a rate of 1 mU/kg min. The insulin levels were ~100 U/ml (~600 pmol/L) [19]. Blood glucose concentrations and electrolyte and fluid balances were monitored as described [19]. HG-HI clamps were continued for 6 h in 6 patients and for 24 h in 3 patients.

Type 2 Diabetes mellitus patients

A 20% glucose solution was infused intravenously at variable rates, adjusted to maintain plasma glucose at ~200 mg/dl (~11 mmol/l) [18]. Blood samples (0.25 ml) were collected every 30-60 min initially and every 1-2 h later for measurement of blood glucose. Serum insulin levels rose from 148 ± 49 pmol/L at 0 h to a mean of 390 ± 139 pmol/L over 24 h.

Effects of hyperglycemia and hyperinsulinemia in vitro on TF-PCA and TLR4

Blood was obtained from 5 healthy nondiabetic subjects (mean age 31.2 ± 6.1 years; 4 males) who were not on any medications. They were recruited from the staff at Temple University Health Sciences Center. All subjects provided an informed consent and the studies were approved by the Institutional Review Board of Temple University.

Whole blood (105 ml) was drawn into one-tenth volume of 3.8% sodium citrate by venipuncture from healthy, medication-free volunteers (n=5). Aliquots of blood (7 ml) were incubated at 37°C for 4 hr with added 200 mg/dL glucose alone, insulin alone (1 nM, 10 nM or 100 nM), or combinations of glucose and insulin, in the presence and absence of LPS (1 μg/ml) [31]. Aliquots (1.5 ml) of blood were centrifuged for 20 min at 1590 g to obtain cell-free plasma for glucose and insulin measurements. Aliquots (1 ml) of whole blood were frozen at baseline (0 hr). Remaining blood was incubated at 37° C and aliquots (1 ml - 1.5 ml) were removed at 2 and 4 hr for TF-PCA and TLR4 and for harvesting plasma. Samples were stored at −70° C until assayed.

Assays

Whole-blood tissue factor-procoagulant activity (TF-PCA) assay

This was measured as described previously [15, 20]. Blood lysates and cellular components were collected from aliquots of whole blood subjected to three cycles of freezing and thawing followed by centrifugation and washing, and they were finally suspended in HBSA buffer (10 mmol/l HEPES, 137 mmol/l NaCl, 5.38 mmol/l KCl, 5.55 mmol/l glucose, and 0.1% bovine serum albumin, pH 7.5) [15]. TF-PCA was measured in whole-blood cell lysates with a two-stage clotting assay using recombinant factor VIIa, factor X, and normal human plasma containing phospholipid vesicles [15]. This assay measures cell-bound and microparticle-associated TF in lysed whole blood. Recombinant human TF (Hemosil Recombiplastin) was used as a standard. The clotting times were measured using a Coagulometer (Amelung KC-4, Trinity Biotech). This assay has been validated by Key et al (15) and by us to show that it is specific for TF; the activity is entirely dependent on the addition of FVIIa and is inhibited ~95% by an anti-TF antibody. Moreover, the activity is highly inducible by LPS (Fig 5) (15), which further supports that TF is being measured. The TF-PCA levels during infusion clamps are from our published studies [18-20].

Figure 5.

Figure 5

Open in a new tab

A. Effect of hyperglycemia and hyperinsulinemia on whole blood TLR4 in the absence and presence of LPS. Whole blood was incubated with added glucose (200 mg/dl), insulin (1 or 100 nM), alone or in combination for 4 hr. This was performed in the presence or absence of added LPS (1 μg/ml). Aliquots were removed at baseline and 4 hr for measurement of TLR4. Each experiment was performed using blood from 5 health subjects. * p< 0.05 compared to control (basal).

Figure 5B. Correlation between TF-PCA and TLR4 levels. Shown are the TF-PCA and TLR4 levels at baseline and 4 hr in samples from experiments shown in Fig 5A, and the Pearson correlation coefficient.

Other Assays

TLR4 levels were measured in the same samples as used for TF-PCA using an ELISA (USCN Life Science, Inc., Wuhan, China); the whole-blood lysates were solubilized by overnight incubation with 0.1% Triton X-100. Plasma glucose was measured with a glucose analyzer using the glucose oxidase method and serum insulin by RIA (Linco Research, St. Charles, MO).

