Peptide-Mediated Disruption of Calmodulin–Cyclin E Interactions Inhibits Proliferation of Vascular Smooth Muscle Cells and Neointima FormationNovelty and Significance
Rationale: Cell cycle progression in vascular smooth muscle cells (VSMCs) is a therapeutic target for restenosis.
Objective: Having discovered that calmodulin (CaM)-dependent cyclin E/CDK2 activity underlies Ca2+-sensitive G1-to-S phase transitions in VSMCs, we sought to explore the physiological importance of the CaM–cyclin E interaction.
Methods and Results: A peptide based on the CaM binding sequence (CBS) of cyclin E was designed to interfere with CaM–cyclin E binding. Compared with control peptides, CBS blocked activating Thr160 phosphorylation of CDK2, decreased basal cyclin E/CDK2 activity, and eliminated Ca2+-sensitive cyclin E/CDK2 activity in nuclear extracts from mouse VSMCs. Nucleofection with CBS, or treatment with CBS conjugated to the HIV-1 TAT protein transduction domain to improve bioavailability, inhibited G1-to-S cell cycle progression in a dose-dependent manner. These effects were not observed with control peptides. TAT-CBS inhibited 3H-thymidine incorporation in primary human aortic SMCs (HA-SMCs) in vitro, manifested greater transduction into HA-SMCs compared with endothelial cells in vitro, and limited decreased SM22α expression, neointima formation, and medial thickening without affecting collagen deposition or reendothelialization in a mouse model of carotid artery injury in vivo. The antiproliferative effects of CBS remained evident in mouse embryonic fibroblasts derived from wild-type mice but not cyclin E1/E2 double knockout mice.
Conclusions: A synthetic peptide designed to disrupt CaM–cyclin E binding inhibits Ca2+/CaM-dependent CDK2 activity, cell cycle progression, and proliferation in VSMCs and limits arterial remodeling following injury. Importantly, this effect appears to be cyclin E–dependent and may form the basis of a potentially novel therapeutic approach for restenosis.
Vascular smooth muscle cells (VSMCs) normally proliferate at very low rates in the media of adult arteries, remaining in the growth arrested (G0) phase of the cell cycle. A shift in the balance between growth stimulatory and inhibitory factors can lead to cell cycle reentry and transformation from contractile and quiescent to proliferative and synthetic phenotypes. Thus activated, VSMCs can remodel the artery by altering the extracellular matrix, replicating in the media, and migrating to the intima to undergo further cycles of proliferation. Indeed, unregulated proliferation of VSMCs is a principal mechanism underlying the pathogenesis of common vascular diseases, such as atherosclerosis and restenosis.1,2
Decades of work have implicated ionic calcium (Ca2+) as a regulator of eukaryotic cell cycle progression.3 In VSMCs, we previously made 3 related discoveries regarding Ca2+-mediated cell cycle regulation: (1) a coordinated increase in the free intracellular Ca2+ concentration is required for G1-to-S phase cell cycle transition4,5; (2) this occurs through cell cycle–associated expression and activation of specific Ca2+ pumps and channels5,–,7; and (3) is at least partly mediated by Ca2+/calmodulin (CaM)-dependent cyclin E/CDK2 activity.8
Our findings suggested that Ca2+ sensitivity of the G1-to-S phase cell cycle transition requires the direct binding of the major Ca2+ signal transducer CaM to cyclin E, through a specific and conserved CaM-binding motif in cyclin E. The functional importance of this motif was accentuated by the observation that a cyclin E mutant lacking this motif was unable to produce Ca2+/CaM-stimulated activity of CDK2.8 These data shed light on a mechanistic basis for Ca2+-sensitive cell cycle progression and predicted other possible Ca2+/CaM-sensitive cell cycle targets.9
Based on the discovery of a functional CaM–cyclin E interaction, we hypothesized that blocking CaM–cyclin E binding through the use of a synthetic peptide will inhibit Ca2+-sensitive G1-to-S phase transitions and slow the proliferation of VSMCs by competing with cyclin E for binding to CaM. The present report details our characterization of such a peptide and provides data supporting the physiological importance of CaM–cyclin E interactions in VSMCs. Together, the results suggest a novel therapeutic approach to inhibiting proliferation in VSMCs.
