Mechanisms of Ageing and Development, 66 (i 992) 107-114
107
Elsevier ScientificPublishers Ireland Ltd.
EPIDERMAL GROWTH FACTOR SUPPRESSES IN VITRO SENESCENCE IN THE ABILITY OF HUMAN UMBILICAL VEIN ENDOTHELIAL CELLS TO PROLIFERATE, BUT NOT IN THE ABILITY TO PRODUCE PROSTACYCLIN
NOBUHIKO HASEGAWA a and KIYOTAKA YAMAMOTO b ayakult Central Institute for Microbiological Research, Tokyo 186 and bDepartment of Cell Biology, Tokyo Metropolitan Institute of Gerontology, Tokyo 173 (Japan)
(Received February 25th, 1992) (Revision receivedJune 9th, 1992) SUMMARY Addition of epidermal growth factor (EGF) to culture medium extended the replicative life span of human umbilical vein endothelial (HUVE) cells in culture. In brief, EGF suppresses the age-related decrease (in vitro senescence) in cell proliferative ability. However, the addition of EGF did not extend the culture period in which prostacyclin (PGI2) is actively produced by the cells. Therefore, EGF does not suppress the age-related decrease (in vitro senescence) in the ability of cultured HUVE cells to produce PGI2. These results suggest that the process of in vitro senescence in the cell proliferative ability is not necessarily correlated with that of in vitro senescence in the ability to produce PGI2.
Key words: In vitro senescence; Endothelial cells; Epidermal growth factor; Suppres-
sion of in vitro senescence; Cell proliferation; PGI2 production INTRODUCTION In 1961, Hayflick [1,2] showed that human diploid fibroblasts have a limited division potential in culture. Ever since, long-term culture of the cells has been used as an experimental model for studying cellular senescence. A large amount of useful and important information has been obtained from studies using cell culture [3,4]. Further, it has been found that vascular endothelial cells also have a limited Correspondence to: Nobuhiko Hasegawa, Yakult Central Institute for MicrobiologicalResearch, 1796 Yaho, Kunitachi-shi,Tokyo 186, Japan.
004%6374/92/$05.00 © 1992 Elsevier ScientificPublishers Ireland Ltd. Printed and Published in Ireland
108
replicative life span in culture [5-11]. There have been many investigations into long-term culture of endothelial cells. We also found an age-related decrease in prostacyclin (PGI2) production by human umbilical vein endothelial (HUVE) cells in culture [12] and no relationship between the age-related decrease in PGI2 production and intracellular lipid peroxide of these cells [13]. In 1983, Thornton et al. [10] reported that addition of heparin plus endothelial cell growth factor (ECGF) to the medium stimulated the growth of cultured HUVE cells and extended the replicative life span. Hence, medium containing heparin and ECGF has been widely used to maintain endothelial cells. However, the co-addition of heparin and ECGF greatly inhibits PGI2 production by HUVE cells as we previously reported [14]. PGI2 has important biological activities such as non-thrombogenicity and vasodilation. PGI2 production is one of the main functions specific to endothelial cells and is performed in the vessel wall mainly by endothelial cells. Masotti et al. [15] and Tokunaga et al. [16] reported an age-related decrease in PGI 2 production in human vessel wall and by human aortic endothelial cells in vivo. On the basis of these facts, the agerelated decrease in PGI2 production by endothelial cells is thought to be one of the main causes of age-related vascular diseases (for example, atherosclerosis). Therefore, we have been using a medium without heparin to study the age-related decrease in PGI2 production by HUVE cells in culture. Under this condition, ECGF was the only defined growth factor added to the medium and HUVE cells had a replicative life span of about 30 population doubling levels (PDL) [12]. Recently, we started to subculture HUVE cells in a medium further supplemented with epidermal growth factor (EGF). Now, the HUVE cells have a replicative life span of about 40 PDL. These findings suggest that the addition of EGF extends the in vitro replicative life span of HUVE cells. Therefore, we checked the possibility that the addition of EGF can suppress the age-related decrease (in vitro senescence) in the ability of HUVE cells to proliferate in culture. If EGF has a strong inhibitory effect on PGI2 production, as heparin does, the culture system with EGF cannot be a useful experimental model to study PGI 2 production. Therefore, we determined the effect of the addition of EGF on PGI~ production by HUVE cells. We think suppression of the age-related decrease in PGI 2 production can make it possible to prevent age-related vascular diseases. Therefore, we further investigated whether the addition of EGF can simultaneously suppress the age-related decrease (in vitro senescence) in PGI2 production by HUVE cells in culture. MATERIALSAND METHODS Cell culture We cultured two strains of endothelial cells (HUVE-17,-8) which we obtained from two human umbilical veins. For HUVE-17 cells, the growth medium was minimal essential medium (MEM) containing 20% heat-inactivated fetal bovine
109 serum (hi-FBS), 100 or 200 gg of ECGF (culture grade; Boehringer Mannheim) per ml and antibiotics (AB: 100 units of penicillin G and 100 #g of streptomycin per ml). For HUVE-8 cells, the growth medium was MEM containing 20% hi-FBS, 150 #g of ECGF per ml, 10 ng of EGF (human, recombinant, Boehringer Mannheim) per ml and AB. These HUVE cells were grown at 37°C in collagen-coated 60-ram tissue culture dishes (Vitrogen 100, collagen type I; Collagen Corp.) in a water-saturated atmosphere of 5% CO2 in air. The growth medium was renewed every 2 or 3 days. Cultures were split 1:4 at every passage. The HUVE-17 strain has a limited replicative life span of about 30 PDL [12] and the HUVE-8 strain has a span of about 55 PDL, under the above culture conditions. Since these two strains were positive for Factor VIII-related antigen during in vitro ageing, we identified them as strains of endothelial cells [17].
