Atherosclerosis 220 (2012) Contents lists available at SciVerse ScienceDirect. Atherosclerosis

Transcription

1 Atherosclerosis 220 (2012) Contents lists available at SciVerse ScienceDirect Atherosclerosis jo ur nal homep age : Role of serotonin in angiogenesis: Induction of angiogenesis by sarpogrelate via endothelial 5-HT1B/Akt/eNOS pathway in diabetic mice Masaaki Iwabayashi a, Yoshiaki Taniyama a,b,, Fumihiro Sanada a, Junya Azuma a,b, Kazuma Iekushi a,b, Hiroshi Kusunoki a,b, Amarnath Chatterjee a, Keita Okayama a, Hiromi Rakugi b, Ryuichi Morishita a, a Department of Clinical Gene Therapy, Osaka University Graduate School of Medicine, Suita, Osaka , Japan b Department of Geriatric Medicine and Nephrology, Osaka University Graduate School of Medicine, Suita, Osaka , Japan a r t i c l e i n f o Article history: Received 25 April 2011 Received in revised form 30 September 2011 Accepted 31 October 2011 Available online 9 November 2011 Keywords: Serotonin Diabetes mellitus Sarpogrelate 5-HT1B enos a b s t r a c t Serotonin (5-hydroxytryptamine, 5-HT) plays a crucial role in peripheral artery disease (PAD) and diabetes mellitus (DM). In these conditions, the balance between the 5-HT2A receptor in smooth muscle cells and the 5-HT1B receptor in endothelial cells (ECs) regulates vascular tonus. In the present study, we focused on the role of 5-HT in endothelial dysfunction using a selective 5-HT2A receptor blocker, sarpogrelate. In human EC, 5-HT markedly stimulated enos expression and the phosphorylation of enos, Akt and ERK1/2. In addition, a dose-dependent increase in tubule-formation on Matrigel was observed after 5- HT treatment. In contrast, high glucose significantly inhibited tubule formation and enos expression through inactivation of Akt, while 5-HT significantly attenuated these actions of high glucose (P < 0.01). These results indicate that 5-HT stimulated angiogenesis through activation of Akt in ECs. However, in clinical situations, 5-HT seems to act as the devil. To examine the role of 5-HT in diabetic PAD, a hindlimb ischemia model was created in diabetic mice. The blood flow ratio of the ischemic to non-ischemic limb was significantly lower in DM mice than in normal mice, while sarpogrelate significantly attenuated the decrease in the blood flow ratio compared to control (P < 0.01). Consistently, the decrease in enos expression and Akt activity in DM mice was significantly attenuated by sarpogrelate. Overall, the present study demonstrated that selective inhibition of 5-HT2A by sarpogrelate significantly restored ischemic limb blood perfusion in a severe diabetic mouse model through stimulation of the enos/akt pathway via the endothelial 5-HT1B receptor. Enhancement of vasodilation and angiogenesis by sarpogrelate might provide a unique treatment for PAD and DM patients Elsevier Ireland Ltd. All rights reserved. 1. Introduction Progression of atherosclerosis results in the occlusion of major limb arteries and diminishes limb perfusion, leading to peripheral artery disease (PAD). Diabetes mellitus (DM) is a strong risk factor for cardiovascular disease [1,2]. The prevalence of PAD in patients with DM is 5-fold higher than that in patients without DM [3]. In the pathological conditions of PAD [4] and DM [5], it is well known that platelet activation occurs and platelet aggregation in the vascular wall is increased. As activated platelets in atherosclerotic lesions release large amounts of serotonin (5-hydroxytryptamine, Corresponding author at: Department of Clinical Gene Therapy, Osaka University Graduate School of Medicine, 2-2 Yamada-oka, Suita , Japan. Tel.: ; fax: Corresponding author. addresses: taniyama@cgt.med.osaka-u.ac.jp (Y. Taniyama), morishit@cgt.med.osaka-u.ac.jp (R. Morishita). 