|Home | About | Journals | Submit | Contact Us | Français|
Thrombin, a G protein-coupled receptor agonist, induced a biphasic expression of cyclin D1 in primary vascular smooth muscle cells. Although both phases of cyclin D1 expression require binding of the newly identified cooperative complex, NFATc1·STAT-3, to its promoter, the second phase, which is more robust, depends on NFATc1-mediated recruitment of p300 onto the complex and the subsequent acetylation of STAT-3. In addition, STAT-3 is tyrosine-phosphorylated in a biphasic manner, and the late phase requires NFATc1-mediated p300-dependent acetylation. Furthermore, interference with acetylation of STAT-3 by overexpression of acetylation null STAT-3 mutant led to the loss of the late phase of cyclin D1 expression. EMSA analysis and reporter gene assays revealed that NFATc1·STAT-3 complex binding to the cyclin D1 promoter led to an enhanceosome formation and facilitated cyclin D1 expression. In the early phase of its expression, cyclin D1 is localized mostly in the cytoplasm and influenced cell migration. However, during the late and robust phase of its expression, cyclin D1 is translocated to the nucleus and directed cell proliferation. Together, these results demonstrate for the first time that the dual function of cyclin D1 in cell migration and proliferation is temperospatially separated by its biphasic expression, which is mediated by cooperative interactions between NFATc1 and STAT-3.
Vascular smooth muscle cells exist in G0 phase in the healthy vasculature. They exit the G0 phase in response to growth factors, mechanical injury, or inflammation, enter the cell cycle, and migrate to a new space (1, 2). The transition of cells from G0 to S phase is the most regulated process in the cell cycle (3). Among the several molecules involved in the regulation of G0 exit, cyclin D1/CDK4 or cyclin D1/CDK6 are the most important gatekeepers (4), and any dysfunction in their expression or function has been known to result in cancer (5–7). Deregulated cyclin D1 expression is well documented in breast, colon, and prostate cancers (8–10). However, the correlation of cyclin D1 overexpression with E2F target gene regulation linking to tumor progression has not been identified. Toward this end, in recent years the work from many laboratories reported that in addition to its function in cell cycle progression, cyclin D1 plays a role in cell migration (11–13).
Cyclin D1 expression is tightly regulated by several factors (14, 15), and it is rapidly degraded by the ubiquitin-dependent proteasome pathway (16). Recent studies in our laboratory have shown that cyclin D1 expression is regulated by the NFATs2 and STATs family of transcription factors (17–19). NFATs are a family of transcription factors that belong to the Rel/NF-κB group and consists of five members, namely NFATc1, NFATc2, NFATc3, NFATc4, and NFAT5 (20). NFATs contain a regulatory region in the N terminus and a Rel homology comprising the DNA binding domain in the C terminus, and they are activated by calcium-calcineurin-mediated dephosphorylation with the exception of NFAT5 (21, 22). In addition to their well studied function in the regulation of cytokine gene expression (23), NFATs have been shown to play a role in cardiac development, skeletal muscle differentiation, and maintenance of neuronal plasticity (24–27). Toward delineating the mechanisms underlying the NFAT-mediated VSMC growth and migration, we have previously reported that NFATc1 mediates cyclin A2 and cyclin D1 expression in human aortic smooth muscle cells (17, 28).
STATs are a family of transcription factors that consist of seven members. They contain a dimerization domain, a coiled-coil domain, a DNA-binding domain, a Src homology 2 domain, and a conserved single tyrosine residue whose phosphorylation is required for its activation in the N-terminal region, whereas a transcriptional activation domain is present in the C-terminal region (29). Interestingly, the STATs are the only characterized transcription factors thus far that are activated by tyrosine phosphorylation, and this post-translational modification is essential for their Src homology 2-mediated homo- or heterodimer formation (30, 31). Thus, it is interesting to note that although both NFATc1 and STAT-3 belong to two different groups of transcription factors and are activated by opposite post-translational modifications, both target the same gene, cyclin D1, in mediating cell proliferation and migration (17–19, 32, 33).
Because cyclin D1 plays a permissive role in cell cycle progression, it is not surprising that cells employ multiple mechanisms in its regulation in response to a plethora of stimuli (14, 15, 18, 34). However, what is not known is whether there is any cross-talk among the multiple transcription factors that are activated by various signaling mechanisms in the induction of cyclin D1 expression in mediating cell cycle progression and cell migration in response to an agonist. Toward this end, we report here for the first time that NFATc1 and STAT-3, members of two different families of transcription factors, exist as a heterodimer in quiescent cells and in response to thrombin bind to the cyclin D1 promoter in a cooperative and biphasic manner. In addition, NFATc1 via recruiting p300 facilitates STAT-3 acetylation. In response to thrombin, STAT-3 is tyrosine-phosphorylated in a biphasic manner in which the late phase phosphorylation requires NFATc1-dependent p300-mediated acetylation. Furthermore, the cooperative interaction between these two transcription factors leads to a biphasic expression of cyclin D1, with the first phase influencing cell migration and the second phase directing cell cycle progression.
Cyclosporin A (A-195) was bought from Biomol (Plymouth Meeting, PA). Recombinant human thrombin (194083) was from MP Biomedicals Inc. (Minneapolis, MN). Matrigel was obtained from BD Biosciences. Anti-NFATc1 (SC-13033), anti-p300 (SC-48343), anti-p53 (SC-6243), anti-STAT-3 (SC-482), anti-DsRed2 (SC-101526), anti-GFP (SC-9996), and anti-β-tubulin (SC-9104) antibodies, normal rabbit serum (SC-2338), and normal mouse serum (SC-45051) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Anti-cyclin D1 antibodies (RB-010-P) were bought from NeoMarkers (Fremont, CA). Anti-NFATc1 antibodies (MA3–024) were from Affinity BioReagents (Golden, CO). Anti-pSTAT-3 (Tyr-705) and anti-acetyl STAT-3 (Lys-685) antibodies were obtained from Cell Signaling Technology (Beverly, MA). Culture inserts for wound healing assays and ibiTreat-35mm μ-dishes with Cell Culture Inserts (80206) were purchased from ibidi LLC (Verona, WI). Hoechst 33342 (10 mg/ml) solution (H3570), Alexa Fluor 488-conjugated goat anti-rabbit secondary antibodies, Lipofectamine 2000 reagent, NuPAGE Novex 3–8% gradient Tris acetate midi gels, and BLOCK-iT polymerase II miR RNAi expression vector kit with EmGFP and Pfx50 Taq polymerase were from Invitrogen. pGEM-T vector was purchased from Promega (Madison, WI). pTRE-Tight, pTet-On Advanced and pDsRed2-N1 vectors were obtained from Clontech. QuickChange site-directed mutagenesis kit was bought from Stratagene (La Jolla, CA). RNase A was purchased from Roche Applied Science. T4 polynucleotide kinase was obtained from New England Biolabs (Ipswich, MA). [γ-32P]ATP (specific activity 3000 Ci/mmol) was from MP Biomedicals (Irvine, CA). [3H]Thymidine (specific activity 20 Ci/mmol) was obtained from PerkinElmer Life Sciences. Protein-A-Sepharose (CL-4B) was purchased from Amersham Biosciences. Rat NFATc1 siRNA (sense, 5′-CUA CUA AUG AGC AGC GAA AUU-3′; antisense, 5′-UUU CGC UGC UCA UUA GUA GUU-3′), rat cyclin D1 siRNA (L-089285-01-0010), rat p300 siRNA (L-101658-01-0010), and scrambled control siRNA (D-001810-10-20) were obtained from Dharmacon (Lafayette, CO). All the primers and oligonucleotides were synthesized by IDT (Coralville, IA).
