J. Cell Set. 44, 123-133 (1980)
Printed in Great Britain © Company of Biologists Limited 1980
I2-j
MICROTUBULE-ORGANIZING CENTRES IN
BINUCLEATE CELLS AND HOMOSYNKARYONS
FIONA M. WATT,* E. SIDEBOTTOM AND H. HARRIS
Sir William Dunn School of Pathology, South Parks Road,
Oxford OXi 3RE, England
SUMMARY
Immunofluorescence studies showed that most binucleate Vero cells formed by virus-induced
fusion or by inhibition of cytokinesis had a single microtubule-organizing centre (MTOC) when
examined during the reassembly of microtubules after chilling, but two or more organizing
centres when examined after exposure to colcemid. These findings suggest that although
binucleate cells initially contain more MTOC than mononucleate cells, the extra MTOC are
normally aggregated, so that the number of MTOC in binucleate cells tends to be reduced
very quickly to that in mononucleate cells.
INTRODUCTION
In the preceding paper, we used immunofluorescence microscopy to observe microtubule-organizing centres (MTOC) in cultured cells under different experimental
conditions (Watt & Harris, 1980). In the present paper we use the same techniques
to study MTOC in binucleate cells produced by virus-induced fusion or by inhibition
of cytokinesis. Such cells, of course, initially contain more MTOC than normal
mononucleate cells, and our aim was to discover what happens to these extra organizing
centres. In addition, the number of MTOC in mononucleate homosynkaryons was
compared with the number in unfused mononucleate cells.
MATERIALS AND METHODS
Cells and cell culture
The method of culture and source of Vero cells are given in the preceding paper (Watt &
Harris, 1980).
Cells were fused in suspension by inactivated Sendai virus, essentially as described by Harris
& Watkins (1965). Vero homokaryons were formed by mixing 2-4 xio 1 cells with 40-50 haemagglutinating units of inactivated Sendai virus in a volume of 1 ml. Immediately after fusion,
the cells were allowed to attach and spread on n-mm diameter round glass coverslips: 20—30 %
of the cells had more than one nucleus, and about half of these were binucleate. The proportion
of binucleate cells in an unfused population was approximately O'3~o-8 %.
At different times after fusion, cells were either chilled on ice or incubated with o-5 /tg/ml
colcemid, and then allowed to recover partially so that MTOC could be visualized by immunofluorescence microscopy. The precise techniques are given in the preceding paper.
• Present address: Room 56-543, Department of Biology, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139, U.S.A.
9-2
124
F. M. Watt, E. Sidebottom and H. Harris
Counting microtubule-organizing centres
The method used for counting MTOC in cells on coverslips is given in the preceding paper.
MTOC were only counted in cells with a interphase nuclei. Each set of results was derived
from at least 3 experiments involving different populations of cells. Replicate experiments did
not show significant variation by x1 analysis.
Calculating the expected number of microtubule-organizing centres in
binucleate cells resulting from cell fusion
After fusion, a binucleate cell should contain the sum of the number of MTOC in the 2 cells
from which it was derived. Assuming that 3 cells at any stage of the cell cycle can fuse (Yamanaka
& Okada, 1966; Johnson & Harris, 1969), the expected distribution of organizing centres in
cells with a nuclei can be calculated as follows. Let
Px = probability that a single cell has x MTOC
and
Bpx = probability that a binucleate cell has x MTOC.
Since all unfused Vero cells have at least 1 MTOC and none has more than 4 (Watt & Harris,
1980):
Bpi = o,
Bps = zpipt + xptps,
Bpt = pi,
Bp, = p\ + 2pipi,
Bps = zpipt,
Bp, = zpipit
Bpl=pl + 2psp3,
Bpi=p\.
Only binucleate cells with both nuclei in interphase are considered.
In the preceding paper we showed that, after 1 h in 0-5 /tg/ml colcemid followed by 30 min
recovery in fresh medium, 57-4 % of interphase Vero cells had 1 MTOC, 42-0 % had 2
MTOC, 0-4 % had 3 MTOC and 0-2 % had 4 MTOC. Thus for colcemid-treated cells:
•Pi = 0-574,
P* = 0-004,
P2
P 4 = 0002.
= 0-420,
After chilling on ice for 30 min and 30 s recovery at 37 °C:
Px = 0-895,
P3 = Pt = o-ooo.
