Continuous Cyclin E Expression Inhibits Progression Through Endoreduplication Cycles in Drosophila

Curr Biol. Author manuscript; available in PMC 2009 Sep 29.

Published in final edited form as:

PMCID: PMC2754250

NIHMSID: NIHMS142591

Sister Chromatids Fail to Separate during an Induced Endoreplication Cycle in Drosophila Embryos

The publisher's final edited version of this article is available at Curr Biol

See other articles in PMC that cite the published article.

Summary

When mitosis is bypassed, as in some cancer cells or in natural endocycles, sister chromosomes remain paired and produce four-stranded diplochromosomes or polytene chromosomes. Cyclin/Cdk1 inactivation blocks entry into mitosis and can reset G2 cells to G1, allowing another round of replication [1]. Reciprocally, persistent expression of Cyclin A/Cdk1 or Cyclin E/Cdk2 blocks Drosophila endocycles [2, 3]. Inactivation of Cyclin A/Cdk1 by mutation or overexpression of the Cyclin/Cdk1 inhibitor, Roughex (Rux), converts the 16^th embryonic mitotic cycle to an endocycle [4–6]; however, we show that Rux expression fails to convert earlier cell cycles unless Cyclin E is also downregulated. Following induction of a Rux transgene in Cyclin E mutant embryos during G2 of cell cycle 14 (G2₁₄), Cyclins A, B, and B3 disappeared and cells reentered S phase. This rereplication produced diplochromosomes that segregated abnormally at a subsequent mitosis. Thus, like the yeast CKIs Rum1 and Sic1, Drosophila Rux can reset G2 cells to G1 [7–9]. The observed cyclin destruction suggests that cell cycle resetting by Rux was associated with activation of the anaphase-promoting complex (APC), while the presence of diplochromosomes implies that this activation of APC outside of mitosis was not sufficient to trigger sister disjunction.

Results and Discussion

Cyclin A prevents G2 nuclei from resetting to G1 and rereplicating. Thus, when Drosophila embryos mutant for Cyclin A come to rely on zygotic expression of Cyclin A in cell cycle 16, their cells lose the normally abundant mitotic cyclins, reset to G1, and respond to a subsequent rise in the activity of Cyclin with its kinase partner, cyclin-dependent kinase (CDK), by entering S phase ([6, 10]; and F. Sprenger and P.H.O., unpublished data). Interestingly, induction of a Rux transgene under the control of a heat shock promoter (HS-Rux) in embryos containing both cycle 15 and cycle 16 cells caused abrupt loss of Cyclin A only in the cycle 16 cells (N.Y. and P.H.O., unpublished data). Cycle 16 differs from 15 and earlier cycles in the expression of the S phase cyclin, Cyclin E; it is continuously present early and is first downregulated in cycle 16 as cell cycle-regulated expression becomes evident [11, 12].

Downregulation of Cyclin E Causes Cyclin Disappearance in Cycle 14 Cells upon Rux Expression

To address whether Cyclin E/Cdk2, which is insensitive to Rux [9], modifies the response to Rux expression, we compared the responses of wild-type and Cyclin E mutant embryos. Flies heterozygous for Cyclin E^AR95 and homozygous for HS-Rux (AR95; HS-Rux) generated progeny that were either homozygous for the Cyclin E mutation (Cyclin E embryos) or carried a wild-type allele of Cyclin E on a balancer chromosome tagged with a β-galactosidase-expressing transgene (Cyclin E ⁺ embryos). By late cycle 14, Cyclin E mutant embryos have lower Cyclin E levels than wild-type embryos; nonetheless, maternal supplies of Cyclin E suffice for the first 16 cell divisions [11].

