Journal of Cell Science 101, 267-275 (1992)Printed in Great Britain © The Company of Biologists Limited 1992

267

Visualization of centromeric and nucleolar DNA in fission yeast by

fluorescence in situ hybridization

SATORU UZAWA* and MITSUHIRO YANAGIDAt

Department of Biophysics, Faculty of Science, Kyoto University. Kitashirakawa, Sakyo-ku, Kyoto 606, Japan

•Present address: Department of Molecular Biology, Graduate School of Medical Science, Kyusyu University, Fukuoka 812, JapantAuthor for correspondence

Summary

The nucleolar and centromeric DNAs of the fission yeastSchizosaccharomyces pombe were visualized in thenucleus by fluorescence in situ hybridization usingrepetitive ribosomal and centromeric DNAs as theprobes. The rDNAs were seen in the nuclear domainpreviously assigned as nucleolar, that is, the region intowhich the rod-like chromatin protrudes from thehemispherical chromosomal domain. Using mitotically-arrested cells containing condensed chromosomes, it wasdemonstrated that the rDNAs were present on thesmallest chromosome HI, consistent with genetic data.Using a centromeric repetitive element as the hybridiz-ation probe, the centromere of chromosome HI, cen3,

which contains the largest number of the repetitiveelements, was visualized. The centromere in interphasecells is located near the periphery of the nucleus as asingle dot. Early in mitosis, however, it divides into twoand is situated in the middle of the short mitotic spindle.After spindle extension in anaphase, the centromericDNA is present at both ends of the spindle, that is, nearthe spindle pole bodies. The movement of cen3 duringmitosis (anaphase A and B) is discussed in relation tospindle dynamics and chromosome separation.

Key words: fluorescence in situ hybridization (FISH),nucleolus, centromere, kinetochore, anaphase, mitosis.

Introduction

Fluorescence in situ hybridization (FISH) is a powerfultechnique used to localize specific DNA sequences inthe chromosomes of higher eukaryotes (Rudkin andStollar, 1977; Trask, 1991a and b). DNA sequences arefirst labelled with reporter molecules. The chromo-somes in the specimen cells and the probe DNA aredenatured by briefly exposing them to a high pH todisrupt the DNA base pairing. Complementary se-quences in the target chromosome and the probe arethen allowed to reanneal. After washing and incubationin fluorescently labelled affinity reagents, a discretefluorescent signal is visible at the site of probehybridization. The highest resolution and sensitivitywould allow one to locate a single unique gene inindividual chromosomes. FISH methods have also beenused to reveal the distribution of specific RNAmolecules in cells and within tissues. Striking patternsof differential gene expression have been observed indeveloping fruit fly and mouse embryos. In spite of thewealth of information about higher eukaryotic chromo-somes or gene expression obtained via FISH, little isknown about the application of FISH technology tolower eucaryotes such as yeast. The techniques arepotentially valuable for understanding the spatial

organization of the yeast genome in the nucleus and theregulation of RNA transport from the nucleus to thecytoplasm.

We have been investigating the higher order chromo-some structure of the fission yeast Schizosaccharomycespombe (Yanagida, 1990). The interphase nucleus ofgrowing S. pombe cells is spherical with a diameter ofapproximately 3 fan. The chromosomal DNA occupiesroughly half of the nucleus, leaving the other hemi-spherical half for the RNA-rich, non-chromosomaldomain (Toda et al., 1981). The haploid organism hasonly three chromosomes (Kohli et al., 1977; Umesonoet al., 1983), and its genome size is approximately14xlO6 base pairs (bp) (Fan et al., 1989). Thereforechromosomal DNA is compacted roughly 2000-fold,even in the interphase nucleus. However, a histone Hl-like protein, thought to be important for chromatincompaction in other eukaryotes, has not been identifiedin S. pombe (Yanagida, 1990). During the cell divisioncycle, chromatin undergoes striking morphologicalalterations including compaction and segregation dur-ing mitosis as analyzed by fluorescence microscopy(Toda et al., 1981; Tanaka and Kanbe, 1986), and insome mitotic mutants the chromosomes are highlycondensed (Hirano et al., 1988; Ohkura et al., 1989).For a detailed understanding of chromosomal behav-

