INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

VASCULAR SMOOTH MUSCLE cells (VSMC) do not proliferate in normal blood vessels from healthy adults. However, excessive VSMC proliferation does occur in vascular diseases, including atherosclerosis and chronic pulmonary hypertension, and it contributes to the pathogenic process in both diseases. A link between hypoxia and smooth muscle cell (SMC) proliferation was first suggested by certain findings in pulmonary hypertension. Chronic hypoxia is often a trigger to pulmonary hypertesion in humans, and the increase in pulmonary vascular resistance is mediated in part via SMC hyperplasia within the media of small pulmonary arteries (27, 50). SMC proliferation has also been demonstrated in the pulmonary vasculature in animal models of chronic hypoxia (44, 62).

More recently, hypoxia has been proposed to play a role in the development of atherosclerotic lesions (15). Cigarette smoking, carbon monoxide exposure, and chronic sleep apnea are associated with hypoxia and increased risk of atherosclerosis (1, 16, 52). Another mechanism that has been proposed to link hypoxia and atherosclerosis involves thrombus formation within the vasa vasorum, leading to hypoperfusion and hypoxia of the adventitia and ultimately inducing the release of cytokines that induce SMC proliferation (43). The cytokines that may contribute to SMC proliferation in this setting have not been identified.

Several in vitro studies have addressed possible mechanisms whereby hypoxia may induce VSMC proliferation. Early studies demonstrated that hypoxic endothelial cells (EC) release a factor(s) that enhances SMC proliferation (56, 57), and subsequent studies have shown that hypoxic EC can release basic fibroblast growth factor and promitogenic prostanoids (45). However, hypoxic EC can also release potential inhibitors of VSMC proliferation (29, 30); thus, the overall effect of EC-derived factors on SMC mitogenesis during hypoxia is unclear. Exposure to hypoxia can also stimulate proliferation of bovine VSMC directly. Bovine SMC proliferate faster when exposed to moderate hypoxia (3% O₂) but only if the cells are costimulated with phorbol ester (18, 17) or serum (19). These results suggest that hypoxia provides a comitogenic stimulus in the presence of other growth factors. However, it is not known whether hypoxia has either mitogenic or comitogenic effects on human VSMC or whether hypoxia induces human VSMC to produce an autocrine growth factor.

The proinflammatory cytokine interleukin (IL)-1 is an intriguing candidate for a hypoxia-induced SMC-derived autocrine growth factor. Human VSMC produce both the  and  forms of IL-1 (6, 39), and both IL-1 and IL-1 are potent VSMC growth factors (40, 4). IL-1 is also a potent autocrine growth factor, as indicated by the markedly enhanced proliferative rate of human VSMC that have been stably transfected with IL-1 expression plasmids that direct the production of low (pg) levels of IL-1 (4). Finally, hypoxia can induce IL-1 production, as demonstrated by in vitro studies in which exposure to severe hypoxia (0% O₂) induces the release of bioactive IL-1 from human EC (54). However, the effect of hypoxia on IL-1 production may be cell type specific, since human macrophages, unlike EC, do not produce IL-1 when exposed to hypoxia but produce it upon subsequent reoxygenation (24, 35).

The present studies first assessed whether exposure to severe hypoxia, or alternatively reoxygenation after hypoxia, induces IL-1 production by human VSMC. Severe hypoxia itself stimulated IL-1 production in human VSMC, reoxygenation reversed this effect, and the levels of IL-1 produced were similar to those that stimulate proliferation in human VSMC that have been stably transfected with IL-1 expression plasmids. Therefore, further studies were designed to test the hypothesis that either moderate or severe hypoxia stimulates proliferation of human VSMC by inducing autocrine production of IL-1.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Human VSMC culture. Segments of pulmonary artery, aorta, and saphenous vein were obtained from organ donors (National Disease Research Interchange, Philadelphia, PA). Segments of saphenous vein were also obtained from coronary bypass patients at New England Medical Center (approved by the Human Investigation Research Committee). Human VSMC were cultured from explants as previously described (5). Cells were grown in DMEM supplemented with 10% FCS, penicillin, streptomycin, and fungizone and were passaged every 2-4 wk using trypsin-EDTA. Cells were passaged at a 1:3 ratio and were used after reaching confluence, at passage 1-4. In some experiments, human VSMC were serum deprived in DMEM-Ham's F-12 media (1:1) supplemented with insulin (1 µM) and transferrin (5 µg/ml; DMEM-F-12-IT) for 48 h before exposure to hypoxia, and then the medium was changed to fresh DMEM-F-12-IT. In other experiments, human VSMC were exposed to hypoxia in DMEM supplemented with 1-10% FCS.

