Cell, Vol.
112, 407–421, February 21, 2003, Copyright 2003 by Cell Press
Centromeres and Kinetochores:
From Epigenetics to Mitotic
Checkpoint Signaling
Don W. Cleveland,1,* Yinghui Mao,1
and Kevin F. Sullivan2
1
Ludwig Institute for Cancer Research and
Department of Cellular and Molecular Medicine
University of California, San Diego
9500 Gilman Drive
La Jolla, California 92093
2
The Scripps Research Institute
10555 N. Torrey Pines Road
La Jolla, California 92037
The centromere is a chromosomal locus that ensures
delivery of one copy of each chromosome to each
daughter at cell division. Efforts to understand the
nature and specification of the centromere have dem-
onstrated that this central element for ensuring inheri-
tance is itself epigenetically determined. The kineto-
chore, the protein complex assembled at each
centromere, serves as the attachment site for spindle
microtubules and the site at which motors generate
forces to power chromosome movement. Unattached
kinetochores are also the signal generators for the
mitotic checkpoint, which arrests mitosis until all
kinetochores have correctly attached to spindle mi-
crotubules, thereby representing the major cell cycle
control mechanism protecting against loss of a chro-
mosome (aneuploidy).
Introduction
Chromosome movement on the spindle during mitosis
and meiosis is powered and regulated by the centro-
mere, a discrete locus on each chromosome. Originally
identified as the primary constriction, this region is a
complex chromosomal substructure (Figure 1A). While
the term centromere is generally taken to refer to the
DNA segment that confers centromere function, the cy-
tologically visible centromere is more complex . In mito-
sis, a proteinaceous structure, the kinetochore, assem-
bles at the surface of the centromere and acts as the site
of spindle microtubule binding. By electron microscopy,
the kinetochore appears as a narrow band of dense
chromatin just at the surface of the primary constriction,
the inner kinetochore, and a laminar outer kinetochore
domain (frequently misnamed a “plate”) that contains
many of the microtubule binding and signal transduction
molecules discussed below. The monotonous hetero-
chromatin that links two sister kinetochores has
emerged as a distinct functional entity within the centro-
mere-kinetochore complex, important for cohesion and
kinetochore regulation. The boundaries of the centro-
mere locus therefore extend beyond the kinetochore-
forming region. In this review, we discuss how centro-
meres and their kinetochores assemble, how they power
mitotic chromosome movements, and how, as signaling
*Correspondence: dcleveland@ucsd.edu
Review
elements of the mitotic checkpoint, they control cell
cycle advance during cell division.
Defining the Locus
The centromere challenges the classic view of a genetic
locus. Dogma and experience suggest that a chromo-
somal locus is defined by its DNA sequence, and its
function is contained with the information content pres-
ent in the primary sequence, e.g., recognition sites for
DNA binding proteins. This simple situation is correct for
the centromere of budding yeast, where point mutations
abolish activity. In other organisms, however, including
fission yeast, specific centromere-nucleating DNA se-
quences have not been found, and centromere function
exhibits properties of an epigenetic locus, behaving as
a self-replicating protein complex that resides on cen-
tromere DNA but is not determined by it.
Although the details differ markedly, resolution of bio-
chemically and structurally distinct centromere domains
in many organisms has revealed common aspects of
organization (Figure 1B), providing a sketch of a simple
“universal centromere architecture.” Chromatin is the
key feature and the centromere domain is built on a
distinct type of nucleosome found nowhere else in the
genome, in which histone H3 is replaced by a divergent
(50% identity) homolog usually referred to as CENP-A
(Smith, 2002). This core is flanked by another chromatin
domain, a highly phased nucleosome array in S. cerevis-
iae and heterochromatin in other species that plays a
role in cohesion of sister centromeres (Bernard et al.,
2001). Both domains are required for complete centro-
mere function.
The Single Nucleosomal Centromere
of Budding Yeast
The budding yeast centromere is by far the simplest
and most dissected one (Figure 2). Spanning only ⵑ125
bp of DNA, S. cerevisiae centromeres contain three dis-
tinct DNA sequence elements (Figure 2A). Two of these,
CDE I (purple, Figure 2A) and CDE III (red, Figure 2A),
are conserved at each chromosome and function as
sequence-specific elements by the criterion that single
point mutations abolish their activity. The other, CDE II
(blue in Figure 2A), is not conserved in sequence among
the yeast chromosomes, but the approximate length
(76–84 bp) and A⫹T rich content (90%) are maintained
(reviewed in detail by Cheeseman et al., 2002b). The
125 bp centromere DNA is flanked on either side by
a uniquely positioned (i.e., highly phased) nucleosome
array rich in cohesins and thought to provide a functional
context for centromere activity (Laloraya et al., 2000;
Tanaka et al., 1999).
Somewhat ironically, CDE II, the unconserved DNA
element, interacts with Cse4 (Stoler et al., 1995), the
homolog of CENP-A, the most highly conserved protein
motif known for the centromere (Figures 2A and 2B).
CENP-A/Cse4 nucleosomes are specifically associated
with centromere DNA in vivo (Meluh et al., 1998), strongly
suggesting their assembly as a nucleosomal complex at
Cell
408

Figure 1. Organization of Centromeres

(A) Overall organization of the centromere. A mitotic chromosome has been sectioned along the plane of the spindle axis, revealing the
symmetric bipolar organization of a chromosome fully engaged on the spindle. (Right) Key elements have been pseudo colored. (Violet) The
inner centromere, a heterochromatin domain that is a focus for cohesins and regulatory proteins such as Aurora B and Kin I. (Red) The inner
kinetochore, a region of distinctive chromatin composition attached to the primary constriction. (Yellow) The outer kinetochore, the site of
microtubule binding, is comprised of a diverse group of microtubule motor proteins, regulatory kinases, microtubule binding proteins, and
mitotic checkpoint proteins.
(B) Schematic illustration of centromere loci. Organization of centromeric DNA sequences from the four example organisms. (Top) Budding
yeast with a 125 bp centromere comprised of three sequence domains (pink, red, yellow). Fission yeast centromeres show an organized
structure, with a nonconserved central core (red), flanking inner repeats (pink arrows) at which the CENP-A-containing nucleosomes assemble,
and conserved outer repeats (stippled purple). The Drosophila centromere spans ⵑ400 kb (red) embedded in constitutive heterochromatin
(purple). (Bottom) Human centromeres have sizes approaching 10 Mb and are comprised of ␣-I satellite DNA (red) and a more divergent, less
regular ␣-II satellite (pink), flanked by heterochromatin (purple).

the centromere (Figure 2B). A second CDE II interacting
protein, Mif2 (Meluh and Koshland, 1995), is the homolog
of an essential mammalian kinetochore chromatin bind-
ing protein CENP-C (Figure 2B).
Centromere function in yeast also depends critically
on sequence specific DNA-protein interactions. CDE III
binds an essential four subunit protein complex, Cbf3
(Gardner et al., 2001) (Figures 2B and 2C). CDE I binds
a nonessential bHLH DNA binding protein, Cbf1, that
interacts with Cbf3 components (Hemmerich et al.,
2000). The geometry dictates that the Cbf1-Cbf3 interac-
tion must take place across the DNA gyres wound over
the Cse4 nucleosome, perhaps stabilizing the higher
order complex like a protein clamp (Figure 2B). The
Cbf3 complex produces an extended footprint (80 bp)
on centromere DNA (Russell et al., 1999; Espelin et al.,
1997). The resulting broad interaction surface may be
important for stable centromere binding or to provide a
large binding surface for more distal kinetochore pro-
teins, or both. The flanking chromatin domain functions
in cohesion (Megee et al., 1999; Tanaka et al., 1999).
How is the centromeric nucleoprotein complex at-
tached stably to a single, highly dynamic (e.g., He et al.,
2000) spindle microtubule? Four key protein complexes
comprised of ⬎20 components are central players in
outer kinetochore assembly and microtubule binding
(Figures 2C and 2D). The Ctf19 complex and the Ndc80
complexes appear to represent “adaptors” that interact
with both core centromere and distal kinetochore or
spindle components (Ortiz et al., 1999; Wigge and Kilm-
artin, 2001; Janke et al., 2001). The Dam1p complex
(Cheeseman et al., 2001; Janke et al., 2002) (also known
as DASH [Li et al., 2002]) may be the central component
of microtubule binding (Figures 2C and 2D), with its
activity regulated by the yeast Aurora B kinase (Ipl1)
(Tanaka et al., 2002; Biggins et al., 1999). Present in a
conserved complex with the yeast homologs of INCENP
(Sli15) and survivin (Bir1), respectively (Bolton et al.,
2002), Ipl1-dependent phosphorylation of at least three
Dam1p components are essential for microtubule cap-
ture (Cheeseman et al., 2002a).
From the model in Figure 2D, with the components
drawn to scale, it is immediately apparent that connect-
ing a large microtubule (25 nm) with a small, 10 nm
nucleosomal centromere requires multiple attachments,
especially considering that the microtubule-kinetochore
linkage is dynamic, with individual connections broken
and reformed as the microtubule grows or shrinks.
Fission Yeast Centromere: Chromatin
Not a Specific DNA Sequence
The critical importance of the chromatin environment
surrounding the core CENP-A chromatin domain has
been most clearly elucidated in S. pombe. Each of the
three centromeres, which range in size from 40–100 kb,
possess a pair of repeated sequence arrays that are
Review
409

Figure 2. Molecular Details of a Simple Kinetochore

(A) The centromere DNA of S. cerevisiae is shown at top with the three sequence elements, CDE I (purple), CDE II (blue), and CDE III (red),
highlighted. Numbering corresponds to chromosome 11. Proteins that bind to the centromere DNA elements are color coded to match the
corresponding DNA binding domain and are drawn to scale assuming a globular structure. Human homologs, where known, are denoted in
italics. CBF3 contains two copies of Ctf3 and two dimers of Ndc10, but their arrangement in the complex is not known and is likely to be
more asymmetric than shown here.
(B) The nucleosomal centromere. Cse4 (blue and purple spacefill) forms a specialized nucleosome at the centromere, modeled here on the
structure of the nucleosome-containing histone H3 (Luger et al., 1997). Other histones (gray) are shown only as backbone traces. The position
of centromere DNA on the nucleosome is derived from Keith and Fitzgerald-Hayes (2000) by placing CDE I (purple) in contact with one of the
two Cse4 molecules. Most of the remaining histone core surface is in contact with CDE II (blue), while CDE III (red, behind) projects into the
internucleosomal linker DNA. The divergent N-terminal tail of Cse4 is drawn in purple. (Right) The nucleosome is rotated 180⬚ and is shown
with Cbf1 (purple), Mif2 (blue), and Cbf3 (red) placed on DNA elements CDE I, -II and -III, respectively, illustrating the relative sizes and
positions of the core centromere components.
(C) Kinetochore complexes associate with the centromere nucleoprotein core. Four key kinetochore protein complexes discussed in the text
are illustrated. Interactions, genetic or biochemical, are indicated by black arrows, while Ipl1 substrates are shown with red arrows.
(D) A model of budding yeast centromere-microtubule interaction. A microtubule (green) is shown binding to the centromere in the context
of a 30 nm fiber.

