Cloning of a human multispanning membrane protein cDNA: evidence for a new protein family

We report the cloning of a human cDNA encoding a protein of calculated 68.8 kDa molecular mass, named hMP70. The deduced protein sequence shows a large N-terminal hydrophilic part and a C-terminal part with nine putative hydrophobic regions characteristic of integral transmembrane domains. Computer searches with sequence databases revealed homologies with three complete yeast proteins and with at least 19 human, 10 plant and one nematode short unidentified protein sequences translated from Expressed Sequence Tags (ESTs). Remarkably, this hMP70 protein retains between 27 and 31% overall sequence identity with the yeast proteins. We propose that hMP70 and related genes have evolved from a common ancestral gene and form a new multispanning membrane protein family which we call the MP70 protein family. Gene expression of hMP70 appears to be ubiquitous, as the mRNA is detectable in all human tissues analysed so far, as shown by Northern blot analysis. Furthermore, a protein of about 70 kDa is detectable in different mammalian cell lines, as shown by immunoblot analysis. From its widespread expression and conservation from yeast, plants to mammals, it is likely that hMP70 has a fundamental biological function in the cell.     

Indication of medical abortion in cases of fetal abnormalities diagnosed in utero

OBJECTIVE: To search the novel gene related to human hepatoma. METHODS: Northern blot analysis was used to isolate from EST (expressed sequence tags) clones a cDNA fragment differentially expressed in hepatocellular carcinoma versus its surrounding noncancerous hepatic tissue. Human fetal liver cDNA library was screened with EST fragment probe. RESULTS: Northern blot analysis showed that EST clone F9391 expressed at high level in all the 6 hepatocellular carcinoma samples but low level in the surrounding noncancerous hepatic tissues and 2 normal liver tissues. In further screening of human fetal liver cDNA library, we obtained a positive clone named FL6. Nucleotide sequencing indicated that FL6 contained 1464 base pairs and the matched DNA sequence was not found in the gene data bases (EMBL 96.6). Northern blot analysis of various fetus tissues with FL6 probe showed a high level expression in lung, small intestine, skin and placenta, middle level expression in liver, stomach, colon and muscle but low level expression in brain, thyroid, thymus, heart, kidney, pancreas and bladder. CONCLUSION: The results suggested that FL6 gene related to cell proliferation and may play an important role in the process of oncogenesis.     

Cloning and identification of human Sca as a novel inhibitor of osteoclast formation and bone resorption

Increased osteoclast activity is responsible for the enhanced bone destruction in postmenopausal osteoporosis, Paget's disease, bone metastasis, and hypercalcemia of malignancy. However, the number of known inhibitory factors that block osteoclast formation and bone resorption are limited. Therefore, we used an expression-cloning approach to identify novel factors produced by osteoclasts that inhibit osteoclast activity. A candidate clone was identified and isolated from a human osteoclast-like multinucleated cell (MNC) cDNA library, named osteoclast inhibitory peptide-1 (OIP-1), and the cDNA sequence was determined. This sequence matched that of the recently identified human stem cell antigen, was structurally similar to the mouse Ly-6 gene family, and the sequence predicted it was a glycosyl phosphatidyl inositol (GPI)-anchored protein that had a cleavable COOH-terminal peptide. Western blot analysis of conditioned media from 293 cells transfected with the OIP-1 cDNA clone confirmed that OIP-1 was released into the media as a membrane-bound GPI-linked protein. Interestingly, both recombinant OIP-1 expressed in Escherichia coli (which does not have GPI linker) and OIP-1 expressed by mammalian cells significantly reduced osteoclast-like MNC formation induced by 1,25-dihydroxyvitamin D3 or PTH-related protein in mouse and human bone marrow cultures, and inhibited 45Ca release from prelabeled bone in fetal rat organ cultures. In contrast, recombinant OIP-1 did not inhibit the growth of a variety of other cell types. These data indicate that OIP-1 is a novel, specific inhibitor of osteoclast formation and bone resorption.     

