Cks1 Mediates Vascular Smooth Muscle Cell Polyploidization*

Vascular smooth muscle cells (VSMC) at capacitance arteries of hypertensive individuals and animals undergo dramatic polyploidization that contributes to-ward their hypertrophic phenotype. We report here the identification of a defective mitotic spindle cell cycle checkpoint in VSMC isolated from capacitance arteries of pre-hypertensive rats. These cells demonstrated a high predisposition to polyploidization in culture and failed to maintain cyclin B protein levels in response to colcemid, a mitotic inhibitor. Furthermore, this altered mitotic spindle checkpoint status was associated with the overexpression of Cks1, a Cdc2 adapter protein that promotes cyclin B degradation. Cks1 up-regulation, cyclin B down-regulation, and VSMC polyploidization were evidenced at the smooth muscle of capacitance arteries of genetically hypertensive and Goldblatt-oper-ated rats. In addition, angiotensin II infusion dramati-cally increased Cks1 protein levels at capacitance arteries of normotensive rats, and angiotensin II treatment of isolated VSMC abrogated their ability to down-regulate Cks1 and maintain cyclin B protein expression in response to colcemid. from mitosis. Mitotic spindle cell cycle checkpoint signals orig-inated at kinetochore/microtubule inter- action points activate a pathway that down-regulates Cks1 expression, delays cyclin B degradation, and arrests the progression of M phase. In VSMC with al- tered mitotic spindle checkpoint status, Cks1 levels are not down-regulated; mi- totic exit may proceed in the absence of segregation and karyokinesis, leading to cell cycle re-entry and polyploidization.

Hypertension is accompanied by changes in vascular smooth muscle cell (VSMC) 1 growth that are specific for different vascular territories. Vascular smooth muscle hypertrophy predominates at capacitance arteries, those of high compliance, and is associated to VSMC polyploidization in hypertensive individu-als (1) and animals (2)(3)(4)(5)(6)(7)(8). Tetraploid and octaploid VSMC of hypertensive rats have 2.4-and 4.8-fold, respectively, the protein content of diploid VSMC of normotensive rats (4). Additionally, on a per cell basis, polyploid VSMC express higher levels of platelet-derived growth factor A, fibronectin, and collagen than their diploid counterparts (9). Importantly, the hypertrophy of vascular smooth muscle at capacitance arteries causes arterial stiffness and promotes left ventricular overload and altered coronary blood perfusion (10).
Mammalian cells are protected from cell cycle re-entry at mitosis by the activity of the mitotic spindle cell cycle checkpoint (20 -22). This pathway delays the exit from mitosis if the chromosomal segregation cannot be properly completed by preventing the inactivation of the M-phase promoting complex (Cdc2, cyclin B, and associated proteins) (23)(24)(25). Recent data indicate that, in mammalian cells, M-phase growth arrest is accomplished in part by down-regulation of Cks1 (26), a Cdc2 adapter protein that promotes cyclin B metabolism (27)(28)(29). Cells in which the mitotic checkpoint fails to down-regulate Cks1 expression cannot maintain cyclin B protein expression and M-phase growth arrest, and are predisposed to undergo cell cycle re-entry and polyploidization (26).
The incidence of polyploidy in vascular smooth muscle of hypertensive individuals prompted us to investigate the activity of the mitotic spindle cell cycle checkpoint in VSMC. The status of this pathway was studied in cultures of VSMC isolated from multiple vascular beds of normal and hypertensive rats. We provide functional and biochemical evidence of a specific mitotic spindle cell cycle checkpoint defect in cultures of VSMC isolated from capacitance arteries of SHR. These cells express high levels of Cks1 protein and fail to down-regulate Cks1 in response to mitotic inhibitors. VSMC isolated from resistance arteries of SHR had low Cks1 expression and normal mitotic checkpoint status. Treatment of SHR with the angiotensin converting enzyme inhibitor captopril reduced Cks1 and ploidy levels in aortic smooth muscle. Furthermore, activation of the renin-angiotensin system in the normotensive rat strain WKY by renal artery clipping, or angiotensin II infusion, induced Cks1 protein levels and VSMC polyploidization at aortic smooth muscle. In addition, treatment of primary cultures of WKY VSMC with angiotensin II induced Cks1 up-regulation and failure to control cyclin B expression in response to a mitotic spindle inhibitor. Finally, ectopic expression of Cks1 in VSMC isolated from normotensive rats reproduced the altered mitotic spindle cell cycle checkpoint phenotype observed in VSMC of hypertensive rats. In summary, these data demonstrate that Cks1 regulates VSMC ploidy and suggest that Cks1 up-regulation may contribute to the phenomena of VSMC polyploidization during hypertension.

