- Email: [email protected]
Aging dependent nucleolar and chromatin changes in cultivated fibroblasts
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
61
AGING DEPENDENT NUCLEOLAR AND CHROMATIN CHANGES IN CULTIVATED FIBROBLASTS F. PUVION-DUTILLEUL Institut de Recherches Scientifiques 94802 Villejuif C&?tex, France. A. MACIEIRA-COELHO Institut de Canc&ologie 94804 Villejuif &dex,
sur le Cancer,
et d'Immunog&&ique, France.
CNRS,
INSEAM U 50,
SUMMARY Adult fibroblasts have already at low population doubling level (PDL) a significant percentage of cells with modified chromatin as compared to embryonic cells where altered chromatin is seen only at high PDL. In fibroblast populations from a patient with Werner's
syndrome, the percentage
of cells
with both altered
chromatin
and
nucleoli was much higher than in the control cultures from normal donors. The data show for the first time nucleoprotein changes in cells from donors with an aging syndrome and reinforce the concept that serial replication of fibroblasts in vitso causes lesions
identical
to those of aging in vivo.
INTRODUCTION Human embryonic and adult fibroblasts are widely used as an in vitro model for aging research (Murphy, 1975). The model supposes that the changes observed after serial subcultivation of embryonic fibroblasts should be present to a certain extent in postnatal fibroblasts during the early stages of their lifespan in vitro; furthermore, these changes should be more pronounced in cell populations from older donors and from patients with aging syndromes. In fact, changes in cell proliferation (Goldstein, 1969; Macieira-Coelho and Pontgn, 1969; Martin et al., 1970; Schneider and Mitsui, 1976), in the cell membrane (Aizawa et al., 1980; Simons and Van den Broek, 1970), RNA content (BeMiller et al., 1981), enzyme thermolability (Goldstein and Singal, 1974; Holliday et al., 1974), radiosensitivity (Macieira-Coelho et al., 1978)~ DNA chain elongation (Fujiwara et al., 1977; Petes et al., 1974), and autofluorescence (Rattan et al., 1982), all have shown an identical evolution with in vitro and in vivo aging. Analysis of chromatin of serially subcultured human embryonic fibroblasts on alkaline sucrose gradients revealed an accumulation of fragile sites during cellular senescence (Icard et al., 1979). These findings were extended (Puvion-Dutilleul and Macieira-Coelho, 1982; Puvion-Dutilleul et al., 1982) at the ultrastructural level with a recently described technique that allows the partial loosening of nucleoproteins (Puvion-Dutilleul and Puvion, 1980). It showed during aging in vitro a less tufty appearance of the filamentous masses 0309-1651/83/010061-1
l/$03.00/0
@ 1983
Academic
Press
Inc. (London)
Ltd.
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
containing the nascent nucleolar RNA and a marked rarefaction of the chromatin threads. With Miller's spreading technique (Miller and Bakken, 1972), chromatin fibers were largely unbeaded and frequently of shorter size. These changes seem to be specifically associated with cellular aging since they have never been observed in a variety of biological situations (Puvion-Dutilleul and Puvion, 1980, 1981). We have now applied the same techniques to compare embryonic with postnatal fibroblasts from normal donors and from a donor with Werner's syndrome. MATERIALSAND METHODS and source of the cell lines Tables 1 and 2 describe the derivation used in the present report. Rnbryonic lung fibroblast line ICIG-7 and fetal skin 'fibroblast line ICIG-9 were started by ourselves from human embryos obtained after volontary abortion. MVL strain was obtained from the skin of an adult donor. AG 780, AG 4353 and AG 4525 were purchased from the Human Genetic Mutant Cell Repository (Camden,N.J., USA). All cultures were grown in 30-cm* plastic bottles in 7 ml minimum essential medium (MEM) supplemented with 10% fetal calf serum and 16 ug/ml gentsmycin. Cells were incubated at 37'C in the presence of 5% CO2 and serially passaged at I:2 split ratio. Confluent cultures in resting phase were prepared for electron microscopic observations. in 0.1 M Monolayers were fixed for 1 hour in 1.6% glutaraldehyde Sijrensen phosphate buffer (pH 7.4), washed in buffer, dehydrated in ethanol and embedded in Epon. For loosening procedure (Puvion-Dutilleul and Puvion, 1980), monolayers were washed in water adjusted to pH 9.0 with 0.1 M borate buffer and then covered with 1% formaldehy0.1 M sucrose and 0.4% Photo flo de aqueous solution containing (Kodak-Path6, France) and adjusted to pH 8.5-8.7 with the borate buffer. During the l-hour treatment, cells were scraped from the bottles and centrifuged. The translucent pellets were then dehydratea and embedded as above. In order to determine the possible effects on nuclear structures of the high pH and of the detergent of the loosening medium, the latter was also performed either at neutral pH or with a detergent-free solution (1% formaldehyde aqueous solution containing 0.1 M sucrose and adjusted to pH 8.5-8.7 with the borate buffer). Ultrathin sections were stained with uranyl acetate and lead citrate and examined on a Siemens Elmiskop 101 electron microscope at 80 kV. For the spreading procedure, monolayers were covered with 0.02 ml of 0.5% Joy detergent (Procter and Gamble, Cincinnati, OH, USA) adjusted to pH 8.5 with the borate buffer. The cells were detached from their substratum, diluted according to a slightly modified original techversion (Puvion-Dutilleul et al., 1977) of Miller's in a microcentrifuge nique (Miller and Bakken, 1972) and centrifuged chamber containing a freshly glow-discharged carbon-coated copper grid. Preparations were observed after rotary shadow casting with platinum at an angle of 6-8O. DNA was measured as described by Burton (1956). The radioactivity incorporated into DNA was measured adding to confluent cells O.S!.ICi /ml 'I&C-thymidine (specific activity : 50 mCi/mM) in fresh nutrient
Cell Biology
International
Reports,
Vol. 7, No. 1, January
63
1983
population douTable 1 : Frequency of altered nuclei at different bling levels (PDL) during the serial subcultivation of the human embryonic lung fibroblast line ICIG-7, revealed by the loosening procedure. Cell strain
Origin
ICIG-7
Fetal
% of sectioned nuclei wing altered nucleoli chromatin.