arranged in an inverted repeat around an unconserved
central core sequence (Figure 1B) (Clarke, 1998). Both
central core and inner repeat sequences are bound by
the CENP-A homolog (Cnp1) and two conserved pro-
teins, Mis6 and Mis12, homologs of subunits of the Ctf19
complex from budding yeast (Goshima et al., 1999; Par-
tridge et al., 2000; Takahashi et al., 2000).
Unlike budding yeast, no unique sequence within it is
sufficient in S. pombe, with flanking sequences critical
for the establishment and maintenance of centromere
function (Partridge et al., 2000; Ngan and Clarke, 1997;
Ekwall et al., 1997). These are modified by methylation
of lysine 9 of histone H3, which in turn recruits Swi6,
the fission yeast equivalent of vertebrate HP1 (Lachner
et al., 2001). These are characteristic biochemical mark-
ers of heterochromatin, which likely require elements of
the RNAi system in S. pombe by analogy with Cen-
homology-dependent heterochromatic silencing at the
mating type locus (Volpe et al., 2002; Partridge et al.,
2002). A key function of this heterochromatin domain
is the recruitment of the fission yeast cohesin (Rad21)
(Bernard et al., 2001; Tomonaga et al., 2000), the central
component of the protein glue that holds duplicated
chromatids together.
Metazoans: The Mega Multisubunit Centromere
Metazoan centromere organization is in much more
complex than that in the yeasts (Figure 1B). Spanning
0.3–5 Mbp of DNA, human centromeres contain exten-
sive (1,500 to ⬎30,000 copies), tandemly repeated
arrays of a 171 bp sequence element termed ␣ satellite
(Figure 3A). Centromere function has been mapped to ␣
satellite arrays by centromeric deletions, either naturally
occurring on the X chromosome (Schueler et al., 2001)
or those induced by telomere insertion into the Y (Brown
et al., 1994). Centromeric satellite DNAs are not con-
Cell
410

Figure 3. Organization of Human Centromeres

(A) DNA organization. The hierarchic organization of an ␣ satellite DNA is illustrated, with a 171 bp monomer sequence shown at top. Monomer
sequences are iterated with nucleotide sequence variations to form a higher order repeat (colored arrows), which itself is tandemly iterated
with high (⬎99%) sequence conservation to form extensive arrays of higher order repeats. At bottom is a diagram of the centromere region
of chromosome 10, illustrating the ⵑ2 Mb ␣ satellite array with surrounding pericentric satellite arrays (SAT2 and SAT3).
(B) Discontinuous distribution of centromere proteins. Hypotonic stretching of chromosomes followed by immunofluorescence with human
anti-centromere antibodies revealed a punctuate distribution of centromere antigens (Blower et al., 2002), suggesting a multiple subunit
organization at the centromere/kinetochore complex. (Right) Extended chromatin fibers from human interphase cells demonstrate discontinuous
CENP-A domains (red) interspersed with histone H3 domains (green). Images provided by Beth Sullivan (Boston University).
(C) Folding centromeric chromatin in mitotic chromosomes. The discontinuous CENP-A chromatin fiber (left) forms a single domain at the
surface of the chromosome and therefore must be folded or coiled to achieve the polarized distribution of CENP-A and H3 sites in mature
mitotic kinetochores (Sullivan et al., 2001). The coil shown illustrates only one possible folding pathway in the context of an idealized centromere
region, showing that CENP-A (red) assembles onto regular ␣-I satellite sequences (green), with ␣-II satellite (teal blue) and pericentric satellite
sequences (orange, purple) flanking this site. At right, a sinuous path of chromosomal DNA through the centromere region is proposed on
the basis of the differential localization of ␣-I and ␣-II sequences to the outer and inner centromere regions.

served in sequence among metazoans, but share: (1)
their presence in very large tandem arrays, and (2) unit
repeat lengths that tend toward multiples of the nucleo-
some repeat length (Henikoff et al., 2001). (An exception
is Drosophila, where the functionally mapped centro-
mere consists of simple [5 bp] satellite sequences inter-
spersed with various transposons [Sun et al., 1997]).
CENP-A nucleosomes are bound to ␣ satellite DNA
in human centromeres (Vafa and Sullivan, 1997), but
these are not uniformly distributed. Stretched chromatin
fibers reveal interspersed CENP-A- and histone H3-con-
taining nucleosomes (Figures 3B and 3C; Blower et al.,
2002). These foci may represent kinetochore “subunits”
(for which the budding yeast represents a unitary exam-
ple) that assemble together (Zinkowski et al., 1991) to
form multiple binding sites for the multiple microtubules
that attach to metazoan centromeres.
Primate centromeres frequently have two major ␣ sat-
ellite families adjoining each other (He et al., 1998): a
highly regular ␣-I and an ␣-II that varies widely in mono-
mer sequences and repeat structure. CENP-A is bound
primarily to ␣-I satellite sequences, at the surface of the
chromosome (Ando et al., 2002). In contrast, ␣-II satellite
is localized in the interior or central domain of the centro-
mere, where components such as INCENP, Aurora B,
and cohesin are concentrated (Figure 3C). This two-
component organization is evident throughout eukary-
otes. A core domain built around CENP-A and centro-
mere-specific chromatin binding proteins establishes
the kinetochore-forming component of the centromere,
while flanking domains are enriched in proteins involved
in chromatid cohesion.
The Epigenetic Centromere
While in budding yeast centromere DNA alone can nu-
cleate centromere formation de novo, centromeres in
S. pombe depend strongly, and those of metazoan cells
primarily, on epigenetic factors rather than DNA se-
quence for activity. Three lines of evidence support this
statement. First, centromere sequences are evolving at
an unusually high rate, coevolving with their essential
partner CENP-A, and show no obvious sequence con-
servation that links divergent species or even different
chromosomes in the case of Drosophila (Henikoff et al.,
2001). Second, centromere DNA sequences by them-
selves are unable to specify centromere function: stable
dicentric chromosomes have been found in which one
centromere has been silenced with no obvious re-
arrangement of centromere DNA (Sullivan and Willard,
1998). Third, acquisition of centromere function has
been found on certain rearranged chromosomes lacking
an endogenous centromere (Figure 4A). In a well studied
case, a stable derivative of human chromosome 10 with
a deletion spanning the centromere was found to have a
functional centromere at position 10q25. Mitotic stability
through development and adult life in this patient dem-
onstrated that this neocentromere functions and binds
to ⬎20 kinetochore, centromere, and chromatin proteins
(Saffery et al., 2000). This region shows no centromeric
DNA or any other distinctive feature (Barry et al., 1999).
CENP-A is present over a 460 kb domain (Lo et al., 2001),
similar in size to the minimal Drosophila centromere (420
kb; Sun et al., 1997) and the smallest natural human
centromeres (500 kb) (found on Y chromosomes, Tyler-
Smith et al., 1993). Thus, noncentromeric DNA segments
can acquire full centromere function without DNA se-
quence changes and be maintained at those sites indefi-
nitely—even through multiple human generations (Tyler-
Smith et al., 1999).
Chromatin rather than DNA may be a major determi-
nant of centromere identity (Williams et al., 1998). During
Review
411

Figure 4. Epigenetics and DNA Sequences in Centromere Formation

(A) A human neocentromere. (Top) Human chromosome 10. (Below) Rare internal recombination has produced the two subfragments in one
patient. One (left) is a ring chromosome (r(del)10) that carries the original centromere (red). The other is a linear deletion product, mar(del)10,
that has acquired a new centromere at 10q25 (orange). The CENP-A domain of this neocentromere spans about 400 kb.
(B) Generation of neocentromeres. A Drosophila chromosome derivative, Dp1187, was mapped for centromere activity using x-irradiation to
introduce chromosomal breaks (top). In one product, an inversion placed the normally euchromatic sequences on the left arm (cyan) directly
in cis to the functional centromere. Centromeric chromatin is proposed to have spread into the now flanking euchromatic DNA. Subsequent
x-irradiation produced deletion that entirely lacks the initial centromere DNA but, nevertheless, retains substantial centromere activity (bottom)
by virtue of the centromeric chromatin that has spread into the region.
(C) Construction of a human minichromosome. A single copy of the 16mer higher order repeat of chromosome 17 ␣ satellite DNA was amplified
to generate arrays of up to 64 repeat units in a bacterial artificial chromosome vector. These were further amplified to generate fragments
up to ⵑ1 Mb in length. Synthetic ␣ satellite arrays were combined with a ␤-geo resistance marker gene (green), telomere DNA (yellow), and
human genomic DNA (gray) for transfection into HT1080 cells. Linear minichromosomes (6–10 Mb in length, 5%–10% the size of the smallest
human chromosomes) containing 17 ␣ satellite (without detectable host chromosomal DNA) were identified at a low frequency (Harrington et
al., 1997).
(D) Dissecting DNA sequence requirements by minichromosome engineering. (Top) The ␣ satellite domain of human chromosome 21, including
a highly regular ␣21-I array with a high frequency of CENP-B boxes (green ovals) and a flanking ␣21-II array that is highly irregular in sequence
organization and has a low frequency of CENP-B boxes. (Middle) Construction of large arrays containing (␣21-ICENPB⫹) or lacking (␣21-
Icenpb-) functional, 11mer CENP-B B boxes were produced in a BAC vector, amplified, and transfected into HT1080 cells. ␣21-ICENPB⫹
yielded minichromosomes in 40%–50% of transformants; ␣21-Icenpb⫺ yielded none. Amplification of a 340 bp non-␣ satellite sequence into
which CENP-B boxes were embedded (V340:CENPB⫹) yielded no minichromosomes (Ohzeki et al., 2002).