268 S. Uzawa and M. Yanagida

iour during the cell division cycle, it is essential to locatespecific chromosomal DNAs in the nucleus at differentcell cycle stages. In the present paper, we have made aninitial attempt to locate such chromosomal DNAs in thefission yeast Schizosaccharomycespombe by FISH. TheDNA probes used for in situ hybridization were thehighly repetitive rDNA and a middle-repetitive centro-meric DNA. The former is a major component of thenucleolus and the latter might be complexed withproteins to form a part of the kinetochore, which is avital structure for the segregation and separation of thechromosomes in mitosis and meiosis.

Materials and methods

Strains and mediaSchizosaccharomyces pombe haploid strains, derivatives ofh~ leul, were used. A cold-sensitive (cs) mutant nda3-KM311has been previously described (Hiraoka et al., 1984). Mutantcells were highly synchronously arrested at the restrictivetemperature (20°C) for 8 h. Cells were cultured in rich YPD(1% yeast extract, 2% polypeptone, 2% glucose) medium.

Indirect immunofluorescence microscopyFor immunofluorescence microscopy, the procedure de-scribed by Hagan and Hyams (1988) was followed forpreparing and fixing cells. For microtubule staining, themonoclonal antibody raised against Trypanosoma brucel a-tubulin TAT-1 (Woods et al., 1989; kindly donated by Dr. K.Gull) was used. For second antibodies, Texas red- and FITC-conjugated goat anti-mouse and rabbit antibodies (EY Lab)were used. Fluorescence was observed by a Zeiss Axiophotowith the 100W light source (HBO 100W/2).

Fluorescence in situ hybridization methodS. pombe cells were fixed according to the procedures used forimmunofluorescence microscopy (Hagan and Hyams, 1988).All the reagents used were filtered with a 0.22 /an pore-sizedmembrane. Novozyme 234 (0.1 mg/ml, Novo) and Zymolyase100T (0.6 mg/ml, Seikagaku) was used to digest the cell wallfor 2 h, and cells were treated with RNAase A (0.1 mg/ml,Sigma) at 37°C for 1 h. The RNA-digested cells could bestored at 4°C for approximately one month without anydeleterious effects on the signal. To alkali-denature nuclearDNA, lxlO7 to 2xlO7 cell pellets were resuspended in 0.1 MNaOH and kept at room temperature for 2 min, followed bycentrifugation at 10,000 revs/min for 5 s. Cells wereresuspended in 100-200 /il heat-treated (65°C for 10 min)hybridization buffer (50% formamide, 2xSSC, 10% dextransulfate, 5xDenhardt's solution, 0.5 mg/ml salmon spermDNA) containing a digoxigenin-labelled probe DNA (Boehr-inger; Lichter et al., 1990) and incubated for hybridization at37°C for 12-16 h. Cells were washed three times in 2xSSC atroom temperature, resuspended in PBS-BAG (PBS, 1%bovine serum albumin, 0.1% sodium azide, 0.5% cold waterfish skin gelatin (Sigma)) and kept at room temperature for 30min. Cells were pelleted by centrifugation and resuspended in100 )A PBS-BAG containing anti-digoxin antibodies (5,000-and 200-fold dilutions, respectively, for mouse monoclonaland rabbit polyclonal antibodies). A random priming label-ling kit (Boehringer) was used to prepare the probes usingDNA previously fragmented with restriction enzymes. Non-reacted nucleotides were removed by gel filtration (SephadexG-50). The probes could be kept at -30°C for up to 6 months

without damage. The standard probe concentration was 1ng/̂ 1 in the hybridization buffer. The probe was denatured at65°C for 10 min prior to annealing. To detect hybridized DNAprobes, rabbit anti-digoxin antibody (polyclonal, EY Lab)and mouse anti-digoxin antibody (monoclonal, Sigma) wereused as primary antibodies. The secondary antibodies weregoat antibodies. The procedures for in situ hybridizationcoupled with antigen detection by indirect immunofluor-escence were described by Masumoto et al. (1989).

