Upregulation of 5-Hydroxytryptamine Receptor Signaling in Coronary Arteries after Organ Culture

Background 5-Hydroxytryptamine (5-HT) is a powerful constrictor of coronary arteries and is considered to be involved in the pathophysiological mechanisms of coronary-artery spasm. However, the mechanism of enhancement of coronary-artery constriction to 5-HT during the development of coronary artery disease remains to be elucidated. Organ culture of intact blood-vessel segments has been suggested as a model for the phenotypic changes of smooth muscle cells in cardiovascular disease. Methodology/Principal Findings We wished to characterize 5-HT receptor-induced vasoconstriction and quantify expression of 5-HT receptor signaling in cultured rat coronary arteries. Cumulative application of 5-HT produced a concentration-dependent vasoconstriction in fresh and 24 h-cultured rat coronary arteries without endothelia. 5-HT induced greater constriction in cultured coronary arteries than in fresh coronary arteries. U46619- and CaCl2-induced constriction in the two groups was comparable. 5-HT stimulates the 5-HT2A receptor and cascade of phospholipase C to induce coronary vasoconstriction. Calcium influx through L-type calcium channels and non-L-type calcium channels contributed to the coronary-artery constrictions induced by 5-HT. The contractions mediated by non-L-type calcium channels were significantly enhanced in cultured coronary arteries compared with fresh coronary arteries. The vasoconstriction induced by thapsigargin was also augmented in cultured coronary arteries. The decrease in Orai1 expression significantly inhibited 5-HT-evoked entry of Ca2+ in coronary artery cells. Expression of the 5-HT2A receptor, Orai1 and STIM1 were augmented in cultured coronary arteries compared with fresh coronary arteries. Conclusions An increased contraction in response to 5-HT was mediated by the upregulation of 5-HT2A receptors and downstream signaling in cultured coronary arteries.

Introduction 5-Hydroxytryptamine (5-HT) is an important signaling molecule in regulation of the cardiovascular system. It is stored primarily in platelets and is released into plasma if activated at injury sites [1] and in patients with coronary artery disease (CAD) [2]. 5-HT can produce harmful acute and chronic effects. It promotes platelet aggregation, vasoconstriction, and proliferation of vascular smooth muscle cells (VSMCs) [3]. Hence, high levels of 5-HT in plasma are associated with accelerated cardiovascular events [4]. 5-HT is a powerful constrictor of coronary arteries in several species (including humans). 5-HT is considered to be involved in the pathophysiological mechanisms of coronary-artery spasm.
The vascular organ-cultured system has the distinct advantage of maintaining the differentiated contractile phenotype of smooth muscle cells because of better preservation of tissue architecture, cell-to-cell interactions, extracellular matrix, and cell morphology. Some studies have shown that organ culture of segments of intact blood vessels could be a model for the phenotypic changes that occur in VSMCs during the development of cardiovascular disease [8,9]. Pathological conditions such as hypertension, atherosclerosis and diabetes are often associated with proliferation and differentiation of vascular smooth muscle cells and subsequent structural changes in the vascular wall [10]. Organ culture of rat coronary arteries induces upregulation of expression of contractile endothelin type-B receptors on VSMCs, thereby mimicking atherosclerosis and CAD [11,12]. Organ culture also induces downregulation of the contractility of angiotensin-II receptors, thereby reflecting the phenotypic changes in human coronary arteries from patients with ischemic heart disease [13,14]. Expression of 5-HT 1B/1D receptors is also upregulated after organ culture of rat cerebral arteries, which resembles the alterations in SMC function after subarachnoidal hemorrhage [15,16,17]. So far, this model of vascular disease has been applied to study the pharmacological characteristics and underlying molecular and cellular mechanism of vascular receptor alterations.
We investigated the vasoconstriction and expression of 5-HT receptors of rat coronary arteries after 24-h culture in vitro. Our results suggested that upregulation of expression of 5-HT 2A receptors and the downstream phospholipase C (PLC)/Orai1 signaling pathway could be responsible for an augmented contraction in response to 5-HT.

