Atherosclerosis, 86 (1991) 219-226 ‘V 1991 Elsevier Scientific Publishers ADONIS 002191509100072Y
ATHERO
219 Ireland,
Ltd. 0021-9150/91/$03.50
04594
Secretion of a potent new migration factor for smooth muscle cells (SMC) by cultured SMC Noriyuki
Koyama,
Second Department
Tomoko
Koshikawa, Nobuhiro and Sho Yoshida
of Internal Medicine, School of Medicine,
Chiha University.
Morisaki,
Yasushi
Saito
I -8-I Inohana, Chrbo 280 (Japan)
(Received 20 June. 1990) (Revised. received 18 October. 1990) (Accepted 19 October. 1990)
Summary Migration of smooth muscle cells (SMC) in the arterial wall is important in the formation of intimal thickening. In this work, cultured SMC from the rat and rabbit aortic media at 2nd to 12th passages were found to secrete a potent migration factor for SMC which was named SMC-derived migration factor (SDMF). This factor stimulated the migration of SMC dose-dependently and its maximum activity was 2-8 times that of PDGF. Checker board analysis showed that SDMF was chemotactic, but not chemokinetic. In further studies, SDMF was found to be inactivated at 100°C for 10 min or by trypsinization, but not inactivated by mercaptoethanol. This factor was not dialyzable. Molecular weight was approximately 500 kDa by a gel filtration. The activity was not inhibited by an anti-PDGF antibody or a fibronectin antiserum. These data suggest that SDMF is a potent migration factor for SMC and that SDMF is distinct from PDGF, fibronectin or other known migration factors. This autocrine system of secretion of SDMF by SMC and its induction of SMC migration may contribute to intimal thickening of the arterial wall in atherosclerosis.
Key words:
SMC migration;
Autocrine
system;
Chemotactic
Introduction Migration of arterial smooth muscle cells (SMC) from the media to the intima (SMC) is a key
Correspondence to: Dr. Yasushi Saito. The Second Dept. of Internal Medicine. School of Medicine, Chiba University, 1-8-1 Inohana, Chiba 280, Japan.
factor;
Intimal
thickening
process in the formation of intimal thickening in arteriosclerotic lesions [1,2]. Several factors that affect this process have been reported. Platelet derived growth factor (PDGF) has been shown to enhance both the migration and the proliferation of SMC [3], and to be secreted by various cells in atheromatous lesions, such as aggregated platelets [4], endothelial cells [5] and macrophages [6]. Therefore, this factor may act as a mediator for
220 the migration of SMC by a paracrine mechanism in the arterial wall. Other factors secreted from SMC themselves may induce SMC migration by an autocrine mechanism, resulting in the formation of intimal thickening. Previous studies from our laboratory have shown that cultured SMC secreted a new growth factor named SDGF for SMC as autocrine [7]. However, the autocrine mechanism of SMC migration is not yet understood. Here, we reported a potent migration factor for SMC secreted from SMC themselves, which we named SMC-derived migration factor (SDMF) and an autocrine system of SMC migration probably involved in the formation of intimal thickening. Materials and methods Reagent Purified human PDGF (purity > 95%, Lot No. 0711) was purchased from R&D System Inc. (Minneapolis, MN) and purified rat fibronectin (purity > 95% Lot No. 519) was from Biomedical Technologies Inc. (Stoughton, MA), respectively. Polyclonal anti-PDGF antibody (Lot No. 89-0414) and anti-fibronectin antiserum (Lot No. 88-1181) were purchased from Collaborative Research Inc. (Lexington, MA). Sepharose CLdB was purchased from Pharmacia Fine Chemicals (Uppsala, Sweden). Culture SMC Wistar by the reported
of SMC were explanted from the thoracic aorta of rats or Japanese white rabbits essentially method of Fischer-Dzoga et al. [8], as in detail elsewhere [7,9].
Culture of macrophages (M+) M+ were collected from the rat peritoneum by the method of Goldstein et al. [lo]. Ten ml 3% thioglycorate solution was injected into the peritoneum of 2 SD rats. After 4 days, peritoneal M+ were harvested in phosphate buffered saline (PBS), subjected to centrifugation, resuspended in Dulbecco’s modified Eagle’s medium (DME) containing 10% fetal bovine serum and cultured for 6 h in 6 wells of 3.5 cm diameter (Coming Co., NY, U.S.A.). After the adherent cell layer was washed with serum-free DME twice vigorously, 2 ml serum-free DME was added to each well. Condi-
tioned medium 8 days.
