Secondary Logo

Journal Logo

Basic Science Aspects

Reduced Insulin Resistance Contributes to the Beneficial Effect of Protein Tyrosine Phosphatase-1B Deletion in a Mouse Model of Sepsis

Delile, Eugénie*; Nevière, Rémi; Thiébaut, Pierre-Alain*; Maupoint, Julie*; Mulder, Paul*; Coquerel, David*; Renet, Sylvanie*; Rieusset, Jennifer; Richard, Vincent*; Tamion, Fabienne*,§

Author Information
doi: 10.1097/SHK.0000000000000853

Abstract

INTRODUCTION

Septic shock is a major healthcare issue and the first cause of both mortality and costs in intensive care units (1). Sepsis-induced metabolic alterations are considered major components in outcome. Hyperglycemia frequently complicates this illness and is associated with worse prognosis (2, 3). Thus, modulation of glucose metabolism in septic shock is considered to be an effective therapeutic approach (4), and adequate glucose control by intravenous insulin administration has been shown to improve outcomes and reduce mortality in septic patients (5). However, subsequent ICU trials failed to confirm a benefit of tight control of glycemia on prognosis in critically ill patients, while emphasizing the potential role of hypoglycemia in explaining the divergent results (6).

During septic shock, inflammatory processes cause important modifications in insulin signaling pathways (7). Pro-inflammatory cytokines, as well as endotoxins via Toll-like receptor 4, participate in the development of insulin resistance by stimulating hepatic glucose production and altering insulin signaling (8). Pro-inflammatory cytokines are responsible for the down-regulation of GLUT4 transcription and inhibitory PI3K (phosphoinositide-3-kinase/serine threonine kinase)/Akt pathway. Inhibition of PI3K/Akt signaling suppresses the translocation of GLUT4 from an intracellular component to the plasma membrane. GLUTs are membrane proteins that facilitate glucose transport over plasma cell membranes. Expression and activity of GLUT4 are stimulated by insulin (9), and alteration of GLUT4 activity plays a central role in insulin resistance (10). Importantly, this is also the case in sepsis, where both the inhibition of GLUT4 expression and suppression of translocation reduce glucose uptake and lead to insulin resistance (11, 12).

Protein tyrosine phosphatase 1B (PTP1B) is a known negative regulator of insulin signaling. Gene deletion of PTP1B in mice stimulates insulin sensitivity and prevents diet-induced obesity, suggesting that this enzyme downregulates insulin signaling (13). Recognition that PTP1B is a negative regulator of Janus kinase-signal transducer and activator of transcription signaling places it as a key link between metabolic diseases and inflammation. In parallel, PTP1B is an important mediator of endothelial dysfunction; PTP1B inhibition/gene deletion induces potent endothelial protective effects in various cardiovascular diseases (14–16) or sepsis (17). We have shown previously that PTP1B gene deletion improved the cardiovascular consequences of endotoxin shock (17). However, whether PTP1B inhibition modulates insulin-resistance and glucose uptake in sepsis is still unknown. This hypothesis however cannot be assessed in the context of endotoxin shock since LPS is known to induce resistant hypoglycaemia in rodents (18). Therefore, the main objective of our study was to determine the effect of PTP1B gene deletion on glucose metabolism and insulin resistance, and its consequences on vascular and myocardial dysfunction in sepsis. For this purpose we used a mouse model of cecal ligation and puncture (CLP), a clinically relevant model of sepsis characterized by profound insulin resistance (19).

MATERIALS AND METHODS

Animals

The PTP1B−/− mouse line was obtained from Dr Michel L. Tremblay (Goodman Cancer Centre, McGill University, Montreal, QC, Canada) and bred in our laboratory. PTP1B−/− mice and wild-type (WT) littermates were obtained by crossing Balb/c and PTP1B+/− mice. The genotype was determined by PCR of tail genomic DNA. All animals were housed in plastic cages maintained on a 12-h light/dark cycle in a controlled temperature (24 ± 2°C) and humidity (50 ± 5%), and allowed free access to standard mice chow and water. We used 8- to 12-week-old PTP1B−/− and WT male mice.

Model of cecal ligation and puncture (CLP)

Mice were slowly induced to anesthesia with 2% isoflurane and maintained at 1% isoflurane during surgical procedure. Mice were shaved and scrubbed with betadine. A middle abdominal incision was made. The caecum was mobilized, ligated with 6-O silk suture below the ileocecal valve, and punctured twice with a 21-gauge needle. A small amount of fecal material was extruded from the puncture site, the caecum was repositioned, and the abdomen was closed with 7-O silk suture for the peritoneum and 6-O silk suture for the skin. Following surgery, the mice were given buprenorphine (0.05 mg/kg, s.c.) for analgesia and received fluid resuscitation (30 mL/kg, s.c.) with 0.9% saline solution at 1, 5, and 9 h after surgery to correct hypovolemia (20). Sham-operated control animals underwent the same surgical protocol but the caecum was neither ligated nor punctured.

Study design

The following four experimental groups were studied: Control (Sham-operated), either WT or PTP1B−/− (n = 50 per group) and CLP-induced sepsis, either WT or PTP1B−/− (n = 60 per group). For all studies, mice were sacrificed 16 h (H-16) after induction of CLP, to collect blood, mesenteric arteries, and heart. In another set of experiments, vascular function, glucose tolerance, isolated perfused heart experimentation was evaluated at H-16.