Expression of Rux during G2₁₄ had very different outcomes in Cyclin E ⁺ and Cyclin E mutant embryos. In Cyclin E ⁺ embryos, mitotic cyclins persisted following Rux expression, but the bulk of Cyclin A relocated from cytoplasm to nucleus, and the exclusively cytoplasmic Cyclin B became substantially nuclear (compare Figures 1C and 1D to Figures 1A and 1B and compare Figures 1H and 1K to Figures 1G and 1J). No shift was noted in the location of the already nuclear Cyclin B3 (data not shown). Mitosis was delayed in these embryos (data not shown), consistent with inhibition of mitotic Cyclin/Cdk1 activities. The relocalization of Cyclins A and B in response to Rux and the temporary block to mitosis are consistent with previous observations [9, 13]. The observed changes in localization presumably involve a shift in the steady state of nuclear import and export, perhaps by direct interaction of Rux with the Cyclin/Cdk1 complexes or indirect modulation of import/export by Rux.

Cyclins A and B Are Downregulated 45 min after Expression of Rux in Cyclin E Mutant Embryos

(A–L) Embryos in (A), (C), and (E) are stained for Cyclin A, while those in (B), (D), and (F) are stained for Cyclin B. Small sections of embryos in (A), (C), and (E) are magnified in (G), (H), and (I), respectively. Similarly, small sections of embryos in (B), (D), and (F) are magnified in (J), (K), and (L), respectively. (A) and (B) are wild-type embryos. Note that both the cyclins are normally cytoplasmic in interphase. This is best seen in (G) and (J), where central clearings representing nuclei are surrounded by rims of bright cytoplasmic staining. The bright foci of nuclear Cyclin A staining represent prophase cells, when Cyclin A becomes nuclear. (C), (D), (H), and (K) represent Cyclin E ⁺ embryos pulsed with Rux. Cyclin A in (C) and (H) appears predominantly nuclear, with some nuclei showing a high prophase level of cyclin A and others showing modest but, nonetheless, nuclear cyclin A (H). Cyclin B in (D) and (K) appears both nuclear and cytoplasmic. (E), (F), (I), and (L) represent Cyclin E mutant embryos pulsed with Rux, in which the cyclins are predominantly absent.

In Cyclin E mutant embryos, all three mitotic cyclins disappeared within 45 min of Rux expression (Figures 1E, 1F, 1I, 1L, and data not shown). Several lines of evidence show that this cyclin disappearance occurred without mitosis. Cyclins disappeared uniformly throughout the embryo rather than following the normal pattern of mitosis [14]. Wheat germ agglutinin (WGA) staining and antibody staining, respectively (data not shown), detected no nuclear membrane breakdown or mitotic histone phosphorylation. Mitotic chromosome condensation was not detected within 90 min of Rux expression (data not shown). Furthermore, centrioles did not exhibit stereotypical mitotic changes in conjunction with cyclin disappearance (S.J.V. and P.H.O., unpublished data). Lastly, nuclei retained interphase₁₄ size and density (Figures 2E–2G) (cell cycle identity can be inferred from nuclear size because cell divisions in the embryo occur in the absence of cell growth, resulting in a progressive decline in cell size; compare Figures 2F and 2G). We conclude that Rux overexpression induces interphase destabilization of the mitotic cyclins and that wild-type levels of Cyclin E block this effect.

G2₁₄ Cells in Cyclin E Mutant Embryos Rereplicate and Undergo Mitosis upon Rux Expression

(A) A timeline of events in Cyclin E mutant and Cyclin E ⁺ embryos with times indicated in minutes after a heat shock pulse that induced Rux during G2₁₄ (see Figure 1). In Cyclin E ⁺ embryos, Rux expression delays entry into M₁₄ so that the first cells to enter mitosis do so about 45 min after heat shock (data not shown). In Cyclin E mutant embryos, mitotic cyclins disappear between 45 and 75 min, BrdU incorporation (S phase) is observed between 90 and 135 min, and mitosis is first seen between 150 and 180 min.

