Positional cloning of rmd-1. Plasmids containing the indicated genomic fragments were tested for rescue of the Psa phenotype of rmd-1(os21). The results are shown at the right. The numbers at the top indicate positions in the T05G5 cosmid. The asterisk represents the position of a mutation in os21 mutants.

RMD-1 and hRMD-1 localize to spindle microtubules and spindle poles during mitosis. (A) Wild-type embryos were fixed and stained for RMD-1 and microtubules (MTs) during pronuclear migration (a–c), at the pronuclear meeting (d–f), metaphase (g–i), anaphase (j–l), telophase (m–o), and at the two-cell stage (p–r). Images of rmd-1(RNAi) embryos at the prometaphase stained for RMD-1 and microtubules (s–u). In the merged images, DNA is blue, RMD-1 is red, and microtubules are green. (B) Western blotting with anti–RMD-1. The lysates from wild-type embryos (N2) and rmd-1(RNAi) embryos were subjected to Western blotting using an anti–RMD-1 antibody. The numbers on the left show the positions of molecular weight markers. (C) HeLa cells expressing HA–hRMD-1 were fixed and stained for HA and microtubules during metaphase. In the merged image, hRMD-1 is red, and microtubules are green. Bars, 5 μm.

Abnormal attachments of microtubules to kinetochores in rmd-1(RNAi) embryos. (A) HCP-1–GFP was distributed in two lines in both wild-type and rmd-1(RNAi) embryos during prometaphase before NEBD. The number at the left in each panel is the elapsed time after NEBD. (B) In prometaphase before NEBD, KLP-7/MCAK was observed as two lines on the condensed chromosome in both wild-type and rmd-1(RNAi) embryos. In the merged images, MCAK is red, and DNA is blue. (C) Embryos in anaphase were fixed and processed for FISH using a 5S rDNA probe. In both wild-type and rmd-1 embryos at anaphase, four discrete FISH signals were observed, indicating that sister chromatid cohesion was resolved. (D) Attachment of microtubules to kinetochores in wild-type and rmd-1(RNAi) embryos. Embryos were fixed and stained to label microtubules (MTs), MCAK, and DNA (with DAPI). In the merged images, microtubules are green, MCAK is red, DNA is blue, and microtubules that overlap with chromosomes are represented in light blue. In wild-type embryos at prometaphase after NEBD, chromosomes interacted with microtubules derived only from the closest spindle poles, but, in rmd-1(RNAi) embryos, microtubules (yellow arrows) crossed over chromosomes and kinetochores (stained with anti-MCAK). Broken lines in the magnified image show microtubules that crossed over chromosomes and kinetochores. At anaphase, kinetochores and chromosomes (light blue and yellow arrowheads) appeared to be pulled toward the opposite pole along incorrectly attached microtubules (light blue and yellow arrows). In addition, microtubules (white arrows) that interacted with the more distant kinetochores and chromosomes (white arrowheads) were observed. The regions indicated by dashed boxes are magnified. The asterisks represent chromosomes that are laterally located. Bars (A–C), 5 μm; (D) 2.5 μm.

Depletion of RMD-1 induces abnormal spindle organization. (A) Time-lapse images of GFP–β-tubulin in wild-type and rmd-1(RNAi) embryos. Time after NEBD is indicated at the left. (B) Embryos were fixed and stained for tubulin using anti–α-tubulin. At the one-cell stage, long asters were observed in wild-type embryos. In contrast, short astral microtubules were observed in the rmd-1(RNAi) embryos. (C) The distance between the spindle poles was tracked in wild-type (n = 9; white squares), rmd-1(RNAi) (n = 13; black squares), knl-1(RNAi) (n = 11; white circles), and knl-1(RNAi) + rmd-1(RNAi) (n = 8; crosses) embryos. The plots show the mean distances between the spindle poles versus time after NEBD. Bars, 5 μm.

