2016 Volume 39 Issue 9 Pages 1475-1481
In cellular events, endoplasmic reticulum (ER) stress has an important role in the development of various diseases including cardiovascular diseases. Tunicamycin, an inhibitor of N-linked glycosylation, is known to be an inducer of ER stress. However, the extent to which tunicamycin affects the vasorelaxant function is not completely understood. Thus, we investigated the effect of tunicamycin on relaxations induced by various vasorelaxant agents, including acetylcholine (ACh; endothelium-dependent vasodilator), sodium nitroprusside (SNP; endothelium-independent vasodilator), isoprenaline (ISO; beta-adrenoceptor agonist), forskolin (FSK; adenylyl cyclase activator), and cromakalim [ATP-sensitive K+ (KATP) channel activator] in organ-cultured superior mesenteric arteries of rats, which are treated with either a vehicle [dimethyl sulfoxide (DMSO)] or tunicamycin (20 µg/mL for 22–24 h). Protein levels of the ER stress marker binding immunoglobulin protein (BiP) were determined by Western blotting. Tunicamycin increased the expression of BiP in organ-cultured arteries. Tunicamycin impaired ACh-induced relaxation, but did not alter SNP-induced relaxation. Tunicamycin also impaired vasorelaxation induced by ISO, FSK, and cromakalim; moreover, it reduced basal nitric oxide (NO) formation. In conclusion, short-term treatment with tunicamycin not only caused endothelial dysfunction but also impaired cAMP- and KATP-mediated responses in the superior mesenteric arteries of rats. These alterations in tunicamycin-treated arteries may be due to reduced basal NO formation. This work provides new insight into ER stress in vascular dysfunction.
Organelle dysfunction, including endoplasmic reticulum (ER) stress, has an important role in the development of various diseases, such as cancer, and neuronal, metabolic, and cardiovascular diseases.1–10) ER stress is caused by various conditions, including glucose deprivation or glucose overload, calcium depletion or overload, hypoxia or oxidative stress, fatty acid overload, protein aggregation, and exposure to chemicals, such as tunicamycin or thapsigargin.10,11) Recent evidence has shown a relationship between ER stress and vascular function in various disease states.8,11,12) For instance, inhibition of ER stress could decrease blood pressure, aortic apoptosis, and fibrosis, and normalize vascular dysfunction in the aorta and mesenteric resistance artery of angiotensin II (Ang II)-treated hypertensive animals.13,14) Spitler et al. found that increased ER stress was present in the aorta of spontaneously hypertensive rats (SHR) and that the suppression of ER stress reduced acetylcholine (ACh)-induced endothelium-dependent contraction in the aorta of SHR.15) Moreover, we recently suggested that a relationship existed between ER stress and renal arterial function in type 2 diabetic rats.16) Therefore, ER stress is a key intracellular event in the pathogenesis of vascular dysfunction and is an important target against such vasculopathies. However, the direct linkage between ER stress and vascular function especially vasorelaxation, remains unclear.
Tunicamycin, an antibiotic, is known to inhibit N-linked glycosylation.17,18) Moreover, tunicamycin is also a major ER stress inducer in vitro and in vivo.13,14,19–21) In vivo treatment with tunicamycin for several weeks reduced ACh-induced endothelium-dependent relaxation in large and small arteries.13,14) Moreover, in vitro acute treatment with tunicamycin (for 1 h) impaired insulin-mediated vasomotor activity, such as by causing contraction but no relaxation in the aorta through endothelin-1 (ET-1) release.22) However, there are no reports regarding the in vitro effect of short-term treatment with tunicamycin on relaxations induced by various types of mechanisms in the superior mesenteric artery of rats.
The aim of this study was to investigate the effect of short-term tunicamycin treatment on vasorelaxant responses induced by various vasodilators, including an endothelium-dependent dilator (ACh), endothelium-independent dilator [nitric oxide (NO) donor, sodium nitroprusside (SNP)], beta-adrenoceptor agonist [isoprenaline (ISO)], adenylyl cyclase (AC) activator [forskolin (FSK)], and ATP-sensitive K+ channel (KATP) activator (cromakalim). We performed organ culture of the entire vessel in this study23–25) because the vessel could be easily incubated with a constant concentration of tunicamycin over a prolonged period.
