J Mol Med DOI 10.1007/s00109-015-1318-7

ORIGINAL ARTICLE

Inhibition of TLR4 attenuates vascular dysfunction and oxidative stress in diabetic rats Maria Alicia Carrillo-Sepulveda 1,5 & Kathryn Spitler 2 & Deepesh Pandey 3 & Dan E. Berkowitz 3 & Takayuki Matsumoto 4

Received: 12 February 2015 / Revised: 24 June 2015 / Accepted: 26 June 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract Hyperglycemia-induced reactive oxygen species (ROS) production plays a major role in the pathogenesis of diabetic vascular dysfunction. However, the underlying mechanisms remain unclear. Toll-like receptor 4 (TLR4), a key component of innate immunity, is known to be activated during diabetes. Therefore, we hypothesize that hyperglycemia activates TLR4 signaling in vascular smooth muscle cells (VSMCs) that triggers ROS production and causes vascular dysfunction. Rat mesenteric VSMCs exposed to high glucose (25 mmol/l) increased TLR4 expression and activated TLR4 signaling via upregulation of myeloid differentiation factor 88 (MyD88). TLR4 inhibitor CLI-095 significantly attenuated elevated levels of ROS and nuclear factor-kappa B (NF-κB) activity in VSMCs exposed to high glucose. Mesenteric arteries from streptozotocin-induced diabetic rats treated with CLI-095 (2 mg/kg/day) intraperitoneally for 2 weeks exhibited reduced ROS generation and attenuated noradrenaline-induced contraction. These results suggest that hyperglycemia-induced

* Maria Alicia Carrillo-Sepulveda [email protected] 1

Department of Physiology, Georgia Regents University, Augusta, GA, USA

2

Department of Internal Medicine, University of Iowa, Iowa City, IA, USA

3

Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA

4

Department of Physiology and Morphology, Institute of Medicinal Chemistry, Hoshi University, Shinagawa-ku, Tokyo 142-8501, Japan

5

Department of Biomedical Sciences, College of Osteopathic Medicine, New York Institute of Technology, Northern Blvd., Old Westbury, NY 11568, USA

ROS generation and NF-κB activation in VSMCs are at least, in part, mediated by TLR4 signaling. Therefore, strategies to block TLR4 signaling pathways pose a promising avenue to alleviate diabetic-induced vascular complications. Key messages & High glucose-induced TLR4 activation in vascular smooth muscle cells. & Inhibition of TLR4 attenuated high glucose-induced ROS production and NF-κB activity in VSMC. & Suppression of TLR4 signaling attenuated mesenteric contraction in diabetic rat. Keywords Diabetes . High glucose . NF-κB . ROS . TLR4 . Vascular smooth muscle cell

Introduction Vascular complications often arise as an important contributing factor in diabetes-associated pathologies, often leading to early disability and rising mortality and morbidity in diabetic patients worldwide [1–4]. Hyperglycemia, characterized by high glucose (HG) levels, is a hallmark of both type 1 and type 2 diabetes. Hyperglycemia-induced alterations in vascular smooth muscle cells (VSMCs) play a major role in the development of diabetic vascular complications [5, 6]. Nuclear factor-κB (NF-κB) is a family of transcription factors expressed in arterial VSMCs which is activated by HG to alter cellular signaling in VSMC that affects its function. NF-κB is also known as a major reactive oxygen species (ROS)-sensitive nuclear factor [7] and is a downstream component of TLR4 signaling [8]. It has been shown that HG promotes exaggerated NF-κB activation, which in turn results in inflammation, proliferation, and migration of VSMCs [9, 10].

J Mol Med

Although hyperglycemia plays a primary and causative role in the development of vascular dysfunction, other factors such as increased ROS, cytokines, and endocrine factors also contribute to dysfunction in both the micro- and macro-vasculature during diabetes [1, 3, 4]. Therefore, a better understanding of the link between hyperglycemia and these other factors will unravel new therapeutic targets for the prevention and treatment of diabetic-associated vascular complications. Toll-like receptor 4 (TLR4) is a key component of innate immune responses [11, 12]. There is a growing body of evidence implicating TLR4 in cardiovascular diseases including hypertension and diabetes [13–20]. Expression of TLR4 were found to be significantly upregulated in mesenteric arteries of Spontaneously Hypertensive Rats (SHR) and neutralizing antibody blocking TLR reduced circulating interleukin-6 (IL-6) levels, contractile responses to noradrenaline in mesenteric arteries, and blood pressure in these rats [13]. Further, deletion of TLR4 prevented NG-nitro-L-arginine methyl ester (LNAME)-induced hypertension, release of damage associated molecular patterns (DAMPs), ROS production, and impairment of the nitric oxide (NO)-soluble guanylate cyclase (sGC)-cyclic GMP (cGMP) signaling pathway [21]. Moreover, it has recently been suggested that TLR4 signaling contributes to the pathophysiology of diabetes. Deletion of TLR4 has been shown to (1) prevent high fat dietinduced insulin resistance and vascular inflammation [22], (2) attenuate circulating pro-inflammatory cytokines in streptozotocin (STZ)-induced diabetic mice [23], and (3) protect arteries from endothelial dysfunction in type 2 diabetic mice [24]. Inhibition of MyD88 (myeloid differentiation factor 88), a downstream molecule of the TLR4 signaling pathway, has also been shown to decrease ROS-induced diabetic retinal complications [25]. However, the knowledge regarding direct relationship between hyperglycemia, ROS generation, and TLR4 activation in the vasculature during diabetes is scarce. The present study aims to identify the cellular signaling mechanisms linking HG-mediated TLR4 signaling leading to downstream ROS generation, and NF-κB activity in rat mesenteric VSMCs.

