1851
Journal of Cell Science 110, 1851-1866 (1997)
Printed in Great Britain © The Company of Biologists Limited 1997
JCS3558
Evidence for cell cycle-specific, spindle pole body-mediated, nuclear
positioning in the fission yeast Schizosaccharomyces pombe
Iain Hagan1,* and Mitsuhiro Yanagida2
1School of Biological Sciences, 2.205 Stopford Building, University of Manchester, Oxford Road, Manchester
2Department of Biophysics, Faculty of Science, Kyoto University, Kitashirakawa, Sakyo-ku, Kyoto 606, Japan
M13 9PT, UK
*Author for correspondence (e-mail: [email protected])
SUMMARY
Specific changes in spatial order occur during cell cycle
progression in fission yeast. Growth of the rod-shaped cells
is highly regulated and undergoes a cell cycle and sizeregulated switch from monopolar to bipolar tip extension.
During both phases of growth, the interphase nucleus is
maintained in a central location. Following the separation
of the genome to the cell tips in mitosis, the two nuclei
migrate back towards the cell equator before stopping in
two new positions that will become the middle of the two
new cells. Here we use simultaneous labeling of microtubules, chromatin and spindle pole bodies in wild-type and
cdc mutants, to show that nuclear positioning is achieved
by regulation of spindle pole body-mediated nuclear
migration. We show that the number and location of
nuclear positioning signals is regulated in a cell cyclespecific manner and that spindle pole body-mediated forces
are likely to be responsible for maintaining correct nuclear
position once the nuclei have reached the appropriate
position in the cell. Accentuating the movement of the
nuclei back towards the cell equator after mitosis by artificially increasing cell length shows that the spindle pole
body leads the nucleus during this migration. When
multiple spindle pole bodies are associated with the same
or different nuclei they all go to the same point indicating
that the different spindle pole bodies are responding to the
same positional cue. In a septation-defective mutant cell,
which contains four nuclei, the spindle pole bodies on the
four different nuclei initially group as two pairs in regions
that would become the middle of the new cells, were the cell
able to divide. In the subsequent interphase, the nuclei
aggregate as a group of four in the centre of the cell. The
presence of two or three clusters of spindle pole bodies in
larger cells with eight nuclei suggests that the mechanisms
specifying the normally central location for multiple nuclei
may be unable to operate properly as the cells get larger.
Perturbation of microtubules with the microtubule poison
thiabendazole prevents the spindle pole body clustering in
septation mutants, demonstrating that nuclear positioning
requires a functional microtubule cytoskeleton.
INTRODUCTION
provided considerable insight into how nuclear movements can
be generated at the molecular level. Movement of nuclei out of
germinating spores into growing Aspergillus nidulans hyphae
is a microtubule-mediated process (Oakley and Morris, 1980;
Oakley and Reinhart, 1985; Meyer et al., 1988). It appears to
be driven, at least in part, by a microtubule motor protein
complex containing cytoplasmic dynein (Xiang et al., 1994,
1995a) and may involve a mechanism that is conserved in
eukaryotes in general (Xiang et al., 1995b). Similarly, in
Neurospora crassa the normal regular spacing of the nuclei
along the fungal hypha is abolished by mutating the heavy
chain of cytoplasmic dynein and functionally related genes
(Plamann et al., 1994; Robb et al., 1995; Tinsley et al., 1996;
Bruno et al., 1996). Whilst these studies clearly indicate that a
motor complex is responsible for nuclear movements in filamentous ascomycetes, the key questions of how these motors
generate nuclear movements and how are they regulated to
establish the even distribution of the nuclei along the hyphae
remain to be addressed.
Further levels of complexity have been revealed in the
The cytoskeleton plays a key role in the establishment and
maintenance of the complex patterns of organelle distribution
that underlie cellular architecture (Bray, 1992). Microtubules
have pivotal functions in the establishment of polarity in many
processes such as embryogenesis (Sullivan and Therkauf,
1995) and the accurate positioning of the nucleus and mitotic
spindle (Gönczy and Hyman, 1996; Reinsch and Karsenti,
1994; Reeve and Kelly, 1983).
A role for the microtubule cytoskeleton in nuclear positioning has been proposed in many simple eukaryotic systems such
as diatoms, algae and fungi (for example, Wordeman et al.,
1984; Doonan et al., 1986; Morris et al., 1995). In addition,
correlative phase-contrast and electron microscopy has
suggested that the microtubule-organising centre may be intimately involved in nuclear migration (e.g. Wilson and Aist,
1967; Dowding and Bakerspigel, 1954; Pickett-Heaps and
Wetherbee, 1987).
The genetic tractability of the fungal systems has recently
Key words: Schizosaccharomyces pombe, Spindle pole body,
Microtubule, MTOC, Nuclear migration
1852 I. Hagan and M. Yanagida
budding yeast Saccharomyces cerevisiae by elegant experiments using drugs and multiple mutant backgrounds. Not only
is the positioning of the undivided nucleus in the bud neck at
the junction between the mother and daughter cells dependent
upon microtubule integrity (Jacobs et al., 1988), but there is
also an intimate relationship between cytoplasmic microtubules and the actin cytoskeleton that involves dynein (Li et
al., 1993; Clark et al., 1994; Eshel et al., 1993; Palmer et al.,
1992; Snyder et al., 1991). This interaction determines both
spindle orientation and the path of elongation during mitosis
and results in the tethering the spindle to one point at the base
of the mother cell and to another at the tip of the bud. Similar
mechanisms may underlie spindle orientation in higher
systems (Gönczy and Hyman, 1996). Further evidence for a
role of the actin cytoskeleton in nuclear positioning comes
from studies such as those of sporulation in Sordaria where
large actin cables appear to wind around the nuclei and are
assumed to assist in the ejection of spores upon ripening
(Thompson-Coffe and Zickler, 1995).
Less is known about the relationship between the cytoskeleton and nuclear positioning in the mitotic cell cycle of the
fission yeast Schizosaccharomyces pombe. The fission yeast
interphase microtubule array resembles those of filamentous
fungi such as N. crassa and A. nidulans (McKerracher and
Heath, 1987; Osmani et al., 1988) and consists of a basket of
microtubules extending between the cell ends (Hagan and
Hyams, 1988). In the few studies of this cytoskeleton at the
level of electron microscopy, interphase microtubules are
generally seen as bundles of three microtubules (Hereward,
1974; Tanaka and Kanbe, 1986; Streiblova and Girbardt,
1980), suggesting that the patterns seen by immunofluorescence represent bundles rather than single microtubules. Disruption of the microtubule cytoskeleton either by mutation of
tubulin genes or exposure to anti-microtubule agents can lead
to branching of the normally linear rod-shaped cell and disruption of the central positioning of the nucleus (Walker, 1982;
Hiraoka et al., 1984; Umesono et al., 1983). Since these manipulations also cause a mitotic arrest that can last for 1.5 generation times or more, it is unclear whether the effects on
branching and nuclear positioning in these instances arise
directly from the disruption of the cytoskeleton or are a
secondary effect of a previous mitotic delay. However, recent
data suggest that microtubules are indeed important for the
establishment of cell polarity as several branched and curved
mutants have altered microtubule cytoskeletons (Verde et al.,
1995). One of these mutants, ban5-3, which is allelic to the α
tubulin gene, atb2 (Yaffe et al., 1996), was reported to also
have defects in nuclear positioning (Verde et al., 1995).
Here we describe experiments that use antibodies that specifically recognise the spindle pole body (SPB) (Hagan and
Yanagida, 1995) to characterise the basis of nuclear positioning
in fission yeast. Using different genetic backgrounds to manipulate the intracellular environment, we show that interphase
nuclear positioning in fission yeast is achieved by the response
of the SPB to cell cycle-specific signals. Whilst the nature of this
regulation of positioning remains obscure, data showing that
several signals are generated in cells unable to divide for three
generations are discussed in terms of fungal nuclear positioning
in general. We also discuss the role of a previously unrecorded
activation of the SPB at the end of anaphase in the rapid
movement of the nuclei towards the cell centre after division.
MATERIALS AND METHODS
Cell culture and strains
Cell culture and maintenance was carried out according to the method
of Moreno et al. (1991). The isolation of 972h−, h90 (Gutz et al., 1974)
cdc7.24, cdc14.118, cdc25.22 (Nurse et al., 1976) and cut1.206
cdc11.123 (Uzawa et al., 1990), has been described previously.
cdc7.A20 was a gift from Dr V. Simanis (ISREC, Lausanne). Mitotic
cultures were grown in rich YPD or YES medium, with the exception
of the cdc7.24 synchronous culture, which was grown in minimal
EMM2 medium (Moreno et al., 1990). Mating was carried out in SSL
following growth in SSL plus 5 g l−1 ammonium acetate (Egel, 1971).