5-HT), plasma 5-HT level is elevated [6]. Released 5-HT binds to the 5-HT2A receptor in vascular smooth muscle cells (VSMC), and accelerates calcium influx and VSMC proliferation [7,8], leading to vascular constriction. Alternatively, 5-HT initiates autocrine activation, and aggregation of platelets through its receptor, 5-HT2A, on the cell surface, occurs [9]. It is reported that vascular sensitivity to 5-HT is significantly enhanced in patients with DM [10], suggesting this vicious cycle is tightly linked to the pathogenesis and development of PAD. Under such conditions, inhibition of 5- HT2A in both VSMC and platelets might be beneficial to treat such patients. Indeed, a selective 5-HT2A receptor blocker, sarpogrelate, has been widely used clinically to treat PAD patients [11,12], as sarpogrelate suppressed 5-HT-mediated vasoconstriction [13] and increased limb blood perfusion [14,15] in experimental animal models. However, in the endothelium, 5-HT is known to mediate NO production through 5-HT1B on endothelial cells, leading to vasodilation [16,17]. Nevertheless, the role of 5-HT in the vasculature is thought to lean towards vascular constriction, as /$ see front matter 2011 Elsevier Ireland Ltd. All rights reserved. doi: /j.atherosclerosis

2 338 M. Iwabayashi et al. / Atherosclerosis 220 (2012) Fig. 1. Effects of 5-HT on HAEC. (A) Effect of 5-HT (10 M) on the phosphorylation of Akt and enos in HAECs. (B) Effect of 5-HT (0 100 M) on protein expression of enos in HAECs at 24 h after treatment. (C) Representative photographs of tubule formation in HAEC on Matrigel at 24 h after treatment. Scale bar = 200 m. (D) Quantitative analysis of tubule length after 5-HT stimulation cells were seeded on Matrigel-coated 24-well plates. Tubule length was determined at 24 h after stimulation with various concentrations of 5-HT. n = 3, *P = , **P < vs. control. Tubule length was determined in 5 random fields. hypersensitivity to 5-HT-induced vascular constriction is increased in advanced atherosclerosis [13]. Thus, the precise molecular mechanisms of how 5-HT causes progression of vascular dysfunction are still an enigma. In the present study, we focused on the role of 5-HT in endothelial function both in vivo and in vitro using a selective 5-HT2A receptor blocker, sarpogrelate. Here, we demonstrated that the selective inhibition of the 5-HT2A receptor by sarpogrelate enhanced therapeutic angiogenesis in a severe diabetic mouse model. 2. Materials and methods 2.1. Cell culture Human aortic endothelial cells (HAECs) purchased from LONZA (Portsmouth, NH) were cultured in endothelial basal medium-2 (Clonetics, Walkersville, MD, USA) supplemented with EGM and 5% fetal bovine serum (FBS). Prior to stimulation, cells were starved with 0.5% FBS-containing medium for 24 h Western blotting Protein extracts (10 g) were fractionated on 8% SDS PAGE gel and transferred to a PVDF membrane (Millipore, USA) [18] Matrigel tubule formation assay The detailed procedure is described in Supplemental information section Animal model The detailed procedure is described in Supplemental information section Materials The materials used in this study are described in Supplemental information section Statistical analysis Statistical analysis was done using one-way analysis of variance (ANOVA) using JMP9.0 (SAS Institute, Inc., Cary, NC, USA). Normality and homoscedasticity were checked visually. Analysis was performed after log-transformation for non-normally distributed data. Within each ANOVA, the P-values used to assess the statistical significance of pairwise comparisons were adjusted for multiplicity by the Tukey Kramer method. Differences were regarded as significant at the P < 0.05 level. All data were reported as mean ± standard error (SE). 3. Results 3.1. Induction of angiogenesis by 5-HT in HAEC To examine the role of 5-HT in endothelial function, we initially examined the effects of 5-HT on enos and Akt activity using an in vitro culture model. As shown in Fig. 1, 5-HT markedly stimulated the phosphorylation of enos (Ser1177) and Akt (Ser473) (Fig. 1A). Similarly, 5-HT also stimulated the phosphorylation of ERK1/2. Also, an increase in the expression of enos was still observed at 24 h after 5-HT treatment (Fig. 1B). As an increase in enos is reported

3 M. Iwabayashi et al. / Atherosclerosis 220 (2012) Fig. 2. Effect of 5-HT on phosphorylated Akt (A) and enos (B) under high glucose condition in HAEC. Upper panel: Representative photographs of western blotting; lower panel: Quantitative analysis of phosphorylated Akt or enos. HAECs were pretreated with LY294002, a specific Akt PI3K inhibitor (2 M), for 1 h and 5-HT (10 M) for 4 h, and then stimulated with high glucose (33 mm) for 24 h. The concentration of LY (2 M) was chosen to block Akt activity sufficiently. Y-axis of the graph is displayed logarithmically. (A) n = 5, *P = vs. control, P = vs. high glucose, P < vs. high glucose with 5-HT. (B) n = 4 6, *P = vs. control, **P < vs. control, P = vs. high glucose, P = vs. high glucose with 5-HT. to stimulate angiogenesis [19], we examined the effects of 5-HT on angiogenic properties such as tubule formation in endothelial cells. After stimulation of HAEC on Matrigel with 5-HT, tubule formation was observed at 24 h after 5-HT treatment (Fig. 2A). 