Determination of the effect of EGF on the replicative life span of HUVE cells in culture We started to add 10 ng of EGF per ml to the culture of HUVE-17 cells at about 19 PDL. The cells were subcultured in the presence of EGF in collagen-coated 16mm tissue culture wells until the cells stopped growing. For HUVE-8 cells, we stopped the addition of EGF at about 31 or 22 PDL and subcultured the cells in the absence of EGF until the limit of the replicative life span was reached. The cells were counted with a hemocytometer and the PDL was calculated at each passage.
Determination of prostacyclin production We determined PGI2 production by the HUVE cells as follows: HUVE-8 cells were grown to confluency in the presence or absence of 10 ng of EGF per ml and used at about 25 PDL, or were subcultured in the presence of EGF and grown to confluency at each passage. These cells were incubated at 37°C for 2 days after replacement of the growth medium with fresh medium. Then the medium was removed and stored and PGI2 secreted into the medium was measured by radioimmunoassay for 6-keto-prostaglandin Flu (6-keto-PGFlot), a stable metabolite of PGI2, by Amersham's assay system. RESULTS AND DISCUSSION
Replicative life span of HUVE-17 cells subcultured in the presence of EGF We added EGF to a culture of HUVE-17 cells at 19 PDL which had been subcultured without EGF. Then we continued to subculture the cells in the presence of EGF. As shown in Fig. 1, the HUVE-17 cells stopped proliferating at about 40 PDL. Thus, the replicative life span of the cells was extended by about 10 PDL, compared with that of HUVE-17 cells previously subcultured in the absence of EGF (the range of the life span is from 28.3 to 29.1 PDL as shown by the bar in Fig. 1 [12]). These results show that the addition of EGF can extend the in vitro replicative life span of HUVE cells.
110
45 40 35 00
30 ._l E3 EL
A4bO 0
0
25 20 15
to ° ho
10
AO
£
5 0--' 0
40
80
120
160
200
Days Fig. 1. Replicative life span of HUVE- 17 cells in the presence or absence of EGF. We subeultured HUVE17 cells in MEM-20% hi-FBS containing 100 (O) or 200 #g (A) of ECGF per ml in the absence of EGF. In some cultures of HUVE-17 cells in the presence of 100 #g of ECGF per ml, we changed the growth medium to MEM-20% hi-FBS containing 150 #g of ECGF and 10 ng of EGF per ml at about 19 PDL. Then, we continued to subculture the cells in the presence of EGF ( • ) as described in Materials and Methods. The points represent means from 2-3 individual dishes.
55 0
0
0
50
0 0
_J 0 13.
45
0 0 0
40
0
o
Oe
35 30 ~ 0
i
i
I
i
20
40
60
80
I
i
100 120
Days Fig. 2. Replicative life span of HUVE-8 cells in the presence or absence of EGF. We subeultured HUVE-8 cells in MEM-20% hi-FBS containing 150 #g of ECGF and 10 ng of EGF per ml (O). In some cultures, EGF was withheld at about 31 PDL and we continued to subculture the cells in the absence of EGF (O). This subeulturing was carried out as described in Materials and Methods. The points represent means from 2-8 individual dishes.
