The 2 micron plasmid of laboratory yeast strains is a type-1/type-2 hybrid

Industrial yeast strains carry one of two homeologous 2 microns plasmids designated as type-1 or type-2. The 2 microns plasmid, Scp1, found in common laboratory strains of Saccharomyces cerevisiae is considered a type-2 plasmid, since the ori, STB, RAF and REP1 loci and intergenic sequences of the right-unique region of Scp1 are homologous to the corresponding loci in industrial strain type-2 plasmids. However, within both its 599 bp inverted repeats Scp1 has 142-bp sequences homologous to the bakers' yeast type-1 plasmid. DNA sequence analyses and oligonucleotide hybridizations indicate that the 142-bp insertion in Scp1 was probably due to homeologous recombination between type-1 and type-2 plasmids. These results suggest that some of the plasmid and chromosomal sequence polymorphisms seen in laboratory yeast strains result from homeologous recombination in their ancestral breeding stock.

Essential region for self-replication of Coryneform bacteria plasmid pXZ10145

A pTSK series of recombinant plasmids were constructed by cloning DNA fragments of pXZ10145 or its deleted deriviate pATN65 into plasmid vector pACYC177 of E. coli. Experiment results of Coryneform bacteria transformation with these pTSK plasmids allowed us to localize the essential region for self-replication on plasmid pXZ10145. The minimal replication region of the pXZ10145 was located on a 1.2kb Nael-Nrul DNA fragment in which only one open reading frame was found. This ORF was believed to be encoded a trans-acting replication factor. The replication origin (oriV) was locate on a 0.3kb NaeI-SalI fragment which was within the ORF region.

Modulation of interferon (IFN)-inducible gene expression by retinoic acid. Up-regulation of STAT1 protein in IFN-unresponsive cells

Interferons (IFN) and retinoids failed to inhibit the growth of a number of breast tumor cell lines. However, a combination of these two biological response modifiers significantly suppressed the cell growth at pharmacologically achievable doses. The molecular basis for such enhancement was investigated in MCF-7, a breast tumor cell line resistant to growth inhibition by IFN-beta. Pretreatment of cells with retinoic acid (RA) for 16 h followed by IFN-beta, but not the converse, induced cytotoxic effects in the cells. Continuous presence of RA was not necessary, although it enhanced the degree of cell death when present. Further analyses revealed that IFN-beta failed to activate IFN-stimulated gene transcription. However, IFN-beta strongly up-regulated the gene expression in RA-pretreated cells. Both IFN-beta- and IFN-gamma-inducible gene expression were enhanced via a modulation of the transcriptional factor IFN-stimulated gene factors-3 and GAF binding to respective cognate regulatory elements. STAT1 was undetectable in these cells prior to RA treatment. RA increased the levels of this crucial regulator, thereby restoring IFN responses. Thus, RA augmentation of STAT1 may be an early step in the cooperative anti-tumor effects of IFN and RA.

Mutagenicity and toxicity of the DNA alkylation carcinogens 1,2-dimethylhydrazine and azoxymethane in Escherichia coli and Salmonella typhimurium