EXPERIMENTAL PROCEDURES
Animals-WKY and SHR were from Charles River, Wilmington, MA. Harlan Sprague-Dawley and Zucker lean rats were from Harland Harlan Sprague-Dawley, Indianapolis, IN. Two-kidney one-clip Goldblatt and sham-operated WKY rats were from Taconic, Germantown, NY. Goldblatt-operated rats (10 weeks old) were sacrificed 2 months after surgery. Captopril (RBI) was given in tap water ad libitum for 2 months to 8-week-old SHR rats at a dose of 100 mg/kg/day. Angiotensin II (Sigma) was infused at 50 ng/kg/min using subcutaneous osmotic pumps (Alza, Palo Alto, CA). Blood pressure was determined using tail cuffs and a programmed electrosphygmomanometer (Narco, Austin, TX). For immunohistochemistry, 5-m formalin-fixed paraffin-embedded arterial sections were deparaffinized in xylene, rehydrated, and processed for periodic acid-Shiff-hematoxylin staining. Immunohistochemistry was carried out using Vectastain-ABC kits (Vector, Burlingame, MA) and antibodies (Santa Cruz Biotechnologies, Santa Cruz, CA) at dilution 1:50 (Cks1, MPP2) or 1:100 (cyclin B, proliferating cell nuclear antigen (PCNA)). Cells from at least five biopsies were scored for quantitative analysis by two independent observers. Pooled data were analyzed for statistical significance using ANOVA followed by Fisher's exact test (30).
Isolation, Infection, and Culture of VSMC-VSMC and vascular fibroblasts were isolated as described previously (31). After isolation, cells were seeded at a density of 3,500 cells/cm 2 and grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum (FBS) and penicillin/streptomycin (Life Technologies, Inc.). The VSMC phenotype was verified by immunostaining using specific antibodies against smooth muscle ␣-actin and smooth muscle myosin heavy chain. Population doubling times (PDL) were determined by plating the cells at low density and determining cell counts over 2-12 days. The pBabe, LXSN, pBabe-Cks1, pBabe-p53 RSCmut (pBabe-p53 143A), and LXSN-E6 retroviruses have been described previously (26). WKY aortic VSMC were infected by incubation in Polybrene-supplemented medium containing 10 7 viral plaque-forming units. Selection was carried out in puromycin (3 g/ml, Sigma) and/or G418 (300 g/ml, Life Technologies, Inc.) media. VSMC were used at passages 1-3 following drug selection.
Analysis of Cell Cycle Distribution of DNA Content-Matched sets of VSMC and fibroblasts were incubated for 2 PDL in the presence of colcemid (100 ng/ml) or nocodazole (20 ng/ml) (Sigma). Cell cycle distribution of DNA content was determined by flow cytometry (22,26). When total and newly synthesized DNA were determined, cells were labeled with 10 M bromodeoxyuridine (BrdUrd, Sigma) for 4 h, trypsinized, counted and fixed in 70% ethanol. Fixed cells were treated with 0.08% pepsin for nuclei isolation. The nuclear pellet was incubated for 30 min in 100 l of a 1:5 dilution of anti-BrdUrd fluorescein isothiocyanate-conjugated antibody (Becton & Dickinson, Franklin Lakes, NJ), stained with 50 g/ml propidium iodide (Aldrich, Natick, MA), and analyzed by flow cytometry using a Coulter Elite ESP flow cytometer (Beckman Coulter, Fullerton, CA) and CellQuest software (Becton-Dickinson). Data were analyzed for statistical significance using ANOVA followed by Student's t test (30). For karyotyping, cells were exposed to 100 ng/ml colcemid, incubated for 30 min in hypotonic solution, fixed, and stained with Giemsa (Sigma). Pooled karyotypic data were analyzed for significance using ANOVA followed by Fisher's test (30).
