Complementation between temperature-sensitive (ts) variants of Balb/c-3T3 defective in the G1 phase of its cell cycle was measured in the [1H]thymidine-labelling indices of the multinucleated cells during incubation at the restricted temperature (38 °C) following cell fusion. One ts variant from each group along the length of the G1 phase was tested for complementation. Varying degrees of complementation were observed between the 4 ts variants tested, judging by the time of entry into S-phase and the degree of synchrony attained. At least 3 complementation groups were discernible.

Evidence has been accumulating over the past few years to suggest (Prescott, 1976) that some critical regulatory changes (controUing both cell proliferation and differentiation) take place during the G1 phase of a mammalian cell cycle (Howard & Pelc, 1953; Mueller, 1971). The study of the genetics of the G1 phase is thus a prerequisite to elucidate the functions controlled in G1. We constructed a tentative ‘map’ of the G1 phase of Balb/c-3T3 using temperature-sensitive (ts) variant clones (Naha, Meyer & Hewitt, 1975) by measuring their time of entry into 5 phase (after temperature shift-down from 38 to 33 °C). I report here the preliminary results of complementation tests performed with one ts variant from each group along the length of the G1 phase.

Cell lines

The cell lines used in these experiments and cultuie conditions of their maintenance were reported before (Naha et al. 1975). The G1 phase ts variants (A83, A123, A92 and A8) were recloned before complementation tests were performed, the reversion frequencies for plating efficiency being in the order of 1–3 × 10− 4 at the cell density of 1 × 106/ml.

Cell fusion and complementation

Cell fusion was conducted following the technique described by Harris & Watkins (1965). Equal numbers of cells from each parent (2 × 10 6) grown at 33 °C were fused in the presence of β-propiolacton-inactivated Sendai virus (HAU 600/ml) and planted in 1-ml volumes irt Leighton tubes at a density of 1 × 105/ml and incubated at 38 °C. In a parallel series of experiments asynchronous log phase cultures growing at 33 °C were incubated at 38 °C for 18 h (roughly one cell generation time). Cells exposed to 38 °C were then fused as above.

Complementation between the ts variants defective in G1 phase was tested in separate experiments: (i) by prelabelling of one parent with [14C]thymidine (14C-TdR, with low specific activity’) followed by pulse-labelling with [1H]thymidine (1H-TdR, with high specific activity’) of cells fused and incubated at the restricted temperature of 38 °C; and (ii) by measuring the time of entry into S phase (by pulse labelling with [3H]thymidine only) of multinucleated cells (Johnson & Harris, 1969) at 38 °C following cell fusion. Pulse labelling with 3H-TdR was performed every 3 h on replicate cultures; the coverslips were washed, fixed, autoradiographed and stained with Giemsa according to the method described before (Naha et al. 1975). Radioactive chemicals were obtained from the Radiochemical Centre, Amersham.

Since the ts variants were inhibited (Naha et al. 1975) from entering the S phase at 38 °C (the labelling indices of homokaryotic parents after fusion were between 5 and 20 % depending on the leakiness of the parental cultures) an increase in the percentage of labelled multinucleated cells over unlabelled multinucleated cells would indicate (Johnson & Harris, 1969) the extent of complementation and also denote a measure of their time of entry into S phase and progress through the S phase. It should be emphasized that co-ordination of nuclear labelling in entering S phase between different cells and synchrony in S are independent and separate phenomena, the latter (Johnson & Harris, 1969) being ascribed to an indirect effect of the fusion method. The percentage of multinucleated cells in a normal population of unfused wild type 3T3 and its ts variants were counted to be less than 0·1, compared to the fused cells where the percentage was recorded to be between 5·8 and 10 ·2 in different experiments. Only’ those multinucleated cells (3 or 4 nuclei) showing co-ordination of labelling between nuclei were counted for positive complementarity. Less than 20 % of the multinucleated cells were ([3H]thymidine) labelled after fusion within the first 32 h of incubation at 38 °C which was presumed to be the background indicating no effective complementation during this period, due probably to some effect of the fusion (Johnson & Harris, 1969).

In the double-labelling experiment where one parent was pre-labelled with 14C-TdR, the radioactive 14C, because of longer track lengths of the ß particles, resulted in more diffuse labelling over the entire cell contrasting with 1H-TdR labelling which was strictly nuclear. This would render identification of the source of each nucleus in the β- and multinucleated cells possible. The strongest indication of complementation between the heterokaryotic G1-phase variants was thought to be provided by the unequal radioactive labelling of the nuclei in the multinucleated cells after fusion. Johnson & Harris (1969) first reported co-ordination of DNA synthesis among the nuclei of asynchronous cultures produced after fusion of homo- and heterokaryotic cells in absence of a genetic marker. The fusion products of the G 1-phase variants used in our experiments were in the strictest term ‘homoheterokaryons’, since the ts variants were phenotypically identical with and derived from Balb/c-3T3 cells. Complementation between cells from different parents held up at different stages of the G1 phase at 38 °C would induce each parent to enter S phase at different times and, consequently, show different degrees of radioactive labelling during their progress (early or late) through the S phase. This is exemplified in Fig. 1. (Relative sequence of the ts variants is shown in the insets of Figs. 2 and 3). Ideally, it should be possible to map the G1 phase ts variants by radioactive grain counts in the nuclei of the multinucleated cells as a measure of the distance between these variants at the time of their entry’ into S phase. This, of course, was not the immediate objective of the present set of experiments.

