INTRODUCTION

Top Abstract Introduction Materials and methods Results Discussion References

VASCULAR SMOOTH MUSCLE cells (VSMC) do not normally divide but begin to proliferate in response to injury. This process of VSMC activation and proliferation, known as neointimal hyperplasia, is responsible for the failure of many vascular reconstructive surgeries, including balloon angioplasty, coronary artery and leg vein bypass surgery, and insertion of vascular access grafts, and also contributes to the pathogenesis of vascular diseases such as atherosclerosis (1, 33-35). Classical growth factors, including platelet-derived growth factor and basic fibroblast growth factor, vasoconstrictors such as angiotensin II, and proinflammatory cytokines such as interleukin-1 (IL-1) can stimulate VSMC proliferation; however, the role of each factor in neointimal hyperplasia is not clear (31). In human VSMC, IL-1 is a particularly effective mitogen, markedly stimulating proliferation upon chronic exposure (23). Consequently, IL-1 released locally from activated macrophages within the blood vessel wall has been proposed to contribute to neointimal hyperplasia (23). VSMC themselves can also produce IL-1, including the -form, when stimulated in vitro with proinflammatory cytokines (4, 41). In vivo studies lend support to the notion that IL-1 may play a role in neointimal hyperplasia that occurs after coronary bypass surgery. After surgery, immunoreactive IL-1 was found in spindle-shaped cells of saphenous vein bypass grafts that had become stenotic but was absent in internal mammary arteries that had remained patent and in normal arteries and veins (8). However, it is not known whether IL-1 produced intrinsically by VSMC can function as an autocrine growth factor. The primary objective of the present study was to test the hypothesis that IL-1 promotes VSMC proliferation when produced intrinsically by VSMC at levels comparable to those produced after activation with pathophysiologically relevant stimuli.

IL-1 has been proposed to exert autocrine effects on cell function by several distinct mechanisms. IL-1 is synthesized as a 271-amino acid precursor molecule [IL-1-(1271)] that lacks a classical signal sequence (11). Subsequent processing and release of IL-1 may vary between different cell types. In macrophages, IL-1-(1271) is cleaved by a calpainlike enzyme to generate a NH₂-terminal domain [IL-1-(1112)] and mature IL-1 [IL-1-(113271)], and IL-1-(113271) is released by unknown mechanisms into the extracellular space (9, 19). In contrast, keratinocytes do not process IL-1-(1271) but release the full-length IL-1 precursor in a regulated fashion upon mechanical perturbation of the cell membrane (22). Although there is no evidence to date that human VSMC either process or release IL-1-(1271), upon release either IL-1-(1271) or IL-1-(113271) can activate the type I IL-1 receptor (28). Membrane-associated IL-1-(1271) is also thought to activate the type I IL-1 receptor on adjacent cells via juxtacrine mechanisms (2, 18, 24), and thus IL-1-(1271) could act as a membrane-anchored growth factor. Recent studies have suggested that IL-1-(1271) may also act within the cell, by a mechanism involving direct localization to the nucleus (16, 25, 27, 42). Finally, IL-1-(1112), which contains the putative nuclear localization sequence, may be biologically active by itself, since it appears to act as a transforming nuclear oncoprotein in rat mesangial cells (37).

The present study addressed whether IL-1 is an autocrine growth factor by assessing the proliferative rate of human VSMC stably transfected with expression plasmids that direct constitutive expression of full-length IL-1-(1271). To assess which domains of IL-1 precursor are required for autocrine growth activity, the effects of stable expression of plasmids that direct constitutive expression of either IL-1-(1112) or IL-1-(113271) were compared. Because IL-1-(113271) lacks the nuclear localization sequence (amino acids 79-86) found in IL-1-(1271) and does not localize to the nucleus (25, 42) or associate with the plasma membrane (5, 7, 13), its expression should reveal effects mediated by release from the cell and activation of type I IL-1 receptors.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and methods Results Discussion References

Culture of human saphenous vein VSMC. Primary cultures of human saphenous vein VSMC (HSVSMC) were grown by explant technique from unused portions of saphenous veins harvested for coronary artery bypass surgery at New England Medical Center (approved by the Human Investigation Research Committee, New England Medical Center). Briefly, segments of vein were cleared of endothelium and adventitia, cut into 3-mm² pieces, and placed in gelatin-coated six-well plates. VSMC typically grew out of the explants within 21-28 days. VSMC were also grown from segments of human pulmonary artery obtained from organ donors (National Disease Research Interchange, Philadelphia, PA). Cells were grown in DMEM supplemented with 10% FCS, glutamine, penicillin, streptomycin, and Fungizone (growth media). Chronic IL-1-stimulated cells were cultured in growth media supplemented with IL-1 or IL-1 (1 ng/ml) for 7-110 days. Growth media with or without IL-1 or IL-1 were changed twice a week, and cells were passaged with trypsin (0.05%) and EDTA (0.53 mM).

