The Cohesion Protein MEI-S332 Localizes to Condensed Meiotic and Mitotic Centromeres until Sister Chromatids Separate.

Moore, D. P., Page, A. W., Tang, T. T., Kerrebrock, A. W., and Orr-Weaver, T. L. (1998) Journal of Cell Biology 140: 1003-1012.

Paper reviewed by Kelly E. Westbrook

Previous research has indicated that, in spermatocytes, the Drosophila protein MEI-S332 is both essential for cohesion between sister chromatids and localized to the centromeric regions of meiotic chromosomes. Cohesion between sister chromatids is essential for chromosome segregation that occurs during both mitosis and meiosis, and cohesion defects have been implicated in oncogenesis and meiotic errors. Studies have also shown that, in males, MEI-S332 is maintained in the centromeric regions of meiotic chromosomes through the metaphase I-anaphase I transition but is no longer present after anaphase II, the time when sister chromatids separate. Along with genetic assays showing that mutant mei -S332 males have normal segregation in the first meiotic division and high levels of missegregation in the second division, the absence of MEI-S332 after the second meiotic division indicates that the protein is required primarily for proper separation between metaphase II and anaphase II.

Most previous studies have been conducted on MEI-S332 in Drosophila spermatocytes. However, Drosophila male meiosis has some unusual features, while the process of meiosis in female flies more closely resembles that in most eukaryotes. Therefore, this paper examines the localization of MEI-S332 during meiosis in Drosophila females. The fate of the protein at the second meiotic division is also examined, and consequently leads researchers to look at its localization during mitosis in developing embryos.

In order to observe the presence of MEI-S332, researchers created a fusion transgene that caused cells to express Green Flourscense Protein (GFP) on the amino terminus of MEI-S332. This fusion protein allowed them to localize MEI-S332 in the various stages of meiosis. Figure 1 shows the fusion protein in fixed oocytes that were also stained for DNA, making the chromatin appear in red. The orientation of the green MEI-S332 in relation to the red-flourescing chromatin indicates that the protein is localized to the centromeric regions of the chromosomes from metaphase I through metaphase II. The first panel (Figure 1-A) shows the oocyte arrested at metaphase I. Here, MEI-S332 appears as two caps at opposite ends of the neucleus, where previous research has indicated the centromeres are positioned during metaphase I. Thus, MEI-S332-GFP likely represents centromeric localization.

Oocytes expressing the fusion protein were activated in vitro to complete meiosis, and the rest of the panels in Figure 1 represent the various meiotic stages. In anaphase I (Figure 1-B) researchers claim to observe eight pairs of sister chromatids, with MEI-S332 detectable at the leading edges of each pair. However, I can only detect 6 "dots" of MEI-S332 in panel B. These dots are, however, positioned at the leading edge of the chromosomes, which supports the argument that MEI-S332 is localized at chromosome centromeres. Panels C,D, and E show the oocyte at late anaphase I, the interim between the first and second meiotic divisions, and metaphase II, respectively. In each of these panels, MEI-S332-GFP appears as a "dot" associated with a pair of sister chromatids. The release of sister chromatid cohesion in anahpase II coincides with the lack of MEI-S332-GFP, as shown in panel F. As well, the protein was not detectable during postmeiotic interphase (Figure 1-G). Thus the researchers indicate that MEI-S332 is localized to the centromeric region of the chromosomes and is present in meiosis from metaphase I through the metaphase II-anaphase II transition. Figure 1 supports these conclusions rather well, and studies are cited to indicate the position of centromeric regions throughout meiosis.

The timing of MEI-S332 centromeric localization in oocytes was examined to asses any difference with that of spermatocytes. In spermatocytes the protein is found in the cytoplasm during prophase I and localized to chromosomes during prometaphase I. The researchers indicate that MEI-S332 is not observed in oocyte chambers during prophase I, but do not show the data to support this. Although this is not the most important point in their argument, I feel it would strengthen the paper to present all data to which they refer. Figure 2 shows the presence of MEI-S332-GFP in stages 13 and 14 of oocyte development. The chromatin is also stained in red and the tubulin stained in blue, so that DNA and spindle formation can be compared with localiztion of the protein. Researchers claim that the earliest stage at which MEI-S332 was observed coincided with the beginning of spindle formation, and they present this as the first panel in Figure 2 (panel A). In panel A, MEI-S332-GFP appears scattered through the chromosomal mass. However, as the spindle (blue staining) appears longer and becomes more bipolar, the protein is concentrated more at the ends of the chromatin. Thus, although the first observation of MEI-S332 is not indicated with data, we can see that the protein is not localized to the centromeric region of the chromosomes until stage 14 of oocyte development, which corresponds wtih prometaphase I of the meiotic stages.

