Immunocytochemistry demonstrates that CcUbc9, CcTopII and CcPCNA localize with CcLim15 in meiotic nuclei during leptotene to zygotene when synaptonemal complex is formed and when homolog
Trang 1Meiosis and small ubiquitin-related modifier
(SUMO)-conjugating enzyme, Ubc9
Kengo Sakaguchi, Akiyo Koshiyama and Kazuki Iwabata
Department of Applied Biological Science, Faculty of Science and Technology, Tokyo University of Science, Chiba, Japan
Introduction
Small ubiquitin-related modifier (SUMO) modification,
known as sumoylation is a post-translational protein
modification like ubiquitination, and appears to play important roles in many diverse processes [1–13] SUMO family proteins and ubiquitin are similar in terms of both structure and the enzymatic reactions
Keywords
DNA polymerase; Lim15 ⁄ Dmc1; meiosis;
PCNA; Rad51; SUMO; sumoylation;
topoisomerase II; Ubc9
Correspondence
K Sakaguchi, Department of Applied
Biological Science, Faculty of Science and
Technology, Tokyo University of Science,
2641 Yamazaki, Noda-shi, Chiba-ken 278,
Japan
Fax: +81 471 23 9767
Tel: +81 471 24 1501 (ext 3409)
E-mail: kengo@rs.noda.tus.ac.jp
(Received 16 March 2007, revised 22 May
2007, accepted 30 May 2007)
doi:10.1111/j.1742-4658.2007.05905.x
In this review, we describe the role of a small ubiquitin-like protein modi-fier (SUMO)-conjugating protein, Ubc9, in synaptonemal complex forma-tion during meiosis in a basidiomycete, Coprinus cinereus Because its meiotic cell cycle is long and naturally synchronous, it is suitable for molecular biological, biochemical and genetic studies of meiotic prophase events In yeast two-hybrid screening using the meiotic-specific cDNA lib-rary of C cinereus, we found that the meiotic RecA homolog CcLim15 interacted with CcUbc9, CcTopII and CcPCNA Moreover, both TopII and PCNA homologs were known as Ubc9 interactors and the targets of sumoylation Immunocytochemistry demonstrates that CcUbc9, CcTopII and CcPCNA localize with CcLim15 in meiotic nuclei during leptotene to zygotene when synaptonemal complex is formed and when homologous chromosomes pair We discuss the relationships between Lim15⁄ Dmc1 (CcLim15), TopII (CcTopII), PCNA (CcPCNA) and CcUbc9, and subse-quently, the role of sumoylation in the stages We speculate that CcLim15 and CcTopII work in cohesion between homologous chromatins initially and then, in the process of the zygotene events, CcUbc9 works with factors including CcLim15 and CcTopII as an inhibitor of ubiquitin-mediated deg-radation and as a metabolic switch in the meiotic prophase cell cycle After CcLim15–CcTopII dissociation, CcLim15 remains on the zygotene DNA and recruits CcUbc9, Rad54B, CcUbc9, Swi5-Sfr1, CcUbc9 and then CcPCNA in rotation on the C-terminus Finally during zygotene, CcPCNA replaces CcLim15 on the DNA and the free-CcLim15 is probably ubiquiti-nated and disappears CcPCNA may recruit the polymerase The idea that CcUbc9 intervenes in every step by protecting CcLim15 and by switching several factors at the C-terminus of CcLim15 is likely At the boundary
of the zygotene and pachytene stages, CcPCNA would be sumoylated CcUbc9 may also be involved with CcPCNA in the switch from the repli-cative polymerase being recruited at zygotene to the repair-type DNA polymerases being recruited at pachytene
Abbreviations
DSB, double-strand break; SC, synaptonemal complex; SSB, single-stranded break; SUMO, small ubiquitin-related modifier.
Trang 2underlying their conjugation [10,11] Furthermore, the
enzymes involved in SUMO conjugation have
sequences with similarities to their counterparts that
mediate ubiquitin conjugation [5] However, both
sum-oylation and ubiquitination have distinct
nonoverlap-ping functions [1–13]
The functions of sumoylation depend on the target
proteins A comprehensive survey of sumoylated
pro-teins was recently performed by Wykoff and O’Shea
[14] Utilizing a collection of epitope-tagged yeast
strains and immunoprecipitation of a large fraction of
the proteome, they developed a novel approach for the
identification of sumoylated proteins At least 82
pro-teins were found to be candidate SUMO targets,
inclu-ding many of low abundance Based on their results, it
is noteworthy that it is not only control processes of
chromosome segregation and cell division, DNA
repli-cation and repair, nuclear protein import, protein
targeting to and formation of certain subnuclear
struc-tures that involve sumoylation, but processes involved
in the mammalian inflammatory response and plant
flowering time have also been described as involving
this protein modification
Since the discovery of SUMO about 10 years ago
many excellent reviews with detailed discussions of
SUMO research have been published [1–13] To
bor-row Ju¨rgen Dohmen’s phrase [10], ‘these reviews
attempt to summarize the current status of the rapidly
increasing knowledge of the mechanisms and functions
of SUMO systems in various eukaryotic model
organ-isms with an emphasis on the enzymes mediating
SUMO conjugation and deconjugation A few
insight-ful examples point to one mode in which sumoylation
is antagonistic to ubiquitination for some substrates,
and to another mode in which sumoylation is either
required for protein interaction or inhibitory to it.’ On
the whole, the situation is similar in 2007 Because the
roles of sumoylation are so many, it is difficult to
pre-sent a summary of the whole field in anything other
than a partly chaotic manner Therefore, we would like
to summarize one aspect of the field, namely meiosis
and sumoylation, because we recently found
sumoyla-tion of a meiosis-specific RecA homolog, Lim15⁄
Dmc1, via interaction with the SUMO-conjugating
enzyme Ubc9 at a particular stage of meiosis [15]
Lim15⁄ Dmc1 is a most important key protein in the
meiotic cell cycle, particularly during the stages when
homologous chromosomes pair and recombine
Meiosis
In meiosis, as is well known, homologous
chromo-somes are paired and recombined during meiotic
