Transglutaminase (TGase)-induced activation of small G proteins via 5-hydroxytryptamine (HT)2A receptor signaling leads to platelet aggregation (Cell115:851–862, 2003). We hypothesize that stimulation of 5-HT2A receptors in neurons activates TGase, resulting in transamidation of serotonin to a small G protein, Rac1, thereby constitutively activating Rac1. Using immunoprecipitation and immunoblotting, we show that, in rat cortical cell line A1A1v, serotonin increases TGase-catalyzed transamidation of Rac1. This transamidation occurs in both undifferentiated and differentiated cells. Treatment with a 5-HT2A/2C receptor agonist 2,5-dimethoxy-4-iodoamphetamine, but not the 5-HT1A receptor agonist 5-hydroxy-2-dipropylamino tetralin, increases transamidation of Rac1 by TGase. In A1A1v cells, 5-HT2A receptors mediate the transamidation reaction because expression of 5-HT2C receptors was not detectable and the selective 5-HT2A receptor antagonist blocked transamidation. Time course studies demonstrate that transamidation of Rac1 is significantly elevated after 5 and 15 min of serotonin treatment, but returns it to control levels after 30 min. The activity of Rac1 is also transiently increased following serotonin stimulation. Inhibition of TGase by cystamine or small interfering RNA reduces TGase modification of Rac1, and cystamine also prevents Rac1 activation. Serotonin itself is bound to Rac1 by TGase following 5-HT2A receptor stimulation as demonstrated by coimmunoprecipitation experiments and a dose-dependent decrease of serotonin-associated Rac1 by cystamine. These data support the hypothesis that Rac1 activity is transiently increased due to TGase-catalyzed transamidation of serotonin to Rac1 via stimulation of 5-HT2A receptors. Activation of Rac1 via TGase is a novel effector and second messenger of the 5-HT2A receptor-signaling cascade in neurons.
5-HT2A receptors are G protein-coupled receptors that are widely expressed in the brain, peripheral vasculature, platelets, and skeletal muscle. 5-HT2A receptors are involved in diverse physiological functions, from platelet aggregation to neuroendocrine release (Van de Kar et al., 2001; Walther et al., 2003). Pathophysiologically, 5-HT2A receptor signaling is implicated in several disorders, including hypertension, atherosclerosis, anxiety, and depression (Roth, 1994; Weisstaub et al., 2006). The classic signal transduction pathway of 5-HT2A receptors is Gq/11-coupled activation of phospholipase C (PLC) (Sanders-Bush et al., 2003), which hydrolyzes phosphatidylinositol 4,5-bisphosphate to inositol 1,4,5-trisphosphate (IP3) and diacylglycerol. IP3 mobilizes calcium from the endoplasmic reticulum and thereby increases intracellular Ca2+ (Julius, 1991).
TGases (EC 126.96.36.199) are a family of Ca2+-dependent enzymes. Once activated, TGases can catalyze the cross-linking of proteins via the γ-carboxamide group of peptide-bound glutamine and the ϵ-amino group of peptide-bound lysine, forming an interior intramolecular isodipeptide bond (Griffin et al., 2002). TGases can also covalently link biogenic amines and polyamines, such as serotonin or spermine, to a peptide-bound glutamine residue in a transamidation reaction (Folk et al., 1980; Dale et al., 2002). In platelets, activation of 5-HT2A receptors stimulates TGase-catalyzed transamidation of serotonin to small G proteins, such as RhoA and Rab4, rendering them constitutively active (Walther et al., 2003). Furthermore, neuronal differentiation of SH-SY5Y cells induced by retinoic acid increases the expression/activation of TGases, resulting in transamidation of putrescine to RhoA and activation of RhoA (Singh et al., 2003).
Rac1 belongs to the Rho family of small G proteins, a subgroup of the Ras superfamily. Members of the Rho family (e.g., RhoA, Rac1, and Cdc42) are associated with a wide array of cellular processes, such as cytoskeletal organization, vesicular transport, cell cycle progression, cell adhesion and migration, and neuronal differentiation, and a variety of enzymatic activities (Etienne-Manneville and Hall, 2002; Burridge and Wennerberg, 2004). Like other small G proteins, Rac1 functions as a molecular switch that cycles between two conformational states: a GTP-bound, active state and a GDP-bound, inactive state. The GTP/GDP cycling is controlled by many regulator molecules such as GTPase-activating proteins (GAPs), guanine nucleotide exchange factors (GEFs), and GDP dissociation inhibitors (GDIs) (Hakoshima et al., 2003). Post-translation modification of Rac1 in the regulator-interaction sites or in the GTP-hydrolyzing domain may have an enormous effect on its activation cycle. Bearing five glutamine residues in the amino acid sequence (Matos et al., 2000), Rac1 might serve as a suitable substrate of TGases, resulting in the transamidation of primary amines to Rac1 and rendering this small G protein constitutively active.