An expanded Methods section is available in the Online Data Supplement at http://circres.ahajournals.org.
Isolation and culture of primary mouse aortic smooth muscle cells (SMCs) have been described.6 Cells and reagents for primary human aortic SMC and primary human aortic endothelial cell (EC) culture were purchased from Invitrogen. Wild-type and cyclin E1/E2 double knockout mouse embryonic fibroblasts (MEFs) were kindly provided by Dr P. Sicinski (Harvard Medical School, Boston, MA).
Cyclin E/CDK2 Assay
Histone H1 kinase activity of complexes immunoprecipitated with anti–cyclin E antibody were performed as described previously.8 Nucleofection and TAT fusion were used to deliver CBS peptide into cells.
Primary antibodies, including anti-CDK2 (07-631; Upstate Biotech, Billerica, MA), anti–cyclin E (ab7959; Abcam, Cambridge, MA), anti–phosphorylated CDK2 on Thr160 (2561; Cell Signaling Technology, Danvers, MA), anti-CaM (sc-1989; Santa Cruz Biotechnology, Santa Cruz, CA), anti–phosphorylated CDK2 on Thr14/Tyr15 (sc-28435-R; Santa Cruz Biotechnology), and anti-actin (sc-7210; Santa Cruz Biotechnology) were used to detect levels of CaM-binding and CDK2 phosphorylation and to confirm peptide delivery.
Confocal microscopy was used to track fluorescein isothiocyanate (FITC)-labeled peptides in cells following nucleofection and to characterize the phenotype of TAT-CBS–treated SMCs in human aortic SMCs and mouse carotid artery by immunostaining for SMC marker proteins. TAT domain–mediated delivery of His-tagged peptides into human aortic SMCs and mouse carotid artery was also confirmed by immunostaining. To assess reendothelialization, carotid artery sections harvested 7 days after wire-denudation injury were immunostained with anti-CD31 antibody.
Cell counting and the MTT and the 3H-thymidine incorporation assays were used to measure cell number and proliferation of CBS peptide–treated cells.
LDH, Caspase-3, and TUNEL Assays
The LDH-based in vitro toxicology assay kit (Sigma) was used to assess cytotoxicity of nucleofected peptides and TAT-conjugated peptides.10 The caspase-3 Colorimetric Assay Kit (Sigma) and In Situ Cell Death Detection Kit (Roche) were used to assess apoptosis of TAT-peptide–treated cells.
To track cell cycle stages, flow cytometric analysis was performed as described previously.8
Mouse Carotid Artery Injury
All animal experimentation conformed to protocols approved by the institutional animal care committee. The model used has been described previously.11
Tissue Processing and Analysis
At 14 days after injury, lethally anesthetized mice were perfusion fixed. Previously injured and sham common carotid arteries were harvested, postfixed, and embedded in paraffin or OCT. Sections were then stained, imaged, and quantified for analyses.
Design of the CBS Peptide
We hypothesized that a 22-aa synthetic peptide based on the Calmodulin Binding Sequence (CBS) of human cyclin E1 would compete with cyclin E for binding to CaM and thus inhibit the CaM–cyclin E interaction (Figure 1A). A “5A” peptide, in which each hydrophobic residue of CBS was substituted with alanine, and a negative control peptide (NC) composed of a random sequence of 22 aa were also synthesized. First, we tested the effects of these peptides on cyclin E/CDK2 activity as measured by histone H1 kinase activity of nuclear extracts immunoprecipitated with anti–cyclin E antibody.8 In these experiments, all peptides were added during the immunoprecipitation (IP) step. CBS reduced cyclin E/CDK2 activity by 40%, whereas the NC peptide had no effect as compared with untreated controls. 5A had less inhibitory effect on cyclin E/CDK2 activity than CBS, suggesting functional importance of the hydrophobic residues in CBS (Figure 1B). Co-IP analysis revealed that cyclin E/CDK2 complex formation was not affected by addition of the CBS peptide (Figure 1C). By measuring histone H1 kinase activity over a range of Ca2+ concentrations, we found that Ca2+-sensitive enhancement of cyclin E/CDK2 activity was abolished by CBS but not NC (Figure 1D).