PDL lung
I
10th to 40th 41th to 49th 50th to 56th
shoand
0
Table 2 : Frequency of altered nuclei found after the loosening cedure, in low PDL cultures of human skin fibroblasts. % of sectioned showing Cell strain
Origin
PDL
altered nucleoli
ICIG-9
Fetal
4th
-
AG 4525
Fetal
6th
-
nuclei
altered chromatin
pro-
Maximal number of doublings.
Adult
(30 years)
MVL
12th
17
30
Adult AG 4353
(59 years)
6th
-
7
40
$?gears)
6th
89
89
19
AG 780 ~~~~~~l~
1
64
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
of confluent cultures in resting phase after Fig. 1 Monolayers loosening procedure. Comparison between nucleolar areas from the young type (a) and the old type (b) fibroblast. Comparison between peripheral nuclear regions from the young type (c) and the old type (d) fibroblast. Bars = 0.5 Urn.
Cell Biology International
Reports, Vol. 7, No. 1, January 1983
medium; after conventional fixation the cells were washed three times in pellet was digested in 1 ml Soluene CO.) was added and the radioactivity tillation counter.
65
or loosening as described above, 5% cold perchloric acid and the (Packard Co.). Instagel (Packard was measured in a.liquid scin-
RESULTS Two types of nuclear organization were observed in serially subcultured human embryonic lung fibroblasts corresponding to young and old cells respectively (Puvion-Dutilleul and Macieira-Coelho, 1982). The young-type nuclei displayed a chromatin which was condensed at the nuclear periphery and around the nucleolus as shown by standard electron microscopy fixation. The loosening procedure dissociated the nucleoproteins with a pattern similar to that regularly observed with other cell lines (Puvion-Dutilleul and Puvion, 1980, 198I), i.e. nucleolar transcription complexes contained within the fibrillar component of the nucleoli, appeared in thin sections as filamentous masses (Fig. la), while chromatin became dissociated into closely spaced threads measuring 20-25 nm in thickness (Fig. lc). The old-type nuclei, on the other hand, mainly observed at the end of the culture (Table l), were characterized by prominent nucleoli and rare condensed chromatin (Puvion-Dutilleul and Macieira-Coelho, 1982; Puvion-Dutilleul et al., 1982); other nuclear changes, also revealed by the loosening procedure, were related to the coarse aspect of the nucleolar masses due to the enlargement of the diameter of their entangled fibrils (Fig. lb). Indeed, the fibrils which constituted the masses of young-type nuclei, were only 6-7 nm thick whereas they reached lo-15 nm in old-type nuclei (compare Fig. lb to Fig. la), these nucleolar alterations were associated with a marked spacing of the chromatin threads (Fig. ld). The intervals separating the threads reached 0.5 urn in old-type nuclei whereas they were no larger than 0.1 urn in young-type nuclei (compare Fig. Id to Fig. lc). Another aspect of the nuclei of embryonic cells aged in vitro was an increased lobulation (Fig. 2).
Fig.
2 Multilobed procedure.
nucleus from AG 4353 cells Bar = 1 urn.
after
loosening
66
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
In the present work, skin fibroblasts from fetal and postnatal origin from normal donors and from a patient with Werner's syndrome (Table 2) at low PDL were screened for the frequency of the two types of nuclei described above as revealed by the mild loosening of nucleoproteins according to three criteria: altered nucleoli, altered chromatin and nuclear lobulation. Results presented in Tables 1 and 2 show that in low PDL fetal lung and skin fibroblasts there are no altered nuclei. In cultures of postnatal origin however there is a significant percentage of cells with old-type nuclei. Among the three criteria, altered chromatin was found in a low percentage of the nuclei of both normal donors at early PDL and nuclear lobulation only in the 59-year-old normal donor (Fig. 2). In these nuclei with altered chromatin, the nucleoli presented a normal pattern. In cells from the Werner's syndrome patient, however, 89% of the nuclei had altered chromatin associated with altered nucleoli. Nuclei were also heavily lobulated. No attempt was made to quantify lobulation since it depends on the plane of section. Chromatin fibers were analyzed in detail with Miller's spreading technique (1972) as illustrated in Fig. 3. DNA fibers from young type chromatin displayed a typical nucleosomal organization whereas in old-type chromatin most of the DNA fibers were punctuated by highly spaced nucleosomes or were entirely extended. In addition in spreads of Werner's syndrome nuclei, short pieces of unbeaded DNA fibers with lengths distributed between 0.1 and 0.2 urn were frequently seen, occasionally forming a circle. To see if the detergent present in the loosening medium was responsible for the nuclear structures of Werner's syndrome cells described above, these cells were also treated with a detergent-free This medium did not induce a good dispersion of the nucleosolution. lar components, however, chromatin was resolved into highly spaced threads similar to those obtained with a detergent-containing medium. In addition, to determine if the nuclear changes may result from the release of enzymes by the numerous lysosomes present in the cytoplasm of the Werner's syndrome cells, ICIG-7 cells at low PDL were mixed with Werner's syndrome cells previously suspended in the complete loosening medium. This treatment did not induce the rarefaction of the chromatin threads within the swollen nuclei of ICIG-7 cells demonstrating that the increased spacing of the chromatin threads was not an artefact produced either by the action of the detergent or by the cellular enzymes released by the hypotonic shock. The alkaline pH of the loosening medium was also not implicated since we have obtained similar changes in the nuclei of ICIG-7 cells at high PDL with a treatment performed at neutral pH. Measurements, before and after the loosening procedure, of the aincorporated mount of DNA (Burton, 1956) and of the radioactivity after 3H-TdR labelling, revealed that there was no loss of DNA during the preparation of the cells for electron microscopy.