Results

FISH of nucleolar rDNAThe interphase nucleus of 5. pombe stained by a DNA-specific fluorescent probe, DAPI, reveals the hemi-spherical chromosomal DNA domain (Toda et al.,1981). Short rods protrude from the flat surface of thechromosomal hemisphere and embed in the RNA-richnon-chromosomal hemisphere. These protrusions arethought to be the rDNA clusters as they are intenselystained with antibodies directed against the largestsubunit of RNA polymerase I (Hirano et al., 1989)which produces rRNA molecules. rDNAs are tandemlyrepeated (Schaak et al., 1982) and present in the distalregions of chromosome III (Toda et al., 1984; Fan etal., 1989; Yanagida et al., 1991). We applied FISH atfirst to rDNA, because it is the most abundantlyrepetitive DNA in the genome of S. pombe (the copynumber of rDNA is estimated to be ~ 100-150).

We have done FISH of wild-type cells using digoxi-genin-labelled rDNA repeat units and anti-digoxinantibodies by the procedures described in Materials andmethods. We found that probe hybridization took placein a particular area in the nucleus as shown in colorfluorescence micrographs (Fig. 1). The nucleolar regionwas heavily stained by Texas red-conjugated goatantibody against mouse monoclonal anti-digoxin anti-body (middle panel). The whole chromosomal DNAwas stained by blue DAPI (top panel). The micrographobtained by the double exposure (bottom panel) clearlydisplayed the region of probe hybridization in thenucleolar DNA. The anti-digoxin antibody-stainingregions are indistinguishable from those stained by anti-RNA polymerase I antibodies in immunofluorescencemicroscopy (Hirano et al., 1989). No hybridization tookplace when the vector DNA was used as the hybridiz-ation probe or when the rDNA probe was not labelledwith degoxigenin (data not shown).

In nda3-KM31I mutant cells in which chromosomescondense, after incubation at a restrictive temperature(20°C) for 8 h, the nucleolar rDNAs were alsoobserved, by FISH (Umesono et al., 1983; Hiraoka etal., 1984). The nda3+ gene encodes /3-tubulin, and itsmutation causes a highly synchronous block in mitosis.Individual chromosomes were seen in the nda3-KM31Imutant cells, so allowing the chromosomal assignmentof hybridizing sequences by FISH. As seen in Fig. 2(left, FISH; middle, DAPI; right, a schematic represen-tation of superimposed images of FISH and DAPI), theintense fluorescence of the hybridized probe wasassociated with the smallest chromosomes (the filled

Fig. 1. Fluorescence in situ hybridization of rDNA in wild-type fission yeast cells. Top panel, the chromosomaldomain in the nucleus is stained with DAPI, a DNA-specific fluorescent probe. Middle panel, the chromosomalareas of the same cells are denatured and hybridized withthe digoxigenin-labelled rDNA probe, which is reactedwith anti-digoxin antibodies and stained with Texas red-conjugated second antibodies. Bottom panel, doubleexposure for DAPI and rDNA hybridization of the samecells. Bar, 10 jan.

Fluorescence in situ hybridization in yeast 269

Fig. 2. Localization of rDNA in nda3-KM311 cells incubated at the restrictive temperature for 8 h. Left panel, FISHprobed with rDNA; middle panel, DAPI stain; right panel, a schematic representation of superimposed images. The filledregions indicate the hybridized areas. In this mutant, chromosomes are condensed. Short protrusions of the smallchromosomes are hybridized. The hybridized domains are often split into two. Bar, 10 /im.

areas in the illustration). This is consistent with the factthat the shortest chromosome III has the rDNA clusters(Toda et al., 1984; Fan et al., 1989; Yanagida et al.,1991).