111
Replicative life span of HUVE-8 cells subcultured in the presence or absence of EGF We next subcultured another strain of HUVE cells, HUVE-8, in medium supplemented with ECGF and EGF from the beginning of culture. At that time, the replicative life span of the cells was about 55 PDL (Fig. 2), which is much higher than about 30 PDL for HUVE-17 cells subcultured in the absence of EGF. Therefore, we assumed that the addition of EGF can extend the replicative life span of HUVE cells. To confirm this assumption, we withheld EGF from the culture at 31 PDL and subcultured the cells in the absence of EGF until they stopped growing. Under these conditions, HUVE-8 cells ceased cell proliferating at about 36 PDL (Fig. 2). We further observed that HUVE-8 cells stopped dividing at about 35 PDL when we stopped adding EGF at about 22 PDL (data not shown). Together with Fig. 1, Fig. 2 indicates that the addition of EGF can extend the replicative life span of HUVE cells. In other words, the addition of EGF can suppress the age-related decrease (in vitro senescence) in the proliferative ability of HUVE cells. Such large extension by the addition of EGF was also observed for human epidermal keratinocytes [18]. For human diploid fibroblasts, however, their replicative life span is not extended by the addition of EGF [19,20], though their growth rate is stimulated by the growth factor [21,22]. This observation for fibroblasts is very interesting to us because of our findings as follows: EGF at concentrations (0.6-1000 ng per ml), which are usually applied for cell culture, hardly stimulated the growth rate of endothelial cells compared with ECGF [14], but 10 ng of EGF per ml can extend the replicative life span of the cells as described above. These results also indicate that the ability to stimulate the cell growth rate is not equalled to the ability to extend the replicative life span. Effect of EGF on PGI2 production by HUVE-8 cells We found that the production of PGI2 is greatly inhibited by heparin [14] which was reported by Thornton et al. to extend the in vitro replicative life span of HUVE cells [10]. Furthermore, several investigators reported that EGF stimulates production of prostaglandins by several kinds of cells other than endothelial cells [23-28]. On the basis of these observations, we investigated whether the addition of EGF has an effect on PGI2 production by HUVE cells. Figure 3 shows that there is no statistically significant difference between the production by HUVE cells at about 25 PDL in the presence and absence of EGF (mean 4- S.D.: 304.1 4- 18.3 ng/105 cells and 305.2 4- 7.7 ng/105 cells, respectively). This finding shows that the addition of EGF has no effect on PGI2 production by HUVE-8 cells. Therefore, we think that we can use HUVE cells cultured in the presence of EGF for our studies on PGI2 production. Effect of in vitro senescence on PGI2 production by HUVE-8 cells Further, we determined PGI2 production by HUVE-8 cells at various PDL subcultured in the presence of EGF. PGI2 production by the cells shows an age-related
112
300 ¢/)
o
200
E
1oo
0
+EGF
-EGF
Fig. 3. Effect of EOF on PGI 2 production by HUVE-8 cells at about 25 PDL. We cultured HUVE-8 cells to confluency in the presence (+EGF: 10 ng of EGF per ml) or absence (-EGF) of EGF. The medium was removed from the cultures and assayed for 6-keto-PGFla as described in Materials and Methods. The data represent means ± S.D. from 4 individual dishes.
decrease between about 23 and 35 PDL (Fig. 4). Then, there is a long period (from 35 PDL to 51 PDL) in which PGI2 production is at a very low level (the production is likewise at a very low level at about 51 PDL; data not shown). For HUVE- 17 cells, PGI2 production also decreases with increasing in vitro age in the absence of EGF
400 o
300
tt') o ¢-
lie.
200
c,i (5 13.
I 100
0
24
'
'
'
'
26
28
30
32
I
~
'1
34
36
~
38
~_
40
i
42
.l_
44
46
PDL Fig. 4. PGI 2 production by HUVE-8 cells at various PDLs. We subcultured HUVE-8 cells in the presence of EGF. The medium was removed from the cultures at each passage and assayed for 6-ketoPGFla as described in Materials and Methods.
113
[12]. At that time, however, there is no period in which PGI 2 production by the cells is barely observed (the level of production is very low only at the last few PDLs). From the point of view of in vitro age, the limit of the ability of the HUVE17 cells to produce PGI2 is consistent with the replicative limit. In the absence of EGF, HUVE-8 cells stop proliferating at about 35 to 36 PDL as stated above. Thus, the replicative limit of HUVE-8 cells in the absence of EGF is almost the same as the limit of PGI2 production as shown in Fig. 4. These findings indicate that the addition of EGF cannot extend the limit of the ability of HUVE cells to produce PGI2. In other words, the addition of EGF cannot suppress the age-related decrease (in vitro senescence) in the cells' ability to produce PGI2. We think that we should investigate the age-related decrease in PGI2 production under not only the serum-stimulation but also more physiological stimulations to understand the mechanism of the decrease further. We examined the effect of EGF on in vitro senescence of HUVE cells, represented by the ability to proliferate and the ability to produce PGI2. The addition of EGF suppresses in vitro senescence in the cell proliferative ability, but not the ability to produce PGI2. Therefore, the two processes of in vitro senescence are not necessarily correlated, though both processes advance with every subculture or cell division. Further, these results indicate that when we investigate a suppressive effect on cellular senescence, we must take account of not only cell proliferation but also specific functions. When we find a suppressive effect on senescence of various cell types, it may lead to prevention of various types of age-related diseases. ACKNOWLEDGMENTS
