Differential Expression of α1 Type VIII Collagen in Injured Platelet-Derived Growth Factor-BB–Stimulated Rat Carotid Arteries
Migration of smooth muscle cells from media to intima is critical for the development of neointimal thickening after balloon catheter injury of the rat carotid artery. The present experiments were designed to identify molecules expressed by smooth muscle cells migrating in vivo in the injured artery. Cell migration was maximized by infusing recombinant platelet-derived growth factor-BB (PDGF-BB) after a minimal filament denudation of the rat carotid artery, whereas cell proliferation was minimized by injecting an antibody against basic fibroblast growth factor (bFGF). This treatment caused an eightfold increase in smooth muscle cell migration into the intima but only a twofold increase in intimal smooth muscle cell replication rates. Differential display screening was used to isolate cDNAs that were overexpressed in the injured PDGF-BB–treated versus unmanipulated rat carotids. One of the clones isolated hybridized to a 4.2-kb mRNA species and shared 90% sequence homology to mouse α1 type VIII collagen. Northern and Western blots confirmed overexpression of type VIII collagen in the injured PDGF-BB–treated vessels. In a separate series of experiments, we performed filament denudation injury and administered antibodies to inhibit the actions of endogenous bFGF and PDGF-BB, thereby decreasing smooth muscle cell migration, and found that type VIII collagen mRNA expression varied with migration. Using a different arterial injury model (balloon catheter injury), we showed that expression of type VIII collagen was maximal 2 to 4 days after injury, in coincidence with cell migration from the media to the intima. This molecule constitutes an important component of smooth muscle cell response to vessel injury and may play an important functional role in mediating migration.
Migration of SMCs from media to intima is critical for the development of neointimal thickening after balloon catheter injury of the rat carotid artery. Once in the intima, SMCs proliferate and synthesize a new matrix, resulting in a dramatically thickened intima within a few weeks after injury. The kinetics of SMC proliferation have been characterized in this model,1 but much less is known about SMC migration. We have developed assays to quantify SMC migration by virtue of the first appearance of cells in the intima at 4 days after injury2 3 ; however, this analysis is limited to the single time point, and it is likely that SMC migration within the media and intima precedes and follows this time. The present experiments were conceived in an attempt to identify molecules expressed in vivo by SMCs migrating in the injured artery. To achieve this goal, we studied gene expression in injured rat carotid arteries where SMC migration was stimulated by PDGF BB-chain infusion, since recent evidence from our laboratory and others has suggested that PDGF-BB is an important regulator of SMC migration. Infusion of PDGF-BB caused a 20-fold increase in intimal area and the number of intimal SMCs after arterial injury but had minimal effects on SMC proliferation (no increase in intimal cell replication rates and only a 2- to 3-fold increase in medial cell replication rates).4 Conversely, neointimal thickening was dramatically reduced in platelet-deprived rats5 and in rats treated with anti–PDGF-BB antibodies after ballooning,2 6 but neither treatment affected SMC replication, and SMC migration (as measured by scanning electron microscopy of SMCs in the intima) was decreased by treatment with anti–PDGF-BB antibodies.2
In the present experiments, cell migration was maximized by infusing recombinant PDGF-BB after a minimal filament denudation of the rat carotid artery. Cell proliferation was minimized by injecting an antibody against bFGF immediately before surgery.7 Differential display screening was used to isolate cDNAs that were overexpressed in the injured, PDGF-BB–stimulated, anti-bFGF–treated rat carotid arteries (migrating) compared with unmanipulated rat carotid arteries (stationary). One of the clones isolated encoded rat α1 type VIII collagen mRNA, and Northern and Western blots showed overexpression of mRNA and protein for this molecule in the migrating group. Further experiments showed that type VIII collagen mRNA expression varied with SMC migration in the vessel; when injured vessels were treated with anti–growth factor antibodies to inhibit migration, mRNA expression decreased. Thus, our evidence suggested that type VIII collagen was overexpressed after injury and may play an important role in mediating SMC response to injury.
