|
FIGURE 1:. Live imaging of meiotic spindles in mei-1 homozygous viable mutants. (A) The schematic drawing shows a metaphase meiotic spindle with microtubule bundles extending from the poles to beyond the spindle midpoint. The microtubule bundles increase in density during the initial stage of spindle shortening. Time-lapse images of worms expressing GFP:tubulin and grown at 25°C indicate that the homozygous alleles P99L,A338S and P99L,P235S result in long, bipolar spindles which undergo normal anaphase shortening. Of the spindles shown, mei-1(P99L) and mei-1(P99L,A338S) are MII; all others are MI. The line above the schematic drawing of a metaphase spindle indicates the pole-to-pole spindle length reported in B and in Table 1. Bar = 4 μm. (B) Average spindle lengths determined from time-lapse images for metaphase I spindles (dark bars) and metaphase II spindles (light bars) for the indicated mei-1 genotypes. Error bars indicate the SE of the mean. n for each measurement is presented in Table 1.
|
|
FIGURE 2:. Expression of wild-type and mutant katanins in Xenopus tissue culture cells. Xenopus fibroblasts and A6 cells were cotransfected with GFP-tagged wild-type or mutant MEI-1 and with GST:MEI-2. Control cells were cotransfected with GFP-tagged human KATNA1 and with GST:KATNB1. Cells were fixed and immunostained with antibodies directed against α-tubulin, and images of high GFP expressers were captured. (A) Representative images of transfected fibroblasts. Expression of wild-type MEI-1 sometimes resulted in fragmented microtubules with both ends clearly visible (top left) but more frequently caused a reduction in the density of microtubules still emanating from the centrosome (top right) as did expression of the human homologue, KATNA1 (bottom right). Microtubule density was somewhat reduced by MEI-1(P99L,P235S). No reduction in microtubule density was seen with the homozygous viable allele, P99L;A338S, or with any of the maternal-effect lethal alleles L231F, P99L;G473D, or P225L. Bar = 25 μm. (B) Graph of relative microtubule density in transfected fibroblasts and A6 cells. For every frame captured, the average α-tubulin pixel intensity in transfected cells was expressed as a percentage of the average α-tubulin pixel intensity in adjacent nontransfected cells. Columns represent the average of the percentages for wild-type and mutant MEI-1 and KATNA1 in Xenopus fibroblasts (light shading) and A6 cells (dark shading). N = the number of transfected cells analyzed; bars = SEM. Relative GFP fluorescence is the average GFP pixel intensity of transfected cells for each experiment divided by the average GFP pixel intensity of wild-type MEI-1âexpressing cells. A number greater than one indicates that the GFP construct was expressed at a greater level than GFP-MEI-1(wild-type).
|
|
FIGURE 3:. ASPM-1 immunolocalization on meiotic spindles of mei-1 mutants. Worms were grown overnight at 25°C, and embryos were fixed and stained with antibodies directed against ASPM-1 and α-tubulin. Bipolar spindles containing a bright focus of ASPM-1 at each of two poles were present in wild-type, P99L, P99L;A338S, and P99L;P235S embryos. No ASPM-1 staining was observed on the meiotic spindles of the maternal-effect lethal mutants L231F, P99L;G473D, P99L;E308K, or P99L;R414K. Green alleles are homozygous viable. Red alleles are homozygous maternal-effect lethal. Each phenotype was observed in 100% of embryos of that genotype, and the number of spindles examined for each genotype is indicated. Bar = 5 μm.
|
|
FIGURE 4:. ASPM-1 is not present in meiotic spindles of klp-18(RNAi), mei-1(null) double mutants. Worms were grown overnight at 25°C, and embryos were fixed and stained with antibodies directed against ASPM-1 and α-tubulin. A bright ring of ASPM-1 is present at the single spindle pole in klp-18(RNAi) embryos, but no significant ASPM-1 staining occurs in mei-1(P99L;ct101) embryos or klp-18(RNAi), mei-1(P99L;ct101) double-mutant embryos. ct101 is a nonsense allele of mei-1 and behaves as a complete null (Mains et al., 1990). Each phenotype was observed in 100% of embryos of that genotype, and the number of spindles examined for each genotype is indicated. Bar = 5 μm.
|
|
FIGURE 5:. Spindle rotation and postrotation spindle shortening in mei-1(P99L;A338S). (A) Time-lapse images of embryos expressing GFP:tubulin indicate that meiotic spindle rotation and postrotation shortening occurred in 7/7 mei-1(P99L, A338S) embryos and in 6/6 wild-type embryos. Bar = 3 μm. (B) Wild-type and mei-1(P99L, P235S) worms were grown overnight at 25°C, and embryos were fixed and stained with antibodies directed against ASPM-1 and α-tubulin. Wild-type embryos (22/22) and mutant embryos with anaphase meiotic spindles (10/10) had bright foci of anti-ASPM-1 staining at the spindle poles. A small amount of anti-tubulin staining was also observed at the poles in embryos with the brightest staining. Bar = 3 μm.
|
|
FIGURE 6:. The N-terminal domain of MEI-1 is necessary and sufficient for binding to MBP-MEI-2. (A) Lanes 1â3 are amylose beadâbound fractions from binding reactions containing MBP-MEI-2, whereas lanes 4â6 are equivalent control reactions containing MBP. Binding reactions contained either the MEI-1 N-terminal domain (1â198) (lanes 1 and 4), MEI-1 missing a small conserved portion of the N-terminal domain (36â475) (lanes 2 and 5), or full-length MEI-1 (1â475) (lanes 3 and 6). (B) The N-terminal domain of human KATNAL1 is necessary and sufficient for binding human MEI-2 orthologues. Lanes 1â3 are amylose beadâbound fractions from binding reactions containing MBP-c15orf29, lanes 4â6 are bound fractions containing an MBP fusion to amino acids 412â655 of human KATNB1, and lanes 7â9 are bound fractions containing MBP only. Binding reactions contained either the KATNAL1 N-terminal domain (1â213) (lanes 1, 4, and 7), KATNAL1 missing a small conserved portion of the N-terminal domain (32â490) (lanes 2, 5, and 8), or full-length KATNAL1 (1â490) (lanes 3, 6, and 9).
|
|
FIGURE 7:. A complex of the N-terminal domain of KATNAL1 and c15orf29 binds microtubules in vitro. The N-terminal domain of KATNAL1 (1â213) or a MBP fusion to c15orf29 (0.6 μM of either), or a mixture of both proteins, was bound to increasing concentrations of polymerized tubulin. Microtubule-bound and unbound katanin were separated by centrifugation and quantified with anti-KATNAL1 immunoblots. Results of hyperbolic curve fits are shown. Kd = apparent equilibrium binding constant; Bmax = fraction bound at saturation, and R2 = quantification of how well the data fit the hyperbolic curve (with 1.0 being a perfect fit). Data for 1â213 and 1â213 + c15orf29 are combined data from two independent experiments.
|
|
FIGURE 8:. The N-terminal domain of MEI-1 binds microtubules in vivo. Xenopus fibroblast cells were cotransfected with a GFP fusion to the 198 N-terminal amino acids of MEI-1 and GST:MEI-2. Images of two transfected cells show tracks of GFP:MEI-1(1â198) that overlap or are interspersed with anti-tubulin staining. Arrows point to sites where individual microtubules are dotted with GFP:MEI-1(1â198). Bar = 20 μm.
|
