The effect of heterochromatin on synapsis of the sex chromosomes of Peromyscus (Rodentia, Cricetidae)

The pairing behavior of the sex chromosomes in male and female individuals representing seven species of Peromyscus was analyzed by electron microscopy of silver-stained zygotene and pachytene configurations. Six species possess submetacentric or metacentric X chromosomes with heterochromatic short arms. Sex-chromosome pairing in these species is initiated during early pachynema at an interstitial position on the X and Y axes. Homologous synapsis then progresses in a unidirectional fashion towards the telomeres of the X short arm and the corresponding arm of the heterochromatic Y chromosome. The distinctive pattern of synaptic initiation allowed a late-synapsing bivalent in fetal oocytes to be tentatively identified as that of the X chromosomes. In contrast to the other species, Peromyscus megalops possesses an acrocentric X chromosome and a very small Y chromosome. Sex-chromosome pairing in this species is initiated at the proximal telomeric region during late zygonema, and then proceeds interstitially towards the distal end of the Y chromosome. These observations suggest that the presence of X short-arm heterochromatin and corresponding Y heterochromatin interferes with late-zygotene alignment of the pairing initiation sites, thereby delaying XY synaptic initiation until early pachynema. The pairing initiation sites are conserved in the vicinity of the X and Y centromeres in Peromyscus, and consequently the addition of heterochromatin during sex-chromosome evolution essentially displaces these sites to an interstitial position.     

The initiation of homologous chromosome synapsis in mouse fetal oocytes is not directly driven by centromere and telomere clustering in the bouquet

We investigated the behaviour of centromeres and distal telomeres during the initial phases of female meiosis in mice. In particular, we wished to determine whether clustering of centromeres and telomeres (bouquet formation) played the same crucial role in homologous chromosome pairing in female meiosis as it does in the male. We found that synapsis (intimate homologous chromosome pairing) is most frequently initiated in the interstitial regions of homologous chromosomes, apparently ahead of the distal regions. The proximal ends of the chromosomes appear to be disfavoured for synaptic initiation. Moreover, initiation of synapsis occurred in oocytes that showed little or no evidence of bouquet formation. A bouquet was present in a substantial proportion of cells at mid to late zygotene, and was still present in some pachytene oocytes. This pattern of bouquet formation and pairing initiation is in stark contrast to that previously described in the male mouse. We propose that although dynamic movements of centromeres and telomeres to form clusters may facilitate alignment of homologues or homologous chromosome segments during zygotene, in the female mouse positional control of synaptic initiation is dependent on some other mechanism.     

A pathway for synapsis initiation during zygotene in Drosophila oocytes

Formation of the synaptonemal complex (SC), or synapsis, between homologs in meiosis is essential for crossing over and chromosome segregation [1-4]. How SC assembly initiates is poorly understood but may have a critical role in ensuring synapsis between homologs and regulating double-strand break (DSB) and crossover formation. We investigated the genetic requirements for synapsis in Drosophila and found that there are three temporally and genetically distinct stages of synapsis initiation. In "early zygotene" oocytes, synapsis is only observed at the centromeres. We also found that nonhomologous centromeres are clustered during this process. In "mid-zygotene" oocytes, SC initiates at several euchromatic sites. The centromeric and first euchromatic SC initiation sites depend on the cohesion protein ORD. In "late zygotene" oocytes, SC initiates at many more sites that depend on the Kleisin-like protein C(2)M. Surprisingly, late zygotene synapsis initiation events are independent of the earlier mid-zygotene events, whereas both mid and late synapsis initiation events depend on the cohesin subunits SMC1 and SMC3. We propose that the enrichment of cohesion proteins at specific sites promotes homolog interactions and the initiation of euchromatic SC assembly independent of DSBs. Furthermore, the early euchromatic SC initiation events at mid-zygotene may be required for DSBs to be repaired as crossovers.     