Cell Biology
Fig.
3
international
Reports,
Spread preparations
Vol. 7, No. 1, January
of ICIG-9
(a),
1983
AC 4353 (b)andAG 4
67
780 (c 1
Beaded chromatin fibe 3x5 are abundant in young-type nuclei.
b. Most of the DNA of old-type nuclei is extended. However chromatin fibers with a "bead-on-a-string" configuration (empty arrow) could be found
C.
Most DNA fibers are fully extended without any nucleosomes in Werner's syndrome cells although typical beaded chromatin fibers can be found (empty arrow). In addition, short pieces of unbeaded DNA fibers are dispersed sometimes forming a circle (arrow). Bars = 0.5 urn.
68
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
DISCUSSION In our study, nuclear alterations were revt?aled by three electron microscopic techniques. Results shown by conventional fixation and related to the irregular nuclear shape and to the disappearance of marginated chromatin, especially in senescent and pathologic cell strains, were in agreement with several studies on matnmalian(Evans et al., 1978; Basler et al., 1979; Van Gansen et al., 1979; PuvionDutilleul and Macieira-Coelho, 1982; Puvion-Dutilleul et al., 1982) and avian (Brock and Hay, 1971) fibroblasts aged in vitro. The loosening procedure showed that unusual spacing of the chromatin threads which are present only in high PDL embryonic cells are already present in some of the cells of adult origin at low PDL and in almost all cells from the donor with pathologic aging. Disorganization of the nucleoli, however, typical of in vitro aging, was not found in the low PDL cells from normal donors but occurred always together with chromatin changes in fibroblasts from Werner's syndrome. These nuclear alterations seem to be specific of aging since they have never been found in a variety of conditions where the same techniques were used (Puvion-Dutilleul and Puvion, 1980, 1981). The abundance of unbeaded DNA fibers in Miller's spreads from old-type nuclei and of short unbeaded DNA fragments in spreads from Werner's syndrome cells suggested an increased fragility of the DNA with aging under our experimental conditions, resulting in a destabilization of most of the nucleosomes as well as breakage of the DNA fibers. The DNA fibers could be seen occasionally forming circles; an increase in circular DNA during fibroblast aging has been previously found (Shmookler Reis et al., 1982). For the best of our knowledge, these changes in chromatin fibers have never been reported before associated with aging syndromes. They agree with the previously described decline in the rate of DNA chain elongation in in vitro aged embryonic cells (Petes et al., 1974) and in Werner's syndrome fibroblasts (Fujiwara et al., 1977) and favor the idea that serial divisions of fibroblast-like cells originate changes identical to aging in vivo. ACKNOWLEGEMENTS The authors thank Mrs. C. Bournoutian, S. Lepelletier for their efficient technical assistance.
and J. Pgdron
This work was partly supported by grants from the Centre National de la Recherche Scientifique (A.T.P. no 60 82 751), the Fondation pour National de la Recherla Recherche MEdicale Franqaise, the Institut the Scientifique et MEdicale (no C.R.L. 802030), and EURATOM(B 10 -352-F (Sl)). REFERENCES Aizawa, S., Mitsui, Y., Kurimoto, F. and Matsuoka, K. (1980). Cell surface changes accompanying aging in human diploid fibroage and senescence revealed by concablasts. Division, navalin A-mediated blood cell adsorption.
Ceil Biology
international
Reports,
Vol. 7, No. 1, January
Mechanisms of Ageing and Development,
l3,
69
1983 297-306.
Basler,
J.W., David, J.D. and Agris, P.F. (1979). Deteriorating collagen synthesis and cell ultrastructure accompanying senescence of human normal and Werner's syndrome fibroblast Experimental Cell Research, 118, 73-84. cell strains.
BeMiller,
J.N. and Pappelis, A.J. P.M., Lee, S.C.L., BeMiller, Determination of the nuclear RNA content of cells from donors of three different ages during in vitro senescence. Mechanisms of Ageing and Development, l5, 349-354. (1981).
Brock, M.A. and Hay, R.J. (1971). Comparative ultrastructure of chick fibroblasts in vitro at early and late stages during their growth span. Journal of Ultrastructural Research, 3, 291-311.
