Plasma membrane associated metabolic parameters and the aging of human diploid fibroblasts

Mechanisms of Ageing and Development, 7 (1978) 151-160

151

©Elsevier Sequoia S.A., Lausanne - Printed in the Netherlands

PLASMA MEMBRANE ASSOCIATED METABOLIC PARAMETERS AND THE AGING OF HUMAN DIPLOID FIBROBLASTS

PETER POLGAR, LINDA TAYLOR and LAWRENCE BROWN Department of Biochemistry, Boston University School of Medicine, Boston, Mass. (U.S.A.)

(Received February 16, 1977;in revised form July 20, 1977)

SUMMARY Substrate uptake, portions of the cyclic AMP system, membrane fluidity and cellular phospholipid content are some of the parameters which are structurally associated with the plasma membrane and which have been linked to the control of cell proliferation. These parameters were studied with respect to cellular aging of human embryo lung fibroblasts (HELF) in culture. We observed in late passage an increase in the rate of uridine transport and in cellular cyclic AMP levels. These results were examined in relation to the increase in cell volume which occurs in senescing HELF. We also observed an increase in Vmax of uridine transport, and a decrease in the K m of cyclic AMP phosphodiesterase (PDE) as quiescent, passage 18-25, HELF were stimulated to divide with fresh serum. A similar effect of serum occurred in late passage (p. 43) cells despite the inability of these late passage cultures to undergo further proliferation. There was no change in cAMP-PDE activity with increasing passage number suggesting that the observed alterations of the cAMP levels, basal and in response to extracellular effectors, were due to alterations in the adenyl cyclase system. We observed no change in senescent HELF in membrane fluidity or phospholipid and neutral fat content.

INTRODUCTION The plasma membrane, inclusive of its receptors for external signals and its enzyme systems, is proving an important intermediary in the differentiation and growth of the developing cell, as well as in the determination of a myriad of cell functions, including cell division [1-4]. Cellular aging, which may represent the ultimate step in cellular differentiation, is often accompanied by a reduced function of the given cell type, including an increased reluctance of the cell to undergo cell division [5-7]. Changes in the morphology of the plasma membrane with age have been demonstrated [8]. Adhesiveness and glucosamine content of the plasma membrane are altered in late passage fibroblasts [9]. Lactoperoxidase iodination of chick fibroblasts indicates changes in

152 exposed surface proteins in senescent cultures [10]. Studies of hormone receptors also indicate alterations with age [ 11, 12]. The series of experiments described herein were designed to begin to examine the role of the plasma membrane in cellular senescence. Human embryo lung fibroblasts (HELF) were chosen for study. The in vitro aging of HELF is characterized by a finite doubling number, with a decline in the number of cells capable of division with advancing passage of the culture. We chose to examine parameters which have been implicated in the control of cell division. These include substrate transport, the cyclic AMP (cAMP) system, membrane fluidity and phospholipid content. Our results indicate that certain parameters are altered with passage, while others are not. Also, that certain parameters which are activated by serum are affected in early and late passage cells alike.

MATERIALS AND METHODS

Eight 3H-guanosine 3', 5'-cyclic phosphate, ammonium salt, 10-30 Ci/mmol was obtained from Amersham/Searle Corp., Arlington Height, Illinois. 3H-(G) adenosine 3', 5'-cyclic phosphate, anamonium salt, 30-50 Ci/mmol, and 3H-uridine, 25 Ci/mmol were purchased from New England Nuclear Corp., Boston, Mass. The radioimmunoassay kits for cyclic AMP were purchased from Swartz/Mann, Orangeburg, N.Y. Prostaglandin (PG) E1 was a gift from Dr. J. Pike of the Upjohn Company, Kalamazoo, Michigan. PGE1 was placed into aqueous solution with the aid of NaHCO3. Snake venom (king cobra), Lepinephrine and diphenylhexatriene were obtained from Sigma Chemical Co., St. Louis, Missouri.