was collected
every 48 h for up to
Collection of conditioned medium (CM) from SMC and treatment of CM Confluent SMC in T-75 flasks at the 2nd passage were subcultured successively at a 1 : 2 split ratio in T-75 flasks. CM was collected as follows: Confluent SMC in T-75 flasks at the 2nd to 12th passage were washed twice with 10 ml DME and incubated with 10 ml fresh DME. CM was collected every 2 days with medium change to 10 ml fresh DME. Where indicated, samples of CM were subjected to the following treatment: (1) heat treatment at 56°C for 30 min or at 100°C for 10 min; (2) freeze thawing by freezing at -20°C for 24 h and thawing at 22’C; (3) a pH-stability test by adjusting the preparation to either pH 2.5 with 1 M HCl or pH 10.0 with 1 M NaOH, incubating it for 30 min at 22°C and then readjusting it to pH 7.4; (4) dialysis in a Spectra/Par No. 3 membrane (molecular weight cut off: approximately 3500) against PBS for 24 h at 4°C; (5) trypsinization by incubation with a solution of trypsin (Sigma Chemical Co., St. Louis, MO, U.S.A.) in PBS at a final concentration of 0.01% or 0.1% trypsin for 30 min at 37°C and then inactivation of trypsin by addition of soybean trypsin inhibitor (Sigma Chemical Co.); (6) reductive alkylation by treatment with 2 mM mercaptoethanol for 30 min at 22°C and then dialyzed against PBS for 24 h. Migration assay of SMC Migration of SMC was assayed by a modification of Boyden’s chamber method using a polycarbonate filter (Nuclepore Co., U.S.A.) with pores of 5.0 pm diameter [ll]. Cultured SMC were trypsinized and suspended at a concentration of 5.0 X lo5 cells/ml in DME supplemented with 10% fetal bovine serum. Then 1 ml SMC suspension was placed in the upper chamber and 0.9 ml DME containing a migration factor(s) was placed in the lower chamber. The chamber was incubated at 37°C in 5% CO, and 95% air for 4 h. SMC on the upper side of the filter were scraped off and the filter was removed. Then SMC that had migrated to the lower side of the filter were fixed in ethanol, stained with hematoxylin and counted
221 by microscopy observation (X 400) for quantitation of SMC migration. Migration activity is expressed as the mean number of migrated cells seen in 10 high-power fields (HPF). In this work, the migration activity of CM examined was expressed as the relative value to the highest activity of PDGF. defined as PDGF unit. Gel filtrution chromatography of CM from SMC A sample of 3 ml CM from SMC was applied to a 1.9 x 22 cm column of Sepharose CL-6B (Pharmacia Fine Chemicals, Sweden) and materials were eluted with PBS, pH 7.4, at 6.0 ml/h. The eluate was collected in 2.0-ml fractions and the migration activity of each fraction was measured by the modified Boyden’s chamber method. The molecular weight was calibrated with standard proteins from Pharmacia Fine Chemicals: 669 kDa, thyroglobulin; 440 kDa, ferritin; 232 kDa, catalase; 158 kDa, aldolase; 43 kDa, ovalbumin. Results Migration
activity of CM from SMC Migration activities of CM of cultured SMC from rabbit and rat aortic media for SMC were detected and the activities of 50% CM were compared with those of control medium and PDGF 2.0 ng/ml (Fig. 1). The migration activity of PDGF was the highest in the dose range of 2.0-5.0 ng/ml and 4-10 times that of control as reported elsewhere. All the CMs examined had much higher migration activities than that of control medium: rabbit CMs had 57.2 i 46.5 times more (mean + SD; n = 26) and rat CMs had 81.8 + 65.5 times more (n = 12). The migration activities of the CMs were also much higher than that of PDGF: rabbit CMs had 4.44 + 1.05 PDGF units (mean &- SD; n = 26) and rat CMs had 4.60 _t 1.71 PDGF units ( n = 12). These data indicated that cultured SMC secreted a potent migration factor for SMC, which we named SDMF. SDMF from rats and rabbits SMC stimulated the migration of both rats and rabbits SMC, indicating cross-reactivity of the factor between rats and rabbits (data not shown).
Dose effects of SDMF und PDGF on the migration of SMC As shown in Fig. 2, SDMF in CM stimulated the migration of SMC dose-dependently and
:
_-_--__-____..__-_--_--
blank
PDGF
rabbit SW
rat SMC
-L
Fig. 1. Migration activity of CM from rabbit and rat SMC. SMC suspension was placed in the upper chamber and DME containing 2 q/ml of PDGF or 50$ of CM from rabbit (0) and rat (0) SMC was in the lower chamber, respectively. The migration activities of PDGF and CM were assayed in duphcate as described in Materials and Methods. Each point in the figure represents the mean value of the migration activity as PDGF unit.
showed maximal activity at concentrations of CM of over 50%. The figure also shows that CM from SMC has much higher migration activities than PDGF. In the following experiments 50% CM was used for the migration of SMC. unless otherwise indicated. All CMs collected from SMC at various passages, from the 2nd to the 12th, had migration activity and the activity of the CM was not affected by the passage number of SMC (data not shown). A checker board ana!vsis of migrution activicv of CM from SMC and PDGF Checker board analysis made using one chemotaxis model and two chemokinesis models was used to investigate the mechanism of induction of SMC migration by SDMF (Fig. 3). When only the lower chamber contained CM from SMC, the migration activity was 2.41 PDGF unit. But when only the upper chamber contained CM, or both chambers contained CM, the migration of SMC in the upper chamber was less and the migration
222 PDGF
5 25
2
8 120> .Y .;
80-
.-5
f
4;jFyu-y 75 Concentration
010
015
of Conditioned
1 :O
Concentration
115 of PDGF
100 Medium
210
W
215
(w/ml)
Fig 2. Dose-dependence of CM from SMC and PDGF on the migration of SMC. SMC suspension was placed in the upper chamber and DME containing each concentration of CM from rat SMC (0) and PDGF (0) was in the lower chamber, respectively. The migration activities of CM and PDGF at each concentration were assayed in duplicate as described in Materials and Methods. Each point in the figure represents the mean value of the migration activity as PDGF unit. Similar results were obtained from a repeated experiment.
activities (in PDGF units) were 0.18 (not significant, vs. blank) and 0.34 (P < 0.05, vs. blank), respectively. Similar results were obtained with PDGF as a migration factor. These results show that the migration by SDMF contained the major chemotactic component with the minor chemokinetic one, and that its mechanism was similar to that of PDGF. Characterization of SDMF The physicochemical properties of SDMF are summarized in Table 1. This factor was not inactivated by heat-treatment at 56°C for 30 min, but was inactivated by heat-treatment at 100°C for 10 min. Its activity was not affected by freezethawing. It was stable at pH 2.5 or pH 10.0 at 22°C for 30 min. It was not dialyzable through membranes with a cut-off molecular weight of
Fig. 3. Checker board analysis of migration activity of CM from SMC and PDGF. Checker board analysis was made with one chemotaxis model and two chemokinetics models. SMC suspension was placed in the upper chamber and DME was placed in the lower chamber. CM from rat SMC or PDGF was added only to the lower chamber (a chemotaxis model), added only to the upper chamber or added to both the chambers (chemokinetics models). The migration activities of CM from rat SMC and PDGF in the models were assayed in the 4 chambers as described in Materials and Methods. The migration activity was expressed as the meanf SD of the migrated cells in 10 high power fields.
