Isolation and Cytogenetic Characterization of Male Meiotic Mutants of Drosophila melanogaster

Corresponding author: Masa-Toshi Yamamoto, Kyoto Institute of Technology, Saga-Ippongi-cho, Ukyo-ku, Kyoto, 616-8354, Japan., yamamoto{at}ipc.kit.ac.jp (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Proper segregation of homologous chromosomes in meiosis I is ensured by pairing of homologs and maintenance of sister chromatid cohesion. In male Drosophila melanogaster, meiosis is achiasmatic and homologs pair at limited chromosome regions called pairing sites. We screened for male meiotic mutants to identify genes required for normal pairing and disjunction of homologs. Nondisjunction of the sex and the fourth chromosomes in male meiosis was scored as a mutant phenotype. We screened 2306 mutagenized and 226 natural population-derived second and third chromosomes and obtained seven mutants representing different loci on the second chromosome and one on the third. Five mutants showed relatively mild effects (<10% nondisjunction). mei(2)yh149 and mei(2)yoh7134 affected both the sex and the fourth chromosomes, mei(2)yh217 produced possible sex chromosome-specific nondisjunction, and mei(2)yh15 and mei(2)yh137 produced fourth chromosome-specific nondisjunction. mei(2)yh137 was allelic to the teflon gene required for autosomal pairing. Three mutants exhibited severe defects, producing >10% nondisjunction of the sex and/or the fourth chromosomes. mei(2)ys91 (a new allele of the orientation disruptor gene) and mei(3)M20 induced precocious separation of sister chromatids as early as prometa-phase I. mei(2)yh92 predominantly induced nondisjunction at meiosis I that appeared to be the consequence of failure of the separation of paired homologous chromosomes.

MEIOSIS consists of two successive cell divisions following a single DNA replication, resulting in the production of haploid cells. Chromosome behavior in meiosis is complex and shows notable differences from that in mitosis. The orderly reduction of chromosome number is accomplished by segregation of homologous chromosomes at meiosis I. Sister chromatids segregate at meiosis II as in mitosis. To ensure proper orientation of chromosomes and the subsequent disjunction in meiotic divisions, two processes are essential: one is the pairing of homologous chromosomes at meiosis I and the other is the maintenance of sister chromatid cohesion at the centromere through metaphase II.

Male meiosis of Drosophila melanogaster is unusual in some respects. Genetic recombination is absent (MORGAN 1912 ) and no chiasmata are formed in bivalents (COOPER 1964 ). Ultrastructural analyses have failed to demonstrate structural entities of meiotic pairing such as the synaptonemal complex between paired homologs (MEYER 1964 ; RASMUSSEN 1973 ; AULT et al. 1982 ; AULT and RIEDER 1994 ). However, homologs pair with each other and segregate regularly to the opposite poles. The mechanism has been studied by determining chromosome regions important for chromosome pairing. GERSHENSON 1933 first pointed out that the centric heterochromatin of the X chromosome is important for sex chromosome meiotic pairing. Only part of the centric X heterochromatin pairs with the Y chromosome. The pairing regions are not evenly distributed throughout the X heterochromatin, but are restricted to particular regions (in blocks hB, hC, and hD; COOPER 1959 , COOPER 1964 ). A mini-X chromosome consisting almost exclusively of hA does not pair with a copy of itself or with the Y chromosome, indicating that sex chromosome pairing requires special chromosome entities called "pairing sites" (YAMAMOTO and MIKLOS 1977 ). APPELS and HILLIKER 1982 and MCKEE and LINDSLEY 1987 proposed that the rDNA region functions as an X-Y pairing site. MCKEE and KARPEN 1990 demonstrated the ability of a single copy rDNA to restore the pairing and disjunction of a heterochromatin-deleted X chromosome. MCKEE et al. 1992 delimited the sequence responsible for pairing to the 240-bp repeats in the nontranscribed region of the genes. Although the repeats function as a pairing site, other X heterochromatin regions in which rDNA is absent also promote X-Y pairing at a certain level. X chromosomes completely deleted for rDNA, such as In(1)sc^4Lsc^8R, pair at a frequency between 55 and 80% depending on the genetic background with the normal Y chromosome (COOPER 1964 ; PEACOCK et al. 1975 ; YAMAMOTO and MIKLOS 1977 ; MCKEE 1996 ). Little is known about the nature of the X chromosome pairing sites other than the 240-bp repeats. Autosomal pairing in males also depends on the homology of limited sites of euchromatin (YAMAMOTO 1979 , YAMAMOTO 1981 ; MCKEE et al. 1993 ).

Meiotic mutants that show high frequencies of nondisjunction would help to clarify the genetic mechanisms of homologous chromosome pairing and sister chromatid cohesion. Previous studies have demonstrated that in Drosophila females and males homologous chromosomes pair and segregate by different mechanisms (reviewed in ORR-WEAVER 1995 ; KARPEN and ENDOW 1998 ). All meiotic mutants that exclusively affect meiosis I exhibit a sex-specific effect. Because, in the male, chromosome-specific pairing sites play a crucial role in the association of the homologs and because nonhomologous pairing is totally absent (YAMAMOTO 1979 ; HILLIKER et al. 1982 ), there must be a mechanism of homolog recognition and holding for each chromosome pair. However, the male-specific meiotic mutants recovered so far that disrupt chromosome pairing, such as mei-O81, mei-1223, and teflon, affect all or a subset of chromosomes rather than just one chromosome pair (SANDLER et al. 1968 ; YAMAMOTO et al. 1993 ; TOMKIEL et al. 2001 ; FLYBASE 2003 ). These mutants emphasize the complexity of meiotic pairing in males and suggest that there must be a common aspect in the genetic control of bivalent formation among all chromosomes (CHURCH and LIN 1988 ).

The other issue to be solved is the mechanism of cohesion of meiotic sister chromatids. Although cohesion along the chromosome arms is lost during meiosis I, centromeric cohesion is maintained until the transition from metaphase II to anaphase II. The meiotic mutants mei-S332 (SANDLER et al. 1968 ; KERREBROCK et al. 1992 ) and orientation disruptor (MASON 1976 ; MIYAZAKI and ORR-WEAVER 1992 ; BICKEL et al. 1997 ) cause nondisjunction owing to premature separation of sister chromatids during meiosis I. These two mutants affect meiosis in both sexes.

The Drosophila genome has not yet been saturated for male meiotic genes, mainly because screenings for such mutants have been carried out only a few times (SANDLER et al. 1968 ; BAKER and CARPENTER 1972 ; GETHMANN 1974 ; YAMAMOTO et al. 1993 ). In this study, we have screened the second and the third chromosomes for male meiotic mutants. We examined flies carrying EMS-treated chromosomes or single P-element insertions and flies collected from natural populations. Nondisjunction of the sex and the fourth chromosomes was assayed. To assay the former, we also used a free minichromosome, Dp(1;f)YP223, which pairs with the compound-XY chromosome and disjoins faithfully during male meiosis (PARK and YAMAMOTO 1993 ). Because this minichromosome carries no 240-bp spacer repeats of rDNA detected by fluorescence in situ hybridization (M.-T. YAMAMOTO, unpublished data) and yet retains the X chromosome pairing site within the part of hB (Fig 1), we designed experiments to isolate mutants that specifically affect the function of this pairing site. Altogether, eight male meiotic mutants were isolated. In this report we describe the genetic and cytological properties of these meiotic mutants.