5-HT significantly increased tubule length in a dose-dependent manner (Fig. 1D, P < 0.001). To study the physiological role of 5-HT, we employed an in vitro diabetic model. A high glucose condition (33 mm) significantly inhibited the phosphorylation of Akt in HAECs (Fig. 2A, P = ). However, 5-HT significantly increased Akt activity under high glucose condition (P = ). As shown in Fig. 2B, although a high glucose condition significantly decreased enos expression (P = 0.030), 5-HT significantly restored enos expression (P = 0.047). This protective effect of 5-HT on enos expression was significantly inhibited by LY294002, a specific Akt PI3K inhibitor (P = ). High glucose also significantly inhibited tubule formation (Fig. 3, P < 0.001), while 5-HT significantly attenuated the decrease in tubule formation induced by high glucose (P = ). However, the decrease in Akt activity by LY significantly diminished the protective effect of 5-HT on tubule formation (P < 0.001). These results demonstrated that 5-HT stimulated angiogenesis through the enos/akt pathway in HAEC Improvement of ischemic limb perfusion by sarpogrelate in DM model However, in in vivo and clinical situations, 5-HT seems to act as the devil, in an apparently opposite manner to those in vitro experiments in endothelial cells. The explanation for this gap is the presence of different subtypes of 5-HT receptor. As described above, the 5-HT2A receptor in VSMC accelerates VSMC proliferation and vascular constriction. Similarly, the 5-HT2A receptor on platelets enhances their aggregation. However, the 5-HT1B receptor on endothelial cells mediates vasodilation through NO production [16,17], consistent with the present study. To examine the role of 5-HT in diabetic PAD, a hindlimb ischemia model was created in severely diabetic mice. As shown in Fig. 4B, the blood flow ratio of the ischemic to non-ischemic limb was significantly lower in DM mice than in normal mice at 28 days. Therefore, we tested the selective inhibition of 5-HT2A using sarpogrelate. Selective inhibition of 5-HT2A would enhance the actions of 5-HT1B in endothelial cells, possibly leading to an increase in NO production and a vasodilator response (see Fig. 5). As expected, treatment with sarpogrelate significantly attenuated the decrease in blood flow ratio of the ischemic to non-ischemic limb at 28 days after operation as compared to control (Fig. 4B, P = at 28 days). This effect was not due to changes in blood glucose and body weight, since there was no significant difference in blood glucose between vehicle- and sarpogrelate-treated DM mice (492.3 ± 47.4 mg/dl), while blood glucose level was significantly higher in diabetic mice than in normal mice (normal mice: ± 5.05 mg/dl vs. DM mice: ± 31.7 mg/dl, P < 0.001). Body weight of DM mice was significantly lower than that of non-dm mice (normal mice: 26.5 ± 0.28 g vs. DM mice: 20.3 ± 0.73 g, P < 0.001). However, there was no significant difference in body weight between vehicle- and sarpogrelate-treated DM mice (21.3 ± 0.77 g). The beneficial effect of sarpogrelate to improve ischemic limb perfusion in a diabetic condition has been reported previously [14]. However, the dosage of STZ used in the present study was much higher than that used in the previous study. Thus, it is becoming more evident that sarpogrelate can improve ischemic limb perfusion even in a more severe diabetic condition. Therefore, we finally explored how sarpogrelate stimulated blood flow. We focused on endothelial function. As shown in Fig. 4C, enos expression in the ischemic limb in DM mice was significantly decreased as compared to that in normal mice. However, the decrease in enos expression in DM mice was significantly attenuated by sarpogrelate (Fig. 4C, P = 0.022). The increase in enos expression may be due to an increase in phosphorylation of Akt, since Akt activity was significantly increased by sarpogrelate (Fig. 4D, P = ). In contrast, Akt activity was significantly lower in DM mice as compared to normal mice (Fig. 4D, P < 0.001). These