IL-1 expression plasmids. The IL-1-(1271) construct was generated by PCR amplification of a plasmid (obtained from American Type Culture Collection) that contained IL-1 cDNA cloned from human peripheral blood cells that had been stimulated with bacterial lipopolysaccharide. The upstream primer (CGGGATATGGCCAAAGTTCCAGAC) contained a BamH I restriction site, and Kozak (20) consensus sequence (underlined) immediately upstream from the ATG start site. The downstream primer (CGGGATCCCTACGCCTGGTTTTCCAG) added a BamH I site immediately downstream from the stop codon. The PCR product was purified (Qiagen), ligated directly into pCRII (TA cloning kit; Invitrogen), then excised with BamH I and inserted into pcDNA3 (Invitrogen). The correct orientation of IL-1-(1271) cDNA was confirmed by restriction endonuclease analysis.

Western blot analysis. Cells were lysed in Tris-buffered saline (TBS) by three cycles of freezing and thawing, boiled for 3 min in loading buffer containing 2% -mercaptoethanol, separated by electrophoresis through a SDS-12% polyacrylamide gel, and transferred to nitrocellulose membranes (Schleicher & Schuell) using an electrotransfer apparatus (Trans blot cell; Bio-Rad). The membranes were blocked in TBS solution containing 5% nonfat dry milk and 0.05% Tween 20 and incubated sequentially with rabbit antiserum against human recombinant IL-1, biotinylated goat anti-rabbit IgG, and streptavidin-biotinylated alkaline phosphatase complex. Blots were developed with nitro blue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate (Bio-Rad).

Localization of IL-1 in transfected VSMC. A7r5 cells were plated on glass coverslips placed in 6-cm dishes and transfected by calcium chloride precipitation (38) with 8 µg/plate of pcDNA alone, IL-1-(1271)/pcDNA, or IL-1-(113271)/pcDNA. Cells were incubated overnight, media were replaced with fresh DMEM supplemented with 20% FCS, and cells were incubated an additional 24 h. Cells were then fixed in 3.7% formaldehyde for 45 min and permeabilized in 0.2% Triton X-100 for 5 min, and nonspecific binding was blocked by incubating 45 min in PBS containing 1% normal goat serum and 0.5% BSA. Cells were then incubated sequentially in rabbit polyclonal anti-human recombinant IL-1 antisera (1:600) for 1 h, PBS/0.5% BSA for 1 h, goat anti-rabbit IgG FITC conjugate (Vector) for 1 h, and PBS/0.5% BSA for 20 min and were observed by epifluoresence microscopy.

Stable transfection of HSVSMC. HSVSMC (passages 1-3) were transfected with pcDNA3 alone, IL-1-(1271)/pcDNA3, IL-1-(113271)/pcDNA3, or IL-1-(1112)/pcDNA3 by electroporation with 100 µg of plasmid at 230 V and 960 µF. After electroporation, cells were incubated overnight in media containing 5 mM sodium butyrate, then washed thoroughly to remove traces of butyrate and grown in normal growth media for 3 days. VSMC expressing the neomycin resistance gene in pcDNA3 were then selected by growing the cell populations for 4 wk in growth media supplemented with neomycin (200 mg/ml geneticin, Sigma, St. Louis, MO). This concentration of neomycin was the lowest concentration that killed all nontransfected cells.

Proliferation rates. HSVSMC were plated in 24-well plates at a density of 5,000 cells/cm², in complete growth media containing 10% FCS. The medium was replaced every 3-4 days. Quadruplicate wells of each cell line (nontransfected and transfected) were trypsinized at regular intervals beginning the day after plating (day 0), and cell number was determined using a Coulter counter. Population doubling times (PDT) were calculated using the following formula: (T₂  T₁)/(log₂ N₂  log₂ N₁), where T is time and N is cell counts.

Bromodeoxyuridine incorporation. HSVSMC were plated 5,000 cells/well into 96-well plates in complete growth media. After 3 days, the medium was replaced with DMEM supplemented with 5-bromo-2'-deoxyuridine (BrdU; Boehringer Mannheim) and either insulin (1 µM) and transferrin (5 µg/ml) or FCS. The cells were fixed after 24 h and were incubated sequentially with mouse monoclonal BrdU antibody conjugated to peroxidase (Boehringer Mannheim) and tetramethylbenzidine as substrate. BrdU incorporation was measured as the absorbance at 405 nm. To test the role of IL-1 receptor type I, we determined the ability of exogenous IL-1 receptor antagonist (IL-1RA) to reverse chronic IL-1-induced proliferation. IL-1RA was added to complete media containing 10% FCS at the time of plating the cells and was added again when the medium was changed after 3 days to BrdU-containing media.

IL-1 ELISA. Cell lysates were prepared by three cycles of freeze-thawing cells 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 was determined using a specific enzyme immunoassay that detects both IL-1 precursor and mature IL-1, at a detection limit of 1-2 pg/ml (Cayman Chemical, Ann Arbor, MI).

RT-PCR. Total RNA was isolated using RNAzol B (Biotecx Laboratories, Houston, TX). RNA, 2 µg, was reverse-transcribed for 1 h at 37°C with 200 U MuMLV reverse transcriptase (GIBCO), using an oligo(dT) primer and the buffer supplied by the manufacturer, in a total volume of 20 µl. The reaction was terminated by heating to 95°C for 5 min, and 2 µl of this first-strand cDNA were added to each PCR reaction. PCR mixes contained 50 mM KCl, 10 mM Tris · HCl (pH 8.3), 1.5 mM MgCl₂, 0.01 mg/ml gelatin, 200 µM of each dNTP, 250 nM of each primer, and 1 U Taq DNA polymerase (GIBCO) in a total volume of 20 µl. The PCR primers used (sense: TCTCTGAATCAGAAATCCTTC and antisense: GTCAAATTTCACTGCTTCATC) amplify native IL-1-(1271) mRNA (nucleotides 121-537) as well as corresponding segments of mRNA derived from IL-1-(1271) or IL-1-(1112) expression plasmids but do not amplify mRNA derived from the IL-1-(113271) expression plasmid. Amplification was performed as previously described (4), with primers annealed for 2 min at 53°C. Products were size separated by electrophoresis in a 5% polyacrylamide gel and visualized by ethidium bromide staining.

