JOURNAL OF CELLULAR PHYSIOLOGY 150:510-516 (1992)
Decreased Response to Epidermal Growth Factor During Cellular Senescence in Cultured Human Microvascular Endothelial Cells TAKA0 MATSUDA,' KAZUKI OKAMURA, YASUFUMI SATO, AKlO MORIMOTO, MAYUMI ONO, KlMlTOSHl KOHNO, AND MlCHlHlKO KUWANO Departments of Biochemistry (T.M., K.O., A.M., M.O., K.K., M.K.) and of Medicine (Y.S.), Oita Medical School, Hasamamachi, Oita 879-55, lapan We have previously demonstrated that epidermal growth factor (EGF) induces cell migration, tissue-type plasminogen activator synthesis, as well as tubular formation in microvascular endothelial cells from human omental tissue. In this study, we compared the responsiveness to EGF of late passaged (senescent) human omental microvascular endothelial (HOME) cells with that of early passaged (young) HOME cells. We have employed HOME cells derived from surgically resected omental samples from 14 patients. EGF stimulated cell migration significantly more in the young cells than in the senescent cells during serial cultivation (aging) in vitro. Scatchard analysis demonstrated that the number for both high and low affinity receptors for EGF in HOME cells was decreased dramatically during serial cultivation. The expression of EGF receptor mRNA was also decreased in the senescent HOME cells. Treatment of HOME cells with EGF significantly increased cellular mRNA levels of tissue-type plasminogen activator, and two protooncogenes, c-for and c-myc, in young HOME cells, but not in senescent HOME cells. Thus HOME cells aged in vitro show a decreased responsiveness to EGF, resulting in decreased migration of human endothelial cells. The serial cultivation of human endothelial cells in vitro may downregulate EGF receptor and decrease responsiveness to exogenous EGF, a potent angiogenic factor.
Vascular endothelial cells form the luminal surface vitro appears to be genetically programmed (Orgel, of blood vessels and play a n important role in maintain- 1973; Martin et al., 1974). One may argue that the loss ing the integrity of the vascular wall. In response to of proliferative capacity is due to a diminished response injury of large vessels, adjacent endothelial cells to growth factors or to increased sensitivity to inhibipromptly migrate and proliferate for re-endotheliza- tory factors. Maier et al. (1990) have proposed that setion. In contrast, in response to the wound, capillary nescent human endothelial cells increase expression of endothelial cells migrate, proliferate, and form a new interleukin-la, a potent inhibitor of endothelial cell capillary network (angiogenesis) for wound healing. proliferation. Shimada et al. (1990) showed a n age-deThese processes are facilitated by a variety of growth pendent change in the responses of human umbilical vein endothelial cells in vitro to a growth-inhibitory factors (Folkman and Klagsbrum, 1986). Epiermal growth factor (EGF) is a polypeptide factor, TNF. In the present study, we examined whether in vitro growth factor composed of 53 amino acids, which induces angiogenesis in vivo (Gospodarowicz et al., 1979; senescence of HOME cells and their responsiveness to Schreiber et al., 1986). We have previously demon- exogenous EGF are correlated. Our results show that strated that EGF induces human microvascular endo- senescent HOME cells become less responsive to EGF thelial (HOME) cells from omental tissue to form the and express reduced numbers of cell surface EGF recepvessel-like structures in type I collagen gels (Mawatari tor molecules. The observed decrease in cell migration, et al., 1989). EGF enhances cell migration and tissue- t-PA synthesis, and c-fos or c-myc expression will be type plasminogen activator (t-PA)synthesis (Mawatari possibly correlated with decreased responsiveness to et al., 1991);moreover, addition of tumor necrosis factor EGF during serial cultivation of endothelial cells. (TNF) inhibits EGF-induced migration of HOME cells MATERIALS AND METHODS as well as EGF-induced enhancement of t-PA gene exChemicals and DNA probe pression (Mawatari et al., 1991). These results suggest Epidermal growth factor (EGF) was purchased from a possible linkage between the initial step for angiogenToyobo Corp., Osaka, Japan. pLSX plasmid encoding esis and expression of t-PA. Normal human diploid WI-38 cells in culture exhibit a finite proliferation capacity (Hayflick and Moorehead, 1961; Hayflick, 1965). Age-dependent loss of the Received June 3,1991; accepted September 30,1991. proliferative potential during cellular senescence in *To whom reprint requestsicorrespondence should be addressed. 0 1992 WILEY-LISS. INC.
AGING IN VITRO OF ENDOTHELIAL CELLS
511
Fig. 1. Microscopic observation of HOME cells in culture on type I collagen coated dishes. Cell morphology of 69 F HOME cells at 2-passage (young) (A and B) and 7-passage (senescent) (C and D). Magnifications: 40x (A, C) and lOOx (€3, D).
the entire human EGF receptor was obtained from Dr. A. Ullrich (Max-Planck Institut fur Biophysik, Frankfurt am Main, Germany) (Livneth et al., 1986; Ono et al., 1991). Human t-PA cDNA from Dr. W-D. Schleuning (Schering Aktiengeselshaft Pharma Forshung, Berlin, Germany), pc-fos-1 (human c-fos) from the Japanese Cancer Research Resources Bank (JCRB), and pc5-8 (human c-myel from Dr. K. Yokoyama (Tsukuba Life Science Center, Inst. Phys. Chem. Res) were used in this study.
cases of benign diseases such as gastric ulcer and cholecystolithiasis.
Cell migration assay Cell migration assay was performed as described previously (Sat0 and Rifkin, 1988; Mawatari et al., 1991). Briefly, a confluent monolayer of HOME cells grown on a 35 mm dish was wounded by a razor blade and incubated in M-199 medium containing 0.1% bovine serum albumin for 24 h a t 37°C. After incubation, the migrated cell number was determined by counting the Cells cells in successive 100-pm sections from the wound Human omental microvascular endothelial (HOME) edge. The effect of EGF on HOME cell migration was detercells were isolated according to Kern’s method (Kern et al., 1983) and grown in Medium 199 (Nissui Pharma- mined by counting the number of migrated cells within cential Co., Tokyo, Japan) containing 10% fetal bovine a lane of 1,000 x 100 pm in 100-pm intervals (see Figs. serum (FBS) as described previously (Mawatari et al., 2 and 3) in 5-6 fields. The results were presented as 1989, 1991). HOME cells were cultivated on collagen- mean cell number per each field, and each result varied coated dishes (CORNING) and subcultivated at 1:3 less than 15% from the mean value. split ratio. In this study, we have mainly used HOME lZ5I-EGFbinding assay cells derived from two patients with stomach cancer, a 69-year-old female (69F) and a 22-year-old female 1251-EGFbinding assay was performed as described (22F). Both HOME cells had been serially cultured. previously (Hamanaka et al., 1990; Ono et al., 1991). Two-passaged and 7-passaged cells were used a s young Briefly, HOME cells were grown to be confluent and senescent cells, respectively. We have also exam- (2 x lo5 cells) in a 24-well, collagen-coated plate ined HOME cells derived from other 12 patients. These (CORNING). HOME cells a t 2 x lo5 per well were incu14 patients contained 10 cases of gastric cancer and 4 bated with 400,000 cpm of lZ5I-EGFin MEM contain-
512
MATSUDA ET AL.
ing 20 mM HEPES and 0.5% BSA at 4 “C for 2 hr. A 100-fold excess of unlabeled EGF was added to determine specific binding. After the incubation, cells were washed with ice-cold PBS 3 times, and cell associated radioactivity was counted.
