,1. gen. ViroL (1979), 42, 4o9-414
409
Printed in Great Britain
SV40-Transformed Human Diploid Cells that Remain
Transformed throughout Their Limited Lifespan
(Accepted 13 September I978)
SUMMARY
Of ninety-one subcultured loci of SV4o-transformed WI-3 8 human diploid
fibroblasts, none yielded cells that grew indefinitely, but all cells in each subculture
continued to produce T antigen and to look morphologically transformed throughout their lifespan. These results are consistent with the commitment theory of
fibroblast senescence, but predict that a special transformation event is necessary
to account for the rare survivors.
Cells transformed by simian virus 4° (SV4o) show the profound changes in growth and
physiology characteristic of the transformed state (Shein & Enders, 1962; Risser & Pollack,
I974). It has not been determined, however, whether these changes are related to the
transition of normal ceils to indefinite growth. This ambiguity has persisted because
transformation studies have most often started with established, indefinitely growing cell
lines (like BALB c/3T3) or with cells easily established in culture (Syrian hamster, for
example) which have tumorigenic capacities of their own (Earle, 1943 ; Stoker & Macpherson,
1964).
Ordinarily, diploid human fibroblasts cultured in vitro have a limited lifespan before
they deteriorate, cease dividing and die (Hayflick & Moorhead, 1961). Koprowski et al.
(I966a, b) have shown that SV4o infection of fibroblast cultures will induce the survival
of rare cells which can subsequently grow into permanent cell lines (see also Potter et al.
197o; Shein & Enders, I962; Todaro et aL 1966). However, it is not known what connection exists between the 'rescue' of the dying fibroblasts and transformation per se.
One possible explanation for the rarity of surviving cells, despite the relative frequency
of apparent transformants, is that the overwhelming majority of these loci are abortive
transformants. The abortive transformants would lose the transformed phenotype before
they reached crisis and thus cease cellular division much as uninfected cells. As a first step
in testing this possibility, we infected human diploid fibroblasts, identified and subcultured
apparently transformed loci and observed their properties as they approached crisis. The
results reported here have a bearing on some current models for cellular senescence and
argue against abortive transformation as a reason for the rarity of transformed cells that
grow indefinitely.
Normal human diploid fabroblasts (WI-38) were obtained at passage 18 from the American
Type Culture Collection and inoculated into 6o mm Petri dishes (6 × lOs ceils per dish).
The cells were grown in alpha-modified Eagle's minimal essential medium (MEM-atpha)
supplemented with zo % foetal calf serum (FCS: K.C. Biological, Inc., Lenexa, Kan.) in
a 5 % COz, 95 % air environment at 37 °C. The medium was changed every other day.
At passage 26, based on 1:2 split ratios, the medium was removed and 5 × IOv p.f.u, of
SV4o (obtained as a plaque-purified small plaque variant from Dr D. Nathans, Johns
Hopkins University, Baltimore) were added in o'5 ml MEM-alpha at a multiplicity of
5° p.f.u./cell. After an adsorption period of 3 h, the virus suspension was removed, fresh
oo22-1317/79/oooo-3316 $o2.oo© 1979 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:20:41
410
Short communications
ilfl ....
25
2o
1o
' [Itll
,,l
38
42
I,[ , | , |
62 64
46
50
54
58
Culture generation number
Fig. I. Lifespan of subcultures of SV4o-transformed WI-38 Cells. Ninety-one transformed WI-38
foci were subcultured and followed until division ceased. The histogram represents the number
of foci whose subcultures died at each culture generation number, Three foci inadvertently lost at
generation 45 were excluded from the figure.
medium added and incubation resumed. BALB c/3T 3 cells, grown in Dulbecco's modified
Eagle's medium supplemented with IO ~ calf serum (Colorado Serum Co., Denver, Colo.),
were infected and handled in a similar manner to serve as transformation controls.
Three to 6 weeks after treatment of subconfluent WI-38 cells with virus, 91 well-separated
foci of heaped-up cells were identified morphologically, 5 to Io on each dish. Each focus
was encircled with a 4 mm glass cylinder and the ceils were trypsinized and transferred to
a single 6 mm well in a Falcon microtitre plate. When cells had reached confluence in the
6 mm well, they were split 1:2, a ratio that was maintained as long as the subcultures
continued to grow and expand to larger cultures. The ' cell age' of each subcultured focus
was conveniently estimated as the total number of passages at I : z dilution, augmented by
the approximate number of divisions required for the initial encircled cells to reach
confluence in a 6 mm microtitre well. The latter number was calculated from: ( 0 The
number of confluent established SV4o-transformed WI-38 cells [WI-38(K) cells], kindly
supplied by Dr H. Koprowski (Wistar Institute, Philadelphia, Pa.), contained in the area
of the cloning cylinder; and (2) the assumption that each focus originated from a single
transformed cell. Because each focus certainly also contained non-transformed cells of
lower viability, and because the plating efficiency was only 3o to 6o ~ , the total number
of cell divisions is likely underestimated. However, this underestimation does not affect
the inferences in this study.