We thank Dr. H. Ooka (Tokyo Metropolitan Institute of Gerontology), Dr. T. Sato, Dr. A. Hirashima and Dr. T. Sako (Yakult Central Institute for Microbiological Research) for helpful discussion. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. REFERENCES L. Hayflick and P.S. Moorhead, The serial cultivation of human diploid cell strains. Exp. Cell Res., 25 (1961) 585-621. 2 L. Hayflick, The limited in vitro lifetime of human diploid cell strains. Exp. Cell Res., 37 (1965) 614-636. 3 L. Hayflick, Recent advances in the cell biology of aging. Mech. Ageing Dev., 14 (1980) 59-79. 4 B.M. Stanulis-Praeger, Cellular senescence revisited: a review. Mech. Ageing Dee., 38 (1987) 1-48. 5 G.S. Duthu and J.R. Smith, In vitro proliferation and lifespan of bovine aorta endothelial cells: effect of culture conditions an~t fibroblast growth factor. J. Cell. Physiol., 103 (1980) 385-392. 6 S.N. Mu¢ller, E.M. Rosen and ,E.M. Lcvin¢, Cellular senescence in a cloned strain of bovine fetal aortic endothelial cells. Science, 207 (1980) 889-891. 7 T. Maciag, G.A. Hoover, M.B. Stemerman and R. Weinstein, Serial propagation of human endothelial cells in vitro. J. Cell Biol.~ 91 (1981) 420-426. 1
114 8 9
10 11
12
13
14
15
16 17
18 19 20 21 22 23 24 25
26
27 28
E.M. Rosen, S.N. Mueller, J.P. Noveral and E.M. Levine, Proliferative characteristics of clonal endothelial cell strains. J. Cell. PhysioL, 107 (1981) 123-137. L.K. Johnson and J.P. Longenecker, Senescence of aortic endothelial cells in vitro: influence of culture conditions and preliminary characterization of the senescent phenotype. Mech. Ageing Dev., 18 (1982) 1-18. S.C. Thornton, S.N. Mueller and E.M. Levine, Human endothelial cells: use of heparin in cloning and long-term serial cultivation. Science, 222 (1983) 623-625. H. Hoshi and W.L. McKeehan, Isolation, growth requirements, cloning, prostacyclin production and life-span of human adult endothelial cells in low serum culture medium. In Vitro, 22 (1986) 51-56. N. Hasegawa, M. Yamamoto, T. Imamura, Y. Mitsui and K. Yamamoto, Evaluation of long-term cultured endothelial cells as a model system for studying vascular ageing. Mech. Ageing Dev., 46 (1988) 111-123. N. Hasegawa and K. Yamamoto, No relationship between the age-related decrease in prostacyclin production and the level of intracellular lipid peroxidation in human umbilical vein endothelial cells in culture. Mech. Ageing Dev., 60 (1991) 35-42. N. Hasegawa, M. Yamamoto and K. Yamamoto, Stimulation of cell growth and inhibition of prostacyclin production by heparin in human umbilical vein endothelial cells. J. Cell. Physiol., 137 (1988) 603-607. G. Masotti, L. Poggesi, G. Galanti, F. Trotta and G.G.N. Serneri, Prostacyclin production in man. In P.J. Lewis and J. O'Grady (eds.), Clinical Pharmacology of Prostacyclin, Raven Press, New York, 1981, pp. 9-20. O. Tokunaga, T. Yamada, J.L. Fan and T. Watanabe, Age-related decline in prostacyclin synthesis by human aortic endothelial cells. Quantitative analysis. Am. J. Pathol., 138 (1991) 941-949. E.A. Jaffe, R.L. Nachman, C.G. Becker and C.R. Minick, Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J. Clin. Invest., 52 (1973) 2745-2756. J.G. Rheinwald and H. Green, Epidermal growth factor and the multiplication of cultured human epidermal keratinocytes. Nature, 265 (1977) 421-424. K. Kaji and M. Matuo, Responsiveness of human lung diploid fibroblast ageing in vitro to epidermal growth factor: saturation density and lifespan. Mech. Ageing Dev., 22 (1983) 129-133. J.C. Angello, Replicative potential and the duration of the cell cycle in human fibroblasts: coordinate stimulation by epidermal growth factor. Mech. Ageing Dev., 62 (1992) 1-12. G. Carpenter and S. Cohen, Human epidermal growth factor and the proliferation of human fibroblasts. J. Cell. Physiol., 88 (1976) 227-237. J. Huey, A.S. Narayanan, K. Jones and R.C. Page, Effect of epidermal growth factor on the synthetic activity of human fibroblasts. Biochim. Biophys. Acta, 632 (1980) 227-233. L. Levine and A. Hassid, Epidermal growth factor stimulates prostaglandin biosynthesis by canine kidney (MDCK) cells. Biochem. Biophys. Res. Commun., 76 (1977) 1181-1187. A.H. Tashjian, Jr. and L. Levine, Epidermal growth factor stimulates prostaglandin production and bone resorption in cultured mouse calvaria. Biochem. Biophys. Res. Commun., 85 (1978) 966-975. K. Yokota, M. Kusaka, T. Ohshima, S. Yamamoto, N. Kurihara, T. Yoshino and M. Kumegawa, Stimulation of prostaglandin E2 synthesis in cloned osteoblastic cells of mouse (MC3T3-E1) by epidermal growth factor. J. Biol. Chem., 261 (1986) 15410-15415. N. Takasu, S. Sato, T. Yamada and Y. Shimizu, Epidermal growth factor (EGF) and tumor promoter 12-o-tetradecanoylphorbol 13-acetate (TPA) stimulate PG synthesis and thymidine incorporation in cultured porcine thyroid cells. Biochem. Biophys. Res. Commun., 143 (1987) 880-884. M.L. Casey, K. Korte and P.C. MacDonald, Epidermal growth factor stimulation of prostaglandin E2 biosynthesis in amnion cells. J. BioL Chem., 263 (1988) 7846-7854. J.A. Handler, R.M. Danilowicz and T.E. Eling, Mitogenic signaling by epidermal growth factor (EGF), but not platelet-derived growth factor, requires arachidonic acid metabolism in BALB/c 3T3 cells. J. BioL Chem., 265 (1990) 3669-3673.
