Analysis of the BiP gene and identification of an ER retention signal in Schizosaccharomyces pombe.

We have cloned the gene for the resident luminal ER protein BiP from the fission yeast, Schizosaccharomyces pombe. The predicted protein product is equally divergent from the budding yeast and mammalian homologues. Disruption of the BiP gene in S. pombe is lethal and BiP mRNA levels are regulated by a variety of stresses including heat shock. Immunofluorescence of cells expressing an epitope-tagged BiP protein show it to be localized to the nuclear envelope, around the cell periphery and in a reticular structure through the cytoplasm. Unexpectedly, we find the BiP protein contains an N-linked glycosylation site which can be utilized. The C-terminal four amino acids of BiP are Ala-Asp-Glu-Leu, a new variant of the XDEL sequence found at the C-termini of luminal endoplasmic reticulum proteins. To determine whether this sequence acts as a sorting signal in S.pombe we expressed an acid phosphatase fusion protein extended at its C-terminus with the amino acids ADEL. Analysis of the sorting of this fusion protein indicates that the ADEL sequence is sufficient to cause the retention of proteins in the endoplasmic reticulum. The sequences DDEL, HDEL and KDEL can also direct ER-retention of acid phosphatase in S.pombe.     

Human nucleoporin p62 and the essential yeast nuclear pore protein NSP1 show sequence homology and a similar domain organization

NSP1 is an essential nuclear pore protein in yeast. We observed that anti-NSP1 antibodies label mammalian nuclear pore complexes and recognize nucleoporin p62. Also peptide antibodies raised against the NSP1 carboxy-terminal end cross-react with p62, a conserved component of the nuclear pore complex in higher eukaryotes. To further analyze the structural and functional similarity between NSP1 and mammalian nucleoporins, we cloned and sequenced nucleoporin p62 from a HeLa cDNA library. Human p62 consists of a carboxy-terminal domain homologous to the essential yeast NSP1 carboxy-terminal domain and an amino-terminal half resembling the repetitive middle domain of NSP1. The full-length p62 and a fusion protein consisting of cytosolic mouse dihydrofolate reductase (DHFR) and the p62 carboxy-terminal domain were expressed in transfected HeLa cells. Only overexpressed full-length p62, but not the DHFR-C-p62 fusion protein, binds wheat germ agglutinin (WGA). This suggests that modification by N-acetylglucosamine is mainly restricted to the repetitive amino-terminal half of p62 and implies a role of this type of repetitive sequences in nuclear transport. In the transfected HeLa cells, the DHFR-C-p62 fusion protein forms patchy aggregates that accumulate at the nuclear periphery but are also scattered through the cytoplasm. It is suggested that nucleoporin p62 may be targeted and anchored to the pore complex via its carboxy-terminal domain which reveals a hydrophobic heptad repeat organization similar to that found in lamins and other intermediate filament proteins.     

Identification of a signal that distinguishes between the chloroplast outer envelope membrane and the endomembrane system in vivo

Certain small outer envelope membrane proteins of chloroplasts are encoded by the nuclear genome without a cleavable N-terminal transit peptide. We investigated in vivo the targeting mechanism of AtOEP7, an Arabidopsis homolog of the small outer envelope membrane protein. AtOEP7 was expressed as a fusion protein with the green fluorescent protein (GFP) either transiently in protoplasts or stably in transgenic plants. In either case, fluorescence microscopy of transformed cells and protein gel blot analysis of fractionated proteins confirmed that the AtOEP7:GFP fusion protein was targeted to the chloroplast outer envelope membrane. In vivo targeting experiments revealed that two regions, the transmembrane domain (TMD) and its C-terminal neighboring seven-amino acid region, were necessary and sufficient for targeting to the chloroplast outer membrane. Substitution of aspartic acid or lysine residues with glycine residues or scrambling of the amino acid sequence of the seven-amino acid region caused mistargeting to the plasma membrane. Although the amino acid sequence of the TMD is not important for targeting, amino acid residues with large side chains inhibited targeting to the chloroplasts and resulted in the formation of large aggregates in the protoplasts. In addition, introduction of a proline residue within the TMD resulted in inhibition of targeting. Finally, a fusion protein, AtOEP7:NLS:GFP, was targeted efficiently to the chloroplast envelope membranes despite the presence of a nuclear localization signal. On the basis of these results, we conclude that the seven-amino acid region and the TMD are determinants for targeting to the chloroplast outer envelope membrane. The seven-amino acid region plays a critical role in AtOEP7 evading the endomembrane system and entering the chloroplast pathway, and the TMD plays critical roles in migration to the chloroplasts and/or subsequent insertion into the membrane.     