DNA alkylating agents such as 1,2-dimethylhydrazine (SDMH) and azoxymethane (AOM) are potent carcinogens and are widely used to induce colon tumors in experimental animals. However, standard bacterial mutagenesis assays have failed to detect the mutagenic effects of these chemicals. Using derivatives of a set of Escherichia coli test strains developed by Cupples and Miller (Proc. Natl. Acad. Sci. USA, 86, 5345, 1989), we hve demonstrated that under two conditions, SDMH and AOM induced point mutations by several-fold in a dose-dependent manner: (i) of six possible base substitutions, they only induced GC-->AT transitions; and (ii) the cells must be deficient in O6-methylguanine (O6MeG) DNA methyltransferase (MTase) activity. SDMH and AOM up to 200 microg/ml were unable to induce His+ revertants in a Salmonella Ames test strain TA1535 (GC-->AT); however, in the absence of mammalian S9 extract, His+ revertants increased up to 55-fold upon treatment of an isogenic Salmonella strain deficient in MTase activity. These results indicate that SDMH and AOM are indeed bacterial mutagens and that lesions induced by them are the target of DNA repair MTases, which probably include mutagenic and carcinogenic lesions such as O6MeG. Furthermore, variable responses of bacterial species to SDMH- and AOM-induced mutagenicity suggests a difference either in the metabolism of potential mutagens or in the repair of specific lesions. Since O6MeG is not only a mutagenic lesion but also a lethal lesion if left unrepaired, we compared the mutagenicity and toxicity of SDMH and AOM with an SN-type methylating carcinogen. N-methyl-N'-nitro-N-nitrosoguanidine, and conclude that SDMH and AOM are weak bacterial mutagens.

Particle-mediated delivery of recombinant expression vectors to rabbit skin induces high-titered polyclonal antisera (and circumvents purification of a protein immunogen)

Polyclonal antibodies were generated in rabbits by delivery to skin of gold particles coated with mammalian expression vectors encoding a cytoplasmic (beta-galactosidase) or a nuclear (L1 capsid of cottontail rabbit papillomavirus) protein. One primary and one booster immunization of 30 micrograms DNA per rabbit yielded specific antisera with titers from 1:24 000 to 1:120 000 in each of eight rabbits, as detected by ELISA and Western blot analysis. Genetic immunization requires relatively small amounts of DNA, eliminates the need to purify the protein immunogen, and does not require irritating adjuvants.

[Changes of peripheral blood stem cells of patients with malignant hematopoietic tumor after chemotherapy]

To investigate the most appropriate condition for collecting stem cells during peripheral blood stem cells transplantation (PBSCT), we used colony forming unit-granulocyte, macrophage (CFU-GM) as index to observe the change of PBSC after chemotherapy in 12 patients with malignant hematopoietic tumor. The results showed that it was more suitable to collect PBSC when the number of leucocytes was 2 x 10(9)/L, neutrophils 0.8 x 10(9)/L and platelets 110 x 10(9)/L after chemotherapy for 2 weeks. After chemotherapy, the number of leucocytes were suppressed evidently. On the contrary, a lot of CFU-GM appeared during the recovery phase of peripheral hemogram. At the nadir of peripheral hemogram, the number of leucocytes correlated negatively with the number of CFU-GM (P < 0.05). The longer the time of chemotherapy, the less the CFU-GM that in the same patients. After chemotherapy the CD34+ cells which appeared early might not form CFU-GM, but the CD34+ cells at the recovery phase of peripheral hemogram had the same trend of change with CFU-GM.

Low-temperature femtosecond-resolution transient absorption spectroscopy of large-scale symmetry mutants of bacterial reaction centers

Reaction centers isolated from three large-scale symmetry mutants sym0, sym2-1, and sym5-2 described in the previous article of this issue [Taguchi, A. K. W., Eastman, J. E., Gallo, D. M., Jr., Sheagley, E.. Xiao, W., & Woodbury, N. W. (1996) Biochemistry 35, 3175-3186] have been investigated by low-temperature ground state and ferntosecond-resolution transient absorption spectroscopy. All three of these large-scale symmetry mutants undergo electron transfer at 20 K. The mutants sym0 and sym5-2 have yields and dominant rates of charge separation comparable to wild type. However. the sym2-mutant shows a roughly 35%, quantum yield at this temperature, and the major kinetic component of the initial electron transfer is slower than wild type by nearly a factor of 100. The sym0 mutant showed substantial changes in the monomer bacteriochiorophyll ground state and transient spectra, and both sym0 sym2-1 showed changes in the bacteriopheophyll ground state and transient spectra. In particular, sym2-1 shows a small absorbance decrease in the region of the Qx band of the B side bacteriopheophytin which could be attributed to 10%-20% electron transfer along the B pathway.