Generation of cells synchronised for cell cycle
progression
Lactose gradient synchronisation according the methodology of
Mitchison (1970) was used to isolate early G2 cells for generation of
cultures synchronised for cell cycle progression. A two-step synchronisation procedure was used: cells were initially synchronised on a 45
ml 10-40% lactose gradient and then immediately passed over a
second (12 ml) 10-40% lactose gradient in order to achieve optimal
size uniformity. Following washing, cells were inoculated into fresh
medium. After removal of a 20 ml sample for incubation at 25°C to
follow synchrony, the remainder of the culture was incubated at 36°C.
107 cells were harvested every 30 minutes from the 36°C culture and
prepared for immunofluorescence microscopy. The cell plate index at
25°C was monitored using Calcofluor staining (Moreno et al., 1990)
and scoring only those cells with an equatorial brightly staining bar
as having septa. cdc14.118h− cells proved to be too leaky for synchronous culture analysis as many cells septated at 36°C in rich or
minimal medium. cdc7.24 synchronised cells only arrested efficiently
when the experiment was conducted in minimal medium.
Immunofluorescence microscopy
TAT1 anti-α tubulin antibody (Woods et al., 1989) or affinity-purified
Sad1 antibodies (AP9.2) (Hagan and Yanagida, 1995) were used to
visualise microtubules and the SPBs, respectively. Cells were fixed
with 3.6% formaldehyde and 0.2% glutaraldehyde as described in
Hagan and Hyams (1988), with the exception that, at the end of the
procedure, after the first wash in antibody buffer, cells were washed
once in phosphate-buffered saline (PBS), resuspended in 0.5 µg ml−1
4,6-diamidino-2-phenylindole (DAPI) in PBS, pelleted and resuspended in 25 µl PBS containing 0.1% sodium azide. Fluorescein
isothiocyanate (FITC) conjugated goat anti-mouse antibodies (EY
labs) were used to localise TAT1 primary antibodies and CY3-conjugated goat anti-rabbit antibodies (Sigma) for AP9.2 primaries. Double
staining was performed by sequential incubation in mixtures of both
primary and then both secondary antibodies. False colour images were
produced as described in Lange et al. (1995).
Drug treatment
Thiabendazole (TBZ) was prepared as a 400× stock solution in
dimethyl sulphoxide (DMSO) as described by Walker (1982) and
added directly to the culture to a final concentration of 50 µg ml−1.
Hydroxyurea (HU) was added to the culture from a 1 M stock solution
in water to a final concentration of 10 mM.
RESULTS
Nuclear movements during and after fission yeast
mitosis
During fission yeast mitosis the nuclei are separated to the cell
ends at a rate of 1.6 µm per minute by an extensive anaphase
B movement (McCully and Robinow, 1971; Hagan et al.,
1990). When anaphase is complete, two microtubule-organis-
Nuclear positioning in S. pombe 1853
A
B
C
D
E
F
G
Fig. 1. False colour images of wild-type fission yeast
cells in mitosis. Wild-type fission yeast cells were
processed for immunofluorescence microscopy to
visualise their SPBs (green), microtubules (red) and
chromatin (blue) after growth to mid-logarithmic
growth phase at 25°C. The lower panels represent a
combination of the fluorescence information, which is
shown in the upper panel, and the phase-contrast image
of the same cell. Following spindle formation (A), the
spindle undergoes an extended anaphase B (B,C) to
achieve maximal nuclear separation. Thereupon two
MTOCs are activated in the middle of the cell (D,E)
giving rise to the PAA (F,G). As the nuclei move
towards the middle of the daughter cells, they assumed
a stretched profile with the SPB at the tip (F, upper
nucleus).
ing centres (MTOCs) are activated in the middle of the cell and
nucleate a structure referred to as the post-anaphase array
(PAA) (Hagan and Hyams, 1988; Horio et al., 1991).
Initially the PAA is seen as two dots at the cell equator
(Fig. 1E) but, as more microtubules are nucleated from the
organising centres, the dots are replaced by two simple asters.
One such array is shown in the focal plane in Fig. 1F. With
time, the microtubules extending from these MTOCs develop
into complex overlapping arrays with microtubules extending
to the cell tips. Thus the complexity of the microtubulestaining pattern of the PAA increases with time and this
increasing complexity can be used as a marker of the time
that has elapsed since spindle dissolution. It has been
proposed that microtubules of the PAA form a ring around
the cell equator (Pichová et al., 1995); however, we are
restricting our description to that of the nucleating capacity
of the PAA MTOCs. Coincident with PAA development is the
movement of the nuclei back towards the equator, at the same
rate as during anaphase B, until they stop at the locations that
will become the middle of the daughter cells (Hagan et al.,
1990).
Simultaneous staining of cells to reveal the location of the
SPB, microtubules and chromatin showed that, in many cases,
whilst the SPB was on the side of the nucleus facing away from
the cell equator during mitosis (Fig. 1A-D), it was often in the
reverse orientation facing towards the equator following
mitosis when the microtubules were in a PAA. In many cases
when the PAA was still relatively unelaborated, nuclei often
appeared stretched with the SPB at the tip of the chromatin
facing the PAA at the cell equator (upper nucleus, Fig. 1F).
This would be consistent with the SPB pulling the nucleus
towards the central organisers. In cells with more complex, presumably later, PAAs, the nuclei attained their correct position
and, although still associated with microtubules, the SPB was
no longer at the very edge of the chromatin (lower nucleus,
Fig. 1).
These data suggested that the SPB may be the transmission
point for forces that pull the nucleus during its migration back
towards the cell centre after mitosis.
Increasing the distance moved by nuclei after
anaphase highlights the role of the SPB in nuclear
migration
As spindle break down can occur before the nuclei have
migrated to the cell tips (Fig. 1D,E), concrete conclusions
about the role of the SPB in post-anaphase nuclear movements
could not be drawn from observations in wild-type cells. Cells
were therefore manipulated so as to increase the distance that
the nucleus must move to accentuate nuclear movement after
spindle dissolution. The cdc25.22 mutation delays mitotic
commitment and concomitantly increases cell and spindle
length at division (Hagan et al., 1990), which means that the
nuclei have greater distances to move after anaphase to reach
the correct position in the middle of the daughter cells in
cdc25.22 strains grown at 26°C.
As in wild-type cells, the SPB in cdc25.22 cells always faced
the cell tips during anaphase B and during the initial stages of
spindle breakdown (Fig. 2A-C). PAA maturation saw more and,
on average, longer, microtubules, which extended beyond the
SPBs to the cell ends. This tubulin staining often appeared to
kink when contacting the SPBs (Fig. 2E). In many cases, the
SPBs were orientated towards the equator and the nuclei had a
cone-shaped profile which formed a point at the SPB, consistent with a role for the SPBs in pulling the nucleus along microtubules (Fig. 2E,F). Once the nuclei were correctly positioned,
the chromatin lost its pointed appearance and resumed the
crescent shape typical of interphase (left nucleus in Fig. 2G).
Although these data are consistent with a role for the SPB
in nuclear positioning following mitosis, they are qualitative
rather than quantitative. Quantitative data were obtained by
following cytoskeletal and nuclear morphology in a synchronous culture. The advantage of following the frequency of
different characteristics in a synchronous culture lies in the fact
that at the start of the experiment all of the cells are in, more
or less, the same stage of the cell cycle and that all cells
progress through the subsequent cell cycle with similar
kinetics. Therefore, by monitoring the phenotypic changes that
occur in the culture with time, we can see what course a typical
cell takes (Mitchison, 1970).
1854 I. Hagan and M. Yanagida
Fig. 2. Increasing the distance migrated by nuclei following mitosis by mutation of the cdc25 gene shows that the SPBs lead the nuclei towards
the cell equator. The figure shows a series of cells illustrating (from top to bottom) microtubules, SPBs, chromatin and the position of the
chromatin within the cell. The information is summarised in cartoons at the bottom of each figure set. (A) A cell in the final stages of anaphase
B just before the central MTOCs nucleate the PAA (B,C). As the PAA matures some microtubules contact the SPBs which start to move
towards the cell centre (D). In cells with more complex PAAs, the nuclei had a stretched appearance pointing towards the equator, with the
SPBs at the apex. This suggests the SPBs are at, or extremely close to, the point of force transduction that pulls the nuclei back towards the
centre of the cell (E,F). Once the nuclei were correctly located the crescent shape was resumed and the SPB positioning once again became
randomised (G). cdc25.22h− cells were grown to mid-log phase at 26°C and fixed for combined anti-tubulin and anti-Sad1 immunofluorescence
microscopy. Bar, 5 µm.
Nuclear positioning in S. pombe 1855
spindle breakdown, when neither SPB faces the cell centre (3050 minutes), was followed by an interval during which the
SPBs face the cell centre before a more random distribution is
adopted. Since the nuclei move towards the cell centre
following spindle breakdown, these data are consistent with the
images in Fig. 2 and show that the SPB leads the nucleus back
towards the centre of the cell immediately after spindle dissolution.
