Current Biology 21, 1314–1319, August 9, 2011 ª2011 Elsevier Ltd All rights reserved
DOI 10.1016/j.cub.2011.06.052
Report
Divergent Strategies for Controlling
the Nuclear Membrane Satisfy Geometric
Constraints during Nuclear Division
Candice Yam,1,2 Yue He,1,2,5 Dan Zhang,1,2,5
Keng-Hwee Chiam,3,4 and Snezhana Oliferenko1,2,*
1Temasek Life Sciences Laboratory, National University of
Singapore, 1 Research Link, Singapore 117604
2Department of Biological Sciences, National University of
Singapore, Singapore 117543
3Institute of High Performance Computing, Agency for
Science, Technology and Research, Singapore 138632
4Mechanobiology Institute, National University of Singapore,
Singapore 117411
Summary
Eukaryotes segregate chromosomes in ‘‘open’’ or ‘‘closed’’
mitosis, depending on whether their nuclear envelopes
(NEs) break down or remain intact. Here we show that the
control of the nuclear surface area may determine the choice
between these two modes. The dividing nucleus does not
expand its surface in the fission yeast Schizosaccharomyces
japonicus, confining the mitotic spindle and causing it
to buckle. The NE ruptures in anaphase, releasing the
compressive stress and allowing chromosome segregation. Blocking the NE expansion in the related species
Schizosaccharomyces pombe that undergoes closed
mitosis induces spindle buckling and collapse in the
absence of an intrinsic NE rupture mechanism. We propose
that scaling considerations could have shaped the evolution
of eukaryotic mitosis by necessitating either nuclear surface
expansion or the NE breakdown.
Results and Discussion
The fission yeast Schizosaccharomyces pombe is a wellstudied unicellular model eukaryote with strictly closed
mitosis. In this case, the nuclear envelope (NE) markers such
as the nucleoporin Nup85-GFP [1, 2] remain at the nuclear
periphery as the nucleus undergoes a progression of shape
changes, from a sphere to a spherocylinder to a dumbbell,
that eventually resolves into two spherical daughter nuclei
(Figure 1A). However, imaging the NE in the related fission
yeast species S. japonicus using Nup85-GFP revealed an
entirely different evolution of mitotic nuclear shapes (Figure 1B). As mitosis progressed, the nuclei assumed an
elongated diamond-like morphology followed by an abrupt
resolution into daughter nuclei without the dumbbell stage.
This unusual nuclear shape was originally observed in 1984
[3]. The initially uniformly distributed Nup85-GFP relocalized
toward the pointed leading edges of the nucleus and remained
there as the half-nuclei abruptly relaxed into a more spherical
morphology. It eventually redistributed around the entire
periphery of daughter nuclei (Figure 1B). Other proteins associated with the nuclear pore complexes (NPCs), including the
nucleoporins Nup132 and Nup189 [1, 2] and the tubular
5These authors contributed equally to this work
*Correspondence: [email protected]
endoplasmic reticulum (ER) protein Tts1 [4, 5], exhibited
similar behavior (see Figure S1A available online). Because
the NPCs mark a subdomain of the NE, we examined the distribution of the inner nuclear membrane (INM) protein Lem2-GFP
[6] and the cisternal ER markers Ost1-GFP [4] and Sec63-GFP
at various stages of the cell cycle. Unlike the NPC-associated
proteins, Lem2-GFP localized along the nuclear periphery
and to the spindle pole bodies (SPBs). At the end of mitosis,
Lem2-GFP formed foci at the intersection between the spindle
and the NE, opposite the SPBs (Figure 1C). The mitotic
S. japonicus nucleus was initially spherical but later appeared
to assume a diamond shape. At later stages of mitosis, the
nucleus exhibited a more relaxed rounded aspect and two
constrictions that delimited the daughter nuclei and what
appeared to be a transient medial compartment. At this stage,
we observed discontinuities in Lem2-GFP localization.
We confirmed the unusual pattern of the NE division using
Ost1-GFP and Sec63-GFP as markers (Figures S1B–S1D).
Similar to the filamentous ascomycete Aspergillus nidulans
[7], the short-lived medial compartment contained the
transient remnant mother nucleolus structure marked by
Erb1-mCherry and Fib1-mCherry (Figures S1E and S1F).
Early in mitosis, the GFP-tagged chromatin-binding factor
Nhp6 [8] localized to the nucleoplasm and was enriched on
condensed chromosomes (Figure 2A). To our surprise, the
abrupt relaxation of the diamond-shaped NE coincided with
a sudden redistribution of the nuclear Nhp6-GFP throughout
the cellular volume. Within minutes, Nhp6-GFP reaccumulated
in the newly formed daughter nuclei (Figure 2A; see Figure 2B
for quantification). Similarly, an artificial nuclear marker
GFP-GST-NLS-GFP exhibited an abrupt loss of nuclear
compartmentalization at the same stage of mitosis (Figure 2C).