Male Wistar rats (13–19 weeks old) were housed under standard laboratory conditions with free access to food and water. This study was approved by the Hoshi University Animal Care and Use Committee, and all studies were conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health and the Guide for the Care and Use of Laboratory Animals adopted by the Committee on the Care and Use of Laboratory Animals of Hoshi University (which is accredited by the Ministry of Education, Culture, Sports, Science and Technology of Japan). The superior mesenteric arteries of rats were isolated for the following assays.
Experimental ProtocolThe methods of organ-culture and vascular functional experiments were essentially based on our previous reports.23,24) Briefly, non-fasted rats were anesthetized with isoflurane (initially at 5% and then maintained at 2.5%) via a nose cone for surgical procedures, and euthanized by thoracotomy and exsanguination via cardiac puncture in all experiments. After euthanasia, the superior mesenteric arteries were isolated, cleaned, and cut into rings under sterile conditions and placed in an ice-cold, oxygenated, modified Krebs–Henseleit solution (KHS). Some rings were placed in 300 µL of low-glucose (5.5 mM) Dulbecco’s modified Eagle’s medium (Gibco BRL, Grand Island, NY, U.S.A.) supplemented with 1% penicillin streptomycin (Gibco BRL) and 1% fetal bovine serum (Biological Industries, Kibbutz Beit Kaemek, Israel) in vehicle [dimethyl sulfoxide (DMSO)] or tunicamycin (20 µg/mL22)). The rings were kept at 37°C in an atmosphere of 95% air and 5% CO2 for 22–24 h. After incubation, we 1) conducted a vascular functional study, 2) performed Western blotting, and 3) performed basal NO measurements.
For the vascular functional studies, the vascular isometric force of the superior mesenteric artery was measured as described in our previous papers.26–33) The arterial rings were stretched until an optimal resting tension of 9.8 mN was loaded and then allowed to equilibrate; subsequently, the high-K+ (80 mM)-induced contraction was measured. For the relaxation studies, mesenteric arterial rings were pre-contracted with an equi-effective concentration of phenylephrine (PE) (0.6–50 µM; the contractile force was similar between vehicle- and tunicamycin-treated groups). When PE-induced contraction had reached a plateau, ACh (10−9–10−5 M),28–32) SNP (10−10–10−5 M),33,34) ISO (10−9–10−6 M),27,35) FSK (10−9–10−5 M),27) or cromakalim (10−9–10−5 M)36) was added in a cumulative manner.
Western BlottingThe superior mesenteric arteries were treated with vehicle (DMSO) or tunicamycin (20 µg/mL) for 22–24 h under the conditions used in the functional study; consecutively, the arteries were freeze-clamped in liquid nitrogen and stored at −80°C for Western blotting. The protein levels of BiP, a major maker of ER stress, were measured as described previously.16,24) To normalize the data, we used β-actin as a housekeeping protein. The bands were analyzed by using CS Analyzer 3.0 software (ATTO, Tokyo, Japan).
Assessment of Basal NO FormationTo investigate the basal formation of NO, we measured the NO synthase (NOS) inhibitor-induced contraction37,38) and NO metabolites.30,31) Two superior mesenteric arterial rings (2 mm/rat) were divided into two groups: the vehicle and tunicamycin groups. After treatment with DMSO (vehicle group) and tunicamycin (20 µg/mL, tunicamycin group) for 22.5 h, the rings were mounted in an organ bath, as mentioned above. To obtain contraction induced by NG-nitro-L-arginine (L-NNA) (10−4 M), we first applied 50 nM PE, the contraction of which was small and similar between the vehicle (DMSO) and tunicamycin groups. Subsequently, the time–course contractions after a single application of L-NNA (10−4 M) were measured for 30 min.