FISH of a centromeric repetitive DNAThe fission yeast centromeres, cenl, cen2 and cen3,respectively, contain two, three and thirteen copies of arepeating element designated dg (Nakaseko et al.,1986; Chikashige et al., 1989; Murakami et al., 1991).We used the ~3.2 kb long dg sequence derived fromcenl (Nakaseko et al., 1986) as the hybridization probefor the centromeres. The nucleotide sequence identitiesamong different dg elements in different centromeresare approximately 98% (Nakaseko et al., 1986;Chikashige et al., 1989) so that hybridization shouldoccur with all three centromere regions. The probesequence has previously been shown to exist only in thecen regions (Nakaseko et al., 1986; Chikashige et al.,1989).

We first performed FISH of the dg sequence for thecondensed chromosomes in nda3-KM311 cells incu-bated at the restrictive temperature for 8 h. As seen inFig. 3 (A, B and C), one small dot was consistently seenin the small chromosome, probably corresponding tochromosome III. It is relatively easy to identifychromosome III, because it is not only the smallestamong the three chromosomes (3.5xlO6 bp long,whereas chromosomes I and II are 5.7xlO6 and4.6xlO6 bp, respectively; Fan et al., 1989), but alsooften displays the filamentous rDNA protrusions byDAPI-staining (Fig. 3; Umesono et al., 1983). Morethan 90% of nda3-KM311 cells examined showed onedot in chromosome III.

The reason why we failed to observe consistently thedots in chromosomes I and II is probably due to thelower copy numbers of the dg sequences in theircentromeres. The copy number ratios of dg in cenl,

cen2 and cen3 were 2:3:13 (Chikashige et al., 1989;Murakami et al., 1991). Occasionally we could observesingle dots, produced apparently by probe hybridiz-ation, in another chromosome, probably II (indicatedby arrows in Fig. 4). However, they were not alwaysobserved, and con-ect conditions for reproducibleidentification of the centromeres of the larger chromo-somes remain to be established.

Location of centromeric DNAs in the interphasenucleusWe tried to observe the centromeric repeat element dgin interphase cells by FISH. The single hybridizing dotwas frequently (in more than 50% of cells) seen at theperiphery of the nucleus (Fig. 5). Such a dot was neverobserved when the vector sequence was used as theprobe, nor when the dg sequence was not labelled bydegoxigenin and otherwise identical procedures wereapplied (data not shown).

The location of the hybridizing dot might be similarto that of the spindle pole body (SPB; Tanaka andKanbe, 1986; Kanbe et al., 1990; Horio et al., 1991).However, experiments to determine the mutual lo-cations of centromeric DNA and the SPB have so farfailed; the use of anti-gamma tubulin antibodies tolocate the SPB (Horio et al., 1991) produced a numberof artifactual, weakly fluorescing dots for specimensmade by using antibodies against the hybridizationprobe (data not shown). To date, anti-gamma-tubulinantibodies are the only probes available to determinethe location of SPB in the interphase cells (Horio et al.,1991). Thus the relationship between the centromericDNA and the SPB in the interphase cells remains to beestablished.

Centromeric DNA movement during mitosisFISH allowed us to observe the centromeric repeatelements in wild-type mitotic cells. As shown in the

270 S. Uzawa and M. Yanagida

Fig. 3. FISH of thecentromeric repeat element innda3-KM311 cells incubated atthe restrictive temperature for8 h. The digoxigenin-labelledcentromeric DNA probe(pSS160; Nakaseko et al.,1986) was used forhybridization. Rabbit anti-digoxin antibodies and sheepTexas red-conjugated anti-rabbit antibodies were used asprimary and secondaryantibodies, respectively. (A, Band C) Top is DAPI staining;middle is FISH (the hybridizeddots are indicated by thearrows or the arrowhead;bottom is a schematicrepresentation of thesuperimposed images (thelocation of the centromericDNA is indicated by the smallcircles). The hybridizedstructures were occasionallyseen as two dots (indicated bythe arrowhead in the middlepanel of A). Bar, 10 ^m.

series of micrographs in Fig. 6 A-E, using doubleimmunostaining of anti-digoxin and anti-tubulin anti-bodies, the centromeric element could be located onthe spindle axis. In the left panel, the micrographsobtained by FISH are shown (the dots indicated by thearrows are thought to be the centromeric DNA),whereas in the center left panel, anti-tubulin immuno-fluorescence microscopy of the same cells is displayed.In the center right panel, chromosomal DNA of thesame cells is stained by DAPI and in the right panel,illustrations of the superimposed images for the centro-meric DNA (indicated by the dots), the spindle(indicated by the lines) and the whole chromosomalDNA (indicated by the hatched nuclear areas) arepresented. In most mitotic cells containing the spindle,the two centromeric DNA dots were seen on the spindleaxis; their precise location was dependent upon thelength of the spindle.

