InteractiveFly: GeneBrief

matrimony: Biological Overview | References

Many meiotic systems in female animals include a lengthy arrest in G2 that separates the end of pachytene from nuclear envelope breakdown (NEB). However, the mechanisms by which a meiotic cell can arrest for long periods of time (decades in human females) have remained a mystery. The Drosophila Matrimony (Mtrm) protein is expressed from the end of pachytene until the completion of meiosis I. Loss-of-function mtrm mutants result in precocious NEB. Coimmunoprecipitation experiments reveal that Mtrm physically interacts with Polo kinase (Polo) in vivo, and multidimensional protein identification technology mass spectrometry analysis reveals that Mtrm binds to Polo with an approximate stoichiometry of 1:1. Mutation of a Polo-Box Domain (PBD) binding site in Mtrm ablates the function of Mtrm and the physical interaction of Mtrm with Polo. The meiotic defects observed in mtrm/+ heterozygotes are fully suppressed by reducing the dose of polo+, demonstrating that Mtrm acts as an inhibitor of Polo. Mtrm acts as a negative regulator of Polo during the later stages of G2 arrest. Indeed, both the repression of Polo expression until stage 11 and the inactivation of newly synthesized Polo by Mtrm until stage 13 play critical roles in maintaining and properly terminating G2 arrest. These data suggest a model in which the eventual activation of Cdc25 (Drosophila Twine) by an excess of Polo at stage 13 triggers NEB and entry into prometaphase (Xiang, 2007).

The mechanism of the lengthy arrest in G2 that separates the end of pachytene from nuclear envelope breakdown (NEB) (which is a characterization of many female meiotic systems) has remained a mystery. One can imagine that both the maintenance and the termination of this arrest might involve either or both of two mechanisms: the transcriptional or translational repression of a protein that induces NEB, and thus meiotic entry, or the presence of an inhibitory protein that precludes entry into the first meiotic division. Because Drosophila females exhibit a prolonged G2 arrest and are amenable to both genetic and cytological analyses, they provide an ideal system in which to study this problem (Xiang, 2007).

The ovaries of Drosophila females are composed of a bundle of ovarioles, each of which contains a number of oocytes arranged in order of their developmental stages. For the purposes of this study, the process of oogenesis may be said to consist of three separate sets of divisions: the initial stem cell divisions, which create primary cystoblasts; four incomplete cystoblast divisions, which create a 16-cell cyst that contains the oocyte; and the two meiotic divisions. Although a great deal is known regarding the mechanisms that control cystoblast divisions and oocyte differentiation, relatively little is known about the mechanisms by which the progression of meiosis is controlled (Xiang, 2007).

As is the case in many meiotic systems, female meiosis in Drosophila involves preprogrammed developmental pauses. The two most prominent pauses during Drosophila meiosis are an arrest that separates the end of pachytene at stages 5-6 from NEB at stage 13, and a second pause that begins with metaphase I arrest at stage 14 and continues until the egg passes through the oviduct. It is the release of this second preprogrammed arrest event that initiates anaphase I and allows the completion of meiosis I followed by meiosis II. The end of meiotic prophase by dissolution of the synaptonemal complex (SC) at stages 5-6 is separated from the beginning of the meiotic divisions, which is defined by NEB at stage 13, by approximately 40 h to allow for oocyte growth (Xiang, 2007).

Elucidating the mechanisms that arrest meiotic progression at the end of prophase, but then allow onset of NEB and the initiation of meiotic spindle formation some 40 h later, was of great interest. One intriguing possibility is that during this period of meiotic arrest, the oocyte actively blocks the function of cell cycle regulatory proteins such as cyclin dependent kinase 1 (Cdk1), the phosphatase Cdc25, and Polo kinase (Polo), all of which promote meiotic progression just as they do during mitotic growth. Recently, Polo was shown to be expressed in the germarium and required for the proper entry of Drosophila oocytes into meiotic prophase, as defined by the assembly of the SC (Mirouse, 2006). Decreased levels of Polo resulted in delayed entry into meiotic prophase, whereas overexpression of Polo caused a dramatic increase in the number of cystocyte cells entering meiotic prophase, indicating that Polo is involved both in the initiation of SC formation and in the restriction of meiosis to the oocyte. How then is Polo, which is known to play multiple roles in promoting meiotic and mitotic progression (Lee, 2003a; Lee, 2003b), prevented from compelling the differentiated oocyte to proceed further into meiosis? One component of this regulation may well lie in the fact that Polo is not expressed during much of oogenesis. Polo is clearly visible in the germarium but is then absent until stage 11, when it begins to accumulate to high levels in the oocyte. A second component of Polo regulation is mediated by binding to the protein product of the matrimony (mtrm) gene, which occurs from stage 11 until the onset of NEB at stage 13. This binding serves to inhibit Polo in the early stages of its expression, and thus prevents precocious nuclear envelope breakdown (Xiang, 2007).