Glucose and insulin tolerance

Mice were fasted 6 h before surgery, and conventional glucose and insulin tolerance tests were performed at H16. A blood sample was collected from the tail vein and blood glucose was measured with a glucometer (StatStrip Xpress, Nova Biomedical UK, Hercules, Calif) before glucose tolerance test, defined as T0. Glucose tolerance test was performed by intravenous injection of 2 g D-glucose/kg body weight and blood samples were collected from the tail vein and blood glucose was measured for 2 h. Insulin tolerance was then assessed by intravenous injection of human insulin Actrapid (0.5 u/kg). Serum insulin levels were measured using a mouse insulin enzyme-linked immunosorbent assay kit (Ultra Sensitive Insulin mouse Assay; Crystal Chem) in blood samples taken from orbital vein immediately before and then again 15 min into the glucose tolerance test.

An index of insulin resistance estimated by the homeostasis model assessment (HOMA) was calculated using the relationships between blood glucose and insulin levels according to the following formula: HOMA-IR = [fasting insulin (mUI/L) × fasting blood glucose (mmol/L)]/22.5. HOMA beta, indicator of pancreatic beta cells function was calculated according to the following formula: HOMA beta = fasting plasma insulin (μUI/mL)/fasting plasma glucose (mmol/L).

Vascular function

Mesenteric arteries, 2 mm in length and <300 μm in diameter, were cannulated at both ends and placed in a video-monitored perfusion system (Living Systems Instrumentation, Burlington, Vt), as previously described (15). First, contractility was assessed by concentration response curves to phenylephrine (Phe; 10−9 to 10−5 mol/L, Sigma) before and after incubation with the selective inducible nitric oxide (NO) synthase (iNOS) inhibitor 1400 W (10−5 mol/L; Sigma). Vascular dilatation was assessed under 1,400 W incubation in preconstricted vessels (Phe 10−5mol/L). Endothelium dependent response was assessed by flow mediated-dilatation (FMD) induced by stepwise increases in intraluminal flow (0 to 200 μl/min), or by the response to insulin (10−9–10−5M). To investigate the mechanisms involved in the endothelial effect of CLP and PTP1B deletion, particularly with regard to the role of endothelial NO Synthase (eNOS), all of the above vascular experiments were repeated in the presence of a non-selective NOS inhibitor (N-nitro-L-arginine [L-NNA], 10−4mol/L, Sigma-Aldrich, St. Louis, Mo).

Isolated perfused hearts

Freshly excised mouse hearts were placed into ice-cold Krebs–Henseleit (KH) buffer solution and immediately mounted onto a Langendorff apparatus. Hearts were perfused in a retrograde fashion via the aorta at a constant flow rate of 2 mL/min with aerated (95% O2, 5% CO2) KH bicarbonate buffer containing NaCl 120 mM, KCl 4.8 mM, KH2PO4 1.2 mM, MgSO4 1.2 mM, NaHCO3 25 mM, CaCl2 1.25 mM, glucose 11 mM (37 °C; pH 7.35–7.40). After 30 min of initial equilibration, hearts were perfused using either conventional KH buffer (glucose as substrate) or KH buffer supplemented with 1.2 mM palmitate bound to 3% BSA (glucose + palmitate as substrates) for 30 min. Cardiac contractile function was assessed using a metal hook inserted into the heart apex to control and record developed force and heart rate (17). Myocardial O2 uptake (MVO2) and cardiac efficiency were calculated using standard formulas.

GLUT-4 translocation

Subcellular membrane fractionation protocols

Mouse hearts were removed and immediately freeze-clamped in liquid nitrogen. Upon thawing, hearts were diced and incubated for 30 min in 1 mL of a high salt solution prior to homogenization (2 mmol/L NaCl, 5 mmol/L NaN3 and 20 mmol/L HEPES, pH 7.4) at 4°C (21). The tissue was recovered by centrifugation at 1,000 g for 5 min and the pellet homogenized with a Potter glass homogenizer in 1.7 mL of buffer A (1 mmol/L MgCl2, 250 mmol/L sucrose, 20 mmol/L HEPES, and 2 mmol/L EDTA, pH 7.4). The resulting homogenate was centrifuged at 1,000 g for 5 min. After, the pellet was homogenized in 1.3 mL of buffer A and then recombined with the previous supernatant. This solution was centrifuged at 100 g for 10 min (22). The pellet named P1 was resuspended in 300 μL of buffer A. The supernatant was centrifuged at 5,000 g for 10 min. The pellet named P2 was resuspended in 100 μL of buffer A. The supernatant was centrifuged at 20,000 g for 20 min. The pellet named P3 was resuspended in 100 μL of buffer A. The supernatant was centrifuged at 48,000 g for 30 min. The pellet named P4 was resuspended in 50 μL of buffer A. The supernatant was centrifuged at 250,000 g for 65 min. The pellet named P5 was resuspended in 30 μL of buffer A. All resuspended pellets including the supernatant named P6 were frozen in liquid nitrogen and were kept at −80°C. We decided to refer to P2 as plasma membrane fraction and to P5 as low-density membrane fraction.