Materials and Methods
The study protocol was approved by the Experimental Animal Ethics Committee of Guangdong General Hospital (Guangzhou, P.R. China).
Tissue collection and organ culture [9] Male Sprague-Dawley rats (220-280 g; 8-10 weeks) were used. Rats were housed under 12-h light-dark conditions with ad libitum access to water and food. Rats were euthanized with CO 2 and decapitated. Hearts were removed immediately and chilled in icecold Kreb's solution (in mM): NaCl 119, KCl 4.7, CaCl 2 2.5, MgCl 2 1, NaHCO 3 25, KH 2 PO 4 1.2, and D-glucose 11.1. The left anterior descending coronary artery was excised from the myocardium. Adherent connective tissues were removed. The endothelium was also removed, and cut into two ringed segments (length, <2 mm, diameter, <250 mm.). Arterial segments were placed in wells containing Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U mL 21 penicillin, and 100 mg mL 21 streptomycin. Wells were incubated at 37uC in an atmosphere of 5% CO 2 for 24 h, and thereafter mounted in myographs to record arterial tone. Other arterial segments used for real-time polymerase chain reaction (PCR) and western blotting experiments were frozen on dry ice and stored at 280uC.

Measurement of vessel force
Measurement of vessel force was undertaken as described previously [18]. Each segment was mounted in a Multi Myograph system (Danish Myo Technology, Aarhus, Denmark) and changes in arterial tone recorded. Briefly, two tungsten wires (each with a diameter of 40 mm) were inserted through the lumen of the segment, and each wire fixed to the jaws of a myograph. The organ chamber was filled with 5 mL Krebs solution, which was bubbled constantly with a mixture of 95% O 2 and 5% CO 2 and maintained at 37uC. Each ring was stretched initially to 1.5 mN (optimal tension) and then allowed to stabilize at this baseline tone for 90 min before the start of each experiment. Each experiment was conducted using rings from different rats. The endothelium was removed mechanically by rubbing the luminal surface of the ring with small stainless-steel wire. 5-HT at 2 mM was used to precontract the vessels. Functional removal of the endothelium was verified if the relaxant effect of 1 mM acetylcholine was absent. In experiments using a high-K + solution (60 mM), an equimolar amount of K + replaced Na + to retain constant ionic strength. We determined the contractile sensitivity to Ca 2+ . Rings were rinsed thrice using a Ca 2+ -free solution containing 50 mM Na 2 -EGTA, then incubated in Ca 2+ -free and 60 mM K + solution before cumulative addition of CaCl 2 .

Culture of VSMCs from rat coronary arteries
Coronary arteries were harvested immediately as mentioned above. Then, they were placed in ice-cold Kreb's solution (in mM): NaCl 119, KCl 4.7, CaCl 2 2.5, MgCl 2 1, NaHCO 3 25, KH 2 PO 4 1.2, and D-glucose 11.1. After removal of fat and connective tissue, coronary arteries were cut into small pieces (length, <0.2 mm) and placed in a 35-mm culture dish. Dulbecco's modified essential medium supplemented with 20% fetal bovine serum (FBS), 100 U mL 21 penicillin, and 100 mg mL 21 streptomycin was added. Cells were cultured at 37uC in an atmosphere of 5% CO 2 for 3 days, after which the dish was replaced with fresh culture medium. SMCs migrated from the specimens in the vessel in 7-10 days. After reaching 80% confluence, cells were subcultured with 0.25% trypsin with 0.02% ethylenediamine tetra-acetic acid.

Ex vivo lentivirus infection in coronary VSMCs
Before the day of infection, coronary VSMCs were seeded at 1610 5 cells mL 21 in six-well plates. Upon reaching 30% confluence, cells were infected with 10 MOI LV-Orai1 shRNA in 1 mL complete culture medium (DMEM with 10% FBS) for 8 h at 37uC. Then, the supernatant was discarded and replaced with fresh culture medium, and VSMCs cultured for an additional 96 h. The effect of Orai1 knockdown was detected using western blotting.
Loading of Fluo-4/AM dye and confocal laser scanning microscopy for recording of intracellular levels of calcium (Ca 2+ ) in coronary VSMCs The intracellular Ca 2+ concentration in coronary SMCs was monitored using the fluorescent dye Fluo 4-AM. Cells were loaded for 25 min in DMEM with 5 mM Fluo 4-AM, which were dissolved in dimethyl sulfoxide (DMSO; final concentration, 0.1%) and Pluronic F-127 (final concentration, ,0.025%). Then, culture dishes were rinsed in standard extracellular solution containing (in mM): NaCl 132, KCl 4.8, MgCl 2 1.2, glucose 5, HEPES 10, and CaCl 2 1.8. Fluo-4 fluorescence was monitored using an inverted Confocal Laser Scanning Microscope (SP5-FCS, Leica, Mannheim, Germany). For fluorescence excitation, the 488-nm band of an argon laser was used, and the emission wavelength was 525 nm. The time-dependent change in mean fluorescence along the scanning line was used to record the intracellular Ca 2+ concentration. The Ca 2+ level was reported as F/F 0 (where F 0 is the resting fluorescence of Fluo-4).

