Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Arterioscler Thromb Vasc Biol. Author manuscript; available in PMC 2014 January 1.
Published in final edited form as:
PMCID: PMC3569043

Loss of CDKN2B promotes p53-dependent smooth muscle cell apoptosis and aneurysm formation



Genome wide association studies have implicated allelic variation at 9p21.3 in multiple forms of vascular disease, including atherosclerotic coronary heart disease and abdominal aortic aneurysm. As for other genes at 9p21.3, human eQTL studies have associated expression of the tumor suppressor gene CDKN2B with the risk haplotype, but its potential role in vascular pathobiology remains unclear.

Methods and Results

Here we employed vascular injury models and found that Cdkn2b knockout mice displayed the expected increase in proliferation after injury, but developed reduced neointimal lesions and larger aortic aneurysms. In situ and in vitro studies suggested that these effects were due to increased smooth muscle cell apoptosis. Adoptive bone marrow transplant studies confirmed that the observed effects of Cdkn2b were mediated through intrinsic vascular cells and were not dependent on bone marrow-derived inflammatory cells. Mechanistic studies suggested that the observed increase in apoptosis was due to a reduction in MDM2 and an increase in p53 signaling, possibly due in part to compensation by other genes at the 9p21.3 locus. Dual inhibition of both Cdkn2b and p53 led to a reversal of the vascular phenotype in each model.


These results suggest that reduced CDKN2B expression and increased SMC apoptosis may be one mechanism underlying the 9p21.3 association with aneurysmal disease.

Keywords: CDKN2B, apoptosis, smooth muscle, remodeling, abdominal aortic aneurysm, genome wide association studies, p53

As much as half of the risk for atherosclerotic coronary heart disease (CHD) is genetic in nature1, 2. Through genome-wide association (GWA) studies, approximately 30 loci have now been associated with CHD 3, 4. To date, the most robust GWA finding for CHD is a group of highly correlated variants in an ~ 58 kilobase region on chromosome 9 at p21.3, the Chromosome 9p21.3 CHD-Associated Region (C9CAR)5, 6. C9CAR has been associated with atherosclerosis burden and myocardial infarction4, 7, 8, as well as extracardiac phenotypes such as abdominal aortic aneurysm (AAA), peripheral arterial disease, and stroke9-12. As much as 20% of the attributable risk for coronary heart disease is contributed by variation at C9CAR5. The simultaneous association of 9p21.3 variation with non-atherosclerotic berry aneurysms suggests that the unifying mechanisms of association may not occur via a classical inflammatory pathway, but rather through some process that governs the structural makeup of the diseased vessel wall11.

To date, the causal gene(s) at 9p21.3 remain unclear. The closest genes to C9CAR include a group of three cancer-related factors that reside over 50 kilobases telomeric to the risk associated region. Two of these genes, CDKN2A and CDKN2B, are cyclin-dependent kinase inhibitors which interact with CDK4 and CDK6 and have been linked to cell cycle regulation. The third is a unique splice variant of CDKN2A termed p14/ARF that is involved in regulation of apoptosis through interactions with p5313. All of these genes are considered tumor suppressors, and are commonly silenced by methylation in a variety of cancers13, 14. Both CDKN2A and CDKN2B have been implicated in biological processes such as senescence, apoptosis, stem cell renewal, and adult onset diabetes13. CDKN2A and ARF have been linked to CHD through gene expression and animal models of experimental atherosclerosis15-17, but no study has yet examined the effects of genes near 9p21 on aneurysm formation. Because several recent genetics-of-gene-expression studies have associated C9CAR variants with altered expression of CDKN2B17-20 (and/or its antisense long noncoding RNA, ANRIL)20, 21, we employed a series of in vitro and in vivo murine vascular disease models to investigate the mechanisms by which CDKN2B might regulate heritable vascular risk, with a particular focus on AAA disease.


Murine vascular disease models

All animal protocols used were approved by the Stanford University Administrative Panel on Laboratory Animal Care. All experiments were performed in 12-14 week old male Cdkn2b+/+ (n = 107) and Cdkn2b-/- (n = 85) mice on a C57BL/6 background.

Negative vascular remodeling was induced by performing complete carotid artery ligation (CAL). In some experiments, mice were injected with 2.2 mg/kg intraperitoneal pifithrin-α one hour prior to ligation, and then every 48 hours until sacrifice, as above. Animals were euthanized 2, 4, 7, 14 or 28 days after the surgery and both the ligated- and non-ligated carotid artery samples were harvested for either RNA, total protein, or histomorphometric analysis as described in the Supplementary Methods.

Porcine pancreatic elastase (PPE) infusion was used to generate aortic aneurysms. Abdominal aortic diameter (AAD) was measured at baseline and 7 and 14 days after aneurysm induction, with B-mode ultrasound (US) imaging. In some experiments, mice were injected with 2.2 mg/kg intraperitoneal pifithrin-α one day prior to surgery, and then every 48 hours until sacrifice. Animals were euthanized 5 or 14 days after surgery, and analyzed as described in the Supplementary Methods.