Western Blots-Confluent cell cultures at passage 3 (4 -5 ϫ 10 4 cells/cm 2 ) were synchronized by a 3-day incubation in low serum media (0.5% calf serum). Cells were then grown at low density (1-2 ϫ 10 4 cells/cm 2 ) in 10% FBS (Life Technologies, Inc.) in the absence or presence of colcemid (100 ng/ml), harvested at the indicated intervals, and extracts prepared as described previously (26). Western blots were performed using Cks1 (Santa Cruz), cyclin B1 (Santa Cruz), and ␤-actin (Sigma) antibodies at dilutions 1:100, 1:500, and 1:10,000, respectively. For quantification, gels were scanned using an LKB Ultrascan XL densitometer and GelScan XL software (LKB, Uppsala, Sweden). When cell cycle time-course studies were carried out, the onset of protein down-regulation was determined as the experimental time point at which a Ͼ50% decrease in protein levels was detected. Significant differences in the distribution of data were determined by G test (30). G values were compared using a 2 distribution with one degree of freedom. A p Ͻ 0.05 was considered significantly different.

VSMC from Prehypertensive Animals Are Predisposed to
Polyploidization-The ability of VSMC to arrest growth at M phase in response to mitotic spindle depolymerizing agents was analyzed using a flow cytometry-based functional assay (22). Primary cultures were generated from VSMC isolated from multiple vascular territories of 3-week-old SHR. This age was selected because it precedes the onset of VSMC polyploidization and hypertrophy in the SHR (4). Cultures of skin and vascular fibroblasts isolated from SHR as well as of VSMC isolated from age-matched WKY, Harlan Sprague-Dawley, and Zucker lean rats were also investigated. Cells were incubated in media containing colcemid for an incubation period equal to 2 PDL, in order to compare equal number of mitotic events among cell groups, and the percentage of cells with Ͼ4 N DNA content determined using flow cytometry (Table I). A small fraction of polyploid SHR skin fibroblasts was detected (5% of Ͼ4 N cells). This number is within the range previously found in mouse and human fibroblasts (21,22). Strikingly, VSMC isolated from multiple vascular beds showed significant differences in their ability to control their DNA ploidy. Cultures of VSMC from capacitance arteries of SHR, such as carotid arteries or aorta, progressed to significantly higher levels of polyploidy than skin fibroblasts (14 -28% of Ͼ4 N cells, p Ͻ 0.01). In contrast, VSMC from mesenteric superior arteries, polyploidization The table shows the percentage of cells that reach a higher of 4N DNA content after a 2-PDL incubation period in media with colcemid. VSMC and fibroblasts were isolated from 3-week-old rats as indicated under "Experimental Procedures." Each cell population represents a pool of cells from 9 -12 animals. n indicates the number of independent cell populations analyzed. Cells were incubated at approximately 30% confluence (2 ϫ 10 4 cells/cm 2 ) in the presence of 100 ng/ml colcemid for an incubation period equal to twice their PDL. All experiments were performed with cells at passages 1-3. DNA content was determined by flow cytometry as described previously (26). Data were analyzed by ANOVA followed by Student's t test. Asterisk denotes p Ͻ 0.01. mesenteric branches, and adventitial fibroblasts from the SHR proximal aorta maintained a low ploidy content (6 -7% of Ͼ4 N cells). VSMC isolated from capacitance and resistance arteries of the normotensive strains WKY, Harlan Sprague-Dawley, and Zucker lean rats also demonstrated active growth arrest at 4 N (5-6% of Ͼ4 N cells). Incubation of SHR proximal aorta VSMC with nocodazole or higher concentrations of colcemid (up to 1 g/ml) did not result in sustained 4 N growth arrest (data not shown). Additionally, in the absence of mitotic inhibitors, only 1-2% of Ͼ4 N cells were detected in all cell groups (data not shown). Importantly, the phenotypic differences observed were independent of the proliferation rate, since cells were analyzed at an equal number of PDL (22). Additionally, some cell populations, such as SHR skin fibroblasts, with a higher growth rate than VSMC, displayed normal ability to control DNA ploidy (Table I). In addition, infection of VSMC isolated Western blot of cyclin B and ␤-actin proteins in extracts prepared from synchronized cultures of VSMC and skin fibroblasts isolated from 3-week-old SHR and WKY rats. a and b, confluent cell cultures (4 ϫ 10 4 to 5 ϫ 10 4 cells/cm 2 ) were synchronized by a 2-day incubation in low serum medium (0.5% calf serum), incubated at low density (1 ϫ 10 4 to 2 ϫ 10 4 cells/cm 2 ) in 10% FBS in the absence (no colcemid) or presence (colcemid) of 100 ng of colcemid/ml, and harvested at the indicated intervals. c-i, cells were synchronized as above and incubated with 100 ng/ml of colcemid. A representative Western blot of three to five experiments per each cell population is shown.

