Identification and characterization of non-dividing cell populations in phase II cultures of human glial cells

Mechanisms of Ageing and Development, 26 (1914) Elsevier Scientific Publishers Ireland Ltd.

1~ 12

IDENTIFICATION AND CHARACTERIZATION OF NON-DIVIDING CELL POPULATIONS IN PHASE H CULTURES OF HUMAN GLIAL CELLS

V. PETER COLLINSa>*, JAN S. WILLEMSa, JEAN M. ALVA-WILLEMSa H. THAWb aDepartment of Tumour Pathology, Karolinska Institute, Stockholm, II, University of LinkGping, Linkiiping (Sweden) (Received (Revision

July 18th, 1983) received December

19th,

and HOWARD

and bDepartment

of Pathology

1983)

SUMMARY

It has been reported that cultures of normal cells in phase II contain a non-multiplying cell population, the size of which increases with passage number. In phase II cultures of normal human glial cells we have found two subpopulations

of non-proliferating

cells, one

of which has a characteristic morphology, and differs from the actively dividing cells in a number of respects: (1) they are larger although of various sizes and are well spread over a very large substratum area: (2) they contain a great number of granules showing acid phosphatase activity, being heavy metal positive and displaying the characteristic natural fluorescence of lipofuscin pigment; and (3) they frequently contained a central somewhat structures.

irregular nucleus with various numbers of darkly staining nucleolar-like Cytophotometric nuclear DNA measurements of the described “large” cell

population

show a decreased

proportion

of diploid

cells as compared

to their smaller

sister cells. Moreover, with increasing passage number, the DNA values for large cells shift towards higher ploidy levels resulting in a scattered aneuploid pattern in the oldest passage. This “large” cell subpopulation consists of between 2% and 3% of all passages and becomes

greatly

decreased

following

subcultivation.

non-dividing cells is generally morphologically with passage number and is the more important

The other

subpopulation

of

similar to the dividers, increases in size in the phase III phenomenon.

Key words: Cultured cells, human; Glial cells; Lysosomes, cell proliferation;

DNA

INTRODUCTION

Cultivated normal human glial cells have a limited life span and after a variable number of population doublings they enter phase II [ 11, and cease to proliferate [2,3]. In this *To whom

correspondence

0047-6374/84/$03.00 Printed and Published

should

be addressed.

0 1984 Elsevier Scientific in Ireland

Publishers

Ireland

Ltd.

and other systems an increasing proportion of non-dividing or slowly dividing ceils with passage number has been reported [4-8] eventually leading to the phase III phenomenon with the subsequent demise of the culture. A cell and nuclear size increase in the later, pre-phase III, passages of normal ceils has been well documented [5,9,10] and has subsequently been correlated with the occurrence of non-dividers [ 11 ]. We have frequently observed distinct groups of "large" cells appearing in phase II cultures of normal human glial cells and to our knowledge there have been no descriptions of such subpopulations, particularly in relation to the phase III phenomenon. We have thus studied cultures of glial cells at various passages in phase II in order to investigate the possible origins of these "large" cells, and to identify and subsequently cytochemically characterize non-dividing cell populations. MATERIAL AND METHODS Cell lines and culture conditions The experiments were performed on a line of cultivated diploid human glial cells (U-787 CG), which were derived and maintained in culture as described previously [17,18]. The cells were grown routinely in 270 ml of Nuclon ® plastic bottles and refed twice weekly with Eagle's minimal essential medium (EMEM) with 10% newborn bovine serum and antibiotics (100 IU/ml penicillin, 50 /ag/ml streptomycin and 2.5 /lg/ml amphotericin). Cells in passages 20, 25, 34, 42 and 47 (this cell strain has maximally reached 48 passages [19]) were plated at approximately 8000 cells/cm 2 in 35 mm Nuclon ® Petri dishes containing either a 20 mm × 20 mm coverslip or a Biirker chamber coverslip. The cultures were studied after they had reached a post-confluent state and were density-dependent growth inhibited. The time period that the individual cultures required to reach this state varied between 8 days (passage 20) and 28 days (passage 47) ([7] ; V.P. Collins, unpublished results). EMEM, prepared with or without 0.01 /~Ci/ml [3H]thymidine ([3H]) (specific activity 25 Ci/nM), was added 48 h following plating, and the cells received new medium every other day until the cultures were post-confluent. Some cukures (passage 25) were subcultivated 1:2 at this time and then fixed for autoradiography after an additional 48 h in EMEM without [3H]TdR. Cultures destined for autoradiography were washed four times in phosphate-buffered saline (PBS), fixed for 1 h in methanol-acetic acid (3:1) and subsequently processed for autoradiography as previously described [20]. A minimum of 1000 cells was counted in all autoradiograms, and parallel cultures were fixed for light microscopy, cytochemistry and cytophotometric DNA measurements as described below. Light microscopy and cytochemistry Those cultures intended for the cytochemical demonstration of acid phosphatase and May-Griinewald Giemsa staining were fixed as described earlier using 2% glutaraldehyde in 0.1 M sodium cacodylate HC1 buffer with 0.1 M sucrose [21]. Acid phosphatase was