View larger version (13K): In this window In a new window Download PPT slide   Figure 1. Structure of the normal X chromosome and its derivative, Dp(1;f)YP223. (A) Normal X chromosome. The heterochromatin is divided into four blocks, hA, hB, hC, and hD, and the primary constriction (COOPER 1959 ), under which are indicated the rDNA region and pairing sites required for normal pairing of the sex chromosomes in male meiosis. The primary (strong) sites are in the rDNA region and hB, and the partial (weak) pairing site is in hD. There are no pairing sites in hA and euchromatin (COOPER 1964 ; YAMAMOTO and MIKLOS 1977 ). (B) Dp(1;f)YP223. This minichromosome consists of the centromere of the X chromosome and some of the proximal X heterochromatin plus a portion of the euchromatic tip marked with the yellow+ gene (PARK and YAMAMOTO 1993 ). Dp(1;f)YP223 retains a pairing site located in hB. Solid lines, euchromatin; white blocks, heterochromatin; open circles, centromeres.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Chromosomes:
Description of genetic markers, balancers, deletions, and compound chromosomes used in this work can be found in LINDSLEY and ZIMM 1992 or in FlyBase (http://flybase.bio.indiana.edu/).

Ethyl methanesulfonate-treated chromosomes: EMS-mediated mutagenesis was carried out basically following the method of LEWIS and BACHER 1968 . We used two different concentrations of EMS (0.025 M and 0.013 M in 1% sucrose solution). Lines each carrying a separately mutagenized second and third chromosome were made using balancer chromosomes (CyO for the second chromosome and TM3 for the third chromosome).

P-element-inserted chromosomes: P{lacW} lines were gifts from E. Nitasaka (Kyushu University) and R. Murakami (Yamaguchi University) and P{GS} lines were from T. Aigaki (Tokyo Metropolitan University). P-element-inserted lines were also newly established, using a P{EP} insertion at cytological location 32D, EP(2)2478, provided by H. Kose (Tokushima University). EP(2)2478 was activated in the male germ line by using the TMS chromosome (obtained from the Bloomington Drosophila Stock Center) carrying the 2-3 transposase source.

Chromosomes from natural populations: Iso-female lines established from flies caught in Ishigaki and Iriomote islands (Okinawa, Japan) in 1997 and 1998 and in Katsunuma (Yamanashi, Japan) in 1997 were kindly provided by M. Itoh and T. K. Watanabe (Kyoto Institute of Technology). From each line, balancer chromosomes were used to extract one second chromosome and/or one third chromosome.

Minichromosome: We used the predominantly heterochromatic free X duplication chromosome Dp(1;f)YP223, y⁺ (hereafter referred to as Dp223; Fig 1). Dp223 was generated by deleting a large portion of the In(1)sc^L8Lsc^8R chromosome using X-ray irradiation (PARK and YAMAMOTO 1993 ). Dp223 is about 0.4 times the size of a fourth chromosome (about 2 Mbp DNA) and retains the proximal half of hA and the distal half of hB. Dp223 pairs with the compound-XY chromosome C(1;Y)6 (XY^L·Y^S, y² sc cv v f) and they disjoin from one another normally in male meiosis (see Table 2). Because hA does not have pairing ability (YAMAMOTO and MIKLOS 1977 ), Dp223 must pair with the compound-XY only by the pairing site located in the half of hB. We consider this pairing site to be different from the 240-bp rDNA spacer repeats, because the repeats could not be detected by fluorescence in situ hybridization (M.-T. YAMAMOTO, unpublished data). To search for a mutation that specifically affects the minichromosome pairing, we compared segregation of the sex chromosomes in the males carrying C(1;Y)6 and Dp223 to those carrying the normal X and Y chromosomes.

Meiotic mutants: The following previously known meiotic mutants, mei-S332¹ (2-99.5, 59D), orientation disruptor¹ (ord; 2-103.5, 58B), and Df(2R)PC4 that uncovers the subito locus (2-82.6, 54D-54F), which were in the collection of the Drosophila Genetic Resource Center in Kyoto, were used in complementation tests. An allele of the teflon gene (2-80.0, 53F-54A), tef^Z5549 (an EMS-induced nonsense mutation; J. E. TOMKIEL, personal communication), and Df(2R)P803-15 deleted for the tef locus were kindly provided by J. E. Tomkiel (University of North Carolina, Greensboro, NC).

Genetic analyses:
Flies were reared on a standard cornmeal-glucose-yeast-agar medium at 24 ± 1°. Three- to 5-day-old females and males were used for each cross. The parents were transferred to new vials on day 3 (day 0 is the day of setup) and then were discarded on day 7. Progeny were scored on days 12 and 17 after the establishment of the cross for each vial.

Initial screenings were made by examining 5 single-pair matings for each line. Candidate lines (>2% nondisjunction, as defined below) were then rescreened by examining 10 single-pair matings. Those that behaved consistently were maintained and examined in this study.

Nondisjunction tests were carried out as described below. For simultaneous examination of sex and fourth chromosome segregation in C(1;Y)6, y² sc cv v f/Dp223, y⁺; spa⁺/spa⁺ males, these flies were crossed to y/y; C(4)RM, spa^pol/O females. Single-pair matings were performed.

The sex chromosomes [C(1;Y)6, y² sc cv v f/Dp223, y⁺]: The tester females produce haplo-X (y) ova. The sperm produced by normal segregation bear either C(1;Y)6 or Dp223, which results in y² females and y⁺ males, respectively. The sperm produced by nondisjunction at meiosis I bear either both C(1;Y)6 and Dp223 or neither of them, resulting in y⁺ female and y male progeny, respectively. In this cross, exceptional sperm bearing two compound-XY chromosomes could not be recovered and those bearing two Dp223 chromosomes were phenotypically indistinguishable from regular ones. Nondisjunction frequency of the sex chromosomes was calculated as [(y⁺ females + y males) x 100/total].

The fourth chromosome: The tester females produce compound-4, [C(4)RM, spa^pol]-bearing, and nullo-4 ova. The sperm produced by normal segregation result in spa⁺ progeny, trisomy-4, and monosomy-4, respectively. Nondisjunction of the fourth chromosome at meiosis I results in two classes of sperm, diplo-4 and nullo-4. Progeny showing spa^pol phenotype are clearly the descendants of the nullo-4 sperm, indicating meiotic nondisjunction of the fourth chromosome. Progeny that arose from diplo-4 sperm are indistinguishable from regular ones. Flies lacking fourth chromosomes are inviable and those carrying a single fourth chromosome are weak and show strong Minute phenotype. Monosomy-4 progeny are subviable. Although the haplo-4 Minutes were counted (see Table 2 and Table 3), they were excluded from any calculations because viability varied between females and males. On the assumption that all exceptional sperm result from meiosis I nondisjunction and that those bearing diplo-4, triplo-4, and tetra-4 show equivalent viability, nondisjunction frequency was calculated as [spa^pol progeny x 2 x 100/spa⁺ progeny]. In the case of mei(2)yh137/tef^Z5549 and mei(2)yh137/Df(2R)P803-15, spa^pol/spa⁺ males were crossed with C(4)RM, ci ey^R/O females. Nondisjunction frequencies were calculated on the basis of the number of ci ey^R progeny derived from nullo-4 sperm.