Materials and Methods
Surgery and SMC Migration and Replication
Male Sprague-Dawley rats (3 to 4 months old) from Bantin and Kingman Laboratories (Edmonds, Wash) were used in all experiments. Immediately before surgery, rats were given a single injection of a goat anti-rat bFGF antibody (60 mg per rat via the tail vein). Rats were anesthetized by intraperitoneal injection of xylazine (Anased, 4.6 mg/kg body wt, Lloyd Laboratories) and ketamine (Ketaset, 70 mg/kg body wt, Aveco Co Inc). A midline incision in the neck was made to expose the left and right external carotid arteries, and the carotid arteries were injured by gentle filament denudation with a loop of 5/0 nylon monofilament suture (Dermalon, Davis and Geck) introduced into the external carotid artery via a trocar made of polyethylene tubing.8 Both carotid arteries were injured in each rat. A catheter (inner diameter, 0.025 in; outer diameter, 0.047 in; Silastic Laboratory Tubing, Dow Corning Corp) was attached to an Alzet 2001 osmotic mini-pump (Alza Corp) containing 5 mg/mL rPDGF-BB (Zymogenetics Inc) dissolved in 10 mmol/L acetic acid in Ringer's USP solution (Baxter). The catheter was placed in the left external jugular vein and advanced until the tip reached the cranial vena cava. The osmotic pumps were tunneled under the skin and placed on the rat's back between the shoulder blades. The pumps delivered PDGF-BB at a rate of 0.3 mg/kg per day for 4 days. To label all cells entering S phase during the last 24 hours before death, three injections of BrdU (Boehringer Mannheim Corp) were given subcutaneously to all rats: 25 mg/kg body weight at 17, 9, and 1 hour before they were killed at 4 days after injury by intravenous injection of sodium pentobarbitol (Anthony Products Co). Lactated Ringer's injection USP (Baxter) was infused at a pressure of 110 mm Hg retrogradely via a catheter placed in the abdominal aorta, followed by perfusion with 0.1 mol/L phosphate-buffered 4% paraformaldehyde. A 1-cm length was excised from the middle of the common carotid artery and used for the SMC migration assay described below. Adjacent sections 5 mm in length were cut and embedded in paraffin. Cross sections were cut and immunostained for BrdU as previously described.9 SMC migration into the intima was quantified as previously described.3 Briefly, intimal cells on the surface of fixed common carotid artery segments were immunostained with an antibody against histone H1. The number of intimal cell nuclei per square millimeter of surface area were counted by light microscopy at 4 days after injury, a time when SMCs first appear in the intima.2
RNA Isolation and Differential Display PCR
To obtain the tissues used in differential display analysis, left and right rat common carotid arteries were subjected to surgery and PDGF-BB/anti-bFGF treatment as described above, except PDGF-BB infusion was carried out for 7 days after arterial injury, at which time the rats were killed (migrating group). The left and right common carotid arteries were excised, adhering adventitia and connective tissue were removed, and the vessels were snap-frozen in liquid nitrogen and then stored at −80°C. To obtain unmanipulated control vessels (stationary group), rats were killed and perfused with Ringer's solution as described above, the carotid arteries were excised, stripped of adventitia, and cut open longitudinally, and the endothelial cells were scraped from the surface using a piece of polytetrafluoroethylene (Teflon). These vessels were also snap-frozen and stored at −80°C.
