Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The normal pulmonary circulation is a low resistance, high capacitance vascular bed whose function is to match ventilation to perfusion, thus optimizing blood and tissue oxygenation. Pulmonary hypertension results from pathologic elevation in pulmonary vascular resistance due to a combination of increased vascular tone and vessel wall remodeling, ultimately leading to right ventricular hypertrophy and premature death. Vessel wall remodeling is a hallmark of severe, advanced pulmonary hypertension, presenting histologically as neointimal proliferation, medial and adventitial hyperplasia and hypertrophy, muscularization of peripheral vessels, and vasoocclusive plexiform lesions (1). Pulmonary vascular remodeling can be initiated by a variety of stimuli including chronic hypoxia, increased pulmonary blood flow, collagen vascular disease, anorectic and other drugs, and idiopathic causes.

Because most vascular cells in the adult pulmonary circulation are in a quiescent state (6), the development of vascular remodeling requires cells arrested in G₀ or G₁ to enter the cell cycle. Progression through the cell cycle requires the coordinated interaction of cyclin-dependent kinases (cdk) and their regulatory subunits, the cyclins, to drive cells through G₁ into S phase to ultimately result in cell division (7, 8). Cyclin-cdk complexes activate transcription factors important in cell cycle progression. Important G₁ cyclin-cdk complexes include cyclin D-cdk4/ cdk6 and cyclin E-cdk2 which phosphorylate (and inactivate) retinoblastoma, an antitumor and antiproliferative protein which limits E2F-mediated gene transcription (9). Cdk inhibitors are proteins that bind cyclin-cdk complexes, inhibit hyperphosphorylation of retinoblastoma, cause G₁ arrest, and limit cell proliferation (7, 8). Two families of cdk inhibitors have been described, INK and Cip/Kip. Loss of INK and Cip/Kip cdk inhibitors has been implicated in tumor development (10). In contrast, only Cip/Kip family members have been shown to regulate vascular smooth muscle cell proliferation (13, 14).

p27^Kip1 is a member of the Cip/Kip family of cdk inhibitors, which also includes p21^Cip1/Waf1 and p57^Kip2. p27^Kip1 has been shown to be an important regulator of aortic smooth muscle cell proliferation in vitro (13, 14) and to limit vascular remodeling in vivo in the intact systemic circulation following balloon injury (13, 14). To determine the importance of p27^Kip1 in regulating vascular smooth muscle cell proliferation in the pulmonary circulation, we studied the temporal expression of p27^Kip1 as well as the mitogenic and proliferative characteristics of wild type, p27^Kip1-overexpressing, and p27^Kip1-deficient pulmonary arterial smooth muscle cell (PASMC) following serum stimulation. Effects of serum on cell cycle, retinoblastoma phosphorylation, and DNA synthesis, as well as protein-protein interaction between p27^Kip1 and cdk-cyclin targets, were assessed. Our results demonstrate that p27^Kip1 is an important modulator of PASMC proliferation during mitogenic stimulation, suggesting that decreased functional p27^Kip1 may contribute to the vascular remodeling associated with pulmonary hypertension.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

We used Dulbecco's modified Eagle's medium (DMEM; Sigma, St. Louis, MO), Trypsin-ethylenediaminetetraacetic acid (EDTA) (Gibco, Grand Island, NY), fetal bovine serum (FBS; Gemini, Woodland, CA), L-glutamine (Gibco), Ipegal (Sigma), propridium iodide (Sigma), RNAse (Sigma), polyvinylidene difluoride (PVDF) membrane (Amersham, Buckinghamshire, UK), and ECL-plus (Amersham). Antibodies used included p21^Cip1/Waf1 (c-19), cdk2 (m2), cdk4 (c-22), PCNA (pc-10), cyclin D (R-124), cyclin A (c-19), and cyclin E (M-20) all from Santa Cruz Biotechnology (Santa Cruz, CA). p27^Kip1 and retinoblastoma (both mouse monoclonal) from PharMingen (San Diego, CA). Secondary HRP-conjugated antibodies used were donkey -mouse, donkey -goat, and donkey -rabbit (Jackson Labs, West Grove, PA).