Burton,
K. (1956). A study of the conditions and mechanisms of the diphenylsmine reaction for the calorimetric estimation of deoxyribonucleic acid. Biochemical Journal, 62, 315-323.
Evans, C.H., Van Gansen, P. and Rasson, I. (1978). Transcription in early and late passage embryonic mouse fibroblasts in pria correlated biochemical autoradiographical mary cultures: and ultrastructural study. Biologie Cellulaire, 33, 117-128. Fujiwara, Y., Higashikawa, T. and Tatsumi, M. (1977). A retarded rate of DNA replication and normal level of DNA repair in Werner's syndrome fibroblasts in culture. Journal of Cellular Physiology, 92, 365-374. Goldstein,
S. (1969). Lifespan Lancet, I, 424.
of cultured
cells
in Progeria.
The
Goldstein,
S. and Singal, D.P. (1974). Alteration of fibroblast gene products in vitro from a subject with Werner's syndrome. Nature, 251, 719-721.
Hayflick,
L. (1965). cell strains.
The limited in vitro lifetime of human diploid Experimental Cell Research, 37, 614-636.
Holliday,
R., Porterfields, J.S. and Gibbs, D.D. (1974). Premature ageing and occurence of altered enzyme in Werner's syndrome fibroblasts. Nature, 248, 762-763. Icard, C., Beaupain, R., Diatloff, C. and Macieira-Coelho, A. (1979). Effect of low dose rate irradiation on the division potential of cells in vitro, VI. Changes in DNA and in radiosensitivity during aging of human fibroblasts. Mechanisms of Ageing and Development, 11, 269-278. Macieira-Coelho, A., Diatloff, C., Billard, M., Fertil, B., Malaise, E. and Fries, D. (1978). Effect of low dose rate irradiation on the division potential of cells in vitro. IV.Rnbryonit and adult human lung fibroblast-like cells. Journal of Cellular Physiology, 95, 235-238. Macieira-Coelho,
A. and Pont&,
J. (1969).
Analogy in growth between
Cell Biology
70
International
Reports,
Vol. 7, No. 1, January
1983
late passage human embryonic and early passage human adult fibroblasts. Journal of Cell Biology, 43, 374-377. Martin,
G.M., Sprague, C.A. and Epstein, C.J. (1970). Replicative lifespan of cultivated human cells. Laboratory Investigation, 23, 86-92.
Miller,
Morphological studies of Jr., O.L. and Bakken, B.R. (1972). transcription. In: E. Diczfaluzy (ea.) Gene Transcription Karolinska Institutet, in Reproductive?issue. pp. 155-177. Stockholm.
Murphy, D.G. (1975). NIH gerontology research support: policy evaluation of in vitro cell lineage studies. Mechanisms of Ageing and Development, 4, 317-323. Petes, T.D., Harberd, R.A., Tarrant, G.M. and Holliday, R. (1974). Altered rate of DNA replication in ageing human fibroblasts. Nature, a, 434-436. Puvion-Dutilleul, F., Azzarone, B. and Macieira-Coelho, A. (1982). Comparison between proliferative changes and nuclear events during ageing of human fibroblasts in vitro. Mechanisms of Ageing and Development (in press). Puvion-Dutilleul,
F., Bernadac, A., Puvion, E. and Bernhard, W. Visualization of two different types of nuclear transcriptional complexes in rat liver cells. Journal of Ultrastructural Research, 58, 107-117. (1977).
Puvion-Dutilleul, F. and Macieira-Coelho, A. (1982). Ultrastructural organization of nucleoproteins during aging of cultured human embryonic fibroblasts. Experimental Cell Research, 138, 423-429. New aspects of intraPuvion-Dutilleul, F. and Puvion, E. (1980). cellular structures following partial decondensation of autoradiochromatin: a cytochemical and high-resolution graphical study. Journal of Cell Science, 42, 305-322. Relationship between Puvion-Dutilleul, F. and Puvion, E. (1981). chromatin and perichromatin granules in cadmium-treated isolated hepatocytes. Journal of Ultrastructural Research, 74, 341-350. R. Rattan, F.I.S., Keeler, K.D., Buchanan, J.H. and Holliday, (1982). Autofluorescence as an index of aging in human Bioscience Reports (in press). fibroblasts in culture. Shmookler Reis, R.J., Lumpkin, C.K., McGill, J.R., Riabowol, K.T. S. (1982). Genome instability during in and Goldstein, of extrachromzomal vitro and in vivo aging: amplification sequence of variable repeat frequency. Cold Spring Harbor Symposium on Quantitative Biology (in press). Schneider, E.L. and Mitsui, Y. (1976). The relationship between in vitro cellular aging and in vivo human age. Proceedings of the National Academy of Sciences, USA, 73, 3584-3588.
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
71
Simons, J.W. and Van den Broek (1970). Comparison of ageing in vitro and ageing in vivo by means of cell size analysis using a Coulter counter. Gerontologin, l6, 340-351. Van Gansen, P., Devos, L., Ozoran, Y. and Roxburgh, C. (1979). Phenotypes des fibroblastes d'embryons de souris vieillissant in vitro (SEM et TEM). Biologie Cellulaire, 34, 255-270.