Identification in chickens of an evolutionarily conserved cellular ets-2 gene (c-ets-2) encoding nuclear proteins related to the products of the c-ets proto-oncogene.

In chicken cells, we previously identified a set of proteins (p58-64) structurally related to, but distinct from, the products encoded by the c-ets proto-oncogene. We report here the isolation and nucleotide sequence of a cDNA encoding nuclear products of mol. wt 58, 60, 62 and 64 kd, indistinguishable from those detected in chicken cells. The p60 and p64 species appear to represent phosphorylated versions on serine and threonine residues of p58 and p62. The homology of p58-64 to other ets-related proteins, including the v-ets encoded domain of the transforming protein of avian leukemia virus E26 and p54c-ets, the translation product of the chicken (Ck) c-ets gene, is confined to two regions of 175 and 96 amino acid residues localized respectively at the carboxy-terminal domain and close to the amino-terminal domain of these molecules. This cDNA corresponds to a gene localized in a locus distinct from that of c-ets which is transcribed as a 4.0-kb RNA species in most chicken tissues. We also identified the human (Hu) c-ets-2-encoded products as two proteins of 60 and 62 kd, highly related to chicken p58-64. This, together with the fact that the amino acid sequence of the cDNA encoding p58-64 is 95% identical to the reported partial sequence of a Hu-c-ets-2 cDNA, indicates that p58-64 are the translation products of the Ck-c-ets-2 gene.     

Cell surface anchorage and ligand-binding domains of the Saccharomyces cerevisiae cell adhesion protein alpha-agglutinin, a member of the immunoglobulin superfamily.

alpha-Agglutinin is a cell adhesion glycoprotein expressed on the cell wall of Saccharomyces cerevisiae alpha cells. Binding of alpha-agglutinin to its ligand a-agglutinin, expressed by a cells, mediates cell-cell contact during mating. Analysis of truncations of the 650-amino-acid alpha-agglutinin structural gene AG alpha 1 delineated functional domains of alpha-agglutinin. Removal of the C-terminal hydrophobic sequence allowed efficient secretion of the protein and loss of cell surface attachment. This cell surface anchorage domain was necessary for linkage to a glycosyl phosphatidylinositol anchor. A construct expressing the N-terminal 350 amino acid residues retained full a-agglutinin-binding activity, localizing the binding domain to the N-terminal portion of alpha-agglutinin. A 278-residue N-terminal peptide was inactive; therefore, the binding domain includes residues between 278 and 350. The segment of alpha-agglutinin between amino acid residues 217 and 308 showed significant structural and sequence similarity to a consensus sequence for immunoglobulin superfamily variable-type domains. The similarity of the alpha-agglutinin-binding domain to mammalian cell adhesion proteins suggests that this structure is a highly conserved feature of adhesion proteins in diverse eukaryotes.     

Functional dissection of the ETS transcription factor MEF

We previously indicated that myeloid elf-1-like factor (MEF) but not elf-1, specifically activated lysozyme gene expression in epithelial cells. MEF is highly homologous at the nucleotide and amino acid level, with elf-1 especially in the ETS domain. Here, we report the functional analysis of the nuclear localization and transactivation properties of MEF. To investigate the intracellular localization of MEF, we transiently transfected MEF-green fluorescence protein (GFP) fusion protein expression vector into HeLa cells. A region spanning residues 177-291 is required for nuclear localization. We produced deletion mutants of MEF to determine the transactivation domain. The data showed that the N-terminal region, encompassing amino acids 1-52 is a potent transactivation domain. The C-terminal region spanning residues 477-663 can also mediate transactivation but not as strongly as the N-terminal region. The activity of the amino acid residues 1-52 was confirmed by experiments with fused constructs of MEF to the DNA binding-domain of the yeast GAL4 protein. These results, which determined the localization of the functional domains of MEF, will provide us with new clues to its transactivation mechanisms to regulate lysozyme gene expression in epithelial cells.     