Asymmetry requirements in the photosynthetic reaction center of Rhodobacter capsulatus

Nine large-scale symmetry reaction center mutants were constructed in Rhodobacter capsulatus by replacing segments of the M subunit gene with the homologous region of the L subunit gene. Between them, the mutations resulted in symmetrization of essentially the entire region from the carboxy terminal portion of the C helix through most of the E helix. The amino acids in this region define about 80% of the environment of the reaction center cofactors. These studies show that roughly 80% of the amino acids that come in close contact with the cofactors involved in initial electron transfer can be made symmetric in a piecewise manner without loss of the ability to grow photoheterotrophically. However, the amino acid regions near the quinones and iron atom are much more sensitive to symmetrization and most of the large-scale changes in this region resulted in the loss of photosynthetic viability, probably due to loss of stable reaction centers from the photosynthetic membrane. More detailed analysis of the isolated photosynthetic membranes from these mutants showed that in all cases but one, there was some amount of charge separation occurring in the mutant reaction centers. This bank of mutants serves as a useful starting point for more detailed studies of the differential molecular interactions which occur between the two reaction center subunits and their associated cofactors.

Neurobiological mechanisms of the meridian and the propagation of needle feeling along the meridian pathway

The present experiments attempt to find the meridian phenomenon and how the needle feeling propagates along the given meridian channels. The neurobiological mechanisms of the meridian were studied with neuroelectrical recording from the motor neurons and CB-HRP retrograde histochemistry technique in both rats and cats. The results demonstrated that most, but not all, of alpha motor neurons supplying a muscle group of a given meridian were selectively activated by afferent inputs originating not only from homonymous or heterogeneous, but synergistic muscle, but also from the skin nerve overlying the muscle group of the homonymous meridian. However, the afferent inputs from the heterogeneous meridian have very weak or no effect. On the other hand, the labeled motor neurons supplying a given meridian muscles from a discrete longitudinal column with a definite bound in the lateral ventral horn. There are oriented dendro-dendristes projections between the labeled motor neurons. The characteristics of both selective responses of the motor neurons to afferent inputs and their neuro-anatomical arrangements in spinal cord offer neurobiological evidence for the meridian phenomenon.

Synthesis and characterization of composite nucleic acids containing 2', 5'-oligoriboadenylate linked to antisense DNA

Composite nucleic acids, known as 2-5A antisense chimeras, cause the 2-5A-dependent ribonuclease (RNase L) to catalyze the specific cleavage of RNA in cell free systems and in intact cells. Such 2-5A antisense chimeras are 5'-monophosphorylated, 2,'5'-linked oligoadenylates covalently attached to antisense 3',5'-oligodeoxyribonucleotides by means of a linker containing two residues of 1,4-butanediol phosphate. Here we report a fully automated synthesis of 2-5A antisense chimeras on a solid support using phosphoramidite methodology with specific coupling time modifications and their subsequent purification by reverse-phase ion-pair and anion exchange HPLC. Purified 2-5A antisense chimeras were characterized by [1H]NMR and [31P]NMR, MALDIMS, and capillary gel electrophoresis. The synthetic 2',5'-linked oligoadenylate showed no phosphodiester isomerization to 3',5' during or after synthesis. In addition, we have developed facile methodologies to characterize the chimeras using digestion with various hydrolytic enzymes including snake venom phosphodiesterase I and nuclease P1. Finally, Maxam-Gilbert chemical sequencing protocols have been developed to confirm the entire sequence of these chimeric oligonucleotides.