3

demonstrated

using a Gomori-type

phide method performed

reaction

as previously

and heavy metals by a modified

reported

silver sul-

[22] .

Demonstration of autofluorescence The living cells (on coverslips) were rinsed, mounted in PBS and rapidly examined in a Leitz fluorescence microscope as previously reported [ 161. Cytophotometric DNA analysis Cytophotometric

DNA analysis was carried out on duplicate

cultures (from passages

20, 25, 34, 42 and 47) grown on Biirker chamber coverslips. The cultures were fixed in 10% neutral buffered formalin for a minimum of 24 h and stored at 4°C until staining. Acid hydrolysis was performed at room temperature the Feulgen-naphthol yellow S staining [23] . Since the material hydrolysis

under

study

displayed

(22°C) in 5 M HCl, followed by

a variation

time of 1 h was used, this being optimal

in chromatin

compactness

for both compact

a

and dispersed

chromatin [24] . Cytophotometric measurements were performed using a rapid scanning and integrating microspectrophotometer [25] operated at a wavelength of 546 run. Fifty cells from each culture were morphologically identified as either small or large cell type and classed separately blood leucocytes,

in the measurements.

Measurements

on human, normal peripheral

taken as controls, were used to obtain diploid (2~) DNA values.

RESULTS

Light microscopy and long-term / “HJ TdR incorporation Those cultures studied following confluency and density-dependent were composed

of monolayered,

nuclear overlapping.

Occasionally,

regularly distributed

polygonal

small groups consisting

growth inhibition

cells absent of apparent

of 2-4 cells which were much

larger than those surrounding them could be seen (Fig. 1). These “large cells” most often contained a central nucleus with a number of nucleolar-like structures, around which the cytoplasm was evenly nuclei. The substratum

and thinly spread. A limited number of “large cells” had two area taken up by such cells was up to six times that taken up by

adjacent smaller cells, which henceforth will be referred to as “small cells”. In the later passages (>40) the “large cells” were more difficult to discern from the “small cells” as the latter approached

the former with regards to size. Sometimes a lacuna could be seen in the monolayer, surrounded by “small cells”, suggesting that a “large cell” had died or loosened from the substratum. The incidence of “large cells” remained constant (between 2% and 3% of all cells) and did not vary with passage number. Cells grown in the presence of i3H]TdR from 48 h after subcultivation until density growth inhibition showed [3H]TdR incorporation to various degrees, depending on passage number (see Figs. 2 and 5). There was an increase in the number of unmarked cells with increasing passage number. In passage 25, 20% of the cells lacked [3H] TdR incorporation, while passages 34, 42 and 47 contained 21%, 34% and 49% unmarked cells, at confluency, respectively.

b

anl

Ill~ !'.

f

q

IJ"

q

[

I

L_large cells unmarked [ ] large cells marked ~'~small cells unmarked []small cells marked 7° 100-

9080-

70 6050.