Redistribution of the NPCs toward the leading edges of the
nucleus substantially preceded the efflux of the nuclear
markers (by w2.5 min) and therefore did not appear directly
related to this process. The cytosolic marker GST-NES-GFP
behaved in a manner opposite to that of nuclear proteins, suggesting an overall transient loss of the nucleocytoplasmic
integrity during mitosis (Figure S2A). An extremely rapid
kinetics of intermixing between the nuclear and cytosolic
markers together with the observed discontinuities of the NE
(Figure 1C; Figures S1C and S1D) suggested that the NE physically broke. To better visualize this process, we constructed
S. japonicus cells carrying an artificial ER marker GFP-AHDL
[4]. Time-lapse imaging of GFP-AHDL Nhp6-mCherry cells
revealed that the diamond-shaped nucleus often bent further
to assume a bow-like configuration and abruptly broke in a
single expanding tear close to the nuclear equator. The tearing
coincided with the relaxation of the nuclear envelope and
release of the nuclear protein Nhp6 into the cytoplasm (Figure 2D; Figure S2B; Movie S1; Movie S2; n = 20 cells). This
phenomenon occurred in late anaphase B as judged by timelapse imaging of cells coexpressing Nhp6-mCherry and the
kinetochore marker Mis6-GFP (Figure S2C). We concluded
that the nuclear membrane breaks in late anaphase in
S. japonicus cells.
We constructed the GFP-tagged a-tubulin, GFP-Atb2, and
visualized spindle dynamics in conjunction with those of
Divergent Mitotic Strategies in Fission Yeast
1315
Figure 1. S. japonicus and S. pombe Exhibit Markedly Dissimilar Progression of Nuclear Shapes during Mitosis
(A and B) Time-lapse maximum-projection images of mitotic S. pombe (A)
and S. japonicus (B) cells expressing endogenous Nup85-GFP proteins as
nuclear pore complex markers. Note an unusual extended nuclear aspect
and visible ‘‘sliding’’ of nuclear pores toward the opposite ends of the
dividing nucleus in S. japonicus. Time is in minutes and seconds.
(C) A collection of scanning confocal maximum-projection images of cells
expressing Lem2-GFP (top) and Nup85-mCherry (middle) in various stages
of mitosis. Overlay of two markers shown in the bottom panel. Scale bars
represent 5 mm.
Nhp6-mCherry. As spindles elongated, they dramatically
buckled within the confines of the nucleus (Figure 3A). We
observed spindle bending in both GFP-Atb2-expressing and
untagged wild-type strains immunostained with anti-a-tubulin
antibodies (data not shown). Strikingly, when the nuclei
ruptured, releasing Nhp6-mCherry into the cytoplasm, the
spindles sprang back to a straight configuration (Figure 3A).
Spindle straightening coincided with the transition of nuclear
morphology from a highly stretched aspect to a relaxed one
(Figure S3).
Spindle buckling could result from the compressive stress
exerted on the elongating spindle by the NE. We sought
to weaken the NE integrity to test whether a more compliant
NE will exert a smaller compressive stress and hence result
in lesser or no spindle buckling. The LEM domain protein
Lem2 was reported as crucial for maintaining the NE integrity
in mammalian cells [9], and we wondered whether such function was conserved in fungi, which lack the lamin-centered
nuclear lamina. We visualized the mitotic NE behavior in cells
lacking Lem2 using Nup85-GFP as a marker. The NE in
lem2D cells was unusually ruffled and failed to assume the
typical diamond shape (Figure 3B). The mitotic spindles in
lem2D cells elongated in a straight line, suggesting that they
were no longer constrained by the NE (Figure 3C, upper panel,
26 out of 26 cells). Interestingly, the nucleoplasmic markers
such as Nhp6-mCherry (Figure 3C, bottom panel) and GFPGST-NLS-GFP (data not shown) seeped out from the nucleus
immediately following the onset of anaphase B, once the
spindle length exceeded the nuclear diameter. This was clearly
different from the wild-type, where the nucleoplasm was contained inside a nucleus until late in mitosis, when it virtually
instantaneously redistributed throughout the cell (Figure 3A).
The postmitotic nuclear recruitment of Nhp6 was delayed in
lem2D cells, suggesting that Lem2 might also function in
resealing of the NE after mitosis. We concluded that the
INM protein Lem2 was important for maintaining the NE
integrity during mitosis and that the NE in S. japonicus could
exert sufficient compressive stress to cause the spindle to
buckle.
We wondered whether the spindle was required to tear apart
the NE during normal mitosis. We constructed cells lacking the
spindle assembly checkpoint protein Mad2 to force them
through mitosis in the absence of the microtubule-kinetochore
attachments. We treated mad2D cells with thiabendazole
(TBZ) to depolymerize microtubules and tested whether the
nucleoplasmic markers such as Nhp6-mCherry redistributed
throughout the cell. All cells lost the nucleocytoplasmic
compartmentalization in spite of the absence of the spindle
(Figure 3D; 21 cells). We observed similar behavior in mitotic
wild-type cells treated with TBZ (data not shown). The fluorescence later recovered in several micronuclei, each formed
around a cluster of the Nhp6-mCherry-labeled chromatin
(albeit with slower kinetics as compared to the wild-type:
27.5 6 7.4 min versus 3 min for TBZ-treated and control cells,
respectively). We concluded that the S. japonicus cells
evolved a specialized mechanism triggering the NE breakdown in anaphase, independent of the forces exerted by the
elongating mitotic spindle.