Pa
= 0105,
RESULTS
Microtubule-organizing centres
Our previous experiments with normal mononucleate cells suggested that chilling
was the gentlest, and probably the most reliable, way to reveal MTOC. This method
was therefore used first to study MTOC in binucleate cells. The cells were placed
on ice for 30 min to disrupt microtubules, transferred to medium at 37 °C for 30 s,
and then fixed and stained with antibodies (Fig. 1).
Table 1 gives the expected number of MTOC in binucleate cells (see Materials
and Methods), and the actual number observed at different times after cell fusion.
The cells required 2-5 h in which to attach and spread on the coverslips, and this is
therefore the earliest time after fusion when MTOC could be counted.
The majority of unfused Vero cells had a single microtubule-organizing centre
during recovery from cold treatment (see preceding paper), and one would therefore
expect most binucleate cells to have two. However, 2-5 h after fusion over 90% of
the binucleate cells had only one MTOC. The number of organizing centres remained
fairly constant during the first 2 days after fusion, although the proportion of cells
Microtubule-organizing centres infused cells
Fig. i. Binucleate Vero cell with i MTOC after cold treatment for 30 min followed
by 30 s recovery at 37 °C. x 1000.
Fig. 2. Microtubules in a binucleate Vero cell, 2-5 h after fusion, x 1000.
with more than 1 MTOC did increase slightly. The display of microtubules in untreated
binucleate cells 2-5 h after fusion resembled the pattern characteristic of mononucleate
cells. Microtubules were most dense in the perinuclear region and spread from there
throughout the cytoplasm to the cell periphery (Fig. 2).
The observed reduction in the number of microtubule-organizing centres in
binucleate cells could be brought about in two ways: the extra MTOC might be
destroyed, or they might aggregate and function as a single organizing centre. We
were able to distinguish between these two possibilities by making use of the discovery
126
F. M. Watt, E. Sidebottom and H. Harris
Table i. MTOC in Vero cells after recovery from cold treatment
Total no.
MTOC per cell (%)
rpllo
Unfused cells*
Binucleate cells: expected
number
Binucleate cells: observed
number at different times after
fusion
2'Sh
8h
i day
ecus
i
2
3
4
89'S
io - 5
8o-i
—
18-8
IT
—
6-2
O'S
0-2
405
93
3-6
2-1
386
i5'O
S-4
3-6
4-9
167
—
93'i
85-0
76-0
77-0
counted
1209
z days
6if
H-8
3-3
• Data from preceding paper.
f After 2 days the number of 1binucleate: cells in culture had declined and the sample size
was therefore smaller.
Table 2. MTOC in Vero cells after colcemid treatment
MTOC per cell (%)
Total no.
11
A.
Unfused cells*
Binucleate cells: expected
number
3
4
5
ceus
counted
1666
i
2
S7'4
420
0-4
O"2
329
48-2
18-1
c6
O-2
—
10-4
24-1
io-o
266
393
311
24-8
25-0
272
27-6
1-2
4-5
10-5
o-6
336
489
1 os
>5
Binucleate cells: observed
number at different times
after fusion:
2-5 h
i day
2 days
2 0 0
6-7
• Data from preceding paper.
io-s
that colcemid increases the number of MTOC in cells probably by causing aggregated
organizing centres to separate (see preceding paper). If the extra organizing centres
in binucleate cells are destroyed, then the number of MTOC in colcemid-treated
binucleate cells should be the same as the number in mononucleate cells exposed to
colcemid. However, if the extra MTOC aggregate, the number seen after colcemid
treatment should more closely resemble the sum of the numbers seen in two single cells.
Two and a half hours after fusion, Vero cells were exposed to 0-5 /<g/ml colcemid
for 1 h and placed in medium lacking colcemid for 30 min, so that the number of
MTOC in binucleate cells could be determined (Fig. 3). Table 2 shows the actual
and the expected distribution of organizing centres. As expected, the majority of
binucleate cells 2-5 h after fusion had 2, 3 or 4 MTOC, whereas most mononucleate
cells had 1 or 2 MTOC (see preceding paper). The extra MTOC in binucleate cells
were not therefore destroyed after cell fusion, but, in the absence of colcemid, were
Microtubule-organizing centres infused cells
ra?
Fig. 3. Binucleate Vero cell with 4 M T O C after 1 h in 0-5 /tg/ml colcemid and
30 min recovery, x 1000.
apparently aggregated together. At later times, the number of binucleate cells with
more than 4 MTOC increased slightly. About 10 % of the binucleate cells still had
a single MTOC after colcemid treatment. This suggests that not all the organizing
centres that had aggregated were separated by this treatment.