A1A1v cells derived from embryonic rat cortex (Scalzitti et al., 1998) provide an excellent experimental model to study 5-HT2A receptor signaling. In the present study, we use A1A1v cells to test whether 5-HT2A receptor stimulation increases TGase-catalyzed transamidation and activation of Rac1 because they endogenously express 5-HT2A/1A receptors, Gαq/11, PLC-β, TGase2, the serotonin transporter, and small G proteins, including Rac1, RhoA, and Cdc42.
Materials and Methods
Cell Culture. A1A1v cells, a rat cortical cell line, were grown on 100-mm2 plates coated with poly-l-ornithine (Sigma-Aldrich, St. Louis, MO) and maintained in 5% CO2 at 33°C, in Dulbecco's modified Eagle medium (Fisher Scientific, Pittsburgh, PA) containing 10% fetal bovine serum (Fisher Scientific). Cells were induced to differentiate by incubation at 37°C for 4 days. Before each experiment, cells were maintained in Dulbecco's modified Eagle medium with 10% charcoal-treated fetal bovine serum for 48 h. Charcoal adsorption removes most but not all serotonin in the medium. The maximal final concentration of serotonin in the medium is approximately 3 nM (Unsworth and Molinoff, 1992). Cells from passages 8 to 15 were used for all experiments.
Drugs. The following drugs were used in this study: (–)-1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane HCl (DOI) and serotonin (Sigma-Aldrich), (R)-(+)-8-hydroxy-2-dipropylamino tetralin hydrobromide (DPAT) (Tocris Cookson Inc., Ellisville, MO), 2-aminoethyl disulfide dihydrochloride (cystamine) (MP Biomedicals, Irvine, CA), and MDL 100907, a gift from Hoechst Marion Roussel Research Institute (Cincinnati, OH). Serotonin was dissolved in 10 μM HCl, and MDL 100907 was dissolved in a minimal volume of dimethyl sulfoxide and then diluted with saline. The remaining drugs were dissolved in purified water. All compounds were further diluted (at least 1:100) in cell culture media before they were applied to the cells and washed away before lysing the cells.
Immunoprecipitation of TGase-Modified Protein. A1A1v cells were harvested and lysed using lysis buffer A [25 mM Tris-HCl, pH 7.5, 250 mM NaCl, 5 mM EDTA, 1% Triton X-100, and 1:1000 protease inhibitor cocktail (Sigma-Aldrich) containing 104 μM 4-(2-aminoethyl)benzenesulfonyl fluoride, 0.08 μM aprotinin, 2 μM leupeptin, 4 μM bestatin, 1.5 μM pepstatin A, and 1.4 μM E-64]. Protein concentration was determined using the BCA Protein Assay kit (Pierce Chemical, Rockford, IL). Immunopurification of proteins containing TGase-catalyzed bonds was performed using 81D4 monoclonal antibody (mouse IgM) prebound to Sepharose beads (Covalab, Lyon, France) using a protocol developed by Covalab and as described previously (Norlund et al., 1999; Zainelli et al., 2005). The 81D4 antibody is well characterized, and it has been previously shown to be specific for the N-ϵ-(γ-l-glutamy)-l-lysine isopeptide and N-ϵ-(γ-l-glutamy)-l-lysine isopeptide cross-link generated by TGase (Sárvári et al., 2002; Thomas et al., 2004). The 81D4 antibody has been extensively used to demonstrate increases in TGase-catalyzed bonds (Citron et al., 2002; Junn et al., 2003; Andringa et al., 2004). The use of the 81D4 antibody in a competitive enzyme-linked immunosorbent assay was shown to result in isodipeptide cross-link measurements that correlate well to those measured by high-performance liquid chromatography analysis but provide more sensitivity than the high-performance liquid chromatography approach (Sárvári et al., 2002). Transfection of cells to overexpress TGases, including TGase1, 2, and 3, and a substrate protein such as mutant huntingtin protein results in increased presence of the isodipeptide cross-link in mutant huntingtin protein as demonstrated with immunoprecipitation with the 81D4 antibody (Zainelli et al., 2005).
In brief, 20 μl of Sepharose-81D4 beads was washed three times in TBS/0.1% Tween 20 with gentle shaking for 15 min, followed by adding 200 μg of cell lysate (1 μg/μl) to the washed beads and incubating for 2 h at 37°C. After incubation, the pellets were washed four times in TBS/0.1% Tween 20 for 15 min. Then, 20 μl of loading buffer (50 mM Tris-HCl, pH 6.8, 2% SDS, 0.1% bromphenol blue, 10% glycerol, and 5% β-mercaptoethanol) was added to the washed pellets followed by 5-min incubation at 90°C. The samples were then centrifuged at 9,000g for 2 min, and the supernatant was transferred and stored at –80°C until immunoblot analysis.