Fig. 3. The frequency of cells with equator-oriented SPBs peaks just
after spindle breakdown. The figure shows the frequency of events in
a synchronous cdc25.22 culture in rich YES medium that was held at
the restrictive temperature for 145 minutes after selection of the
small G2 cells and returned to the permissive temperature in rich
YES medium at Time 0 on the graph. (a) Interphase nuclei (j)
decline as mitotic spindles (h) start to appear after 10 minutes.
Subsequently cells containing both PAA and spindles (n) peak at 40
minutes, followed by those with PAAs alone (d) at 80 minutes.
n>281 for each time point. (b) Scoring cells with PAAs but no
spindles as to whether or not there was any chromatin between the
SPB and the centre of the cell in either or both of the two nuclei
showed that whilst the SPB is not initially facing the centre (r) at 40
minutes, as would be expected immediately after spindle breakdown,
cells with their SPB closer than the chromatin to the cell centre (s)
predominate between 50 and 70 minutes after breakdown. These are
replaced by cells with no SPBs facing the cell centre at 70 minutes.
(n=119 at 40 minutes and 263 at 100 minutes; for all other time
points n>320).
The ability to synchronise cultures by size selection of
smaller G2 cells was combined with the ability to synchronise
mitosis further with a transient arrest at the G2/M boundary by
brief inactivation of the cdc25.22 gene product (Hagan, 1988).
cdc25.22 cells isolated from a lactose gradient were grown at
36°C before being returned to 24°C after 145 minutes
(probably consistent with a maximum arrest at the G2/M transition of 40 minutes). Within 10 minutes of return to the permissive temperature spindles were formed with a peak of 72%
(Fig. 3 open squares combined with open triangles). The wave
of mitosis (Fig. 3A, open squares) was followed successively
by activation of the central organising centres (Fig. 3A, open
triangles) and the more extensive PAAs (Fig. 3A, circles).
Scoring cells with PAAs as to whether either SPB was facing
the centre of the cell (Fig. 3B, circles) or not (Fig. 3B,
diamonds) showed that the brief period immediately after
The SPB is activated as spindle microtubules
depolymerise
The combination of unprecedented mitotic synchrony and the
ability to detect the relative location of the SPBs and the microtubules revealed a previously unrecorded microtubule pattern.
Several faint microtubules extended from the SPBs of cells in
which the PAA was very immature. This staining pattern was
seen in cdc25.22 cells, which were released from temporary
arrest at 36°C (data not shown) as well as in the control synchronous culture maintained at 24°C (Fig. 4A-F, arrows
indicate the SPBs from which the arrays extend). Re-examination of wild-type cells stained with increased concentrations
of anti-tubulin antibodies revealed similar patterns (Fig. 4G-I).
These microtubules were often much fainter than the strongly
staining astral microtubule bundles, which contain three microtubules (Tanaka and Kanbe, 1986), raising the possibility that
they may represent individual microtubules. This new pattern
is consistent with activation of the SPB immediately after
mitosis and is reminiscent of the cytoskeletal changes that
accompany spindle dissolution in Schizosaccharomyces
japonicus (Alfa and Hyams, 1990).
Multiple SPBs within the same nucleus respond to
the same signal
The involvement of the SPB in nuclear positioning immediately following anaphase led us to question whether the SPB
might also be responsible for mediating nuclear positioning
during interphase. We reasoned that we could determine
whether the SPB responds to specific signals to determine
interphase nuclear position by putting multiple SPBs in a single
nucleus. If the SPB is used to direct nuclear positioning all of
the SPBs would be expected to respond in concert to the
nuclear positioning signal and cluster at a single point. If, on
the other hand, the SPB plays no role in nuclear positioning in
interphase they would be more likely to maintain independent
positions.
Despite undergoing multiple S phases and forming an apparently normal spindle, cut1 mutants are unable to segregate their
DNA and multiple SPBs accumulate in single nuclei (Uzawa
et al., 1990). As septation continues, the nuclei in these cells
are randomly cleaved by the division apparatus and the cell
dies. Cell cleavage can be avoided by the introduction of additional mutations which inhibit septation and so allow a single
nucleus to continue through multiple cell cycles, accumulating
SPBs as it does so (Uzawa et al., 1990). Fig. 5 shows such a
double mutant, cut1.206 cdc11.123. Whilst many SPBs were
seen during the defective mitoses (Fig. 5G-I), only a single
large dot of Sad1 staining was seen in interphase cells that had
been through several abortive attempts at nuclear division (Fig.
5A-F), suggesting that all SPBs were now responding to the
same signal and clustering at one point. In every case in which
we could manipulate the cell cycle to increase SPB numbers
1856 I. Hagan and M. Yanagida
A
Fig. 4. Post-anaphase nuclear movement is
accompanied by the appearance of previously
undetected SPB-associated microtubules. The
figure shows images of wild-type (G-I) and
cdc25.22 (A-F) cells grown at the permissive
temperature of 24°C in rich YES medium. The
cdc25.22 cells were synchronised by size
selection from two successive lactose gradients.
Different representations of the staining of the
cells are presented in sets of three, separated by
the white dividing lines. False colour images in
A,D,G show microtubules (red) and SPBs
(green) with DAPI (lilac) added in B,E,H. The
locations of these structures relative to the cell
outline can be seen by comparing the combined
DAPI and phase-contrast images of the cells in
C,F,I with the fluorescence images. Arrows
indicate SPBs that have short microtubules
associated with them. The data are consistent
either with the spindle fraying upon spindle
breakdown, or new nucleation, presumably from
its cytoplasmic face, by the SPB. The central
tubulin-staining mass in each case represents the
PAA. The fact that the microtubules appear
longer in B where the PAA is more advanced
than H suggest that the array in B is more mature
implying that it is nucleation rather than fraying
that is recorded.
B
C
D
G
E
H
F
I
in a single nucleus, all the SPBs appeared to respond to the
same positional signal (data not shown).
Multiple SPBs on different nuclei in the same cell
respond to the same signals
One caveat to the explanation for SPB clustering given above
is related to our previous demonstration that the centromeres
are always associated with the SPBs during interphase
(Funabiki et al., 1993). If it were possible for the centromeres
of one chromosome to simultaneously interact with two SPBs,
the clustering of SPBs in a single nucleus could be due to centromere-mediated “cross linking”.
To test this possibility, we looked at the SPB and nuclear
localisation in septation mutants in which a single cell has
multiple SPBs that are each associated with a single, different,
nucleus. If all of the SPBs were responding to the same signal,
they would again cluster and this would be independent of any
intranuclear centromere interaction.
Following the first mitosis at the restrictive temperature in
any of the septation-deficient mutants, cdc14.118, cdc7.24 or
cdc7.A20, the nuclei migrated back towards the cell equator
(Fig. 6A-D). Thus, in the subsequent interphase the nuclei
“paired up” in the centre of the cell with the SPBs at the site
where the two nuclei contacted each other, often with the SPBs
no longer distinguishable as separate entities. These observations showed that two SPBs associated with two separate nuclei
in the same cell responded to the same nuclear positioning
signal. It may be significant that these paired nuclei were
invariably in a “back-to-back” configuration with their
nucleoli, as determined by the morphology of the DAPIstaining region (Toda et al., 1981), facing the cell ends.
Cryptic controls over nuclear positioning revealed in
septation mutants
Analysis of septation mutants undergoing a second mitosis at
the restrictive temperature suggested that the signal directing
nuclear positioning to the middle of the daughter cells directly
after mitosis may still be present. However, the nuclei in these
cells do not stay at these two positions but, instead, subsequently move to the centre of the cell, as would be appropriate for a normal interphase cell.
The morphology of the microtubules during these transitions
is graphically shown in the micrographs of cdc14.118 cells in
Fig. 7, whilst the false colour images of cdc7.24 cells in Fig.
8 show the locations of microtubules, SPBs and chromatin
relative to the outline of the cell periphery. Following two
Nuclear positioning in S. pombe 1857
Fig. 5. All the SPBs in cut1.206 cdc11.123
double mutants containing multiple SPBs in one
nucleus go to the same point during interphase.
(A,D,G) Anti-tubulin; (B,E,H) anti-Sad1;
(C,F,I) DAPI-phase contrast images. Cells were
incubated for 6 hours at the restrictive
temperature. Whilst multiple SPBs can be seen
in the defective mitosis in G-I, all the SPBs are
clustered in the two interphase cells shown in AC and D-F. Bar, 5 µm.
normal mitoses, cells contained four, round, evenly spaced
nuclei (Figs 7A, 8A-C) whose nucleoli and SPBs were
randomly oriented (Fig. 7B). As PAA complexity increased,
the nuclei became pointed at the SPBs and the microtubules
appeared to kink when contacting the SPBs (Figs 7C,D, 8D).