DNA mismatch repair mutants do not increase N-methyl-N'-nitro-N-nitrosoguanidine tolerance in O6-methylguanine DNA methyltransferase-deficient yeast cells

Treatment of cells with N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) produces, among other lesions, mutagenic and carcinogenic lesions such as O6-methylguanine (O6MeG) and O4-methylthymine in DNA. An O6MeG DNA methyl-transferase (MTase) specifically and efficiently repairs such lesions. MTase-deficient bacterial, yeast and mammalian cells exhibit increased sensitivity not only to MNNG-induced mutagenesis, but also to MNNG-induced killing, suggesting that O6MeG-type lesions are also lethal to the cells. The lethal effect caused by O6MeG is not clear. Results from several recent experiments indicate that some MNNG-tolerant cell lines exhibit a loss of DNA mismatch binding/repair activity, suggesting that functional mismatch repair is probably responsible for the cellular sensitivity to DNA methylating agents. We tested this abortive O6MeG-T mismatch repair hypothesis in a well-defined lower eukaryote, Saccharomyces cerevisiae. We found that while mgt1-deleted MTase-deficient yeast strains are hypersensitive to MNNG-induced killing, combination of this mutation with any of the mlh1, msh2 or pms1 mutations did not render cells more tolerant to killing. msh3 mutation also did not rescue MNNG-induced genotoxicity. Furthermore, through the isolation and characterization of MNNG-tolerant cell lines from the MTase-deficient mutants we demonstrated that a DNA mismatch repair defect is neither sufficient nor required for this process. Since both DNA repair MTases and mismatch repair proteins are highly conserved between yeast and mammalian cells, our results could suggest alternative mechanisms in the cellular tolerance to O6MeG-induced killing.

Catalytic cleavage of an RNA target by 2-5A antisense and RNase L

2-5A antisense (2-5A-AS) molecules are chimeric oligonucleotides that cause 2-5A-dependent RNase (RNase L) to catalyze the selective cleavage of RNA in human cells. These composite nucleic acids consist of a 5'-monophosphorylated, 2',5'-linked oligoadenylate known as 2-5A (an activator of RNase L) covalently attached to antisense 3',5'-oligodeoxyribonucleotides. Here, we characterize the targeted cleavage of the double-stranded RNA-dependent protein kinase (PKR) mRNA by purified, recombinant human RNase L. A 2-5A-AS chimera, which contains complementary sequence to PKR mRNA, and unmodified 2-5A, which causes general RNA decay, were about 20- and 40-fold more active, respectively, than 2-5A-AS chimeras in which the DNA domains are not complementary to sequences in PKR mRNA. Directed cleavage was efficient because each 2-5A-AS chimera targeted many RNA molecules. Moreover, RNase L caused the catalytic cleavage of the RNA target (kcat of approximately 7 s-1). The precise sites of PKR mRNA cleavage caused by 2-5A-AS were mapped, using a primer extension assay, to phosphodiester bonds adjacent to the 3' terminus of the chimera binding site (5' on the RNA target) as well as within the chimera's oligonucleotide binding site itself. The selectivity of this approach is shown to be provided by the antisense arm of the chimera, which places the RNA target in close proximity to the RNase.

Papilloma formation in human foreskin xenografts after inoculation of human papillomavirus type 16 DNA

A mouse model of high-risk human papillomavirus infection was developed in which human papillomavirus (HPV) type 16 DNA was inoculated into human foreskin grafted to the skin of severe combined immunodeficient (scid) mice. Grafted skin contained human epidermis and dermis and, like normal human skin, expressed involucrin in differentiating keratinocytes. HPV type 16 DNA, attached to gold particles, was delivered directly into human epidermal cells and induced exophytic papilloma with histologic features of papillomavirus infection, including koilocytosis and expression of papillomavirus capsid antigen. This model should be useful for determining in vivo the functions of viral genes and for developing strategies to prevent and treat HPV-associated disease. It may also be of value in developing animal models of other human skin diseases.

