Stable expression of blood group H determinants and GDP-L-fucose: beta-D-galactoside 2-alpha-L-fucosyltransferase in mouse cells after transfection with human DNA

We report here the application of a genetic approach to identify and isolate human DNA sequences controlling the expression of a GDP-L-fucose: beta-D-galactoside 2-alpha-L-fucosyltransferase [alpha-1,2)fucosyltransferase). Mouse L cells were chosen as host cells for this scheme since they express the necessary substrate and acceptor molecules for surface display of blood group H Fuc alpha 1----2 G al linkages constructed by (alpha-1,2) fucosyltransferases. However, they do not express cell surface blood group H structures nor detectable (alpha-1,2)fucosyltransferase activity. We therefore asked if (alpha-1,2)fucosyltransferase activity could be expressed and detected in these cells after transfection with human DNA sequences. These cells were transfected with genomic DNA isolated from a human cell line (A431) that expresses (alpha-1,2)fucosyltransferase. A panning procedure and fluorescence-activated cell sorting were used to isolate a mouse transfectant cell line that expresses cell surface H Fuc alpha 1----2 Gal linkages and a cognate (alpha-1,2)fucosyltransferase. Southern blot analysis showed that the genome of this cell line contains several hundred kilobase pairs of human DNA. Genomic DNA from this primary transfectant was used to transfect mouse L cells, and several independent, H-expressing secondary transfectants were isolated by immunological selection. Each expresses an (alpha-1,2)fucosyltransferase. Southern blot analysis demonstrated that the genome of each secondary transfectant contains common, characteristic human DNA restriction fragments. These results show that transfected human DNA sequences determine expression of the (alpha-1,2)fucosyltransferases in the mouse transfectants, that these sequences represent a single locus, and that they are within or linked to specific human restriction fragments identifiable in each secondary transfectant. These sequences may represent a human (alpha-1,2)fucosyltransferase gene.     

Identification of CRG-2. An interferon-inducible mRNA predicted to encode a murine monokine

In order to identify novel proteins produced by activated macrophages, a cDNA library was made from cultures of the mouse macrophage-like cell line RAW 264.7 that had been treated with conditioned medium from mitogen-stimulated spleen cells, and the library was screened by differential plaque hybridization. A cDNA clone was isolated that detected a 1.4-kilobase mRNA that accumulated dramatically in response to the spleen cell conditioned medium. The 1.4-kilobase mRNA encodes a predicted protein of 98 amino acids, designated CRG-2, molecular weight (Mr) 10,781, with a 21-residue signal peptide. The amino acid sequence indicates that CRG-2 is a member of the platelet factor 4 family (PF4) of cytokines. The CRG-2 mRNA was induced by alpha-, beta-, and gamma-interferons (IFNs) and by lipopolysaccharide. In response to IFN-gamma, the CRG-2 mRNA level reached a peak between 3 and 6 h. The accumulation of CRG-2 mRNA was not blocked by cycloheximide. Among the known members of the PF4 family, CRG-2 is most closely related to the interferon-inducible human protein IP-10. The 5'-flanking region of the crg-2 gene was isolated, and comparisons between crg-2 and IP-10 genes, mRNAs, and proteins reveal conserved features of possible functional importance. CRG-2 may play a role in host defense, particularly in the response to viral infection.     

The human thyroglobulin gene contains two 15-17 kb introns near its 3''-end.

We have cloned overlapping segments of the human thyroglobulin gene from a genomic cosmid library. Restriction mapping and electron microscopy show that a region of 38 kb at or near the 3'-end of this gene encodes only 850 nucleotides or 10% of the messenger RNA (mRNA) sequence. The region contains five exons of 130-210 nucleotides, split by introns of 1 to 15-17 kb. This represents the lowest ratio of coding to non-coding DNA (2.2%) found thus far in any eukaryotic gene. Blot hybridization under non-stringent conditions shows the presence of only one copy of this gene in the human genome and the absence of other closely related sequences.     