pro-phase I (also called synapsis) and then segregated into tetrads [16] Prophase I is divided into five stages, namely leptotene, zygotene, pachytene, diplotene and diakinesis Chromosomes condense from the dispersed state typical of interphase during early meiotic pro-phase (leptotene) to form long thin threads and each acquires a proteinaceous axial core to which the two sister chromatids are attached Then, homologous chromosomes become aligned during zygotene and form the synaptonemal complex (SC), a proteinaceous framework assembled between homologous chromo-somes, and required for the subsequent maintenance of synapses SC polymerization ensures continuous and stable association along the homologous chromosomes throughout pachytene, during which time completion
of reciprocal strand-exchange events takes place [16]
At pachytene, nonsister chromatids of the completely paired chromosomes recombine by forming chiasmata which become visible during diplotene This is followed
by two cell divisions, namely reductional segregation
of homologous chromosomes and equational segrega-tion of sister chromatids, resulting in four gametes
In Saccharomyces cerevisiae, SC polymerization initi-ates at sites undergoing meiotic recombination and requires the activities of an enzyme induced by double-strand breaks (DSBs) and double-strand-exchange proteins [17] It should also be noted that the zygotene and pachytene stages, which are the most important pro-phase stages when homologous chromosomes pair and recombine, tend to be intermixed By contrast, in higher plants and mammals, the SC forms exactly at zygotene, and when this is finished recombination begins at pachytene, as ascertained by cytogenetic research in the 20th century The end of SC formation must be an initiation signal of recombination [16] Moreover, in studies of meiosis using higher plant (lily) and mouse spermatocytes, the initiation of pachytene DNA recombination was shown to be related to single-stranded breaks (SSBs) rather than DSBs [16,18–20] Since the 1980s, there have been few new insights into the role of SSBs in pachytene DNA recombination However, meiotic recombination by nicks and⁄ or gaps in Schizosaccharomyces pombe has been reported [21] It was proposed that meiotic recombination could be initiated by DSBs, as well as
by non-DSB lesions, such as nicks and gaps At pre-sent, Spo11 is identified as the protein that catalyses DSBs and is widely conserved in eukaryotes [22] Prior
to the DSB-repair model of Resnick [23], it was sug-gested that DNA nicks or gaps induced meiotic recom-bination [24] These SSB recomrecom-bination models lost favor after the publication of another DSB repair model by Szostack et al [25], the observation of
Trang 3meiosis-specific DSBs at a recombination hotspot in
S cerevisiae [26], and the identification of Spo11 as a
DSB enzyme [27] Despite such differences between
yeast and other eukaryotes in the meiotic cell-cycle
pattern, the molecular machinery of meiotic DNA
recombination is likely to be conserved
Meiotic DNA recombination is composed of several
steps First, meiosis-specific DSBs or SSBs appear to
be introduced and this is followed by formation of
single-stranded DNA The formed single-stranded ends
then invade regions of homology in the other allele
After strand invasion and initial repair synthesis, the
crossover and the noncrossover pathways diverge [28]
These reactions are mediated by the coordinated
activ-ity of various proteins including RecA-like protein, an
ATPase playing a central role in the strand-exchange
reaction [29] In eukaryotes, Lim15⁄ Dmc1 and RAD51
have been identified as RecA homologs Whereas
RAD51 is expressed in both meiotic and somatic cells
and is required in the DNA repair reaction,
Lim15⁄ Dmc1 expression is restricted to meiotic cells
[29] Lim15⁄ Dmc1 has a critical role in meiotic
chro-mosome events, but its molecular functions and
differ-ences from Rad51 are not well understood [30]
In this review, we would like to discuss CcLim15 in
terms of its function and interactions In particular, we
would like to discuss the relationship with CcTopII
[31], CcPCNA [32] and CcUbc9 [15] at the meiotic
prophase of Coprinus cinereus and subsequently, the
role of sumoylation at zygotene and pachytene The
role of sumoylation in meiosis is still largely unclear
except for its involvements in the synaptonemal
com-plex [33–35], chromosome segregation [36] and
sperma-togenesis [37–40] We propose CcUbc9-mediated
sumoylation as a novel regulator of meiotic
chromo-some paring and recombination
Biomaterials for meiotic studies
In the biochemical study of meiosis, important
consid-eration should be paid to the choice of biomaterials,
because meiosis is a distinct part of sexual
develop-ment which occurs only at a certain point in time The
meiotic cell cycle must be synchronous and usable over
a year for such meiotic study Lilium microsporocytes
during the 1980s and before [16,41–47], and S
cerevisi-ae more recently [17], have mainly been used for the
studies The former system is not usable over a year
and has a genome that is too large for modern
genet-ics The latter system is very convenient for genetic
engineering but differs in the process of zygotene and
pachytene, two crucial stages for pairing and
recombi-nation, from the other eukaryotes
To avoid such problems, we have long used a basidi-omycete, C cinereus, as a model organism in studies
of sexual development and meiosis Despite the rapid morphogenesis of its multicellular structure, its meiotic cell cycle is long [48–50] and meiotic cells develop syn-chronously after photoinduction Each fruiting cap is extremely rich in meiotic cells at the same stage [48–50] Moreover, as is the case for yeasts, the gen-ome project for C cinereus has been completed and the genome is not so large