All carotid arteries from the migrating group were pooled, as were the stationary group carotid arteries. Frozen arterial tissue was ground to a fine powder under liquid nitrogen, and total cellular RNA was prepared by acid thiocyanate extraction.10 For differential display, total RNA (100 ng) was denatured at 65°C for 15 minutes and reverse-transcribed at 37°C for 1 hour. Reverse-transcription reaction conditions were as follows: arbitrary primer 1 (20 nucleotides long, 1 μmol/L), MgCl2 (1.5 mmol/L), dNTP (8 μmol/L GTP, CTP, and TTP; 4 μmol/L ATP), dithiothreitol (2 mmol/L), RNAsin (20 U), and MMLV reverse transcriptase (200 U, BRL) in PCR buffer with a total reaction volume of 20 μL. After reverse transcription, the cDNA was PCR-amplified by adding the following ingredients to the mix: arbitrary 20-mer primer 2 (1 μmol/L), MgCl2 (1.5 mmol/L), Amplitaq (2.5 U), and [35S]dATP (12.5 μCi) in PCR buffer, bringing the reaction volume to 40 μL. PCR cycling was carried out as follows: once at 39°C for 5 minutes and 72°C for 5 minutes and then 35 cycles at 95°C for 30 seconds, 60°C for 30 seconds, and 72°C for 1.5 minutes, followed by cooling to 4°C. Samples from migrating and stationary RNA were amplified together and in duplicate, and the labeled cDNAs obtained from these reactions were run side by side and separated on a 6% acrylamide sequencing gel, which was subsequently dried and exposed to autoradiographic film (12 to 36 hours). Some differential display primers were based on random primer sequences made available by Operon and included restriction sites for Cla I (underlined) to facilitate subcloning. Sequences for these primers were as follows: BIH3, 5′-CTGCCAGGCCCTTC-3′; BIH4, 5′-CTGCTGCCGAGCTG-3′; BIH5, 5′-CTGCCTGCTGGGAC-3′; MBDDa, 5′-CTGCGTCCAGACGG-3′; and MBDDb, 5′-CTGCGGTGACGCAG-3′. We also used primers containing short degenerate consensus sequences to highly conserved regions of MMP enzymes and cell surface integrin receptors to direct our selection of primers toward detecting these molecules, which play important functional roles in mediating migration. Primer sequences used were as follows: MMP consensus primers: MBDD1, 5′-GCCA(T/C)TT(T/C)GA(T/C)GA-3′; MBDD2, 5′-GC(A/G)TC(A/G)TC(A/G)AA(A/G)TG-3′; MBDD3, 5′-GCGITG(T/G)GGIGTICCIGA-3′; and MBDD4, 5′-GCTTIGC(C/T)TT(A/G)TC-3′; integrin consensus primers: MBDD5, 5′-GCGGIGA(A/G)CAG(A/C)TIG-3′; MBDD6, 5′-GCA(C/T)IGCIA(C/T)(A/G)TC-3′; MBDD7, 5′-GCCCI(A/G)TIGA(C/T)(A/C/T)TI-3′; and MBDD8, 5′-GCICCICC(C/T)TCIGG-3′. Restriction sites for Xba I (underlined) were included to facilitate subcloning. Fig 1⇓ contains a flow chart outlining the number of primer pairs that were used in the differential display analysis (47 primer pairs), the total number of differentially expressed fragments excised, and further screening steps that were used to confirm differential expression.
Recovery, Reamplification, and Cloning of cDNAs
Selected cDNAs that showed increased expression in the migrating group were excised from the sequencing gel and eluted at 95°C for 10 minutes in 50 μL dH2O; 5 μL of this reaction was reamplified with MgCl2 (1.5 mmol/L), dNTP (50 μmol/L each GTP, CTP, TTP, and ATP), Amplitaq (2 U), and the same two primers that were used in the differential display reaction (0.5 μmol/L each) in a total reaction volume of 40 μL PCR buffer. PCR cycling was carried out as follows: 35 cycles at 92°C for 30 seconds, 60°C for 30 seconds, and 72°C for 1.5 minutes and a final elongation of 72°C for 5 minutes, followed by cooling to 4°C. The reamplified PCR products were visualized on 1.5% agarose gels stained with ethidium bromide and recovered by filtration (Spin-ex, 0.2 μm, Costar Corp). The products were then cloned into the Bluescript plasmid (PCRscript, Stratagene Inc) according to the manufacturer's directions. Plasmids were used to infect DH5a cells, and the inserts were colony-screened by PCR and screened by restriction mapping to assess the uniformity of the ligated inserts (Fig 1⇑).