We noted that in cells thought to be at early stages ofmitosis and containing the short spindle (A and B), thecentromeric DNA dots seemed to be present in itscentral region. The dots were often seen as a pair. Theirlocation was always in the chromosomal hemisphere,near the periphery of the nucleus. After extensivespindle elongation (C, D, E) in anaphase, the centro-meric DNAs were separated and moved to the two endsof the spindle. They were located at the periphery

of the divided nuclei (D, E). In post-anaphase, aftercomplete spindle degradation, single centromeric DNAdots were observed for each divided nucleus (data notshown). In these cells, the nucleus was rotated as thespindle was degraded. The centromeres are no longerlocated on the spindle axis (see Fig. 7).

Discussion

FISH techniques have hardly been applied in yeast. Tothe best of the authors' knowledge, the present report isthe first application. Two classes of repetitive DNAs,namely rDNA and a centromeric repeat element, werelabelled with the reporter molecules, and used asprobes to visualize these elements in vivo by antibodiesagainst the hybridized probe DNA. Nucleolar rDNAwas repeatedly and clearly observed, but only thecentromere of chromosome III was consistently ob-served. More than ten copies of repetitive sequenceswere required for visualization under the experimentalconditions employed in the present study (the numberof dg sequences in cen3 is 13; Murakami et al., 1991).The fission yeast centromeres are very large comparedto those of budding yeast, and functional analysesindicate that the repetitive centromeric sequences are

Fluorescence in situ hybridization in yeast 271

necessary for normal segregation of the chromosomesin mitosis and meiosis (Niwa et al., 1989; Clarke andBaum, 1990; Matsumoto et al., 1990). We are trying toimprove our experimental conditions to increase theefficiency and sensitivity of the assay in order to detectDNAs with less repetitive sequences. In fact cenl andcen2 are now consistently observed if a better exper-imental conditions are employed (Funabiki, H. et al.,unpublished result).

The repetitive rDNAs have been localized to onechromosome in nda3-KM311 mutant cells. This isconsistent with the fact that rDNA is exclusivelypresent on chromosome III (Toda et al., 1984; Fan etal., 1989; Chikashige et al., 1989; Yanagida et al., 1991;Kurooka, H., unpublished result). The total number ofrDNA repeats was approximately 150. The images ofrDNAs visualized by FISH resembled rods consisting ofclusters of globules. There were often two rods, whichmay correlate with the two rDNA clusters present atboth ends of chromosome III (Yanagida et al., 1991).

Fig. 4. FISH of nda3-KM311 cells which showprobe hybridization to a dot in a largerchromosome (arrows) as well as the smallestchromosome III (arrowheads). (A) DAPI;(B) FISH; (C) an illustration of superimposedimages. Bar, 10 ^m.

Visualization of the centromeric DNA in fission yeasthas opened a way for studying the movement of thekinetochore DNAs during the cell cycle. It will beparticularly valuable for the examination and compari-son of movements in different mitotic mutants defectivein sister chromatid separation (Yanagida, 1989). Thecentromeric repetitive element dg used as the probe isexclusively localized in S. pombe centromeres (Naka-seko et al., 1986; Chikashige et al., 1989). In cen3, it islocalized within a 110 kb long inverted region (Murak-ami et al., 1991); four copies in the left arm and ninecopies in the right arm flank the ~20 kb long centralregion. We assume that all the copies of dg in cen3 werevisualized as a dot by FISH. In interphase cells, thecentromeric DNA was observed as a single dot, but inmitotic cells, two dots were observed. These obser-vations were consistent with the expected behaviour ofthe centromere DNAs in the cell cycle.