Isolation of a human tissue-type plasminogen-activator genomic DNA clone and its expression in mouse L cells

We have isolated a cDNA clone corresponding to a substantial portion of the human tissue-type plasminogen activator (t-PA) protein. It encodes almost all of the protein B chain and part of the 3' untranslated region. We have used this clone to screen bacteriophage lambda and cosmid libraries of human genomic DNA. Several related genomic clones were isolated. One of these, a cosmid clone, carried approx. 40 kb of human DNA. Mapping experiments indicate that the region containing the protein-coding exons is approx. 20 kb in length. The cosmid, containing the t-PA gene and the aminoglycosyl-3'-phosphotransferase dominant-selection marker, was introduced into mouse L cells. Approximately half of the transformants were shown to produce human t-PA. We demonstrated that the fibrinolytic t-PA activity could be specifically quenched by anti-t-PA antibody and that the recombinant t-PA was of similar size (by SDS-polyacrylamide gel electrophoresis) to the t-PA produced by the human Bowes melanoma cell line. Our results suggest that the cosmid clone carries the whole t-PA coding region together with the regulatory elements necessary for its expression.     

Differential regulation of interferon-induced mRNAs and c-myc mRNA by alpha- and gamma-interferons

The extent and kinetics of induction have been determined for eight mRNAs induced by alpha- and gamma-interferons in HeLa cells. The mRNAs which code for a 2-5A synthetase, metallothionein II, HLA class I antigen and five unknown proteins were induced 2 - greater than 100-fold by alpha-interferons. In the continued presence of alpha-interferon some mRNAs were maintained at the induced levels until at least 40 h, whereas others were induced only transiently. When the effects of alpha- and gamma-interferons were compared, the induced levels and kinetics were very similar for one mRNA (1-8) but were significantly different for the others. One mRNA (6-16) was induced more than 100-fold by alpha-interferon but not significantly by gamma-interferon. Parallel analysis of the level of c-myc mRNA showed it to decrease twofold in response to alpha-interferon, but to increase more than threefold in response to gamma-interferon, despite a more profound inhibition of cell growth by the latter. There must, therefore, be differences in how the levels of different mRNAs are sustained by alpha-interferons and how alpha- and gamma-interferons regulate the levels of the same mRNAs.     

Simplified cosmid vectors for gene transfer to cultured mammalian cells: isolation of the gene for elongation factor 2 from the mouse

We constructed a series of cosmid vectors that carry two tandemly arranged lambda cos and mammalian selective markers. We achieved cloning efficiencies of 1-3 x 10(7) and greater than 10(6) colony-forming units per microgram of insert, using a cloned 42-kb BamHI fragment and Sau3AI fragments of 40-50 kb from mouse genomic DNA, respectively. The modified Ca.phosphate coprecipitation method [Ishiura et al., Mol. Cell. Biol. 2 (1982) 607-616] considerably improved the efficiency of gene transfer of cosmids into cultured mammalian cells: when genes encoding thymidine kinase from herpes simplex virus type 1 and aminoglycoside 3'-phosphoribosyltransferase from Tn5 were selected, the efficiencies of gene transfer into mouse L cells were about 10(-6). The mouse genome contains one copy of the functional gene for elongation factor 2 (EF2) per haploid genome and multiple copies of the EF2-related gene. We isolated a cosmid that carried functional full-length mouse EF2 from a cosmid library of L-cell genomic DNA, by colony hybridization and subsequent gene transfer of candidate cosmids into human 143B cells.     

Isolation of a functional human interleukin 2 gene from a cosmid library by recombination in vivo

A method has been developed that allows the isolation of genomic clones from a cosmid library by homologous recombination in vivo. This method was used to isolate a human genomic interleukin 2 (IL2) gene. The genomic cosmid library was packaged in vivo into lambda phage particles. A recombination-proficient host strain carrying IL2 cDNA sequences in a non-homologous plasmid vector was infected by the packaged cosmid library. After in vivo packaging and reinfection, recombinants carrying the antibiotic resistance genes of both vectors were selected. From a recombinant cosmid clone the chromosomal IL2 gene was restored. After DNA mediated gene transfer into mouse Ltk- cells human IL2 was expressed constitutively.     

The human thyroglobulin gene is over 300 kb long and contains introns of up to 64 kb.