Cell culture

Human embryo lung fibroblasts (WI38 and IMR90) were maintained in culture as described previously, with some modifications [13]. The cells were grown in 80 cm 2 bottles for maintenance and passage, and in 20 cm 2 petri dishes for incubation (Falconware, Fisher Scientific Co., Medford, Mass.). The cultures were kept at 37 °(7 in a humidified incubator and gassed with CO2 at CO2/air, 5:95. The culture medium was composed of basal medium Eagle (BME), 90%; fetal calf serum (FCS), 10%; streptomycin, 50 /ag/ml; and penicillin, 50 units/ml. The fetal calf serum was virus screened (Grand Island Biological Co.). The cells were judged mycoplasma-free by the method of Schneider et al. [14]. Uridine/uracil ratios above 400 were obtained routinely. The cultures were placed in a parasynchronized state by allowing them to become quiescent in the growth medium. This occurred a few days after contact density, with one medium change at day 5 after plating at a density of 104 cells/era 2. Quiescent cell cultures were stimulated to divide by replacement of old growth medium with fresh. Under these conditions, maximmn DNA increase as determined by DNA assay, was obtained at 22 hours after stimulation.

153

Cell growth Cell growth in the cultures was followed by cell counts with a Coulter Counter, Model FM, by DNA determination with the diphenylamine method of Burton [ 15], and by autoradiography, as described by Cristofalo and Sharf [I 6].

Cell sizing Cell sizing was done with a Size Distributor Analyzer Model P128 attachment to the Coulter Counter.

Assay for cyclic AMP Cellular cAMP content was determined as described previously [13]. Incubations were initiated by the addition of a tenfold concentration of "hormone" solution to the cell culture. At the end of the incubation, the medium was removed by aspiration, and the cell cultures fixed with 1 m! of 5% perchloric acid, containing a trace of 3H-cAMP. The cell debris was removed from the dish by scraping, the supernatant containing the cAMP was obtained by centrifugation at 1000 X g for 30 min. Following neutralization with KOH, centrifugation and chromatography on a Dowex 50 column, the supernatant containing the cAMP fraction was assayed for the cyclic nucleotide content by radioimmunoassay [17, 18]. A standard curve was performed for each cell assay.

PDE activity To determine PDE activity, a monolayer culture of cells was washed 4 times with 0.15 M NaC1 (25 °C) and scraped from the culture dish in ice cold 40 mM Tris-HC1, pH 7.5, containing 5 mM MgSO4 and 1 mM dithiothreitol. The cells were homogenized in a glass Dounce homogenizer with a Teflon pestle, 20 strokes at 4 °C. The resulting extract was centrifuged at 1000 × g. The supernatant, or the pellet, was assayed for cAMP or cGMP hydrolytic activity. The activities were determined by the radiodisplacement method of Thompson et al. [19]. Briefly, the given fraction of the above broken cell preparation was added to a preincubated mixture of lyophilized king cobra venom, with various quantities of 3H-cAMP or 3H-cGMP dissolved in the same 40 mM Tris-HC1 solution as above. After the incubation, the reaction was terminated with a slurry containing 50% Bio-Rad AG l-X2 (200-400 mesh) resin in the chloride form. The presence of 3H-adenosine or aH-guanosine was then determined.

Substrate uptake To determine uptake of substrate, the method of Cunningham and Pardee [20] was followed, with slight modifications [21]. The cultures were incubated either directly in the culture medium or in phosphate buffered saline, pH 7.4 (PBS), as indicated. To stop incubation, the incubation medium was quickly removed by aspiration. The cultures were then washed four times with ice-cold isotonic saline and fixed with 10% trichloroacetic acid (TCA). The whole fixing procedure was done on ice and took less than 30 seconds. Acid soluble and insoluble labeled fractions were separated by centrifugation. Radioactivity in the acid-insoluble material was determined by either solubilization or

154 trapping on a Millipore filter. Radioactivity was determined in Aquasol (New England Nuclear Corporation). A Packard Tri-Carb liquid scintillation counter was used. The quenching in various fractions showed negligible variation.