Quantitative real-time PCR (real-time PCR)
Total RNA was isolated from coronary SMCs or coronary artery tissue using Trizol reagent (Molecular Research Center, Cincinnati, OH, USA) according to manufacturer instructions. Total RNA (1 mg) was reverse-transcribed in a total volume of 20 mL, and real-time PCR carried out using SYBR green fluorescence. The specific primers for real-time PCR analyses were (forward and reverse, respectively): 5-HT 2A

Statistical analysis
All of the data are expressed as the mean 6 SEM. Responses in each segment are presented as percentage of the K + -induced contraction. E max refers to the maximal contraction induced by an agonist, expressed as a percentage of the 60 mM K+-induced response. The negative logarithm of the agonist concentration that elicited 50% contraction of E max (pEC 50 ) was determined by linear regression analysis using values immediately above and below the half-maximal response. Target gene mRNA levels were expressed in relation to GAPDH levels. Target protein expression was determined relative to GAPDH protein levels. For statistical analysis, a two-tailed Student's t-test or one-way analysis of variance followed by a Newman-Keuls test was used when more than two groups were compared. Individual concentrationresponse curves were also compared using a two-way analysis of variance followed by Bonferronic posttests. The data and statistical analysis was calculated using Sigmaplot v10.0 (Systat Software, Chicago, IL, USA). P,0.05 was considered to be statistically significant.

5-HT induced vasocontractile responses in fresh and cultured coronary arteries
The organ-culture procedure was performed as described previously [9]. First, we observed the morphological change of the cultured arteries. The width of the artery wall was not significantly changed between fresh and cultured arteries, but after organ culture the smooth muscle cells were disarranged and some nuclei of the cells turned round ( Figure S1).
To ascertain if contractile responses were altered in cultured rat coronary arteries, we determined vascular constrictions induced by different vasoconstrictors. The cumulative application of 5-HT produced a concentration-dependent vasoconstriction in fresh and 24 h-cultured rat coronary arteries without endothelia ( Figure 1A). 5-HT induced significantly greater contractions in cultured coronary arteries than in fresh coronary arteries (E max : 138.362.38% in fresh, 204.262.63% in culture, P,0.01; EC 50 : 0.4360.02 mM in fresh, 0.1560.03 mM in culture, P,0.05, n = 12). U46619-(E max : 126.964.5% in fresh, and 125.961.5% in culture, P.0.05; EC 50 : 0.8160.06 nM in fresh, 0.9360.05 nM in culture, P.0.05, n = 8) and high-K + (60 mM)-induced contractions (tension: 6.260.5 mN in fresh, and 6.360.6 mN in culture, P.0.05, n = 10) were comparable in the two groups ( Figures 1B  and C). In addition, we also observed the 5-HT induced vasoconstriction in serum-free organ culture coronary arteries or in cultured arteries with endothelia. The increased contractility to 5-HT was comparable between serum-free cultured arteries and 10% FBS-cultured arteries ( Figure S2), which excluded the possibility that FBS contributes to the augment of 5-HT induced vasoconstriction after organ culture for 24 h. The effects of 5-HTinduced contraction in the vessels with intact endothelium were not significantly different from in vessels without endothelium ( Figure S3).
The 5-HT 2A receptor was functionally linked to a Gq protein whose activation stimulates PLC, so the PLC inhibitor U73122 was tested. U73122 (50 mM) significantly prevented 5-HT-induced contraction ( Figure 2B). These data confirm that 5-HT stimulates the 5-HT 2A receptor and PLC cascade to induce coronary vasoconstriction.
Role of Ca 2+ in the enhanced vasoconstriction induced by 5-HT in cultured coronary arteries 5-HT-induced contractions were suppressed significantly by 2 mM nifedipine (specific L-type calcium blocker) in the control compared with nifedipine (E max : 128.762.2% in fresh, 200.563.3% in culture, P,0.05, n = 8), but incompletely ( Figure 3A). Next, we explored the other Ca 2+ -permeable channels involved.
To explore the component that was independent of L-type Ca 2+ channels, nifedipine first inhibited the contractions induced by high-K + , and subsequent addition of 5-HT still induced significant contractions in the presence of nifedipine. The 5-HT-induced contraction was blocked by 50 mM 2-APB, the non-selective Ca 2+ release-activated Ca 2+ (CRAC) channel blocker ( Figure 3B). These results suggested that Ca 2+ influx through L-type calcium channels In the presence of 2 mM nifedipine, the contractions in coronary arteries induced by 5-HT were greater in cultured coronary arteries (E max : 129.461.4%) than in fresh coronary arteries (E max : 27.660.3%, P,0.05, n = 8) ( Figure 3C). This result suggested that the contractions mediated by non-L-type calcium channels were enhanced.
Furthermore, we ascertained if 5-HT-induced activation of the PLC/inositol 1,4,5-trisphosphate (IP 3 ) signaling cascade would stimulate a functional store-operated Ca 2+ entry (SOC) pathway in coronary arteries. We investigated the implication of SOC in coronary vasoconstriction. After 5-HT-induced contraction, coronary arteries were washed in Ca 2+ free-EGTA buffer to deplete intracellular Ca 2+ stores in the presence of thapsigargin. Then intracellular Ca 2+ stores were loaded by 1.6 mM Ca 2+ buffer. Contractile responses induced by thapsigargin were determined during the loading period. Thapsigargin-induced contractions were augmented in cultured coronary arteries compared with fresh coronary arteries ( Figure 3D).