Expression of the human MGMT O6-methylguanine DNA methyltransferase gene in a yeast alkylation-sensitive mutant: its effects on both exogenous and endogenous DNA alkylation damage

Common Mer- cell lines deficient in O6-methylguanine DNA methyltransferase (MTase) activity probably result from the down-regulation of, rather than mutations in, the MGMT gene. However, the down-regulation of other unrelated genes was also observed in some of these cell lines, making it difficult to determine the precise functions of the MGMT MTase gene. To study the biological function of human MGMT MTase, we seek to utilize a newly created yeast mgt1 mutant deficient in the DNA repair MTase activity. The human MGMT cDNA was cloned into yeast expression vectors so that the MGMT gene is under the control of either an inducible GAL1 promoter or a constitutive ADH1 promoter. Upon galactose induction, the PGAL1-MGMT transformant had about 40-fold MTase activity compared to the wild-type strain. MGMT overexpression protected the yeast mgt1 mutant against alkylation-induced killing and mutation. Limited expression of the MGMT gene in the mgt1 mutant still provides significant alkylation resistance, albeit at a reduced level. The yeast mgt1 mutants increase spontaneous mutation rate, whereas constitutive expression of the MGMT gene lowered the spontaneous mutation rate in the mgt1 mutant to the wild-type level. We suggest that MGMT MTase may play the same role in human cells as the MGT1 MTase in yeast cells. Thus our results demonstrate that the human MGMT gene functionally complements the yeast MTase-deficient mutant in the protection against exogenous and endogenous DNA alkylation damage, which provides a useful tool for the study of in vivo mammalian MTase functions.

UBP5 encodes a putative yeast ubiquitin-specific protease that is related to the human Tre-2 oncogene product

A gene from chromosome V of the yeast Saccharomyces cerevisiae has been cloned and sequenced. The deduced amino acid sequence encoded by this gene is similar to several ubiquitin-specific proteases from yeast, especially at the highly conserved domain. It is thus named UBP5. UBP5 is also closely related to the human Tre-2 and the mouse Unp oncogene products. This study adds a new member to the ubiquitin protease family and suggests that alteration of ubiquitin protease activity may result in cancer in mammals. However, disruption of the UBP5 gene in a haploid strain did not result in a noticeable phenotypic alteration.

Development of 2',5'-Oligonucleotides as Therapeutic Agents

Abstract:

The unique 2',5'-phosphodie.ster bond-linked oligonucleo­ tide known as 2-5A (Pn5'A2'(p5'A2')mp5'A) plays a key role in mediation of the anti-encephalomyocarditis virus action of interferon. 2-5A acts as a potent inhibitor of translation through the activation of a constituent latent endonuclease, the 2-5A-dependent ribonuclease (RNase) , which degrades RNAs. This 2-5A system, as part of a natural defense mechanism against virus infection, provides a paradigm for a new approach to the regulation of gene expression. Realization of this poten­ tial requires an understanding of the 2-5A oligoribonucleotide-associated structural parameters which govern its lifetime in biological systems and its interaction with the 2-5A-dependent ribonuclease responsible for RNA destruction. In this review, we describe the partial realization of such an understanding and the resulting development of a new approach to the specific and targeted cleavage of RNA by directing 2-5A-dependent RNase action to a precise target with an antisense DNA. The synthesis and mechanism of action of these novel composite nucleic acids permits exploration of the potent RNA destruction ability of the 2-5A-dependent RNase coupled with the specificity of antisense oligonucleotides as potential therapeutic agents for a variety of diseases.

Transpogenes: the transposition-like integration of short sequence DNA into the yeast 2 micron plasmid creates the STB locus and plasmid-size polymorphism

The type-2 2 mu plasmid of industrial yeast strains exhibits extensive size polymorphism in the STB (plasmid stability) locus and IR (inverted repeat)-right region. Comparative DNA sequence analyses of STB alleles identified a 38-bp sequence flanked by a 25-bp direct repeat as the underlying structural motif. Variable unequal recombination within the direct repeat accounted for the observed polymorphism of STB alleles. IR-right polymorphism was observed to result from tandem duplication of a 22-bp sequence flanked by a 9-bp direct repeat. The flanking direct repeats marked both loci as originating from the transposition-like integration of short DNA fragments. We call these structures transpogenes and note that these are hybrid structures of host and foreign DNA which can evolve into functional loci.