Because Figure 1 indicated that MEI-S332 was not present on the sister chromatids after they separated at anaphase II, researchers hypothesized that it was degraded at the metaphase II-anaphase II transition. In order to test this hypothesis, they examined the levels of protein in the oocytes by generating a polyclonal antibody against the carboxyl terminus of the MEI-S332 protein. Figure 3-A shows an illustration of the MEI-S332 protein. The large arrow designates the site of antibody generation, and the small arrow designates the site of a nonsense mutation in a mutant strain of flies. This strain, mei-S332⁷, has the designated mutation which causes it to stop translation of the MEI-S332 protein before reaching the carboxyl terminus. Oocytes from this mutant strain were used as a negative control on a Western blot for MEI-S332 in oocytes, ovaries and embryos (Figure 3-B). After determining that the antibodies labeled the protein, researchers could assess whether or not it is degraded after metaphase II. Because the samples of wild type ovaries and embryos all revealed a band of the same length (~55kD), researchers concluded that the band must represent the MEI-S332 protein. They claim that the band was not evident in samples from the mutant strain (the negative control), however, I observe a faint band in at least one of these negative control lanes. This figure is also lacking a posative control that yields a band in every lane, proving that equal amounts of sample were added to each lane. Never-the-less, the probe is more intense in each of the wild type lanes and significantly more intense in lane 1, which contains a sample of oocytes from a strain with 6 copies of the fusion transgene (thus it should be productin 6 times as much MEI-S332-GFP). Thus, the researchers accept that the ~55kD bands correspond with MEI-S332 and use them to examine the fate of the protein after meiosis.

Panel C in figure 3 compares protein levels between unactivated oocytes and oocytes that have completed meiosis (therefore passed through the metaphase II - anaphase II transition). Again, the mutant mei-S332⁷ strain was used as a negative control, and a postaive control was added to demonstrate equal loading of samples in all lanes. Because the bands in lanes 3 and 4 have the same intensity, they indicate that the level of MEI-S332 remains unchanged from the beginning of meiosis through the metaphase II- anaphase II transition. This suggests that even though the protein dissociates from the chromosomes at anaphase II (shown in figure 1) it is not degraded. In order to control for the possibility that the protein levels may be equal simply due to continuing translation of MEI-S332, the researchers activated the oocytes in the presence of a translational inhibitor (Fig. 3-C, lanes 5 & 6). However, even when the oocytes were incubated in the translantional inhibitor, the protein was evident in equal amounts both before and after meiosis. The conclusion that the protein is not degraded is thus supported by the data.

After finding that the MEI-S332 protein was not degraded after meiosis (Figure 3-C) but persisted into embryonic development (Figure 3-B, lanes 6&7), researchers attempted to localize it during mitosis. They again used the fusion protein and labeled the DNA in red. They observed the MEI-S332-GFP in embryos on "polar body rosettes," which result from unused meiotic products (Figure 4-A). Specifically, MEI-S332 was seen on the inside ring of the rosette, where the centromeres are believed to be (Fig. 4-B). Figures 4-C and 5-B show the protein as discrete dots on the metaphase plate, and 5-B localizes MEI-S332 to the leading edge of the chromosomes in early anaphase. Thus, they conclude that MEI-S332 is also localized to the centromeric regions of chromosones during mitosis. They indicate that the protein dissociates when the sister chromatids separate in mitosis (as in meiosis), because it is no longer detectable on mid-anaphase chromosomes but is evident in clouds surrounding the interphase neuclei (Fig. 4-D&E).

Because MEI-S332 was not previously indicated in mitosis, researchers are left to speculate about its function in this process. Much of the paper's discussion is devoted to such speculation. The researchers argue that mei-S332 mutants do not show a phenotype difference, and therefore the protein must not be necessary for mitosis. They indicate that the system may be redundant, having two proteins that are localized to the centromeres during mitosis, so no phenotypic alterations would be evident with mei-S332 mutation. Future studies should search for such a protein by looking at MEI-S332 protein interactions. The two-hybrid system would allow researchers to isolate and clone proteins that interact with MEI-S332. The levels of these proteins could then be assesed by flourescense in the same way MEI-S332 was examined throughout the cell cycles (creation of a fusion protein with GFP).

The researchers also examine the role of the protein at the metaphase-anaphase transitions, since they observed its dispersal at this stage in both meiosis and mitosis. They suggest the possibilities that either MEI-S332 dissociation from the centromeres triggers chromatid separation or that it is dispersed as a result of the separation that occurs in anaphase. They support this second theory with the fact that MEI-S332 was found on the centromeres of chromosomes in early anaphase, but at lower levels than in metaphase. Measuring the differences in expression of the protein throughout each of the different stages may give them a more concrete indication as to which of these theories is more accurate. Future studies might use pulse-chase analysis to more closely follow the levels of MEI-S332 in both oocytes and in early embryonic development. Additionally, the question of whether MEI-S332 is necessary for establishment of sister-chromatid cohesion was raised. Flourescence In Situ Hybridization could be used to determine exactly which stage cohesion is established. The feasibility of MEI-S332 involvement in that process could then be assessed.