Construction of Ad-GFP, Ad-GFPVIVIT, and Ad-dnSTAT-3 was described previously (35, 36). Expression vector for the acetyl mutant of STAT-3, pcDNA3-mSTAT-3(K685R), was generated by site-directed mutagenesis (37). Cyclin D1 miRNA coding sequence was cloned under a constitutively active CMV promoter as well as tetracycline-responsive TRE-Tight promoter. Initially, the scrambled miRNA coding sequence consisting of XmaI and HindIII sites (sense, 5′-TGCTGCCCGGGGCCCTGCAGAAAGCGTGATAAGTTTTGGCCACTGACTGACTTATCACGCTTTCTGCAGGGCAAGCTT-3′, antisense, 5′-CCTGAAGCTTGCCCTGCAGAAAGCGTGATAAGTCAGTCAGTGGCCAAAACTTATCACGCTTTCTGCAGGGCCCCGGGC-3′) was annealed and ligated with prelinearized pcDNA6-EmGFP-miR vector to generate pcDNA6-EmGFP-Scr miR. EmGFP-Scr miR cassette was PCR-amplified using primers forward 5′-TTTAAGGTACCAACCATGGTGAGCAAGGGCGAGGAGC-3′ and reverse 5′-ACTTTGTACAAGAAAGCTGGGTCTAGATATC-3′ and pcDNA6-EmGFP-Scr miR as a template. The PCR product was digested with KpnI and EcoRV and cloned into pTRE-Tight vector at KpnI/EcoRV site to generate pTRE-Tight-EmGFP-Scr miR vector. EmGFP coding sequence from pcDNA6-EmGFP-Scr miR was removed by digesting with DraI and SalI. DsRed2 coding sequence was PCR-amplified from pDsRed2-N1 using primers forward 5′-GCAGGCTTTAAAGCCACCATGGCCTCCTCCGAGAACGTCA-3′ and reverse 5′-CCACTGGTCGACGTGCTTAGCCTACAGGAACAGGTGGTG-3′. The PCR product was digested with DraI and SalI and cloned into pcDNA6-EmGFP-Scr miR at the same sites to generate pcDNA6-DsRed-Scr miR. Cyclin D1 miRNA coding sequence (sense, 5′-CCGGGACTTGAAGTAAGAAACGGAGGGTTTTGGCCACTGACTGACCCTCCGTTTTACTTCAAGTA-3′, and antisense, 5′-AGCTTACTTGAAGTAAAACGGAGGGTCAGTCAGTGGCCAAAACCCTCCGTTTCTTACTTCAAGTC-3′) was annealed and ligated with XmaI/HindIII-digested pcDNA6-DsRed-Scr miR or pTRE-Tight-EmGFP-Scr miR to generate pcDNA6-DsRed-CCND1 miR or pTRE-Tight-EmGFP-CCND1 miR, respectively. The nucleotide sequence of each construct was verified by DNA sequencing.
VSMCs were isolated from male rats and subcultured as described previously (38). VSMCs were used between 4 and 12 passages.
VSMC migration with and without an appropriate treatment was measured by modified Boyden chamber method as described previously (39). Wherever adenoviral vectors were used, cells were transduced with the respective adenovirus at an m.o.i. of 40, growth-arrested, and subjected to migration assay. Cell migration was expressed as the number of migrated cells per field. In the case of wound healing assays, 80 μl of DMEM containing 2 × 105 cells was added in each chamber of ibidi culture inserts, and after 12 h, cells in the left chamber were transfected with either pcDNA6-DsRed-Scr miR or pcDNA6-DsRed-CCND1 miR vectors, and the cells in the right chamber were transfected with either pTRE-Tight-EmGFP-Scr miR or pTRE-Tight-EmGFP-CCND1 miR vectors. Following a 24-h quiescent period, the culture inserts were removed using sterile tweezers, and 2 ml of DMEM containing 5 mm hydroxyurea was added to the culture dish. Cells in the right and left chambers were treated with and without thrombin (0.5 unit/ml) in the presence and absence of doxycycline (1 μg/ml) for 6–8 h. Migrated cells were observed under a Zeiss inverted microscope (Zeiss AxioObserver Z1, type, plan-Apochromat; magnification ×10/0.45 NA), and fluorescence images were captured with a Zeiss AxioCam MRm camera using the microscope operating and image analysis software AxioVision Version 4.7.2 (Carl Zeiss Imaging Solutions GmbH).
VSMC DNA synthesis with and without an appropriate treatment was measured by pulse-labeling cells for the last 42 h of the 48-h incubation period with 2 μCi/ml [3H]thymidine as described previously (17).
VSMCs with and without an appropriate treatment were harvested, and cell extracts were prepared. An equal amount of protein from control and treatment samples was analyzed by Western blotting for the protein of investigation using its specific antibodies as described previously (28).
VSMCs were transfected with scrambled or specific siRNA molecules at a final concentration of 100 nm using Lipofectamine 2000 transfection reagent according to the manufacturer's instructions. Wherever adenovirus vectors were used, cells were transduced with adenovirus harboring GFP or a target molecule at 40 m.o.i. overnight in complete medium. After transfections or transductions, cells were growth-arrested for 48 h and used as needed.