C cinereushas been analyzed using forward genetics approaches because of the ease of mutagenesis by transformation of an asexual spore of the haploid mycelium known as oidium [51–55] We have also suc-ceeded in performing gene repression by double-stran-ded RNA-mediated gene silencing as an alternative reverse genetics technique in C cinereus [56] Zolan
et al also reported molecular analyses of the C
cinere-usmeiotic recombination process [57–62]
By taking advantage of the properties of this organ-ism as described above, we succeeded in establishing cDNA libraries from mRNAs from C cinereus meiotic cells at leptotene, zygotene and pachytene and have studied 3R (DNA replication, repair, recombination) enzymes from each stage [31,63–69] We found that transcripts of the 3R enzymes as described below are abundant at meiotic prophase I and we have previ-ously discussed the roles of the 3R enzymes during meiosis The enzymes are PCNA (CcPCNA) [63], DNA ligase I [64], DNA ligase IV [65], Flap endonuc-lease-1 [66], Lim15⁄ Dmc1 (CcLim15) [67], Rad51 (CcRad51) [68,69] and DNA topoisomerase II (CcTop-II) [31]
We have also investigated the C cinereus DNA polymerase group in the database, and know that the
C cinereus genome has genes homologous to DNA polymerases a, d, e, f, k and l at a minimum and lacks genes homologous to DNA polymerase b and g [70–72] (A Sakamoto et al., unpublished results) Meiocytes at zygotene express at a minimum DNA polymerase a (CcPola), k (CcPolk) and l (CcPoll) and pachytene cells express CcPolk and CcPoll [72] (A Sakamoto et al., unpublished results) With regard
to DNA polymerase d, e and f, their expression has not been examined in C cinereus meiotic prophase According to biochemical studies of lily meiosis, a small amount of DNA replicates at zygotene (zygotene DNA synthesis) and repair synthesis of DNA occurs
at pachytene (pachytene DNA synthesis) [16,41,45] Zygotene and pachytene DNA syntheses are thought
to be the molecular basis of SC formation and recom-bination repair, respectively, and play a role in the progression of meiosis [16,19,46] As is well known,
Trang 4Pola is involved in replication, and Polk and Poll are
repair-type polymerases [73,74] Taking the meiotic
expression patterns of these DNA polymerases and the
two sequential DNA syntheses into consideration,
CcPola may contribute to zygotene DNA synthesis,
and CcPolk and CcPoll may be involved in pachytene
DNA synthesis in C cinereus
in meiosis
In a series of studies, we investigated RecA homologs
Two homologs of RecA-like protein, Lim15⁄ Dmc1
and Rad51 are known to be present in eukaryotes
According to Stassen et al [58], phylogenetic analyses
of eukaryotic RecA homologs reveal gene duplication
early in eukaryotic evolution giving rise to two
puta-tively monophyletic groups of RecA-like genes
Excep-tionally, higher plants possess one more additional
RecA-like protein, RadA [75] As in other eukaryotes,
with the exception of higher plants, we found
Lim15⁄ Dmc1 and Rad51 homologs in C cinereus
cDNA cloning and characterization of CcLim15 and
CcRad51 have been performed and both have been
shown to be expressed in meiotic prophase, especially
late leptotene to early zygotene [67,69] CcLim15 is
transcribed only in meiosis [67], whereas CcRad51 is
present in both somatic and meiotic cells [58] In
two-hybrid assays and in vitro protein–protein interaction
assays, both CcLim15 and CcRad51 homotypically
interact via their C-terminal domains [68] As
des-cribed previously [69], these two proteins exist in
mei-otic nuclei predominantly during late leptotene to
zygotene According to Lilium microsporocyte studies,
two different types of DNA synthesis occur at
zygo-tene and pachyzygo-tene [16,41,45] Because these DNA
synthetic processes appear to be for juxtaposing
homologous DNAs at zygotene and for exchanging
between homologous DNAs at pachytene, both
proces-ses would independently require D-loop formation If
this is the case, neither CcLim15 nor CcRad51 would
be involved in recombination between homologous
chromosomes at pachytene but rather in strand
arrangement (or SC formation) at zygotene
In order to understand the roles of these two RecA
homologs in meiosis, meiotic protein factors that
inter-act with them should be looked for Recent studies
imply that Rad51 interacts with various nuclear factors
such as RPA [76,77], Rad52 [78–80], Rad54 [81–83],
BRCA2 [84–87], the Rad55–Rad57 heterodimer [88]
and others By contrast, only a few proteins are known
to interact with Lim15⁄ Dmc1 The Rad54 homolog
proteins, Rdh54⁄ Tid1 in yeast and Rad54B in human,
interact with Lim15⁄ Dmc1 as well as Rad51 In
S cerevisiaeRdh54⁄ Tid1 is involved in crossover inter-ference [89,90], while Rad54B in human enhances Lim15⁄ Dmc1-mediated DNA-strand exchange The Mei5–Sae3 complex has also been identified as a new assembly factor for meiotic-specific Lim15⁄ Dmc1 in
S cerevisiae [91], while the Swi5–Sfr1 complex, the Mei5–Sae3 homolog in Schizosaccharomyces pombe, physically interacts with both RecA homologs [92] According to biochemical studies of Swi5–Sfr1, the complex stimulates strand exchange mediated by Lim15⁄ Dmc1, which indicates that Swi5–Sfr1 also acts
as a Lim15⁄ Dmc1 mediator [92] In addition, the Hop2–Mnd1 complex functionally associates with both RecA homologs and stimulates D-loop formation and strand exchange in yeast and mammals [93–95] Fur-ther screening for Lim15⁄ Dmc1 interactors would shed light on the machinery of meiotic chromosome paring and recombination This concept prompted us to screen for such proteins
As a result, we were successful in finding novel CcLim15-interacting proteins, namely DNA topoisom-erase II (CcTopII) [31], PCNA (CcPCNA) [32] and Ubc9 (CcUbc9) [15], through yeast two-hybrid screen-ing usscreen-ing the meiotic stage-specific cDNA library of
C cinereus This led us to find the possible involve-ment of sumoylation in meiosis Ubc9 is the E2 type enzyme for SUMO conjugation to targets In C cine-reus, CcLim15 is a target protein of sumoylation both