TABLE
1
PHYSICOCHEMICAL PROPERTIES TION FACTOR IN CM FROM SMC
OF
THE
MIGRA-
The SMC suspension was placed in the upper chamber and DME containing untreated or treated CM from rabbit SMC was placed in the lower chamber. After the various treatments indicated in the table, the migration activity of CM was assayed in the 3 to 7 chambers as described in Materials and Methods. The relative value of control was represented as the mean f SD. Migration activitv (%)
Treatment
Control
(no treatment)
Heated at 56°C for 30 min 100°C for 10 min Freeze-thawing pH 2.5 at 22’C for 30 min pH 10.0 at 22°C for 30 min Dialysis (cut off Mw 3500) Trypsin (0.01%) at 37OC for 30 min (0.1%) at 37°C for 30 min Mercaptoethanol(2 mM) at 22°C for 30 min
lOO.Ok
4.2
85.1 f 0.8+ 107.6 f 97.5 + 96.8 + 75.1 + 20.8* 2.8* 88.6*
1.1 0.2 13.2 16.3 16.9 11.1 2.0 0.7 6.7
223
rabbit SMC-CM
rat SW-CM
5
10
15
20
25
30
35
Fraction Number
Fig. 4. Gel filtration chromatography of CM from SMC. CM from rat SMC was applied to the column of Sepharose CL-6B gel and material was eluted with phosphate buffered saline. The eluate was collected and the migration activity was measured as described in Materials and Methods. Each point represents the mean value of duplicated assay. 6 repeated experiments gave similar results. The arrows indicate the void position and the elution positions of standard proteins.
s .g
40
&I ._ =
20
0
6
1’0 io Anti-PDGF
Antibody
sb (a/ml)
3500. Its activity was almost completely lost on treatment with trypsin but not on treatment with 2 mM mercaptoethanol at 22°C for 30 min. CM from SMC was applied to the column of Sepharose CL-6B gel, the migration activity was eluted in a fraction corresponding to a molecular weight 500 kDa (Fig. 4).
Fig. 5. Dose-dependence of anti-PDGF antibody on the migration of SMC. Human polyclonal anti-PDGF antibody was added to CM from rabbit (0) or rat (W) SMC. CM from rat Me (0) or human PDGF (0) and then migration activities were assayed in the 6 chambers as described in Materials and Methods. The figure represents the mean&SD of the relative value to the migration activity without antibody as percentage.
Inhibition of migration activities of CM by an antiPDGF antibody A human polyclonal anti-PDGF antibody inhibited the migration activity of human PDGF dose-dependently. The antibody inhibited the migration activity of PDGF almost completely at 50 pg/ml. This antibody dose-dependently inhibited also that of CM from rat M$ which contains rat PDGF [6]. But this antibody did not inhibit the migration activity of CM from rat SMC or rabbit SMC (Fig. 5). These results indicate that this polyclonal antibody crossreacts with rat PDGF and that the migration activity of CM from rat SMC and rabbit SMC is not due to PDGF.
SMC [12] and to have chemotactic activity for various cells [13,14]. Fibronectin has a molecular weight of approximately 450 kDa which is similar to that of SDMF. To examine whether the migration factor in CM from SMC is due to fibronectin or not, the migration activity of rat fibronectin was compared with CM from rat SMC (Fig. 6). Fibronectin induced the migration of SMC dosedependently at a dose of 5-100 pg/ml. However, its maximal activity was 1.58 PDGF units observed at a dose of 40 pg/ml, which was about one fourth of that of SDMF in the same experiment. Next. the dose effects of human polyclonal anti-fibronectin antiserum on the migration activity of CM from SMC were compared with that of rat fibronectin and that of PDGF (Fig. 7). This antiserum inhibited the migration activity of fibronectin dose-dependently. The antiserum at
Relationship of the migration factor in CM from SMC with fibronectin Fibronectin is also reported to be produced by
224 ‘Gi % o r 8 f 2 = Q)
300
250
z .= > ‘3
200
:
150
C 0 0 6
100
fibronectin
._ I
fibronsctin
50
0
I 0.0
,
3
1.0
2.0
Anti-fibronectin
io
b I’0
4b
fibronectin
r
I
0510
I
lb0 (dml)
1
20 SMC-CM
1
50 (%I
Fig. 6. Dose-dependence of CM from SMC and fibronectin on the migration of SMC. SMC suspension was placed in the upper chamber and DME containing each concentration of CM from rat SMC and rat fibronectin was in the lower chamber, respectively. The migration activities of CM (0) and fibronectin (A) at each concentration were assayed in duplicate as described in Materials and Methods. Each point in the figure represents the mean value of the migration activity as the migrated cells in 10 high power fields. A repeat experiment gave similar results.
the dose which inhibited 87% of the migration activity of fibronectin inhibited only 20% of that of CM from rat SMC as well as 13% of that of PDGF and these slight inhibitions were not dosedependent. These results indicate that this antiserum crossreacts with rat fibronectin and that the major part of the migration activity of CM from rat SMC is not due to fibronectin. Discussion In this work, we reported that cultured SMC from rat and rabbit aortic media at the 2nd to the 12th passage secreted a migration factor(s) for
2.5
3.0
Antiserum (unit)
Fig. 7. Dose-dependence of anti-fibronectin antiserum on the migration of SMC. Human polyclonal anti-fibronectin antiserum was added to CM from rat SMC (o), rat fibronectin (0) and human PDGF (A) and the migration activities of them were assayed in the 6 chambers as described in Materials and Methods. The figure represents the mean+ SD of the relative value to the migration activity without antiserum as a percentage.