Received:
21st
October
1982
Accepted:
12th November 1982
International
Reports,
Vol. 7, No. 1, January
1983
61
AGING DEPENDENT NUCLEOLAR AND CHROMATIN CHANGES IN CULTIVATED FIBROBLASTS F. PUVION-DUTILLEUL Institut de Recherches Scientifiques 94802 Villejuif C&?tex, France. A. MACIEIRA-COELHO Institut de Canc&ologie 94804 Villejuif &dex,
sur le Cancer,
et d'Immunog&&ique, France.
CNRS,
INSEAM U 50,
SUMMARY Adult fibroblasts have already at low population doubling level (PDL) a significant percentage of cells with modified chromatin as compared to embryonic cells where altered chromatin is seen only at high PDL. In fibroblast populations from a patient with Werner's
syndrome, the percentage
of cells
with both altered
chromatin
and
nucleoli was much higher than in the control cultures from normal donors. The data show for the first time nucleoprotein changes in cells from donors with an aging syndrome and reinforce the concept that serial replication of fibroblasts in vitso causes lesions
identical
to those of aging in vivo.
INTRODUCTION Human embryonic and adult fibroblasts are widely used as an in vitro model for aging research (Murphy, 1975). The model supposes that the changes observed after serial subcultivation of embryonic fibroblasts should be present to a certain extent in postnatal fibroblasts during the early stages of their lifespan in vitro; furthermore, these changes should be more pronounced in cell populations from older donors and from patients with aging syndromes. In fact, changes in cell proliferation (Goldstein, 1969; Macieira-Coelho and Pontgn, 1969; Martin et al., 1970; Schneider and Mitsui, 1976), in the cell membrane (Aizawa et al., 1980; Simons and Van den Broek, 1970), RNA content (BeMiller et al., 1981), enzyme thermolability (Goldstein and Singal, 1974; Holliday et al., 1974), radiosensitivity (Macieira-Coelho et al., 1978)~ DNA chain elongation (Fujiwara et al., 1977; Petes et al., 1974), and autofluorescence (Rattan et al., 1982), all have shown an identical evolution with in vitro and in vivo aging. Analysis of chromatin of serially subcultured human embryonic fibroblasts on alkaline sucrose gradients revealed an accumulation of fragile sites during cellular senescence (Icard et al., 1979). These findings were extended (Puvion-Dutilleul and Macieira-Coelho, 1982; Puvion-Dutilleul et al., 1982) at the ultrastructural level with a recently described technique that allows the partial loosening of nucleoproteins (Puvion-Dutilleul and Puvion, 1980). It showed during aging in vitro a less tufty appearance of the filamentous masses 0309-1651/83/010061-1
l/$03.00/0
@ 1983
Academic
Press
Inc. (London)
Ltd.
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
containing the nascent nucleolar RNA and a marked rarefaction of the chromatin threads. With Miller's spreading technique (Miller and Bakken, 1972), chromatin fibers were largely unbeaded and frequently of shorter size. These changes seem to be specifically associated with cellular aging since they have never been observed in a variety of biological situations (Puvion-Dutilleul and Puvion, 1980, 1981). We have now applied the same techniques to compare embryonic with postnatal fibroblasts from normal donors and from a donor with Werner's syndrome. MATERIALSAND METHODS and source of the cell lines Tables 1 and 2 describe the derivation used in the present report. Rnbryonic lung fibroblast line ICIG-7 and fetal skin 'fibroblast line ICIG-9 were started by ourselves from human embryos obtained after volontary abortion. MVL strain was obtained from the skin of an adult donor. AG 780, AG 4353 and AG 4525 were purchased from the Human Genetic Mutant Cell Repository (Camden,N.J., USA). All cultures were grown in 30-cm* plastic bottles in 7 ml minimum essential medium (MEM) supplemented with 10% fetal calf serum and 16 ug/ml gentsmycin. Cells were incubated at 37'C in the presence of 5% CO2 and serially passaged at I:2 split ratio. Confluent cultures in resting phase were prepared for electron microscopic observations. in 0.1 M Monolayers were fixed for 1 hour in 1.6% glutaraldehyde Sijrensen phosphate buffer (pH 7.4), washed in buffer, dehydrated in ethanol and embedded in Epon. For loosening procedure (Puvion-Dutilleul and Puvion, 1980), monolayers were washed in water adjusted to pH 9.0 with 0.1 M borate buffer and then covered with 1% formaldehy0.1 M sucrose and 0.4% Photo flo de aqueous solution containing (Kodak-Path6, France) and adjusted to pH 8.5-8.7 with the borate buffer. During the l-hour treatment, cells were scraped from the bottles and centrifuged. The translucent pellets were then dehydratea and embedded as above. In order to determine the possible effects on nuclear structures of the high pH and of the detergent of the loosening medium, the latter was also performed either at neutral pH or with a detergent-free solution (1% formaldehyde aqueous solution containing 0.1 M sucrose and adjusted to pH 8.5-8.7 with the borate buffer). Ultrathin sections were stained with uranyl acetate and lead citrate and examined on a Siemens Elmiskop 101 electron microscope at 80 kV. For the spreading procedure, monolayers were covered with 0.02 ml of 0.5% Joy detergent (Procter and Gamble, Cincinnati, OH, USA) adjusted to pH 8.5 with the borate buffer. The cells were detached from their substratum, diluted according to a slightly modified original techversion (Puvion-Dutilleul et al., 1977) of Miller's in a microcentrifuge nique (Miller and Bakken, 1972) and centrifuged chamber containing a freshly glow-discharged carbon-coated copper grid. Preparations were observed after rotary shadow casting with platinum at an angle of 6-8O. DNA was measured as described by Burton (1956). The radioactivity incorporated into DNA was measured adding to confluent cells O.S!.ICi /ml 'I&C-thymidine (specific activity : 50 mCi/mM) in fresh nutrient
Cell Biology
International
Reports,
Vol. 7, No. 1, January
63
1983
population douTable 1 : Frequency of altered nuclei at different bling levels (PDL) during the serial subcultivation of the human embryonic lung fibroblast line ICIG-7, revealed by the loosening procedure. Cell strain
Origin
ICIG-7
Fetal
% of sectioned nuclei wing altered nucleoli chromatin.