Further development of the PAA, so that it more closely
resembled an interphase cytoskeleton, resulted in two different
patterns. In one, the four nuclei paired up in twos (Fig. 7E,F,
8E) whereas in the other three nuclei clustered and one
remained isolated (Fig. 7G). It may be significant that, in the
class with three nuclei grouped together, many cells had predominantly one or two very strong tubulin-staining filaments
(probably microtubule bundles) along which all SPBs were
aligned. Comparing the relative positions of the SPBs to the
Fig. 6. “Back-to-back” nuclear positioning in cdc14.118 mutants
incubated at the restrictive temperature for 3 hours to inhibit
cytokinesis. (A-D) Anti-tubulin, anti-Sad1, DAPI and DAPI phasecontrast images, respectively, of the same cdc14.118 cells after the
first mitosis at the restrictive temperature of 36°C. As in wild-type
and cdc25.22 cells, the SPB appeared to lead the nucleus along the
microtubules of the PAA back towards the cell equator immediately
following mitosis (cell on right). Once an interphase microtubule
array had been established, a typical “back-to-back” configuration
was achieved as the two nucleoli faced the cell tips and the SPBs were
tightly associated into one brightly staining dot in the cell centre.
Cells were grown in rich YPD medium to early log phase and then
shifted from the permissive to the restrictive temperature. Bar 5 µm.
chromatin in Fig. 8D and E with those in Fig. 8F and G
revealed a subtle difference. In Fig. 8D,E, the SPBs of each
nuclear pair lie between the two “back-to-back” nuclei,
whereas those in Fig. 8F,G were on the same side of the nuclei,
facing the cell centre. This suggested that, in Fig. 8D and E,
the direction of movement was of the two nuclei towards each
other but that, in Fig. 8F and G, both nuclei of each pair were
moving in concert towards the cell centre. In the final class,
1858 I. Hagan and M. Yanagida
Fig. 7. Complex patterns of nuclear positioning are seen following the second mitosis at the restrictive temperature in the septation mutant
cdc14.118. As in Fig. 2, each panel is composed of separate images to show from top to bottom: microtubules, SPBs, chromatin and the
relationship of the chromatin to the cell outlines. Following a double mitosis (A), a PAA was initiated at the cell centre (B). As the PAA
developed, the nuclei began to take on a pointed appearance with SPBs at the tips of the points (C) which, in cells with more mature PAAs (D),
seemed to be pulling them to two common points (E,F). As the microtubules returned to less complex, more strongly staining patterns (G),
some cells are seen with three nuclei clustered and one free. Interestingly, in this situation the SPBs are virtually always on the same
microtubule (in addition to the microtubule in the focal plane in G there are others in different focal planes). As many cells show four nuclei
clustered in the cell centre (H,I), this is assumed to be the terminal phenotype (see Fig. 9). Cells were grown to early-log phase in rich YPD
medium at the permissive temperature prior to incubation at the restrictive temperature for 4G hours. Bar, 5 µm.
Nuclear positioning in S. pombe 1859
which predominated in the culture with prolonged incubation,
and thus is presumably the final state of cells containing four
nuclei, all four SPBs clustered in the cell centre (Figs 7H,I,
8H).
Synchronous culture analysis shows that nuclei
move successively to two different locations in a
cell cycle-dependent manner
To determine whether there was a temporal order to the phenotypic classes described above, these classes were quantified
in a synchronous culture of cdc7.A20 at the restrictive temperature of 36°C (Fig. 9). The synchrony of the culture was
reflected in successive mitotic peaks (Fig. 9A). The frequency
of first and second cycle phenotypes are shown separately in
Fig. 9B and C, respectively.
As expected, the first mitotic peak (110 minutes) was immediately followed by a peak in cells in which the nuclei were
post-mitotic but not in a “back-to-back” configuration (Fig. 9B
circles). Subsequently cells with “back-to-back” interphase
nuclei peaked at 160 minutes (Fig. 9B, triangles).
In the early stages of the second cycle, there were peaks both
of cells in which there were no apparent SPB/SPB clustering
(Fig. 9C, squares) and those in which SPBs grouped as one or
two off centre pairs (Fig. 9C, diamonds). These peaks were
followed by a peak of cells in which all four nuclei were
clustered in the cell centre (Fig. 9C, circles). The progressive
order of peaks in Fig. 9C indicated that, in the individual
members of the population, the phenotypes with all four nuclei
were initially separate, then paired up off centre before all clustering in the middle of the cell. The greater area defined by the
peak representing all four nuclei centrally located indicates that
this stage persists for longer than the others.
The curves defined by scoring cells with three and four
clustered nuclei were symmetrical showing that they represent
transient states with no particular delay at any stage. In
contrast, the curve defined by scoring paired nuclei declined
much more slowly indicating either that central clustering was
slow or that a subpopulation did not make the transition from
paired nuclei to central clustering. To differentiate between
these possibilities, a delay to cell cycle progression beyond the
four nuclei stage was imposed to determine whether this delay
would allow sufficient time for all four nuclei to cluster in the
middle, which would be consistent with the first possibility.
Because S phase immediately follows spindle dissolution in
logarithmically growing cultures, the addition of the DNA
synthesis inhibitor hydroxyurea (HU) after the first wave of
mitosis will not affect the ability of the cells to execute the
second mitosis (because the DNA replication for this division
is already complete). It will, however, block further cell cycle
progression into a third mitosis due to checkpoint-dependent
inhibition of mitosis (Carr, 1995).
The culture incubated at 36°C in Fig. 9 was split after the
first mitosis (150 minutes) and HU added to one half (Fig. 9D).
The resulting extension of interphase led to fewer cells with
paired nuclei but it did not result in a complete loss of the phenotypic class. This shows that some cells in the actively cycling
population scored for Fig. 9C did not implement the second
stage of nuclear positioning to move all nuclei to the cell
centre.
The inability of some cells to cluster all four nuclei after the
second mitosis anticipates the patterns seen following the next,
third, mitosis as the nuclei aggregate in two or three clumps
(Fig. 10), indicating that at this stage cells were no longer able
to establish a single “nuclear positioning zone” in the centre of
the cell.
Data from these experiments showed that, in the majority of
cells following a second mitosis in a single cell, the nuclei
initially moved to two non-central locations before all grouping
in the cell centre. This suggests that the SPBs were responding to cell cycle-regulated nuclear positioning signals.
Nuclear clustering in septation mutants is disrupted
by anti-microtubule agents
As the preceding staining patterns suggest that microtubules
are used as the “rails” along which the SPBs move nuclei, the
effect of disrupting microtubule integrity on nuclear positioning was examined. The data in Figs 7, 8 and 9 suggested that
the most sensitive test would be the ability of the four nuclei
to cluster following a second mitosis in a septation mutant. A
synchronised culture of cdc7.24 cells was therefore treated
with the anti-microtubule drug thiabendazole (TBZ) at the
restrictive temperature (Walker, 1982).
Synchronised cdc7.24 cells were inoculated into minimal
medium at 36°C to inhibit septation and the culture was split
into three aliquots after the first mitotic peak (210 minutes).
TBZ was added to a final concentration of 50 µg ml−1 to one
aliquot, DMSO (the solvent for TBZ) to another and the third
was left as an untreated control (Fig. 11C,B and A, respectively). The profile of the DMSO-treated aliquots mimicked the
untreated sample; the frequency of cells with one or two nuclei
decreased whilst those with four and then eight increased, indicating that DMSO had no effect upon cell cycle progression
(Fig. 11A,B). In contrast, in the TBZ-treated culture, the
decline in cells with one and two nuclei and the rise of those
with four nuclei was much less pronounced, and there was no
commitment to a third mitosis (Fig. 11C) showing that cell
cycle progression was severely compromised by drug
treatment.
Interphase microtubule cytoskeletons were largely intact
despite the drug treatment, although they were considerably
more disorganised (data not shown). Fig. 11E shows the
relative frequencies of nuclear positioning profiles in TBZtreated cells containing an interphase cytoskeleton and four
nuclei at 480 minutes. In stark contrast to the predominance of
cells with all four SPBs clustered and the low frequency of
cells with all four SPBs separate in controls (Fig. 11D), the
majority of interphase cells with four nuclei in the TBZ-treated
culture (40%) showed no apparent SPB clustering at all, and it
was the minority that positioned all four SPBs together (Fig.
11E).
These data indicate that microtubule integrity is required for
nuclear positioning throughout interphase in fission yeast.
Central MTOCs are not observed following meiosis I
The possibility that the SPB may be responsible for directing
nuclear migration following mitosis suggests that the function
of the PAA may be to ensure that an interphase cytoskeleton
is re-established sufficiently rapidly to orchestrate correct
nuclear positioning before cytokinesis, thus preventing inappropriate partitioning or cleavage of the nucleus by the septum.