(B–K) Micrographs of embryos (B–D) labeled with BrdU to show S phase, (E–G) stained with wheat germ agglutinin to reveal nuclear size, and (H–K) stained with Hoechst to show DNA. The relevant genotypes are indicated on the panels. (B) Note that BrdU labeling (a 30-min pulse starting 90 min after heat shock) in Cyclin E ⁺ embryos occurs in a pattern reminiscent of S₁₅, (C) but labeling in Cyclin E mutant embryos occurs uniformly and (D) does not occur at all in embryos arrested in G2₁₄ by a Cdc25^string mutation. Nuclear size in a (E) Rux-expressing Cyclin E mutant embryo during rereplication is similar to that of a (F) normal interphase₁₄ embryo. The DNA stain shows that nuclei in Rux-expressing embryos are smaller and more numerous in (H) Cyclin E ⁺ embryos than in the (I) Cyclin E mutants. A higher magnification view reveals mitotic cells in the (K) Cyclin E mutant embryos at this time (180 min after heat shock) as compared to smaller interphase cells in (J) Cyclin E ⁺ embryos.

(L) Progress through mitosis is represented by the ratio of prometaphase plus metaphase figures to anaphase plus telophase figures. Since each stage lasts on the order of 1 min, progression though a normal mitosis gives a ratio of about 1, as do Rux-expressing embryos heterozygous for Cyclin E. Cyclin E mutant embryos expressing Rux exhibited a higher ratio (3.5), indicating a delay in mitotic progression.

Disappearance of Mitotic Cyclins Is Followed by Rereplication

We labeled embryos with the nucleotide analog bromo deoxy uridine (BrdU) to follow the consequence of Rux expression on DNA replication. In undisturbed wild-type or Cyclin E mutant embryos, nuclei labeled equally during the synchronous S phase of cell cycle 14 (S₁₄) and labeled in a pattern similar to the spatial program of mitosis of cell cycle 14 (M₁₄) during S phase of cell cycle 15 (S₁₅) [15]. Following a control heat shock, M₁₄ began within 10 min, and cells entered S₁₅ immediately upon completion of mitosis (data not shown). Following induction of Rux in Cyclin E ⁺ embryos, which delayed M₁₄ (above), there was a corresponding delay in S₁₅, which was detected by labeling between 90 and 120 min after heat shock (Figure 2B). This delayed S₁₅ was patterned, a feature that is attributable to S phase occurrence secondary to a patterned mitosis [15]. Consistent with these nuclei having progressed through M₁₄, they are the size of interphase₁₅ nuclei and exhibit normal ploidy (similar to that in Figure 2G; data not shown).

The Cyclin E mutant embryos incorporated BrdU between 90 and 135 min after Rux induction (Figure 2C) without a preceding mitosis (above). Consistent with the absence of the patterned M₁₄, the incorporation was uniform rather than patterned (Figure 2C). Furthermore, nuclei retained a cycle 14 size and density (Figure 2E) and exhibited increased ploidy (below). The duration (about 45 min) and progression of incorporation (early diffuse labeling and late focal labeling) resembled a normal S₁₄. Since prolongation of G2 (using a Cdc25 ^string mutation) is not in itself sufficient to allow another round of S phase (Figure 2D), we conclude that Rux promotes rereplication, presumably by resetting the cell cycle to G1. The ability of Rux to reset G2 cells to G1 parallels the action of CKIs Sic1 and Rum1 in budding and fission yeast, respectively [7, 8], and emphasizes the similarities in cell cycle control between the yeasts and Drosophila. Perhaps this regulatory conservation will extend to mammals.