Reciprocal bone marrow transplant studies were performed in lethally irradiated mice, as described in the Supplementary Methods. After 14 days of recovery, engraftment was confirmed by flow cytometry and the mice were subjected to PPE infusion, as above.

Human vascular sample acquisition and preparation

Human aortic samples were taken from patients during open surgical AAA repair (n =13), or from organ donors at the time of explant (n = 5) and subjected to quantitative real time reverse transcriptase polymerase chain reaction mRNA expression analysis. Immunohistochemical analyses were performed on aneurysmal (n = 7) and non-aneurysmal aortic sections (n = 7) from a second confirmation cohort.

Cell culture methods

Human coronary, pulmonary and aortic smooth muscle cells (SMC) (Lonza) were grown in media provided by the supplier. Endothelial cells, monocytes, macrophages and fibroblasts were studied as described in the Supplementary Methods. To induce growth arrest and the expression of differentiation genes, SMC were serum starved in basal media (SmBM) for 72 hours. In vitro p53 inhibition studies were done by adding 10μM pifithrin-α (Calbiochem) to the cell culture media. For knockdown experiments, HCASMC were transfected with 300 nM of anti-CDKN2B or high-GC negative control siRNA (Ambion) using the Amaxa Nucleofector system (Lonza).

Cellular proliferation was quantified with a modified MTT assay as well as cell counting and FACS analysis. To analyze SMC migration, a modified Boyden chamber assay was performed.

Rates of programmed cell death were assessed with three independent in vitro assays. In each assay, 1×105 transfected HCASMC were treated with 1 μM staurosporine (Sigma) in serum free media for 6 hours prior to analysis. Caspase-3 and -7 activity was measured using a commercially available luminometric assay (Promega). In the second assay, the cells were FACS sorted for FITC annexin V and propidium iodide (BD FACSCaliber). In the final assay, cells were fixed in 10% formalin prior to TUNEL staining with the Cell Death Detection Kit (Roche).

Phopsho-proteomic antibody microarray profiling and bioinformatics analyses were performed in CDKN2B-deficient cells, as described in the Supplementary Methods.

Expression analysis

Standard methodology was applied for mRNA and protein extraction from cell lysates and vascular samples, as well as for subsequent gene expression analysis and Western blotting, as described in the Supplementary Methods.

Statistical analysis

Data are presented as mean ± SEM. Data were subjected to the Kolmogorov-Smirnov test to determine distribution. Groups were compared using the Mann-Whitney U test for non-parametric data or the Students t-test for parametric data. When comparing multiple groups, data were analyzed by analysis of variance with Bonferroni's post-test. Statistical analysis was performed with GraphPad Prism 5.

Additional methods

Detailed methodology and primary citations are provided in the Supplementary Methods.


Cdkn2b regulates both negative vascular remodeling and aneurysm formation in vivo

To investigate the impact of a loss of Cdkn2b in vivo, we studied the CAL injury model and the elastase-infusion AAA model in Cdkn2b-/- and Cdkn2b+/+ mice. No difference in blood pressure, heart rate, lipid level or glucose level was observed between genotypes (Suppl. Fig. I).

In the CAL model, Cdkn2b-/- mice exhibited reduced neointimal areas (41.4% reduction, P<0.03), medial areas (10.0% reduction, P<0.03), and intimal-to-medial (IM) ratios (37.2% reduction, P<0.05), compared to control Cdkn2b+/+mice four weeks after vascular injury (Fig. 1A). No measurable differences in luminal area (P = 0.24) or total vessel area (P = 0.15) were observed. Cdkn2b-/- mice displayed accelerated dropout of SMC, and had less vascular SMA staining than Cdkn2b+/+ animals (31.6% reduction, P<0.03), as well as fewer total neointimal cells, as assessed by number of DAPI-positive nuclei (212.9 fewer cells/vessel, P<0.03) (Figs. 1B-1C).

Figure 1
Cdkn2b regulates murine vascular disease. (A) Cdkn2b-/- mice display alterations in vascular remodeling in the carotid ligation model 28 days after vascular injury. Compared to Cdkn2b+/+ mice (right, n=15), Cdkn2b-/- mice (left, n=19) have smaller neointimal ...

In the AAA model, Cdkn2b-/- mice were found to develop significantly larger aortic aneurysms than Cdkn2b+/+control mice at both the one week (44.4% larger, P< 0.01) and two week (29.3% larger, P<0.01) timepoints following elastase infusion (Fig. 1E). In keeping with the CAL model findings above, Cdkn2b-/-mice had significantly less SMA staining (46.0% reduction, P<0.05, Fig. 1F). No significant differences in elastin degradation score (P = 0.23, Fig. 1G), Mac-3 (P = 0.24), CD-3 (P = 0.87), or collagen staining (P = 0.92) were observed between genotypes (data not shown).

In both models, the Cdkn2b-/- mice displayed increased cellular proliferation in response to vessel injury. In the CAL model, significant differences were observed one week (3.81 fold increase, P<0.01) and four weeks (5.65 fold increase, P<0.0001, Fig. 1D) after ligation. In the AAA model, there was a marked increase in proliferation five days after elastase infusion (1.66 fold increase, P<0.01, Fig. 1H) as measured by the number of PCNA positive vascular cells.