Fluorescence polarization determination The procedure followed has been described by Fuchs et al. [22]. Plates were washed with PBS and the cells were then incubated with 2 × 10-6M diphenylhexatriene (DPH) in PBS. The cells were labeled for 60 minutes at 37 °C, and then washed with PBS. The cells were then treated with trypsin, suspended, collected and resuspended in PBS after centrifugation. Control cells were treated identically, except for the inclusion of DPH. Polarization determinations were done within one hour of this procedure. Fluorescence polarization and intensity were measured in a MPF-2A Hitachi PerkinElmer spectrofluorimeter, equipped with polarizers and a thermostated cell holder. The fluorescence polarization value, O, was calculated from the following equation:

/by

/by

P = I w - T ~ Ivhll~v + - ~ Ihv,

where I is the corrected fluorescence (fluorescence of DPH labeled cells minus fluorescence of the same number of unlabeled cells) and the subscripts v and h indicate the orientation, vertical and horizontal, of the excitation and analyzer polarizers.

Determination of phospholipids and neutral fats The procedures followed for the assay of phospholipids and neutral fats were described by Downing et aL [23]. Cells were washed 3 times in calcium, magnesium free (CMF) PBS, trypsinized and pelleted. The cell pellet was washed twice in CMF PBS. Lipids from the cell pellet were extracted in choroform/methanol (2:1) overnight at room temperature. The chloroform/methanol was evaporated to dryness under nitrogen, and the lipids were resuspended in small volumes of hexane. Lipids were spotted on silicacoated t.l.c, plates. Neutral lipids were run in hexane/ether/acetic acid (70:30:1) and phospholipids in chloroform/methanol/water/acetic acid (25:14:4:1). The plates were sprayed with sulfuric acid, charred on a hot plate and scanned with a photodensitometer attached to a stripchart recorder. The relative areas under the peaks were assumed proportional to the relative amounts of the various lipids. Known amounts of standards were run on each plate to calculate the absolute amounts of the lipids.

RESULTS The uptake of uridine was examined in early passage (23) and late passage (42) HELF. Table I illustrates the Km and Vmax determined with initial rates of uptake in quiescent cultures and quiescent cultures stimulated to divide with fresh culture medium. Under these conditions, 83% of passage 23 and 5% of passage 42 cultures went on to

155 TABLE I KINETIC PARAMETERS OF URIDINE UPTAKE IN EARLY AND LATE PASSAGE W138 CELLS Cultures

WI38 p. 23 W138 p. 43 WI38-VA13-2RA*

Quiescent

Serum stimulated

K m +-S.E.

Vma x +-S.E.

K m +-S.E.

Vraax +-S.E.

50 +- 4.1 58 +-5.2 12.5 +- 0.93

14 +- 0.98 20+-.1.3 26 +- 2.2

50 +- 5.2 58+-6.0 -

25 +- 2.2 38+-2.9 -

Quiescent cultures were stimulated with fresh serum for 18 hours. The cultures were then washed with PBS and incubated in the presence of various concentrations of cold uridine containing a trace of 3Huridine. After a 1 rain incubation, the cultures were treated as described in text, All values were obtained from triplicate culture plates. Km:WVI, Vmax:pmol/10 s ceUs/min. Each curve contained 6 points. *The WI38-SV40 cultures were not treated as quiescent or serum stimulated ceils.

T A B L E II KINETIC PARAMETERS OF LOW K m cAMP-PDE ACTIVITY Cultures

WI38 p. 25 WI38 p. 43 IMR90 p. 10 WI38-VA13-2RA

Quiescent

Stimulated

Krn(IJM) +-S.E.

Km(WVI) +-S.E.