40 30

10 25 34 Passage nr

42

47

Fig. 5. A comparison of cell populations in each of the passages studied after reaching densitydependent growth inhibition. The relationship between the various large and small glial cell subpopulations (expressed as a percentage of the total population), including the fraction of each incorporating [ 3H] TdR. Cells were considered m a r k e d if t h e y had 5 grains m o r e t h a n the background, over the nucleus. The labelling indices o f b o t h the " l a r g e " and " s m a l l " cell types, following long-term [3HI T d R labelling, decreased w i t h passage n u m b e r (Fig. 5). The labelling o f small cells was characterized b y either m a n y evenly dispersed grains or the t o t a l absence Fig. 1. A 14-day-old glial cell culture (passage 25) stained with May-Griinewald Giemsa showing a number of large ceils surrounded by smaller glial cells. Note the centrally placed large nucleus with three nucleolus-like structures and the evenly and circularly spread cytoplasm with its very thin periphery where the adjacent small glial cells make contact, x 70. Fig. 2. A 14-day-old glial cell culture (passage 25) which has undergone long-term exposure to [3H ITdR (for autoradiography) and subsequently counterstained with May-Griinewald Giemsa. Note the signs of [3H] TdR incorporation in almost all cells, except for the large cell (compare with Fig. 1). ×170. Fig. 3. Demonstration of acid phosphatase (passage 25, 14-day-old culture). Note the very positive perinuclear granules in the central large cell. The surrounding cells show only very weak staining at this incubation time. ×270. Fig. 4. Demonstration of heavy metals by the silver sulphide method (passage 25, 14-day-old culture). Note the large central cell, again containing a great number of highly positive, perinuclear granules and the surrounding cells which are very weakly stained in comparison. ×270.

of grains. The large cells, in contrast, if marked at all, contained only very few, evenly dispersed grains over the nucleus. The large cell populations in all passages had a labelling index approximately half that of the small cells (Fig. 5). Cultures from passage 25 which had undergone long-term (14 days) labelling with [3H] TdR, followed by subcultivation and log phase growth in non-[3H] TdR-containing medium showed little trace of the large cell population. At 48 h the "large cells" were found to amount to less than 0.5% of the total population while the fraction of small cells with no signs of [3H]TdR incorporation had increased from 19% to 25%.

Acid phosphatase demonstration The incubation times were made particularly short ( 4 5 - 6 0 min). At 45 min the "large cells" showed numerous acid phosphatase positive granules whereas the surrounding "small cells" showed varying degrees of "staining" which was, however, generally much less at this time (see Fig. 3). Controls incubated with NaF or without substrate showed no activity. Demonstration of heavy metals The incubation period for the silver sulphide method was also shortened to 1 h, and this resulted in a distribution pattern which was similar to that of acid phosphatase (compare Figs. 3 and 4). The surrounding "small" glial cells had fewer positive granules (Fig. 4) following this short incubation time. There was no staining in the non-sulphidated controls. Demonstration of autofluorescence Large numbers of fluorescent perinuclear granules were observed, and their distribution was similar to both the acid phosphatase and heavy metal positive granules which were visible in the "large-cell" population. Various numbers of similar, but smaller granules could be seen in the adjacent "small cells". DNA The Feulgen-naphthol yellow S staining allowed easy identification (prior to measuring the nuclear DNA) of the smaller, more uniform appearing cells or the larger, irregular, occasionally multinucleated cells. After 20 passages (Fig. 6) nuclear DNA values of the "small cells" were distinctly diploid (2c), while the "large cells" had DNA modal values of 2c, 4c and 8c (intermediate 3c DNA level). At passage 25 the "small cells" again showed largely diploid (2c) DNA values, although there were single cells with 3c and 4c values. The "large cell" population was only to a small proportion diploid (2c), the cells predominantly being tetraploid (4c) with a few having octaploid (8c) values. In passage 34 the small cells consisting of both diploid (2c) and tetraploid (4c) DNA made up 1/3 of the small cell population. The large cells' heteromorphous nuclei show an

SMALL CELLS of c, ~ler num

LARGE CELLS

Control

40.

Control

20.

.

I

I

tt

i

20th p

20 t h p

25thp

25th p

40

20.

40

20

J

.UL,__

i 'i 4~0[

-

.

, ,

47t h p

DNA (rel.units)

Fig. 6. A comparison of DNA histograms for small and large glial cell subpopulations for each of the various passage numbers studied. Controls consisted of leucocytes.

