A mutant cytochrome b5 with a lengthened membrane anchor escapes from the endoplasmic reticulum and reaches the plasma membrane.

Many resident membrane proteins of the endoplasmic reticulum (ER) do not have known retrieval sequences. Among these are the so-called tail-anchored proteins, which are bound to membranes by a hydrophobic tail close to the C terminus and have most of their sequence as a cytosolically exposed N-terminal domain. Because ER tail-anchored proteins generally have short (< or = 17 residues) hydrophobic domains, we tested whether this feature is important for localization, using cytochrome b5 as a model. The hydrophobic domain of cytochrome b5 was lengthened by insertion of five amino acids (ILAAV), and the localization of the mutant was analyzed by immunofluorescence in transiently transfected mammalian cells. While the wild-type cytochrome was localized to the ER, the mutant was relocated to the surface. This relocation was not due to the specific sequence introduced, as demonstrated by the ER localization of a second mutant, in which the original length of the membrane anchor was restored, while maintaining the inserted ILAAV sequence. Experiments with brefeldin A and with cycloheximide demonstrated that the extended anchor mutant reached the plasma membrane by transport along the secretory pathway. We conclude that the short membrane anchor of cytochrome b5 is important for its ER residency, and we discuss the relevance of this finding for other ER tail-anchored proteins.

Human plectin: organization of the gene, sequence analysis, and chromosome localization (8q24).

Plectin, a 500-kDa intermediate filament binding protein, has been proposed to provide mechanical strength to cells and tissues by acting as a cross-linking element of the cytoskeleton. To set the basis for future studies on gene regulation, tissue-specific expression, and pathological conditions involving this protein, we have cloned the human plectin gene, determined its coding sequence, and established its genomic organization. The coding sequence contains 32 exons that extend over 32 kb of the human genome. Most of the introns reside within a region encoding the globular N-terminal domain of the molecule, whereas the entire central rod domain and the entire C-terminal globular domain were found to be encoded by single exons of remarkable length, >3 kb and >6 kb, respectively. Overall, the organization of the human plectin gene was strikingly similar to that of human bullous pemphigoid antigen 1 (BPAG1), confirming that both proteins belong to the same gene family. Comparison of the deduced protein sequences for human and rat plectin revealed that they were 93% identical. By using fluorescence in situ hybridization, we have mapped the plectin gene to the long arm of chromosome 8 within the telomeric region. This gene locus (8q24) has previously been implicated in the human blistering skin disease epidermolysis bullosa simplex Ogna. Detailed knowledge of the structure of the plectin gene and its chromosome localization will aid in the elucidation of whether this or any other pathological conditions are linked to alterations in the plectin gene.

Defective dimerization of von Willebrand factor subunits due to a Cys-> Arg mutation in type IID von Willebrand disease.

The same heterozygous T -> C transition at nt 8567 of the von Willebrand factor (vWF) transcript was found in two unrelated patients with type III) von Willebrand disease, with no other apparent abnormality. In one family, both alleles were normal in the parents and one sister; thus, the mutation originated de novo in the proposita. The second patient also had asymptomatic parents who, however, were not available for study. The structural consequences of the identified mutation, resulting in the CyS2010 -> Arg substitution, were evaluated by expression of the vWF carboxyl-terminal domain containing residues 1366-2050. Insect cells infected with recombinant baculovirus expressing normal vWF sequence secreted a disulfide linked dimeric molecule with an apparent molecular mass of 150 kDa before reduction, yielding a single band of 80 kDa after disulfide bond reduction. In contrast, cells expressing the mutant fragment secreted a monomeric molecule of apparent molecular mass of 80 kDa, which remained unchanged after reduction. We conclude that CyS2010 is essential for normal dimerization of vWF subunits through disulfide bonding of carboxyl-terminal domains and that a heterozygous mutation in the corresponding codon is responsible for defective multimer formation in type III) von Willebrand disease.

Cell-cell communication regulates the effects of protein aspartate phosphatases on the phosphorelay controlling development in Bacillus subtilis.

Rap phosphatases are a recently discovered family of protein aspartate phosphatases that dephosphorylate the Spo0F--P intermediate of the phosphorelay, thus preventing sporulation of Bacillus subtilis. They are regulators induced by physiological processes that are antithetical to sporulation. The RapA phosphatase is induced by the ComP-ComA two-component signal transduction system responsible for initiating competence. RapA phosphatase activity was found to be controlled by a small protein, PhrA, encoded on the same transcript as RapA. PhrA resembles secreted proteins and the evidence suggests that it is cleaved by signal peptidase I and a 19-residue C-terminal domain is secreted from the cell. The sporulation deficiency caused by the uncontrolled RapA activity of a phrA mutant can be complemented by synthetic peptides comprising the last six or more of the C-terminal residues of PhrA. Whether the peptide controls RapA activity directly or by regulating its synthesis remains to be determined. Complementation of the phrA mutant can also be obtained in mixed cultures with a wild-type strain, suggesting the peptide may serve as a means of communication between cells. Importation of the secreted peptide required the oligopeptide transport system. The sporulation deficiency of oligopeptide transport mutants can be suppressed by mutating the rapA and rapB genes or by introduction of a spo0F mutation Y13S that renders the protein insensitive to Rap phosphatases. The data indicate that the sporulation deficiency of oligopeptide transport mutants is due to their inability to import the peptides controlling Rap phosphatases.