Fig. 1.

Result of pulse labelling with 3H-TdR (16 μCi/ml, 32 Ci/mmol) of cells fused and incubated at 3 8 °C. Multinucleated cells (× 3000) of A83 + A8, A83 + A92 and A83 + A123 are compared with A83 + WT to indicate the unequal radioactive labelling during 5 phase.

Fig. 1.

Result of pulse labelling with 3H-TdR (16 μCi/ml, 32 Ci/mmol) of cells fused and incubated at 3 8 °C. Multinucleated cells (× 3000) of A83 + A8, A83 + A92 and A83 + A123 are compared with A83 + WT to indicate the unequal radioactive labelling during 5 phase.

Fig. 2.

Labelling indices of multinucleated cells fused and incubated at 38 °C. Complementarity of clone A83+WT (▴ · · · ▴) is compared with A83+A8 (•—•), A83 + A92 (×–×) and A83 + A123 (○–.–○).

Fig. 2.

Labelling indices of multinucleated cells fused and incubated at 38 °C. Complementarity of clone A83+WT (▴ · · · ▴) is compared with A83+A8 (•—•), A83 + A92 (×–×) and A83 + A123 (○–.–○).

Fig. 3.

Labelling indices of multinucleated cells fused and incubated at 38 °C. Complementarity of A8 + A83 (• —•) is compared with. A8 + A92 (× · · · ×) and A8 + A123 (○–.–○).

Fig. 3.

Labelling indices of multinucleated cells fused and incubated at 38 °C. Complementarity of A8 + A83 (• —•) is compared with. A8 + A92 (× · · · ×) and A8 + A123 (○–.–○).

The results (Fig. 2) of the first series of experiments showed varying degrees of complementation of A83 with A123, A92 and A8. Judging from the level of ‘one-way’ complementation observed between A83 and WT (wild-type), A83 and A8 acquired near optimum level of complementation, though the onset of the 5 phase was delayed by about 6 h in the latter combination. A generation time of 18 h (measured between 2 S periods) was, however, observed in both cases. These data suggested that the degree of complementation between A83 and A8 was nearly 100% as compared with A83 and WT. Since the duration of S phase of wild type 3T3 cells (Naha et al. 1975) was about 9 · 5 h at 38 °C, only half (around 50%) the population would ideally be labelled at any particular time. The fact that a degree of synchrony (48 · 2% labelled) was retained in the multinucleated cells of A83 and A8 during the second 5 phase, as compared to A83 and WT, is possibly a reflexion of complementarity between these parents. It is difficult from these experiments to postulate alternative explanations for these bimodal curves, as it is difficult to understand how multinucleated cells traversed 2 cycles, including mitosis, and appeared in the second S phase as multinucleated cells. Incubation at the restricted temperature might possibly contribute to this phenomenon when complementary partners were still under the effect of growth restriction.

The percentage of labelled multinucleated cells in A83 and A123, and A83 and A92, showed (Fig. 2) delayed entry into 5 phase, compared with that of A83 and A8. Two rather more significant results observed in these 2 combinations were: (1) in the degree of synchrony attained; and (2) in the duration of the S phase. While in the combination of A83 and A123 the percentage of labelled multinucleated cells in S phase reached 51 · 8, that in A83 and A92 was 59·2. This is probably another indication of close complementarity between these variants. The homokaryons during this period did not exceed 20% (data not presented). The labelling index curve in the combination of A83 and A123 failed to show 2 distinct peaks; the duration the multinucleated cells were in 5 phase seemed to indicate a longer 5 phase (12 h). This could happen if the 5 phase of the complementing parents did closely overlap, or else if perfect complementarity due to close proximity of their genetic functions was extended in time. In the combination of A83 and A123 sharp peaks are discernible, the 5 periods separated by 15 h. Since the cell generation time was calculated to be about 18 h, the 2 peaks separated by 15 h probably mean a marginal overlapping of the 2 S phases.