To determine which meiotic division is disrupted in a given meiotic mutant, y w/y⁺Y males were crossed singly to females carrying a compound-X chromosome [C(1)RM, y v f/O]. In this cross all classes of sperm, two regular and seven exceptional, can be recovered upon fertilization with either C(1)RM (diplo-X) or nullo-X ova produced by the tester females. Normal segregation produces X-bearing and Y-bearing sperm. Nondisjunction at meiosis I produces XY and nullo-XY sperm and at meiosis II produces XX, YY, and nullo-XY sperm. Three classes of sperm, XXY, XYY, and XXYY, are diagnostic of nondisjunction in both meiosis I and meiosis II. If complete loss of sister chromatid cohesion occurs, all seven classes of exceptional sperm, as well as the regular two, will be produced. All classes were distinguished by the phenotypes of zygotes, except sperm bearing one Y chromosome and those bearing two. Provided that sufficient numbers of exceptional progeny are obtained in crosses using a mutant stock, we were able to distinguish whether the mutant affected predominantly, if not wholly, meiosis I or meiosis II. While metafemales [C(1)RM/X; A/A, where A represents a set of autosomes] usually die as larvae, escapers can be identified by their characteristic phenotype. Metafemales, triploid females [C(1)RM/X(/Y); A/A/A] and triploid intersexes [C(1)RM(/Y); A/A/A] were scored but were omitted from the table because of their highly variable recovery. The nondisjunction frequency was calculated as [exceptional progeny x 100/total].

In females, nondisjunction of the X and the fourth chromosomes was assayed separately. For the X chromosome, y/y females were crossed with y pn/B^SY males. Females were tested individually for mei(2)yh15 and mei(2)yh149, while semisterile females of mei(2)ys91 (ord^ys91) and mei(3)M20 were tested in mass matings. Regular X (y) ova yielded y females and y B^S males. Two classes of exceptional ova were recoverable in this cross. Diplo-X ova that were fertilized by Y (B^S) sperm and nullo-X ova that were fertilized by X (y pn) sperm were recovered as y B^S females and y pn males, respectively. The nondisjunction frequency was calculated as [(y B^S females + y pn males) x 2 x 100/(total + y B^S females + y pn males)]. For fourth chromosome segregation spa⁺/spa⁺ females were crossed with C(4)RM, spa^pol/O males and examined as described above for males.

Cytology:
We made meiotic chromosome preparations without colchicine treatment using the air-dry procedure (YAMAMOTO 1992 ; YAMAMOTO et al. 1993 ). We used testes of 0- to 3-day-old adults. We stained the chromosomes with 4',6-diamidino-2-phenylindole (DAPI) or Giemsa. We scored prometaphase I and metaphase I cells for the sex, second, and third chromosomes, but not for the fourth chromosome, since it is not always visible due to its small, dot-like morphology. y w/y⁺Y males were used as the control.

Embryo preparation:
Females mated with y w/y⁺Y males were allowed to lay eggs on apple juice agar plates for 3 hr. Eggs were dechorionated in 50% bleach 2–3 hr later. Vitelline membrane permeabilization in heptane and fixation and devitellinization in a mixture of methanol/heptane were performed before staining with DAPI following the procedures described in ROTHWELL and SULLIVAN 2000 . In separate tests, eggshell morphology was examined under the dissection microscope. All flies and embryos were kept at 24 ± 1°.

Inverse PCR:
Genomic DNA preparation, restriction enzyme digestions, ligations, and inverse PCR were performed essentially following the protocol of HUANG et al. 2000 . Purified DNA was digested with MspI, which makes cuts within the P{GS} vector sequence as well as in the 5' flanking sequence. Following self-ligation, it was PCR amplified with primers 5' CTGAATAGGGAATTGGGAATTCGACTAGTT and 5' CTCCGTAGACGAAGCGCCTCTATTT. The product was then directly sequenced using ABI310 sequencer (Perkin-Elmer, Norwalk, CT) with a BigDye Terminator cycle sequencing kit (Applied Biosystems, Foster City, CA) and a sequencing primer, 5' CACTGAATTTAAGTGTATACTTCGG.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Screening:
We first generated 1100 EMS-treated second chromosome lines and 913 such third chromosome lines. Of these, 493 second and 331 third chromosome lines that were homozygous viable and fertile in the male were screened for the presence of male meiotic mutants with increased nondisjunction frequencies. The males tested here carried a compound-XY chromosome [C(1;Y)6 = XY^L·Y^S] and a free miniduplication chromosome (Dp223), because an aim was to recover mutations that might specifically affect the sex chromosome pairing site carried by the Dp223 chromosome (see Fig 1). C(1;Y)6, y²/Dp223, y⁺; spa⁺/spa⁺ males were single-pair mated to y/y; C(4)RM, spa^pol/O females. This cross makes it possible to examine the segregation of the sex and the fourth chromosomes simultaneously. Three mutants were recovered. These were mei(2)yh92, mei(2)yh149, and mei(2)yh217 (Table 1). Using the same mating scheme, we next screened 67 second and 33 third chromosome lines from the Ishigaki and Iriomote natural populations. Two mutants, mei(3)M19 and mei(3)M20, were obtained (Table 1).

We also screened 1482 second and third chromosome P-element-insertion lines. Males that had unmarked X and Y chromosomes and were homozygous for a P-insertion chromosome were single-pair mated to y/y; C(4)RM, spa^pol/O females. Here only fourth chromosome nondisjunction could be detected. Two mutants, mei(2)ys91 and mei(2)yoh7134, were recovered (Table 1). Similarly, 126 second chromosomes from the Katsunuma natural population were screened and two mutants, mei(2)yh15 and mei(2)yh137, were recovered (Table 1).

For the sake of comparison, the mutants recovered solely on the basis of abnormal segregation of the fourth chromosome (Table 1) were reexamined in the mating scheme by which the other mutants (Table 1) were obtained. Table 2 shows the results for eight of the nine mutants recovered. The remaining mutant, mei(2)ys91, was not included in Table 2 since it was shown to be an allele of the well-characterized meiotic mutant orientation disruptor (ord), as described below. The two third chromosomal mutants, mei(3)M19 and mei(3)M20, showed the same high frequencies of nondisjunction of the sex and the fourth chromosomes. Since they were recovered from the same Ishigaki natural population and since they showed the same high frequencies of nondisjunction in mei(3)M19/mei(3)M20 males, they most probably are two independent isolates of the same mutation. We chose mei(3)M20 for further characterization.

Table 3 shows the results of reexamination of all eight mutants recovered for segregation of the X and the Y chromosomes (instead of the compound-XY and Dp223) and of the fourth chromosomes by crossing y/y⁺Y; spa⁺/spa⁺ males to y/y; C(4)RM, spa^pol/O females. Nondisjunction frequencies were calculated in a manner similar to that employed for C(1;Y)6/Dp223; spa⁺/spa⁺ males. As can be seen in Table 2 and Table 3, each mutant examined showed similar nondisjunction frequencies in both C(1;Y)6/Dp223 and X/Y males. Three mutants, mei(2)ys91, mei(2)yh92, and mei(3)M20, may be called severe meiotic mutants, producing >10% nondisjunction. They all affected the segregation of both the sex and the fourth chromosomes. The other five mutants recovered may be called mild meiotic mutants, producing <10% nondisjunction. Some of the mild mutants exhibited a possible chromosome-specific effect.