107
Elsevier ScientificPublishers Ireland Ltd.
EPIDERMAL GROWTH FACTOR SUPPRESSES IN VITRO SENESCENCE IN THE ABILITY OF HUMAN UMBILICAL VEIN ENDOTHELIAL CELLS TO PROLIFERATE, BUT NOT IN THE ABILITY TO PRODUCE PROSTACYCLIN
NOBUHIKO HASEGAWA a and KIYOTAKA YAMAMOTO b ayakult Central Institute for Microbiological Research, Tokyo 186 and bDepartment of Cell Biology, Tokyo Metropolitan Institute of Gerontology, Tokyo 173 (Japan)
(Received February 25th, 1992) (Revision receivedJune 9th, 1992) SUMMARY Addition of epidermal growth factor (EGF) to culture medium extended the replicative life span of human umbilical vein endothelial (HUVE) cells in culture. In brief, EGF suppresses the age-related decrease (in vitro senescence) in cell proliferative ability. However, the addition of EGF did not extend the culture period in which prostacyclin (PGI2) is actively produced by the cells. Therefore, EGF does not suppress the age-related decrease (in vitro senescence) in the ability of cultured HUVE cells to produce PGI2. These results suggest that the process of in vitro senescence in the cell proliferative ability is not necessarily correlated with that of in vitro senescence in the ability to produce PGI2.
Key words: In vitro senescence; Endothelial cells; Epidermal growth factor; Suppres-
sion of in vitro senescence; Cell proliferation; PGI2 production INTRODUCTION In 1961, Hayflick [1,2] showed that human diploid fibroblasts have a limited division potential in culture. Ever since, long-term culture of the cells has been used as an experimental model for studying cellular senescence. A large amount of useful and important information has been obtained from studies using cell culture [3,4]. Further, it has been found that vascular endothelial cells also have a limited Correspondence to: Nobuhiko Hasegawa, Yakult Central Institute for MicrobiologicalResearch, 1796 Yaho, Kunitachi-shi,Tokyo 186, Japan.
004%6374/92/$05.00 © 1992 Elsevier ScientificPublishers Ireland Ltd. Printed and Published in Ireland
108
replicative life span in culture [5-11]. There have been many investigations into long-term culture of endothelial cells. We also found an age-related decrease in prostacyclin (PGI2) production by human umbilical vein endothelial (HUVE) cells in culture [12] and no relationship between the age-related decrease in PGI2 production and intracellular lipid peroxide of these cells [13]. In 1983, Thornton et al. [10] reported that addition of heparin plus endothelial cell growth factor (ECGF) to the medium stimulated the growth of cultured HUVE cells and extended the replicative life span. Hence, medium containing heparin and ECGF has been widely used to maintain endothelial cells. However, the co-addition of heparin and ECGF greatly inhibits PGI2 production by HUVE cells as we previously reported [14]. PGI2 has important biological activities such as non-thrombogenicity and vasodilation. PGI2 production is one of the main functions specific to endothelial cells and is performed in the vessel wall mainly by endothelial cells. Masotti et al. [15] and Tokunaga et al. [16] reported an age-related decrease in PGI 2 production in human vessel wall and by human aortic endothelial cells in vivo. On the basis of these facts, the agerelated decrease in PGI2 production by endothelial cells is thought to be one of the main causes of age-related vascular diseases (for example, atherosclerosis). Therefore, we have been using a medium without heparin to study the age-related decrease in PGI2 production by HUVE cells in culture. Under this condition, ECGF was the only defined growth factor added to the medium and HUVE cells had a replicative life span of about 30 population doubling levels (PDL) [12]. Recently, we started to subculture HUVE cells in a medium further supplemented with epidermal growth factor (EGF). Now, the HUVE cells have a replicative life span of about 40 PDL. These findings suggest that the addition of EGF extends the in vitro replicative life span of HUVE cells. Therefore, we checked the possibility that the addition of EGF can suppress the age-related decrease (in vitro senescence) in the ability of HUVE cells to proliferate in culture. If EGF has a strong inhibitory effect on PGI2 production, as heparin does, the culture system with EGF cannot be a useful experimental model to study PGI 2 production. Therefore, we determined the effect of the addition of EGF on PGI~ production by HUVE cells. We think suppression of the age-related decrease in PGI 2 production can make it possible to prevent age-related vascular diseases. Therefore, we further investigated whether the addition of EGF can simultaneously suppress the age-related decrease (in vitro senescence) in PGI2 production by HUVE cells in culture. MATERIALSAND METHODS Cell culture We cultured two strains of endothelial cells (HUVE-17,-8) which we obtained from two human umbilical veins. For HUVE-17 cells, the growth medium was minimal essential medium (MEM) containing 20% heat-inactivated fetal bovine
109 serum (hi-FBS), 100 or 200 gg of ECGF (culture grade; Boehringer Mannheim) per ml and antibiotics (AB: 100 units of penicillin G and 100 #g of streptomycin per ml). For HUVE-8 cells, the growth medium was MEM containing 20% hi-FBS, 150 #g of ECGF per ml, 10 ng of EGF (human, recombinant, Boehringer Mannheim) per ml and AB. These HUVE cells were grown at 37°C in collagen-coated 60-ram tissue culture dishes (Vitrogen 100, collagen type I; Collagen Corp.) in a water-saturated atmosphere of 5% CO2 in air. The growth medium was renewed every 2 or 3 days. Cultures were split 1:4 at every passage. The HUVE-17 strain has a limited replicative life span of about 30 PDL [12] and the HUVE-8 strain has a span of about 55 PDL, under the above culture conditions. Since these two strains were positive for Factor VIII-related antigen during in vitro ageing, we identified them as strains of endothelial cells [17].