The mtrm gene was first identified in a deficiency screen for loci that were required in two doses for faithful meiotic chromosome segregation (Harris, 2003). mtrm/+ heterozygotes display a significant defect in achiasmate segregation (the meiotic process that ensures the segregation of those homologs that, for various reasons, fail to undergo crossingover). As a result of this defect, mtrm/+ heterozygotes exhibit high levels of achiasmate nondisjunction. As homozygotes, mtrm mutants are fully viable but exhibit complete female sterility. This study shows that the Mtrm protein prevents precocious NEB. Indeed, the effects of reducing the dose of mtrm on meiotic progression and on chromosome segregation are easily explained as the consequence of precocious NEB at stages 11 or 12, and can be suppressed by simultaneously reducing the copy number of polo⁺. In addition, the effects of heterozygosity for loss-of-function alleles of mtrm can be phenocopied by increasing the copy number of polo⁺. These genetic interactions suggest that Mtrm negatively regulates Polo in vivo (Xiang, 2007).

Interestingly, Mtrm was shown to interact physically with Polo by a global yeast two-hybrid study. This yeast two-hybrid finding reflects a true physical interaction in vivo by both coimmunoprecipitation studies and by multidimensional protein identification technology (MudPIT) mass spectrometry experiments, which indicate that Mtrm binds to Polo with an approximate stoichiometry of 1:1. Moreover, ablating one of the two putative Polo binding sites on Mtrm by mutation prevents the physical interaction between Polo and Mtrm and renders the mutated Mtrm protein functionless. This experiment, along with genetic interaction studies, provides compelling evidence that the function of the binding of Mtrm to Polo is to inhibit Polo, and not vice versa (Xiang, 2007).

The analysis of mtrm mutants allows us examination of the effects of premature Polo function during oogenesis. The evidence shows that in the absence of Mtrm, newly synthesized Polo is capable of inducing NEB from stage 11 onward. As a result of this precocious NEB, chromosomes are not properly compacted into a mature karyosome and they are released prematurely onto the meiotic spindle. In many cases, the centromeres of achiasmate bivalents subsequently fail to co-orient (Xiang, 2007).

The mtrm gene was first identified as a dosage-sensitive meiotic locus; heterozygosity for a loss-of-function allele of mtrm specifically induced the failed segregation of achiasmate homologs (Harris, 2003). The mtrm gene encodes a 217-amino acid protein with two Polo-Box Domain (PBD) binding sites (STP and SSP) and a C-terminal SAM/Pointed domain. The studies reported in this paper rely primarily on a null allele of mtrm (mtrm¹²⁶), which removes 80 bp of upstream sequence and the sequences encoding the first 41 amino acids of the Mtrm protein (Xiang, 2007).

Western blot analysis using an antibody to Mtrm reveals that Mtrm can be detected only in ovaries. This is consistent with a previous report by Arbeitman (2002), which showed that the expression profile of the mtrm gene product was strictly maternal and that its expression was reduced greater than 10-fold over 0-6.5 h of embryonic development. The specificity of this antibody is demonstrated by the fact that no signal was detected by either Western blotting or by immunofluorescence of ovarioles homozygous for the mtrm¹²⁶ mutant. Immunofluorescence studies using the same antibody reveal that Mtrm is expressed as a diffuse nuclear protein in the oocytes and nurse cells beginning at stage 4-5. The Mtrm signal was not restricted to the karyosome itself; but rather Mtrm seems to fill the space in the entire nucleus. Although Mtrm is restricted to the nucleus until approximately stage 10, it localizes throughout the oocyte in later stages. Mtrm brightly stains both the oocyte nucleus and cytoplasm between stage 11 and stage 12, but staining is greatly reduced at stage 13, the stage at which NEB occurs (Xiang, 2007).

