Crk-associated substrate (CAS), a 130-kDa adaptor protein, was discovered as a tyrosine kinase substrate of Src that was important to cellular motility and actin filament formation. As the tyrosine kinase Src is utilized by the 5-hydroxytryptamine (5-HT)2A receptor in arterial contraction, we tested the hypothesis that CAS was integral to 5-HT2A receptor-mediated vasoconstriction. Rat thoracic aorta was used as a model of the arterial 5-HT2A receptor. Western and immunohistochemistry analyses validated the presence of CAS in the aorta, and tissue bath experiments demonstrated reduction of contraction to 5-HT (13.5 ± 5% control maximum) and the 5-HT2 receptor agonist α-methyl-5-HT (6 ± 2% maximum) by latrunculin B (10−6 mol/l), an actin disruptor. In aorta contracted with 5-HT (10−5 mol/l), tyrosine phosphorylation (Tyr410) of CAS was significantly increased (∼225%), and both contraction and CAS phosphorylation were reduced by the 5-HT2A/2C receptor antagonist ketanserin (3 × 10−8 mol/l). Src is one candidate for 5-HT-stimulated CAS tyrosyl-phosphorylation as 5-HT promoted interaction of Src and CAS in coimmunoprecipitation experiments, and the Src tyrosine kinase inhibitor PP1 (10−5 mol/l) abolished 5-HT-induced tyrosyl-phosphorylation of CAS and reduced 5-HT- and α-methyl-5-HT-induced contraction. Antisense oligodeoxynucleotides delivered to the aorta reduced CAS expression (33% control) and arterial contraction to α-methyl-5-HT (45% of control), independent of changes in myosin light chain phosphorylation. These data are the first to implicate CAS in the signal transduction of 5-HT.
- antisense oligodeoxynucleotides
- signal transduction
the activation of protein kinases in phosphorylation cascades is critical to signal transduction. This process is facilitated by adaptor proteins that contain Src homology 2 (SH2) and Src homology 3 (SH3) protein regions that allow for docking of other proteins and facilitation of protein-protein interactions. One such adapter protein is Crk-associated substrate or CAS (also known as p130 or p130 CAS). This 130-kDa protein has been recognized as a substrate for the tyrosine kinase Crk (23), and CAS is highly tyrosine phosphorylated in response to activation of Crk or Src (25, 26). Docking of and tyrosyl-phosphorylation by these proteins, as well as other regulators, including focal adhesion kinase (FAK) and protein tyrosine phosphatase 1B (PTP1B), enables CAS to participate in important physiological events, including cell migration (14, 18, 27), mitosis (30), adhesion (42), and apoptosis (48).
The involvement of CAS in contraction is logical because CAS is essential for actin filament reorganization (12, 38). CAS is a regulator of the actin-associated protein profilin (14), which is necessary for actin-filament polymerization in smooth muscle cells (37). Importantly, the involvement of CAS in contraction is independent of myosin light chain (MLC) phosphorylation, an event appreciated to be permissive in vascular smooth muscle contraction (19, 20). Thus involvement of CAS in contraction is relatively novel in its MLC independence.
The present paper tests the hypothesis that CAS is important to arterial contraction elicited by serotonin (5-HT, 5-hydroxytryptamine). 5-HT has multiple effects in arterial smooth muscle, including vasoconstriction, endothelium-dependent relaxation in some beds, and smooth muscle/endothelial cell mitogenesis (46). Whereas many mechanisms by which 5-HT mediates arterial smooth muscle contraction and smooth muscle cell mitogenesis have been investigated, the ability of 5-HT to utilize CAS in this process is unknown and is logical to consider. 5-HT utilizes c-Src to mediate arterial contraction, as evidenced by the ability of PP-1 to reduce 5-HT-induced contraction (2). Thus we consider here whether CAS is a downstream effector of Src as it pertains to arterial contraction. Ultimately, this knowledge is also relevant because in multiple cardiovascular diseases including atherosclerosis and hypertension, arterial responsiveness to 5-HT is significantly increased and arterial smooth muscle remodeling, including hyperplasia and hypertrophy, is involved and is potentially dependent on CAS (6, 13, 18, 29, 39, 41).