Animals

Adult male Sprague-Dawley rats were obtained from Harlan Sprague Dawley (Indianapolis, IN) and housed in the UCHSC animal facility before use. The p27^Kip1 mice were bred on a c57B6/sv129 background (original breeding pair was obtained from Dr. Andrew Koff at Memorial Sloan-Kettering Cancer Center). Genotype was confirmed by polymerase chain reaction (PCR) analysis using standard techniques (15). Because the p27^Kip1/ female is infertile all breeding pairs were heterozygotes. Paired p27^Kip1+/+ and p27^Kip1/ littermates were compared to minimize any potential age or genetic differences. Animals were fed standard rodent chow and water ad libitum.

Cell Culture

Smooth muscle cells were isolated by elastase and collagenase digestion of main (extralobar) pulmonary arteries from adult (24- 28 wk) Sprague Dawley male rats or adult (5-8 wk) mice (see above). The pulmonary artery was dissected free from surrounding tissue and the adventitial layer removed. The pulmonary artery was cut into small pieces and placed in an elastase (Boehringer Mannheim, Indianapolis, IN) and collagenase (Worthington, Lakewood, NJ) digest media supplemented with albumin (Sigma) and soybean trypsin inhibitor (Worthington) for 60 min at 37°C. The cells were cultured in DMEM supplemented with 100 U/ml penicillin, 0.1 mg/ml streptomycin, 2 mM L-glutamine, and 10% FBS and grown in humidified incubators (Forma Scientific, Marietta, OH) at 37°C in 5% CO₂, 16% O₂, and balance nitrogen. Cells were used between passages 2 and 6. Smooth muscle cell identity was verified by positive staining for smooth muscle -actin (mouse monoclonal antibody; Sigma) at each passage (> 95% of cells stained positive for smooth muscle -actin).

[³H]Thymidine Incorporation

Smooth muscle cells were plated in DMEM-10% FBS at a density of 20,000 cells/well in 24-well plates. After 24 h, cells were growth-arrested by addition of DMEM with 0.1% FBS. After 72 h of low-serum, cells were labeled with [methyl]-³H-thymidine at 25 µCi/ml following the addition of DMEM with either 0.1% or 10% FBS. Cells were harvested 24 h later in 1% sodium dodecyl sulfate (SDS)/0.01 N NaOH. [³H]thymidine incorporation was determined in a Becton (Franklin Lakes, NJ) LS6500 scintillation counter and normalized to cell number (cpm/cell). For [³H]thymidine incorporation measured at Days 3 and 5, cells were stimulated by addition of DMEM-10% FBS on Day 0, and [³H]thymidine added for the 24-h period before harvesting.

Cell Proliferation

Smooth muscle cells were plated at 20,000 cells/well, growth- arrested for 72 h, and then grown in DMEM supplemented with 10% FBS. Cells were removed from the wells by 0.05% trypsin/ 0.53 mM EDTA digestion and counted at Days 1, 3, and 5. Cells from 4 wells were counted using a Fischer (Pittsburgh, PA) hemocytometer and the results averaged to obtain a single cell count (± standard error) for each time point.

Flow Cytometry

Cell cycle was determined by flow cytometry in propridium iodide-stained cells. At each time point, cells were digested with trypsin-EDTA from culture plates and the trypsin inactivated by addition of DMEM with 10% FBS. Cells were collected by low centrifugation, washed with PBS, recollected by centrifugation, stained with Krishan's solution (propridium iodide, sodium citrate.2H₂O, Ipegal (Sigma)), boiled RNAse (16), and incubated overnight at 4°C. The cells were then analyzed in the University of Colorado Health Science Flow cytometry core using a Beckman (Miami, FL) Epics-XL flow cytometer. Histograms of DNA content were analyzed using Modfit LT software (Verity Software, Topsham, ME) to determine fractions of the population in each phase of the cell cycle (G₀/G₁, S, G₂/M).