Western blot analysis of subcellular membrane fraction

The amount of proteins loaded on the gel was verified by a Bradford assay and was in each case 30 μg. The subcellular membrane fraction was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis for 30 min at 150 V (Criterion TGX Stain Free Precast Gels, Bio-Rad Laboratories, Hercules, Calif) in migration solution (Tris/Glycine Buffer, Bio-Rad Laboratories). Then, proteins were transferred on membranes (Trans-Blot Turbo Nitrocellulose, Bio-Rad Laboratories) in transfer solution (Trans-Blot Turbo 5X Transfert Buffer, Bio-Rad Laboratories) for 10 min (Trans-Blot Turbo Transfert System, Bio-Rad Laboratories). After transfer, membranes were activated by a U.V exposition (Bio-Rad Laboratories). To avoid nonspecific link, membranes were placed in a blockade solution (milk, 3%) for 30 min at room temperature. Then, membranes were incubated with the primary antibodies: anti-GLUT-4 (Anti-glucose Transporter GLUT-4 antibody ab654, ABCAM, 1/2,000, 4°C, overnight). Proteins were visualized with the use of a Chemiluminescence kit (Clarity Western ECL Substrate, peroxide and luminol/enhancer solutions, Bio-Rad Laboratories) and revealed with imaging system (Bio-Rad Laboratories). Analyses were performed with Biorad software (Image Lab software, Bio-Rad Laboratories, Hercules, Calif).

Insulin signaling pathway

Mice were injected intravenously with human insulin Actrapid (0.5 u/kg) or similar volume of NaCl. Mouse hearts were removed 15 min after injection and all hearts were immediately frozen in liquid nitrogen. Hearts were homogenized using a precellys (Bertin Technologies). Then, the amount of proteins loaded on the gel was verified by a Bradford assay and was 20 μg for Akt analysis and 50 μg for AMPK analysis. The fraction was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis for 30 min at 200 V (Criterion TGX Stainfree gels, Biorad) in migration solution (Tris-Glycine-SDS solution, Euromedex). Then, proteins were transferred on PVDF membranes (Biorad) in transfer solution (Tris-Glycine solution from Euromedex supplemented with 20% ethanol) for 10 min (Trans-blot Turbo transfer system, BioRad). To avoid nonspecific link, membranes were placed in a blockade solution (BSA, 5%) for 60 min at room temperature. Then, membranes were incubated with the primary antibodies: anti-Akt (Cell Signaling 4691, 1/1,000), antiphosphoAkt (Cell Signaling 4060, 1/2,000), anti-AMPK (Cell Signaling 2532, 1/1,000), and antiphosphoAMPK (Cell Signaling 2535, 1/1,000). Proteins were visualized with the use of a Chemiluminescence kit (luminata classic or forte, Millipore) and revealed with the ChemiDoc XRS+ system, BioRad. Analyses were performed with Biorad software (Image Lab software, Bio-Rad Laboratories).

Polymerase chain reaction experiments

Total mRNA was extracted from left ventricular tissue with the Trizol reagent (Gibco life science) according to the manufacturer's instructions. The cardiac expression of interleukin-1β (IL1-β), interleukin-6 (IL-6), interleukin-10 (IL-10), tumor necrosis factor-α (TNF-α), intercellular adhesion molecule 1 (ICAM-1), vascular cell adhesion molecule 1 (VCAM-1), Gp91phox, and cluster of differentiation 45 (CD45) mRNA was assessed by quantitative real-time reverse transcription-polymerase chain reaction with a light cycler (Roche, Basel, Switzerland) using SYBR green I. The housekeeping gene used was Eef2. The primers were obtained from Sigma-Aldrich and had the following sequences:

IL1-β: Forward 5′-CAGGCAGGCAGTATCACTCA-3′; Reverse 5′-TGTCCTCATCCTGGAAGGTC-3′; IL-6: Forward 5′-CCGGAGAGGAGACTTCACAG-3′; Reverse 5′-TCCACGATTTCCCAGAGAAC-3′; IL-10: Forward 5′-CCAGGGAGATCCTTTGATGA-3′; Reverse 5′-AACTGGCCACAGTTTTCAGG-3′; TNF-α: Forward 5′-TAGCCAGGAGGGAGAACAGA-3′; Reverse 5′-TTTTCTGGAGGGAGATGTGG-3′; ICAM-1: Forward 5′-ACTGGCAGTGGTTCTCTGCT-3′; Reverse 5′-AAAGTAGGTGGGGAGGTGCT-3′; VCAM-1: Forward 5′-TCTTACCTGTGCGCTGTGAC-3′; Reverse 5′-ACCTCCACCTGGGTTCTCTT-3′; Gp91phox: Forward 5′-AAAGGTGGTCATCACCAAGG-3′; Reverse 5′-ACTGTCCCACCTCCATCTTG-3′; CD45: Forward 5′-TCGTGCCCAAACAAATTAACA-3′; Reverse 5′-ATCCCCAAATCTGTCTGCAC-3′; Eef2: Forward 5′-GCGAGGACAAAGACAAGGAG-3′; Reverse 5′-GGGATGGTAAGTGGATGGTG-3′

Statistical analysis

All results are given as mean ± SEM. Normality was verified using the Kolmogorov–Smirnov test. Fasting blood glucose level, HOMA index, HOMA beta (Fig. 1), and cardiac inflammation (Fig. 2) were compared by one-way ANOVA followed by Bonferroni post-test. Insulin concentrations, glucose tolerance test, area under curve (Fig. 1), GLUT-4 expression, PKB, and AMPK phosphorylation in heart (Fig. 3), cardiac contractile function (Fig. 2), vascular contractility, and endothelial function (Fig. 4) were analyzed by two-way ANOVA followed by Bonferroni post-test with comparison of each group with all the other groups. P < 0.05 was considered statistically significant.