Orai1 involvement in 5-HT-evoked entry of Ca 2+ into coronary VSMCs
Once we had determined the role of SOC in 5-HT-mediated contractions, we investigated the participation of Orai1 (an essential pore subunit of the SOC channel) in 5-HT effects. We verified that RNA knockdown of Orai1 specifically reduced the expression of target proteins as assessed by western blotting ( Figure 4A). Downregulation of the expression of Orai1 significantly suppressed 5-HT-elicited Ca 2+ influx in the presence of 1 mM nifedipine in coronary VSMCs ( Figure 4B).

Upregulation of 5-HT receptor signaling in cultured coronary arteries
The 5-HT 2A receptor, Orai1 and STIM1 mRNA levels were augmented in cultured coronary arteries ( Figure 5A) but mRNA levels of the 5-HT 2B receptor and L-type calcium channel were not affected. Cultured coronary arteries exhibited the increases in protein levels of the 5-HT 2A receptor, Orai1 and STIM1, but those of the 5-HT 2B receptor and L-type calcium channel were not affected ( Figure 5B). These results suggested that upregulation of expression of 5-HT 2A receptor and downstream signaling were involved in contractile hyperactivity in cultured coronary arteries. In addition, the protein expression of 5-HT 2A receptor, Orai1 and STIM1 in serum-free cultured rat coronary arteries were also examined, the results showed they were all increased ( Figure S4), which indicated that organ culture procedure could induce the upregulation of 5-HT receptor signaling.