During subculture, cells were grown in conditioned medium which contained anti-SV4o
serum to prevent re-infection by any virus particles shed during growth. Conditioned
medium was prepared by plating z x Io 7 WI-38(K) cells (or WI-38 cells younger than culture
generation z2) in a flask and removing the medium 24 h later (Holley & Kiernan, I968 ).
This medium was filtered through a Millipore filter (o.45 #m) and supplemented with
I ml goat anti-SV4o serum (titre I : Iooo; Biological Assoc., Bethesda, Md.) per litre before use.
Of the 9I loci isolated, only 46 grew sufficiently to be expanded to a confluent cuIture in
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:20:41
Short communications
i
i
I
I
"1
i
4~
I
i
I
I
L
I
i
i
|
10
15
20
25
38
D
P
~
42
I
I
46
I
.I
50
54
58
Culture generation number
I
62
t
Fig. z. Subculture doubling times as a function of culture age. Culture doubling times versus cell
age for four representative subcultured foci (A to D) are plotted. Subculture A was followed b o t h
continuously ( 0 - - 0 ) and after four m o n t h s storage in liquid nitrogen at culture generation 54
(x --x).
a 60 mm Petri dish. In general, the fastest-growing (earliest-appearing) foci were the most
successfully subcultured. These foci could be subcultured at dilutions of up to I:5o, while
the slower growing foci died out even after they were plated without dilution.
The initial criterion for transformation was morphological; every focus showed characteristics of transformed cells comparable to WI-38(K ) cells. These cells also contained
SV4o T-antigen, which was not detected in control WI-38 cells.
[The T-antigen positivity was scored by growing cells on cover slips and staining for
indirect immunofluorescence (Pope & Rowe, I964); reagents, including hamster anti-SV4o
T-antigen serum and fluorescein-eonjugated rabbit anti-hamster serum, were kindly
supplied by Dr K. Takemoto (NIAID, NIH, Bethesda, Md.)]. Saturation density, a third
criterion for transformation, was also increased in these subcultures as compared to
control cultures. For example, at confluency, untransformed WI-38 cells numbered 9"3 × Io~
cells/cm~; WI-38(K ) cells, 3"6 × Io 5 cells/cm"-; and a subculture of a typical transformed
focus (52) averaged 3"8 + o-I × ]o ~ cells/era 2 at culture generations 49, 53, 55 and 57.
None of the apparently transformed SV4o-infected human fibroblast foci were able to
grow in soft agar, even in conditioned medium fortified with 20 ~ foetal calf serum (Macpherson & Montagnier, I964). However, cloning in soft agar could easily be accomplished
with both the SV4o-infected BALB c/3T3 foci and the WI-38(K ) cells. It was this inability
to grow in soft agar that necessitated the use of glass cylinders for the isolation of transformed WI-38 loci.
The longevity of the subcultured foci varied widely. Twenty-four grew only through
passage 4I (Fig. I). Confluent uninfected WI-38 cells were also isolated in cylinders in
Io experiments; none of those subcultures survived any longer than the mass culture.
The remaining 22 subcultures survived up to a maximum of 64 to 65 passages at 1:2
dilution (Fig. I, 2). The doubling time of cultures increased sharply in the last few generations
(Fig. 2); this was especially precipitous and occurred within I generation in the longestlived subculture (A). When cells of this subculture frozen at passage 54 were thawed and
passaged once again, crisis occurred in the same abrupt fashion. The cell density of confluent
cultures also fell in the last few passages. For example, the subculture from focus 52 at
confluence dropped from 3"6 to 3"7 x Io 5 cells/cm 2 to z.I × Io 5 cells/cm 2 at generation 58
and I-8 × Io 5 at generation 6L
Two independent mass cultures of SV4o-infected WI-38 cells passaged without intermediate subculture survived no longer than the individual loci, with division ceasing at
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:20:41
412
Short communications
culture generations 50 and 5r. These findings contrast markedly with the results of three
independent trials with WI-38(K) cells, which were easily subcultured and passaged for
more than I4o generations.
During the entire growth period, including the time in crisis, the cells of each subculture
uniformly showed the morphology and SV4o T-antigen positivity of transformed cells.