Reagents. Human recombinant IL-1 (amino acids 117-269 of the IL-1 precursor protein) was a gift of Dr. Richard Dondero, Cistron Biotechnology (Pine Brook, NJ), and had a specific activity of 10 U/ng (based on a murine thymocyte costimulation assay). Human recombinant IL-1RA, human recombinant IL-1, and rabbit antisera to human IL-1 were a gift of Dr. Charles Dinarello, University of Colorado (Denver, CO).

	RESULTS

Top Abstract Introduction Materials and methods Results Discussion References

Chronic exposure to exogenous IL-1 induces marked HSVSMC proliferation. Chronic exposure to human recombinant IL-1 induced marked proliferation of HSVSMC. HSVSMC grown in media supplemented with 5% FCS alone proliferated slowly compared with HSVSMC grown in media supplemented with 5% FCS and IL-1. In three experiments with HSVSMC derived from different patients, the average PDT of HSVSMC incubated 15-22 days with 5% FCS alone was 9.6 ± 1.0 days, whereas the average PDT of HSVSMC incubated with 5% FCS and IL-1 (2 ng/ml) was 4.8 ± 0.5 days. IL-1 was an equally effective mitogen in all three HSVSMC lines, shortening PDT by ~0.5 in all three experiments. A representative experiment with HSVSMC studied at passage 3 is shown in Fig. 1. Similar results were obtained with passage 5 and 7 HSVSMC (data not shown). Exogenous IL-1 enhanced HSVSMC proliferation to a comparable degree at all serum concentrations tested. In the experiment shown in Fig. 2, IL-1 enhanced proliferation rates ~3.5-fold in the presence of either 1, 2, 5, 10, or 20% FCS.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Chronic exposure to exogenous interleukin-1 (IL-1) induces proliferation of human saphenous vein vascular smooth muscle cells (HSVSMC). HSVSMC were plated at 4,000 cells/cm2; after 24 h, media were replaced with DMEM supplemented with 5% FCS, with or without IL-1 (2 ng/ml). Cells were counted at 3-day intervals with a Coulter counter. Population doubling times during days 3-9 of IL-1 exposure were 13.0 and 3.0 days in control (Con) and IL-1-treated cells, respectively.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Exogenous IL-1 enhances proliferation at low and high serum concentrations. HSVSMC were plated at 5,000 cells/cm2; after 24 h, media were replaced with DMEM supplemented with varying concentrations of FCS, with or without IL-1 (1 ng/ml). Cells were counted by Coulter counter on days 0 and 13. *P < 0.01 compared with HSVSMC incubated without IL-1 and with same concentration of FCS.

Western blot analysis of IL-1 produced by transfected VSMC. Immunoreactive IL-1 of the expected size was produced following transient transfection of VSMC with IL-1 expression plasmids (Fig. 3). A7r5 cells transfected with vector alone contained proteins that cross-reacted with the IL-1 antiserum. The identify of these proteins, detected at ~35-40 kDa, is not known. A7r5 cells transfected with human IL-1-(1271) expression plasmids contained an immunoreactive band that migrated consistent with a molecular mass of ~31 kDa. In contrast, A7r5 cells transfected with IL-1-(113271) expression plasmids contained an immunoreactive IL-1 band consistent with the expected size of mature IL-1 (17 kDa).

View larger version (61K): [in this window] [in a new window]   Fig. 3.   Vascular smooth muscle cells transfected with IL-1 expression plasmids produce immunoreactive IL-1 of appropriate size. A7r5 cells were transfected with pcDNA3 alone (lane 1), IL-1-(1271)/pcDNA3 (lane 2), or IL-1-(113271)/pcDNA3 (lane 3) expression plasmids by electroporation. After 48 h, whole cell lysates were prepared, separated on a SDS-12% polyacrylamide gel, and immunoblotted with antisera to IL-1. Molecular mass markers are indicated on left.

Localization of IL-1 in VSMC transfected with IL-1 expression plasmids. IL-1 preferentially localized to the nucleus in A7r5 cells transiently transfected with IL-1-(1271) expression plasmids (Fig. 4, A and B). In contrast, immunoreactive IL-1 was distributed throughout the cytosol and nucleus (Fig. 4, C and D) in A7r5 cells transiently transfected with IL-1-(113271) expression plasmids. Approximately 5% of cells stained positively for immunoreactive IL-1 (data not shown). Specific nuclear localization of immunoreactive IL-1 was also apparent in HSVSMC that had been stably transfected with IL-1-(1271) expression plasmids (Fig. 4, E and F).