Exposure of human VSMC to hypoxia/reoxygenation. VSMC were exposed to low O₂ tension in a humidified modular incubator chamber (Billups-Rothenberg, Del Mar, CA) that was maintained at 37°C. To expose VSMC to low O₂, the incubation medium was pregassed for 1 h with 95% N₂-5% CO₂ before the experiment. VSMC were washed two times with PBS, the medium was replaced with pregassed low O₂ medium, and the culture plates placed in the incubator chamber, which was sealed and purged with 95% N₂-5% CO₂ for 20 min. Chambers were reequilibrated with 95% N₂-5% CO₂ at 24-h intervals. Reoxygenation involved transferring the cells to 95% ambient air-5% CO₂ at 37°C. VSMC were exposed to intermediate (5%) O₂ levels as described above, except the medium was not pregassed, and the incubator chamber was purged with 90% N₂-5% O₂-5% CO₂ for 20 min. Normoxic control VSMC were exposed to 95% ambient air-5% CO₂ for the entire incubation period.

A portable O₂ analyzer (Hudson, Ventronics Division) was used to determine the O₂ content of the air inside the chamber at the end of the hypoxic exposure period. When chambers were purged with 0% O₂, O₂ readings were either 0 or 1% (to the nearest percentage) at the end of the hypoxia period (referred to as low O₂ in text). O₂ readings were consistently 5% when chambers were purged with 5% O₂. One experiment was excluded from analysis, since low O₂ was not sustained at the end of the incubation period. Hypoxia had no effect on the pH of the medium. The viability of human pulmonary artery SMC (HPASMC), human aortic SMC (HASMC), and human saphenous vein SMC (HSVSMC), assessed by trypan blue exclusion and by measuring lactate dehydrogenase activity in cell supernatants (Promega, Madison, WI), was not affected by incubation in low O₂, and cell morphology was unchanged.

RT-PCR. Total RNA was extracted from human VSMC using RNAzol B (Biotecx Laboratories, Houston, TX). RNA was reverse transcribed for PCR analysis in a total volume of 20 µl containing 2 µg RNA, 200 units of Moloney murine leukemia virus reverse transcriptase (GIBCO-BRL), 25 µg/ml oligo(dT) primer (GIBCO-BRL), 0.5 mM dNTP, and 10 mM dithiothreitol in the manufacturer's recommended buffer. The reactions were carried out at 37°C for 1 h and were terminated by heating to 95°C for 5 min. PCR mixtures contained 2 µl of cDNA, 50 mM KCl, 10 mM Tris · HCl (pH 8.3), 1.0-1.5 mM MgCl₂, 0.01 mg/ml gelatin, 200 µM of each deoxynucleotide phosphate, 0.15 µCi [³²P]dCTP (20 nM), 250 nM of each primer, and 1.25 units of Taq polymerase (Pharmacia) in a final volume of 20 µl. Primer sequences and annealing temperatures are given in Table 1. Amplification reactions were begun with an initial melt of 94°C for 2 min, consisted of 16-32 cycles of 1 min at 94°C, 1 min at the annealing temperature, and 2 min at 72°, and were completed with a final extension at 72°C for 5 min. Each sample was analyzed in two or three reactions of differing cycle number, within the exponential phase of amplification. Products were separated by gel electrophoresis on a 5% polyacrylamide gel and were visualized by ethidium bromide staining. PCR products were quantitated by cutting the band from the gel and counting the incorporated [³²P]dCTP by liquid scintillation counting. Levels of mRNA were determined by normalizing the amount of each PCR product (counts/min) to the amount of -actin product obtained for the same cDNA sample. The level of mRNA in cells exposed to hypoxia was expressed relative to the value for VSMC exposed to 20% O₂ alone for the same time period.

IL-1, IL-1, and IL-1-receptor antagonist measurements. Cell lysates were prepared by three successive freeze-thaw cycles in buffer containing 10 mM sodium phosphate, 0.15 M NaCl, 0.25% BSA, and 0.05% sodium azide. Both lysates and supernatants were cleared of cellular debris by centrifugation at 500 g for 10 min. Immunoreactive IL-1 in cell lysates and cell supernatants was measured by an enzyme immunometric assay (EIA) that detects both IL-1 precursor and mature IL-1 (Cayman Chemical, Ann Arbor, MI) at a detection limit of 1-2 pg/ml (26). IL-1 in cell supernatants was measured by an EIA (Cayman Chemical) that detects mature IL-1 (the active form of IL-1) preferentially, at a detection limit of 1-2 pg/ml. Immunoreactive IL-1-receptor antagonist (ra) in cell lystes was measured by a specific RIA that detects both secreted and intracellular (ic) forms of IL-1ra with a detection limit of 35-70 pg/ml (49).