Synaptonemal complex components persist at centromeres and are required for homologous centromere pairing in mouse spermatocytes

Recent studies in simple model organisms have shown that centromere pairing is important for ensuring high-fidelity meiotic chromosome segregation. However, this process and the mechanisms regulating it in higher eukaryotes are unknown. Here we present the first detailed study of meiotic centromere pairing in mouse spermatogenesis and link it with key events of the G2/metaphase I transition. In mouse we observed no evidence of the persistent coupling of centromeres that has been observed in several model organisms. We do however find that telomeres associate in non-homologous pairs or small groups in B type spermatogonia and pre-leptotene spermatocytes, and this association is disrupted by deletion of the synaptonemal complex component SYCP3. Intriguingly, we found that, in mid prophase, chromosome synapsis is not initiated at centromeres, and centromeric regions are the last to pair in the zygotene-pachytene transition. In late prophase, we first identified the proteins that reside at paired centromeres. We found that components of the central and lateral element and transverse filaments of the synaptonemal complex are retained at paired centromeres after disassembly of the synaptonemal complex along diplotene chromosome arms. The absence of SYCP1 prevents centromere pairing in knockout mouse spermatocytes. The localization dynamics of SYCP1 and SYCP3 suggest that they play different roles in promoting homologous centromere pairing. SYCP1 remains only at paired centromeres coincident with the time at which some kinetochore proteins begin loading at centromeres, consistent with a role in assembly of meiosis-specific kinetochores. After removal of SYCP1 from centromeres, SYCP3 then accumulates at paired centromeres where it may promote bi-orientation of homologous centromeres. We propose that, in addition to their roles as synaptonemal complex components, SYCP1 and SYCP3 act at the centromeres to promote the establishment and/or maintenance of centromere pairing and, by doing so, improve the segregation fidelity of mammalian meiotic chromosomes.     

The nature of the 1;29 translocation in cattle as revealed by synaptonemal complex analysis using electron microscopy

Synaptonemal complex analyses were carried out by electron microscopy on surface-spread spermatocytes of one normal bull and two bulls that were heterozygous for the so-called 1;29 translocation. The autosomal bivalents of the normal karyotype, which could be arranged by size in a series, demonstrated kinetochores at the terminally located attachment plaques. One autosomal bivalent was clearly larger than the rest and apparently consisted of the long arm of the 1;29 translocation. The 1;29 translocation was the longest autosome in the set and had a kinetochore in a subtelocentric position. Some of the autosome pairs had nucleolus organizer regions in telomeric regions. The X and Y chromosomes, which were not paired at zygotene, demonstrated association in a very short segment at early pachytene; in no cells could a synaptonemal complex be seen between the X and Y. Very often the sex chromosomes were dissociated. At zygotene, a few, usually large, bivalents were unpaired proximally. This always also involved the proximal parts of the arms of the 1;29 translocation and their normal homologs. At early pachytene, the 1;29 trivalent, although to a less extensive degree, was also unpaired in the pericentric region. Configurations in which one chromosome, either 1 or 29, was completely paired with its corresponding arm in the 1;29 translocation chromosome also occurred. When unpaired proximally, the size of chromosome 1 agreed fairly well with the size of its corresponding arm, but the size of chromosome 29 was considerably larger than the corresponding arm of the 1;29 translocation chromosome. During late zygotene and early pachytene, the percent difference between chromosome 29 and its corresponding arm decreased, and at mid and late pachytene there had been a complete synaptic adjustment. The size difference and pairing behavior indicated that a deletion of the kinetochore and the most proximal segment of chromosome 29 had preceded the fusion with chromosome 1 into the 1;29 translocation. The unique structural appearance of the 1;29 translocation chromosome compared to that of other centric fusion translocations in cattle lends support to the theory of a monophyletic origin of the 1;29 translocation. The importance of the pairing behavior observed in governing recombination and chromosome disjunction is briefly discussed.     

Terminal regions of wheat chromosomes select their pairing partners in meiosis

Many plant species, including important crops like wheat, are polyploids that carry more than two sets of genetically related chromosomes capable of meiotic pairing. To safeguard a diploid-like behavior at meiosis, many polyploids evolved genetic loci that suppress incorrect pairing and recombination of homeologues. The Ph1 locus in wheat was proposed to ensure homologous pairing by controlling the specificity of centromere associations that precede chromosome pairing. Using wheat chromosomes that carry rye centromeres, we show that the centromere associations in early meiosis are not based on homology and that the Ph1 locus has no effect on such associations. Although centromeres indeed undergo a switch from nonhomologous to homologous associations in meiosis, this process is driven by the terminally initiated synapsis. The centromere has no effect on metaphase I chiasmate chromosome associations: homologs with identical or different centromeres, in the presence and absence of Ph1, pair the same. A FISH analysis of the behavior of centromeres and distal chromomeres in telocentric and bi-armed chromosomes demonstrates that it is not the centromeric, but rather the subtelomeric, regions that are involved in the correct partner recognition and selection.     