Adenoviral Transfection

Rat PASMCs were plated at 20,000 cells/well in 24-well plates. Cells were transfected for 2 h at a multiple of infectivity of 200 with a replication-deficient adenovirus serotype 5 containing a human p27^Kip1 cDNA driven by a CMV promoter (14) (gift from Dr. Elizabeth Nabel, National Institutes of Health, Bethesda, MD). Cells were then washed and growth-arrested for 72 h before being stimulated. An adenovirus encoding the human placental alkaline phosphatase cDNA was used as a control.

Western Blot Analysis

Subconfluent cells at the scheduled time points were washed with cold phosphate-buffered saline (PBS), lysed in Radioimmunoprecipitation (RIPA) buffer (PBS, 1% Ipegal, 0.5% sodium deoxycholate, 0.1% SDS, PMSF [10 mg/ml], aprotinin [30 µL/ml], and sodium orthovanadate [1 mM]), pelleted at 14,000 × g and the protein concentration of the supernatant determined by Bradford assay. Equal amounts of protein were separated by SDS-polyacrylamide gel electrophoresis (PAGE) under reducing conditions (1% -mercaptoethanol) using 14% gels. The proteins were then transferred to PVDF membrane in 20% methanol. Membrane was blocked in 1% nonfat milk and 0.1% Tween-20, probed with appropriate antibodies (see MATERIALS), detected with appropriate secondary antibodies conjugated to HRP at 1:10,000 dilution, and detected using ECL-plus (Amersham). Protein expression was quantified using NIH image 1.63 and expressed as arbitrary density units relative to baseline (cells incubated with DMEM supplemented with 0.1% FBS).

Immunoprecipitation

A quantity of 200 µg of cell lysate was incubated with 2 µg of cyclin E antibody (rabbit polyclonal) and Protein A sepharose beads (Santa Cruz Biotechnology) overnight at 4°C. Beads were washed twice with RIPA buffer, placed in sample buffer containing -mercaptoethanol, boiled, separated on a 14% gel, and transferred to PVDF paper as described above. Rabbit nonimmune immunoglobulin (Ig)G (2 µg) was used as a negative control.

Kinase Activity Assay

Cyclin E was immunoprecipitated as described above. Beads were washed twice with RIPA buffer and twice in cold kinase buffer (50 mM HEPES [pH 7.5], 10 mM MgCl₂, 2.5 mM EGTA, 1 mM DTT, 10 mM -glycerophosphate, 1 mM NaF, 0.1 mM sodium orthovanadate, and 20 µM ATP). Samples were resuspended in 30 µl of kinase buffer containing 2 µg of GST-Rb fusion protein (17) and 10 µCi of [-³²P]ATP (Amersham Pharmacia Biotech); after incubation at 30°C for 30 min with occasional mixing, reactions were stopped by adding 30 µL of hot 2× Laemmli sample buffer and boiling for 5 min. Samples were resolved by SDS-PAGE in a 10% gel and phosphorylated proteins were detected by autoradiography.

Statistical Methods

Data are expressed as means ± SEM. Thymidine incorporation, cell cycle, and cell growth were compared using ANOVA with Fisher post-hoc test, with values of P < 0.05 considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

DNA Synthesis in Stimulated PASMCs Peaks at 24 h

PASMCs were growth-arrested in 0.1% FBS for 72 h, stimulated with 10% FBS, and followed for 5 d. After addition of 10% FBS, [³H]thymidine incorporation (Figure 1A) and percentage of cells in S phase (Figure 2A) peaked at 24 h and then returned to baseline by Day 3. Cell growth was logarithmic between Days 1 and 3 and then the rate of proliferation decreased (Figure 1B).

View larger version (12K): [in this window] [in a new window]   Figure 1.   [3H]thymidine incorporation and cell growth of rat PASMCs following serum stimulation. Growth-arrested rat PASMCs were exposed to 10% serum and followed over 5 d. (A) [3H]thymidine incorporation. (B) Cell growth at Days 1, 3, and 5 (n = five separate experiments from five different animals).