We currently investigate the expression and localization of CAS in rat arterial tissue, as well as contraction to 5-HT in the presence of an actin disruptor, latrunculin B, and the src-family tyrosine kinase inhibitor 4-amino-5-(4-methylphenyl)-7-(t-butyl)pyrazole[3,4-d]pyrimidine (PP1). We used the rat aorta as an arterial model in these particular studies to connect with previous studies done in experimental models of hypertension, such as the deoxycorticosterone acetate (DOCA) salt rat. In the rat aorta, the 5-HT2A receptor has been established as the sole mediator of 5-HT contraction. Though present, the 5-HT2B receptor is not functional and the 5-HT2C receptor has not been detected (2, 32). Thus use of the 5-HT2 receptor agonist α-methyl-5-HT and 5-HT2A/2C receptor antagonist ketanserin allows us to draw connections between 5-HT2A receptor activation and CAS. To investigate the hypothesis that CAS, specifically, was involved in 5-HT contraction, antisense oligodeoxynucleotides ex vivo were delivered for knockdown of CAS mRNA and thus protein. This work, collectively, is the first to investigate the potential of CAS to be involved in arterial signal transduction elicited by 5-HT.
MATERIALS AND METHODS
Normal male Sprague-Dawley rats (0.300–0.350 kg; Charles River Laboratories, Portage, MI) were used. All animal protocols were approved by the Michigan State University Institutional Animal Care committee.
Isolated tissue bath.
Rats were euthanized using 60 mg/kg ip pentobarbital. Aortas were removed, placed in physiological salt solution (PSS) (containing in mmol/l: 103 NaCl, 4.7 KCl, 1.18 KH2PO4, 1.17 MgSO4·7H2O, 1.6 CaCl2·2H2O, 14.9 NaHCO3, 5.5 dextrose, and 0.03 CaNa2 EDTA), cleaned of fat and connective tissue, and cut into endothelium-intact helical strips, and isometric contraction was measured as described previously (8). Strips incubated with vehicle (0.1% dimethylsulfoxide), latrunculin-B (10−6 or 10−7 mol/l), PP1 (10−5 mol/l), or ketanserin (3 × 10−8 nmol/l) for 1 h before addition of vehicle (H2O), 5-HT (10−9 to 3 × 10−4 mol/l; or 10−5 mol/l) or α-methyl 5-HT (10−9 to 3 × 10−5 mol/l). In some experiments, strips were removed and promptly frozen and homogenized in liquid nitrogen for Western blot analyses after a 3-min incubation with vehicle or 5-HT.
Arterial sections of formaldehyde-fixed, paraffin-embedded aorta (8 μm) were cut, air dried overnight, and taken through standard protocol using a Vector kit. After being blocked with 1.5% serum in phosphate-buffered saline (PBS), aortic sections were incubated 18 h with total anti-p130CAS antibody (mouse, 5 μg/ml, Santa Cruz Technologies with 1.5% blocking serum in PBS) or no antibody. Sections were washed three times with PBS, incubated 30 min with secondary antibody, washed again, and incubated 30 min with Vector ABC Elite reagant. Antibody binding was detected by incubating sections 1 min with a DAB developing solution (Vector Laboratories, Burlingame, CA). Binding is represented by a dark brown/black precipitate. Sections were photographed using an inverted Nikon T2000 microscrope connected to a SPOT Insight color camera using MetaMorph software. Images were processed using Adobe Photoshop.
Western blot analysis and coimmunoprecipitation.