Delivery of the CBS Peptide to VSMCs
To investigate the effects of the CBS peptide on cell cycle progression and proliferation of VSMCs, CBS and NC were delivered to primary mouse aortic SMCs via nucleofection. For tracking purposes, both peptides were tagged with FITC. Confocal microscopy for FITC confirmed delivery of CBS to VSMC nuclei with approximately 40% efficiency (data not shown). The MTT cell viability assay12 was used to determine the effect of peptide nucleofection on VSMC number, as an indirect indicator of proliferation (Online Figure I).
Because the absence of NC effect on cyclin E/CDK2 kinase activity (Figure 1B) may have resulted from differences in secondary structure and biochemical properties (pH, hydrophobicity, polarity) of the NC peptide, MTT assays were also performed with a scrambled CBS, harboring all of the original residues at different locations. CBS treatment significantly decreased the number of VSMCs compared with untreated and NC–treated cells as measured by MTT assay (Figure 2A). Importantly, the scrambled peptide was unable to decrease VSMC number, confirming that the amino acid sequence of CBS is critical to its antiproliferative function (Figure 2A).
The CBS peptide also delayed and inhibited the number of primary mouse aortic SMCs in a dose-dependent manner (Figure 2B). To define an IC50, different concentrations of CBS were delivered to asynchronous primary mouse aortic SMCs tracked with the MTT assay. A sigmoidal dose response curve (Boltzmann model) yielded an IC50 of 186 μmol/L for the parent CBS peptide (Online Figure II).
Mechanisms Underlying the Observed Effects of CBS
To explore mechanisms underlying CBS-induced inhibition of VSMC number, LDH levels were examined as a measure of cytotoxicity. LDH release into the culture media of each treatment group did not differ (untreated-, CBS-, and NC-treated VSMCs), indicating that nucleofection of CBS peptide did not increase cell death (Online Figure III).
Densitometric quantification of co-IP and Western blot data revealed that CBS inhibited both (1) CaM–cyclin E interactions (2.25 versus 1.11) without affecting levels of CaM (1.84 versus 1.68) or cyclin E (2.36 versus 2.48); and (2) phosphorylation of CDK2 on Thr160, a critical event in the activation of CDK2 (2.63 versus 0.94). It was also observed that the amount of inhibitory CDK2 phosphorylation at Thr14/Tyr15 sites (0.41 versus 0.38), and the total amount of CDK2 (1.04 versus 0.98) were not altered by treatment with CBS. Furthermore, binding of CaM to calcineurin, another CaM target protein, was unaltered by CBS (1.44 versus 1.40), suggesting selectivity of CBS for its target interaction (Figure 3A).
FACS-driven analysis of cell cycle progression confirmed that the effect of CBS on cell number was attributable to inhibition of proliferation by G1 cell cycle arrest. Starvation of primary mouse aortic SMCs resulted in 80% cell synchronization in G0/G1. CBS versus NC peptides were delivered to starved cells via nucleofection followed by serum stimulation with 10% FBS +50 ng/mL PDGF for 24 hours. Without peptide treatment, 22.4% of serum-stimulated cells entered S phase, whereas only 11.7% of CBS-treated cells entered S phase (Figure 3B).
We next examined whether CBS acts to inhibit cell proliferation exclusively through cyclin E, or nonselectively through other proteins. Although cyclin E–deficient VSMCs would have been ideal for such an investigation, cyclin E1−/−E2−/− embryos do not survive past embryonic day 11.5.13 Alternatively, wild-type and cyclin E double knockout mouse embryonic fibroblasts (MEFs) were obtained. Absence of cyclin E1 protein in double knockout cells was confirmed by Western blot (Online Figure IV). MTT assays revealed that wild-type MEF cell number was reduced by CBS at a level similar to that observed in VSMCs. However, the number of MEFs from cyclin E1−/−E2−/− mice was unaffected by CBS, supporting a cyclin E–dependent mechanism of action (Figure 4A).