F1-13
Fig. 1:
Insulin-resistance and hyperglycemia studies.Representation of fasting blood glucose level (A) n = 8 mice per group; ††† P < 0.001 vs. WT CLP; # P < 0.05 vs. KO Sham. Evolution of plasmatic insulinemia before or 15 min after glucose intravenous injection (B) n = 6 to 8 mice per group; * P < 0.05 vs. WT CLP 0 min; ## P < 0.01 vs. WT Sham 0 min; P < 0.05 vs. WT CLP 15 min. Representation of HOMA index (C) and HOMA beta (D) n = 8 to 13 mice per group (C) n = 8 mice per group (D) * P < 0.05 vs. WT Sham; # P < 0.05 vs. KO Sham. Evolution of glycemia during glucose tolerance tests expressed as absolute values at 16 h after CLP induction (E) (n = 10–13 mice per group) or as area under the curve (AUC) (F) (n = 8–13 mice per group) between Sham and CLP mice * P < 0.05 vs. WT Sham; P < 0.05 vs. WT CLP; †† P < 0.01 vs. WT CLP. CLP indicates cecal ligation and puncture; HOMA, homeostasis model assessment; WT, wild type.
F2-13
Fig. 2:
Evaluation of cardiac function in isolated perfused heart (n = 5–7 mice per group) and cardiac inflammation.Left ventricular developed force (A) and cardiac efficacy (B) are displayed in KH buffer (standard Krebs) and palmitate-supplemented KH buffer (Krebs palmitate) perfused hearts. MVO2 was calculated by using standard formula with specific coefficients of O2 solubility and O2 density at 37°C (MVO2 = coronary flow/g of heart tissue × (PO2 perfusate – PO2 effluent) × oxygen solubility, in which oxygen solubility was 24 μL/mL H2O at 760 mm Hg). LVDF × heart rate divided by MVO2 was calculated as cardiac efficiency. * P < 0.05 vs. Sham; P < 0.05 vs. palmitate. Heart mRNA expression of interleukin (IL)-1β, IL-6, IL-10, tumor necrosis factor (TNF)-α, intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), Gp91phox and cluster of differentiation (CD) 45 (reverse transcription polymerase chain reaction) (C–J) * P < 0.05; ** P < 0.01; *** P < 0.001 vs. WT Sham; P < 0.05; ††† P < 0.001 vs. WT CLP. IL indicates interleukin; TNF-α, tumor necrosis factor-α.
F3-13
Fig. 3:
GLUT-4 expression, Akt phosphorylation, and AMPK phosphorylation in heart (Western blot, n = 5 mice per group) without or with insulin.GLUT-4 expression in plasma membrane fraction (heart) (A) and GLUT-4 expression in low density membrane fraction (heart) (B). Phosphorylation of Akt in heart (C) and phosphorylation of AMP-activated protein kinase (AMPK) in heart (D) * P < 0.05 vs. without insulin; ** P < 0.01 vs. without insulin. AMPK indicates AMP-activated protein kinase; GLUT-4, glucose transporter 4.
F4-13
Fig. 4:
Vascular contractility and endothelial function (n = 6–7 mice per group).Concentration–response curve to phenylephrine (Phe) at baseline (A) and after incubation with the inducible nitric oxide synthase (iNOS) inhibitor 1400 W (10−5 mol/L) (B). Insulin dilatation in mesenteric arteries (in the presence of 1400 W 10−5 mol/L) before and after incubation with the NOS inhibitor Nω-nitro-L-arginine (L-NNA; 10−4 mol/L) (C, D). Flow-mediated dilatation (FMD) in mesenteric arteries (in the presence of 1400 W 10−5 mol/L) before and after incubation with the NOS inhibitor Nω-nitro-L-arginine (L-NNA; 10−4 mol/L) (E, F). Acetylcholine dilation in mesenteric arteries (in the presence of 1400 W 10−5 mol/L) (G). Sodium nitroprusside in mesenteric arteries (in the presence of 1400 W 10−5 mol/L) (H). * P < 0.05, ** P < 0.01, *** P < 0.001 vs. WT Sham; # P < 0.05, ### P < 0.001 vs. KO Sham. Additional file 1: SDS-PAGE pictures of AMPK and P-AMPK in heart. Additional file 2: SDS-PAGE pictures and analysis of stain free to control totals proteins. Additional file 3: eNOS expression and eNOS phosphorylation in mesenteric arteries (Western blot, n = 10 mice per group). eNOS expression in mesenteric arteries (A) and eNOS phosphorylation in mesenteric arteries (B). All data are statistically none significantly different. AMPK indicates AMP-activated protein kinase; CLP, cecal ligation and puncture; eNOS, endothelial nitric oxide synthase; HOMA, homeostasis model assessment; WT, wild type.

RESULTS

Metabolic effects: insulin resistance and glucose tolerance

PTP1B deletion did not modify fasting blood glucose levels in control (sham) mice. CLP did not significantly affect glucose levels in WT mice; however, PTP1B deletion markedly reduced these levels in CLP mice, compared with WT (Fig. 1A).

PTP1B deletion did not modify fasting plasma insulin levels in control (sham) mice. In WT, CLP significantly increased plasma insulin levels both before (time 0) and 15 min after glucose injection; however, these increases were significantly less marked in PTP1B−/− CLP mice (Fig. 1B). CLP also significantly increased HOMA index in WT, but not in PTP1B−/− mice (Fig. 1C). HOMA beta was increased by CLP to the same extent in WT and PTP1B−/− mice, suggesting that deletion of this enzyme did not affect the ability of beta pancreatic cells to secrete insulin (Fig. 1D).

In control (Sham) mice, glucose tolerance was moderately increased in PTP1B−/− mice compared with WT. In WT mice, glucose tolerance was considerably altered by CLP, and this was almost fully prevented by PTP1B deletion (Fig. 1, E and F).