In another set of experiments to understand basal NO formation, NO metabolites (NOx; nitrite+nitrate) were measured. Two superior mesenteric arterial rings (4 mm/rat) were divided into two groups: the vehicle and tunicamycin groups. After treatment with DMSO (vehicle group) and tunicamycin (20 µg/mL, tunicamycin group) for 22 or 22.5 h, the rings were transferred into KHS at 37°C for 5 min and then removed and weighed. Effluent samples were assayed by using a NOx analyzer (ENO20; Eicom, Kyoto, Japan).30,31)
ReagentsThe functional experiments were performed in KHS with the following composition (mM): 118.0, NaCl; 4.7, KCl; 25.0, NaHCO3; 1.8, CaCl2; 1.2, NaH2PO4; 1.2, MgSO4; and 11.0, glucose in H2O. L-NNA, PE, ISO and FSK, and monoclonal β-actin antibody were all purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). ACh was purchased from Daiichi-Sankyo Pharmaceuticals (Tokyo, Japan). Cromakalim was purchased from Toronto Research Chemicals (North York, ON, Canada). DMSO, SNP, and tunicamycin were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). All concentrations were expressed as the final molar concentration of the base in the organ bath. Horseradish-peroxidase (HRP)-linked secondary anti-mouse antibody was purchased from Promega (Madison, WI, U.S.A.). Primary antibody against the BiP was purchased from Cell Signaling Technology (Danvers, MA, U.S.A.).
Data and Statistical AnalysisData are presented as the mean±standard error of the mean (S.E.M.), and n represents the number of rats. The high-K+ (80 mM)-induced contractions were similar between the vehicle [7.21±0.44 mN (n=23)] and tunicamycin [7.12±0.39 mN (n=24)] groups. The relaxation responses are shown as a percent decrease in the PE-induced pre-contraction level, and individual concentration–response curves were analyzed using a nonlinear regression curve for relaxation–drug concentration relationships to determine the Emax (the maximal response generated by the relaxant drugs), pD2 (a negative logarithm of EC50, which is the concentration of the agonist producing 50% Emax), and area under the curve using Graph Pad Prism software (ver. 5.0 for Mac, San Diego, CA, U.S.A.). Significant differences were calculated using Student’s t-test (Figs. 1–4) or the paired t-test (Fig. 5). * p<0.05 was considered significant.
As shown in Fig. 1, when the PE-induced contraction had reached a plateau level, ACh (10−9–10−5 M) induced a concentration-dependent relaxation in both groups that was significantly weaker in the tunicamycin-treated rings than in the vehicle-treated rings (Fig. 1A). The endothelium–independent relaxation induced by the NO donor SNP (10−10–10−5 M) did not differ significantly between the vehicle- and tunicamycin-treated groups (Fig. 1B).
The arterial rings were preincubated with vehicle (DMSO) and tunicamycin (20 µg/mL) in Dulbecco’s modified Eagle’s medium for 22–24 h. The ordinate shows the relaxation response normalized by phenylephrine (PE)-induced pre-contraction. Each point of the curves represents the mean±S.E. n=6 or 7. * p<0.05, vehicle vs. tunicamycin.
To verify the effect of tunicamycin on relaxation in the cAMP pathway, the mesenteric arterial rings were exposed to two drugs, ISO and FSK, which are related to cAMP signaling. As shown in Fig. 2A, the relaxation induced by ISO (10−9–10−6 M) was significantly weaker in the tunicamycin-treated mesenteric arteries than in the vehicle-treated ones. In the tunicamycin-treated arteries, the relaxation induced by FSK (10−9–10−5 M) was also significantly decreased relative to that in the vehicle-treated ones (Fig. 2B).
The arterial rings were pre-incubated with vehicle (DMSO) or tunicamycin (20 µg/mL) in Dulbecco’s modified Eagle’s medium for 22–24 h. The ordinate shows the relaxation response normalized by PE-induced pre-contraction. Values are the mean±S.E. n=6 or 7. * p<0.05, vehicle vs. tunicamycin.
To investigate the effect of tunicamycin on KATP channel-mediated relaxation, we added cromakalim (10−9–10−5 M) cumulatively to rings pre-contracted by PE (Fig. 3). The relaxation induced by cromakalim was also significantly weaker in the tunicamycin-treated group than that in the vehicle-treated group.
The arterial rings were pre-incubated with vehicle (DMSO) or tunicamycin (20 µg/mL) in Dulbecco’s modified Eagle’s medium for 22–24 h. The ordinate shows the relaxation response normalized by PE-induced pre-contraction. Values are the mean±S.E. n=7. * p<0.05, vehicle vs. tunicamycin.
To investigate whether ER stress was increased in organ-cultured mesenteric artery exposed to tunicamycin, we examined the protein expression of BiP.5–7,12,15,21) As shown in Fig. 4, the expression of BiP was significantly increased in the tunicamycin-treated group relative to that in the vehicle-treated group.