View larger version (27K): [in this window] [in a new window]   Figure 2.   Cell cycle phase of growth-arrested rat PASMCs following serum stimulation over 5 d. (A) Percentage of cells in G0/G1, S phase, and G2/M; (B) representative cell cycle histograms at baseline and Days 1 through 5 (n = five separate experiments from five different animals).

p27^Kip1 Expression Is Biphasic in Serum-Stimulated PASMCs

p27^Kip1 protein levels dropped to 40% of baseline after 1 d of serum stimulation, but returned to growth-arrested levels by 3 d. In contrast, expression of p21^Cip1/Waf1 peaked following 1 d of serum stimulation and then decreased to baseline by 3 d (Figure 3). Expression of proliferating cell nuclear antigen (PCNA), a marker of cell stimulation, had an inverse (temporal) pattern relative to that seen with p27^Kip1, peaking at 24 h and then decreasing over 5 d (Figure 3). Cdk2 levels increased following serum stimulation. In addition, a faster-migrating cdk2 band appeared at 24 h (cdk2-P), which is indicative of cdk2 phosphorylation and activation on threonine 160 (Figure 3) (18).

View larger version (69K): [in this window] [in a new window]   Figure 3.   Western blot analysis of p21Cip1/Waf1, p27Kip1, cdk2, and proliferating cell nuclear antigen (PCNA) expression in rat PASMCs during 5 d of 10% FBS. cdk2-P indicates phosphorylation of the cdk2 kinase on threonine 160 (18) (representative blot from four separate experiments).

Cyclin E-cdk2 Kinase Activity Peaks at 24 h

Cell progression through G₁ and the G₁/S transition requires active cyclin E complexes. Increases in p27^Kip1 binding to cyclin E can inhibit cyclin E kinase activity and lead to G₁ arrest. We therefore determined (by coimmunoprecipitation) whether the changes in total p27^Kip1 protein levels (Figure 3) resulted in similar changes in p27^Kip1 binding to cyclin E, and whether this correlated with cyclin E kinase activity. Consistent with our functional data, which established peak mitogenic activity at 24 h, p27^Kip1 binding to cyclin E decreased at 24 h, increased above baseline by Day 3 and then plateaued (Figure 4A). Hyperphosphorylation of retinoblastoma, an important cyclin E substrate critical in controlling G₁/S transition (9, 17), peaked at 24 h and declined to baseline by Day 3 (Figure 4B), which correlated with the increase and subsequent decrease in [³H]thymidine incorporation and the percent of cells in S phase seen at these time points. Cyclin E kinase activity measured in vitro also peaked at 24 h and returned to background activity by Day 3 (Figure 4C).

View larger version (57K): [in this window] [in a new window]   Figure 4.   Cyclin E immunoprecipitation and kinase activity following serum stimulation. (A) p27Kip1 binding to cyclin E was lowest at 24 h and increased above baseline by Day 3; (B) hyperphosphorylation of retinoblastoma (Rb-P) peaked at 24 h and then returned to baseline by Day 3; (C) autoradiograph of 32P incorporation into GST-Rb indicates that the kinase activity of cyclin E peaked at 12 to 24 h following serum stimulation and then returned to background levels by 72 h (representative of three different experiments from three different animals. Exposure time for autoradiograph was 4 h).

p27^Kip1 Overexpression Blunts DNA Synthesis and Cell Proliferation

We tested the ability of p27^Kip1 overexpression to block or blunt mitogenesis and cell proliferation in PASMC. Using a replication-deficient adenovirus, we infected rat PASMCs at a multiple of infectivity (MOI) of 200 and compared these cells with control infected (adenovirus encoding the human placental alkaline phosphatase cDNA) or untreated cells. p27^Kip1overexpression prevented the drop in p27^Kip1 levels associated with serum (Figure 5A) and resulted in a tenfold increase in cylin E-p27^Kip1 binding (data not shown). The increase in p27^Kip1 blunted [³H]thymidine incorporation, increased the % of cells in G₀/G₁, and reduced cell growth at 3 and 5 d (Figures 5B-5D).