If the nuclear volume remains constant during mitosis, a
simple scaling argument shows that when r1 is the radius of
the mother nucleus and r2 is the radius of each of the two
daughter nuclei, 4/3pr13 = 2 3 4/3pr23 and (r2/r1)3 = 1/2. The
ratio of the total surface area of the two daughter nuclei
to that of the mother nucleus is then 2 3 4pr22/4pr12 = 2 3
(r2/r1)2 = 21/3 = 1.26. Indeed, similar ratios were previously reported for S. pombe [10–12]. However, when the extra surface
area is not added, an alternative is to reduce the nuclear
volume to form two smaller daughter nuclei. In this case,
constancy of surface area implies that 4pr12 = 2 3 4pr22,
and (r2/r1)2 = 1/2. The ratio of the total volume of the two
daughter nuclei to that of the mother nucleus is then 2 3
4/3pr23/4/3pr13 = 2 3 (r2/r1)3 = 221/2 = 0.71. We carefully
estimated the expansion of the S. pombe and S. japonicus
nuclei during mitosis. The measurements were performed
from time 0 when the mitotic nuclei were still spherical to
a diamond/bow-like stage immediately preceding the NE
rupture in S. japonicus and a corresponding dumbbell stage
in S. pombe (n = 10 cells, at time 0 the surface area and volume
are set to 100%; Figure 4A; Figure S4A). Linear least-squares
fitting indicated that the nuclear surface area is S. pombe
indeed grew at an average rate of 3.6%/min (p = 0.0345).
The nuclear volume remained relatively constant (Figure S4A;
volume growth was statistically insignificant, p = 0.236). In
line with previously published data, upon completion of
Current Biology Vol 21 No 15
1316
Figure 2. The Nuclear Envelope Breaks and
Reseals during Mitosis in S. japonicus
(A) Time-lapse maximum-projection images
of cells expressing the nucleoporin Nup189mCherry (top) and the high-mobility group
protein Nhp6-GFP (bottom) show an abrupt
redistribution of Nhp6-GFP fluorescence from
the nucleus throughout the cell during mitosis,
followed by its eventual reimport to the nucleus.
(B) Quantification of nuclear and cytoplasmic
fluorescence intensities of Nhp6-GFP at three
indicated stages of mitosis (n = 5; error bars
indicate standard deviation).
(C) Time-lapse maximum-projection images of
S. japonicus cells expressing GFP-GST-NLSGFP (top) and Nup85-mCherry (bottom).
(D) Time-lapse single plane images of mitotic
cells coexpressing the endoplasmic reticulum
(ER) marker GFP-AHDL (top) and the nuclear
protein Nhp6-mCherry (bottom). Note the abrupt
rupture of the nuclear envelope (NE) (at time point
30 1200 ) that coincides with leakage of Nhp6mCherry from the nucleus. Time is in minutes
and seconds. Scale bars represent 5 mm.
nuclear division, the combined surface area and volume of
daughter nuclei were 133% 6 11% and 105% 6 9%, respectively, relative to the mother nucleus. Thus, the S. pombe
nucleus indeed adds the extra surface area as it prepares to
divide, suggesting the existence of a connected membrane
reservoir.
In contrast, we found that both the surface area and the
volume of S. japonicus nuclei did not increase prior to the
nuclear rupture, suggesting that there was no addition of
new membranes during mitosis (Figure 4B; Figure S4A; n =
10 cells). Upon formation of the daughter nuclei, their
combined surface area and volume were 99% 6 7% and
69% 6 7% respectively, relative to the mother nucleus. Therefore, we concluded that in order to divide, the S. japonicus
nucleus undergoes major restructuring; i.e., breaking and reassembling the NE.
The NE in fungi is reinforced by the SPBs that appear
to counter the spindle-induced membrane deformations
[10, 11]; therefore, when the elongating spindle pushes against
the NE, it will experience compressive stress along its longitudinal direction. In S. pombe, the NE grows concomitantly
with spindle elongation so the compressive stress is negligible
and the spindle remains straight. However, in S. japonicus,
where the nuclear surface area does
not increase, the compressive stress
would eventually exceed the spindle’s
buckling threshold, and the spindle will
buckle. We hypothesized that spindle
buckling should occur in all nuclei
that do not add new membranes. To
test this prediction, we restricted the
membrane availability in S. pombe by
pretreating cells with the fatty acid synthase inhibitor cerulenin (Figure S4B;
[13, 14]) for 30 min prior to recording
the time-lapse sequences of mitosis.
