A Deficiency Screen of the Major Autosomes Identifies a Gene (matrimony) That Is Haplo-insufficient for Achiasmate Segregation in Drosophila Oocytes

Corresponding author: R. Scott Hawley, 1000 E. 50th St., Kansas City, MO 64110., rsh{at}stowers-institute.org (E-mail)

Communicating editor: K. GOLIC

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In Drosophila oocytes, euchromatic homolog-homolog associations are released at the end of pachytene, while heterochromatic pairings persist until metaphase I. A screen of 123 autosomal deficiencies for dominant effects on achiasmate chromosome segregation has identified a single gene that is haplo-insufficient for homologous achiasmate segregation and whose product may be required for the maintenance of such heterochromatic pairings. Of the deficiencies tested, only one exhibited a strong dominant effect on achiasmate segregation, inducing both X and fourth chromosome nondisjunction in FM7/X females. Five overlapping deficiencies showed a similar dominant effect on achiasmate chromosome disjunction and mapped the haplo-insufficient meiotic gene to a small interval within 66C7-12. A P-element insertion mutation in this interval exhibits a similar dominant effect on achiasmate segregation, inducing both high levels of X and fourth chromosome nondisjunction in FM7/X females and high levels of fourth chromosome nondisjunction in X/X females. The insertion site for this P element lies immediately upstream of CG18543, and germline expression of a UAS-CG18543 cDNA construct driven by nanos-GAL4 fully rescues the dominant meiotic defect. We conclude that CG18543 is the haplo-insufficient gene and have renamed this gene matrimony (mtrm). Cytological studies of prometaphase and metaphase I in mtrm hemizygotes demonstrate that achiasmate chromosomes are not properly positioned with respect to their homolog on the meiotic spindle. One possible, albeit speculative, interpretation of these data is that the presence of only a single copy of mtrm disrupts the function of whatever "glue" holds heterochromatically paired homologs together from the end of pachytene until metaphase I.

THE accurate disjunction of homologs during meiosis is accomplished by a series of highly coordinated processes beginning in meiotic prophase. In the chiasmate system of segregation, centromeres are oriented toward opposite poles by chiasmata, which restrict the movement of the chromosomes and serve to orient the centromeres toward the poles (NICKLAS 1974 ; HAWLEY 1988 ). However, Drosophila melanogaster females also employ a highly efficient backup mechanism, known as homologous achiasmate segregation, to ensure the separation of homologous chromosomes that, for one reason or another, fail to undergo exchange (HAWLEY and THEURKAUF 1993 ). In this system, the physical association of homologs at the end of pachytene is maintained not by chiasmata, but rather by the perdurance of heterochromatic pairing, which is both necessary and sufficient to ensure centromere co-orientation (HAWLEY et al. 1992 ; KARPEN et al. 1996 ). These heterochromatic pairings are established in early prophase and maintained until metaphase I, despite the dissolution of euchromatic synapsis at the end of pachytene (CARPENTER 1979 ; DERNBURG et al. 1996 ).

The first evidence that heterochromatic homology is crucial to the segregation of achiasmate homologs came from studies of the effects of homologous duplications on the segregation of two normal fourth chromosomes (HAWLEY et al. 1992 ). These duplications, designated as Dp(1;4)'s, carry some or all of the heterochromatic base of the fourth chromosome and range in size from the size of a normal fourth chromosome to 3.5 times its length. The more fourth chromosome heterochromatin carried by a Dp(1;4), the higher the level of induced fourth chromosome nondisjunction, regardless of the size of the duplication. Removing a block of the fourth chromosome heterochromatin from the largest of these duplications ablates its ability to induce Dp(1;4) 44 disjunctions.On the basis of these data, HAWLEY et al. 1992 suggested that fourth chromosome segregation is facilitated by the pairing of fourth chromosome heterochromatin. Similar results were also obtained using duplications for the pericentric heterochromatin of the X chromosome, validating this conclusion for larger achiasmate chromosomes as well.

KARPEN et al. 1996 presented stronger evidence that homologous achiasmate segregation is dependent on heterochromatic homology. They showed that the frequency with which two achiasmate deletion derivatives of Dp(1;f)1187 segregate from each other in female meiosis is proportional to the amount of centric heterochromatic homology. Normal segregation requires 800 kb of overlap in the centric heterochromatin, whereas nearly random disjunction is observed with only 300 kb of overlap. A linear correlation between the amount of heterochromatic homology and segregation efficiency was observed in the range of 300–800 kb. They concluded that sequences found throughout the centric heterochromatin of Dp(1;f)1187 act additively to ensure achiasmate meiotic segregation. Finally, DERNBURG et al. 1996 used three-dimensional fluorescent hybridization to demonstrate that although euchromatic pairings dissolve following pachytene, heterochromatic pairings are preserved within the karyosome until prometaphase. Thus, in Drosophila female meiosis, heterochromatic pairings persist beyond the dissolution of the synaptonemal complex at pachytene until centromere co-orientation at prometaphase.

The fact that heterochromatic pairings are maintained after the dissolution of euchromatic synapsis suggests the existence of specific functions required to maintain heterochromatic associations. Obviously, the identification of such functions might provide important insights into the mechanisms by which heterochromatic alignments facilitate achiasmate segregations. Unfortunately, little is known about the mechanisms that underlie the maintenance of such homologous heterochromatic associations in the oocyte nucleus. In an attempt to identify the structural proteins that maintain these heterochromatic pairings, we set out to identify genes that showed a clear dosage effect on the efficacy of achiasmate segregation.

Our screen was partly based on the assumption that a reduction in the quantity of some essential "glue" or "binder" protein by 50% might have dramatic and direct effects on the ability of the oocyte to maintain heterochromatic associations and thus disrupt achiasmate segregation. Toward that end, we screened the Bloomington deficiency kit for deficiencies that exhibited a dominant effect on X and/or on fourth chromosome nondisjunction in females of the genotype FM7/X; Df/+; spa^pol. We focused our attention on deficiencies that impaired the proper segregation of both achiasmate X and fourth chromosomes, for which we could subsequently demonstrate a direct effect on fourth chromosome nondisjunction in females bearing chiasmate X chromosomes (i.e., we were specifically interested in identifying genes whose ability to induce fourth chromosome nondisjunction as hemizygotes did not require high levels of X chromosome missegregation). This latter stipulation is meant to remove from consideration those mutants in which the effect on fourth chromosome segregation appears to be an indirect consequence of X chromosome nondisjunction (O'TOUSA 1982 ; ZITRON and HAWLEY 1989 ; WHYTE et al. 1993 ).