Northern blot analysis Northern blot analysis was carried out as described (Mizoguchi et al., 1990; Mawatari et al., 1991). Briefly, HOME cells were incubated with EGF (10 ng/ml) for indicated times. After the incubation, total cellular RNA was extracted, electrophoresed in 1%agarose gel containing 2.2 M formaldehyde, and blotted to NYTRAN filter (Schleicher & Schnell, Inc). The filter was hybridized with 32P-labeledprobe, and autoradiography was carried out.
RESULTS Cell migration of young and senescent HOME cells in response to EGF HOME cells were subcultivated at a 1:3 split ratio. Without adding any growth factor, the cells could be passaged about 8 times under our culture conditions. Thus the finite lifespan of the cells was estimated to be 15-25 population doublings. As shown in Figure 1A,B, HOME cells exhibit a cobblestonelike appearance with homologous cell size when the 2-passaged (“young”), 69F HOME cells reached confluence, and the cell density was 73.7 ? 7.0 x lo3 cells/cm2. After serial subcultivation, the 7-passaged (“senescent”)69 F HOME cells showed heterogenous morphology, several giant polynuclear cells appeared (Fig. lC,D), and the cell density a t confluent stage was decreased to 39.3 +- 5.5 x lo3 cells/cm2. Similar morphological changes were observed when 22 F HOME cells a t 2-passage and 7-passage were examined (data not shown). We have previously reported that the migration of HOME cells (3-5 passages) is significantly stimulated by EGF (Mawatari et al., 1991). In this work, we compared the effect of EGF on the migration of young and senescent cells. Confluent monolayers of “young” and “senescent” HOME cells were wounded with a razor blade and further incubated in medium in the absence or presence of 10 ng/ml EGF for 24 h r at 37°C (Fig. 2). Both young and senescent 69 F HOME cells migrated into denuded areas in collagen-coated plastic dishes. Cell migration of young and senescent cells was stimulated over control 2.27-fold and 1.53-fold, respectively, in the presence of EGF (Fig. 2). Similar results were obtained with 22F HOME cells (data not shown). As seen in Figure 3, migration of both young 69 F and 22 F HOME cells increased as a function of EGF concentration: 69 F HOME cells and 22 F HOME cells migrated 2.27-fold and 1.80-fold, respectively, in the presence of 10 ng/ml EGF. In contrast, the migration of both senescent 69 F and 22 F HOME cells was stimulated 1.53and 1.48-fold, respectively, in the presence of 10 ng/ml EGF. Both young and senescent cells migrated a t a higher rate with EGF. However, the augmentation of cell migration appeared to be higher in young than in senescent cells. We further confirmed the patterns of migration of young and senescent HOME cells in the presence of 10ng/ml EGF in other omental samples (Fig. 4). We
examined HOME cells derived from several individuals young and senescent HOME cells of the same donor were obtained from 4 individuals (29 F, 40 M, 58 F, and 79 F), and the remaining cell samples were obtained from different individuals. Both young and senescent cells showed increased migration in the presence of EGF. The difference in migration values (1.92 and 1.36, respectively) is significant (P < 0.01). Young (1-3 passages) HOME cells showed increased migration in response to EGF a s compared to senescent (4-8 passages) HOME cells. It is also apparent in Figure 4 that the differences in EGF-stimulated migration is unrelated to the age of the donor. The differences were also unrelated to donor’s diseases whether they were malignant or benign (data not shown).
Binding kinetics of EGF receptor To determine the basis of the altered migration in response to EGF during serial cultivation of HOME cells, we compared the EGF binding capacity of young and senescent HOME cells. Scatchard analysis from the saturation kinetics of lZ5I-EGFbinding to the cell surface of young and senescent HOME cells showed high and low affinity EGF binding sites (Fig. 5). Table 1 summarizes receptor numbers and affinities for both high and low EGF receptors. In both 69F and 22F HOME cells a dramatic decrease in the high and low affinity populations of EGF receptors was observed during in vitro senescence. Similar results were also observed in cells from two other female patients (58 years and 79 years) (data not shown). Cellular responses of young and senescent HOME cells to EGF HOME cells at 3-5 passages showed a 3- to 4-fold increase of t-PA mRNA levels 6 to 12 h r after EGF addition (Mawatari et al., 1991). Furthermore, serum or endothelial cell growth supplement could induce expression of c-fos and c-myc protooncogene mRNA in human umbilical vein endothelial cells (Lampugnani et al., 1990). We examined whether cellular expression of t-PA, c-fos and c-myc genes were changed in response to EGF during the serial cultivation of HOME cells. As seen in Figure 6A,B, there is a 4-fold increase of t-PA mRNA in young 69 F and 22 F HOME cells after 9-hr incubation with EGF, whereas only a very slight increase of t-PA mRNA levels was observed in the 7-passaged senescent HOME cells. The time kinetics of EGF induction of both c-fos and c-myc mRNA in young 69F HOME cells shows a marked increase (more than 10fold) of the c-fos mRNA level a t 30-60 min and a 2-fold increase of c-myc mRNA level at 120 min (Fig. 7). In comparison, the level of c-fos and c-myc mRNA in unstimulated senescent cells was much lower than in young cells, and the level of c-fos and c-myc transcripts in response to EGF was much lower. In this figure, we also demonstrated a dramatic difference in the steadystate levels of EGF receptor mRNA between young and senescent HOME cells, consistent with the Scatchard data. No significant down regulation of EGF receptor gene expression was detected after exposure of HOME cells to EGF for short periods of time.
AGING IN VITRO OF ENDOTHELIAL CELLS
513
,
-
A -60
- 50 - 40 - 30 - 20
-
-
- 10 -0
-
B -60
-
-
T
7
I-
c
t-
-20
5
-10
E
10
z
3
460 i50
- 40 - 30 - 20
-
-
- 10 0
c
D 460
t, 0
- 50 - 40 - 30
SB:
1
2
3
4
I
IU
5
Distance Fig. 2. Wound migration assay of young and senescent HOME cells. Confluent 69 F HOME cells (A, B) 2-passage and (C, D) 7-passage were scraped with a razor blade, washed with PBS, and incubated for 24 hr in the absence (A, C) or presence (B, D) of 10 ngiml EGF. They were fixed with methanol, stained with Giemsa, and photographed.
6
7
8
9
1
0
10
( x ldprn)
The arrows indicates the wound edge. Migration was quantitated by counting the number of cells within 1,000 x 100 Fm area in 5-6 fields. The results are presented as mean number of cells per field. Bars tSD.
MATSUDA ET AL.
514 400-
'
,
69F
350.
,
,
22,F
,
T
Fig. 3. The effect of EGF on migration of young and senescent HOME cells. Cell migration of both 69 F and 22 F HOME cells a t 2-passage (Y) and 7-passage (S)was compared in the presence of EGF. Cell migration was determined a s described in Figure 2. Each value represents the cell number average within 1,000 x 100 pm area in 5-6 fields. Bars f SD.
