Isolation and characterization of cDNA clones corresponding to two different human apoC-III alleles

We have recently reported that the human apolipoprotein A-I (apoA-I) and apolipoprotein C-III (apoC-III) genes are physically linked and that the presence of a DNA insertion in the apoA-I gene is correlated with apoA-I-apoC-III deficiency in patients with premature atherosclerosis. In addition, the presence of a polymorphic restriction endonuclease site (SacI) in the 3' noncoding region of apoC-III mRNA has been correlated with hypertriglyceridemia in humans. In this study, we report the isolation and characterization of cDNA clones containing the entire apoC-III mRNA coding sequence. The nucleotide-derived apoC-III amino acid sequence indicates that the apoC-III primary translational product contains a 20 amino acid N-terminal extension, which conforms with the general properties of known signal peptides, and is highly homologous to the recently reported rat apoC-III signal peptide. The DNA-derived apoC-III amino acid sequence differs from the previously reported apoC-III amino acid sequence at four amino acid residues. More specifically, at positions +32, +33, +37, +39, the DNA sequence predicts Glu, Ser, Gln, Ala, respectively, while the previously reported sequence specifies Ser, Gln, Ala, Gln, respectively. Finally, isolation and characterization of apoC-III cDNA clones, with or without the polymorphic SacI restriction site, indicated that the apoC-III nucleotide sequence corresponding to the Sac+ and Sac- clones differs at three nucleotide sites; however, the amino acid sequence specified by the Sac+ and Sac- alleles is identical.     

Nucleotide sequence of transforming human c-sis cDNA clones with homology to platelet-derived growth factor.

Three c-sis cDNA clones were obtained from polyadenylated RNA of a human T-cell lymphotropic virus (HTLV) type I transformed cell line. Two clones, designated pSM-1 and pSM-2, have cDNA inserts of 2498 and 2509 base pairs (bp), respectively, excluding the sizes of the guanylate tails, and the polyadenylate tracts. These clones are shorter than the estimated size of the c-sis mRNA of 4200 bp. Both of these clones can transform NIH 3T3 cells. The third clone, designated pSM-3 has a cDNA insert of 1421 bp and lacks transforming activity. The sequence of clone pSM-1 reveals a single long open reading frame (nucleotides 118-840) encoding chain A of platelet-derived growth factor, and two segments with homology to v-sis (nucleotides 182-871 and 1021-1325). Sequence homology is noted in the 3' untranslated region to the corresponding regions of the beta 1 interferon (IFN), human and murine beta-nerve growth factor (NGF), human interleukin 2 (IL2) genes, and tubulin pseudogenes. However, no typical AATAAA polyadenylation signal is present. An alternating (dCdA)n X (dGdT)n sequence is present in the 3' flanking cellular sequences similar to those in the corresponding position of the human proenkephalin gene, in the first intron of the gamma-IFN gene, and the second intron of the beta-NGF gene.     

Fibronectin and vitronectin regulate the organization of their respective Arg-Gly-Asp adhesion receptors in cultured human endothelial cells

Human umbilical vein endothelial cells (ECs) adhere in vitro to proteins of the extracellular matrix including fibronectin (fn) and vitronectin (vn). Specific receptors for fn and vn have been previously characterized. These receptors belong to a family of membrane glycoproteins characterized (a) by being a transmembrane complex of two noncovalently linked subunits and (b) by recognizing the tripeptide Arg-Gly-Asp on their respective ligands. In this paper we investigated how vn and fn control the organization of their respective receptors over the surface of ECs. It was found that the clustering of individual receptors and the organization thereafter of focal contacts occurred only when ECs were exposed to the specific ligand and did not occur on the opposite ligand. The shape of receptor clusters was slightly different and a colocalization of the two receptors was found when ECs were cultured on a mixed matrix of fn plus vn. Adhesion was selectively inhibited by vn or fn receptor antibodies on their respective substrates. The clustering of both receptors preceded the association of vinculin with focal contacts and stress fiber formation. Also, the vn receptor, in the absence of associated fn receptor, was capable of inducing the organization of the membrane-microfilament interaction complex. Overall, these results indicate that individual matrix ligands induce only the clustering of their respective membrane receptors. The clustering of only one receptor is capable of supporting the subsequent formation of focal contacts and the local assembly of related cytoskeletal proteins.     