The rat cyclin D1 promoter sequence encompassing −1234 to +37 nt relative to the transcription start site was amplified from rat genomic DNA using forward primer rCCND1p-F1, 5′-TCATCTGGTACCAGACCAAGGTGGTGGAACCTC-3′, incorporating a KpnI restriction enzyme site at the 5′-end and a reverse primer rCCND1p-R, 5′-AGTCTACTCGAGCTCTCTGCTACTGCGCCAAC-3′, incorporating an XhoI restriction enzyme site at the 5′-end. The PCR product was digested with KpnI and XhoI, and the released fragment was cloned into KpnI and XhoI sites of the pGL3 basic vector (Promega) to yield pGL3-rCCND1p-(1.3 kb)-Luc. To generate 5′-truncated promoter fragments starting from −664 nt (pGL3-rCCND1p-(0.7 kb)-Luc), −378 nt (pGL3-rCCND1p-(0.4 kb)-Luc), and −234 nt (pGL3-rCCND1p-(0.27 kb)-Luc) to +37, forward primers 5′-TCATCTGGTACCTTTGAGCGAGTACATGCCAGG-3′, 5′-TCATCTGGTACCTTCTTTCTTGGCTTGCGTGTGG-3′, and 5′-TCATCTGGTACCAGCTTAGAGAGAAGCAGTCC-3 and reverse primer rCCND1p-R, 5′-AGTCTACTCGAGCTCTCTGCTACTGCGCCAAC-3′, were used in the PCR. The PCR products were digested with KpnI and XhoI and cloned into KpnI and XhoI sites of the pGL3 basic vector. Site-directed mutations within the NFAT-binding elements at −742 (pGL3-rCCND1p-(1.3 kb)-NFAT7m-Luc), −890 (pGL3-rCCND1p-(1.3 kb)-NFAT8m-Luc), −1037 (pGL3-rCCND1p-(1.3 kb)-NFAT9m-Luc), and STAT binding element at −970 (pGL3-rCCND1p-(1.3 kb)-STATm-Luc) were introduced using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer's instructions using the following primers: forward 5′-AGGAGGAATAGATGAAAAAAATAATGGCCACCAT-3′ and reverse 5′-ATGGTGGCCATTATTTTTTTCATCTATTCCTCCT-3′; forward 5′-GCCTATCGTGTCTCAATTTTTAATTCAAAGATTT-3′ and reverse 5′-AAATCTTTGAATTAAAAATTGAGACACGATAGGC-3′; forward 5′-CAGCTGCAATTCTAAAAATGGGGGAACACAAAAC-3′ and reverse 5′-GTTTTGTGTTCCCCCATTTTTAGAATTGCAGCTG-3′; forward 5′-TCCGGTGGTCTGGTGCCTGCAAGGGCAAC-3′ and reverse 5′-GTTGCCCTTGCAGGCACCAGACCACCGGA-3′, respectively. The boldface letters indicate the mutated bases. Similarly, to generate cyclin D1 promoter-luciferase constructs containing NFAT- and STAT-binding elements transposed with each other (TR) or positioned closer to each other (NX), pGL3-rCCND1p-(1.3 kb)-Luc was used as a template along with the following primers: forward 5′-CCAGCTGCAATTCTttcctggaaTGGGGGAACACAAAACACACCCCCAACGAAGCCAATCGGGAAGCTTCCGGTGGTCTGGaaaggGGGCAACTAATACT-3′ and reverse 5′-AGTATTAGTTGCCCcctttCCAGACCACCGGAAGCTTCCCGATTGGCTTCGTTGGGGGTGTGTTTTGTGTTCCCCCAttccaggaaAGAATTGCAGCTGG-3′; forward 5′-CCAGCTGCAATTCTTGGGGGAACACAAAACACACCCCCAACaaaggGAAGttcctggaaCCAATCGGGAAGCTTCCGGTGGTCTGGGGGCAACTAATACT-3′ and reverse 5′-AGTATTAGTTGCCCCCAGACCACCGGAAGCTTCCCGATTGGttccaggaaCTTCcctttGTTGGGGGTGTGTTTTGTGTTCCCCCAAGAATTGCAGCTGG-3′, respectively. The nucleotide sequence of each construct was verified by DNA sequencing.
VSMCs were transfected with pGL3 basic vector or cyclin D1 promoter-luciferase constructs using Lipofectamine 2000 reagent. After growth arrest in serum-free medium for 48 h, the cells were treated with and without thrombin (0.5 unit/ml) for 8 h, washed once with cold PBS, and lysed in 200 μl of lysis buffer. The cell extracts were collected into microcentrifuge tubes and centrifuged at 12,000 × g for 2 min at 4 °C. The supernatants were assayed for luciferase activity using the Luciferase Assay System (Promega) and a single tube luminometer (TD20/20 Turner Designs, Sunnyvale, CA). The values are expressed as relative luciferase units.
Nuclear extracts of VSMCs with and without an appropriate treatment were prepared and analyzed for DNA binding activity using 32P-labeled double-stranded oligonucleotide probes as described previously (17). Double-stranded oligonucleotides that encompass NFAT-binding elements located between −752 and −735 nt (GATGAAGGAAATAATGGC), −900 and −883 nt (TCTCAACCTTTCCTTCAA), −1047 and −1030 nt (AATTCTAAAGGTGGGGGA), and STAT-binding element located between −983 and −966 nt (TCTGGTTCCTGGAAGGGC) of rat cyclin D1 promoter were used as 32P-labeled probes to measure NFAT- and STAT-DNA binding activities. In the case of gel supershift assays, after incubation of the nuclear extract with the radiolabeled probe for 30 min at 4 °C, appropriate antibodies were added, and incubation was continued for another 60 min, and the complexes were analyzed by EMSA. To study cooperative binding of NFATc1 and STAT-3 by EMSA, a large probe spanning from −1064 to −948 nt was generated by PCR using the following primers: forward 5′-AAGGCTGCCCCAGCTGCAAT-3′ and reverse 5′-TCGCTGCAAGTATTAGTTGC-3′ and pGL3-rCCND1p-(1.3 kb)-Luc as a template. To generate probes containing NFAT- and STAT-binding elements transposed with each other (TR) or moved closer to each other (NX), pGL3-rCCND1p-(1.3 kb)-Luc-TR or pGL3-rCCND1p-(1.3 kb)-Luc-NX were used as templates, respectively. The PCR products were gel-purified, radiolabeled, and used as probes.