500
T
51 400
B /
Fig. 4. Cell migration of young and senescent HOME cells from different donors. Wound migration assays were performed at early (A) and late passages (B) in the absence and presence of 10 ngiml EGF and quantitated. Among 8 samples, HOME cells at early (A) and late (B) passage were derived from the same samples of 4 individuals (29 F, 40 M, 58 F, and 79 F) and remaining samples for HOME cells at (A) were different from those at (B). Bars f SD. Samples from 36F, 41F, 69M, 65M, 36M, 71M, 72M, and 81M were not enough to assay both young and senescent cells.
DISCUSSION Migration, plasminogen activator synthesis, and proliferation of endothelial cells are essential for both reendothelization of large vessels after denudation and angiogenesis of small vessels during wound healing (Pepper et al., 1987; Lindner et al., 1989). Wound heal-
ing is impaired with age. The injury of arterial endothelial cells is considered to be the first step in atherosclerosis (Ross, 1986), and the incidence of this disease increases also with aging. Therefore, it is of interest to understand cellular senescence of endothelial cells. EGF is a potent mitogen for endothelial cells in vitro (McAuslan et al., 1985) and induces angiogenesis in vivo (Gospodarowicz et al., 1979, Schreiber et al., 1986). An EGF-like peptide is present in platelets (Oka and Orth, 1983) and also at the sites of tissue regeneration (Grotendorst, 1989). We have previously reported that EGF stimulates migration, t-PA synthesis, and proliferation of HOME cells (Mawatari et al., 1989,1991). In this study we investigated the responsiveness of young and senescent HOME cells to EGF in vitro. Migration, t-PA synthesis, c - m y and c-fos expression of HOME cells in response to EGF diminished during cellular senescence in vitro. Interestingly, Seshadri and Campisi (1990) have shown the repression of c-fos transcription in response to serum in senescent, late-passaged human fibroblasts. Senescent and no-proliferating human fibroblasts are blocked a t the GllS boundary (Olashaw et al., 1983; Rittling et al., 19861, suggesting that the loss of some cycle-dependent genes is a feature of senescence. EGF enhanced significantly expression of the t-PA gene in young cells, but poorly in senescent cells (Fig. 6). Incubation of HOME cells with anti-t-PA antibody inhibited the EGF-mediated cell migration of HOME cells (Mawatari et al., 1991). The loss of EGF-responsiveness in the senescent HOME cells correlates with diminished production of t-PA, resulting in decreased migration of the senescent cells. Similarly, Sato and Rifkin (1988) demonstrated that basic fibroblast growth factor (bFGF) stimulates both cell migration and expression of urokinase-type plasminogen activator in bovine endothelial cells. Sat0 and Rifkin (1989) have further reported that transforming growth factorbeta inhibits both migration and plasminogen activator in the bovine endothelial cells. These results suggest that plasminogen activator plays a n important role in modulation of endothelial cell migration in culture. Senescent fibroblasts become progressively less responsive to mitogenic signals in culture. By using a human fetal fibroblast cell line WI-38, Phillips et al. (1983) have demonstrated that the cells become less responsive to EGF during serial cultivation, although EGF binding capacity remains stable throughout the replicative lifespan of WI-38 cells. In these cells, a marked reduction of tyrosine specific autophosphorylation of the EGF receptor in the senescent WI-38 cells is observed (Brooks et al., 1987). By contrast, expression of EGF receptor was dramatically decreased during serial culture of HOME cells. Our results thus suggest that senescent HOME cells may lose EGF responsiveness by a different mechanism; decreased expression of EGF receptor molecules. The contrast between endothelial cells and fibroblasts is striking and may indicate a t least two distinct pathways to senescence. However, it remains to be determined whether a decrease in EGF receptor number or its mRNA in the late passage HOME cells is the cause or effect of cellular senescence. Experiments evaluating these events in HOME cells are currently in progress.
AGING IN VITRO OF ENDOTHELIAL CELLS
515
‘;J
0 7
X
LL \
m
0
100
200
300
LOO
Bound ( pg / lo5 cells ) Fig. 5. Comparison of Scatchard analysis of ““I-EGF binding between young and senescent HOME cells. lZ5I-EGFbinding to 69 F (A) and 22 F (B) HOME cells at young, 2-passaged ( 0 ) and senescent, 7-passaged ( 0 )states at 4°C as measured, and results were expressed as specific binding after subtracting nonspecific binding in the presence of excess amount of unlabeled EGF.
TABLE 1. Determination of dissociation constants and EGF receptor number in young and senescent HOME cell Donor 69F 22F
Aging’
Y S Y S
High Kd(M)’ Nolcell 3.1 x lo-’’ 1.2 x lo-’’ 1.8 x lo-’’ 8.1 x lo-’’
112,000 33,000 91,000 23,000
Low Kd(M) No/cell 1.4 x 2.8 x 2.7 x 6.5 x
lo-’ lo-’’ lo-’ lo-’’
399,000 45,000 409,000 90,000
‘Y (young) and S (senescent)show 1-3 passaged and4-8passaged cell during serial cultivation. PDissociation constants (Kd) and number (No) are determined from Scatchard analysis as seen in Figure 5.
Fig. 7. Time kinetics of c-fos, c-myc, and EGF receptor (EGF-R) mRNA synthesis in 69 F HOME cells treated with EGF. Confluent 69 F HOME cells at 2-passage (young, Y) and 7-passage (senescent, S) were incubated for indicated times (0, 30, 60, 120 min) with 10 ng/ml EGF and total cellular RNAs were hybridized with 3”P-labeled c-fos, c-myc and EGF receptor cDNA probes; 28 S rRNAs, which were loaded on the gels, were presented. Fig. 6. Comparison of t-PA mRNA synthesis in the presence of EGF between young and senescent HOME cells. Confluent 69 F and 22 F HOME cells at 2-passage (young: Y) and 7-passage (senescent: S) were incubated for 24 hr with 10 n g h l EGF, and total RNA (15 pg) was fractioned on a 1%agarose gel and transferred to a Nylon filter. Northern blot hybridization was performed with 3ZP-labeled t.PA cDNA. In the lower panel 28 S rRNAs which were loaded on the gel were presented after ethidium bromide staining.
Shimada et al. (1990) have reported that the growthinhibitory effect of tumor necrosis factor (TNF) on human umbilical vein endothelial cells is enhanced during aging in vitro’ In this system> no significant change in the TNF receptors between the young and SX?nescent cells is observed. TNF inhibits the EGF-stimulated mi-
516
MATSUDA ET AL.
gration and t-PA production in HOME cells (Mawatari et al., 1991). We have recently reported that TNF enhances expression of bFGF gene in HOME cells (Okamura et al., 1991), but bFGF does not efficiently enhance both cell migration and t-PA production in young HOME cells (unpublished data). It remains to be known if TNF can differentially inhibit both migration and t-PA production in young and senescent HOME cells and by which mechanism. Maier et al. (1990) have demonstrated that senescent human umbilical vein endothelial cells contain high amounts of interleukin-la (IL-la) transcripts. IL-la is a potent inhibitor of endothelial cell proliferation in vitro. IL-la down-regulates the receptor for basic fibroblast growth factor, another potent mitogen for endothelial cells (Cozzolinoet al., 1990). The attempt is now underway to clarify whether senescent HOME cells produce increasing amounts of IL-1 a,and whether IL-1 a affects the synthesis of EGF receptor in HOME cells during aging in vitro. Our present study with HOME cells will propose a useful system to understand molecular mechanisms for “senescence” in culture as well as angiogenesis. Further study should be required to determine how HOME cells loses the EGF responsiveness during serial cultivation.