Our initial screen of 123 deficiencies identified a single deletion, Df(3L)66C-G28, that exhibited a reproducible dominant defect in achiasmate segregation in females, inducing both X and fourth chromosome nondisjunction in FM7/X females. We were able to determine that this increase in meiotic nondisjunction is indeed the consequence of hemizygosity for a gene identified in the Drosophila genome sequence as CG18543 (CELNIKER et al. 2002 ). We have renamed this gene matrimony (mtrm) because it appears to hold paired chromosomes together. In mtrm hemizygotes, achiasmate chromosomes are often improperly positioned on the spindle, relative to their homologs during both prometaphase and metaphase, suggesting that proper heterochromatic alignments or pairings either failed to occur or were not properly maintained. One possible interpretation of these data is that the presence of only a single copy of mtrm disrupts functioning of whatever "glue" holds heterochromatically paired homologs together from the end of pachytene until metaphase I.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Drosophila stocks:
Most of the deficiency lines screened in this study were obtained from the Bloomington Drosophila Stock Center. A list of the deficiencies tested is provided in Table 1. In addition to those deficiencies, we examined a number of other deficiencies that delete all or part of the 66C region. Df(3L)66C-I65 (BL#209), Df(3L)66C-G28 (BL #1541), and Df(3L)66C-T2-10 were generated by loss of the transposon (p{white-un2}30) in the GR881 insertion line (J. E. NATZLE, unpublished data). Df(3L)66C-I65 and Df(3L)66C-G28 were generated by imprecise transposon excision of the p{white-un2}30 insertion by 2-3-mediated mobilization. Df(3L)66C-T2-10 was generated from the GR881 line by X-ray mutagenesis. Df(3L)66C-B2-2 and Df(3L)66C-E3-1 were recovered in a screen for diepoxybutane-induced mutations that were lethal in trans with the Df(3L)66C-I65 deletion. Both deletions were induced on an isogenized third chromosome carrying the markers mwh red ebony. We also tested Df(3L)BSC13, which uncovers the region 66B12-C1;66D2-4. This deletion was kindly provided by Kevin Cook at the Bloomington Drosophila Stock Center, who produced this deficiency by recombining two P elements located at positions 66B10-11 and 66D6 on the map presented in Fig 2. Df(3L)ZP1 (66A17-66C5) was received courtesy of P. Maroy (Szeged, Hungary). l(3)F6 and l(3)J1 were isolated by J. NATZLE (unpublished data). l(3)L3852, EP(3)3729, and EP(3)3616 were obtained from the Bloomington Drosophila Stock Center.

View larger version (52K): In this window In a new window Download PPT slide   Figure 1. X and fourth chromosome nondisjunction of regions screened for meiotic haplo-insufficiency. The effects of the deficiencies listed in Table 1 on the frequencies of X and fourth chromosome nondisjunction.

View larger version (26K): In this window In a new window Download PPT slide   Figure 2. A cytogenetic map of region 66C showing the breakpoints of the indicated deletions in reference to the polytene map (top); the known and predicted genes that lie within the limits defined by the deficiencies that are haplo-insufficient for achiasmate segregation (middle); and the site of insertion of the P(SUPor-P)KG08051 element relative to exo70 and CG18543 (bottom).

The basic cross scheme:
To introduce marked X and fourth chromosomes, males from each of the second or third chromosome deficiency stocks were crossed to y w; Sp¹ Bl¹ L^rm Bc¹ Pu²/SM6a; spa^pol or y w; D/TM3; spa^pol, respectively. To screen for haplo-insufficient meiotic loci, deficiency-bearing males resulting from these crosses were crossed to FM7w; spa^pol females to generate y w/FM7w; Df/+; spa^pol females. For each deficiency stock, we scored progeny from at least 20 such y w/FM7w; Df/+; spa^pol females, crossed individually to attached-XY, y⁺ v f B; C(4),ci ey^R males, and assessed the frequency of X and fourth chromosome nondisjunction. (The symbol attached-XY denotes the chromosome C(1;Y), IN(1)EN, whose structure is Y^SX·Y^L.) The FM7w chromosome is marked with y, w, and B. Regular female progeny from this cross are yellow-plus Bar (FM7w/attached-XY) and yellow-plus Bar (y w/attached-XY). Regular male progeny are yellow-white Bar (FM7,w/0) and yellow-white (yw/0). All regular progeny carry three copies of chromosome 4, the attached-4 chromosome [C(4)RM], and a maternal fourth chromosome marked with spa^pol, and are wild type for spa^pol, ci, and ey^R.

X chromosome nondisjunction at meiosis I in the female results in yellow-white Bar females (FM7,w/y w, arising from diplo-X ova) or vermillion forked Bar males (attached-XY/0, arising from nullo-X ova). X chromosome nondisjunction at meiosis II results in FM7/FM7 or yw/yw daughters. Meiosis II nondisjunction is extremely rare, and no such events were observed in the crosses reported here. The adjusted total is corrected for the inviability or reduced viability of certain progeny classes. Diplo-X ova are viable only when fertilized by a nullo-XY-bearing sperm, and nullo-X ova are viable only when fertilized by a sperm carrying the attached-XY. Therefore, only half of the nondisjunctional ova are recovered as exceptions. In the adjusted totals for crosses involving y/y and FM7/y females, we correct for this inequality by doubling the number of X chromosome exceptions. The frequency of nondisjunction was calculated as described in HAWLEY et al. 1992 .

Fourth chromosome nondisjunction results in cubitus interruptus eyeless Russian progeny, arising from the fertilization of a nullo-4 ovum by a sperm carrying the attached-4 chromosome, or sparkling poliert offspring arising from the fertilization of a diplo-4 ova by a nullo-4 sperm. Nullo-4 flies do not survive. Tetra-4 flies are both rare and indistinguishable from regular progeny. Some crosses occasionally produce haplo-4 Minute progeny. The survival of these flies is poor and erratic. We recorded these flies, but do not list them in the tables or include them in our calculations. The cross involving the P{lacW}exo^mp1 insertion was done using females heterozygous for spa^pol; therefore the diplo-4 exceptions could not be scored. In this cross, the calculated frequency of fourth chromosome nondisjunction was based on doubling the number of ci ey^R (nullo-4) exceptions and thus may represent an underestimate of the actual rate of fourth chromosome nondisjunction events. In several crosses, it became apparent that one or more of the tested females was triplo-4, as evidenced by a number of sparkling poliert exceptions that approximated 30% among progeny. Such vials also showed regular segregation of the sex chromosomes and produced few or no cubitus interruptus exceptions. The progeny of these females were excluded from consideration.

For crosses involving the 66C deletions we verified that the observed effects on achiasmate nondisjunction were not the effects of secondary nondisjunction events mediated by a cryptic Y chromosome by demonstrating that the expected regular XO sons were indeed sterile. In crosses involving the P(SUPor-P)KG08051 insertion, the presence of both y⁺ and w⁺ genes on the P element precluded the identification of 50% of the yellow-white matroclinous exceptions. To account for this, the number of yellow-white exceptions was doubled and the actual number of yellow-white exceptions was subtracted from the corresponding class of yellow-plus white-plus progeny. The numbers of yellow-white matroclinous exceptions observed were as follows: yellow-white females (18), yellow-white cubitus interruptus eyeless Russian females (17), and yellow-white sparkling poliert females (1). The corrected values were then used to create the list of progeny presented in Table 4.