Statistical analysis

Comparisons of the values obtained within each group at different time points were performed using the Student's t test. Statistical significance was defined as p < 0.05. All results are presented as means ± SE. The relationships between variables were assessed using Pearson correlation coefficients.

RESULTS

Basal TLR4 levels in non-diabetic and diabetic subjects

Basal TLR4 levels in T2DM patients (0.81 ± 0.08 ng/ml) were markedly decreased to ~35% of those in healthy subjects (2.23 ± 0.30 ng/ml, p<0.001) (Fig 1A). The levels were lower than in T1DM subjects (3.47 ± 0.98 ng/ml, p<0.05). The corresponding basal TF-PCA levels in the same subjects are shown in Fig 1B.

Figure 1.

Figure 1

Open in a new tab

Basal TLR4 (A) and TF-PCA (B) levels in healthy nondiabetic subjects, and patients with type 1 and type 2 diabetes mellitus

In vivo studies on TLR4 levels during hyperglycemia-hyperinsulinemia clamps

We measured TLR4 levels in samples obtained from our studies in healthy subjects and DM patients during combined HG + HI infusion clamps [18-20]. These studies showed marked differences in the TLR4 responses between the 3 groups. In non-diabetic healthy subjects there was a striking elevation in TLR4 with ~2-fold increase by 6 hr and 6-fold by 24 hr (Fig 2A). T2DM patients started with markedly lower basal levels as compared to non-diabetic and T1DM subjects and showed a 25% increase over basal levels. The levels in T2DM at baseline and 24 hr were 0.67 ± 0.05 ng/ml and 0.84 ± 0.07 ng/ml, respectively (p<0.05) (Fig 2A). In contrast, in T1DM subjects, TLR4 declined from 3.33 ± 1.36 ng/ml at baseline to 2.24 ± 1.30 ng/ml (p=0.03, n=6) at 6 hr and to 1.18 ± 0.10 at 24 hr (p=0.02, n=3). Thus, there was a striking elevation in TLR4 in non-diabetic subjects, a decrease in T1DM subjects, and a small increase (25%) in T2DM patients.

Figure 2. TLR4 (left panel) and TF-PCA (right panel) in whole blood samples during combined hyperglycemia and hyperinsulinemia clamps in healthy nondiabetic subjects, and patients with type 1 and type 2 diabetes mellitus.

Figure 2

Open in a new tab

Shown are results in 9 nondiabetic subjects (squares), with levels at 6 hr in 5 subjects and at 24 hr in 4 subjects; in 6 T1DM subjects (circles) with levels in 3 subjects at 0, 6 and 24 hr and 6 subjects at 0 and 6 hr; and in 5 T2DM subjects (diamonds) with levels at 0, 6 and 24 hr in all subjects. TF-PCA results are from our previously published studies in non-diabetic subjects [20], T1DM [19] and T2DM [18]. *p<0.05 compared to 0 hr; # p<0.05 compared to 6 hr.

The pattern of changes in TF-PCA (Fig 2B) and TLR4 (Fig 2A) following HG+HI clamps were identical both in non-diabetic subjects and in T1DM with increases over 24 hr in the former group and a decline in the latter. There was a good correlation between TLR4 and TF-PCA levels (T1DM, r=0.55, p=0.03; nondiabetic controls, r=0.91, p<0.0001) (Fig 3), even though TFPCA and TLR4 levels increased in nondiabetic subjects and declined in T1DM. In T2DM patients, TFPCA levels rose with HG+HI by ~30% over already elevated basal levels – this was asociated with a 25% increase in TLR4 (Fig 2). The correlation between TLR4 and TF-PCA (r=0.43, p=NS) was not significant.

Figure 3. Correlation between TF-PCA and TLR4 levels in healthy non-diabetic subjects, and patients with T1DM.

Figure 3

Open in a new tab

Shown are levels obtained before and during hyperglycemiahyperinsulinemia infusion clamps in 9 healthy subjects and 5 T1DM patients and the Pearson correlation coefficients.

We explored the relationship between TLR4 and insulin and glucose levels during HG+HI clamps (0, 6, 24 hr) in the 3 groups (Table 1). The relationship between TLR4 and insulin in these three groups was distinctly different. In healthy non-diabetic subjects TLR4 levels were positively correlated with insulin levels (r=0.60, p<0.01); in T1DM there was an inverse relationship (r= −0.68; p=0.01, with exclusion of one subject with extremely high TLR4 levels at 0 and 6 hr). In T2DM there was no relationship observed (r=0.07, p=NS). TLR4 levels were positively related to glucose levels in non-diabetic subjects (r=0.73; p<0.001). No relationship was noted in T1DM or T2DM subjects.