in vivoand in vitro, via interaction with CcUbc9 Inter-estingly, another RecA protein Rad51 was also repor-ted to associate with Ubc9, particularly in pachytene chromosomes in mouse spermatocytes [96] and was shown to be sumoylated in vitro [97] Furthermore, both TopII and PCNA also interact with Ubc9 and are well known targets of sumoylation [98,99] These properties add clarity to what is known about the con-trol of the meiotic chromosome events through post-translational modifications such as sumoylation
The role of Ubc9 in meiosis
Ubc9 in mitosis Ubc9 is known as a SUMO-conjugating enzyme (E2), which receives activated SUMO (SUMO-GG) from the Uba2 subunit of SUMO-activating enzyme (E1) and forms a SUMO–Ubc9 intermediate in the sumoy-lation pathway [2,7,9–13] Crystal structure analysis showed Ubc9 to have a domain similar to the core domain of ubiquitin-conjugating enzymes [100,101] The surface of Ubc9, however, is positively charged by two sequence insertions, while the corresponding
Trang 5regions in ubiquitin E2 enzymes have negative or
neut-ral charge [102–104] Thus, Ubc9 binds to SUMO but
not ubiquitin Furthermore, Ubc9 was reported
pre-viously to interact with many sumoylation targets
Within the hydrophobic groove of Ubc9, Asp127
appears to engage in hydrogen bonding with a Lys
residue within the sumoylation consensus motif
-Y-K-X-E ⁄ D-, where Y is a large hydrophobic amino
acid and X is any amino acid [103,105] Therefore,
complexes of Ubc9–Lim15⁄ Dmc1, Ubc9–TopII and
Ubc9–PCNA may be intermediates in the production
of SUMO–Lim15⁄ Dmc1, SUMO–TopII and SUMO–
PCNA complexes, respectively It is known that
sumoylated proteins can escape from immediate
ubiqu-itin-dependent degradation when both modifications
target the same lysine within the substrate [1–13]
Taken together with the manner of the substrate
recog-nition by Ubc9, the Ubc9-conjugated intermediates
may also be protected from ubiquitination However,
in addition to a role in mediating sumoylation, another
role of Ubc9 has been reported Binding of Ubc9 to a
nuclear-localization signal contributes to nuclear
local-ization of the homeobox protein Vsx1 [106]
Ubc9 in meiosis
In meiosis, a few roles of Ubc9 and sumoylation were
known Analysis of the Drosophila Ubc9 mutant
less-wright has implicated SUMO modification in the
dis-junction of homologous chromosomes in meiotic M1
[36] Ubc9 was shown to localize on meiotic
chromo-somes in S cerevisiae and mice and bind to the
consti-tutive proteins of the synaptonemal complex [33,96]
Recently, Cheng et al clarified the relationship
between SC formation and Ubc9-mediated
sumoyla-tion [35] In S cerevisiae Zip3, a protein involved in
the initiation of SC formation, is a SUMO E3 ligase
[33,35] In a Zip3-lacking mutant, a polycomplex was
formed instead of the SC Moreover, their results
sug-gested that Zip1, a building block of the SC, binds to
SUMO-conjugated proteins These interactions may be
important for homology sorting during early prophase,
as well as in triggering extensive SC polymerization
As described, meiosis is a special cell cycle
associ-ated with homologous chromosome pairing and
recombination [16] In mitosis, TopII is sumoylated in
a cell-cycle-controlled manner indicating that SUMO
modification serves to synchronize the function of
many of its substrates with the mitotic cell cycle
[107,108] By contrast, sumoylated PCNA has been
observed in the S phase but not in G2⁄ M [99] We
found that CcLim15, CcRad51, CcTopII, CcPCNA
and CcUbc9 are all present at meiotic prophase in
C cinereus and that each of CcLim15, CcTopII, CcPCNA and CcRad51 has the potential to interact with CcUbc9 Moreover, CcLim15 can also independ-ently interact with either CcTopII or CcPCNA at zygotene [31,32]
CcUbc9 is expressed from the premeiotic S phase through the tetrad stages, suggesting that CcUbc9 acts
in concert with many of the meiotic events [15] Expression of CcPCNA temporarily becomes most prominent at the transition between leptotene and zyg-otene, although small amounts of CcPCNA are con-stantly detected in nuclei from the premeiotic S phase through the tetrad stages [63] In contrast, CcLim15 and CcRad51 are expressed from late leptotene to early zygotene with CcLim15 and CcRad51 proteins present
at the same stages, then rapidly disappearing by early pachytene [67,69] CcTopII transcripts begin to accu-mulate during late leptotene, slightly earlier than the CcLim15 transcript, becoming most abundant at early zygotene [31] Thus, the interaction of CcLim15 with each of CcTopII, CcPCNA and CcUbc9 is always lim-ited around the transition between leptotene and zygo-tene, which is the point at which the homologous chromosomes pair (zygotene)
Taking the localization during meiotic prophase I and interactions of these proteins into consideration, CcLim15, CcRad51, CcTopII and CcPCNA may be the meiotic target proteins of sumoylation Because of the mechanism of Ubc9-mediated SUMO conjugation, analysis of the interaction between CcUbc9 and each
of CcLim15, CcRad51, CcTopII and CcPCNA would give a clue to homologous chromosomes pairing in relation to sumoylation For example, in late leptotene
or early zygotene, which of CcPCNA or CcUbc9 inter-acts the earliest with CcLim15 or is it a simultaneous interaction?
CcLim15–CcUbc9 complex in meiosis CcLim15 is distributed on the chromosomes in the nuclei at meiotic prophase, and becomes most pro-minent in late leptotene to zygotene [15,31,69] A CcLim15-repressed strain shows defects in SC forma-tion and abnormal homologous chromosome pairing during meiosis [56] CcLim15 is not detected after the late pachytene stages at all, whereas CcUbc9 is con-stant throughout meiosis, indicating that the CcLim15–CcUbc9 complex must occur and separate only for a limited period, namely during late leptotene
to zygotene [15] CcRad51 is also likely to behave in the same way, because its expression profile is the same
