Table 1.
Embryonic lethality of par mutants depleted of MATH-33 by RNAi.
Table 2.
Polarity phenotypes of math-33; par loss-of-function embryos.
Figure 1.
Paralogs math-33 and usp-47 have synthetic polarity phenotypes.
(A) Bars indicate percent embryo lethality. >300 embryos were scored for each genotype in two trials. Error bars indicate standard error of the mean; asterisks indicate significant difference compared to controls, p<0.01, determined by Student's t-test. K09A9.4 and T24B8.7 are negative control DUBs. (B) Un-rooted phylogenetic tree of 25 UCH domains in C. elegans and two human homologs of math-33 and usp-47. We did not perform RNAi on the three genes indicated by ‘. Human homologs USP7 and USP47 are included to highlight that there is evolutionary conservation of the DUBs between species. (C) DIC time-lapse of one-cell to four-cell embryos of the indicated genotypes at 22°C. Embryos are oriented with the anterior at the left in this and subsequent figures. The spindle orientation of P1 as it divides is usually longitudinal to the embryo axis, but is transverse in polarity-defective mutants such as par-2(lw32). The scale bar represents 10 µm.
Figure 2.
Polarity defects in math-33(tm3561);usp-47(RNAi) embryos.
(A) Confocal micrographs of embryos immunostained for PAR-3 (green) and PAR-2 (red). Compared to controls, PAR-3 occupies a larger region of the cortex and PAR-2 a smaller region in the math-33(tm3561); usp-47(RNAi) embryo. (B) Averages of PAR-3 and PAR-2 domain sizes in immunostained embryos as a percentage of embryo length. n indicates the number of anterior and posterior domains examined respectively. Asterisks mark statistical significance compared to wild type, p<0.01, by Student's t-test. (C) Time-lapse images displaying the dynamic localization of PAR-2::GFP in the indicated genotypes. The embryo genotype of the control row (top) showing wild-type behavior is par-2::gfp; math-33(tm3561)/nT1. math-33(tm3561); usp-47(RNAi) embryos weakly recruit PAR-2 and could be grouped into two classes: those that maintain a PAR-2 domain throughout the cell cycle (class I), and those in which the weak PAR-2 domain is not maintained (class II). Scale bars represent 5 µm.
Figure 3.
PAR-2 function is not required to mediate loss of polarity in math-33(RNAi) usp-47(RNAi).
(A) Embryonic lethality of the indicated genotypes. Error bars indicate the standard error of the mean. The experimental class marked with an asterisk is significantly different from all three controls, p<0.05, by Student's t-test. (B) LGL-1::GFP time-lapse images of embryos of the indicated genotypes. Upon knockdown of the DUBs MATH-33 and USP-47, the LGL-1::GFP domain size is smaller than in control embryos and is not maintained, similar to previous results with PAR-2::GFP in Figure 2C, and transverse spindles occur in P1. Scale bar represents 5 µm.
Figure 4.
The posterior to anterior myosin clearing is defective in math-33(tm3561); usp-47(RNAi).
(A) Time-lapse confocal images of embryos expressing NMY-2::GFP. The maximum projections of sections through the cortex of the embryo show the localization of NMY-2::GFP foci at three stages. Kymographs illustrate the pattern of myosin clearing and flow over time. About halfway through the kymographs (from top to bottom), large myosin foci transition into smaller myosin puncta. (B) The extent of myosin absence from the posterior (clearing) was measured as a proportion of the embryo length and the maximum level of clearing was recorded; each data point represents one embryo. (C) Confocal stacks of embryos immunostained for NMY-2 (green) and tubulin (red). Arrows indicate the location of the centrosome, red dotted circles mark the position of pronuclei, and the extent of myosin clearing is indicated by arrowheads. Scale bars represent 5 µm.
Figure 5.
Deubiquitylases regulate the position of the centrosome in early one-cell embryos.
(A) Micrographs showing tubulin::GFP in embryos at the time the centrosome was first detected. Centrosomes are marked with a white arrowhead and the position of the paternal pronucleus is indicated by red dotted circles. The DIC micrograph column shows representative embryos of the same genotypes at the time of pseudocleavage (an indirect indicator of the extent of myosin clearing). The white arrows indicate the location of pseudocleavage furrows. (B) Distance in micrometers of the centrosome from the embryo cortex at the time of earliest detection for the indicated genotypes. Each data point represents a single embryo. Black dots are embryos that had normal or weak pseudocleavage, red dots are embryos that displayed no pseudocleavage. Blue bars mark the mean distance. The asterisk indicates a significant difference from WT controls, p<0.01, by Student's t-test. (C) The number of embryos lacking pseudocleavage increases as centrosome distance from the cortex increases. Embryos were pooled from five RNAi experiment replicates. Differences were statistically significant according to a Student's t-test, p<0.01. (D) The distance of the centrosome from the cortex measured in the indicated genetic backgrounds. p<0.01 for column 4 compared to WT distances from multiple experiments, but p = 0.156 for column 4 compared to math-33(tm3561) in multiple experiments. (E) A single experiment examining absence of pseudocleavage resulting from depletion of DUBs. The absence of pseudocleavage phenotype is completely suppressed by dhc-1(RNAi), p<0.05. Scale bar represents 5 µm.
Figure 6.
MATH-33 and USP-47 are not enriched at cortex or centrosome.
The left panels are representative images of embryos showing the distribution of GFP::MATH-33 and GFP::USP-47 in one-cell embryos. Both proteins are present in the cytoplasm, but MATH-33 is enriched in nuclei. The right panels are images of embryos showing immunostaining of endogenous MATH-33(red) in the cytoplasm and the nucleus; USP-47(green) primarily in the cytoplasm. Scale bar represents 5 µm.
Figure 7.
Mutation of rpn-10 suppresses lethality and polarity defects in math-33(tm3561); usp-47(RNAi) embryos.
(A) Embryonic lethality of math-33(tm3561); usp-47(RNAi) is reduced by either of two rpn-10 mutations. Standard error of the mean is indicated by the error bars. n>350 embryos for N2 controls and n>600 for the other genotypes. Asterisks indicate significance compared to math-33(tm3561) single mutants, p<0.01, Student's t-test (B) Data showing the suppression of phenotypic defects in early math-33(tm3561);usp-47(RNAi) embryos by rpn-10 mutations. (C) Distance in micrometers of the centrosome from the embryo cortex when it is first detectable. rpn-10 mutation suppresses the absence-of-pseudocleavage phenotype and the mislocalization of the centrosome compared to math-33(tm3561); usp-47(RNAi) controls. None of the 18 centrosomes observed in column 2 were detached from the paternal pronucleus, indicating that the detachment phenotype was also completely suppressed. Results in column 4 were significantly different in a Student's t-test p<0.01 compared to column 1 controls. Data from columns 1, 3, and 4 are also displayed in Figure 5C.
Figure 8.
usp-46 acts redundantly with math-33 and usp-47.
(A) Two-cell embryos at interphase (top) illustrate unequal vs. equal first divisions and at P1 mitosis (bottom) show spindle orientations. Maternal genotypes are indicated. Scale bars for each genotype represent 5 µm. (B) Embryonic lethality measured after depleting math-33 in usp-46 and usp-47 mutants. Bars marked with an asterisk are significantly different from the RNAi controls in a t-test, p<0.01. Scale bar represents 5 µm.
Table 3.
Effect of RNAi/mutant combinations of math-33, usp-46, and usp-47 on P1 spindle orientation.