The extensive homology between mRNA sequences of normal and SV40-transformed human fibroblasts.

The poly(A)-containing messenger RNA of normal diploid fibroblast and SV40-transformed progeny cells are compared by cross-hybridizing cDNA. We find a high degree of homology between the mRNA from normal and transformed cells. Despite imperfections in the procedure, the technique permits the conclusion that, at most, 3% of the mRNA in the transformed cell has sequences not present in the normal parental cell. Furthermore, much of the difference appears to occur in low and intermediate complexity classes of mRNA molecules. Extension homology in the mRNA sequences of disparate cell lines may be a general phenomenon, and even HeLa cell mRNA is nearly identical to that of diploid human fibroblasts.     

Isolation and sequence determination of cDNA clones for porcine and human lipoamide dehydrogenase. Homology to other disulfide oxidoreductases

A 2.3-kilobase cDNA clone encoding lipoamide dehydrogenase was isolated from a porcine adrenal medulla library in the vector pCD by screening with four synthetic oligonucleotide probes corresponding to amino acid sequence from tryptic peptides of porcine lipoamide dehydrogenase. A 450-bp fragment of the porcine cDNA was used to screen a human small cell lambda gt10 library at reduced stringency. Overlapping human cDNA clones of various lengths were isolated, the largest of which was again 2.3 kilobases in length. Sequencing of both porcine and human cDNAs revealed a short 5'-untranslated region followed by 1530-bp of coding region and 700 bp of 3'-untranslated region preceding a poly(A) tail. The porcine cDNA displayed coding regions corresponding to the known tryptic peptides and a 35-amino acid leader sequence involved in targeting of the protein to the mitochondria. The human lipoamide dehydrogenase cDNA is 96% identical to the porcine at the amino acid level. Alignment of the deduced amino acid sequence of human lipoamide dehydrogenase with human erythrocyte glutathione reductase and mercuric reductase from Tn501 revealed extensive homologies throughout the primary sequence, suggesting that secondary and tertiary structure is also similar among these three enzymes.     

Insect immunity. Isolation of cDNA clones corresponding to attacins and immune protein P4 from Hyalophora cecropia

Diapausing pupae of the Cecropia moth (Hyalophora cecropia) respond to an injection of live bacteria by the selective synthesis of certain types of RNA and immune proteins (designated P1-P9). The in vitro translation products of RNA from both injured and infected pupae showed specific patterns with a defined number of extra bands. Some proteins characteristic of the normal RNA were reduced in the immune RNA translation products. Antibody reaction was used to show the selective synthesis of immune proteins P4 and P5 with mRNA from pupae subjected to injury or infection. The protein synthesized in vitro, which cross-reacted with P5 antibodies, is most likely a precursor of the attacins described in the preceding paper. A cDNA clone bank was prepared and two clones were isolated and shown to contain 750 bp corresponding to P4 and 250 bp of attacin information. These clones were used to estimate the sizes of the mRNAs by Northern blotting and to estimate, by RNA/DNA hybridization, the levels of P4 and P5 mRNA. In vivo incorporation of [35S]methionine into attacins and P4 during different conditions was compared with the levels of the corresponding mRNA.     

Insect immunity. Isolation and sequence of two cDNA clones corresponding to acidic and basic attacins from Hyalophora cecropia

The Cecropia moth has three known classes of antibacterial immune proteins, attacins, lysozyme and cecropins (earlier referred to as P5, P7 and P9, respectively). Six attacins with different isoelectric points have been purified. The N-terminal sequences for five of these forms imply that only two different genes exist. We have now isolated and sequenced two cDNA clones, one for the basic attacin and one for the acidic form. The two mature proteins show 76% homology at the nucleotide level, while the regions beyond the stop codons are 36% homologous. The differences in the content of aspartic acid accounts for the difference in net charge between the acidic and basic attacin. Further differences in charge can be obtained by post-translational removal of a lysine-containing tetrapeptide at the C-terminal end of the two proteins. Evidence for a prepro form of the basic attacin is presented.     

Isolation of cDNA clones encoding an enzyme from bovine cells that repairs oxidative DNA damage in vitro: homology with bacterial repair enzymes.