The difference between the number of MTOC seen after exposure to low temperature and the number seen after colcemid treatment was even more marked in
cells with 3 or more nuclei. In chilled cells the nuclei were often arranged round 1
large organizing centre (Fig. 4), whereas in colcemid-treated cells many MTOC were
distributed among the nuclei or further away in the cytoplasm (Fig. 5).
Microtubule-organizing centres in binucleate cells produced by inhibition of cytokinesis
When cytokinesis is inhibited, the binucleate cells formed should contain the same
number of MTOC as pairs of daughter cells. It was of interest to see whether the
extra MTOC in such cells were aggregated, as occurred in fused cells. Cytokinesis
of Vero cells was therefore prevented by exposure to 1 /tg/ml cytochalasin B (supplied
by Aldrich Chemical Co., Milwaukee, Wisconsin, U.S.A.). This mould metabolite
appears to inhibit the separation of daughter cells after nuclear division by interfering
with microfilaments in the cleavage furrow (Schroeder, 1969, 1970). Table 3 shows
the number of nuclei in Vero cells cultured in cytochalasin B for up to 7 days.
Cells incubated in cytochalasin B for different lengths of time were chilled or
treated with colcemid in order to determine the number of MTOC in binucleate
cells with two interphase nuclei (Tables 4, 5). After 24 h in cytochalasin B most
binucleate cells had a single organizing centre during recovery from cold treatment,
F. M. Watt, E. Sidebottom andH. Harris
Fig. 4. Tetranucleate Vero cell with 1 MTOC after cold treatment for 30 min followed
by 30 s recovery at 37 °C. x 1000.
Fig. 5. Trinucleate Vero cell with 6 MTOC after 1 h in 0-5 fig/ml colcemid and
30 min recovery, x 1000.
Microtubule-organizing centres in fused cells
129
Table 3. Effect of exposure to cytochalasin B on number of nuclei in Vero cells
Nuclei per cell
Duration of exposure to
i /ig/ml cytochalasin B
Total no.
cells
counted
(%)
A
Untreated cells
i day
3 days
7 days
I
2
994
o-6
57'7
562
57-5
4°-5
I3-9
194
17-5
179
5
4
3
1276
560
486
273
04
I2-I
c-7
4-4
Table 4. MTOC in binucleate Verocells after recovery from cold treatment
MTOC per cell (%)
Total no.
cells
counted
"A
Duration of exposure to
cytochalasin B
1 day
2 days
5 days
7 days
I
2
839
76-4
652
58-8
12-7
18-0
252
3
4
1-8
4-7
7-5
i-6
109
>4
685
556
I-I
1-2
o-6
3 "4
1-7
161
119
Vero cellsafter colcemidtreatment
Table 5., MTOC in binucleate
i
Duration of
exposure to
cytochalasin B
MTOC per cell (%)
1
2
3
4
1 day
2 days
3 days
5 days
7 days
18-1
12-5
10-3
7-0
4-3
323
26-5
31-4
320
299
299
329
18'•1
27 •5
25 • 1
27 1
12 •9
2I-I
151
I4'3
5
6
7
8
9
0-4
40
39
0-3
0-3
4-9
57
o-i
03
3-2
—
—
0-7
—
0-7
9-1
8-i
200
Total
no. cells
counted
601
690
284
140
but 2 or more after exposure to colcemid. This finding supports the view that in
binucleate cells, whether they are formed by failure of cytokinesis or by virus-induced
fusion, the number of functional organizing centres is approximately the same as that
in mononucleate cells, because the extra organizing centres aggregate.
Binucleate cells formed by virus-induced fusion divide or are overgrown by unfused
cells in a relatively short time. However, after 3 days in cytochalasin B, the number
of binucleate Vero cells remains fairly constant; further nuclear division occurs in
only a small proportion of cells (Table 3). Thus, MTOC can be studied in binucleate
cells which persist for several days.
Tables 4 and 5 show that after 2 days in cytochalasin B, the number of MTOC
in binucleate cells tended to increase (Fig. 6). By 7 days, 36 % of the colcemid-treated
cells had 5 or more MTOC, whereas no cells had this number of MTOC after 24 h.
After cold treatment, the proportion of cells with more than 2 MTOC rose from 3 %
after 24 h to 16 % after 7 days.
