148
ited similar sizes to the in vivo equivalent. Furthermore, an upper limit to spindle
size was also shown to occur in Xenopus embryos: during the first four cell cycles,
the spindle size is relatively constant at an upper limit similar to that observed in
spindles formed in vitro using an early mitotic or meiotic extract (Wühr et al. 2008 ;
Good et al. 2013 ; Hazel et al. 2013 ). The small size of the mitotic spindle compared
to the cell size requires some special adaptation for proper DNA segregation. While
the mitotic spindle still is responsible for the initial separation of sister chromatids,
the majority of the DNA movement into the center of the future daughter cell is
executed by cell-spanning anaphase/telophase asters (see Sect. 4.3.2.3).
Starting at the fifth cell cycle, spindle length begins to scale with blastomere
length, exhibiting an approximately linear relationship. These observations showed
a transition between spindle length control mechanism, with very early and late
blastomeres exhibiting different control mechanisms. The observations that the
upper spindle length limit is observed in in vitro-reconstituted spindles (Wühr et al.
2008 ) as well as in embryos where cells are too large to be contacted by metaphase
spindle asters (Wühr et al. 2010 ; see above) suggest the presence of a length-
determining mechanism intrinsic to the spindle. The precise nature of this mecha-
nism remains unknown, but it is thought to depend on a balance between microtubule
nucleation dynamics and the function of microtubule-associated motors, as pro-
posed for the meiotic spindle (Burbank et al. 2007 ; Cai et al. 2009 ; Dumont and
Mitchinson 2009 ; Reber et al. 2013 ). Subsequently in smaller cells, spindle length
does become reduced coordinately with blastomere cell size.
As stated above, spindle size does scale with blastomere size during the later
blastomere cycles (Wühr et al. 2008 ). A simple model by which subcellular struc-
tures may scale to the decreasing size of embryonic blastomeres invokes a limiting-
component mechanism, developed through studies in the nematode C. elegans
embryo (Decker et al. 2011 ). Under this mechanistic model, the size of subcellular
structures, in this case centrosomes, scales according to cell volume due to the
inheritance during cell division of a limiting amount of structure precursor material
that necessarily decrements with each cell division.
The limiting-component mechanism appears to also apply to spindle formation
in Xenopus laevis, as shown by the analysis of spindles forming in cytoplasmic
compartments produced by microfluidic technology (Good et al. 2013 ; Hazel et al.
2013 ). In these compartments, spindle size correlates with cell volume. By deform-
ing the compartments to maintain cell volume while changing droplet diameter, the
authors showed that spindle length depends on a volume-sensing mechanism as
opposed to a boundary-sensing mechanism. Direct measurements shows a decrease
in free cytoplasmic tubulin in smaller blastomeres, consistent with spindle scaling
being dependent on the concentration of a limiting spindle factor precursor (Good
et al. 2013 ). Interestingly, encapsulated cytoplasm from early stages shows an upper
limit in spindle length in vivo as well as in within large droplets (Good et al. 2013 ;
Hazel et al. 2013 ). Together, these findings indicate that spindle size is regulated by
two separate yet interacting systems, one dependent on cytoplasmic composition
which imparts an upper limit to spindle length in large cells and a second which
A. Hasley et al.