Immunoprecipitation of Rac1. Two hundred micrograms of protein from each sample was brought up to a total volume of 100 μl with immunoprecipitation buffer (50 mM Tris-HCl, pH 7.4, 10 mM EGTA, 100 mM NaCl, 0.5% Triton X-100, and 0.1% protease inhibitor cocktail), and then the samples were precleared using 10 μlof recombinant protein G (rProtein G) agarose (Invitrogen, Carlsbad, CA) for 1 h. The samples were centrifuged at 9000g for 10 min, and the supernatant was incubated for 1.5 h at 4°C with either 2 μgof Rac1 antibody (Millipore, Billerica, MA) or 2 μg of normal mouse IgG (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) as a control for nonspecific binding. The immunocomplexes were precipitated using 20 μl of rProtein G agarose at 4°C for 1 h. The agarose-immunocomplexes were washed three times in immunoprecipitation buffer and centrifuged after each wash at 100g for 3 min. After the last wash, bound proteins were eluted by adding 2× PAGE sample buffer and heating for 5 min at 90°C. The samples were centrifuged at 13,000g for 5 min, and the supernatant was transferred and stored at –80°C until immunoblot analysis.
Immunoblot. Immunoaffinity-purified proteins and cell lysates were separated on 12% SDS-polyacrylamide gels and then electrophoretically transferred to nitrocellulose membranes. Membranes were then incubated in blocking buffer (5% nonfat dry milk, 0.1% Tween 20, and 1× TBS) for 1 h at room temperature. Membranes were incubated overnight at 4°C with primary antibodies on a shaker. Primary antibodies [Millipore: anti-Na+/K+ ATPase, mouse IgG, 1:10,000; Abcam Inc., Cambridge, MA: anti-serotonin, rabbit IgG, 1:1000; BD Biosciences Pharmingen, San Jose, CA: anti-5-HT2C receptor, mouse IgG, 1:300; Cell Signaling Technology Inc., Danvers, MA: anti-phosphorylated ERK1/2 at residues Thr202/Tyr204, mouse IgG, 1:1000; anti-ERK1/2, rabbit IgG, 1:1000; MP Biomedicals, Aurora, OH: anti-actin, mouse IgG, 1:20,000; Covalab, Lyon, France: 81D4, mouse IgM, 1:500; NeoMarkers, Fremont, CA: TG100, mouse IgG, 1:100] were diluted in antibody buffer (1% nonfat dry milk, 0.1% Tween 20, and 1× TBS). The next day, membranes were washed with TBS/0.1% Tween 20, and then they were incubated with goat anti-mouse or goat anti-rabbit secondary antibody conjugated to horseradish peroxidase (Jackson ImmunoResearch Laboratories Inc., West Grove, PA) diluted in antibody buffer. Membranes were washed, and signal was detected using enhanced chemiluminescence Western blotting detection reagents (GE Healthcare, Chalfont, St. Giles, UK). Using Scion Image for Windows (Scion Corporation, Frederick, MD), immunoblots were quantified by calculating the integrated optical density (IOD) of each protein band on the film.
Expression and Purification of Glutathione Transferase-PAK1. The pGEX-2T plasmid was used to express Rac1 interactive domain of human PAK1 (residues 51–135) fused to GST. The BL21 (DE3) Escherichia coli (Invitrogen) were transformed with plasmid expressing GST-PAK1, and then they were grown overnight at 37°C on LB-ampicillin plates. A single colony was used to inoculate 25 ml of LB-ampicillin (100 μg/ml), which was shaken overnight at 37°C. Ten milliliters of the overnight culture was used to inoculate 500 ml of LB-ampicillin, which was grown for 3 h at 37°C while shaking. After induction with 0.4 mM isopropyl β-d-1-thiogalactopyranoside for 2 h at 30–32°C, bacteria were lysed in lysis buffer B (50 mM HEPES, pH 7.6, 1% Triton X-100, 100 mM NaCl, 5 mM MgCl2, 10% glycerol, 1 mM phenylmethylsulfonyl fluoride, and 0.1% protease inhibitor cocktail). The lysates were sonicated and centrifuged for 15 min at 9000g at 4°C. Then, the supernatant was added to glutathione-Sepharose 4B (GE Healthcare), and the mixture was incubated at 4°C for 2 h. The beads were then centrifuged at 4°C at 2000g and washed four times with lysis buffer B. The beads were resuspended in lysis buffer B, and then they were stored at –80°C for at most 2 weeks.