Bromodeoxyuridine incorporation. Human VSMC were plated at 2,000 cells/well in 96-well plates in DMEM supplemented with 10% FCS. After the cells were attached to the plastic, the medium was replaced with DMEM supplemented with 2% FCS, and the plates were transferred to low, 5%, or 20% O₂. After 3 days, the medium was changed to fresh DMEM containing 2% FCS and 5-bromo-2'-deoxyuridine (BrdU; Boehringer Mannheim). After 24 h, the cells were fixed, and DNA was denatured; next, the fixed cells were incubated 2 h at room temperature with mouse monoclonal BrdU antibody conjugated to peroxidase (Boehringer Mannheim). Tetramethylbenzidine was used as substrate, and absorbance at 405 nm was determined. In some of the experiments, human recombinant IL-1ra (10 µg/ml) or IL-1 (1 ng/ml) was added to the media before exposure to hypoxia or normoxia and was also added to the fresh media containing BrdU.

Proliferation rates. Human VSMC were plated in 24-well plates at a density of 5,000 cells/cm², in complete growth media containing 10% FCS. After the cells had adhered to the plastic, the medium was replaced with media supplemented with 1, 2, 5, or 10% FCS, and the plates were transferred to low, 5%, or 20% O₂. Fresh medium was replaced every 3-4 days. Quadruplicate wells of cells exposed to each O₂ concentration were trypsinized on the day after plating (day 0) and after 4-13 days of exposure to low O₂, and cell number was determined using a Coulter Counter.

Data analysis. Values presented are means ± SE of four to eight replicate cultures for each treatment group in a representative experiment. Values that are below the limit of detection of the EIA are indicated by ND (nondetectable). Within experiments, the significance of treatment-induced differences was determined either by t-test or by ANOVA followed by Dunnett's procedure to compare multiple means with a single control value. P < 0.05 was considered statistically significant. Each experiment was repeated in VSMC derived from different blood vessels and different patients or donors, as described.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IL-1 and IL-1 gene expression are enhanced in human VSMC incubated in low O₂. Because the pulmonary vasculature is sensitive to the effects of hypoxia in vivo, the effects of low (~1%) O₂ and subsequent reoxygenation on cytokine gene expression were initially assessed in SMC derived from HPASMC. Both IL-1 and IL-1 mRNA were detectable in HPASMC before exposure to low O₂. After exposure to low O₂ tension for 48 h, IL-1 mRNA levels increased 3.7-fold compared with the levels in HPASMC incubated in 20% O₂, and IL-1 mRNA levels increased 1.7-fold (Fig. 1). Subsequent reoxygenation rapidly reversed the effects of low O₂ on IL-1 and IL-1 mRNA. Both IL-1 and IL-1 mRNA levels remained moderately elevated after 4 h of reoxygenation (1.9- and 1.4-fold, respectively) but were not different from time controls after 8 h of reoxygenation. Levels of -actin mRNA were similar in HPASMC incubated in either 20% O₂ or low O₂ and were not affected by reoxygenation.

View larger version (80K): [in this window] [in a new window]   Fig. 1.   Exposure to low O2 increases levels of interleukin (IL)-1, IL-1, and intracellular (ic) IL-1-receptor antagonist (ra) mRNA in human pulmonary artery smooth muscle cells (HPASMC). RNA was isolated from HPASMC after 48 h of exposure to low O2 alone (lane 2) or after 4 or 8 h of reoxygenation (lanes 4 and 6, respectively). Levels of mRNA were analyzed by RT-PCR with amplification within the exponential phase [cycles of PCR amplification: IL-1, (28); IL-1 (28); tumor necrosis factor- (TNF-; see Ref. 30), icIL-1ra (32); and -actin (18)]. Corresponding time controls (48, 52, or 56 h in 20% O2) are shown in lanes 1, 3, and 5, respectively. RT-PCR products obtained from HSVSMC treated with IL-1 (2 ng/ml) for 24 h are shown in lane 7.