SMC of rat and rabbit. This factor, SMC derived migration factor (SDMF), stimulated the migration of SMC dose-dependently and acted as a chemotactic factor, but not as a chemokinetic one. We characterized SDMF physicochemically to determine its relation with other migration factors. SDMF was not inactivated by heat treatment at 56°C for 30 min, but was inactivated by heat treatment at 100°C for 10 min. It was not inactivated at pH 2.2 or 10.0 at 22°C for 30 min, by dialysis. On treatment with 2 mM mercaptoethanol, its activity was not lost, suggesting that S-S bonds are not important for its migration activity. The result that SDMF was inactivated on treatment with trypsin indicates that SDMF is a polypeptide. Molecular weight was approximately 500 kDa. Our previous study has shown that cultured SMC secreted a new growth factor named SDGF [7]. SDGF has a molecular weight of 8.7 kDa which is lower than that of SDMF and its growth
225 activity is lost on treatment of 2 mM mercaptoethanol, suggesting that SDMF is different from SDGF. Several factors that enhance the migration of SMC have been reported, such as 12-L-hydroxy5,8,10.14-eicosatetraenoic acid (12-HETE) [15]. leukotriene B4 (LTB,) [16]. interleukin-1 (IL-l) [17], elastin peptide [ll], PDGF [3] and fibronectin (Fig. 6). 12-HETE and LTB, have low molecular weights and are dialyzable. IL-1 has a molecular weight of 12-20 kDa [18] and elastin peptide, formed by digestion of elastin has a molecular weight of less than 25 kDa [19]. Therefore, as these four factors differ from SDMF in molecular weights, it is unlikely that SDMF is one of these factors. PDGF is reported to be secreted by SMC [20] and to have chemotactic activity for SMC [3]. However, the molecular weight (32 kDa) is much lower than that of SDMF, and it is not inactivated by heat treatment at 100°C but is inactivated by mercaptoethanol [21]. These physicochemical properties differ from those of SDMF shown in Table 1. Furthermore, the migration activity of SDMF was not inhibited by human polyclonal anti-PDGF antibody, which inhibited that of purified human PDGF and that of PDGF derived from rat Me. These results indicate that this polyclonal antibody crossreacts with rat PDGF. Thus at least the migration activity of CM from rat SMC is not due to PDGF. Fibronectin has a molecular weight of approximately 450 kDa, which is also similar to that of SDMF. Fibronectin is also reported to be produced by SMC [12] and to have chemotactic activity for SMC (Fig. 6). However, its maximal activity was one fourth of that of SDMF. Moreover. the migration activity of rat SDMF was not inhibited by human anti-fibronectin rabbit antiserum, which strongly inhibited the activity of purified rat fibronectin. These results indicate that this antiserum crossreacts with rat fibronectin and that the migration activity of CM from rat SMC is not due to fibronectin. The above results indicate that SDMF is a new migration factor. distinct from various factors reported previously. This work showed that cultured SMC secrete a potent, new migration factor for SMC, which we
named SDMF. In atheromatous lesions, SDMF may enhance the migration of the medial SMC to the intima. This autocrine system of SMC migration may contribute to the development of intimal thickening in arteriosclerotic lesions. Acknowledgement
This work was supported in part by a grant from the Ministry of Education, Japan (No. 01570351). References
1 Ross. R.. The pathogenesis 2
3
4
5
6
7
8
9
10
of atherosclerosis - an update. N. Engl. J. Med.. 314 (1986) 488. Clowes. A.W.. and Schwartz. SM.. Significance of quiescent smooth muscle migration in the inJured rat carotid artery. Cir. Res.. 56 (1985) 139. Grotendorst. G.R., Seppa. H.E.J., Kleinman. H.K. and Martin. G.R.. Attachment of smooth muscle cells to collagen and their migration toward platelet-derived growth factor. Proc. Natl. Acad. Sci. USA. 78 (1981) 3669. Heldin. C.H.. Westermark. B. and Wasteson. A.. Plateletderived growth factor: purrfication and partial characterization, Proc. Natl. Acad. Sci. USA. 76 (1979) 3722. DiCorleto. P. and Bowen-Pope. D.. Cultured endothehal cells produce a platelet-derived growth factor-like protein. Proc. Natl. Acad. SCI. USA. 80 (1983) 1919. Shimokado. K.. Raines. E.W.. Madtes. D.K.. Barrett, T.B.. Bonditt, E.P. and Ross, R.. A significant part of macrophage-derived growth factor consists of at least two forms of PDGF, Cell. 43 (1985) 277. Morisaki. N.. Kanzaki, T.. Koshikawa. T.. Sarto. Y. and Yoshida. S.. Secretion of a new growth factor, smooth muscle cell derived growth factor. dtstinct from platelet derived growth factor by cultured rabbit aorttc smooth muscle cells. FEBS Lett., 230 (198X) 186. Fischer-Dzoga. K.. Jones. R.M.. Vesselinovttch, D. and Wissler. R.W.. Ultrastructural and immunohistochemical studies of primary cultures of aortrc medial cells. Exp. Mol. Pathol.. 18 (1973) 162. Mortsaki. N., Koyama, N., Mori. S.. Kanzaki. T.. Koshikawa, T., Saito. Y. and Yoshida. S.. Effects of smooth muscle cell derived growth factor (SDGF) in combination with other growth factors on smooth muscle cells, Atherosclerosis. 78 (1989) 61. Goldstein. J.L., Ho. Y.K.. Basu. SK. and Brown. M.S..
Binding site on macrophage that mediates uptake and degradation of acetylated low density lipoprotein. producing massive cholesterol deposition. Proc. Natl. Acad. Sci. USA, 76 (1979) 333. 11 Ooyama, T.. Fukuda. K.. Oda. H.. Nakamura. H. and Hikita. Y.. Substration-bound elastin peptide Inhibits aortic smooth muscle cell migration in vitro, Arteriosclerosis. 7 (1987) 593.
226 12 Holderbaum, D. and Ehrhart, L.A., Substratum influence on collagen and fibronectin biosynthesis by arterial smooth muscle cell in vitro, J. Cell. Physiol., 126 (1986) 216. 13 Postlethwaite, A.E., Keski-Oja, J., Balian, G. and Kang, A.H., Induction of fibroblast chemotaxis by fibronectin, J. Exp. Med., 153 (1981) 494. 14 Bowersox, J.C. and Sorgente, N., Chemotaxis of aortic endothelial cells in response to fibronectin, Cancer Res., 42 (1982) 2547. 15 Nakao, J., Ooyama, T., Ito, H., Chang, W.C. and Murota, S., Comparative effect of lipoxygenase products of arachidonic acid on rat aortic smooth muscle cell migration, Atherosclerosis, 44 (1982) 339. 16 Hirosumi, J., Nomoto, A., Ohkubo, Y., Sekigucbi, C., Mutoh, S., Yamaguchi, I. and Aoki, H., Inflammatory responses in cuff-induced atherosclerosis in rabbits, Atherosclerosis, 64 (1987) 243.
17 Nomoto, A., Mutoh, S., Hagihara, H. and Yamaguchi, I., Smooth muscle cell migration induced by inflammatory cell products and its inhibition by a potent calcium antagonist, nilvadipine, Atherosclerosis, 72 (1988) 213. 18 Oppenheim, J.J., Kovacs, E.J. and Matsushima, K., There is more than one interleukin 1, Immunol. Today, 7 (1986) 45. 19 Senior, R.M., Griffin, G.L. and Mecham, R.P., Chemotactic activity of elastin-derived peptides, J. Clin. Invest., 66 (1980) 859. 20 Nilsson, J., Sjolund, M., Palmberg, L., Thyberg, J. and Heldin, C.H., Arterial smooth muscle cells in primary culture produce a platelet-derived growth factor-like protein, Proc. Natl. Acad. Sci. USA, 82 (1985) 4418. 21 Raines, E.W. and Ross, R., Platelet-derived growth factor, I. High yield purification and evidence for multiple forms, J. Biol. Chem., 257 (1982) 5154.