View larger version (20K): [in this window] [in a new window]   Figure 5.   p27Kip1 overexpression reduced DNA synthesis and cell growth in rat PASMCs. (A) Western blot following p27Kip1 adenoviral-infection; (B) [3H]thymidine incorporation; (C) % cells in G0/G1; and (D) cell growth. , untreated cells; , control adenovirus infection; , adenovirus-p27Kip1 (three separate experiments from three different animals).

p27^Kip1 Null PASMCs Have Increased [³H]thymidine Incorporation and Cell Growth

We tested the effect of p27^Kip1 deficiency on cell proliferation using cultured PASMCs from individual p27^Kip1 null and wild-type mice. Paired littermates were chosen to minimize any potential age and genetic differences, and animal genotype determined by PCR analysis of genomic DNA. As an additional control, the cultured cells were also genotyped by PCR and Western blots performed to confirm the absence of p27^Kip1 in cells cultured from p27^Kip1-deficient mice (data not shown). Like wild-type cells, p27^Kip1/ PASMCs did not grow when placed in 0.1% FBS (data not shown). Compared with wild-type controls, p27^Kip1 null PASMCs had increased [³H]thymidine incorporation at baseline and in response to serum stimulation (Figure 6A). PASMCs from p27^Kip1 null mice had increased cell growth compared with PASMCs from wild-type animals (Figure 6B). Overexpression of p27^Kip1 in p27^Kip1 null cells decreased both baseline and serum-stimulated [³H]thymidine incorporation (Figure 7A) and decreased cell proliferation (Figure 7B).

View larger version (17K): [in this window] [in a new window]   Figure 6.   Loss of p27Kip1 increased DNA synthesis and cell growth following serum stimulation of mouse PASMCs. (A) [3H]thymidine incorporation; (B) cell growth of PASMC isolated from p27Kip1/ and wild-type mice. , p27Kip1+/+ PASMC; , p27Kip1/ PASMC (three separate experiments from three different animals in each group. *P < 0.05).

View larger version (17K): [in this window] [in a new window]   Figure 7.   p27Kip1 overexpression blunts DNA synthesis and cell growth in p27Kip1/ mouse PASMCs. (A) [3H]thymidine incorporation; (B) cell growth. , untreated cells; , vector control infection; , adenovirus-p27Kip1 (two separate experiments in two different animals. *P < 0.05).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Pulmonary hypertension is a heterogeneous disorder with the vascular phenotype dependent upon the mechanism and severity of injury. What is clear is that with increasing severity and duration, pulmonary hypertension is associated with progressively more significant vessel wall remodeling, ultimately resulting in fixed, irreversible lesions. Rabinovitch and colleagues (1) demonstrated that chronic hypoxia led to medial hypertrophy with peripheral vessel muscularization in rats (the severity depending on sex and age), whereas Fagan and coworkers (19) found much less impressive vessel wall remodeling in mice. In other animal studies, mechanical ventilation (20), the plant alkaloid monocrotaline (21), and monocrotaline plus increased flow (22, 23) have been shown to cause progressively more severe vascular remodeling, respectively. In humans, Chazova and Meyrick (3) demonstrated arterial obliteration, adventitial thickening, and evidence of recanalization in primary pulmonary hypertension while Rabinovitch and associates (5) found pulmonary artery endothelial lesions in patients with congenital heart disease, which correlated with the severity of disease. These and other reports suggest that additional study of the pulmonary vascular mitogenic response to hypoxia and injury are required. As various cdk inhibitors have been found important in controlling cellular proliferation in a variety of cell types, we undertook this study to determine the importance of the cyclin-dependent kinase inhibitor p27^Kip1 in controlling proliferation of PASMC.

This study demonstrates an important and specific role for p27^Kip1 in controlling cell proliferation in PASMC. We found that serum stimulation decreased p27^Kip1 protein expression at 24 h, which correlated with an increase in PASMC mitogenesis as determined by [³H]thymidine incorporation and percent of cycling cells. The peak of mitogenic activity at 24 h was associated with decreased p27^Kip1 binding to cyclin E with a corresponding increase in kinase activity. Consistent with subsequent restoration of p27^Kip1, the phosphorylation of retinoblastoma, a known cyclin E/cdk2 substrate (9, 17), was reduced both in vivo and in vitro by Day 3. Overexpression of p27^Kip1 blunted, while its absence augmented, the serum-stimulated increase in mitogenesis and cell cycle progression.