We therefore infer that SV4o transformation p e r se is insufficient to confer indefinite
growth on human diploid fibroblasts. The limited life span of these SV4o-transformed cells
is thus distinct from the abortive transformation of mouse cells by SV 4o. In the latter
case (Smith et al. I972), T-antigen is lost and cell morphology reverts to normal with time;
and SV4o genetic information is also lost.
Since we were unable to 'rescue' any of the transformed cells, we cannot exclude the
possibility that the failure of our transformed cells to survive was related to some technical
problem. However, the ease with which WI-38(K) cells were subcultured, even after great
dilution, and their subsequent long-term survival argues that the technique was probably
adequate. It is likely, then, that the finite lifespan of these transformed cells is an intrinsic
property.
Why is the survival of SV4o-transformed cells so rare ? To answer this question, it is
first necessary to consider how these results relate to current theories of senescence in
untransformed fibroblasts.
The commitment theory of cellular aging (Holliday et al. I977) suggests that fibroblasts
tend to lose the capacity for indefinite growth by an irreversible change ('commitment')
which occurs many generations before their cell division actually ceases. The finite lifespan
of human fibroblasts in culture is due, according to this view, to dilution and loss of
remaining uncommitted cells during subculture. Consistent with this theory are the observations of considerable variation in the growth potential of untransformed human fibroblasts separated at an early passage (Smith & Hayflick, I974). Likewise, our findings that
subcultured foci did not survive crisis can be explained if each focus was derived from
cells already committed at the time of transformation, and the variation in lifespans
(Fig. I) could again reflect the different division potential remaining in the cells at the
time of transformation.
The commitment theory has emphasized that the gradual increase in doubling time of
a mass culture reflects its content of cells of markedly different reproductive capacities
(Holliday et al. ~977). In our experiments, however, each subcultured focus is essentially
a clone consisting of cells of the same 'age'. Cell division would then be expected to cease
abruptly. This is in agreement with our observation (see particularly subculture A, Fig. 2).
It strongly supports commitment as opposed to theories which propose the cessation of
cellular division as a result of cell damage accumulating from heterogeneous, random
events even in cohorts of cells of the same age (see Hayflick, I973).
The commitment theory does not, however, explain how SV4o transformation promotes
the survival of any cells from mass culture. From the studies of Jensen et al. (1963), one
can infer that SV4o transformation provides some function required for the indefinite
growth of human fibroblasts. Our data suggest that the cell which survives must be special
in some other way. Perhaps a human fibroblast destined to survive indefinitely must be
'uncommitted' at the time of transformation. Transformation could then specifically
increase the growth rate of the uncommitted cell, or lessen the probability for later commitment, or both, thereby tending to conserve these cells during subculture. And our study,
beginning at passage I8 and following only 92 subcultured foci, would have been unlikely
to produce a transformed uncommitted cell. However, Koprowski et al. (T966a, b) were
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:20:41
Short communications
4I 3
able to rescue some fibroblasts with SV4o transformation even in the crisis phase, at a
time when the commitment theory states that no uncommitted cells remain. Thus, either
uncommitted cells can exist in cultures in crisis but cannot overgrow the mass of undividing
cells; or SV4o transformation can reverse the terminal cessation of cell division in rare
cells.
That SV4o transformation can selectively increase growth potential is directly indicated
by the findings that mass cultures of transformed fibroblasts survive, on the average,
many generations longer than cultures of untransformed cells (Koprowski et al. I966a)
and by our additional finding that the fastest-growing transformed subcultures were
precisely those that continued to divide the longest.
If reversal of commitment is required to promote survival, a special transformation
event must also be postulated or survivors would be more common. Among the possibilities
that might produce indefinite growth are integration at a rare site (Botchan et al. 1976;
Gelb & Martin, 1973; Ketner & Kelly, I976; Prasad et al. I975) or infection of a cell in
a rare physiological state. Transformation, then, would be a relatively frequent event;
but the indefinite growth associated with tumorigenesis would be much more rare and
only occasionally induced by the transforming virus. The only hint of a possible physiological correlate of this 'special' transforming event noted was the inability of our 91 subcultures, unlike SV4o(K) cells, to grow in soft agar. Perhaps selection of transformants in
soft agar would reveal rare cells destined to survive indefinitely.
We thank Dr J. Huang for technical aid and discussion. This work was supported in
part by grants from the Damon Runyon Memorial Fund for Cancer Research (DRG-I233),
the National Institutes of Health (GM 21357), N I H Training Grant 5ToI Aioo459 and
the National Science Foundation (VMS-o7oI3Ao0. Dr Gelb is an investigator of the
Howard Hughes Medical Institute.