Cdkn2b mediates its effects through resident vascular cells independent of bone marrow derived cells

To confirm that Cdkn2b was mediating its effect solely through intrinsic blood vessel cells and not via an inflammatory mechanism, we next performed elastase infusions in mice that had been lethally irradiated and undergone adoptive bone marrow transplantation (BMT) (Suppl. Fig. II). Cdkn2b-/- mice reconstituted with wild type marrow developed significantly larger AAA's than Cdkn2b+/+ mice that had been reconstituted with either Cdkn2b-/- or Cdkn2b+/+ marrow (176.9% vs.155.1% and 145.8% aortic diameter increase, respectively, P<0.05, Fig. 2A). Only the Cdkn2b-/- recipients displayed a reduction in SMA staining, confirming an effect of this gene on resident SMC survival (58.0% reduction relative to Cdkn2b+/+/Cdkn2b+/+ mice, P<0.03). No difference in T cell or macrophage infiltrate was observed across genotypes in either the CAL or elastase model (P=NS for each, data not shown).

Figure 2
Cdkn2b regulates apoptosis in vivo and potentiates disease through intrinsic vessel wall cells. (A) Bone marrow transplant studies revealed that Cdkn2b's vascular effects are not mediated through bone marrow-derived cells, with enhanced aneurysms only ...

Cdkn2b regulates vascular disease in vivo by inhibiting programmed cell death

To explain how the knockout animals could have enhanced proliferation yet develop smaller neointimal lesions and aneurysms with fewer SMC, we next evaluated the rates of apoptosis in each genotype. The Cdkn2b-/- mice in the CAL model displayed a striking increase in apoptosis observed as early as two days after injury (3.32 fold increase, P<0.03, Fig. 2B) that persisted one week post ligation (1.96 fold increase, P<0.05). Also, the Cdkn2b-/- mice displayed increased aortic apoptosis relative to Cdkn2b+/+ mice five days after elastase infusion (4.64 fold increase, P<0.0001) in the AAA model (Fig. 2C). In each model, the differences were no longer observed at the terminal timepoint, several weeks after injury (P>0.22 for each).

CDKN2B is expressed in vascular SMC and is downregulated in diseased vascular tissue

To define the localization of CDKN2B in human vascular tissue and understand its regulation in vascular disease, we investigated the expression of CDKN2B in normal and pathological vascular samples. Compared to explants from organ donors, aortic samples taken from patients undergoing open AAA repair showed markedly reduced CDKN2B mRNA levels (4.8 fold reduction, P<0.03, Suppl. Fig. IIIA). Similarly, studies revealed that Cdkn2b expression was decreased in an animal model of vascular injury, with a three-fold reduction observed in murine carotid arterial tissue two weeks after carotid ligation (P<0.01). In keeping with these findings, we found that CDKN2B was expressed at a high level in cultured SMC at baseline (in contrast to cultured endothelial cells and macrophages) and was significantly downregulated as cells assumed the de-differentiated phenotype (7.2 fold reduction, P<0.01).

Immunolocalization of CDKN2B expressing cells was performed with human aortic sections from a separate AAA cohort, and confirmed that CDKN2B was highly expressed in medial vascular smooth muscle cells in vivo, but decreased in aneurysm tissue (Suppl. Fig. IIIB). Semi-quantitative analysis of these data confirmed the mRNA expression findings, revealing that CDKN2B was highly downregulated in aneurysmal compared to non-aneurysmal tissue (P<0.01). Further staining in human vascular sections revealed colocalization of CDKN2B and smooth muscle α-actin (SMA) staining. Colocalization of CDKN2B and vWF expression was present but less intense (Suppl. Fig. IIIC).

CDKN2B regulates proliferation, migration and apoptosis in cultured SMC

Given findings implicating the tunica media as the site of action of CDKN2B, we next evaluated the role of this gene in vascular SMC physiology in vitro. HCASMC were rendered deficient in CDKN2B through treatment with siRNA (10.5 fold knockdown, P<0.01). These cells revealed a significantly higher rate of proliferation than control transfected cells, as measured using an MTT assay (0.028 vs. 0.019 relative proliferation units, P<0.01, Fig. 3A) and hemocytometer cell counts (25.9% increase, P<0.03). Consistent with these findings, cells deficient in CDKN2B were also found to have higher rates of mitosis by FACS analysis, as evidenced by a higher frequency of G2/M phase cells (4.6 vs. 3.5%, P<0.01, Fig. 3B). Also, siRNA induced suppression of CDKN2B was associated with significant enhancement of SMC migration in Boyden Chamber assays (15.4 vs. 9.6 cells/hpf, P<0.0001, Fig. 3C). In keeping with the in vivo data, CDKN2B was then also found to inhibit programmed cell death using three independent assays. Both the percentage of TUNEL-stained apoptotic cells grown on chamber slides (38.5 vs. 13.2%, P<0.01, Fig. 3D), and the caspase-3/7 activity (13,414 vs. 6,715 relative light units, P<0.0001, Fig. 3E) were significantly increased in CDKN2B-deficient cells relative to control cells. Similarly, CDKN2B knockdown increased the percentage of cells stained for annexin V compared to control-transfected cells (65.8 vs. 46.3%, P<0.05) after exposure to the proapoptotic global kinase inhibitor staurosporine (Fig. 3F). These apoptotic differences were confirmed in HAoSMC (P<0.01, data not shown). CDKN2B was not expressed in cultured macrophages, did not modulate monocyte differentiation and did not regulate apoptosis in cultured fibroblasts (P=NS, data not shown). Knockdown of CDKN2B had no measurable effect on the expression of SMC-differentiation markers (Suppl. Fig. IV), inflammatory cytokines or extracellular matrix collagens in SMC (P= NS for each, data not shown).

