Extracellular mutations of protease-activated receptor-1 result in differential activation by thrombin and thrombin receptor agonist peptide.

The protease-activated thrombin receptor-1 (PAR-1) can be activated by both the tethered ligand exposed by thrombin cleavage and a synthetic peptide having the tethered ligand sequence (thrombin receptor agonist peptide or TRAP). We conducted a mutational analysis of extracellular residues of the receptor potentially involved in interaction with both the tethered ligand and the soluble peptide agonist. Agonist-stimulated calcium efflux in X. laevis oocytes or inositol phosphate accumulation in COS-7 cells was used to assess receptor activation. We have also examined the binding of a radiolabeled TRAP for the wild-type and mutant PAR-1 receptors. Our results indicated that most of the mutations strongly affected TRAP-induced responses without significantly altering thrombin-induced responses or TRAP binding. Several point mutations and deletion of extracellular domains (DeltaEC3, DeltaNH3) drastically altered the ability of mutant receptors to respond to TRAP, but not to thrombin, and did not affect the affinity for the radiolabeled TRAP by these mutant receptors. Only mutations that disrupted the putative disulfide bond or substitution of multiple acidic residues in the second extracellular loop by alanine had a significant effect on both ligand binding and thrombin activation. These results suggest that although both agonists can activate PAR-1, there are profound differences in the ability of thrombin and TRAP to activate PAR-1. In addition, we have found PAR-1 mutants with the ability to dissociate receptor-specific binding from functional activity.

Ligand cross-reactivity within the protease-activated receptor family.

Recently, a second member of the protease-activated receptor (PAR) family, named PAR-2, has been identified. Similar to the thrombin receptor, PAR-2 appears to be activated by proteolytic-mediated exposure of a "tethered ligand" sequence and can also be activated by the corresponding synthetic peptides. Similarities in the amino acid sequence of the receptors' tethered ligand sequences suggest that their respective agonist peptides might not be absolutely specific for their particular receptors. To test this, the receptor specificity of each agonist has been determined by measuring the responses of Xenopus oocytes expressing the thrombin receptor or PAR-2 to agonist peptides or enzymes. Thrombin receptors responded to thrombin, the human thrombin receptor-activating peptide SFLLRNP-NH2 (TRAP) (EC50 = 0.1 microM), and Xenopus TRAP, TFRIFD-NH2 (EC50 = 1 microM), but did not show any increase in calcium efflux over control levels with trypsin (50 nM) or PAR-2 agonist peptides (100 microM). Human and murine PAR-2 receptors responded comparably to human and murine PAR-2 agonist peptides (SLIGKVD and SLIGRL, respectively) (EC50 = 0.5-2.0 microM) and trypsin, but not to thrombin. PAR-2 was also found to be responsive to TRAP (EC50 = 1 microM) but was unresponsive to Xenopus TRAP (50 microM). Responses to additional peptide agonist analogs suggest that an amino-terminal serine is critical for PAR-2 agonist activity.

Characterization of recombinant human apoB-48-containing lipoproteins in rat hepatoma McA-RH7777 cells transfected with apoB-48 cDNA. Overexpression of apoB-48 decreases synthesis of endogenous apoB-100.

We studied the effect of overexpression of apolipoprotein (apo) B-48 on the synthesis and secretion of endogenous apoB-100 in rat hepatoma McA-RH7777 cell lines stably transfected with human apoB-48 cDNA under the control of the cytomegalovirus promoter. Three cell lines that secrete 40 to 60 ng human apoB.mg cell protein-1.h-1 were used. The recombinant human apoB-48 exhibited physicochemical characteristics (buoyant density, 1.06 to 1.21 g/mL; beta-electrophoretic mobility and diameters, 16 to 20 nm) indistinguishable from those of endogenous rat apoB-48. Overexpression of the recombinant human apoB-48 resulted in a 50% decrease in the secretion of endogenous apoB-100 but did not affect the secretion of apoE or apoA-I. Several possible mechanisms for the decreased secretion of apoB-100 were evaluated. First, recruitment of lipids into lipoproteins was shown to be unaffected since no major changes in the physicochemical properties of apoB-100-containing lipoproteins were observed. Second, the intracellular degradation of apoB-100 was not altered as the intracellular retention half-time and secretion efficiency remained unaffected by apoB-48 overexpression. Third, the posttranslational regulatory mechanisms for apoB-100 remained normal, as demonstrated by a twofold increase in apoB-100 secretion after supplementation with oleic acid. Unexpectedly, a 35% to 50% decrease in the steady-state synthesis of endogenous apoB-100 was observed in apoB-48-transfected cells compared with control cells. These data suggested that decreased secretion of apoB-100 was secondary to decreased synthesis. The decreased apoB-100 synthesis was not due to decreased steady-state levels of rat apoB-100 mRNA. These results suggest that overexpression of recombinant human apoB-48 may interfere with posttranscriptional events, possibly at the translation-translocation level, and decrease translational yield of apoB-100. These posttranscriptional events prior to the complete synthesis of the apoB-100 polypeptide can be important in the control of apoB-100 secretion.