Rac1 Activity Assay. A1A1v cells were treated for 5 or 15 min with 14 μM serotonin or vehicle (10 μM HCl). The cells were harvested in lysis buffer C (50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 10 mM MgCl2, 1% Triton X-100, 0.1% SDS, 1 mM phenylmethylsulfonyl fluoride, and 0.1% protease inhibitor cocktail). Cell lysates were centrifuged at 14,500g for 20 min at 4°C, and 200 μl of the supernatant was incubated with 20 μl of GST-PAK1-Sepharose beads for 40 min at 4°C under constant agitation. The beads were washed three times with lysis buffer C, and 2× Laemmli sample buffer was added to the washed beads followed by 5-min incubation at 90°C. The samples were then centrifuged at 9000g for 2 min, and equivalent amounts of proteins in the supernatants were loaded on 12% SDS-PAGE as well as 20 μg of cell lysates (to detect total Rac1 protein levels), followed by immunoblot analysis as described above.
Small Interfering RNA. To induce TGase2 gene silencing, two siRNA duplexes to target the coding sequence of rat TGase2 mRNA were designed and synthesized by QIAGEN (Germantown, MD). The target sequence is 5′-AAGAGCGAGATGATCTGGAAT-3′ for siRNA1 and is 5′-AGAGCCAACCACCTGAACAAA-3′ for siRNA2. The sense strand of each siRNA was labeled at 3′ end with Alexa Fluor 488 to monitor transfection efficiency. At approximately 30 to 50% confluence, A1A1v cells were transfected with siRNA at a final concentration of 90 nM using Lipofectamine 2000 (QIAGEN) according to the manufacturer's instructions. Twenty-four hours after transfection, siRNA-lipid complexes were removed by changing medium. Transfection efficiency was assessed based on the percentage of fluorescent cells observed under Nikon Eclipse TE 2000U microscope. Forty-eight hours after transfection, cells were stimulated with DOI or vehicle for 5 min, and then cell lysates were collected to measure TGase2 expression and TGase-modified Rac1 by immunoblot and immunoprecipitation. Cells incubated with Lipofectamine 2000 alone were used as the nontransfected control.
Statistical Analyses. All data are presented as group mean ± S.E.M. and analyzed by one- or two-way ANOVA. Post hoc tests were conducted using Newman-Keuls multiple comparison test. SYSTAT 11 (Systat Software, Inc., San Jose, CA) and GraphPad Prism 5.0 (GraphPad Software Inc., San Diego, CA) were used for all statistical analyses. A probability level of p < 0.05 was considered to be statistically significant for all statistical tests.
Serotonin Treatment Induces Rac1 Transamidation in A1A1v Cells. To determine whether TGase-catalyzed transamidation of Rac1 is increased after serotonin treatment, we treated A1A1v cells with 14 μM serotonin for 5, 15, or 30 min. The transamidation of Rac1 is significantly elevated after 5 or 15 min of serotonin treatment by approximately 2 to 2.5-fold compared with vehicle (HCl)-treated cells (Fig. 1). However, there is no significant difference in the amount of TGase-modified Rac1 in cells treated with serotonin for 30 min compared with vehicle-treated cells (Fig. 1, A and B).
5-HT2A Receptor Stimulation Increases the Transamidation of Rac1. To explore the serotonin receptor specificity for induction of TGase-catalyzed transamidation, A1A1v cells were stimulated with 14 μM serotonin, 3 μM DOI (5-HT2A/2C receptor agonist), 10 nM DPAT (5-HT1A receptor agonist), or 10 μM HCl (vehicle) for 15 min. Treatment with serotonin or DOI significantly (p < 0.05) increases the amount of TGase-modified Rac1 2-fold compared with vehicle-treated cells, whereas treatment with DPAT has no effect on the amount of TGase-modified Rac1 compared with vehicle-treated cells (Fig. 2, A and B). Expression of 5-HT2A receptors in A1A1v cells has been previously confirmed using antisense oligodeoxynucleotide strategies and radioligand binding assays (Scalzitti et al., 1998). Here, 5-HT2C receptor expression in undifferentiated and differentiated A1A1v cells was examined by immunoblot analysis. Choroid plexus tissue from the fourth ventricle of a rat was used as a positive control, and HEK293 cell lysates were used as a negative control. We found that 5-HT2C receptors are expressed in rat choroid plexus, but 5-HT2C receptor expression was not detected in either HEK293 cells or undifferentiated or differentiated A1A1v cells (Fig. 2C). To exclude the possibility that the lack of involvement of the 5-HT1A receptors in Rac1 transamidation is not due to insufficient concentration of DPAT, we treated cells with 14 μM serotonin and increasing concentrations of DPAT (1, 10, and 100 nM) for 5 min, and then we detected ERK phosphorylation by immunoblot. Treatment with 14 μM serotonin and 10 or 100 nM DPAT significantly increased phosphorylation of ERK (Fig. 2, D and E), indicating that 10 nM DPAT is enough to activate 5-HT1A receptors in A1A1v cells. To further confirm that the effect of DOI on Rac1 transamidation in A1A1v cells is due to stimulation of the 5-HT2A receptors, cells were pretreated with 100 nM MDL 100907 (a selective 5-HT2A receptor antagonist) for 15 min, and then they were stimulated with DOI for 5 min. We found the pretreatment of MDL 100907 significantly reduced DOI-induced Rac1 transamidation without itself having any effect on Rac1 transamidation (Fig. 2F).