Table 1.

Relationship between TLR4 and insulin and between TLR4 and glucose levels in 9 non-diabetic subjects, 5 T2DM patients and 6 T1DM patients before and during HG+HI infusion clamps.

TLR4 vs Glucose r (p) vs Insulin r (p)
Non-diabetic healthy subjects 0.73 (<0.001) 0.60 (<0.01)
Type 2 diabetes mellitus patients 0.22 (0.44) 0.07 (0.8)
Type 1 diabetes mellitus patients −0.21 (0.44) −0.68 (0.01)
Open in a new tab

Shown are the Pearson correlation coefficients and p values.

Effect in vitro of high glucose and high insulin on TF-PCA in healthy non-diabetic subjects

To better understand the effects of high glucose and high insulin on TF-PCA and TLR4 we performed studies in vitro in the presence and absence of LPS. The mean basal glucose and insulin levels prior to addition of exogenous glucose or insulin were 75 mg/dL ± 4.2 (mean±SE, n=3) and 99 + 24.4 pmol/L, respectively. With added 200 mg/dL glucose, TF-PCA increased by ~2-fold (Fig 4A) at 2 hr and by ~2.5-fold by 4 hr (Fig 4B). Addition of insulin, 1, 10 or 100 nM increased TF-PCA by 2-fold at 2 hr and 2-4 fold by 4 hr. Thus, with added glucose or insulin, there was an increase in TF-PCA. The combination of high glucose (200 mg/dL) with insulin (1, 10 or 100 nM) increased TF-PCA at 2 hr by 2-3 fold, and by 3-4 fold at 4 hr. In general, the highest levels were with the combination of HG and HI. TF-PCA level with high glucose plus high insulin (100 nM) was significantly higher than with glucose alone at both 2 hr (69.6 ± 10.8 vs 45 ± 6.7, p=0.02) and 4 hr (104.0 ± 18.5 vs 62 ± 10, p=0.01). At 2 hr TF-PCA with the HG+HI combination was higher than with respective insulin concentration alone (Fig 4A). At 4 hr, it was significantly higher in combination with10 nM insulin (Fig 4B).

Figure 4. Effect of hyperglycemia and hyperinsulinemia on whole blood TF-PCA in the absence (A, B) and presence of lipopolysaccharide (LPS) (C, D) at 2 hr (A, C) and 4 hr (B, D).

Figure 4

Open in a new tab

Whole blood was incubated with added glucose (200 mg/dl), insulin (1, 10, or 100 nM), alone or in combination for up to 4 hr, in the presence or absence of added LPS (1 μg/ml).

Aliquots were removed at baseline, 2 and 4 hr for measurement of TF-PCA. Each experiment was performed using blood from 5 health subjects.

Figure 4A and 4B: * p< 0.05 compared to baseline (control), ** p< 0.05 compared to added glucose alone.

Figure 4C and 4D: * p< 0.05 compared to LPS alone; and ** p< 0.05 compared to LPS plus glucose.

Effect in vitro of high glucose and high insulin on TF-PCA in the presence of LPS

LPS (1 μg/ml) alone markedly increased the mean TF-PCA by 32 fold at 2 hr and by 25-fold at 4 hr (Figure 5C and D). With added glucose TF-PCA was higher than with LPS alone but the difference was not statistically significant. However, added insulin inhibited LPS-induced increase in TF-PCA levels. Insulin 10 nM inhibited the levels by ~35% (p=0.04) at 2 hr and 30% (p=0.01) by 4 hr. This was in contrast to that observed in the absence of LPS, where insulin by itself increased TF-PCA (Fig 4A). Most strikingly, with the combination of LPS, high glucose and high insulin (1 nM or 100 nM) there was a marked and further increase by ~3 fold in TF-PCA over the already high levels with LPS alone. These elevated TF-PCA levels were higher than with LPS plus glucose and LPS plus insulin (Fig 4C and D). Thus, glucose reversed the inhibitory effect of insulin and enhanced the levels even higher.