View larger version (119K): [in this window] [in a new window]   Fig. 4.   Cytosolic and nuclear localization of immunoreactive IL-1 in HSVSMC transfected with either IL-1-(1271) or IL-1-(113271) expression plasmids. A7r5 cells were transfected with pcDNA3 (data not shown), IL-1-(1271)/pcDNA3 (A and B), or IL-1-(113271)/pcDNA3 (C and D) expression plasmids by calcium phosphate coprecipitation. HSVSMC stably transfected with IL-1-(1271)/pcDNA3 expression plasmids were plated for immunostaining (E and F). After 48 h, cells were stained by indirect immunofluoresence with IL-1-specific antisera and photographed at ×400 magnification.

HSVSMC transfected with either IL-1-(1271) or IL-1-(113271) expression plasmids proliferate rapidly. Transfection with either IL-1-(1271) or IL-1-(113271) expression plasmids markedly enhanced the proliferative rate of HSVSMC. The results of all experiments are summarized in Table 1; growth curves of two representative experiments are shown in Fig. 5. Nontransfected HSVSMC proliferated slowly in the presence of 10% FCS, with variability between lines derived from different patients. The average PDT of the nontransfected group was 7.0 ± 1.4 days (n = 3). The average PDT of pcDNA3-transfected HSVSMC was higher, 13.5 ± 4.6 days (n = 4), since two of three pcDNA3-transfected cell lines proliferated slower than the corresponding nontransfected cell line. This may be due to the fact that the process of selecting stable transfectants generated cells that had undergone a higher number of cumulative population doublings, bringing pcDNA3-transfected cell lines closer to cellular senescence. In all four HSVSMC cell lines, cells transfected with IL-1-(1271) expression plasmids proliferated faster than HSVSMC transfected with vector alone and also faster than nontransfected HSVSMC. PDT were consistently low in HSVSMC stably transfected with IL-1-(1271) expression plasmids (3.6 ± 0.5 days). Transfection with IL-1-(113271) expression plasmids was almost as effective; PDT were consistently shortened to 4.1 ± 0.6 days. The effect of IL-1 expression was most marked in HSVSMC cell lines in which pcDNA3-transfected cells proliferated very slowly. For example, among pcDNA3-transfected cell lines, HSVSMC line 1 had the slowest proliferation rate and responded the most to transfection with either IL-1-(1271) or IL-1-(113271) (Fig. 5A). In comparison, HSVSMC line 4 had a relatively high proliferation rate after stable transfection with pcDNA3. Proliferation was enhanced by stable transfection with IL-1-(1271) expression plasmids, but the magnitude of the effect was less (Fig. 5B). Similar results were obtained in a line of VSMC derived from a human pulmonary artery (data not shown).

View this table: [in this window] [in a new window]   Table 1.   Population doubling time, final cell density, and cell-associated and supernatant IL-1

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Autocrine production of IL-1-(1271) or IL-1-(113271) promotes proliferation of HSVSMC. HSVSMC from different patients (A, patient 1; B, patient 4) were transfected by electroporation with pcDNA3 alone, IL-1-(1271)/pcDNA3, IL-1-(113271)/pcDNA3, or IL-1-(1112)/pcDNA3 expression plasmids and stable transfectants selected in neomycin for 4 wk. NT, nontransfected. Cells were plated (5,000 cells/cm2) in DMEM supplemented with 10% FCS, and cell counts were determined at times shown.

View larger version (38K): [in this window] [in a new window]   Fig. 6.   DNA synthesis is increased in HSVSMC stably transfected with IL-1-(1271) or IL-1-(113271) expression plasmids; exogenous IL-1 receptor antagonist (IL-1RA) partially reverses DNA synthesis. HSVSMC were plated into wells with or without IL-1RA at various concentrations. After 72 h, media were changed to DMEM supplemented with insulin, transferrin, 0.5% FCS, and 5-bromo-2'-deoxyuridine (BrdU), with or without IL-1RA. A: HSVSMC stably transfected with pcDNA3 alone, IL-1-(1271)/pcDNA3 (precursor), IL-1-(113271)/pcDNA3 (mature), or IL-1-(1112) expression plasmids. B: nontreated (NT) HSVSMC or HSVSMC treated chronically with IL-1 (1 ng/ml; added fresh at each media change). C: nontreated HSVSMC or HSVSMC stimulated acutely with IL-1 (1 ng/ml) 1 h after plating and again when media were changed to BrdU-containing media. BrdU incorporation is expressed relative to value for either pcDNA3 stable transfectants (A) or nontreated HSVSMC (B and C) incubated without IL-1RA.

HSVSMC transfected with either IL-1-(1271) or IL-1-(113271) expression plasmids demonstrate a distinct morphology. HSVSMC grown in media containing IL-1 exhibited a distinct morphology distinguished by long spindle-shaped cells that assembled in multiple layers of densely packed cells, and achieved very high final cell densities (Fig. 7F), as previously described (23). The pattern of growth was characterized by the formation of whorls, which were particularly striking when cells were observed at their final cell density. HSVSMC transfected with either IL-1-(1271) or IL-1-(113271) expression plasmids demonstrated a similar morphology of elongated cells (Fig. 7, C and D, respectively), which was indistinguishable from the morphology of HSVSMC treated chronically with exogenous IL-1. In contrast, nontransfected HSVSMC (Fig. 7A), pcDNA3-transfected HSVSMC (Fig. 7B), and HSVSMC transfected with IL-1-(1112) (Fig. 7E) were morphologically heterogeneous, with cultures consisting mostly of cells which spread out on the substratum and displayed asymmetric, irregular shapes.