increasingly polyploid pattern with modal DNA peaks of 2c, 4c, 8c and 16c with some intermediate values at 3c and 6c. The proportion of diploid (2c) "large cells" was diminished compared to earlier passages (20 and 25). In the 45th passage, small and large glial cell varieties displayed a grossly similar nuclear DNA pattern to that observed in passage 34. The 47th passage consisted of small cells that are diploid (2c) with a tetraploid (4c) peak. The DNA content of the large cells shows a shift to a higher ploidy level with diminishing diploid (2c) and tetraploid (4c) DNA levels, and a tendency to aneuploidy with single values of up to 25c. To summarize, the small cells show at a certain stage (between passage 25 and 34) transition from a unimodal diploid (2c) to a bimodal (2c-4c) pattern. The larger cells display a polyploid DNA content which evolves with increasing passage number towards proportionately higher ploidy values and in the final passages to aneuploidy. DISCUSSION Considerable interest has been focused on the diminished replication capacity of diploid cell populations as they enter phase III [1,26,27]. That slowly dividing or nondividing subpopulations exist in phase II cultures has been reported for human glial and other cell systems [4,6,7]. Increased cell size as well as increased nuclear size has been found in the late passages of WI-38 cells [10,11] and separation of WI-38 cells on the basis of volume revealed that the cell fraction with the largest volumes contained the highest percentage of slow or non-replicating cells [28]. The present study describes a characterization of two non-dividing subpopulations. One has a very characteristic morphology and makes up an unusually stable fraction (between 2% and 3%) of all passages studied. The fact that this cell population decreases markedly following subcultiration, along with the finding that these cells often appear in small groups, suggests that this phenomenon may reflect a parallel determination of two daughter cells occurring with unchanged probability regardless of passage number. This fraction also shows a decreased [3H]TdR incorporation incidence with increased passage number, reflecting the general trend in the culture as the cells approach phase lII. Thus, the concept that after a period of division, certain cells arise which get irreversibly committed to senescence and death [29,30], may be in accord with the present observations for this subpopulation. While comparisons may not directly be made between the present study and that of Yanishelsky e t al. [31] on human fibroblasts, the trend in the relationship between proliferating and non-proliferating cells is similar. We have never observed cell fusion in these cultures despite many thousands of hours time-lapse cinemicrography in various studies (for example, refs. 19-21). How the "large cells" become polyploid is uncertain (Fig. 6); however, certain speculations can be made and our observations would suggest a number of possible pathways from which the large cells could be derived (Fig. 7). Cells of the earliest passage studied (passage 20) could increase in size to become diploid "small" or "large" cells. Another possibility is

PLOIDY

STEM CELLS

CELL SIZE SMALL ~ LARGE

CELL DEATH e

NOT OBSERVED

4+C

I

II

~

Ill

Fig. 7. A schematic diagram showing the possible pathways (including stem cells) from which the large cell subpopulation may have been derived. Consistent with the data from Figs. 5 and 6: (1) lla to llIa: occurs only up to passage 42 (since no large cells with 2c DNA content are present after this passage); (2) Illc to IIId: probably occurs only after passage 25 (prior to passage 25 there are no small cells with 4c DNA content in growth inhibited cultures); (3) IIb,c to Illb,c: appears to be the main pathway after passage 34 (no small cells with ploidy value greater than 4c have been observed in this study). Cell death or disappearance of cells is mainly associated with subcultivation.

that they increase in size from the G2 or tetraploid state to become "large cells" (pathways IIa to Ilia, or IIb,c to IIIb,c - see Fig. 7). The decrease in the marking index of the large cell group might be explained by an increased progression from the llc state to the IIIc state as the IIc group becomes significant after passage 34. In the higher passages it appears that many o f the large cells are derived from other large cells since we have never seen the small type with a polyploidy content o f greater than 4c. Endoreduplication is the most likely underlying cause, alone, or even in combination with a G2 block. The low grain numbers over "large cells" nuclei would tend to support this hypothesis. However, in this study, chromosome analysis and correlation between DNA level and quantitative [3H] TdR incorporation have not been performed. Our findings of a cell group with aberrant DNA amounts is in accordance with reports o f a group of 2 - 6 % of cells in phase II cultures o f normal human fibroblasts which show chromosome aberrations and polyploidy [ 3 1 - 3 3 ] . Phase III cells contain large numbers o f secondary lysosomes o f the residual b o d y type, which are comparable (both ultrastructurally and histochemically) to those found in aged post-mitotic cells in vivo [ 1 2 - 1 4 ] . Residual bodies also occur in actively dividing