the mapping of a Drosophila centromere by X-ray-
induced chromosome breakage, a chromosome was re-
covered that placed the centromere directly next to a
euchromatic DNA segment (Figure 4B). A subsequent
round of x-irradiation produced stable chromosome de-
rivatives that lacked centromere DNA but retained only
the euchromatic sequence, indicative of a centromere-
determining chromatin structure that spread into the
euchromatin and which retained full centromere func-
tion after subsequent X-ray-mediated severing of the
natural centromere.
The universal association of centromere function with
a distinct histone also supports a role for chromatin in
centromere determination. CENP-A appears to be at
the root of the kinetochore assembly process and is
required for assembly of most distal kinetochore com-
ponents examined (Howman et al., 2000; Oegema et al.,
2001). CENP-A chromatin does not form a specialized
domain of DNA replication (Shelby et al., 2000; Sullivan
and Karpen, 2001), and its loading is uncoupled from
that of the conventional histones (Shelby et al., 1997;
Takahashi et al., 2000). CENP-A is distributed conserva-
tively, however, to the two daughter chromatids during
S phase (Shelby et al., 2000), providing the potential for
carrying a DNA sequence-independent heritable mark
through a CENP-A directed, self-replicating chromatin
assembly pathway that bears little reference to the ac-
tual DNA upon which it sits (for more detailed account,
see Sullivan et al., 2001).
Determinants of De Novo Centromere Formation
It is clear that non-DNA elements play a major role in
centromere determination, but do DNA sequences not
play a role at all? This has been answered by engineering
Cell
412
artificial mammalian chromosomes. Minichromosomes
were initially formed de novo (Harrington et al., 1997)
by transfecting a mixture of a large synthetic chromo-
some 7 ␣ satellite array, telomeric sequences, a se-
lectable marker, and genomic DNA fragments (for origin
and other functions) (Figure 4C). This produced stably
transmitted microchromosomes with functional centro-
meres containing chromosome 7 ␣ satellite sequences
and typical kinetochore components, but only in a hand-
ful of extremely rare instances. Subsequently, a yeast
artificial chromosome (YAC) carrying chromosome 21 ␣
satellite arrays was retrofitted with telomeres by recom-
bination and introduced into human cells, yielding a
much enhanced frequency of microchromosomes con-
taining newly assembled functional centromeres, but
only when the YAC contained ␣-I type satellite arrays
(Ikeno et al., 1998).
The homogeneous, highly regular ␣-I DNA has a high
frequency of sites for the centromere-associated DNA
binding protein CENP-B; ␣-II does not. This suggested
a role for CENP-B in centromere formation, an idea pre-
viously dismissed by demonstration that CENP-B null
mice are viable and fertile with only mild phenotypic
effects on gonad size (Hudson et al., 1998). By engi-
neering two ␣-I satellite arrays such that they differed
only by the presence of functional CENP-B boxes (Fig-
ure 4D), CENP-B was shown to be necessary for de
novo centromere formation, but it functions efficiently
only in the context of ␣-satellite DNA. Since heterochro-
matin formation is a highly cooperative process driven
by a series of mutually reinforcing reactions (Richards
and Elgin, 2002), components such as CENP-B likely
function in mediating centromeric chromatin modifica-
tion, as has been seen for the three CENP-B homologs
in S. pombe (Nakagawa et al., 2002).
Kinetochores Power Chromosome
Movements in Mitosis
Since the term “mitosis” was introduced over 100 years
ago by Fleming, a challenge has been to understand
how cells divide and faithfully transmit chromosomes at
each cell division. In a typical somatic cell cycle (Figure
5A), chromosomes in prophase initiate condensation,
then, upon disassembly of the nuclear envelope and
the interphase microtubule array, the fully compacted
chromosomes spill into what was the cytoplasm to pro-
duce prometaphase. As a nascent mitotic spindle as-
sembles, a dynamic process of repetitive search by un-
stable microtubules ensues for capture of chromosomes
at their kinetochores (Figure 1A). Initial capture is fre-
quently by binding of one kinetochore of a duplicated
chromosome pair along the side of a spindle microtubule
(Figure 5B, violet chromatid pair), allowing rapid (up to
1 ␮m/s) poleward translocation along that microtubule
powered by a minus-end directed, kinetochore bound
microtubule motor, almost certainly cytoplasmic dynein
(Rieder and Alexander, 1990). This is followed by attach-
ment of additional microtubules (up to 30 in humans
[Rieder, 1981] or seven at mouse kinetochores [Putkey
et al., 2002]), motor action at attachment sites, and oscil-
latory movements linked to continued growth and
shrinkage of those kinetochore bound microtubules
(Figures 5B and 5C, orange chromatid pair).
Subsequent capture by the unattached kinetochore of
a microtubule from the opposite spindle pole produces
biorientation and congression to the cell center in a
rapid, discontinuous series of movements, again medi-
ated by kinetochore motors (Figure 5B, red chromatids).
The presence at kinetochores of active plus and minus
end motors has been demonstrated, with net direction
of movement switchable in vitro by phosphorylation (Hy-
man and Mitchison, 1991). In metazoans (Figure 6A),
known kinetochore motors include the kinesin family
member CENP-E and cytoplasmic dynein.
Motors in Congression and Anaphase
A long perplexing question has been how congression
is achieved. With a common set of motors at both sister
kinetochores, how is it that movement of a bioriented
pair to the center is dominant, especially for those (like
the red pair in Figure 5B) that can have many more
microtubules bound to the kinetochore first attached
and that are initially much closer to one pole? Laser
ablation to disconnect the two chromatids or destroy
either kinetochore established that almost all of the force
is generated by the leading kinetochore (Khodjakov and
Rieder, 1996), i.e., the one whose bound microtubules
are shortening and whose motors are moving toward
the microtubule minus ends.
Two soluble dynein inhibitors (p50 dynamitin and a
dynein antibody) disrupt the alignment of kinetochores
at metaphase (Sharp et al., 2000). Mutations in the teth-
ers (Rough deal [Rod] and Zeste white 10 [ZW10]) that
link dynein to kinetochores attenuate the rate of pole-
ward chromosome movement (Savoian et al., 2000), im-
plicating dynein as a likely primary motor for con-
gression.
Other motors, especially the plus end motor CENP-E,
stabilize microtubule capture at both sisters (McEwen
et al., 2001; Putkey et al., 2002) in part through a contri-
bution of CENP-E in maintaining attachment of kineto-
chores to the end of a disassembling microtubule (Lom-
billo et al., 1995). Chromosome alignment is precluded
by disruption of CENP-E function in vitro (Wood et al.,
1997) or in vivo (Putkey et al., 2002). A final class of
kinetochore component, exemplified by the Kin I sub-
group of the kinesin family (including XKCM1 in Xenopus
[Walczak et al., 1996] and MCAK in mammals [Worde-
man and Mitchison, 1995]), is a microtubule disassem-
blase [Desai et al., 1999]. This novel action acts at and
between the two sister kinetochores, using the normal
kinesin ATPase cycle to power removal of tubulin sub-
units.
Chromosomes also experience forces exerted along
the chromosomes arms as a result of spindle microtu-
bule interaction with plus-end directed microtubule mo-
tors bound to chromatin (chromokinesins). The first
identified, Drosophila Nod, is required for proper align-
ment of meiotic chromosomes that have not undergone
recombination (Afshar et al., 1995). Immunodepleting
another (Kid) prevents normal chromosome alignment
(Antonio et al., 2000; Funabiki and Murray, 2000), while
antibody-induced inhibition of human Kid blocks chro-
mosome oscillations (like those of the red chromatid
pair, Figure 5C), with chromosome arms atypically ex-
tending toward spindle poles during congression (Lev-
esque and Compton, 2001).
Review
413
Modeling Chromosome Congression
Among the early models to explain congression were
proposals that the force generated was proportional to
the length of the attached microtubule (Hays et al., 1982;
Ostergren, 1951). The explanation was as perplexing as
the initial problem, as no molecular mechanism was
apparent for how such a length-dependent force could
actually be generated through action at kinetochores.
With the demonstration that ZW10 and Rod apparently
stream poleward after microtubule capture (Williams et
al., 1996), as do many of the kinetochore bound compo-
nents of the mitotic checkpoint (Howell et al., 2000,
2001), it seems to us that a model compatible with the
key evidence is that congression is powered, at least in
part, by cytoplasmic dynein asymmetrically bound to
the leading kinetochore of a chromatid pair.
The 1-50 pN force generated during chromosome
movement (Alexander and Rieder, 1991; Nicklas, 1983,
1988) is within the capacity of one or a few dynein mole-
cules (ⵑ2–6 pN per motor molecule [Shingyoji et al.,
Figure 5. Chromosome Movements in Mi-
tosis
(A) Summary of the stages of mitosis. (Red)
DNA; (green) microtubules. Images provided
by Andreas Merdes (University of Edinburgh).
(B) Chromosome movements during mitosis:
(violet) one kinetochore of a duplicated chro-
matid pair becomes laterally attached to a
single microtubule and rapidly powers the
pair poleward. (Orange) The mono-oriented
chromosome acquires additional microtu-
bules and then oscillates near the pole. (Red)
After capture by the unattached kinetochore
of a microtubule from the opposite pole, the
now bioriented chromosome congresses to
the spindle equator. (Blue) At anaphase, the
sister chromatids separate and undergo pole-
ward motion.
(C) The speeds of chromosome movements
for the chromatids in (B). (Violet) Mono-ori-
ented rapid poleward translocation; (orange)
oscillatory movements prior to biorientation;
(red) bioriented congression and oscillation;
and (blue) poleward motion in anaphase.
Adapted from Skibbens et al. (1993). Repro-
duced from The Journal of Cell Biology, 1993,
vol. 122, pp. 859–875 by copyright permission
of the Rockefeller University Press.
1998]). Considering that dynein is tethered to the under-
lying kinetochore by ZW10/Rod (see model in Figure
6B), congression would be achieved if the affinity of
these tethers for kinetochore sites was weakened by
increasing microtubule occupancy at that kinetochore.
In other words, if distortion of the kinetochore occurs
as more microtubules bind and this decreases the
strength of ZW10/Rod binding, then dynein’s action
would simply pull itself and ZW10/Rod off of the more
highly occupied, trailing kinetochore, producing the
streaming of ZW10 so prominently seen on both sides
of the sister kinetochores after congression (Williams
et al., 1996). The continuing oscillations at metaphase
(Figures 5B and 5C, red chromatid pair) would simply
reflect stoichastic changes in microtubule number on
the two sides, thereby switching which kinetochore
could more tightly hang onto its dynein.
A cautionary note, however, is that there are multiple
contributors and multiple solutions to congression so
that the model put forward here is likely only be one
Cell
414

Figure 6. Kinetochore Microtubule Capture and Chromosome Congression

(A) Kinetochore microtubules are captured by motor proteins, including CENP-E and cytoplasmic dynein, the latter in associated with dynactin,
ZW10, and Rod. Kin I acts as a microtubule disassemblase at and between the two sister kinetochores. Microtubule plus-end tracking proteins
(such as EB1, APC) track microtubule “plus” ends and contribute to kinetochore microtubule stability and attachment. Plus-end directed
chromokinesins are localized on chromosome arms and are responsible for driving the chromosome arms away from the poles (the “polar
ejection” force).
(B) A model for chromosome congression in prometaphase driven by dynein-dependent minus-end directed kinetochore motility. Increasing
microtubule occupancy at the trailing kinetochore lowers ZW10/Rod affinity for that kinetochore causing dynein to pull itself and ZW10/Rod
out of the kinetochore and to stream along the kinetochore microtubule. On the leading kinetochore with fewer microtubules bound, ZW10/
Rod is tethered more tightly, and the dynein power stroke moves the chromosome toward the spindle equator.