The data presented in this study that Mtrm serves to inactivate newly synthesized Polo during the period of meiotic progression that precedes NEB. An excess of functional (unbound) Polo, produced by reducing the amount of available Mtrm, causes the early onset of NEB. This early entry into prometaphase releases an immature karyosome into the cytoplasm, which then fails to properly align the centromeres of achiasmate chromosomes on the prometaphase spindle. These observations raise a number of questions ranging from the role of Polo in mediating the G2/M transition in oogenesis to the role of the karyosome structure in facilitating the proper segregation of achiasmate chromosomes (Xiang, 2007).

The trigger for the G2/M transition is activation of Cdk1 by Cdc25, and multiple lines of evidence suggest that Polo can activate Cdc25. First, in Caenorhabditis elegans, RNAi experiments demonstrate that ablation of Polo prevents NEB. Second, the Xenopus Polo homolog Plx1 is activated in vivo during oocyte maturation with the same kinetics as Cdc25. Additionally, microinjection of Plx1 accelerates the activation of both Cdc25 and cyclinB-Cdk1. Moreover, microinjection of either an antibody to Plx1 or kinase-dead mutant of Plx1 inhibits the activation of Cdc25 and its target cyclinB-Cdk1. Another demonstrated that injection of a constitutively active form of Plx1 accelerated Cdc25 activation. These studies support "the concept that Plx1 is the 'trigger' kinase for the activation of Cdc25 during the G2/M transition." Finally, a small molecule inhibitor of Polo kinase (BI 2536) also results in extension of prophase (Lenart, 2007). These data are consistent with the view that the presence of functional (unbound) Polo plays a critical role in ending the extended G2 that is characteristic of oogenesis in most animals (Xiang, 2007).

In light of these data, it is tempting to suggest that in wild-type Drosophila oocytes, the large quantity of Mtrm deposited into the oocyte from stage 10 onward inhibits the Polo that is either newly synthesized or transported into the oocyte during stages 11-12. However, at stage 13, an excess of functional Polo is created when the number of Polo proteins exceeds the available amount of inhibitory Mtrm proteins. This unencumbered and thus functional Polo then serves to activate Cdc25, initiating the chain of events that leads to NEB and the initiation of prometaphase. In the absence of a sufficient amount of Mtrm, an excess of Polo causes the precocious activation of Cdc25, and thus an early G2/M transition. A model describing this hypothesis is presented. Based on this model, one can visualize that decreasing the dose of Mtrm or increasing the dose of Polo will hasten NEB, whereas simultaneous reduction in the dose of both proteins should allow for proper timing of NEB (Xiang, 2007).

Mtrm's first PBD binding site (T40) is required for its interaction with Polo. Mtrm T40 has to be first phosphorylated by a priming kinase, such as one of the Cdks or MAPKs, and was indeed detected as phosphorylated in the mass spectrometry dataset. The NetPhosK algorithm predicts T40 to be a Cdk5 site, and the serines immediately distal to T40 (S48 and S52), which were also detected as phosphorylated are sites for proline-directed kinases such as Cdk or MAPK sites as well. The other prominent phosphorylation event occurs at S137, which could be a Polo phosphorylation site since it falls within a Polo consensus (D/E-X-S/T-Ø-X-D/E). Although the combined sequence coverage for Mtrm was 44%, indicating that some phosphorylated sites might have been missed, Mtrm S137 is a suitable binding site for activated Polo, in agreement with the processive phosphorylation model. At this point of our studies, Mtrm T40 priming kinase or the kinase responsible for Polo activating phosphorylation on T182 has not been identified (Xiang, 2007).

