Recombination and nondisjunction in humans and flies

Recent studies of Drosophila and humans indicate that aberrant genetic recombination is an important component of nondisjunction in both species. In both, a proportion of nondisjunction is associated with failure to pair and/or recombine and in both, exchanges which are either too distal or too proximal increase the likelihood of malsegregation. In this review we provide two perspectives on these observations: first, a review of exchange and chromosome segregation in model organisms, focusing on Drosophila, and secondly an overview of nondisjunction in humans. This format allows us to describe the paradigms developed from studies of model organisms and to ask whether these paradigms apply to the human situation.

Direct evidence of a role for heterochromatin in meiotic chromosome segregation

We have investigated the mechanism that enables achiasmate chromosomes to segregate from each other at meiosis I in D. melanogaster oocytes. Using novel cytological methods, we asked whether nonexchange chromosomes are paired prior to disjunction. Our results show that the heterochromatin of homologous chromosomes remains associated throughout prophase until metaphase I regardless of whether they undergo exchange, suggesting that homologous recognition can lead to segregation even in the absence of chiasmata. However, partner chromosomes lacking homology do not pair prior to disjunction. Furthermore, euchromatic synapsis is not maintained throughout prophase. These observations provide a physical demonstration that homologous and heterologous achiasmate segregations occur by different mechanisms and establish a role for heterochromatin in maintaining the alignment of chromosomes during meiosis.

Chromosomal control of meiotic cell division

Chromosomes have multiple roles both in controlling the cell assembly and structure of the spindle and in determining chromosomal position on the spindle in many meiotic cells and in some types of mitotic cells. Moreover, functionally significant chromosome-microtubule interactions are not limited to the kinetochore but are also mediated by proteins localized along the arms of chromosomes. Finally, chromosomes also play a crucial role in control of the cell cycle.

Identification of the chromosome localization domain of the Drosophila nod kinesin-like protein

The nod kinesin-like protein is localized along the arms of meiotic chromosomes and is required to maintain the position of achiasmate chromosomes on the developing meiotic spindle. Here we show that the localization of ectopically expressed nod protein on mitotic chromosomes precisely parallels that observed for wild-type nod protein on meiotic chromosomes. Moreover, the carboxyl-terminal half of the nod protein also binds to chromosomes when overexpressed in mitotic cells, whereas the overexpressed amino-terminal motor domain binds only to microtubules. Chromosome localization of the carboxyl-terminal domain of nod depends upon an 82-amino acid region comprised of three copies of a sequence homologous to the DNA-binding domain of HMG 14/17 proteins. These data map the two primary functional domains of the nod protein in vivo and provide a molecular explanation for the directing of the nod protein to a specific subcellular component, the chromosome.

The Drosophila meiotic recombination gene mei-9 encodes a homologue of the yeast excision repair protein Rad1

Meiotic recombination and DNA repair are mediated by overlapping sets of genes. In the yeast Saccharomyces cerevisiae, many genes required to repair DNA double-strand breaks are also required for meiotic recombination. In contrast, mutations in genes required for nucleotide excision repair (NER) have no detectable effects on meiotic recombination in S. cerevisiae. The Drosophila melanogaster mei-9 gene is unique among known recombination genes in that it is required for both meiotic recombination and NER. We have analyzed the mei-9 gene at the molecular level and found that it encodes a homologue of the S. cerevisiae excision repair protein Rad1, the probable homologue of mammalian XPF/ERCC4. Hence, the predominant process of meiotic recombination in Drosophila proceeds through a pathway that is at least partially distinct from that of S. cerevisiae, in that it requires an NER protein. The biochemical properties of the Rad1 protein allow us to explain the observation that mei-9 mutants suppress reciprocal exchange without suppressing the frequency of gene conversion.

The mei-41 gene of D. melanogaster is a structural and functional homolog of the human ataxia telangiectasia gene

The D. melanogaster mei-41 gene is required for DNA repair, mitotic chromosome stability, and normal levels of meiotic recombination in oocytes. Here we show that the predicted mei-41 protein is similar in sequence to the ATM (ataxia telangiectasia) protein from humans and to the yeast rad3 and Mec1p proteins. There is also extensive functional overlap between mei-41 and ATM. Like ATM-deficient cells, mei-41 cells are exquisitely sensitive to ionizing radiation and display high levels of mitotic chromosome instability. We also demonstrate that mei-41 cells, like ATM-deficient cells, fail to show an irradiation-induced delay in the entry into mitosis that is characteristic of normal cells. Thus, the mei-41 gene of Drosophila may be considered to be a functional homolog of the human ATM gene.

Induction of metaphase arrest in Drosophila oocytes by chiasma-based kinetochore tension

In normal Drosophila melanogaster oocytes, meiosis arrests at metaphase I and resumes after oocyte passage through the oviduct. Thus, metaphase arrest defines a control point in the meiotic cell cycle. Metaphase arrest only occurs in oocytes that have undergone at least one meiotic exchange. Here it is shown that crossovers between homologs attached to the same centromere do not induce metaphase arrest. Hence, exchanges induce metaphase arrest only when they physically conjoin two separate kinetochores. Thus, the signal that mediates metaphase arrest is not the exchange event per se but the resulting tension on homologous kinetochores.

DNA binding and meiotic chromosomal localization of the Drosophila nod kinesin-like protein

The Drosophila no distributive disjunction (nod) gene encodes a kinesin-like protein that has been proposed to push chromosomes toward the metaphase plate during female meiosis. We report that the nonmotor domain of the nod protein can mediate direct binding to DNA. Using an antiserum prepared against bacterially expressed nod protein, we show that during prometaphase nod protein is localized on oocyte chromosomes and is not restricted to either specific chromosomal regions or to the kinetochore. Thus, motor-based chromosome-microtubule interactions are not limited to the centromere, but extend along the chromosome arms, providing a molecular explanation for the polar ejection force.

Separation anxiety: the etiology of nondisjunction in flies and people

Two new studies examine the recombinational history of human chromosomes that nondisjoin at the first meiotic division in females. Our analysis of these studies suggests two possible etiologies of nondisjunction in terms of well-understood properties of chromosome mechanics. For both the X chromosome and for chromosome 21, 60-70% of nondisjoined chromosomes are derived from chiasmate bivalents, many of which display unusual patterns of exchange. The patterns of exchange and nondisjunction observed for human chromosome 21 parallel those exhibited by a mutation in Drosophila that impairs spindle assembly and function. Based on these similarities, we propose that nondisjunction of chromosome 21 in human females results from an age-dependent loss of spindle-forming ability. The recombinational histories of nondisjoining human X chromosomes are quite different from those of chromosome 21, but rather parallel those obtained for spontaneous nondisjunction in Drosophila females. The data for X chromosome disjunction in both species can be explained by a model in which nondisjunction is the consequence of the age-dependent movement of transposable elements. According to this model, nondisjunction is explained as the consequence of the repair of transposon-induced breaks in the DNA. Both models provide reasonable alternatives to biologically implausible explanations such as the 'production line hypothesis'.

A structure-function analysis of NOD, a kinesin-like protein from Drosophila melanogaster

We have analyzed a collection of 12 mutations in the Drosophila melanogaster nod locus, which encodes a kinesin-like protein involved in female meiotic chromosome segregation. The kinesin-like domain is at the N-terminus of the protein, while the C-terminal portion of the protein is unique. Four of the mutations are missense and affect highly conserved domains of the kinesin-like portion of the predicted protein, and thus demonstrate that the sequence conservation is biologically relevant. Surprisingly, two other mutations, which behave genetically as null alleles, are the result of mutations in the last exon of the nod gene. Thus, these two mutations affect the most C-terminal residues in the unique portion of the predicted protein. Based on these mutations, we suggest that this part of the protein may also be essential for wild-type function. The mutations were induced by either gamma-rays or ethyl methanesulfonate (EMS). All of the gamma-ray induced mutations were small or large chromosomal rearrangements, while all of the EMS mutations were G-->A transitions. These findings are consistent with the biochemical basis of the mode of action of each mutagen.

Requiem for distributive segregation: achiasmate segregation in Drosophila females

The segregation of achiasmate chromosome pairs at meiosis I is not brought about by a single 'distributive system' as previously thought, but rather by two separate mechanisms. One system uses the pairing of proximal heterochromatic sequences to mediate the segregation of achiasmate homologs-an observation that, at long last, defines a function for heterochromatin. The other system facilitates the segregation of heterologous chromosomes, by an as yet undiscovered mechanism.

The genetic analysis of achiasmate segregation in Drosophila melanogaster. III. The wild-type product of the Axs gene is required for the meiotic segregation of achiasmate homologs

The regular segregation of achiasmate chromosomes in Drosophila melanogaster females is ensured by two distinct segregational systems. The segregation of achiasmate homologs is assured by the maintenance of heterochromatic pairing; while the segregation of heterologous chromosomes is ensured by a separate mechanism that may not require physical association. AxsD (Aberrant X segregation) is a dominant mutation that specifically impairs the segregation of achiasmate homologs; heterologous achiasmate segregations are not affected. As a result, achiasmate homologs frequently participate in heterologous segregations at meiosis I. We report the isolation of two intragenic revertants of the AxsD mutation (Axsr2 and Axsr3) that exhibit a recessive meiotic phenotype identical to that observed in AxsD/AxsD females. A third revertant (Axsr1) exhibits no meiotic phenotype as a homozygote, but a meiotic defect is observed in Axsr1/Axsr2 females. Therefore mutations at the AxsD locus define a gene necessary and specific for homologous achiasmate segregation during meiosis. We also characterize the interactions of mutations at the Axs locus with two other meiotic mutations (ald and ncd). Finally, we propose a model in which Axs+ is required for the normal separation of paired achiasmate homologs. In the absence of Axs+ function, the homologs are often unable to separate from each other and behave as a single segregational unit that is free to segregate from heterologous chromosomes.

Mechanical basis of meiotic metaphase arrest

Control of the metaphase to anaphase transition is a central component of cell-cycle regulation. Arrest at either metaphase I or II before fertilization is a common component of oogenesis in many organisms. In Drosophila melanogaster females, this arrest occurs at meiosis I with the chiasmate bivalents tightly massed at the metaphase plate and the nonexchange chromosomes positioned between the plate and the poles on long tapered spindles. Meiosis resumes only after passage through the oviduct. Thus, metaphase arrest defines an important checkpoint in the meiotic cell cycle. We report here that this arrest results from the balancing of chiasmate bivalents at the metaphase plate. Two meiotic mutations, mei-9b and mei-218a4, both of which greatly reduce the frequency of chiasma formation, bypass the metaphase block and allow stage 14 oocytes to finish both meiotic divisions without arrest. We conclude that metaphase arrest results from the balancing of kinetochore forces due to chiasmata.

Purification and characterization of an endo-exonuclease from adult flies of Drosophila melanogaster

An endo-exonuclease (designated nuclease III) has been purified to near homogeneity from adult flies of Drosophila melanogaster. The enzyme degrades single- and double-stranded DNA and RNA. It has a sedimentation co-efficient of 3.1S and a strokes radius of 27A The native form of the purified enzyme appears to be a monomer of 33,600 dalton. It has a pH optimum of 7-8.5 and requires Mg2+ or Mn2+ but not Ca2+ or Co2+ for its activity. The enzyme activity on double-stranded DNA was inhibited 50% by 30 mM NaCl, while its activity on single-stranded DNA required 100 mM NaCl for 50% inhibition. Under the latter conditions, its activity on double-stranded DNA was inhibited approximately 98%. The enzyme degrades DNA to complete acid soluble products which are a mixture of mono- and oligonucleotides with 5'-P and 3'-OH termini. Supercoiled DNA was converted by the enzyme to nicked and subsequently to linear forms in a stepwise fashion under the condition in which the enzyme works optimally on single-stranded DNA. The amino acid composition and amino acid sequencing of tryptic peptides from purified nuclease III is also reported.

Meiotic spindle assembly in Drosophila females: behavior of nonexchange chromosomes and the effects of mutations in the nod kinesin-like protein

Mature Drosophila oocytes are arrested in metaphase of the first meiotic division. We have examined microtubule and chromatin reorganization as the meiosis I spindle assembles on maturation using indirect immunofluorescence and laser scanning confocal microscopy. The results suggest that chromatin captures or nucleates microtubules, and that these subsequently form a highly tapered spindle in which the majority of microtubules do not terminate at the poles. Nonexchange homologs separate from each other and move toward opposite poles during spindle assembly. By the time of metaphase arrest, these chromosomes are positioned on opposite half spindles, between the metaphase plate and the spindle poles, with the large nonexchange X chromosomes always closer to the metaphase plate than the smaller nonexchange fourth chromosomes. Nonexchange homologs are therefore oriented on the spindle in the absence of a direct physical linkage, and the spindle position of these chromosomes appears to be determined by size. Loss-of-function mutations at the nod locus, which encodes a kinesin-like protein, cause meiotic loss and nondisjunction of nonexchange chromosomes, but have little or no effect on exchange chromosome segregation. In oocytes lacking functional nod protein, most of the nonexchange chromosomes are ejected from the main chromosomal mass shortly after the nuclear envelope breaks down and microtubules interact with the chromatin. In addition, the nonexchange chromosomes that are associated with spindles in nod/nod oocytes show excessive poleward migration. Based on these observations, and the structural similarity of the nod protein and kinesin, we propose that nonexchange chromosomes are maintained on the half spindle by opposing poleward and anti-poleward forces, and that the nod protein provides the anti-poleward force.

There are two mechanisms of achiasmate segregation in Drosophila females, one of which requires heterochromatic homology

There are numerous examples of the regular segregation of achiasmate chromosomes at meiosis I in Drosophila melanogaster females. Classically, the choice of achiasmate segregational partners has been thought to be independent of homology, but rather made on the basis of availability or similarities in size and shape. To the contrary, we show here that heterochromatic homology plays a primary role in ensuring the proper segregation of achiasmate homologs. We observe that the heterochromatin of chromosome 4 functions as, or contains, a meiotic pairing site. We show that free duplications carrying the 4th chromosome pericentric heterochromatin induce high frequencies of 4th chromosome nondisjunction regardless of their size. Moreover, a duplication from which some of the 4th chromosome heterochromatin has been removed is unable to induce 4th chromosome nondisjunction. Similarly, in the absence of either euchromatic homology or a size similarity, duplications bearing the X chromosome heterochromatin also disrupt the segregation of two achiasmate X chromosome centromeres. Although heterochromatic regions are sufficient to conjoin nonexchange homologues, we confirm that the segregation of heterologous chromosomes is determined by size, shape, and availability. The meiotic mutation Axs differentiates between these two processes of achiasmate centromere coorientation by disrupting only the homology-dependent mechanism. Thus there are two different mechanisms by which achiasmate segregational partners are chosen. We propose that the absence of diplotene-diakinesis during female meiosis allows heterochromatic pairings to persist until prometaphase and thus to co-orient homologous centromeres. We also propose that heterologous disjunctions result from a separate and homology-independent process that likely occurs during prometaphase. The latter process, which may not require the physical association of segregational partners, is similar to those observed in many insects, in Saccharomyces cerevisiae and in C. elegans males. We also suggest that the physical basis of this process may reflect known properties of the Drosophila meiotic spindle.

The lethal(1)TW-6cs 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

The l(1)TW-6cs mutation is a cold-sensitive recessive lethal mutation in Drosophila melanogaster, that affects both meiotic and mitotic chromosome segregation. We report the isolation of three revertants of this mutation. All three revert both the meiotic and mitotic effects as well as the cold sensitivity, demonstrating that all three phenotypes are due to a single lesion. We further show that these revertants fail to complement an amorphic allele of the nod (no distributive disjunction) locus, which encodes a kinesin-like protein. These experiments demonstrate that l(1)TW-6cs is an antimorphic allele of nod, and we rename it nodDTW. Sequencing of the nod locus on a nodDTW-bearing chromosome reveals a single base change in the putative ATP-binding region of the motor domain of nod. Recessive, loss-of-function mutations at the nod locus specifically disrupt the segregation of nonexchange chromosomes in female meiosis. We demonstrate that, at 23.5 degrees, the meiotic defects in nodDTW/+ females are similar to those observed in nod/nod females; that is, the segregation of nonexchange chromosomes is abnormal. However, in nodDTW/nodDTW females, or in nodDTW/+ females at 18 degrees, we observe a more severe meiotic defect that apparently affects the segregation of both exchange and nonexchange chromosomes. In addition, nodDTW homozygotes and hemizygous males have previously been shown to exhibit mitotic defects including somatic chromosome breakage and loss. We propose that the defective protein encoded by the nodDTW allele interferes with proper chromosome movement during both meiosis and mitosis, perhaps by binding irreversibly to microtubules.

Genetic analysis of microtubule motor proteins in Drosophila: a mutation at the ncd locus is a dominant enhancer of nod

The nod (no distributive disjunction) and the ncd (non-claret disjunctional) mutations are both female-specific, recessive meiotic mutations in Drosophila melanogaster. Mutations at either locus show high frequencies of nondisjunction at meiosis I and both have been shown to encode kinesin-like proteins. Unlike the ncd mutation, which affects all chromosome pairs, the nod mutation affects only the disjunction of nonexchange chromosomes. Although both the nod and ncd mutations are fully recessive, females doubly heterozygous for nod and ncd mutations show levels of X and fourth chromosome nondisjunction that are 6- to 35-fold above those observed in control females. Exchange between chromosomes can suppress this effect; thus, only nonexchange chromosomes segregating via the distributive system are sensitive in double heterozygotes. Since the phenotype of double heterozygotes mimics that of the nod mutation, we infer that ncd is a dominant enhancer of nod. Failure of ncd to fully complement nod reveals the chromosome segregation machinery to be dosage sensitive. The probability that the distributive system will fail is enhanced in females simultaneously haploinsufficient at the nod and ncd loci.

A kinesin-like protein required for distributive chromosome segregation in Drosophila

The nod gene is required for the distributive segregation of nonexchange chromosomes during meiosis in D. melanogaster. Loss-of-function nod mutations cause nondisjunction and loss of nonrecombinant chromosomes both at meiosis I and during subsequent mitotic divisions. We have cloned the nod locus, examined its expression patterns, and determined its coding sequence. In adults the nod transcript is only present in females, consistent with the observation that males do not use the distributive segregation system. However, the nod locus is also transcribed in the embryonic, larval, and pupal stages of development, and possibly in all dividing cells. Finally, the N-terminal domain of the predicted nod protein has amino acid similarity to the mechanochemical domain of kinesin heavy chain; however, the C-terminal domain is unlike that of kinesin heavy chain or of any previously reported protein. Thus, the nod protein is a member of the kinesin superfamily and may be a microtubule motor.

The genetic analysis of distributive segregation in Drosophila melanogaster. II. Further genetic analysis of the nod locus

In Drosophila melanogster females the segregation of nonexchange chromosomes is ensured by the distributive segregation system. The mutation noda specifically impairs distributive disjunction and induces nonexchange chromosomes to undergo nondisjunction, as well as both meiotic and mitotic chromosome loss. We report here the isolation of seven recessive X-linked mutations that are allelic to noda. As homozygotes, all of these mutations exhibit a phenotype that is similar to that exhibited by noda homozygotes. We have also used these mutations to demonstrate that nod mutations induce nonexchange chromosomes to nondisjoin at meiosis II. Our data demonstrate that the effects of noda on meiotic chromosome behavior are a general property of mutations at the nod locus. Several of these mutations exhibit identical phenotypes as homozygotes and as heterozygotes with a deficiency for the nod locus; these likely correspond to complete loss-of-function or null alleles. None of these mutations causes lethality, decreases the frequency of exchange, or impairs the disjunction of exchange chromosomes in females. Thus, either the nod locus defines a function that is specific to distributive segregation or exchange can fully compensate for the absence of the nod+ function.