Mammalian protein secretion without signal peptide removal. Biosynthesis of plasminogen activator inhibitor-2 in U-937 cells

Plasminogen activator inhibitor-2 (PAI-2) is a serine protease inhibitor that regulates plasmin generation by inhibiting urokinase and tissue plasminogen activator. The primary structure of PAI-2 suggests that it may be secreted without cleavage of a single peptide. To confirm this hypothesis we have studied the glycosylation and secretion of PAI-2 in human monocytic U-937 cells by metabolic labeling, immunoprecipitation, glycosidase digestion, and protein sequencing. PAI-2 is variably glycosylated on asparagine residues to yield intracellular intermediates with zero, one, two, or three high mannose-type oligosaccharide units. Secretion of the N-glycosylated species began by 1 h of chase and the secreted molecules contained both complex-type N-linked and O-linked oligosaccharides. Enzymatically deglycosylated PAI-2 had an electrophoretic mobility identical to that of the nonglycosylated precursor and also to that of PAI-2 synthesized in vitro in a rabbit reticulocyte lysate from synthetic mRNA derived from full length PAI-2 cDNA. The amino-terminal protein sequence of secreted PAI-2 began with the initiator methionine residue. These results indicate that PAI-2 is glycosylated and secreted efficiently without the cleavage of a signal peptide. PAI-2 shares this property with its nearest homologue in the serine protease inhibitor family, chicken ovalbumin, and appears to be the first well characterized example of this phenomenon among natural mammalian proteins.     

cDNA cloning and expression in Escherichia coli of a plasminogen activator inhibitor from human placenta

Two nearly full-length cDNAs for placental plasminogen activator inhibitor (PAI) have been isolated from a human placenta lambda gt11 cDNA library. One positive (lambda PAI-75.1) expressed a protein that could adsorb and purify anti-PAI antibodies. The expressed protein inhibited the activity of human urokinase in a fibrin autography assay, and formed a 79-kDa (reduced) covalent complex with 125I-urokinase that could be immunoprecipitated with anti-PAI. The cDNA insert of the longer isolate (lambda PAI-75.15) consisted of 1909 base pairs, including a 5'-noncoding region of 55 base pairs, an open reading frame of 1245 base pairs, a stop codon, a 3'-noncoding region of 581 base pairs, and a poly(A) tail. The size of the mRNA was estimated to be 2.0 kilobases by Northern blot analysis. The translated amino acid sequence consisted of 415 amino acids, corresponding to a 46.6-kDa protein. The sequence was related to members of the serpin gene family, particularly ovalbumin and the chicken gene Y protein. Like these avian proteins, placental PAI appears to lack a cleavable NH2-terminal signal peptide. Residues 347-376 were identical to the partial sequence reported recently for a PAI isolated from the human monocytic U-937 cell line. Placental PAI mRNA was apparently expressed at low levels in human umbilical vein endothelial cells, but was not detectable in HepG2 hepatoma cells. It was present in U-937 cells and was inducible at least 10-fold by phorbol 12-myristate 13-acetate. Thus placental PAI is a unique member of the serpin gene family, distinct from endothelial-type PAI. It is probably identical to monocyte-macrophage PAI.     

Isolation of a cDNA clone for human antithrombin III

Antithrombin III (ATIII) is an important plasma protease inhibitor with a central role in the coagulation system. On the basis of its protein sequence, ATIII is one member of a "super family" of protease inhibitors that includes alpha 1-antitrypsin and chicken ovalbumin. An increased risk of thromboembolism is associated with inherited ATIII deficiency. To study the structure and expression of the human ATIII gene, we have isolated complementary (cDNA) clones for ATIII from human liver mRNA. ATIII cDNA clones were identified by hybridization to a mixture of synthetic oligodeoxynucleotides encoding amino acids 251-256 of the ATIII protein sequence. The largest cDNA clone (1.4 kilobases) included the coding region of ATIII mRNA from codon 10 through a 3'-untranslated region. Comparison of ATIII cDNA clones from two different sources revealed a sequence polymorphism at an internal PstI restriction site. Analysis of both total genomic DNAs and an ATIII gene cloned in a bacteriophage Charon 4A showed that the ATIII gene is present once per haploid genome and is distributed over 10-16 kilobases of DNA. Computer-assisted comparison of the cDNA sequence with those for baboon alpha 1-antitrypsin and chicken ovalbumin revealed homologies consistent with their inclusion in the protease inhibitor superfamily.     