In the second series of experiments where clone A8 was fused with clones A92, A123 and A83, the results (Fig. 3) indicated that the multinucleated cells of A8 and A92, and A8 and A123, entered S phase earlier than those of A8 and A83; this would again, to a certain extent, confirm the relative position of the variants in their functional time scale. Consistent with the previous experiment, the multinucleated cells of clones A83, A123, or A92 combination with A8 retained (Fig. 3) a degree of synchrony during the second 5 phase. The labelling indices also showed nearly uniform complementarity between clone A8 and the others (between 42 · 6 and 54 · 4%). That the S phases (measured during 73 h of incubation) in all 3 combinations were separated by 18 h was probably an indication suggesting that the clone A8 controls an independent genetic function unrelated to clones A83, A123 and A92; the same is possibly true of clone A92 which complemented almost equally with A83 A123. The extended S phase in A83 and A123 observed in the previous experiment (Fig. 2) was probably an indication that they are part of the same genetic function.

The results presented above were from cell fusion experiments where parental cultures were grown at 33 °C. In the parallel series of experiments where cultures growing at 33 °C were preincubated for 18 h at 38 °C before cell fusion, the results showed no qualitative or quantitative difference in the pattern of co-ordinated (unequal) labelling of nuclei in the multinucleated cells compared to those presented above. The only difference observed in these experiments were in the lower level of labelling indices (by nearly a factor of 2) in both the homokaryons and the heterokaryons. This is attributed to the loss of viability of cells due to incubation at 38 °C for 18 h (Naha et al. 1975) prior to fusion. The general similarities in the results of 2 parallel series of experiments indicated that the unequal labelling of nuclei in the multinucleated cells was not due to the physiological condition of the cells prior to fusion but more probably due to ‘genetic’ complementarity.

In the double-labelling experiment where one parent was pre-labelled with 14C-TdR before fusion, pulse labelling with 3H-TdR was performed at 24, 36, 48 and 60 h after cell fusion. As in the case of single labelling experiments, 3H-labelling could be detected in less than 5 % of the multinucleated cells up to 36 h after fusion. At 48 h onwards 3H-TdR was observed (around 50%) on all combinations of complementary partners, except the combination of ts A83 and ts A123 where 3H-labelling was noted to reach 42 percent only at 60 h after fusion. In these experiments also, unequal co-ordinate labelling between all different ts variants was observed.

Although the G1 phase of a mammalian cell cycle has not yet been explained by any specific event, there is evidence to suggest that there are some rate controlling signals during this period which have regulatory function (Johnson & Harris, 1969; Rao & Johnson, 1970). The accepted parameter of initiation of DNA synthesis (5 phase) as the only measurement of G1 controls is made doubly ambiguous by the variability of the G1 phase (Prescott, 1976). It has thus become necessary to decipher the genetic controls of the G1 phase in order to understand the controls in cell proliferation. Temperature-sensitive variants have provided a powerful tool for the study of the G1 controls.

The results of the experiments presented above confirm our previous observations (Naha et al. 1975) on the relative positions of the ts variants on a functional time-scale map. The data also suggest (subject to further verification) that possibly half (up to 4 h) of the early G1 phase in Balb/c-3T3 cells is made up of a complex set of functions forming possibly part of the same gene; the rest perform at least 2 independent genetic functions as could be detected by the complementation pattern of A92 and A8. We have cloned multinucleated cells (hybrids?) from each of these combinations growing at 38 °C and are presently studying the pattern of segregation in culture, and also the electrophoretic distribution of their nuclear proteins.

The author acknowledges the technical assistance provided by Mrs K. A. Hewitt. This work was supported by grants from the Medical Research Council and the Cancer Research Campaign.

Harris
,
H.
&
Watkins
,
J. F.
(
1965
).
Hybrid cells derived from mouse
and man
:
artificial heterokaryons of mammalian cells from different species
.
Nature, Land
.
205
,
640
645
.
Howard
,
A.
&
Pelc
,
S. R.
(
1953
).
Synthesis of deoxyribonucleic acid in normal and irradiated cells and its relation to chromosome breakage
.
Heredity
, uppl.
6
,
261
273
.
Iohnson
,
R. T.
&
Harris
,
H.
(
1969
).
DNA synthesis and mitosis in fused cells
.
F. Cell Sci
.
5
,
603
624
.
Mueller
,
G. C.
(
1971
).
Biochemical perspective of the G1 and S intervals in the replication cycle of
animal cells
:
a study in the control cell growth
.
In The Cell Cycle and Cancer
(ed.
R.
Baserga
), pp.
269
307
.
New York
:
Dekker
.
Naha
,
P. M.
,
Meyer
,
A. L.
&
Hewitt
,
K.
(
1975
).
Mapping of the G1 phase of a mammalian cell cycle
.
Nature, Land
.
258
,
49
53
.
Prescott
,
D. M.
(
1976
).
The cell cycle and the control of cellular reproduction
.
Adv. Genet
.
18
,
100
177
.
Rao
,
P. N.
Johnson
,
R. T.
(
1970
).
Mammalian cell fusion: studies on the regulation of DNA synthesis and mitosis
.
Nature, Land
.
225
,
15 9
164
.