CHMP7, a novel ESCRT-III-related protein, associates with CHMP4b and functions in the endosomal sorting pathway

All CHMPs (charged multivesicular body proteins) reported to date have common features: they all contain approx. 200 amino acid residues, have coiled-coil regions and have a biased distribution of charged residues (basic N-terminal and acidic C-terminal halves). Yeast orthologues of CHMPs, including an ESCRT-III component Snf7, are required for the sorting of cargo proteins to intraluminal vesicles of multivesicular bodies. We have characterized a novel human ESCRT-III-related protein, designated CHMP7, which consists of 453 amino acid residues. CHMP7 contains an SNF7 domain and a distantly SNF7-related domain in its C-terminal half and N-terminal half respectively. Among the ten CHMP proteins classified previously in six subfamilies (CHMP1-CHMP6), the C-terminal SNF7 domain of CHMP7 is most similar to the SNF7 domain of CHMP6, which associates with CHMP4 proteins and EAP20, a component of ESCRT-II. Pull-down assays using lysates of HEK-293T (human embryonic kidney) cells that overexpressed Strep-tagged CHMP7 and GFP (green fluorescent protein)-fused CHMP4b (also named Shax1) revealed a positive interaction between the C-terminal half of CHMP7 and CHMP4b. However, interaction was not observed between CHMP7 and EAP20. Confocal fluorescence microscopic analyses revealed that FLAG-CHMP7 is distributed in HeLa cells diffusely throughout the cytoplasm, but with some accumulation, especially in the perinuclear area. The distribution of FLAG-CHMP7 was altered to a cytoplasmic punctate pattern by overexpression of either CHMP4b-GFP or GFP-Vps4B(E235Q), a dominant-negative mutant of the AAA (ATPase associated with various cellular activities) Vps4B, and partially co-localized with them. Ubiquitinated proteins and endocytosed EGF accumulated in GFP-CHMP7-expressing cells. A dominant-negative effect of overexpressed GFP-CHMP7 was also observed in the release of virus-like particles from HEK-293T cells that transiently expressed the MLV (murine leukaemia virus) Gag protein. These results suggest that CHMP7, a novel CHMP4-associated ESCRT-III-related protein, functions in the endosomal sorting pathway.     

A novel FK506- and rapamycin-binding protein (FPR3 gene product) in the yeast Saccharomyces cerevisiae is a proline rotamase localized to the nucleolus

The gene (FPR3) encoding a novel type of peptidylpropyl-cis-trans- isomerase (PPIase) was isolated during a search for previously unidentified nuclear proteins in Saccharomyces cerevisiae. PPIases are thought to act in conjunction with protein chaperones because they accelerate the rate of conformational interconversions around proline residues in polypeptides. The FPR3 gene product (Fpr3) is 413 amino acids long. The 111 COOH-terminal residues of Fpr3 share greater than 40% amino acid identity with a particular class of PPIases, termed FK506-binding proteins (FKBPs) because they are the intracellular receptors for two immunosuppressive compounds, rapamycin and FK506. When expressed in and purified from Escherichia coli, both full-length Fpr3 and its isolated COOH-terminal domain exhibit readily detectable PPIase activity. Both fpr3 delta null mutants and cells expressing FPR3 from its own promoter on a multicopy plasmid have no discernible growth phenotype and do not display any alteration in sensitivity to the growth-inhibitory effects of either FK506 or rapamycin. In S. cerevisiae, the gene for a 112-residue cytosolic FKBP (FPR1) and the gene for a 135-residue ER-associated FKBP (FPR2) have been described before. Even fpr1 fpr2 fpr3 triple mutants are viable. However, in cells carrying an fpr1 delta mutation (which confers resistance to rapamycin), overexpression from the GAL1 promoter of the C-terminal domain of Fpr3, but not full-length Fpr3, restored sensitivity to rapamycin. Conversely, overproduction from the GAL1 promoter of full- length Fpr3, but not its COOH-terminal domain, is growth inhibitory in both normal cells and fpr1 delta mutants. In fpr1 delta cells, the toxic effect of Fpr3 overproduction can be reversed by rapamycin. Overproduction of the NH2-terminal domain of Fpr3 is also growth inhibitory in normal cells and fpr1 delta mutants, but this toxicity is not ameliorated in fpr1 delta cells by rapamycin. The NH2-terminal domain of Fpr3 contains long stretches of acidic residues alternating with blocks of basic residues, a structure that resembles sequences found in nucleolar proteins, including S. cerevisiae NSR1 and mammalian nucleolin. Indirect immunofluorescence with polyclonal antibodies raised against either the NH2- or the COOH-terminal segments of Fpr3 expressed in E. coli demonstrated that Fpr3 is located exclusively in the nucleolus.     