as CcLim15 [69] Therefore, the meiotic expression data for CcUbc9 indicates that chromosome paring,
Trang 6which is closely related to the function of CcLim15
and⁄ or CcRad51, may be partly controlled by
SUMO-mediated regulation Meanwhile, each of the CcUbc9
complexes may function independently in sumoylation
As described above, Ubc9 can conjugate to the lysine
residue within a sumoylation consensus motif in the
sumoylation pathway [105] This conjugation may
inhi-bit ubiquitin-mediated proteolysis RecA homologs,
Lim15⁄ Dmc1 and Rad51, promote strand exchange
with a donor DNA in an ATP-dependent manner [29]
CcLim15 is abundantly distributed on the
chromo-somes in late leptotene to zygotene CcRad51 is also
distributed in a similar way to CcLim15 (unpublished
data) The roles of CcLim15 and CcRad51 are likely
to overlap but be independent of each other Both
pro-teins are likely to be required at zygotene It was
sug-gested that the CcUbc9 complexes are protected from
degradation by ubiquitination at zygotene until strand
arrangement between the homologous chromatins is
complete (Fig 1)
There are two potential consensus motifs within
CcLim15, the sequences surrounding Lys78 (-AKVE-)
and Lys223 (-DKDF-) Although it is not clear
whe-ther Lys223 is the target site, the sumoylation target
sites are in the C-terminal part of CcLim15 (amino
acids 105–347), which contains the ATPase domain
[15] Intriguingly, the sumoylation target domain of
CcLim15 coincides with the domain that binds to
CcUbc9 [15] This correlation suggests that CcLim15
may be protected from degradation in the form of a
CcLim15–CcUbc9 intermediate before the regulation
of the functional activity by sumoylation
CcLim15 and CcTopII Previously the only known role for TopII in meiosis was in untying the entangling between chromatins, mainly at M1 [109,110] Immunocytochemistry of
C cinereus meiotic cells shows that CcTopII is locali-zed on chromosomes in nuclei during the premeiotic
S phase and also throughout the meiotic divisions, and that CcTopII signal culminated from leptotene to pachytene [31] Furthermore, CcTopII and CcLim15 colocalized during leptotene and zygotene, suggesting that the CcLim15–CcTopII complex may be related to specific events in early stages of meiosis [31]
As reported previously, CcLim15 and CcTopII influ-ence the activities of each other CcLim15 can potently activate the relaxation⁄ catenation activity of CcTopII
in vitro, but CcTopII suppresses CcLim15-dependent strand-transfer activity [31] CcLim15’s DNA-depend-ent ATP digestion potDNA-depend-ential was strongly enhanced by the CcTopII protein with ssDNA The ATPase activity
of DNA topoisomerase II is suppressed by using ssDNA as the cofactor We also measured DNA-dependent ATPase activity of CcTopII using double-stranded M13 DNA as a cofactor Although CcLim15 itself had subtle DNA-dependent ATPase activity in the presence of 1 mm Ca2+, the ATPase activity of CcTopII was significantly inhibited by addition of CcLim15 in the presence of 1 mm Ca2+ The interac-tion between CcLim15 and CcTopII could easily form during meiotic pairing between homologous chromo-somes at the boundary of leptotene to zygotene, i.e at the beginning of SC formation [31] Therefore, the
Fig 1 Model of the sequential molecular machinery involved in the meiotic chromo-some events from leptotene to zygotene Several steps in meiotic prophase are shown schematically Initially Lim15 inter-acts with TopII and homologous chromatins initiate pairing After dissociation of TopII, Lim15 remains on DNA and recruits Ubc9, Rad54B, Swi5–Sfr1 and PCNA After PCNA replaces Lim15 on the zygotene DNA, the free-Lim15 disappears via ubiquitin-mediated degradation The zygotene DNA is synthes-ized by Pola At the end of zygotene PCNA
is sumoylated and recruits Poll or ⁄ and Polk The pachytene DNA synthesis is occurred
by Poll or ⁄ and Polk.
Trang 7zygotene role of CcTopII may differ from the role in
M1, when it appears to control chromosome
disjunc-tion and pachytene chromosome segregadisjunc-tion
Interestingly, the C-terminus (amino acids 1066–
1569) of CcTopII as well as CcUbc9 binds to the
C-terminus (amino acids 104–345) of CcLim15,
indica-ting that CcTopII and CcUbc9 share the binding
domain within CcLim15 [15,31] In early meiotic
pro-phase, either CcTopII or CcUbc9 is likely to interact
at a similar if not the same site at the C-terminus of
CcLim15 The question is therefore raised as to which
of CcTopII or CcUbc9, binds to CcLim15 earlier? It
seems, although the evidence is weak (stage-dependent
expression order), that the CcTopII–CcLim15
interac-tion occurs at late leptotene to early zygotene [31]
while the CcLim15–CcUbc9 interaction appears to
occur throughout the whole of zygotene [15] Initially
the C-terminus of CcTopII binds to the C-terminus of
CcLim15, and then with progression through the
stages within zygotene, CcUbc9 may replace CcTopII
on the C-terminus of CcLim15 (Fig 1) The released
CcTopII molecules may also simultaneously be
sumo-ylated by CcUbc9
In meiosis, whether TopII is sumoylated or not is as
yet unclear However, the sumoylation of CcTopII may
occur through meiotic prophase and at M1, because of
its coexistence with CcUbc9 Three roles of sumoylated
CcTopII could be considered First, sumoylation may
contribute to the stability of CcTopII protein
Accord-ing to our studies, CcTopII would be involved in
chro-mosome pairing by interacting with CcLim15 Even if
CcTopII is released by CcLim15 and becomes unstable,
it may be protected by the immediate sumoylation from
ubiquitin-mediated degradation Second, sumoylated
CcTopII during the zygotene stage may be related to SC
initiation In S cerevisiae, SC formation is controlled
by sumoylation during assembling proteins and
chro-mosomes [34] TopII is a candidate for a
SUMO-conju-gated protein which binds to Zip1, a building block of
SC [33,35] Sumoylated CcTopII may interact with the
Zip1 homolog and form the foundation of SC A third