Cytological studies:
Oocytes were prepared and examined as previously described with minor modifications (THEURKAUF and HAWLEY 1992 ; MATTHIES et al. 2000 ). Egg chambers from 3- to 7-day-old females were extracted by quick pulses of a blender using a modified Robb's medium. The mixture was passed sequentially through a loose and fine mesh to separate late-stage oocytes. The oocytes were fixed for 20 min on a rotator at room temperature in a hypertonic solution, thereby preventing hypotonic activation of the mature oocytes. After removal of the follicle cells, chorion, and vitelline membranes, the oocytes were permeabilized with 1% Triton X-100 in PBS. Oocytes were labeled with MAS078P (Harlan Sera-Lab at 1:500) and MAB1864 [Chemicon (Temecula, CA) at 1:500] rat antitubulin monoclonal antibodies. These antibodies were then labeled with Cy3 conjugated anti-rat secondary antibodies (1:250) purchased from Jackson Immunoresearch (West Grove, PA). Chromosomes were stained with 4'6-diamidino-2-phenylindole. Data were collected in 0.2-µm Z steps 3–5 µm above and below the subject of interest. The resulting data were deconvolved using the Softworx package (API).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Screening deficiency heterozygotes for their effects on achiasmate segregation:
We initially screened 123 autosomal deficiencies obtained from the Bloomington Drosophila Stock Center, 66 second chromosome deficiencies, and 57 third chromosome deficiencies. An estimate of the autosomal genome coverage represented by this deficiency collection is 64–73%. The effects of heterozygosity for each of these 123 deficiencies on X and fourth chromosome nondisjunction in FM7/X; Df/+; spa^pol females are listed in Table 1 and displayed graphically in Fig 1. Only four of the deficiencies, Df(3L)66C-G28, Df(2L)r10, Df(3L)Cat, and Df(3R)ea, exhibited a strong dominant effect on achiasmate chromosome nondisjunction.

The highest frequencies of achiasmate chromosome nondisjunction were exhibited by females heterozygous for Df(3L)66C-G28. Similar high frequencies of X and fourth chromosome nondisjunction were also exhibited by five smaller deficiencies that overlap Df(3L)66C-G28 (Table 2 and see below). None of the three smaller deficiencies that partially overlap the Df(2L)r10 region [Df(2L)osp29, Df(2L)TE35Bc-24, or Df(2l)H20] exerted a dominant effect on chromosome segregation. This might suggest the presence of a dosage-sensitive region in the interval that is removed only by Df(2L)r10 (35F-36A). The gene c(2)M, which encodes a synaptonemal complex component, lies within this interval (MANHEIM and MCKIM 2003 ). However, heterozygosity for a strong allele of this gene does not elevate the frequency of X and fourth chromosome nondisjunction in FM7/X females (data not shown). Similarly, the meiotic effects of Df(3L)Cat are not exhibited by the overlapping deficiency Df(3L)W4. Again, this might suggest the presence of a dosage-sensitive region in the interval that is removed only by Df(3L)Cat (75C-D). There are no obvious candidate genes in this interval. Df(3R)ea was not studied further.

The effects of Df(3L)66C-G28 are also generated by overlapping deficiencies in region 66C:
Df(3L)66C-G28 removed all (or most) of region 66C. To map the haplo-insufficient site, we tested a number of deficiencies that overlapped Df(3L)66C-G28 for their effects on meiotic chromosome segregation (see Fig 2). As shown in Table 2a, Df(3L)ZP1, which uncovers the interval 66A17; 66C5, did not exhibit a dominant effect on meiotic nondisjunction, nor was any such effect observed for Df(3L)pbl-X1 and Df(3L)h-i22. However, four overlapping deficiencies [Df(3L)66C-I65, Df(3L)66C-T2-10, Df(3L)66C-B2-2, and Df(3L)66C-E3-1] did show a dominant defect in achiasmate chromosome disjunction similar to that observed in Df(3L)66C-G28 heterozygotes (Table 2a). These data restrict the putative haplo-insufficient meiotic gene to a small interval within 66C.

Because all five nondisjunction-inducing deficiencies were derived by mutagenesis of a P-element insertion in 66C (see MATERIALS AND METHODS), it is possible that observed defects were the result of some other dominant mutation carried by the original P-insertion chromosome and not due to haplo-insufficiency for one or more genes in region 66C. To eliminate this possibility, we tested an independently derived deficiency [Df(3L)BSC13] uncovering the region 66B12-C1;66D2-4 that was kindly provided by Kevin Cook. As shown in Table 2b, FM7/X; Df(3L)BSC13/+ females displayed frequencies of X and fourth chromosome nondisjunction of 50.8 and 11.9%, respectively (N = 751). The fact that this independently derived deficiency shares no known background elements with the other tested deficiencies confirms that the effects of these deficiencies truly are a consequence of the deletion of some element in region 66C7-12.

As is obvious from Table 2b, the observed frequencies of X and fourth chromosome nondisjunction varied considerably among the six deficiencies tested (from 23.3 to 50.8% for the X chromosome and from 6.7 to 30.1% for the fourth chromosome). To some degree, this may well reflect the influence of genetic background. However, during the course of these experiments we also noted substantial quantitative variations in the observed levels of X chromosome and fourth chromosome nondisjunction exhibited by several of these deficiencies, but most notably for Df(3L)66C-I65 and Df(3L)B2-2. For these deficiencies, the observed frequencies of X and fourth chromosome nondisjunction could vary as much as twofold from experiment to experiment. The highest and lowest values observed for several of these deficiencies over the course of numerous trials are displayed in Table 3 as experiment 1 or experiment 2. We can suggest only that our assays are extremely sensitive and that the experiment-to-experiment variation that we observe reflects the effects of subtle environmental variation and/or genetic background effects.

Characterizing the deficiency breakpoints:
To determine the limits of the region that included the putative dosage-sensitive meiotic gene, we set out to position the ends of the deficiencies portrayed in Fig 1 on the physical map of the fly genome. This was accomplished by PCR analysis of homozygous deficiency-bearing embryos using pairs of primers selected to define specific intervals in the proximal region of 66C. As shown in Fig 3, the distal breakpoints of Df(3L)66C-I65, Df(3L)66C-B2-2, and Df(3L)66CE3-1 lie distal to ImpE1 (CG7116 or CG32356). For Df(3L)66C-T2-10, the distal breakpoint lies immediately proximal to ImpE1. Thus the distal limit of the region containing the haplo-insufficient gene may be approximated to lie between 8,349,000 and 8,351,000 on the Release 3.1 sequence of the Drosophila chromosome 3L map (CELNIKER et al. 2002 ).

View larger version (13K): In this window In a new window Download PPT slide   Figure 3. PCR mapping of the endpoints of the deficiencies in 66C. DNA obtained from embryos homozygous for the indicated deletions was amplified using various primer combinations corresponding to the positions noted. Deficiency homozygotes were distinguished from siblings using a green fluorescent protein marker carried on the balancer chromosome. All primer pairs spanned an interval of no more than 1 kb. + indicates a clear positive signal and - indicates a failure to observe amplification. Numbering of the sequence is based on Release 3.1 of the Drosophila genome annotation (CELNIKER et al. 2002 ).