ChIP assay was performed on VSMCs by using a kit following the supplier's protocol (Upstate Biotechnology Inc., Lake Placid, NY). NFATc1 or STAT-3·DNA complexes were immunoprecipitated using anti-NFATc1 or STAT-3 antibodies. Preimmune mouse or rabbit serum was used as a negative control. The immunoprecipitated DNA was uncross-linked, subjected to proteinase K digestion, purified using QIAquick columns (catalog no. 28104, Qiagen, Valencia, CA), and used as a template for PCR amplification. To study NFATc1 binding, primers, forward 5′-TAA ATA TCA CCT TAT CGG CTC ACA-3′ and reverse 5′- TCA GCA ACA GCT CAA GAT GG-3′, that would amplify 418 bp encompassing the −742, −890, and −1037 NFAT-binding elements were used. To study STAT-3 binding, primers, forward 5′-CAA CGA AGC CAA TCG GGA AGC TTC-3′ and reverse 5′-CCT CTG GTA TCC CCC TCC TCC ACT-3′, that would amplify 174 bp encompassing the STAT-binding element at −970 were used. The resulting PCR products were resolved on 1.8% acrylamide gels and stained with ethidium bromide, and images were taken using the AlphaEase digital imaging system (Alpha Innotech Corp.).
VSMCs were grown on cell culture grade coverslips to 40% confluence, quiesced for 48 h, and treated with and without thrombin (0.5 unit/ml) for various time periods. Cells were then washed with PBS, fixed with 3% paraformaldehyde for 10 min at 37 °C, blocked, and permeabilized in PBS containing 3% BSA and 0.5% Triton X-100 for 15 min at room temperature. The permeabilized cells were incubated first with anti-cyclin D1 antibodies (1:500 dilution in PBS) followed by incubation with Alexa Fluor 488-conjugated goat anti-rabbit secondary antibodies, counterstained with Hoechst 33342 (1:3000 dilution in PBS) for 1 min at room temperature, and mounted onto glass slides with Prolong Gold antifade mounting medium. Fluorescence images of cells were captured using an inverted Zeiss fluorescence microscope (AxioObserver Z1) via a ×40/0.6 NA objective and AxioCam MRm camera without any enhancements.
All the experiments were repeated three times, and data are presented as means ± S.D. The treatment effects were analyzed by Student's t test, and the p values <0.05 were considered statistically significant. In the case of EMSA, ChIP assay, Western blotting, and fluorescence imaging, one representative set of data is shown.
VSMC growth and migration play a major role in obstructive vascular diseases such as atherosclerosis and restenosis (40, 41). Many chemical cues that are generated at the site of vascular injury have been linked to the pathogenesis of these vascular wall diseases (42–44). Toward understanding the mechanisms of vascular diseases, we have previously reported that NFATc1 mediates PDGF-BB-induced cyclin A2 and cyclin D1 expression facilitating VSMC growth and migration (17, 28). Because NFATs are activated by both receptor tyrosine kinase and G protein-coupled receptor agonists (45, 46) and NFATc1 mediates PDGF-BB-induced cyclin A2 and cyclin D1 expression, we wanted to find whether NFATc1 also plays a role in thrombin-induced cyclin D1 expression. Thrombin (0.5 unit/ml) induced the expression of cyclin D1 in a biphasic manner with a 2-fold increase at 1 h and a 3-fold increase at 8 h of treatment at mRNA and protein levels (Fig. 1, A and B). Based on these observations, we next examined the role of cyclin D1 in thrombin-induced VSMC migration and DNA synthesis. Down-regulation of cyclin D1 levels by its siRNA inhibited thrombin-induced VSMC migration and DNA synthesis (Fig. 1, C and D). To test the role of NFATs in thrombin-induced cyclin D1 expression, we applied a decoy peptide-based approach to intervene with NFAT activation. Adenoviral vector-mediated expression of GFPVIVIT, a decoy pentapeptide fused to GFP (47), significantly suppressed thrombin-induced cyclin D1 expression (Fig. 2A).
To identify the specific NFAT involved in thrombin-induced cyclin D1 expression, we next used an siRNA approach. Depletion of NFATc1 levels by its siRNA also suppressed thrombin-induced cyclin D1 expression (Fig. 2B). These results indicate that NFATc1 mediates thrombin-induced cyclin D1 expression. To understand the role of NFATs in thrombin-induced cell migration and proliferation, VSMCs that were transduced with either Ad-GFP or Ad-GFPVIVIT and quiesced were treated with and without thrombin (0.5 unit/ml), and cell migration and DNA synthesis were measured. Thrombin induced VSMC migration by about 2-fold as measured by a modified Boyden chamber method, and overexpression of GFPVIVIT attenuated this effect by about 90% (Fig. 2C). Similarly, thrombin induced VSMC DNA synthesis by about 2-fold as measured by [3H]thymidine incorporation, and it was completely inhibited by Ad-GFPVIVIT (Fig. 2D). In addition, depletion of NFATc1 levels by its siRNA also blocked thrombin-induced VSMC migration and DNA synthesis (Fig. 2, E and F). These observations indicate that thrombin-induced VSMC migration and proliferation require NFATc1-mediated cyclin D1 expression.
Studies from our laboratory as well as others have shown that cyclin D1 expression is regulated at both transcriptional and post-transcriptional levels (18, 19, 48, 49). To understand the mechanisms by which NFATc1 regulates thrombin-induced cyclin D1 expression, we first cloned 1.3-kb rat cyclin D1 promoter and analyzed the sequence for transcription factors binding elements using TRANSFAC (50). We identified nine putative NFAT-binding elements, one putative STAT-binding element, and several other cis-acting elements in the 1.3-kb promoter region (Fig. 3A). To find whether thrombin induces cyclin D1 promoter activity and, if so, to identify the regulatory elements, we performed serial promoter deletion analysis. By PCR, the rat cyclin D1 promoter sequences starting from −664, −378 nt and −234 to +37 nt were generated using the 1.3-kb promoter region as a template and cloned into the pGL3 basic vector. VSMCs were transfected with full-length as well as truncated cyclin D1 promoter-luciferase constructs, quiesced, and treated with and without thrombin (0.5 unit/ml) for 8 h, and cell extracts were prepared and analyzed for luciferase activity. Compared with control, thrombin induced a 2-fold increase in luciferase activity with a full-length pGL3-rCCND1p-(1.3 kb)-Luc construct, and deletion of sequences from −1234 to −663 nt in the promoter region resulted in the loss of a response to thrombin, indicating the presence of the thrombin-responsive elements in this region (Fig. 3B).