Molecular cloning of a novel cytokine cDNA encoding the ninth member of the fibroblast growth factor family, which has a unique secretion property

Glia-activating factor (GAF) is a novel heparin-binding growth factor purified from the culture supernatant of a human glioma cell line. It shows a spectrum of activity slightly different from those of other known growth factors. We have isolated the cDNA which encodes human GAF. A homology search revealed that GAF would be the ninth member of the FGF family, and we therefore call it FGF-9. The human FGF-9 cDNA cloned by using oligonucleotide probes encoded a polypeptide consisting of 208 amino acids. Sequence similarity to other members of the FGF family was estimated to be around 30%. Two cysteine residues and other consensus sequences in family members were also well conserved in the FGF-9 sequence. FGF-9 was found to have no typical signal sequence in its N terminus like those in acidic FGF and basic FGF. Acidic FGF and basic FGF are known not to be secreted from cells in a conventional manner. However, FGF-9 was found to be secreted from cells after synthesis despite its lack of a typical signal sequence. It could be detected exclusively in the culture medium of cDNA-transfected COS cells. The amino acid sequence of proteins purified from culture supernatant of the CHO cell line, which was cDNA transfected and selected as a high producer of FGF-9, showed that no peptides were cleaved from the N terminus except the initiation methionine. The rat FGF-9 cDNA was also cloned, and the structural analysis indicated that the PGF-9 gene is highly conserved. Expression of the FGF-9 gene could be detected in the brain and kidney of the adult rat. Restricted gene expression in organs and the unique secretion nature of the protein suggest that FGF-9 plays a physiological role which differs from those of well-characterized acidic FGF and basic FGF.     

Molecular cloning of a full-length cDNA for human type 3 adenylyl cyclase and its expression in human islets

The GK (Goto-Kakizaki) rat is a lean model of type 2 diabetes in which the diabetic state was spontaneously induced. We recently demonstrated the presence in GK rats of two functional point mutations in the promoter region of the type 3 adenylyl cyclase (AC3) gene that resulted in overexpression of AC3 mRNA associated with increased cAMP generation. The AC3 gene promoter mutations are the first molecular changes to be described in any specific gene in the GK rat. Here we report cloning of a full-length cDNA encoding human AC3 from a human fetal brain cDNA library using a PCR-based screening method. This 4142-bp cDNA predicts an open reading frame encoding 1144 amino acids containing putative 12 transmembrane-spanning domains which are typically found in other mammalian AC isoforms. Comparison of the translated amino acid sequence of the AC3 gene between human and rat shows 95% homology. Using RT-PCR, clear AC3 expression was detected in isolated human islets as well as a cDNA panel containing templates from eight different tissues (brain, heart, kidney, liver, lung, pancreas, placenta, and skeletal muscle). This wide distribution of AC3 expression may involve a number of physiological and pathophysiological metabolic processes.     

beta ig-h3: a transforming growth factor-beta-responsive gene encoding a secreted protein that inhibits cell attachment in vitro and suppresses the growth of CHO cells in nude mice