The mei(2)yh92 gene was mapped to 2-40.9 (202 recombinants between Sternopleural and Tufted were scored), and mei(3)M20 was mapped to 3-40 (37 recombinants between Roughened and Dichaete were scored). A mild mutant mei(2)yoh7134 was shown to be induced by a P-element insertion (see below) and mapped to the cytological interval 37A4-6 by inverse PCR. The remaining EMS-induced and natural population-derived mutants were difficult to map because they gave only mild nondisjunction. Complementation tests among the seven meiotic mutations on the second chromosome were carried out. In all cases, nondisjunction frequencies of the sex and the fourth chromosomes did not differ from the values observed in the respective heterozygous controls (data not shown). Thus the seven meiotic mutations represent different genes on the second chromosome.

We next asked if these second chromosomal mutants were allelic to the four male meiotic genes previously reported, mei-S332 (SANDLER et al. 1968 ), ord (MASON 1976 ), subito (GIUNTA et al. 2002 ), and teflon (tef; TOMKIEL et al. 2001 ). Two mutants recovered in this study, mei(2)ys91 and mei(2)yh137, were found to be new alleles of ord and tef, respectively (see below). The remaining five mutants were shown not to be allelic to any of the genes already known. Further, there is no known meiotic mutant around the mei(3)M20 gene on the third chromosome. Thus the male meiotic mutants isolated in this study, five on the second and one on the third, represent novel genes. Below we present the results of genetic analyses of each of these mutants and cytological analyses for the three severe mutants and a tef allele, mei(2)yh137.

Severe mutants:
mei(2)ys91: The mutation, recovered from a P-element-insertion line, was found to be allelic to ord, which is known to be required for normal sister chromatid cohesion in meiosis (MASON 1976 ; MIYAZAKI and ORR-WEAVER 1992 ; BICKEL et al. 1997 ). A heteroallelic combination, mei(2)ys91/ord¹, produced 26.2% nondisjunction between the X and Y chromosomes (total 237). The severity of nondisjunction seen in the homozygotes and the hemizygotes (over a deficiency) was equivalent (Table 3). We cytologically examined and confirmed the ord phenotype in primary spermatocytes of mei(2)ys91. In the control prometaphase I cells, sister chromatids were attached to each other at the centromeres (Fig 2A). In the mutant cells, precocious separation of sister chromatids was observed for all chromosomes. Fig 2B shows an example of mei(2)ys91 mutant cells in which the early separation in autosomes is evident. We also observed severely reduced fertility associated with this mutant [ Table 3, compare N (number of pair matings) and Total (total number of M⁺ progeny)], most probably reflecting the production of aneuploid sperm for the second and third chromosomes at high frequencies, which should generate lethal aneuploid zygotes. We will designate mei(2)ys91 as ord^ys91. ord mutants have been shown to cause higher levels of nondisjunction in meiosis I than in meiosis II (MIYAZAKI and ORR-WEAVER 1992 ).

View larger version (22K): In this window In a new window Download PPT slide   Figure 2. Chromosome behavior in mutant meiosis I cells. (A) Wild type. (B) Homozygous mei(2)ys91 (ordys91). (C) Homozygous mei(2)yh92. (D) Homozygous mei(3)M20. (E) Hemizygous mei(2)yh137 (tefyh137). (A, B, D, and E) Prometaphase I. (C) Metaphase I. Chromosomes were stained with DAPI. In the control, homologous chromosomes are paired and sister chromatids are attached to each other at the centromeres (arrows in A). Arrows in B and D point to prematurely separated chromatids. Bivalents are of normal appearance and congress properly in mei(2)yh92 (C). Univalents of a major autosomal homologous pair are observed in the cell mutant for the teflon gene (arrows in E). X, X chromosome; Y, Y chromosome; A, second or third chromosome; 4, fourth chromosome. Bars, 10 µm.

mei(2)yh92: The EMS-induced mutant mei(2)yh92 showed relatively high nondisjunction (20%). To examine which meiotic division this mutant affects, y w/y⁺Y; mei(2)yh92/mei(2)yh92 males were mated to compound-X females, in which XX sperm that result from nondisjunction in meiosis II as well as XY sperm that result from nondisjunction in meiosis I are detectable by using the genetic markers we employed. Essentially all of the exceptional sperm were the XY and nullo-XY classes, although a small number of the XX and XXY classes were produced (Table 4). We conclude that mei(2)yh92 predominantly disrupts meiosis I.

Cytological examinations of the mutant showed that homologous chromosomes faithfully paired in prometaphase-metaphase I cells (Fig 2C, number of cells examined is 405). We noticed characteristic abnormalities in chromosome behavior in late stages of meiosis I. In the control, anaphase I cells showed synchronous chromosome movement to each pole (Fig 3A). In mei(2)yh92, however, some pairs of chromosomes showed a delay in migration to the poles or remained in the vicinity of the equator in anaphase I cells, while others had already moved a considerable distance to the poles (Fig 3B). Such disrupted chromosome segregation was observed in 43% of the anaphase I cells (n = 83), whereas it was never observed in the control cells (n = 54). The mutant phenotype can be characterized by the presence of daughter nuclei connected by a thin chromatin bridge (Fig 3C) or of nuclei associated with chromatin trailing behind at telophase I (Fig 3D). The chromatin lagging is likely a cytological basis of the nondisjunction induced by mei(2)yh92. The mutant shows a unique defect in separation of paired homologous chromosomes at the onset of anaphase I.

View larger version (72K): In this window In a new window Download PPT slide   Figure 3. Aberrant chromosome segregation in mei(2)yh92 meiosis I. (A) Control. (B, C, and D) mei(2)yh92. (A and B) Anaphase I. (C and D) Telophase I. Chromosomes were stained with Giemsa. (A) Chromosomes migrate to each pole synchronously. (B) A cell showing asynchronous chromosome segregation. The large autosomes have moved to the poles. The sex chromosomes (solid arrows) and the fourth chromosomes (open arrows) show delays in migration. Note that the bivalent sex chromosome remaining in the vicinity of the equator appears to be oriented to the poles. (C) A thin chromatin bridge connecting daughter nuclei is visible. (D) The upper nucleus is associated with chromatin trailing behind. Bar, 10 µm.

mei(3)M20: The natural population-derived mutant mei(3)M20 induced 40% nondisjunction of the sex and the fourth chromosomes (Table 2 and Table 3). Genetic analysis similar to that described above for mei(2)yh92 indicated that mei(3)M20 primarily affected meiosis I, but the proportion of meiosis II nondisjunctional sperm was significantly higher in this mutant than in the other mutants (Table 4). Not included in Table 4 is the appearance of triploid intersex progeny [nine C(1)RM/y⁺Y; A/A/A and two C(1)RM/O; A/A/A] from the cross. Their occurrence indicates that the mei(3)M20 mutation also affects the disjunction of the second and the third chromosomes at high frequencies.