If this were the case the central microtubule organisers might
not be expected to be required after a nuclear division that is
1860 I. Hagan and M. Yanagida
Fig. 8. False colour images
showing a range of cytoskeletal
arrangements in cells in which
cytokinesis has been inhibited
by incubation of cdc7.24 cells
at the restrictive temperature
for 300 minutes. In each panel,
three images of the same cell
are presented: one shows
microtubules (red), chromatin
(blue) and SPBs (green – in a
single channel but this appears
as yellow when the red and
green channels co-incide and it
appears white when overlaying
both red and blue); another
shows SPBs in red and
chromatin in green for
maximum contrast of these two
structures; the final image
shows a standard DAPI/ phasecontrast image. In all cells
shown, the SPBs are all in the
same focal plane. Following
the first mitosis, the nuclei
assumed a typical “back-toback” configuration (Fig. 6)
which was maintained (A)
until the next mitosis (B). The
relative positions of the SPBs
and the nuclei suggested that,
following a double mitosis
within a single cell (C), the
nuclei migrated towards each
other to take up positions at
two points (E) along the
microtubules of the PAA (D).
Where pairs of nuclei were
closer than those in D the two
SPBs were no longer between
the two nuclei (E) but rather
orientated towards the equator,
and appeared to be leading
pointed nuclei towards the cell
centre (F). These images,
combined with the data in Fig.
9, suggested that, after pairing,
the nuclei move back to the
cell equator (G) to cluster with
their SPBs as close as the bulk
of the chromatin will let them
be to a point roughly in the
centre of the cell’s long axis
(H).
A
B
C
D
E
F
G
not succeeded by cytokinesis or post-anaphase nuclear
migration: the first meiotic division.
When homothallic cells are starved for a source of nitrogen,
they initially undergo a series of mitotic divisions that result in
a reduction in cell size, then they conjugate and undergo
meiosis. Fig. 12 shows two zygotes at successive stages at the
end of meiosis I. In contrast to cells after the completion of
mitotic anaphase, no MTOCs were seen, rather microtubules
appeared to nucleate at random in the cytoplasm. The lack of
H
a PAA and nuclear migration after this nuclear division is consistent with the primary role for the PAA being to ensure that
the nuclei are not cleaved by the incipient septum rather than
being required for the final stages of nuclear division per se.
DISCUSSION
A well-established strength of genetic systems is the ability to
Nuclear positioning in S. pombe 1861
Fig. 9. Progressive appearance of phenotypes in a synchronous
cdc7.A20 culture grown at the restrictive temperature shows that
nuclei pair up in two non-equatorial positions prior to clustering
equatorially. Data from a single culture are presented in three
separate panels each containing different information.
(A) Successive waves of mitoses at the restrictive temperature.
Cells were scored for being in their first (j), second (r) or third
(d) mitosis within a single cell and the results plotted against time.
The synchrony and sequential appearance of the phenotypes is
apparent. n>220 for each time point. (B) First cycle characteristics.
The following phenotypes are indicated: j, interphase; r, first
B
D
mitotic spindle; d, post-first mitosis but not “back-to-back”; m,
“back-to-back”. Clearly individuals within the population were
progressing from interphase into mitosis and then the nuclei took a
little while to pair up finally as “back-to-back” nuclei in the centre
of the cell. (C) Second cycle characteristics. n>250 for each time
point. The following phenotypes are indicated: j, 4 nuclei all
SPBs separate; r, cells containing one or two off centre paired
SPBs; n, cells containing three SPBs in one cluster and one free;
d, cells containing all four SPBs clustered together in the centre of
the cell. As the frequency of the “four SPBs separate” and “cells
containing one or two pairs” rose almost immediately after the
second mitotic peak to simultaneously peak around 210 minutes,
these two classes predominate immediately after the second
mitosis and were transitory rather than terminal phenotypes. The
terminal phenotype following the second mitosis appeared to be four SPBs all clustered in the cell centre as its peak was later and higher. Whilst
the “three SPBs clustered and one free” and “all four nuclei centrally located” phenotypic classes rose and fell with similar kinetics, the peak of
cells with paired nuclei (r) declines more gradually than it rises. n>290 for each time point (for t=210, n=664). (D) The addition of hydroxyurea
shows the inability of some cells to reposition their nuclei centrally. The culture in A was split into two and HU was added to a final
concentration of 10 mM at t=150 minutes to provide the data for D. The symbols are the same as in C with the exception that those for ‘none
clustered’ and paired are exchanged. The data show a much greater accumulation of cells with four centrally located nuclei, but more
significantly, the frequency of cells with pairs of nuclei does not decline to zero, rather, it appears to decay slowly suggesting that not all cells go
on to locate their nuclei centrally. Cells were grown to mid-log phase in YES medium before selection of small G2-phase cells on lactose
gradients and washing in and inoculation into fresh YES at 36°C at the beginning of the experiment. n>330 for each time point
C
A
Fig. 10. Nuclei cluster in either two
or three groups after the third
mitosis at the restrictive temperature
in cdc7.A20 cells. Two
representative fields of cells from
the culture used in Fig. 9 after 390
minutes at the restrictive
temperature. The representation of
the signals as false colour images is
the same as that for Fig. 8.
(A-C) One cell with nuclei clustered
in two groups and another in three,
whilst both cells in D-F have two
clusters. As in Fig. 8, the impression
given by the relative location of the
SPB and chromatin signals is that
the SPBs are actively involved in the
generation and maintenance of the
clusters. The relative frequency of
the cells were scored as to whether
there was no obvious grouping of
the interphase nuclei, they were
found in two separate clusters or
they were grouped in three distinct
clusters. At 360 minutes; these
frequencies were 32:53:15,
respectively, while at 390 minutes
the relative ratios were 22:55:23.
A
B
C
D
E
F
1862 I. Hagan and M. Yanagida
A
C
B
D
E
Fig. 11. The addition of the microtubule inhibitor thiabendazole
(TBZ) slows mitotic progression and disrupts nuclear clustering in
the cdc7.24 cytokinesis mutant. A synchronous culture was split into
three after the first mitotic peak at 240 minutes and, whilst one
aliquot was untreated, DMSO was added to another and TBZ (to a
final concentration of 50 µg ml−1 in DMSO) was added to the third.
Each culture was then fixed after various incubation times and scored
for the number of interphase nuclei as a marker for cell cycle
progression. The data are presented graphically: (u) single nuclei,
(e) two nuclei, (s) four nuclei and (n) eight nuclei. (A) The control
with no treatment, (B) the control with DMSO treatment, and (C) the
effect of TBZ treatment. Although DMSO has no gross effect on the
appearance of successive waves of mitoses, the rate of mitotic
progression, as shown by the appearance of cells in the subsequent
interphase, is severely attenuated in the aliquot treated with TBZ.
The lines are drawn by eye to indicate the trend in the frequencies of
the respective classes. (D) The frequency of cells with four nuclei
either separate (d) or clustered (j) in the culture used for A. (E) The
relative configurations of nuclei in the interphase cells containing an
interphase cytoskeleton and four nuclei at time point 480 minutes.
Instead of most cells exhibiting four centrally located nuclei as is the
case for cells incubated in the absence of TBZ (D), such cells were in
a minority in the drug-treated culture and the majority of cells (40%)
show no apparent SPB grouping. Cells were propagated in minimal
EMM2 medium throughout the experiment.
alter cellular environments to test hypotheses. Here we have
exploited mutations in a number of cell division cycle genes in
order to reveal the otherwise cryptic role of the SPB in positioning the interphase nucleus to different locations at
different cell cycle stages. The SPB uses microtubule tracks to
position the interphase nucleus
Fig. 12. Microtubule nucleation after meiosis I is not mediated by
central organising centres. (A,B and C,D) Anti-tubulin and
DAPI/phase contrast images, respectively, of two wild-type fission
yeast zygotes at the end (A,B) and immediately after (C,D) meiosis I.
Whilst cytoplasmic microtubules were seen in the presence of a
complete, strongly staining, spindle at the end of meiosis I in A, no
organising centres are observed. The random arrays in C contrast
sharply with the MTOC organised patterns seen at similar stages of
mitosis (Fig. 1F,G). Bar, 5 µm.
Our initial observations in wild-type cells and those dividing
at an increased size show that the nucleus is preceded in its
movements around the cell by the SPB. We conclude from
experiments in which nuclear positioning is disrupted
following treatment with microtubule depolymerising drugs
that the SPB is migrating along microtubule tracks. This conclusion is consistent with defects in nuclear positioning arising
from overproduction of an SPB component and mutation of the
atb2+ gene (Hagan and Yanagida, 1995; Verde et al., 1995;
Yaffe et al., 1996). Because the centromeres are always associated with the interphase SPB (Funabiki et al., 1993), the
genome follows the SPB in its migrations and thus nuclear
positioning is achieved.