Cks1 Mediates VSMC Polyploidization
from the proximal aorta of WKY rats with a retroviral vector that drives the expression of a tumor suppressor gene p53 RSC mutant (WKY-pBabe-p53RSCmut; RSC, relaxed mitotic spindle checkpoint allele) or the human papilloma virus E6 protein (WKY-LXSN-E6) generated cultures of VSMC with high proliferation rates. However, only those VSMC infected with the mitotic checkpoint deficient p53 RSC vector accumulated an elevated fraction of polyploid cells (24%, p Ͻ 0.001 related to SHR skin fibroblasts).
The results shown in Table I suggested that VSMC from capacitance arteries of SHR undergo polyploidization in culture because of a defective mitotic spindle cell cycle checkpoint. This pathway delays the degradation of mitotic proteins, such as cyclin B, if the segregation of chromosomes cannot be properly completed (21). In exponentially growing cells, cyclin B protein oscillates during the cell cycle. The level of this protein raises at the transition between the G 2 and M cell cycle phases and declines sharply when cells exit M-phase. Activation of the mitotic spindle cell cycle checkpoint by spindle depolymerizing agents inhibits cyclin B degradation and originates a prolonged, although not permanent, period of cyclin B protein expression (26). We observed that, in synchronized cultures of SHR skin fibroblasts, cyclin B protein levels oscillated in 48-h cycles (Fig. 1a), indicating normal entry and exit from M phase. When colcemid was added to the cultures, cyclin B protein was detected up to 120 h after cell cycle entry, indicating transient mitotic arrest, and thus, normal mitotic spindle checkpoint activity. In the absence of inhibitor, no differences in the expression of cyclin B protein (48-h peaks) were observed among cultures of VSMC from multiple vascular territories (Fig. 1b, no colcemid and data not shown). However, incubation of these cells with colcemid revealed striking differences in their ability to regulate cyclin B protein expression. Fig. 1 (b-i) shows a representative gel for each vascular territory investigated. In order to quantify these results, the time of onset of cyclin B down-regulation was determined as the experimental time point at which a Ͼ50% decline in cyclin B levels was detected by gel densitometry. A summary of these experiments is shown in Table II. Significant differences in the onset of cyclin B down-regulation were observed between cultures of VSMC isolated from capacitance arteries (proximal aorta VSMC, carotid arteries VSMC, and distal thoracic aorta VSMC) of SHR rats and skin fibroblasts isolated from the same animals (p Ͻ 0.05). Thus, VSMC from capacitance arteries of SHR failed to delay cyclin B degradation in response to the mitotic inhibitor, indicating defective mitotic checkpoint activity. In contrast, normal mitotic checkpoint activity was found in cultures of VSMC isolated from mesenteric arteries and branches of SHR (Fig. 1  (h and i) and Table II). Significant differences were found also between cultures of VSMC from capacitance arteries of SHR and their WKY counterparts (p Ͻ 0.05). In addition, VSMC isolated from aortas of Harlan Sprague-Dawley rats demonstrated sustained delay of cyclin B metabolism in response to colcemid (data not shown). Taken together, the results shown in Fig. 2 and Tables I and II demonstrate a defective mitotic spindle cell cycle checkpoint in cultures of VSMC isolated from capacitance arteries of rats predisposed to hypertension.