beta ig-h3 is a novel gene first discovered by differential screening of a cDNA library made from A549 human lung adenocarcinoma cells treated with transforming growth factor-beta 1 (TGF-beta 1). It encodes a 683-amino-acid protein containing a secretory signal sequence and four homologous internal domains. Here we show that treatment of several types of cells, including human melanoma cells, human mammary epithelial cells, human keratinocytes, and human fibroblasts, with TGF-beta resulted in a significant increase in beta ig-h3 RNA. A portion of the beta ig-h3 coding sequence was expressed in bacteria, and antisera against the bacterially produced protein was raised in rabbits. This antisera was used to demonstrate that several cell lines secreted a 68-kD beta IG-H3 protein after treatment with TGF-beta. Transfection of beta IG-H3 expression plasmids into Chinese hamster ovary (CHO) cells led to a marked decrease in the ability of these cells to form tumors in nude mice. The beta IG-H3 protein was purified from media conditioned by recombinant CHO cells, characterized by immunoblotting and protein sequencing and shown to function in an anti-adhesion assay in that it inhibited the attachment of A549, HeLa, and WI-38 cells to plastic in serum-free media. Sequencing of cDNA clones encoding murine beta ig-H3 indicated 90.6% conservation at the amino acid level between the murine and human proteins. Finally, the beta ig-h3 gene was localized to human chromosome 5q31, a region frequently deleted in preleukemic myelodysplasia and leukemia. The corresponding mouse beta ig-h3 gene was mapped to mouse chromosome 13 region B to C1, which confirms a region of conservation on human chromosome 5 and mouse chromosome 13. We suggest that this protein be named p68 beta ig-h3.     

Cloning, expression and chromosome mapping of adducin-like 70 (ADDL), a human cDNA highly homologous to human erythrocyte adducin

From a human fetal-brain cDNA library we isolated a novel human cDNA, termed human adducin-like 70 (gene symbol ADDL), whose predicted amino acid sequence showed a high degree of homology to adducins. This cDNA clone (ADDL), which contained an open reading frame of 2,022 nucleotides encoding 674 amino acids, revealed 54%, 53%, and 59% identity in predicted amino acid sequence with alpha and beta components of human adducin and rat adducin 63, respectively. Human adducin-like 70 is likely to play an important role in the skeletal organization of the cell membrane. Northern blot analysis indicated ubiquitous expression of this gene in adult human tissues. We localized the gene to chromosome bands 10q24.2-->q24.3 by fluorescence in situ hybridization (FISH).     

Cloning of a human cDNA for CTP-phosphoethanolamine cytidylyltransferase by complementation in vivo of a yeast mutant

CTP-phosphoethanolamine cytidylyltransferase (ET) is the enzyme that catalyzes the formation of CDP-ethanolamine in the phosphatidylethanolamine biosynthetic pathway from ethanolamine. We constructed a Saccharomyces cerevisiae mutant of which the ECT1 gene, putatively encoding ET, was disrupted. This mutant showed a growth defect on ethanolamine-containing medium and a decrease of ET activity. A cDNA clone was isolated from a human glioblastoma cDNA expression library by complementation of the yeast mutant. Introduction of this cDNA into the yeast mutant clearly restored the formation of CDP-ethanolamine and phosphatidylethanolamine in cells. ET activity in transformants was higher than that in wild-type cells. The deduced protein sequence exhibited homology with the yeast, rat, and human CTP-phosphocholine cytidylyltransferases, as well as yeast ET. The cDNA gene product was expressed as a fusion with glutathione S-transferase in Escherichia coli and shown to have ET activity. These results clearly indicate that the cDNA obtained here encodes human ET.     

Expression of structure-specific recognition protein mRNA in fetal kidney and Fe-nitrilotriacetate-induced renal carcinoma in the rat

Specific expression of the structure-specific recognition protein (SSRP) gene was investigated in rat fetal, adult, and tumor tissues using a 2.0-kb partial sequence of rat SSRP cDNA isolated from a cDNA library of rat renal cell carcinoma. The results revealed that it was rather specifically expressed in rat fetal kidney and renal cell carcinoma induced by Fenitrilotriacetate, but not in adult kidney, when various organs were tested by Northern blot analysis. In situ hybridization further demonstrated that it was located in the neoplastic cells of renal cell carcinoma and in the epithelial cells of fetal kidney but undetectable in any cells of normal adult kidney. These observations seem to imply the involvement of SSRP gene, which is believed to recognize structural alterations of DNA, in kidney development and carcinogenesis of certain types of kidney cancer.     