PDL lung
I
10th to 40th 41th to 49th 50th to 56th
shoand
0
Table 2 : Frequency of altered nuclei found after the loosening cedure, in low PDL cultures of human skin fibroblasts. % of sectioned showing Cell strain
Origin
PDL
altered nucleoli
ICIG-9
Fetal
4th
-
AG 4525
Fetal
6th
-
nuclei
altered chromatin
pro-
Maximal number of doublings.
Adult
(30 years)
MVL
12th
17
30
Adult AG 4353
(59 years)
6th
-
7
40
$?gears)
6th
89
89
19
AG 780 ~~~~~~l~
1
64
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
of confluent cultures in resting phase after Fig. 1 Monolayers loosening procedure. Comparison between nucleolar areas from the young type (a) and the old type (b) fibroblast. Comparison between peripheral nuclear regions from the young type (c) and the old type (d) fibroblast. Bars = 0.5 Urn.
Cell Biology International
Reports, Vol. 7, No. 1, January 1983
medium; after conventional fixation the cells were washed three times in pellet was digested in 1 ml Soluene CO.) was added and the radioactivity tillation counter.
65
or loosening as described above, 5% cold perchloric acid and the (Packard Co.). Instagel (Packard was measured in a.liquid scin-
RESULTS Two types of nuclear organization were observed in serially subcultured human embryonic lung fibroblasts corresponding to young and old cells respectively (Puvion-Dutilleul and Macieira-Coelho, 1982). The young-type nuclei displayed a chromatin which was condensed at the nuclear periphery and around the nucleolus as shown by standard electron microscopy fixation. The loosening procedure dissociated the nucleoproteins with a pattern similar to that regularly observed with other cell lines (Puvion-Dutilleul and Puvion, 1980, 198I), i.e. nucleolar transcription complexes contained within the fibrillar component of the nucleoli, appeared in thin sections as filamentous masses (Fig. la), while chromatin became dissociated into closely spaced threads measuring 20-25 nm in thickness (Fig. lc). The old-type nuclei, on the other hand, mainly observed at the end of the culture (Table l), were characterized by prominent nucleoli and rare condensed chromatin (Puvion-Dutilleul and Macieira-Coelho, 1982; Puvion-Dutilleul et al., 1982); other nuclear changes, also revealed by the loosening procedure, were related to the coarse aspect of the nucleolar masses due to the enlargement of the diameter of their entangled fibrils (Fig. lb). Indeed, the fibrils which constituted the masses of young-type nuclei, were only 6-7 nm thick whereas they reached lo-15 nm in old-type nuclei (compare Fig. lb to Fig. la), these nucleolar alterations were associated with a marked spacing of the chromatin threads (Fig. ld). The intervals separating the threads reached 0.5 urn in old-type nuclei whereas they were no larger than 0.1 urn in young-type nuclei (compare Fig. Id to Fig. lc). Another aspect of the nuclei of embryonic cells aged in vitro was an increased lobulation (Fig. 2).
Fig.
2 Multilobed procedure.
nucleus from AG 4353 cells Bar = 1 urn.
after
loosening
66
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
In the present work, skin fibroblasts from fetal and postnatal origin from normal donors and from a patient with Werner's syndrome (Table 2) at low PDL were screened for the frequency of the two types of nuclei described above as revealed by the mild loosening of nucleoproteins according to three criteria: altered nucleoli, altered chromatin and nuclear lobulation. Results presented in Tables 1 and 2 show that in low PDL fetal lung and skin fibroblasts there are no altered nuclei. In cultures of postnatal origin however there is a significant percentage of cells with old-type nuclei. Among the three criteria, altered chromatin was found in a low percentage of the nuclei of both normal donors at early PDL and nuclear lobulation only in the 59-year-old normal donor (Fig. 2). In these nuclei with altered chromatin, the nucleoli presented a normal pattern. In cells from the Werner's syndrome patient, however, 89% of the nuclei had altered chromatin associated with altered nucleoli. Nuclei were also heavily lobulated. No attempt was made to quantify lobulation since it depends on the plane of section. Chromatin fibers were analyzed in detail with Miller's spreading technique (1972) as illustrated in Fig. 3. DNA fibers from young type chromatin displayed a typical nucleosomal organization whereas in old-type chromatin most of the DNA fibers were punctuated by highly spaced nucleosomes or were entirely extended. In addition in spreads of Werner's syndrome nuclei, short pieces of unbeaded DNA fibers with lengths distributed between 0.1 and 0.2 urn were frequently seen, occasionally forming a circle. To see if the detergent present in the loosening medium was responsible for the nuclear structures of Werner's syndrome cells described above, these cells were also treated with a detergent-free This medium did not induce a good dispersion of the nucleosolution. lar components, however, chromatin was resolved into highly spaced threads similar to those obtained with a detergent-containing medium. In addition, to determine if the nuclear changes may result from the release of enzymes by the numerous lysosomes present in the cytoplasm of the Werner's syndrome cells, ICIG-7 cells at low PDL were mixed with Werner's syndrome cells previously suspended in the complete loosening medium. This treatment did not induce the rarefaction of the chromatin threads within the swollen nuclei of ICIG-7 cells demonstrating that the increased spacing of the chromatin threads was not an artefact produced either by the action of the detergent or by the cellular enzymes released by the hypotonic shock. The alkaline pH of the loosening medium was also not implicated since we have obtained similar changes in the nuclei of ICIG-7 cells at high PDL with a treatment performed at neutral pH. Measurements, before and after the loosening procedure, of the aincorporated mount of DNA (Burton, 1956) and of the radioactivity after 3H-TdR labelling, revealed that there was no loss of DNA during the preparation of the cells for electron microscopy.