Cytological analysis revealed that, although homologous chromosomes were apparently paired, sister centromeres were prematurely separated as early as prometaphase I in mei(3)M20 cells (Fig 2D). Although mei(3)M20 bears close resemblance to ord with respect to these phenotypes, the two genes are located on different chromosomes. Some ord alleles have been shown to disrupt mitotic segregation in the gonial cells, resulting in aneuploidy of primary spermatocytes (LIN and CHURCH 1982 ; MIYAZAKI and ORR-WEAVER 1992 ). However, this was not the case for at least this allele of the mei(3)M20 gene. In >100 mutant prometaphase-metaphase I cells examined, all chromosomes paired as bivalents, and neither univalents nor trivalents were observed.

Mild mutants:
The five mild mutants were subdivided into three categories defined by the chromosomes showing nondisjunction. Two mutants—mei(2)yoh7134, which was recovered from P-element-insertion lines (TOBA et al. 1999 ), and mei(2)yh149, which was induced with EMS—affected both the sex and the fourth chromosomes (Table 2 and Table 3). P-element-excision experiments were performed for mei(2)yoh7134. Males carrying a P-element-excised chromosome, when examined in the mating scheme employed in Table 3, did not produce nondisjunction at an appreciable frequency (total 512). Thus the mutation was clearly caused by the P-element insertion. This mutant exclusively affected meiosis I because mei(3)yoh7134 males produced XY sperm but not XX sperm (Table 4). In contrast, mei(2)yh149 induced either nondisjunction at meiosis II or chromosome loss at meiosis I and/or II, because no XY sperm were recovered (Table 3 and Table 4).

The EMS-induced mutant mei(2)yh217 elicited a stronger effect on the sex chromosomes than on the fourth chromosome in C(1;Y)6/Dp223 males (Table 2), suggesting the possibility that this mutant specifically impairs the function of the pairing site in hB (Fig 1). If this is true, X-Y segregation might be compensated by other pairing sites on the X chromosome. However, nondisjunction took place between the X and the Y chromosomes (Table 3). Thus it is unlikely that the effect is specific to the pairing site in hB. This mutant was found to cause meiosis I-specific nondisjunction (Table 4). Another striking feature of mei(2)yh217 was meiotic drive, that is, a discrepancy in the recovery of reciprocal products of meiotic segregation (Table 3). The recovery of X-bearing sperm (1816) exceeded that of Y-bearing sperm (1175), and recovery of nullo-XY sperm (77) exceeded that of XY sperm (7). The meiotic drive coefficients [X/(X + Y) and O/(XY + O)] were 0.61 and 0.92, which were significantly different from those of the controls, +/+ (0.50 and 0.67) and mei(2)yh217/SM1 (0.52 and 0.80), and from those of the other mutants that predominantly caused meiosis I nondisjunction, mei(2)yh92 (0.51 and 0.61) and mei(3)yoh7134 (0.51 and 0.75). Consistent results were obtained in separate tests (Table 4).

Two mutants derived from Katsunuma natural population, mei(2)yh15 and mei(2)yh137, caused nondisjunction of the fourth chromosome but not of the sex chromosomes (Table 2 and Table 3). mei(2)yh15 represents a new locus as noted above, but was not examined further because of its only mild effect on nondisjunction. The mei(2)yh137 mutation turned out to be allelic to a known male-specific meiotic gene, tef, which is required for the maintenance of homolog pairing (TOMKIEL et al. 2001 ). This gene has been characterized as having an autosome-specific effect. mei(2)yh137/tef^Z5549 males showed an increased frequency of fourth chromosome nondisjunction (6.7%) compared to the value (4.6%) observed in homozygous mei(2)yh137, while sex chromosome segregation was unaffected (Table 3). Similar results (9.7% fourth chromosome nondisjunction, total 1100) were obtained in the male of mei(2)yh137/Df(2R)P803-15. Df(2R)P803-15/+ males showed normal segregation (0.7% fourth chromosome nondisjunction, total 578), indicating no dominant effect of tef. The increased level of nondisjunction prompted us to carry out cytology. We observed univalents of autosomes at meiosis I in mutant cells (Fig 2E). Either one or both of the major autosomes were evidently unpaired in 12/102 cells hemizygous for mei(2)yh137. The sex chromosome pairing was intact. Such a defect in autosomes was never observed in 265 control cells examined (Fig 2A). We can infer from the genetic result that the fourth chromosomes also fail to form a stable bivalent at an increased frequency in mutant cells. Frequencies of fourth chromosome pairing were not scored because, when the punctiform chromosomes are visible, they frequently appear to be unconjoined even in control cells, although they are generally close to each other relative to other chromosomes (GUYENOT and NAVILLE 1929 ; GOLDSTEIN 1980 ). We will designate mei(2)yh137 as tef^yh137.

Effects of male meiotic mutants on female gamete:
We examined the effect of male meiotic mutants on female fertility: mei(2)yh15 and mei(2)yh149 were fully fertile; ord^ys91 and mei(3)M20 were semisterile; and mei(2)yh137 (tef^yh137), mei(2)yh92, mei(2)yh217, and mei(2) yoh7134 were completely sterile. We tested the two fertile mutants for their X and fourth chromosome segregations at meiosis (see MATERIALS AND METHODS). Control females of the genotype y w/y w; spa⁺/spa⁺ produced no nondisjunction in the X and the fourth chromosomes (total 2724 and 1760, respectively). In females homozygous for mei(2)yh15 or mei(2)yh149, nondisjunction frequencies were comparable to those of the control: 0.1% X chromosome and no fourth chromosome nondisjunction in mei(2)yh15 (total 1545 and 997, respectively) and no nondisjunction in either chromosome pair in mei(2)yh149 (total 1799 and 1227, respectively). Thus, these two meiotic mutants are male specific.

When mated with wild-type males, females homozygous for ord^ys91 and for mei(3)M20 laid a large number of eggs. Gross morphology of the eggs was normal. Hatchability of the eggs laid by ord^ys91 females (36/635) and mei(3)M20 females (35/657) was 5%. X chromosome nondisjunction of ord^ys91 and mei(3)M20 was estimated to be 40.0% (total 953) and 52.2% (total 408), respectively. No gynandromorph (X/X-X/O) was produced. We also examined embryonic development by staining with DAPI. The final preparation included embryos 2–6 hr old. Development beyond the syncytial blastoderm stage was observed in 97.1% embryos from the control females (number of embryos examined is 239), but only 57.7% embryos from ord^ys91 females (total 286) and 51.0% embryos from mei(3)M20 females (n = 204) reached the stage. Nondisjunction in mei(3)M20 females and males appears to be caused by the same mutation, because the abnormalities were also manifested when mei(3)M20 was placed over a chromosomal deficiency Df(3L)vin6 (data not shown), although the possibility that two separate mutations are closely linked and located within the deletion (68C8-11; 69A4-5) remains.