Figure 3
CDKN2B regulates SMC functions in cultured human vascular smooth muscle cells. HCASMC deficient in CDKN2B (siCDKN2B) have higher rates of proliferation than control cells (siControl) as assessed by (A) MTT assay, and (B) percentage of cells in G2/M phase ...

9p21 genes are regulated in response to stress

Because compensatory ‘back-up’ roles have been described for genes at the 9p21 locus previously14, we next evaluated the expression of each of the genes near C9CAR in response to stress. A pattern of reduced CDKN2B with increased CDKN2A and Arf expression was observed both in vitro and in vivo after injury (control cells treated with an apoptotic stimuli (grey bars, Fig. 4A) and wild type carotid tissue after ligation (grey bars, Fig. 4B)). MTAP fell after injury in vivo, but not in vitro. When evaluating the changes which occurred during conditions of CDKN2B-deficiency, a similar pattern was observed (compare black bars to grey bars in Fig. 4A and 4B), with the exception of Cdkn2a which was elevated in knockout mice at baseline and did not augment in response to stress. A direct comparison of the relative responses to stress across genotypes revealed a significantly larger increase in the pro-apoptotic factor Arf in Cdkn2b-/- mice compared to Cdkn2b+/+ mice after carotid injury (P <0.05, Fig. 4B), but only a trend in vitro (P =0.07). The relative compensation of all other 9p21-locus genes was similar across genotypes (P>0.05 for each).

Figure 4
Compensation occurs at the 9p21 locus in response to stress and loss of CDKN2B promotes the activity of apoptotic factors, including MDM2 and p53. (A) In response to an apoptotic stimulus, control-transfected HCASMC (grey bars) display reduced expression ...

CDKN2B regulates SMC apoptosis via an MDM2 and p53-dependent mechanism

To further investigate the mechanism by which CDKN2B regulates SMC survival, we next evaluated the expression of apoptosis-regulating genes in HCASMC lacking CDKN2B and in the injured vessel wall of CAL mice lacking Cdkn2b. Western blot analysis revealed that CDKN2B knockdown in HCASMC resulted in increased expression of p53 (2.9 fold increase, P<0.03) and its downstream product p21 (3.7 fold increase, P<0.05) as compared to control transfected cells (Fig. 4C). These differences were associated with a concomitant increase in the pro-apoptotic factor BAX (6.3 fold increase, P = 0.18) and a decrease in the anti-apoptotic factor BCL2 (1.8 fold decrease, P = 0.056), although these changes did not reach statistical significance. Modulation of CDKN2B in vitro did not significantly alter total RB protein levels (1.9 fold increase, P = 0.11). In the CAL model, Cdkn2b-/- mice showed a significant increase inTrp53 (2.8 fold increase, P<0.03), p21 (3.1 fold increase, P<0.05), p19/Arf (2.8 fold increase, P<0.03) and Bax (2.2 fold increase, P<0.03), with a non-significant decrease in Bcl2 mRNA expression relative to Cdkn2b+/+ ligated vessels, confirming the cell-based findings (Fig. 4D).

To further study the mechanism by which a deficiency in CDKN2B resulted in enhanced p53 expression, we performed a phospho-antibody protein microarray analysis of 196 factors related to the p53 signaling and apoptotic pathways. Intensity signals representing protein levels in apoptosing CDKN2B-deficient and control-transfected cells were evaluated with Significance Analysis of Microarrays, and the genes responsible for differentially regulated proteins (FDR <1%) were evaluated by overabundance analysis using DAVID (P<0.05, 2 genes/category minimum). Enriched KEGG pathways revealed that CDKN2B regulates genes related to SMC function, cell cycling, malignancy and, as predicted, the p53 pathway (Fig. 4E). Interestingly, the MDM2 protein was the third most significantly regulated candidate on the array (ahead of p53 itself), suggesting that CDKN2B may regulate apoptosis by targeting an MDM2-stabilizing kinase to enhance p53 ubiquitination and/or degradation. Subsequent flow cytometry and western blot studies confirmed the array-based findings and revealed that apoptosing CDKN2B-deficient cells expressed significantly lower levels of both total MDM2 and phospho-MDM2 than control-transfected cells, explaining the higher levels of p53 in the knockdown cells (P<0.01 for each, Figs 4F, 4G).