DOI Increases Rac1 Transamidation in Both Undifferentiated and Differentiated A1A1v Cells. After differentiation of A1A1v cells, the cytoskeleton undergoes a dramatic reorganization, and the cells acquire a neuronal-like cell shape with long processes similar to axons and dendrites, compared with undifferentiated cells (Fig. 3A). Rho family GTPases are critical regulators of the actin cytoskeleton organization (Etienne-Manneville and Hall, 2002; Burridge and Wennerberg, 2004). This prompted us to explore the effects of cell differentiation on Rac1 transamidation stimulated by 5-HT2A receptor activation. Stimulation of 5-HT2A receptors with DOI increased TGase-catalyzed transamidation of Rac1 in both undifferentiated and differentiated cells (Fig. 3B). Treatment with DOI in A1A1v cells significantly increased the amount of TGase-modified Rac1 by approximately 2 to 2.5-fold compared with vehicle-treated cells (p < 0.01). In vehicle-treated groups, the amount of TGase-modified Rac1 in differentiated cells was similar to the amount in undifferentiated cells, indicating that cell differentiation has no effect on basal level of Rac1 transamidation. Although the amount of TGase-modified Rac1 in DOI-treated differentiated cells was increased compared with DOI-treated undifferentiated cells, this increase was not significant, suggesting that cell differentiation also does not have a significant effect on DOI-induced transamidation of Rac1 (Fig. 3C).
Activity of Rac1 Is Transiently Increased following Serotonin Stimulation. Racl is GDP-bound in the inactive state and GTP-bound when activated. Activated Rac1 binds to its downstream effectors such as PAK1; therefore, a GST-PAK1 fusion protein purified from E. coli was used to determine whether serotonin stimulation increases the activated from of Rac1. A1A1v cells were treated with serotonin for 5 or 15 min, the cells were harvested, and cell lysates were incubated with GST-PAK1 prebound to glutathione-Sepharose beads. Then, equivalent amounts of the purified proteins were resolved by SDS-PAGE. Immunoblot analysis was performed using an antibody against Rac1. As shown in Fig. 4, the activity of Rac1 is increased by 80% after 5 min of serotonin treatment, but it returns to baseline after 15 min in the presence of serotonin, indicating that Rac1 becomes transiently activated after serotonin treatment in A1A1v cells. In addition, there is no significant change in the total amount of Rac1 in the cell lysate after serotonin treatment (Fig. 4A), suggesting that the activity increase is not due to synthesis of new Rac1.
TGase Inhibition Reduces Rac1 Transamidation and Activity. The activity of Rac1 is under the direct control of a large set of regulatory proteins. To exam whether the serotonin-stimulated increase of Rac1 activity is due to TGase-catalyzed transamidation, A1A1v cells were treated with increasing concentrations (0, 100, 500, and 1000 μM) of the TGase inhibitor cystamine for 1 h, followed by treatment with 14 μM serotonin for 5 min. Precipitation of TGase-modified proteins and the activated Rac1 was performed as described in previous experiments. Then, immunopurified proteins and activated Rac1 were examined on immunoblots using anti-Rac1 and 81D4 antibodies. We found that pretreatment with cystamine significantly decreases Rac1 transamidation (Fig. 5) and activity (Fig. 6) in a dose-dependent manner. Treatment with 100, 500, or 1000 μM cystamine decreases the amount of TGase-modified Rac1 by approximately 50, 70, or 80%, respectively, compared with untreated cells (Fig. 5, A and C). A similar dose-dependent decrease in TGase-modified Rac1 was also observed when cells were treated with serotonin for 15 min following cystamine inhibition (data not shown). Treatment with 500 or 1000 μM cystamine decreases the amount of activated Rac1 by approximately 20 or 40%, respectively, compared with untreated cells (Fig. 6, A and C). These results indicate that TGase-catalyzed transamidation of Rac1 contributes to the increase in Rac1 activity upon serotonin stimulation. The ubiquitously expressed Na+/K+ ATPase is a well established plasma membrane marker, and its function underlies essentially all of mammalian cell physiology (Kaplan, 2002). To verify that the dose-dependent decrease of Rac1 transamidation and activation was not due to cystamine-induced total Rac1 reduction or cellular toxicity, cell lysates from the same experiment were examined on Western blots with antibodies for Rac1 or Na+/K+ ATPase. There is no significant difference in total Rac1 or Na+/K+ ATPase levels between cystamine-treated and untreated cells, indicating that the concentrations of cystamine were not toxic to the cells (Figs. 5B and 6B).