possibility is that sumoylation of CcTopII may be
rela-ted to chromosome segregation at M1 In mitosis, TopII
was found to be SUMO-modified and sumoylation of
TopII inhibits its ability to promote centromeric
cohe-sion [108] Sister chromatid cohecohe-sion at the centromere
is suggested to be specifically regulated by SUMO-1
modification of TopII Meanwhile, the disjunction of
homologous chromosomes in meiotic M1 occurred in
the Drosophila ubc9 mutant lesswright [36] Thus, as
seen in mitosis, CcTopII itself is also involved in untying
DNA entangling and may be inhibited in function by
sumoylation at M1
From a series of biochemical studies, we propose a hypothesis about the sequential molecular machinery related to the meiotic chromosome events from lepto-tene to zygolepto-tene Initially, CcLim15 finds and binds to CcTopII to bring homologous chromatins closer together Then some of the CcLim15–CcTopII com-plex are replaced by CcUbc9 resulting in a division into CcLim15–CcUbc9 and CcTopII–CcUbc9 com-plexes The SUMO-mediated CcTopII may prepare to form the SC Furthermore, CcLim15–Ccubc9 may need to be protected from proteolysis for it to still function in the next step In the zygotene process, Rad54B, a member of the Swi2⁄ Snf2 family of DNA translocases and homolog of yeast Rdh54⁄ Tid1, pos-sesses the ability to generate negative supercoils in duplex DNA, leads to the transient opening of the DNA strands in the duplex [111–113] and interacts with both Rad51 and Lim15⁄ Dmc1 The CcLim15– CcUbc9 complex may recruit Rad54B and CcUbc9 may be replaced by Rad54B in the complex In recent studies, Rad54B bound to the terminus of the Lim15⁄ Dmc1–ssDNA complex and caused stimulation
of Lim15⁄ Dmc1-mediated DNA-strand exchange [113] The CcLim15–Rad54B complex may stabilize the CcLim15–ssDNA complex and begin to pair homolog-ous zygotene DNAs (Fig 1) Shortly after, Rad54B is released from the complex by recruitment of CcUbc9, again, to protect CcLim15 on the DNA from ubi-quitin-mediated proteolysis (Fig 1) Next, the new CcLim15–CcUbc9 complex recruits the pairing elonga-tion factors (Swi5–Sfr1) with replacement of CcUbc9, and the CcLim15–Swi5–Sfr1 complex elongates the SC (Fig 1) Even if homologous chromosomes pair incor-rectly, CcLim15–Swi5–Sfr1 homology searching could contribute to correct pairing [92] Biochemical studies using yeast have provided evidence that Swi5–Sfr1 sti-mulates the strand exchange activity of Lim15⁄ Dmc1 [92] Finally at zygotene, the SC begins to dissociate (Fig 1)
It is well-known that purified Hop2–Mnd1 stimu-lates the strand invasion activity of Dmc1 in vitro in yeast, mouse and human [93,94,114] However, Hop2– Mnd1 has strand-exchange activity itself [115] and is required at pachytene according to fluorescence in situ hybridization of spread chromosomes [116] Although the interaction between Hop2 and Mnd1 in yeast and human was easily detected, they failed to detect any measurable interaction between Hop2–Mnd1 and Rad51 or Lim15⁄ Dmc1 [95,114,116] Hop2–Mnd1 appears to be able to form a complex and localize to chromosomes independent of Lim15⁄ Dmc1, suggesting that it might be required for the strand invasion pro-cess at pachytene
Trang 8Therefore, one of the roles of CcUbc9 is to protect
the C-terminus of CcLim15 on the zygotene DNA
from ubiquitination, since the dissociation of each
complex may lead to CcLim15 degradation triggered
by a specific proteolytic pathway such as the
ubiqu-itin–proteasome pathway (Fig 1) CcUbc9 is unable to
ubiquitinate this site Another role for CcUbc9 maybe
to serve to synchronize the zygotene cell cycle at each
point (Fig 1), as SUMO modification serves to
syn-chronize the function of many substrates with the
mitotic cell cycle [10]
CcPCNA–CcUbc9 complex in meiosis
In our previous study, CcPCNA was indicated to
inter-act with CcLim15 CcPCNA is detected in nuclei from
the premeiotic S phase through the tetrad stages
[32,63] Importantly, a significant proportion of
CcLim15 and CcPCNA colocalizes on chromosomes
from leptotene to zygotene Unlike CcTopII, however,
no enhancement of CcLim15-dependent strand transfer
or DNA-dependent ATPase activities by CcPCNA
have been found [32] We suggest that the
strand-trans-fer reaction by CcLim15 and the association between
CcLim15 and CcPCNA may be temporally separable
events in vivo Furthermore, CcLim15 binds to the
N-terminus of CcPCNA and CcPCNA binds to the
C-ter-minus of CcLim15 [32], suggesting that CcTopII and
CcPCNA compete on the C-terminus of CcLim15 One
possible hypothesis is that CcTopII and CcPCNA
would alternately bind to the C-terminus of CcLim15
at late leptotene to early zygotene, and each complex
would function for the cohesion (CcTopII–CcLim15)
and chromosome rearrangement (CcPCNA–CcLim15)
(Fig 1) Because chromosome rearrangement is
thought to accompany zygotene DNA replication [16],
CcPCNA may be involved in recruiting the replication
polymerase (Fig 1)
PCNA is known to interact with Ubc9 and is
sumo-ylated or ubiquitinated at Lys164 Ubiquitination of
PCNA is involved in the DNA-damage-tolerance
path-way, although the function of sumoylated PCNA is as
yet unclear [117] There are a few interesting reports
[118–120] that SUMO-modified PCNA may inhibit
Rad51-mediated DNA recombination after recruiting
SRS2, which then leads to gross chromosome
rear-rangement Genetic evidence also suggests that
sumoy-lation of PCNA on Lys164 inhibits Rad52-dependent
recombinational repair, which may reduce the risk of
chromosome rearrangements during the S phase [121]
It has not yet been examined whether PCNA is
sumo-ylated in meiosis In C cinereus, however, because
CcPCNA and CcUbc9 exist together in meiotic nuclei,
the interaction between CcPCNA and CcUbc9 and sumoylation may occur at a certain point of meiosis It
is suggested that sumoylation of CcPCNA may prevent premature chromosomal recombination from late lep-totene to early zygotene, until the end of the strand arrangement between homologous chromatins by CcLim15