Rat thoracic aorta were removed de novo, placed in PSS, and cleaned as described above or taken down from the tissue bath. Tissues were quick frozen, pulverized, and homogenized in standard protease and phosphatase inhibitor-rich buffer (0.5 mol/l Tris·HCl, pH 6.8, 10% SDS, 10% glycerol, 0.5 mmol/l phenylmethylsulfonyl fluoride, 10 μg/μl aprotinin, 10 μg/ml leupeptin, and 1 mmol/l sodium orthovanadate). Equivalent amounts of total protein lysate (50 μg) were separated using standard SDS-PAGE and transfer procedures. Blocked nitrocellulose membranes were probed overnight with either a total anti-p130CAS antibody (1:1,000, mouse antibody, Santa Cruz Biotechnology, Santa Cruz, CA) or phosphospecific anti-CAS antibody (rabbit, 1:500, Tyr410, Cell Signalling, Beverly, MA). Blots were incubated 1 h with the appropriate secondary antibody and then with ECL reagents (Amersham Life Sciences, Arlington Heights, IL) to visualize bands. Blots were reprobed with smooth muscle α-actin primary antibody (1:1,000; Oncogene Research Products, Boston, MA) to ensure equal protein loading. CAS phosphorylation is reported as a percentage of total CAS.
For coimmunoprecipitation of CAS and Src, 400 μg of total protein were incubated for 3 h with 8 μl of Src antibody (mouse, clone GD11, Upstate, Lake Place, NY) on ice and then tumbled overnight at 4°C with 60 μl of protein A/G agarose beads (Santa Cruz, CA). Beads were washed three times with protease-phosphatase inhibitor-enriched PBS and boiled in 2× sample buffer before the sample was halved and loaded onto 10% SDS gels. Once transferred, blots were probed with either the total p130CAS antibody or the Src antibody (antibodies described above).
Loading of oligodeoxynucleotides and organ culture.
Antisense oligodeoxynucleotides (ODNs) with the following sequence were designed to selectively suppress CAS expression in rat aorta: 5′-GTGACTGGATGCTGCTGG-3′. The sequence for sense ODNs was 5′-CCAGCAGCATCCAGTCAC-3′. According to sequence-matching results obtained from The National Center for Biotechnology Information, these sequences are not homologous to sequences of any other contractile proteins, cytoskeletal proteins, or actin-associated proteins. Phosphorothioate ODNs were synthesized and purified by Invitrogen (Carlsbad, CA).
The ODNs were introduced into rat aortic rings by the method of chemical loading (also referred to as reversible permeabilization; 7, 37). After determination of initial contractile response in the isolated tissue bath to 80 mmol/l KCl, each ring was placed into an autoclaved test tube and incubated successively in each of the following solutions: solution 1 (600 μl, at 4°C for 120 min) containing (mmol/l) 10 EGTA, 5 Na2ATP, 120 KCl, 2 MgCl2, and 20 N-tris-(hydroxymethyl) methyl-2-aminoethanesulfonic acid (TES); solution 2 (300 μl, at 4°C overnight) containing (mmol/l) 0.1 EGTA, 5 Na2ATP, 120 KCl, 2 MgCl2, and 20 TES, with 15 μmol/l antisense or sense ODNs; and solution 3 (600 μl, at 4°C for 30 min) containing (mmol/l) 0.1 EGTA, 5 Na2ATP, 120 KCl, 10 MgCl2, and 20 TES. Solutions 1-3 were maintained at pH 7.1 and aerated (100% O2). The tissues were then transferred to aerated (95% O2-5% CO2) 25-ml organ baths containing solution 4 (at 22°C for 60 min; in mmol/l: 110 NaCl, 3.4 KCl, 0.8 MgSO4, 25.8 NaHCO3, 1.2 KH2PO4, and 5.6 dextrose; pH = 7.4). After 30 min in solution 4, CaCl2 was added gradually to reach a final concentration of 2.4 mmol/l. The vessels were then incubated for 2 days in DMEM containing (in mmol/l) 5 Na2ATP, 100 U/ml penicillin, 100 μg/ml streptomycin, and 15 μmol/l antisense or 15 μmol/l sense ODNs, kept at 37°C, 5% CO2. The addition of ODNs in media was to compensate for the degradation of ODNs in cells (34, 38, 43). ODNs can be taken up into cells by endocytosis, maintaining appropriate intracellular ODN concentration (34, 38). Vessels were then placed in isolated tissue baths for addition of α-methyl-5-HT or KCl. Contraction of each ring after organ culture was normalized to force generated to KCl by each ring before organ culture.