To understand the role of NFATc1 in the regulation of cyclin D1 promoter activity, VSMCs were cotransfected with both NFATc1 siRNA and pGL3-rCCND1p-(1.3 kb)-Luc construct, and luciferase activity was measured. Thrombin increased the luciferase activity by 2.5-fold in scrambled siRNA-transfected control cells, and this increase was attenuated by depletion of NFATc1 levels, indicating a role for NFATc1 in thrombin-induced cyclin D1 promoter activity (Fig. 3C). Because thrombin-responsive cyclin D1 promoter region (from −663 to −1234 nt) contains three putative NFAT-binding sites and a putative STAT-binding site, we performed site-directed mutagenesis to identify which of these elements are involved in thrombin-induced promoter activity. Site-directed mutagenesis of NFAT-binding elements either at −1037 or −742 abolished thrombin-induced cyclin D1 promoter activity. Mutation in the NFAT-binding element at −890 had no effect on thrombin-induced cyclin D1 promoter activity (Fig. 3D). Surprisingly, mutation in STAT-binding element at −970 also completely blocked thrombin-induced cyclin D1 promoter activity (Fig. 3D). To find whether NFATc1 binds to all three putative NFAT-binding elements or any specific site, we performed EMSA using the −742, −890, and −1037 NFAT-binding elements as 32P-labeled double-stranded oligonucleotide probes. Thrombin induced protein-DNA binding activity only with −1037 NFAT-binding element as a probe (Fig. 3E). To find whether NFATc1 binds to the −1037 NFAT-binding element, supershift EMSA was performed. As shown in Fig. 3E, the anti-NFATc1 antibodies supershifted the −1037 protein·DNA complex. To gain mechanistic insights into the role of the STAT-binding element in thrombin-induced cyclin D1 promoter activity, we next studied the time course effect of thrombin on STAT-DNA binding activity by EMSA using the STAT-binding element as a 32P-labeled double-stranded oligonucleotide probe. Thrombin induced STAT-DNA binding activity in a time-dependent manner (Fig. 3F). Furthermore, the supershift resulting from the incubation of these protein·DNA complexes with anti-STAT-3 antibodies confirmed the binding of STAT-3 to this element (Fig. 3F).
Many studies have shown that NFATs interact with other transcription factors, such as AP-1, Egr1, NF-κB, and Sp1, and influence their transcriptional transactivation activity synergistically (20, 51). In this study, we found that both NFATc1 and STAT-3 bind to the cyclin D1 promoter and mediate its activity in response to thrombin. Therefore, to find whether there is any interaction between these two transcription factors, we performed coimmunoprecipitation experiments. As shown in Fig. 4A, NFATc1 was found to exist in complex with STAT-3 constitutively, and thrombin treatment did not affect this interaction. It was well established that STAT-3 is activated by phosphorylation at Tyr-705 (52, 53), and its DNA binding as well as transcriptional transactivation activity are enhanced by acetylation at Lys-685 (54). Hence, we tested whether NFATc1 interaction with STAT-3 has any influence on STAT-3 phosphorylation or acetylation. Surprisingly, NFATc1-associated STAT-3 was phosphorylated and acetylated with peak effects between 4 and 8 h of thrombin treatment (Fig. 4A). To find whether there is any connection between STAT-3 phosphorylation and its acetylation, first we determined the time course effect of thrombin on STAT-3 phosphorylation (Tyr-705) and acetylation (Lys-685). We found that STAT-3 is phosphorylated in a biphasic manner with its first peak between 1 and 5 min and a second peak between 4 and 8 h of thrombin treatment (Fig. 4B). However, STAT-3 is acetylated in a delayed manner peaking between 4 and 8 h of thrombin treatment (Fig. 4B).
Next, we tested the role of NFATc1 on STAT-3 phosphorylation and acetylation using two approaches. First, we blocked thrombin-induced NFAT activation by Ad-GFPVIVIT and tested its effect on STAT-3 phosphorylation and acetylation. Blockade of NFAT activation did not inhibit STAT-3 phosphorylation at its early phase (Fig. 4C). However, overexpression of GFPVIVIT blocked phosphorylation and acetylation at 8 h of thrombin treatment (Fig. 4D). Blockade of thrombin-induced phosphorylation of STAT-3 by its dominant negative mutant did not have any effect on its acetylation. These observations led to the conclusion that NFATs mediate both phosphorylation and acetylation of STAT-3 at 8 h of thrombin treatment, and STAT-3 acetylation occurs independent of its phosphorylation. In the second approach, NFATc1 levels were down-regulated by its siRNA, and its effect was tested on thrombin-induced STAT-3 phosphorylation and acetylation. Depletion of NFATc1 levels also led to a decrease in thrombin-induced phosphorylation and acetylation of both total and NFATc1-associated-STAT-3 (Fig. 4, E and F). These results infer that thrombin-induced STAT-3 phosphorylation and acetylation at 8 h of thrombin treatment require NFATc1 activation.
To understand the role of STAT-3 acetylation on its tyrosine phosphorylation, we used an acetylation null mutant of STAT-3, STAT-3K685R (37). Interference of STAT-3 acetylation by STAT-3K685R attenuated both total and NFATc1-associated STAT-3 phosphorylation almost to the same level (Fig. 5, A and B). This finding indicated that STAT-3 acetylation is required for its second phase of tyrosine phosphorylation. To understand the role of acetylation on thrombin-induced cyclin D1 expression, VSMCs were cotransfected with STAT-3K685R (pcDNA3-mSTAT-3) and pGL3-rCCND1p(1.3 kb)-Luc construct, quiesced, treated with and without thrombin (0.5 unit/ml) for 8 h, and analyzed for cyclin D1 promoter activity. STAT-3K685R mutant reduced thrombin-induced cyclin D1 promoter-luciferase activity (Fig. 5C). STAT-3K685R also attenuated thrombin-induced VSMC migration and DNA synthesis (Fig. 5, D and E).
Previous studies have demonstrated that STAT-3 acetylation is mediated by p300 in response to oncostatin-M (37, 54, 55). Based on this information, we speculated that NFATc1 may recruit p300 and thereby facilitate thrombin-induced STAT-3 acetylation. To test this assumption, we studied the association of NFATc1 and p300 in control versus thrombin-treated VSMCs by coimmunoprecipitation. As expected, p300 was associated with NFATc1 in response to thrombin (Fig. 6A). Maximum increase in NFATc1 and p300 association was detected between 4 and 8 h of thrombin treatment, and this was correlated with the time course of STAT-3 acetylation. In addition, blockade of NFAT activation by cyclosporin A (10 μm), a potent inhibitor of calcineurin (20), or Ad-GFPVIVIT resulted in the loss of NFATc1 association with p300 (Fig. 6, B and C). Moreover, knockdown of p300 levels by its siRNA also reduced thrombin-induced total and NFATc1-associated STAT-3 acetylation and its late phase phosphorylation (Fig. 6, D and E). In addition, down-regulation of p300 levels attenuated thrombin-induced cyclin D1 promoter activity and its expression (Fig. 6, F and G). These observations strongly support a role for NFATc1-recruited p300 in thrombin-induced STAT-3 acetylation and phosphorylation leading to cyclin D1 expression. To understand the functional significance of p300, we also tested its role in thrombin-induced VSMC migration and proliferation. Down-regulation of p300 levels by its siRNA attenuated thrombin-induced VSMC migration and proliferation (Fig. 6, H and I).