SPB-mediated movement could be achieved either via
motors, or via static non-motor attachment to moving microtubules (Morris et al., 1995). Support for the existence of SPBbound motor proteins comes from the localisation of known
motor proteins to, or near to, SPBs in both budding and fission
yeasts (Page et al., 1994; Yeh et al., 1995; Hagan and Yanagida,
1992). Their presence is also strongly suggested by the SPBled nuclear migration in fission yeast during karyogamy and
meiotic prophase (Chikashige et al., 1994).
The nuclear profiles after the second mitosis in septation
mutants (Fig. 8D) suggest that the SPBs initially move towards
two non-central points even though their eventual destination
is the cell centre. This seems to be the case even if they started
out close to the cell centre immediately after division and
Nuclear positioning in S. pombe 1863
suggests active movement to these two points. If one assumes
that all of the single microtubules of the early PAA (Horio et
al., 1981) extend from the cell equator with the same polarity,
then activating a unidirectional motor would result in simultaneous migration of all four nuclei either towards, or away from,
the MTOCs of the PAA. As this is not observed, the SPBs must
be capable of moving in either direction along the microtubules
of the PAA.
There are two ways in which the SPB might use microtubule
motors to generate such movement. Firstly, the SPB could
posses both plus-end and minus-end directed motors or a single
motor with regulatable directionality of motility (Euteneuer et
al., 1988). A second, less attractive, possibility is that the SPB
may contain only one unidirectional motor. Such an SPB could
still move towards and away from the cell centre but only if
the PAA consisted of anti-parallel microtubule bundles, which
is not consistent with the one electron micrograph of the PAA
published to date (Horio et al., 1981).
It is also possible that there are no motor proteins at the
SPB and that the SPB is just a tethering point to attach the
nucleus to moving microtubules which are being pulled
around the cell by either plasma membrane-bound motors, or
motors generating microtubule/microtubule sliding within the
cytoplasmic bundles (Plamann et al., 1994; Morris et al.,
1995). In this case, the SPB must be able to attach and detach
depending on where it is relative to a positioning signal. In
other words, it must be able to attach to the moving microtubule if the nucleus is in the wrong place, and detach when
correctly positioned. It might not be necessary (or desirable)
for the SPB to release completely from the microtubule,
however, if the attachment was via a regulatable ratchet it
would allow alternating tight binding, when the SPB would
be pulled along by the MT, with loose binding, where the
microtubule slides past the SPB whilst maintaining long-term
attachment. As ratcheting has been described for dynein (Vale
et al., 1989), this model can be viewed as a variation on the
SPB-bound motor model.
Of these two possibilities, bi-directional, motor-mediated
SPB motility seems most likely to account for the tight SPB
clustering that we report in fission yeast septation mutants.
Increasing the number of SPBs in a cell reveals
cryptic controls
One of our key observations demonstrating the fidelity of SPBmediated nuclear positioning was the finding that multiple
Fig. 13. A cartoon summarising nuclear
positioning in cytokinesis mutants. The SPBs
are represented as dots on the nuclei. The
temporal order of the images is from left to
right. The two arrows indicate the location of
the two proposed nuclear positioning signals
which occur sequentially following the end of
mitosis. This signal pattern is highly
reminiscent of that which must occur in wildtype cells.
SPBs on one nucleus, or single SPBs on different nuclei within
a single cell aggregate at the same point/s. This showed that
the different SPBs responded to common signals.
Immediately following the second mitosis in septation
mutants, the signals for nuclear localisation ensure that the four
nuclei were positioned in two regions roughly a third of the
way from either tip (Fig. 13). Later, all four nuclei moved to
the centre of the cell. This suggests that instead of being an
irreversible commitment to keep the two points defined as the
cell mid-points prior to cytokinesis, this signal is turned off and
a subsequent signal turned on to define a single central point
as the cell cycle progresses. These data strongly suggest that
the similar positioning of nuclei in wild-type cells does not
arise from a default maintenance of the two post-mitotic
positions after cytokinesis but that there is a switch to a central
location at some point in early G2 phase of the cell cycle after
cytokinesis is completed.
Microtubules, SPBs, the PAA and nuclear
positioning in fission yeast
The temporal coupling of nuclear division and cytokinesis
means that it is important to establish the correct nuclear positioning quickly after mitosis to avoid fatal random cleavage of
the nucleus by the septum (the “cut” phenotype; Hirano et al.,
1986), or the generation of binucleate and anucleate cells.
Thus it may be significant that two MTOCs are specifically
activated at the end of nuclear division in mitosis but not in
meiosis. Analysis of the cell cycle mutant wee1.50 has established that PAA formation is not required for cytokinesis
(Hagan and Hyams, 1988) indicating that the difference in the
microtubule organisation after mitotic and meiotic divisions
must be for some other function, perhaps nuclear positioning.
Nuclear positioning in the meiotic cycle is not as important as
in the mitotic cycle because the positional requirements
following meiosis are that the nucleus is encapsulated within
the spore wall rather than positioned to avoid cleavage by a
septum. Spore encapsulation is guaranteed by the initiation of
the spore wall from the outer face of the SPB during
metaphase of meiosis II (Hirata and Tanaka, 1982), thus the
nucleus itself contains the inherent information to ensure
packaging into a spore: nuclear position within meiotic cells
is therefore largely irrelevant.
It is clear both from analysis of meiosis and the regrowth
of microtubules after reversible depolymerisation by cold in
the mitotic cell cycle (our unpublished observations and cited
1864 I. Hagan and M. Yanagida
by Verde et al., 1995) that an apparently normal interphase
cytoskeleton can be re-established by virtually random microtubule nucleation throughout the cytoplasm. The function of
the PAA may therefore be to increase the fidelity of re-establishing a cytoskeleton throughout the cell and so aiding rapid
nuclear positioning and hence survival in the mitotic cell
cycle. The microtubules of the PAA may scan the cytoplasm
to find SPBs in a way similar to the scanning of the nucleoplasm for kinetochores by the microtubules of the prometaphase centrosome (Kirschner and Mitchison, 1986).
The coincidence of the activation of microtubule nucleation
of both the PAA and the cytoplasmic face of the SPB directly
after mitosis maximally enhances the fidelity of nuclear positioning at the end of anaphase B. Following the breakdown of
the mitotic spindle, the SPB needs to interact with PAA microtubules to migrate to its correct location, but it is on the wrong
side of the nucleus. It is therefore faced with a problem similar
to that faced for microtubule capture by kinetochores (Nicklas
and Ward, 1994); the bulk of the nucleus will sterically block
PAA microtubules from reaching the SPB. The microtubules
polymerised from the SPB at the end of mitosis extend beyond
the chromatin and can thus interact with microtubules
extending from the PAA. If we assume that such contacts
enable microtubule/microtubule sliding, the SPB would be
dragged, or pushed round to face the PAA enabling further or
stronger interactions of PAA microtubules directly with the
SPB.
A requirement for rapid and accurate nuclear positioning
directly after mitosis is not so obvious under logarithmic
growth in liquid culture where the cell is sufficiently large
that the nuclei could wander a considerable distance from the
middle of the daughter cells without producing a “cut” or binucleate cell. In contrast, nutrient limitation, which is often
encountered in the centre of a colony and in the natural environment, leads to considerable reduction in cell size
(Nasmyth, 1979) and thus a significant increase in the likelihood of cleavage of either nucleus by the septum if nuclear
positioning immediately following mitosis is not ensured.
The possible nature of the nuclear positioning
signal
The nature of the signal responsible for directing SPB
movement remains obscure and awaits the identification of
mutants defective in nuclear positioning but it is likely to be
generated in response to morphogenic or ion gradients
(Harold, 1990; Nurse, 1994). The fact that overexpression of
a fission yeast protein phosphatase 1B homologue results in
aberrant SPB positioning suggests that protein phosphorylation cycles are involved (Yoshida et al., 1994). The continual
rocking motion of nuclei in time-lapse sequences of mitotic
cells suggests that the SPB may use such cycles to constantly
assess its position (Hagan et al., 1990; Hagan, 1988).
A clue as to the identity of the positioning signal may come
from studies of the cdc11+ gene because, in the cdc11.136
mutant, the nuclei do not always cluster centrally (Hagan and
Hyams, 1988). In contrast, nuclear clustering has been seen in
all other septum mutants studied to date: cdc7, cdc14, plo1 and
skp1 (Hagan and Hyams, 1988; Ohkura et al., 1995; Plyte et
al., 1996). This difference may be significant especially since
the MTOCs of the PAA in cdc11.136 mutants are similarly
unrestricted to the central zone.