ACKNOWLEDGMENTS We thank J. Gopas a t NCI-FCRDC for fruitful discussion and critical reading of this manuscript. We thank M. Kobayashi, Y. Uchida, and their colleagues at Oita Medical School for supplying omental tissue samples. LITERATURE CITED Brooks, K.M., Phillips, P.D., Carlin, C.R., Knowles, B.B., and Cristofalo, V.J. (1987) EGF-dependent phosphorylation of the EGF receptor in plasma membranes. J . Cell Physiol., 133.523-531. Cozzolino, F., Torcia, M., Aldinucci, D., Ziche, M., Almerigogna, F., Bani, D., and Syern, D.M. (1990)Interleukin 1is an autocrine regulator of human endothelial cell erowth. Proc. Natl. Acad. Sci. U.S.A., 87t6487-6491. Folkman, J . and Klagsbrum, M. (1986) Angiogenic Factors Science, 235t442447. Gospodarowicz, D., Bialecki, H., and Thakral, T. (1979) The angiogic activity of the fibroblast and epidermal growth factor. Exp. Eye Res., 28:501-514. Grotendorst. G.R.. Soma. Y.. Takehara. K., and Charette, M. (1989) EGF and TGF-a are potent chemoattractants for endothelial cells and EGF-like peptides are present at sites of tissue regeneration. J. Cell. Physiol., 139t617-623. Hamanaka, R., Ono, M., Kuratomi, Y., Mizoguchi, H., Hirai, R., Kohno, K., and Kuwano, M., (1990)Epidermal growth factor (EGF)nonresponsive variant of normal rat kidney cell line: Response to EGF and transforming growth factor-p. Exp. Cell Res., 186:83-89. Hayflick, L. (1965) The limited in vitro lifetime of human diploid cell strain. Exp. Cell Res., 37t614-636. Hayflick, L. and Moorehead, P.S. (1961) The serial cultivation of human diploid cell strains. Exp. Cell Res., 25585-621. Kern, P., Knedler, A,, and Eckel, R.H. (1983) Isolation and culture of human microvascular endothelial cell strains. J. Clin. Invest., 71:1822-1829. Lampugnani, M.G., Polentarutti, N., Pedenovi, M., Mantovani, A,, Dejana, E., and Colotta, F. (1990) c-fos and c-myc expression in human endothelial cells as a function of different culture conditions. Exp. Cell Res., 186t381-384. Lindner, V., Ready, M.A., and Fingerle, J. (1989) Regrowth of arterial endothelium denudation with minimal trauma leads to complete endothelial cell growth. Lab. Invest., 61556-563.
-
~~
Livneth, E., Prywea, R., Kashles, O., Reiss, N., Sasson, I., Mory, Y., Ullrich, A,, and Schlessinger, J . (1986) Reconstitution of human epidermal growth factor receptor and its deletion mutants in cultured hamster cells. J . Biol. Chm., 261t1249O-12497. Maier, J.A.M., Hla, T., and Maciag, T. (1990) Cycrooxygenase is an immediate gene induced by interleukin-1 in human endothelial cells. J. Biol. Chem., 265t10805-10808. Martin, G.M., Sprague, C.A., Norwood, T.H., and Penderglass, W.R. (1974)Colon selection; attenuation and differentiation in an in vitro model of hyperplasia. Am. J . Pathol., 74:137-150. Mawatari, M., Kohno, K., Mizoguchi, H., Matsuda, T., Asoh, K., Van Damme, J., Welgus, H.G., Kuwano, M. (1989) Effect of tumor necrosis factor and epidermal growth factor on cell morphology, cell surface receptor, and the production of tissue inhibitor of metalloproteinases and IL-6 in human microvascular endothelial cells. J . Immunol., 143:1619-1627. Mawatari, M., Okamura, K., Matsuda, T., Hamanaka, R., Higashio, K., Kohno, K., and Kuwano, M. (1991) Tumor necrosis factor and epidermal growth factor modulate migration of human microvascular endothelial cells and production of tissue-type plasminogen activator and its inhibitor. Exp. Cell Res., 192574-580. McAuslan, B.R., Bender, V., Reilly, W., and Moss, B.A. (1985) New functions of epidermal growth factor: stimulation of capillary endothelial cell mieration and matrix deuendent oroliferation. Cell Biol. Int. Rep., 9;17%-183. Mizoguchi, H., Uchiumi, T., Ono, M., Kohno, K., and Kuwano, M. (1990) Enhanced production of tissue-type plasminogen activator by estradiol in a novel type variant of human breast cancer MCF-7 cell line. Biochim. Biophys. Acta, 1052:475A82. Oka, Y. and Orth, D.N. (1983)Human plasma epidermal growth factor p-urogastron is associated with blood platelets. J . Clin. Invest., 72t249-259. Okamura, K., Sato, Y., Matsuda, T., Hamanaka, R., Ono, M., Kohno, K., and Kuwano, M. (1991) Endogenous basic fibroblast growth factor-deoendent induction of collaeenase and interleukin-6 in tumor necrosis factor-treated human &crovascular endothelial cells. J . Biol. Chem., 266:19162-19165. Olashaw, N.E., Kress, E.D., and Cristofalo, V.J. (1983) Thvmidine triphosphate synthesis in senescent WI-38 cells. Exp. Cell Res., 149547454, Ono, M., Kuwano, M., and Kung, H.F. (1991) Malignant transformation of mouse Balbi3T3 cells by polyoma middle T antigen requires epidermal growth factor receptor expression. Cell Growth & Diff., 2:317-322. Orgel, L.E. (1973) Aging of clones of mammalian cells. Nature, 243t441-445. Pepper, M.S., Vassalli, J.D., Montessano, R., and Orci, L. (1987) Urokinase-type plasminogen activator is induced in migrating capillary endothelial cells. J . Cell Biol., 105t2535-2541. Phillips. P.D.. Kuhnle. E.. and Cristofalo. V.J. (1983) P511EGF binding-ability is stable throughout the replicative lifespan of WI-38 cells. J . Cell. Physiol., 114:311-316. Rittling, S.R., Brooks, K.M., Cristofalo, V.J., and Baseraa, R. (1986) Expression of cell cycle-dependent genes in young and senescent WI-38 fibroblast. Proc. Natl. Sci. U.S.A., 83t3316-3320. Ross, R. (1986) The pathogenesis of atherosclerosisan update. New Engl. J. Med., 314t488-500. Sato, Y. and Rifkin, D.B. (1988) Autocrine activities of basic fibroblast growth factor: regulation of endothelial cell movement, plasminogen activator synthesis, and DNA synthesis. J . Cell Biol., 107:11991205. Sato, Y., and Rifkin, D.B. (1989) Inhibition of endothelial cell movement by pericytes and smooth muscle cells: Activation of latent TGF-Pl by plasmin during coculture. J. Cell Biol., 109:309-315. Schreiber, A.B., Winkler, M.E., and Deryn ck, R. (1986)Transforming growth factor-a: A more potent angiogenic mediator than epidermal growth factor. Science,232t1250-1253. Seshardri, T., and Campisi, J. (1990) Repression of c-fos transcription and an altered genetic program in senescent fibroblasts. Science, 247t205-209. Shimada, Y., Kaji, K., Ito, H., Noda, K., and Matsuo, M. (1990) Growth-inhibiting effect of tumor necrosis factor on human umbilical vein endothelial cells is enhanced with advancing age in vitro. J. Cell Physiol., 142t3138.