130
F. M. Watt, E. Sidebottom and H. Harris
Fig. 6. Six MTOC in a binucleate Vero cell grown in i /ig/ml cytochalasin B for
3 days, then incubated in 0-5 /ig/ml colcemid for 1 h and allowed to recover for
30 min. x 1000.
Fig. 7. Two MTOC in a Vero cell containing five 1-6-/4111 diameter latex beads after
1 h in 05 /ig/ml colcemid and 30 min recovery, x 1000.
Microtubule-organizing centres in fused cells
131
The persistence of binucleate cells in cultures kept in cytochalasin B for 7 days
suggests that the nuclei are incapable of further division. Such cells increase in size
and resemble irradiated 'giant' cells (Puck & Marcus, 1956). It is possible that the
observed increase in the number of organizing centres in these cells is due to persistence of the centriolar cycle in the presence of prolonged inhibition of nuclear
division.
Microtubule-organizing centres in mononucleate homosynkaryons
The experiments which have been described suggest that the number of active
microtubule-organizing centres in binucleate cells is strictly controlled. It was therefore
of interest to see whether mononucleate daughters of cells with 2 or more nuclei
(homosynkaryons) contain the same number of organizing centres as normal
mononucleate cells.
Homosynkaryons were identified by the presence in the one cell of 2 different kinds
of fluorescent latex bead (obtained from Polysciences Inc., Warrington, Pennsylvania,
U.S.A.). The cells to be fused were grown in medium containing either o-8-/<m or
i-6-pm diameter beads. The smaller beads showed green fluorescence and the larger
ones yellow-orange fluorescence when exposed to light of the wavelength optimal for
fluorescein excitation.
After 2 days about 90% of the cells contained beads; that they had, indeed, been
ingested was confirmed by electron microscopy (Watt, 1979). Many small beads were
taken up but the number of large beads per cell was often less than 10 (Fig. 7). The
presence of the beads did not affect the number of MTOC seen in single or binucleate
cells.
Cells containing either large or small beads were fused after extensive washing to
remove beads that had not been ingested. One or two days later, the cells were exposed
to 0-5 /^g/ml colcemid for 1 h and allowed to recover for 30 min, so that MTOC
could be counted. Control experiments, in which cells containing one or other type
of bead were grown together in culture, indicated that cells with only one nucleus
but 2 or more beads of each type could only have arisen by cell fusion. The number of
homosynkaryons that could thus be positively identified was, however, low, partly
because of the small number of i-6-ju.m diameter beads originally ingested and partly
because of the low plating density necessary to prevent overgrowth by unfused cells.
Ninety-five positively identified homosynkaryons were counted in 5 separate fusion
experiments. Thirty-eight of these had 1 MTOC; 38 had 2 MTOC; 14 had 3 MTOC;
4 had 4 MTOC; and 1 had 6 MTOC. The majority of homosynkaryons thus had 1
or 2 MTOC, but the proportion with 3 or more MTOC was significantly higher than
in normal mononucleate cells (see preceding paper).
DISCUSSION
Most binucleate cells formed by virus-induced fusion or by inhibition of cytokinesis
have a single MTOC during microtubule regrowth after chilling, but two or more
organizing centres after exposure to colcemid. Our interpretation of these results is
132
F. M. Watt, E. Sidebottom and H. Harris
that although binucleate cells initially contain more MTOC than single cells, the
extra MTOC are normally aggregated and function as a single centre.
The aggregation of MTOC in binucleate cells could ensure that the three-dimensional array of microtubules is the same as in cells with a single nucleus, and, in
addition, it might influence the events of mitosis. When spontaneous or artificially
induced binucleate cells enter mitosis the chromosomes of both nuclei are commonly
associated with a single mitotic spindle. The spindle formed is sometimes bipolar,
but multipolar spindles are also common (Fell & Hughes, 1949; Oftebro & Wolf,
1967; Oftebro, 1968). If the aggregation of MTOC observed in binucleate Vero cells
is a general phenomenon, it might ensure that a single spindle is formed. Multipolar
spindles might arise from incomplete aggregation of MTOC. Evidence for aggregation
of MTOC in a different situation comes from the work of Spiegelman, Lopata &
Kirschner (1979). They found that undifferentiated neuroblastoma cells have an
average of 12 MTOC close to the nucleus, but when neurite formation is stimulated
the MTOC aggregate and form a single large organizing centre at the point where
neurite outgrowth occurs.