We next tested whether the effects of CBS on cell cycle progression of MEFs were also cyclin E–dependent. CBS or NC peptides were delivered to starved MEFs via nucleofection, followed by serum stimulation with 10% FBS for 24 hours. FACS analysis indicated that cell cycle progression into S phase was inhibited by CBS in wild-type, but not in cyclin E1/E2 double knockout, MEFs. By contrast, calmidazolium, a nonselective CaM-inhibitor, was able to inhibit cell cycle progression of double knockout MEFs, possibly through other CaM-dependent pathways (Figure 4B). These results further supported cyclin E–target specificity of the CBS peptide.
Modifications to Increase Bioavailability
Because nucleofection is not possible in vivo, we fused the TAT protein transduction domain from HIV-1 to CBS to enable potential in vivo therapy (Online Figure V, A). This approach was based on reports of the ability of the TAT domain to successfully deliver a size-independent variety of molecules into cell nuclei. Although the exact mechanism of TAT-mediated protein transduction is unknown, the large cationic charge of arginine residues is believed to allow TAT-conjugates to effectively cross cell membranes by receptor-mediated endocytosis.14,15 Cells were exposed to TAT-CBS-His peptide in serum-free cell culture media for 1 hour, as serum decreases the transfection efficiency of the TAT domain.14 Delivery of TAT-CBS-His or TAT-NC-His was confirmed by both Western blot and immunostaining analyses using an anti–His tag antibody (Online Figure V, B and C).
To investigate putative antiproliferative effects of TAT-CBS-His, primary mouse aortic SMCs were treated with peptide, washed twice with PBS, supplied with fresh media containing 10% FBS, and analyzed by MTT assay 2 days later. As expected, the TAT-CBS-His peptide inhibited the number of VSMCs, whereas TAT-NC-His peptide showed no effect (Figure 5A). Treatment with various concentrations of peptide revealed a dose-dependent relationship between TAT-CBS-His and decrease in cell number. Nonlinear regression analysis (Boltzmann sigmoid curve model) of 3 separate experiments revealed an IC50 of 8.89±1.24 μmol/L (Online Figure VI, A), demonstrating that the antiproliferative effect of CBS was significantly enhanced by TAT domain fusion compared with nucleofection (IC50: 8.9 versus 186.2 μmol/L). The maximum level of inhibition was also increased by TAT-CBS fusion as compared with CBS delivered by nucleofection (0.70 versus 0.50).
Effects of CBS Peptide Treatment on Human Vascular Cells
To validate results in mouse and explore physiological relevance of TAT-CBS in human, we demonstrated an antiproliferative effect of the peptide in human aortic SMCs by cell counting (Online Figure VI, B). Confirming that the antiproliferative effect of TAT-CBS was attributable to cell cycle arrest and not cell death, neither extracellular LDH release, caspase-3 activation, nor TUNEL staining were detected in human aortic SMCs at 24 hours (data not shown) or 72 hours following treatment (Online Figure VII). Moreover, immunostaining for contractile SMC marker proteins smooth muscle 22-α (SM22-α), smooth muscle myosin heavy chain (sm-MHC), and α-smooth muscle actin (α-SMA) revealed that TAT-CBS treatment did not significantly alter the phenotype of human aortic SMCs in vitro (Online Figure VIII).
To further establish that CBS works via an inhibitory effect on S-phase entry and to ascertain whether this was evident in human VSMCs, we next examined DNA synthesis as measured by 3H-thymidine incorporation in human aortic SMCs treated with TAT-CBS. Compared with untreated cells, TAT-CBS, but not TAT-NC, produced a dose-dependent inhibition of S-phase entry in human aortic SMCs (Figure 5B).