10 human glial cells in phase II, accumulating during density-dependent inhibition of growth, only to decrease in amount on the return of the culture to log growth phase following subcultivation [1536,34]. The similarity between the cytochemical pattern achieved for demonstration of heavy metals together and the accumulation of secondary lysosomes containing material with the characteristic autofluorescence of lipofuscin probably reflects this cell group's non-proliferative state [15,22]. This "large cell" population, although obvious and easily identifiable, probably only represents a variant of the phase III phenomenon at the cellular level. The non-dividing "small cell" population was not easily morphologically distinguishable from the proliferating "small cells". It is possible that the "small" non-dividers have a greater content of lipofuscin, but quantitative correlations between the lipofuscin/cell and the [3H] TdR incorporation/cell would be required to show this. There is, however, a trend in the fact that later passages have increasing non-dividing "small cell' fractions and an increasing proportion of small cells with a 4c DNA value, a finding which might suggest that a G2 block may be present in this group. In studies of WI-38 cells, MacieraCoelho et al. [35] found evidence for prolonged G1 and G2 periods in late passages. Direct correlations between proliferative activity and cytochemical measurements on individual cells were not possible with the present experimental design and thus this could not be ascertained. The finding that there was a slight increase in the incidence of unmarked "small cells" at 48 h after subcultivation may suggest that at least some of the "small cell" non-dividing group may be capable of proliferation, but have a lower probability of entering the cell cycle [36]. The steadily increasing fraction of the small non-dividers with passage number in the cultures suggests that this cell group is of importance in the phase III phenomenon [37]. We thus conclude that "large cells" with abnormal cytological appearance and DNA content occur in cultures of glial cells with a stable probability which does not change with passage number. Most of this subpopulation is made up of non-dividers; however, it is conceivable that a great number of these cells may also be slow dividers. The "small" cells make up a subpopulation that is cytologically more difficult to distinguish from dividers and has a lower probability of entering the cell cycle with increased passage number. The latter subpopulation is probably the more interesting as regards the phase III phenomenon. ACKNOWLEDGEMENTS We are grateful to Prof. T. Caspersson for making available the use of his facilities and equipment. This study was supported by Stockholm's Cancer Society and the Swedish Cancer Society. REFERENCES 1 L. Hayflick, Senescence and cultured cells. Perspect. Exp. Gerontol., 14 (1966) 195-211. 2 B. Westermark, Growth control of normal and neoplastic human glia-like cells in culture. Acta Univ. Ups., 164 (1973).

11 3 U.T. Brunk, J.L.E. Ericsson, J. Pont6n and B. Westermark, Residual bodies and "aging" in cultured human glia cells. Effect of entrance into phase III and prolonged periods of confluence. Exp. Cell Res., 79 (1973) 1-14. 4 V.J. Cristofalo and B. Sharf, Cellular senescence and DNA synthesis. Exp. Cell Res., 76 (1973) 419-427. 5 P.D. Bowman, R.L. Meck and C.V. Daniel, Aging of human fibroblasts in vitro. Exp. Cell Res., 93 (1975) 184-190. 6 G. Merz and J. Ross, Clone size variation in the human diploid cell strain, WI-38. J. Cell. Physiol., 82 (1969) 75-80. 7 V.P. Collins, E. Arro, E. Blomquist, U.T. Brunk, B.-A. Fredriksson and B. Westermark, Cell motility and proliferation in relation to available substratum area, serum concentration and culture age. In O. Johari (ed.), Scanning Electron Microscopy, Vol. III, liT Research Institute, Chicago, 1979, pp. 411-420. 8 E. Blomquist, B. Westermark and J. Pont6n, Ageing of human glial ceils in culture: Increase in the fraction of non-dividers as demonstrated by a minicloning technique. Mech. Ageing Dev., 12 (1980) 173-182. 9 V.J. Cristofalo and D. Kritchevsky, Cell size and nucleic acid content in the diploid human cell line WI-38 during aging. Med. Exp., 19 (1969) 313-320. 10 G.B. Greenberg, G.L. Grove and V.J. Cristofalo, Cell size in ageing monolayer culture. In Vitro, 13 (1977) 297-300. 11 Y. Mitsui and E.L. Schneider, Increased nuclear size in senescent human diploid fibroblast culture. Exp. CellRes., 100 (1976) 147-152. 12 M.A. Brock and R.J. Hay, Comparative ultrastructure of chick fibroblasts in vitro at early and late stages during their growth span. J. Ultrastruct. Res., 36 (1971) 291-311. 13 J. Lipetz and V.J. Cristofalo, Ultrastructural changes accompanying the ageing of human diploid cells in culture. J. Ultrastruct. Res., 39 (1972) 43-56. 14 T. Robbins, E.M. Levine and H. Eagle, Morphological changes accompanying senescence of culture human diploid cells. J. Exp. Med., 131 (1970) 1211-1222. 15 V.P. Collins and U.T. Brunk, Quantitation of residual bodies in cultured human glial cells during stationary and logaritbmic growth phases. Mech. Ageing Dev., 8 (1978) 139-152. 16 V.P. Collins and H.H. Thaw, The measurement of lipid peroxidation products (lipofuscin) in individual cultivated human glial cells. Mech. Ageing Dev., 23 (1983) 199 214. 17 J. Pont6n, B. Westermark and R. Hugosson, Regulation of proliferation and movement of human glial like cells in culture. Exp. Cell Res., 58 (1969) 393--400. 18 J. Pont6n, Human glial cells. In P.F. Kruse and M.K. Patterson (eds.), Tissue Culture Methods and Applications, Academic Press, New York and London, 1973, pp. 50-53. 19 V.P. Collins, Cultured human glial and glioma cells. Int. Rev. Exp. Pathol., 24 (1983) 135-202. 20 V.P. Collins, B. Arborgh and U.T. Brunk, A comparison of the effects of three widely used glutaraldehyde fixatives on cellular volume and structure. Acta Pathol. Mierobiol. Scand. Sect. A, 85 (1977) 157-168. 21 V.P. Collins, U.T. Brunk, B. Westermark and B.-A. Fredriksson, Cell proliferation and plasma membrane motility. Stationary and proliferating human glia and glioma cells at various densities. In O. Johari (ed.), Scanning Electron Microscopy, Vol. II, l i t Research Institute, Chicago, 1977, pp. 1-11. 22 V.P. Collins and U.T. Brunk, Characterization of residual bodies formed in phase II cultivated human glia cells. Mech. AgeingDev., 5 (1976) 193-207. 23 J. Gaub, G. Auer and A. Zetterberg, Quantitative aspects of a combined Feulgen-naphthol Yellow S procedure for the simultaneous determination of nuclear and cytoplasmic proteins and DNA in mammalian cells. Exp. Cell Res., 92 (1975) 323-332. 24 A. Zetterberg and P.L. Esposti, Cytophotometric DNA analysis of aspirated cells from prostatic carcinoma. Acta Cytol., 20 (1976) 46-57. 25 T. Caspersson and L. Zech, Identification of mammalian chromosomes. Karolinska Symposia (5th Symposium). Acta Endocrinol. (Copenhagen) SuppL, 168 (1972) 65-83. 26 V.P. Collins, The re-entry of long-term density dependent growth-inhibited cells into the cell cycle. Cell Biol. Int. Rep., in press. 27 L. Hayflick, Recent advances in the cell biology of ageing. Mech. Ageing Dev., 14 (1980) 59-79.