At the beginning of zygotene, the possible role of the CcPCNA–CcLim15 interaction may be to recruit free CcPCNA onto the zygotene CcLim15–CcTopII cohesion region with CcPCNA replacing CcTopII (Fig 1) Alternatively, through sumoylation immedi-ately after cohesion, CcTopII–CcLim15 may separate into CcUbc9–CcTopII and CcLim15–CcUbc9, with CcLim15–CcUbc9 left on the cohesion regions leading
to the recruitment of CcPCNA into the regions (Fig 1) In our model, initially CcLim15–CcTopII occurring at late leptotene is involved in the coherence
of the homologous chromatins at the boundary and after dissociation, CcLim15 or CcLim15–CcUbc9 remain on the zygotene DNA to recruit CcPCNA at early zygotene, and finally, the nonmodified CcPCNA
is left there (Fig 1) Then, a replicative-type of DNA polymerase, for example CcPola, may be recruited in order to replicate the zygotene DNA sequence (Fig 1)
Of course, some CcLim15 must be left for binding to other factors as described above, and alternatively be used for various events at zygotene With progression
of the zygotene stage, CcTopII and CcPCNA on the complexes may replace CcUbc9 and be sumoylated For the next related-event to occur, each of CcLim15, CcTopII and CcPCNA has to be kept from the ubi-quitin-mediated degradation for a while (Fig 1)
As is well known, PCNA is closely related to DNA polymerases And the modification states of PCNA eli-cit different responses and select the types of DNA polymerases Unmodified PCNA acts as a processivity clamp for replicative DNA polymerases d and e [122] Monoubiquitination of PCNA is induced by DNA damage and activates DNA polymerases f and g for translesion synthesis [123] From S phase studies, it has been proposed that SUMO-modified PCNA may recruit DNA polymerase f in order to overcome repli-cation fork blocks not caused by DNA damage These suggest that PCNA may play a role as a switchboard
to shift DNA polymerases
Taking these observations into consideration, we would like to discuss the relationship between sumoy-lation of CcPCNA and meiotic DNA synthesis At zygotene, no repair-type DNA synthesis is observed, but replication-type does occur [41,46] Although PCNA is not modified during this stage in our model,
it is unclear whether DNA polymerases d and e, which
Trang 9are closely related to PCNA, are present in meiotic
prophase (A Sakamoto et al., unpublished results) In
C cinereus meiocytes CcPola is expressed at zygotene
and its primase-lacking form is mostly functional,
sug-gesting that this polymerase replicates the zygotene
DNA sequence [70–72] At the end of zygotene,
SUMO conjugation of CcPCNA should occur after
dissolution of the CcLim15–CcPCNA complex Next,
homologous chromosomes recombine and typical
repair-type DNA synthesis occurs at pachytene,
sug-gesting that the repair-type of DNA polymerases could
be recruited We demonstrated that the X family DNA
polymerases, namely the repair-type enzymes, CcPoll
and CcPolk localized in meiotic nuclei and that their
signal culminated at pachytene These two enzymes
may be recruited by sumoylated CcPCNA and
syn-thesize the pachytene DNA sequence As reported
previously, however, DNA polymerase k homolog
functionally and physically interacted with
nonmodi-fied PCNA [124] As yet there are no reports about the
interaction between sumoylated PCNA and DNA
po-lymerases including CcPolk and CcPoll and what is
more, it is not clear as yet whether PCNA continues to
be sumoylated through the pachytene stage These
points remain to be confirmed
References
1 Melchior F (2000) SUMO – nonclassical ubiquitin
Annu Rev Cell Dev Biol 16, 591–626
2 Yeh ET, Gong L & Kamitani T (2000)
Ubiquitin-like proteins: new wines in new bottles Gene 248,
1–14
3 Jackson PK (2001) A new RING for SUMO: wrestling
transcriptional responses into nuclear bodies with
PIAS family E3 SUMO ligases Genes Dev 15, 3053–
3058
4 Muller S, Hoege C, Pyrowolakis G & Jentsch S (2001)
SUMO, ubiquitin’s mysterious cousin Nat Rev Mol
Cell Biol 2, 202–210
5 Wilson VG & Rangasamy D (2001) Intracellular
tar-geting of proteins by sumoylation Exp Cell Res 271,
57–65
6 Wilson VG & Rangasamy D (2001) Viral interaction
with the host cell sumoylation system Virus Res 81,
17–27
7 Kim KI, Baek SH & Chung CH (2002) Versatile
pro-tein tag, SUMO: its enzymology and biological
func-tion J Cell Physiol 191, 257–268
8 Seeler JS & Dejean A (2003) Nuclear and unclear
func-tions of SUMO Nat Rev Mol Cell Biol 4, 690–699
9 Verger A, Perdomo J & Crossley M (2003)
Modifica-tion with SUMO A role in transcripModifica-tional regulaModifica-tion
EMBO Rep 4, 137–142
10 Dohmen RJ (2004) SUMO protein modification Biochim Biophys Acta 1695, 113–131
11 Gill G (2004) SUMO and ubiquitin in the nucleus: dif-ferent functions, similar mechanisms? Genes Dev 18, 2046–2059
12 Johnson ES (2004) Protein modification by SUMO Annu Rev Biochem 73, 355–382
13 Muller S, Ledl A & Schmidt D (2004) SUMO: a regu-lator of gene expression and genome integrity Onco-gene 23, 1998–2008
14 Wykoff DD & O’Shea EK (2005) Identification of sumoylated proteins by systematic immunoprecipita-tion of the budding yeast proteome Mol Cell Proteo-mics 4, 73–83
15 Koshiyama A, Hamada FN, Namekawa SH, Iwabata
K, Sugawara H, Sakamoto A, Ishizaki T & Sakaguchi
K (2006) Sumoylation of a meiosis-specific RecA homolog, Lim15⁄ Dmc1, via interaction with the small ubiquitin-related modifier (SUMO)-conjugating enzyme Ubc9 FEBS J 273, 4003–4012
16 Stern H & Hotta Y (1985) Molecular biology of mei-osis: synapsis-associated phenomena Basic Life Sci 36, 305–316
17 Haber JE (1998) Meiosis: avoiding inappropriate rela-tionships Curr Biol 8, R832–R835
18 Hotta Y & Stern H (1974) DNA scission and repair during pachytene in Lilium Chromosoma 46, 276– 296
19 Hotta Y & Stern H (1976) Persistent discontinuities
in late replicating DNA during meiosis in Lilium Chromosoma 55, 171–182