Effect in vitro of high glucose and high insulin on TLR4

TLR4 levels were measured at 4 hr. In general, the effects of glucose and insulin on TLR4 levels (Fig 5A) appeared to parallel the TF-PCA levels (Fig 4) and there was a strong correlation between TF-PCA and TLR4 (r=0.71, p=<0.0001) (Fig 5B). In the absence of LPS, there was a ~4-5 fold increase in TLR4 with added high glucose and with added high insulin (1 nM) (Fig 5A). Combined high glucose plus high insulin increased TLR4 over basal, but not over that with added glucose or insulin alone. LPS caused a ~10- fold increase in TLR4 over basal (Fig 5A). In presence of LPS, added glucose did not further increase TLR4. However, insulin (1 and 100 nM) inhibited LPS-induced TLR4 by ~65%. The highest TLR4 levels were noted with the combination of LPS, high glucose and high insulin suggesting that high glucose overrides the inhibitory effect of insulin on LPS-stimulated TLR4.

Discussion

The combination of our in vivo studies following infusion of HG and HI in healthy non-diabetic subjects and DM patients and the in vitro studies provides novel insights into the inter-play between TF-PCA and TLR4. Our in vivo infusion studies provide the novel and important evidence that HG of a relatively short duration (6 hr), even in healthy non-diabetic subjects, induces a striking increase in TLR4 and TF-PCA, both key players in inflammation, thrombosis and atherosclerosis [24-26, 34]. Our previous studies have shown that the TF-PCA elevation induced by HG+HI is associated with a strong evidence of an in vivo procoagulant state with elevation in plasma TAT complexes and factor VIII, increased expression of CD40 ligand on platelets and TF on monocytes, and increased circulating platelet-monocyte aggregates [20, 35]. High glucose in vitro has been shown to induce TLR expression in human monocytes [36]. Our previous studies [18-20] involving combined HG+HI infusion clamps showed striking differences between the 3 subject groups studied with respect to TF-PCA, and now we provide new evidence of a parallel effect on TLR4 (Fig 2). Nondiabetic healthy subjects showed marked increases in TF-PCA and TLR4 levels (Fig 2) with a good correlation between the two (Fig 3A). Previous studies have established that activation of monocyte TLR4 leads to downstream activation of NF-κβ and TF production [24-26]. Our findings, therefore, are consistent with an in vivo regulatory effect of TLR4 on TF-PCA expression and as the basis for downstream TF levels.

In vitro, both HG and HI individually increased TF-PCA in blood from healthy subjects with a further increase with combined HG+HI (Fig 4A, 4B). However, insulin inhibited by ~30% (Fig 4) LPS-induced marked increase in TF-PCA, which was in contrast to an increase with insulin alone (Fig 4A). More importantly, in the presence of LPS, the combination of HG + HI induced the highest TF-PCA levels (Figs 4C, 4D) indicating that HG overrides the insulin inhibition and increases levels further. The changes in TLR4 levels paralleled and strongly correlated with TF-PCA (Fig 4). HG and HI alone induced 4-5-fold increases over basal TLR4 levels. As with TF-PCA, insulin alone inhibited (~65%) LPS-induced marked increase in TLR4, and added HG reversed this inhibition and further increased TLR4 levels (Fig 4A). These parallel changes in TF-PCA and TLR4 are in line with the established role of TLR4 in regulating TF expression on activation by ligands including LPS [24-26]. Our findings are consistent with the paradigm that insulin inhibits TLR4 and TF-PCA while high glucose stimulates their expression, and that this is influenced by LPS levels, which are known to be increased in T1DM and T2DM [21-23].

HG-induced upregulation of TLR4 is also relevant in processes beyond recognition of pathogen associated molecular patterns (PAMPs), innate immunity and cytokine release [24]. TLR4 is activated by host-derived molecules and is implicated in pathogenetic effects. Free fatty acids activate TLR4 on macrophages, adipocytes and liver to induce pro-inflammatory cytokine expression and insulin resistance [29], and this is particularly relevant in DM. TLR4 downregulation is associated with a decrease in inflammation and clinical DM [30]. Hypercholesterolemia-induced prothrombotic state and monocyte TF expression is driven by engagement of TLR4/TLR6 complex by oxidized low-density lipoproteins [37]. Moreover, TLR4 is implicated in cardiac ischemia and failure [38]. There is an increase in TLR4 in myocardial tissue of diabetic mice as early as 3 days after appearance of hyperglycemia [39]. HG-induced myocardial apoptosis was prevented by silencing of TLR4 gene expression [39]. These studies reflect on the significance of HG-induced elevation of TLR4.