We found two types of centromere DNA movementsduring mitosis, namely those at early (probably related

272 S. Uzawa and M. Yanagida

Fig. 5. FISH of the centromeric DNA in wild-type interphase cells. Left panel, FISH. Middle, DAPI. Right, schematicrepresentation of the superimposed images (centromeric dots indicated by open circles, and the chromosome domains byhatched areas). The dots are frequently observed at the periphery of the nucleus. Bar, 10 j<m.

to anaphase A) and later stages (related to anaphaseB). In early mitotic cells containing the short spindle, apair of the dots were seen on the central region of thespindle. Because only a single dot was found in thenuclear periphery of interphase cells, the dots seen inearly mitotic cells might represent sister centromeresdiametrically opposed by the kinetochore micro-tubules. The metaphase-like stage has been demon-strated in the arrested cells of nuc2 mutant (Hirano etal., 1988); condensed chromosomes in nuc2 form theplate-like structure at a restrictive temperature. Thespindle runs perpendicularly through the middle of theclosely arranged chromosomes. In wild-type cells, asimilar stage can be observed (S. Uzawa, unpublishedresult). The short spindle shows a thinner central partwith the thicker sides toward both ends (see Ohkura etal., 1988). The kinetochore microtubules and thekinetochores are thought to be present in the thickersides and the central part of the spindle, respectively.Because the centromeric DNAs were found at bothspindle ends in later mitotic stages, the stage analogousto anaphase A should occur in fission yeast.

The distance between the two centromeric dots wasstrikingly increased in late mitosis by the full extensionof the spindle. This is analogous to anaphase B in highereukaryotic cells. The fully extended spindle is ~15 //mlong, five times longer than that of the short spindle.The centromeric dots were found in the divided nucleiat the spindle ends, that is, the site of the SPB. The

fission yeast SPBs in mitotic cells can be seen by anti-cyclin antibodies (Alfa et al., 1991) and anti-MPM2antibodies (Masuda et al., 1990) as well as by anti-y-tubulin (Horio et al., 1991). The centromeric DNAsappeared to be closely associated with the SPBs in latemitosis and also possibly in interphase, but, in earlymitosis, they were associated with microtubules andmoved to the center of the short spindle, followed byseparation, each moving toward the SPBs.

Thus it might be considered that the movements ofthe centromeric DNAs during the cell cycle of fissionyeast are changes of positioning relative to the SPBs(Fig. 7). However, microtubule association between theinterphase SPBs and the centromeric DNAs is unlikely,because microtubules were not seen in the interphasenucleus (Tanaka and Kanbe, 1986). It is worth notingthat in Drosophila, centromeres are clustered at the topof the nucleus in interphase cells (Hiraoka et al., 1990).The centromeric DNAs visualized in the present studyare only those of cen3 so the behaviour of the two othercen DNAs is unknown. Furthermore, the techniquesemployed did not permit determination of the relativelocation of the SPB and centromeric DNA in interphasecells by antibodies against y-tubulin and probe DNA.Thus the present study is an attempt to visualize thecentromeric DNA in fission yeast. Further study will beneeded to clarify the detailed spatial relationships of themovements of the centromeric DNA, the spindle andkinetochore microtubules and the SPB during mitosis.

o

ITO

Fig. 6. Location of the centromeric DNA in wild-type mitotic cells. Because of double immunostaining for centromeric DNA and tubulin, the micrograph quality ofin situ hybridization was inferior to that observed with single immunostaining. Left end panel, FISH. Second left panel, anti-tubulin antibodies. Right panel, DAPIstain. Right end panel, a schematic representation of superimposed images. Bar, 10 ixm.

5'

i

274 S. Uzawa and M. Yanagida

Fig. 7. Schematic representation of FISH probed with thecentromeric repetitive probe in the cell cycle of fissionyeast. The centromeric DNA visualized by FISH isindicated by the open circle. The nuclear chromosomaldomains and the mitotic spindle are shown by the hatchedareas and the bars, respectively.

We thank Dr Hiroshi Masumoto of Nagoya University forinstructions in the fluorescence in situ hybridization techniqueand Dr Iain Hagan for helpful comments. This work wassupported by grants from the Ministry of Education, Scienceand Culture and the Riken Biodesign Project.