Fluorescent in situ hybridization (FISH) analysis of the relationship between chromosome location and nuclear morphology in human neutrophils

Human neutrophil nuclei typically consist of three of four large heterochromatic lobes joined by thin, DNA-containing filaments. In addition, some lobes exhibit appendages of various sizes and shapes. Classical genetic and cytological studies suggest that some appendages contain specific chromsomes. The studies reported here provide the first detailed analysis of the spatial relationship between individual chromosomes and recognizable structures in neutrophil nuclei using fluorescent in situ hybridization. Analysis of DNA sequences in chromosomes 2, 18, X, and Y demonstrate that specific lobes in a population of neutrophil nuclei do not have a fied chromosome content. This result implies that chromosomes partition randomly among lobes during neutrophil differentiation. However, neutrophil nuclear topography is not entirely fortuitous. For instance, none of the sequences probed in this study mapped to a filament and most centromeres lie in clusters near the nuclear periphery. In addition, one of the X chromosome centromeres in females and the Y chromosome centromere in males consistently associate with specific nuclear appendages found in a subset of neutrophil nuclei. Chromosomes 2 and 18 occupy discrete nd separate territories within individual lobes and neither territory ever extends into a filament. Surprisingly, the sizes of these territories are not proportional to chromosome length, suggesting that individual neutrophil chromosomes vary in their degree of compaction. These results are discussed in the light of models that attempt to explain nuclear morphology in terms of chromosome spatial organization. Received: 10 April 1997 / Accepted: 14 April 1997     

Presence of abnormal synaptonemal complexes in heterothallic species of Neurospora

Synaptonemal complex abnormalities are frequent in reconstructed meiotic prophase nuclei of Neurospora crassa and Neurospora intermedia. Three kinds of synaptonemal complex anomalies were seen: lateral component splits, lateral component junctions, and multiple complexes. The anomalies apparently are formed during or after the pairing process, as they were not seen in the largely unpaired early zygotene chromosomes. Their presence at all the other substages from mid-zygotene to late pachytene indicates that they are not eliminated before the synaptonemal complex decomposes at diplotene. Abnormal synaptonemal complexes were seen in all 19 crosses of N. crassa and N. intermedia that were examined, including matings between standard laboratory strains, inversions, Spore killers, and strains collected from nature. The frequency of affected nuclei and degree of abnormality within a nucleus varied in different matings. No abnormalities were present in the homothallic species Neurospora africana and Neurospora terricola. Structural chromosome aberrations, introgression, and heterozygosity have been eliminated as causes for pairing disorder. The abnormal synaptonemal complexes seemingly do not interfere with normal ascus development and ascospore formation. The affected nuclei are not aborted during meiotic prophase, nor are they eliminated by abortion of mature asci. The abnormal meiocytes do not lead to aneuploidy, as judged by the low frequency of white ascospores in crosses between wild type strains that have many abnormalities. Thus, the abnormal synatonemal complexes do not appear to prevent chiasma formation between homologues.     

Centromere and telomere redistribution precedes homologue pairing and terminal synapsis initiation during prophase I of cattle spermatogenesis

Alterations in nuclear topology associated with meiotic chromosome pairing were studied in premeiotic cells and spermatocytes I of adult bovine males. To this end, we performed FISH with chromosome, pericentromeric satellite-DNA and telomere-specific probes in combination with immunostaining of synaptonemal complex proteins (SCP3, SCP1) on testis tissue sections. Nuclei of premeiotic cells (spermatogonia) exhibited a scattered telomere distribution while pericentromeres formed a few intranuclear clusters. We observed that the chromosome pairing process in cattle prophase I is preceded by repositioning of centromeres and telomeres to the nuclear periphery during preleptotene. Clustering of chromosome ends (bouquet formation) was observed during the transition from leptonema to zygonema and coincided with pairing of a sub-centromeric marker of bovine chromosomes 7. Dissolution of bouquet topology during zygonema left perinuclear telomeres scattered over the nuclear periphery at pachynema. SCP3 staining in frozen tissue sections revealed the appearance of this axial element protein in intranuclear aggregates during preleptotene, followed by extensive axial element formation during leptotene. Synapsis as revealed by SCP1 staining initiated peripherally at earliest zygotene, at this stage nuclei still contained numerous SCP3 clusters. Our observations reveal prominent non-homologous satellite-DNA associations in spermatogonia and indicate the conservation of topological features of the meiotic chromosome pairing process among mammals. The comparison of telomere dynamics in mouse and cattle prophase I suggests that a larger number of chromosomes prolongs the duration of the bouquet stage.     