ATHERO
219 Ireland,
Ltd. 0021-9150/91/$03.50
04594
Secretion of a potent new migration factor for smooth muscle cells (SMC) by cultured SMC Noriyuki
Koyama,
Second Department
Tomoko
Koshikawa, Nobuhiro and Sho Yoshida
of Internal Medicine, School of Medicine,
Chiha University.
Morisaki,
Yasushi
Saito
I -8-I Inohana, Chrbo 280 (Japan)
(Received 20 June. 1990) (Revised. received 18 October. 1990) (Accepted 19 October. 1990)
Summary Migration of smooth muscle cells (SMC) in the arterial wall is important in the formation of intimal thickening. In this work, cultured SMC from the rat and rabbit aortic media at 2nd to 12th passages were found to secrete a potent migration factor for SMC which was named SMC-derived migration factor (SDMF). This factor stimulated the migration of SMC dose-dependently and its maximum activity was 2-8 times that of PDGF. Checker board analysis showed that SDMF was chemotactic, but not chemokinetic. In further studies, SDMF was found to be inactivated at 100°C for 10 min or by trypsinization, but not inactivated by mercaptoethanol. This factor was not dialyzable. Molecular weight was approximately 500 kDa by a gel filtration. The activity was not inhibited by an anti-PDGF antibody or a fibronectin antiserum. These data suggest that SDMF is a potent migration factor for SMC and that SDMF is distinct from PDGF, fibronectin or other known migration factors. This autocrine system of secretion of SDMF by SMC and its induction of SMC migration may contribute to intimal thickening of the arterial wall in atherosclerosis.
Key words:
SMC migration;
Autocrine
system;
Chemotactic
Introduction Migration of arterial smooth muscle cells (SMC) from the media to the intima (SMC) is a key
Correspondence to: Dr. Yasushi Saito. The Second Dept. of Internal Medicine. School of Medicine, Chiba University, 1-8-1 Inohana, Chiba 280, Japan.
factor;
Intimal
thickening
process in the formation of intimal thickening in arteriosclerotic lesions [1,2]. Several factors that affect this process have been reported. Platelet derived growth factor (PDGF) has been shown to enhance both the migration and the proliferation of SMC [3], and to be secreted by various cells in atheromatous lesions, such as aggregated platelets [4], endothelial cells [5] and macrophages [6]. Therefore, this factor may act as a mediator for
220 the migration of SMC by a paracrine mechanism in the arterial wall. Other factors secreted from SMC themselves may induce SMC migration by an autocrine mechanism, resulting in the formation of intimal thickening. Previous studies from our laboratory have shown that cultured SMC secreted a new growth factor named SDGF for SMC as autocrine [7]. However, the autocrine mechanism of SMC migration is not yet understood. Here, we reported a potent migration factor for SMC secreted from SMC themselves, which we named SMC-derived migration factor (SDMF) and an autocrine system of SMC migration probably involved in the formation of intimal thickening. Materials and methods Reagent Purified human PDGF (purity > 95%, Lot No. 0711) was purchased from R&D System Inc. (Minneapolis, MN) and purified rat fibronectin (purity > 95% Lot No. 519) was from Biomedical Technologies Inc. (Stoughton, MA), respectively. Polyclonal anti-PDGF antibody (Lot No. 89-0414) and anti-fibronectin antiserum (Lot No. 88-1181) were purchased from Collaborative Research Inc. (Lexington, MA). Sepharose CLdB was purchased from Pharmacia Fine Chemicals (Uppsala, Sweden). Culture SMC Wistar by the reported
of SMC were explanted from the thoracic aorta of rats or Japanese white rabbits essentially method of Fischer-Dzoga et al. [8], as in detail elsewhere [7,9].
Culture of macrophages (M+) M+ were collected from the rat peritoneum by the method of Goldstein et al. [lo]. Ten ml 3% thioglycorate solution was injected into the peritoneum of 2 SD rats. After 4 days, peritoneal M+ were harvested in phosphate buffered saline (PBS), subjected to centrifugation, resuspended in Dulbecco’s modified Eagle’s medium (DME) containing 10% fetal bovine serum and cultured for 6 h in 6 wells of 3.5 cm diameter (Coming Co., NY, U.S.A.). After the adherent cell layer was washed with serum-free DME twice vigorously, 2 ml serum-free DME was added to each well. Condi-
tioned medium 8 days.
was collected
every 48 h for up to
Collection of conditioned medium (CM) from SMC and treatment of CM Confluent SMC in T-75 flasks at the 2nd passage were subcultured successively at a 1 : 2 split ratio in T-75 flasks. CM was collected as follows: Confluent SMC in T-75 flasks at the 2nd to 12th passage were washed twice with 10 ml DME and incubated with 10 ml fresh DME. CM was collected every 2 days with medium change to 10 ml fresh DME. Where indicated, samples of CM were subjected to the following treatment: (1) heat treatment at 56°C for 30 min or at 100°C for 10 min; (2) freeze thawing by freezing at -20°C for 24 h and thawing at 22’C; (3) a pH-stability test by adjusting the preparation to either pH 2.5 with 1 M HCl or pH 10.0 with 1 M NaOH, incubating it for 30 min at 22°C and then readjusting it to pH 7.4; (4) dialysis in a Spectra/Par No. 3 membrane (molecular weight cut off: approximately 3500) against PBS for 24 h at 4°C; (5) trypsinization by incubation with a solution of trypsin (Sigma Chemical Co., St. Louis, MO, U.S.A.) in PBS at a final concentration of 0.01% or 0.1% trypsin for 30 min at 37°C and then inactivation of trypsin by addition of soybean trypsin inhibitor (Sigma Chemical Co.); (6) reductive alkylation by treatment with 2 mM mercaptoethanol for 30 min at 22°C and then dialyzed against PBS for 24 h. Migration assay of SMC Migration of SMC was assayed by a modification of Boyden’s chamber method using a polycarbonate filter (Nuclepore Co., U.S.A.) with pores of 5.0 pm diameter [ll]. Cultured SMC were trypsinized and suspended at a concentration of 5.0 X lo5 cells/ml in DME supplemented with 10% fetal bovine serum. Then 1 ml SMC suspension was placed in the upper chamber and 0.9 ml DME containing a migration factor(s) was placed in the lower chamber. The chamber was incubated at 37°C in 5% CO, and 95% air for 4 h. SMC on the upper side of the filter were scraped off and the filter was removed. Then SMC that had migrated to the lower side of the filter were fixed in ethanol, stained with hematoxylin and counted