aspect. Chromosome fragments produced by laser cut-
ting still congress with only a single kinetochore (Khodja-
kov et al., 1997). Similarly, in some cells, dynein inhibi-
tors apparently do not affect congression, although they
disrupt mitotic checkpoint signaling (Howell et al., 2001).
Mitotic Complexities: Nonmotor Attachments
In budding yeast where the intranuclear spindle forms
prior to centromere duplication during S phase, nonmo-
tor microtubule-associated proteins (MAPs) seem to be
primary in microtubule capture, while dynein plays a role
in spindle positioning, but not chromosome movement
(Yeh et al., 1995). The Dam1p complex (Figures 2C and
2D) interacts physically with central kinetochore pro-
teins of both the Ctf3 and Ndc80 complexes and binds
to microtubules directly in vitro (Cheeseman et al., 2001),
consistent with a direct role in mediating kinetochore-
microtubule attachments.
A group of microtubule plus-end binding proteins
(e.g., the EB1 protein family) might also be involved in
mediating interactions between microtubules and kinet-
ochores (Figure 6A). The yeast EB1 homolog BIM1 local-
izes to the plus ends of cytoplasmic microtubules and
increases dynamic instability (Tirnauer et al., 1999). Re-
moval of EB1 has demonstrated that it is important for
spindle microtubule stabilization (Rogers et al., 2002).
During congression of a chromatid pair, EB1 is found
at the ends of kinetochore bound microtubules that are
growing, but not at those that are disassembling (Tir-
nauer et al., 2002). EB1 interacts (Su et al., 1995) with
the human adenomatous polyposis coli (APC) tumor
suppressor protein (not to be confused with APC/C, the
E3 ubiquitin ligase used in controlling anaphase onset—
see below). APC also binds to and stabilizes microtu-
bules (Zumbrunn et al., 2001), localizes to the ends of
microtubules embedded in kinetochores, forms a com-
plex with mitotic checkpoint proteins Bub1 and Bub3,
and is a substrate for both of the Bub1 and BubR1
kinases in vitro (Kaplan et al., 2001). One possible expla-
nation for these observations is that EB1/APC complex
may be one of the nonmotor linker(s) that connect micro-
tubule attachment and the spindle checkpoint signaling
machinery on the kinetochore.
Anaphase Movements Using Motors and Flux
Two mechanisms for poleward chromosome movement
in anaphase, the primary function of mitosis, are now
Review
415
known: kinetochore motors and microtubule flux. That
motors are used is hardly surprising, given their contri-
bution to congression. For motors, the most obvious
candidate is dynein. Although not required for chromo-
some segregation in budding yeast (Yeh et al., 1995),
disruption of dynein clearly shows an anaphase chromo-
some segregation defect in Tetrahymena (Lee et al.,
1999) and in kinetochore- and chromatid-to-pole move-
ments in flies (Sharp et al., 2000).
Flux, however, is much less intuitive. Discovered by
photobleaching fluorescent microtubules (Gorbsky et
al., 1987) and confirmed using fluorescence photoac-
tivation of tubulin assembled into kinetochore bound
microtubules (Mitchison and Salmon, 1992), flux repre-
sents continuous addition of tubulin subunits at kineto-
chores, coupled to disassembly at the poles driven by
plus-end directed, pole bound microtubule motors pre-
sumably pulling on kinetochore microtubules and sliding
them poleward. Subunits are lost from the minus ends of
the microtubules, which must be tethered to the spindle
pole, but in a way that subunits can be disassembled.
The major mechanism for chromosome movement in
anaphase in vertebrate somatic cells is motor-powered
kinetochore movement coupled to microtubule disas-
sembly at the kinetochore (Mitchison and Salmon, 1992;
Walters et al., 1996). Here, poleward flux makes a rela-
tively minor contribution with chromosome-to-pole
movement three to eight times faster than flux. Yeast
appear to lack microtubule depolymerization at poles
and poleward microtubule flux during anaphase (Malla-
varapu et al., 1999). In Xenopus egg extracts, however,
anaphase A movement occurs at rates similar to pole-
ward spindle microtubule flux (Desai et al., 1998), consis-
tent with flux as the predominant mechanism. Else-
where, the situation is controversial: fluorescent speckle
microscopy has been used to claim a dominant role
for flux (Maddox et al., 2002) in syncytial Drosophila
embryos, while other efforts have found that dynein in-
hibitors disrupt chromatid-to-pole movement during
anaphase A (Savoian et al., 2000; Sharp et al., 2000). It
seems safe to conclude that in general, both mecha-
nisms are used to produce anaphase force at kineto-
chores.
Unattached Kinetochores as Signaling Devices
for the Mitotic Checkpoint
In order to assure accurate segregation, the mitotic
checkpoint (also known as the spindle assembly check-
point) acts to block entry into anaphase until both kineto-
chores of every duplicated chromatid pair have attached
correctly to spindle microtubules. What has emerged in
the decade since its initial description is that unattached
kinetochores and/or those not under microtubule-
induced tension are the central signaling elements that
produce a “wait anaphase” signal. A trio of cell-biologi-
cal experiments provided the primary insights that a
single, unattached kinetochore is enough to inhibit ana-
phase onset. By filming mitoses, it was initially found
that anaphase ensues about 20 min after the last kineto-
chore attaches to the spindle (Rieder et al., 1994) and
that by repeated detachment of a meiotic chromosome
from a spindle by manipulation with a microneedle de-
layed anaphase indefinitely (Li and Nicklas, 1995). That
it was an unattached kinetochore that was responsible
came from the seminal demonstration that laser ablation
of the last unattached kinetochore produces anaphase
onset within about 15 min (Rieder et al., 1995). A kineto-
chore-dependent wait anaphase signal is also sug-
gested in budding yeast: blocking centromere assembly
(by destruction of the Cbf3 component Ndc10) elimi-
nates mitotic delay in the presence of microtubule as-
sembly inhibitors (Gardner et al., 2001).
Generating a Diffusible Inhibitor
of Advance to Anaphase
Genetics in yeast initially identified seven components
of the mitotic checkpoint, Mad1–Mad3 (Mitotic arrest
defective) (Li and Murray, 1991), Bub1–Bub3 (Budding
uninhibited by benomyl) (Hoyt et al., 1991), and Mps1,
a kinase that is also essential for spindle pole body
duplication (Weiss and Winey, 1996). There are verte-
brate homologs of all of these except Bub2. Initially
demonstrated for Mad2 (Chen et al., 1996; Li and Be-
nezra, 1996), all have now been identified to bind to and
act at unattached kinetochores including Mad1 (Chen
et al., 1998), Bub1 (Taylor and McKeon, 1997), Bub3
(Taylor et al., 1998), BubR1 (the mammalian Mad3) (Tay-
lor et al., 1998), and Mps1 (Abrieu et al., 2001). Additional
required contributors without yeast counterparts are
known in metazoans. Without the kinetochore-associ-
ated microtubule motor protein CENP-E, a binding part-
ner of BubR1, the checkpoint cannot be established or
maintained in vitro (Abrieu et al., 2000) or in mice (Putkey
et al., 2002). Inhibition by mutation (Basto et al., 2000)
or antibody injection (Chan et al., 2000) of ZW10 and Rod
have revealed both to also be required for checkpoint
signaling. A final component with extensive sequence
homology to and overlapping function with Bub3 is Rae1
(Babu et al., 2003), initially identified with an additional
role in nuclear transport.
Although the nature of the direct molecular interac-
tion(s) between checkpoint proteins and kinetochores
and the interdependencies of checkpoint proteins have
not been determined (see Musacchio and Hardwick
[2002] for details), the basic plan of the signaling cas-
cade is established (Figure 7). Mad2 is recruited to unat-
tached kinetochores in a complex with Mad1 (Chen et
al., 1998). BubR1 and Bub1, both kinases, are required
for generation and then rapid release from kinetochores
of one or more inhibitors of Cdc20 (Fizzy in flies [Dawson
et al., 1993], p55 in mammals [Kallio et al., 1998] or Slp1p
in fission yeast [Kim et al., 1998]). This sequesters or
inactivates Cdc20, which is required for substrate recog-
nition by the anaphase promoting complex/cyclosome
(APC/C) (substrates include securin [Zur and Brandeis,
2001] and cyclin B [Raff et al., 2002]), whose ubiquitin-
mediated destruction by the proteosome is required for
anaphase onset (Cohen-Fix et al., 1996; Fang et al.,
1998; Zou et al., 1999).
The identity of the kinetochore-derived wait anaphase
inhibitor (or inhibitors) is not established. An activated
form of Mad2 (frequently referred to as Mad2*) has long
been thought to be the inhibitory signal released. Micro-
injection of antibodies to Mad2 first revealed premature
anaphase onset and chromosome missegregation
(Gorbsky et al., 1998). Mad2 only binds to unattached
Cell
416
kinetochores and the association at those kinetochores
is extremely dynamic, with release and rebinding to un-
attached kinetochores within 25 s (Howell et al., 2000).
Since a tetrameric form of recombinant Mad2 forms a
ternary complex with Cdc20 and APC/C in vitro, blocks
APC/C ubiquitination activity, and arrests Xenopus em-
bryos and egg extracts in mitosis (Fang et al., 1998), it
has been proposed as a candidate for the wait anaphase
inhibitor. However, no checkpoint-dependent Mad2
oligomer generation has yet been identified in vivo.
The recruitment of Mad2 at kinetochores has often
been taken as a measure of ongoing checkpoint signal
generation (and release). However, inhibition of ZW10/
Rod yields an inactive checkpoint despite prominent
Mad2 binding at kinetochores (Chan et al., 2000). Dimi-
nution of Hec1, the homolog of yeast Ndc80, on the
other hand, yields a chronically activated checkpoint
with no Mad2 bound to kinetochores (Martin-Lluesma
et al., 2002). Thus, it is now abundantly clear that the
steady-state level of Mad2 bound to kinetochores is not
a faithful reporter for checkpoint activation or inacti-
vation.
Another candidate for the wait anaphase signal is
BubR1, which directly binds Cdc20 and APC/C elements
Cdc16, Cdc27, and APC7 (Chan et al., 1999; Wu et al.,
2000) and by doing so can block Cdc20 activation of
APC/C for mitotic substrates. The inhibitory activity has
been argued to be BubR1 alone without a contribution
of Mad2 (Tang et al., 2001) or in a complex (named MCC)
that apparently contains stoichiometric amounts of
BubR1, Bub3, Mad2, and Cdc20 (Sudakin et al., 2001).
An equivalent quaternary complex containing Mad3 (the
yeast BubR1), Bub3, Mad2, and Cdc20 has also been
observed in budding yeast (Fraschini et al., 2001; Hard-
wick et al., 2000). Further, the much higher APC/C inhibi-
tory activity in vitro of MCC (⬎3,000-fold more potent
than tetrameric Mad2) (Sudakin et al., 2001) would make
this complex an attractive candidate for a diffusible in-
hibitory signal were it not that it (and its yeast counter-
part) is formed in a kinetochore-independent manner
Figure 7. Activating and Silencing the Mitotic