GLUT-4 expression, insulin receptor signaling, and AMPK pathways in heart

In Sham mice, insulin increased GLUT-4 expression in plasma membrane fractions in WT but not in PTP1B−/− mice. This insulin-mediated GLUT-4 expression in plasma membrane was absent in WT CLP mice; however, this effect was partially restored in PTP1B−/− CLP mice (Fig. 3, A and B and supplemental digital content, https://links.lww.com/SHK/A561).

At baseline, the cardiac P-Akt/Akt ratio was similar in all groups. Fifteen minutes after insulin, this ratio increased significantly and similarly in all groups (Fig. 3C). Finally, PTP1B deletion increased the cardiac P-AMPK/AMPK ratio in Sham and CLP mice. Insulin did not modify this ratio in WT mice; however, it increased it in PTP1B−/− CLP mice (Fig. 3D and supplemental digital content, https://links.lww.com/SHK/A562).

Cardiac contractile function and metabolic response

In standard KH buffer perfused hearts, cardiac functional parameters were similar between WT Sham and PTP1B−/− Sham. CLP induced major reductions in systolic performance (LVDF) and cardiac efficiency in WT mice, which were not prevented in PTP1B−/− CLP when isolated hearts were perfused with standard KH buffer (Fig. 2, A and B). Compared with standard KH buffer, palmitate perfusion had no effects on cardiac efficiency in WT CLP mice, whereas palmitate perfusion further reduced cardiac efficiency in WT Sham, PTP1B−/− Sham, and PTP1B−/− CLP (Fig. 2B).

Cardiac inflammation

No significant difference in cardiac inflammation markers was observed between WT and PTP1B−/− Sham mice (Fig. 2, C–J). In WT mice, CLP significantly increased the cardiac expression of IL-6, IL-10, TNF-α, ICAM-1, and Gp91phox (Fig. 2, C–J). In CLP mice, PTP1B deletion attenuated cardiac inflammation as evaluated by TNF-α, IL-1β, and IL-6 levels, yet this was only statistically significant for TNF-α. In these CLP mice, PTP1B deletion attenuated cardiac adhesion and oxidative stress as evaluated by ICAM-1, VCAM-1, and Gp91phox levels, yet this was statistically significant only for VCAM-1 and Gp91phox (Fig. 2, C–J).

Arterial constriction and NO-mediated vasodilation

In mesenteric arteries, the constriction induced by phenylephrine was significantly increased in CLP mice at the concentrations 10−7 mol/L and 10−6 mol/L and this constriction was reduced by the selective iNOS inhibitor 1,400 W. This impaired contractile response was not significantly reversed in PTP1B−/− mice (Fig. 4, A and B).

In Sham mice (either WT of PTP1B−/−), insulin induced concentration-dependent dilatations that were abolished by the NOS inhibitor L-NNA suggesting that they were entirely due to NO in these arteries (Fig. 4, C and D). CLP abolished this NO-mediated dilation to insulin, however this alteration was absent in PTP1B−/− mice.

Compared with WT Sham, WT CLP mice displayed a complete abolition of FMD, demonstrating severe endothelial dysfunction. L-NNA abolished FMD in Sham arteries, suggesting that CLP abolished flow-induced NO-mediated dilatation. This CLP-induced alteration of FMD was absent in PTP1B−/− mice. In PTP1B−/− CLP mice, L-NNA abolished this response to flow, suggesting that the increased FMD induced by PTP1B deficiency is due to restored (endothelial) NO production (Fig. 4, E and F).

The dilatation induced by acetylcholine was significantly reduced in PTP1B−/− CLP mice at the concentrations 10−7mol/L and 3 × 10−7mol/L (Fig. 4G). No significant difference in the endothelium-independent dilation to the NO donor sodium nitroprusside was found between groups (Fig. 4H).

DISCUSSION

In a previous study, we demonstrated the beneficial effect of PTP1B gene deletion in experimental models of septic shock (i.e., LPS and CLP). Here, we demonstrate in CLP mice that this PTP1B deletion also limits septic shock-induced insulin resistance, possibly partly via its protective effect on endothelial dysfunction and on “vascular insulin resistance.” This modulation of insulin resistance may contribute to the beneficial effect of PTP1B blockade in septic shock, especially in terms of inflammation and cardiac metabolism.

Insulin sensitivity

Abnormal carbohydrate metabolism, as reflected by hyperglycemia and insulin resistance, is implicated in many adverse effects and in poor outcome in septic patients (23). PTP1B, which is increased in experimental sepsis, is an essential negative regulator of insulin signaling. Indeed, PTP1B gene deletion in mice increases insulin sensitivity and prevents the development of obesity (13–16). Whether this is the case in sepsis is unknown, thus we evaluated the effects of PTP1B gene deletion on hyperglycemia and insulin resistance in CLP septic mice. We confirmed that CLP induced insulin resistance, as shown by the blunted response to glucose injection despite increased insulin levels and the marked increases in HOMA index and HOMA beta. In this context, PTP1B deletion in CLP mice significantly improved these parameters, suggesting that this enzyme plays a critical role in the development of insulin resistance in sepsis.

We then attempted to explore the molecular mechanisms of the PTP1B-mediated regulation of insulin resistance in sepsis. Insulin-mediated glucose uptake requires increased GLUT-4 translocation mediated by PI3K/Akt pathway (24, 25). We found that CLP profoundly affected basal and insulin-mediated GLUT-4 expression in cardiac plasma membranes, as previously reported in the skeletal muscle of CLP mice (26), while PTP1B deletion restored both basal and insulin-mediated GLUT-4 expression. This probably markedly participates in the restored glucose uptake observed in PTP1B−/− CLP mice. Of note, in normal (non-CLP) mice, PTP1B deletion also reduced cardiac GLUT-4 expression, suggesting that this enzyme is essential for normal insulin signaling.