Transport of proteins to the mitochondrial intermembrane space: the ''matrix-targeting'' and the ''sorting'' domains in the cytochrome c1 presequence.

We reported earlier that the yeast cytochrome c1 presequence (length: 61 amino acids) directs attached proteins to the mitochondrial intermembrane space and that it appears to contain two functional domains: a 'matrix-targeting' domain, and a 'sorting' domain. We have now used gene manipulation together with two different in vivo import assays to map these two domains within the cytochrome c1 presequence. The 'matrix-targeting' domain is contained within the N-terminal 16 residues (or less); by itself, it directs attached proteins to the matrix. The 'sorting' domain extends into the C-terminal 13 residues of the presequence; while it does not mediate intracellular protein transport by itself, it acts together with the preceding 'matrix-targeting' sequence in sorting attached proteins into the intermembrane space. On replacing the authentic 'matrix-targeting' sequence with artificial sequences of different lengths we found that sorting of proteins between the outer membrane and the intermembrane space is not exclusively determined by the length of the N-terminal 'matrix-targeting' sequence.     

Mutations in the membrane-spanning domain of the human immunodeficiency virus envelope glycoprotein that affect fusion activity.

A chimeric protein consisting of the human immunodeficiency virus type 1 (HIV-1) envelope protein (Env) ectodomain joined to the transmembrane and cytoplasmic-tail domains of vesicular stomatitis virus G protein lost the ability to fuse CD4+ HeLa cells yet was transported to the cell surface and cleaved normally. These results suggested some critical role of the HIV gp41 transmembrane or cytoplasmic domain in fusion. Subsequent mutagenic analysis of the HIV-1 Env transmembrane domain revealed that the sequence of amino acid residues from positions 696 to 707 of the transmembrane domain was important for fusion function but was not required for anchoring of the Env protein in the lipid bilayer or for transport to the cell surface. Further analysis indicated that the basic residues at positions 696 and 707 were critical for membrane fusion activity, as was the spacing between these residues. These results demonstrate that in addition to providing an anchoring function, the specific amino acid sequence in the transmembrane domain plays a crucial role in the membrane fusion process.     

Sun2 is a novel mammalian inner nuclear membrane protein

Sun protein (Sun1 and Sun2) cDNAs were previously cloned based on the homology of their C-terminal regions (SUN (Sad1 and UNC) domain) with the Caenorhabditis elegans protein UNC-84 whose mutation disrupts nuclear migration/positioning. In this study, we raised an anti-Sun2 serum and identified Sun2 in mammalian cells. In HeLa cells, Sun2 displays a nuclear rim-like pattern typical for a nuclear envelope protein. The Sun2 antibody signal co-localizes with nuclear pore and INM markers signals. The rim-like pattern was also observed with the recombinant full-length Sun2 protein fused to either EGFP or V5 epitopes. In addition, we found that a recombinant truncated form of Sun2, extending from amino acids 26 to 339, is sufficient to specify the nuclear envelope localization. Biochemical analyses show that Sun2 is an 85-kDa protein that is partially insoluble in detergent with high salt concentration and in chaotropic agents. Furthermore, Sun2 is enriched in purified HeLa cell nuclei. Electron microscopy analysis shows that Sun2 localizes in the nuclear envelope with a sub-population present in small clusters. Additionally, we show that the SUN domain of Sun2 is localized to the periplasmic space between the inner and the outer nuclear membranes. From our data, we conclude that Sun2 is a new mammalian inner nuclear membrane protein. Because the SUN domain is conserved from fission yeast to mammals, we suggest that Sun2 belongs to a new class of nuclear envelope proteins with potential relevance to nuclear membrane function in the context of the involvement of its components in an increasing spectrum of human diseases.     