2.6 2.13 2.18 4.1

1.3+-0.11 1.2 +- 0.09 1.5 +- 0.13 1.9 +- 0.21

+-0.29 +- 0.18 +- 0.25 +- 0.38

The kinetic determinations were carried out using three confluent (60 cm 2) plates for each point on the curve. Each curve contained 6 points. The quiescent cells were stimulated with a change of medium, containing 10% fresh fetal calf serum. The cells were then allowed to remain for 18 hours at 37 °C, in a water-saturated 95% air-5% CO2 incubator. The plates were washed 3 times with PBS, scraped, and homogenized in a Dounce homogenizer. The assay for PDE activity was carried out as described in text.

divide. The addition o f fresh culture m e d i u m to the quiescent H E L F , increased the Vmax o f u p t a k e within 18 hours o f the stimulation. This increase occurred in b o t h the early and late passage cells. The Vmax o f the late passage cultures was a p p r o x i m a t e l y 1.5 that o f the early passage cells. The K m o f these cells remained unchanged with serum stimulation and w i t h passage n u m b e r o f the cultures. Uridine transport in SV40 transformed H E L F showed a lower K m t h a n that in the n o r m a l H E L F . Serum stimulation o f quiescent WI38 cells in culture also has an effect on the high affinity, l o w K i n , PDE for cAMP. This is illustrated in Table II. As quiescent WI38 cells are exposed to fresh serum, the K m o f the c A M P - P D E is l o w e r e d b y a p p r o x i m a t e l y 80%, at 22 hours after the stimulation w i t h the serum. This alteration occurs in passage 25 cultures as well as passage 45 cultures. The activity o f the c A M P - P D E in late passage cells remains unchanged. When g r o w t h o f the early and late passage cultures was c o m p a r e d ,

156 TABLE III CYCLIC AMP LEVELS EXPRESSED AS PMOL cAMP/')"DNA Days in culture

3 5 6 8

Basal cAMP

PGE l stimulated cAMP

Stimulated/basal

P18

P43

P18

P43

P18

P43

2.07 2.03 1.85 2.36

5.38 7.66 8.05 7.71

14.7 42.5 44.2 46.8

15.7 80.9 107 101

7.10 20.9 23.9 19.9

2.92 10.6 13.4 13.1

Cells were plated at a density of 1 × 105/20 cm 2 dish. Days in culture refers to days after plating. DNA was determined with the procedure of Burton [ 16 ]. Cyclic AMP was determined as described in text. Each determination was run in triplicate culture plates. Each value is an average. Maximum deviation did not exceed 12%.

TABLE IV EFFECT OF L-EPINEPHRINE ON cAMP LEVELS IN HELF Passage

Cell type

Stimulated cAMP/basal cAMP

11 18 25 40 55

IMR90 IMR90 WI38 WI38 W138

2.2 9.5 10.8 18.4 17.7

Confluent (quiescent) plates of W138 or IMR90 human embryo fibroblasts were treated as described in Table III. Epinephrine, final concentration 50 p34, was used. All points were done in duplicate. Maximum deviation did not exceed 10%. Cyclic AMP is expressed in pmol. passage 25 cultures underwent active cell growth following the addition of the fresh serum; however, only a minimal amount of cell division occurred in passage 45 cultures. The effect of serum on PDE activity in intact WI38 cells occurred within 2 hours of addition. The activity of the PDE for cGMP is not affected by serum. Cyclic GMP-PDE activity was determined in cells treated with fresh medium at 15 min, 2, 6 and 22 hours after serum stimulation. No alteration in activity was observed at any time interval following the change of medium. We also found no change in activity in the low affinity, high K m cAMP-PDE as quiescent cells were treated with serum. To examine if changes occur in the cAMP system with increasing passage number, we determined basal cAMP levels and cAMP levels in response to extracellular effectors. Extracellular effectors, such as PGEI, increase cAMP levels and inhibit cell division in WI38 cells [13]. Cell division in these cells is also inhibited by cAMP analogs added to the culture [13, 24]. Cellular cAMP content, basal and in response to PGEI stimulation was determined in early and late passage WI38 cultures (Table III). The basal cAMP levels remained essentially unchanged from 3 to 8 days after passage of the cells. At day 8, the