Two new X-autosome Robertsonian translocations in the mouse

The influence of X-autosome Robertsonian (Rb) translocation hemizygosity on meiotic chromosome behaviour was investigated in male mice. Two male fertile translocations [Rb(X.2)2Ad and Rb(X.9)6H] and a male sterile translocation [Rb(X.12)7H] were used. In males of all three Rb translocation types, the acrocentric homologue of the autosome involved in the rearrangement regularly failed at pachytene to pair completely with its partner in the Rb metacentric. The centric end of the acrocentric autosome was found regularly to associate either with the proximal end of the Y chromosome or with the ends of nonhomologous autosomal bivalents; the proportions of cells with such configurations varied between pachytene substages and genotypes. Various other categories of synaptic anomaly, such as nonhomologous synapsis, foldback pairing and interlocks, affected the sex chromosome multivalent in a substantial proportion of cells. In one of the Rb(X.12)7H males screened, an unusual, highly aneuploid spermatocyte that contained trivalent and bivalent configurations was found. Rb translocation hemizygosity did not appear to increase to a significant extent the incidence of X-Y pairing failure at pachytene, although the incidence was elevated at metaphase I in Rb(X.12)7H animals. Overall, a comparison of the frequencies and types of chromosome pairing anomalies did not suggest that these were important factors in the aetiology of infertility in males carrying the Rb(X.12)7H translocation.     

Molecular analysis of recombination in a family with Duchenne muscular dystrophy and a large pericentric X chromosome inversion.

It has been demonstrated in animal studies that, in animals heterozygous for pericentric chromosomal inversions, loop formation is greatly reduced during meiosis. This results in absence of recombination within the inverted segment, with recombination seen only outside the inversion. A recent study in yeast has shown that telomeres, rather than centromeres, lead in chromosome movement just prior to meiosis and may be involved in promoting recombination. We studied by cytogenetic analysis and DNA polymorphisms the nature of meiotic recombination in a three-generation family with a large pericentric X chromosome inversion, inv(X)(p21.1q26), in which Duchenne muscular dystrophy (DMD) was cosegregating with the inversion. On DNA analysis there was no evidence of meiotic recombination between the inverted and normal X chromosomes in the inverted segment. Recombination was seen at the telomeric regions, Xp22 and Xq27-28. No deletion or point mutation was found on analysis of the DMD gene. On the basis of the FISH results, we believe that the X inversion is the mutation responsible for DMD in this family. Our results indicate that (1) pericentric X chromosome inversions result in reduction of recombination between the normal and inverted X chromosomes; (2) meiotic X chromosome pairing in these individuals is likely initiated at the telomeres; and (3) in this family DMD is caused by the pericentric inversion.     

Synaptonemal complex-dependent centromeric clustering and the initiation of synapsis in Drosophila oocytes

The pairing of homologous chromosomes and the intimate synapsis of the paired homologs by the synaptonemal complex (SC) are essential for subsequent meiotic processes including recombination and chromosome segregation. Here we show that the centromere clustering plays an important role in initiating homolog synapsis during meiosis in Drosophila females. Although centromeres are not clustered prior to the onset of meiosis, all four pairs of centromeres are actively clustered into one or two masses during early meiotic prophase. Within the 16-cell cyst, centromeric clustering appears to define the first step in the initiation of synapsis. Clustering is restricted to the nuclei that form the SC and is dependent on all known SC proteins. Surprisingly, both centromeric clusters and the SC components associated with them persist long after the disassembly of the euchromatic SC at the end of pachytene. The initiation of homologous recombination through the formation of programmed double-strand breaks (DSBs) is not required for either the formation or the maintenance of the centromeric clusters. Our data support a view in which the SC-mediated clustering at the centromeres is the initiating event for meiotic synapsis.     