Structure of the RMD proteins. (A) Schematic presentation of RMD proteins. Black boxes indicate coiled-coil domains. The total length of each protein is indicated at the right. hRMD-1/cgi-90 was identified by comparative proteomics (Lai et al., 2000). hRMD-2/BLOCK18 was identified by a computational screen for secreted proteins (Clark et al., 2003). hRMD-3/cerebral protein-10 (GenBank/EMBL/DDBJ accession no. AB000782) and hRMD-4 (GenBank/EMBL/DDBJ accession no. AK095462) were identified from a study of human transcriptome and functional genomics (Ota et al., 2004). Mouse homologues mRMD-1 (GenBank/EMBL/DDBJ accession no. AK010421), mRMD-2 (GenBank/EMBL/DDBJ accession no. BC024059), and mRMD-3 (GenBank/EMBL/DDBJ accession no. BC055754; Strausberg et al., 2002); Xenopus homologues xRMD-1 (GenBank/EMBL/DDBJ accession no. BC054253) and xRMD-2 (GenBank/EMBL/DDBJ accession no. BC090235); and zebrafish homologue zRMD-1 (GenBank/EMBL/DDBJ accession no. BX936455) are presented. (B) Comparison of C. elegans, human, mouse, Xenopus, and zebrafish RMD-1 proteins. Identical amino acids among RMD-1 homologues are shaded with black. Similar amino acids are shaded with gray. The asterisk shows the position of the mutation in rmd-1(os21). The regions marked with lines were used as antigens to prepare an anti–RMD-1 antibody. (C) Results of cosedimentation assay. Affinity-purified GST-fused RMD-1, -2, and -3, hRMD-1, -2, and -3, and GST as indicated at the top of the panels were incubated in the presence (+) or absence (−) of taxol-stabilized microtubules (MTs) and subjected to centrifugation. The pellets (P) and supernatants (S) were immunoblotted.

Defects of rmd-1(RNAi) embryos in chromosome segregation. (A) Time-lapse analyses of wild-type and rmd-1(RNAi) embryos expressing GFP-histone. In this rmd-1(RNAi) embryo, two maternal pronuclei (arrows), which failed to be extruded as polar bodies during meiosis, were detected, but they did not participate in the zygotic mitosis. In the rmd-1(RNAi) embryos, the chromosomes were not aligned on the metaphase plate in metaphase, and, in anaphase, the chromosomes were stretched along the axis of the spindle and failed to segregate. The position of the cleavage furrow in rmd-1(RNAi) embryos is indicated by arrowheads. The numbers on the left indicate the amount of time after NEBD. (B) Time-lapse images of zyg-9(b244) embryos expressing GFP-histone upshifted from 15 to 25°C at the stages indicated at the top of each column. Control embryos were observed at 16°C. The numbers on the left indicate the elapsed time after the start of observation (within 10 s after the upshift). The loss of ZYG-9 activity before metaphase caused lagging chromosomes, which are indicated by the arrows. (C) Percentages of embryos showing lagging chromosomes or a stretched mass of chromosomes in rmd-1(RNAi) (n = 19) and zyg-9(b244) embryos. Chromosomes that were uniformly stretched along the spindle were defined as a stretched mass of chromosomes. Chromosomes that were separated into two masses interconnected by chromatin bridges were scored as lagging chromosomes. The zyg-9(b244) embryos were upshifted from 15 to 25°C at the pronuclear meeting (n = 7), before chromosome alignment (n = 14), or during late metaphase (n = 28). Bars, 5 μm.

RMD-1 interacts with aurora B kinase. FLAG-tagged AIR-2 was coprecipitated with GST–RMD-1 but not with GST, indicating that RMD-1 interacts with AIR-2. Input shows 1/25 of cell lysate used in the experiments. Asterisks show degradation products.

RMD-1 is a novel MAP that regulates the dynamics of microtubules. (A) Examples of tracking EBP-2–GFP dots obtained in live analyses. Each arrow indicates the position of a single dot of EBP-2–GFP whose movement was traced. (B) Microtubule growth rates calculated by speed of EBP-2 movement before NEBD (n = 48 for wild type and n = 48 for rmd-1(RNAi)), after NEBD (n = 70 for wild type, n = 65 for rmd-1(RNAi), n = 21 for zyg-9(RNAi), and n = 21 for zyg-9 rmd-1(RNAi)), and during anaphase (n = 30 for wild type and n = 40 for rmd-1). In rmd-1(RNAi) embryos, the growth rate of astral microtubules was significantly lower than in wild type (P < 0.0001 before NEBD, P < 0.0001 after NEBD, and P = 0.00013 during anaphase). In the zyg-9(RNAi) embryos, the additional RNAi against rmd-1 had no significant additional effect on the microtubule growth rate after NEBD (P = 0.82). P-values were calculated by t test assuming unequal variances. Error bars represent SD. Bars, 5 μm.