Checkpoint
Unattached kinetochores bind a collection of
mitotic checkpoint components. These in-
clude at least four kinases [BubR1 (Chan et
al., 1999), Bub1 (Roberts et al., 1994), Mps1
(Abrieu et al., 2001), and Mapk (Takenaka et
al., 1998)]. These and the microtubule motor
CENP-E are required for rapid recruitment of
the Mad1/Mad2 complex. Mad2 cycles rap-
idly, releasing from kinetochores in a modi-
fied form or complex that sequesters Cdc20,
thus inhibiting recognition by the ubiquitin li-
gase APC/C of substrates, including securin,
whose destruction is required for advance to
anaphase. ZW10/Rod, the kinetochore tar-
geting components for cytoplasmic dynein,
are required for generation of the inhibitory
signal, possibly because they mitigate the
rapid release of the inhibitor from kineto-
chores. After microtubule attachment, ten-
sion is generated between the bioriented ki-
netochores, checkpoint signaling is silenced,
and anaphase ensues after decay of the pre-
viously made inhibitor.
(Sudakin et al., 2001; Fraschini et al., 2001). This has led
to the suggestion that the kinetochore contribution may
modify APC/C itself to increase its affinity for MCC (Su-
dakin et al., 2001). While possible, this is unattractive
as the only kinetochore-derived inhibitor because in or-
der to keep APC/C inactive, such a model would require
a single kinetochore to be capable of rapidly recruiting,
and modifying, all cellular APC/C—a daunting task.
More plausible are models with built-in signal amplifica-
tion, for example, where a very abundant component
(like Mad2) is activated and released by kinetochores,
thereby producing a high concentration of an inhibitor
to saturate available Cdc20.
Lastly, kinetochores directly attract and may modify
the target for inhibition, Cdc20, which cycles at kineto-
chores (Kallio et al., 2002) with kinetics even faster than
reported for Mad2 (complete turnover within about 5 s).
How this relates to checkpoint signaling is made less
clear by unaltered cycling after checkpoint signal pro-
duction is silenced by microtubule attachment.
The Checkpoint Signal Has Limited Diffusibility
While the identity(ies) and mechanism of generation at
unattached kinetochores of the Cdc20-APC/C inhibi-
tor(s) remains imperfectly defined, the activated form
must be diffusible. The primary evidence for this is sim-
ple logic: in order to delay anaphase so as to prevent
loss of a single chromosome, one unattached kineto-
chore (for example, on a chromatid pair trapped by
chance behind one of the spindle poles) must be able
to inhibit APC/C-dependent destruction of securin. This
requires that the lone unattached kinetochore release a
sufficiently strong (diffusible) signal to inhibit all Cdc20-
APC/C.
But, there are strict limits to such diffusion. By analysis
of cells containing two distinct spindles (after cell fu-
sion), one or a few unattached kinetochores have been
found unable to prevent anaphase in an adjacent spindle
in which all kinetochores are attached (Rieder et al.,
1997). The limited range of the APC/C inhibition almost
Review
417
certainly reflects the limited signal strength produced
by a single unattached kinetochore, coupled to the short
natural half-life of the inhibitor(s), whose destruction is
complete within about 15 min after silencing production
(Rieder et al., 1995). In addition, after release and local
diffusion the inhibitor may be concentrated by trafficking
to spindle poles by cytoplasmic dynein (see below),
where APC/C is enriched (Tugendreich et al., 1995). This
is compatible with the finding that many checkpoint
components are rapidly trafficked poleward by dynein
after release (Howell et al., 2001).
Silencing the Checkpoint Signal:
Attachment and Tension
There is a continuing controversy as to whether the
mitotic checkpoint is silenced by microtubule attach-
ment (Rieder et al., 1995) or by the tension exerted be-
tween bioriented kinetochore pairs after attachment (Li
and Nicklas, 1995) and whether activities of subsets of
the known components are selectively silenced by one
or the other (Skoufias et al., 2001; Zhou et al., 2002).
McIntosh formally proposed the tension model under
which the mechanical tension generated by poleward-
directed forces acting on both kinetochores of a bi-
oriented chromatid pair turns off checkpoint signaling
(McIntosh, 1991). Compelling evidence for a tension re-
quirement initially emerged using praying mantid sper-
matocytes in meiosis I, which have a Y chromosome
and two genetically different X chromosomes. Although
the Y is supposed to pair with both X chromosomes,
occasionally it pairs only with one, thus yielding a mono-
oriented X. When this occurs, anaphase onset is delayed
by up to 9 hr but can be triggered to initiate almost
immediately by application of mechanical tension ap-
plied across the mono-oriented kinetochore by use of
a force-calibrated microneedle (Li and Nicklas, 1995).
This is a general finding in meiosis: eliminating tension
between homologous chromosomes by preventing re-
combination delayed anaphase onset in yeast, presum-
ably from checkpoint activation. This delay was elimi-
nated by genetically allowing sister kinetochores to
inappropriately separate during meiosis I, thereby
allowing each homolog to biorient in the absence of
recombination and restore tension across the sisters
(Shonn et al., 2000). Similarly, in cells that enter mitosis
without a prior round of DNA replication, the unrepli-
cated chromatids attach to spindle microtubules, but
no tension can be developed and the mitotic checkpoint
is chronically activated (Stern and Murray, 2001).
This is not the whole story, however. In the initial
demonstration that kinetochores are the signaling ele-
ments for the checkpoint, destruction of the last unat-
tached kinetochore eliminated checkpoint signaling de-
spite lack of tension on the kinetochore of the sister
chromatid (Rieder et al., 1995). Similarly, in maize, satis-
fying the checkpoint requires tension in meiosis, but in
mitosis, attachment is sufficient (Yu et al., 1999). Further-
more, in PtK1 cells, loss of Mad2 recruitment to kineto-
chores depends on microtubule attachment, not tension
(Waters et al., 1998). Similarly, very low doses of the
microtubule inhibitor vinblastine produce loss of tension
and kinetochore bound Mad2 and a checkpoint arrest
that, unlike the case after inhibition of spindle assembly
(Gorbsky et al., 1998), is insensitive to inactivation with
Mad2 antibody injection (Skoufias et al., 2001).
This has lead to the proposal that there are two
branches of checkpoint signaling and silencing. One, in
which the signal released is an activated, inhibitory form
of Mad2, is silenced by microtubule attachment, while
the other, presumably involving conversion of BubR1
into a Cdc20 inhibitor, is silenced by tension. This is not
unappealing, but is at best a very murky argument, since
attachment and tension are intimately interrelated. It has
been long known that tension stabilizes attachment (Ault
and Nicklas, 1989). CENP-E elimination, for example,
yields loss of kinetochore tension and fewer (McEwen
et al., 2001; Putkey et al., 2002), less stably bound micro-
tubules. Moreover, BubR1 (Mad3), Bub1, or Mad2 can-
not suppress the Cdc20-APC/C activity independently
of the others. Thus, it seems to us that rather than sepa-
rate parts of checkpoint silencing, attachment and ten-
sion really represent two halves of the same coin, inte-
grally interrelated.
Mitotic Checkpoint Defects Promote Aneuploidy
and Tumorigenesis
The mitotic checkpoint in budding yeast, where the in-
tranuclear spindle forms quickly after centromeres are
duplicated in S, is a real checkpoint activated only to
arrest mitosis in the relatively rare instances when at-
tachment is delayed. In most other organisms, it is not
a checkpoint at all. Rather, it is an essential cell cycle
control pathway activated at every mitosis/meiosis im-
mediately upon nuclear envelope disassembly. Loss of
Mad2, Bub3, or Rae1 in mice is lethal early, with cells
accumulating mitotic errors and undergoing apoptosis
by embryonic day 5 or 6 (Dobles et al., 2000; Kalitsis et
al., 2000; Babu et al., 2003). Similarly, in Drosophila,
loss of Bub1 causes chromosome missegregation and
lethality (Basu et al., 1999). Microinjection of antibodies
to Mad2 yields premature anaphase onset and chromo-
some missegregation (Gorbsky et al., 1998). Haplo-
insufficiency in Mad2 provokes late onset, self-limiting
lung tumors (Michel et al., 2001). Reduction in Bub3
or Rae1 generates aneuploidy in vitro and a sharply
increased susceptibility to chemically induced tumori-
genesis (Babu et al., 2003). Thus, the primary mission of
the checkpoint is to prevent such errors in chromosome
segregation, a hallmark of human tumor progression
(Hartwell and Kastan, 1994).
Conclusions
It is now clear that the centromere and its associated
kinetochore are much more than simple attachment
sites for spindle microtubules. Mammalian centromeres
are much more complex than initially imagined, repre-
senting repeated assemblies of the simple, one nucleo-
some centromeres of budding yeast. Central to genetic
inheritance, in almost all examples known they are them-
selves determined not by DNA sequence, but by one or
more epigenetic elements. They are active components
in microtubule capture, stabilization, and in powering
chromosome movements essential to proper segrega-
tion. More than that, they are the signaling elements for
controlling cell cycle advance through mitosis.
Cell
418
References
Abrieu, A., Kahana, J.A., Wood, K.W., and Cleveland, D.W. (2000).
CENP-E as an essential component of the mitotic checkpoint in
vitro. Cell 102, 817–826.
Abrieu, A., Magnaghi-Jaulin, L., Kahana, J.A., Peter, M., Castro, A.,
Vigneron, S., Lorca, T., Cleveland, D.W., and Labbe, J. (2001). Mps1
is a kinetochore-associated kinase essential for the vertebrate mi-
totic checkpoint. Cell 106, 83–93.
Afshar, K., Barton, N.R., Hawley, R.S., and Goldstein, L.S. (1995).
DNA binding and meiotic chromosomal localization of the Drosoph-
ila nod kinesin-like protein. Cell 81, 129–138.
Alexander, S.P., and Rieder, C.L. (1991). Chromosome motion during
attachment to the vertebrate spindle: initial saltatory-like behavior
of chromosomes and quantitative analysis of force production by
nascent kinetochore fibers. J. Cell Biol. 113, 805–815.
Ando, S., Yang, H., Nozaki, N., Okazaki, T., and Yoda, K. (2002).
CENP-A, -B, and -C chromatin complex that contains the I-type
alpha-satellite array constitutes the prekinetochore in HeLa cells.
Mol. Cell. Biol. 22, 2229–2241.
Antonio, C., Ferby, I., Wilhelm, H., Jones, M., Karsenti, E., Nebreda,
A.R., and Vernos, I. (2000). Xkid, a chromokinesin required for chro-
mosome alignment on the metaphase plate. Cell 102, 425–435.
Ault, J.G., and Nicklas, R.B. (1989). Tension, microtubule re-
arrangements, and the proper distribution of chromosomes in mito-
sis. Chromosoma 98, 33–39.
Babu, J.R., Jeganathan, K.B., Baker, D.J., Wu, X., Kang-Decker, N.,
and van Deursen, J.M. (2003). Rae1 is an essential mitotic check-