4 340 M. Iwabayashi et al. / Atherosclerosis 220 (2012) Fig. 3. Effect of 5-HT on tubule formation under high glucose condition in vitro. (A) Representative photographs of tube formation in HAEC on Matrigel at 24 h after treatment. Scale bar = 200 m. (B) Quantitative analysis of tubule length HAECs were seeded on Matrigel-coated 24 well plates. Cells were pretreated with LY (2 M) for 1 h and 5-HT (10 M) for 4 h, and then stimulated with high glucose (33 mm) for 16 h. HG, high glucose; LY, LY n = 4, *P = vs. control, **P < vs. control, P < vs. high glucose, # P < vs. high glucose with 5-HT. Tubule length was determined in 5 random fields. Fig. 4. Upper panel: Effect of sarpogrelate on ischemic limb blood perfusion and protein expression in adductor muscle in vivo. (A) Representative LDI images of limb perfusion at 28 days. Yellow arrows indicate ischemic foot. Red color shows blood perfusion. (B) Quantitative analysis of ischemic/non-ischemic limb perfusion ratio by LDI. Administration of sarpogrelate (50 mg/kg/day) was started from 1 week before mice were operated on to induce hind limb ischemia. Normal mice, limb perfusion of normal (non-dm) mice; DM control mice, limb perfusion of DM control mice; DM mice with sarpogrelate, limb perfusion of DM mice with sarpogrelate. n = 6 8, **P < vs. normal mice, *P = vs. DM control mice, P = vs. normal mice, P = vs. DM control mice, # P = vs. normal mice, P = vs. DM control mice. Values are expressed as mean ± SE. Lower panel: Western blotting analysis of phosphorylated Akt (C) and enos (D) in adductor muscle at 28 days after surgery in vivo. Upper panel: Representative photographs of western blotting; lower panel: Quantitative analysis of phosphorylated Akt or enos. Y-axis of the graph is displayed logarithmically. Normal mice, ischemic limb of normal (non-dm) mice; control, ischemic limb of DM control mice; sarpogrelate, ischemic limb of DM mice with sarpogrelate. (C) n = 3, *P < vs. control, P = vs. control (DM control mice). (D) n = 3, *P < vs. control, P = vs. control (DM control mice).

5 M. Iwabayashi et al. / Atherosclerosis 220 (2012) Fig. 5. Actions of sarpogrelate in blood vessels. In the normal vasculature, the vascular tonus is regulated by the balance between 5-HT1B on EC and 5-HT2A on VSMC. (A) In DM and PAD, platelets are aggregated and activated, leading to increased serum and tissue levels of 5-HT. Hypersensitivity of 5-HT2A to 5-HT of VSMC and platelets mediated vasoconstriction and stimulated atherosclerosis, while vasodilation induced by 5-HT through 5-HT1B was impaired. (B) Selective blockage of 5-HT2A by sarpogrelate inhibited vasoconstriction and platelet aggregation in VSMC and platelets. Increased 5-HT only stimulated the binding to 5-HT1B in endothelium, leading to an increase in NO production via the Akt/eNOS pathway. Thus, 5-HT under sarpogrelate increases blood flow, and induces vasodilation and angiogenesis. data suggest that the improvement of lower blood flow perfusion by sarpogrelate might have been due to an increase in endothelial enos expression through the Akt pathway. 4. Discussion 5-HT is synthesized by enterochromaffin cells from tryptophan and is taken up by, and stored in platelets [9]. During platelet aggregation, 5-HT is released from these intracellular stores. 5-HT promotes vascular constriction and VSMC proliferation via the 5- HT2A receptor in VSMC. 5-HT2A expression is elevated in ischemic limbs, and is further augmented in DM [14]. 5-HT also stimulates platelets aggregation through 5-HT2A on the platelet surface. Given the increase in serum 5-HT concentration in patients with DM and PAD, 5-HT is believed to play a crucial role in DM and PAD as the bad guy. On the other hand, 5-HT is believed to mediate vasodilation through enos production via 5-HT1B in endothelial cells and endothelium. Consistent with previous reports [20], the present study confirmed that 5-HT increased the activity of enos through the phosphorylation of Akt. Moreover, we demonstrated that 5- HT stimulated angiogenesis in human endothelial cells. These data suggest that 5-HT may act as the good guy in endothelial function. Apparently, 5-HT has double faces in regulation of the vascular system (see Fig. 5). In clinical practice, the blockade of 5-HT might have therapeutic values to treat DM and PAD patients, since sarpogrelate, a selective 5-HT2A