A chicken middle-repetitive DNA sequence which shares homology with mammalian ubiquitous repeats.

We have identified and sequenced two members of a chicken middle repetitive DNA sequence family. By reassociation kinetics, members of this family (termed CRl) are estimated to be present in 1500-7000 copies per chicken haploid genome. The first family member sequenced (CRlUla) is located approximately 2 kb upstream from the previously cloned chicken Ul RNA gene. The second CRl sequence (CRl)Va) is located approximately 12 kb downstream from the 3' end of the chicken ovalbumin gene. The region of homology between these two sequences extends over a region of approximately 160 base pairs. In each case, the 160 base pair region is flanked by imperfect, but homologous, short direct repeats 10-15 base pairs in length. When the CRl sequences are compared with mammalian ubiquitous interspersed repetitive DNA sequences (human Alu and Mouse Bl families), several regions of extensive homology are evident. In addition, the short nucleotide sequence CAGCCTGG which is completely conserved in ubiquitous repetitive sequence families from several mammalian species is also conserved at a homologous position in the chicken sequences. These data imply that at least certain aspects of the sequence and structure of these interspersed repeats must predate the avian-mammalian divergence. It seems that the CRl family may possibly represent an avian counterpart of the mammalian ubiquitous repeats.     

The ovalbumin gene family: complete sequence and structure of the Y gene.

The "ovalbumin Y" gene, one of three which constitute the ovalbumin gene family in chicken has been completely sequenced. The exact location of exons can be derived from the comparison with the ovalbumin gene sequence and from the map previously established by electron microscopy analysis. During evolution of the Y gene, selective pressure has operated to retain a sequence coding for an ovalbumin-like protein. The location of splice junctions, the length of protein coding exons and the reading phase are as in the ovalbumin gene. The overall homology between the Y and ovalbumin protein coding sequences is 72.6% (resulting in a 58% homology for the amino acid sequences). A significantly high number of base changes within coding sequences are present in clusters, which appear in several cases to be correlated with the occurrence of direct repeats. The 3' untranslated sequences of the Y and ovalbumin mRNAs have diverged much more, and the Y sequence contains a peculiar U(T) rich region. Corresponding introns of the ovalbumin and Y genes differ extensively both in sequence and in length. They share however characteristic biases in their base distribution.     

Functional activity of eukaryotic signal sequences in Escherichia coli: the ovalbumin family of serine protease inhibitors

It is widely assumed that the functional activity of signal sequences has been conserved throughout evolution, at least between Gram-negative bacteria and eukaryotes. The ovalbumin family of serine protease inhibitors (serpins) provides a unique tool to test this assumption, since individual members can be secreted (ovalbumin), cytosolic (leukocyte elastase inhibitor, LEI), or targeted to both compartments (plasminogen activator inhibitor 2, PAI-2). The facultative secretion of PAI-2 is mediated by a signal sequence proposed to be inefficient by design. We show here that the same internal domain that promotes an inefficient translocation of murine PAI-2 in mammalian cells is a weak signal sequence in Escherichia coli. In contrast, the ovalbumin signal sequence is much more efficient, whereas the corresponding sequence elements from LEI, maspin and PI-10 are entirely devoid of signal sequence activity in E.coli. Mutations that improve the activity of the PAI-2 signal sequence and that convert the N-terminal regions of maspin and PI-10 into efficient signal sequences have been characterized. Taken together, these results indicate that several structural features contribute to the weak activity of the PAI-2 signal sequence and provide new insights into the plasticity of the "hydrophobic core" of signal sequences. High-level expression of two chimeric proteins containing the PAI-2 signal sequence is toxic, and the reduced viability is accompanied by a rapid decrease in the membrane proton motive force, in ATP levels and in translation. In unc- cells, which lack the F0F1 ATP-synthase, the chimeric proteins retain their toxicity and their expression only affected the proton motive force. Thus, the properties of these toxic signal sequences offer a new tool to dissect the interactions of signal sequences with the protein export machinery.     