To explore the clinical potential of TAT-CBS as a novel therapeutic agent, we also tested the effect of TAT-CBS on 3H-thymidine incorporation in human aortic ECs. Although TAT-CBS did appear to dose-dependently inhibit human aortic EC proliferation, the effect only differed from TAT-NC–treated cells at the highest concentration administered (1 mmol/L), which may be cytotoxic (Figure 5B). To compare transduction efficiency of TAT-CBS in HA-SMCs versus HA-ECs, immunostaining for the His-tagged version of the peptide was performed at several time points following treatment in vitro. Results showed that HA-SMCs manifest a similar extent but greater degree of His-staining than HA-ECs (Online Figure IX). Consistent with the results from nucleofection experiments with CBS (MTT assays and flow cytometry; Figure 3C and 3D), the ability of TAT-CBS to block S-phase entry as measured by 3H-thymidine incorporation was only evident in wild-type MEFs (Online Figure X, A) and not cyclin E double knockout MEFs (Online Figure X, B). Together, these data strongly support a cyclin E–dependent effect of the CBS peptide sequence on S-phase progression in both mouse and human cell types.
CBS Inhibits Neointima Formation In Vivo
The TAT-CBS-His peptide was also tested in a mouse common carotid artery injury model via pluronic gel administration. In aqueous solution, the surfactant pluronic F-127 transforms from liquid to nonfluid hydrogel above room temperature. This property enables in vivo peptide delivery, because the semisolid solution of peptide and F-127 allows the peptide to remain concentrated and protected by the surfactant matrix. Carotid artery injury was performed by wire denudation. Injuries were performed on the left common carotid artery, with the right common carotid artery serving as an uninjured control. After injury, common carotid arteries were immersed in 100 μL of 25% (wt/vol) pluronic F-127 gel, either containing peptide (TAT-CBS-His or TAT-NC-His, 250 μmol/L) or no peptide (6 mice per treatment group) before wound closure. Delivery of TAT-CBS-His into VSMCs of the mouse carotid artery was confirmed by immunostaining (Online Figure XI).
Because migration of VSMCs from media to intima following initial proliferation in media is an essential step in the development of restenosis or atherosclerosis, our quantitative analysis of arterial sections included: total area of arterial (1) media, (2) intima, and (3) intima to media (I/M) ratio. H&E staining of arteries harvested 14 days after injury showed that treatment with TAT-CBS-His significantly decreased neointima formation and arterial media thickening compared with treatment with F-127 alone or TAT-NC-His control peptide groups (Figure 6A). Morphometry confirmed that injured arteries treated with TAT-CBS-His manifest reduced medial and intimal areas and I:M ratio, indices consistent with reduced VSMC proliferation and migration (Figure 6B).
Proliferating cell nuclear antigen (PCNA) staining was performed as an additional measure of the level of proliferation in injured arteries. Arteries treated with TAT-CBS-His had a decreased percentage of PCNA-positive nuclei, further confirming the ability of TAT-CBS-His peptide to inhibit cell proliferation in injured arteries (Figure 7A). Separate analyses of intima versus media revealed significantly more PCNA-positive nuclei in the intima than media of injured arteries. Indeed, TAT-CBS-His treatment specifically decreased the percentage of PCNA-positive nuclei in the intima as compared with F-127–only and TAT-NC-His–treated controls, with differences in the percentage of PCNA-positive nuclei in the media not achieving statistical significance. Moreover, in vivo SM22-α immunostaining revealed that TAT-CBS–treated arteries manifest significantly greater SM22-α expression, similar to uninjured arteries, than injured arteries treated with TAT-NC (Figure 7B). This finding is consistent with the notion that CBS-induced growth-arrest may prevent injury-driven decreases in the expression of VSMC-specific contractile markers (ie, “dedifferentiation”) in vivo, a strategy recently shown to have promise in a similar model in rat.16
CBS Does Not Affect Extracellular Matrix Formation or Reendothelialization In Vivo
Because extracellular matrix deposition also contributes to the formation of restenosis, we quantified collagen in injured carotid arteries (Online Figure XII). Compared with controls, TAT-CBS treatment did not significantly alter type I, type III, or total collagen deposition.