Cell-associated IL-1 is increased in human VSMC incubated in low O₂. Incubation in a low-O₂ environment stimulated IL-1 production by human VSMC. Cell-associated IL-1 levels increased progressively in HPASMC exposed to low O₂ for 12-48 h (Fig. 2A). In contrast, the level of cell-associated IL-1 in control HPASMC decreased throughout the course of the experiment, an effect that may be due to progressive serum deprivation (48 h serum deprivation at the start of the experiment vs. 96 h by the end of the experiment). When compared with the corresponding time control HPASMC, cell-associated IL-1 levels were increased in hypoxic HPASMC by 1.5-, 2.5-, and 18-fold, respectively, after 12, 24, and 48 h of exposure to low O₂.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   A: time-dependent increase in cell-associated IL-1 in HPASMC exposed to low O2 for 4-48 h. B: cell-associated IL-1 levels decreased in HPASMC that were reoxygenated after 48 h of exposure to low O2 and were nondetectable in the corresponding time controls. HPASMC were serum deprived for 48 h before exposure to low O2 in fresh serum-free media. At the times indicated, cells were lysed in enzyme immunometric assay (EIA) buffer, and cellular levels of IL-1 were determined by EIA. ND, nondetectable (<0.04 ng/106 cells). IL-1 was not detectable in cell supernatants. * P < 0.002, significantly different from 20% O2 at same time point.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Exposure to low O2 increases cell-associated IL-1 in VSMC derived from human aorta (HASMC) or saphenous vein (HSVSMC). A: HASMC were exposed to low or 20% O2 for 48 h, followed by 0 or 9 h of reoxygenation or continued oxygenation, as in Fig. 2B, except serum deprivation was begun upon exposure to low O2. B: HSVSMC (passage 2) derived from 4 different patients were exposed to 20% or low O2 for 48 h, as in Fig. 2A. ND, <0.02 ng/106 cells. * P < 0.01, significantly different from 20% O2 at same time point.

View this table: [in this window] [in a new window]   Table 2.   Effect of low O2 on IL-1, IL-1, and IL-1ra

Proliferation of human VSMC in low O₂ is not influenced by IL-1 receptor activation. To test whether IL-1 produced during exposure to low O₂ had a proliferative effect on human VSMC, BrdU incorporation was measured in cells that had been exposed to low O₂ for 72-96 h to allow sufficient time for induction of IL-1 synthesis and for the produced IL-1 to enhance DNA synthesis. Exposure to low O₂ did not significantly affect the proliferative rate of human VSMC. In seven experiments with different human VSMC cell lines (HPASMC, HASMC, and HSVSMC), BrdU incorporation was not significantly enhanced by exposure to low O₂ in the presence of 2% FCS (1.08 ± 0.24 relative to normal O₂). Although there was a tendency for BrdU incorporation to be increased in passage 1 human VSMC exposed to low O₂ (1.54 ± 0.42 relative to 20% O₂, n = 3 experiments) and attenuated in passage 3-4 human VSMC that were exposed to low O₂ (0.74 ± 0.12 relative to normal O₂, n = 4 experiments), neither of these effects was statistically significant. A representative experiment with passage 1 HSVSMC is shown in Fig. 4.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Exposure to 5% O2, but not low O2, stimulates DNA synthesis in HSVSMC; proliferation is not affected by exogenous IL-1ra. HSVSMC (passage 1) were plated in 96-well plates in complete media. After cells had attached to the plastic, medium was changed to DMEM supplemented with 2% FCS, and some of the plates were transferred to either 5% or low O2 for 72 h. At 72 h, the medium was replaced with fresh DMEM supplemented with 2% FCS and bromodeoxyuridine (BrdU). Some of the cells were exposed to IL-1ra (10 µg/ml) from the time of plating, with fresh IL-1ra added with the BrdU-containing media. BrdU incorporation was determined at 96 h. * P < 0.002, significantly different from 20% O2 without IL-1ra.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Exposure to low O2 does not attenuate the mitogenic response to exogenous IL-1 in human VSMC. Human VSMC were plated as described in Fig. 4, and medium was changed to DMEM supplemented with either 2% FCS, 2% FCS and IL-1 (1 ng/ml), or 10% FCS. After 72 h, medium was replaced with fresh medium containing BrdU, and BrdU incorporation was determined at 96 h. BrdU incorporation is expressed relative to values for cells incubated in the presence of 2% FCS and 20% O2. * P < 0.0005, significantly different from control at the same O2 level. + P < 0.0001, significantly different from 10% FCS and 20% O2.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Reoxgenation after exposure to low O2 significantly enhances proliferation of human VSMC; proliferation is not affected by exogenous IL-1ra. Human VSMC were plated as described in Fig. 4, and medium was changed to DMEM supplemented with 2% FCS with or without IL-1ra (10 µg/ml). Some of the plates were transferred to low O2 for 72 h, and then medium was replaced with fresh DMEM supplemented with 2% FCS and BrdU, with or without IL-1ra, and all plates were returned to 20% O2. BrdU incorporation was determined at 96 h. Values shown are means ± SE of 4 experiments with HSVSMC, HASMC, and HPASMC. BrdU incorporation is expressed relative to values for cells incubated in 20% O2 without IL-1ra. * P < 0.02, significantly different from 20% O2 without IL-1ra.