221 by microscopy observation (X 400) for quantitation of SMC migration. Migration activity is expressed as the mean number of migrated cells seen in 10 high-power fields (HPF). In this work, the migration activity of CM examined was expressed as the relative value to the highest activity of PDGF. defined as PDGF unit. Gel filtrution chromatography of CM from SMC A sample of 3 ml CM from SMC was applied to a 1.9 x 22 cm column of Sepharose CL-6B (Pharmacia Fine Chemicals, Sweden) and materials were eluted with PBS, pH 7.4, at 6.0 ml/h. The eluate was collected in 2.0-ml fractions and the migration activity of each fraction was measured by the modified Boyden’s chamber method. The molecular weight was calibrated with standard proteins from Pharmacia Fine Chemicals: 669 kDa, thyroglobulin; 440 kDa, ferritin; 232 kDa, catalase; 158 kDa, aldolase; 43 kDa, ovalbumin. Results Migration
activity of CM from SMC Migration activities of CM of cultured SMC from rabbit and rat aortic media for SMC were detected and the activities of 50% CM were compared with those of control medium and PDGF 2.0 ng/ml (Fig. 1). The migration activity of PDGF was the highest in the dose range of 2.0-5.0 ng/ml and 4-10 times that of control as reported elsewhere. All the CMs examined had much higher migration activities than that of control medium: rabbit CMs had 57.2 i 46.5 times more (mean + SD; n = 26) and rat CMs had 81.8 + 65.5 times more (n = 12). The migration activities of the CMs were also much higher than that of PDGF: rabbit CMs had 4.44 + 1.05 PDGF units (mean &- SD; n = 26) and rat CMs had 4.60 _t 1.71 PDGF units ( n = 12). These data indicated that cultured SMC secreted a potent migration factor for SMC, which we named SDMF. SDMF from rats and rabbits SMC stimulated the migration of both rats and rabbits SMC, indicating cross-reactivity of the factor between rats and rabbits (data not shown).
Dose effects of SDMF und PDGF on the migration of SMC As shown in Fig. 2, SDMF in CM stimulated the migration of SMC dose-dependently and
:
_-_--__-____..__-_--_--
blank
PDGF
rabbit SW
rat SMC
-L
Fig. 1. Migration activity of CM from rabbit and rat SMC. SMC suspension was placed in the upper chamber and DME containing 2 q/ml of PDGF or 50$ of CM from rabbit (0) and rat (0) SMC was in the lower chamber, respectively. The migration activities of PDGF and CM were assayed in duphcate as described in Materials and Methods. Each point in the figure represents the mean value of the migration activity as PDGF unit.
showed maximal activity at concentrations of CM of over 50%. The figure also shows that CM from SMC has much higher migration activities than PDGF. In the following experiments 50% CM was used for the migration of SMC. unless otherwise indicated. All CMs collected from SMC at various passages, from the 2nd to the 12th, had migration activity and the activity of the CM was not affected by the passage number of SMC (data not shown). A checker board ana!vsis of migrution activicv of CM from SMC and PDGF Checker board analysis made using one chemotaxis model and two chemokinesis models was used to investigate the mechanism of induction of SMC migration by SDMF (Fig. 3). When only the lower chamber contained CM from SMC, the migration activity was 2.41 PDGF unit. But when only the upper chamber contained CM, or both chambers contained CM, the migration of SMC in the upper chamber was less and the migration
222 PDGF
5 25
2
8 120> .Y .;
80-
.-5
f
4;jFyu-y 75 Concentration
010
015
of Conditioned
1 :O
Concentration
115 of PDGF
100 Medium
210
W
215
(w/ml)
Fig 2. Dose-dependence of CM from SMC and PDGF on the migration of SMC. SMC suspension was placed in the upper chamber and DME containing each concentration of CM from rat SMC (0) and PDGF (0) was in the lower chamber, respectively. The migration activities of CM and PDGF at each concentration were assayed in duplicate as described in Materials and Methods. Each point in the figure represents the mean value of the migration activity as PDGF unit. Similar results were obtained from a repeated experiment.
activities (in PDGF units) were 0.18 (not significant, vs. blank) and 0.34 (P < 0.05, vs. blank), respectively. Similar results were obtained with PDGF as a migration factor. These results show that the migration by SDMF contained the major chemotactic component with the minor chemokinetic one, and that its mechanism was similar to that of PDGF. Characterization of SDMF The physicochemical properties of SDMF are summarized in Table 1. This factor was not inactivated by heat-treatment at 56°C for 30 min, but was inactivated by heat-treatment at 100°C for 10 min. Its activity was not affected by freezethawing. It was stable at pH 2.5 or pH 10.0 at 22°C for 30 min. It was not dialyzable through membranes with a cut-off molecular weight of
Fig. 3. Checker board analysis of migration activity of CM from SMC and PDGF. Checker board analysis was made with one chemotaxis model and two chemokinetics models. SMC suspension was placed in the upper chamber and DME was placed in the lower chamber. CM from rat SMC or PDGF was added only to the lower chamber (a chemotaxis model), added only to the upper chamber or added to both the chambers (chemokinetics models). The migration activities of CM from rat SMC and PDGF in the models were assayed in the 4 chambers as described in Materials and Methods. The migration activity was expressed as the meanf SD of the migrated cells in 10 high power fields.