12 28 Y. Mitsui and E.L. Schneider, Relationship between cell replication and volume in senescent human fibroblasts. Mech. Ageing Dev., 5 (1976) 45-56. 29 R. Holiday, L.I. Huschtscha, G.M. Tarrant and T.B.L. Kirkwood, Testing the committment theory of cellular ageing. Science, 198 (1977) 366-372. 30 T.B.L. Kirkwood and R. Holliday, A stochastic model for the committment of human cells to senescence. In A.J. Valleran and P.D.M. MacDonald (eds.), Biomathematics and Cell Kinetics, Elsevier/North-Holland, 1978, pp. 161-172. 31 R. Yanishevsky, L. Mortimer, L. Mendelsohn, B.H. Mayall and V.J. Cristofalo, Proliferative capacity and DNA content of aging human diploid cells in culture: A cytophotometric and autoradiographic analysis. J. Cell. Physiol., 84 (1974) 165-170. 32 Z.K. Kadanka, D.J. Sparkles and H.G. MacMorine, A study of the cytogenetics of the human cell strain Wl-38 in vitro. In Vitro, 8 (1973) 353-361. 33 R.C. Miller, W.W. Nichols, J. Pattash and M.M. Aronson, In vitro ageing. Exp. Cell Res., 110 (1977) 63-73. 34 A. Maciera-Coelho, E. Garcia-Giralt and M. Adrian, Changes in lysosomal associated structures in human fibroblasts kept in resting phase. Proc. Soc. Exp. Biol. Med., 138 (1971) 712-718. 35 A. Maciera-Coelho, J. Pont6n and L. Philipson, The division cycle and RNA-synthesis in diploid human cells at different passage levels in vitro. Exp. Cell Res., 42 (1966) 673-684. 36 J.A. Smith and L. Martin, Do cells cycle? Proc. Natl. Acad. Sci. USA, 70 (1973) 1263-1267. 37 S. Shall and W.D. Stein, A mortilization theory for the control of cell proliferation and for the origin of immortal cell lines. J. Theor. Biol., (1979) 219-231.