Sterility of mei(2)yh137 (tef^yh137) females is caused by another mutation on the same chromosome, because females tef^yh137/Df(2R)P803-15 restored fertility. The tef mutation has been shown to have no effect on female meiosis (TOMKIEL et al. 2001 ). No further examination of mei(2)yh137 females was thus performed. The remaining three mutants, mei(2)yh92, mei(2)yh217, and mei(3)yoh7134, laid a large number of eggs with normal eggshell morphology. In >100 embryos examined for each genotype, nuclear divisions became abnormal by the syncytial blastoderm stage. We do not have any evidence to demonstrate that the female sterility of mei(2)yh92 and mei(2)yh217 is caused by the same mutation responsible for male meiotic nondisjunction. The P-element-excision experiment for mei(2)yoh7134 mentioned above showed that the female sterility was also attributable to the P insertion.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We screened 2532 second and third chromosome lines, derived from mutagenesis or from natural populations, that were homozygous viable and fertile in the male, and recovered nine male meiotic mutants. The efficiency was 3/824 for EMS-treated chromosomes, 4/226 for chromosomes from natural populations, and 2/1482 for P-element insertions. The rates were nearly equivalent to those reported previously: 2/160 for EMS-treated chromosomes (GETHMANN 1974 ), 4/423 for chromosomes from natural populations (SANDLER et al. 1968 ), and 41/18,558 for P-element insertions (SEKELSKY et al. 1999 , data on female meiotic mutants). Seven of the mutants recovered in this study were located on the second chromosome, each representing a different locus. Only one mutant [two lines, mei(3)M19 and mei(3)M20] was on the third. Among the known autosomal male meiotic mutants, eight are on the second chromosome [teflon (tef, thought to be allelic to the extinct mei-S8), mei-O81, mei-G17, mei-S332, orientation disruptor (ord), mei-G87, subito (sub), and sting] and three are on the third chromosome [mei-1223 (mei-I1), mei-I3, and homeless (hls); FLYBASE 2003]. Because six of eight newly recovered mutants represent new loci, further screenings are needed to fulfill a set of genes required for male meiosis in D. melanogaster.

In this study we examined the segregation of the sex and the fourth chromosomes, but not of the second and the third chromosomes, in male meiosis. mei(2)ys91 (named ord^ys91), mei(2)yh92, and mei(3)yoh7134 disrupted both sex and fourth chromosome segregation. Preferential effects on specific chromosomes were observed in other mutants: mei(2)yh217 showing a higher rate of nondisjunction in the sex chromosomes and mei(2)yh15 and mei(2)yh137 (named tef^yh137) showing a higher rate of nondisjunction in the fourth chromosome. Because, in male meiosis, pairing of homologs is mediated by chromosome-specific pairing sites, one can expect to recover pairing-defective mutants that exhibit a chromosome-specific effect. However, it has not been shown that Drosophila spermatocytes have such a chromosome-specific control of pairing by trans-acting proteins. Among meiotic genes analyzed to date, only two, mei-1223 and tef, function specifically in the process of homologous chromosome pairing in meiosis of the male. All chromosome complements are affected in mei-1223 mutant cells, albeit with different frequencies (YAMAMOTO et al. 1993 ). tef exclusively affects autosomes but not the sex chromosomes (TOMKIEL et al. 2001 ; this study). It has been suggested that homologous pairing in male meiosis is related to the somatic pairing observed in spermatogonial cells (COOPER 1950 ; VAZQUEZ et al. 2002 ). However, the mei-1223^m144 and the tef^Z5549 mutants, which severely disrupt meiotic pairing of all autosomes, have no detectable somatic pairing defects (K. HIRAI and M.-T. YAMAMOTO, unpublished observations). Regulatory mechanisms of homologous chromosome pairing mediated by chromosome-specific pairing sites remain obscure.

Three mutants described here, mei(2)yh217, mei(2)yh15, and tef^yh137, showed potential chromosome specificity, although the effects were mild. BAKER and CARPENTER 1972 recovered 20 mutants on the X chromosome that induced sex chromosome-specific nondisjunction at rates of <10%. It should be mentioned that all male meiotic mutants causing significant anomalies in pairing and segregation have never been shown to affect any specific chromosomes. Thus we would postulate that the chromosome specificity in homologous chromosome pairing in the male of Drosophila may be generated by the results of an additive effect of multiple genes with mild influences.

To obtain mutants that exhibit a specific effect on the pairing site in hB carried by Dp223, but not on the other pairing sites of the X chromosome, we compared nondisjunction frequencies of the sex chromosomes between compound-XY/Dp223 and X/Y males. Such mutants, if isolated, would have altered disjunction between the compound-XY and Dp223 chromosomes (Table 2), but not between the normal X and Y chromosomes (Table 3). One mild mutant, mei(2)yh217, did produce possible sex chromosome-specific nondisjunction in meiosis I, but it showed the same levels of nondisjunction in both compound-XY/Dp223 and X/Y. mei(2)yh217 may thus be a mutation affecting all pairing sites on the X chromosome or may be involved in a meiotic process other than pairing. The Stellate (Ste) elements (arrays of partially homologous and tandemly repeated sequences with an open reading frame encoding a 19,500-D protein) are expressed when the Suppressor of Stellate on the Y (also composed of tandemly repeated sequences) is deleted (LIVAK 1984 , LIVAK 1990 ). The expression of Ste causes, among other effects, chromosome nondisjunction in male meiosis. Nondisjunction is observed for the sex, second, and third chromosomes but not for the fourth chromosome (HARDY et al. 1984 ). How this type of chromosome-specific nondisjunction occurs is not known. Meiosis is disrupted after the formation of bivalents (PALUMBO et al. 1994 ). Expression of the Ste elements is also regulated by other mutants such as hls. In the hls mutant males nondisjunction is produced in both meiotic divisions (STAPLETON et al. 2001 ), whereas it is restricted to meiosis I in mei(2)yh217 males. Double or nothing is an additional example of meiotic mutant in which nondisjunction frequency of the X and Y chromosomes is appreciably higher than that of the fourth chromosome (MOORE et al. 1994 ). This mutation is an antimorphic allele of the sub gene encoding a kinesin motor protein required for normal spindle assembly (GIUNTA et al. 2002 ).

We also determined which meiotic division was defective in each recovered meiotic mutant by examining sex chromosome segregation. Three mutants, mei(2)yh92, mei(2)yh217, and mei(3)yoh7134, cause meiosis I nondisjunction almost exclusively. Normal function of these genes may be involved in homologous chromosome behavior. Two mutants, ord^ys91 and mei(3)M20, disrupted both meiotic divisions, suggesting sister chromatid cohesion defects. Because almost all exceptional sperm produced by mei(2)yh149 males were nullo-XY class, simple chromosome loss during meiotic divisions may be the major cause of the nondisjunction.