Department of Microbiology and Immunology
Division of Biology and Biomedical Sciences
Washington University School of Medicine
St Louis, Missouri 63110, U.S.A.
SADAO GOTOH
LAWRENCE GELB
DAVID SCHLESSINGER
REFERENCES
BOTCHAN, M., TOPP, W. & SAMBROOK,J. (1976). The arrangement of simian virus 40 sequences in the D N A
of transformed cells. Cell 9, 269-287 •
EARLE, W. (1943). Production of malignancy in vitro. IV. The mouse fibroblast cultures and changes seen
in the living cells. Journal of the National Cancer Institute 4, 165-212.
GELB, L. D. & MARTIN,M. A. (1973)- Simian virus 4o D N A integration within the genome of virus-transformed
mammalian ceils. Virology 5I, 351-357.
HAYFLICK, L. (1973). The biology of human aging. American Journal of Medical Science 265, 432-445,
HAY~LICK, L. & MOORHEAD, V. (1961). The serial cultivation of human diploid cell strains. Journal of Cell
Biology 62, 48-53.
HOLLEr, R. W. & JClERNAN,J. A. (1968). 'Contact inhibition' of cell division in 3T3 cells. Proceedings
of the National Academy of Sciences of the United States of America 6o, 3oo-3o4.
HOLLIDAY, R., HUSCHTSCHA, L.I.~ TARRANT, G.M. & KIRKWOOD, T.B.L. (I977). Testing the commitment
theory of cellular aging. Science x98, 366-372.
JENSEN, •., KOPROWSKI,1-I. & VO~qTEN,J. A. (1963). Rapid transformation of human fibroblast cultures by
simian virus 4o. Proceedings of the National Academy of Sciences of the United States of America 5o,
343-348.
ICETNER, a. & KELLY, T. J,, JR. (1976). Integrated simian virus 4o sequences in transformed cell D N A :
analysis using restriction endonucleases. Proceedings of the National Academy of Sciences of the United
States of America 73, I Io2-I IO6.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:20:41
414
Short communications
KOPROWSKI, H., JENSEN, F., GIRARDI, A. & KOPROWSKA, L (I966a). Neoplastic transformation.
Cancer Research
26, 198o-I993.
KOPROWSKI, H., PONTEN, J. R., JENSEN, F., RAVDIN, R. G., MOOR.HEAD, P. & SAKSELA, E. (I966b). Transformation
of cultures of human tissue infected with simian virus SV4o. Journal of Cellular and Comparative
Physiology 59, 281-292.
MACPHERSON, I. & MONTAGNIER, L. (I964). Agar suspension culture for the selective assay of cells transformed by polyoma virus. Virology 23, 291-294.
PORE,J. z. & ROWE,W. P. 0964)- Detection of specific antigen in SV4o transformedcellsby immunofluorescence.
Journal of Experimental Medicine x2o, 12I-I27.
rOTTER, C. W., rOTTER, A. M. & OXFORD,J. S. (I970). Comparison of transformation and T antigen induction
in human cell lines. Journal of Virology 5, 293-298.
VRASAD, L, ZOUZIAS, D. & BASILI¢O, ¢. (I975). Simian virus 40 integration sites in the genome of virustransformed mouse cells. Journal of Virology x6, 897-9o4.
toSSER, R. & POLLACK,R. (I974). A nonselective analysis of SV4o transformation of mouse 3T3 cells. Virology
59, 477-489.
SHEEN, H. & ENDERS, S. 096Z). Transformation induced by simian virus 40 in human renal cell cultures:
1. Morphology and growth characteristics. Proceedings of the National Academy of Sciences of the
United States of America 48, I I64-I I72.
SMITH, H.S., GELB, L.O. & MARTIN, M.a. 0972). Detection and quantitation of simian virus 4o genetic
material in abortively transformed Balb/3T3 clones. Proceedings of the National Academy of Sciences
of the United States of America 69, I52-I56.
SMtTH, J. R. & HAYFLtCK, L. (L974). Variations in the lifespan of clones derived from human diploid cell
strains. Journal of Cell Biology 62, 48-53.
STOKER, M. & MACVH~RSON,I. (I964). Syrian hamster fibroblast cell line B H K and its derivatives. Nature,
London 203, I355-1357.
TODARO, 6. S., GREEN, n. & swIvi, i . R. (I966). Susceptibility of human diploid fibroblast strains to transformation by SV4o. Science x53, I252-1254.
(Received 2 June I978)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 20:20:41