Y-autosome translocation and infertility: usefulness of molecular, cytogenetic and meiotic studies

An apparently balanced reciprocal translocation 46,X,t(Y;6) (q11.23 ∼ q12;p11.1) was observed in an infertile man with severe oligozooteratozoospermia. Different mitotic chromosome banding patterns were performed and fluorescence in situ hybridization indicated a breakpoint in the fluorescent Yq heterochromatin. Molecular genetic deletion experiments for the azoospermia factor region in distal Yq11 showed the retention of the DAZ gene and meiotic pairing configurations suggested that the man’s infertility could be due to the pairing behaviour of the Y;6 translocation chromosome with the X chromosome visualised by synaptonemal complex analysis at the electron microscopy level. The morphological appearance of the normal chromosome 6 and the Y;6 translocated chromosome included in the compartment of the sex vesicle may allow an explanation of the degeneration of most spermatocytes after the pachytene stage. Received: 1 August 1997 / Accepted: 25 September 1997     

The synaptic behaviour of the wild forms of Triticum turgidum and T. timopheevii.

Different wild allopolyploid species of Triticeae show extensive bivalent formation at zygotene while a considerable number of multivalents is present in cultivated polyploid wheats. To study the chromosome behaviour at early meiotic stages in wild forms of tetraploid wheats Triticum turgidum and T timopheevii (2n = 4x = 28) we have analysed the synaptic pattern in fully traced spread nuclei at mid- and late zygotene and at pachytene of wild accessions of these species. The mean number of synaptonemal complex (SC) bivalents at mid-zygotene ranged from 12.22 to 13.14 among the accessions studied indicating a strong restriction of synapsis initiation to homologous chromosomes. The mean of bivalents increased at pachytene because of the transformation of multivalents into bivalents. Ring bivalents observed at metaphase I support that SC bivalents were formed by homologous chromosomes. The average values of SC bivalents at mid-zygotene in the wild forms are much higher than the average values observed in the cultivated tetraploid wheats but similar to that of a mutant line of T turgidum with a duplication that includes Ph1, the major homoeologous pairing suppressor locus. These results suggest that the efficiency of the mechanism operating in the homologous recognition for synapsis is higher in wild wheat populations than in cultivated varieties. Apparently, a relatively detrimental modification of the pairing regulating genetic system accompanied the domestication of the wild wheat forms.     

Centromeric behaviour in wheat with high and low homoeologous chromosomal pairing

Control of homoeologous chromosomal pairing in hexaploid wheat stems from a balance between a number of suppressor and promoter genes. This study used centromeric behaviour as a tool to investigate the mechanism. Fluorescent in situ hybridization employing centromeric and telomeric sequences as probes was applied to pollen mother cells of wheat and wheat/alien hybrids having different pairing gene combinations. It showed: association of centromeres during pre-meiotic interphase; decondensation of centromeric structure; sister chromatid disjunction of univalent chromosomes in homoeologous pairing situations at anaphase I; and centromeric stretching between univalent sister chromatids in wheat/rye hybrids deficient for pairing genes. The implications of these results are discussed. Received: December 1996 / in revised form: 11 March 1997 / Accepted: 14 March 1997     

Centromere separation and association in the nuclei of an interspecific hybrid between Torenia fournieri and T. baillonii (Scrophulariaceae) during mitosis and meiosis

In the nuclei of some interspecific hybrid and allopolyploid plant species, each genome occupies a separate spatial domain. To analyze this phenomenon, we studied localization of the centromeres in the nuclei of a hybrid between Torenia fournieri and T. baillonii during mitosis and meiosis using three-dimensional fluorescence in situ hybridization (3D-FISH) probed with species-specific centromere repeats. Centromeres of each genome were located separately in undifferentiated cells but not differentiated cells, suggesting that cell division might be the possible force causing centromere separation. However, no remarkable difference of dividing distance was detected between chromatids with different centromeres in anaphase and telophase, indicating that tension of the spindle fiber attached to each chromatid is not the cause of centromere separation in Torenia. In differentiated cells, centromeres in both genomes were not often observed for the expected chromosome number, indicating centromere association. In addition, association of centromeres from the same genome was observed at a higher frequency than between different genomes. This finding suggests that centromeres within one genome are spatially separated from those within the other. This close position may increase possibility of association between centromeres of the same genome. In meiotic prophase, all centromeres irrespective of the genome were associated in a certain portion of the nucleus. Since centromere association in the interspecific hybrid and amphiploid was tighter than that in the diploid parents, it is possible that this phenomenon may be involved in sorting and pairing of homologous chromosomes.     