Insulin binding induces the recruitment of insulin receptor substrates (IRSs) which can activate several downstream pathways including the PI3K/Akt and MAPK pathways (27). We observed that insulin-mediated increase in Akt phosphorylation was affected neither by CLP nor by PTP1B deletion, suggesting that the observed changes in insulin sensitivity and GLUT-4 translocation are not related to changes in PI3K/Akt.

In addition to the PI3/Akt pathway, alternative mechanisms modulate GLUT-4 translocation, such as AMPK phosphorylation and inflammatory processes (28, 29). Indeed, PTP1B deletion increased insulin-induced cardiac AMPK phosphorylation in CLP. A similar stimulation of AMPK signaling after PTP1B deletion has been already reported in a high-fat diet model of insulin resistance (30). Although it is not clear how these AMPK modulating events are activated, global deletion of PTP1B increases AMPK α2 protein expression and activity without affecting AMPKα1 expression or activity in brown adipose tissues and skeletal muscles of mice with diet-induced obesity (31). Thus, the changes in GLUT-4 expression and insulin response observed in PTP1B−/− CLP mice may be mediated in part by stimulation of the AMPK signaling. In parallel, inflammatory cytokines produced by the adipose tissue, such as TNF-α and IL-6, are able to trigger insulin resistance, by altering insulin signaling, e.g., via reduced IR or IRS-1 phosphorylation (9), and by reducing GLUT-4 expression (32). Indeed, we observed that PTP1B deletion limited the CLP-induced increases in cardiac expression of TNF-α and IL-1β mRNA, as well as that of the adhesion molecule VCAM-1 and of the Gp91phox subunit of the oxygen-radical producing enzymes NADPH oxidases. Of note, previous studies showed that AMPK is a key target of this inflammatory response; indeed, AMPK inhibits the production of proinflammatory cytokines via modulation of the NF-κB pathway, while LPS reduces AMPK activity (33).

Altogether, we suggest that PTP1B deletion in the CLP-induced sepsis model activates AMPK and reduces inflammatory processes, leading to increased GLUT-4 translocation and glucose uptake, and finally to reduced insulin resistance.

Heart metabolism

In isolated perfused hearts or in vivo, insulin-mediated increases in glucose utilization and oxidation are associated with reduced fatty acid oxidation on the basis of the Randle phenomenon (34). As a consequence, free fatty acid (palmitate) perfusion in control hearts is associated with blunted glucose utilization and oxidation with a parallel increase in myocardial oxygen uptake (MVO2) that reduces cardiac efficiency. Although glucose uptake measurements in cardiac tissue were not performed in our study, isolated heart studies showing that palmitate perfusion reduced cardiac efficiency in PTP1B−/− CLP mice but not in WT CLP mice suggested that flexibility in substrate utilization was restored by PTP1B deletion. Despite this, no changes in cardiac intrinsic performance in PTP1B−/− CLP mice were observed. Overall, these observations are consistent with known inhibitory effects of palmitate perfusion on cardiac glucose uptake and MVO2, which are only observed if substrate utilization flexibility exists (35).

Endothelial dysfunction and “vascular insulin resistance”

We reported previously that PTP1B deletion improved endothelial dysfunction (especially flow-dependent, NO-mediated vasodilation of isolated perfused mesenteric arteries) after LPS (17). Here, we extend this observation and show that this endothelial protective effect of PTP1B deletion is also observed in the CLP model of sepsis. Interestingly, comparison of these two studies revealed that the severity of endothelial dysfunction was more marked after CLP than after LPS (i.e., complete abolition of flow-mediated dilatation after CLP vs. partial reduction after LPS). In this context, the fact that PTP1B deletion completely reversed endothelial dysfunction in CLP mice in the present study reinforces the concept that blockade of this enzyme induces powerful endothelial protection in septic shock, independently of the experimental model used. This observation is important since loss of endothelial function in sepsis is known to promote impaired local regulation of vascular responsiveness and to trigger inflammation and thus is a key factor in mortality in experimental and human sepsis.

Importantly, we also report here that arteries isolated from CLP mice display a complete abolition of the NO-mediated vasodilatory response to insulin, thus demonstrating a potent sepsis-induced “vascular insulin resistance.” There is ample evidence that insulin-mediated NO production and the resulting vasodilatory response is essential for its effects on glucose uptake and that impairment of this endothelial response strongly contributes to insulin resistance (36–39) (supplemental digital content, https://links.lww.com/SHK/A563). Thus, the restoration of insulin-mediated vasodilation induced by PTP1B deletion that we observed in CLP mice most likely contributes to the overall reduction of insulin resistance observed in this model.

CONCLUSION

Our study reveals for the first time that PTP1B deletion prevents insulin resistance associated with sepsis, an effect that may be attributed partly to decreased production of pro-inflammatory cytokines and preservation of AMPK signaling and GLUT-4 translocation. These beneficial effects may be due in part to the endothelial protective effects associated with PTP1B deficiency. Given the known importance of insulin resistance in the development and aggravation of septic shock, it is likely that these observed metabolic effects of PTP1B deletion contribute to the overall cardiovascular protection induced by PTP1B blockade. Overall, our results reinforce the hypothesis that PTP1B is an attractive target for the treatment of sepsis, and this hypothesis now deserved to be tested in septic patients.