157 cultures (early and late passage) were quiescent. Basal cAMP levels rose approximately 3- to 4-fold per cell from passage 18 to passage 43. The cAMP response to PGE1 increased with days in culture after plating in both young and old cells, up to day 6 after plating. Late passage cells showed a decrease in stimulated/basal ratios. Table IV demonstrates the cAMP response to epinephrine in confluent cultures of HELF in various passages. In this case the response to epinephrine (stimulated cAMP/ basal cAMP) increases with increasing passage. . Similar results were demonstrated by Haslam and Goldstein in human skin fibroblasts [25]. The average cell volume in our cultures increased with passage from 1800/al 3 at passage 18, to 6400/al 3 at passage 43. The increase in volume of senescing cells has also been observed by others [26, 27]. If this increase is taken into consideration and the data in Table II are expressed as cAMP/cell volume, then, as shown in Table V, the basal cAMP levels of the old and young cells remains approximately the same. The response of the old cells to PGE1, however, remains less than in the cultures of early passage. In relation to the differences in cell volume and response to prostaglandins in the early and late passage cultures, it should be pointed out that after the initial seeding, the cell densities in the early and late passage cultures were never the same, both during log growth and quiescence. Because of the recent reported correlations between membrane fluidity and cellular phospholipid content and receptor-adenyl cyclase interaction [28, 29], we looked at these parameters in early and late passage HELF. The phospholipid and neutral lipid content of passage 18 and 48 cultures is shown in Table VI. Our results are very similar to those reported by Kritchevsky and Howard [30]. Essentially we found no changes in lipid content in late passage cells. We also found no change in membrane fluidity in late passage. The fluorescence polarization value, p, at 22 °C was 0.21 at passage 23 and 0.20 at passage 48. DISCUSSION Our studies on uridine transport indicate a net increase in the number of transport sites in late passage cells with the efficiency of each site remaining unchanged. This TABLE V CYCLICAMP LEVELSEXPRESSEDAS pmol cAMP/mm3 CELLS Days after plating

3 5 6 8

Basal cAMP

PGE 1 stimulated cAMP

Stimulated/cAMP basal/cAMP

P18

P43

P18

P43

P18

P43

8.35 8.18 7.46 9.52

6.10 8.67 9.11 8.73

59.3 171 178 189

17.8 91.6 121 114

7.10 20.9 23.9 19.9

2.92 10.6 13.4 13.1

The data are identical to those in Table III only expressed per cell volume.

158 TABLE VI PHOSPHOLIPID AND NEUTRAL LIPID CONTENT IN EARLY AND LATE PASSAGE W138 CELLS Phospholipids

% o f total phospholipids p. 18

p. 48

sphingomyelin lecithin phosphatidylinositol phosphatidylethanolamine

16 47 8 29

17 47 8 28

Neutral lipids

% of total neutral lipids

cholesterol fatty acids triglycerides cholesterol esters monoglycerides diglycerides

34 10 10 27 12 7

32 9 11 31 11 6

The procedures followed for the assay of phospholipids and neutral fats were as described in text. Confluent 75 cm 2 tissue culture flasks were used. Data from a representative experiment are shown.

increase appears as a partial compensation by the cell for the observed increase in cell volume. In fact, when the increase in volume is considered, the net number o f sites per surface area actually decreases in senescing cells. The cellular response to growth stimulation by serum is to increase the number o f available transport sites. Interestingly, this occurs in cells o f early and late passage alike. Since the late passage cultures selected for these studies underwent only minimal growth following serum stimulation, these results suggest that at least a part o f the metabolic system responding to extracellular growth activation is operating in " o l d " cells despite the inability o f these cells to go on and divide. A similar event occurs with the high affinity, low K m PDE for cAMP. Addition o f fresh serum to quiescent cultures lowers the K m o f this enzyme in early passage. This response o f the PDE to serum remains unaltered in late passage cells. Incidentally, a low K i n , high cAMP affinity enzyme has been reported to be bound to the plasma membrane in chick embryo fibroblast ceils in culture by Pastan [31 ]. We have not been able to locate this enzyme in particulate preparations from the WI38 cells. It is possible, however, that this enzyme is being solubilized under our isolation conditions. The observed changes in the effect o f epinephrine and PGE~ on cAMP levels in late passage cells indicate that an alteration in adenyl cyclase enzyme system is occurring. Our results on cAMP phosphodiesterase, showing no changes in activity with increasing passage, further support this conclusion. It is not clear from these studies which member o f the adenyl cyclase enzyme complex has changed. However, our results in combination with those reported b y others, give us certain clues. F o r example, the study o f Barnes et