Given our data with ECs in vitro, we also investigated the effect of CBS on the reendothelialization of injured carotid arteries in vivo. CD31 staining of sectioned arteries revealed that TAT-CBS-His peptide delivery did not affect reendothelialization at 7 days after injury, as compared with gel-only and TAT-NC-His administered arteries (Figure 8).
In this study, we have shown that CBS, a specific 22-aa peptide inhibits (1) the binding of CaM to cyclin E; (2) Ca2+-sensitive cyclin E/CDK2 activity; (3) G1-to-S cell cycle progression of VSMCs; and (4) VSMC and EC proliferation in vitro, with (5) greater peptide transduction in VSMCs versus ECs, and without increasing (6) cytotoxicity, (7) apoptosis, or (8) dedifferentiation of VSMCs. It was also shown that CBS inhibits (9) the activating phosphorylation of CDK2 at Thr160 without altering the inhibitory phosphorylation on Thr14/Tyr15 by (10) selectively interfering with CaM–cyclin E interactions. Importantly, (11) the binding of CaM to another target protein, calcineurin, was not altered by CBS peptide; and (12) the proliferation and S-phase progression of cyclin E double knockout MEFs were not affected by treatment with CBS. Finally, in a mouse model of carotid arterial endothelial denudation, in vivo delivery of the TAT-CBS-His peptide to VSMCs prevented loss of SM22α expression and demonstrated (13) decreased neointima formation with (14) suppression of intimal VSMC proliferation, (15) without adversely affecting extracellular matrix formation or (16) reendothelialization. Taken together, these findings suggest that the CBS peptide inhibits (1) cyclin E–specific, Ca2+/CaM-dependent, CDK2 activity; and (2) Ca2+-sensitive cell cycle progression and cell proliferation in VSMCs.
Although several previous studies of restenosis have therapeutically targeted cell cycle regulators,17 our study is unique in its selective disruption of a protein-protein interaction (CaM–cyclin E) to effect a novel therapy for vascular proliferative disorders. Such disorders include restenosis, a relatively frequent consequence of balloon angioplasty performed on occluded or narrowed coronary arteries and/or coronary artery bypass grafts.18,19 Indeed, before the development and widespread use of intracoronary stents, studies of early percutaneous coronary interventions revealed that as many as 30% of dilated arteries underwent restenosis, often requiring repeat procedures.20 However, based on research showing that restenosis was characterized by rapid proliferation of VSMCs,21 drugs that interrupt this process and techniques for the elution of such drugs from deployed stents have resulted in a dramatic reduction in the rates of clinical restenosis.22,23 As drug eluting stents continue to be based on the local release of high concentrations of powerful cell toxins that may delay and or impair endothelial healing as well as inhibiting VSMC proliferation, persistent endothelial dysfunction manifesting itself in the form of residual vasodilatory deficits and/or long-term thrombosis risks remain a concern. Indeed, alternative strategies for more cost-effective and less toxic treatments of conditions such as restenosis remain of interest to interventional cardiology.
A distinguishing characteristic of the CBS peptide versus antiproliferative agents used in existing drug eluting stents is its focused mechanism of action. Targeting cyclin E–dependent Ca2+-sensitive CDK2 activity, the actions of CBS peptide are restricted to proliferating cells, whereas the actions of agents such as sirolimus and paclitaxel (on mTOR and microtubules respectively) are not. As such, the CBS peptide may be a less toxic alternative to these agents. Although we cannot state with certainty that CBS does not alter the binding of CaM to any of the >300 known target proteins of CaM,24 our experiments in cyclin E knockout cells strongly support the target specificity of CBS. In these studies, CBS only reduced proliferation of MEFs with intact cyclin E, suggesting that even if CBS does alter CaM-binding to proteins other than cyclin E, these off-target effects do not affect proliferation.
Our dose–response studies of TAT-CBS–treated human aortic ECs and SMCs in vitro reveal IC50 values that suggest similar antiproliferative potency in these cell types. However, adventitial delivery of TAT-CBS in vivo did not retard reendothelialization following injury. This discrepancy may be attributable to 2 factors. First, the ability of adventitial TAT-CBS to gain access to the luminal EC compartment may be less than its ability to penetrate VSMCs. Second, circulating EC progenitors have been shown to contribute to reendothelialization,25 and their attachment and spread to denuded areas may be insensitive to TAT-CBS.