blocker, inhibits platelet aggregation, vasoconstriction and VSMC proliferation. Recent clinical studies demonstrated that sarpogrelate treatment is effective for thrombosis, coronary artery spasm, atherosclerosis, PAD and DM [21]. Currently, therapeutic angiogenesis is the center of interest in the treatment of PAD, since stimulation of angiogenesis using angiogenic growth factors such as HGF (hepatocyte growth factor) [22] and stem/progenitor cells such as BMMNCs (bone marrow mononuclear cells) [23] demonstrated clinical efficacy in critical limb ischemia. As 5-HT is also reported to stimulate proliferation of bone marrow cells [24], stimulation of migrated BMMNC might contribute to the enhanced angiogenesis by sarpogrelate. The present data suggest that single treatment with sarpogrelate and/or a combination of sarpogrelate with angiogenesis therapy might be considered to treat critical limb ischemia, since the natural history of critical limb ischemia has been well documented to have an inexorable downhill course. These beneficial effects of sarpogrelate might be further important in treating PAD patients with DM, since the expression of 5-HT2A is up-regulated in VSMC under conditions of insulin resistance such as DM [25]. It is also reported that insulin inhibited 5-HT-induced extracellular calcium influx [26]. Therefore, the bad actions of 5-HT could be enhanced in DM or metabolic syndrome patients. Previous papers reported that high glucose down-regulated Akt activity and induced endothelial cell apoptosis, while over-expression of Akt prevented apoptosis [27]. In contrast, inhibition of the PI3K/Akt/eNOS pathway was reported to enhance endothelial cell apoptosis induced by high glucose. Also in this study, we found that 5-HT pretreatment prevents increased endothelial cell apoptosis induced by high glucose (Fig. S2). It is noteworthy that the expression of phosphorylated Akt and enos in ischemic limb adductor muscles was significantly increased by sarpogrelate, in parallel with recovery of perfusion. These results suggest that the binding of 5-HT to 5-HT1B on endothelial cells might be enhanced by sarpogrelate, leads to restoration of the endothelial function in ischemic limb in severe diabetic mice, not only evoking angiogenesis. Overall, the present study demonstrated that selective inhibition of 5-HT2A by sarpogrelate significantly restored ischemic limb blood perfusion even in a severe diabetic mice model. As sarpogrelate is widely used to treat PAD patients in clinical practice, the

6 342 M. Iwabayashi et al. / Atherosclerosis 220 (2012) enhancement of vasodilation and angiogenesis, in addition to the well known inhibition of vasoconstriction, by sarpogrelate might provide unique treatment combination with therapeutic angiogenesis. Conflict of interest None. Acknowledgments We greatly appreciate Dr. Toshimitsu Hamasaki (Department of Biomedical Statistics, Osaka University Graduate School of Medicine) for statistical advice. This work was partially supported by a Grant-in-Aid from the Organization for Pharmaceutical Safety and Research, a Grant-in-Aid from the Ministry of Public Health and Welfare, a Grant-in-Aid from Japan Promotion of Science, and through special coordination funds of the Ministry of Education, Culture, Sports, Science and Technology, of the Japanese Government. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi: /j.atherosclerosis References [1] Kannel WB, McGee DL. Diabetes and cardiovascular disease: the Framingham study. Journal of the American Medical Association 1979;241:2035. [2] Gaede P, Vedel P, Larsen N, Jensen GVH, Parving HH, Pedersen O. Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes. New England Journal of Medicine 2003;348:383. [3] Second European Consensus Document on chronic critical leg ischemia. Circulation 1991;84:IV1 26. [4] Barradas M, Gill D, Fonseca V, Mikhailidis D, Dandona P. Intraplatelet serotonin in patients with diabetes mellitus and peripheral vascular disease. European Journal of Clinical Investigation 1988;18: [5] Malyszko J, Urano T, Knofler R, et al. Daily variations of platelet aggregation in relation to blood and plasma serotonin in diabetes. Thrombosis Research 1994;75: [6] Hara K, Hirowatari Y, Yoshika M, Komiyama Y, Tsuka Y, Takahashi H. The ratio of plasma to whole-blood serotonin may be a novel marker of atherosclerotic