Exposure to low O₂ induces modest increases in production of icIL-1ra. Previous studies have shown that HSVSMC produce icIL-1ra, a potential inhibitor of IL-1 action, and its production is stimulated by various stimuli, including IL-1 and platelet-derived growth factor (6). Expression of icIL-1ra mRNA was enhanced in HPASMC after exposure to low O₂ tension for 48 h. cDNA prepared from HPASMC that had been exposed to low O₂ tension for 48 h yielded visible PCR product of the expected size (510 bp) after 32 cycles of amplification, whereas cDNA prepared from HPASMC incubated in 20% O₂ did not yield visible product (Fig. 1). However, the magnitude of the hypoxia-induced increase in icIL-1ra was modest compared with the magnitude of the induction elicited by IL-1 (Fig. 1), a known potent activator of icIL-1ra production (6). The level of icIL-1ra transcript was 10-fold lower in HPASMC exposed to low O₂ compared with HSVSMC stimulated with IL-1 (2 ng/ml) for 24 h, as determined by quantitation of the PCR bands.

Exposure to 5% O₂ enhances human VSMC proliferation by an IL-1-independent mechanism. Exposure to 5% O₂ tension consistently enhanced BrdU incorporation in human VSMC, including HPASMC, HASMC, and HSVSMC. The effect of 5% O₂ was similar in SMC derived from different blood vessels. BrdU incorporation was consistently increased after 72 h of exposure to 5% O₂ in the presence of 2% FCS (1.55 ± 0.11 relative to 20% O₂, P < 0.002, n = 9 experiments). Passage 1 human VSMC responded more to the effect of 5% O₂; BrdU incorporation increased by 1.76 ± 0.13 relative to 20% O₂ (n = 5 experiments, P < 0.005). A representative experiment with passage 1 HSVSMC is shown in Fig. 4. BrdU incorporation was also enhanced in passage 3 or 4 human VSMC incubated in 5% O₂; however, the magnitude of the effect was less (1.28 ± 0.03 relative to 20% O₂, n = 4, P < 0.003).

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Exposure to 5% O2 stimulates proliferation of HPASMC and HSVSMC in media containing either low or high concentrations of FCS. A: HSVSMC (passage 1) were plated in DMEM supplemented with 10% FCS overnight. Initial cell counts were determined by Coulter counter (day 0 = 2,330 cells/cm2), and the medium was changed on the other 2 plates to DMEM supplemented with varying concentrations of FCS. One of the plates was then transferred to 5% O2 for 13 days. Fresh medium was replaced every 3-4 days. Increments in cell counts (day 13  day 0) are shown. B: HPASMC (passage 1) were studied as in A, except the cells were incubated 4 days in 5% O2 with no media changes. Increments in cell counts (day 4  day 0) are shown; cell counts on day 0 = 5,060 cells/cm2. * P < 0.001 compared with human VSMC incubated in 20% O2 and the same concentration of FCS.

View this table: [in this window] [in a new window]   Table 3.   Effect of 5% O2 on IL-1 and IL-1

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates that exposure to very low O₂ levels induces IL-1 gene expression and increases the cell-associated levels of IL-1 in VSMC derived from human aorta, pulmonary artery, and saphenous vein. Cell-associated levels of IL-1 increased progressively after exposure to low O₂ and returned to control levels upon subsequent reoxygenation. IL-1 levels were not increased in human VSMC incubated in 5% O₂, indicating that a severe hypoxic stimulus is required to stimulate IL-1 production. IL-1 began to increase after 12 h of exposure to 0% O₂ and markedly increased after 48 h. In contrast, human EC produce IL-1 after only 4 h of exposure to 0% O₂ (54), and human mononuclear phagocytes produce IL-1 when reoxygenated after only 9 h of exposure to 0% O₂. Shorter periods of low O₂ tension appear to induce IL-1 production by EC compared with SMC. It is possible that 0% O₂ represents a stronger hypoxic stimulus for EC, since they are exposed to a higher range of O₂ tension in vivo compared with SMC. In vivo, EC are exposed directly to O₂ tensions that occur in blood. In contrast, VSMC can be relatively distant from the blood, depending on the thickness of the blood vessel, their location in the vessel wall, and the presence or absence of vasa vasorum, and thus can be exposed to O₂ tensions that are significantly lower than that of blood.