We reasoned that it could prevent new
membrane biosynthesis at the mitotic
entry without major disruption to cell
growth. We used GFP-Atb2 to visualize spindle structure and
GST-NLS-mCherry and Tts1-mCherry to mark the nucleoplasm and the ER. In control cells, the mitotic spindles remained straight throughout nuclear division (Figure 4C,
upper panel). In contrast, many anaphase spindles in cerulenin-treated cells severely buckled (15 out of 23 cells). Buckling
frequently led to spindle breakage into two half-spindles (8 out
of 15 spindles). In such cases, the nuclei extended into a diamond shape but later collapsed back into a single sphere of
the original diameter (Figure 4C, bottom panel). The rest of
the nuclei divided into unequally sized daughters exhibiting
a limited increase in their combined surface area, similar to
what was reported previously (data not shown and [14]). The
difference between the two outcomes could be due to the
incomplete exhaustion of the membrane reservoir in some
cerulenin-treated cells. Similar treatment of S. japonicus cells
with cerulenin did not interfere with mitotic progression
(Figures S4B and S4C). Thus, it appears that the initial
sequence of events during anaphase spindle elongation
is similar in cerulenin-treated S. pombe and wild-type
S. japonicus cells, with spindles buckling within the NE.
However, because the mechanism that regulates the mitotic
NE rupture is absent from the S. pombe lineage, the spindles
Divergent Mitotic Strategies in Fission Yeast
1317
in cerulenin-treated S. pombe eventually collapse under the
compressive stress.
In summary, we show that two closely related species have
settled on strikingly divergent approaches to mitotic nuclear
division (Figure 4D). The S. pombe cells appear to assemble
a membrane reservoir sufficient for the nuclear surface area
increase during anaphase that allows a ‘‘closed’’ division of
the nucleus. In contrast, the nuclear surface area remains
constant during mitosis in S. japonicus. The mechanisms
responsible for this variation could include differential regulation of ER membrane biogenesis at the G2/M boundary or
structural dissimilarities in the organization of the ER and
the NE. In the absence of an available membrane reservoir,
S. japonicus uses an active mechanism to break down the
NE in order to avoid mitotic spindle collapse. Unlike in many
metazoan cells where the NE breakdown occurs at the end
of prophase [15], S. japonicus ruptures its nuclear membrane
in anaphase B, suggesting that these mechanisms have
been acquired independently. However, the two systems
might use similar means to compromise the NE integrity,
including the cell-cycle-entrained regulation of the nuclear
membrane proteins. In general, the so-called ‘‘open’’ and
‘‘closed’’ mitoses can be considered as extremes of phenotypic variation in mitotic mechanisms [16–18]. The fact that
the closely related species exhibit pronounced differences
in their mitotic programs suggests that the emergence of
fundamentally different phenotypic traits may sometimes
require a relatively modest modification of the basic cell physiology. The comparative studies of the mechanisms regulating
nuclear division in the two fission yeast species may provide
important insights into both physiology and evolution of
mitosis as a major eukaryotic condition.
Experimental Procedures
Figure 3. Elongating Mitotic Spindles in S. japonicus Are Confined and
Grossly Deformed by the Nuclear Envelope Prior to the Loss of Nuclear
Integrity during Anaphase
(A) The elongating spindle buckles inside the bow-shaped nucleus as
shown by these time-lapse maximum-projection images of GFP-Atb2 and
Nhp6-mCherry. Note an abrupt spindle relaxation as Nhp6-mCherry redistributes throughout the cell.
(B) Time-lapse maximum-projection images of Nup85-GFP in lem2D cells
show extensive ruffling of the NE during mitosis.
(C) Mitotic spindles remain straight during mitosis in cells lacking Lem2, as
shown by this time-lapse maximum-projection sequence of lem2D cells expressing GFP-Atb2 and Nhp6-mCherry. Note that the nucleoplasmic marker
Nhp6-mCherry leaks out of the nucleus immediately following anaphase
spindle elongation.
(D) The time-lapse maximum-projection sequence of mad2D cells progressing through mitosis without the mitotic spindle (in the presence of
a microtubule poison thiabendazole) shows that the nucleoplasmic protein
Nhp6-mCherry is released and reimported into the daughter nuclei. Time is
in minutes and seconds. Scale bars represent 5 mm.
Schizosaccharomyces japonicus Strains and Drug Treatments
The original wild-type auxotrophic S. japonicus strains were kindly provided by H. Niki [19]. All strains constructed in the course of our study
are listed in Table S1. The S. japonicus open reading frames used in this
study were as follows: Nup85 (SJAG_00471), Nup132 (SJAG_01374),
Nup189 (SJAG_03299), Tts1 (SJAG_05671), Lem2 (SJAG_01745),
Ost1 (SJAG_01006), Sec63 (SJAG_05394), Nhp6 (SJAG_03603.4),
Mis6 (SJAG_00684), Atb2 (SJAG_02509), Erb1 (SJAG_02830), Fib1
(SJAG_00943), and Mad2 (SJAG_05642). To depolymerize microtubules,
we added 150 mM TBZ (thiabendazole, Sigma T-8904) dissolved in dimethyl
sulfoxide (DMSO) to exponentially growing cells for 30 min at 24! C. To
inhibit fatty acid biosynthesis prior to mitosis, we added 10 mM cerulenin
dissolved in DMSO (Sigma C-2389) to exponentially growing cells for
30 min at 24! C followed by immediate time-lapse imaging. For details of
S. japonicus strain construction, cell growth, imaging, and image analysis
methods, see Supplemental Experimental Procedures.
Supplemental Information
Supplemental Information includes four figures, one table, Supplemental
Experimental Procedures, and two movies and can be found with this article
online at doi:10.1016/j.cub.2011.06.052.