Next, a second and more specific approach was used to determine the significance of TGase in Rac1 transamidation; we designed two siRNA duplexes to inhibit endogenous TGase expression. Several isoenzymes of TGase are found in the brain, of which TGase2 is the most abundant; therefore, siRNAs were developed to silence rat TGase2 gene. At 24 h after transfection, transfection efficiency of both siRNAs reached approximately 90%. Forty-eight hours after transfection, siRNA1 and siRNA2 transfection of A1A1v cells resulted in 95 and 65% down-regulation of TGase2 protein expression, respectively (data not shown). Reprobing of membrane with anti-Rac1 antibody revealed that neither siRNA changed Rac1 protein level, indicating no off-target effect on Rac1 (Fig. 7A). DOI-induced TGase-modification of Rac1 was significantly reduced by knocking down TGase2 expression with siRNA1 or siRNA2. Compared with DOI-stimulated nontransfected cells, siRNA1 and siRNA2 transfection caused 70 and 30% decreases in TGase-modified Rac1 responding to DOI, respectively (Fig. 7, B and C). These results confirm that 5-HT2A receptor-mediated Rac1 transamidation is dependent on TGase2 expression in A1A1v cells.
TGase-Mediated Transamidation of Serotonin to Rac1. TGase-modified Rac1 did not show a significant upward shift on immunoblots compared with native Rac1 in cell lysates (data not shown), indicating that the molecular weight of Rac1 does not significantly increase after modification by TGase. Therefore, a small amine such as serotonin is most likely incorporated into Rac1 upon stimulation of 5-HT2A receptors. To test this hypothesis, we stimulated A1A1v cells with serotonin or vehicle for 5 min, and then cells were harvested and cell lysates were used to immunoprecipitate Rac1 with an anti-Rac1 antibody or to immunoprecipitate TGase-modified proteins with the 81D4 antibody. The immunopurified proteins were examined on immunoblots using antibodies directed against serotonin. Immunoprecipitation of Rac1 and probing for serotonin reveals an association between Rac1 and serotonin in A1A1v cells (Fig. 8A). However, we were unable to detect serotonin in TGase-modified proteins (Fig. 8A). This may be due to a compromise of the serotonin antibody epitope during the immunoprecipitation of TGase-modified proteins, because this procedure uses higher temperatures and longer incubation times compared with the Rac1 immunoprecipitation procedure. Therefore, to confirm that serotonin is associated with Rac1 by a TGase-catalyzed covalent bond, we treated cells with increasing concentrations (0, 100, 500, and 1000 μM) of cystamine for 1 h followed by stimulating the cells with serotonin for 5 min. Then, we immunoprecipitated Rac1 and detected associated serotonin by immunoblot, as described above. We found that treatment with cystamine decreases the serotonin-associated Rac1 in a dose-dependent manner (Fig. 8, B and D), supporting the hypothesis that serotonin is incorporated into Rac1 by a TGase-catalyzed covalent bond. To compare the effects of different agonists on the incorporated serotonin, cells were exposed to either serotonin or DOI for 5 min. Then, Rac1 was immunoprecipitated, and the incorporated serotonin was detected by immunoblot. There was no significant difference in serotonin incorporation between serotonin and DOI-treated cells. Pretreatment with cystamine also inhibited DOI-induced serotonin incorporation (Fig. 9).
In platelets, 5-HT2A receptor activation causes TGase-catalyzed transamidation of RhoA and Rab4, leading to activation of these proteins and platelet aggregation (Walther et al., 2003). In Aplysia californica ganglia, serotonin treatment induces the activation of Cdc42 and its downstream effector PAK to regulate the actin cytoskeleton (Udo et al., 2005). In the present study, we extend these findings to a rat cortical cell model, and we find that 5-HT2A receptor stimulation increased TGase-catalyzed transamidation of Rac1 and Rac1 activity, both of which can be suppressed by TGase inhibition. In elucidating the mechanism further, we identify serotonin as an amine that becomes transamidated to Rac1 by TGase. Activation of Rac1 via TGase is a novel effector and second messenger of the 5-HT2A receptor signaling pathway in neurons.
To examine the serotonin receptor specificity on induction of Rac1 transamidation, we treated cells with different serotonin receptor agonists, and we found that the 5-HT2A/2C receptor agonist DOI, but not the 5-HT1A receptor agonist DPAT, induces a significant increase in TGase-modified Rac1, with a magnitude similar to that observed in serotonin-treated cells. Because there is no 5-HT2C receptor expression in A1A1v cells and MDL 100907 reversed the effect of DOI on Rac1 transamidation (Fig. 2), serotonin or DOI-induced Rac1 transamidation is selectively mediated by activation of 5-HT2A receptors in A1A1v cells.