Introns of the chicken ovalbumin gene promote nucleosome alignment in vitro.

A defined in vitro chromatin assembly system was used to examine the nucleosome alignment induced by histone H5 throughout a 12 kilobase pair chicken genomic DNA fragment containing the ovalbumin gene. In contrast with total fragmented chicken DNA and several anonymous cloned fragments, much of the gene permitted histone H5 to space nucleosomes at physiological intervals in an extended array. Nucleosomes at the 3'-end of the gene and on approximately 4 kilobase pairs of 5'-flanking ovalbumin sequence did not become aligned to appreciable extents. Analysis of cloned 2-3 kilobase pair subfragments suggested that a strong nucleosome alignment signal, specifying a 196 +/- 5 base pair repeat exists in intron E. A second discrete region of the gene, which mapped approximately to intron A, exhibited nucleosome alignment with a spacing periodicity of about 200 base pairs. The ovalbumin cDNA did not permit nucleosome alignment. These findings suggest that some of the introns contain signals that direct nucleosome alignment over the ovalbumin gene in a way conducive to its regulation.     

The ovalbumin gene family: Structure of the X gene and evolution of duplicated split genes

The X, Y and ovalbumin genes, which are found within a 40 kb region of the chicken genome, are all expressed in oviduct under steroid hormone control, and share some sequence homologies. We have now cloned the complete X gene and have analyzed its structure. It codes for two RNA species, X and X′; both are coded by eight exons and appear to differ only by the size of their 3′ untranslated region, X′ RNA being 1400 nucleotides longer than X RNA. The striking similarity in the number and length of the exons which constitute the X, Y or ovalbumin genes establishes that they have evolved from a common ancestor gene by duplication events. Comparison of selected regions of the X and ovalbumin genes indicates that the exon sequences coding for protein and the location of the splice junctions have been well-conserved. The introns and the 3′ untranslated exonic sequences have diverged much more rapidly. Four regions of apparently unrelated repetitive sequences are found both outside the X gene and within it (in two introns and in the sequence coding for the 3′ untranslated part of X′RNA). The intragenic repetitive sequences have no counterpart in the ovalbumin and Y genes.     

The nucleotide sequence of the chicken thymidine kinase gene and the relationship of its predicted polypeptide to that of the vaccinia virus thymidine kinase.

The entire DNA nucleotide sequence of a 3.0 kilobase pair Hind III fragment containing the chicken cytoplasmic thymidine kinase gene was determined. Oligonucleotide linker insertion mutations distributed throughout this gene and having known effects upon gene activity ( Kwoh , T.J., Zipser , D., and Wigler , M. 1983. J. Mol. Appl. Genet. 2, 191-200), were used to access regions of the Hind III fragment for sequencing reactions. The complete nucleotide sequence, together with the positions of the linker insertion mutations within the sequence, allows us to propose a structure for the chicken thymidine kinase gene. The protein coding sequence of the gene is divided into seven small segments (each less than 160 base pairs) by six small introns (each less than 230 base pairs). The proposed 244 amino acid polypeptide encoded by this gene bears strong homology to the vaccinia virus thymidine kinase. No homology with the thymidine kinases of the herpes simplex viruses was found.