cardiovascular disease. Journal of Laboratory and Clinical Medicine 2004;144:31 7. [7] Pakala R, Willerson JT, Benedict CR. Effect of serotonin, thromboxane A2, and specific receptor antagonists on vascular smooth muscle cell proliferation. Circulation 1997;96:2280. [8] Lee S, Wang W, Moore B, Fanburg B. Dual effect of serotonin on growth of bovine pulmonary artery smooth muscle cells in culture. Circulation Research 1991;68:1362. [9] Maurer-Spurej E. Serotonin reuptake inhibitors and cardiovascular diseases: a platelet connection. Cellular and Molecular Life Sciences 2005;62: [10] Tanaka N. Pharmacological studies on responsiveness of 5-hydroxytryptamine in overcoming perioperative spasm of coronary artery bypass graft. Yakugaku zasshi: Journal of the Pharmaceutical Society of Japan 2009;129:1049. [11] Hara H, Osakabe M, Kitajima A, Tamao Y, Kikumoto R. MCI-9042, a new antiplatelet agent is a selective S2-serotonergic receptor antagonist. Thrombosis and Haemostasis 1991;65:415. [12] Miyazaki M, Higashi Y, Goto C, et al. Sarpogrelate hydrochloride, a selective 5-HT2A antagonist, improves vascular function in patients with peripheral arterial disease. Journal of Cardiovascular Pharmacology 2007;49:221. [13] Nishihira K, Yamashita A, Tanaka N, et al. Inhibition of 5-hydroxytryptamine2A receptor prevents occlusive thrombus formation on neointima of the rabbit femoral artery. Journal of Thrombosis and Haemostasis 2006;4: [14] Bir S, Fujita M, Marui A, et al. New therapeutic approach for impaired arteriogenesis in diabetic mouse hindlimb ischemia. Circulation Journal: Official Journal of the Japanese Circulation Society 2008;72:633. [15] Li T, Furutani A, Takahashi M, et al. Impaired potency of bone marrow mononuclear cells for inducing therapeutic angiogenesis in obese diabetic rats. American Journal of Physiology - Heart and Circulatory Physiology 2006;290:H1362. [16] Hoyer D, Clarke D, Fozard J, et al. International Union of Pharmacology classification of receptors for 5-hydroxytryptamine (Serotonin). Pharmacological Reviews 1994;46:157. [17] Fujita M, Minamino T, Sanada S, et al. Selective blockade of serotonin 5-HT2A receptor increases coronary blood flow via augmented cardiac nitric oxide release through 5-HT1B receptor in hypoperfused canine hearts. Journal of Molecular and Cellular Cardiology 2004;37:1219. [18] Iekushi K, Taniyama Y, Azuma J, et al. Novel mechanisms of valsartan on the treatment of acute myocardial infarction through inhibition of the antiadhesion molecule periostin. Hypertension 2007;49:1409. [19] Murohara T, Asahara T, Silver M, et al. Nitric oxide synthase modulates angiogenesis in response to tissue ischemia. Journal of Clinical Investigation 1998;101:2567. [20] Mcduffie JE, Coaxum SD, Maleque MA. 5-Hydroxytryptamine evokes endothelial nitric oxide synthase activation in bovine aortic endothelial cell cultures. Proceedings of the Society for Experimental Biology and Medicine 1999;221: [21] Park K, Chung WY, Seo JB, et al. The prevention of contrast induced nephropathy by sarpogrelate in patients with chronic kidney disease: a study protocol for a prospective randomized controlled clinical trial. Trial 2010;11:122. [22] Morishita R, Makino H, Aoki M, et al. Phase I/IIa clinical trial of therapeutic angiogenesis using hepatocyte growth factor gene transfer to treat critical limb ischemia. Arteriosclerosis, Thrombosis, and Vascular Biology 2011;31: [23] Lawall H, Bramlage P, Amann B. Stem cell and progenitor cell therapy in peripheral artery disease. Thrombosis and Haemostasis 2010;103: [24] Yang M, Li K, Ng PC, et al. Promoting effects of serotonin on hematopoiesis: ex vivo expansion of cord blood CD34+ stem/progenitor cells, proliferation of bone marrow stromal cells, and antiapoptosis. Stem Cells 2007;25: [25] Ohkura M, Tanaka N, Kobayashi H, Wada A, Nakai J, Yamamoto R. Insulin induces internalization of the 5-HT2A receptor expressed in HEK293 cells. European Journal of Pharmacology 2005;518: [26] Kahn AM, Allen JC, Seidel CL, Song T. Insulin inhibits serotonin-induced Ca2+ influx in vascular smooth muscle. Circulation 1994;90:384. [27] Ho F, Lin W, Chen B, et al. High glucose-induced apoptosis in human vascular endothelial cells is mediated through NF-[kappa] B and c-jun NH2-terminal kinase pathway and prevented by PI3K/Akt/eNOS pathway. Cellular Signalling 2006;18:391 9.