Evidence from in vitro studies suggests that IL-1 produced by vascular cells versus mononuclear phagocytes may be produced by distinct stimuli and may have distinct roles in vivo. Monocytes do not produce IL-1 during hypoxia; however, newly synthesized IL-1 is released during the subsequent reoxygenation period, and the stimulus for IL-1 synthesis is the generation of O₂ free radicals (35). Because IL-1 is produced during exposure to low O₂ tension in both EC (54) and VSMC, the stimulus to induction of IL-1 may not be generation of O₂ free radicals. VSMC-derived IL-1 may play a role in pathophysiology associated with chronic ischemia or hypoxia, whereas mononuclear phagocyte-derived IL-1 may contribute to reperfusion injury.

Hypoxia also induced a modest increase in the levels of IL-1 mRNA in HPASMC. IL-1, like IL-1, is initially synthesized as a precursor protein that is cleaved to the mature form by IL-1-converting enzyme and subsequently released. However, in contrast to IL-1, the precursor form of IL-1 is inactive (46); IL-1 is only active when released as a mature molecule that activates IL-1 receptors. In the present study, lowering O₂ tension from 20% to either 5 or ~1% did not enhance IL-1 release from human VSMC. Thus IL-1 is unlikely to have autocrine effects in hypoxic human VSMC. Likewise, TNF-, which is also produced by human VSMC (58) and induces VSMC proliferation (48, 51), is unlikely to mediate the effects of hypoxia or reoxygenation in cultured human VSMC, since TNF- mRNA was not expressed by HPASMC that were exposed to low O₂ or subsequent reoxygenation.

IL-1 produced by human VSMC that were exposed to low O₂ remained cell associated and was not released from the cells either during the hypoxia or reoxygenation periods. Human EC also produce IL-1 when exposed to 0% O₂; however, unlike VSMC, IL-1 is released from hypoxic EC into the supernatant (54). When stimulated with IL-1, TNF, or lipopolysaccharide, human EC likewise release IL-1, whereas human VSMC produce IL-1 but do not release it (6, 41). Together these findings suggest that IL-1 may be more efficiently released by human EC compared with human VSMC. It is not entirely clear whether cellular release of IL-1 is required for its subsequent bioactivity, since IL-1 has been proposed to affect cellular function by alternative pathways. The precursor form of IL-1 is thought to associate with the plasma membrane of human VSMC and macrophages in a form that can stimulate IL-1 receptors on cells that directly contact IL-1-producing cells (13, 38, 41, 63). IL-1 precursor may also affect gene expression by localizing directly to the nucleus, without prior release from the cell. IL-1 precursor contains a nuclear localization sequence in the precursor domain of the molecule and localizes to the nucleus of EC, fibroblasts, and VSMC that have been transiently transfected with IL-1 precursor expression plasmids (4, 42, 60). Also, stable transfection of human EC and VSMC with IL-1 precursor expression plasmids produces effects similar to exogenous IL-1; however, these effects are not reversed by exogenous IL-1ra in human EC (42) and are only partially reversed in human VSMC (4). Therefore, it is possible that IL-1 produced by hypoxic human VSMC is biologically active in the absence of cellular release.

Previous studies have documented that IL-1 is a potent growth factor for VSMC, including those subcultured from human pulmonary artery or saphenous vein (40, 4) and primary cultures of rat aortic SMC (RASMC; see Ref. 10). However, IL-1 can also induce growth-inhibitory pathways, and the overall effect on proliferation is highly dependent on the type of VSMC tested, the specific assay conditions used, and the length of IL-1 exposure. IL-1 can inhibit proliferation of subcultured RASMC by inducing the expression of inducible nitric oxide (NO) synthase and the generation of growth-inhibitory NO (14, 23, 53). Human VSMC do not express inducible NO synthase when stimulated with IL-1 (5) and thus provide a more suitable model for studies of the pro-proliferative actions of IL-1. IL-1 can also induce the production of growth-inhibitory prostanoids, which suppress the early mitogenic response; IL-1 induces DNA synthesis only in SMC treated with indomethacin when assessed 24-48 h after initial exposure (40, 36). In the present study, exogenous IL-1 was an effective mitogen in the absence of indomethacin when assessed after longer exposures to IL-1 (72-96 h), consistent with earlier findings with RASMC (31). Also, inhibition of prostanoid synthesis with indomethacin did not significantly affect the proliferative responses produced by either IL-1 or low O₂ (data not shown).