Decreased Response to Epidermal Growth Factor During Cellular Senescence in Cultured Human Microvascular Endothelial Cells TAKA0 MATSUDA,' KAZUKI OKAMURA, YASUFUMI SATO, AKlO MORIMOTO, MAYUMI ONO, KlMlTOSHl KOHNO, AND MlCHlHlKO KUWANO Departments of Biochemistry (T.M., K.O., A.M., M.O., K.K., M.K.) and of Medicine (Y.S.), Oita Medical School, Hasamamachi, Oita 879-55, lapan We have previously demonstrated that epidermal growth factor (EGF) induces cell migration, tissue-type plasminogen activator synthesis, as well as tubular formation in microvascular endothelial cells from human omental tissue. In this study, we compared the responsiveness to EGF of late passaged (senescent) human omental microvascular endothelial (HOME) cells with that of early passaged (young) HOME cells. We have employed HOME cells derived from surgically resected omental samples from 14 patients. EGF stimulated cell migration significantly more in the young cells than in the senescent cells during serial cultivation (aging) in vitro. Scatchard analysis demonstrated that the number for both high and low affinity receptors for EGF in HOME cells was decreased dramatically during serial cultivation. The expression of EGF receptor mRNA was also decreased in the senescent HOME cells. Treatment of HOME cells with EGF significantly increased cellular mRNA levels of tissue-type plasminogen activator, and two protooncogenes, c-for and c-myc, in young HOME cells, but not in senescent HOME cells. Thus HOME cells aged in vitro show a decreased responsiveness to EGF, resulting in decreased migration of human endothelial cells. The serial cultivation of human endothelial cells in vitro may downregulate EGF receptor and decrease responsiveness to exogenous EGF, a potent angiogenic factor.
Vascular endothelial cells form the luminal surface vitro appears to be genetically programmed (Orgel, of blood vessels and play a n important role in maintain- 1973; Martin et al., 1974). One may argue that the loss ing the integrity of the vascular wall. In response to of proliferative capacity is due to a diminished response injury of large vessels, adjacent endothelial cells to growth factors or to increased sensitivity to inhibipromptly migrate and proliferate for re-endotheliza- tory factors. Maier et al. (1990) have proposed that setion. In contrast, in response to the wound, capillary nescent human endothelial cells increase expression of endothelial cells migrate, proliferate, and form a new interleukin-la, a potent inhibitor of endothelial cell capillary network (angiogenesis) for wound healing. proliferation. Shimada et al. (1990) showed a n age-deThese processes are facilitated by a variety of growth pendent change in the responses of human umbilical vein endothelial cells in vitro to a growth-inhibitory factors (Folkman and Klagsbrum, 1986). Epiermal growth factor (EGF) is a polypeptide factor, TNF. In the present study, we examined whether in vitro growth factor composed of 53 amino acids, which induces angiogenesis in vivo (Gospodarowicz et al., 1979; senescence of HOME cells and their responsiveness to Schreiber et al., 1986). We have previously demon- exogenous EGF are correlated. Our results show that strated that EGF induces human microvascular endo- senescent HOME cells become less responsive to EGF thelial (HOME) cells from omental tissue to form the and express reduced numbers of cell surface EGF recepvessel-like structures in type I collagen gels (Mawatari tor molecules. The observed decrease in cell migration, et al., 1989). EGF enhances cell migration and tissue- t-PA synthesis, and c-fos or c-myc expression will be type plasminogen activator (t-PA)synthesis (Mawatari possibly correlated with decreased responsiveness to et al., 1991);moreover, addition of tumor necrosis factor EGF during serial cultivation of endothelial cells. (TNF) inhibits EGF-induced migration of HOME cells MATERIALS AND METHODS as well as EGF-induced enhancement of t-PA gene exChemicals and DNA probe pression (Mawatari et al., 1991). These results suggest Epidermal growth factor (EGF) was purchased from a possible linkage between the initial step for angiogenToyobo Corp., Osaka, Japan. pLSX plasmid encoding esis and expression of t-PA. Normal human diploid WI-38 cells in culture exhibit a finite proliferation capacity (Hayflick and Moorehead, 1961; Hayflick, 1965). Age-dependent loss of the Received June 3,1991; accepted September 30,1991. proliferative potential during cellular senescence in *To whom reprint requestsicorrespondence should be addressed. 0 1992 WILEY-LISS. INC.
AGING IN VITRO OF ENDOTHELIAL CELLS
511
Fig. 1. Microscopic observation of HOME cells in culture on type I collagen coated dishes. Cell morphology of 69 F HOME cells at 2-passage (young) (A and B) and 7-passage (senescent) (C and D). Magnifications: 40x (A, C) and lOOx (€3, D).
the entire human EGF receptor was obtained from Dr. A. Ullrich (Max-Planck Institut fur Biophysik, Frankfurt am Main, Germany) (Livneth et al., 1986; Ono et al., 1991). Human t-PA cDNA from Dr. W-D. Schleuning (Schering Aktiengeselshaft Pharma Forshung, Berlin, Germany), pc-fos-1 (human c-fos) from the Japanese Cancer Research Resources Bank (JCRB), and pc5-8 (human c-myel from Dr. K. Yokoyama (Tsukuba Life Science Center, Inst. Phys. Chem. Res) were used in this study.
cases of benign diseases such as gastric ulcer and cholecystolithiasis.
Cell migration assay Cell migration assay was performed as described previously (Sat0 and Rifkin, 1988; Mawatari et al., 1991). Briefly, a confluent monolayer of HOME cells grown on a 35 mm dish was wounded by a razor blade and incubated in M-199 medium containing 0.1% bovine serum albumin for 24 h a t 37°C. After incubation, the migrated cell number was determined by counting the Cells cells in successive 100-pm sections from the wound Human omental microvascular endothelial (HOME) edge. The effect of EGF on HOME cell migration was detercells were isolated according to Kern’s method (Kern et al., 1983) and grown in Medium 199 (Nissui Pharma- mined by counting the number of migrated cells within cential Co., Tokyo, Japan) containing 10% fetal bovine a lane of 1,000 x 100 pm in 100-pm intervals (see Figs. serum (FBS) as described previously (Mawatari et al., 2 and 3) in 5-6 fields. The results were presented as 1989, 1991). HOME cells were cultivated on collagen- mean cell number per each field, and each result varied coated dishes (CORNING) and subcultivated at 1:3 less than 15% from the mean value. split ratio. In this study, we have mainly used HOME lZ5I-EGFbinding assay cells derived from two patients with stomach cancer, a 69-year-old female (69F) and a 22-year-old female 1251-EGFbinding assay was performed as described (22F). Both HOME cells had been serially cultured. previously (Hamanaka et al., 1990; Ono et al., 1991). Two-passaged and 7-passaged cells were used a s young Briefly, HOME cells were grown to be confluent and senescent cells, respectively. We have also exam- (2 x lo5 cells) in a 24-well, collagen-coated plate ined HOME cells derived from other 12 patients. These (CORNING). HOME cells a t 2 x lo5 per well were incu14 patients contained 10 cases of gastric cancer and 4 bated with 400,000 cpm of lZ5I-EGFin MEM contain-
512
MATSUDA ET AL.
ing 20 mM HEPES and 0.5% BSA at 4 “C for 2 hr. A 100-fold excess of unlabeled EGF was added to determine specific binding. After the incubation, cells were washed with ice-cold PBS 3 times, and cell associated radioactivity was counted.