Implications for nuclear positioning in fungi in
general
There is a striking similarity between the distribution of
clusters of nuclei in the cdc7 cells containing eight nuclei and
the regular distribution of clusters of groups of nuclei along the
hyphae of the nudF7 Aspergillus nidulans mutant grown at the
permissive temperature. This may suggest that similar underlying nuclear positioning mechanisms may underlie nuclear
positioning in unicellular and coencytic fungi (Woklow et al.,
1996). Indeed, it has long been suggested, based on correlative
phase-contrast and electron microscopic images, that the SPB
is intimately involved in nuclear migration. The resistance of
Nectaria haematococa interphase nuclei to optical tweezergenerated pulling forces has been interpreted as indicating
positional anchoring by the SPB (Berns et al., 1992). Thus
SPB-mediated positioning of nuclei is an attractive mechanism
not only for bulk nuclear migration but also for determining
the accurate nuclear positioning in fungi in general. Given the
sensitivity of the clustering assay in fission yeast and the
organism’s established genetics, the way is open for genetic
analysis of this complex phenomenon.
We are grateful to Prof. K. Gull for the generous gift of TAT1
antibody and for use of image processing equipment. We thank
Viesturs Simanis for strains. We would like to extend our thanks to
D. Drummond and A. Bridge for critical reading of the manuscript
and to V. Allan, C. Roghi and A. Carr for stimulating discussions.
This work was supported by grants from the Ministry of Education,
Science and Culture of Japan (a Specially Promoted Research) to M.
Y., the International Human Frontier Science Programme Organisation and Cancer Research Campaign [CRC] to I. H. and a Welcome
Trust equipment grant.
REFERENCES
Alfa, C. E. and Hyams, J. S. (1990). Distribution of tubulin and actin through
the cell division cycle of the fission yeast Schizosaccharomyces japonicus
var. versitilis: a comparison with Schizosaccharomyces pombe. J. Cell Sci.
96, 71-77.
Berns, M. W., Aist, J. R., Wright, W. H. and Liang, H. (1992). Optical
trapping in Animal and fungal cells using a tunable near-infrared TitaniumSapphire laser. Exp. Cell Res. 198, 375-378.
Bray, D. (1992). Cell Movements. N. York: Garland Press.
Bruno, K. S., Tinsley, J. H., Minke, P. F. and Plammann, M. (1996). Genetic
interactions among cytoplasmic dynein, dynactin and nuclear distribution
mutants of Neurospora crassa. Proc. Nat. Acad. Sci. USA 93, 4775-4780.
Carr, A. M. (1995). DNA structure checkpoints in fission yeast. Semin. Cell
Biol. 6, 65-72.
Chikashige, Y., Ding, D. Q., Funabiki, H., Haraguci, T., T., Mashiko, S.,
Yanagida, M. and Hiraoka, Y. (1994). Telomere led premeiotic
chromosome movement in fission yeast. Science 264, 270-273.
Clark, S. W. and Meyer, D. I. (1994). A putative centractin homologue in S.
cerevisiae is required for proper orientation of the mitotic spindle. J. Cell
Biol. 127, 129-138.
Doonan, J. H., Jenkins, G. I., Cove, D. J. and Lloyd, C. W. (1986).
Microtubules connect the migrating nucleus to the prospective division site
during side branch formation in the moss Physcomitrella patens. Eur. J. Cell
Biol. 41, 157-164.
Dowding, E. S. and Bakerespigel, A. (1954). The migrating nucleus. Can. J.
Microbiol. 1, 68-78.
Egel, R. (1971). Physiological aspects of conjugation in fission yeast. Planta
98, 89-96.
Eshel, D., Urrestarazu, L. A., Vissers, S., Jauniaux, J.-C., van VlietReedijk, J. C., Planta, R. J. and Gibbons, I. (1993). Cytoplasmic dynein is
required for normal nuclear segregation in yeast. Proc. Nat. Acad. Sci. USA
90, 11172-11176.
Nuclear positioning in S. pombe 1865
Euteneuer, U., Koonce, M. P., Pfister, K. K. and Schliwa, M. (1988). An
ATPase with properties expected for the organelle motor of the giant amoeba,
Reticulomyxa. Nature 332, 176-178.
Funabiki, H., Hagan, I., Uzawa, S. and Yanagida, M. (1993). Cell-cycledependent specific positioning and clustering of centromeres and telomeres
in fission yeast. J. Cell Biol. 121, 961-976.
Gönczy, P. and Hyman, A. A. (1996). Cortical domains and the mechanisms of
asymmetric cell division. Trends Cell Biol. 6, 382-387.
Gutz, H., Heslot, H., Leupold, U. and Loprieno, N. (1974).
Schizosaccharomyces pombe. In Handbook of Genetics (ed. R. C. King), pp.
395-445. Plenum Press, New York.
Hagan, I. M. (1988). A Study of the Behaviour of Microtubules and the Mitotic
Spindle in the Fission Yeast Schizosaccharomyces pombe. PhD. thesis,
University of London.
Hagan, I. M. and Hyams, J. S. (1988). The use of cell division cycle mutants
to investigate the control of microtubule distribution in the fission yeast
Schizosaccharomyces pombe. J. Cell Sci. 89, 343-357.
Hagan, I. M., Riddle, P. N. and Hyams, J. S. (1990). Intramitotic controls in
fission yeast: The effect of cell size upon the length and timing of mitotic
events. J. Cell Biol. 110, 1617-1621.
Hagan, I. and Yanagida, M. (1992). Kinesin related Cut7 protein associates
with mitotic and meiotic spindles in fission yeast. Nature 356, 74-76.
Hagan, I. and Yanagida, M. (1995). The product of the spindle formation gene
sad1+ associates with the fission yeast spindle pole body and is essential for
viability. J. Cell Biol. 129, 1033-1047.
Harold, F. M. (1990). To shape a cell: an inquiry into the causes of
morphogenesis of micro-organisms. Microbiol Rev. 54, 381-431.
Hereward, F. V. (1974). Cytoplasmic microtubules in a yeast. Planta 117, 355360.
Hiraoka, Y., Toda, T. and Yanagida, M. (1984). The NDA3 gene of fission
yeast encodes β−tubulin: A cold sensitive nda3 mutation reversibly blocks
spindle formation and chromosome movement in mitosis. Cell 39, 349-358.
Hirano, T., Funahashi, S., Uemura, T. and Yanagida, M. (1986). Isolation
and characterisation of Schizosaccharomyces pombe cut mutants that block
nuclear division but not cytokinesis. EMBO J. 5, 2973-2979.
Hirata, A. and Tanaka, K. (1982). Nuclear behaviour during conjugation and
meiosis in the fission yeast Schizosaccharomyces pombe. J. Gen. Appl.
Microbiol. 28, 2653-274.
Horio, T., Uzawa, S., Jung, M. K., Oakley, B. R., Tanaka, K. and Yanagida,
M. (1991). The fission yeast γ-tubulin is essential for mitosis and is localised
at microtubule organising centres. J. Cell Sci. 99, 693-700.
Jacobs, C. W., Adams, A. E. M., Szaniszlo, P. J. and J. R. (1988). Functions
of microtubules in the Saccharomyces cerevisiae cell cycle. J. Cell Biol. 107,
1409-1426.
Kirschner, M. and Mitchison, T. (1986). Beyond self assembly: from
microtubules to morphogenesis. Cell 45, 329-342.
Lange, B. H., Sherwin, T., Hagan, I. M. and Gull, K. (1995). The basics of
immunofluorescence video-microscopy for mammalian cells and microbial
systems. Trends Cell Biol. 5, 328-332.
Li, Y.-Y., Yeh, E., Hay, T. S. and Bloom, K. (1993). Disruption of mitotic
spindle orientation in a yeast dynein mutant. Proc Nat. Acad. Sci. USA 90,
10096-10100.
Meyer, S. L. F., Kamiskyj, S. G. W. and Heath, I. B. (1988). Nuclear
migration in a nud mutant of Aspergillus nidulans is inhibited in the presence
of a quantitatively normal population of cytoplasmic microtubules. J. Cell
Biol. 106, 773-778.
McCully, K. and Robinow, C. F. (1971). Mitosis in the fission yeast
Schizosaccharomyces pombe: a comparative study with light and electron
microscopy. J. Cell Sci. 9, 475-307.
McKerracher, L. J. and Heath, I. B. (1987). Cytoplasmic migration and
intracellular organelle movements during tip growth of fungal hyphae. Exp.
Mycol. 11, 79-100.
Mitchison, J. M. (1970). Physiological and cytological methods for
Schizosaccharomyces pombe. Meth. Physiol. 4, 131-165.
Moreno, S., Klar, A. and Nurse, P. (1990). An introduction to molecular
analysis of the fission yeast Schizosaccharomyces pombe. Meth. Enzymol.
194, 793-823.
Morris, N. R., Xiang, X. and Beckwith, S. (1995). Nuclear migration
advances in fungi. Trends Cell Biol. 5, 278-283.
