Genetic synthetic lethality screen at the single gene level in cultured human cells.

Recently, we demonstrated the feasibility of a chemical synthetic lethality screen in cultured human cells. We now demonstrate the principles for a genetic synthetic lethality screen. The technology employs both an immortalized human cell line deficient in the gene of interest, which is complemented by an episomal survival plasmid expressing the wild-type cDNA for the gene of interest, and the use of a novel GFP-based double-label fluorescence system. Dominant negative genetic suppressor elements (GSEs) are selected from an episomal library expressing short truncated sense and antisense cDNAs for a gene likely to be synthetic lethal with the gene of interest. Expression of these GSEs prevents spontaneous loss of the GFP-marked episomal survival plasmid, thus allowing FACS enrichment for cells retaining the survival plasmid (and the GSEs). The dominant negative nature of the GSEs was validated by the decreased resident enzymatic activity present in cells harboring the GSEs. Also, cells mutated in the gene of interest exhibit reduced survival upon GSE expression. The identification of synthetic lethal genes described here can shed light on functional genetic interactions between genes involved in normal cell metabolism and in disease.

Functional conservation between the human, nematode, and yeast CK2 cell cycle genes.

Protein kinase CK2 (formerly casein kinase II) is a highly conserved serine/threonine protein kinase ubiquitous in eukaryotic organisms. Previously, we have shown that CK2 is required for cell cycle progression and essential for the viability of the yeast Saccharomyces cerevisiae. We now report that either the human or the nematode Caenorhabditis elegans CK2alpha catalytic subunit can substitute for the yeast catalytic subunits. Additionally, expression of the human CK2 regulatory subunit (CK2beta) can suppress the temperature sensitivity of either of the two yeast CK2 mutant catalytic subunits. Taken together, these observations reinforce the view that the CK2 cell cycle progression genes have been highly conserved during evolution from yeast to humans, not only in structure but also in function.

Establishment of a chemical synthetic lethality screen in cultured human cells.

The synthetic lethality screen is a powerful genetic method for unraveling functional interactions between proteins in yeast. Here we demonstrate the feasibility of a chemical synthetic lethality screen in cultured human cells, based in part on the concept of the yeast method. The technology employs both an immortalized human cell line, deficient in a gene of interest, which is complemented by an episomal survival plasmid expressing the gene of interest, and the use of a novel double-label fluorescence system. Selective pressure imposed by any one of several synthetic lethal metabolic inhibitors prevented the spontaneous loss of the episomal survival plasmid. Retention or loss over time of this plasmid could be sensitively detected in a blind test, while cells were grown in microtiter plates. Application of this method should thus permit high throughput screening of drugs, which are synthetically lethal with any mutant human gene of interest, whose normal counterpart can be expressed. This usage is particularly attractive for the search of drugs, which kill malignant cells in a gene-specific manner, based on their predetermined cellular genotype. Moreover, by replacing the chemicals used in this example with a library of either DNA oligonucleotides or expressible dominant negative genetic elements, one should be able to identify synthetic lethal human genes.

Molecular cloning of a novel human gene encoding a 63-kDa protein and its sublocalization within the 11q13 locus.

A human cDNA previously isolated by virtue of its ability to complement partially the ultraviolet sensitivity of a xeroderma pigmentosum cell line was further characterized. The transcription unit is expressed as a single 4.0-kb mRNA that encodes a novel 63-kDa cytoplasmic protein, possibly initiating from an internal AUG codon. The gene encoding this protein, named UVRAG, has been extremely well conserved during evolution, implying an important role for this gene product in cell metabolism. The transcribed mRNA is constitutively expressed in a wide variety of human tissues. The protein encoded by this gene is predicted to contain a coiled-coil structure and is likely to be metabolically unstable based on the occurrence of a strong PEST domain. UVRAG was assigned to human chromosome 11 by Southern hybridization to a somatic cell hybrid panel. Fluorescence in situ hybridization coupled with PCR analysis of human/rodent somatic cell hybrids containing segments of human chromosome 11 has localized this gene to a subregion of 11q13 in between the D11S916 and the D11S906 loci. Importantly, this region has been shown to be amplified in a variety of human malignancies, including breast cancer.

Isolation, chromosomal localization, and sequence analysis of human chromosome 21 zinc finger domains.

The zinc finger element is a conserved motif among a group of proteins involved in binding to nucleic acid. This motif has been detected in many regulatory factors and is highly represented in the human genome. To investigate the presence of zinc-finger-encoding genes on human chromosome 21, chromosome-specific libraries were screened with an oligodeoxynucleotide probe representing the conserved H-C link region between adjacent fingers. Three distinct genomic clones, designated ZF21-1, ZF21-2, and ZF21-3, were isolated and mapped to the long arm of chromosome 21 as well as to the heterochromatic short arm of several other chromosomes. DNA sequence analysis has shown that these genomic clones contain multiple zinc finger elements of the Kruppel type with only partial similarity to other known zinc finger genes. However, in each clone, few fingers were degenerated; they contain inframe stop codons and frameshifts that would preclude their translation. It seems therefore, that these chromosome 21 zinc finger sequences are not parts of functional genes. Nevertheless, the possibility that these domains are transcribed, and thus might have a regulatory role, is considered.