To understand the role of STAT-3 acetylation and its biphasic phosphorylation on NFATc1 and STAT-3 binding to the cyclin D1 promoter, we studied their DNA binding activities by EMSA using a −1037 NFAT-binding element and a −970 STAT-binding element as 32P-labeled double-stranded oligonucleotide probes. Surprisingly, both NFATc1 and STAT-3 showed a time-dependent biphasic DNA binding activity to their respective probes with an initial peak at 1 h and a second peak at 8 h of thrombin treatment (Fig. 7, A and B). To obtain further evidence for the binding of NFATc1 and STAT-3 to the cyclin D1 promoter, we also performed ChIP assays. ChIP assays showed that NFATc1 binds to the cyclin D1 promoter region that contains the −1037 NFAT-binding element in response to thrombin in a time-dependent biphasic manner (Fig. 7C). STAT-3 also bound to cyclin D1 promoter in vivo in a biphasic manner (Fig. 7D).
To investigate the functional significance of the NFATc1-STAT-3 interaction with their binding to the cyclin D1 promoter, we studied the role of their activation on each other's DNA binding capacities by EMSA. Blockade of NFAT activation by Ad-GFPVIVIT and STAT-3 by dnSTAT-3 attenuated thrombin-induced NFAT and STAT binding to their respective binding elements (Fig. 7, E and F). Although interfering with NFAT activation did not affect the first phase of STAT-DNA binding activity at the 1-h time period of thrombin treatment, it blocked the second phase of STAT-DNA binding activity (Fig. 7F). Furthermore, ChIP analysis data show that blockade of NFAT activation by Ad-GFPVIVIT did not affect STAT-3 binding to the cyclin D1 promoter at an early phase in vivo. However, as observed by the EMSA, interfering with NFAT activation completely blocked STAT-3 binding to the cyclin D1 promoter at 8 h of thrombin treatment (Fig. 7H). Interestingly, blockade of STAT-3 activation attenuated NFATc1 binding to the cyclin D1 promoter at both the phases in vivo (Fig. 7G). These observations indicate that although NFATs are essential for binding of STAT-3 to the cyclin D1 promoter at 8 h of thrombin treatment, STAT-3 appears to be involved in mediating the priming effect.
Next, we wanted to evaluate the influence of NFATc1 and STAT-3 on each other's DNA binding activity in thrombin-induced cyclin D1 expression. The blockade of NFATs or STAT-3 completely inhibited both phases of cyclin D1 expression by thrombin (Fig. 7, I and J). Even though these results signified the activation of both interacting partners in facilitating thrombin-induced cyclin D1 expression, they did not explain the role of NFATc1-mediated STAT-3 acetylation in thrombin-induced cyclin D1 expression. Therefore, VSMCs were transfected with the STAT-3K685R mutant, quiesced, and treated with and without thrombin for 1 or 8 h, and the cyclin D1 expression was analyzed. Although blockade of STAT-3-acetylation did not affect the early phase of thrombin-induced cyclin D1 expression, it attenuated the second phase of cyclin D1 expression in response to thrombin (Fig. 7K). Together, these results emphasize that the late phase of cyclin D1 expression depends on acetylation-dependent STAT-3 phosphorylation and its subsequent binding to the cyclin D1 promoter.
To obtain additional evidence in support of their cooperative binding to the cyclin D1 promoter, we next designed a probe that spans from −948 to −1064 nt and contains both NFAT- and STAT-binding elements. Using this sequence as a probe, nuclear extracts from thrombin-treated VSMCs formed a unique protein·DNA complex in a biphasic manner, and it was supershifted by both NFATc1- and STAT-3-specific antibodies (Fig. 8A). This result showed that both NFATc1 and STAT-3 bind to the cyclin D1 promoter simultaneously. If both NFATc1 and STAT-3 bind to their respective elements in the promoter in a cooperative manner and associate with each other, then one would expect that the intervening DNA sequence would loop out. The loop formation is a primary event in any enhanceosome formation by multiple transcription factors (56–58). In addition, the relative position of the transcription factor-binding site is a determinant factor in enhanceosome formation (59). To understand the interactions between NFATc1 and STAT-3 in binding to the cyclin D1 promoter and forming an enhanceosome, we developed 32P-labeled DNA probes that span cyclin D1 promoter region from −948 to −1064 nt with either transposing their binding elements or moving them closer to each other (Fig. 8B). The EMSA analysis using these probes revealed that thrombin-treated VSMC nuclear extracts formed a protein·DNA complex with intact probe as well as the probe with NFAT- and STAT-binding elements transposed (Fig. 8C). However, no protein·DNA complex was observed with the probe in which the NFAT- and STAT-binding elements were moved closer to each other (Fig. 8C). Next, we introduced the same modifications in cyclin D1 promoter-luciferase vector pGL3-rCCND1p-(1.3 kb)-Luc. Although transposing NFAT- and STAT-binding elements did not affect thrombin-induced cyclin D1 promoter activity, moving these elements closer to each other abolished the promoter activity (Fig. 8D). These results suggest that activation-dependent binding of NFATc1 and STAT-3 complex to cyclin D1 promoter leads to looping out of the intervening DNA sequence to initiate enhanceosome formation (Fig. 8E).
Cyclin D1 functions are primarily linked to cell cycle progression (4, 6, 60). Recently, many studies using transformed cells or cells derived from cancer patients have demonstrated that cyclin D1 plays a role in cell migration (8, 12, 13). These reports and the fact that cells in the late S or G2/M phase do not migrate (61, 62) tempted us to test the role of early and late phases of thrombin-induced cyclin D1 expression in VSMC migration and proliferation, respectively. Cytoskeleton reorganization events that takes place in the cytoplasm are necessary to provide force for cells to migrate (63), and most of the effector molecules involved in cell motility are confined to the cytoplasm of migrating cells (16, 64, 65). In contrast, molecules such as cyclin D1 that are linked to cell cycle progression most likely translocate to the nucleus (66). Therefore, we first conducted immunofluorescence staining studies to observe the localization of cyclin D1 during the early and late phases of its thrombin-induced expression. As seen in Fig. 9A, cyclin D1 was detected primarily in the cytoplasm during the early phase of its expression and translocated to the nucleus during the late phase of its expression. To confirm this result, we prepared cytoplasmic and nuclear extracts from thrombin-treated VSMCs and analyzed for cyclin D1 levels in both the fractions. Cyclin D1 was present only in the cytoplasmic fraction at the 1-h period of thrombin treatment, and at 8 h it was detected in both the fractions (Fig. 9B). These results provide further clues supporting our hypothesis that the early phase cyclin D1 expression, during which it is mostly located in the cytoplasm, may be mediating cell migration.