The proximal breakpoint of Df(3L)66C-T2-10 lies immediately distal to or within CG7015, consistent with the failure of this deficiency to complement lethal alleles of the Cbl gene (CG7037). For Df(3L)66C-B2-2, the proximal breakpoint appears to lie within the CG7127 (exo70) gene. Indeed, the most distal "positive" PCR results were obtained using primers corresponding to the CG18543 gene that lies imbedded in the second intron of exo70. The breakpoints of Df(3L)66C-I65 and Df(3L)66C-E3-1 are more ambiguous and appear to lie between those of Df(3L)66C-B2-2 and Df(3L)66C-T2-10. Nonetheless, the proximal limit of the region containing the haplo-insufficient gene may be approximated to lie between 8,371,000 and 8,373,000 on the Release 3.1 sequence of the Drosophila chromosome 3L map (CELNIKER et al. 2002 ).

Identifying the haplo-insufficient gene:
As shown in Fig 3, the region defined by our deficiency analysis is composed of 20–24 kb of genomic DNA. To identify the haplo-insufficient locus, we have tested a number of mutations that fall within the 66C10-11 interval for their ability to mimic the effects of the tested deficiencies with respect to achiasmate nondisjunction. The first mutations to be tested were l(3)F6, l(3)J1, l(3)L3852, EP(3)3729, and EP(3)3616. None of these mutants increased the level of X or fourth chromosome nondisjunction in FM7/X females above that observed in controls (data not shown).

However, as shown in Table 4, a P insertion designated P(SUPor-P)KG08051 fully mimicked the effects of deficiencies for 66C in terms of its dominant effect on achiasmate segregation. [ Table 4 provides detailed segregation data for Df(3L)T2-10 for comparison to P(SUPor-P)KG08051]. As shown in Fig 2, the P(SUPor-P)KG08051 insertion falls into the third intron of the exo70 gene, which encodes a component of a secretory structure known as the exocyst. However, a second predicted gene (CG18543) also falls within this intron and the insertion site for P(SUPor-P)KG08051 is 90 bp upstream of the predicted start codon for CG18543. To demonstrate that the P(SUPor-P)KG08051 insertion is indeed the basis of the segregational defect, we created both precise and imprecise excisions of this P insertion. P(SUPor-P)KG08051-exc21 is a precise excision of the P element, as verified by DNA sequencing. As shown in Table 4, P(SUPor-P)KG08051-exc21 no longer causes a defect in achiasmate segregation, confirming that the high levels of nondisjunction observed in females heterozygous for P(SUPor-P)KG08051 are indeed a consequence of the P insertion.

The P(SUPor-P)KG08051 insertion also exhibits recessive female sterility (associated with defects in oogenesis) and this sterility is also alleviated in the precise excision (exc21) described above. However, the P(SUPor-P)KG08051-exc43 excision derivative, which retains the sterility defect but has lost the dominant meiotic defect, allows us to separate these two defects. Although P(SUPor-P)KG08051-exc43 is associated with the loss of both y⁺ and w⁺ markers, it retains at least the two ends of the P element, and attempts to amplify the remaining P-element sequence by PCR using primers that span the insertion site were unsuccessful. Thus, this derivative may well be the result of an internal deletion that causes loss of the y⁺ and w⁺ markers while maintaining a substantial amount of the original sequence. Regardless of the exact nature of the lesion born by this derivative, the fact that it reverts only one of the two phenotypes exhibited by the original insertion is consistent with the view that the two types of defects may result from the disruption of two different genes (exo70 and CG18543) by the P(SUPor-P)KG08051 insertion.

Expression of CG18543 in the germline from a transgenic insertion construct rescues the meiotic defects exerted by heterozygosity for P(SUPor-P)KG08051:
The meiotic defects of P(SUPor-P)KG08051 are largely ameliorated in females that also carry a transgenic construct bearing a CG18543 cDNA expressed under the control of a UAS promoter and driven by nanos-GAL4 (VAN DOREN et al. 1998 ). The coding region of the mtrm cDNA was amplified by PCR using a 5' primer tagged with a KpnI restriction site and a 3' primer tagged with an XbaI restriction site. The product was then digested and cloned into the pUASP vector using the above restriction enzymes. FM7/X; P(SUPor-P)KG08051/+ females bearing both the UAS-CG18543 construct and the nanos-GAL4 driver display only 2.7% X nondisjunction and 7.1% fourth chromosome nondisjunction (adjusted total is 1266), demonstrating both that CG18543 is in fact the dosage-sensitive locus revealed in the original deficiency screen and that CG18543 exerts its effects in the female germline. The CG18543 transcription unit encodes a predicted protein of 217 amino acids. Although this gene is conserved within the Drosophila species group, at least as far as D. pseudoobscura, no significant homologs are observed in mosquitoes, and there are no obvious homologs in any other nonfly species (W. GILLILAND, personal communication). We have named this gene matrimony (mtrm).

The identification of the matrimony gene as the meiotic haplo-insufficient locus is consistent with several other observations. First, mutants in other exocyst components (sec5 and exo84) failed to exhibit dominant effects on meiotic chromosome segregation (data not shown) so it would have been surprising had such an effect been observed for exo70. Second, a P-element insertion (P{lacW}exo70^mp1) that lies in the 5' untranslated region of exo70, 52 bp upstream of the predicted start codon, fails to exhibit a dominant effect on meiotic chromosome segregation (see Table 4). (The isolation and characterization of this mutant is described in MATERIALS AND METHODS.) Third, the excision derivative, P(SUPor-P)KG08051-exc13, is associated with a large deletion of material upstream of the P insertion, including the first three exons of exo70. However, despite what might be considered to be a far more damaging mutational lesion in exo70, the effects of this derivative on meiotic chromosome segregation are similar, if not identical, to those observed for P(SUPor-P)KG08051 (see Table 4 and Table 5). The P(SUPor-P)KG08051-exc13 derivative does display recessive semilethality. Less than 50% of the expected number of P(SUPor-P)KG08051-exc13 adults hatch out of their pupal cases, and all of these die within 1–2 days. These phenotypes are similar to those observed for a loss-of-function mutant in another component of the exocyst (sec5), which also displays defects in oogenesis and recessive lethality (MURTHY et al. 2003 ; M. MURTHY and T. L. SCHWARZ, personal communication).

The observed segregational defects are the result of an impairment of the achiasmate segregation system:
One could imagine that the reduction in the dosage of mtrm might occur by one of three general mechanisms. First, hemizygosity for mtrm might reduce the frequency of crossing over and thus increase the frequency of X and fourth chromosome nondisjunction in FM7/X females by overloading the distributive system. However, the recombination frequency along the length of the third chromosome did not show significant deviation from normal in females carrying a multiply marked third chromosome and either Df(3L)66C-G28 or Df(3L)B2-2. [The map distances for the ru–st interval on 3L were 46.7, 45.0, and 52.2 cM, respectively, for females of the genotypes +/ru h th st (N = 169), Df(3L)66C-G28/ru h th st (N = 258), and Df(3L)B2-2 /ru h th st (N = 143)]. Second, although one could imagine a generalized defect in spindle function, none of these deficiencies increased the frequency of meiosis II nondisjunctional events or induced meiotic chromosome nondisjunction in the male, suggesting that the defect in segregation is restricted to meiosis I in females. Finally, it is possible that the observed defect represents a specific impairment of achiasmate segregation.