Data are presented as means ± SE for the number of animals indicated in parentheses. When two groups were compared, the appropriate Students t-test was used. When three or more groups are compared, ANOVA followed by Student-Newman-Keuls post hoc was performed. A P value < 0.05 was considered statistically significant.
Latrunculin B inhibited 5-HT- and α-methyl-5-HT-induced contraction in normal rat thoracic aorta.
In an effort to connect 5-HT-induced contraction to actin reorganization, we inhibited actin polymerization in rat thoracic aorta by using latrunculin B and stimulated the aorta with 5-HT or α-methyl-5-HT. Aortas incubated with latrunculin B (10−7 and 10−6 mol/l) showed a significant concentration-dependent inhibition of 5-HT-induced contraction (43 ± 4% and 13 ± 5% of control maximum, respectively). A similar inhibition was observed when the aortas were stimulated with the 5-HT2 receptor agonist α-methyl-5-HT (35.9 ± 2.0% and 6.2 ± 3.0% of control maximum, respectively).
CAS protein immunostaining was observed in rat thoracic aorta.
In aortic sections incubated with a total anti-p130 CAS primary antibody, a dark staining was observed in smooth muscle cell layers (Fig. 1A); this same area stains intensely when a smooth muscle cell α-actin antibody is used (not shown). This dark staining was competed off when a five times excess of competing peptide (CP) was preincubated with primary antibody (Fig. 1B). Sections treated with no primary antibody were counterstained with hematoxylin and used as a control for the secondary antibody (Fig. 1C).
5-HT stimulated Src-dependent tyrosyl-phosphorylation of CAS during contraction of rat thoracic aorta.
Addition of 5-HT for 3 min significantly increased arterial tone (89 ± 2% phenylephrine contraction), and this was paralleled by an increase in tyrosine phosphorylation (Tyr410) of CAS over basal levels (Fig. 2A). 5-HT-stimulated contraction was abolished by addition of the 5-HT2A/2C receptor antagonist ketanserin (3 × 10−8 mol/l) and tyrosyl-phosphorylation of CAS significantly reduced (−ketanserin = 228 ± 38% vehicle; +ketanserin = 134 ± 33% vehicle). Similar results were obtained when the nonselective 5-HT2 receptor antagonist LY-53857 was used (data not shown). Src is one candidate for a kinase that tyrosyl-phosphorylates CAS in response to 5-HT. When Src was immunoprecipitated from aortic strips incubated with either vehicle or 5-HT in the isolated tissue bath, a greater amount of CAS was immunoprecipitated with Src in those aortas exposed to 5-HT (153 ± 33% vehicle; Fig. 2B). Additionally, 5-HT-induced contraction (Fig. 3A) and tyrosyl-phosphorylation of CAS (Fig. 3, B and C) was reduced by the Src inhibitor PP1 (10−5 mol/l).
Depletion of CAS decreased α-methyl-5-HT-induced contraction of rat thoracic aorta, whereas MLC phosphorylation remained normal.
CAS antisense or sense ODNs were introduced into aortic tissue. CAS protein expression was significantly lower in tissues treated with antisense ODNs compared with tissues treated with sense ODNs or not treated with ODNs (Fig. 4). Tissues were stimulated before and after incubation with ODNs, and isometric force was recorded. Isometric contraction to a maximal concentration of the 5-HT2 receptor agonist α-methyl 5-HT (10−5 mol/l) was significantly reduced in tissues in which CAS was depleted (Fig. 5A). However, contraction was normal in tissues in which CAS expression was not reduced (CAS S, no ODNs). Similarly, KCl (80 mmol/l)-induced active force in no ODN-treated tissues was 92.5 ± 6.6% of preincubation response to KCl; CAS sense-treated rings were 92.7 ± 6.9%; CAS antisense-treated rings were 26.3 ± 11.4% (P < 0.05, n = 6). Despite attenuation of α-methyl 5-HT-induced contraction in tissues with CAS depleted, downregulation of CAS did not affect the levels of MLC phosphorylation. α-Methyl-5-HT stimulated MLC phosphorylation, and the level of MLC phosphorylation was not significantly different in any group (Fig. 5B). Importantly, in normal tissues not treated with any kind of ODN, α-methyl-5-HT contraction was reduced by PP1 (10−5 mol/l; Fig. 5C), whereas KCl-induced contraction was not reduced by PP1 (Fig. 5D).