Dual inhibition studies confirm the role of p53 in CDKN2B-mediated apoptosis and vascular remodeling

Finally, to determine if the observed CDKN2B-related apoptotic differences were dependent on p53 signaling, we performed dual targeting studies that included the pharmacological p53 inhibitor, pifithrin-α. Simultaneous knockdown of p53 augmented the enhanced proliferation in CDKN2B deficient HCASMC based on FACS analyses, with the ratio of cells in G2/M increasing from 1.28 at baseline to 1.66 after pifithrin-α treatment (Fig. 5A). In contrast, inactivation of p53 eliminated the difference in apoptosis between CDKN2B knockdown and control transfected cells, as assessed by caspase 3/7 activity, TUNEL staining, and FACS analysis of annexin V positivity (Figs. 5B-5D). Cdkn2b-/- mice administered pifithrin-α parenterally no longer showed increased rates of apoptosis after carotid ligation (Fig. 5E) and actually exhibited a reversal of the remodeling phenotype after CAL, with an increase in the ratio of neointimal areas and I/M ratios (Fig. 5F). Similar results were observed in the AAA model, where the rate of aneurysm expansion was no longer different between genotypes after pifithrin-α treatment (P>0.71 at each timepoint, Fig 5G).

Figure 5
CDKN2B's effects on remodeling and apoptosis are dependent on p53. In vitro, inhibition of both CDKN2B and p53 eliminated the difference in apoptosis while enhancing the difference in proliferation. (A) Inhibition of p53 with pifithrin-α accentuated ...


The present study provides a hypothesis for how CDKN2B might contribute to heritable cardiovascular risk. The data suggest that it does so, at least in part, by inducing multiple functional alterations in vascular SMC that could impact disease development, progression, and/or end stage clinical consequences. First, we show that reduced Cdkn2b expression in murine vascular remodeling and AAA models accelerates SMC proliferation while paradoxically leading to both smaller neointimal lesions and larger aortic aneurysms. Second, we show that these changes are secondary to an increase in vascular apoptosis, which is the result of an interaction between CDKN2B and the MDM2-p53 pathway, possibly due at least in part to upregulation of ARF in the presence of vascular injury. These CDKN2B-related effects on programmed cell death appear to be the major determinant of the observed vascular phenotype, have the capacity to overwhelm the concomitant proliferative differences, occur through an effect on resident vascular cells, and can be fully reversed by simultaneously inhibiting both the p53 and CDKN2B pathways. Although further investigation is required in additional models, these studies suggest one mechanism for how a gene which is regulated by polymorphisms at C9CAR may potentiate risk for abdominal aortic aneurysm 5, 22.

The identity of the causal vascular disease gene(s) at 9p21 remains the topic of significant debate. To date, a number of expression quantitative trait locus (eQTL) and allelic imbalance studies have been performed in human tissues, and altered expression of each of the genes (alone or in combination) near the 9p21 locus has been described in carriers of the risk allele, including CDKN2A, ARF, ANRIL and MTAP17, 20, 23-25. For CDKN2B, eQTL studies have associated non-coding C9CAR risk SNPs with reduced CDKN2B expression in man, including adipose tissue and circulating leukocytes4, 17. Importantly, recent studies have extended these mapping efforts into vascular tissue, and found reduced CDKN2B expression in atherosclerotic plaque and vascular smooth muscle cells derived from carriers of the risk allele, consistent with association in the end organ18, 19. At a mechanistic level, much of the observed reduction in CDKN2B expression may be occurring indirectly through the local long non-coding RNA, ANRIL, which has been shown to epigenetically suppress CDKN2B transcription26, 27, and has been associated with C9CAR allelic variation20, 21, 23. Indeed, a recent study revealed that a lead C9CAR polymorphism ablates a STAT1 binding site in a 9p21 enhancer element, resulting in enhanced ANRIL expression, and presumably reduced CDKN2B in the vessel wall28. Finally, recent association studies in African Americans have identified risk variants within the CDKN2B 3′-UTR29, and one of these variants has now been independently associated in a large meta-analysis of 63,746 CHD cases and 130,681 controls by the CARDIoGRAM+C4D consortium30.

The observation that CDKN2B is strongly linked to SMC apoptosis may explain the association of 9p21 with both AAA disease and non-atherosclerotic intracranial aneurysms. A central feature of these diseases, and for berry aneurysms in particular, is a decrease in vascular wall integrity related to loss of vascular wall medial SMC number31. Aneurysmal dilatation has also been associated with SMC apoptosis and p53 upregulation in human aneurysms32. For carriers of the C9CAR risk haplotype, pathological downregulation of CDKN2B would enhance p53-dependent apoptosis, and ultimately promote medial thinning. Because vascular SMC turnover occurs at an exceedingly low rate33, it is possible that even a slight imbalance in the ratio of proliferation to apoptosis might have a dramatic impact on AAA progression, particularly when considered over a period of several decades 34, 35.

An important point is that the effects observed here are mediated solely through intrinsic vessel-wall cells, and not bone-marrow derived macrophages. This consideration is important given that prior studies have implicated Cdkn2b in monocytic myeloproliferation36 and previous observations that most cardiovascular diseases are inflammatory in nature34. However, the link between risk haplotype and the non-atherosclerotic berry aneurysm points to a mechanism which governs blood vessel composition over one that promotes vascular inflammation. Indeed, this hypothesis was borne out by our observation that the Cdkn2b knockout mice displayed a severe phenotype in each animal model, with no increase in inflammatory cell burden, as well as a null effect after transplanting knockout marrow into otherwise healthy animals. The fact that Cdkn2b does not alter vascular inflammation may explain its independence from classical risk factors such as smoking, dyslipidemia and diabetes as these processes appear to primarily work through inflammatory pathways, rather than medial cell fate decision making37.