point regulator that cooperates with Bub3 to prevent chromosome
missegregation. J. Cell Biol. 160, in press. Published online January
27, 2003. 10.1083/jbc.200211048.
Basu, J., Bousbaa, H., Logarinho, E., Li, Z., Williams, B.C., Lopes, C.,
Sunkel, C.E., and Goldberg, M.L. (1999). Mutations in the essential
spindle checkpoint bub1 cause chromosome missegregation and
fail to block apoptosis in Drosophila. J. Cell Biol. 46, 13–28.
Barry, A.E., Howman, E.V., Cancilla, M.R., Saffery, R., and Choo,
K.H. (1999). Sequence analysis of an 80 kb human neocentromere.
Hum. Mol. Genet. 8, 217–227.
Basto, R., Gomes, R., and Karess, R. (2000). Rough Deal and ZW10
are required for the Metaphase checkpoint in Drosophila. Nat. Cell
Biol. 2, 939–943.
Bernard, P., Maure, J.F., Partridge, J.F., Genier, S., Javerzat, J.P.,
and Allshire, R.C. (2001). Requirement of heterochromatin for cohe-
sion at centromeres. Science 294, 2539–2542.
Biggins, S., Severin, F.F., Bhalla, N., Sassoon, I., Hyman, A.A., and
Murray, A.W. (1999). The conserved protein kinase Ipl1 regulates
microtubule binding to kinetochores in budding yease. Genes Dev.
13, 532–544.
Blower, M.D., Sullivan, B.A., and Karpen, G.H. (2002). Conserved
organization of centromeric chromatin in flies and humans. Dev.
Cell 2, 319–330.
Bolton, M.A., Lan, W., Powers, S.E., McCleland, M.L., Kuang, J.,
and Stukenberg, P.T. (2002). Aurora B kinase exists in a complex
with survivin and INCENP and its kinase activity is stimulated by
survivin binding and phosphorylation. Mol. Biol. Cell 13, 3064–3077.
Brown, K.E., Barnett, M.A., Burgtorf, C., Shaw, P., Buckle, V.J., and
Brown, W.R. (1994). Dissecting the centromere of the human Y chro-
mosome with cloned telomeric DNA. Hum. Mol. Genet. 3, 1227–1237.
Chan, G.K., Jablonski, S.A., Sudakin, V., Hittle, J.C., and Yen, T.J.
(1999). Human BUBR1 is a mitotic checkpoint kinase that monitors
CENP-E functions at kinetochores and binds the cyclosome/APC.
J. Cell Biol. 146, 941–954.
Chan, G.K., Jablonski, S.A., Starr, D.A., Goldberg, M.L., and Yen,
T.J. (2000). Human Zw10 and ROD are mitotic checkpoint proteins
that bind to kinetochores. Nat. Cell Biol. 2, 944–947.
Cheeseman, I.M., Brew, C., Wolyniak, M., Desai, A., Anderson, S.,
Muster, N., Yates, J.R., Huffaker, T.C., Drubin, D.G., and Barnes, G.
(2001). Implication of a novel multiprotein Dam1p complex in outer
kinetochore function. J. Cell Biol. 155, 1137–1145.
Cheeseman, I.M., Anderson, S., Jwa, M., Green, E.M., Kang, J.,
Yates, J.R., Chan, C.S., Drubin, D.G., and Barnes, G. (2002a). Phos-
pho-regulation of kinetochore-microtubule attachments by the au-
rora kinase ipl1p. Cell 111, 163–172.
Cheeseman, I.M., Drubin, D.G., and Barnes, G. (2002b). Simple cen-
tromere, complex kinetochore: linking spindle microtubules and
centromeric DNA in budding yeast. J. Cell Biol. 157, 199–203.
Chen, R.H., Waters, J.C., Salmon, E.D., and Murray, A.W. (1996).
Association of spindle assembly checkpoint component XMAD2
with unattached kinetochores. Science 274, 242–246.
Chen, R.H., Shevchenko, A., Mann, M., and Murray, A.W. (1998).
Spindle checkpoint protein Xmad1 recruits Xmad2 to unattached
kinetochores. J. Cell Biol. 143, 283–295.
Clarke, L. (1998). Centromeres: proteins, protein complexes, and
repeated domains at centromeres of simple eukaryotes. Curr. Opin.
Genet. Dev. 8, 212–218.
Cohen-Fix, O., Peters, J.M., Kirschner, M.W., and Koshland, D.
(1996). Anaphase initiation in Saccharomyces cerevisiae is con-
trolled by the APC-dependent degradation of the anaphase inhibitor
Pds1p. Genes Dev. 10, 3081–3093.
Dawson, I.A., Roth, S., Akam, M., and Artavanis-Tsakonas, S. (1993).
Mutations of the fizzy locus cause metaphase arrest in Drosophila
melanogaster embryos. Development 117, 359–376.
Desai, A., Maddox, P.S., Mitchison, T.J., and Salmon, E.D. (1998).
Anaphase A chromosome movement and poleward spindle microtu-
bule flux occur at similar rates in Xenopus extract spindles. J. Cell
Biol. 141, 703–713.
Desai, A., Verma, S., Mitchison, T.J., and Walczak, C.E. (1999). Kin
I kinesins are microtubule-destabilizing enzymes. Cell 96, 69–78.
Dobles, M., Liberal, V., Scott, M.L., Benezra, R., and Sorger, P.K.
(2000). Chromosome missegregation and apoptosis in mice lacking
the mitotic checkpoint Mad2. Cell 101, 635–645.
Ekwall, K., Olsson, T., Turner, B.M., Cranston, G., and Allshire, R.C.
(1997). Transient inhibition of histone deacetylation alters the struc-
tural and functional imprint at fission yeast centromeres. Cell 91,
1021–1032.
Espelin, C.W., Kaplan, K.B., and Sorger, P.K. (1997). Probing the
architecture of a simple kinetochore using DNA-protein crosslinking.
J. Cell Biol. 139, 1383–1396.
Fang, G., Yu, H., and Kirschner, M.W. (1998). The checkpoint protein
MAD2 and the mitotic regulator CDC20 form a ternary complex with
the anaphase-promoting complex to control anaphase initiation.
Genes Dev. 12, 1871–1883.
Fraschini, R., Beretta, A., Sironi, L., Musacchio, A., Lucchini, G., and
Piatti, S. (2001). Bub3 interaction with Mad2, Mad3 and Cdc20 is
mediated by WD40 repeats and does not require intact kineto-
chores. EMBO J. 20, 6648–6659.
Funabiki, H., and Murray, A.W. (2000). The Xenopus chromokinesin
Xkid is essential for metaphase chromosome alignment and must
be degraded to allow anaphase chromosome movement. Cell 102,
411–424.
Gardner, G.D., Poddar, A., Yellman, C., Tavormina, P.A., Monte-
agudo, M.C., and Burke, D.J. (2001). The spindle checkpoint of the
yeast Saccharomyces cerevisiae requires kinetochore function and
maps to the CBF3 domain. Genetics 157, 1493–1502.
Gorbsky, G.J., Sammak, P.J., and Borisy, G.G. (1987). Chromosomes
move poleward in anaphase along stationary microtubules that co-
ordinately disassemble from their kinetochore ends. J. Cell Biol.
104, 9–18.
Gorbsky, G.J., Chen, R.H., and Murray, A.W. (1998). Microinjection
of antibody to Mad2 protein into mammalian cells in mitosis induces
premature anaphase. J. Cell Biol. 141, 1193–1205.
Goshima, G., Saitoh, S., and Yanagida, M. (1999). Proper metaphase
spindle length is determined by centromere proteins Mis12 and
Mis6 required for faithful chromosome segregation. Genes Dev. 13,
1664–1677.
Hardwick, K.G., Johnston, R.C., Smith, D.L., and Murray, A.W. (2000).
MAD3 encodes a novel component of the spindle checkpoint which
interacts with Bub3p, Cdc20p, and Mad2p. J. Cell Biol. 148, 871–882.
Review
419
Harrington, J.A., Van Bokkelen, G., Mays, R.W., Gustashaw, K., and
Willard, H.F. (1997). Formation of de novo centromeres and con-
struction of first-generation human artificial microchromosomes.
Nat. Genet. 15, 345–355.
Hartwell, L.H., and Kastan, M.B. (1994). Cell cycle control and can-
cer. Science 266, 1821–1828.
Hays, T.S., Wise, D., and Salmon, E.D. (1982). Traction force on a
kinetochore at metaphase acts as a linear function of kinetochore
fiber length. J. Cell Biol. 93, 374–389.
He, D., Zeng, C., Woods, K., Zhong, L., Turner, D., Busch, R.K.,
Brinkley, B.R., and Busch, H. (1998). CENP-G: a new centromeric
protein that is associated with the alpha-1 satellite DNA subfamily.
Chromosoma 107, 189–197.
He, X., Asthana, S., and Sorger, P.K. (2000). Transient sister chroma-
tid separation and elastic deformation of chromosomes during mito-
sis in budding yeast. Cell 101, 763–775.
Hemmerich, P., Stoyan, T., Wieland, G., Koch, M., Lechner, J., and
Diekmann, S. (2000). Interaction of yeast kinetochore proteins with
centromere-protein/transcription factor Cbf1. Proc. Natl. Acad. Sci.
USA 97, 12583–12588.
Henikoff, S., Ahmad, K., and Malik, H.S. (2001). The centromere
paradox: stable inheritance with rapidly evolving DNA. Science 293,
1098–1102.
Howell, B.J., Hoffman, D.B., Fang, G., Murray, A.W., and Salmon,
E.D. (2000). Visualization of mad2 dynamics at kinetochores, along
spindles fibers, and at spindle poles in living cells. J. Cell Biol. 150,
1233–1250.
Howell, B.J., McEwen, B.F., Canman, J.C., Hoffman, D.B., Farrar,
E.M., Rieder, C.L., and Salmon, E.D. (2001). Cytoplasmic dynein/
dynactin drives kinetochore protein transport to the spindle poles
and has a role in mitotic spindle checkpoint inactivation. J. Cell Biol.
155, 1159–1172.
Howman, E.V., Fowler, K.J., Newson, A.J., Redward, S., MacDonald,
A.C., Kalitsis, P., and Choo, K.H. (2000). Early disruption of centro-
meric chromatin organization in centromere protein A (Cenpa) null
mice. Proc. Natl. Acad. Sci. USA 97, 1148–1153.
Hoyt, M.A., Totis, L., and Roberts, B.T. (1991). S. cerevisiae genes
required for cell cycle arrest in response to loss of microtubule
function. Cell 66, 507–517.
Hudson, D.F., Fowler, K.J., Earle, E., Saffery, R., Kalitsis, P., Trowell,
H., Hill, J., Wreford, N.G., de Kretser, D.M., Cancilla, M.R., et al.
(1998). Centromere protein B null mice are mitotically and meiotically
normal but have lower body and testis weights. J. Cell Biol. 141,
309–319.
Hyman, A.A., and Mitchison, T.J. (1991). Two different microtubule-
based motor activities with opposite polarities in kinetochores. Na-
ture 351, 206–211.
Ikeno, M., Grimes, B., Okazaki, T., Nakano, M., Saitoh, K., Hoshino,
H., McGill, N.I., Cooke, H., and Masumoto, H. (1998). Construction
of YAC-based mammalian artificial chromosomes. Nat. Biotechnol.
16, 431–439.
Janke, C., Ortiz, J., Lechner, J., Shevchenko, A., Shevchenko, A.,
Magiera, M.M., Schramm, C., and Schiebel, E. (2001). The budding
yeast proteins Spc24p and Spc25p interact with Ndc80p and Nuf2p
at the kinetochore and are important for kinetochore clustering and
checkpoint control. EMBO J. 20, 777–791.
Janke, C., Ortiz, J., Tanaka, T.U., Lechner, J., and Schiebel, E. (2002).
Four new subunits of the Dam1-Duo1 complex reveal novel func-
tions in sister kinetochore biorientation. EMBO J. 21, 181–193.
Kalitsis, P., Earle, E., Fowler, K.J., and Choo, K.H. (2000). Bub3
gene disruption in mice reveals essential mitotic spindle checkpoint
function during early embryogenesis. Genes Dev. 14, 2277–2282.
Kallio, M., Weinstein, J., Daum, J.R., Burke, D.J., and Gorbsky, G.J.
(1998). Mammalian p55CDC mediates association of the spindle
checkpoint protein Mad2 with the cyclosome/anaphase-promoting
complex, and is involved in regulating anaphase onset and late
mitotic events. J. Cell Biol. 141, 1393–1406.
Kallio, M.J., Beardmore, V.A., Weinstein, J., and Gorbsky, G.J.
(2002). Rapid microtubule-independent dynamics of Cdc20 at kinet-
ochores and centrosomes in mammalian cells. J. Cell Biol. 158,
841–847.
Kaplan, K.B., Burds, A.A., Swedlow, J.R., Bekir, S.S., Sorger, P.K.,
and Nathke, I.S. (2001). A role for the Adenomatous Polyposis Coli
protein in chromosome segregation. Nat. Cell Biol. 3, 429–432.
Keith, K.C., and Fitzgerald-Hayes, M. (2000). CSE4 genetically inter-
acts with the Saccharomyces cerevisiae centromere DNA elements
CDE I and CDE II but not CDE III. Implications for the path of the
centromere dna around a cse4p variant nucleosome. Genetics 156,
973–981.
Khodjakov, A., and Rieder, C.L. (1996). Kinetochores moving away
from their associated pole do not exert a significant pushing force
on the chromosome. J. Cell Biol. 135, 315–327.
Khodjakov, A., Cole, R.W., McEwen, B.F., Buttle, K.F., and Rieder,