Pathway for CAS-mediated 5-HT-induced contraction.
The SH2 and SH3 binding sites of CAS suggest that CAS participates in the signal transduction cascades of multiple extracellular proteinaceous ligands (36). Our results indicate that 5-HT utilizes CAS in the signaling events leading to smooth muscle contraction, and thus understanding the role of CAS in vascular smooth muscle function is important. This represents a new signaling paradigm for 5-HT and the 5-HT2A receptor.
The vasoconstrictive effects of 5-HT are mediated through smooth muscle 5-HT2 receptors in most tissues (10, 46). In our studies, the connection between the 5-HT2 receptor and CAS was strengthened with the use of α-methyl-5-HT, a specific agonist of the 5-HT2 receptor. We demonstrated that 5-HT-induced CAS tyrosyl-phosphorylation and contraction could be reduced by the 5-HT2A/2C receptor antagonist ketanserin, and α-methyl-5-HT-induced contraction was reduced by CAS depletion. Protein kinases activated by 5-HT, such as c-Src, have been described as signaling mediators for smooth muscle 5-HT2A receptors (4, 47). The ability of the Src tyrosine kinase inhibitor PP1 to reduce 5-HT-induced contraction, α-methyl-5-HT-induced contraction, and 5-HT-stimulated CAS tyrosyl-phosphorylation suggests that Src is upstream of CAS and serves the 5-HT2A receptor. This is further supported by the finding that a greater magnitude of CAS tended to coimmunoprecipitate with Src in arteries exposed to 5-HT. Src activation has been implicated in the stimulation of Erk mitogen-activated protein kinase (MAPK) via the 5-HT2A receptor in the rat aorta (3), but it is not yet clear whether CAS is necessary for 5-HT-induced, Src-dependent activation of Erk MAPK.
5-HT, CAS, and Src.
Src participates in a number of diverse physiological processes stimulated by 5-HT. These include, but are not limited, to the upregulation of acetyl-CoA enzyme in human macrophages (35), inhibition of chloride/hydroxide anion exchange (33), growth of the biliary tree (17), mediation of Erk MAPK activation in multiple tissues (3, 49), contraction of various smooth muscle (1, 3, 40), mediation of the synergistic effect of 5-HT in mitogenesis caused by urotensin (44), generation of 5-HT-stimulated reactive oxygen species in CHO cells (21), and 5-HT-stimulated cell cycle progression (24). It is reasonable to suggest that some of these effects of Src may be through CAS-mediated events. In its substrate domain, CAS possesses 15 sites of YXXP motifs of potential for tyrosine phosphorylation by Src (4). The antibody used presently recognizes residue 410, whereas other studied residues include Tyr165 and Tyr249. Other proteins like FAK, which itself has a weak capacity to tyrosyl-phosphorylate CAS, can facilitate the engagement of Src with CAS and thus promote Src phosphorylation of CAS (31). We demonstrate presently that 5-HT activates tyrosyl-phosphorylation of CAS in a Src-dependent manner, and that contraction elicited by both 5-HT and the 5-HT2 receptor agonist α-methyl-5-HT is inhibited by PP1. A fair criticism of these contractile experiments is that the concentration of PP1 used (10−5 mol/l) is not selective among the Src family (such as the ubiquitously expressed Fyn and Yes) and thus our conclusion should be constrained to the family of Src tyrosine kinases. There is no evidence that PP1 has affinity for 5-HT receptors and thus it is unlikely that PP1 inhibited 5-HT-induced contraction through receptor antagonism.