References

Alfa, C. E., Dncommnn, B., Beach, D. and Hyams, J. S. (1991).Distinct nuclear and spindle pole body population of cyclin-cdc2 infission yeast. Nature 347, 680-682.

Chikashige, Y., Klnoshlta, N., Nakaseko, Y., Matsumoto, T.,Murakami, S., Nlwa, O. and Yanagida, M. (1989). Compositemotifs and repeat symmetry in S. pombe centromeres: directanalysis by integration of NotI restriction sites. Cell 57, 739-751.

Clarke, L. and Baum, M. P. (1990). Functional analysis of acentromere from fission yeast: a role for centromere-specificrepeated DNA sequences. Mol. Celt. Biol. 10, 1863-1872

Fan, J.-B., Chikashige, Y., Smith, C. L., Nlwa, O., Yanagida, M. andCantor, C. R. (1989). Construction of a NotI restriction map of thefission yeast Schizosaccharomyces pombe genome. Nucl. AcidsRes. 17, 2801-2818.

Hagan, I. M. and Hyams, J. S. (1988). The use of cell division cycle

mutants to investigate the control of microtubule distribution in thefission yeast Schizosaccharomyces pombe. J. Cell Sci. 89, 343-357.

EDrano, T., Hlraoka, Y. and Yanagida, M. (1988). A temperature-sensitive mutant of the Schizosaccharomyces pombe gene nuc2+

that encodes a nuclear scaffold like protein blocks spindleelongation in mitotic anaphase. J. Cell Biol. 106, 1171-1183.

Hlrano, T., Konoha, G., Toda, T. and Yanagida, M. (1989). Essentialroles of the RNA polymerase I largest subunit and DNAtopoisomerases in the formation of fission yeast nucleolus. J. CellBiol. 108, 243-253.

Hiraoka, Y., Agard, D. A. and Sedat, J. W. (1990). Temporal andspatial coordination of chromosome movement, spindle formation,and nuclear envelope breakdown during prometaphase inDrosophila mehnogaster embryos . J. Cell Biol. I l l , 2815-2828.

Hlraoka, Y., Toda, T. and Yanagida, M. (1984). The NDA3 gene offission yeast encodes /S-tubulin: A cold sensitive nda3 mutationreversibly blocks spindle formation and chromosome movementduring mitosis. Cell 39, 349-358.

Horio, T., Uzawa, S., Jung, M. K., Oakley, B. R., Tanaka, K. andYanagida, M. (1991). The fission yeast Hubulin is essential formitosis and is localized at microtubule organizing centers. J CellSci. 99, 693-700.

Kanbe, T., Hiraoka, Y., Tanaka, K. and Yanagida, M. (1990). Thetransition of cells of the fission yeast /S-tubulin mutant nda3-311 asseen by freeze-substirution electron microscopy. Requirement offunctional tubulin for spindle pole body duplication. J. Cell Sci. 96,275-282.

Kohli, J., Hottinger, H., Munz, P., Strauss, A. and Thuriaux, P.(1977). Genetic mapping in Schizosaccharomyces pombe by mitoticand meiotic analysis and induced haploidization. Genetics 87, 471-489.

Lichter, L., Tang, C. C , Call, K., Hermanson, G., Evans, G. A.,Housman, D. and Ward, D. C. (1990). High-resolution mapping ofhuman chromosome 11 by in situ hybridization with cosmid clones.Science 24, 64-69.

Masuda, H., Hirano, T., Yanagida, M. and Cande, W. Z. (1990). Invitro reactivation of spindle elongation in fission yeast nuc2 mutantcells. J. Cell Biol. 110, 417-425.

Masumoto, H., Sugimoto, K. and Okazaki, T. (1989). Alphoidsatellite DNA is tightly associated with centromere antigens inhuman chromosomes throughout the cell cycle. Exp. Cell Res. 181,181-196.

Matsumoto, T., Murakami, S., Niwa, O. and Yanagida, M. (1990).Construction and characterization of centric circular and acentriclinear chromosomes in fission yeast. Curr. Genet. 18, 323-330.

