Nuclear transport in vitro.

In this paper, progress towards the goal of understanding communication between the nucleus and cytoplasm using an in vitro system is reviewed. To probe the mechanism of nuclear targeting, we developed an in vitro transport system and have begun to dissect the highly selective process of nuclear transport. The basic parameters of transport were defined using an easily isolated nuclear protein, nucleoplasmin. To study the interaction of nuclear targeting signals with the pore, an artificial nuclear transport substrate was constructed, which consists of human serum albumin coupled to the signal sequence of the SV40 T-antigen. A similar peptide-protein conjugate was made using a mutant signal sequence. These conjugates were fluorescently labeled and/or tagged with gold and tested for transport in the in vitro system. High levels of nuclear transport of the wild-type signal sequence-containing protein were observed, while no transport of the mutant signal sequence-containing protein was seen. Thus, the in vitro system correctly recognizes the single amino acid change between the wild-type and mutant signal sequences. We found that the observed nuclear transport was completely dependent on the presence of ATP. Using the in vitro system we identified a specific inhibitor of nuclear transport, the lectin wheat germ agglutinin (WGA), which we find binds directly to the nuclear pore. Probing blots of nuclear proteins with 125I-WGA identified a family of nuclear pore glycoproteins, including one major glycoprotein of 62K (K = 10(3)Mr) molecular weight. With the inhibitor and the in vitro assay, it has been possible to experimentally separate nuclear transport into two steps: (1) a step in which the signal sequence-bearing protein binds to the pore, followed by (2) a step in which the protein translocates through the pore. It is this second step which is the ATP-dependent step of transport, since pore binding but not translocation was seen to occur in the absence of ATP.