The nuclear envelope and the architecture of the nuclear periphery.

By definition, the eukaryotic cell is characterized by a compartmentation structure that divides the intracellular space into two different regions: (a) the nucleus, which contains the genome and the structures involved in transcription and processing of transcription products; and (b) the cytoplasm, which contains the translational apparatus, the cell organelles, the endomembranes, and a variety of other particles . This nucleocytoplasmic compartmentation is maintained by a specific eukaryotic membrane structure, the "nuclear envelope" (exceptions are some special situations, such as some forms ofnuclear division, certain sperm cells, and some pathological conditions ; see below) . The existence of a membranelike structure between nucleus and cytoplasm had been indicated in early light microscope studies. In his description of the cell nucleus in 1833, Brown (1) had already mentioned the possibility that it might be surrounded by a membranelike structure . After decades oflively discussion of the existence, real or artifact of preparation, of a distinct boundary layer between nucleus and cytoplasm, it was Flemming (2), who, in 1882 in a thorough review, summarized the accumulated evidence for "the existence of a special achromatinous lamella, that is a real-though in most types ofnuclei very thin-layer of substance, which . . . is not merely the expression of the region of contact between nuclear substance and cytoplasmic substance." Extending earlier observations of Hertwig (3) in nuclei of certain protozoa and of Soltwedel (4) in plant cell nuclei, Flemming (2) also clearly distinguished between the nuclear membrane proper as the "outer, achromatinous layer" and an inner "usually interrupted layer of peripheral chromatin" (the "nuclear cortex layer" sensu Hertwig and Soltwedel). Hertwig (3) also described certain fine punctate interruptions in cross sections of the achromationus nuclear membrane structures and discussed the possibility that these might represent pores, which allow exchange between nucleus and cytoplasm, an idea critically discussed by Flemming (2), who correctly pointed to the lack of evidence for the existence of such pores ("Poorn in der Kernmembran," 1882; in reference 2) . Several authors (2, 5) also noted the plasticity and viscosity of the nuclear membrane, properties that were