Biologic effects of introduction of wild-type p53 cDNA into a human ovarian cancer cell line SKOV-3

OBJECTIVE: To study the effects on biologic behavior in cells obtained from human ovarian cancer cell line SKOV-3 into which the wild-type p53 cDNA was introduced. METHOD: Recombinant eukaryotic expression vector pC53-SN3 containing full-length human wild-type p53 cDNA and vector containing neomycin resistance gene only were introduced by lipofectamine-mediated gene transfection into SKOV-3 cell line which does not express endogenous p53. The clones obtained were observed for their biologic behavior. RESULTS: (1) 2 clones named pC53 and 2 clones named pNeo were obtained after pC53-SN3 and vector transfection respectively; (2) The morphology of cells either from pC53 or from pNeo did not change significantly with respect to their parental SKOV-3; (3) The growth rate of cells from pC53 was much slower than that from SKOV-3, while the cell growth curve of pNeo was similar to that of SKOV-3; (4) The number of colones formed in the soft-agar by pC53 was significantly less than that by SKOV-3 or by pNeo; (5) The percentage of phase G1/G0 of pC53 was much higher than that of SKOV-3 and pNeo. CONCLUSION: Wild-type p53 cDNA may be considered as one of the target genes for the gene therapy of ovarian cancer.     

Structure of human estrogen and aryl sulfotransferase gene. Two mRNA species issued from a single gene

Estrone sulfate is the predominant form of estrogens found in the circulation in women and could thus serve as precursor for active estrogens in target tissues by removal of the sulfate group through the action of endogenous steroid sulfatase. Recently, we isolated a cDNA encoding human placental estrogen sulfotransferase that differs from brain aryl sulfotransferase only in the 5'-noncoding sequence. To increase our knowledge of the regulation and tissue-specific expression of sulfotransferase gene, we screened a lambda EMBL3 library of human leucocyte genomic DNA using the estrogen sulfotransferase cDNA as probe and isolated a clone containing almost the whole gene sequence. Sequencing of the gene indicates that it is included in approximately 7.7 kilobases and contains nine short exons separated by eight introns. The two first exons, named exon 1a and exon 1b, are noncoding and correspond to the 5'-untranslated sequences of human brain and human placental estrogen sulfotransferase cDNAs, respectively. Transfection of chloramphenicol acetyltransferase reporter gene vectors containing the 5'-flanking sequence upstream from exon 1a and exon 1b in human adrenal adenocarcinoma cells indicates that both sequences possess promoter activity. The present results thus indicate that brain aryl sulfotransferase and placental human placental estrogen sulfotransferase mRNA species are transcribed from a single gene by alternate exon 1a and exon 1b promoters, respectively. Using DNA from panels of human/rodent somatic cell hybrids and amplification of the gene by polymerase chain reaction, the human placental estrogen sulfotransferase gene was assigned to chromosome 16.     

Structure and expression of human fibroblast growth factor-10

We isolated the cDNA encoding a novel member of the human fibroblast growth factor (FGF) family from the lung. The cDNA encodes a protein of 208 amino acids with high sequence homology (95.6%) to rat FGF-10, indicating that the protein is human FGF-10. Human FGF-10 as well as rat FGF-10 has a hydrophobic amino terminus ( approximately 40 amino acids), which may serve as a signal sequence. The apparent evolutionary relationships of human FGFs indicate that FGF-10 is closest to FGF-7. Chromosomal localization of the human FGF-10 gene was examined by in situ hybridization. The gene was found to map to the 5p12-p13 region. Human FGF-10 (amino acids 40 to 208 with a methionine residue at the amino terminus) was produced in Escherichia coli and purified from the cell lysate. Recombinant human FGF-10 (approximately 19 kDa) showed mitogenic activity for fetal rat keratinizing epidermal cells, but essentially no activity for NIH/3T3 cells, fibroblasts. The specificity of mitogenic activity of FGF-10 is similar to that of FGF-7 but distinct from that of bFGF. In structure and biological activity, FGF-10 is similar to FGF-7.