Studying essential genes has revealed the core machinery that powers mitosis, but much remains unclear. For example, the microtubule motor dynein is required for many processes including mitosis, but little is known about the requirements for its accessory chains or how it is regulated to perform multiple tasks. We therefore hypothesize that nonessential and conserved genes also contribute to cell division. We have identified twenty genes that, when reduced in function, restore viability to conditional dynein heavy chain mutants (O'Rourke et al. PLoS Genetics 2007). Sixteen of these suppressors are not essential and 15 have human orthologs. We recorded the first mitosis in embryos expressing fluorescent protein fusions to mark mitotic spindles while reducing gene function. We find distinct requirements for eight. Two suppressors, dylt-1
, encode dynein light chains that inhibit DHC-1 because depleting either gene restores viability to dhc-1ts
null mutants are fully viable but exhibit spindle rocking defects in P0, while spindle function otherwise appears normal. By contrast, most dyrb-1
null mutants fail to hatch, exhibit enlarged P0 centrosomes, transverse P0 spindles, lagging chromosomes, and an absence of spindle rocking; in contrast to dhc-1
mutants, robust mitotic spindles still form and chromosomes eventually segregate to the daughter cells. We conclude that specific dynein subunits contribute both positively and negatively to heavy chain function and the chains exhibit distinct and only partially overlapping requirements. The remaining suppressors have not been previously implicated in dynein function, but we have discovered related phenotypes for about half of them. We have found thus far that F10E7.8-depleted embryos exhibit transverse P0 spindles and lagging chromosomes while some embryos lack maternal pronuclei. Reducing NPP-22/nucleoporin function causes lagging chromosomes. Knockdown of either F45H11.3 or T23D8.3 causes metaphase spindle positioning defects. Embryos lacking EFA-6/GEF exhibit P0 centrosome hyper-separation during pronuclear migration, anterior spindle pole flattening, and reduced spindle rocking. The centrosome hyper-separation is consistent with EFA-6 negatively regulating dynein, as centrosome separation fails in dhc-1
mutants. Thus many non-essential genes do indeed have clear cell division requirements.