Discussion
In the present work, the phenotypic changes in 5-HT receptors in cultured coronary arteries were examined for the first time. Our data suggested that the augmented constriction induced by 5-HT results from upregulation of the 5-HT 2A receptor signaling pathway in cultured coronary arteries.
First, we found that 5-HT induced an increase in vasocontractile responses in cultured rat coronary arteries. 5-HT induced coronary vasoconstriction in concentration-dependent manner. The maximal contractile response was significantly greater in cultured coronary arteries, but EC 50 was much lower in cultured coronary arteries than fresh coronary arteries. So the potency of 5-HT increased in cultured coronary arteries. But the contraction curves induced by U46619 and CaCl 2 showed no significant differences. These results showed that the thromboxane A2 receptor signaling pathway and vascular sensitivity to Ca 2+ were not responsible for the enhancement of vasoconstriction after vessel culture. We further explored the mechanism of the increased vascular reactivity to 5-HT in cultured coronary arteries. We found that sarpogrelate (specific 5-HT 2A receptor antagonist) almost completely reversed the vasoconstriction induced by 5-HT, but that SB204741 (specific 5-HT 2B receptor antagonist) did not affect it in cultured rat coronary arteries. DOI (specific 5-HT 2A receptor agonist) also produced augmented constriction in cultured rat coronary arteries. The PLC inhibitor U73122 completely prevented 5-HT-induced vasoconstriction. These results suggest that 5-HT stimulates the 5-HT 2A receptor and PLC cascade to induce an increase in coronary vasoconstriction. Past researches showed that the 5-HT receptors mediating constriction in coronary arteries from different species were mostly heterogeneous, consisting of a predominant 5-HT 2 receptor and a smaller population of 5-HT 1 receptors [19]. The uniform population of 5-HT 2 receptors in rat coronary arteries is unique [7,20]. Our results also support the concept that the 5-HT 2A receptor signaling pathway mediates the constriction induced by 5-HT in rat coronary arteries. It has been reported that the vasoconstriction induced by 5-HT was enhanced after organ culture in rat cerebral arteries. But this response was inhibited by 5-HT 1B/1D receptor antagonists [15]. 5-HT stimulated the 5-HT 2A receptor to induce enhanced vasoconstriction after the organ culture of rat mesenteric artery ring segments removed of their endothelium [21]. Therefore, upregulation of the expression of the 5-HT 2A receptor contributes to the increase in contractile response to 5-HT in cultured coronary arteries.
However, previous studies have focused almost entirely on changes in the 5-HT receptor in organ cultures [9,15]. In this experiment, we further determined changes in downstream signaling pathways after activation of the 5-HT 2A receptor in cultured coronary arteries. We know that the 5-HT 2A receptor activates PLC through G q and leads to accumulation of IP 3 , Increase in the cytoplasmic level of IP 3 causes the release of Ca 2+ from the endoplasmic reticulum (ER) [22]. Defective regulation of intracellular Ca 2+ is a hallmark of vascular dysfunction and plays a key part in the increased vascular reactivity [23]. 5-HT elicits an increase in intracellular Ca 2+ to induce vasoconstriction. We observed that nifedipine partially inhibited the effect of 5-HT. In the presence of nifedipine, 5-HT-induced constriction is completely inhibited by 2-APB (putative non-selective cationic channel blocker) [24]. Those results suggest that L-type and non-L-type Ca 2+ influx channels could be involved in 5-HT 2A receptormediated constriction in rat coronary arteries. Furthermore, we found that 5-HT-induced constriction of coronary arteries through non-L-type calcium channels was enhanced in cultured coronary arteries.
Ca 2+ influx mediated by non-L-type calcium channels was involved in SOC in VSMCs [23]. In this mechanism, the ER acts as a capacitor. SOC carries a highly Ca 2+ -selective, non-voltagegated, inwardly rectifying current termed ''CRAC current''. Activation of the 5-HT 2A receptor induces IP 3 -mediated release of Ca 2+ in the ER. After depletion of Ca 2+ from the ER, Ca 2+ channels are activated in the plasma membrane to refill internal Ca 2+ stores. Thapsigargin was used to inhibit Ca 2+ -ATPase in the sarcoplasmic reticulum and to promote depletion of intracellular Ca 2+ stores [25]. We found that incubation with thapsigargin induced constriction during the Ca 2+ -loading period in both fresh and cultured coronary artery rings. Constriction was greater in cultured coronary artery rings. These results suggested that activation of CRAC channels increased in cultured coronary artery rings, thereby contributing to augmented influx of extracellular Ca 2+ . New advances in SOC research have identified STIM and Orai proteins to be key signaling molecules in several cell types, and that their interaction promotes SOC. Recently, some researches demonstrated that STIM1-and Orai1-mediated Figure 5. Upregulation of signaling of the 5-HT receptor is involved in contractile hyperactivity in cultured coronary arteries. A, 5-HT 2A receptor, Orai1, STIM1, 5-HT 2B receptor and L-type calcium channel mRNA expression levels in Fresh and Cultured coronary arteries. 