In our studies, the effects of combined HG + HI on TLR4 and TF-PCA are strikingly different between T1DM and T2DM (Fig 2 and 3). In T2DM patients, TLR4 levels started low and showed a small (~25%) increase by 24 hr, as also noted with TF-PCA (Fig 2). We postulate that the lack of a marked increase in TLR4 in T2DM patients (as was noted in healthy subjects) is due to an inhibitory effect of chronic HI. In contrast to nondiabetic and T2DM subjects, T1DM subjects showed a decline in TLR4 at 6 and 24 hr, which was accompanied by a decrease in TF-PCA (Fig 2,3). Moreover, there was an inverse relationship between insulin and TLR4 levels (Table 1) consistent with an inhibitory effect of insulin.

We found that the basal TLR4 levels were distinctly different between the three groups. They were lower in T2DM patients than in T1DM and non-diabetic individuals (Fig 1). The two subjects with the highest TLR4 levels (>9.5 ng/ml) had T1DM suggesting that TLR4 levels may be elevated in T1DM patients, as noted previously in monocytes from T1DM patients [40]. High glucose levels are reported to increase monocytes TLR4 levels in vitro [36]. Because, in general, T1DM subjects have HG but not HI, the normal or possibly elevated TLR4 levels may reflect a lack of chronic inhibitory effect of HI. Insulin has been shown to decrease monocyte TLR4 mRNA levels in T2DM [41], and based on our in vitro studies (Fig 4C, 4D), we postulate that chronic HI present in T2DM decreases basal TLR4 levels. Low TLR4 levels have been observed in tissue from skin ulcers in T2DM patients [42]. In T2DM the basal TF-PCA was elevated (Fig 2) and this may be related to a stimulatory effect via pathways other than TLR4, such as through the advanced glycation end products (AGEs) and oxidized lipoproteins, which can stimulate TF-PCA via receptor for AGE (RAGE) and CD36, respectively [43, 44] (Fig 2).

TLR4 levels were measured in our studies in whole blood lysates, which includes membranes from blood cells, including monocytes and platelets, both of which express TLR4. Both TF-PCA and TLR4 are cell surface membrane proteins and were measured in the same membrane samples, which supports our conclusions regarding their expression and relationship. Moreover, we have previously shown that HG+HI associated increase in whole blood TF-PCA is accompanied by a parallel increase in TF surface expression by flow cytometry and in monocyte TF mRNA [20, 35].

Overall, our findings document a hitherto unrecognized proinflammatory and prothrombotic state, with elevated TLR4 and TF-PCA, induced by HG of a relatively short duration even in healthy subjects. Our studies are consistent with the paradigm that high glucose elevates and insulin inhibits TLR4 and TF-PCA with a potential effect of endogenous LPS, a TLR4 ligand. They document marked differences in the circulating levels and on the effect of HG and HI on TLR4 and TF-PCA in healthy and T1DM and T2DM subjects, and provide a cogent explanation for the strikingly different effects on TF-PCA we have observed [18-20] in these subjects.

What is known on this topic

  • Diabetes mellitus (DM) patients are recognized to be in a prothrombotic state with an increased incidence of cardiovascular events.

  • Blood tissue factor-procoagulant activity (TF-PCA), the initiating mechanism for blood coagulation, is elevated in DM, along with elevated plasma lipopolysaccharide (LPS), a ligand for the toll-like receptor (TLR)4.

What this paper adds

  • In healthy nondiabetic subjects, 24 hr of hyperglycemia – hyperinsulinemia induced by glucose infusion results in a 6-fold increase in TLR4, which correlates with TF-procogulant activity.

  • This is the first evidence that even in healthy subjects, hyperglycemia of short duration increases TLR4 and TF-PCA, key players in inflammation and thrombosis.

  • TLR4-TF interplay is strikingly different in healthy non-diabetic subjects and patients with T1DM and T2DM.

Acknowledgments

This work was supported by the National Institutes of Health grant R01-HL-073367. The excellent assistance of Ms. Denise Tierney in the preparation of the manuscript is gratefully acknowledged.