The finding that Polo not only is able to bind to Mtrm in vivo in a 1:1 ratio, but also is fully phosphorylated on T182 in its activation loop suggests a method by which Mtrm serves to inhibit Polo. In general, enzymes are usually not recovered from affinity purifications at levels similar to their targets. They do not form stable complexes, but rather form transient interactions with their substrates, which is how efficient catalysis is achieved. In this instance, Mtrm is able to sequester activated Polo away in a stable binary complex over a long period of time. It is only when this equilibrium is disturbed at the onset of stage 13 by the production of an excess of Polo or by degradation of Mtrm that Polo can be released. The molecular determinants of the Mtrm::Polo sequestration event are not clear, but it would be interesting to test whether the serines found phosphorylated in the vicinity of Mtrm PBD binding sites play a role in locking the binary complex into place (Xiang, 2007).

The data demonstrate that a reduction in the levels of Mtrm results in the release of an incompletely compacted karyosome that rapidly dissolves into individual bivalents during the early stages of spindle formation. For chiasmate bivalents, this is apparently not a problem, because they still co-orient correctly. However, the nonexchange bivalents frequently fail to co-orient properly, such that both homologs are oriented toward the same pole (but often occupy two different arcs of the spindle). This initial failure of proper co-orientation leads to high frequencies of nondisjunction as demonstrated by the genetic studies and analysis of metaphase I images (Xiang, 2007).

Although achiasmate homologs are properly co-oriented in wild-type oocytes, it has been noted that such homologs can often vacillate between the poles such that two achiasmate homologs are often found on the same arc of the same half-spindle during mid to late prometaphase (Gilliland, 2007). These chromosomes are often observed to be physically associated. This situation is quite different from the defect observed in mtrm/+ heterozygotes, where the homologs are neither physically associated nor on the same arc of the spindle (Xiang, 2007).

It is tempting to suggest that the chromosome segregation defects observed in mtrm/+ heterozygotes are simply the result of precocious release of an incompletely re-compacted karyosome. According to this explanation, the defects observed in meiotic chromosome segregation are solely the consequence of premature NEB. (Implicit in this model is the assumption that it is the events that occur during karyosome re-compaction, at stages 11 and 12, that serve to initially bi-orient achiasmate chromosomes, and there is no direct evidence to support such a hypothesis) (Xiang, 2007).

Alternatively, Polo plays multiple roles in the meiotic process (Lee, 2003a; Lee, 2003b), and it is possible that the chromosome segregation defects we see represent effects of excess Polo that are un-related to the precocious breakdown of the nuclear envelope. Such a view is supported by two observations. First, the bivalent individualization observed after NEB in mtrm/+ oocytes does not disrupt FM7-X pairings. Second, although heterozygosity for twine in mtrm¹²⁶/+ heterozygotes suppresses the frequency of precocious NEB from 42%, two alleles of twine tested (twe¹ and twe^k08310) failed to suppress the levels of meiotic nondisjunction observed in FM7/X; mtrm¹²⁶/+ heterozygotes. These data suggest that the effects of excess Polo on nondisjunction may not be regulated via Cdc25/Twine, but rather by the effects of excess Polo on some other as-yet-unidentified Polo target. This suggests that the effects of Mtrm on the level of Polo might affect multiple Polo-related processes (Xiang, 2007).

Support for such an idea that Mtrm can inhibit Polo-regulated proteins that are unrelated to NEB comes from the observation that the ectopic expression of Drosophila Mtrm in Schizosaccharomyces pombe blocks karyokinesis, producing long multi-septate cells with only one or two large nuclei (Edgar, 1994; Edgar, personal communication to Xiang, 2007). This phenotype is similar, if not identical to, that exhibited by mutants in the S. pombe Polo homolog plo1 (Plo1), which fail in later stages of mitosis due to the role of Plo1 in activating the septation initiation network to trigger cytokinesis and cell division. However, Plo1 also plays a role in bipolar spindle assembly that might also be inhibited in the Mtrm expressing cells, but this function of Plo1 is less well understood (Xiang, 2007).

Thus the possibility exists that the effect of mtrm mutants on meiotic chromosome segregation may well not be the direct consequence of early NEB, but rather may be due to the role of Polo in other meiotic activities, such as spindle formation or the combined effects of these defects with precocious NEB. Efforts to identify such processes and their components are underway (Xiang, 2007).