Western blots for BiP/β-actin in the organ-cultured arteries exposed to vehicle (DMSO) or tunicamycin (20 µg/mL) for 22–24 h. Upper: Representative western blots. Lower: Bands were quantified as described in Materials and Methods. Values are the mean±S.E. n=6 or 7. * p<0.05, vehicle vs. tunicamycin.
To investigate whether basal NO generation in the superior mesenteric artery was affected by treatment with tunicamycin, we assessed the contraction induced by the NOS inhibitor [L-NNA (10−4 M)], because this response is an index of basal NO formation.37,38) The L-NNA-induced contraction was significantly reduced in the tunicamycin-treated group compared with that in the vehicle-treated group (Fig. 5A).
(A) Contraction induced by L-NNA (10−4 M), an inhibitor of NOS, as an index of basal NO formation in vehicle (DMSO) or tunicamycin-treated superior mesenteric arteries (for 22.5 h). The contraction is shown as a percentage of the maximum high-K+ (80 mM)-induced contraction. Each data point represents are the mean±S.E. n=3. (B) Basal release of nitrite plus nitrate (NOx) in vehicle (DMSO) or tunicamycin-treated superior mesenteric arteries (for 22 or 22.5 h). The levels of NOx expressed as fold decrease (relative to vehicle). Each column presents the mean±S.E. n=6. * p<0.05, vehicle vs. tunicamycin.
We also measured the basal release of NO metabolites (namely, nitrite, nitrate; NOx) in the superior mesenteric arteries treated with tunicamycin (Fig. 5B). The level of NOx was significantly decreased in the tunicamycin-treated group compared with that in the vehicle-treated group (Fig. 5B).
Although the ER has pivotal roles in the physiological regulation of various processes, the relationship between the disturbance of ER function (i.e., ER stress) and vascular function remains unclear. The findings of the present study show that in vitro short-term treatment with tunicamycin (for 22–24 h) led to increased BiP protein and reduced ACh-induced endothelium-dependent relaxation, cAMP-mediated relaxation induced by ISO and FSK, and KATP-mediated relaxation induced by cromakalim in organ-cultured superior mesenteric arteries.
Vascular tone is tightly regulated by various endogenous factors, and there are diverse signaling pathways involved in the dilation of vasculature in smooth muscle cells.39–50) So far, few studies have investigated the direct effect of tunicamycin on vascular relaxation. In vivo treatment with tunicamycin for several weeks increased ER stress and reduced ACh-induced relaxation in the arteries of rats14) and mice.13) On the other hand, in vitro acute direct exposure of the aorta to tunicamycin (i.e., for 1 h) caused contraction rather than relaxation induced by insulin, and reduced ACh-induced relaxation but not SNP-induced relaxation.22) Using organ-cultured superior mesenteric arteries, we showed for the first time that short-term (i.e., 22–24 h) treatment with tunicamycin impaired vasorelaxation induced by manipulations of different signaling pathways.
Three endothelium-derived relaxing factors are currently known: NO, prostacyclin, and endothelium-derived hyperpolarizing factor (EDHF). These factors affect vasorelaxant activity to various degrees in different vessel types and species.48–51) It has been reported that tunicamycin impaired NO-dependent relaxation in large arteries (i.e., aorta).22) Moreover, Galan et al.52) demonstrated that inhibition of ER stress improved the expression and phosphorylation of endothelial NOS in the mesenteric arteries of streptozotocin-induced type 1 diabetic mice and could partly but not completely restore the ACh-induced endothelium–dependent relaxation in diabetic arteries. Our present study showed that in vitro exposure to tunicamycin caused reduced relaxation to ACh but not SNP, an NO donor. Moreover, we found that basal NO production was reduced in the tunicamycin-treated superior mesenteric arteries. Taken together with other relevant evidence and the present study results, tunicamycin may impair ACh-induced relaxation by reducing NO signaling. However, it remains to be determined if signaling of EDHF is altered in the superior mesenteric artery when treated with tunicamycin because both NO and EDHF have been shown to contribute to vasorelaxation induced by ACh in rat superior mesenteric artery.26,28–32) A future study will be required to answer this question.