159 al. suggests that this observed alteration in the response o f adenyl cyclase to hormones is

not due to a simple change in basal adenyl cyclase activity. Broken cell studies o f basal adenyl cyclase activity in human diploid fibroblasts conducted by this group showed no change in late passage [32]. We observed no changes in membrane fluidity or in phospholipid content in late passage cells. To the extent that these parameters can be used as indicators o f receptor-enzyme interaction, our results suggest that no such alteration is taking place during senescence. On the other harrd, changes in hormone-receptor interaction in senescing cells have been demonstrated [11, 12], and may be related to the observed alteration in cAMP response to PGE and epinephrine. It remains to be resolved whether the net increase in cAMP content per cell in late passage is significant and contributes to the inability o f senescing cells to divide; or is merely a reflection o f the increase in cell volume. Certainly the proliferation o f HELF is sensitive to increases in cAMP levels and to extracellular effectors, such as PGE~ which increase cAMP levels [21]. Our more recent studies [33] with cAMP levels and cell growth suggest that a very small rise in cAMP per cell accompanies the inhibition o f cell division by low concentrations o f PGE. Thus, even small increases in cAMP cell content in late passage HELF, particularly if this increase is compartmentalized, may be contributory to the inability o f late passage cells to undergo cell division.

ACKNOWLEDGEMENTS We wish to thank Dr. D. Downing for many hours o f help. This work was supported by grants from the NIH, AG-00455 and T32HL07052.

REFERENCES 1 G. E. W. Wolstenholme and J. Knight (eds.), Growth Control in Cell Culture, Ciba Foundation Symposium, Churchill Livingstone, Edinburgh and London, 1971. 2 L Sachs, Regulation of membrane changes, differentiation and malignancy in carcinogenesis, The Harvey Lectures Series 68, Academic Press, New York, 1972-1973, pp. 1-35. 3 S. Roth, A molecular model for cell interactions, Q. Rev. Biol., 48 (1973) 541-563. 4 R. W. Holley and J. A. Kieman, Control of the initiation of DNA synthesis in 3T3 ceils, Proc. Nat. Acad. Sci. U.S.A., 71 (1974) 2942-2945. 5 H. E. Swim and R. F. Parker, Culture characteristics of human fibroblasts propagated serially, Am. Z Hyg., 66 (1957) 235-243. 6 L. Hayflick and P. S. Moorhead, The serial cultivation of human diploid cell strains, Exp. CelIRes., 25 (1961) 585-621. 7 I. L. Cameron, Call proliferation in aging mice, J. Gerontol., 27 (1972) 162-172. 8 P. D. Bowman and C. W. Daniel, Aging of human fibroblasts in vitro: surface features and behavior of aging WI38 cells. Mech. Aging Dev., 4 (1975) 147-158, 9 R. Azencott, C. Hughes and Y. Courtois, Cell surface alterations and in vitro aging in animal cells, Adv. Exp. Med. Biol., 53 (1975) 147-153. 10 Y. Courtois and R. E. Hughes, Surface labelling of senescent chick fibroblasts by lactoperoxidasecataiysed iodination, Gerontologia, 22 (1976 ) 371-379. 11 G. S. Roth and R. C. Addman, Age related changes in hormone binding by target cells and tissues: possible role in altered adaptive responsiveness, Exp. Gerontol., 10 (1975) 1-11.