Acknowledgments
We are very grateful to H. Niki for S. japonicus strains and advice. We would
like to thank A. Vjestica for discussions and help with the artwork; M. Sato
for sharing reagents; M. Kopecká for rare early reprints; and S. Cohen,
M. Balasubramanian, A. Vjestica, J. Yew, and E. Makeyev for comments
on the manuscript. This work has been supported by Singapore Millennium
Foundation.
Current Biology Vol 21 No 15
1318
Figure 4. Increase in the Nuclear Surface Area Allows ‘‘Closed’’ Mitosis
(A and B) Plots representing experimentally determined values for the nuclear surface area during nuclear division in S. pombe (A) and S. japonicus (B). At
time 0, the surface area is set to 100% and the changes are shown in relationship to this benchmark (n = 10 cells). Measurements chart nuclear shape
changes in both organisms expressing the ER marker GFP-AHDL (top panels).
Divergent Mitotic Strategies in Fission Yeast
1319
Received: March 24, 2011
Revised: May 25, 2011
Accepted: June 22, 2011
Published online: July 28, 2011
References
1. Baı̈, S.W., Rouquette, J., Umeda, M., Faigle, W., Loew, D., Sazer, S., and
Doye, V. (2004). The fission yeast Nup107-120 complex functionally
interacts with the small GTPase Ran/Spi1 and is required for mRNA
export, nuclear pore distribution, and proper cell division. Mol. Cell.
Biol. 24, 6379–6392.
2. Chen, X.Q., Du, X., Liu, J., Balasubramanian, M.K., and Balasundaram,
D. (2004). Identification of genes encoding putative nucleoporins and
transport factors in the fission yeast Schizosaccharomyces pombe: A
deletion analysis. Yeast 21, 495–509.
3. Gabriel, M. (1984). Karyokinesis and septum formation during
the regeneration of incomplete cells walls in protoplasts of
Schizosaccharomyces japonicus var. versatilis: A time-lapse microcinematographic study. Microbiol. 130, 625–630.
4. Zhang, D., Vjestica, A., and Oliferenko, S. (2010). The cortical ER
network limits the permissive zone for actomyosin ring assembly.
Curr. Biol. 20, 1029–1034.
5. Chadrin, A., Hess, B., San Roman, M., Gatti, X., Lombard, B., Loew, D.,
Barral, Y., Palancade, B., and Doye, V. (2010). Pom33, a novel
transmembrane nucleoporin required for proper nuclear pore complex
distribution. J. Cell Biol. 189, 795–811.
6. Brachner, A., Reipert, S., Foisner, R., and Gotzmann, J. (2005). LEM2 is
a novel MAN1-related inner nuclear membrane protein associated with
A-type lamins. J. Cell Sci. 118, 5797–5810.
7. Ukil, L., De Souza, C.P., Liu, H.L., and Osmani, S.A. (2009). Nucleolar
separation from chromosomes during Aspergillus nidulans mitosis
can occur without spindle forces. Mol. Biol. Cell 20, 2132–2145.
8. Paull, T.T., and Johnson, R.C. (1995). DNA looping by Saccharomyces
cerevisiae high mobility group proteins NHP6A/B. Consequences for
nucleoprotein complex assembly and chromatin condensation.
J. Biol. Chem. 270, 8744–8754.
9. Ulbert, S., Antonin, W., Platani, M., and Mattaj, I.W. (2006). The inner
nuclear membrane protein Lem2 is critical for normal nuclear envelope
morphology. FEBS Lett. 580, 6435–6441.
10. Zheng, L., Schwartz, C., Magidson, V., Khodjakov, A., and Oliferenko, S.
(2007). The spindle pole bodies facilitate nuclear envelope division
during closed mitosis in fission yeast. PLoS Biol. 5, e170.
11. Lim, H.W.G., Huber, G., Torii, Y., Hirata, A., Miller, J., and Sazer, S.
(2007). Vesicle-like biomechanics governs important aspects of nuclear
geometry in fission yeast. PLoS ONE 2, e948.
12. Neumann, F.R., and Nurse, P. (2007). Nuclear size control in fission
yeast. J. Cell Biol. 179, 593–600.
13. Awaya, J., Ohno, T., Ohno, H., and Omura, S. (1975). Substitution of
cellular fatty acids in yeast cells by the antibiotic cerulenin and exogenous fatty acids. Biochim. Biophys. Acta 409, 267–273.
14. Saitoh, S., Takahashi, K., Nabeshima, K., Yamashita, Y., Nakaseko, Y.,
Hirata, A., and Yanagida, M. (1996). Aberrant mitosis in fission yeast
mutants defective in fatty acid synthetase and acetyl CoA carboxylase.
J. Cell Biol. 134, 949–961.
15. Guttinger, S., Laurell, E., and Kutay, U. (2009). Orchestrating nuclear
envelope disassembly and reassembly during mitosis. Nat. Rev. Mol.
Cell Biol. 10, 178–191.
16. Pickett-Heaps, J. (1974). The evolution of mitosis and the eukaryotic
condition. Biosystems 6, 37–48.
17. Heath, I.B. (1980). Variant mitoses in lower eukaryotes: Indicators of the
evolution of mitosis. Int. Rev. Cytol. 64, 1–80.