In the vascular system, the second messenger cAMP is produced from ATP through the action of AC in response to many extracellular signals41,42) and has an important role in controlling vascular tone.40,44,46) It is well known that activation of beta-adrenoceptors in arteries leads to AC activation, which subsequently causes intracellular cAMP accumulation and activation of protein kinase A (PKA).39,43) Therefore, relaxation of vascular smooth muscle by non-selective beta-adrenoceptor ISO involves cAMP-dependent mechanisms because the beta-adrenoceptors are coupled to AC through Gs proteins.40) In non-vascular tissues/cells, ER stress induction including tunicamycin treatment can affect cAMP signaling.53,54) Although no study has investigated the relationship between tunicamycin and/or ER stress and vasorelaxation mediated by cAMP signaling, our data indicate that impairment of the cAMP pathway is related to the relaxation observed in the tunicamycin-treated superior mesenteric artery.
KATP channels are widely distributed in the vascular system and have a pivotal role in the regulation of vascular tone.45,47) In various disease conditions, KATP channel activator-induced relaxation in arteries is impaired.36,55–57) Our present data showed for the first time that cromakalim-induced relaxation was impaired in tunicamycin-treated superior mesenteric arteries.
The underlying mechanisms of impaired vasorelaxation in tunicamycin-treated superior mesenteric arteries remain unclear. One possibility may be the reduction of basal NO production, which is supported by several reports. For example, in addition to ACh-induced endothelium-dependent relaxation, NO has been shown to positively modulate cAMP signaling and/or cAMP-mediated relaxation, including beta-adrenoceptor stimulation and KATP channel activity.27,58–60) However, further investigations of other signaling pathways associated with vasodilator responses are required.
There were some limitations of present study. Several reports have stated that tunicamycin acts as an inducer of ER stress. Another possibility that might explain the deleterious effect of tunicamycin observed in the present study is the direct modification of N-glycosylation of related vasorelaxation proteins, including receptors, enzymes, and channels. N-glycosylation plays important roles in protein trafficking and requires proper localization for its activity.61) For example, Romero-Fernandez et al. found that tunicamycin inhibited transport of human M3 muscarinic ACh receptor to cell membranes through inhibition of N-glycosylation.62) Wu et al. found that tunicamycin blocked N-glycosylation of type VI AC and reduced FSK-induced cAMP accumulation.63) Tunicamycin reduced membrane sulfonylurea receptors, which are components of KATP via inhibition of N-glycosylation.64) Indeed, in the present study, we found increased ER stress in the tunicamycin-treated arteries. However, tunicamycin may induce ER stress not only by excess accumulation of unfolded/misfolded proteins aggregating into insoluble structures, but also by directly modulating the mechanisms that regulate vascular tone via inhibition of N-glycosylation of the proteins. Therefore, although we could not rule out the possibility that tunicamycin modulates vascular function, future investigations would be needed to confirm this functionality.
There is an emerging body of evidence suggesting that ER stress can be a pathogenic phenomenon in diabetes- and/or hypertension-associated vascular complications. Increased ER stress in vasculature has been observed in models of diabetes or hypertension at different stages of disease.12–15,21) In addition to increased ER stress in vasculature, suppression of ER stress could improve and/or prevent vascular dysfunction in such diseases.13–15,21,50,65,66) We recently stated that ER stress was present in renal arteries obtained from type 2 diabetic Otsuka Long-Evans Tokushima fatty (OLETF) rats throughout the early and later stages of disease.16) Moreover, short-term treatment with tauroursodeoxycholic acid (TUDCA), a suppressor of ER stress, has been shown to improve some vascular functions in renal arteries of OLETF rats in the chronic stage of diabetes.16) On the other hand, we (present study) and others14,21,22,67,68) have observed that induction of ER stress impaired vascular functions and increased blood pressure. Considering the evidence from those studies and the present study, manipulation of ER stress may become a therapeutic goal of diabetes/hypertension-associated vasculopathies.
In conclusion, short-term treatment with tunicamycin was found to cause vascular dysfunction, including not only endothelial dysfunction but also impaired AC/cAMP- and KATP-mediated responses in superior mesenteric arteries of rats. This work provides new insight into the effect of ER stress on vascular dysfunction. We believe that these findings should stimulate further interest in comprehensively understanding the molecular basis of the pathophysiology associated with ER stress.
This work was supported in part by JSPS KAKENHI Grant Numbers 26460107, 15K21419, and 15K07975. The authors would also like to thank M. Takeuchi, M. Takahashi, A. Suwa, J. Nomoto, S. Hotozuka, M. Majima, and H. Sashikubi for the excellent technical assistance.
The authors declare no conflict of interest.