Cell Biology
Fig.
3
international
Reports,
Spread preparations
Vol. 7, No. 1, January
of ICIG-9
(a),
1983
AC 4353 (b)andAG 4
67
780 (c 1
Beaded chromatin fibe 3x5 are abundant in young-type nuclei.
b. Most of the DNA of old-type nuclei is extended. However chromatin fibers with a "bead-on-a-string" configuration (empty arrow) could be found
C.
Most DNA fibers are fully extended without any nucleosomes in Werner's syndrome cells although typical beaded chromatin fibers can be found (empty arrow). In addition, short pieces of unbeaded DNA fibers are dispersed sometimes forming a circle (arrow). Bars = 0.5 urn.
68
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
DISCUSSION In our study, nuclear alterations were revt?aled by three electron microscopic techniques. Results shown by conventional fixation and related to the irregular nuclear shape and to the disappearance of marginated chromatin, especially in senescent and pathologic cell strains, were in agreement with several studies on matnmalian(Evans et al., 1978; Basler et al., 1979; Van Gansen et al., 1979; PuvionDutilleul and Macieira-Coelho, 1982; Puvion-Dutilleul et al., 1982) and avian (Brock and Hay, 1971) fibroblasts aged in vitro. The loosening procedure showed that unusual spacing of the chromatin threads which are present only in high PDL embryonic cells are already present in some of the cells of adult origin at low PDL and in almost all cells from the donor with pathologic aging. Disorganization of the nucleoli, however, typical of in vitro aging, was not found in the low PDL cells from normal donors but occurred always together with chromatin changes in fibroblasts from Werner's syndrome. These nuclear alterations seem to be specific of aging since they have never been found in a variety of conditions where the same techniques were used (Puvion-Dutilleul and Puvion, 1980, 1981). The abundance of unbeaded DNA fibers in Miller's spreads from old-type nuclei and of short unbeaded DNA fragments in spreads from Werner's syndrome cells suggested an increased fragility of the DNA with aging under our experimental conditions, resulting in a destabilization of most of the nucleosomes as well as breakage of the DNA fibers. The DNA fibers could be seen occasionally forming circles; an increase in circular DNA during fibroblast aging has been previously found (Shmookler Reis et al., 1982). For the best of our knowledge, these changes in chromatin fibers have never been reported before associated with aging syndromes. They agree with the previously described decline in the rate of DNA chain elongation in in vitro aged embryonic cells (Petes et al., 1974) and in Werner's syndrome fibroblasts (Fujiwara et al., 1977) and favor the idea that serial divisions of fibroblast-like cells originate changes identical to aging in vivo. ACKNOWLEGEMENTS The authors thank Mrs. C. Bournoutian, S. Lepelletier for their efficient technical assistance.
and J. Pgdron
This work was partly supported by grants from the Centre National de la Recherche Scientifique (A.T.P. no 60 82 751), the Fondation pour National de la Recherla Recherche MEdicale Franqaise, the Institut the Scientifique et MEdicale (no C.R.L. 802030), and EURATOM(B 10 -352-F (Sl)). REFERENCES Aizawa, S., Mitsui, Y., Kurimoto, F. and Matsuoka, K. (1980). Cell surface changes accompanying aging in human diploid fibroage and senescence revealed by concablasts. Division, navalin A-mediated blood cell adsorption.
Ceil Biology
international
Reports,
Vol. 7, No. 1, January
Mechanisms of Ageing and Development,
l3,
69
1983 297-306.
Basler,
J.W., David, J.D. and Agris, P.F. (1979). Deteriorating collagen synthesis and cell ultrastructure accompanying senescence of human normal and Werner's syndrome fibroblast Experimental Cell Research, 118, 73-84. cell strains.
BeMiller,
J.N. and Pappelis, A.J. P.M., Lee, S.C.L., BeMiller, Determination of the nuclear RNA content of cells from donors of three different ages during in vitro senescence. Mechanisms of Ageing and Development, l5, 349-354. (1981).
Brock, M.A. and Hay, R.J. (1971). Comparative ultrastructure of chick fibroblasts in vitro at early and late stages during their growth span. Journal of Ultrastructural Research, 3, 291-311.