C.L. (1997). Chromosome fragments possessing only one kineto-
chore can congress to the spindle equator. J. Cell Biol. 136, 229–240.
Kim, S.H., Lin, D.P., Matsumoto, S., Kitazono, A., and Matsumoto,
T. (1998). Fission yeast Slp1: an effector of the Mad2-dependent
spindle checkpoint. Science 279, 1045–1047.
Lachner, M., O’Carroll, D., Rea, S., Mechtler, K., and Jenuwein, T.
(2001). Methylation of histone H3 lysine 9 creates a binding site for
HP1 proteins. Nature 410, 116–120.
Laloraya, S., Guacci, V., and Koshland, D. (2000). Chromosomal
addresses of the cohesin component Mcd1p. J. Cell Biol. 151, 1047–
1056.
Lee, S., Wisniewski, J.C., Dentler, W.L., and Asai, D.J. (1999). Gene
knockouts reveal separate functions for two cytoplasmic dyneins
in Tetrahymena thermophila. Mol. Biol. Cell 10, 771–784.
Levesque, A.A., and Compton, D.A. (2001). The chromokinesin Kid
is necessary for chromosome arm orientation and oscillation, but
not congression, on mitotic spindles. J. Cell Biol. 154, 1135–1146.
Li, R., and Murray, A.W. (1991). Feedback control of mitosis in bud-
ding yeast. Cell 66, 519–531.
Li, X., and Nicklas, R.B. (1995). Mitotic forces control a cell-cycle
checkpoint. Nature 373, 630–632.
Li, Y., and Benezra, R. (1996). Identification of a human mitotic
checkpoint gene: hsMAD2. Science 274, 246–248.
Li, Y., Bachant, J., Alcasabas, A.A., Wang, Y., Qin, J., and Elledge,
S.J. (2002). The mitotic spindle is required for loading of the DASH
complex onto the kinetochore. Genes Dev. 16, 183–197.
Lo, A.W., Magliano, D.J., Sibson, M.C., Kalitsis, P., Craig, J.M., and
Choo, K.H. (2001). A novel chromatin immunoprecipitation and array
(CIA) analysis identifies a 460-kb CENP-A-binding neocentromere
DNA. Genome Res. 11, 448–457.
Lombillo, V.A., Nislow, C., Yen, T.J., Gelfand, V.I., and McIntosh,
J.R. (1995). Antibodies to the kinesin motor domain and CENP-E
inhibit microtubule depolymerization-dependent motion of chromo-
somes in vitro. J. Cell Biol. 128, 107–115.
Luger, K., Mader, A.W., Richmond, R.K., Sargent, D.F., and Rich-
mond, T.J. (1997). Crystal structure of the nucleosome core particle
at 2.8 Å resolution. Nature 389, 251–260.
Maddox, P., Desai, A., Oegema, K., Mitchison, T.J., and Salmon,
E.D. (2002). Poleward microtubule flux is a major component of
spindle dynamics and anaphase a in mitotic Drosophila embryos.
Curr. Biol. 12, 1670–1674.
Mallavarapu, A., Sawin, K., and Mitchison, T. (1999). A switch in
microtubule dynamics at the onset of anaphase B in the mitotic
spindle of Schizosaccharomyces pombe. Curr. Biol. 9, 1423–1426.
Martin-Lluesma, S., Stucke, V.M., and Nigg, E.A. (2002). Role of
hec1 in spindle checkpoint signaling and kinetochore recruitment
of mad1/mad2. Science 297, 2267–2270.
McEwen, B.F., Chan, G.K.T., Zubrowski, B., Savoian, M.S., Sauer,
M.T., and Yen, T.J. (2001). CENP-E is essential for reliable bioriented
spindle attachment, but chromosome alignment can be achieved
via redundant mechanisms in mammalian cells. Mol. Biol. Cell 12,
2776–2789.
McIntosh, J.R. (1991). Structural and mechanical control of mitotic
progression. Cold Spring Harb. Symp. Quant. Biol. 56, 613–619.
Cell
420
Megee, P.C., Mistrot, C., Guacci, V., and Koshland, D. (1999). The
centromeric sister chromatid cohesion site directs Mcd1p binding
to adjacent sequences. Mol. Cell 4, 445–450.
Meluh, P.B., and Koshland, D. (1995). Evidence that the MIF2 gene
of Saccharomyces cerevisiae encodes a centromere protein with
homology to the mammalian centromere protein CENP-C. Mol. Biol.
Cell 6, 793–807.
Meluh, P.B., Yang, P., Glowczewski, L., Koshland, D., and Smith,
M.M. (1998). Cse4p is a component of the core centromere of Sac-
charomyces cerevisiae. Cell 94, 607–613.
Michel, L.S., Liberal, V., Chatterjee, A., Kirchwegger, R., Pasche, B.,
Gerald, W., Dobles, M., Sorger, P.K., Murty, V.V., and Benezra, R.
(2001). Mad2 haplo-insufficiency causes premature anaphase and
chromosome instability in mammalian cells. Nature 409, 355–359.
Mitchison, T.J., and Salmon, E.D. (1992). Poleward kinetochore fiber
movement occurs during both metaphase and anaphase-A in newt
lung cell mitosis. J. Cell Biol. 119, 569–582.
Musacchio, A., and Hardwick, K.G. (2002). The spindle checkpoint:
structural insights into dynamic signalling. Nat. Rev. Mol. Cell Biol.
3, 731–741.
Nakagawa, H., Lee, J.K., Hurwitz, J., Allshire, R.C., Nakayama, J.,
Grewal, S.I., Tanaka, K., and Murakami, Y. (2002). Fission yeast
CENP-B homologs nucleate centromeric heterochromatin by pro-
moting heterochromatin-specific histone tail modifications. Genes
Dev. 16, 1766–1778.
Ngan, V.K., and Clarke, L. (1997). The centromere enhancer mediates
centromere activation in Schizosaccharomyces pombe. Mol. Cell.
Biol. 17, 3305–3314.
Nicklas, R.B. (1983). Measurements of the force produced by the
mitotic spindle in anaphase. J. Cell Biol. 97, 542–548.
Nicklas, R.B. (1988). The forces that move chromosomes in mitosis.
Annu. Rev. Biophys. Biophys. Chem. 17, 431–449.
Oegema, K., Desai, A., Rybina, S., Kirkham, M., and Hyman, A.A.
(2001). Functional analysis of kinetochore assembly in Caenorhab-
ditis elegans. J. Cell Biol. 153, 1209–1226.
Ohzeki, J., Nakano, M., Okada, T., and Masumoto, H. (2002). CENP-B
box is required for de novo centromere chromatin assembly on
human alphoid DNA. J. Cell Biol. 159, 765–775.
Ortiz, J., Stemmann, O., Rank, S., and Lechner, J. (1999). A putative
protein complex consisting of Ctf19, Mcm21, and Okp1 represents
a missing link in the budding yeast kinetochore. Genes Dev. 13,
1140–1155.
Ostergren, G. (1951). The mechanism of co-orientation in bivalents
and multivalents. Hereditas 37, 85–156.
Partridge, J.F., Borgstrom, B., and Allshire, R.C. (2000). Distinct
protein interaction domains and protein spreading in a complex
centromere. Genes Dev. 14, 783–791.
Partridge, J., Scott, K., Bannister, A., Kouzarides, T., and Allshire,
R. (2002). cis-acting DNA from fission yeast centromeres mediates
histone H3 methylation and recruitment of silencing factors and
cohesin to an ectopic site. Curr. Biol. 12, 1652–1660.
Putkey, F.R., Cramer, T., Morphew, M.K., Silk, A.D., Johnson, R.S.,
McIntosh, J.R., and Cleveland, D.W. (2002). Unstable kinetochore-
microtubule capture and chromosomal instability following deletion
of CENP-E. Dev. Cell 3, 351–365.
Raff, J.W., Jeffers, K., and Huang, J.Y. (2002). The roles of Fzy/
Cdc20 and Fzr/Cdh1 in regulating the destruction of cyclin B in
space and time. J. Cell Biol. 157, 1139–1149.
Richards, E.J., and Elgin, S.C. (2002). Epigenetic codes for hetero-
chromatin formation and silencing: rounding up the usual suspects.
Cell 108, 489–500.
Rieder, C.L. (1981). The structure of the cold-stable kinetochore
fiber in metaphase PtK1 cells. Chromosoma 84, 145–158.
Rieder, C.L., and Alexander, S.P. (1990). Kinetochores are trans-
ported poleward along a single astral microtubule during chromo-
some attachment to the spindle in newt lung cells. J. Cell Biol. 110,
81–95.
Rieder, C.L., Schultz, A., Cole, R., and Sluder, G. (1994). Anaphase
onset in vertebrate somatic cells is controlled by a checkpoint that
monitors sister kinetochore attachment to the spindle. J. Cell Biol.
127, 1301–1310.
Rieder, C.L., Cole, R.W., Khodjakov, A., and Sluder, G. (1995). The
checkpoint delaying anaphase in response to chromosome mono-
orientation is mediated by an inhibitory signal produced by unat-
tached kinetochores. J. Cell Biol. 130, 941–948.
Rieder, C.L., Khodjakov, A., Paliulis, L.V., Fortier, T.M., Cole, R.W.,
and Sluder, G. (1997). Mitosis in vertebrate somatic cells with two
spindles: implications for the metaphase/anaphase transition
checkpoint and cleavage. Proc. Natl. Acad. Sci. USA 94, 5107–5112.
Roberts, B.T., Farr, K.A., and Hoyt, M.A. (1994). The Saccharomyces
cerevisiae checkpoint gene BUB1 encodes a novel protein kinase.
Mol. Cell. Biol. 14, 8282–8291.
Rogers, S.L., Rogers, G.C., Sharp, D.J., and Vale, R.D. (2002). Dro-
sophila EB1 is important for proper assembly, dynamics, and posi-
tioning of the mitotic spindle. J. Cell Biol. 158, 873–884.
Russell, I.D., Grancell, A.S., and Sorger, P.K. (1999). The unstable
F-box protein p58-Ctf13 forms the structural core of the CBF3 kinet-
ochore complex. J. Cell Biol. 145, 933–950.
Saffery, R., Irvine, D.V., Griffiths, B., Kalitsis, P., Wordeman, L., and
Choo, K.H. (2000). Human centromeres and neocentromeres show
identical distribution patterns of ⬎20 functionally important kineto-
chore-associated proteins. Hum. Mol. Genet. 9, 175–185.
Savoian, M.S., Goldberg, M.L., and Rieder, C.L. (2000). The rate of
chromosome poleward motion is attenuated in Drosophila zw10 and
rod mutants. Nat. Cell Biol. 2, 948–952.
Schueler, M.G., Higgins, A.W., Rudd, M.K., Gustashaw, K., and Wil-
lard, H.F. (2001). Genomic and genetic definition of a functional
human centromere. Science 294, 109–115.
Sharp, D.J., Rogers, G.C., and Scholey, J.M. (2000). Cytoplasmic
dynein is required for poleward chromosome movement during mi-
tosis in Drosophila embryos. Nat. Cell Biol. 2, 922–930.
Shelby, R.D., Vafa, O., and Sullivan, K.F. (1997). Assembly of CENP-A
into centromeric chromatin requires a cooperative array of nucleo-
somal DNA contact sites. J. Cell Biol. 136, 501–513.
Shelby, R.D., Monier, K., and Sullivan, K.F. (2000). Chromatin assem-
bly at kinetochores is uncoupled from DNA replication. J. Cell Biol.
151, 1113–1118.
Shingyoji, C., Higuchi, H., Yoshimura, M., Katayama, E., and Yanag-
ida, T. (1998). Dynein arms are oscillating force generators. Nature
393, 711–714.
Shonn, M.A., McCarroll, R., and Murray, A.W. (2000). Requirement
of the spindle checkpoint for proper chromosome segregation in
budding yeast meiosis. Science 289, 300–303.
Skibbens, R.V., Skeen, V.P., and Salmon, E.D. (1993). Directional
instability of kinetochore motility during chromosome congression
and segregation in mitotic newt lung cells: a push-pull mechanism.
J. Cell Biol. 122, 859–875.
Skoufias, D.A., Andreassen, P.R., Lacroix, F.B., Wilson, L., and Mar-
golis, R.L. (2001). Mammalian mad2 and bub1/bubR1 recognize dis-
tinct spindle-attachment and kinetochore-tension checkpoints.
Proc. Natl. Acad. Sci. USA 98, 4492–4497.
Smith, M.M. (2002). Centromeres and variant histones: what, where,
when and why? Curr. Opin. Cell Biol. 14, 279–285.
Stern, B.M., and Murray, A.W. (2001). Lack of tension at kinetochores
activates the spindle checkpoint in budding yeast. Curr. Biol. 11,
1462–1467.
Stoler, S., Keith, K.C., Curnick, K.E., and Fitzgerald-Hayes, M. (1995).
A mutation in CSE4, an essential gene encoding a novel chromatin-
associated protein in yeast, causes chromosome nondisjunction
and cell cycle arrest at mitosis. Genes Dev. 9, 573–586.
Su, L.K., Burrell, M., Hill, D.E., Gyuris, J., Brent, R., Wiltshire, R.,
Trent, J., Vogelstein, B., and Kinzler, K.W. (1995). APC binds to the
novel protein EB1. Cancer Res. 55, 2972–2977.
Sudakin, V., Chan, G.K., and Yen, T.J. (2001). Checkpoint inhibition
of the APC/C in HeLa cells is mediated by a complex of BUBR1,
BUB3, CDC20, and MAD2. J. Cell Biol. 154, 925–936.
Sullivan, B., and Karpen, G. (2001). Centromere identity in Drosophila
Review
421