Although TAT-CBS treatment was effective at inhibiting VSMC proliferation and neointima formation in mice, our peptide did not confer therapeutic benefits on other aspects of vascular remodelling such as collagen deposition and endothelial repair. Therefore, as a potential next generation agent in drug-eluting stents, treatment with CBS, by itself, may not be sufficient to prevent adverse remodeling in more complex large animal models or clinical cases of atherosclerosis. Combining the CBS peptide with therapies promoting endothelial repair and/or inhibiting extracellular matrix deposition may be required for synergistic benefits.
Based on our mouse carotid injury model data, we anticipate that the CBS peptide could be delivered to patients at the time of coronary intervention and continuously administered afterward as a novel agent in drug eluting stents. VSMC-targeted delivery of CBS peptide (or its next generation surrogates) may have the potential to selectively block rapidly proliferating VSMCs, while not interfering with reformation of the antithrombotic endothelial cell lining at the site of injury. For instance, a gene therapy method could be used in which a plasmid containing the CBS sequence is under the transcriptional control of a SMC-specific promoter. This approach has proven effective with the SM22-α promoter for SMC-specific transfection in a rat model of carotid injury.26 Additionally, virus retargeting techniques that modify surface proteins and moieties for binding to a cell surface receptor specific to VSMCs may further enhance delivery.27,–,29
When small molecules such as peptides bind to proteins in a critical region of protein–protein interaction, “complementarity” is altered and the association kinetics and thermodynamics of protein complex formation are disrupted. Despite this logic, creating a small molecule capable of blocking specific protein interactions has traditionally been difficult. However, successful antagonists of protein–protein interactions are currently being marketed and many more inhibitors are in various stages of development (reviewed30). Recently, a peptide inhibitor of NFAT (nuclear factor of activated T cells) was developed and observed to selectively inhibit NFAT-mediated proliferation and inflammation of VSMCs.31 Small peptides that block the interaction of cyclin A/CDK2 with substrates such as E2F1 have also been investigated in a number of tumor cell lines. These inhibitory peptides induced S-phase arrest and abrupt apoptosis. Cell death was selective to transformed cells; although a normal human fibroblast cell line did not undergo apoptosis, a T antigen–transformed subclone derived from it was killed.32 Similar to the CBS peptide, other molecules have been developed to inhibit the binding of CaM to its target proteins, such as ATPase33 and MLCK.34 The success of these related drugs supports further drug development based on CBS.
Sources of Funding
J.C. was supported in part by doctoral student stipend awards from the Ontario Graduate Scholarship (OGS) and Canadian Institute of Health Research (CIHR). S.H. and S.K.S. were supported in part by trainee awards from the Heart & Stroke Richard Lewar Centre of Excellence. M.H. is a recipient of a Career Investigator Award of the Heart & Stroke Foundation of Ontario (CI5503). This work was supported in part by a CIHR operating grant to M.H. (MOP14648) and a CIHR Proof-of-Principle grant to M.H. (82571).
J.C. and M.H. are coinventors of the CBS and have filed a provisional patent for its use as an antiproliferative agent. The authors have no other conflicts of interest to disclose.
We thank Keith Brunt and Anton Mihic for helpful discussions.
In January 2011, the average time from submission to first decision for all original research papers submitted to Circulation Research was 13.3 days.
Non-standard Abbreviations and Acronyms
- α-smooth muscle actin
- ionic calcium
- calmodulin binding sequence
- endothelial cell
- fluorescein isothiocyanate
- growth-arrested phase of cell cycle
- mouse embryonic fibroblast
- negative control
- proliferating cell nuclear antigen
- smooth muscle 22-α
- smooth muscle cell
- smooth muscle myosin heavy chain
- vascular smooth muscle cell
- Received December 20, 2010.
- Revision received February 17, 2011.
- Accepted February 23, 2011.
- © 2011 American Heart Association, Inc.