TABLE
1
PHYSICOCHEMICAL PROPERTIES TION FACTOR IN CM FROM SMC
OF
THE
MIGRA-
The SMC suspension was placed in the upper chamber and DME containing untreated or treated CM from rabbit SMC was placed in the lower chamber. After the various treatments indicated in the table, the migration activity of CM was assayed in the 3 to 7 chambers as described in Materials and Methods. The relative value of control was represented as the mean f SD. Migration activitv (%)
Treatment
Control
(no treatment)
Heated at 56°C for 30 min 100°C for 10 min Freeze-thawing pH 2.5 at 22’C for 30 min pH 10.0 at 22°C for 30 min Dialysis (cut off Mw 3500) Trypsin (0.01%) at 37OC for 30 min (0.1%) at 37°C for 30 min Mercaptoethanol(2 mM) at 22°C for 30 min
lOO.Ok
4.2
85.1 f 0.8+ 107.6 f 97.5 + 96.8 + 75.1 + 20.8* 2.8* 88.6*
1.1 0.2 13.2 16.3 16.9 11.1 2.0 0.7 6.7
223
rabbit SMC-CM
rat SW-CM
5
10
15
20
25
30
35
Fraction Number
Fig. 4. Gel filtration chromatography of CM from SMC. CM from rat SMC was applied to the column of Sepharose CL-6B gel and material was eluted with phosphate buffered saline. The eluate was collected and the migration activity was measured as described in Materials and Methods. Each point represents the mean value of duplicated assay. 6 repeated experiments gave similar results. The arrows indicate the void position and the elution positions of standard proteins.
s .g
40
&I ._ =
20
0
6
1’0 io Anti-PDGF
Antibody
sb (a/ml)
3500. Its activity was almost completely lost on treatment with trypsin but not on treatment with 2 mM mercaptoethanol at 22°C for 30 min. CM from SMC was applied to the column of Sepharose CL-6B gel, the migration activity was eluted in a fraction corresponding to a molecular weight 500 kDa (Fig. 4).
Fig. 5. Dose-dependence of anti-PDGF antibody on the migration of SMC. Human polyclonal anti-PDGF antibody was added to CM from rabbit (0) or rat (W) SMC. CM from rat Me (0) or human PDGF (0) and then migration activities were assayed in the 6 chambers as described in Materials and Methods. The figure represents the mean&SD of the relative value to the migration activity without antibody as percentage.
Inhibition of migration activities of CM by an antiPDGF antibody A human polyclonal anti-PDGF antibody inhibited the migration activity of human PDGF dose-dependently. The antibody inhibited the migration activity of PDGF almost completely at 50 pg/ml. This antibody dose-dependently inhibited also that of CM from rat M$ which contains rat PDGF [6]. But this antibody did not inhibit the migration activity of CM from rat SMC or rabbit SMC (Fig. 5). These results indicate that this polyclonal antibody crossreacts with rat PDGF and that the migration activity of CM from rat SMC and rabbit SMC is not due to PDGF.
SMC [12] and to have chemotactic activity for various cells [13,14]. Fibronectin has a molecular weight of approximately 450 kDa which is similar to that of SDMF. To examine whether the migration factor in CM from SMC is due to fibronectin or not, the migration activity of rat fibronectin was compared with CM from rat SMC (Fig. 6). Fibronectin induced the migration of SMC dosedependently at a dose of 5-100 pg/ml. However, its maximal activity was 1.58 PDGF units observed at a dose of 40 pg/ml, which was about one fourth of that of SDMF in the same experiment. Next. the dose effects of human polyclonal anti-fibronectin antiserum on the migration activity of CM from SMC were compared with that of rat fibronectin and that of PDGF (Fig. 7). This antiserum inhibited the migration activity of fibronectin dose-dependently. The antiserum at
Relationship of the migration factor in CM from SMC with fibronectin Fibronectin is also reported to be produced by
224 ‘Gi % o r 8 f 2 = Q)
300
250
z .= > ‘3
200
:
150
C 0 0 6
100
fibronectin
._ I
fibronsctin
50
0
I 0.0
,
3
1.0
2.0
Anti-fibronectin
io
b I’0
4b
fibronectin
r
I
0510
I
lb0 (dml)
1
20 SMC-CM
1
50 (%I
Fig. 6. Dose-dependence of CM from SMC and fibronectin on the migration of SMC. SMC suspension was placed in the upper chamber and DME containing each concentration of CM from rat SMC and rat fibronectin was in the lower chamber, respectively. The migration activities of CM (0) and fibronectin (A) at each concentration were assayed in duplicate as described in Materials and Methods. Each point in the figure represents the mean value of the migration activity as the migrated cells in 10 high power fields. A repeat experiment gave similar results.
the dose which inhibited 87% of the migration activity of fibronectin inhibited only 20% of that of CM from rat SMC as well as 13% of that of PDGF and these slight inhibitions were not dosedependent. These results indicate that this antiserum crossreacts with rat fibronectin and that the major part of the migration activity of CM from rat SMC is not due to fibronectin. Discussion In this work, we reported that cultured SMC from rat and rabbit aortic media at the 2nd to the 12th passage secreted a migration factor(s) for
2.5
3.0
Antiserum (unit)
Fig. 7. Dose-dependence of anti-fibronectin antiserum on the migration of SMC. Human polyclonal anti-fibronectin antiserum was added to CM from rat SMC (o), rat fibronectin (0) and human PDGF (A) and the migration activities of them were assayed in the 6 chambers as described in Materials and Methods. The figure represents the mean+ SD of the relative value to the migration activity without antiserum as a percentage.