160 12 A. L. Rosenbloom, S. Goldstein and C. C. Yip, Insulin binding to cultured human fibroblasts increases with normal and precocious aging, Science, 193 (1976) 412-415. 13 M. J. Kurtz, P. R. Polgar, L. Taylor and A. M. Rutenburg, The role of cyclic AMP in DNA synthesis and cell division of the W138 cell line in culture, Biochem. J., 142 (1974) 339-344. 14 E. L. Schneider, E. J. Stanbridge and C. J. Epstein, Incorporation of 3H-uridine and 3H-uracil into RNA, Exp. Cell Res., 84 (1974) 311-318. 15 K. Burton, A study of the conditions and mechanisms of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid, Biochem. J., 62 (1956) 315-323. 16 V. J. Cristofalo and B. B. Sharf, Cellular senescence and DNA synthesis, Exp. Cell Res., 76 (1973) 419-427. 17 A. L. Steiner, D. M. Kipnis, R. Utinger and C. W. Parker, Radioimmunoassay for the measurement of adenosine 3', 5'-cyclic phosphate, Proc. Nat. Acad. Sci. U.S.A., 64 (1969) 367-373. 18 P. Polgar, J. C. Vera, P. R. Kelley and A. M. Rutenburg, Adenylate cyclase activity in normal and leukemic human leukocytes as determined by a radioimmunoassay for cyclic AMP, Biochim. Biophys. Acta, 297 (1973) 378-383, 19 W. J. Thompson, G. Brooker and M. M. Appleman, Assay of cyclic nucleotide phosphodiesterases with radioactive substrates, in J. G. Hardman and B. W. O'Malley (eds.), Methods in Enzymology, Vol. 38 Part C, Academic Press, New York, 1974, pp. 205-212. 20 D. D. Cunningham and A. B. Pardee, Transport changes rapidly initiated by serum addition to "contact inhibited" 3T3 cells, Proc. Nat. Acad. Sci. U.S.A., 64 (1969) 1049-1056. 21 P. Polgar and L. Taylor, Effects of prostaglandin on substrate uptake and cell division in human diploid fibroblasts, Biochem. J., 162 (1977) 1-8. 22 P. Fuchs, A. Parola, P. W, Robbins and E. R. Blout, Fluorescence polarization and viscosities of membrane lipids of 3T3 cells, Proc. Nat. Acad. Sci. U.S.A., 72 (1975) 3351-3354. 23 D. T. Downing, Photodensitometry in the thin-layer chromotographic analysis of neutral lipids, J. Chromatog., 38 (1968) 91-99. 24 J. E. Froehlich and M. Rachmeler, Effect of adenosine 3', 5'-cyclic monophosphate on cell proliferation, J. Cell Biol., 55 (1972) 19-31. 25 R. J. Haslam and S. Goldstein, Adenosine 3', 5'-cyclic monophosphate in young and senescent human fibroblasts during growth and stationary phase in vitro, Biochem. J., 144 (1974) 253-263. 26 J. W. I. M. Simons, The use of frequency distributions of cell diameters to characterize cell populations in tissue culture, Exp. CellRes., 45 (1967) 336-350. 27 P. D. Bowman, R. L. Meek and C. W. Daniel, Aging of human fibroblasts in vitro, Exp. Cell. Res., 93 (1975) 184-190. 28 V. H. Engelhard, J. D. Esko, D. R. Storm and M. Glaser, Modification of adenylate cyclase activity in LM cells by manipulation of the membrane phospholipid composition in vivo, Proc. Nat. A cad. Sci. U.S.A., 73 (1976) 4482-4486. 29 T. V. Zenser, V. J. Petrella and F. Hughes, Spin-labeled stearates as probes for microenvironment of murinethymocyte adenylate-cyclase adenosine 3', 5'-monophosphate system, J. Biol. Chem., 251 (1976) 7431-7436. 30 D. Kritchevsky and B. V. Howard, Lipid metabolism in human diploid cells, in E. Holeckowi and V. J. Cristofalo (eds.), Aging in Cell and Tisgue Culture, Plenum Press, New York, 1970, pp. 57-82. 31 T. Russel and I. Pastan, Plasma membrane cyclic adenosine 3', 5'-monophosphate phosphodiesterase of cultured ceils and its modifications after trypsin treatment of intact cells, J. BioL Chem., 248 (1973) 5835-5840. 32 E. Barnes, D. Brandes, P. Barnes, H. Ryser and S. Barnard, Adenyl cyclase activity during in vitro aging of Wl-38 diploid cell strain, In Vitro, 7 (1972) 274-275. 33 L. Taylor and P. Polgar, Autoregulation of growth by human diploid fibroblasts via prostaglandin production, In Vitro, 13 (1977) 164.