REFERENCES

1. Angus DC, Pires Pereira CA, Silva E. Epidemiology of severe sepsis around the world. Endocr Metab Immune Disord Drug Targets 2006; 6 2:207–212.
2. Li L, Messina JL. Acute insulin resistance following injury. Trends Endocrinol Metab 2009; 20 9:429–435.
3. El-Osta A, Brasacchio D, Yao D, Pocai A, Jones PL, Roeder RG, Cooper ME, Brownlee M. Transient high glucose causes persistent epigenetic changes and altered gene expression during subsequent normoglycemia. J Exp Med 2008; 205 10:2409–2417.
4. Vlasselaers D, Mesotten D, Langouche L, Vanhorebeek I, van den Heuvel I, Milants I, Wouters P, Wouters P, Meyns B, Bjerre M, et al. Tight glycemic control protects the myocardium and reduces inflammation in neonatal heart surgery. Ann Thorac Surg 2010; 90 1:22–29.
5. van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, Vlasselaers D, Ferdinande P, Lauwers P, Bouillon R. Intensive insulin therapy in critically ill patients. N Engl J Med 2001; 345 19:1359–1367.
6. Jeschke MG, Klein D, Herndon DN. Insulin treatment improves the systemic inflammatory reaction to severe trauma. Ann Surg 2004; 239 4:553–560.
7. Hagiwara S, Iwasaka H, Hasegawa A, Asai N, Noguchi T. Hyperglycemia contributes to cardiac dysfunction in a lipopolysaccharide-induced systemic inflammation model. Crit Care Med 2009; 37 7:2223–2227.
8. Savage DB, Petersen KF, Shulman GI. Mechanisms of insulin resistance in humans and possible links with inflammation. Hypertension 2005; 45 5:828–833.
9. Shepherd PR, Kahn BB. Glucose transporters and insulin action—implications for insulin resistance and diabetes mellitus. N Engl J Med 1999; 341 4:248–257.
10. Thompson LH, Kim HT, Ma Y, Kokorina NA, Messina JL. Acute muscle-type specific insulin resistance following injury. Mol Med 2008; 14 (11–12):715–723.
11. Orban JC, Deroche D, Ichai C. Septic shock: blood glucose regulation. Ann Fr Anesth Reanim 2006; 25 3:275–279.
12. Gamelli RL, Liu H, He LK, Hofmann CA. Alterations of glucose transporter mRNA and protein levels in brain following thermal injury and sepsis in mice. Shock 1994; 1 6:395–400.
13. Elchebly M, Payette P, Michaliszyn E, Cromlish W, Collins S, Loy AL, Normandin D, Cheng A, Himms-Hagen J, Chan CC, et al. Increased insulin sensitivity and obesity resistance in mice lacking the protein tyrosine phosphatase-1B gene. Science 1999; 283 5407:1544–1548.
14. Vercauteren M, Remy E, Devaux C, Dautreaux B, Henry JP, Bauer F, Mulder P, Hooft van Huijsduijnen R, Bombrun A, Thuillez C, et al. Improvement of peripheral endothelial dysfunction by protein tyrosine phosphatase inhibitors in heart failure. Circulation 2006; 114 23:2498–2507.
15. Maupoint J, Besnier M, Gomez E, Bouhzam N, Henry JP, Boyer O, Nicol L, Mulder P, Martinet J, Richard V. Selective vascular endothelial protection reduces cardiac dysfunction in chronic heart failure. Circ Heart Fail 2016; 9 4:e002895.
16. Thiebaut PA, Besnier M, Gomez E, Richard V. Role of protein tyrosine phosphatase 1B in cardiovascular diseases. J Mol Cell Cardiol 2016; 101:50–57.
17. Coquerel D, Neviere R, Delile E, Mulder P, Marechal X, Montaigne D, Renet S, Remy-Jouet I, Gomez E, Henry JP, et al. Gene deletion of protein tyrosine phosphatase 1B protects against sepsis-induced cardiovascular dysfunction and mortality. Arterioscler Thromb Vasc Biol 2014; 34 5:1032–1044.
18. Raetzsch CF, Brooks NL, Alderman JM, Moore KS, Hosick PA, Klebanov S, Akira S, Bear JE, Baldwin AS, Mackman N, et al. Lipopolysaccharide inhibition of glucose production through the Toll-like receptor-4, myeloid differentiation factor 88, and nuclear factor kappa b pathway. Hepatology 2009; 50 2:592–600.
19. Heuer JG, Bailey DL, Sharma GR, Zhang T, Ding C, Ford A, Stephens EJ, Holmes KC, Grubbs RL, Fynboe KA, et al. Cecal ligation and puncture with total parenteral nutrition: a clinically relevant model of the metabolic, hormonal, and inflammatory dysfunction associated with critical illness. J Surg Res 2004; 121 2:178–186.
20. Dejager L, Pinheiro I, Dejonckheere E, Libert C. Cecal ligation and puncture: the gold standard model for polymicrobial sepsis? Trends Microbiol 2011; 19 4:198–208.
21. Fuller W, Eaton P, Medina RA, Bell J, Shattock MJ. Differential centrifugation separates cardiac sarcolemmal and endosomal membranes from Langendorff-perfused rat hearts. Anal Biochem 2001; 293 2:216–223.
22. Luiken JJ, Koonen DP, Willems J, Zorzano A, Becker C, Fischer Y, Tandon NN, Van Der Vusse GJ, Bonen A, Glatz JF. Insulin stimulates long-chain fatty acid utilization by rat cardiac myocytes through cellular redistribution of FAT/CD36. Diabetes 2002; 51 10:3113–3119.
23. Brierre S, Kumari R, Deboisblanc BP. The endocrine system during sepsis. Am J Med Sci 2004; 328 4:238–247.
24. Watson RT, Pessin JE. GLUT-4 translocation: the last 200 nanometers. Cell Signal 2007; 19 11:2209–2217.
25. Bae SS, Cho H, Mu J, Birnbaum MJ. Isoform-specific regulation of insulin-dependent glucose uptake by Akt/protein kinase B. J Biol Chem 2003; 278 49:49530–49536.
26. Matsuda N, Yamamoto S, Yokoo H, Tobe K, Hattori Y. Nuclear factor-kappaB decoy oligodeoxynucleotides ameliorate impaired glucose tolerance and insulin resistance in mice with cecal ligation and puncture-induced sepsis. Crit Care Med 2009; 37 10:2791–2799.
27. Engelman JA, Luo J, Cantley LC. The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet 2006; 7 8:606–619.
28. Sriwijitkamol A, Coletta DK, Wajcberg E, Balbontin GB, Reyna SM, Barrientes J, Eagan PA, Jenkinson CP, Cersosimo E, DeFronzo RA, et al. Effect of acute exercise on AMPK signaling in skeletal muscle of subjects with type 2 diabetes: a time-course and dose-response study. Diabetes 2007; 56 3:836–848.
29. Richter EA, Hargreaves M. Exercise, GLUT-4, and skeletal muscle glucose uptake. Physiol Rev 2013; 93 3:993–1017.
30. Kandadi MR, Panzhinskiy E, Roe ND, Nair S, Hu D, Sun A. Deletion of protein tyrosine phosphatase 1B rescues against myocardial anomalies in high fat diet-induced obesity: role of AMPK-dependent autophagy. Biochim Biophys Acta 2015; 1852 2:299–309.
31. Xue B, Pulinilkunnil T, Murano I, Bence KK, He H, Minokoshi Y, Asakura K, Lee A, Haj F, Furukawa N, et al. Neuronal protein tyrosine phosphatase 1B deficiency results in inhibition of hypothalamic AMPK and isoform-specific activation of AMPK in peripheral tissues. Mol Cell Biol 2009; 29 16:4563–4573.
32. Rotter V, Nagaev I, Smith U. Interleukin-6 (IL-6) induces insulin resistance in 3T3-L1 adipocytes and is, like IL-8 and tumor necrosis factor-alpha, overexpressed in human fat cells from insulin-resistant subjects. J Biol Chem 2003; 278 46:45777–45784.
33. Su RY, Chao Y, Chen TY, Huang DY, Lin WW. 5-Aminoimidazole-4-carboxamide riboside sensitizes TRAIL- and TNFα-induced cytotoxicity in colon cancer cells through AMP-activated protein kinase signaling. Mol Cancer Ther 2007; 6 5:1562–1571.
34. Randle PJ, Garland PB, Hales CN, Newsholme EA. The glucose fatty-acid cycle. Its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 1963; 1 7285:785–789.
35. Schrauwen P, Timmers S, Hesselink MK. Blocking the entrance to open the gate. Diabetes 2013; 62 3:703–705.
36. Steinberg HO, Chaker H, Leaming R, Johnson A, Brechtel G, Baron AD. Obesity/insulin resistance is associated with endothelial dysfunction. Implications for the syndrome of insulin resistance. J Clin Invest 1996; 97 11:2601–2610.
37. Roy D, Perreault M, Marette A. Insulin stimulation of glucose uptake in skeletal muscles and adipose tissues in vivo is NO dependent. Am J Physiol 1998; 274 (4 pt 1):E692–E699.
38. Duplain H, Burcelin R, Sartori C, Cook S, Egli M, Lepori M, Vollenweider P, Pedrazzini T, Nicod P, Thorens B, et al. Insulin resistance, hyperlipidemia, and hypertension in mice lacking endothelial nitric oxide synthase. Circulation 2001; 104 3:342–345.
39. Cook S, Hugli O, Egli M, Ménard B, Thalmann S, Sartori C, Perrin C, Nicod P, Thorens B, Vollenweider P, et al. Partial gene deletion of endothelial nitric oxide synthase predisposes to exaggerated high-fat diet-induced insulin resistance and arterial hypertension. Diabetes 2004; 53 8:2067–2072.
Keywords:

Glucose; hyperglycemia; insulin resistance; PTP1B; severe sepsis; vascular dysfunction; Akt; kinase protein B; AMPK; adenosine monophosphate-activated protein kinase; CD45; cluster of differenciation 45; CLP; cecal ligation and puncture; DNA; desoxyribo nucleotide acid; eNOS; endothelial nitric oxide synthase; FMD; flow-mediated dilatation; GLUT-1; glucose transporter 1; GLUT-4; glucose transporter 4; GLUTs; glucose transporters; HOMA; homeostasis model assessment; ICAM-1; intercellular adhesion molecule 1; ICU; intensive care unit; IL-1β; interleukin-1β; IL-10; interleukin-10; IL-6; interleukin-6; iNOS; inducible nitric oxide synthase; IR; insulin receptors; IRS-1; insulin receptor substrate 1; IRSs; insulin receptor substrates; KH; Krebs–Henseleit; L-NNA; Nω-nitro-L-arginine; LPS; lipopolysaccharide; mRNA; messenger ribonucleic acid; MVO2; myocardial oxygen uptake; NF-κB; nuclear factor-kappa B; NO; nitric oxide; PCR; polymerase chain reaction; Phe; phenylephrine; PI3K; phospho-inositol-3-kinase; PTP1B; protein tyrosine phosphatase 1B; TLR4; toll-like receptor 4; TNF-α; tumor necrosis factor-α; VCAM1; vascular cell adhesion molecule 1; WT; wild type

Supplemental Digital Content

© 2017 by the Shock Society