The structure of the human liver-type phosphofructokinase gene.

We have isolated the gene for the human liver-type phosphofructokinase, from upstream to the 5' mRNA terminus to beyond the polyadenylation site. The gene is at least 28 kb long and is divided into 22 exons; it contains conventional splice-junction sequences and one polyadenylation signal. Exons and introns are quite rich in G and C residues; some 60% of all nucleotides are either G or C. Five possible sites of polymorphism have been found. The gene structure reveals no signs of internal similarities despite protein sequence evidence which suggests that the PFK molecule is divided into two similar halves. The structure and organization of the human liver-type PFK gene are shown to be extremely similar to those of the rabbit muscle-type PFK.

The primary structure of human liver type phosphofructokinase and its comparison with other types of PFK.

The complete mRNA sequence of the human liver-type phosphofructokinase (hPFKL) was determined. The sequence included 55 nucleotides of 5' and 515 of 3' noncoding regions, as well as 2,337 nucleotides encoding the 779 amino acids of the hPFKL. Extensive similarity (approximately 90%) in the coding region was observed between the hPFKL and the mouse PFKL, whereas the degree of similarity between different types of PFK, i.e., hPFKL and human muscle-type PFK (hPFKM), was merely 68%. Nevertheless, striking similarity between these different types of PFK was noticed when the amino acid residues creating the various active sites of the enzyme were compared. Human PFK L- and M-specific probes were constructed and used to quantitate the mRNA levels in fetal and adult brains and fetal liver. It was found that while relative amount of PFKL mRNA in adult brain was one-fourth of that detected in fetal brain the level of PFKM mRNA in adult brain was slightly higher than in fetal tissue, suggesting that PFK expression might be controlled at the transcriptional level.

Hyperoxic exposure alters gene expression in the lung. Induction of the tissue inhibitor of metalloproteinases mRNA and other mRNAs.

Exposure to high concentrations of oxygen can result in tissue damage, particularly in the lung. Lung pathology induced by hyperoxia includes changes in lung cell populations and morphology. Presumably, alterations in gene expression underlie some of these cellular changes. In order to better understand the molecular basis of these events, a cDNA library was constructed from the mRNA of the lungs of a hyperoxia-exposed rabbit and differentially screened for clones corresponding to hyperoxia-induced messages. This approach has led to the isolation of four clones, three of which are presented in this communication. One clone corresponds to a message whose steady state levels were induced 6-fold and encodes the tissue inhibitor of metalloproteinases, a protein that plays a key role in the regulation of connective tissue turnover in some cells and potentiates erythroid development in others.

Construction of a cDNA clone containing the entire coding region of the human liver-type phosphofructokinase.

Genomic fragments containing coding sequences of the human liver phosphofructokinase (PFKL) were used to screen cDNA libraries and several clones corresponding to PFKL were isolated. Three overlapping cDNA clones spanning the entire coding region were stitched together yielding the plasmid pG-cPFKL3.0 containing a 3Kb PFKL cDNA down stream of the T7 promoter. To assess its coding capacity pG-cPFKL3.0 was transcribed by T7 RNA polymerase and the RNA product was used to program an in vitro translation system. An 80KDa polypeptide which co-migrated with an in vivo labeled PFKL and was recognized by anti PFKL antibodies was synthesized. These results show that the cDNA clone pG-cPFKL3.0 contains the entire coding region of the human PFKL.

Genomic clones of the human liver-type phosphofructokinase.

Genomic clones of human liver phosphofructokinase (PFK) were isolated by screening a gene bank enriched for chromosome 21 sequences with two synthetic oligonucleotide probes designed from peptide sequences of purified human liver PFK. A 3.3 Kb fragment derived from the genomic clones was sub-cloned and designated to pG-PFKL 3.3. It hybridized with a 3.5 Kb mRNA on Northern blots and was able to enrich selectively for liver PFK mRNA by hybrid-selection. These results demonstrated that the isolated clones contain sequences homologous to human PFKL mRNA. When hybridized to genomic DNA blots pG-PFKL 3.3 reacted with the same 3.3 Kb BamHI fragment in both human DNA and DNA of the mouse/human hybrid line WA17 which contains human chromosome 21 as the only human chromosome. These data confirm the assignment of the PFKL gene to chromosome 21.

Human Cu/Zn superoxide dismutase gene family: molecular structure and characterization of four Cu/Zn superoxide dismutase-related pseudogenes.

Four distinct human Cu/Zn superoxide dismutase (SOD1; EC 1.15.1.1)-related sequences were isolated from genomic DNA libraries. Genomic blots, heteroduplex analyses, and DNA sequencing showed that they are processed pseudogenes not residing on chromosome 21. Three of them originated from the 0.7-kilobase SOD1 mRNA, while the fourth was derived from the 0.9-kilobase mRNA species. Comparison between the coding sequences of the functional gene and two of the processed genes suggested that they integrated into the genome about 25 million years ago.