Thyroglobulin (Tg), the precursor of thyroid hormones, is a 660.000 Da dimeric glycoprotein synthesized exclusively in the thyroid gland. We have cloned the human thyroglobulin gene from cosmid and phage libraries and constructed a complete restriction map. The gene encodes an 8.7 kb mRNA, covers at least 300 kb DNA and contains at least 37 exons separated by large introns of up to 64 kb. A striking difference in structure between the 5' and 3' part of the gene suggests that it is composed of two evolutionarily different regions. The first 30 kb DNA encode 3 kb of the mRNA, yielding an exon:intron ratio of 1:10, whereas the remaining 270 kb encodes 5.7 kb of the mRNA with an exon:intron ratio of 1:47. In thyroid cells, the Tg gene is not rearranged and nuclear RNA homologous with sequences internal to the 64 kb intron is present, suggesting that the Tg gene is transcribed as a 300 kb RNA.     

Isolation of the functional human excision repair gene ERCC5 by intercosmid recombination

The complete human nucleotide exicision repair gene ERCC5 was isolated as a functional gene on overlapping cosmids. ERCC5 corrects the excision repair deficiency of Chinese hamster ovary cell line UV135, of complementation group 5. Cosmids that contained human sequences were obtained from a UV-resistant cell line derived from UV135 cells transformed with human genomic DNA. Individually, none of the cosmids complemented the UV135 repair defect; cosmid groups were formed to represent putative human genomic regions, and specific pairs of cosmids that effectively transformed UV135 cells to UV resistance were identified. Analysis of transformants derived from the active cosmid pairs showed that the functional 32-kbp ERCC5 gene was reconstructed by homologous intercosmid recombination. The cloned human sequences exhibited 100% concordance with the locus designated genetically as ERCC5 located on human chromosome 13q. Cosmid-transformed UV135 host cells repaired cytotoxic damage to levels about 70% of normal and repaired UV-irradiated shuttle vector DNA to levels about 82% of normal.     

Glycosylation in Saccharomyces cerevisiae: cloning and characterization of an alpha-1,2-mannosyltransferase structural gene.

A gene encoding an alpha-1,2-mannosyltransferase from Saccharomyces cerevisiae was cloned and sequenced. The alpha-1,2-mannosyltransferase which utilizes alpha-methylmannoside as acceptor of mannose from GDP-mannose was purified. The enzyme activity was shown to correspond to a 41 kDa protein band on sodium dodecyl sulphate-polyacrylamide gel electrophoresis. This protein band was digested in situ with trypsin and amino acid sequence information was obtained from four peptides. Degenerate oligonucleotide primers corresponding to the amino acid sequences were designed and used for polymerase chain reactions on yeast genomic DNA. A specific reaction product was used to screen a genomic library of S.cerevisiae. A fragment of approximately 5.7 kb was isolated, of which a 2.9 kb fragment was sequenced. It contained a 1329 base pair open reading frame encoding the peptide sequences of the purified alpha-1,2-mannosyltransferase. The gene, designated MNT1, is located on the right arm of chromosome 4. It encodes a 442 amino acid polypeptide with a calculated mol. wt of 51.4 kDa. The corresponding mRNA has a length of approximately 1.6 kb. Overexpression of the MNT1 gene increased this alpha-1,2-mannosyltransferase activity approximately 2.5-fold. The protein was shown to be modified with N-linked carbohydrate chains and its sequence contains one N-glycosylation site. The enzyme contains a putative membrane-spanning domain near its N-terminus and its topology is thus similar to that of mammalian Golgi glycosyltransferases. This is the first report of the cloning and sequencing of a yeast Golgi mannosyltransferase.     

The murine gene encoding the highly conserved Sm B protein contains a nonfunctional alternative 3' splice site.

The cDNA and partial genomic nucleotide (nt) sequences were derived for the mouse Sm B polypeptide and compared to the cDNA and genomic sequences encoding human Sm B. The deduced amino acid (aa) sequences from the mouse and human genes are identical with the exception of a single conserved aa substitution, accounting for the ability of anti-Sm antibodies to recognize the Sm polypeptides from a broad range of species. The genomic sequence of mouse B gene is similar to the human B genomic locus that extends from exon 6 to exon 7. These loci include conservation of both 3' alternative splice sites and putative branch points required to process B and B' mRNAs in human cells. However, the nt sequence downstream from the putative distal 3' splice junction and single nt flanking the 3' splice site consensus sequence, differ between mouse and human B. This results in a murine mRNA with a different predicted secondary structure around the distal 3' splice site when compared to humans. Thus, secondary structural constraints in the mRNA or changes in the exon sequence might prevent recognition of this alternative splice site to form B' mRNA in murine tissues.