The percentage of mononucleate Vero homosynkaryons with 3 or more MTOC is
higher than the percentage seen in unfused Vero cells. In view of the small sample
size, it is difficult to assess the significance of this difference, but even if it were a
general phenomenon the difference is probably not large enough to account for the
failure of many synkaryons to divide (Yamanaka & Okada, 1968; Gordon & Cohn,
1970) and yield clones. There is one interesting report of a permanent alteration in
the number of MTOC following cell fusion (Shay, Peters & Fuseler, 1978). Cytoplasts
of one cell line were fused with whole cells from another. Both parental lines and most
of the cybrid clones had only one MTOC per cell during interphase, but all the cells
in one of the cybrid clones had 2 MTOC.
Many features of MTOC remain to be investigated. It would, for example, be
interesting to discover whether the aggregation of MTOC found in Vero homokaryons is a general feature of multinucleate cells. Holmes & Choppin (1968) have
described syncytia of BHK-21 cells in which the nuclei are aligned in rows by microtubules, and this might reflect a different distribution of MTOC. Our studies of cells
cultured in cytochalasin B suggest that prolonged inhibition of nuclear division may
affect the number of MTOC per cell. Finally, if centrioles are aggregated in multinucleate cells, this would provide further evidence for a relationship between
centrioles and MTOC.
F.M.W. was a recipient of a Medical Research Council Studentship for training in research
methods.
REFERENCES
FELL, H. B. & HUCHES, A. F. (1949). Mitosis in the mouse: a study of living and fixed cells in
tissue cultures. Q.Jlmicrosc. Set. 90, 355-380.
GORDON, S. & COHN, Z. (1970). Macrophage-melanocyte heterokaryons. I. Preparation and
properties. J. exp. Med. 131, 981-1003.
Microtubule-orgamzing centres in fused cells
133
H. & WATKINS, J. F. (1965). Hybrid cells derived from mouse and man: artificial
heterokaryons of mammalian cells from different species. Nature, Lond. 205, 640—646.
HOLMES, K. V. & CHOPPIN, P. W. (1968). On the role of microtubules in movement and
alignment of nuclei in virus-induced syncytia._7. Cell Biol. 39, 526—543.
JOHNSON, R. T. & HARRIS, H. (1969). DNA synthesis and mitosis in fused cells. I. HeLa
homokaryons. J. Cell Sci. 5, 603-624.
OFTEBRO, R. (1968). Further studies on mitosis of bi-and multinucleate HeLa cells. Scand.J.
din. Lab. Invest., Suppl. 106, 79—96.
OFTEBRO, R. & WOLF, I. (1967). Mitosis of bi- and multinucleate HeLa cells. Expl Cell Res.
48, 39-52.
PUCK, T. T. & MARCUS, P. I. (1956). Action of X-rays on mammalian cells. J. exp. Med. 103,
653-666.
SCHROEDER, T. E. (1969). The role of 'contractile ring' filaments in dividing Arbacia egg.
Biol. Bull. mar. biol. Lab., Woods Hole 137, 413-414 (abstr.).
SCHROEDER, T. E. (1970). The contractile ring. I. Fine structure of dividing mammalian (HeLa)
cells and the effects of cytochalasin B. Z. Zellforsch. mikrosk. Anat. 109, 431-449.
SHAY, J. W., PETERS, T. T. & FUSELER, J. W. (1978). Cytoplasmic transfer of microtubule
organizing centers in mouse tissue culture cells. Cell 14, 835-842.
SPIECELMAN, B. M., LOPATA, M. A. & KIRSCHNER, M. W. (1979). Aggregation of microtubule
initiation sites preceding neurite outgrowth in mouse neuroblastoma cells. Cell 16, 253-263.
WATT, F. M. (1979). Microtubule Organizing Centres in Cells in Culture and in Hybrids Derived
from Tliem. Thesis presented for D.Phil, in University of Oxford.
WATT, F. M. & HARRIS, H. (1980). Microtubule-organizing centres in mammalian cells in
culture. J. Cell Sci. 44, 103-121.
YAMANAKA, T. & OKADA, Y. (1966). Cultivation of fused cells resulting from treatment of cells
with HVJ. I. Synchronization of the stages of DNA synthesis of nuclei involved in fused
multinucleated cells. Biken'sJ. 9, 159-175.
YAMANAKA, T. & OKADA, Y. (1968). Cultivation of fused cells resulting from treatment of cells
with HVJ. II. Division of binucleated cells resulting from fusion of two KB cells by HVJ.
Expl Cell Res. 49, 461-469.
(Received 11 February 1980)
HARRIS,