Dot Blot and Northern Blot Screening and Sequencing of PCR Products
Plasmid preps of the cDNA clones were applied by dot blot to Zeta probe nylon membranes (Bio-Rad) (Fig 1⇑). Duplicate membranes were prepared and hybridized overnight with 32P-labeled cDNA derived by reverse transcription from the original migrating or stationary RNA. To prepare these probes, first-strand cDNA synthesis was carried out from 5 μg of total RNA using an oligo(dT)12-18 primer (Life Technologies) at a concentration of 1.25 μmol/L, 8 μmol/L dithiothreitol, 1 U/μL RNAsin, 100 nmol/L dNTP (ATP, CTP, GTP, and TTP), and Superscript reverse transcriptase (5 U/μL) in a volume of 20 μL. A radioactive probe was obtained by random primer extension of first-strand cDNA after alkaline denaturation and ethanol precipitation (Multiprime, Amersham). Blots were hybridized overnight with 2×106 cpm/mL of hybridization buffer (50% formamide, 0.75 mol/L NaCl, 50 mmol/L Tris [pH 7.4], 1× Denhardt's solution [Sigma], 1% SDS, 10% dextran sulfate, and 200 μg/mL denatured salmon sperm DNA [Sigma]). Autoradiographic analysis was carried out by the Phosphorimager Facility of the Markey Molecular Medicine Center at the University of Washington.
Clones that were differentially expressed on the dot blots were taken through a second screening on a Northern blot (Fig 1⇑). Northern blots were prepared by subjecting total RNA from migrating and stationary groups (15 μg per lane) to agarose gel electrophoresis and transfer to nylon membranes (Zeta probe, Bio-Rad). Northern blots were photographed under UV light to verify equal loading in each lane. They were hybridized using cDNA clones labeled with [32P]dCTP by random primer extension, using hybridization and wash conditions identical to those described for dot blots. Expression was visualized by autoradiographic film exposure (1 to 14 days) or phosphorimaging (1 to 5 days). Clones that were differentially expressed on both dot blots and Northern blots were sequenced using a Sequenase DNA sequencing kit (version 2.0, USB) according to the manufacturer's directions. Reactions were run in both the forward (5′ to 3′) and reverse (3′ to 5′) directions using M13 primers. Sequences were read from standard 6% acrylamide sequencing gels, which were dried and exposed to autoradiographic film (12 to 24 hours). After identifying homology of clone 2M2/5 with type VIII collagen, we probed an additional migrating versus stationary Northern blot with a 540-bp cDNA clone for rat α1 type VIII collagen obtained from Dr N. Rosenblum at the University of Toronto.11 The longer cDNA gave stronger signals on Northern blots.
Migration-Enhanced Versus Migration-Inhibited Northern Blots
Clones that were differentially expressed both on the dot blot and migrating versus stationary Northern blot were hybridized to a five-group Northern blot. In addition to migrating and stationary RNA, these Northern blots contained RNA from rats subject to gentle filament denudation of the carotid alone, gentle filament denudation+anti-bFGF antibody, and gentle filament denudation+anti-bFGF+anti-PDGF antibodies (anti–PDGF-BB antibodies were provided by Dr Charles Hart of Zymogenetics Inc, Seattle, Wash). These five-group Northern blots were hybridized, washed, and visualized using the same conditions as described above. mRNA levels were analyzed using data obtained from the phosphorimager. Intensity of the 28S ribosomal RNA band was measured using a scanning densitometer. The signal due to mRNA in each lane was normalized for loading by expressing the signal as a percentage of 28S rRNA.
Type VIII collagen protein was detected by probing Western blots with a guinea pig polyclonal antibody raised against bovine type VIII collagen. The antibody was kindly provided by Helene Sage, University of Washington, Seattle. A Western blot with arterial extracts from stationary carotids and injured PDGF-BB–treated carotid arteries was prepared as previously described,12 with 7.4 μg total protein loaded in each lane, and it was probed with the antibody. Purified type VIII collagen extracted from bovine Descement's membrane was used as a positive control sample, and the antibody reacted with a band of ≈50 kD. A second blot was prepared with arterial extracts from control vessels and vessels taken at 1, 2, 4, 7, 14, and 42 days after balloon catheter injury of the carotid artery. On this blot, 10 μg total protein was added in each lane.