5-HT 2A receptor, Orai1, STIM1, 5-HT 2B receptor and L-type calcium channel were expressed relative to GAPDH levels. Expression of the 5-HT 2A receptor, STIM1 and Orai1 gene was augmented in cultured coronary arteries compared with fresh coronary arteries. The graph shows the means 6 SEM of 4-6 experiments on samples from different rats. *P,0.05, **P,0.01 vs fresh coronary arteries. B, 5-HT 2A receptor, Orai1, STIM1, 5-HT 2B receptor and L-type calcium channel protein expression levels in Fresh and Cultured coronary arteries. 5-HT 2A receptor, Orai1, STIM1, 5-HT 2B receptor and L-type calcium channel were expressed relative to GAPDH levels. Expression of the 5-HT 2A receptor, STIM1 and Orai1 protein was increased in cultured coronary arteries compared with fresh coronary arteries. The graph shows the means 6 SEM of 4-6 experiments on samples from different rats. *P,0.05, **P,0.01 vs fresh coronary arteries. doi:10.1371/journal.pone.0107128.g005 Effects of 5-HT on Vasoconstriction PLOS ONE | www.plosone.org SOC and the existence of endogenous STIM1 and Orai1 proteins in rat aortic and coronary VSMCs [23,25]. We showed that lentivirus-Orai1 shRNA infection could knockdown of Orai1 expression in coronary artery SMCs, which attenuated 5-HTelicited Ca 2+ influx. We suggested that, as a potential component of CRAC channels, Orai1 may contribute to the increase in intracellular Ca 2+ induced by 5-HT in cultured coronary arteries.
Possible mechanisms associated with an abnormal function of 5-HT/PLC/Orai signaling pathways include increased activation/ decreased inactivation of proteins. We observed that expression of the 5-HT 2A receptor, STIM1 and Orai1 were increased in cultured coronary arteries. Increased protein levels may account for the increased functional contractile responses observed after depletion of intracellular Ca 2+ . Hoel et al. reported that expression of 5-HT 1B/1D receptors is upregulated after organ culture in rat cerebral arteries [15], which resembles the alterations in SMC function after subarachnoidal hemorrhage [16,17]. There is an increased level of 5-HT 2A receptor mRNA and of contractile 5-HT 2A receptors after organ culture of rat mesenteric artery ring segments in which the endothelium has been removed [21].
Our results showed that upregulation of the 5-HT 2A receptor and downstream signaling pathway was found to contribute to the increase in 5-HT-induced contraction, and it may be involved in the development of cardiovascular disease. Activation of 5-HT 2A receptor signaling elicited contraction of vascular smooth muscle to induce spasms in coronary arteries. The 5-HT 2A receptor appears to play a crucial part in the formation of occlusive thrombi in diseased arteries by platelet aggregation and vasoconstriction [1]. Selective 5-HT 2A receptor antagonists might help reduce the onset of acute coronary events and of acute coronary occlusion after the intervention [26]. Sarpogrelate, a selective 5-HT 2A receptor antagonist, dose-dependently inhibited the 5-HT-induced coronary spasm in porcine model [27]. Sarpogrelate increased both baseline and the maximal coronary blood flow (CBF) without changing the systemic hemodynamics in patients with coronary artery disease [28]. Sarpogrelate has been introduced clinically as a therapeutic agent for the treatment of ischemic diseases associated with thrombosis [1]. Here we showed for the first time that STIM1 and Orai1 proteins are upregulated in cultured coronary artery. Oari1 channel inhibitors also may represent a new therapeutic approach to treat coronary artery dysfunction [23]. Figure S1 The morphological changes of the rat coronary artery after organ culture. A, Light micrographs of hematoxylin and eosin sections of the fresh and 24 h-cultured rat coronary arteries (4006magnification). B, There was no change in the width of the media of rat coronary arteries after organ culture. (TIFF) Figure S2 5-HT-induced contraction in serum-free cultured arteries and FBS cultured arteries. The increased contractility to 5-HT was comparable between serum-free organ culture and serum organ culture (n = 5). (TIF)