In addition to the initial decrease in p27^Kip1, we also found that p27^Kip1 expression in response to serum was biphasicthe initial decline followed by a return to baseline by 3 d. Expression inversely tracked the changes in cell activity as measured by cell cycle, [³H]thymidine incorporation, changes in cyclin E kinase activity, and hyperphosphorylation of retinoblastoma. The return of p27^Kip1 levels to that seen in growth-arrested cells occurred despite the continued presence of serum and the absence of cell confluence. The importance of this finding is uncertain, but suggests that additional homeostatic mechanisms may operate in this system, ultimately re-establishing p27^Kip1 levels and cell cycle control. Possible mechanisms for this phenomenon include an increase in TGF- (24) or cyclic AMP (25) (known inducers of p27^Kip1), a decrease in ubiquitin-mediated degradation (26), or some other autocrine or paracrine counter-regulatory pathway.

Evidence from other model systems suggests that p27^Kip1 is important in determining the response of the cell to proliferative stimuli and can direct cells toward proliferation, hypertrophy, or apoptosis (27). The mesangial cell hypertrophy characteristic of diabetic nephropathy has been shown to be mediated at least partially through p27^Kip1 (30, 31). Increased glucose concentrations cause mesangial cell hypertrophy which is associated with increased p27^Kip1 levels; however, in cells treated with antisense to p27^Kip1 or in p27^Kip1/ mesangial cells, increased glucose caused cell proliferation rather than hypertrophy. In the absence of serum, p27^Kip1/ mesangial cells underwent apoptosis, unlike wild-type cells (28).

The fate of stimulated vascular smooth muscle cells also appears to be dependent on p27^Kip1 levels. Servant and colleagues (29) established that cultured rat aortic cells underwent hypertrophy when exposed to angiotensin II, whereas they proliferated when exposed to PDGF, and that while exposure to PDGF decreased p27^Kip1 levels sufficiently to generate active cyclin E complexes, angiotensin II failed to do so. As a result, angiotensin II-treated cells were arrested in late G₁, while the PDGF-treated cells progressed through the cell cycle. Braun-Dullaeus and colleagues (27) confirmed these findings and also concluded that growth factors are needed in addition to angiotensin II to promote cell proliferation. In an in vivo model of vascular injury in porcine femoral arteries, Tanner and coworkers (14) showed that overexpression of p27^Kip1 reduced intimal proliferation following mechanical injury. Therefore p27^Kip1 appears to be important in limiting cell proliferation in response to injury and mitogens, with the role of p27^Kip1 determined by the cell type and the specific stimulus involved. Because myofibroblasts can also express smooth muscle cell -actin it is possible that our cell culture may have included a mixed population of pulmonary artery myofibroblasts and smooth muscle cells. Proliferation of both cell types has been implicated in the pulmonary vascular remodeling associated with pulmonary hypertension. Our findings would indicate that both cell types would be susceptible to regulation by p27^Kip1.

Our results suggest an important role for p27^Kip1 in modulating PASMC proliferation. It is possible that a decrease in p27^Kip1 expression due to excessive mitogenic stimulation or inactivation of p27^Kip1 by the adenovirus oncoprotein E1A (32) following a viral infection might predispose the pulmonary circulation to excessive vascular remodeling in response to normally controlled stimuli. Similarly, genetic differences in control of p27^Kip1 could modulate the pulmonary vascular response to injury. Whether decreases in p27^Kip1 expression or activity contribute to the pathologic structural changes associated with pulmonary hypertension, and whether increases in p27^Kip1 can block remodeling, remain to be determined. However, our studies implicate p27^Kip1 as a key regulator of PASMC cell cycle and suggest that intrinsic or acquired differences in regulation of p27^Kip1, in concert with increases in local growth factors, are likely to modulate the development of pulmonary vascular disease.