SMC of rat and rabbit. This factor, SMC derived migration factor (SDMF), stimulated the migration of SMC dose-dependently and acted as a chemotactic factor, but not as a chemokinetic one. We characterized SDMF physicochemically to determine its relation with other migration factors. SDMF was not inactivated by heat treatment at 56°C for 30 min, but was inactivated by heat treatment at 100°C for 10 min. It was not inactivated at pH 2.2 or 10.0 at 22°C for 30 min, by dialysis. On treatment with 2 mM mercaptoethanol, its activity was not lost, suggesting that S-S bonds are not important for its migration activity. The result that SDMF was inactivated on treatment with trypsin indicates that SDMF is a polypeptide. Molecular weight was approximately 500 kDa. Our previous study has shown that cultured SMC secreted a new growth factor named SDGF [7]. SDGF has a molecular weight of 8.7 kDa which is lower than that of SDMF and its growth
225 activity is lost on treatment of 2 mM mercaptoethanol, suggesting that SDMF is different from SDGF. Several factors that enhance the migration of SMC have been reported, such as 12-L-hydroxy5,8,10.14-eicosatetraenoic acid (12-HETE) [15]. leukotriene B4 (LTB,) [16]. interleukin-1 (IL-l) [17], elastin peptide [ll], PDGF [3] and fibronectin (Fig. 6). 12-HETE and LTB, have low molecular weights and are dialyzable. IL-1 has a molecular weight of 12-20 kDa [18] and elastin peptide, formed by digestion of elastin has a molecular weight of less than 25 kDa [19]. Therefore, as these four factors differ from SDMF in molecular weights, it is unlikely that SDMF is one of these factors. PDGF is reported to be secreted by SMC [20] and to have chemotactic activity for SMC [3]. However, the molecular weight (32 kDa) is much lower than that of SDMF, and it is not inactivated by heat treatment at 100°C but is inactivated by mercaptoethanol [21]. These physicochemical properties differ from those of SDMF shown in Table 1. Furthermore, the migration activity of SDMF was not inhibited by human polyclonal anti-PDGF antibody, which inhibited that of purified human PDGF and that of PDGF derived from rat Me. These results indicate that this polyclonal antibody crossreacts with rat PDGF. Thus at least the migration activity of CM from rat SMC is not due to PDGF. Fibronectin has a molecular weight of approximately 450 kDa, which is also similar to that of SDMF. Fibronectin is also reported to be produced by SMC [12] and to have chemotactic activity for SMC (Fig. 6). However, its maximal activity was one fourth of that of SDMF. Moreover. the migration activity of rat SDMF was not inhibited by human anti-fibronectin rabbit antiserum, which strongly inhibited the activity of purified rat fibronectin. These results indicate that this antiserum crossreacts with rat fibronectin and that the migration activity of CM from rat SMC is not due to fibronectin. The above results indicate that SDMF is a new migration factor. distinct from various factors reported previously. This work showed that cultured SMC secrete a potent, new migration factor for SMC, which we
named SDMF. In atheromatous lesions, SDMF may enhance the migration of the medial SMC to the intima. This autocrine system of SMC migration may contribute to the development of intimal thickening in arteriosclerotic lesions. Acknowledgement
This work was supported in part by a grant from the Ministry of Education, Japan (No. 01570351). References
1 Ross. R.. The pathogenesis 2
3
4
5
6
7
8
9
10
of atherosclerosis - an update. N. Engl. J. Med.. 314 (1986) 488. Clowes. A.W.. and Schwartz. SM.. Significance of quiescent smooth muscle migration in the inJured rat carotid artery. Cir. Res.. 56 (1985) 139. Grotendorst. G.R., Seppa. H.E.J., Kleinman. H.K. and Martin. G.R.. Attachment of smooth muscle cells to collagen and their migration toward platelet-derived growth factor. Proc. Natl. Acad. Sci. USA. 78 (1981) 3669. Heldin. C.H.. Westermark. B. and Wasteson. A.. Plateletderived growth factor: purrfication and partial characterization, Proc. Natl. Acad. Sci. USA. 76 (1979) 3722. DiCorleto. P. and Bowen-Pope. D.. Cultured endothehal cells produce a platelet-derived growth factor-like protein. Proc. Natl. Acad. SCI. USA. 80 (1983) 1919. Shimokado. K.. Raines. E.W.. Madtes. D.K.. Barrett, T.B.. Bonditt, E.P. and Ross, R.. A significant part of macrophage-derived growth factor consists of at least two forms of PDGF, Cell. 43 (1985) 277. Morisaki. N.. Kanzaki, T.. Koshikawa. T.. Sarto. Y. and Yoshida. S.. Secretion of a new growth factor, smooth muscle cell derived growth factor. dtstinct from platelet derived growth factor by cultured rabbit aorttc smooth muscle cells. FEBS Lett., 230 (198X) 186. Fischer-Dzoga. K.. Jones. R.M.. Vesselinovttch, D. and Wissler. R.W.. Ultrastructural and immunohistochemical studies of primary cultures of aortrc medial cells. Exp. Mol. Pathol.. 18 (1973) 162. Mortsaki. N., Koyama, N., Mori. S.. Kanzaki. T.. Koshikawa, T., Saito. Y. and Yoshida. S.. Effects of smooth muscle cell derived growth factor (SDGF) in combination with other growth factors on smooth muscle cells, Atherosclerosis. 78 (1989) 61. Goldstein. J.L., Ho. Y.K.. Basu. SK. and Brown. M.S..
Binding site on macrophage that mediates uptake and degradation of acetylated low density lipoprotein. producing massive cholesterol deposition. Proc. Natl. Acad. Sci. USA, 76 (1979) 333. 11 Ooyama, T.. Fukuda. K.. Oda. H.. Nakamura. H. and Hikita. Y.. Substration-bound elastin peptide Inhibits aortic smooth muscle cell migration in vitro, Arteriosclerosis. 7 (1987) 593.
226 12 Holderbaum, D. and Ehrhart, L.A., Substratum influence on collagen and fibronectin biosynthesis by arterial smooth muscle cell in vitro, J. Cell. Physiol., 126 (1986) 216. 13 Postlethwaite, A.E., Keski-Oja, J., Balian, G. and Kang, A.H., Induction of fibroblast chemotaxis by fibronectin, J. Exp. Med., 153 (1981) 494. 14 Bowersox, J.C. and Sorgente, N., Chemotaxis of aortic endothelial cells in response to fibronectin, Cancer Res., 42 (1982) 2547. 15 Nakao, J., Ooyama, T., Ito, H., Chang, W.C. and Murota, S., Comparative effect of lipoxygenase products of arachidonic acid on rat aortic smooth muscle cell migration, Atherosclerosis, 44 (1982) 339. 16 Hirosumi, J., Nomoto, A., Ohkubo, Y., Sekigucbi, C., Mutoh, S., Yamaguchi, I. and Aoki, H., Inflammatory responses in cuff-induced atherosclerosis in rabbits, Atherosclerosis, 64 (1987) 243.
17 Nomoto, A., Mutoh, S., Hagihara, H. and Yamaguchi, I., Smooth muscle cell migration induced by inflammatory cell products and its inhibition by a potent calcium antagonist, nilvadipine, Atherosclerosis, 72 (1988) 213. 18 Oppenheim, J.J., Kovacs, E.J. and Matsushima, K., There is more than one interleukin 1, Immunol. Today, 7 (1986) 45. 19 Senior, R.M., Griffin, G.L. and Mecham, R.P., Chemotactic activity of elastin-derived peptides, J. Clin. Invest., 66 (1980) 859. 20 Nilsson, J., Sjolund, M., Palmberg, L., Thyberg, J. and Heldin, C.H., Arterial smooth muscle cells in primary culture produce a platelet-derived growth factor-like protein, Proc. Natl. Acad. Sci. USA, 82 (1985) 4418. 21 Raines, E.W. and Ross, R., Platelet-derived growth factor, I. High yield purification and evidence for multiple forms, J. Biol. Chem., 257 (1982) 5154.