View larger version (228K): [in this window] [in a new window]   Fig. 7.   Phase-contrast photomicrographs of HSVSMC. HSVSMC were plated at 15,000 cells/cm2 into 24-well plates and photographed at ×200 magnification after 10 days growth. Nontransfected HSVSMC (A) and HSVSMC stably transfected with pcDNA3 alone (B) demonstrated polyglonal morphology and low densities (25,450 ± 550 and 18,025 ± 330 cells/cm2, respectively), as did HSVSMC stably transfected with IL-1-(1112)/pcDNA3 (E). HSVSMC stably transfected with IL-1-(1271)/pcDNA3 (C) or IL-1-(113271)/pcDNA3 (D) demonstrated elongated morphology and dense cultures (177,500 ± 2,350 and 78,300 ± 2,150 cells/cm2, respectively), as did HSVSMC cultured in media supplemented with IL-1 (1 ng/ml) (F).

Cell-associated and supernatant levels of IL-1. IL-1 was not detectable in any of the lysates or cell supernatants of either nontransfected HSVSMC or HSVSMC that had been stably transfected with pcDNA3. In contrast, HSVSMC transfected with either IL-1-(1271) or IL-1-(113271) expression plasmids contained elevated levels of cell-associated IL-1, measured in freeze-thaw extracts of the cells (Table 1). IL-1 levels determined on day 7 of the growth experiments are shown in Table 1; similar values were obtained on day 14. The measured level of cell-associated IL-1 was variable between different lines of stable transfectants, and the increment in proliferation rate did not correlate with the measured level of cell-associated IL-1. Immunoreactive IL-1 was also detectable in the supernatants of IL-1-(1271)- or IL-1-(113271)-expressing cells, ranging between 2 and 17 pg/ml. The levels of IL-1 measured in cell supernatants also did not correlate with the magnitude of the proliferative response. HSVSMC stably transfected with the IL-1-(1112) expression plasmid, like pcDNA3-transfected HSVSMC, did not contain detectable immunoreactive IL-1 in cell lysates or supernatants.

Low levels of exogenous IL-1 induce HSVSMC proliferation. Concentration-response experiments revealed that IL-1 and IL-1 are potent mitogens for human VSMC when added exogenously to the media for 1 wk. In two experiments, one with HSVSMC and the other with smooth muscle cells derived from human pulmonary artery (HPASMC), the pro-proliferative effects of human recombinant IL-1 and IL-1 were near-maximal at concentrations of 2 ng/ml. IL-1 was also more potent than IL-1 in both experiments. In HPASMC, half-maximal proliferative responses were obtained with 53 pg/ml IL-1 or 220 pg/ml IL-1 (Fig. 8). In a similar experiment with HSVSMC, half-maximal proliferative responses were obtained with 36 pg/ml IL-1 or 72 pg/ml IL-1 (data not shown). Both IL-1 and IL-1 were ineffective at 2 pg/ml.

View larger version (19K): [in this window] [in a new window]   Fig. 8.   Concentration dependence of exogenous IL-1-induced proliferation. Human pulmonary artery smooth muscle cells were plated at 5,000 cells/cm2 in DMEM supplemented with 10% FCS and varying concentrations of IL-1 or IL-1. Cells were counted by Coulter counter on days 0 and 7. Population doubling times were as follows: IL-1: 10.38, 9.53, 5.81, 3.54, 3.00, and 3.07 days; IL-1: 8.69, 8.61, 9.19, 4.33, 3.01, and 2.88 days for 0, 2, 20, 200, 2,000, and 20,000 pg/ml, respectively.

HSVSMC stably transfected with either IL-1-(1271) or IL-1-(113271) expression plasmids express IL-1-(1271). RT-PCR analysis was performed using PCR primers that amplify IL-1-(1271) mRNA [either native or transcribed from stably integrated IL-1-(1271) or IL-1-(1112) expression plasmids] but do not amplify mRNA transcribed from stably integrated IL-1-(113271) expression plasmid. IL-1-(1271) mRNA was not present in nontransfected HSVSMC, HSVSMC stably transfected with pcDNA3, or HSVSMC treated chronically with IL-1 (Fig. 9). In contrast, RT-PCR analysis of cDNA samples from HSVSMC transfected with IL-1-(1271) expression plasmids produced a band of the expected size (417 bp), documenting that the cells contained elevated levels of IL-1-(1271) mRNA. Surprisingly, HSVSMC transfected with IL-1-(113271) expression plasmids also contained elevated levels of IL-1-(1271) mRNA that was not vector derived. As expected, HSVSMC transfected with IL-1-(1112) expression plasmids contained elevated levels of IL-1-(1271) mRNA, documenting that the stably integrated expression plasmid was expressed at the mRNA level.

View larger version (42K): [in this window] [in a new window]   Fig. 9.   HSVSMC transfected with IL-1 expression plasmids express IL-1 precursor mRNA. Total RNA (2 µg) was prepared from HSVSMC stably transfected with pcDNA3, IL-1-(1271), IL-1-(113271), or IL-1-(1112) expression plasmids (lanes 2-5, respectively), nontreated (NT) HSVSMC (lane 6), or chronic IL-1-treated HSVSMC (lane 7). cDNA was amplified for 30 cycles with primers specific for IL-1 precursor cDNA. Molecular size markers are shown in lane 1 (Hae III-digested ØX-174: 1353, 1078, 872, 603, 310, 271-281 doublet, 234, and 194 bp). IL-1 precursor PCR products run between 603- and 310-bp fragments.