Ionizing radiation and radiomimetic compounds, such as hydrogen peroxide and bleomycin, generate DNA strand breaks with fragmented deoxyribose 3' termini via the formation of oxygen-derived free radicals. These fragmented sugars require removal by enzymes with 3' phosphodiesterase activity before DNA synthesis can proceed. An enzyme that reactivates bleomycin-damaged DNA to a substrate for Klenow polymerase has been purified from calf thymus. The enzyme, which has a Mr of 38,000 on SDS-PAGE, also reactivates hydrogen peroxide-damaged DNA and has an associated apurinic/apyrimidinic (AP) endonuclease activity. The N-terminal amino acid sequence of the purified protein matches that reported previously for a calf thymus enzyme purified on the basis of AP endonuclease activity. Degenerate oligonucleotide primers based on this sequence were used in the polymerase chain reaction to generate from a bovine cDNA library a fragment specific for the 5' end of the coding sequence. Using this cDNA fragment as a probe, several clones containing 1.35 kb cDNA inserts were isolated and the complete nucleotide sequence of one of these determined. This revealed an 0.95 kb open reading frame which would encode a polypeptide of Mr 35,500 and with a N-terminal sequence matching that determined experimentally. The predicted amino acid sequence shows strong homology with the sequences of two bacterial enzymes that repair oxidative DNA damage, ExoA protein of S. pneumoniae and exonuclease III of E. coli.     

Characterization of cDNA clones for human myeloperoxidase: predicted amino acid sequence and evidence for multiple mRNA species.

Myeloperoxidase is a component of the microbicidal network of polymorphonuclear leukocytes. The enzyme is a tetramer consisting of two heavy and two light subunits. A large proportion of humans demonstrate genetic deficiencies in the production of myeloperoxidase. As a first step in analyzing these deficiencies in more detail, we have isolated cDNA clones for myeloperoxidase from an expression library of the HL-60 human promyelocytic leukemia cell line. Two overlapping plasmids (pMP02 and pMP062) were identified as myeloperoxidase cDNA clones based on the detection with myeloperoxidase antiserum of 70 kDa protein expressed in pMP02-containing bacteria and a 75 kDa polypeptide produced by hybridization selection and translation using pMP062 and HL-60 RNA. Formal identification of the clones was made by matching the predicted amino acid sequences with the amino terminal sequences of the heavy and light subunits. Both subunits are encoded by one mRNA in the following order: pre-pro-sequences--light subunit--heavy subunit. The molecular weight of the predicted primary translation product is 83.7 kDa. Northern blots reveal two size classes of hybridizing RNAs (approximately 3.0-3.3 and 3.5-4.0 kilobases) whose expression is restricted to cells of the granulocytic lineage and parallels the changes in enzymatic activity observed during differentiation.     

A peptide corresponding to GPIIb alpha 300-312, a presumptive fibrinogen gamma-chain binding site on the platelet integrin GPIIb/IIIa, inhibits the adhesion of platelets to at least four adhesive ligands.

The platelet fibrinogen (Fg) receptor (GPIIb/IIIa) is an integrin which plays a critical role in hemostasis by recognizing at least the four adhesive ligands: Fg, fibronectin (Fn), vitronectin (Vn), and von Willebrand factor (vWf). We reported that residues 309-312 of GPIIb alpha appear to comprise at least part of a Fg binding site on the Fg receptor (Gartner, T. K., and Taylor, D. B. (1990) Thromb. Res. 60, 291-309). Here we report that the peptide GPIIb alpha 300-312 (G13) inhibits platelet aggregation and binds Fg and Vn. Significantly, this peptide inhibits the adhesion of stimulated platelets to Fg, Fn, Vn, and vWf, but not the adhesion of resting platelets to Fn. Thus, GPIIb 300-312 may constitute a specific but common recognition site on GPIIb/IIIa for both LGGAKQAGDV- and RGD-containing ligands.     