If the effects of hemizygosity for mtrm indeed reflect an impairment of achiasmate segregation, one should observe a decrease in the effect on X nondisjunction in females bearing structurally normal (and thus usually chiasmate) X chromosomes. As shown in Table 5, Df(3L)66C-T2-10, P(SUPor-P)KG08051, and P(SUPor-P)KG08051-exc13 were also tested in females bearing two normal sequence X chromosomes. In all three cases, the levels of X chromosome nondisjunction observed in X/X females were 10-fold lower than the frequencies of X nondisjunction observed in FM7/X females. Moreover, the observed frequencies of X chromosome nondisjunction in females carrying two normal sequence X chromosomes (4%) are 40% of the predicted frequency of nonexchange X bivalents (8–10%; HAWLEY et al. 1992 ), consistent with the view that even these low frequencies of nondisjunction arise as a consequence of the failed segregation of achiasmate X chromosome bivalents. These observations suggest that hemizygosity for mtrm primarily, if not exclusively, affects the segregation of nonexchange chromosomes.

A defect in achiasmate segregation that directly impairs partner maintenance would also be expected to exert a strong effect on fourth chromosome segregation that is independent of the effect on achiasmate X chromosomes. As shown in Table 5, the effects of hemizygosity for mtrm on chromosome 4 nondisjunction are similar in both X/X females and FM7/X females, demonstrating a direct effect of mtrm hemizygosity on fourth chromosome nondisjunction. Such a direct effect on fourth chromosome segregation has been previously demonstrated in only two other instances, namely mutants in the nod gene and overexpression of an Axs^D construct (driven by nanos-GAL4). In the case of mutants at the nod gene, nonexchange chromosomes precociously dissociate from the developing spindle during prometaphase (CARPENTER 1973 ; ZHANG and HAWLEY 1990 ; THEURKAUF and HAWLEY 1992 ; MATTHIES et al. 1999 ), while in the case of the overexpression of Axs^D, chromosome organization in the karyosome is obviously disrupted and achiasmate chromosomes are often observed to be mispositioned relative to the center of the main mass (KRAMER and HAWLEY 2003 ).

mtrm hemizygotes are similar to other meiotic mutants affecting achiasmate segregation in that simultaneous X,4 nondisjunctional events usually occur more often than might be expected by chance and most cases of simultaneous nondisjunction result from XX 44 segregations (SEKELSKY et al. 1999 ). As noted in the DISCUSSION, the excess of simultaneous exceptions and the preponderance of such exceptions that involve XX 44 segregations both likely reflect the function of a homology-independent third system of meiotic chromosome segregation. This process, referred to as the nonhomologous achiasmate segregation (HAWLEY and THEURKAUF 1993 ), does not depend on the physical association of segregational partners (DERNBURG et al. 1996 ) and may rather simply reflect the "crowding" of spindle poles by one pair of missegregating homologs forcing the remaining pair to choose the alternative pole (SANDLER and NOVITSKI 1956 ; HAWLEY and THEURKAUF 1993 ).

Finally, we also observe an excess of diplo-4 exceptions relative to nullo-4 exceptions among the progeny of mtrm hemizygotes. This excess of diplo-4 exceptions observed in the progeny of mtrm hemizygotes does not result from the inclusion of triplo-4 mothers in the reported crosses (see MATERIALS AND METHODS) and its basis is not understood. However, a similar excess of diplo-4 exceptions was also observed in females heterozygous for Axs^D and is not understood in either case (ZITRON and HAWLEY 1989 ).

Cytological studies confirm a defect in achiasmate segregation:
Cytological studies confirm a specific defect in the position of achiasmate chromosomes on the meiotic spindles of oocytes produced by FM7/X; Df(3L)66C-T2-10 females (i.e., females that are hemizygous for mtrm). The vast majority of prometaphase and metaphase figures observed in these oocytes are normal bipolar spindles, suggesting that these oocytes are competent to assemble proper bipolar spindles. However, as exemplified in Fig 4, achiasmate X and fourth chromosomes are often improperly placed on metaphase I spindles of these oocytes from females.

View larger version (12K): In this window In a new window Download PPT slide   Figure 4. Metaphase I image of an FM7/X; Df(3L)T2-10 oocyte. The two chromosomes segregating to the upper pole are fourth chromosomes and the two segregating toward the lower pole are X chromosomes. The X chromosome on the left (with two obviously separated blocks of heterochromatin) is FM7. The X on the right is its normal sequence homolog.

Among a sample of 19 metaphase figures in which the achiasmate X and fourth chromosomes were clearly discernible, the achiasmate X chromosomes are found on the same half-spindle 21% (4/19) of the time and the fourth chromosomes are found on the same half-spindle in 31% of the oocytes (6/19). In two of the cases of X chromosome nondisjunction, the two X chromosomes were on one-half of the spindle and the two fourth chromosomes were on the other, indicative of XX 44 segregation events. We also observed two cases involving the fourth chromosome and one involving the X chromosome in which, although the achiasmate homologs were found on opposite halves of the spindle, they were not observed to be on the same arc of the spindle. Thus it seems unlikely that these spindles reflect the segregation of the two homologs from their partners; rather they likely reflect the chance orientation of the two homologs to opposite poles. We did not observe X or fourth chromosomes dissociated from the spindle, as is observed in nod mutant oocytes (THEURKAUF and HAWLEY 1992 ), or defects in spindle assembly such as those observed in Axs^D-bearing oocytes or in oocytes heterozygous for mutants in the -tub67C gene (MATTHIES et al. 1999 ; KRAMER and HAWLEY 2003 ). Indeed, the only obvious defect is the lack of proper positioning of achiasmate chromosomes on the meiotic spindle in a large fraction of the oocytes examined.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

How might the Mtrm protein function?
Both the haplo-insufficiency of the mtrm gene and its rather impressive phenotypic variability suggest to us that it encodes a structural protein whose concentration is critical to meiotic chromosome segregation. We propose that hemizygosity for the mtrm gene creates a "threshold defect," in which the level of Mtrm protein is just below some critical threshold for function, such that minor perturbations in mtrm expression can have dramatic effects on the final phenotype. To further characterize the phenotypic effects of altering the dosage of mtrm, it would be of interest to know the meiotic phenotype of P(SUPor-P)KG08051 homozygotes. Unfortunately, homozygotes for P(SUPor-P)KG08051 are female sterile and display multiple defects in oogenesis that preclude the cytological study of stage 13–14 oocytes.