Exogenous IL-1RA partially reverses proliferation induced by endogenous or exogenous IL-1. Human recombinant IL-1RA, when added exogenously to the media of HSVSMC stably transfected with IL-1 expression plasmids, partially reversed the proliferative action of autocrine IL-1 production (Fig. 6A). High concentrations of IL-1RA (0.1-10 µg/ml) were required to reverse the proliferative effect of IL-1 in stable transfectants. Also, only partial reversal was achieved even though IL-1RA was added at the time of plating the cells, was preincubated with the cells for 72 h, and was readded when the media was changed to BrdU-containing media. The inhibition that was achieved with IL-1RA was significantly greater (P < 0.01) in IL-1-(113271) stable transfectants (26, 34, and 36% inhibition of the proliferative response at 0.1, 1, and 10 µg/ml IL-1RA, respectively) than in IL-1-(1271) stable transfectants (13, 15, and 19% inhibition at 0.1, 1, and 10 µg/ml IL-1RA, respectively). Similar results were obtained in a replicate experiment with stable transfectants derived from a different patient (data not shown). Because IL-1RA was used to reverse rather than prevent the proliferative action of IL-1 in stable transfectants, we compared the ability of IL-1RA to reverse proliferation induced by chronic treatment with exogenous IL-1 with its ability to prevent the proliferative response to acute exposure to exogenous IL-1. IL-1RA prevented induction of proliferation by a 3-day exposure to exogenous IL-1 (41, 71, and 95% inhibition of the proliferative response at 0.1, 1, and 10 µg/ml IL-1RA, respectively) (Fig. 6C). In contrast, a 3-day preincubation with IL-1RA did not readily reverse the proliferation of HSVSMC treated chronically with IL-1 (12, 17, and 18% inhibition of the proliferative response at 0.1, 1, and 10 µg/ml IL-1RA, respectively) (Fig. 6B).

	DISCUSSION

Top Abstract Introduction Materials and methods Results Discussion References

The present studies document that IL-1 is a potent growth factor for human VSMC, when either added exogenously to the culture media or produced endogenously by the cells. The pro-proliferative effect of IL-1 was highly significant after 1 wk of addition to the culture media and occurred at concentrations as low as 20 pg/ml. IL-1 acted synergistically with serum, enhancing proliferation in the presence of either low (1%) or high (20%) serum concentrations. HSVSMC treated chronically with IL-1 achieved very high final cell densities, accumulating in multiple cell layers as they continued to proliferate. These results suggest that IL-1-treated cells have markedly reduced constraint by density-dependent inhibition of cell division.

The mitogenic effect of chronic exposure to exogenous IL-1 was mimicked by stable transfection with IL-1-(1271) expression plasmids. HSVSMC producing IL-1-(1271) under the control of the constitutively active human cytomegalovirus promoter proliferated rapidly compared with HSVSMC transfected with vector alone; cells doubled in <4 days compared with 13 days in pcDNA3-transfected cells. IL-1-transfected HSVSMC also reached final cell densities, as did HSVSMC treated with exogenous IL-1, and proliferated faster in the presence of either low or high serum. The amount of IL-1 produced by HSVSMC stably transfected with IL-1-(1271) expression plasmids was variable between cell lines derived from different patients, ranging between 0.2 and 2 ng/10⁶ cells. In comparison, HSVSMC that have been stimulated 24 h with IL-1 (4) or exposed to hypoxia for 48 h (Cooper and Beasley, unpublished data) produce 0.5-3 ng/10⁶ cells. Thus these studies indicate that IL-1 is an autocrine growth factor for HSVSMC when produced at levels similar to, or even lower than, the levels produced endogenously after activation with pathophysiologically relevant stimuli.

Preparing stable transfectants of HSVSMC required at least 1 mo for expansion of initial explant-derived cell populations to achieve sufficient numbers of cells for transfection and an additional 1 mo for selection of stable transfectants in neomycin. Thus, when proliferation rates were assessed, stable transfectants of HSVSMC were usually at passage 5, a relatively high passage for these cells. A striking finding in this study was that HSVSMC that were treated chronically with IL-1 beginning at a low passage continued to proliferate rapidly for several months, even after up to 20 passages. In addition, high-passage HSVSMC retained the ability to proliferate rapidly when exposed to exogenous IL-1 for the first time. We are not aware of any previous studies that employed stable transfection of primary cultures of human VSMC. The ability to obtain stable transfectants that displayed the IL-1 phenotype for sufficient time to allow studies was highly dependent on the fact that human VSMC retain the ability to proliferate in response to IL-1 for many passages in culture.

Stable transfection of HSVSMC with IL-1-(1112) expression plasmid had no discernible effect on HSVSMC. Neither proliferative rates nor morphology was altered. These results are in contrast to previous findings in which rat mesangial cells displayed a transformed phenotype following stable transfection with IL-1-(1112) expression plasmid (37). Our results suggest that the NH₂-terminal domain of IL-1 precursor is ineffective by itself. In contrast, the COOH terminal or mature portion of the IL-1 precursor molecule is required for the pro-proliferative action of IL-1 in human VSMC. These results may indicate that the activity of IL-1-(1112) is cell-type specific.