Footnotes

Disclosure of Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

  • 1.Grundy SM, Howard B, Smith S, Jr., Eckel R, Redberg R, Bonow RO. Prevention Conference VI: Diabetes and Cardiovascular Disease: executive summary: conference proceeding for healthcare professionals from a special writing group of the American Heart Association. Circulation. 2002;105(18):2231–9. doi: 10.1161/01.cir.0000013952.86046.dd. [DOI] [PubMed] [Google Scholar]
  • 2.Capes SE, Hunt D, Malmberg K, Gerstein HC. Stress hyperglycaemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet. 2000;355(9206):773–8. doi: 10.1016/S0140-6736(99)08415-9. [DOI] [PubMed] [Google Scholar]
  • 3.Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE. Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab. 2002;87(3):978–82. doi: 10.1210/jcem.87.3.8341. [DOI] [PubMed] [Google Scholar]
  • 4.Falciglia M, Freyberg RW, Almenoff PL, D'Alessio DA, Render ML. Hyperglycemia-related mortality in critically ill patients varies with admission diagnosis. Crit Care Med. 2009;37(12):3001–9. doi: 10.1097/CCM.0b013e3181b083f7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kosiborod M, Rathore SS, Inzucchi SE, Masoudi FA, Wang Y, Havranek EP, et al. Admission glucose and mortality in elderly patients hospitalized with acute myocardial infarction: implications for patients with and without recognized diabetes. Circulation. 2005;111(23):3078–86. doi: 10.1161/CIRCULATIONAHA.104.517839. [DOI] [PubMed] [Google Scholar]
  • 6.Targher G, Chonchol M, Zoppini G, Franchini M. Hemostatic disorders in type 1 diabetes mellitus. Semin Thromb Hemost. 2011;37(1):58–65. doi: 10.1055/s-0030-1270072. [DOI] [PubMed] [Google Scholar]
  • 7.Carmassi F, Morale M, Puccetti R, De Negri F, Monzani F, Navalesi R, et al. Coagulation and fibrinolytic system impairment in insulin dependent diabetes mellitus. Thromb Res. 1992;67(6):643–54. doi: 10.1016/0049-3848(92)90068-l. [DOI] [PubMed] [Google Scholar]
  • 8.Bogdanov VY, Osterud B. Cardiovascular complications of diabetes mellitus: The Tissue Factor perspective. Thromb Res. 2010;125(2):112–8. doi: 10.1016/j.thromres.2009.06.033. [DOI] [PubMed] [Google Scholar]
  • 9.Alzahrani SH, Ajjan RA. Coagulation and fibrinolysis in diabetes. Diab Vasc Dis Res. 2010;7(4):260–73. doi: 10.1177/1479164110383723. [DOI] [PubMed] [Google Scholar]
  • 10.Kakouros N, Rade JJ, Kourliouros A, Resar JR. Platelet function in patients with diabetes mellitus: from a theoretical to a practical perspective. Int J Endocrinol. 2011:2011742719. doi: 10.1155/2011/742719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Rauch U, Nemerson Y. Tissue factor, the blood, and the arterial wall. Trends Cardiovasc Med. 2000;10(4):139–43. doi: 10.1016/s1050-1738(00)00049-9. [DOI] [PubMed] [Google Scholar]
  • 12.Tilley R, Mackman N. Tissue factor in hemostasis and thrombosis. Semin Thromb Hemost. 2006;32(1):5–10. doi: 10.1055/s-2006-933335. [DOI] [PubMed] [Google Scholar]
  • 13.Wilcox JN, Smith KM, Schwartz SM, Gordon D. Localization of tissue factor in the normal vessel wall and in the atherosclerotic plaque. Proc Natl Acad Sci U S A. 1989;86(8):2839–43. doi: 10.1073/pnas.86.8.2839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Giesen PL, Rauch U, Bohrmann B, Kling D, Roque M, Fallon JT, et al. Blood-borne tissue factor: another view of thrombosis. Proc Natl Acad Sci U S A. 1999;96(5):2311–5. doi: 10.1073/pnas.96.5.2311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Key NS, Slungaard A, Dandelet L, Nelson SC, Moertel C, Styles LA, et al. Whole blood tissue factor procoagulant activity is elevated in patients with sickle cell disease. Blood. 1998;91(11):4216–23. [PubMed] [Google Scholar]