Previous studies demonstrated that TGase-mediated transamidation of the Rho family of small G proteins blocked the GTP-hydrolyzing activity of these proteins, rendering these small GTPases constitutively active for their respective signaling pathways (Masuda et al., 2000; Walther et al., 2003). Based on these results, we hypothesized that 5-HT2A receptor-stimulated transamidation of Rac1 by TGase will result in prolonged activation of Rac1. We unexpectedly found a transient increase in both Rac1 activity and TGase-catalyzed transamidation that returned to baseline levels following continuous exposure to serotonin (Figs. 1 and 4). The transient response may due to selective degradation of transamidated Rac1. Mammalian cells contain multiple proteolytic systems for degradation of various classes of intracellular proteins. Most abnormal proteins (mutant or misfolded) are degraded in the ubiquitin-proteasome pathway. Rac1 is normally a stable protein, but activation by cytotoxic necrotizing factor (CNF)-1 in rat bladder carcinoma cells (804G) and human umbilical vein endothelial cells resulted in increased sensitization to ubiquitination and proteasomal degradation (Doye et al., 2002; Munro et al., 2004). Furthermore, dominant-positive forms of Rac1 seemed more susceptible to ubiquitin-mediated proteasomal degradation compared with both dominant-negative and wild-type Rac1 (Doye et al., 2002). Activated and transamidated Rac1 may be targeted for proteasomal degradation, resulting in transient activation of Rac1. However, there is no significant change in total Rac1 protein 15 min after serotonin stimulation (Fig. 2A), suggesting that activated or transamidated Rac1 may only account for small fraction of total Rac1.
In this study, we found differences in the time course for Rac1 transamidation and for its increased activity. After serotonin stimulation for 15 min, Rac1 activity is reduced to baseline (Fig. 4), whereas TGase-catalyzed transamidation of Rac1 still remains at a relatively high level (Fig. 1). These results suggest that transamidation may not only occur at the GTPase activity-related residues but also at residues that do not have an effect on Rac1 activity. Essentially, transamidation of Rac1 at some sites may induce activation of the small G proteins, whereas transamidation at other sites may not affect the GTPase activity of Rac1. If activated Rac1 is more sensitive to proteasomal degradation, the activated Rac1 may be degraded earlier than nonactivated Rac1 that is transamidated at sites not involved in activation.
Cystamine has been shown to decrease TGase activity and TGase-catalyzed N-ϵ-(γ-glutamyl)-l-lysine bonds in cultured cells and animal models of neurodegenerative diseases (Karpuj et al., 2002; Ientile et al., 2003; Zainelli et al., 2005). As a primary amine, cystamine is a potential substrate for TGase, and it acts as a competitive inhibitor for TGase by blocking access to the active site of the enzyme for the glutamine residues in proteins that would otherwise participate in forming N-ϵ-(γ-glutamyl)-l-lysine bonds (Lorand et al., 1979). In our study, pretreatment with cystamine reduced both serotonin-stimulated Rac1 transamidation (Fig. 5, A and C) and activation (Fig. 6, A and C) in a dose-dependent manner. Those observations were further supported by the use of TGase2-specific siRNAs, which significantly suppressed DOI-induced TGase-modification of Rac1 (Fig. 7, B and C). These results suggest that cystamine prevented Rac1 transamidation through TGase inhibition and that transamidation of Rac1 by TGase may contribute to the increase in Rac1 activity.
Post-translational modifications, such as transamidation and phosphorylation, of Rac1 may affect its ability to interact with regulatory proteins (e.g., GAPs, GEFs, and GDIs) or to convert GTP back to GDP, resulting in increased activation of Rac1 and stimulation of its signaling cascade in the cells. The Conserved Domain Database (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=cdd) provides an interactive tool to identify conserved domains present in protein sequences. We searched the Conserved Domain Database for GTP/Mg2+ binding sites, and GAP, GEF, and GDI interaction sites in the Rac1 sequence (PSSM-Id 57957) because they will be most likely to modify Rac1 activity after TGase-catalyzed transamidation at these sites. We identified two glutamine residues (Gln61 and Gln74) located within these domains. It has been reported that site-specific deamidation of a Gln residue in the Rho family G proteins (Gln61 in Rac and Cdc42, Gln63 in RhoA) by CNF-1 inhibits both intrinsic and GAPs-stimulated GTP hydrolysis activity, resulting in constitutive activation of these proteins (Flatau et al., 1997; Schmidt et al., 1997). CNF-1 has also been shown to possess in vitro TGase activity. In the presence of primary amines, RhoA is transamidated in vitro at Gln63 by CNF-1 and at positions 52, 63, and 136 by guinea pig liver TGase (Schmidt et al., 1998). Similarly, in addition to Gln61, which is critical in regulating Rac1 activity, the other four glutamine residues in Rac1 may also be modified by TGase. Therefore, it is not surprising that the same dosage of cystamine prevents Rac1 transamidation and its activation to different extents (Figs. 5C and 6C).