18. De Souza, C.P., and Osmani, S.A. (2007). Mitosis, not just open or
closed. Eukaryot. Cell 6, 1521–1527.
19. Furuya, K., and Niki, H. (2009). Isolation of heterothallic haploid and
auxotrophic mutants of Schizosaccharomyces japonicus. Yeast 26,
221–233.
Note Added in Proof
While this manuscript was in press, a manuscript by Niki and colleagues on
S. japonicus mitosis was published online: Aoki, K., Hayashi, H., Furuya, K.,
Sato, M., Takagi, T., Osumi, M., Kimura, A., Niki, H. (2011). Breakage of the
nuclear envelope by an extending mitotic nucleus occurs during anaphase
in Schizosaccharomyces japonicus. Genes Cells, in press. Published online
July 6, 2011. 10.1111/j.1365-2443.2011.01540.x.
(C) Limiting the membrane reservoir prior to mitosis by inhibition of fatty acid biosynthesis in S. pombe triggers pronounced spindle buckling, often leading
to spindle breakage and abortive nuclear division. Shown are time-lapse sequences of maximum-projection images of control (top) and cerulenin-treated
(bottom) S. pombe cells expressing NLS-GST-mCherry, Tts1-mCherry, and GFP-Atb2. Time is in minutes. Scale bars represent 5 mm.
(D) Pictorial models of nuclear behavior during mitosis in S. pombe and S. japonicus. Note that the two organisms are not shown to scale.
Current Biology, Volume 21
Supplemental Information
Divergent Strategies for Controlling
the Nuclear Membrane Satisfy Geometric
Constraints during Nuclear Division
Candice Yam, Yue He, Dan Zhang, Keng-Hwee Chiam, and Snezhana Oliferenko
Supplemental Inventory
1. Supplemental Figures and Table
Figure S1, related to Figure 1
Figure S2, related to Figure 2
Figure S3, related to Figure 3
Figure S4, related to Figure 4
Table S1.
2. Supplemental Experimental Procedures
3. Supplemental References
Figure S1, Related to Figure 1.
(A) Time-lapse maximum projection images of mitotic S. japonicus cells expressing
respectively: nucleoporins Nup132-GFP (top), Nup189-GFP (middle) and the transmembrane ER
protein specifically associated with highly curved membranes and the nuclear pores, Tts1-GFP
(bottom).
(B) Overall, the NE in S. japonicus appears to resolve into two daughter nuclei and a short-lived
medial compartment, as shown by the resident ER protein Ost1-GFP. Shown are maximum
projections images of scanning confocal micrographs of cells expressing Ost1-GFP (top) and
Nup85-mCherry (middle) in various stages of mitosis. Overlay of two markers shown in the
bottom panel. Note that these are not the time-lapse images.
(C) Time-lapse maximum projection images (of three planes with a z-distance of 0.5 Pm) of a
mitotic S. japonicus cell expressing the ER protein Ost1-GFP. The image indicated by the
asterisk shows the nuclear membrane rupture.
(D) Time-lapse single plane images of a mitotic S. japonicus cell expressing the ER protein
Sec63-GFP. The image indicated by the asterisk shows the nuclear membrane rupture.
(E) The remnant of the mother nucleolus is discarded in a short-lived medial compartment
formed during mitosis. Shown are maximum z-projections of scanning confocal micrographs of
cells expressing the ER marker Sec63-GFP (top) and the nucleolus marker Erb1-mCherry
(middle) at various stages of mitosis. Overlay of two images is shown in the bottom panel. Note
that these are not the time-lapse images.
(F) Time-lapse maximum projection images of mitotic S. japonicus cells co-expressing the
nucleolar marker fibrillarin Fib1-mCherry (top) and the nucleoporin Nup85-GFP (bottom). When
appropriate, time is indicated in minutes and seconds. Scale bars represent 5 Pm.
Figure S2, Related to Figure 2.
(A) Time-lapse maximum z-projection images of GST-NES-GFP dynamics in S. japonicus cells
undergoing mitosis. Note that this cytosolic marker redistributes throughout the cell following
nuclear elongation. It is yet again compartmentalized properly following formation of two
daughter nuclei.
(B) Time-lapse single plane images of mitotic cells co-expressing the ER marker GFP-AHDL
and the nuclear protein Nhp6-mCherry. Note the abrupt rupture of the NE (at time point 2’ 48’’)
that coincides with leakage of Nhp6-mCherry from the nucleus.
(C) Time-lapse maximum projection images of cells expressing the kinetochore marker Mis6GFP (top) and Nhp6-mCherry (middle) show that loss of nuclear integrity occurs in late
anaphase B. Early mitotic cells exhibited three bright Mis6-GFP dots representing three
metaphase kinetochore pairs that split into six kinetochores upon the anaphase onset. The
daughter kinetochores rapidly segregated towards spindle poles while the nucleus was still
spherical, marking the end of anaphase A. In anaphase B, the nuclei elongated often exhibiting
bending deformations until cells experienced a sudden loss of the nuclear integrity. An overlay
of two deconvolved time-lapse sequences is shown in the bottom panel. For all panels, time is in
minutes and seconds; scale bar represents 5 Pm.
Figure S3, Related to Figure 3.