Dynamics of homologous chromosome pairing during meiotic prophase in fission yeast

Pairing of homologous chromosomes is important for homologous recombination and correct chromosome segregation during meiosis. It has been proposed that telomere clustering, nuclear oscillation, and recombination during meiotic prophase facilitate homologous chromosome pairing in fission yeast. Here we examined the contributions of these chromosomal events to homologous chromosome pairing, by directly observing the dynamics of chromosomal loci in living cells of fission yeast. Homologous loci exhibited a dynamic process of association and dissociation during the time course of meiotic prophase. Lack of nuclear oscillation reduced association frequency for both centromeric and arm regions of the chromosome. Lack of telomere clustering or recombination reduced association frequency at arm regions, but not significantly at centromeric regions. Our results indicate that homologous chromosomes are spatially aligned by oscillation of telomere-bundled chromosomes and physically linked by recombination at chromosome arm regions; this recombination is not required for association of homologous centromeres.     

A study of cryptic terminal chromosome rearrangements in recurrent miscarriage couples detects unsuspected acrocentric pericentromeric abnormalities

Fifty chromosomally normal couples with three or more miscarriages were examined using fluorescent in situ hybridisation (FISH) and a library of subtelomere-specific probes together with alphoid repeats mapping to the acrocentric centromeres. Six abnormalities were found. Firstly, a cryptic reciprocal subtelomere translocation between the long arm of a chromosome 3 and the short arm of a chromosome 10. The other five cryptic abnormalities involved the acrocentric chromosome pericentromeric regions and in one case also Yp. Two patients had a rearranged chromosome 13, where the centromeric region was found to be derived from the short arm, centromere and proximal long arm of chromosome 15. Another two patients had a derived chromosome 22, where the centromere was replaced by two other centromeres, one derived from chromosome 14 and the other from either chromosome 13 or 21, while one patient had the subtelomere region of Yp translocated onto the short arm of a chromosome 21. These abnormalities may be the underlying cause of the recurrent miscarriages, because they may result in abnormal pairing configurations at meiosis leading to non-disjunction of whole chromosomes at metaphase I. The frequency of rearrangements seen in the recurrent miscarriage patient population was significantly different from that in the control group ( P=0.0096, Fisher's exact test) due to the acrocentric pericentromeric abnormalities.     

Cytological evidence that the Sxr fragment of XY,Sxr mice pairs homologously at meiotic prophase with the proximal testis-determining region

Self-pairing of the Y chromosome at prophase of meiosis in XY,Sxr male mice appears to take place in many cells to the exclusion of pairing between the Y and the X. This phenomenon offers an explanation for the high level of X-Y separation seen in these males at prophase of meiosis, additional separations being evident, however, when metaphase I (MI) cells are examined. A minority of prophase cells show the Y paired both autologously and with a sub-terminal region of the X which could be the normal pairing region. The balloon-like configurations observed when self-pairing occurs suggest that the distal Sxr fragment is inverted on the Y chromosome of Sxr carrier males in relation to the normal proximal testis-determining (Td)-containing region.     

美味猕猴桃原生质体再生植株细胞遗传学研究 Ⅲ .母株减数分裂前期Ⅰ染色体配对的光镜观察

首次报道在光镜下观察美味猕猴桃 (品种 :No.2 6原生质体植株的母株 )花粉母细胞( PMC)染色体在减数分裂前期的配对 ,发现其配对和凝缩有明显不同步性。不同细胞间染色体配对形式变化较大 ,一般以二价联会为主 ,其次由其它多种配对方式 (包括有复合配对、重复配对、着丝点或端粒处联合和多价联会 )形成多价体 ,还有少数未配对或发生内配对 (偶见 )的单价体和几条二价体之间的次级配对。粗线期观察到少数染色体有缺失 (或重复 )、倒位、易位和疏松配对等结构性改变。表明该植株是一个复杂的区段异源六位体 ,少数染色体在结构上累积有变异。还认为该植株是研究减数分裂染色体配对和联会机制的好材料。