Cloned cytochrome P-450 cDNA. Nucleotide sequence and homology to multiple phenobarbital-induced mRNA species

Phenobarbital (PB) treatment of rats of various strains leads to the accumulation of liver mRNAs which encode two or three immunochemically related but electrophoretically separable cytochrome P-450 polypeptides. These mRNAs hybridize efficiently to a single cloned cDNA derived from mRNA of PB-treated rats and, therefore, must have extensive sequence homology. The nucleotide sequence of this cloned cDNA was determined and shown to encode the COOH-terminal 211 amino acids of one of the major cytochrome P-450 isozymes induced in rat liver by PB. Together with the recently reported sequence data of Fujii-Kuriyama et al. (Fujii-Kuriyama, Y., Mizukami, Y., Kawajiri, K., Sagawa, K., and Muramatsu, M. (1982) Proc. Natl. Acad. Sci. U. S. A. 79, 2793-2797) for cloned rat cytochrome P-450 cDNA, our data suggest that differences between two closely related P-450 isozymes are restricted to the COOH-terminal half of the polypeptides, with highly divergent regions flanking a tridecapeptide which has been previously shown to be highly conserved in two dissimilar forms of rabbit liver cytochrome P-450. The significance of other interesting features of the cDNA sequence such as a second long (409 residues) open frame, an unusual poly(A) addition signal, and the absence of long hydrophobic stretches in the encoded polypeptide is discussed.     

Evidence for hybrid rodent and human insulin receptors in transfected cells.

Chinese hamster ovary cells and NIH 3T3 cells overexpressing mutant human insulin receptors were examined for the presence of hybrid receptors composed of human and rodent insulin receptors. In the present studies, most of the endogenous rodent receptors were found to be immunoprecipitated from the transfected cells but not the parental cells with a monoclonal antibody specific for human receptor. These data indicate that in these transfected cells, most of the endogenous rodent receptors are in a hybrid complex with the overexpressed human receptor. These results together with the in vitro studies of Treadway et al. (Treadway, J.L., Morrison, B.D., Soos, M.A., Siddle, K., Olefsky, J., Ullrich, A., McClain, D.A., and Pessin, J.E. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 214-218) showing that hybrid receptors exhibit transdominant inhibition explain the prior finding indicating that overexpression of defective insulin receptors interferes with the normal signaling of endogenous receptors.     

A discordance between the human muscle NCAM sequence and those seen in other NCAM cDNA clones.

A cDNA clone, designated NC7, has been isolated from human foetal kidney that partially codes for the 140 kDa isoform of human NCAM. This clone contains a 6 bp insert that is not present in the human muscle cDNA clone lambda 4.4. This same sequence has also been found in both a cDNA clone obtained from a human Small Cell Lung Carcinoma (SCLC) line and in human genomic DNA. Furthermore, an equivalent sequence to the 6 bp region identified in the above samples is present in mouse, rat and chicken NCAM. The 6 bp insertion does not lie at a predicted intron/exon boundary as extrapolated by homology studies with the chicken and, therefore, the mechanism by which the sequence is deleted from the human muscle clone lambda 4.4 remains unclear.     

Evidence for a novel Kit adhesion domain mediating human mast cell adhesion to structural airway cells

Background

Human lung mast cells (HLMCs) infiltrate the airway epithelium and airway smooth muscle (ASM) in asthmatic airways. The mechanism of HLMC adhesion to both cell types is only partly defined, and adhesion is not inhibited by function-blocking anti-Kit and anti-stem cell factor (SCF) antibodies. Our aim was to identify adhesion molecules expressed by human mast cells that mediate adhesion to human ASM cells (HASMCs) and human airway epithelial cells.

Methods

We used phage-display to isolate single chain Fv (scFv) antibodies with adhesion-blocking properties from rabbits immunised with HLMC and HMC-1 membrane proteins.

Results

Post-immune rabbit serum labelled HLMCs in flow cytometry and inhibited their adhesion to human BEAS-2B epithelial cells. Mast cell-specific scFvs were identified which labelled mast cells but not Jurkat cells by flow cytometry. Of these, one scFv (A1) consistently inhibited mast cell adhesion to HASMCs and BEAS-2B epithelial cells by about 30 %. A1 immunoprecipitated Kit (CD117) from HMC-1 lysates and bound to a human Kit-expressing mouse mast cell line, but did not interfere with SCF-dependent Kit signalling.

Conclusion

Kit contributes to human mast cell adhesion to human airway epithelial cells and HASMCs, but may utilise a previously unidentified adhesion domain that lies outside the SCF binding site. Targeting this adhesion pathway might offer a novel approach for the inhibition of mast cell interactions with structural airway cells, without detrimental effects on Kit signalling in other tissues.