The anion-binding exosite is critical for the high affinity binding of thrombin to the human thrombin receptor.

The thrombin receptor has been shown to be a novel member of the family of G-protein coupled receptors (Vu, T.-K. H., Hung, D.T., Wheaton, V.I., and Coughlin, S.R. (1991) Cell 64, 1057-1068). This receptor appears to be activated through a thrombin-mediated proteolytic mechanism which exposes a "tethered ligand" responsible for receptor activation. In order to investigate the initial interactions of thrombin with this receptor, we have constructed cell lines which express high levels of the human thrombin receptor and studied the binding of various forms of thrombin to the cell surface. Analysis of transfected cells with thrombin receptor monoclonal antibodies identified a particular cell line (clone #5-18) which displayed > 150,000 thrombin receptors per cell. Clone #5-18 appeared to express functional receptors since treatment with thrombin resulted in both a 15-20 fold increase of cytoplasmic phosphoinositide levels and a comparable shift in the EC50 of thrombin-mediated calcium mobilization when compared to non-transfected CHO cells. Binding of 125I-alpha-thrombin to clone #5-18 did not reach equilibrium at 37 degrees C. However, direct binding studies of 125I-alpha-, 125I-diisopropylphospho (DIP)-alpha-, and 125I-beta-thrombin to clone #5-18 demonstrated that binding at 4 degrees C was saturable and reversible for each ligand. Analysis of the binding data revealed Kd's of 0.8 nM, 0.7 nM and 9.7 nM for 125I-alpha-, 125I-DIP-alpha- and 125I-beta-thrombin respectively. Association of 125I-alpha-, DIP-alpha, and beta-thrombin could be competed by unlabelled alpha- and DIP-alpha-thrombin. Unlabelled beta-thrombin, which has a modified anion-binding exosite, was a poor competitor for 125I-alpha- and 125I-DIP-alpha-thrombin, but did compete for 125I-beta-thrombin. In addition, the hirudin54-65 peptide competed at submicromolar concentrations for the binding of alpha- and DIP-alpha-thrombin, but not for beta-thrombin. This peptide binds specifically at the anion-binding exosite of alpha-thrombin and has been shown to have a lower affinity for beta-thrombin. These results demonstrate directly a high affinity interaction between thrombin and its receptor, and suggest that an important component is the high affinity association of the thrombin receptor with the anion-binding exosite of thrombin.

Elimination of apolipoprotein B48 formation in rat hepatoma cell lines transfected with mutant human apolipoprotein B cDNA constructs.

Rat hepatoma McA-RH7777 cell lines transfected with full-length human apolipoprotein (apo) B constructs produce mostly human apoB48 and only small amounts of apoB100, as a result of mRNA editing at codon 2153 (C to U conversion at nucleotide 6666). To abolish the formation of apoB48 and increase the yield of apoB100 and other forms of apoB longer than apoB48, site-specific mutations were introduced at or near the site of apoB mRNA editing. Among four mutations examined, only that in which codon 2153 was converted from CAA (Gln) to CTA (Leu) effectively precluded the formation of apoB48. In this mutant, a stop codon would not be generated even if the C to U conversion occurred. The three other mutations were introduced to disrupt the proposed stem-loop structure encompassing the editing site. Changes made in the third positions of five codons on the 5' side of the edited base or of four codons 3' of the edited base failed to eliminate the production of a protein with the approximate size of apoB48. A construct in which codon 2153 was changed from CAA to GAT (Asp) also failed to eliminate the production of a protein the size of apoB48. Analysis of the region between nucleotides 6200 and 6900 of the cDNA did not detect any prevalent alternate editing sites. Immunoblot analysis using polyclonal antibodies raised against synthetic peptides of human apoB100 indicated that the carboxyl terminus of the apoB48-like proteins probably resides between amino acid residues 2068 and 2129 of apoB100. These results provide some insight into the mechanism of apoB mRNA editing and will facilitate further studies on apoB-containing lipoproteins.