Evidence has been presented that nuclear localization is required for the action of IL-1-(1271) produced by transfected cell lines. In fibroblasts or endothelial cells transfected with the corresponding cDNA, IL-1-(1271) localizes to the nucleus, whereas IL-1-(113271), which lacks the nuclear localization sequence (NLS), remains cytosolic (25, 42). Our results likewise demonstrate that IL-1-(1271) localizes to the nucleus in both human VSMC and embryonic rat aortic smooth muscle cells. In contrast, IL-1-(113271) remains cytosolic. In two previous studies, a human endothelial cell line (EC) stably transfected with IL-1-(1271) expression plasmids had slower proliferation rates, higher levels of plasminogen activator inhibitor-1 and collagenase mRNA, and a lower migratory potential than did EC transfected with IL-1-(113271) expression plasmids or vector alone (25, 27). In addition, a single-base pair mutation in the NLS of IL-1-(1271) reversed its ability to inhibit human EC migratory potential (27). These results were interpreted as evidence that nuclear localization is important in the action of IL-1-(1271). However, IL-1-(1271) may also localize to the plasma membrane in a form that can stimulate adjacent cells via juxtacrine mechanisms, whereas IL-1-(113271) does not (5, 7, 13), and the ability of the NLS mutant to localize to the plasma membrane is unknown. Thus juxtacrine mechanisms may also account for the effectiveness of IL-1-(1271) over IL-1-(113271). In contrast to the ineffectiveness of IL-1-(113271) expression plasmids in human EC, HSVSMC transfected with IL-1-(113271) expression plasmids also proliferated rapidly; in some HSVSMC lines they proliferated as rapidly as HSVSMC transfected with IL-1-(1271) expression plasmids and in other HSVSMC lines somewhat slower. These results appear to suggest that nuclear localization and association with the plasma membrane are not crucial components of IL-1 action in human VSMC. Surprisingly, however, HSVSMC stably transfected with expression plasmids encoding IL-1-(113271) also expressed mRNA encoding IL-1-(1271). These results indicate that some of the IL-1 produced by IL-1-(113271) stable transfectants was not vector derived, but rather endogenous IL-1-(1271), produced upon activation of an autocrine loop whereby IL-1 induces its own production (40). The effectiveness of stable transfection with IL-1-(113271) expression plasmids suggests that even low levels of IL-1 released from cells can act, ultimately, as an autocrine growth factor for human VSMC. However, it is not clear whether upregulation of additional native IL-1-(1271) production is crucial to the ultimate pro-proliferative action of transfected IL-1-(113271). Thus these studies do not rule out the possibility that the NH₂-terminal domain of IL-1-(1271), which contains a nuclear localization sequence (42) as well as sequences required for plasma membrane association (13), is also important in the proliferative action of IL-1-(1271).

To assess whether stably produced IL-1-(1271) exerts its pro-proliferative effect by activation of cell surface IL-1 receptors, stable transfectants were incubated with IL-1RA, a specific inhibitor of IL-1 action, which binds to, but does not activate, the type I IL-1 receptor (26). Incubation with high concentrations of IL-1RA for >72 h caused a partial reversal of the enhanced proliferation of stable transfectants expressing IL-1-(1271). These results indicate that the action of VSMC-derived IL-1-(1271) is at least partially mediated by signaling via the type I IL-1 receptor. Because high concentrations of IL-1RA have been documented to inhibit the action of both soluble and membrane-bound IL-1 (18), the results also implicate either released or membrane-associated IL-1 in the proliferative effect of stable transfection with IL-1-(1271) expression plasmids. It is possible that the component of the proliferative response that is not reversed by exogenous IL-1RA represents actions of IL-1-(1271) that occur within the cell. However, our results support an alternative hypothesis. IL-1RA appears to be less effective when used to reverse the effect of IL-1 that is produced chronically by stable transfectants, compared with when it is used to block the initial action of IL-1. Although numerous studies have documented that IL-1RA added before or with IL-1 prevents the ability of IL-1 to activate cells (26), we are not aware of any studies that document the ability of IL-1RA to reverse IL-1 action. In the present study, preincubation with high concentrations of IL-1RA completely blocked induction of proliferation by acute exposure to IL-1, documenting that the proliferative action of IL-1 in HSVSMC involves the type I IL-1 receptor and that the IL-1RA preparation used was biologically active. In contrast, incubation with exogenous IL-1RA for >72 h only partially reversed the enhanced proliferation rate in HSVSMC that were exposed chronically to exogenous IL-1. IL-1RA was likewise only partially effective in HSVSMC stably transfected with IL-1-(113271) expression plasmids. Because IL-1-(113271) lacks the NLS found in IL-1-(1271), and does not localize to the nucleus (25, 42) or associate with the plasma membrane (5, 7, 13), its effects are mediated by release from the cell and activation of type I IL-1 receptors, and thus, like the effect of exogenous IL-1, should be completely inhibitable by IL-1RA. Therefore, we conclude that the ineffectiveness of IL-1RA appears to relate to the fact that IL-1 action is not readily reversed.