Cytological examination as well as gene mapping were plausible for the severe effect mutants, mei(2)yh92, mei(2)ys91, and mei(3)M20. mei(2)yh92 caused nondisjunction almost exclusively at meiosis I, although homologous chromosomes were paired as normal (Fig 2C). Double staining with a DNA dye and an anti--tubulin monoclonal antibody for chromosomes and microtubules, respectively, showed that meiotic spindles were morphologically normal and bivalents were normally aligned on the metaphase plate (data not shown). The mei(2)yh92 mutation interferes with the fidelity of meiosis I disjunction. When homologous chromosomes begin to move to opposite poles in anaphase I, paired chromosomes are lagged behind (Fig 3B). The lagging chromosome pairs eventually reached both poles (normal disjunction) or a single pole (nondisjunction; Fig 3D). Thus, the ability of kinetochores to bind microtubules appears to be preserved in the mutant cells. Rather, the separation of paired homologs is defective in mei(2)yh92. The normal function of the mei(2)yh92 gene product may thus be involved in proteolysis or dispersal of presumptive adhesive proteins at the pairing sites, for example "segregation bodies" (WOLF 1994 ), in a direct or indirect manner. Alternatively, it may be that transportation of homologous chromosomes upon dissolution of pairing is perturbed in mei(2)yh92 anaphase I cells. Alignment of bioriented chromosome pairs at the metaphase plate is accomplished by the integration of antagonistic poleward forces and antipoleward forces, exerted by microtubule dynamics and microtubule-based motor proteins on the kinetochores and along the chromosome arms. Anaphase onset is permitted by downregulation of the antipoleward forces as well as disassociation of partner chromosomes (reviewed in MCINTOSH et al. 2002 ; CLEVELAND et al. 2003 ; SCHOLEY et al. 2003 ). mei(2)yh92 cells may thus be abnormal in continuous production of antipoleward forces during anaphase, blocking the movement of individualized homologous chromosomes to opposite poles. The predominant, but not exclusive, effect of this mutation on meiosis I could be explained if a larger amount of the gene product is needed to align a bivalent (tetrad) in meiosis I cells than to align a dyad in meiosis II cells, and/or if a redundant pathway exists enabling segregation of sister chromatids in the mutant meiosis II cells.

The remaining two mutants, ord^ys91 and mei(3)M20, are both deficient in meiotic sister chromatid cohesion. The genes ord (MASON 1976 ; GOLDSTEIN 1980 ; LIN and CHURCH 1982 ; MIYAZAKI and ORR-WEAVER 1992 ; BICKEL et al. 1996 , BICKEL et al. 1997 , BICKEL et al. 2002 ; BALICKY et al. 2002 ; this study) and mei-S332 (SANDLER et al. 1968 ; GOLDSTEIN 1980 ; KERREBROCK et al. 1992 , KERREBROCK et al. 1995 ; MOORE et al. 1998 ; TANG et al. 1998 ) are known to be defective in meiotic sister chromatid cohesion. Although a majority of the previously known Drosophila meiotic mutations affect only one sex, mutations in the ord and mei-S332 genes result in nondisjunction in both sexes (reviewed in ORR-WEAVER 1995 ). Similarly, mei(3)M20 causes nondisjunction in both sexes. Meiotic sister chromatid cohesion, not homologous chromosome pairing, should depend on a common mechanism in the two sexes.

In wild-type spermatocytes, separation of sister chromatids along the chromosome arms occurs in mid-G₂, but centromeric cohesion is maintained throughout meiosis I (BALICKY et al. 2002 ; VAZQUEZ et al. 2002 ). The ord and mei-S332 genes differ from each other in the stages at which precocious separation of sister centromeres occurs. Cohesion defects become detectable in late-G₂ in ord cells but in late anaphase I in mei-S332 cells. This difference explains the result of genetic analyses that nondisjunction takes place in both meiosis I and meiosis II in ord but primarily in meiosis II in mei-S332. The ORD and MEI-S332 proteins are essential to maintain centromeric cohesion until the onset of anaphase II. These proteins have no known homologs in other organisms (KERREBROCK et al. 1995 ; BICKEL et al. 1996 ).

The phenotype of the newly recovered mei(3)M20 mutant is similar to that of ord mutants. Prematurely individualized chromatids are observed in prometaphase I cells (Fig 2B and Fig D). Spermatogonia (data not shown) and meiosis I cells (Fig 2D) mutant for mei(3)M20 carry the normal complement of chromosomes, indicating no effect of mei(3)M20 on premeiotic mitosis. The ord alleles, ord¹ and ord², examined in homozygotes have been shown to cause nondisjunction not only in spermatocytes but also in spermatogonia (LIN and CHURCH 1982 ; MIYAZAKI and ORR-WEAVER 1992 ). However, such is not the case for the null allele ord¹⁰ (either over a chromosomal deficiency or over a strong allele of ord⁵; BICKEL et al. 1997 ). Complete loss of ORD activity is more similar to the mei(3)M20 mutation.

A complex of proteins, the cohesin complex, establishes a structural link between sister chromatids during S phase in mitosis and meiosis (reviewed in LEE and ORR-WEAVER 2001 ). The release of the cohesin complex from chromosomes in mitosis is permitted by the cleavage of one of the cohesin subunits, SCC1/MCD1/RAD21, at the transition from metaphase to anaphase. In meiosis this subunit is largely replaced with meiosis-specific REC8, which is not cleaved at the centromeres in meiosis I (KITAJIMA et al. 2003 ). This major effector of meiotic cohesion is known from diverse species, from fission yeast to humans (PARISI et al. 1999 ), but it has not been identified in Drosophila. How the cohesion of meiotic sister chromatids is established in Drosophila has not been understood. The Drad21 gene appears to reside in heterochromatin of an unidentified chromosome (VASS et al. 2003 ). The mei(3)M20 gene is mapped to the left arm of the third chromosome. Thus, mei(3)M20 is not an allele of mitotic Drad21 with strong meiotic effects in surviving mutant individuals. Indeed, we observed no effect of the mei(3)M20 mutation on viability from the egg to the adult (data not shown). The phenotypes of ord (MIYAZAKI and ORR-WEAVER 1992 ; BICKEL et al. 1997 , BICKEL et al. 2002 ; this study) and of mei(3)M20 suggest involvement of these gene products in the establishment as well as the maintenance of meiotic sister chromatid cohesion. However, ORD appears to play roles in chromatin condensation and maintenance of centromeric cohesion. This protein begins to accumulate on the chromatin in mid-G₂ of primary spermatocytes and then remains only at the centromeres of condensed chromosomes until the onset of anaphase II (BALICKY et al. 2002 ). The mechanism of establishment of cohesion between meiotic sister chromatids in Drosophila would be revealed by further genetic and molecular characterization of the mei(3)M20 gene.

	ACKNOWLEDGMENTS

We are grateful to Kugao Oishi, Kyoichi Sawamura, and Ian Boussy for critically reading the manuscript and for helpful advice; to those who kindly provided Drosophila strains (as stated); to John Tomkiel for communicating his unpublished results; and to Hiromi Sato and Mai Kimura for technical help. This work was in part supported by a grant to K.H. from The Japan Science Society and by grants to M.-T.Y. from Ministry of Education, Culture, Sports, Science, and Technology.