CAS and contraction.
The specific pathway by which CAS mediates smooth muscle contraction remains unknown. However, the strong association of CAS with profilin and the actin cytoskeleton suggests that CAS contributes to force development through actin polymerization. Our results indirectly support this mechanism for contraction as latrunculin B, a marine macrolide isolated from the Red Sea sponge Latrunculia magnifica that disrups actin polymerization, reduced contraction. However, our results exclude CAS from the signaling pathway of contraction that leads to MLC phosphorylation, which regulates cross-bridge cycling between myosin and actin. CAS is an upstream mediator in the Rho kinase (22), phosphoinositide-3-kinase (28), and c-Jun NH2-terminal kinase (JNK) pathways (15, 16), and each of these pathways can modify arterial contraction.
The findings of this study demonstrate that 5-HT is one of the agonists that can tap into CAS-dependent contraction. The observation that CAS antisense oligodeoxynucleotides also reduced KCl-induced contraction is testament to the fact that CAS is a protein important to contraction. The lack of PP1 inhibition of KCl-induced contraction does not discount this idea, because multiple other kinases can affect CAS phosphorylation and function. We can state that KCl-induced contraction in the rat aorta is not dependent on Src but still dependent on CAS. This suggests that CAS regulation is complex and may be key to contractile function.
A particularly important effector of CAS may be rac-1 (9), a small GTP-binding protein critical to the function of vascular NAD(P)H oxidase. This enzyme that generates superoxide and other systems that produce free radicals, including hydrogen peroxide and the hydroxyl radical, may itself be able to influence the function of CAS. Gozin et al. (11) reported that the hydroxyl radical increased tyrosyl-phosphorylation of CAS in endothelial cells, an event that would facilitate the activation of signaling pathways. Moreover, Src and CAS mediation of JNK activation was driven by reactive oxygen species (50). Thus one can speculate that CAS may play an important mediator of arterial contractile function in conditions of elevated levels of reactive oxygen species or oxidative stress, such as is found in essential hypertension. This becomes more interesting with the recent report that nitric oxide can attenuate CAS tyrosyl-phosphorylation (5). This work, which includes the determination of the impact of the endothelial cell may have on CAS-dependent contraction, is future work.
Several limitations to this study need to be acknowledged. First, the aorta is not a resistance artery and thus no claims can specifically be made that the CAS signaling pathway as activated by 5-HT is relevant to total peripheral resistance. However, we have demonstrated that the 5-HT2A receptor continues to function in small resistance arteries from DOCA-salt hypertensive rats (45). Thus it is possible that the 5-HT2A receptor and, therefore, CAS play a role in small arterial response to 5-HT in hypertension. Moreover, the thoracic aorta was used because it is has minimal sympathetic innervation, allowing for a relatively nerve-free and simpler system.
Second, because a specific pharmacological inhibitor of CAS is not available, we turned to a different approach using ODNs to decrease CAS expression. Third, use of arteries from a CAS knockout mouse in which to examine 5-HT-induced contraction would be an excellent addition to these studies, but this is not yet possible because of the lethality of the removal of CAS in toto. Finally, we have not investigated direct association of 5-HT or CAS with actin because both of these interactions, while interesting and relevant to this story, were beyond the scope of the present study.
In summary, we demonstrate the presence and function of CAS in the rat aorta to mediate 5-HT-induced arterial contraction through the 5-HT2A receptor. This contraction occurred independently of changes in MLC phosphorylation but dependently on activation of the tyrosine kinase Src. CAS depletion also reduced KCl-induced contraction, suggesting a role for CAS in arterial contraction that is more important than previously appreciated. With the knowledge that CAS is utilized by multiple vasoactive hormones which include 5-HT, CAS presents itself as a candidate for generally modifying arterial contractility.
This study was funded by National Heart, Lung, and Blood Institute Grant HL-081115
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 by the American Physiological Society