Cks1 Is Up-regulated in the Smooth Muscle of Hypertensive Rats-It has been shown that Cks1 plays a key role in the control of the progression through M phase in mammalian cells (26). The expression of this protein was investigated using Western blot analysis in VSMC freshly isolated from the vascular smooth muscle of WKY and SHR rats. Strikingly, a gradient of Cks1 was observed along the vascular tree (Fig. 2a), with significantly higher levels of protein at capacitance arteries of SHR rats than in their WKY counterparts (266 Ϯ 6% increase in SHR, p Ͻ 0.003; Fig. 2a and data not shown). Additionally, immunohistochemistry experiments demonstrated that Cks1 up-regulation is specific for the medial vascular smooth muscle layer, with background Cks1 expression levels at intima and adventitia (Fig. 2b). Moreover, although Cks1 protein was expressed at higher levels in aortic VSMC of SHR rats (33% versus 11% of positively stained cells, p Ͻ 0.003), higher levels of cyclin B protein were found in agematched WKY rats (78% versus 34% of positively stained cells, p Ͻ 0.001). In addition, the mitotic marker M-phase protein 2 (MPP2) was expressed at higher levels in SHR than in WKY rats (36% versus 12% MPP2-positive cells, p Ͻ 0.001), whereas no differences were observed in the level of PCNA protein between both rat strains (Fig. 2b). Thus, these results demonstrate that enhanced Cks1 expression accompanies cyclin B down-regulation and mitotic progression in VSMC of aortas of SHR rats. Aorta SHR VSMC have been previously shown to undergo polyploidization and hypertrophy in vivo (4,5,33,34); our data support the hypothesis that mitotic checkpoints play a role in the regulation of VSMC growth in vivo.
In order to control for genetic divergence between the SHR and WKY strains, two series of experiments were conducted. First, SHR rats were treated for 2 months with captopril, an angiotensin converting enzyme inhibitor. Aortic VSMC were isolated from sham-and captropril-treated SHR rats, and Cks1 and cell ploidy levels determined. A decrease (59 Ϯ 7%, p Ͻ 0.004) in the polyploid VSMC fraction in captopril-treated animals was accompanied by a reduction (64 Ϯ 3%, p Ͻ 001) in Cks1 protein levels ( Fig. 2c and data not shown). In the second group of experiments, the renin-angiotensin system was activated in WKY rats clipping the left renal artery (Goldblatt's surgery). This procedure has been shown previously to induced VSMC hypertrophy and polyploidization at large arteries (17). Enhanced polyploidization (262 Ϯ 21%, p Ͻ 0.006) and increased expression of Cks1 protein (493 Ϯ 18%, p Ͻ 0.002) were observed in Goldblatt WKY rats (Fig. 2d and data not shown). Furthermore, aortic VSMC of Goldblatt-operated animals demonstrated lower cyclin B staining (62% decrease, p Ͻ 0.001) and higher MPP2 expression (649% increase, p Ͻ 0.001) than shamoperated animals (Fig. 2e). Thus, enhanced expression of Cks1 accompanied mitotic progression and polyploidization in an experimental model of renovascular hypertension, whereas in-

Cks1 Mediates VSMC Polyploidization
FIG. 2. Elevated Cks1 levels are associated with VSMC polyploidization. a, Western blot of Cks1 and ␤-actin in VSMC freshly isolated from SHR or WKY rat arteries (1 ϫ 10 6 cells/lane). The figure is representative of three blots. Densitometry was carried out using an LKB densitometer. b, immunohistochemistry of Cks1, cyclin B, MPP2, and PCNA in arterial sections (ϫ250) of 3-week-old WKY and SHR. For quantification, cells from at least five biopsies were scored for quantitative analysis by two independent observers. Pooled data were analyzed for significance using ANOVA followed by Fisher's test (30). Other details as indicated under "Experimental Procedures." c, Western analysis (upper panel) and ploidy content (lower panel) in VSMC freshly isolated from the proximal aorta of control SHR and captopril-treated SHR. Ploidy content was determined by flow cytometry of three preparations (8 -12 rats) of propidium iodide-stained freshly isolated cells. Bars show percentage of cells
These experiments suggested that angiotensin II may induce VSMC polyploidization by up-regulating Cks1. This hypothesis was investigated by the infusion of sub-pressor concentrations of angiotensin II in WKY rats using subcutaneous osmotic pumps. Short term angiotensin II-treatment caused dramatic up-regulation of Cks1 at capacitance arteries (Fig. 4a). Immunohistochemistry revealed a 770% increase in Cks1-positive cells at the aorta smooth muscle related to untreated controls (p Ͻ 0.0001). Furthermore, in the absence of angiotensin II, colcemid induced lower and significantly delayed expression of Cks1 in cultures of WKY VSMC (Fig. 3c, p Ͻ 0.005). However, when angiotensin II was added to the media, peaks of Cks1 and cyclin B protein expression in these cells were observed at 48 h after cell cycle entry, despite the presence of the mitotic inhibitor (Fig. 3, b and c). Thus, these results demonstrate that angiotensin II induces VSMC Cks1 up-regulation and failure to arrest the progression of M phase in response to a mitotic inhibitor.