Isolation of an oxygen-sensitive FNR protein of Escherichia coli: interaction at activator and repressor sites of FNR-controlled genes.

The Escherichia coli fnr gene product, FNR, is a DNA binding protein that regulates a large family of genes involved in cellular respiration and carbon metabolism during conditions of anaerobic cell growth. FNR is believed to contain a redox/O2-sensitive element for detecting the anaerobic state. To investigate this process, a fnr mutant that encodes an altered FNR protein with three amino acid substitutions in the N-terminal domain was constructed by site-directed mutagenesis. In vivo, the mutant behaved like a wild-type strain under anaerobic conditions but had a 14-fold elevated level of transcriptional activation of a reporter gene during aerobic cell growth. The altered fur gene was overexpressed in E. coli and the resultant FNR protein was purified to near homogeneity by using anaerobic chromatography procedures. An in vitro Rsa I restriction site protection assay was developed that allowed for the assessment of oxygen-dependent DNA binding of the mutant FNR protein. The FNR protein was purified as a monomer of M(r) 28,000 that contained nonheme iron at 2.05 +/- 0.34 mol of Fe per FNR monomer. In vitro DNase I protection studies were performed to establish the locations of the FNR-binding sites at the narG, narK, dmsA, and hemA promoters that are regulated by either activation or repression of their transcription. The sizes of the DNA footprints are consistent with the binding of two monomers of FNR that protect the symmetrical FNR-recognition sequence TTGAT-nnnnATCAA. Exposure of the FNR protein or protein-DNA complex to air for even short periods of time (approximately 5 min) led to the complete loss of DNA protection at a consensus FNR recognition site. A model whereby the FNR protein exists in the cell as a monomer that assembles on the DNA under anaerobic conditions to form a dimer is discussed.

Identification of the YopE and YopH domains required for secretion and internalization into the cytosol of macrophages, using the cyaA gene fusion approach.

Pathogenic yersiniae secrete a set of antihost proteins, called Yops, by a type III secretion mechanism. Upon infection of cultured epithelial cells, extracellular Yersinia pseudotuberculosis and Yersinia enterocolitica translocate cytotoxin YopE across the host cell plasma membrane. Several lines of evidence suggest that tyrosine phosphatase YopH follows the same pathway. We analyzed internalization of YopE and YopH into murine PU5-1.8 macrophages by using recombinant Y. enterocolitica producing truncated YopE and YopH proteins fused to a calmodulin-dependent adenylate cyclase. The YopE-cyclase and YopH-cyclase hybrids were readily secreted by Y. enterocolitica. The N-terminal domain required for secretion was not longer than 15 residues of YopE and 17 residues of YopH. Internalization into eukaryotic cells, revealed by cAMP production, only required the N-terminal 50 amino acid residues of YopE and the N-terminal 71 amino acid residues of YopH. YopE and YopH are thus modular proteins composed of a secretion domain, a translocation domain, and an effector domain. Translocation of YopE and YopH across host cell's membranes was also dependent on the secretion of YopB and YopD by the same bacterium. The cyclase fusion approach could be readily extended to study the fate of other proteins secreted by invasive bacterial pathogens.

Palmitoylation of the GluR6 kainate receptor.

The G-protein-coupled metabotropic glutamate receptor mGluR1 alpha and the ionotropic glutamate receptor GluR6 were examined for posttranslational palmitoylation. Recombinant receptors were expressed in baculovirus-infected insect cells or in human embryonic kidney cells and were metabolically labeled with [3H]palmitic acid. The metabotropic mGluR1 alpha receptor was not labeled whereas the GluR6 kainate receptor was labeled after incubation with [3H]palmitate. The [3H]palmitate labeling of GluR6 was eliminated by treatment with hydroxylamine, indicating that the labeling was due to palmitoylation at a cysteine residue via a thioester bond. Site-directed mutagenesis was used to demonstrate that palmitoylation of GluR6 occurs at two cysteine residues, C827 and C840, located in the carboxyl-terminal domain of the molecule. A comparison of the electrophysiological properties of the wild-type and unpalmitoylated mutant receptor (C827A, C840A) showed that the kainate-gated currents produced by the unpalmitoylated mutant receptor were indistinguishable from those of the wild-type GluR6. The unpalmitoylated mutant was a better substrate for protein kinase C than the wild-type GluR6 receptor. These data indicate that palmitoylation may not modulate kainate channel function directly but instead affect function indirectly by regulating the phosphorylation state of the receptor.