Rereplication Is Followed by an Aberrant Mitosis of Tetranemic Chromosomes

Following rereplication, mitotic figures were seen in homozygous Cyclin E mutant embryos between 150 and 180 min after heat shock (Figures 2I and 2K as compared to Figures 2H and 2J). Embryos laid by AR95; HS-Rux mothers were exposed to heat shock during G2₁₄, and individual embryos (n = 30) were squashed 3 hr after heat shock (when Cyclin E mutant embryos undergo mitosis). The squash technique did not permit genotypic classification by β-galactosidase staining. However, we detected a distinctive chromosomal phenotype at a frequency expected for Cyclin E mutant embryos (1/4): in six embryos, many of the chromosomes appeared especially bulky, and four chromatids were resolved in some of these (Figures 3I and 3J). Furthermore, with a frequency (24/30 = 4/5) comparable to that expected for the control genotype (3/4 Cyclin E ⁺), we found embryos with predominantly divalent chromosomes (Figure 3K) and rare groupings of more than two chromatids. We suggest that the exceptional embryos are Cyclin E mutants and that the distinctive chromosomes are tetranemic due to Rux-dependent endoreplication. The existence of these tetranemic chromosomes (also called diplochromosomes) suggests that the sisters produced in the first round of replication did not disjoin when Rux reset the cell cycle to G1 and that rereplication produced the observed four chromatid pairings.

Rux Expression in Cyclin E Mutant Embryos Results in the Generation of Tetranemic Chromosomes

(A–H) Mitotic chromosomes are visualized by staining for the mitosis-specific phospho-histone 3 (PH-3) epitope (blue) (Upstate Biotechnology) and FISH signals representing hybridization to the Y chromosome-specific AATAC (in red) and the X chromosome-specific 359 repeat (in green). (A and C) A total of 80% of prophase- and prometaphase-like figures exhibit one major focus of AATAC-associated Y chromosome staining, indicating that sister chromatids are not disjoined. (F and G) In metaphase, AATAC-staining foci appear as a cluster about half the time, suggesting that some disjunction has occurred. Since the AATAC locus is on the chromosome arm, these data do not indicate whether the centromeres (or just the arms) have disjoined. (B, E, and H) Between 55% and 85% of cells exhibited two prominent 359-associated X chromosome labeling, indicating that sister chromatids are primarily not disjoined. The scale bar represents 5 μm.

(I–K) Images of tetranemic and binemic chromosomes from Cyclin E mutant and Cyclin E+ embryos expressing Rux, respectively.

We used fluorescence in situ hybridization (FISH) to test the ploidy of chromosomes and to follow segregation behavior at mitosis. An X chromosome probe detected a tandem repeat of 359 bp [16], and a Y chromosome probe detected the simple sequence AATAC repeat (A.W.S. and P.H.O., unpublished data). The Y chromosome probe detected a single focus in each interphase nucleus of half the embryos (the males), while the X chromosome probe detected two foci in half the embryos (females), representing the maternal and paternal homologs (Figures 3B, 3E, and 3H). In Cyclin E ⁺ embryos (±Rux), the single Y chromosome focus and each of the X chromosome foci resolved into two upon anaphase disjunction (similar to Figure 4A). Thus, the two sisters of replication are normally paired at both of these loci until anaphase.

Segregation of Disjoined Tetranemic Chromosomes Is Primarily Normal but Inaccurate

Anaphase figures are revealed by staining for PH-3 (in blue), while the Y chromosome is labeled by FISH against the AATAC repeat sequence (in red).

(A) A wild-type anaphase with normal 1:1 segregation. Note that Rux-expressing Cyclin E ⁺ embryos also exhibit a similar pattern of segregation.

(B and C) Representative segregation patterns seen in Rux-expressing Cyclin E mutant embryos. (C) While most of the anaphase figures exhibit 2:2 segregation, (B) about 28% anaphase figures exhibit mostly 3:1 and, rarely, 4:0 segregation (data not shown). The scale bar represents 5 μm.

In Rux-expressing male Cyclin E mutant embryos, most anaphase figures exhibited four Y chromosome foci, consistent with a doubling of ploidy (Figures 4B and 4C). However, earlier in mitosis, the Y chromosome and each of the X chromosomes appeared as single large foci of staining (Figures 3A, 3C, and 3F) or as tight clusters of staining foci (Figures 3D and 3G). The observed single focus of staining leads us to conclude that Rux-dependent rereplication results in four sister chromatids that are paired (or within an unresolved distance from each other) until mitotic disjunction. Thus, like the squash results, the FISH analysis suggests that two rounds of replication occur prior to disjunction.

The abrupt disappearance of mitotic cyclins is generally taken as an indication of activation of the mitotic ubiquitin ligase APC and subsequent degradation. Since Rux-mediated resetting from G2 to G1 was accompanied by cyclin disappearance (Figure 1), we speculate that APC activation occurs during resetting but that it is insufficient to promote sister disjunction in this context. Consistent with this, a recent finding in Saccharomyces cerevisiae suggests that mitotic phosphorylation acts in conjunction with APC-targeted degradation to promote chromatid disjunction [17]. The widespread occurrence of polytene and diplochromosomes [18, 19] suggests that disjunction can be bypassed in diverse systems. The variety of cell cycle defects and polyploid phenotypes in cancer cell lines has suggested that production of diplochromosomes involves the complete bypass of mitosis [18]. Perhaps the dependence of disjunction on mitotic events is general.

Segregation of Tetranemic Chromatids Is Imprecise

Since the mitotic spindle normally segregates binemic chromosomes, its operation might be compromised when faced with tetranemic chromosomes. An accumulation of early mitotic figures suggests that the presence of tetranemic chromosomes slows progress to anaphase (Figure 2L). Additionally, while FISH hybridization failed to reveal intermediates in disjunction of the sisters in binemic chromosomes, tetranemic chromosomes frequently exhibited tightly clustered but partially resolved signals (Figures 3D and 3G), perhaps as a result of protracted disjunction. Note that partially disjoined chromosomes appear similar to partially congressed chromosomes and hence are likely to be classified as prophase (see Figure 2L). A possible explanation for slowed disjunction is suggested by the idea that the kinetochores on the central pair of chromatids in a tetranemic chromosome might not have the opportunity to interact with the spindle until disjunction of the outer pair. In this case, the initial steps of disjunction would expose naïve kinetochores that would trigger a checkpoint arrest so that progress is halting.

The bipolar mitotic spindle separates the two chromatids of a binemic chromosome to opposite poles with very high fidelity, but how does it distribute the four chromatids of a tetranemic chromosome? Using FISH, we scored the anaphase distributions of the four copies of the Y chromosome. Of the anaphases with four distinguishable chromatids, 72% exhibited two chromatids per half anaphase (Figure 4C). The remaining 28% of the anaphases exhibited aberrant (either 3:1 or, rarely, 4:0) segregation (Figure 4B). We conclude that the four sister chromatids are only imprecisely partitioned into the daughter cells, and we note that this will promote aneuploidy. Given the significant level of diplochromosomes in tumor cells, this inaccuracy could be an important source of the karyotypic instability of tumor cells and hence a catalyst of tumor progression.

Supplementary Material

Acknowledgments

We thank Tin Tin Su, Renny M. Feldman, Simon Prochnik, and Edan Foley for comments on the manuscript. We also thank Barbara Thomas for the "HS-Rux" flies. S.J.V. was supported by a predoctoral fellowship from the Howard Hughes Medical Institute, and P.H.O. acknowledges financial support from National Institutes of Health grant GM37193.

Footnotes

References

1. Su TT, Follette PJ, O'Farrell PH. Qualifying for the license to replicate. Cell. 1995;81:825–828. [PMC free article] [PubMed] [Google Scholar]

2. Follette PJ, Duronio RJ, O'Farrell PH. Fluctuations in cyclin E levels are required for multiple rounds of endocycle S phase in Drosophila. Curr Biol. 1998;8:235–238. [PMC free article] [PubMed] [Google Scholar]

3. Weiss A, Herzig A, Jacobs H, Lehner CF. Continuous Cyclin E expression inhibits progression through endoreduplication cycles in Drosophila. Curr Biol. 1998;8:239–242. [PubMed] [Google Scholar]

4. Avedisov SN, Krasnoselskaya I, Mortin M, Thomas BJ. Roughex mediates G(1) arrest through a physical association with cyclin A. Mol Cell Biol. 2000;20:8220–8229. [PMC free article] [PubMed] [Google Scholar]

5. Foley E, O'Farrell PH, Sprenger F. Rux is a cyclin-dependent kinase inhibitor (CKI) specific for mitotic cyclin-Cdk complexes. Curr Biol. 1999;9:1392–1402. [PMC free article] [PubMed] [Google Scholar]

6. Sauer K, Knoblich JA, Richardson H, Lehner CF. Distinct modes of cyclin E/cdc2c kinase regulation and S-phase control in mitotic and endoreduplication cycles of Drosophila embryogenesis. Genes Dev. 1995;9:1327–1339. [PubMed] [Google Scholar]

7. Correa-Bordes J, Nurse P. p25rum1 orders S phase and mitosis by acting as an inhibitor of the p34cdc2 mitotic kinase. Cell. 1995;83:1001–1009. [PubMed] [Google Scholar]

8. Sanchez-Diaz A, Gonzalez I, Arellano M, Moreno S. The Cdk inhibitors p25rum1 and p40SIC1 are functional homologues that play similar roles in the regulation of the cell cycle in fission and budding yeast. J Cell Sci. 1998;111:843–851. [PubMed] [Google Scholar]

9. Thomas BJ, Zavitz KH, Dong X, Lane ME, Weigmann K, Finley RL, Jr, Brent R, Lehner CF, Zipursky SL. roughex down-regulates G2 cyclins in G1. Genes Dev. 1997;11:1289–1298. [PubMed] [Google Scholar]

10. Sigrist SJ, Lehner CF. Drosophila fizzy-related down-regulates mitotic cyclins and is required for cell proliferation arrest and entry into endocycles. Cell. 1997;90:671–681. [PubMed] [Google Scholar]

11. Knoblich JA, Sauer K, Jones L, Richardson H, Saint R, Lehner CF. Cyclin E controls S phase progression and its down-regulation during Drosophila embryogenesis is required for the arrest of cell proliferation. Cell. 1994;77:107–120. [PubMed] [Google Scholar]

12. Richardson H, O'Keefe LV, Marty T, Saint R. Ectopic cyclin E expression induces premature entry into S phase and disrupts pattern formation in the Drosophila eye imaginal disc. Development. 1995;121:3371–3379. [PubMed] [Google Scholar]

13. Gonczy P, Thomas BJ, DiNardo S. roughex is a dose-dependent regulator of the second meiotic division during Drosophila spermatogenesis. Cell. 1994;77:1015–1025. [PubMed] [Google Scholar]

14. Foe VE. Mitotic domains reveal early commitment of cells in Drosophila embryos. Development. 1989;107:1–22. [PubMed] [Google Scholar]

15. Edgar BA, O'Farrell PH. The three postblastoderm cell cycles of Drosophila embryogenesis are regulated in G2 by string. Cell. 1990;62:469–480. [PMC free article] [PubMed] [Google Scholar]

16. Lohe AR, Hilliker AJ, Roberts PA. Mapping simple repeated DNA sequences in heterochromatin of Drosophila melanogaster. Genetics. 1993;134:1149–1174. [PMC free article] [PubMed] [Google Scholar]

17. Uhlmann F, Wernic D, Poupart MA, Koonin EV, Nasmyth K. Cleavage of cohesin by the CD clan protease separin triggers anaphase in yeast. Cell. 2000;103:375–386. [PubMed] [Google Scholar]

18. Levan A, Hauschka TS. Endomitotic reduplication mechanisms in ascites tumors of the mouse. J Natl Cancer Inst. 1953;14:1–43. [PubMed] [Google Scholar]

19. Zybina EV, Zybina TG. Polytene chromosomes in mammalian cells. Int Rev Cytol. 1996;165:53–119. [PubMed] [Google Scholar]

claypoolshough.blogspot.com

Source: https://europepmc.org/articles/pmc2754250/bin/nihms142591-supplement-supplement_1.pdf

Share this post