Determination of the effect of EGF on the replicative life span of HUVE cells in culture We started to add 10 ng of EGF per ml to the culture of HUVE-17 cells at about 19 PDL. The cells were subcultured in the presence of EGF in collagen-coated 16mm tissue culture wells until the cells stopped growing. For HUVE-8 cells, we stopped the addition of EGF at about 31 or 22 PDL and subcultured the cells in the absence of EGF until the limit of the replicative life span was reached. The cells were counted with a hemocytometer and the PDL was calculated at each passage.
Determination of prostacyclin production We determined PGI2 production by the HUVE cells as follows: HUVE-8 cells were grown to confluency in the presence or absence of 10 ng of EGF per ml and used at about 25 PDL, or were subcultured in the presence of EGF and grown to confluency at each passage. These cells were incubated at 37°C for 2 days after replacement of the growth medium with fresh medium. Then the medium was removed and stored and PGI2 secreted into the medium was measured by radioimmunoassay for 6-keto-prostaglandin Flu (6-keto-PGFlot), a stable metabolite of PGI2, by Amersham's assay system. RESULTS AND DISCUSSION
Replicative life span of HUVE-17 cells subcultured in the presence of EGF We added EGF to a culture of HUVE-17 cells at 19 PDL which had been subcultured without EGF. Then we continued to subculture the cells in the presence of EGF. As shown in Fig. 1, the HUVE-17 cells stopped proliferating at about 40 PDL. Thus, the replicative life span of the cells was extended by about 10 PDL, compared with that of HUVE-17 cells previously subcultured in the absence of EGF (the range of the life span is from 28.3 to 29.1 PDL as shown by the bar in Fig. 1 [12]). These results show that the addition of EGF can extend the in vitro replicative life span of HUVE cells.
110
45 40 35 00
30 ._l E3 EL
A4bO 0
0
25 20 15
to ° ho
10
AO
£
5 0--' 0
40
80
120
160
200
Days Fig. 1. Replicative life span of HUVE- 17 cells in the presence or absence of EGF. We subeultured HUVE17 cells in MEM-20% hi-FBS containing 100 (O) or 200 #g (A) of ECGF per ml in the absence of EGF. In some cultures of HUVE-17 cells in the presence of 100 #g of ECGF per ml, we changed the growth medium to MEM-20% hi-FBS containing 150 #g of ECGF and 10 ng of EGF per ml at about 19 PDL. Then, we continued to subculture the cells in the presence of EGF ( • ) as described in Materials and Methods. The points represent means from 2-3 individual dishes.
55 0
0
0
50
0 0
_J 0 13.
45
0 0 0
40
0
o
Oe
35 30 ~ 0
i
i
I
i
20
40
60
80
I
i
100 120
Days Fig. 2. Replicative life span of HUVE-8 cells in the presence or absence of EGF. We subeultured HUVE-8 cells in MEM-20% hi-FBS containing 150 #g of ECGF and 10 ng of EGF per ml (O). In some cultures, EGF was withheld at about 31 PDL and we continued to subculture the cells in the absence of EGF (O). This subeulturing was carried out as described in Materials and Methods. The points represent means from 2-8 individual dishes.