Structural and functional dissection of Sec62p, a membrane-bound component of the yeast endoplasmic reticulum protein import machinery.

SEC62 is required for the import of secretory protein precursors into the endoplasmic reticulum (ER) of Saccharomyces cerevisiae. The DNA sequence of SEC62 predicts a 32-kDa polypeptide with two potential membrane-spanning segments. Two antisera directed against different portions of the SEC62 coding region specifically detected a 30-kDa polypeptide in cell extracts. A combination of subcellular fractionation, detergent and alkali extraction, and indirect immunofluorescence studies indicated that Sec62p is intimately associated with the ER membrane. Protease digestion of intact microsomes and analysis of the oligosaccharide content of a set of Sec62p-invertase hybrid proteins suggested that Sec62p spans the ER membrane twice, displaying hydrophilic amino- and carboxy-terminal domains towards the cytosol. Sec62p-invertase hybrid proteins that lack the Sec62p C terminus failed to complement the sec62-l mutation and dramatically inhibited the growth of sec62-l cells at a normally permissive temperature. The inhibitory action of toxic Sec62p-invertase hybrids was partially counteracted by the overexpression of Sec63p. Taken together, these data suggest that the C-terminal domain of Sec62p performs an essential function and that the N-terminal domain associates with other components of the translocation machinery, including Sec63p.     

An essential yeast protein, CBF5p, binds in vitro to centromeres and microtubules.

Yeast centromere DNA (CEN) affinity column chromatography has been used to purify several putative centromere and kinetochore proteins from yeast chromatin extracts. The single yeast gene (CBF5) specifying one of the major low-affinity centromere-binding proteins (p64'/CBF5p) has been cloned and shown to be essential for viability of Saccharomyces cerevisiae. CBF5 specifies a 55-kDa highly charged protein that contains a repeating KKD/E sequence domain near the C terminus, similar to known microtubule-binding domains in microtubule-associated proteins 1A and 1B, CBF5p, obtained by overexpression in bacterial cells, binds microtubules in vitro, whereas C-terminal deleted proteins lacking the (KKD/E)n domain do not. Dividing yeast cells containing a C-terminal truncated CBF5 gene, producing CBF5p containing only three copies of the KKD/E repeat, delay with replicated genomes at the G2/M phase of the cell cycle, while depletion of CBF5p arrests most cells in G1/S. Overproduction of CBF5p in S. cerevisiae complements a temperature sensitivity mutation in the gene (CBF2) specifying the 110-kDa subunit of the high-affinity CEN DNA-binding factor CBF3, suggesting in vivo interaction of CBF5p and CBF3. A second low-affinity centromere-binding factor has been identified as topoisomerase II.     

Crystallization of a trimeric human T cell leukemia virus type 1 gp21 ectodomain fragment as a chimera with maltose-binding protein.

We present a novel protein crystallization strategy, applied to the crystallization of human T cell leukemia virus type 1 (HTLV-1) transmembrane protein gp21 lacking the fusion peptide and the transmembrane domain, as a chimera with the Escherichia coli maltose binding protein (MBP). Crystals could not be obtained with a MBP/gp21 fusion protein in which fusion partners were separated by a flexible linker, but were obtained after connecting the MBP C-terminal alpha-helix to the predicted N-terminal alpha-helical sequence of gp21 via three alanine residues. The gp21 sequences conferred a trimeric structure to the soluble fusion proteins as assessed by sedimentation equilibrium and X-ray diffraction, consistent with the trimeric structures of other retroviral transmembrane proteins. The envelope protein precursor, gp62, is likewise trimeric when expressed in mammalian cells. Our results suggest that MBP may have a general application for the crystallization of proteins containing N-terminal alpha-helical sequences.     