  • 16.Chou J, Mackman N, Merrill-Skoloff G, Pedersen B, Furie BC, Furie B. Hematopoietic cell-derived microparticle tissue factor contributes to fibrin formation during thrombus propagation. Blood. 2004;104(10):3190–7. doi: 10.1182/blood-2004-03-0935. [DOI] [PubMed] [Google Scholar]
  • 17.Sambola A, Osende J, Hathcock J, Degen M, Nemerson Y, Fuster V, et al. Role of risk factors in the modulation of tissue factor activity and blood thrombogenicity. Circulation. 2003;107(7):973–7. doi: 10.1161/01.cir.0000050621.67499.7d. [DOI] [PubMed] [Google Scholar]
  • 18.Boden G, Vaidyula VR, Homko C, Cheung P, Rao AK. Circulating tissue factor procoagulant activity and thrombin generation in patients with type 2 diabetes: effects of insulin and glucose. J Clin Endocrinol Metab. 2007;92(11):4352–8. doi: 10.1210/jc.2007-0933. [DOI] [PubMed] [Google Scholar]
  • 19.Singh A, Boden G, Homko C, Gunawardana J, Rao AK. Whole-Blood Tissue Factor Procoagulant Activity Is Elevated in Type 1 Diabetes: Effects of hyperglycemia and hyperinsulinemia. Diabetes Care. 2012;35(6):1322–7. doi: 10.2337/dc11-2114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Vaidyula VR, Rao AK, Mozzoli M, Homko C, Cheung P, Boden G. Effects of hyperglycemia and hyperinsulinemia on circulating tissue factor procoagulant activity and platelet CD40 ligand. Diabetes. 2006;55(1):202–8. [PubMed] [Google Scholar]
  • 21.Devaraj S, Dasu MR, Park SH, Jialal I. Increased levels of ligands of Toll-like receptors 2 and 4 in type 1 diabetes. Diabetologia. 2009;52(8):1665–8. doi: 10.1007/s00125-009-1394-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Creely SJ, McTernan PG, Kusminski CM, Fisher M, Da Silva NF, Khanolkar M, et al. Lipopolysaccharide activates an innate immune system response in human adipose tissue in obesity and type 2 diabetes. Am J Physiol Endocrinol Metab. 2007;292(3):E740–7. doi: 10.1152/ajpendo.00302.2006. [DOI] [PubMed] [Google Scholar]
  • 23.Al-Attas OS, Al-Daghri NM, Al-Rubeaan K, da Silva NF, Sabico SL, Kumar S, et al. Changes in endotoxin levels in T2DM subjects on anti-diabetic therapies. Cardiovasc Diabetol. 2009:820. doi: 10.1186/1475-2840-8-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Medzhitov R. Toll-like receptors and innate immunity. Nat Rev Immunol. 2001;1(2):135–45. doi: 10.1038/35100529. [DOI] [PubMed] [Google Scholar]
  • 25.Akira S, Takeda K. Toll-like receptor signalling. Nat Rev Immunol. 2004;4(7):499–511. doi: 10.1038/nri1391. [DOI] [PubMed] [Google Scholar]
  • 26.Bruserud O. Bidirectional crosstalk between platelets and monocytes initiated by Toll-like receptor: an important step in the early defense against fungal infections? Platelets. 2013;24(2):85–97. doi: 10.3109/09537104.2012.678426. [DOI] [PubMed] [Google Scholar]
  • 27.Kim JJ, Sears DD. TLR4 and Insulin Resistance. Gastroenterol Res Pract. 2010:2010. doi: 10.1155/2010/212563. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Jain SK, Rains JL. Toll-like receptor-4 and vascular inflammation in diabetes: editorial. Cytokine. 2011;55(3):446–7. doi: 10.1016/j.cyto.2011.05.009. [DOI] [PubMed] [Google Scholar]
  • 29.Shi H, Kokoeva MV, Inouye K, Tzameli I, Yin H, Flier JS. TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest. 2006;116(11):3015–25. doi: 10.1172/JCI28898. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Kolek MJ, Carlquist JF, Muhlestein JB, Whiting BM, Horne BD, Bair TL, et al. Toll-like receptor 4 gene Asp299Gly polymorphism is associated with reductions in vascular inflammation, angiographic coronary artery disease, and clinical diabetes. Am Heart J. 2004;148(6):1034–40. doi: 10.1016/j.ahj.2004.05.049. [DOI] [PubMed] [Google Scholar]