Ectopic Expression of Cks1 Is Sufficient to Promote VSMC Polyploidization-To test a direct effect of Cks1 on VSMC ploidy content, human Cks1 was expressed in VSMC isolated from the aorta of normotensive rats using a retroviral vector (pBabe-Cks1) (26). Control VSMC were infected with an empty vector (pBabe VSMC). Cultures of pBabe and pBabe-Cks1 VSMC were then growth arrested at G 0 , allowed to enter the cell cycle synchronously in the presence or absence of colcemid, and the levels of Cks1 and cyclin B proteins determined at multiple time points. Cks1 expression in colcemid-treated VSMC cultures was lower and significantly delayed in pBabe-VSMC (p Ͻ 0.05), but not in pBabe-Cks1-VSMC, relative to untreated VSMC (Fig. 4a). Additionally, a significant delay in the onset of cyclin B degradation was observed in colcemid treated pBabe-VSMC, but not in pBabe-Cks1-VSMC (Fig. 4a, p Ͻ 0.05). Thus, cultures of Cks1-transduced VSMC demonstrated a defective mitotic spindle cell cycle checkpoint status, with its corresponding biochemical alterations in the regulation of Cks1 and cyclin B proteins (Fig. 4a), that resembles angiotensin II-treated WKY VSMC (Fig. 3, b and c) or SHR VSMC (Figs. 1 (b-d) and 3c). Furthermore, flow cytometry of total DNA content (PI staining) versus newly synthesized DNA (BrdUrd incorporation) demonstrated that pBabe-Cks1-VSMC are able to undergo cell cycle re-entry at 4 N DNA content either in the presence of colcemid (Fig. 4b) or following cell confluence (Fig. 4c). Finally, the polyploidization rate of cultures of pBabe-VSMC and pBabe-Cks1-VSMC were investigated by karyotypic analysis. pBabe-Cks1-VSMC, but not pBabe-VSMC, accumulated a significantly higher fraction of tetraploid (4 N) and octaploid (8 N) cells than their parental WKY VSMC cultures when incubated for 2 PDLs in the presence of colcemid (Fig. 4d, p Ͻ 0.0001). Thus, these data demonstrate that unregulated Cks1 expression is sufficient to promote VSMC polyploidization. In addition, pBabe-Cks1 VSMC demonstrated a higher rate of spontaneous polyploidization in culture than pBabe-VSMC (Fig. 4e), although the difference was only marginally significant (p Ͻ 0.06).