Studies with transfected human VSMC have shown that IL-1 is also a potent autocrine growth factor for these cells. Human VSMC that have been stably transfected with IL-1 expression plasmids proliferate rapidly compared with human VSMC transfected with vector alone (4). Exposure to low O₂ induces human VSMC to produce IL-1 at levels similar to or greater than the levels produced by IL-1 stable transfectants. In contrast with IL-1 stable transfectants which proliferate rapidly, proliferation was not significantly affected in human VSMC exposed to low O₂. It is not clear why proliferation is not enhanced in hypoxic human VSMC despite the fact that they are producing IL-1 similar to that of IL-1 stable transfectants, which proliferate rapidly. Human EC exposed to severe hypoxia released IL-1 but were insensitive to its effects, and then, upon reoxygenation, the cells continued to release IL-1 and displayed IL-1-dependent expression of leukocyte adhesion (12, 54). In contrast, human VSMC remained responsive to the proliferative effects of IL-1 when incubated in low O₂, suggesting that insensitivity to released IL-1 does not account for the lack of IL-1-dependent proliferation in human VSMC exposed to low O₂. However, we cannot rule out the possibility that hypoxic VSMC are insensitive to nuclear or membrane-associated IL-1 or that IL-1 is not effectively targeted to its active site in hypoxic VSMC. Alternatively, it is possible that IL-1 is only active when released by VSMC. Stable transfectants release low levels of IL-1 (pg/ml), and these low levels can induce human VSMC proliferation (4). In contrast, hypoxic VSMC do not release IL-1 either during exposure to hypoxia or during the subsequent reoxygenation period, and proliferation during either hypoxia or reoxygenation is not dependent on IL-1 receptor activation. Interestingly, human VSMC, including those derived from human pulmonary artery, aorta, and saphenous vein, proliferated faster in a 5% O₂ environment compared with in 20% O₂ when assessed in the presence of either low or high serum concentrations. The finding that serum-induced VSMC proliferation is enhanced by moderate hypoxia is consistent with previous findings. The growth of cells from human saphenous vein explants in the presence of 10% FCS was enhanced in a 5% O₂ environment compared with 20% O₂ (55). Also, exposure to 3% O₂ enhanced proliferation of SMC derived from bovine pulmonary artery that were costimulated with either phorbol ester (18, 17) or serum (19). However, bovine SMC did not proliferate when exposed to 3% O₂ alone, without serum or phorbol ester. In contrast, 5% O₂ induced proliferation of human VSMC under low serum conditions, which maintained cell viability but did not induce proliferation. Thus hypoxia alone is an effective mitogenic stimulus for human VSMC.

Recent studies have documented that arteries are composed of phenotypically diverse subpopulations of SMC and cells with nonmuscle characteristics (20). Four distinct populations of SMC have been identified within the media of bovine pulmonary artery and aorta, based on distinct morphological appearance, differential expression of muscle-specific proteins, and differential localization between the inner, middle, and outer media (21). Exposure to hypoxia in vivo induces proliferation of a meta-vinculin-negative SMC population in bovine neonatal pulmonary arteries (62). Four distinct cell types can also be isolated from bovine pulmonary artery and aorta in vitro, and these cells retain their characteristics over time in culture. Two of the cell types were SMC-like and proliferated slowly in serum, and moderate hypoxia further inhibited their proliferative rate, whereas the other two types were non-muscle-like and proliferated rapidly in serum, and moderate hypoxia further enhanced their proliferative rate (19). SMC derived from human blood vessels are likewise phenotypically heterogeneous with respect to morphological appearance, growth rates (22), responsivenesss to growth factors (8), and expression of muscle-specific proteins (25). Thus it seems likely that the mixed cultures of human VSMC used in the present study contain subpopulations of cells that respond differentially to hypoxia.