Unpaired Student's t tests were used to analyze significant differences in SMC migration and replication between filament-denuded control and anti-bFGF/PDGF-BB–infused rat carotid arteries.
The aim of the present study was to isolate genes overexpressed by migrating SMCs. To maximize the chances of isolating such genes from rat arteries, we induced SMC migration by addition of PDGF-BB and anti-bFGF to rats after gentle denudation injury. This caused an eightfold increase in migration from 31.78±9.70 cells/mm2 on the intimal surface in control rats (gentle denudation injury+vehicle infusion) to 260±25.5 cells/mm2 in the treated rats (Fig 2⇓). By contrast, the medial SMC replication rate was not significantly increased by PDGF-BB and anti-bFGF treatment (Fig 3⇓). Intimal SMC replication rate was 61.74±5.74% in PDGF-BB– and anti-bFGF–treated rats, which was greater than the value of 22.60±10.46% in filament-denuded control rats (Fig 4⇓). Although this increase in SMC replication rate was significant (P<.05), the increase in replication under these treatment conditions (2.7-fold) was far less than the increase in migration (8-fold). Thus, we used the RNA extracted from these arteries to isolate genes that are potentially important for SMC migration.
Rats were treated with PDGF-BB and anti-bFGF as outlined above, and RNA from this migrating condition was compared by differential display to RNA from the media of unmanipulated common carotid arteries with endothelium stripped off (stationary condition). The PCR products from migrating and stationary RNA samples were run side by side on polyacrylamide sequencing gels, and cDNA fragments that were overexpressed in the migrating sample were recovered by excision and elution from the gel. Fig 5⇓ shows the differential overexpression of a 122-bp cDNA band, 2M2/5. In the present study, we report on expression of this cDNA. The fragment was reamplified by PCR using the same primers and reaction conditions as the initial differential display reaction, and the reamplified product was purified by agarose gel electrophoresis, ligated into Bluescript plasmid vector, and grown and plated in DH5α cells. Resultant colonies were screened by PCR to ensure that a single fragment was inserted into the plasmid, and further confirmation was obtained by restriction mapping the plasmids from several colonies on each plate.
Dot Blot and Northern Blot Screening of Migrating cDNA Clones
Since there are many possible artifacts involved in reverse-transcriptase PCR amplification of RNA, differential expression of cDNA clones was confirmed using two additional methods: dot blot and Northern blot screening. A dot blot was prepared from plasmid preps of clone 2M2/5 and 33 other clones. Duplicate nylon membranes were probed with 32P-labeled cDNA generated by reverse transcription of total RNA from migrating and stationary samples. Clone 2M2/5 and 13 other clones hybridized more strongly with the labeled migrating cDNA probe. Clone 2M2/5 was dideoxy-sequenced in both directions and was found to contain a cDNA fragment 90% homologous to mouse α1 type VIII collagen gene. Total RNA derived from migrating and stationary carotids was applied to Northern blots, which were probed with the 32P-labeled cDNA fragment of type VIII collagen. The labeled cDNA hybridized to an mRNA band of ≈4.2 kb (Fig 6⇓), consistent with the transcript size for mouse13 and rat11 α1 type VIII collagen mRNA. Expression of type VIII collagen was higher in the migrating sample than the stationary sample. Differential expression of the five other clones, including osteopontin, fibronectin, tropomyosin, and two clones with no known sequence homologies, was confirmed by Northern blotting. The sequences of these two unknown clones are shown in Fig 7⇓.
Expression of Type VIII Collagen mRNA in Migration-Enhanced Versus Migration-Inhibited Conditions
Our initial Northern blots showed overexpression of type VIII collagen in the injured arteries, but this overexpression may have been due to injury alone or associated with proliferation and not specifically with migration. To address the relationship of type VIII collagen expression to migration, we examined expression in rat arteries subjected to various treatments designed to produce a gradation of the migration response in the vessel wall. Rats were subject to gentle filament denudation injury alone, injury and anti-bFGF antibodies, or injury and anti-bFGF and anti–PDGF-BB antibodies. Total carotid RNA from these three experiments was applied to a Northern blot along with RNA from migrating and stationary samples, and the Northern blots were probed with labeled type VIII collagen cDNA fragments (Fig 8⇓). Type VIII collagen mRNA expression was increased by 194% in the migrating compared with the stationary group and by 73% after gentle filament denudation injury alone. Treatment with anti-bFGF antibody after gentle denudation reduced the increase in mRNA expression to 28%, and treatment with anti-bFGF and anti–PDGF-BB antibodies decreased expression to 29% of control values.