Northern blot analysis Northern blot analysis was carried out as described (Mizoguchi et al., 1990; Mawatari et al., 1991). Briefly, HOME cells were incubated with EGF (10 ng/ml) for indicated times. After the incubation, total cellular RNA was extracted, electrophoresed in 1%agarose gel containing 2.2 M formaldehyde, and blotted to NYTRAN filter (Schleicher & Schnell, Inc). The filter was hybridized with 32P-labeledprobe, and autoradiography was carried out.
RESULTS Cell migration of young and senescent HOME cells in response to EGF HOME cells were subcultivated at a 1:3 split ratio. Without adding any growth factor, the cells could be passaged about 8 times under our culture conditions. Thus the finite lifespan of the cells was estimated to be 15-25 population doublings. As shown in Figure 1A,B, HOME cells exhibit a cobblestonelike appearance with homologous cell size when the 2-passaged (“young”), 69F HOME cells reached confluence, and the cell density was 73.7 ? 7.0 x lo3 cells/cm2. After serial subcultivation, the 7-passaged (“senescent”)69 F HOME cells showed heterogenous morphology, several giant polynuclear cells appeared (Fig. lC,D), and the cell density a t confluent stage was decreased to 39.3 +- 5.5 x lo3 cells/cm2. Similar morphological changes were observed when 22 F HOME cells a t 2-passage and 7-passage were examined (data not shown). We have previously reported that the migration of HOME cells (3-5 passages) is significantly stimulated by EGF (Mawatari et al., 1991). In this work, we compared the effect of EGF on the migration of young and senescent cells. Confluent monolayers of “young” and “senescent” HOME cells were wounded with a razor blade and further incubated in medium in the absence or presence of 10 ng/ml EGF for 24 h r at 37°C (Fig. 2). Both young and senescent 69 F HOME cells migrated into denuded areas in collagen-coated plastic dishes. Cell migration of young and senescent cells was stimulated over control 2.27-fold and 1.53-fold, respectively, in the presence of EGF (Fig. 2). Similar results were obtained with 22F HOME cells (data not shown). As seen in Figure 3, migration of both young 69 F and 22 F HOME cells increased as a function of EGF concentration: 69 F HOME cells and 22 F HOME cells migrated 2.27-fold and 1.80-fold, respectively, in the presence of 10 ng/ml EGF. In contrast, the migration of both senescent 69 F and 22 F HOME cells was stimulated 1.53and 1.48-fold, respectively, in the presence of 10 ng/ml EGF. Both young and senescent cells migrated a t a higher rate with EGF. However, the augmentation of cell migration appeared to be higher in young than in senescent cells. We further confirmed the patterns of migration of young and senescent HOME cells in the presence of 10ng/ml EGF in other omental samples (Fig. 4). We
examined HOME cells derived from several individuals young and senescent HOME cells of the same donor were obtained from 4 individuals (29 F, 40 M, 58 F, and 79 F), and the remaining cell samples were obtained from different individuals. Both young and senescent cells showed increased migration in the presence of EGF. The difference in migration values (1.92 and 1.36, respectively) is significant (P < 0.01). Young (1-3 passages) HOME cells showed increased migration in response to EGF a s compared to senescent (4-8 passages) HOME cells. It is also apparent in Figure 4 that the differences in EGF-stimulated migration is unrelated to the age of the donor. The differences were also unrelated to donor’s diseases whether they were malignant or benign (data not shown).
Binding kinetics of EGF receptor To determine the basis of the altered migration in response to EGF during serial cultivation of HOME cells, we compared the EGF binding capacity of young and senescent HOME cells. Scatchard analysis from the saturation kinetics of lZ5I-EGFbinding to the cell surface of young and senescent HOME cells showed high and low affinity EGF binding sites (Fig. 5). Table 1 summarizes receptor numbers and affinities for both high and low EGF receptors. In both 69F and 22F HOME cells a dramatic decrease in the high and low affinity populations of EGF receptors was observed during in vitro senescence. Similar results were also observed in cells from two other female patients (58 years and 79 years) (data not shown). Cellular responses of young and senescent HOME cells to EGF HOME cells at 3-5 passages showed a 3- to 4-fold increase of t-PA mRNA levels 6 to 12 h r after EGF addition (Mawatari et al., 1991). Furthermore, serum or endothelial cell growth supplement could induce expression of c-fos and c-myc protooncogene mRNA in human umbilical vein endothelial cells (Lampugnani et al., 1990). We examined whether cellular expression of t-PA, c-fos and c-myc genes were changed in response to EGF during the serial cultivation of HOME cells. As seen in Figure 6A,B, there is a 4-fold increase of t-PA mRNA in young 69 F and 22 F HOME cells after 9-hr incubation with EGF, whereas only a very slight increase of t-PA mRNA levels was observed in the 7-passaged senescent HOME cells. The time kinetics of EGF induction of both c-fos and c-myc mRNA in young 69F HOME cells shows a marked increase (more than 10fold) of the c-fos mRNA level a t 30-60 min and a 2-fold increase of c-myc mRNA level at 120 min (Fig. 7). In comparison, the level of c-fos and c-myc mRNA in unstimulated senescent cells was much lower than in young cells, and the level of c-fos and c-myc transcripts in response to EGF was much lower. In this figure, we also demonstrated a dramatic difference in the steadystate levels of EGF receptor mRNA between young and senescent HOME cells, consistent with the Scatchard data. No significant down regulation of EGF receptor gene expression was detected after exposure of HOME cells to EGF for short periods of time.
AGING IN VITRO OF ENDOTHELIAL CELLS
513
,
-
A -60
- 50 - 40 - 30 - 20
-
-
- 10 -0
-
B -60
-
-
T
7
I-
c
t-
-20
5
-10
E
10
z
3
460 i50
- 40 - 30 - 20
-
-
- 10 0
c
D 460
t, 0
- 50 - 40 - 30
SB:
1
2
3
4
I
IU
5
Distance Fig. 2. Wound migration assay of young and senescent HOME cells. Confluent 69 F HOME cells (A, B) 2-passage and (C, D) 7-passage were scraped with a razor blade, washed with PBS, and incubated for 24 hr in the absence (A, C) or presence (B, D) of 10 ngiml EGF. They were fixed with methanol, stained with Giemsa, and photographed.
6
7
8
9
1
0
10
( x ldprn)
The arrows indicates the wound edge. Migration was quantitated by counting the number of cells within 1,000 x 100 Fm area in 5-6 fields. The results are presented as mean number of cells per field. Bars tSD.
MATSUDA ET AL.
514 400-
'
,
69F
350.
,
,
22,F
,
T
Fig. 3. The effect of EGF on migration of young and senescent HOME cells. Cell migration of both 69 F and 22 F HOME cells a t 2-passage (Y) and 7-passage (S)was compared in the presence of EGF. Cell migration was determined a s described in Figure 2. Each value represents the cell number average within 1,000 x 100 pm area in 5-6 fields. Bars f SD.
500
T
51 400
B /
Fig. 4. Cell migration of young and senescent HOME cells from different donors. Wound migration assays were performed at early (A) and late passages (B) in the absence and presence of 10 ngiml EGF and quantitated. Among 8 samples, HOME cells at early (A) and late (B) passage were derived from the same samples of 4 individuals (29 F, 40 M, 58 F, and 79 F) and remaining samples for HOME cells at (A) were different from those at (B). Bars f SD. Samples from 36F, 41F, 69M, 65M, 36M, 71M, 72M, and 81M were not enough to assay both young and senescent cells.