Architecture and anatomy of the chromosomal locus in human chromosome 21 encoding the Cu/Zn superoxide dismutase.

The SOD-1 gene on chromosome 21 and approximately 100 kb of chromosomal DNA from the 21q22 region have been isolated and characterized. The gene which is present as a single copy per haploid genome spans 11 kb of chromosomal DNA. Heteroduplex analysis and DNA sequencing reveals five rather small exons and four introns that interrupt the coding region. The donor sequence at the first intron contains an unusual variant dinucleotide 5'-G-C, rather than the highly conserved 5'-GT. The unusual splice junction is functional in vivo since it was detected in both alleles of the SOD-1 gene, which were defined by differences in the length of restriction endonuclease fragments (RFLPs) that hybridize to the cDNA probe. Genomic blots of human DNA isolated from cells trisomic for chromosome 21 (Down's syndrome patients) show the normal pattern of bands. At the 5' end of gene there are the 'TATA' and 'CAT' promoter sequences as well as four copies of the -GGCGGG- hexanucleotide. Two of these -GC- elements are contained within a 13 nucleotide inverted repeat that could form a stem-loop structure with stability of -33 kcal. The 3'-non coding region of the gene contains five short open reading-frames starting with ATG and terminating with stop codons.

Human Cu/Zn superoxide dismutase gene: molecular characterization of its two mRNA species.

Two cytoplasmic superoxide dismutase (SOD-1) mRNAs of about 0.7 and 0.9 kilobases (Kb.) were previously found in a variety of human cells. The two SOD-1 mRNAs are transcribed from the same gene and the major 0.7 Kb. species is approximately four times more abundant than the minor 0.9 Kb. mRNA. These two mRNAs differ in the length of their 3'-untranslated region and both have multiple 5'-ends. The longer transcript contains 222 additional nucleotides beyond the 3'-polyadenylated terminus of the short mRNA. S1 nuclease mapping and sequence analysis showed that these extra 222 nucleotides are specified by sequences contiguous to those shared by the two SOD-1 mRNAs. The 5'-termini of the two SOD-1 mRNAs were identified and mapped by both primer extension and S1 mapping. The majority of SOD-1 mRNA molecules (90-95%) have a 5'-start site located 23 base pairs (b.p.) downstream of the hexanucleotide -TATAAA-. The rest of the SOD-1 mRNA molecules have 5'-termini 30, 50 and 65 b.p. upstream from the major start region.

Nucleotide sequence and expression of human chromosome 21-encoded superoxide dismutase mRNA.

Cytoplasmic superoxide dismutase (SOD-1; EC 1.15.1.1) is encoded by human chromosome 21. The SOD-1 gene locus is located at chromosomal region 21q22, which is involved in Down syndrome. cDNA clones containing sequences of human SOD-1 were previously isolated. In the present study the nucleotide sequence of one clone, designated pS61-10, was determined. It contains 459 nucleotides representing the entire coding region and 95 nucleotides of the 3' untranslated region. In human cells two poly(A)-containing SOD-1 RNAs of 0.7 and 0.5 kilobases were detected. These two species are also present in monkey cells, whereas mouse cells contain only a 0.5-kilobase RNA. In a mouse/human hybrid line that contains chromosome 21 as the only human chromosome, the two human SOD-1 RNAs were detected, indicating that both are encoded by this chromosome. These RNAs were found in poly(A)-containing polysomal RNA and were translated in vitro to SOD-1 polypeptide; they are therefore functional mRNAs. In normal human fibroblasts 0.002-0.006% of the poly(A)-containing RNA was SOD-1 RNA. The level in monosomic 21 cells was 70% of this value and the level in fibroblasts from Down syndrome patients was about 2 times higher than normal.

Human cytoplasmic superoxide dismutase cDNA clone: a probe for studying the molecular biology of Down syndrome.

The gene locus for human cytoplasmic superoxide dismutase (SOD-1; superoxide:superoxide oxidoreductase, EC 1.15.1.1) is located in or near a region of chromosome 21 known to be involved in Down syndrome. To approach the molecular biology of this genetic disease we have constructed a SOD-1 cDNA clone. Poly(A)-containing RNA enriched for human SOD-1 mRNA was isolated, used to synthesize double-stranded cDNA, and inserted into the endonuclease Pst I site of the plasmid pBR322. The chimeric molecules were used to transform Escherichia coli. Two clones containing SOD-1 cDNA inserts were identified by their ability to hybridize specifically with mRNA coding for SOD-1. Each of these clones carries a 650-base-pair insert, as was determined by restriction enzyme digestion and electron microscopic heteroduplex analysis. Hybridization of labeled cloned cDNA to RNA blots revealed two distinct SOD-1 mRNA classes of 500 and 700 nucleotides. The data suggest that both are polyadenylylated and are coded by chromosome 21.