Time-lapse maximum z-projection images of S. japonicus cells expressing both GFP-Atb2 and
Nup189-mCherry show a drastic change in nuclear shape upon abrupt relaxation of highly
buckled anaphase spindles. Time is in minutes and seconds. Scale bar represents 5 Pm.
Figure S4, Related to Figure 4.
(A) Measurements of the nuclear volume in S. pombe and S. japonicus cells undergoing mitosis
(n=10 cells each). The measurements start at the stage when the nuclei are still spherical (time 0)
and are performed every minute, until the bow/diamond stage (immediately preceding the
nuclear rupture) for S. japonicus. and the corresponding early dumbbell stage for S. pombe.
Representative montages of the time-lapse images are shown on the top of each graph. The final
measurements (at 12 minutes) are taken for fully formed daughter nuclei. The change is shown in
percentages with respect to time 0. Time is in minutes. Note that the nuclear volume remains
fairly constant throughout nuclear division in S. pombe but dramatically drops following nuclear
rupture and reformation of the daughter nuclei in S. japonicus.
(B) Cerulenin inhibits fatty acid biosynthesis in both S. pombe and S. japonicus cells. Shown are
serial dilutions of S. pombe and S. japonicus cultures grown on minimal media agar plates
containing indicated chemicals. Note that in both yeasts the growth inhibition by cerulenin can
be rescued by supplementing the media with exogenous palmitate.
(C) Time-lapse sequences of maximum projection images of GFP-AHDL Nhp6-mCherry
expressing S. japonicus cells that were pretreated with the fatty acid biosynthesis inhibitor
cerulenin for 30 min prior to mitosis. Note that mitosis occurs normally in these cells. Time is in
minutes and seconds. Scale bar represents 5 Pm.
Table S1. List of Strains
Schizosaccharomyces japonicus
Figure
Genotype
1B
Nup85-GFP::ura4+
ura4sj-D3
1C
Lem2-GFP::ura 4+ Nup85-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
2A
Nhp6-GFP::ura4+ Nup189-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
2B
Nhp6-GFP::ura4+ Nup85-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
2C
GFP-GST-NLS-GFP::ura4+ Nup85-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
2D, 4B,
pBip1-linker GFP-AHDL::ura4+
S2B, S4A, S4C Nhp6-mcherry::ura4+ ura4sj-D3 ade6sj-domE?
3A
GFP-Atb2::ura4+ Nhp6-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
3B
lem2'::kanR Nup85-GFP::ura4+
ura4sj-D3 ade6sj-domE?
3C
lem2'kanR GFP-Atb2::ura4+ Nhp6-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
3D
mad2'kanR GFP-Atb2::ura4+ Nhp6-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
S1A
Nup132-mCherry::ura4+
ura4sj-D3 ade6sj-domE
S1A
Nup189-mCherry::ura4+
ura4sj-D3 ade6sj-domE
S1A
Tts1-GFP::kanR
ura4sj-D3
S1B
Ost1-GFP::ura4+ , Nup85-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
S1C
Ost1-GFP::ura4+ ura4sj-D3 ade6sj-domE? h+
S1D
Sec63-GFP::ura4+ ura4sj-D3 ade6sj-domE? h+
S1E
Sec63-GFP::ura4+ Erb1-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
S1F
Nup85-GFP::ura4+ Fib1-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
S2A
GST-NES-GFP::ura4+
ura4sj-D3 ade6sj-domE
S2C
Mis6-GFP::ura4+ Nhp6-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
S3
GFP-Atb2::ura4+ Nup189-mCherry::ura4+
ura4sj-D3 ade6sj-domE?
S4B
ura4sj-D3 ade6sj-domE h-
Collection No.
SOJ54
SOJ469
SOJ400
SOJ238
SOJ427
SOJ501
SOJ359
SOJ268
SOJ387
SOJ452
SOJ253
SOJ380
SOJ17
SOJ260
SOJ241
SOJ257
SOJ472
SOJ332
SOJ429
SOJ414
SOJ398
SOJ11
Schizosaccharomyces pombe
Figure
Genotype
1A
Nup85-linker-GFP::ura4+ h+
ade6-210 leu1-32 ura4-D18
4A, S4A
pBip1-GFP-AHDL::leu1+ h+
ade6? leu1-32 ura4-D18
4C
nmt81-GST-NLS-mCherry::leu1
Tts1-linker-mCherry::ura4+ kanR::nmt81-GFP-Atb2
ade6-? leu1-32 ura4-D18
S4B
ade6-210 leu1-32 ura4-D18 h+
Collection No.
SO3985
SO4808
SO5962
SO2865
Supplemental Experimental Procedures
Construction of Schizosaccharomyces japonicus Strains
The
genome
assembly
for
Schizosaccharomyces
genus
is
available
at
http://www.broadinstitute.org/annotation/genome/schizosaccharomyces_group/MultiHome.html.
To introduce GFP- or mCherry-tagged markers we used the homologous recombination method
to integrate recombination cassettes either at the endogenous or the ura4 (extra-copy) loci. The S.
pombe plasmid backbone pJK210 was modified to include the S. japonicus ura4 gene as a
selection marker. The GFP-Atb2 strain was constructed by introducing an extra copy of Nterminally tagged GFP-Atb2 under the control of its own promoter at the ura4 genomic locus.