Murakami, S., Matsumoto, T., Nlwa, O. and Yanagida, M. (1991).Structure of the fission yeast cen3: direct analysis of the reiteratedinverted region. Chromosoma (in press).

Nakaseko, Y., Adachi, Y., Funahashi, S., Niwa, O. and Yanagida, M.(1986). Chromosome walking shows a highly homologousrepetitive sequence present in all the centromere regions of fissionyeast. EMBO J. 5, 1011-1021.

Niwa, O., Matsumoto, T., Chikashige, Y. and Yanagida, M. (1989).Characterization of Schizosaccharomyces pombe minichromosomedeletion derivatives and a functional allocation of their centromere.EMBO J. 8, 3045-3052.

Ohkura, H., Adachi, Y., Klnoshita, N., Niwa, O., Toda, T. andYanagida, M. (1988). Cold-sensitive and caffeine-supersensitivemutants of the Schizosaccharomyces pombe dis genes implicated insister chromatid separation during mitosis. EMBO J. 7, 1465-1473.

Ohkura, H., Kinoshlta, N., Miyatani, S., Toda, T. and Yanagida, M.(1989). The fission yeast dis2+ gene required for chromosomedisjoining encodes one of two putative type 1 protein phosphatases.Cell 57, 997-1007.

Rudkin, G. T. and StoUar, B. D. (1977). High resolution detection ofDNA-RNA hybrids in situ by indirect immunofluorescence. Nature265, 472-473.

Schaak, J., Mao, J. and Soil, D. (1982). The 5.8S RNA gene sequenceand the nbosomal repeat of Schizosaccharomyces pombe. Nucl.Acids Res. 10, 2851-2864.

Tanaka, K. and Kanbe, T. (1986). Mitosis of the fission yeastSchizosaccharomyces pombe as revealed by freeze-substitutionelectron microscopy. / . Cell Sci. 80, 253-268.

Toda, T., Nakaseko, Y., Niwa, O. and Yanagida, M. (1984). Mapping

Fluorescence in situ hybridization in yeast 275

of rRNA genes by integration of hybrid plasmids inSchizosaccharomyces pombe. Curr. Genet. 8, 93-97.

Toda, T., Yamamoto, M. and Yanagida, M. (1981). Sequentialalterations in the nuclear chromatin region during mitosis of thefission yeast Schizosaccharomyces pombe: video fluorescencemicroscopy of synchronously growing wild-type and cold sensitivecdc mutants by using a DNA-binding fluorescent probe. / . Cell Sci.52, 271-287.

Trask, B. J. (1991a). Fluorescence in situ hybridization: applicationsin cytogenetics and gene mapping. Trends Genet. 7, 149-154.

Trask, B. J. (1991b). Gene mapping by in situ hybridization. Curr.Opin. Genet. Dev. 1, 82-87.

Umesono, K., Hiraoka, Y., Toda, T. and Yanagida, M. (1983).Visualization of chromosomes in mitotic arrested cells of the fissionyeast Schizosaccharomyces pombe. Curr. Genet. 7, 123-128.

Woods, A., Sherwin, T., Sasse, R., McRae, T. H., Balnes, A. J. and

Gull, K. (1989). Definition of individual components within thecytoskeleton of Trypanosoma brucei by a library of monoclonalantibodies. J. Cell Sci. 93, 491-500.

Yanagida, M. (1989). Gene products required for chromosomeseparation. J. Cell Sa. Supplement. 12, 213-229.

Yanagida, M. (1990). Higher-order chromosome structure in yeast. J.Cell Sci. 96, 1-3.

Yanagida, M., Niwa, O., Chikashige, Y., Matsumoto, T., Murakami,S., Kurooka, H., Horio, T. and Shimanuki, M. (1991). Genomeanalysis of Schizosaccharomyces pombe. In Control of Cell Growthand Division (ed. Ishihama, A. and Yoshikawa, H.), pp. 255-262.Springer-Verlag, Berlin.

(Received 11 September 1991 - Accepted 28 October 1991)