Levels of cell-associated IL-1 were not increased in human VSMC incubated in 5% O₂. Studies with exogenous IL-1ra further ruled out the possibility that the proliferative response to 5% O₂ involves a membrane-associated form of IL-1 that may not be detected by EIA analysis of cell lysates. Exogenous IL-1ra, by binding to the type I IL-1 receptor, can inhibit the action of either IL-1, which is released by cells, or membrane-associated IL-1 (32); however, higher concentrations of IL-1ra are required to inhibit the membrane-associated form. In the present study, high concentrations of exogenous IL-1ra did not affect proliferation of human VSMC incubated in either low, 5%, or 20% O₂, ruling out a role for released IL-1 ( or ) or membrane-associated IL-1 in human VSMC proliferation in low or normal O₂ tension. Although the studies with exogenous IL-1ra do not rule out a possible intracellular action of IL-1, this possibility seems unlikely, since IL-1 was not detectable in the lysates of human VSMC incubated in 5% O₂. These results demonstrate two direct and distinct effects of hypoxia on VSMC. Exposure to intermediate O₂ tension induces VSMC proliferation, independent of IL-1, whereas exposure to low O₂ tension induces production of IL-1.

We have previously reported that HSVSMC produce an icIL-1ra (6) that lacks the signal peptide sequence found in the secreted form of IL-1ra (sIL-1ra) and remains cell associated. Production of icIL-1ra is augmented in human VSMC that have been stimulated with exogenous IL-1, phorbol ester, platelet-derived growth factor, or transforming growth factor- (6). In the present study, exposure to low O₂ induced icIL-1ra gene expression in HPASMC; however, the effect was modest, and the amount of IL-1ra produced (1 ng/10⁶ cells) was low compared with that produced by HSVSMC that have been stimulated with IL-1 or phorbol ester (6-14 ng/10⁶ cells; see Ref. 6). icIL-1ra, in contrast to sIL-1ra, has been proposed to act as a unique intracellular inhibitor that attenuates IL-1 signaling at a point downstream from the IL-1 receptor in cancer cell lines (59). It is not known whether icIL-1ra modulates IL-1 signaling in VSMC or whether the modest amounts of icIL-1ra that are produced during hypoxia would be effective.

Chronic alveolar hypoxia is a common cause of pulmonary hypertension (27), and hypertrophy and hyperplasia of VSMC within the media of distal pulmonary arteries is a common feature of this disease (50, 44). Systemic hypoxia may also aggravate the development of atherosclerotic lesions. Cigarette smoking (52), carbon monoxide exposure (1), and chronic sleep apnea (16) are associated with systemic hypoxia and increased incidence of atherosclerosis. Experimentally induced arterial hypoxia enhances the development of atherosclerotic lesions in cholesterol-fed rabbits (34), whereas arterial hyperoxia inhibits lesion development (33). Because both EC and VSMC become hypoxic in the setting of alveolar or systemic hypoxia, both cell types may release VSMC mitogens that contribute to the pathophysiological response, including IL-1, basic fibroblast growth factor, and promitogenic prostanoids (45, 54).

Local hypoxia has also been proposed to be a contributing factor to the development of atherosclerotic lesions, since intimal thickening itself impairs diffusion of O₂ from the vessel lumen. Also, O₂ consumption is increased in diet-induced atherosclerotic lesions of rabbits (61), most likely due to increased consumption in foam cells in which cholesteryl esters are continually hydrolyzed and reesterified (9, 11). It is possible that local hypoxia can also initiate the development of atherosclerotic lesions, since occlusion or removal of the vasa vasorum in experimental animals results in intimal hyperplasia (47, 43, 2, 3). It has been proposed that human atherosclerosis may be enhanced when a thrombus forms within the vasa vasorum, leading to hypoperfusion and hypoxia of the adventitia and ultimately to the release of cytokines that induce SMC proliferation (43). Because high O₂ consumption by foam cells, intimal thickening, or thrombosis within the vasa vasorum would lead to local hypoxia within the vessel wall, and EC would remain well-oxygenated, direct effects of hypoxia on VSMC may predominate in this setting.

In summary, the present studies provide evidence that hypoxia directly stimulates proliferation of human VSMC. Hypoxia-induced VSMC proliferation may contribute to the pathophysiological effects of chronic hypoxia on the vasculature and thus may play a role in the development of pulmonary hypertension or atherosclerosis. Exposure to moderate hypoxia induced human VSMC proliferation without inducing IL-1 production, whereas exposure to severe hypoxia induced IL-1 production. Hypoxia-induced IL-1 production by human VSMC may contribute to the pathophysiological effects of hypoxia on blood vessels.