In the absence of immunolocalization data, it is difficult to speculate on the function of the product of the mtrm gene. At present, we know only that mtrm is highly expressed in oocytes and early embryos (ARBEITMAN et al. 2002 ) and that overexpression of the Mtrm protein in Schizosaccharomyces pombe results in cell cycle arrest (EDGAR 1993 ). Our data suggest that in mtrm hemizygotes the X and fourth chromosomes are no longer competent to segregate from their partners and that the two homologs are often found on separate arcs of the spindle. One possible interpretation of these data is that the presence of only a single copy of mtrm disrupts functioning of whatever glue holds heterochromatically paired homologs together from the end of pachytene until metaphase I. (Our reason for suggesting that Mtrm is not required for euchromatic associations derives from the observation that meiotic recombination is normal in mtrm hemizygotes.) This hypothesis is testable both by immunolocalization and by fluorescence in situ hybridization analysis of heterochromatic pairing in oocytes hemizygous for mtrm, and such studies are underway in the laboratory.

Are there other loci comparable to mtrm in the Drosophila genome?
One could imagine two general means in which haplo-insufficient meiotic mutants might be identified in Drosophila. The first would be through the recovery of dominant loss-of-function meiotic mutants. Unfortunately, the vast majority of the known cases of dominant meiotic mutations (e.g., sub^Dub, nod^DTW, and Axs) clearly reflect the effects of antimorphic mutations (ZITRON and HAWLEY 1989 ; KOMMA et al. 1991 ; RASOOLY et al. 1991 ; MATTHIES et al. 1999 ; GUINTA et al. 2002 ; KRAMER and HAWLEY 2003 ). However, WASSER and CHIA 2003 have recently reported that loss-of-function mutations in the X chromosome east gene exert a dominant effect on achiasmate X chromosome segregation, but do not affect the segregation of the fourth chromosomes. As homozygotes these mutants also exhibit defects in chromosome condensation and organization during the cell cycle in both male meiosis and mitosis. Thus the east gene may well be an additional example of a gene that is haplo-insufficient for achiasmate segregation (at least for the X chromosome).

The second method of identifying haplo-insufficient meiotic loci is the type of deficiency screen reported here. Deficiency-based screens for haplo-insufficient genes have a long history in Drosophila genetics and are epitomized by the heroic screen performed by LINDSLEY et al. 1972 using segmental aneuploids. However, with the exception of the Minute loci, the number of haplo-insufficient loci that produce a discernible phenotype on their own is quite low. Most highly successful deficiency screens for haplo-insufficient genes are based on screening for enhancers or suppressors of mutants with "threshold" phenotypes (HALSELL and KIEHART 1998 ; LEE et al. 2001 ). It is thus not surprising that we were aware of but two examples of loci that were haplo-insufficient for proper meiotic chromosome segregation at the time we initiated our screen: namely Df(1)w^rJ1, a deficiency spanning the zeste-white region (ROBBINS 1977 , ROBBINS 1980 , ROBBINS 1981 ), and Df(3)sdb¹⁰⁵ (HINTON 1966 ). Females heterozygous for Df(1)w^rJ1 show elevated levels of achiasmate nondisjunction and reduced recombination, while females heterozygous for Df(3)sdb¹⁰⁵ also display reduced levels of exchange. In neither case has the effect been narrowed to a single gene. These observations, as well as our own findings, strongly suggest that loci similar to mtrm are few in number.

Our working model for homologous achiasmate segregation in Drosophila:
In most instances, meiosis is composed of three basic cellular processes: pairing, recombination, and segregation. Pairing identifies homologs, recombination acts to lock them together, and segregation moves them to opposite poles at the first meiotic division. In a more mechanistic sense, meiotic segregation can be viewed as the events that mediate the stable co-orientation of the homologous centromeres toward opposite poles of the meiotic spindle. Establishing stable co-orientation requires a balancing of plateward and poleward forces acting on the kinetochores. This balancing of forces is usually achieved by chiasmata, the physical consequence of recombination. Proper orientation results from the ability of chiasmata to constrain homologous kinetochores such that they are oriented in opposite directions (NICKLAS 1974 ).

However, many organisms also possess systems that ensure the segregation of chromosomes that do not undergo exchange (WOLF 1994 ). In Drosophila two systems ensure the segregation of achiasmate chromosomes, homologous achiasmate segregation and nonhomologous achiasmate segregation. Although these systems are generally considered together and denoted by the terms "distributive pairing" or "distributive segregation" (GRELL 1976 ), they differ in two fundamental ways. First, nonhomologous achiasmate segregation does not require or involve the physical association of segregational partners, while homologous achiasmate segregation is dependent on heterochromatic pairings (HAWLEY et al. 1992 ; DERNBURG et al. 1996 ; KARPEN et al. 1996 ). Second, homologous achiasmate segregations are disrupted mutants in genes such as Axs, ald, mei-S51, and tub67C, while nonhomologous achiasmate segregation occurs normally in the presence of such mutants. The observations of very high levels of XX 44 segregations in mtrm homozygotes suggests that this defect is also restricted to homologous achiasmate segregations. Indeed, the only mutants known to affect both homologous and nonhomologous achiasmate segregations are nod, ncd, and subito. nod mutants cause nonexchange chromosomes to dissociate from the developing spindle during prometaphase, while ncd and sub mutants cause a failure to maintain a bipolar meiotic spindle (WALD 1936 ; KIMBLE and CHURCH 1983 ; THEURKAUF and HAWLEY 1992 ; MATTHIES et al. 1996 ; GUINTA et al. 2002 ).

Our focus here is on the mechanisms that underlie homologous achiasmate segregations. We imagine that the pairing events underlying homologous achiasmate segregations are initiated by the same processes that allow the initial meiotic pairing of homologs in early meiosis. Thus in Drosophila females, synaptonemal complex (SC) formation does not require exchange, and nonexchange homologs pair and synapse normally (CARPENTER 1973 , CARPENTER 2003 ; MCKIM et al. 1998 ). However, at the end of pachytene, and concomitant with SC dissolution, the euchromatic arms of the chromosomes desynapse and nonexchange homologs nonetheless remain associated by heterochromatic pairings (DERNBURG et al. 1996 ). Our current model is that the maintenance of these pairings requires the Mtrm protein, which serves to link homologous heterochromatic regions together.

One could imagine, for example, that the Mtrm protein might act by modifying some component of the SC that originally connected the heterochromatic regions of chromosomes during prophase. Although the primary function of the SC lies in mediating interhomolog exchanges (PAGE and HAWLEY 2003 ), in many meiotic systems the segregation of achiasmate chromosomes is mediated by the preservation of modified SC structures that persist after SC breakdown (RASMUSSEN 1977 ). Most recently, PAGE et al. 2003 have shown that proteins normally found in the lateral element of the SC create a "dense plate" that holds the achiasmate sex chromosome bivalent together in marsupial males. It thus does not seem unreasonable to suggest that Mtrm-dependent modifications of the heterochromatic SC might play an important role in achiasmate homologous segregations as well.