Comparative analysis of sequences at the 5' end of the human and mouse apolipoprotein B genes.

A comparison was made between the DNA sequences in two regions of the mouse and the human apolipoprotein B genes: the 5'-flanking sequence and the region between the first exon and the second intron. Considerable homology was observed, particularly in the immediate 5' region and in the second intron. Because promoter and enhancer elements have been previously localized to these regions in the human apolipoprotein B gene, it is proposed that regions of conserved base sequence delineate binding regions for regulatory proteins. In some cases, contiguous regions of homology are longer than expected for regions designed as recognition sites for individual nuclear proteins, and may define regions recognizable by a cluster of interacting proteins. Both the human and mouse genes contain repetitive elements and a hypervariable dinucleotide repeat.

Characterization of tissue-specific enhancer elements in the second intron of the human apolipoprotein B gene.

We report the identification and characterization of tissue-specific transcriptional enhancer elements that influence the expression of the human apolipoprotein B gene. A 704-base pair PstI fragment comprising sequences from the first and second introns of the human apolipoprotein B gene (positions +360 to +1064) possesses tissue-specific transcriptional enhancer elements when assayed in transient transfection experiments using either the apolipoprotein B or thymidine kinase promoter. The majority of the enhancer activity, which was observed in transcriptionally active HepG2 and CaCo-2 cells, but not in transcriptionally inactive Chinese hamster ovary or HeLa cells, was subsequently localized to a 443-base pair SmaI-PvuII fragment (positions +621 to +1064) within the second intron of the apolipoprotein B gene. Gel retention experiments demonstrated that sequence motifs within this region interact with a number of nuclear proteins from HepG2, CaCo-2, and HeLa cells. The actual sequence elements that bound to nuclear proteins from HepG2 cells were identified by DNase I footprinting. Deletion experiments were performed to distinguish those protein-binding regions involved in the enhancer effect. Our data demonstrate that sequences between positions +806 and +940 are essential for this enhancer activity. This segment contains one large 97-base pair footprint, whose sequence has been conserved between the human and mouse genes. Binding sites for the liver-specific transcription factors HNF-1 and HNF-3 are present within this footprint.

Expression of carboxyl-terminally truncated forms of human apolipoprotein B in rat hepatoma cells. Evidence that the length of apolipoprotein B has a major effect on the buoyant density of the secreted lipoproteins.

We examined the relationship between the size of human apolipoprotein (apo) B and the formation and secretion of apoB-containing lipoprotein particles. Stable transformants of the rat hepatoma cell line McA-RH7777 harboring a variety of human apoB cDNA constructs were established, and these produced carboxyl-terminally truncated apoB proteins (apoB18, -B23, -B28, -B31, -B48, and -B53). Immunoblotting of apoB proteins secreted into the culture medium and fractionated by equilibrium density ultracentrifugation revealed that each of the truncated apoB species was secreted from the cells. The peak densities of the apoB-containing particles decreased as the length of the apoB proteins increased. Apolipoproteins B18 and B23 appeared at the bottom of the salt gradient (d = 1.23 g/ml), whereas particles containing apoB28, -B31, -B37, -B48, and -B53 exhibited progressive decreases in density. The density distribution of secreted apolipoproteins was not affected by the expression or secretion of these recombinant apoB species. As determined by nondenaturing gel electrophoresis, apoB28, -B31, -B37, -B48, and -B53 formed their own discrete particles, and there was a direct correlation between the size of the particles and the length of the apoB species. The efficiency and rate of secretion of these truncated forms of apoB were studied by measuring the decrease of immunoprecipitated 35S-labeled apoB proteins in the cells and their accumulation in the medium. Proteins corresponding to apoB28 or larger were rapidly and efficiently secreted, whereas apoB18 and apoB23 were secreted much more slowly. These data imply that the size of these truncated apoB forms governs the lipid content of the apoB-containing lipoproteins formed as well as the kinetics of secretion.