DISCUSSION Migration, plasminogen activator synthesis, and proliferation of endothelial cells are essential for both reendothelization of large vessels after denudation and angiogenesis of small vessels during wound healing (Pepper et al., 1987; Lindner et al., 1989). Wound heal-
ing is impaired with age. The injury of arterial endothelial cells is considered to be the first step in atherosclerosis (Ross, 1986), and the incidence of this disease increases also with aging. Therefore, it is of interest to understand cellular senescence of endothelial cells. EGF is a potent mitogen for endothelial cells in vitro (McAuslan et al., 1985) and induces angiogenesis in vivo (Gospodarowicz et al., 1979, Schreiber et al., 1986). An EGF-like peptide is present in platelets (Oka and Orth, 1983) and also at the sites of tissue regeneration (Grotendorst, 1989). We have previously reported that EGF stimulates migration, t-PA synthesis, and proliferation of HOME cells (Mawatari et al., 1989,1991). In this study we investigated the responsiveness of young and senescent HOME cells to EGF in vitro. Migration, t-PA synthesis, c - m y and c-fos expression of HOME cells in response to EGF diminished during cellular senescence in vitro. Interestingly, Seshadri and Campisi (1990) have shown the repression of c-fos transcription in response to serum in senescent, late-passaged human fibroblasts. Senescent and no-proliferating human fibroblasts are blocked a t the GllS boundary (Olashaw et al., 1983; Rittling et al., 19861, suggesting that the loss of some cycle-dependent genes is a feature of senescence. EGF enhanced significantly expression of the t-PA gene in young cells, but poorly in senescent cells (Fig. 6). Incubation of HOME cells with anti-t-PA antibody inhibited the EGF-mediated cell migration of HOME cells (Mawatari et al., 1991). The loss of EGF-responsiveness in the senescent HOME cells correlates with diminished production of t-PA, resulting in decreased migration of the senescent cells. Similarly, Sato and Rifkin (1988) demonstrated that basic fibroblast growth factor (bFGF) stimulates both cell migration and expression of urokinase-type plasminogen activator in bovine endothelial cells. Sat0 and Rifkin (1989) have further reported that transforming growth factorbeta inhibits both migration and plasminogen activator in the bovine endothelial cells. These results suggest that plasminogen activator plays a n important role in modulation of endothelial cell migration in culture. Senescent fibroblasts become progressively less responsive to mitogenic signals in culture. By using a human fetal fibroblast cell line WI-38, Phillips et al. (1983) have demonstrated that the cells become less responsive to EGF during serial cultivation, although EGF binding capacity remains stable throughout the replicative lifespan of WI-38 cells. In these cells, a marked reduction of tyrosine specific autophosphorylation of the EGF receptor in the senescent WI-38 cells is observed (Brooks et al., 1987). By contrast, expression of EGF receptor was dramatically decreased during serial culture of HOME cells. Our results thus suggest that senescent HOME cells may lose EGF responsiveness by a different mechanism; decreased expression of EGF receptor molecules. The contrast between endothelial cells and fibroblasts is striking and may indicate a t least two distinct pathways to senescence. However, it remains to be determined whether a decrease in EGF receptor number or its mRNA in the late passage HOME cells is the cause or effect of cellular senescence. Experiments evaluating these events in HOME cells are currently in progress.
AGING IN VITRO OF ENDOTHELIAL CELLS
515
‘;J
0 7
X
LL \
m
0
100
200
300
LOO
Bound ( pg / lo5 cells ) Fig. 5. Comparison of Scatchard analysis of ““I-EGF binding between young and senescent HOME cells. lZ5I-EGFbinding to 69 F (A) and 22 F (B) HOME cells at young, 2-passaged ( 0 ) and senescent, 7-passaged ( 0 )states at 4°C as measured, and results were expressed as specific binding after subtracting nonspecific binding in the presence of excess amount of unlabeled EGF.
TABLE 1. Determination of dissociation constants and EGF receptor number in young and senescent HOME cell Donor 69F 22F
Aging’
Y S Y S
High Kd(M)’ Nolcell 3.1 x lo-’’ 1.2 x lo-’’ 1.8 x lo-’’ 8.1 x lo-’’
112,000 33,000 91,000 23,000
Low Kd(M) No/cell 1.4 x 2.8 x 2.7 x 6.5 x
lo-’ lo-’’ lo-’ lo-’’
399,000 45,000 409,000 90,000
‘Y (young) and S (senescent)show 1-3 passaged and4-8passaged cell during serial cultivation. PDissociation constants (Kd) and number (No) are determined from Scatchard analysis as seen in Figure 5.
Fig. 7. Time kinetics of c-fos, c-myc, and EGF receptor (EGF-R) mRNA synthesis in 69 F HOME cells treated with EGF. Confluent 69 F HOME cells at 2-passage (young, Y) and 7-passage (senescent, S) were incubated for indicated times (0, 30, 60, 120 min) with 10 ng/ml EGF and total cellular RNAs were hybridized with 3”P-labeled c-fos, c-myc and EGF receptor cDNA probes; 28 S rRNAs, which were loaded on the gels, were presented. Fig. 6. Comparison of t-PA mRNA synthesis in the presence of EGF between young and senescent HOME cells. Confluent 69 F and 22 F HOME cells at 2-passage (young: Y) and 7-passage (senescent: S) were incubated for 24 hr with 10 n g h l EGF, and total RNA (15 pg) was fractioned on a 1%agarose gel and transferred to a Nylon filter. Northern blot hybridization was performed with 3ZP-labeled t.PA cDNA. In the lower panel 28 S rRNAs which were loaded on the gel were presented after ethidium bromide staining.
Shimada et al. (1990) have reported that the growthinhibitory effect of tumor necrosis factor (TNF) on human umbilical vein endothelial cells is enhanced during aging in vitro’ In this system> no significant change in the TNF receptors between the young and SX?nescent cells is observed. TNF inhibits the EGF-stimulated mi-
516
MATSUDA ET AL.
gration and t-PA production in HOME cells (Mawatari et al., 1991). We have recently reported that TNF enhances expression of bFGF gene in HOME cells (Okamura et al., 1991), but bFGF does not efficiently enhance both cell migration and t-PA production in young HOME cells (unpublished data). It remains to be known if TNF can differentially inhibit both migration and t-PA production in young and senescent HOME cells and by which mechanism. Maier et al. (1990) have demonstrated that senescent human umbilical vein endothelial cells contain high amounts of interleukin-la (IL-la) transcripts. IL-la is a potent inhibitor of endothelial cell proliferation in vitro. IL-la down-regulates the receptor for basic fibroblast growth factor, another potent mitogen for endothelial cells (Cozzolinoet al., 1990). The attempt is now underway to clarify whether senescent HOME cells produce increasing amounts of IL-1 a,and whether IL-1 a affects the synthesis of EGF receptor in HOME cells during aging in vitro. Our present study with HOME cells will propose a useful system to understand molecular mechanisms for “senescence” in culture as well as angiogenesis. Further study should be required to determine how HOME cells loses the EGF responsiveness during serial cultivation.