Western Blot for Type VIII Collagen Protein After Filament Injury and PDGF-BB Infusion and After Balloon Catheter Injury
Western blots containing arterial extracts from stationary and migrating groups were probed with type VIII collagen antibody, and bands at 125 kD and 65 kD and a weaker one at 50 kD were observed in the injured artery. In comparison, the control artery showed weaker staining of the 125- and 65-kD bands (Fig 9⇓). These bands were consistent in size with type VIII collagen synthesized by cultured endothelial cells,14 with the 125-kD species representing the intact type VIII collagen and the 65- and 50-kD species arising from the digestion of the globular end domains of the protein. We also studied protein expression using a different injury model, balloon catheter injury of the rat carotid artery (Fig 10⇓). The same bands were detected in these arteries, although the 125-kD band, thought to be the undegraded protein, was most prominent. The protein was present in the control artery and increased dramatically at 1, 2, and 4 days after injury but decreased to control levels at times later than 7 days after injury. Western blots containing purified type VIII collagen extracted from bovine Descement's membrane with pepsin were used as a positive control, and immunoreactivity with a band of 50 kD was observed (results not shown).
Migration of SMCs from the media to the intima is an important step in arterial lesion development in experimental animals, yet our knowledge of the mechanisms controlling this process is limited. Our aim in the present study was to use differential display screening to identify genes specifically expressed by a migrating SMC population. In vivo, SMC migration occurs in arteries after endothelial denudation, but denuding injury results in widespread SMC activation characterized by a host of changes including loss of contractile apparatus, replication, and migration, making the separation of these events very difficult. Thus, a simple comparison between injured and uninjured arteries will yield those genes induced in response to injury and activation as well as those important for migration and replication. Analysis is further complicated because the time course of migration in a denuded artery is not known. Therefore, our approach in the present study was to use a well-characterized model of denudation in rat arteries that induces a low rate of cell replication,8 to add PDGF-BB to stimulate migration,4 and to block cell replication with the addition of an antibody against bFGF.7 Arteries treated in this manner should be enriched for genes important for cell migration, and these genes were isolated by a comparison with cDNA isolated from arteries in which there was no migration. The cDNAs isolated on this initial screening were then used to screen Northern blots isolated from arteries under conditions in which migration was inhibited by administration of antibodies to two growth factors. Using this approach, we identified several genes that are upregulated after injury, and at least two of these factors have been linked to cell migration.15 16 17
The infusion of PDGF-BB enhanced migration by eightfold, whereas intimal SMC proliferation was increased only twofold by this treatment. These results extend and confirm an earlier study in which the same dose of PDGF-BB was administered for 2 weeks after injury.4 In that study, intimal thickening doubled in the absence of a significant increase in SMC proliferation, and it was inferred that the increases in thickness were due to increased SMC migration. In the present study, we provide direct evidence of a role for PDGF-BB in stimulating migration by quantifying the number of cells that had migrated through the internal elastic lamella. bFGF also plays an important role in mediating SMC replication and migration in injured arteries.7 18 The effects of this factor were minimized by performing a gentle denudation injury (which causes little medial damage and, therefore, little bFGF release8 18 ) and by giving a single injection of anti-bFGF antibody to block any released bFGF. In this way, we hoped to enrich these arteries for PDGF-stimulated migrating cells with a minimum of replication. Our data show that this strategy was successful, because migration was increased significantly more than proliferation, although SMC replication was not totally blocked.