Apoptosis-related signaling downstream of CDKN2B appears to be more complex than previously appreciated. In health, both CDKN2B and CDKN2A are well described inhibitors of tumor formation that signal through the retinoblastoma pathway to inhibit the G1 to S transition. In human cancer, these two genes are coordinately hypermethylated or deleted, allowing tumor growth13, 14. In the vasculature, CDKN2B similarly regulates cell-fate decisions, but appears to have an even greater role in the modulation of apoptosis in response to stress. Rather than doing so through the RB pathway, a loss of CDKN2B results in activation of the p53 axis and several downstream effector molecules. p53 has been implicated in vascular remodeling previously38, but its role in atherosclerotic plaque progression is the subject of significant debate39, 40. There are several possible mechanisms by which signaling between CDKN2B and p53 may occur (Suppl. Fig. IV). In keeping with extensive literature linking ARF to p53, one possible mechanism identified through these studies is the upregulation of vascular ARF expression in response to reduced CDKN2B expression. These data would suggest that either direct modulation of ARF by causal variation at C9CAR, or indirect modulation of ARF through intermediate regulation of CDKN2B could provide a common disease pathway. An argument against a sole role for ARF in the context of CDKN2B down-regulation is provided by data obtained in an Arf knockout mouse in the ApoE model15. In these studies mice specifically targeted for Arf, and shown to have no Cdkn2b compensation, did not display any change in lesion cellularity, aortic SMC proliferation or SMC content. This suggests that CDKN2B itself may have a distinct role in SMC biology above and beyond the effects of ARF. Also, direct crosstalk between RB, MDM2 and p53 is known to exist in other cell types41, 42, suggesting further investigation of direct interaction between CDKN2B and the kinases upstream of p53 in SMC.

This study has several limitations that warrant discussion. First, because this study employed non-atherosclerotic models, it is difficult to draw inferences about a potential role for CDKN2B in coronary artery disease. The CAL model is useful to measure the SMC's response to injury, but does not reflect human plaque formation. Also, Kim and colleagues recently reported that ApoE*Leiden/CDKN2B mice do not display an increase in atherosclerotic plaque burden relative to controls43, raising the possibility that the 9p21-locus genes might have context-specific vascular effects, as has been described for p53 previously39. Indeed, it is possible that multiple cis-regulatory elements, compensation patterns, or gene-by-gene interactions may be active in different disease states with more than one ‘causal’ gene at 9p21.

In summary, this report provides a hypothesis for how variation at 9p21.3 may regulate SMC survival to promote risk for vascular disease. These findings highlight the concept of medial integrity as a determinant of disease progression and rationalize the development of antiapoptotic therapies directed towards reducing aneurysm progression.

Supplementary Material


The authors gratefully acknowledge Linda Wolff of the National Cancer Institute at the National Institutes of Health for providing the Cdkn2b knockout mice and Jan Lindeman for providing human aortic sections used in this study.


This work was supported by NIH grants R01HL103635 (TQ), R01 HL38854 and R01 HL57353 (GKO), K12 HL087746 and K08 HL103605-01 (NJL), AHA grant 10BGIA3290011 (NJL) and support from the LeDucq Foundation (TQ).


Disclosures - None.