Walther et al. (2003) found that, in platelets, TGase covalently cross-links serotonin to RhoA or Rab4 at a position in the phosphate-binding site that is conserved in the sequences of all Ras-related small GTPases. It has also been shown that polyamines such as putrescine, spermidine, and spermine could be incorporated into Rac1 by bacterial TGase in vitro (Masuda et al., 2000). In our study, coimmunoprecipitation assays demonstrated that serotonin is associated with Rac1 in A1A1v cells after serotonin stimulation (Fig. 8A). To explore the association mechanism further, we inhibited TGase by cystamine, and we found a reduction of serotonin-associated Rac1 with increasing concentrations of cystamine (Fig. 8, B and D), indicating that serotonin is incorporated into Rac1 by a TGase-catalyzed bond. Because both serotonin and DOI stimulation can lead to serotonin incorporation to similar levels (Fig. 9), the serotonin bound to 5-HT2A receptors is not likely the source of Rac1-incorporated serotonin; the serotonin bound to Rac1 may originate from endogenously synthesized serotonin or serotonin transported into the cell from serum in the cell culture media.
The first signal transduction mechanism identified for the 5-HT2A receptor was Gq/11-mediated activation of PLC, leading to increased accumulation of IP3 and an increase in intracellular Ca2+ (Boess and Martin, 1994). In addition to activation of PLC, extensive evidence suggests that 5-HT2A receptors couple to other effector pathways, such as phospholipase A2/arachidonic acid cascade (Berg et al., 1994), mitogen-activated protein kinase signaling (Watts, 1998) and phospholipase D/protein kinase C pathway (Mitchell et al., 1998). Conversely, TGases are subjected to transcriptional regulation by retinoic acid and steroid hormones (Fujimoto et al., 1996; Ou et al., 2000), and they require the binding of Ca2+ for their activity (Burgoyne and Weiss, 2001). Evidence indicates that TGase activity can be significantly enhanced in response to increased intracellular Ca2+ through IP3 generation (Zhang et al., 1998). In A. californica, serotonin-mediated activation of Cdc42 was dependent on PLC and phosphatidylinositol 3-kinase (Udo et al., 2005). In retinoic acid-induced neuronal differentiation of SH-SY5Y cells, TGase-catalyzed transamidation is required for activation of RhoA (Singh et al., 2003), whereas activation of Rac1 is mediated by phosphatidylinositol 3-kinase in a transamidation-independent manner (Pan et al., 2005). Further studies are needed to elucidate the underlying molecular mechanisms by which the 5-HT2A receptor signaling regulates TGase-catalyzed activation of Rac1.
The present study provides the first evidence in neurons that stimulation of 5-HT2A receptors induces an increase in TGase-catalyzed transamidation of serotonin into Rac1 and constitutive activation of Rac1. Further studies using alternative approaches to measure Rac1 transamidation, such as mass spectroscopy, are needed to confirm this hypothesis. Rac1 activation via 5-HT2A receptor stimulation may have an important functional impact given the plethora of pathways that use this small G protein. For example, Rac1 is best known as a regulator for the assembly of the actin cytoskeleton, thereby playing a role in neurite outgrowth and neuronal differentiation. We demonstrate that stimulation of 5-HT2A receptor can increase the transamidation of Rac1 in both undifferentiated and differentiated A1A1v cells, suggesting that transamidation and constitutive activation of this signal transducer are not altered by neuronal differentiation of cells. However, the effect of 5-HT2A receptor-mediated activation of Rac1 on cytoskeleton organization, cell cycle progression, transcriptional activation, or other crucial cellular functions in neurons has yet to be explored.
We thank Dr. William Clarke and Kelly Berg (University of Texas Health Science Center, San Antonio, TX) for providing A1A1v cells.
This work was supported by United States Public Health Service Grant MH068612.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: 5-HT, 5-hydroxytryptamine (serotonin); PLC, phospholipase C; IP3, inositol 1,4,5-trisphosphate; TGase, transglutaminase; GAP, GTPase-activating protein; GEF, guanine nucleotide exchange factor; GDI, GDP dissociation inhibitor; DOI, 2,5-dimethoxy-4-iodoamphetamine; DPAT, 5-hydroxy-2-dipropylamino tetralin; MDL 100907, α-(2,3-dimethoxyphenyl)-1-[2-(4-fluorophenyl)ethyl]-4-piperidine methanol; E-64, N-(trans-epoxysuccinyl)-l-leucine 4-guanidinobutylamide; TBS, Tris-buffered saline; PAGE, polyacrylamide gel electrophoresis; ERK, extracellular signal-regulated kinase; IOD, integrated optical density; GST, glutathione transferase; siRNA, small interfering RNA; ANOVA, analysis of variance; HEK, human embryonic kidney; CNF, cytotoxic necrotizing factor.
- The American Society for Pharmacology and Experimental Therapeutics