- Husain M,
- Bein K,
- Jiang L,
- Alper SL,
- Simons M,
- Rosenberg RD
- Afroze T,
- Yang LL,
- Wang C,
- Gros R,
- Kalair W,
- Hoque AN,
- Mungrue IN,
- Zhu Z,
- Husain M
- Gros R,
- Afroze T,
- You XM,
- Kabir G,
- Van Wert R,
- Kalair W,
- Hoque AE,
- Mungrue IN,
- Husain M
- Choi J,
- Chiang A,
- Taulier N,
- Gros R,
- Pirani A,
- Husain M
- You XM,
- Mungrue IN,
- Kalair W,
- Afroze T,
- Ravi B,
- Sadi AM,
- Gros R,
- Husain M
- Dong LH,
- Wen JK,
- Liu G,
- McNutt MA,
- Miao SB,
- Gao R,
- Zheng B,
- Zhang H,
- Han M
- Kibbe MR,
- Billiar TR,
- Tzeng E
- Dangas G,
- Kuepper F
- Serruys PW,
- Luijten HE,
- Beatt KJ,
- Geuskens R,
- de Feyter PJ,
- van den Brand M,
- Reiber JH,
- ten Katen HJ,
- van Es GA,
- Hugenholtz PG
- Hoffmann R,
- Mintz GS,
- Dussaillant GR,
- Popma JJ,
- Pichard AD,
- Satler LF,
- Kent KM,
- Griffin J,
- Leon MB
- Junker JP,
- Rief M
- Urbich C,
- Dimmeler S
- Yu H,
- Sliedregt-Bol K,
- Overkleeft H,
- van der Marel GA,
- van Berkel TJ,
- Biessen EA
- Chen YN,
- Sharma SK,
- Ramsey TM,
- Jiang L,
- Martin MS,
- Baker K,
- Adams PD,
- Bair KW,
- Kaelin WG Jr.
Novelty and Significance
What Is Known?
Tightly coordinated increases in intracellular calcium are required for cell cycle progression in vascular smooth muscle cells (VSMCs).
Calcium dependence of cell cycle in VSMCs is regulated in part by protein–protein interaction between the calcium signal-transducer calmodulin (CaM) and the cell cycle regulator cyclin E.
What New Information Does This Article Contribute?
A synthetic peptide corresponding to the calmodulin binding sequence (CBS) of cyclin E disrupts CaM–cyclin E interactions and inhibits calcium-dependent activation of the cyclin E–cell cycle kinase CDK2 complex in VSMCs.
The CBS peptide exerts a cyclin E–dependent inhibitory effect on cell cycle progression and cell proliferation.
Adventitial application of the CBS peptide following carotid artery denudation injury in the mouse prevented loss of contractile marker gene expression and inhibited VSMC proliferation and neointima formation in vivo without affecting collagen deposition or endothelial repair.
VSMC proliferation contributes to the adverse remodelling of atherosclerosis and restenosis. Elucidating mechanisms regulating cell cycle progression in VSMCs can form the basis of new antiproliferative therapies for these conditions. Here, we demonstrate the physiological significance of the CaM–cyclin E interaction and the relevance of calcium-sensitive cell cycle progression in VSMCs. A synthetic peptide based on the CaM-binding sequence (CBS) of cyclin E disrupted CaM–cyclin E interactions and inhibited cell cycle progression and proliferation of mouse and human VSMCs in vitro. The CBS peptide significantly inhibited proliferation of VSMCs and neointima formation following carotid wire injury in vivo without affecting matrix deposition or endothelial repair. What distinguishes the CBS peptide from agents used in existing drug-eluting stents for restenosis is its focused mechanism of action. CBS inhibits calcium-sensitive but not baseline cyclin E/CDK2 activity and is unable to inhibit proliferation of cells lacking cyclin E. Because existing drugs for restenosis broadly affect mTOR signaling (eg, sirolimus) or microtubule stability (paclitaxel), their toxicity includes impaired endothelial function and repair, which pose long-term risks. The present study identifies a novel antiproliferative agent that prevents adverse vascular remodeling without inducing cell death or loss of contractile marker gene expression.