Materials and methods Animals Twelve-week-old male Sprague–Dawley rats (Harlan) were maintained on a 12-h light–dark cycle with access to standard rodent chow and water ad libitum. All experimental protocols were approved by the Institutional Animal care and Use Committee of Georgia Regents University and conducted by the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Streptozotocin-induced hyperglycemic rats treated with TLR4 inhibitor Diabetes was induced by intraperitoneal (i.p.) injection of STZ at a dose of 65 mg/kg dissolved in 0.01 mol/l sodium citrate, pH 4.5, as described previously [26]. The control group received vehicle injections of sodium citrate. The presence of diabetes was confirmed 48 h after injection of STZ by measuring blood glucose levels using a One Touch Ultra Glucometer (Monitoring System Abbott). Fasting blood glucose of >250 mg/dl was considered to be diabetic. To examine whether TLR4 plays a role in the vasculature, diabetic and vehicle-treated rats were treated by a daily i.p. injection of 2 mg/kg/day of CLI-095, a TLR4 signaling inhibitor (Invivogen, San Diego, CA, USA) [27, 28], for the last 14 days of the treatment protocol. This optimal dose was selected based on previous studies [29]. Rats were divided into four experimental groups: (1) control, (2) control treated with TLR4 inhibitor, (3) diabetic, and (4) diabetic treated with TLR4 inhibitor. Primary VSMC culture VSMCs were isolated from rat second- and third-order mesenteric arteries by an enzymatic procedure, as described previously [30]. Briefly, mesenteric arteries were dissected and perivascular fat was removed. Clean arteries were cut into small pieces and incubated with collagenase, elastase, and soybean trypsin inhibitor at 37 °C for 40 min. The suspension was then centrifuged at 2000 rpm for 5 min. Supernatant was discarded and the pellet containing cells was re-suspended in Dulbecco’s Modified Eagle’s Medium (DMEM) with 10 % fetal bovine serum (FBS) and antibiotics and plated in a humidified incubator set at 37 °C, 5 % CO2, and atmospheric O2. VSMCs exhibited the typical Bhill and valley^ growth morphology and were confirmed positive (>95 %) for smooth muscle α-actin and calponin by immunostaining. Cells at passage 3 were used in all experiments. After 80 % confluence, VSMCs were serum starved for 24 h to reach a quiescent state, followed by treatment with HG (25 mmol/l) for 12, 24, 48, and 72 h. As a control, VSMCs were maintained in normal glucose (NG; 5.5 mmol/l) and mannitol (25 mmol/l for osmotic control). Human aortic smooth muscle cells (Gibco, Eugene, OR, USA) were exclusively used to assay NF-κB activity. In some experiments, cells were pretreated with 3 μmol/l of CLI-095, a TLR4 inhibitor, and with 50 μmol/l pyrrolidine dithiocarbamate (PDTC), a potent NF-κB inhibitor. Western blot Briefly, proteins were extracted from mesenteric arteries from each experimental rat group as well as VSMC cultures treated with or without HG and CLI-095 (3 μmol/l) using RIPA lysis