111
Replicative life span of HUVE-8 cells subcultured in the presence or absence of EGF We next subcultured another strain of HUVE cells, HUVE-8, in medium supplemented with ECGF and EGF from the beginning of culture. At that time, the replicative life span of the cells was about 55 PDL (Fig. 2), which is much higher than about 30 PDL for HUVE-17 cells subcultured in the absence of EGF. Therefore, we assumed that the addition of EGF can extend the replicative life span of HUVE cells. To confirm this assumption, we withheld EGF from the culture at 31 PDL and subcultured the cells in the absence of EGF until they stopped growing. Under these conditions, HUVE-8 cells ceased cell proliferating at about 36 PDL (Fig. 2). We further observed that HUVE-8 cells stopped dividing at about 35 PDL when we stopped adding EGF at about 22 PDL (data not shown). Together with Fig. 1, Fig. 2 indicates that the addition of EGF can extend the replicative life span of HUVE cells. In other words, the addition of EGF can suppress the age-related decrease (in vitro senescence) in the proliferative ability of HUVE cells. Such large extension by the addition of EGF was also observed for human epidermal keratinocytes [18]. For human diploid fibroblasts, however, their replicative life span is not extended by the addition of EGF [19,20], though their growth rate is stimulated by the growth factor [21,22]. This observation for fibroblasts is very interesting to us because of our findings as follows: EGF at concentrations (0.6-1000 ng per ml), which are usually applied for cell culture, hardly stimulated the growth rate of endothelial cells compared with ECGF [14], but 10 ng of EGF per ml can extend the replicative life span of the cells as described above. These results also indicate that the ability to stimulate the cell growth rate is not equalled to the ability to extend the replicative life span. Effect of EGF on PGI2 production by HUVE-8 cells We found that the production of PGI2 is greatly inhibited by heparin [14] which was reported by Thornton et al. to extend the in vitro replicative life span of HUVE cells [10]. Furthermore, several investigators reported that EGF stimulates production of prostaglandins by several kinds of cells other than endothelial cells [23-28]. On the basis of these observations, we investigated whether the addition of EGF has an effect on PGI2 production by HUVE cells. Figure 3 shows that there is no statistically significant difference between the production by HUVE cells at about 25 PDL in the presence and absence of EGF (mean 4- S.D.: 304.1 4- 18.3 ng/105 cells and 305.2 4- 7.7 ng/105 cells, respectively). This finding shows that the addition of EGF has no effect on PGI2 production by HUVE-8 cells. Therefore, we think that we can use HUVE cells cultured in the presence of EGF for our studies on PGI2 production. Effect of in vitro senescence on PGI2 production by HUVE-8 cells Further, we determined PGI2 production by HUVE-8 cells at various PDL subcultured in the presence of EGF. PGI2 production by the cells shows an age-related
112
300 ¢/)
o
200
E
1oo
0
+EGF
-EGF
Fig. 3. Effect of EOF on PGI 2 production by HUVE-8 cells at about 25 PDL. We cultured HUVE-8 cells to confluency in the presence (+EGF: 10 ng of EGF per ml) or absence (-EGF) of EGF. The medium was removed from the cultures and assayed for 6-keto-PGFla as described in Materials and Methods. The data represent means ± S.D. from 4 individual dishes.
decrease between about 23 and 35 PDL (Fig. 4). Then, there is a long period (from 35 PDL to 51 PDL) in which PGI2 production is at a very low level (the production is likewise at a very low level at about 51 PDL; data not shown). For HUVE- 17 cells, PGI2 production also decreases with increasing in vitro age in the absence of EGF
400 o
300
tt') o ¢-
lie.
200
c,i (5 13.
I 100
0
24
'
'
'
'
26
28
30
32
I
~
'1
34
36
~
38
~_
40
i
42
.l_
44
46
PDL Fig. 4. PGI 2 production by HUVE-8 cells at various PDLs. We subcultured HUVE-8 cells in the presence of EGF. The medium was removed from the cultures at each passage and assayed for 6-ketoPGFla as described in Materials and Methods.
113
[12]. At that time, however, there is no period in which PGI 2 production by the cells is barely observed (the level of production is very low only at the last few PDLs). From the point of view of in vitro age, the limit of the ability of the HUVE17 cells to produce PGI2 is consistent with the replicative limit. In the absence of EGF, HUVE-8 cells stop proliferating at about 35 to 36 PDL as stated above. Thus, the replicative limit of HUVE-8 cells in the absence of EGF is almost the same as the limit of PGI2 production as shown in Fig. 4. These findings indicate that the addition of EGF cannot extend the limit of the ability of HUVE cells to produce PGI2. In other words, the addition of EGF cannot suppress the age-related decrease (in vitro senescence) in the cells' ability to produce PGI2. We think that we should investigate the age-related decrease in PGI2 production under not only the serum-stimulation but also more physiological stimulations to understand the mechanism of the decrease further. We examined the effect of EGF on in vitro senescence of HUVE cells, represented by the ability to proliferate and the ability to produce PGI2. The addition of EGF suppresses in vitro senescence in the cell proliferative ability, but not the ability to produce PGI2. Therefore, the two processes of in vitro senescence are not necessarily correlated, though both processes advance with every subculture or cell division. Further, these results indicate that when we investigate a suppressive effect on cellular senescence, we must take account of not only cell proliferation but also specific functions. When we find a suppressive effect on senescence of various cell types, it may lead to prevention of various types of age-related diseases. ACKNOWLEDGMENTS