Burton,
K. (1956). A study of the conditions and mechanisms of the diphenylsmine reaction for the calorimetric estimation of deoxyribonucleic acid. Biochemical Journal, 62, 315-323.
Evans, C.H., Van Gansen, P. and Rasson, I. (1978). Transcription in early and late passage embryonic mouse fibroblasts in pria correlated biochemical autoradiographical mary cultures: and ultrastructural study. Biologie Cellulaire, 33, 117-128. Fujiwara, Y., Higashikawa, T. and Tatsumi, M. (1977). A retarded rate of DNA replication and normal level of DNA repair in Werner's syndrome fibroblasts in culture. Journal of Cellular Physiology, 92, 365-374. Goldstein,
S. (1969). Lifespan Lancet, I, 424.
of cultured
cells
in Progeria.
The
Goldstein,
S. and Singal, D.P. (1974). Alteration of fibroblast gene products in vitro from a subject with Werner's syndrome. Nature, 251, 719-721.
Hayflick,
L. (1965). cell strains.
The limited in vitro lifetime of human diploid Experimental Cell Research, 37, 614-636.
Holliday,
R., Porterfields, J.S. and Gibbs, D.D. (1974). Premature ageing and occurence of altered enzyme in Werner's syndrome fibroblasts. Nature, 248, 762-763. Icard, C., Beaupain, R., Diatloff, C. and Macieira-Coelho, A. (1979). Effect of low dose rate irradiation on the division potential of cells in vitro, VI. Changes in DNA and in radiosensitivity during aging of human fibroblasts. Mechanisms of Ageing and Development, 11, 269-278. Macieira-Coelho, A., Diatloff, C., Billard, M., Fertil, B., Malaise, E. and Fries, D. (1978). Effect of low dose rate irradiation on the division potential of cells in vitro. IV.Rnbryonit and adult human lung fibroblast-like cells. Journal of Cellular Physiology, 95, 235-238. Macieira-Coelho,
A. and Pont&,
J. (1969).
Analogy in growth between
Cell Biology
70
International
Reports,
Vol. 7, No. 1, January
1983
late passage human embryonic and early passage human adult fibroblasts. Journal of Cell Biology, 43, 374-377. Martin,
G.M., Sprague, C.A. and Epstein, C.J. (1970). Replicative lifespan of cultivated human cells. Laboratory Investigation, 23, 86-92.
Miller,
Morphological studies of Jr., O.L. and Bakken, B.R. (1972). transcription. In: E. Diczfaluzy (ea.) Gene Transcription Karolinska Institutet, in Reproductive?issue. pp. 155-177. Stockholm.
Murphy, D.G. (1975). NIH gerontology research support: policy evaluation of in vitro cell lineage studies. Mechanisms of Ageing and Development, 4, 317-323. Petes, T.D., Harberd, R.A., Tarrant, G.M. and Holliday, R. (1974). Altered rate of DNA replication in ageing human fibroblasts. Nature, a, 434-436. Puvion-Dutilleul, F., Azzarone, B. and Macieira-Coelho, A. (1982). Comparison between proliferative changes and nuclear events during ageing of human fibroblasts in vitro. Mechanisms of Ageing and Development (in press). Puvion-Dutilleul,
F., Bernadac, A., Puvion, E. and Bernhard, W. Visualization of two different types of nuclear transcriptional complexes in rat liver cells. Journal of Ultrastructural Research, 58, 107-117. (1977).
Puvion-Dutilleul, F. and Macieira-Coelho, A. (1982). Ultrastructural organization of nucleoproteins during aging of cultured human embryonic fibroblasts. Experimental Cell Research, 138, 423-429. New aspects of intraPuvion-Dutilleul, F. and Puvion, E. (1980). cellular structures following partial decondensation of autoradiochromatin: a cytochemical and high-resolution graphical study. Journal of Cell Science, 42, 305-322. Relationship between Puvion-Dutilleul, F. and Puvion, E. (1981). chromatin and perichromatin granules in cadmium-treated isolated hepatocytes. Journal of Ultrastructural Research, 74, 341-350. R. Rattan, F.I.S., Keeler, K.D., Buchanan, J.H. and Holliday, (1982). Autofluorescence as an index of aging in human Bioscience Reports (in press). fibroblasts in culture. Shmookler Reis, R.J., Lumpkin, C.K., McGill, J.R., Riabowol, K.T. S. (1982). Genome instability during in and Goldstein, of extrachromzomal vitro and in vivo aging: amplification sequence of variable repeat frequency. Cold Spring Harbor Symposium on Quantitative Biology (in press). Schneider, E.L. and Mitsui, Y. (1976). The relationship between in vitro cellular aging and in vivo human age. Proceedings of the National Academy of Sciences, USA, 73, 3584-3588.
Cell Biology
International
Reports,
Vol. 7, No. 1, January
1983
71
Simons, J.W. and Van den Broek (1970). Comparison of ageing in vitro and ageing in vivo by means of cell size analysis using a Coulter counter. Gerontologin, l6, 340-351. Van Gansen, P., Devos, L., Ozoran, Y. and Roxburgh, C. (1979). Phenotypes des fibroblastes d'embryons de souris vieillissant in vitro (SEM et TEM). Biologie Cellulaire, 34, 255-270.
Received:
21st
October
1982
Accepted:
12th November 1982