J Mol Med

buffer. Ten micrograms of cellular protein was resolved by 10 % sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to PVDF membranes (Thermo Scientific), as previously described [31, 32], and probed with specific antibodies [TLR4 (Abcam # 22048), MyD88 (Cell Signaling #4283), Histone H3 (Cell Signaling #9715), GAPDH (Sigma #G9295), cyclooxygenase-2 (COX-2) (BD Transduction Laboratories #610203), and Na+/K+ ATPase α1 (Cell Signaling #3010)]. The stripped membranes were later probed with β-actin antibody as a loading control. Protein expression was detected using chemiluminescence (ECL Plus, Amersham Biosciences). Data are presented as a percent (%) of control after normalized to β-actin. Cellular fractionation for TLR4 localization To determine whether TLR4 is expressed in the different cellular compartments of VSMCs, cells were suspended in hypotonic extraction buffer for 10 min on ice and lysed. Nuclear fractions were saved and utilized to determine NF-κB activation under HG conditions. The remaining lysates were separated into cytosolic, nuclear, and membrane fractions by centrifugation at 720×g for 2 min. Pellets containing nuclei were washed and then re-suspended in the nuclear buffer. Supernatant containing cytosol and membranes were centrifuged at 10,000×g and pellets discarded. The supernatant was again centrifuged at 100,000×g for 1 h and then the cytosol (supernatant) and membrane (pellet) fraction obtained. GAPDH, Histone H3, and Na+/K+ ATPase α1 were used as a marker for the cytosolic, nuclear, and membrane fraction, respectively. Fifty micrograms of protein from each fraction were loaded onto an SDS-PAGE gel and separated by electrophoresis, and proteins were detected by immunoblotting. Immunoprecipitation VSMC treated with HG and mesenteric arteries from diabetic rats were lysed in RIPA lysis buffer for immunoprecipitation (IP). Lysates were centrifuged at 15,000×g for 20 min at 4 °C. Supernatants (500 μg) were incubated with 10 μg of TLR4 antibody with gentle rotation at 4 °C overnight. Then, protein G Dynabeads (Life Technologies) were added for 4 h at 4 °C. Supernatant was removed and the remained TLR4-bound beads were washed three times with IP buffer and protein was eluted by heating at 70 °C for 15 min in Laemmli sample buffer containing 200 mmol/l dithiothreitol (DTT). Immunoprecipitates were further analyzed by western blotting and immunoblotted with MyD88 antibody. Measurement of ROS Superoxide production was detected by 25 μmol/l 2hydroxyethidium (DHE, Invitrogen) and 10 μmol/l 2′,7′-

dichlorodihydrofluorescein diacetate (H 2 DCF-DA, Invitrogen) using methods described previously [32, 33]. Briefly, quiescent VSMCs were pre-incubated with CLI-095 (3 μmol/l) for 30 min followed by HG treatment for 12 h. Subsequently, 25 μmol/l DHE or 10 μmol/l H2DCF-DA was added into the media and incubated at 37 °C for 20 min. After incubation, VSMCs were washed with PBS and fluorescence emitted by the DHE or H2DCF-DA was detected using an Axiovert200 fluorescence microscope fitted with a camera. In situ production of ROS from mesenteric arteries from the four groups was determined using 10 μmol/l H2DCF-DA. Mesenteric segments were embedded in Tissue-Tek OCT compound (Sakura Finetek, Torrance, CA, USA), snap frozen, and stored at −80 °C. Cryosections of 5 μm thickness were cut and placed on Superfrost Plus slides (Menzel-Gläzer, Germany). Sections were dried and stored at −80 °C. H2DCFDA was first diluted in dimethylsulfoxide (DMSO) as a 20 mmol/l solution and then diluted to 2 μmol/l in water before use. Fifty microliters of hydroethidine solution was topically applied to each tissue section and the slide coverslipped. The slides were incubated in a light-protected and humidified chamber (37 °C, 30 min) before visualizing. Semi-quantitative analyses were performed to detect changes in fluorescence in VSMCs or mesenteric arteries using Image Pro Plus software. NF-κB activity assay VSMCs were co-transduced with adenoviruses encoding NF-κB luciferase reporter (30 MOI) and GFP cDNA (30 MOI). Twenty-four hours post-transduction, cells were subjected to overnight serum starvation. The following day, cells were lysed, and luciferase activity (Promega) was determined by measuring luminescence using FlexStation 3 microplate reader (Molecular Devices). GFP fluorescence was used as a normalization control. Vascular reactivity After euthanasia with isoflurane (via nasal 5 % in 100 % O2), mesenteric arteries (second order) were removed and carefully cleared of perivascular fat in oxygenated Krebs buffer (130 mmol/l NaCl, 14.9 mmol/l NaHCO3, 4.7 mmol/l KCl, 1.18 mmol/l KH 2 PO 4 , 1.17 mmol/l MgSO 4 ·7H 2 O, 1.56 mmol/l CaCl2·2H2O, 0.026 mmol/l EDTA, 5.5 mmol/l glucose, pH 7.4). Mesenteric arteries were then cut into rings (2 mm in length), mounted in a wire myograph for isometric tension recordings by a PowerLab 8/SP data acquisition system (ADInstruments Pty Ltd., Castle Hill, Australia), equilibrated in Krebs buffer for 30 min, and gassed with 5 % CO2 in 95 % O2 at 37 °C, as described previously [31]. In all experiments, structural preservation of endothelium was confirmed by determining endothelial dependent relaxation to

J Mol Med

acetylcholine (ACh) in phenylephrine (PE) pre-contracting mesenteric ring preparations. Mesenteric ring integrity was assessed by stimulation with 120 mmol/l KCl. Concentration-response curves to noradrenaline (10 nmol/l– 30 μmol/l) were performed. Statistics Results are expressed as mean±SEM and analyzed with oneway ANOVA followed by Bonferroni’s post hoc test. Student’s t test was used when appropriate. Contractions were recorded as changes in displacement (mN) from baseline. Concentration-response curves were log-transformed, normalized to percent maximal response, and fitted using a nonlinear interactive fitting program (Graph Pad Prism 4.0; GraphPad Software Inc.). P values less than 0.05 (p