Kinetic characterization of the human retinoblastoma protein bipartite nuclear localization sequence (NLS) in vivo and in vitro. A comparison with the SV40 large T-antigen NLS.

The retinoblastoma (RB) tumor suppressor is a nuclear phosphoprotein important for cell growth control and able to bind specifically to viral oncoproteins such as the SV40 large tumor antigen (T-ag). Human RB possesses a bipartite nuclear localization sequence (NLS) consisting of two clusters of basic amino acids within amino acids 860-877, also present in mouse and Xenopus homologs, which resembles that of nucleoplasmin. The T-ag NLS represents a different type of NLS, consisting of only one stretch of basic amino acids. To compare the nuclear import kinetics conferred by the bipartite NLS of RB to those conferred by the T-ag NLS, we used beta-galactosidase fusion proteins containing the NLSs of either RB or T-ag. The RB NLS was able to target beta-galactosidase to the nucleus both in vivo (in microinjected cells of the HTC rat hepatoma line) and in vitro (in mechanically perforated HTC cells). Mutational substitution of the proximal basic residues of the NLS abolished nuclear targeting activity, confirming its bipartite character. Nuclear accumulation of the RB fusion protein was half-maximal within about 8 min in vivo, maximal levels being between 3-4-fold those in the cytoplasm, which was less than 50% of the maximal levels attained by the T-ag fusion protein, while the initial rate of nuclear import of the RB protein was also less than half that of T-ag. Nuclear import conferred by both NLSs in vitro was dependent on cytosol and ATP and inhibited by the nonhydrolyzable GTP analog GTPgammaS. Using an ELISA-based binding assay, we determined that the RB bipartite NLS had severely reduced affinity, compared with the T-ag NLS, for the high affinity heterodimeric NLS-binding protein complex importin 58/97, this difference presumably representing the basis of the reduced maximal nuclear accumulation and import rate in vivo. The results support the hypothesis that the affinity of NLS recognition by NLS-binding proteins is critical in determining the kinetics of nuclear protein import.