An evolutionary perspective on the relationship between kinetochore size and CENP-E dependence for chromosome alignment

Abstract

Chromosome alignment during mitosis can occur as a consequence of bi-orientation or is assisted by the CENP-E (kinesin-7) motor at kinetochores. We previously found that Indian muntjac chromosomes with larger kinetochores bi-orient more efficiently and are biased to align in a CENP-E-independent manner, suggesting that CENP-E dependence for chromosome alignment negatively correlates with kinetochore size. Here, we used targeted phylogenetic profiling of CENP-E in monocentric (localized centromeres) and holocentric (centromeres spanning the entire chromosome length) clades to test this hypothesis at an evolutionary scale. We found that, despite being present in common ancestors, CENP-E was lost more frequently in taxa with holocentric chromosomes, such as Hemiptera and Nematoda. Functional experiments in two nematodes with holocentric chromosomes in which a CENP-E ortholog is absent (Caenorhabditis elegans) or present (Pristionchus pacificus) revealed that targeted expression of human CENP-E to C. elegans kinetochores partially rescued chromosome alignment defects associated with attenuated polar-ejection forces, whereas CENP-E inactivation in P. pacificus had no detrimental effects on mitosis and viability. These data showcase the dispensability of CENP-E for mitotic chromosome alignment in species with larger kinetochores.

Keywords: CENP-E; Chromosome; Holocentric; Kinesin; Kinetochore; Mitosis.

Fig. 1.

Phylogenetic profile of CENP-E across holocentric and monocentric taxa. (A–D) CENP-E conservation in (A) the phylum Nematoda, (B) the phylum Vertebrata, (C) the insect order Hemiptera and (D) the insect order Diptera. Lineages with an inferred CENP-E loss are highlighted with a colored box. Holocentric lineages are indicated with ‘H’ and monocentric lineages with ‘M’, as well as with a graphic depiction of holocentric and monocentric chromosomes. Species silhouettes are from https://www.phylopic.org/, licensed under a CC0 1.0 Universal Public Domain Dedication: Nematoda, Trichinella sp.; Vertebrata, Danio rerio; Hemiptera, Nilaparvata lugens; Diptera, Bradysia coprophila. BUSCO scores provide a quantitative assessment of genome assembly and annotation completeness with single-copy orthologs.

Fig. 2.

Expression of human CENP-E in C. elegans. (A) Schematic representation of the transgenes used in this study: KBP-3 (control) and hCENP-E::KBP-3 (hCENP-E) were fused to mKate2 in order to visualize kinetochores. (B) Selected still images from time-lapse sequences of the first embryonic division showing the kinetochore localization of the two constructs (mKate2, inverted grayscale). Scale bar: 5 µm. Timestamps are shown in seconds (s). (C) Normalized fluorescence intensity of mKate2 fusions at kinetochores. (D) Normalized fluorescence intensity of the cytoplasmic signal. Bars show mean±s.d. ***P<0.001; ****P<0.0001; two-tailed unpaired t-test. Control: n=10 embryos; hCENP-E: n=9 embryos.

Fig. 3.

Expression of hCENP-E does not compromise mitotic progression and rescues chromosome alignment defects due to attenuated polar-ejection forces in C. elegans. (A) Representative live-cell imaging examples of the first embryonic division of control, hCENP-E, klp-19 RNAi and klp-19 RNAi+hCENP-E embryos stably expressing GFP::H2B to visualize chromosomes and GFP::γ-tubulin to label the spindle poles. Red arrowheads indicate lagging chromosomes. Scale bar: 5 µm. (B) Schematics of the two quantitative measurements extracted from the live-cell recordings. (C,D) Spindle pole separation (C) and chromosome alignment kinetics (D) in one-cell embryos, showing the same unperturbed effect in the absence (control) and presence of hCENP-E. (E,F) klp-19 depletion induced a slight premature pole separation at ∼120 s after nuclear envelope breakdown (NEBD; time zero on the plots) (E) and chromosome scattering (F). Repeated measures two-way ANOVA showed no differences in spindle length in embryos expressing hCENP-E [F(20, 400)=1.29, P=0.1775)], but a significant change in chromosome span [F(20, 400)=2.92, P<0.0001)]. Post hoc significant comparisons between klp-19 RNAi and klp-19 RNAi+hCENP-E are highlighted in the graph: 70 s, P=0.0012; 80 s, P=0.0001; 90s, P=0.0018. Distances were measured in images acquired every 10 s, averaged for the total number of embryos, and plotted against time. Error bars represent the 95% c.i. Black dashed lines represent NEBD. Anaphase onset (AO) for each condition is shown in a dashed line of the corresponding color. N=2 independent experiments (control, n=13 embryos; control+klp-19 RNAi, n=11 embryos; hCENP-E, n=12 embryos; hCENP-E+klp-19 RNAi, n=11 embryos).

Fig. 4.

Expression of hCENP-E rescues brood size to control levels in C. elegans with defective polar-ejection forces. (A) Effect of hCENP-E expression on embryo viability and brood size, with and without klp-19 RNAi. The mean of n mothers ±95% c.i. is shown. ns, not significant; **P<0.01; unpaired two-tailed t-test. (B) Frequency of one-cell embryos with anaphases with lagging DNA that persisted for two consecutive frames after anaphase onset; total numbers of analyzed embryos are indicated.