Differential activation of yeast adenylyl cyclase by Ras1 and Ras2 depends on the conserved N terminus.

Although both Ras1 and Ras2 activate adenylyl cyclase in yeast, a number of differences can be observed regarding their function in the cAMP pathway. To explore the relative contribution of conserved and variable domains in determining these differences, chimeric RAS1-RAS2 or RAS2-RAS1 genes were constructed by swapping the sequences encoding the variable C-terminal domains. These constructs were expressed in a cdc25ts ras1 ras2 strain. Biochemical data show that the difference in efficacy of adenylyl cyclase activation between the two Ras proteins resides in the highly conserved N-terminal domain. This finding is supported by the observation that Ras2 delta, in which the C-terminal domain of Ras2 has been deleted, is a more potent activator of the yeast adenylyl cyclase than Ras1 delta, in which the C-terminal domain of Ras1 has been deleted. These observations suggest that amino acid residues other than the highly conserved residues of the effector domain within the N terminus may determine the efficiency of functional interaction with adenylyl cyclase. Similar levels of intracellular cAMP were found in Ras1, Ras1-Ras2, Ras1 delta, Ras2, and Ras2-Ras1 strains throughout the growth curve. This was found to result from the higher expression of Ras1 and Ras1-Ras2, which compensate for their lower efficacy in activating adenylyl cyclase. These results suggest that the difference between the Ras1 and the Ras2 phenotype is not due to their different efficacy in activating the cAMP pathway and that the divergent C-terminal domains are responsible for these differences, through interaction with other regulatory elements.

A cooperative model for receptor recognition and cell adhesion: evidence from the molecular packing in the 1.6-A crystal structure of the pheromone Er-1 from the ciliated protozoan Euplotes raikovi.

The crystal structure of the pheromone Er-1 from the unicellular eukaryotic organism Euplotes raikovi was determined at 1.6 A resolution and refined to a crystallographic R factor of 19.9%. In the tightly packed crystal, two extensive intermolecular helix-helix interactions arrange the Er-1 molecules into layers. Since the putative receptor of the pheromone is a membrane-bound protein, whose extracellular C-terminal domain is identical in amino acid sequence to the soluble pheromone, the interactions found in the crystal may mimic the pheromone-receptor interactions as they occur on a cell surface. Based on this, we propose a model for the interaction between soluble pheromone molecules and their receptors. In this model, strong pheromone-receptor binding emerges as a consequence of the cooperative utilization of several weak interactions. The model offers an explanation for the results of binding studies and may also explain the adhesion between cells that occurs during mating.

Crystal structure of the large fragment of Thermus aquaticus DNA polymerase I at 2.5-A resolution: structural basis for thermostability.

The crystal structure of the large fragment of the Thermus aquaticus DNA polymerase (Klentaq1), determined at 2.5-A resolution, demonstrates a compact two-domain architecture. The C-terminal domain is identical in fold to the equivalent region of the Klenow fragment of Escherichia coli DNA polymerase I (Klenow pol I). Although the N-terminal domain of Klentaq1 differs greatly in sequence from its counterpart in Klenow pol I, it has clearly evolved from a common ancestor. The structure of Klentaq1 reveals the strategy utilized by this protein to maintain activity at high temperatures and provides the structural basis for future improvements of the enzyme.

Oligomeric structure of caveolin: implications for caveolae membrane organization.

A 22-kDa protein, caveolin, is localized to the cytoplasmic surface of plasma membrane specializations called caveolae. We have proposed that caveolin may function as a scaffolding protein to organize and concentrate signaling molecules within caveolae. Here, we show that caveolin interacts with itself to form homooligomers. Electron microscopic visualization of these purified caveolin homooligomers demonstrates that they appear as individual spherical particles. By using recombinant expression of caveolin as a glutathione S-transferase fusion protein, we have defined a region of caveolin's cytoplasmic N-terminal domain that mediates these caveolin-caveolin interactions. We suggest that caveolin homooligomers may function to concentrate caveolin-interacting molecules within caveolae. In this regard, it may be useful to think of caveolin homooligomers as "fishing lures" with multiple "hooks" or attachment sites for caveolin-interacting molecules.

Phosphorylation-dependent conformational changes in OmpR, an osmoregulatory DNA-binding protein of Escherichia coli.