The present findings have important implications for the design of studies to investigate the role of IL-1 type I receptor using IL-1RA, particularly those studies designed to reverse IL-1 action in cells that are already producing the cytokine. Such studies should employ sufficiently long incubations with the antagonist. The fact that exogenous IL-1RA did not reverse the effect of stable transfection with IL-1-(1271) expression plasmids in previous studies with human EC (25, 27) has been presented as evidence that IL-1-(1271) has intracellular actions. However, these studies employed a relatively short incubation with IL-1RA (24 h); therefore, the lack of effect of IL-1RA may be due to its inability to rapidly reverse IL-1 action. Our results may also have important clinical ramifications. It is possible that the ineffectiveness of IL-1RA in recent clinical trials for septic shock (12) is due to the prolonged action of IL-1 and the fact that IL-1RA does not readily reverse IL-1 action.

The magnitude of the pro-proliferative effect of IL-1 did not correlate with the level of IL-1 measured in cell lysates. It is possible that IL-1 present in freeze-thaw lysates of the cells does not represent the active pool. As discussed above, the fact that the proliferative effect of IL-1 in stable transfectants was partially reversed by exogenous IL-1RA argues that IL-1 is active, at least in part, extracellularly, either via a membrane-associated or released form that activates the type I IL-1 receptor on the cell surface. The active pool of IL-1 may be membrane IL-1 (24) that is sequestered on the extracellular surface of the plasma membrane and activates adjacent cells via juxtacrine mechanisms. Alternatively, IL-1 released from the cell may be the active pool. Although the levels of IL-1 measured in cell supernatants were low, in the range of 7-17 pg/ml, concentration-response studies indicate that IL-1 was highly potent when added exogenously to the media. Exposure to 20 pg/ml IL-1 for only 1 wk induced significant HSVSMC proliferation. It seems possible that exposure to IL-1 at levels somewhat less than 20 pg/ml for 4 or more weeks may also induce HSVSMC proliferation. Although the magnitude of the pro-proliferative effect of IL-1 also did not correlate with the level of IL-1 measured in the cell supernatants, HSVSMC derived from different patients may differ in their responsiveness to the pro-proliferative effect of autocrine IL-1. In this regard, the proliferative response to stable transfection with IL-1 expression plasmids was variable in HSVSMC cell lines derived from different patients, but correlated with the proliferative response to chronic exposure to exogenous IL-1 in the same cell line (D. Beasley, unpublished data).

HSVSMC transfected with IL-1-(1271) expression plasmids display a distinct morphology characterized by long spindle-shaped cells that grow in dense multiple layers and form whorls. It is interesting to note that the elongated form of IL-1-treated HSVSMC is the predominant form of cells obtained from primary cultures of enzyme-dispersed human aorta (30). Primary and low-passage cultures of HSVSMC cell lines established in our laboratory contain similar elongated cell types, whereas the percentage of elongated cells is diminished in high-passage cells. In contrast, nontransfected HSVSMC and HSVSMC transfected with vector alone display a higher proportion of polygonal and asymmetric shapes, which are also found in cells cultured by enzymatic dispersal of human aorta (30) and which are more typical of high-passage HSVSMC. It is tempting to speculate that the elongated shape characteristic of a subset of human VSMC may represent VSMC that are producing IL-1.

In vitro studies have documented that IL-1 can have variable effects on proliferation depending on the type of VSMC tested and the length of exposure to IL-1. In the present study, exposure to exogenous IL-1 or IL-1 for 72 or more hours induced proliferation of subcultured human VSMC derived by explant technique from either pulmonary artery or saphenous vein. IL-1 is also pro-proliferative in primary cultures of rat aortic smooth muscle cells (6). However, IL-1 can also induce growth-inhibitory pathways. For example, short-term incubations with IL-1 (i.e., 48 h) stimulate proliferation of human aortic, human saphenous, and rat aortic smooth muscle cells only in the presence of indomethacin, suggesting that growth-inhibitory prostanoids suppress the early mitogenic response to IL-1 (17, 21, 23, 32). In addition, both short-term (10, 14) and long-term (36) exposure to IL-1 can inhibit proliferation of subcultured rat aortic smooth muscle cells by inducing expression of inducible nitric oxide (NO) synthase and generation of growth inhibitory NO. Because IL-1 does not induce expression of inducible NO synthase in subcultured human VSMC (3), human VSMC are a more suitable model for studies of the growth-promoting actions of IL-1.

Several studies are consistent with the hypothesis that IL-1 is produced within intimal hyperplastic lesions. In atherosclerotic lesions of hypercholesterolemic monkeys, IL-1 mRNA was present in endothelial cells, intimal foam cells, and medial VSMC, but absent in uninvolved vascular tissue (29). IL-1 mRNA has also been demonstrated in a primary human atherosclerotic lesion by RT-PCR (39), although the cell type expressing IL-1 mRNA was not identified. Immunoreactive IL-1 was found in spindle-shaped cells and macrophages of sclerotic saphenous veins but was absent in normal saphenous veins and normal internal mammary arteries studied before coronary bypass surgery (8). In blood vessels obtained after coronary bypass surgery, IL-1 was found in saphenous vein bypass grafts that had become stenotic but was absent in internal mammary arteries that had remained patent (8). Together with the present study, which implicates a role for IL-1 as a potent autocrine growth factor for vascular smooth muscle cells, the above studies suggest that IL-1 may contribute to the development of intimal hyperplasia in primary atherosclerotic lesions and in saphenous vein bypass grafts.