A polymorphism in a region with enhancer activity in the second intron of the human apolipoprotein B gene.

A 443-base pair fragment (+622 to +1064) from the second intron of the human apolipoprotein B gene was shown to contain a tissue-specific enhancer when placed in front of an apolipoprotein B promoter-chloramphenicol acetyltransferase construct in transfection experiments. To identify potential regulatory mutations in this region of the gene, DNA from various subjects was examined for the presence of point mutations by means of chemical cleavage of mismatched heteroduplexes. An A----G substitution within the second intron of the gene at position +722 was identified in three unrelated subjects and confirmed by DNA sequencing. Although the base substitution was contained within a nuclear protein-binding site, as determined by DNase I footprinting, it did not appear to affect the protein/DNA interaction in its vicinity, as shown by gel retardation experiments. The single base substitution at position +722 abolishes a StyI restriction site, thus creating a StyI polymorphism. Using allele-specific oligonucleotides, we screened the DNA of 172 subjects for the presence of this polymorphism: two other subjects carrying the polymorphism were found. In each of the five unrelated subjects, the polymorphism was associated with the same haplotype.

New variation on the translocation of proteins during early biogenesis of apolipoprotein B.

Apolipoprotein B (apo B) is crucial for the transport of cholesterol in humans. It is a large secretory protein that mediates the uptake of low-density lipoproteins and renders several forms of lipid droplets soluble in the blood. The binding of lipid by apo B also prevents this hydrophobic protein from precipitating in aqueous solution. In the endoplasmic reticulum, nascent secretory proteins must be translocated through an aqueous channel in the membrane into the aqueous lumen, so some novel form of processing may be necessary to maintain the solubility of apo B during its translocation. We have discovered that the biogenesis of apo B in cell-free systems does indeed involve a new variation on protein translocation: unlike typical secretory proteins, apo B is synthesized as a series of transmembrane chains with large cytoplasmic domains and progressively longer amino-terminal regions that are protected against added proteases during the translocation process. In contrast to typical transmembrane proteins, these transmembrane chains are not integrated into the bilayer. Moreover, the transmembrane chains with the shortest protected domains are precursors of forms whose protection is progressively extended to cover the length of the protein. This stepwise conversion occurs post-translationally for the most part. We propose a model on the basis of these findings for the biogenesis of apo B.

An expression system for human apolipoprotein B100 in a rat hepatoma cell line.

We have designed an in vitro expression system for human apolipoprotein (apo) B. A full-length human apoB minigene was constructed from cDNA and genomic apoB clones and inserted into a vector where its expression was directed by the cytomegalovirus promoter. The apoB minigene was expressed in a rat hepatoma cell line, McA-RH7777. Human apoB100, which is the ligand for the low density lipoprotein receptor, was secreted in low density lipoprotein or very low density lipoprotein particles, depending on the composition of the medium. A protein with the mobility of apoB48, a structurally related protein involved in cholesterol metabolism, was also produced from the human apoB minigene. This in vitro expression system for human apoB will enable investigators to identify which domains of this protein are involved in processes such as lipoprotein assembly and secretion. This system should also allow investigators to identify definitively the domain in apoB that enables the protein to bind to the low density lipoprotein receptor.

DNase I- and micrococcal nuclease-hypersensitive sites in the human apolipoprotein B gene are tissue specific.

We have mapped the DNase I- and micrococcal nuclease-hypersensitive sites present in the 5' end of the human apolipoprotein B (apo-B) gene in nuclei from cells expressing or not expressing the gene. Four DNase I-hypersensitive sites were found in nuclei from liver-derived HepG2 cells and intestine-derived CaCo-2 cells, which express the apo-B gene, but not in HeLa cells, which do not. These sites are located near positions -120, -440, -700, and +760 base pairs relative to the transcriptional start site. Undifferentiated CaCo-2 cells exhibited another site, near position -540. Six micrococcal nuclease-hypersensitive sites were found in nuclei from HepG2 and CaCo-2 cells, but not in HeLa cells or free DNA. These sites are located near positions -120, -390, -530, -700, -850, and +210. HepG2 cells exhibited another site, near position +460. Comparison of the DNA sequence of the 5' flanking regions of the human and mouse apo-B genes revealed a high degree of evolutionary conservation of short stretches of sequences in the immediate vicinity of each of the DNase I- and most of the micrococcal nuclease-hypersensitive sites.