We also imagine that the maintenance of heterochromatic associations requires functions that serve to maintain the overall structure and compact form of the karyosome. We noted above that mutants in the nuclear protein East disrupt achiasmate segregation in meiosis and exhibit defects in chromosome condensation and organization (WASSER and CHIA 2003 ). Perhaps East and other proteins assist in maintaining the integrity of the karyosome prior to nuclear envelope breakdown. The maintenance of karyosome integrity after nuclear envelope breakdown might involve the function of a membranous sheath that was recently shown to ensheath the developing meiotic spindle during prometaphase and metaphase (KRAMER and HAWLEY 2003 ). The product of the Axs gene colocalizes with this sheath and dominant mutants in the Axs gene disrupt the segregation of achiasmate chromosomes during female meiosis.

Although chiasmate chromosomes achieve stability at the metaphase plate by balancing the tension of forces between two oppositely oriented kinetochores and the two poles of the spindle (NICKLAS 1974 ), the segregation of achiasmate chromosomes poses a different problem. However, in principle, the manner in which achiasmate kinetochores become stably oriented should be functionally equivalent to that used by chiasmate chromosomes, i.e., by balancing tension on the kinetochores, and thus should require the balancing of poleward force at the kinetochore by plateward force on the chromosome. Holding the achiasmate chromosomes on the developing spindle and preventing their precocious migration to the poles requires the Nod protein, which is localized along the arms of the meiotic chromosomes (AFSHAR et al. 1995A , AFSHAR et al. 1995B ; MURPHY and KARPEN 1995 ).

The Nod protein is specifically required for the segregation of achiasmate chromosomes and functions to counterbalance the poleward forces acting at their kinetochores. When these poleward forces are unchecked, as a consequence of the absence of Nod or of chiasmata, chromosomes precociously move off the ends of the developing spindle. Mutations at the nod locus specifically impair achiasmate segregation, as assayed both genetically and cytologically (CARPENTER 1973 ; ZHANG and HAWLEY 1990 ; THEURKAUF and HAWLEY 1992 ). Although Nod is a highly divergent member of the kinesin superfamily (ZHANG et al. 1990 ; RASOOLY et al. 1991 , RASOOLY et al. 1994 ), recent biochemical studies demonstrate that although Nod binds to microtubules and exhibits a microtubule-dependent ATPase activity, it is not capable of functioning as a motor (MATTHIES et al. 2001 ). The ability of nod mutants to suppress an effect in kinetochore poleward movement created by mutants in the tub67C gene supports the conclusion that Nod serves to balance the poleward forces acting at the kinetochore (MATTHIES et al. 1999 ). We imagine that Nod serves this function by binding both microtubules and chromosomes as well by anchoring itself to a stable component of the spindle apparatus.

Concluding remarks:
Thus, we imagine that the basic process of achiasmate segregation can be described as chromosome pairing maintenance of heterochromatic pairing Nod-based segregation. We propose that the Mtrm protein plays a critical role in the second step of this process, that of holding homologs together long enough to ensure their co-orientation. We imagine that Mtrm does so by binding paired heterochromatic regions together long after the euchromatic intervals have desynapsed. If we are correct, we should be able to demonstrate that Mtrm protein binds to the heterochromatic regions of the chromosome early in meiotic prophase and stays bound until metaphase I. We are now conducting the experiments necessary to test this hypothesis.

	ACKNOWLEDGMENTS

This article is dedicated to the memory of our friend and co-worker Annette Lentz. Annette was an inspiration who is greatly missed. We thank Kevin Cook, Minx Fuller, Mala Murthy, and P. Maroy for providing us with stocks. We especially thank Kevin Cook for providing us with Df(3L)BDSC13 soon after its synthesis. In addition, we express our great gratitude to Hugo Bellen and the Drosophila P-element insertion project at Baylor University for providing us with P(SUPor-P)KG08051. We also acknowledge Anne Elliot, John Modrow, Christine Orr, and Ryan Miller for their help in the earlier stages of this project at the University of California at Davis and gratefully acknowledge the help of Heather Peters at Stowers Institute for Medical Research (SIMR). Finally, we are enormously grateful to Michael Melko and Michael Coleman at SIMR and an anonymous reviewer for their help in the statistical analysis of the data presented in Fig 1. We also thank Arcady Mushegian for his assistance in the analysis of the CG18543 transcription unit and Bill Gilliland for his help in the comparison of the CG18543 sequence with that of other organisms. M.J.P. was a Jane Coffin Childs postdoctoral fellow. This research was supported by a grant to R.S.H. from the National Science Foundation.

Manuscript received May 7, 2003; Accepted for publication May 30, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AFSHAR, K., J. SCHOLEY, and R. S. HAWLEY, 1995a  Identification of the chromosome localization domain of the Drosophila Nod kinesin-like protein. J. Cell Biol. 131(4):833-843.[Abstract/Free Full Text]

AFSHAR, K., N. BARTON, R. S. HAWLEY, and L. S. B. GOLDSTEIN, 1995b  DNA binding and meiotic chromosomal localization of the Drosophila Nod kinesin-like protein. Cell 81(1):129-138.[Medline]

ARBEITMAN, M. N., E. E. FURLONG, F. IMAM, E. JOHNSON, and B. H. NULL et al., 2002  Gene expression during the life cycle of Drosophila melanogaster. Science 297(5590):2270-2275.[Abstract/Free Full Text]

CARPENTER, A. T. C., 1973  A meiotic mutant defective in distributive disjunction in Drosophila melanogaster. Genetics 73:393-428.[Abstract/Free Full Text]

CARPENTER, A. T. C., 1979  Synaptonemal complex and recombination nodules in wild-type Drosophila melanogaster females. Genetics 92:511-541.[Abstract/Free Full Text]

CARPENTER, A. T. C., 2003  Normal synaptonemal complex and abnormal recombination nodules in two alleles of the Drosophila melanogaster mutant mei-W68.. Genetics 163:1337-1356.[Abstract/Free Full Text]

CELNIKER, S. E., D. A. WHEELER, B. KRONMILLER, J. W. CARLSON, A. HALPERN et al. 2002 Finishing a whole-genome shotgun: release 3 of the Drosophila melanogaster euchromatic genome sequence. Genome Biol. 3 (12): 0079.1–0079.14.

DERNBURG, A. F., J. W. SEDAT, and R. S. HAWLEY, 1996  Direct evidence of a role for heterochromatin in meiotic chromosome segregation. Cell 86:135-146.[Medline]

EDGAR, B., 1993 Drosophila melanogaster Sevelin clone D52 cell cycle arrest protein gene, complete cds. GenBank/EMBL/DDBJ: U03288, FlyBase F Brf011401.

GRELL, R. F., 1976 Distributive pairing segregation, pp. 436–486 in The Genetics and Biology of Drosophila, Vol. 1a, edited by M. ASHBURNER and E. NOVITSKI. Academic Press, London.

GUINTA, 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]

HALSELL, S. R. and D. P. KIEHART, 1998  Second-site noncomplementation identifies genomic regions required for Drosophila nonmuscle myosin function during morphogenesis. Genetics 148:1845-1863.[Abstract/Free Full Text]

HAWLEY, R. S., 1988 Exchange and chromosomal segregation in eukaryotes, pp. 497–527 in Genetic Recombination, edited by G. SMITH and R. KUCHERLAPATI. ASM Press, Washington, DC.

HAWLEY, R. S. and W. E. THEURKAUF, 1993  Requiem for the distributive system: achiasmate segregation in Drosophila females. Trends Genet. 9:310-317.[Medline]

HAWLEY, R. S., H. IRICK, A. E. ZITRON, D. A. HADDOX, and A. LOHE et al., 1992  There are two mechanisms of achiasmate segregation in Drosophila females, one of which requires heterochromatic homology. Dev. Genet. 13:440-467.[Medline]

HINTON, C. W., 1966  Enhancement of recombination associated with the c3G mutant of Drosophila melanogaster.. Genetics 53:157-164.[Free Full Text]

KARPEN, G. H., M. H. LE, and H. LE, 1996  Centric heterochromatin and the efficiency of achiasmate disjunction in Drosophila female meiosis. Science 273:118-122.[Abstract]

KIMBLE, M. and K. CHURCH, 1983  Meiosis and early cleavage in Drosophila melanogaster eggs: effects of the claret non-disjunctional mutation. J. Cell Sci. 62:301-318.[Abstract]

KOMMA, D. J., A. S. HORNE, and S. A. ENDOW, 1991  Separation of meiotic and mitotic effects of claret non-disjunctional on chromosome segregation in Drosophila.. EMBO J. 10(2):419-424.[Medline]

KRAMER, J. J. and R. S. HAWLEY, 2003  Mutants in the spindle-associated transmembrane protein Axs impair meiotic spindle assembly in Drosophila oocytes. Nat. Cell Biol. 5:261-263.[Medline]

LEE, L. A., L. K. ELFRING, G. BOSCO, and T. L. ORR-WEAVER, 2001  A genetic screen for suppressors and enhancers of the Drosophila PAN GU cell cycle kinase identifies cyclin B as a target. Genetics 158:1545-1556.[Abstract/Free Full Text]

LINDSLEY, D. L., L. SANDLER, B. S. BAKER, A. T. CARPENTER, and R. E. DENELL et al., 1972  Segmental aneuploidy and the genetic gross structure of the Drosophila genome. Genetics 71:157-184.[Abstract/Free Full Text]

MANHEIM, E. A. and K. S. MCKIM, 2003  The synaptonemal complex component C(2)M regulates meiotic crossing over in Drosophila.. Curr. Biol. 13(4):276-285.[Medline]

MATTHIES, H. J., H. B. MCDONALD, L. S. GOLDSTEIN, and W. E. THEURKAUF, 1996  Anastral meiotic spindle morphogenesis: role of the non-claret disjunctional kinesin-like protein. J. Cell Biol. 134:455-464.[Abstract/Free Full Text]

MATTHIES, H. J., L. G. MESSINA, R. NAMBA, K. J. GREER, and M. Y. WALKER et al., 1999  Mutations in the alpha-tubulin 67C gene specifically impair achiasmate segregation in Drosophila melanogaster.. J. Cell Biol. 147(6):1137-1144.[Abstract/Free Full Text]

MATTHIES, H. J. G., M. CLARKSON, R. B. SAINT, R. NAMBA and R. S. HAWLEY, 2000 Analysis of meiosis in fixed and live oocytes by light microscopy, pp. 67–85 in Drosophila: Protocols, edited by W. S. SULLIVAN, M. ASHBURNER and R. S. HAWLEY. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MATTHIES, H. J., R. J. BASKIN, and R. S. HAWLEY, 2001  Orphan kinesin NOD lacks motile properties but does possess a microtubule-stimulated ATPase activity. Mol. Biol. Cell 12(12):4000-4012.[Abstract/Free Full Text]

MCKIM, K. S., B. GREEN-MARROQUIN, J. J. SEKELSKY, G. CHIN, and C. STEINBERG et al., 1998  Meiotic synapsis in the absence of recombination. Science 279:876-878.[Abstract/Free Full Text]

MURPHY, T. D. and G. H. KARPEN, 1995  Localization of centromere function in a Drosophila minichromosome. Cell 82:599-609.[Medline]

MURTHY, M., D. GARZA, R. H. SCHELLER, and T. L. SCHWARZ, 2003  Mutations in the exocyst component Sec5 disrupt neuronal membrane traffic, but neurotransmitter release persists. Neuron 37(3):433-447.[Medline]

NICKLAS, R. B., 1974  Chromosome segregation mechanisms. Genetics 78:205-213.[Abstract/Free Full Text]

O'TOUSA, J., 1982  Meiotic chromosome behavior influenced by mutation-altered disjunction in Drosophila melanogaster females. Genetics 102:503-524.[Abstract/Free Full Text]

PAGE, S. L. and R. S. HAWLEY, 2003  Chromosome choreography: the meiotic ballet. Science 301(5634):785-789.[Abstract/Free Full Text]

PAGE, J., S. BERRIOS, J. S. RUFAS, M. T. PARRA, and J. A. SUJA et al., 2003  The pairing of X and Y chromosomes during meiotic prophase in the marsupial species Thylamys elegans is maintained by a dense plate developed from their axial elements. J. Cell Sci. 116:551-560.[Abstract/Free Full Text]

RASMUSSEN, S. W., 1977  The transformation of the synaptonemal complex into the elimination chromatin of Bombyx mori oocytes. Chromosoma 60:205-221.[Medline]

RASOOLY, R. S., C. M. NEW, P. ZHANG, R. S. HAWLEY, and B. S. BAKER, 1991  The lethal(1)TW-6^cs mutation of Drosophila melanogaster is a dominant antimorphic allele of nod and is associated with a single base change in the putative ATP-binding domain. Genetics 129:409-422.[Abstract]

RASOOLY, R. S., P. ZHANG, A. K. TIBOLLA, and R. S. HAWLEY, 1994  A structure-function analysis of NOD, a kinesin-like protein from Drosophila melanogaster.. Mol. Gen. Genet. 242:145-151.[Medline]

ROBBINS, L. G., 1977  The meiotic effect of a deficiency in Drosophila melanogaster with a model for the effects of enzyme deficiency on recombination. Genetics 87:655-684.[Abstract/Free Full Text]

ROBBINS, L. G., 1980  Maternal-zygotic lethal interactions in Drosophila melanogaster: the effects of deficiencies in the zeste-white region of the X chromosome. Genetics 96:187-200.[Abstract/Free Full Text]

ROBBINS, L. G., 1981  The meiotic behavior of some single-cistron mutants in the zeste-white region of the Drosophila melanogaster X chromosome. Mol. Gen. Genet. 183(2):214-219.[Medline]

SANDLER, L. and E. NOVITSKI, 1956  Evidence for genetic homology between chromosomes I and IV in Drosophila melanogaster, with a proposed explanation for the crowding effect in triploids. Genetics 41:189-193.[Free Full Text]

SEKELSKY, J. J., K. S. MCKIM, T. ARBEL, G. CHIN, and R. L. FRENCH