Nasmyth, K. (1979). A control acting over the initiation of DNA replication in
the yeast Schizosaccharomyces pombe. J. Cell Sci. 36, 215-233.
Nicklas, R. B. and Ward, S. C. (1994). Elements of error correction in mitosis:
microtubule capture, release and tension. J. Cell Biol. 126, 1241-1253.
Nurse, P. (1994). Fission yeast morphogenesis - posing the problems. Mol.
Biol. Cell 5, 613-616.
Nurse, P., Thuriaux, P. and Nasmyth, K. (1976). Genetic control of the cell
division cycle of the fission yeast Schizosaccharomyces pombe. Mol. Gen.
Genet. 146, 167-178.
Oakley, B. R. and Morris, N. R. (1980). Nuclear movement is β-tubulin
dependent in Aspergillus nidulans. Cell 19, 255-262.
Oakley, B. R. and Reinhart, J. E. (1985). Mitochondria and nuclei move by
different mechanisms in Aspergillus nidulans. J. Cell Biol. 101, 2392-2397.
Ohkura, H., Hagan, I. M. and Glover, D. (1995). The conserved
Schizosaccharomyces pombe kinase plo1, required to form a bipolar spindle,
the actin ring, and septum, can drive septum formation in G1 and G2 cells.
Genes Dev. 9, 1059-1073.
Osmani, S. A., Engle, D. B., Doonan, J. H. and Morris, N. R. (1988). Spindle
formation and chromatin condensation in cells blocked at interphase by
mutation of a negative cell cycle control gene. Cell 52, 241-251.
Page, B. D., Satterwhite, L. L., Rose, M. and Snyder, M. (1994). Localisation
of Kar3 kinesin heavy chain-related protein requires the Cik1 interacting
protein. J. Cell Biol. 124, 507-519.
Palmer, R. E., Sullivan, D. S., Huffaker, T. and Koshland, D. (1992). Role of
astral microtubules and actin in spindle orientation and migration in the
budding yeast S. cerevisiae. J. Cell Biol. 119, 583-593.
Pichová, A., Kohlwein, S. D. and Yamamoto, M. (1995). New arrays of
microtubules in the fission yeast Schizosaccharomyces pombe. Protoplasma
188, 252-257.
Pickett-Heaps, J. D. and Wetherbee, R. (1987). Spindle function in the green
alga Mougetia: absence of anaphase A correlates with postmitotic nuclear
migration. Cell Motil. Cytoskel. 7, 68-77.
Plamann, M., Minke, P. F., Tinsley, J. H. and Bruno, K. S. (1994).
Cytoplasmic dynein and actin related protein Arp1 are required for normal
nuclear distribution in filamentous fungi. J. Cell Biol 127, 139-149.
Plyte, S. E., Feoktistova, A., Burke, J. D., Woodgett, J.R. and Gould, K. L.
(1996). Schizosaccharomyces pombe skp1+ encodes a protein kinase related
to mammalian glycogen synthase kinase 3 and complements a cdc14
cytokinesis mutant. Mol. Cell Biol. 16, 179-191.
Reeve, W. J. and Kelly, F. P. (1983). Nuclear position in the cells of the mouse
early embryo. J. Embryol. Exp. Morphol. 75, 117-139.
Reinsch, S. and Karsenti, E. (1994). Orientation of spindle axis and
distribution of plasma membrane proteins during cell division in polarised
MDCKII cells. J. Cell Biol. 126, 1509-1526.
Robb, M. J., Wilson, M. A. and Vierula, P. J. (1995). A fungal actin related
protein involved in nuclear migration. Mol. Gen. Genet. 247, 583-590.
Snyder, M., Gehrung, S. and Page, B. D. (1991). Studies concerning the
temporal and genetic control of cell polarity in Saccharomyces cerevisiae. J.
Cell Biol. 114, 515-532.
Streiblova, E. and Girbardt, M. (1980). Microfilaments and cytoplasmic
microtubules in cell division cycle mutants of Schizosaccharomyces pombe.
Can. J. Microbiol. 26, 250-261.
Sullivan, W. and Theurkauf, W. E. (1995). The cytoskeleton and
morphogenesis of the early Drosophila embryo. Curr. Opin. Cell Biol. 7, 1822.
Tanaka, K. and Kanbe, T. (1986). Mitosis in the fission yeast
Schizosaccharomyces pombe as revealed by freeze-substitution electron
microscopy. J. Cell Sci. 80, 253-268.
Thompson-Coffe, C. and Zickler, D. (1993). Cytoskeleton interactions in the
ascus development and sporulation of Sordaria Macrospora. J. Cell Sci. 104,
883-898.
Tinsley, J. H., Minke, P. F., Bruno, K. S. and Plamann, M. (1996). p150glued,
the largest subunit of the dynactin complex is nonessential in Neurospora but
required for nuclear distribution. Mol. Biol. Cell 7, 731-742.
Toda, T., Yamamoto, M. and Yanagida, M. (1981). Sequential alterations in
the nuclear chromatin region during mitosis of the fission yeast
Schizosaccharomyces pombe: video fluorescence microscopy of
synchronously growing wild-type and cold sensitive cdc mutants by using a
DNA-binding fluorescent probe. J. Cell Sci. 52, 271-287.
Umesono, K., Toda, T., Hayashi, S. and Yanagida, M. (1983). Two cell
division cycle genes NDA2 and NDA3 of the fission yeast
Schizosaccharomyces pombe control microtubular organisation and
sensitivity to anti-mitotic benzimidazole compounds. J. Mol. Biol. 168, 271284.
Uzawa, S., Samejima, I., Hirano, T., Tanaka, K. and Yanagida, M. (1990).
The fission yeast cut1+ gene regulates spindle pole body duplication and has
homology to the budding yeast ESP1 gene. Cell 62, 913-925.
1866 I. Hagan and M. Yanagida
Vale, R. D., Soll, D. R. and Gibbons, I. R. (1989). One dimensional diffusion
of microtubules bound to flagellar dynein. Cell 59, 915-925.
Verde, F., Mata, J. and Nurse, P. (1995). Fission yeast cell morphogenesis:
identification of new genes and analysis of their role during the cell cycle. J.
Cell Biol. 131, 1529-1538.
Walker, G. M. (1982). Cell cycle specificity of certain anti-microtubular drugs
in Schizosaccharomyces pombe. J. Gen. Microbiol. 128, 61-71.
Wilson, C. L. and Aist, J. R. (1967). Motility of fungi. Phytopathology 57,
769-771.
Wolkow, T. D., Harris, S. D. and Hammer, J. E. (1996). Cytokinesis in
Aspergillus nidulans is controlled by cell size nuclear positioning and
mitosis. J. Cell Sci. 109, 2179-2188.
Woods, A., Sherwin, T., Sasse, R., MacRae, T. H., Baines, A. J. and Gull, K.
(1989). Definition of individual components within the cytoskeleton of
Trypanosoma brucei by a library of monoclonal antibodies. J. Cell Sci. 93,
491-500.
Wordeman, L., McDonald, K. L. and Cande, W. Z. (1986). The distribution
of cytoplasmic microtubules throughout the cell cycle of the centric diatom
Stephanopyxis turris: their role in nuclear migration and positioning the
mitotic spindle during cytokinesis. J. Cell Biol. 102, 1688-1698.
Xiang, X., Beckwith, S. and Morris, N. R. (1994). Cytoplasmic dynein is
involved in nuclear migration in Aspergillus nidulans. Proc. Nat. Acad. Sci.
USA 91, 2100-2104.
Xiang, X., Roghi, C. and Morris, N. R. (1995a). Characterisation and
localisation of the cytoplasmic dynein heavy chain in Aspergillus nidulans.
Proc. Nat. Acad. Sci. USA 92, 9890-9894.
Xiang, X., Osmani, S. A., Osmani, M., Xin, M. and Morris, N. R. (1995b).
NudF, a nuclear migration gene in Aspergillus nidulans is similar to the
human LIS-1 gene required for neuronal migration. Mol. Biol. Cell 6, 297310.
Yaffe, M. P., Hirata, D., Verde, F., Eddison, M., Toda, T. and Nurse, P.
(1996). Microtubules mediate mitochondrial distribution in fission yeast.
Proc. Nat. Acad. Sci. USA 93, 11664-11668.
Yeh, E., Skibbens, R. V., Cheng, J. W. Salmon, E. D. and Bloom, K. (1995).
Spindle dynamics and cell cycle regulation of dynein in the budding yeast
Saccharomyces cerevisiae. J. Cell Biol. 130, 687-700.
Yoshida, T., Toda, T. and Yanagida, M. (1994). A calcineurin-like gene ppb1+
in fission yeast: mutant defects in cytokinesis cell polarity, mating and
spindle pole body positioning. J. Cell Sci. 107, 1725-1735.
(Received 28 February 1997 - Accepted 6 June 1997)