The first clone isolated and sequenced in our experiment using the differential display approach described above was osteopontin, which has been recently shown to be present in the intima at times when SMCs are migrating.15 Furthermore, this molecule has been shown to influence cell migration and not cell replication in vitro.16 Other clones isolated by this strategy coded for rat α1 type VIII collagen, fibronectin, tropomyosin, and two sequences with no known homologies. In the present study, we report on the cDNA clone 2M2/5, which, on the basis of partial sequence analysis and mRNA size, codes for rat α1 type VIII collagen. On Northern blots, type VIII collagen mRNA was overexpressed in rat carotid arteries in which SMC migration was stimulated by filament denudation injury and PDGF-BB infusion compared with unmanipulated control vessels. Type VIII collagen expression was increased by filament denudation alone, and this increase was diminished by treating arteries with antibodies to block endogenous bFGF alone or in combination with endogenous PDGF-BB (Fig 8⇑, bottom). Both these treatments have also been shown to inhibit SMC migration in the injured rat carotid artery.2 6 18 Therefore, these results suggest that type VIII collagen expression correlates with the degree of SMC migration after arterial injury and demonstrates a role for both endogenous bFGF and PDGF-BB in stimulating expression of type VIII collagen.
Type VIII collagen is a protein with a short central triple-helical domain and nonhelical globular domains at the amino and carboxy termini of the molecule and is composed of two α1 chains and one α2 chain.19 It is found in perichondrium, periosteum, meninges, sclera, mesangium of the kidney, and Descement's membrane of the cornea.20 Immunoreactive protein is produced by endothelial cells during angiogenesis in vitro,14 consistent with production by a motile cell type during remodeling, and in the subendothelial intima of large blood vessels21 22 and kidney arterioles22 in mature animals, consistent with our detection of the protein in control vessels. In our present experiments, we found that type VIII collagen was overexpressed in rat carotid arteries after injury.
Western blots showed that type VIII collagen protein level was increased at 7 days in the filament-injured PDGF-BB–stimulated carotid arteries, confirming that upregulation occurred at the level of mRNA and protein. We also studied a time course of protein expression in rat arteries after balloon catheter injury, because the kinetics of SMC migration and proliferation have been well characterized in the model. Type VIII collagen protein increased transiently, peaking at 2 to 4 days after balloon injury, and disappeared very rapidly with establishment of the neointima. This time course of expression correlates quite closely with early SMC responses to vascular injury, including migration and proliferation. Turnover of type VIII collagen is probably very rapid, since it is easily digested within its globular domains by a variety of serine proteases and in the collagenous domain by MMPs.19 Plasmin2 and MMP activity3 present in rat carotid arteries between 1 and 7 days after injury may provide a mechanism for rapid clearance of type VIII collagen from the vessel wall. We hypothesize that type VIII collagen was laid down as a provisional matrix during SMC remodeling of the vessel wall but was cleared once the cells reached the intima, where they synthesized a new, more permanent matrix composed of fibrillar type I and III collagen and elastin. There is a precedent for the transient expression of type VIII collagen in the developing mouse heart: at embryonic day 11, the protein is present in the myocardium, but just 4 days later, the myocardial deposition is cleared, and type VIII collagen is found in the subendocardium and cardiac jelly.17 23
Although some differential display PCR reactions used primers designed to code for consensus sequences of MMP and integrin molecules, we were not successful in amplifying homologous gene products. It is likely that annealing occurred in an arbitrary, random fashion and that the short primer length was insufficient to allow hybridization to specific sequences.
In conclusion, we have studied differential gene expression in injured PDGF-BB–stimulated rat carotid arteries using a differential display technique. We have identified rat α1 type VIII collagen as a gene that was overexpressed after injury, and we believe that its expression at mRNA and protein levels is coincident with SMC migration. This molecule may serve as a useful tag to identify migrating SMCs; furthermore, it may play an important functional role in mediating SMC migration. Our results also suggest that type VIII collagen expression is regulated by the growth factors PDGF-BB and bFGF. Future studies will concentrate on establishing a functional role for type VIII collagen in tissue remodeling after arterial injury.
