INTRODUCTION Despite the enormous current interest in gene expression and its role in biology, very little is known about the structural and functional environment within which gene expression occurs. The molecular composition and spatial organization of even the most rudimentary extra-chromosomal nuclear components remain unclear. Indeed, the existence of any internal nuclear structural features other than the nucleolus or chromatin (e.g. the so-called nuclear skeleton or matrix) remains controversial (for reviews, see Cook, 1988; Fisher, 1989; Jackson, 1991). In this uncertain situation, the few universally accepted com- ponents of the nucleus (e.g. the nuclear envelope and the nucleolus) have become strong foci of study, in part because they may represent examples of more global nuclear organiz- ing principles. The best characterized of these components is the fibrous, proteinaceous network on the inner (nucleoplas- mic) face of the nuclear envelope, the nuclear lamina. The major proteins of the nuclear lamina, designated nuclear lamins (Gerace and Blobel, 1980), have undergone extensive molecular characterization in the last decade. Lamins are now known to be specialized types of intermediate filaments that undergo differential patterns of expression in the development of vertebrates. The molecular basis for lamin targeting to and assembly into the nuclear lamina is now understood, as well as the mechanism by which lamins depolymerize during mitosis. Indeed, lamins themselves may be substrates of a major regulator of the cell cycle, the p34 cdc2 kinase (for reviews, see Burke, 1990a; McKeon, 1991; Dessev, 1992; Nigg, 1992a,b; Paddy et al., 1992; Moir and Goldman, 1993; Georgatos et al., 1994). Functionally, morphological studies and biochemical evidence have long suggested that lamins interact with inter- phase chromatin in the nuclear periphery (e.g. see Franke et al., 1981; Benavente and Krohne, 1987; Hiraoka et al., 1989; Nigg, 1989; Burke, 1990b; Glass and Gerace, 1990; Laskey and Leno, 1990; Yuan et al., 1991; Ulitzur et al., 1992; Glass et al., 1993). Recently, direct, molecularly based structural analyses have revealed structures consistent with this role (Paddy et al., 1990; Belmont et al., 1993). Though a functional interaction between lamins and chromatin has not been demon- strated unequivocally in vivo, the indirect evidence is suffi- 591 Journal of Cell Science 109, 591-607 (1996) Printed in Great Britain © The Company of Biologists Limited 1996 JCS3281 Time-resolved, two-component, three-dimensional fluor- escence light microscopy imaging in living Drosophila early embryos is used to demonstrate that a large fraction of the nuclear envelope lamins remain localized to a rim in the nuclear periphery until well into metaphase. The process of lamin delocalization and dispersal, typical of ‘open’ forms of mitosis, does not begin until about the time the final, metaphase geometry of the mitotic spindle is attained. Lamin dispersal is completed about the time that the chromosomal movements of anaphase begin. This pattern of nuclear lamina breakdown appears to be intermediate between traditional designations of ‘open’ and ‘closed’ mitoses. These results thus clarify earlier observations of lamins in mitosis in fixed Drosophila early embryos, clearly showing that the observed lamin localization does not result from a structurally defined ‘spindle envelope’ that persists throughout mitosis. During this extended time interval of lamin localization in the nuclear periphery, the lamina undergoes an extensive series of structural rearrangements that are closely coupled to, and likely driven by, the movements of the centrosomes and microtubules that produce the mitotic spindle. Furthermore, throughout this time the nuclear envelope structure is permeable to large macromolecules, which are excluded in interphase. While the functional sig- nificance of these structural dynamics is not yet clear, it is consistent with a functional role for the lamina in mitotic spindle formation. Key words: Nuclear lamin, Mitosis, Mitotic spindle, Dynamics, Imaging, In vivo, Drosophila SUMMARY Time-resolved, in vivo studies of mitotic spindle formation and nuclear lamina breakdown in Drosophila early embryos Michael R. Paddy 1,2, *, Harald Saumweber 3 , David A. Agard 2 and John W. Sedat 2 1 Center for Structural Biology and Department of Anatomy and Cell Biology, University of Florida, Gainesville, FL 32610-0235, USA 2 Structural Biology Unit, Howard Hughes Medical Institute and Department of Biochemistry and Biophysics, University of California at San Francisco, San Francisco, CA 94143-0554, USA 3 Biologische Institute Abt. Zytogenetik, Humboldt Universitaet zu Berlin, D-10115 Berlin, Germany *Author for correspondence (e-mail: paddy@ufom0.health.ufl.edu)