In summary, we have investigated a molecular mechanism of control of ploidy in VSMC of hypertensive animals. Several findings are reported here. First, VSMC isolated from prehypertensive animals are predisposed to undergo polyploidization due to an altered mitotic spindle cell cycle checkpoint status. Second, angiotensin II up-regulates Cks1. Third, ectopic expression of Cks1 is sufficient to abrogate the mitotic spindle cell cycle checkpoint in VSMC. Collectively, these studies provide the first evidence for a molecular mechanism of VSMC with Ͼ4 N DNA content (mean Ϯ S.D.). Blot is representative of three experiments. Data were analyzed for significance using ANOVA followed by Student's t test (30). Mean blood pressures were 186 Ϯ 7 (SHR control) and 121 Ϯ 9 (captopril-treated) mm Hg. d, Western blot and ploidy content in VSMC isolated from the proximal aorta of sham-and Golblatt-operated WKY rats. Mean blood pressures were 131 Ϯ 8 (sham) and 195 Ϯ 11 (Goldblatt) mm Hg. Cks1 levels and ploidy content determined as in c (n ϭ 3). e, immunohistochemistry of Cks1, cyclin B, and MPP2 proteins in sections (ϫ250) of the proximal aorta of sham-and Goldblatt-operated WKY rats. Other experimental details are as in b.  Fig. 1. b, flow cytometry of DNA content in pBabe-and pBabe-Cks1 VSMC. Asynchronous cell cultures at passages 1-3 were incubated in the absence or presence of 100 ng/ml colcemid for 1 PDL. PDL for pBabe VSMC and pBabe-Cks1 VSMC were 68 and 52 h, respectively. Data are representative of three independent experiments. c, Flow cytometry of DNA content in confluent pBabe-and pBabe-Cks1 VSMC. Data are polyploidization. These results extend previous observations by others (2,6,8) that reported a higher level of polyploidization in culture of SHR VSMC than WKY VSMC. In contrast, a previous report indicated that the rate of polyploidization in culture was higher in WKY VSMC than in SHR VSMC (32). However, when these authors tested VSMC from another two normotensive rat strains, Harlan Sprague-Dawley and Fisher rats, lower polyploidization rates than in SHR VSMC were found, which suggests a phenotypic divergence of the WKY colony investigated.
Our data support also previous observations indicating a role for Cks1 in the control of M phase in mammalian cells. It has been shown that Cks1 is a target of the mitotic spindle cell cycle checkpoint pathway, and that down-regulation of Cks1 is essential to promote mitotic arrest in mammalian cells (26). However, the mechanism of action of Cks1 at mitosis is not completely understood. It is known that Cks1 and related proteins work as adapters that modulate substrate recognition by Cdc2 (35)(36)(37). Importantly, Cks1 may drive the activation by Cdc2 of the cyclin B degradation machinery at mitosis. Cks1depleted Xenopus oocytes cannot undergo Cdc2-dependent phosphorylation of the Cdc27 component of the cyclosome, the ubiquitin ligase activity that targets cyclin B for degradation (36). Furthermore, a recent study showed that Cks1 plays a yet undefined role in the activation of the proteasome, the proteolytic complex that degrades cyclosome-ubiquitinated cyclin B (37). Thus, we hypothesize that up-regulation of Cks1 in hypertensive VSMC uncouples the mitotic spindle cell cycle checkpoint with the cell cycle regulatory machinery at M phase, and, consequently, abrogates the ability of these cells to control cyclin B metabolism and arrest the progression of mitosis (Fig. 5). Further studies on the regulation of Cks1 expression and activity by angiotensin and other growth factors may contribute to a better understanding of the pathways that modulate the VSMC phenotype during hypertension (38). Importantly, the fact that growth factors may control the expression of proteins required for M-phase progression in VSMC was postulated previously (39), but direct evidence has been lacking until now.
representative of two experiments. d, karyotypic analysis of the parental WKY VSMC, WKY-pBabe, and WKY-pBabe Cks1 cells incubated for 2 PDL with 100 ng/ml colcemid. Pooled data were analyzed for significance using ANOVA followed by Fisher's test (30). Other details are as indicated under "Experimental Procedures." e, karyotypic analysis of WKY-pBabe and WKY-pBabe Cks1 cells at early (1-3) and late (7-9) passage. Cells were incubated with 100 ng/ml colcemid for 4 h. Other experimental details are as in d.

FIG. 5.
Model of mitotic spindle cell cycle checkpoint activity and Cks1 role in VSMC polyploidization. Cks1 is required for the degradation of cyclin B protein at the exit from mitosis. Mitotic spindle cell cycle checkpoint signals originated at kinetochore/microtubule interaction points activate a pathway that down-regulates Cks1 expression, delays cyclin B degradation, and arrests the progression of M phase. In VSMC with altered mitotic spindle checkpoint status, Cks1 levels are not down-regulated; mitotic exit may proceed in the absence of segregation and karyokinesis, leading to cell cycle re-entry and polyploidization.