Next, we applied miRNA-based RNAi approach (67) to down-regulate cyclin D1 in a phase-specific manner and test its role in VSMC migration and proliferation. We designed a constitutively active CMV promoter-based vector (pcDNA6-DsRed2-CCND1 miR) and a tetracycline-inducible TRE-Tight promoter-based vector (pTRE-Tight-EmGFP-CCND1 miR) to cocistronically express cyclin D1 miRNA along with DsRed2 or EmGFP, respectively (Fig. 10A), and we tested their efficiency to knock down cyclin D1 expression in a constitutive (Fig. 10B) or doxycycline-dependent manner (Fig. 10C). Addition of doxycycline to VSMCs after 2 h of thrombin treatment enabled cells to retain the early phase of cyclin D1 expression intact while knocking down its late phase expression (Fig. 10C). To test the role of the early phase of cyclin D1 expression in cell migration, VSMCs were transfected with pcDNA6-DsRed2-Scr miR, pcDNA6-DsRed2-CCND1 miR, pTRE-Tight-EmGFP-Scr miR, or pTRE-Tight-EmGFP-CCND1 miR vectors, quiesced, and subjected to thrombin-induced migration while inducing down-regulation of cyclin D1 expression from the beginning or 2 h after thrombin treatment by doxycycline. VSMCs that were transfected with pcDNA6-DsRed2-Scr miR or pTRE-Tight-EmGFP-Scr miR and therefore expressing scrambled miRNA migrated in response to thrombin irrespective of the initiation period of doxycycline treatment (Fig. 10D). However, VSMCs that were transfected with pcDNA6-DsRed2-CCND1 miR and thereby expressing cyclin D1 miRNA constitutively did not migrate in response to thrombin (Fig. 10D). Similarly, VSMCs that were transfected with pTRE-Tight-EmGFP-CCND1 miR and exposed simultaneously to both doxycycline and thrombin also did not migrate (Fig. 10D). In these cells, either constitutive or doxycycline-induced expression of cyclin D1 miRNA from the beginning of thrombin treatment would have blocked both the early and late phases of cyclin D1 expression, and therefore cells did not migrate indicating a mere requirement of cyclin D1 for cell migration. To our surprise, VSMCs that were transfected with pTRE-Tight-EmGFP-CCND1 miR vector and exposed to doxycycline after 2 h of thrombin treatment showed migratory response equivalent to those transfected with pcDNA6-DsRed2-Scr miR and treated with thrombin (Fig. 10D). Addition of doxycycline 2 h post-thrombin treatment enabled these cells to retain the early phase of thrombin-induced cyclin D1 expression intact, and this appears to be sufficient to drive VSMC migration, which clearly supports our hypothesis.
To understand the role of early versus late phase of cyclin D1 expression in thrombin-induced cell proliferation, VSMCs were transfected with pTRE-Tight-EmGFP-Scr miR or pTRE-Tight-EmGFP-CCND1 miR vectors, quiesced, and subjected to thrombin-induced DNA synthesis while inducing the down-regulation of cyclin D1 expression from 0, 2, and 8 h after the initiation of thrombin treatment by doxycycline. VSMCs that were transfected with pTRE-Tight-EmGFP-Scr miR vector showed a 3–4-fold increase in DNA synthesis in response to thrombin irrespective of the initiation period of doxycycline treatment (Fig. 10E). However, down-regulation of cyclin D1 expression from 0 or 2 h after the addition of thrombin treatment in pTRE-Tight-EmGFP-CCND1 miR vector-transfected VSMCs blocked DNA synthesis (Fig. 10E). These results suggest that although the first phase of cyclin D1 expression is not sufficient, the late phase of its expression is required for thrombin-induced VSMC DNA synthesis, which indirectly shows that proliferation is mediated by the late phase of cyclin D1 expression.
Studies from our laboratory as well as others have reported that both receptor tyrosine kinase and G protein-coupled receptor agonists activate STAT-3 and NFATc1 in various cell types mediating their migration and/or proliferation (18, 28, 68–70). In addition, many reports have demonstrated that both STAT-3 and NFATc1 regulate agonist-induced cyclin D1 expression in the mediation of cell growth and migration (17, 18, 32, 71). In characterizing the NFATc1-mediated regulatory mechanisms of cyclin D1 promoter activity, we found that it interacts with STAT-3. It is well established that the NFAT family of transcription factors interact with AP-1, Egr1, NFκB and Sp1 transcription factors in the regulation of cytokine and/or cell cycle-dependent kinase inhibitor genes (72–74). In this study, it was found that NFATc1 exists as a complex with STAT-3 even in quiescent cells, and their activation does not affect their physical association. In view of the aforementioned findings as well as the present observations that thrombin-induced STAT-3 acetylation is dependent on NFATc1 activation, and both these transcription factors mediate thrombin-induced VSMC migration and proliferation via enhancing cyclin D1 expression, it may be conceivable that NFATs via their ability to interact with various transcription factors may be involved in the regulation of diverse cellular functions.
STAT-3 activation requires tyrosine phosphorylation at Tyr-705 (30, 31). Recent studies have shown that in addition to its tyrosine phosphorylation, STAT-3 may be acetylated, and this post-translational modification is needed for its transcriptional transactivation activity (37, 54, 55). Furthermore, it was demonstrated that STAT-3 acetylation and phosphorylation occur acutely and independently of each other in response to oncostatin-M (37). In contrast, in this study we show that although STAT-3 tyrosine phosphorylation occurs in a biphasic manner, its acetylation takes place in a delayed manner, and it coincides with the second phase of its tyrosine phosphorylation. Because blockade of STAT-3 phosphorylation had no effect on its acetylation, but interference with its acetylation suppresses the second phase of its tyrosine phosphorylation, STAT-3 acetylation is required for its late phase phosphorylation. In regard to the mechanisms of protein acetylation, besides histone acetyltransferases, transcriptional coactivators such as p300 and CREB-binding protein have been shown to possess acetyltransferase activity (75, 76). In this aspect, both NFATc1 and STAT-3 recruit p300 in the regulation of expression of their target genes (55, 77). In addition, Yuan et al. (37) had established that p300 mediates STAT-3 acetylation and thereby enhances its transcriptional transactivation activity. In our studies, we also found that p300 mediates thrombin-induced STAT-3 acetylation. However, the novel aspect of this study is that NFATc1 and STAT-3 exist as a heterodimer in resting cells, and the recruitment of p300 by the NFATc1·STAT-3 complex requires NFATc1 activation. It is also interesting to note that NFATc1·STAT-3 complex binds to the cyclin D1 promoter in a biphasic manner. It appears that during the early phase of STAT-3 DNA binding, its phosphorylation seems to be sufficient, whereas for the second phase of its DNA binding acetylation and acetylation-dependent phosphorylation are likely required. This may infer that although NFATc1·STAT-3 complex binds to the promoter first, STAT-3 acetylation, which occurs at a much later time point, may be required for its enhanced transcriptional transactivation activity. Because STAT-3 acetylation is required for late phase cyclin D1 expression in VSMCs, and blockade of STAT-3 acetylation did not affect the early phase of cyclin D1 expression, which is sufficient to drive thrombin-induced cell migration, overexpression of acetylation null mutant of STAT-3 should not have affected VSMC migration. Therefore, inhibition of thrombin-induced migration by either overexpression of STAT-3K685R mutant or down-regulation of p300 indicates that besides cyclin D1 cell migration may require the involvement of several other gene products whose expression might require either p300 and/or STAT-3 acetylation. This also implies that cyclin D1 function may be necessary but is not sufficient for VSMC migration.