Finally, it should be noted that while Mtrm is the first known protein that is able to inactivate Polo by physical interaction with Polo itself; there is certainly additional mechanisms of Polo regulation. For example, Archambault (2007) has described mutants in the gene that encodes Greatwall/Scant kinase, which have both late meiotic and mitotic defects. Although there is no evidence for a physical interaction between these two kinases, the authors speculate that the function of the Greatwall kinase serves to antagonize that of Polo. The Scant mutations create a hyperactive form of Greatwall, which might be expected to lower the dosage Polo, and thus perhaps partially suppress the defects observed in mtrm/+ heterozygotes. Indeed, exactly such a suppressive effect has been observed in Scant homozygotes (however, this suppression is much weaker than that obtained by heterozygosity for loss of function alleles of polo) (Xiang, 2007).

The data presented in this study demonstrate that Mtrm acts as a negative regulator of Polo during the later stages of G2 arrest during meiosis. Indeed, both the repression of Polo expression until stage 11 and the inactivation of newly synthesized Polo by Mtrm until stage 13 play critical roles in maintaining and properly terminating G2 arrest. Our data suggest a model in which the eventual activation of Cdc25 by an excess of Polo at stage 13 triggers NEB and entry into prometaphase. Although the data do shed some light on the mechanism by which Mtrm inhibits Polo, it is not entirely clear whether Polo's ability to phosphorylate targets other than Cdc25 might be blocked by Mtrm::Polo binding. These issues will need to be addressed in the future studies. Finally, it is noted that although small molecule inhibitors of Polo have been identified, Mtrm represents the first case of a protein inhibitor of Polo. It would be most exciting to identify functional orthologs of Mtrm outside of the genus Drosophila. Perhaps that might best be accomplished through a screen for oocyte-specific Polo-interacting proteins (Xiang, 2007).

A deficiency screen of the major autosomes identifies a gene (matrimony) that is haplo-insufficient for achiasmate segregation in Drosophila oocytes

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. It is concluded that CG18543 is the haplo-insufficient gene, and this gene has been named 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 (Harris, 2003; full text of article).

Search PubMed for articles about Drosophila Matrimony

Arbeitman, M. N., et al. (2002). Gene expression during the life cycle of Drosophila melanogaster. Science 297: 2270-2275. PubMed citation: 12351791

Archambault, V., Zhao, X., Carpenter, A. T. and Glover, D. M. (2007). Mutations in Drosophila Greatwall/Scant reveal its roles in mitosis and meiosis and suggest interdependence with Polo kinase. PLoS Genet 3(11): e200. PubMed citation: 17997611

Edgar, B. A., et al. (1994). Distinct molecular mechanism regulate cell cycle timing at successive stages of Drosophila embryogenesis. Genes Dev 8: 440-452. PubMed citation: 7510257

Gilliland, W. D., et al. (2007). The multiple roles of Mps1 in Drosophila female meiosis. PLoS Genet 3(7): e113. PubMed citation: 17630834

Harris, D., et al. (2003). A deficiency screen of the major autosomes identifies a gene (matrimony) that is haplo-insufficient for achiasmate segregation in Drosophila oocytes. Genetics 165(2): 637-52. PubMed citation: 14573476

Lee, B. H. and Amon, A. (2003). Polo kinase-meiotic cell cycle coordinator. Cell Cycle 2: 400-402. PubMed citation: 18482691

Lee, B. H. and Amon, A. (2003). Role of Polo-like kinase CDC5 in programming meiosis I chromosome segregation. Science 300: 482-486. PubMed citation: 12663816

Lenart, P., et al. (2007). The small-molecule inhibitor BI 2536 reveals novel insights into mitotic roles of polo-like kinase 1. Curr Biol 17: 304-315. PubMed citation: 17291761

Mirouse, V., Formstecher, E. and Couderc, J. L. (2006). Interaction between Polo and BicD proteins links oocyte determination and meiosis control in Drosophila. Development 133: 4005-4013. PubMed citation: 16971474

Xiang, Y., et al. (2007). The inhibition of polo kinase by matrimony maintains G2 arrest in the meiotic cell cycle. PLoS Biol. 5(12): e323. PubMed citation: 18052611

date revised: 10 July 2008

Home page: The Interactive Fly © 2008 Thomas Brody, Ph.D.