Osmoregulated porin gene expression in Escherichia coli is controlled by the two-component regulatory system EnvZ and OmpR. EnvZ, the osmosensor, is an inner membrane protein and a histidine kinase. EnvZ phosphorylates OmpR, a cytoplasmic DNA-binding protein, on an aspartyl residue. Phospho-OmpR binds to the promoters of the porin genes to regulate the expression of ompF and ompC. We describe the use of limited proteolysis by trypsin and ion spray mass spectrometry to characterize phospho-OmpR and the conformational changes that occur upon phosphorylation. Our results are consistent with a two-domain structure for OmpR, an N-terminal phosphorylation domain joined to a C-terminal DNA-binding domain by a flexible linker region. In the presence of acetyl phosphate, OmpR is phosphorylated at only one site. Phosphorylation induces a conformational change that is transmitted to the C-terminal domain via the central linker. Previous genetic analysis identified a region in the C-terminal domain that is required for transcriptional activation. Our results indicate that this region is within a surface-exposed loop. We propose that this loop contacts the alpha subunit of RNA polymerase to activate transcription. Mass spectrometry also reveals an unusual dephosphorylated form of OmpR, the potential significance of which is discussed.

Structural domains of IS10 transposase and reconstitution of transposition activity from proteolytic fragments lacking an interdomain linker.

All of the DNA cleavage and strand transfer events required for transposition of insertion sequence IS10 are carried out by a 46-kDa IS10-encoded transposase protein. Limited proteolysis demonstrates that transposase has two principal structural domains, a 28-kDa N-terminal domain (N alpha beta; aa 1-246) and a 17-kDa C-terminal domain (C; aa 256-402). The two domains are connected by a 1-kDa proteolytic-sensitive linker region (aa 247-255). The N-terminal domain N alpha beta can be further subdivided into domains N alpha and N beta by a weaker protease-sensitive site located 6 kDa (53 aa) from the N terminus. The N beta and N alpha beta fragments are capable of nonspecific DNA binding as determined by Southwestern blot analysis. None of the fragments alone is capable of carrying out the first step of transposition, assembly of a synaptic complex containing a pair of transposon ends. Remarkably, complete transposition activity can be reconstituted by mixing fragment N alpha beta and fragment C, with or without the intervening linker region. We infer that the structural integrity of transposase during the transitions involved in the chemical steps of the transposition reaction is maintained independent of the linker, presumably by direct contacts between and among the principal domains. Reconstitution of activity in the absence of the linker region is puzzling, however, because mutations that block strand transfer or affect insertion specificity alter linker region residues. Additional reconstitution experiments demonstrate that the N alpha region is dispensable for formation of a synaptic complex but is required for complexes to undergo cleavage.

Identification of a plant serine-arginine-rich protein similar to the mammalian splicing factor SF2/ASF.

We show that the higher plant Arabidopsis thaliana has a serine-arginine-rich (SR) protein family whose members contain a phosphoepitope shared by the animal SR family of splicing factors. In addition, we report the cloning and characterization of a cDNA encoding a higher-plant SR protein from Arabidopsis, SR1, which has striking sequence and structural homology to the human splicing factor SF2/ASF. Similar to SF2/ASF, the plant SR1 protein promotes splice site switching in mammalian nuclear extracts. A novel feature of the Arabidopsis SR protein is a C-terminal domain containing a high concentration of proline, serine, and lysine residues (PSK domain), a composition reminiscent of histones. This domain includes a putative phosphorylation site for the mitotic kinase cyclin/p34cdc2.

One sequence, two folds: a metastable structure of CD2.

When expressed as part of a glutathione S-transferase fusion protein the NH2-terminal domain of the lymphocyte cell adhesion molecule CD2 is shown to adopt two different folds. The immunoglobulin superfamily structure of the major (85%) monomeric component has previously been determined by both x-ray crystallography and NMR spectroscopy. We now describe the structure of a second, dimeric, form present in about 15% of recombinant CD2 molecules. After denaturation and refolding in the absence of the fusion partner, dimeric CD2 is converted to monomer, illustrating that the dimeric form represents a metastable folded state. The crystal structure of this dimeric form, refined to 2.0-A resolution, reveals two domains with overall similarity to the IgSF fold found in the monomer. However, in the dimer each domain is formed by the intercalation of two polypeptide chains. Hence each domain represents a distinct folding unit that can assemble in two different ways. In the dimer the two domains fold around a hydrophilic interface believed to mimic the cell adhesion interaction at the cell surface, and the formation of dimer can be regulated by mutating single residues at this interface. This unusual misfolded form of the protein, which appears to result from inter- rather than intramolecular interactions being favored by an intermediate structure formed during the folding process, illustrates that evolution of protein oligomers is possible from the sequence for a single protein domain.