Chromosome motion and the spindle matrix.

Abstract

During mitosis, three kinds of mitotic movement can be distinguished.Collectively they are responsible for chromosome separation.(For our present purposes, we will ignore several less common motions that are poorly understood, such as those of prophase chromosome movement.)Chromosome motion usually commences during prometaphase when the forming spindle invades the nucleoplasm .The consequent prometaphase interaction between microtubules (MTs)' and chromosomes generates the first type of motion, irregular chromosome oscillations directed initially at either pole that lead to the metaphase plate configuration.Anaphase usually involves two distinct phases (36): anaphase A moves the chromosomes to the pole, and during anaphase B, the spindle elongates.In this paper, we consider mainly prometaphase and anaphase A movements.Once Inoue and collaborators (19) had shown that spindle fibers exist in living cells, cytologists could conceptualize the structural framework associated with mitotic movements.The advent ofglutaraldehyde fixation made it possible to establish that these spindle fibers contain MTs.In most cells, a proportion ofthe MTs terminate in the kinetochores.This structural relationship was immediately correlated with other evidence suggesting that the kinetochores are the site of chromosome attachment to the spindle, where poleward forces are exerted upon the chromosomes.The deduction that these inserted MTs had been nucleated or polymerized by the kinetochore has received widespread support from biochemical (summarized in reference 30) and cytological observations, but recently there has been a reappraisal ofthis viewpoint .Furthermore, the attachment ofMTs to the kinetochore suggests that they are mechanically functional in moving the kinetochore to the pole.Either or both of these conclusions are implicit in several models of mitosis .Although the issues are not yet decided, we believe that these two very influential conclusions are incorrect, and so current models ofthe spindle may be fundamentally flawed (30).Another serious shortcoming of all the well-known models is their inability to explain mitotic phenomena in all but the simplest terms; for example, the complex, erratic prometaphase activity of chromosomes has defied satisfactory explanation .We have chosen two aspects of the many conceptual com-'Abbreviations used in this paper.AP, away from pole; MTL, microtrabecular lattice ; MTs, microtubules.