1. Lloyd-Jones DM, Nam BH, D'Agostino RB, Sr, Levy D, Murabito JM, Wang TJ, Wilson PW, O'Donnell CJ. Parental cardiovascular disease as a risk factor for cardiovascular disease in middle-aged adults: A prospective study of parents and offspring. JAMA. 2004;291:2204–2211. [PubMed]
2. Marenberg ME, Risch N, Berkman LF, Floderus B, de Faire U. Genetic susceptibility to death from coronary heart disease in a study of twins. N Engl J Med. 1994;330:1041–1046. [PubMed]
3. Musunuru K, Kathiresan S. Genetics of coronary artery disease. Annu Rev Genomics Hum Genet. 2010;11:91–108. [PubMed]
4. Schunkert H, Konig IR, Kathiresan S, et al. Large-scale association analysis identifies 13 new susceptibility loci for coronary artery disease. Nat Genet. 2011;43:333–338. [PMC free article] [PubMed]
5. Helgadottir A, Thorleifsson G, Manolescu A, et al. A common variant on chromosome 9p21 affects the risk of myocardial infarction. Science. 2007;316:1491–1493. [PubMed]
6. McPherson R, Pertsemlidis A, Kavaslar N, Stewart A, Roberts R, Cox DR, Hinds DA, Pennacchio LA, Tybjaerg-Hansen A, Folsom AR, Boerwinkle E, Hobbs HH, Cohen JC. A common allele on chromosome 9 associated with coronary heart disease. Science. 2007;316:1488–1491. [PMC free article] [PubMed]
7. O'Donnell CJ, Kavousi M, Smith AV, et al. Genome-wide association study for coronary artery calcification with follow-up in myocardial infarction. Circulation. 2011;124:2855–2864. [PMC free article] [PubMed]
8. Kathiresan S, Voight BF, Purcell S, et al. Genome-wide association of early-onset myocardial infarction with single nucleotide polymorphisms and copy number variants. Nat Genet. 2009;41:334–341. [PMC free article] [PubMed]
9. Cluett C, Tanaka T, Kisialiou A, Guralnik J, Ferrucci L, Frayling TM, Murray A, Melzer D. The rs1333049 snp near cdkn2a/2b, associated with myocardial infarction, is associated with increased risk of peripheral artery disease in older people. American Society of Human Genetics. 2008;2:347–353.
10. Matarin M, Brown WM, Singleton A, Hardy JA, Meschia JF. Whole genome analyses suggest ischemic stroke and heart disease share an association with polymorphisms on chromosome 9p21. Stroke. 2008;39:1586–1589. [PubMed]
11. Helgadottir A, Thorleifsson G, Magnusson KP, et al. The same sequence variant on 9p21 associates with myocardial infarction, abdominal aortic aneurysm and intracranial aneurysm. Nat Genet. 2008;40:217–224. [PubMed]
12. Bown MJ, Braund PS, Thompson J, London NJ, Samani NJ, Sayers RD. Association between the coronary artery disease risk locus on chromosome 9p21.3 and abdominal aortic aneurysm. Circ Cardiovasc Genet. 2008;1:39–42. [PubMed]
13. Gil J, Peters G. Regulation of the ink4b-arf-ink4a tumour suppressor locus: All for one or one for all. Nat Rev Mol Cell Biol. 2006;7:667–677. [PubMed]
14. Krimpenfort P, Ijpenberg A, Song JY, van der Valk M, Nawijn M, Zevenhoven J, Berns A. P15ink4b is a critical tumour suppressor in the absence of p16ink4a. Nature. 2007;448:943–946. [PubMed]
15. Gonzalez-Navarro H, Abu Nabah YN, Vinue A, Andres-Manzano MJ, Collado M, Serrano M, Andres V. P19(arf) deficiency reduces macrophage and vascular smooth muscle cell apoptosis and aggravates atherosclerosis. J Am Coll Cardiol. 2010;55:2258–2268. [PubMed]
16. Kuo CL, Murphy AJ, Sayers S, Li R, Yvan-Charvet L, Davis JZ, Krishnamurthy J, Liu Y, Puig O, Sharpless NE, Tall AR, Welch CL. Cdkn2a is an atherosclerosis modifier locus that regulates monocyte/macrophage proliferation. Arterioscler Thromb Vasc Biol. 2011;31:2483–2492. [PMC free article] [PubMed]
17. Liu Y, Sanoff HK, Cho H, Burd CE, Torrice C, Mohlke KL, Ibrahim JG, Thomas NE, Sharpless NE. Ink4/arf transcript expression is associated with chromosome 9p21 variants linked to atherosclerosis. PLoS One. 2009;4:e5027. [PMC free article] [PubMed]
18. Pilbrow AP, Folkersen L, Pearson JF, Brown CM, McNoe L, Wang NM, Sweet WE, Tang WH, Black MA, Troughton RW, Richards AM, Franco-Cereceda A, Gabrielsen A, Eriksson P, Moravec CS, Cameron VA. The chromosome 9p21.3 coronary heart disease risk allele is associated with altered gene expression in normal heart and vascular tissues. PLoS One. 2012;7:e39574. [PMC free article] [PubMed]
19. Motterle A, Pu X, Wood H, Xiao Q, Gor S, Liang Ng F, Chan K, Cross F, Shohreh B, Poston RN, Tucker AT, Caulfield MJ, Ye S. Functional analyses of coronary artery disease associated variation on chromosome 9p21 in vascular smooth muscle cells. Hum Mol Genet. 2012 [PMC free article] [PubMed]
20. Cunnington MS, Santibanez Koref M, Mayosi BM, Burn J, Keavney B. Chromosome 9p21 snps associated with multiple disease phenotypes correlate with anril expression. PLoS Genet. 2010;6:e1000899. [PMC free article] [PubMed]