Manuscript received August 14, 2003; Accepted for publication December 31, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

APPELS, R. and A. J. HILLIKER, 1982  The cytogenetic boundaries of the rDNA region within heterochromatin of the X chromosome of Drosophila melanogaster and their relation to male meiotic pairing sites. Genet. Res. 39:149-156.[Medline]

AULT, J. G. and C. L. RIEDER, 1994  Meiosis in Drosophila males. I. The question of separate conjunctive mechanisms for the XY and autosomal bivalents. Chromosoma 103:352-356.[Medline]

AULT, J. G., H.-P. P. LIN, and K. CHURCH, 1982  Meiosis in Drosophila melanogaster. IV. The conjunctive mechanism of the XY bivalent. Chromosoma 86:309-317.[CrossRef][Medline]

BAKER, B. and A. T. C. CARPENTER, 1972  Genetic analysis of sex chromosomal meiotic mutants in Drosophila melanogaster.. Genetics 71:255-286.[Abstract/Free Full Text]

BALICKY, E. M., M. W. ENDRES, C. LAI, and S. E. BICKEL, 2002  Meiotic cohesion requires accumulation of ORD on chromosomes before condensation. Mol. Biol. Cell 13:3890-3900.[Abstract/Free Full Text]

BICKEL, S. E., D. W. WYMAN, W. Y. MIYAZAKI, D. P. MOORE, and T. L. ORR-WEAVER, 1996  Identification of ORD, a Drosophila protein essential for sister chromatid cohesion. EMBO J. 15:1451-1459.[Medline]

BICKEL, S. E., D. W. WYMAN, and T. L. ORR-WEAVER, 1997  Mutational analysis of the Drosophila sister-chromatid cohesion protein ORD and its role in the maintenance of centromeric cohesion. Genetics 146:1319-1331.[Abstract]

BICKEL, S. E., T. L. ORR-WEAVER, and E. M. BALICKY, 2002  The sister-chromatid cohesion protein ORD is required for chiasma maintenance in Drosophila oocytes. Curr. Biol. 12:925-929.[CrossRef][Medline]

CHURCH, K., and H.-P. P. LIN, 1988 Drosophila: a model for the study of aneuploidy, pp. 227–255 in Aneuploidy, Part B: Induction and Test Systems, edited by B. K. VIG and A. A. SANDBERG. Alan R. Liss, New York.

CLEVELAND, D. W., Y. MAO, and K. F. SULLIVAN, 2003  Centromeres and kinetochores: from epigenetics to mitotic checkpoint signaling. Cell 112:407-421.[CrossRef][Medline]

COOPER, K. W., 1950 Normal spermatogenesis in Drosophila, pp. 1–61 in Biology of Drosophila, edited by M. DEMEREC. John Wiley & Sons, New York.

COOPER, K. W., 1959  Cytogenetic analysis of major heterochromatic elements (especially Xh and Y) in Drosophila melanogaster, and the theory of "heterochromatin.". Chromosoma 10:535-588.[CrossRef][Medline]

COOPER, K. W., 1964  Meiotic conjunctive elements not involving chiasmata. Proc. Natl. Acad. Sci. USA 52:1248-1255.[Free Full Text]

FLYBASE,, 2003  The FlyBase database of the Drosophila genome projects and community literature. Nucleic Acids Res. 31:172-175.[Abstract/Free Full Text]

GERSHENSON, S., 1933  Studies on the genetically inert region of the X-chromosome of Drosophila. I. Behaviour of an X-chromosome deficient for a part of its inert region. J. Genet. 28:297-313.

GETHMANN, R. C., 1974  Meiosis in male Drosophila melanogaster. I. Isolation and characterization of mutants affecting second chromosome disjunction. Genetics 78:1127-1142.[Abstract/Free Full Text]

GIUNTA, K. L., J. K. JANG, E. A. MANHEIM, G. SUBRAMANIAN, and K. S. MCKIM, 2002  subito encodes a kinesin-like protein required for meiotic spindle pole formation in Drosophila melanogaster.. Genetics 160:1489-1501.[Abstract/Free Full Text]

GOLDSTEIN, L. S. B., 1980  Mechanisms of chromosome orientation revealed by two meiotic mutants in Drosophila melanogaster.. Chromosoma 78:79-111.[CrossRef][Medline]

GUYÉNOT, E. and A. NAVILLE, 1929  Les chromosomes et la réduction chromatique chez Drosophila melanogaster (Cinèses somatiques, spermatogenèse, ovogenèse). Cellule 39:25-82.

HARDY, R. W., D. L. LINDSLEY, K. J. LIVAK, B. LEWIS, and A. L. SIVERSTEN et al., 1984  Cytogenetic analysis of a segment of the Y chromosome of Drosophila melanogaster.. Genetics 107:591-610.[Abstract/Free Full Text]

HILLIKER, A. J., D. G. HOLM, and R. APPELS, 1982  The relationship between heterochromatic homology and meiotic segregation of compound second autosomes during spermatogenesis in Drosophila melanogaster.. Genet. Res. 39:157-168.[Medline]

HUANG, A. M., E. J. REHM and G. M. RUBIN, 2000 Recovery of DNA sequences flanking P-element insertions: inverse PCR and plasmid rescue, pp. 429–437 in Drosophila Protocols, edited by W. SULLIVAN, M. ASHBURNER and R. S. HAWLEY. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

KARPEN, G. H., and S. A. ENDOW, 1998 Meiosis: chromosome behavior and spindle dynamics, pp. 203–247 in Dynamics of Cell Division, edited by S. A. ENDOW and D. M. GLOVER. Oxford University Press, Oxford.

KERREBROCK, A. W., W. Y. MIYAZAKI, D. BIRNBY, and T. L. ORR-WEAVER, 1992  The Drosophila mei-S332 gene promotes sister-chromatid cohesion in meiosis following kinetochore differentiation. Genetics 130:827-841.[Abstract]

KERREBROCK, A. W., D. P. MOORE, J. S. WU, and T. L. ORR-WEAVER, 1995  Mei-S332, a Drosophila protein required for sister-chromatid cohesion, can localize to meiotic centromere regions. Cell 83:247-256.[CrossRef][Medline]

KITAJIMA, T. S., S. YOKOBAYASHI, M. YAMAMOTO, and Y. WATANABE, 2003  Distinct cohesin complexes organize meiotic chromosome domains. Science 300:1152-1155.[Abstract/Free Full Text]

LEE, J. Y. and T. L. ORR-WEAVER, 2001  The molecular basis of sister-chromatid cohesion. Annu. Rev. Cell Dev. Biol. 17:753-777.[CrossRef][Medline]

LEWIS, E. B. and F. BACHER, 1968  Method of feeding ethyl methane sulfonate (EMS) to Drosophila males. Dros. Inf. Serv. 48:193.

LIN, H.-P. P. and K. CHURCH, 1982  Meiosis in Drosophila melanogaster. III. The effect of orientation disruptor (ord) on gonial mitotic and the meiotic divisions in males. Genetics 102:751-770.[Abstract/Free Full Text]

LINDSLEY, D. L., and G. G. ZIMM, 1992 The Genome of Drosophila melanogaster. Academic Press, San Diego.

LIVAK, K. J., 1984  Organization and mapping of a sequence on the Drosophila melanogaster X and Y chromosomes that is transcribed during spermatogenesis. Genetics 107:611-634.[Abstract/Free Full Text]

LIVAK, K. J., 1990  Detailed structure of the Drosophila melanogaster Stellate genes and their transcripts. Genetics 124:303-316.[Abstract]

MASON, J. M., 1976  Orientation disruptor (ord): a recombination-defective and disjunction-defective meiotic mutant in Drosophila melanogaster.. Genetics 84:545-572.[Abstract/Free Full Text]

MCINTOSH, J. R., E. L. GRISHCHUK, and R. R. WEST, 2002  Chromosome-microtubule interactions during mitosis. Annu. Rev. Cell Dev. Biol. 18:193-219.