is not determined in vivo by replication timing. J. Cell Biol. 154, Waters, J.C., Chen, R.H., Murray, A.W., and Salmon, E.D. (1998).
683–690. Localization of Mad2 to kinetochores depends on microtubule at-
Sullivan, B.A., and Willard, H.F. (1998). Stable dicentric X chromo- tachment, not tension. J. Cell Biol. 141, 1181–1191.
somes with two functional centromeres. Nat. Genet. 20, 227–228. Weiss, E., and Winey, M. (1996). The Saccharomyces cerevisiae
Sullivan, B.A., Blower, M.D., and Karpen, G.H. (2001). Determining spindle pole body duplication gene MPS1 is part of a mitotic check-
centromere identity: cyclical stories and forking paths. Nat. Rev. point. J. Cell Biol. 132, 111–123.
Genet. 2, 584–596. Wigge, P.A., and Kilmartin, J.V. (2001). The Ndc80p complex from
Sun, X., Wahlstrom, J., and Karpen, G.H. (1997). Molecular structure Saccharomyces cerevisiae contains conserved centromere compo-
of a functional Drosophila centromere. Cell 91, 1007–1019. nents and has a function in chromosome segregation. J. Cell Biol.
152, 349–360.
Takahashi, K., Chen, E.S., and Yanagida, M. (2000). Requirement of
Mis6 centromere connector for localizing a CENP-A-like protein in Williams, B.C., Gatti, M., and Goldberg, M.L. (1996). Bipolar spindle
fission yeast. Science 288, 2215–2219. attachments affect redistributions of ZW10, a Drosophila centro-
mere/kinetochore component required for accurate chromosome
Takenaka, K., Moriguchi, T., and Nishida, E. (1998). Activation of the
segregation. J. Cell Biol. 134, 1127–1140.
protein kinase p38 in the spindle assembly checkpoint and mitotic
arrest. Science 280, 599–602. Williams, B.C., Murphy, T.D., Goldberg, M.L., and Karpen, G.H.
(1998). Neocentromere activity of structurally acentric mini-chromo-
Tanaka, T., Cosma, M.P., Wirth, K., and Nasmyth, K. (1999). Identifi-
somes in Drosophila. Nat. Genet. 18, 30–37.
cation of cohesin association sites at centromeres and along chro-
mosome arms. Cell 98, 847–858. Wood, K.W., Sakowicz, R., Goldstein, L.S., and Cleveland, D.W.
(1997). CENP-E is a plus end-directed kinetochore motor required
Tanaka, T.U., Rachidi, N., Janke, C., Pereira, G., Galova, M., Schie-
for metaphase chromosome alignment. Cell 91, 357–366.
bel, E., Stark, M.J., and Nasmyth, K. (2002). Evidence that the Ipl1-
Sli15 (Aurora kinase-INCENP) complex promotes chromosome bi- Wordeman, L., and Mitchison, T.J. (1995). Identification and partial
orientation by altering kinetochore-spindle pole connections. Cell characterization of mitotic centromere-associated kinesin, a
108, 317–329. kinesin-related protein that associates with centromeres during mi-
tosis. J. Cell Biol. 128, 95–104.
Tang, Z., Bharadwaj, R., Li, B., and Yu, H. (2001). Mad2-independent
inhibition of APCcdc20 by the mitotic checkpoint protein Bubr1. Dev. Wu, H., Lan, Z., Li, W., Wu, S., Weinstein, J., Sakamoto, K.M., and
Cell 1, 227–237. Dai, W. (2000). p55CDC/hCDC20 is associated with BUBR1 and may
be a downstream target of the spindle checkpoint kinase. Oncogene
Taylor, S.S., and McKeon, F. (1997). Kinetochore localization of mu-
19, 4557–4562.
rine Bub1 is required for normal mitotic timing and checkpoint re-
sponse to spindle damage. Cell 89, 727–735. 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
Taylor, S.S., Ha, E., and McKeon, F. (1998). The human homologue
budding yeast, Saccharomyces cerevisiae. J. Cell Biol. 130,
of Bub3 is required for kinetochore localization of Bub1 and a Mad3/
687–700.
Bub1-related protein kinase. J. Cell Biol. 142, 1–11.
Yu, H.G., Muszynski, M.G., and Kelly Dawe, R. (1999). The maize
Tirnauer, J.S., O’Toole, E., Berrueta, L., Bierer, B.E., and Pellman,
homologue of the cell cycle checkpoint protein MAD2 reveals kinet-
D. (1999). Yeast Bim1p promotes the G1-specific dynamics of micro-
ochore substructure and contrasting mitotic and meiotic localization
tubules. J. Cell Biol. 145, 993–1007.
patterns. J. Cell Biol. 145, 425–435.
Tirnauer, J.S., Canman, J.C., Salmon, E.D., and Mitchison, T.J.
Zhou, J., Panda, D., Landen, J.W., Wilson, L., and Joshi, H.C. (2002).
(2002). EB1 targets to kinetochores with attached, polymerizing mi-
Minor alteration of microtubule dynamics causes loss of tension
crotubules. Mol. Biol. Cell 13, 4308–4316.
across kinetochore pairs and activates the spindle checkpoint. J.
Tomonaga, T., Nagao, K., Kawasaki, Y., Furuya, K., Murakami, A., Biol. Chem. 277, 17200–17208.
Morishita, J., Yuasa, T., Sutani, T., Kearsey, S.E., Uhlmann, F., et al.
Zinkowski, R.P., Meyne, J., and Brinkley, B.R. (1991). The centro-
(2000). Characterization of fission yeast cohesin: essential anaphase
mere-kinetochore complex: a repeat subunit model. J. Cell Biol.
proteolysis of rad21 phosphorylated in the S phase. Genes Dev. 14,
113, 1091–1110.
2757–2770.
Zou, H., McGarry, T.J., Bernal, T., and Kirschner, M.W. (1999). Identi-
Tugendreich, S., Tomkiel, J., Earnshaw, W., and Hieter, P. (1995).
fication of a vertebrate sister-chromatid separation inhibitor in-
CDC27Hs colocalizes with CDC16Hs to the centrosome and mitotic
volved in transformation and tumorigenesis. Science 285, 418–422.
spindle and is essential for the metaphase to anaphase transition.
Cell 81, 261–268. Zumbrunn, J., Inoshita, K., Hyman, A.A., and Nathke, I.S. (2001).
Binding of the Adenomatous Polyposis Coli protein to microtubules
Tyler-Smith, C., Oakey, R.J., Larin, Z., Fisher, R.B., Crocker, M.,
increases microtubules stability and is regulated by GSK3b phos-
Affara, N.A., Ferguson-Smith, M.A., Muenke, M., Zuffardi, O., and
phorylation. Curr. Biol. 11, 44–49.
Jobling, M.A. (1993). Localization of DNA sequences required for
Zur, A., and Brandeis, M. (2001). Securin degradation is mediated
human centromere function through an analysis of rearranged Y
chromosomes. Nat. Genet. 5, 368–375. by fzy and fzr, and is required for complete chromatid separation
but not for cytokinesis. EMBO J. 20, 792–801.
Tyler-Smith, C., Gimelli, G., Giglio, S., Floridia, G., Pandya, A., Terzoli,
G., Warburton, P.E., Earnshaw, W.C., and Zuffardi, O. (1999). Trans-
mission of a fully functional human neocentromere through three
generations. Am. J. Hum. Genet. 64, 1440–1444.
Vafa, O., and Sullivan, K.F. (1997). Chromatin containing CENP-A
and alpha-satellite DNA is a major component of the inner kineto-
chore plate. Curr. Biol. 7, 897–900.
Volpe, T.A., Kidner, C., Hall, I.M., Teng, G., Grewal, S.I., and Martiens-
sen, R.A. (2002). Regulation of heterochromatic silencing and his-
tone H3 lysine-9 methylation by RNAi. Science 297, 1833–1837.
Walczak, C.E., Mitchison, T.J., and Desai, A. (1996). XKCM1: a Xeno-
pus kinesin-related protein that regulates microtubule dynamics
during mitotic spindle assembly. Cell 84, 37–47.
Walters, J.C., Mitchison, T.J., Rieder, C.L., and Salmon, E.D. (1996).
The kinetochore microtubule minus-end disassembly associated
with poleward flux produces a force that can do work. Mol. Biol.
Cell 7, 1547–1558.