Vacuolar protein sorting in fission yeast: cloning, biosynthesis, transport, and processing of carboxypeptidase Y from Schizosaccharomyces pombe.

PCR was used to isolate a carboxypeptidase Y (CPY) homolog gene from the fission yeast Schizosaccharomyces pombe. The cloned S. pombe cpy1+ gene has a single open reading frame, which encodes 950 amino acids with one potential N-glycosylation site. It appears to be synthesized as an inactive pre-pro protein that likely undergoes processing following translocation into appropriate intracellular organelles. The C-terminal mature region is highly conserved in other serine carboxypeptidases. In contrast, the N-terminal pro region containing the vacuolar sorting signal in CPY from Saccharomyces cerevisiae shows fewer identical residues. The pro region contains two unusual repeating sequences; repeating sequence I consists of seven contiguous repeating segments of 13 amino acids each, and repeating sequence II consists of seven contiguous repeating segments of 9 amino acids each. Pulse-chase radiolabeling analysis revealed that Cpy1p was initially synthesized in a 110-kDa pro-precursor form and via the 51-kDa single-polypeptide-chain intermediate form which has had its pro segment removed is finally converted to a heterodimer, the mature form, which is detected as a 32-kDa protein on sodium dodecyl sulfate-polyacrylamide gel electrophoresis under reducing conditions. Like S. cerevisiae CPY, S. pombe Cpy1p does not require the N-linked oligosaccharide moiety for vacuolar delivery. To investigate the vacuolar sorting signal of S. pombe Cpy1p, we have constructed cpy1+-SUC2 gene fusions that direct the synthesis of hybrid proteins consisting of N-terminal segments of various lengths of S. pombe Cpy1p fused to the secreted enzyme S. cerevisiae invertase. The N-terminal 478 amino acids of Cpy1 are sufficient to direct delivery of a Cpy1-Inv hybrid protein to the vacuole. These results showed that the pro peptide of Cpy1 contains the putative vacuolar sorting signal.     

The interferon alpha induced protein ISG12 is localized to the nuclear membrane.

Interferons exert their biological function mainly through the activation of interferon-stimulated genes (ISGs). ISG12 (originally designated p27) belongs to a family of small, interferon alpha inducible genes of unknown function. We have determined the 5' end sequence of ISG12 cDNA from the human cell lines HeLa and AMA by RACE. Comparing this sequence to ISG12 sequences in the expressed sequence tag (EST) database revealed the presence of two alternative splice variants of ISG12 in human cells exhibiting the same open reading frame. We have sequenced the promoter region of the ISG12 gene and found ISRE, IRF1/IRF2, and STAT elements correlating to the interferon alpha inducibility of the gene. Subsequently, we have expressed human ISG12, a 12-kDa hydrophobic protein in the baculovirus expression system and with a C-terminal FLAG-tag in the human cell line 293. Recombinant ISG12 sediments in the nuclear envelope in both cell types. Finally, we have been able to demonstrate the prevalence of the ISG12 gene product in the nuclear envelope of HeLa cells treated with interferon alpha by immunocytochemical analyses. ISG12 is the first interferon induced protein found localizing to the nuclear envelope.     

Identification of p34 and p13, human homologs of the cell cycle regulators of fission yeast encoded by cdc2+ and suc1+

cdc2+ and CDC28 play central roles in the cell division cycles of the widely divergent yeasts Schizosaccharomyces pombe and Saccharomyces cerevisiae, respectively. The genes encode protein kinases that show 62% protein sequence identity and are capable of cross-complementation. Monoclonal antibodies were raised against p34cdc2, and a subset recognize p36cdc28. The cross-reacting antibodies detected a 34 kd homolog of the p34cdc2/p36CDC28, protein in HeLa cells. Human p34 was also recognized by an affinity-purified polyclonal anti-p34cdc2 serum. Peptide mapping of p34cdc2, p36CDC28, and human p34 revealed complete conservation of four tryptophan residues in the three proteins. p34 thus appears to be closely related to the two yeast proteins. In addition, a p34 immune complex showed protein kinase activity in vitro, and HeLa cell p34 interacts with p13, the human homolog of the suc1+ gene product of S. pombe.