21. Jarinova O, Stewart AF, Roberts R, Wells G, Lau P, Naing T, Buerki C, McLean BW, Cook RC, Parker JS, McPherson R. Functional analysis of the chromosome 9p21.3 coronary artery disease risk locus. Arterioscler Thromb Vasc Biol. 2009;29:1671–1677. [PubMed]
22. Ripatti S, Tikkanen E, Orho-Melander M, et al. A multilocus genetic risk score for coronary heart disease: Case-control and prospective cohort analyses. Lancet. 2010;376:1393–1400. [PMC free article] [PubMed]
23. Holdt LM, Beutner F, Scholz M, Gielen S, Gabel G, Bergert H, Schuler G, Thiery J, Teupser D. Anril expression is associated with atherosclerosis risk at chromosome 9p21. Arterioscler Thromb Vasc Biol. 2010;30:620–627. [PubMed]
24. Jarinova O, Stewart AF, Roberts R, Wells G, Lau P, Naing T, Buerki C, McLean BW, Cook RC, Parker JS, McPherson R. Functional analysis of the chromosome 9p21.3 coronary artery disease risk locus. Arterioscler Thromb Vasc Biol. 2009;29:1671–1677. [PubMed]
25. Holdt LM, Sass K, Gabel G, Bergert H, Thiery J, Teupser D. Expression of chr9p21 genes cdkn2b (p15(ink4b)), cdkn2a (p16(ink4a), p14(arf)) and mtap in human atherosclerotic plaque. Atherosclerosis. 2011;214:264–270. [PubMed]
26. Yu W, Gius D, Onyango P, Muldoon-Jacobs K, Karp J, Feinberg AP, Cui H. Epigenetic silencing of tumour suppressor gene p15 by its antisense rna. Nature. 2008;451:202–206. [PMC free article] [PubMed]
27. Kotake Y, Nakagawa T, Kitagawa K, Suzuki S, Liu N, Kitagawa M, Xiong Y. Long non-coding rna anril is required for the prc2 recruitment to and silencing of p15(ink4b) tumor suppressor gene. Oncogene. 2011;30:1956–1962. [PMC free article] [PubMed]
28. Harismendy O, Notani D, Song X, Rahim NG, Tanasa B, Heintzman N, Ren B, Fu XD, Topol EJ, Rosenfeld MG, Frazer KA. 9p21 DNA variants associated with coronary artery disease impair interferon-gamma signalling response. Nature. 2011;470:264–268. [PMC free article] [PubMed]
29. Kral BG, Mathias RA, Suktitipat B, Ruczinski I, Vaidya D, Yanek LR, Quyyumi AA, Patel RS, Zafari AM, Vaccarino V, Hauser ER, Kraus WE, Becker LC, Becker DM. A common variant in the cdkn2b gene on chromosome 9p21 protects against coronary artery disease in americans of african ancestry. J Hum Genet. 2011;56:224–229. [PMC free article] [PubMed]
30. CARDIoGRAMplusC4D C. Coronary artery disease risk loci identified in over 190,000 individuals implicate lipid metabolism and inflammation as key causal pathways. Nat Genet. 2012 Submitted.
31. Minyard AN, Parker JC., Jr Intracranial saccular (berry) aneurysm: A brief overview. South Med J. 1997;90:672–677. [PubMed]
32. Lopez-Candales A, Holmes DR, Liao S, Scott MJ, Wickline SA, Thompson RW. Decreased vascular smooth muscle cell density in medial degeneration of human abdominal aortic aneurysms. Am J Pathol. 1997;150:993–1007. [PubMed]
33. Kockx MM, De Meyer GR, Muhring J, Jacob W, Bult H, Herman AG. Apoptosis and related proteins in different stages of human atherosclerotic plaques. Circulation. 1998;97:2307–2315. [PubMed]
34. Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation. 2002;105:1135–1143. [PubMed]
35. Kockx MM, Herman AG. Apoptosis in atherosclerosis: Beneficial or detrimental? Cardiovasc Res. 2000;45:736–746. [PubMed]
36. Bies J, Sramko M, Fares J, Rosu-Myles M, Zhang S, Koller R, Wolff L. Myeloid-specific inactivation of p15ink4b results in monocytosis and predisposition to myeloid leukemia. Blood. 2010;116:979–987. [PubMed]
37. Cunnington MS, Keavney B. Genetic mechanisms mediating atherosclerosis susceptibility at the chromosome 9p21 locus. Curr Atheroscler Rep. 2011;13:193–201. [PubMed]
38. Mayr U, Mayr M, Li C, Wernig F, Dietrich H, Hu Y, Xu Q. Loss of p53 accelerates neointimal lesions of vein bypass grafts in mice. Circ Res. 2002;90:197–204. [PubMed]
39. Sanz-Gonzalez SM, Barquin L, Garcia-Cao I, Roque M, Gonzalez JM, Fuster JJ, Castells MT, Flores JM, Serrano M, Andres V. Increased p53 gene dosage reduces neointimal thickening induced by mechanical injury but has no effect on native atherosclerosis. Cardiovasc Res. 2007;75:803–812. [PubMed]
40. Mercer J, Bennett M. The role of p53 in atherosclerosis. Cell Cycle. 2006;5:1907–1909. [PubMed]
41. Yap DB, Hsieh JK, Chan FS, Lu X. Mdm2: A bridge over the two tumour suppressors, p53 and rb. Oncogene. 1999;18:7681–7689. [PubMed]
42. Hsieh JK, Chan FS, O'Connor DJ, Mittnacht S, Zhong S, Lu X. Rb regulates the stability and the apoptotic function of p53 via mdm2. Mol Cell. 1999;3:181–193. [PubMed]
43. Kim JB, Deluna A, Mungrue IN, Vu C, Pouldar D, Civelek M, Orozco L, Wu J, Wang X, Charugundla S, Castellani LW, Rusek M, Jakobowski H, Lusis AJ. The effect of 9p21.3 coronary artery disease locus neighboring genes on atherosclerosis in mice. Circulation. 2012 [PMC free article] [PubMed]