The mammalian promoter/enhancer sequences have been shown to contain more than one transcription factor-binding elements (78–80). Furthermore, it is widely observed that multiple transcription factors influence the expression of a single gene such as cyclin D1 in response to various agonists in the mediation of mitogenic and/or motogenic effects (14, 57, 58). In this context, it is important to note that both NFATc1 and STAT-3 bind to cyclin D1 promoter in response to thrombin. Because the deletion of either NFAT- or STAT-binding elements abolished thrombin-induced cyclin D1 promoter activity and both NFATc1 and STAT-3 exist as a complex, it is likely that these transcription factors form an enhanceosome in the cyclin D1 promoter. This conclusion is further strengthened by the observations that transposing the NFAT- and STAT-binding elements did not affect their binding capacities to the cyclin D1 promoter or their influence on the promoter activity in response to thrombin. However, moving the NFAT- and STAT-binding elements closer to each other resulted in the loss of their binding to the promoter and thereby ablated the promoter activity.
It is noteworthy that despite the presence of three putative NFAT-binding elements in the thrombin-responsive cyclin D1 promoter region, NFATc1 binds only to NFAT-binding element at −1037. Conversely, mutagenesis analysis shows that disruption of either NFAT-binding element at −1037 or −742 but not at −890 affects thrombin-induced cyclin D1 promoter activity. Based on these observations, it may be suggested that in addition to the core binding element (A/T)GGAAAA, other nucleotide sequences surrounding it may be crucial for NFATc1 binding to the promoter. This assumption may be corroborated by the observation that although the NFAT site at −890 possesses complete homology to the core-binding element, it does not bind to NFAT in vitro, and its disruption does not affect thrombin-induced cyclin D1 promoter activity in vivo. However, it may be speculated that spatial positions of NFAT-binding elements at −1037 and −742 and STAT-binding element at −970 in the cyclin D1 promoter are crucial for the recruitment and binding of the NFATc1·STAT-3 complex to the promoter leading to its enhanced activity.
Regulation of cyclin D1 expression is extensively investigated to understand its cyclin-dependent kinase-dependent/independent functions (8). In studies where its function in cell cycle progression was examined, its expression in response to a variety of stimuli was studied at time periods closer to late S phase (18, 19, 81, 82). However, when the role of cyclin D1 in cell migration or cyclin-dependent kinase-independent functions was tested in cancer cell models, its levels were either overexpressed or down-regulated using a variety of genetic tools (8, 12, 83). Thus, its biphasic expression was never identified. In contrast, we used primary VSMCs to study cell migration and proliferation, which permitted the detection of a biphasic expression of cyclin D1. The prominent unanswered question in cell biology is as follows: How does the cell cycle regulatory components modulate both cell migration and proliferation? The origin of this question is based on the fact that most of the growth factors and cytokines stimulate both cell migration and proliferation (18, 19, 69, 84), and the cells in the late S or G2/M phase do not migrate (61, 62). Toward this end, the biphasic expression of cyclin D1 that we observed in this study may explain its dual role in both cell migration and proliferation.
The classical function of cyclin D1 is to associate with partner cyclin-dependent kinases and phosphorylate pRb, thereby relieving it from repressing transcriptional transactivation activity of E2Fs (5, 6, 66). But in several cancers cyclin D1 overexpression did not result in the hyperphosphorylation of pRb or higher expression of E2F target genes (5, 85). In addition, the transforming ability of cyclin D1 was not hampered by deletion of the pRb-binding domain located at its N terminus (86), pointing to its role in cellular functions other than cell proliferation. In this direction, we wanted to study the role of the early and late phases of cyclin D1 expression in VSMC migration and proliferation. Although previous studies from our laboratory as well as others have provided clues for its role in cell migration and proliferation (8, 12, 13, 17–19), the approaches that were employed to down-regulate cyclin D1 levels were not sufficient to differentiate its function between cell migration and cell proliferation independently. To overcome this difficulty, we designed a CMV promoter-based vector to facilitate constitutive cocistronic expression of cyclin D1 miRNA with DsRed2 and a TRE-Tight promoter-based vector to drive tetracycline-inducible cocistronic expression of cyclin D1 miRNA with EmGFP. While the usage of these vectors enabled us to down-regulate cyclin D1 expression, the cocistronically expressed fluorescent proteins aided in differentiating the cells in which cyclin D1 levels were affected in a phase-specific manner from the cells with constitutive depletion of cyclin D1 levels. Our findings that the blockade of late phase cyclin D1 expression while negating the DNA synthesis did not inhibit cell migration, and its early phase expression was not sufficient for cell cycle progression, imply that the first phase of its expression is linked to cell migration, and the second phase of its induction is required for the cell cycle progression. Moreover, during the early phase of its expression, cyclin D1 was mostly localized in the cytoplasm leading to the speculation of its possible direct or indirect role in the regulation of cytoskeleton reorganization. This view is strengthened by the recent observations that overexpression of cyclin D1 induced mitotic spindle disorganization (87). During the second phase of its expression, as expected, cyclin D1 is translocated to the nucleus, which supports its role in cell proliferation. Any new insights in the cell cycle regulation or cell migration have a large impact in the treatment of cancer and vascular diseases (11, 60, 88). In this regard, our study for the first time provides a mechanistic insight into how temporally spaced biphasic expression of cell cycle control molecules such as cyclin D1 bridges the cell migration and proliferation.
*This work was supported, in whole or in part, by National Institutes of Health Grant HL069908) from NHLBI (to G. N. R.).
2The abbreviations used are: