QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Doronkin, S. Articles by Beckendorf, S. K. Search for Related Content PubMed PubMed Citation Articles by Doronkin, S. Articles by Beckendorf, S. K.

CSN5/Jab1 mutations affect axis formation in the Drosophila oocyte by activating a meiotic checkpoint

Department of Molecular and Cell Biology, 401 Barker Hall, University of California, Berkeley, CA 94720, USA

View larger version (90K): [in a new window]   Fig. 1. CSN5 expression in wild-type and CSN5 mutants. (A,B) In wild-type ovaries, CSN5 expression is detected in nurse cells beginning in the germarium. (C,D) In CSN5L4032-mutant germline clones (GLC), expression of CSN5 RNA is strongly reduced. Northern blot analysis of ovarian extracts show a similar reduction in CSN5 GLC (data not shown). (E) Expression in early wild-type embryos indicates a strong maternal contribution. (F) In CSN5 heterozygotes that lack maternal CSN5 RNA, zygotic expression is detectable in an anterior stripe during cellular blastoderm (arrow). (G,H) During gastrulation CSN5 RNA is most strongly expressed in the ventral furrow, the cephalic furrow, and both the anterior and posterior midgut invaginations. This expression pattern is absent in maternal and zygotic minus embryos (data not shown). All figures show anterior towards the left. (I) Western blot of wild-type and CSN5-mutant ovarian extracts. A single 37-38 kDa band seen in wild-type ovaries was strongly reduced in CSN5L4032 GLC ovaries. Its size is consistent with the predicted size for Drosophila CSN5 (37 kDa) and no other specifically reduced bands were detected. Heteroallelic combinations of CSN5L4032 with excision derivatives show a gradation of CSN5 protein that correlates with their allelic strength determined from viability, eggshell phenotypes and northern blots.

View larger version (56K): [in a new window]   Fig. 2. Mutations in CSN5 cause eggshell defects. (A) A wild-type eggshell. Note the length of the dorsal appendages (DA) and the separation at their bases. (B-F) Eggshells derived from females carrying CSN5-mutant, germline clones. (B,C) Fused or absent DA in weakly or strongly ventralized eggshells. (D) A partially dorsalized eggshell. Bases of the appendages are more widely separated than in wild type. (E) Short but normally spaced DA. (F) An unusual eggshell with duplicated DA. In addition to these eggshell defects, some eggs had an unusually weak eggshell. These eggs were sometimes destroyed during attempts to move them. This phenotype may be related to observations made during dechorionation or fixation that many eggs from CSN5 GLC mothers fall into pieces.

View larger version (77K): [in a new window]   Fig. 3. CSN5 is required for dorsoventral patterning of the embryo. (A) In wild-type embryos dpp RNA localizes to the dorsal side of the egg as well as the two poles. (B) In eggs from CSN5 GLC mothers, the dpp-expressing domain is often reduced, indicating that the embryo is ventralized. (C) In rare, dorsalized embryos, dpp RNA extends around the DV circumference. (D) In wild-type embryos, twi RNA is expressed in the ventral mesoderm and at both poles. (E,F) twi RNA in embryos derived from CSN5 germline mothers. (E) In ventralized embryos, twi RNA is more broadly expressed ventrally and laterally. (F) In rare, strongly dorsalized embryos, twi RNA fails to accumulate in the mutant egg. (G) In wild-type embryos, rho RNA is expressed in ventrolateral stripes and a dorsal midline stripe. (H) In weakly ventralized eggs derived from CSN5 GLC mothers, the lateral stripe of rho RNA was moved dorsally and the dorsal stripe was reduced. (I) In dorsalized embryos rho RNA accumulates in stripes that are probably derived from the dorsal stripe seen in wild-type embryos. Zygotic expression patterns could not be analyzed in the substantial fraction of embryos that were unfertilized or extremely fragile (see Table 1).

View larger version (55K): [in a new window]   Fig. 4. Effects of CSN5 mutations on the AP axis. (A-H) CSN5 is required for proper localization of bcd and osk RNAs. (A,B) In CSN5-mutant oocytes, bcd RNA is often present diffusely throughout the oocyte. In addition, the amount of bcd RNA at the anterior end of the oocyte may be reduced and there may be central bcd RNA concentrations. (E,F) In embryos the effects of CSN5 mutations on bcd RNA localization are less extreme. Some embryos do show dorsally displaced bcd RNA. (C,D) In CSN5-mutant oocytes, little osk RNA is localized at the posterior pole. Instead it is often present throughout the oocyte cytoplasm or in central concentrations. (G,H) In some mutant embryos, osk RNA is posteriorly localized, but is strongly reduced and sometimes shifted dorsally. (I-K) CSN5 may affect localization and number of pole cells. In agreement with the osk RNA results, anti-Vasa staining often shows a reduced number of pole cells (J). In some cases the pole cells were not tightly clustered (K).

View larger version (143K): [in a new window]   Fig. 5. Germline CSN5 mutations affect follicle cell patterning. (A-D) Reduction of CSN5 function disrupts the specification of the terminal follicle cells. In wild-type egg chambers, the PZ6356 enhancer trap is expressed in border cell and oocyte nuclei (A), while the slbo enhancer trap is expressed only in border cells (C). In CSN5 mutant egg chambers, both enhancer traps are also expressed in posterior follicle cells (arrows, B,D), suggesting that they have taken on anterior fate. (E-H) CSN5 GLCs affect patterning of the dorsoanterior follicle cells. In CSN5-mutant egg chambers, expression of the kekkon enhancer trap is often reduced (F,G) or in rare cases undetectable (not shown). In a small number of egg chambers, kekkon is expressed in a larger than normal patch of dorsal follicle cells (H).

View larger version (78K): [in a new window]   Fig. 6. CSN5 is necessary for Grk protein expression, not for grk transcription. (A-C) In situ localization of grk RNA. At stage 10, grk transcript accumulates at the dorsoanterior corner of wild-type and most CSN5-mutant oocytes (A,B). Occasionally, grk RNA is mislocalized, probably because of mislocalization of the oocyte nucleus (C). (D-F) Grk antibody immunostaining. In CSN5 GLC egg chambers, expression of Grk is reduced or more diffusely distributed than in wild type. (G) Northern blot analysis of grk mRNA levels. In CSN5 GLC ovaries, the level of grk mRNA (extracted from an equal number of ovaries) is similar to wild type and significantly higher than in grk null-mutant ovaries. (H) Western blot analysis of Grk protein level. The Grk antibody recognizes a 46-47 kDa band that is reduced in ovaries homozygous for a hypomorphic grk allele, grkHK36, and strongly reduced in ovaries from CSN5 germline clones.

View larger version (70K): [in a new window]   Fig. 7. CSN5 mutations activate a mei-41-dependent meiotic checkpoint. (A) In wild type, Grk protein is detectable in region 2a of the germarium and is restricted to the oocyte from region 3 onwards. (B) In CSN5ex21/CSN5L4032 germaria, Grk is undetectable. (C) Loss of mei-41 restores Grk expression in CSN5ex21/CSN5L4032 heterozygotes. (D,E) CSN5 mutations affect the mobility of Vasa protein. (D) An anti-Vasa western blot of ovarian extracts shows retarded Vasa migration in CSN5L4032 germline clones. (E) An anti-Vasa western blot of ovarian extracts from several CSN5 allelic combinations. In wild type, Vasa migrates as a single 72 kDa band. Vasa migrates as two bands in CSN5ex13/CSN5L4032 and CSN5ex21/CSN5L4032. Notice, that in the stronger combination (CSN5ex21/CSN5L4032) more Vasa protein migrates slowly than in the weaker combination (CSN5ex13/CSN5L4032). Furthermore, in CSN5ex13/CSN5L4032, Vasa is fully restored to normal mobility by removal of one dose of mei-41, while CSN5ex21/CSN5L4032 requires homozygous mei-41 for normal Vasa mobility. Removing one dose of mei-W68 is sufficient to restore Vasa mobility in both hypomorphic, CSN5-mutant combinations. A second mei-W68 allele (mei-W681) had a weaker effect. As a heterozygote with the CSN5-mutant combinations, it gave only a partial rescue of Vasa mobility, although the mei-W681/mei-W68k05603 combination resulted in full rescue of Vasa mobility (data not shown).

View larger version (15K): [in a new window]   Fig. 8. Model of CSN5/CSN function during oogenesis (modified from Ghabrial and Schupbach, 1999). In region 2a of the germarium, meiotic recombination begins with the formation of DSBs under the control of mei-W68 (McKim and Hayashi-Hagihara, 1998). These breaks are repaired by proteins of the recombination repair pathway, including the spindle-class genes. The progress of DSB repair is monitored by a mei-41-dependent checkpoint. If DNA damage persists, as it does in CSN5 or spindle-class mutants, the checkpoint is activated and Vasa is modified to a form that prevents efficient translation of grk message. The ensuing underproduction of Grk protein leads to axial patterning defects (Ghabrial and Schupbach, 1999). We propose that the role of CSN5 is either to limit the production of DSBs, perhaps by promoting Mei-W68 turnover, or to stabilize one of the repair proteins, thereby promoting repair and bypass of the mei-41-dependent checkpoint.