ACKNOWLEDGMENTS We thank J. Gopas a t NCI-FCRDC for fruitful discussion and critical reading of this manuscript. We thank M. Kobayashi, Y. Uchida, and their colleagues at Oita Medical School for supplying omental tissue samples. LITERATURE CITED Brooks, K.M., Phillips, P.D., Carlin, C.R., Knowles, B.B., and Cristofalo, V.J. (1987) EGF-dependent phosphorylation of the EGF receptor in plasma membranes. J . Cell Physiol., 133.523-531. Cozzolino, F., Torcia, M., Aldinucci, D., Ziche, M., Almerigogna, F., Bani, D., and Syern, D.M. (1990)Interleukin 1is an autocrine regulator of human endothelial cell erowth. Proc. Natl. Acad. Sci. U.S.A., 87t6487-6491. Folkman, J . and Klagsbrum, M. (1986) Angiogenic Factors Science, 235t442447. Gospodarowicz, D., Bialecki, H., and Thakral, T. (1979) The angiogic activity of the fibroblast and epidermal growth factor. Exp. Eye Res., 28:501-514. Grotendorst. G.R.. Soma. Y.. Takehara. K., and Charette, M. (1989) EGF and TGF-a are potent chemoattractants for endothelial cells and EGF-like peptides are present at sites of tissue regeneration. J. Cell. Physiol., 139t617-623. Hamanaka, R., Ono, M., Kuratomi, Y., Mizoguchi, H., Hirai, R., Kohno, K., and Kuwano, M., (1990)Epidermal growth factor (EGF)nonresponsive variant of normal rat kidney cell line: Response to EGF and transforming growth factor-p. Exp. Cell Res., 186:83-89. Hayflick, L. (1965) The limited in vitro lifetime of human diploid cell strain. Exp. Cell Res., 37t614-636. Hayflick, L. and Moorehead, P.S. (1961) The serial cultivation of human diploid cell strains. Exp. Cell Res., 25585-621. Kern, P., Knedler, A,, and Eckel, R.H. (1983) Isolation and culture of human microvascular endothelial cell strains. J. Clin. Invest., 71:1822-1829. Lampugnani, M.G., Polentarutti, N., Pedenovi, M., Mantovani, A,, Dejana, E., and Colotta, F. (1990) c-fos and c-myc expression in human endothelial cells as a function of different culture conditions. Exp. Cell Res., 186t381-384. Lindner, V., Ready, M.A., and Fingerle, J. (1989) Regrowth of arterial endothelium denudation with minimal trauma leads to complete endothelial cell growth. Lab. Invest., 61556-563.
-
~~
Livneth, E., Prywea, R., Kashles, O., Reiss, N., Sasson, I., Mory, Y., Ullrich, A,, and Schlessinger, J . (1986) Reconstitution of human epidermal growth factor receptor and its deletion mutants in cultured hamster cells. J . Biol. Chm., 261t1249O-12497. Maier, J.A.M., Hla, T., and Maciag, T. (1990) Cycrooxygenase is an immediate gene induced by interleukin-1 in human endothelial cells. J. Biol. Chem., 265t10805-10808. Martin, G.M., Sprague, C.A., Norwood, T.H., and Penderglass, W.R. (1974)Colon selection; attenuation and differentiation in an in vitro model of hyperplasia. Am. J . Pathol., 74:137-150. Mawatari, M., Kohno, K., Mizoguchi, H., Matsuda, T., Asoh, K., Van Damme, J., Welgus, H.G., Kuwano, M. (1989) Effect of tumor necrosis factor and epidermal growth factor on cell morphology, cell surface receptor, and the production of tissue inhibitor of metalloproteinases and IL-6 in human microvascular endothelial cells. J . Immunol., 143:1619-1627. Mawatari, M., Okamura, K., Matsuda, T., Hamanaka, R., Higashio, K., Kohno, K., and Kuwano, M. (1991) Tumor necrosis factor and epidermal growth factor modulate migration of human microvascular endothelial cells and production of tissue-type plasminogen activator and its inhibitor. Exp. Cell Res., 192574-580. McAuslan, B.R., Bender, V., Reilly, W., and Moss, B.A. (1985) New functions of epidermal growth factor: stimulation of capillary endothelial cell mieration and matrix deuendent oroliferation. Cell Biol. Int. Rep., 9;17%-183. Mizoguchi, H., Uchiumi, T., Ono, M., Kohno, K., and Kuwano, M. (1990) Enhanced production of tissue-type plasminogen activator by estradiol in a novel type variant of human breast cancer MCF-7 cell line. Biochim. Biophys. Acta, 1052:475A82. Oka, Y. and Orth, D.N. (1983)Human plasma epidermal growth factor p-urogastron is associated with blood platelets. J . Clin. Invest., 72t249-259. Okamura, K., Sato, Y., Matsuda, T., Hamanaka, R., Ono, M., Kohno, K., and Kuwano, M. (1991) Endogenous basic fibroblast growth factor-deoendent induction of collaeenase and interleukin-6 in tumor necrosis factor-treated human &crovascular endothelial cells. J . Biol. Chem., 266:19162-19165. Olashaw, N.E., Kress, E.D., and Cristofalo, V.J. (1983) Thvmidine triphosphate synthesis in senescent WI-38 cells. Exp. Cell Res., 149547454, Ono, M., Kuwano, M., and Kung, H.F. (1991) Malignant transformation of mouse Balbi3T3 cells by polyoma middle T antigen requires epidermal growth factor receptor expression. Cell Growth & Diff., 2:317-322. Orgel, L.E. (1973) Aging of clones of mammalian cells. Nature, 243t441-445. Pepper, M.S., Vassalli, J.D., Montessano, R., and Orci, L. (1987) Urokinase-type plasminogen activator is induced in migrating capillary endothelial cells. J . Cell Biol., 105t2535-2541. Phillips. P.D.. Kuhnle. E.. and Cristofalo. V.J. (1983) P511EGF binding-ability is stable throughout the replicative lifespan of WI-38 cells. J . Cell. Physiol., 114:311-316. Rittling, S.R., Brooks, K.M., Cristofalo, V.J., and Baseraa, R. (1986) Expression of cell cycle-dependent genes in young and senescent WI-38 fibroblast. Proc. Natl. Sci. U.S.A., 83t3316-3320. Ross, R. (1986) The pathogenesis of atherosclerosisan update. New Engl. J. Med., 314t488-500. Sato, Y. and Rifkin, D.B. (1988) Autocrine activities of basic fibroblast growth factor: regulation of endothelial cell movement, plasminogen activator synthesis, and DNA synthesis. J . Cell Biol., 107:11991205. Sato, Y., and Rifkin, D.B. (1989) Inhibition of endothelial cell movement by pericytes and smooth muscle cells: Activation of latent TGF-Pl by plasmin during coculture. J. Cell Biol., 109:309-315. Schreiber, A.B., Winkler, M.E., and Deryn ck, R. (1986)Transforming growth factor-a: A more potent angiogenic mediator than epidermal growth factor. Science,232t1250-1253. Seshardri, T., and Campisi, J. (1990) Repression of c-fos transcription and an altered genetic program in senescent fibroblasts. Science, 247t205-209. Shimada, Y., Kaji, K., Ito, H., Noda, K., and Matsuo, M. (1990) Growth-inhibiting effect of tumor necrosis factor on human umbilical vein endothelial cells is enhanced with advancing age in vitro. J. Cell Physiol., 142t3138.