We thank Dr. H. Ooka (Tokyo Metropolitan Institute of Gerontology), Dr. T. Sato, Dr. A. Hirashima and Dr. T. Sako (Yakult Central Institute for Microbiological Research) for helpful discussion. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. REFERENCES L. Hayflick and P.S. Moorhead, The serial cultivation of human diploid cell strains. Exp. Cell Res., 25 (1961) 585-621. 2 L. Hayflick, The limited in vitro lifetime of human diploid cell strains. Exp. Cell Res., 37 (1965) 614-636. 3 L. Hayflick, Recent advances in the cell biology of aging. Mech. Ageing Dev., 14 (1980) 59-79. 4 B.M. Stanulis-Praeger, Cellular senescence revisited: a review. Mech. Ageing Dee., 38 (1987) 1-48. 5 G.S. Duthu and J.R. Smith, In vitro proliferation and lifespan of bovine aorta endothelial cells: effect of culture conditions an~t fibroblast growth factor. J. Cell. Physiol., 103 (1980) 385-392. 6 S.N. Mu¢ller, E.M. Rosen and ,E.M. Lcvin¢, Cellular senescence in a cloned strain of bovine fetal aortic endothelial cells. Science, 207 (1980) 889-891. 7 T. Maciag, G.A. Hoover, M.B. Stemerman and R. Weinstein, Serial propagation of human endothelial cells in vitro. J. Cell Biol.~ 91 (1981) 420-426. 1
114 8 9
10 11
12
13
14
15
16 17
18 19 20 21 22 23 24 25
26
27 28
E.M. Rosen, S.N. Mueller, J.P. Noveral and E.M. Levine, Proliferative characteristics of clonal endothelial cell strains. J. Cell. PhysioL, 107 (1981) 123-137. L.K. Johnson and J.P. Longenecker, Senescence of aortic endothelial cells in vitro: influence of culture conditions and preliminary characterization of the senescent phenotype. Mech. Ageing Dev., 18 (1982) 1-18. S.C. Thornton, S.N. Mueller and E.M. Levine, Human endothelial cells: use of heparin in cloning and long-term serial cultivation. Science, 222 (1983) 623-625. H. Hoshi and W.L. McKeehan, Isolation, growth requirements, cloning, prostacyclin production and life-span of human adult endothelial cells in low serum culture medium. In Vitro, 22 (1986) 51-56. N. Hasegawa, M. Yamamoto, T. Imamura, Y. Mitsui and K. Yamamoto, Evaluation of long-term cultured endothelial cells as a model system for studying vascular ageing. Mech. Ageing Dev., 46 (1988) 111-123. N. Hasegawa and K. Yamamoto, No relationship between the age-related decrease in prostacyclin production and the level of intracellular lipid peroxidation in human umbilical vein endothelial cells in culture. Mech. Ageing Dev., 60 (1991) 35-42. N. Hasegawa, M. Yamamoto and K. Yamamoto, Stimulation of cell growth and inhibition of prostacyclin production by heparin in human umbilical vein endothelial cells. J. Cell. Physiol., 137 (1988) 603-607. G. Masotti, L. Poggesi, G. Galanti, F. Trotta and G.G.N. Serneri, Prostacyclin production in man. In P.J. Lewis and J. O'Grady (eds.), Clinical Pharmacology of Prostacyclin, Raven Press, New York, 1981, pp. 9-20. O. Tokunaga, T. Yamada, J.L. Fan and T. Watanabe, Age-related decline in prostacyclin synthesis by human aortic endothelial cells. Quantitative analysis. Am. J. Pathol., 138 (1991) 941-949. E.A. Jaffe, R.L. Nachman, C.G. Becker and C.R. Minick, Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J. Clin. Invest., 52 (1973) 2745-2756. J.G. Rheinwald and H. Green, Epidermal growth factor and the multiplication of cultured human epidermal keratinocytes. Nature, 265 (1977) 421-424. K. Kaji and M. Matuo, Responsiveness of human lung diploid fibroblast ageing in vitro to epidermal growth factor: saturation density and lifespan. Mech. Ageing Dev., 22 (1983) 129-133. J.C. Angello, Replicative potential and the duration of the cell cycle in human fibroblasts: coordinate stimulation by epidermal growth factor. Mech. Ageing Dev., 62 (1992) 1-12. G. Carpenter and S. Cohen, Human epidermal growth factor and the proliferation of human fibroblasts. J. Cell. Physiol., 88 (1976) 227-237. J. Huey, A.S. Narayanan, K. Jones and R.C. Page, Effect of epidermal growth factor on the synthetic activity of human fibroblasts. Biochim. Biophys. Acta, 632 (1980) 227-233. L. Levine and A. Hassid, Epidermal growth factor stimulates prostaglandin biosynthesis by canine kidney (MDCK) cells. Biochem. Biophys. Res. Commun., 76 (1977) 1181-1187. A.H. Tashjian, Jr. and L. Levine, Epidermal growth factor stimulates prostaglandin production and bone resorption in cultured mouse calvaria. Biochem. Biophys. Res. Commun., 85 (1978) 966-975. K. Yokota, M. Kusaka, T. Ohshima, S. Yamamoto, N. Kurihara, T. Yoshino and M. Kumegawa, Stimulation of prostaglandin E2 synthesis in cloned osteoblastic cells of mouse (MC3T3-E1) by epidermal growth factor. J. Biol. Chem., 261 (1986) 15410-15415. N. Takasu, S. Sato, T. Yamada and Y. Shimizu, Epidermal growth factor (EGF) and tumor promoter 12-o-tetradecanoylphorbol 13-acetate (TPA) stimulate PG synthesis and thymidine incorporation in cultured porcine thyroid cells. Biochem. Biophys. Res. Commun., 143 (1987) 880-884. M.L. Casey, K. Korte and P.C. MacDonald, Epidermal growth factor stimulation of prostaglandin E2 biosynthesis in amnion cells. J. BioL Chem., 263 (1988) 7846-7854. J.A. Handler, R.M. Danilowicz and T.E. Eling, Mitogenic signaling by epidermal growth factor (EGF), but not platelet-derived growth factor, requires arachidonic acid metabolism in BALB/c 3T3 cells. J. BioL Chem., 265 (1990) 3669-3673.