Selected Abbreviations and Acronyms
|bFGF||=||basic fibroblast growth factor|
|PCR||=||polymerase chain reaction|
|PDGF||=||platelet-derived growth factor|
|SMC||=||smooth muscle cell|
This study was supported by National Institutes of Health grants HL-41103 and HL-03174. Dr Bendeck was supported by a postdoctoral research fellowship from the Medical Research Council of Canada. The authors would like to acknowledge the excellent technical assistance of Jeff Kozlowski. We would like to thank Dr Helene Sage for her generous gifts of purified type VIII collagen and antibodies to type VIII collagen and for helpful discussions and comments on the manuscript. We also acknowledge Dr Norman Rosenblum for the type VIII collagen cDNA probe. Finally, we want to thank Zymogenetics Inc for the generous gift of recombinant PDGF-BB and polyclonal PDGF-BB antibodies.
Reprint requests to Dr Michelle P. Bendeck, PhD, Vascular Biology, 810F, St. Michael's Hospital, 30 Bond St, Toronto, Ontario, Canada M5B 1W8. E-mail firstname.lastname@example.org.
- Received December 29, 1995.
- Accepted May 14, 1996.
Jackson CL, Raines E, Ross R, Reidy MA. Role of endogenous platelet-derived growth factor in arterial smooth muscle cell migration after balloon catheter injury. Arterioscler Thromb. 1993;13:1218-1226.
Bendeck MP, Zempo N, Clowes AW, Galardy RE, Reidy MA. Smooth muscle cell migration and matrix metalloproteinase expression after arterial injury in the rat. Circ Res. 1994;75:539-545.
Jawien A, Bowen-Pope DF, Lindner V, Schwartz SM, Clowes AW. Platelet-derived growth factor promotes smooth muscle cell migration and intimal thickening in a rat model of balloon angioplasty. J Clin Invest. 1992;89:507-511.
Fingerle J, Johnson R, Clowes AW, Majesky MW, Reidy MA. Role of platelets in smooth muscle cell proliferation and migration after vascular injury in rat carotid artery. Proc Natl Acad Sci U S A. 1989;86:8412-8416.
Ferns GAA, Raines EW, Sprugel KH, Montani AS, Reidy MA, Ross R. Inhibition of neointimal smooth muscle cell accumulation after angioplasty by an antibody to PDGF. Science. 1991;253:1129-1132.
Lindner V, Reidy MA. Proliferation of smooth muscle cells after vascular injury is inhibited by an antibody against basic fibroblast growth factor. Proc Natl Acad Sci U S A. 1991;88:3739-3743.
Fingerle J, Au YPT, Clowes AW, Reidy MA. Intimal lesion formation in rat carotid arteries after endothelial denudation in absence of medial injury. Arteriosclerosis. 1990;10:1082-1087.
Lindner V, Olson NE, Clowes AW, Reidy MA. Inhibition of smooth muscle cell proliferation in injured rat arteries: interaction of heparin with basic fibroblast growth factor. J Clin Invest. 1992;90:2044-2049.
Rosenblum ND, Briscoe DM, Karnovsky MJ, Olsen BR. a1-VIII collagen is expressed in the rat glomerulus and in resident glomerular cells. Am J Physiol. 1993;264:F1003-F1010.
Iruela-Arispe ML, Diglio CA, Sage EH. Modulation of extracellular matrix proteins by endothelial cells undergoing angiogenesis in vitro. Arterioscler Thromb. 1991;11:805-815.
Giachelli CM, Bae N, Almeida M, Denhardt DT, Alpers CE, Schwartz SM. Osteopontin is elevated during formation in rat arteries and is a novel component of human atherosclerotic plaques. J Clin Invest. 1993;92:1686-1696.
Liaw L, Almeida M, Hart CE, Schwartz SM, Giachelli CM. Osteopontin promotes vascular cell adhesion and spreading and is chemotactic for smooth muscle cells in vitro. Circ Res. 1994;74:214-224.
Iruela-Arispe ML, Chun LE, Sage EH. Structure and biology of type VIII collagen. Trends Glycosci Glycotechnol. 1992;4:188-199.