  • 31.Gerrits AJ, Koekman CA, Yildirim C, Nieuwland R, Akkerman JW. Insulin inhibits tissue factor expression in monocytes. J Thromb Haemost. 2009;7(1):198–205. doi: 10.1111/j.1538-7836.2008.03206.x. [DOI] [PubMed] [Google Scholar]
  • 32.Ferreira IA, Mocking AI, Feijge MA, Gorter G, van Haeften TW, Heemskerk JW, et al. Platelet inhibition by insulin is absent in type 2 diabetes mellitus. Arterioscler Thromb Vasc Biol. 2006;26(2):417–22. doi: 10.1161/01.ATV.0000199519.37089.a0. [DOI] [PubMed] [Google Scholar]
  • 33.Gerrits AJ, Koekman CA, van Haeften TW, Akkerman JW. Increased tissue factor expression in diabetes mellitus type 2 monocytes caused by insulin resistance. J Thromb Haemost. 2011;9(4):873–5. doi: 10.1111/j.1538-7836.2011.04201.x. [DOI] [PubMed] [Google Scholar]
  • 34.Stark RJ, Aghakasiri N, Rumbaut RE. Platelet-derived Toll-like receptor 4 (Tlr-4) is sufficient to promote microvascular thrombosis in endotoxemia. PLoS One. 2012;7(7):e41254. doi: 10.1371/journal.pone.0041254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Vaidyula VR, Boden G, Rao AK. Platelet and monocyte activation by hyperglycemia and hyperinsulinemia in healthy subjects. Platelets. 2006;17(8):577–85. doi: 10.1080/09537100600760814. [DOI] [PubMed] [Google Scholar]
  • 36.Dasu MR, Devaraj S, Zhao L, Hwang DH, Jialal I. High glucose induces toll-like receptor expression in human monocytes: mechanism of activation. Diabetes. 2008;57(11):3090–8. doi: 10.2337/db08-0564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Owens AP, 3rd, Passam FH, Antoniak S, Marshall SM, McDaniel AL, Rudel L, et al. Monocyte tissue factor-dependent activation of coagulation in hypercholesterolemic mice and monkeys is inhibited by simvastatin. J Clin Invest. 2012;122(2):558–68. doi: 10.1172/JCI58969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Frantz S, Ertl G, Bauersachs J. Toll-like receptor signaling in the ischemic heart. Front Biosci. 2008:135772–9. doi: 10.2741/3114. [DOI] [PubMed] [Google Scholar]
  • 39.Zhang Y, Peng T, Zhu H, Zheng X, Zhang X, Jiang N, et al. Prevention of hyperglycemia-induced myocardial apoptosis by gene silencing of Toll-like receptor-4. J Transl Med. 2010:8133. doi: 10.1186/1479-5876-8-133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Devaraj S, Dasu MR, Rockwood J, Winter W, Griffen SC, Jialal I. Increased toll-like receptor (TLR) 2 and TLR4 expression in monocytes from patients with type 1 diabetes: further evidence of a proinflammatory state. J Clin Endocrinol Metab. 2008;93(2):578–83. doi: 10.1210/jc.2007-2185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Ghanim H, Mohanty P, Deopurkar R, Sia CL, Korzeniewski K, Abuaysheh S, et al. Acute modulation of toll-like receptors by insulin. Diabetes Care. 2008;31(9):1827–31. doi: 10.2337/dc08-0561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Kanhaiya NK, A SK, G Singh K. Differential expression of toll like receptor 4 in type 2 diabetic patients with impaired wound healing. J Diabetes Metab. 2013:4260. [Google Scholar]
  • 43.Zhu W, Li W, Silverstein RL. Advanced glycation end products induce a prothrombotic phenotype in mice via interaction with platelet CD36. Blood. 2012;119(25):6136–44. doi: 10.1182/blood-2011-10-387506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Ichikawa K, Yoshinari M, Iwase M, Wakisaka M, Doi Y, Iino K, et al. Advanced glycosylation end products induced tissue factor expression in human monocyte-like U937 cells and increased tissue factor expression in monocytes from diabetic patients. Atherosclerosis. 1998;136(2):281–7. doi: 10.1016/s0021-9150(97)00221-9. [DOI] [PubMed] [Google Scholar]

RESOURCES