The artificial markers GFP-NLS-GST-GFP and NES-GST-GFP were integrated at the ura4 locus
and expressed from the atb2 promoter. These markers were derived from S. pombe plasmids
pREP81-GST-NLS-GFP and pREP81-GST-NES-GFP that were kindly provided by Dr. M. Sato
(University of Tokyo). The artificial ER marker GFP-AHDL was constructed as in [1] and
integrated at the ura4 locus. The PCR-based recombination method adapted from [2] was used to
generate lem2' and mad2'strains. We used electroporation for yeast transformation [3]. Genetic
crosses were carried out as previously described [4]. Free spore analyses were performed by
treating asci with 0.05% glusulase overnight and plating spores on YES medium plates followed
by replica plating on selection media.
We tested efficacy of two microtubule-depolymerizing drugs in S. japonicus. Unlike in S.
pombe, we did not observe any microtubule depolymerization using methyl benzimidazol-2-ylcarbamate at various concentrations. However, microtubules were efficiently depolymerized
when cells were treated with thiabendazole (TBZ, Sigma T-8904).
Image Acquisition and Analysis
Time-lapse fluorescence microscopy images were acquired using a Zeiss Axiovert 200M
microscope (Plan Apochromat 100X, 1.4NA objective) equipped with an UltraView RS-3
spinning disk confocal system (PerkinElmer Inc., Boston, MA, USA) including a CSU21
confocal optical scanner, 12-bit digital cooled Hamamatsu Orca-ER camera (OPELCO, Sterling,
VA, USA) and krypton-argon triple line laser illumination source (488, 568 and 647 nm), under
the control of Metamorph software (Universal Imaging, Sunnyvale, CA, USA). Typically, we
acquired z-stacks consisting of 13 0.5Pm-spaced planes. The z-stack maximum projection were
obtained using the Metamorph built-in module. Scanning confocal microscopy was performed
using a LSM 510 microscope (Carl Zeiss, Inc.) equipped with a Achroplan 100X 1.25NA
objective, a 488-nm argon and a 543-HeNe lasers. Z-stacks were taken with a spacing of 0.5 Pm
and the maximum projection images were obtained in ImageJ 1.42q (NIH ,USA). For both
spinning disk and scanning confocal microscopy, cells were maintained on freshly sealed 2%
agarose YES pads.
Quantitation of cellular Nhp6-GFP fluorescence was performed using ImageJ as follows.
First, we ascertained that the total fluorescence intensity of Nhp6-GFP remained constant
throughout mitosis. Second, the GFP fluorescence intensity was adjusted for bleaching and
background fluorescence. Nup85-mCherry was used as a marker outlining the NE. The
integrated GFP fluorescence intensity was measured separately in the nucleus and cytoplasm, at
three cell cycle stages: at early mitosis , anaphase B and immediately following exit from
mitosis, and normalized with respect to the total cellular fluorescence (n=5 cells).
In order to quantify surface area and volume of S. japonicus and S. pombe mitotic nuclei,
we analyzed time-lapse sequences of z-stacks of 0.5 Pm–spaced (h) planes covering the entire
GFP-AHDL-labeled nuclei. The nuclear outline was traced in ImageJ, yielding values for
perimeter (Pi) and area (Ai) for each respective plane. The total surface area and volume were
then calculated using trapezoid rule, as S (total surface area) = h/2 * ( P1 + 2*P2 + ... + 2*P(m-1) +
Pm ) and V (total volume) = h/2 * (A1 + 2*A2 +... + 2*A(m-1) + Am ) where m is the number of
planes spanning the nucleus. For S. japonicus, we followed nuclear shape changes starting at
metaphase, when nuclei were spherical, until late anaphase when nuclei assumed a highly
elongated and stretched aspect, just before the breakage of the NE. For S. pombe, we measured
nuclear geometries as cells progressed from metaphase to a point at anaphase B directly
preceding the resolution of a mother nucleus into two daughter half nuclei. Plotting of data and
least square linear fits were performed using Matlab (Mathworks, Natick, MA, USA).
Supplemental References
1.
2.
3.
4.
Zhang, D., Vjestica, A., and Oliferenko, S. (2010). The cortical ER network limits the
permissive zone for actomyosin ring assembly. Curr Biol 20, 1029-1034.
Bahler, J., Wu, J.Q., Longtine, M.S., Shah, N.G., McKenzie, A., 3rd, Steever, A.B.,
Wach, A., Philippsen, P., and Pringle, J.R. (1998). Heterologous modules for efficient
and versatile PCR-based gene targeting in Schizosaccharomyces pombe. Yeast
(Chichester, England) 14, 943-951.
Aoki, K., Nakajima, R., Furuya, K., and Niki, H. (2010). Novel episomal vectors and a
highly efficient transformation procedure for the fission yeast Schizosaccharomyces
japonicus. Yeast (Chichester, England) 27, 1049-1060.
Furuya, K., and Niki, H. (2009). Isolation of heterothallic haploid and auxotrophic
mutants of Schizosaccharomyces japonicus. Yeast (Chichester, England) 26, 221-233.