Genetic and physical interactions between Srp1p and nuclear pore complex proteins Nup1p and Nup2p

Previously identified protein of uncertain function is karyopherin alpha and together with karyopherin beta docks import substrate at nuclear pore complexes

Previously, we had purified a cytosolic protein complex, termed karyopherin, that functions in docking import substrate at the nuclear envelope in digitonin-permeabilized cells and also had molecularly cloned and sequenced its 97-kDa beta subunit. We now report that the karyopherin alpha subunit is the previously identified protein NPI-1/SRP-1 of hitherto uncertain function. Using purified recombinant karyopherin alpha or beta subunit, we showed that neither karyopherin alpha nor karyopherin beta alone was sufficient for docking of import substrate at the nuclear envelope. Docking occurred only when both subunits were present. Moreover, docking of import substrate by the two recombinant karyopherin subunits was productive, as it led to nuclear internalization of the docked substrate in the presence of additional, previously characterized cytosolic factors. In a binding assay using immobilized karyopherin alpha and beta subunits and import substrate as a ligand, we found that only karyopherin alpha bound ligand. We suggest that karyopherin beta functions as an adaptor that binds both to karyopherin alpha and to any of a large number of docking sites that are represented by a repetitive peptide motif containing nucleoporins on both the cytoplasmic and nucleoplasmic side of the nuclear pore complex (NPC), bidirectionally ferrying a complex of karyopherin alpha-substrate across the NPC.

NPI-1, the human homolog of SRP-1, interacts with influenza virus nucleoprotein

We used the yeast interactive trap system to identify a cellular protein which interacts with the nucleoprotein of influenza A viruses. This protein, nucleoprotein interactor 1 (NPI-1) is the human homolog of the yeast protein SRP1. SRP1 was previously identified as a suppressor of temperature-sensitive RNA polymerase I mutations (R. Yano, M. Oakes, M. Yamaghishi, J. Dodd, and M. Nomura, Mol. Cell. Biol. 12, 5640-5651, 1992). A full-length cDNA clone of NPI-1 was generated from HeLa cell poly A + RNA. The viral nucleoprotein, which had been partially purified from influenza A/PR/8/34 virus-infected embryonated eggs, could be coprecipitated from solution by glutathione agarose beads complexed with a bacterially expressed glutathione-S-transferase-NPI-1 fusion protein, confirming the results of the yeast genetic system. Antisera raised against NPI-1 identified a 60-kDa polypeptide from total cellular extracts of both HeLa and MDBK cells. The viral nucleoprotein was coimmunoprecipitated from influenza A/WSN/33 virus-infected MDBK cells by anti-NPI-1 sera, demonstrating an interaction of these two proteins in infected cells. Similarly, NPI-1 was coimmunoprecipitated from MDBK cells by anti-NP sera. These experiments suggest that NPI-1 plays a role during influenza virus replication.

Cell type-specific desmosomal plaque proteins of the plakoglobin family: plakophilin 1 (band 6 protein)

Desmosomes represent a special type of the plaque-bearing adhering junctions, characteristic of certain pathways of cell differentiation, which compositionally are not identical in the various kinds of desmosome-forming cells. While all desmosomes contain the cytoplasmic plaque proteins desmoplakin I and plakoglobin, they can vary in their specific complement of desmosomal cadherins and by the presence of additional plaque proteins. We have raised monoclonal antibodies recognizing one such 'accessory' plaque protein, the cytokeratin-binding, basic protein plakophilin 1, originally introduced as 'band 6 protein' or 'polypeptide D6', which is an abundant desmosomal component in certain epithelia. Using such antibodies, we have isolated cDNA clones encoding the bovine and the human protein and determined their complete amino acid sequences. The mRNAs, which on Northern blot tests appear as two bands corresponding to approximately 4 and 2.4 kb (bovine) or 5 and 2.6 kb (human), code for 727 amino acids (calculated mol. wt. 80,180; IEP 9.25) in bovine and 726 amino acids (mol. wt. 80,496; IEP 9.34) in human plakophilin. Sequence analyses have revealed the presence of 9.2 repeated units of the arm-motif sequence, confirming our previous conclusion that this protein is a member of a larger family of proteins including, inter alia, several membrane-associated plaque proteins such as vertebrate plakoglobin and beta-catenin as well as the product of the armadillo gene of Drosophila. The plakophilin antibodies and cDNA probes have also allowed us to examine its synthesis in various tissues and cell cultures. While we confirm the occurrence of the protein in cytoskeletal fractions from various stratified squamous, complex, glandular duct and bladder epithelia, where it can be localized to desmosomes, we have, surprisingly, also identified the protein, although at lower amounts, in cytoskeletal fractions from several cultured cell lines in which the protein has not been consistently localized to desmosomes by immunofluorescence microscopy. Examples include cultured cells derived from certain simple epithelia such as the kidney-derived line MDBK and cultured calf lens cells. We have also found that, in all plakophilin 1-positive cells examined, a pool of diffusible ('soluble') cytoplasmic plakophilin exists, including cell lines such as human mammary carcinoma MCF-7 cells in which this soluble plakophilin seems to be the only detectable form. In addition, we have identified some soluble proteins conspicuously cross-reacting with plakophilin 1. Possible functions of plakophilin and its potential value as a marker for specific states of cell differentiation are discussed, particularly with respect to tumor diagnosis.

nup1 mutants exhibit pleiotropic defects in nuclear pore complex function

The NUP1 gene of Saccharomyces cerevisiae encodes one member of a family of nuclear pore complex proteins (nucleoporins) conserved from yeast to vertebrates. We have used mutational analysis to investigate the function of Nup1p. Deletion of either the amino- or carboxy-terminal domain confers a lethal phenotype, but partial truncations at either end affect growth to varying extents. Amino-terminal truncation causes mislocalization and degradation of the mutant protein, suggesting that this domain is required for targeting Nup1p to the nuclear pore complex. Carboxy-terminal mutants are stable but do not have wild-type function, and confer a temperature sensitive phenotype. Both import of nuclear proteins and export of poly(A) RNA are defective at the nonpermissive temperature. In addition, nup1 mutant cells become multinucleate at all temperatures, a phenotype suggestive of a defect in nuclear migration. Tubulin staining revealed that the mitotic spindle appears to be oriented randomly with respect to the bud, in spite of the presence of apparently normal cytoplasmic microtubules connecting one spindle pole body to the bud tip. EM analysis showed that the nuclear envelope forms long projections extending into the cytoplasm, which appear to have detached from the bulk of the nucleus. Our results suggest that Nup1p may be required to retain the structural integrity between the nuclear envelope and an underlying nuclear scaffold, and that this connection is required to allow reorientation of the nucleus in response to cytoskeletal forces.