DNA sequence of the human apolipoprotein B gene.

The sequence of the human apolipoprotein B gene comprises 43 kb divided into 29 exons, one of which is unusually long and contains 7572 bp. Comparison of the gene sequence with four complete and three partial cDNA sequences published elsewhere reveals a total of 60 nucleotide substitutions and 39 amino acid substitutions and one small deletion in the signal peptide.

Structure of the human apolipoprotein B gene.

Human apolipoprotein B100 cDNA is 14 kilobases in length and encodes a 4563-amino acid precursor protein. The corresponding human gene has been isolated as a series of overlapping lambda clones and extends over 43 kilobases. The gene comprises 29 exons and 28 introns. The distribution of introns is extremely asymmetrical, most of them appearing in the 5'-terminal one-third of the gene. Although most of the exons fall within the normal size limits for mammalian genes, two are unusually long: 1906 and 7572 base pairs. The latter exon is by far the longest reported for a vertebrate gene.

Analysis of the products of the Myxococcus xanthus frz genes.

The frizzy (frz) genes of Myxococcus xanthus control the ability of cells to reverse direction of gliding motility. The orientation of the frz genes was studied by isolating transcriptional fusions with the transposon derivative Tn5-lac. The frz genes were then cloned in the proper orientation in an expression vector. By using maxicell experiments, we were able to identify several labeled bands which were plasmid encoded. To identify the labeled proteins and their respective genes, we constructed deletion plasmids in which various regions of the insert DNA had been removed. The plasmid-encoded proteins were then labeled in maxicell experiments, and the bands which correspond to the frzCD, frzE, and frzF gene products were identified. The sizes of the gene products agreed with the genetic and physical map of the cloned DNA.

"Frizzy" genes of Myxococcus xanthus are involved in control of frequency of reversal of gliding motility.

Myxococcus xanthus, a Gram-negative bacterium, has a complex life cycle that includes fruiting body formation. Frizzy (frz) mutants are unable to aggregate normally, instead forming frizzy filamentous aggregates. We have found that these mutants are defective in the control of cell reversal during gliding motility. Wild-type cells reverse their direction of gliding about every 6.8 min; net movement occurs since the interval between reversals can vary widely. The frzA-C, -E and -F mutants reverse their direction of movement very rarely, about once every 2 hr. These mutants cannot aggregate normally and give rise to frizzy filamentous colonies on fruiting agar or motility agar. In contrast, frzD mutants reverse their direction of movement very frequently, about once every 2.2 min; individual cells show little net movement and form smooth-edged "nonmotile" type colonies. Genetic analysis of the frzD locus shows that mutations in this locus can be dominant to the wild-type allele and that its gene product(s) must interact with the other frz gene products. Our results suggest that the frz genes are part of a system responsible for directed movement of this organism.

Cloning and complementation analysis of the "Frizzy" genes of Myxococcus xanthus.

Fruiting-body formation in Myxococcus xanthus involves the aggregation of cells into raised mounds, where they sporulate. "Frizzy" mutants fail to aggregate into mounds, but rather aggregate into "frizzy" filaments (D.R. Zusman 1982). The frizzy mutations (frz) were found to be genetically linked. The region of DNA carrying the frz genes was cloned in Escherichia coli by selecting for the kanamycin resistance element present on a transposon Tn5 insertion linked to the frz genes. Phage P1 mediated transduction of the cloned DNA into M. xanthus frizzy mutants showed that the cloned DNA could complement the frz mutations. The cloned DNA was analyzed by isolating and characterizing new Tn5 insertions at short intervals within the M. xanthus DNA and by constructing in vitro deletions. The mutated DNA was then transduced into M. xanthus where the cloned DNA became integrated into the bacterial chromosome as gene replacements or as merodiploids. The gene replacement strains allowed us to define the limits of the frz region, since Tn5 insertions in the frz genes resulted in the frizzy phenotype. The merodiploid strains allowed us to perform complementation analyses. Using appropriate crosses, we were able to identify 5-6 frz complementation groups on 7.5 kb of cloned DNA. One of the complementation groups was separated from the others by 1.4 kb of DNA, whereas the others were contiguous. The different frz loci behave as separate transcriptional groups although interactions between some of the gene products are indicated.