20 Hotta Y, Chandley AC & Stern H (1977) Meiotic crossing-over in lily and mouse Nature 269, 240–242
21 Farah JA, Cromie G, Davis L, Steiner WW & Smith
GR (2005) Activation of an alternative, rec12 (spo11)-independent pathway of fission yeast meiotic recombi-nation in the absence of a DNA flap endonuclease Genetics 171, 1499–1511
22 Keeney S & Neale MJ (2006) Initiation of meiotic recombination by formation of DNA double-strand breaks: mechanism and regulation Biochem Soc Trans
34, 523–525
23 Resnick MA (1976) The repair of double-strand breaks
in DNA; a model involving recombination J Theor Biol 59, 97–106
24 Smith GR (2004) How homologous recombination is initiated: unexpected evidence for single-strand nicks from v(d)j site-specific recombination Cell 117, 146–148
25 Szostak JW, Orr-Weaver TL, Rothstein RJ & Stahl
FW (1983) The double-strand-break repair model for recombination Cell 33, 25–35
26 Sun H, Treco D, Schultes NP & Szostak JW (1989) Double-strand breaks at an initiation site for meiotic gene conversion Nature 338, 87–90
Trang 1027 Keeney S, Giroux CN & Kleckner N (1997)
Meiosis-specific DNA double-strand breaks are catalyzed by
Spo11, a member of a widely conserved protein family
Cell 88, 375–384
28 Lorenz A & Whitby MC (2006) Crossover promotion
and prevention Biochem Soc Trans 34, 537–541
29 Masson JY & West SC (2001) The Rad51 and Dmc1
recombinases: a non-identical twin relationship Trends
Biochem Sci 26, 131–136
30 Shinohara A & Shinohara M (2004) Roles of RecA
homologues Rad51 and Dmc1 during meiotic
recombi-nation Cytogenet Genome Res 107, 201–207
31 Iwabata K, Koshiyama A, Yamaguchi T, Sugawara H,
Hamada FN, Namekawa SH, Ishii S, Ishizaki T,
Chiku H, Nara T et al (2005) DNA topoisomerase II
interacts with Lim15⁄ Dmc1 in meiosis Nucleic Acids
Res 33, 5809–5818
32 Hamada FN, Koshiyama A, Namekawa SH, Ishii S,
Iwabata K, Sugawara H, Nara TY, Sakaguchi K &
Sawado T (2007) Proliferating cell nuclear antigen
(PCNA) interacts with a meiosis-specific RecA
homo-logue, Lim15⁄ Dmc1, but does not stimulate its strand
transfer activity Biochem Biophys Res Commun 352,
836–842
33 Hooker GW & Roeder GS (2006) A role for SUMO
in meiotic chromosome synapsis Curr Biol 16, 1238–
1243
34 De Carvalho CE & Colaiacovo MP (2006)
SUMO-mediated regulation of synaptonemal complex
forma-tion during meiosis Genes Dev 20, 1986–1992
35 Cheng CH, Lo YH, Liang SS, Ti SC, Lin FM, Yeh
CH, Huang HY & Wang TF (2006) SUMO
modifica-tions control assembly of synaptonemal complex and
polycomplex in meiosis of Saccharomyces cerevisiae
Genes Dev 20, 2067–2081
36 Apionishev S, Malhotra D, Raghavachari S, Tanda S
& Rasooly RS (2001) The Drosophila UBC9
logue lesswright mediates the disjunction of
homo-logues in meiosis I Genes Cells 6, 215–224
37 Yan W, Santti H, Janne OA, Palvimo JJ & Toppari J
(2003) Expression of the E3 SUMO-1 ligases PIASx
and PIAS1 during spermatogenesis in the rat Gene
Expr Patterns 3, 301–308
38 Rogers RS, Inselman A, Handel MA & Matunis MJ
(2004) SUMO modified proteins localize to the XY
body of pachytene spermatocytes Chromosoma 113,
233–243
39 Vigodner M & Morris PL (2005) Testicular expression
of small ubiquitin-related modifier-1 (SUMO-1)
sup-ports multiple roles in spermatogenesis: silencing of sex
chromosomes in spermatocytes, spermatid microtubule
nucleation, and nuclear reshaping Dev Biol 282, 480–
492
40 Vigodner M, Ishikawa T, Schlegel PN & Morris PL
(2006) SUMO-1, human male germ cell development,
and the androgen receptor in the testis of men with normal and abnormal spermatogenesis Am J Physiol Endocrinol Metab 290, E1022–E1033
41 Hotta Y & Stern H (1971) Analysis of DNA synthesis during meiotic prophase in Lilium J Mol Biol 55, 337– 355
42 Hotta Y & Stern H (1971) A DNA-binding protein in meiotic cells of Lilium Dev Biol 26, 87–99
43 Moens PB (1970) The fine structure of meiotic chro-mosome pairing in natural and artificial Lilium poly-ploids J Cell Sci 7, 55–63
44 Stern H, Westergaard M & Von Wettstein D (1975) Presynaptic events in meiocytes of Lilium longiflorum and their relation to crossing-over: a preselection hypo-thesis Proc Natl Acad Sci USA 72, 961–965
45 Smyth DR & Stern H (1973) Repeated DNA synthes-ized during pachytene in Lilium henryi Nat New Biol
245, 94–96
46 Hotta Y, Tabata S & Stern H (1984) Replication and nicking of zygotene DNA sequences Control by a mei-osis-specific protein Chromosoma 90, 243–253
47 Hotta Y, Tabata S, Bouchard RA, Pinon R & Stern H (1985) General recombination mechanisms in extracts
of meiotic cells Chromosoma 93, 140–151
48 Casselton LA & Olesnicky NS (1998) Molecular genet-ics of mating recognition in basidiomycete fungi Microbiol Mol Biol Rev 62, 55–70
49 Kues U (2000) Life history and developmental proces-ses in the basidiomycete Coprinus cinereus Microbiol Mol Biol Rev 64, 316–353
50 Kamada T (2002) Molecular genetics of sexual devel-opment in the mushroom Coprinus cinereus Bioessays
24, 449–459
51 Binninger DM, Skrzynia C, Pukkila PJ & Casselton
LA (1987) DNA-mediated transformation of the basidiomycete Coprinus cinereus Embo J 6, 835–840
52 Zolan ME, Tremel CJ & Pukkila PJ (1988) Production and characterization of radiation-sensitive meiotic mutants of Coprinus cinereus Genetics 120, 379–387
53 Zolan ME, Crittenden JR, Heyler NK & Seitz LC (1992) Efficient isolation and mapping of rad genes of the fungus Coprinus cinereus using
chromosome-speci-fic libraries Nucleic Acids Res 20, 3993–3999
54 Pukkila PJ & Skrzynia C (1993) Frequent changes in the number of reiterated ribosomal RNA genes throughout the life cycle of the basidiomycete Coprinus cinereus Genetics 133, 203–211
55 Cummings WJ, Celerin M, Crodian J, Brunick LK & Zolan ME (1999) Insertional mutagenesis in Coprinus cinereus: use of a dominant selectable marker to gener-ate tagged, sporulation-defective mutants Curr Genet
36, 371–382
56 Namekawa SH, Iwabata K, Sugawara H, Hamada
FN, Koshiyama A, Chiku H, Kamada T & Sakaguchi
K (2005) Knockdown of LIM15⁄ DMC1 in the