Papilloma formation by cottontail rabbit papillomavirus requires E1 and E2 regulatory genes in addition to E6 and E7 transforming genes

The present study used the cottontail rabbit papillomavirus DNA-rabbit system to evaluate whether the regulatory genes E1 and E2 and the transforming gene E6 are required for papilloma formation. Frameshift mutations were generated in the individual genes in the context of a full-length cottontail rabbit papillomavirus genome, and the mutant DNAs were intradermally inoculated into domestic rabbits. None of the mutants induced papillomas. Marker rescue experiments confirmed that the defects were due to mutations that we deliberately introduced. Marker rescue also confirmed our previous report that the upstream region of E7 around position 9 was critical for papilloma induction. These results demonstrate that the E1 and E2 regulatory genes as well as the E6 and E7 transforming genes are each required for papilloma formation. Each gene may provide molecular targets for therapeutic intervention.

Blockage of NF-kappa B signaling by selective ablation of an mRNA target by 2-5A antisense chimeras

Activation of 2-5A-dependent ribonuclease by 5'-phosphorylated, 2',5'-linked oligoadenylates, known as 2-5A, is one pathway of interferon action. Unaided uptake into HeLa cells of 2-5A linked to an antisense oligonucleotide resulted in the selective ablation of messenger RNA for the double-stranded RNA (dsRNA)-dependent protein kinase PKR. Similarly, purified, recombinant human 2-5A-dependent ribonuclease was induced to selectively cleave PKR messenger RNA. Cells depleted of PKR activity were unresponsive to activation of nuclear factor-kappa B (NF-kappa B) by the dsRNA poly(I):poly(C), which provides direct evidence that PKR is a transducer for the dsRNA signaling of NF-kappa B.

Femtosecond pump-probe analysis of energy and electron transfer in photosynthetic membranes of Rhodobacter capsulatus

Low-intensity, 295 K, femtosecond pump-probe transient absorption measurements are described that have been performed to investigate energy and electron transfer in photosynthetic membranes from a Rhodobacter capsulatus strain lacking functional light harvesting antenna complex II. Spectral and kinetic similarities between the absorption changes of isolated reaction centers and those of reaction centers in membranes upon 800-nm excitation suggest that the charge separation process in both cases is very similar. An ultrafast energy relaxation process observed near 872 nm when 800-nm excitation is used is interpreted as interexcitonic relaxation within the antenna, though other interpretations, such as vibrational relaxation, are possible. On the basis of global exponential fitting analysis of the time-dependent spectral changes using 800- and 880-nm excitation wavelengths to selectively excite the reaction center and the LHI antenna, respectively, it is found that excitation energy transfer and trapping in Rb. capsulatus is limited by the overall rate of energy transfer between the antenna and the reaction center. This conclusion is supported by the observation that excitation at 800 nm, but not 880 nm, results in absorbance changes indicative of charge separation with a lifetime (3.1 ps) very close to that reported for charge separation in isolated reaction centers (3.5 ps). Thus, most reaction centers that are directly excited undergo charge separation and not backward energy transfer to the LHI antenna complexes. Both a kinetic model analysis and a direct comparison between time-resolved spectra obtained using different excitation wavelengths resulted in an energy-detrapping efficiency of about 15 +/- 10%.

The MAG1* 3-methyladenine DNA glycosylase gene is closely linked to the SPT15 TATA-binding TFIID gene on chromosome V-R in Saccharomyces cerevisiae

The MAG1 gene encodes a 3-methyladenine DNA glycoslyase, which is involved in DNA alkylation repair in Saccharomyces cerevisiae. The mag1 mutant is deficient in 3-methyladenine DNA glycosylase activity and shows enhanced sensitivity to several monofunctional alkylating agents. MAG1 is allelic to MMS5. This gene has been previously located on chromosome V by chromosomal hybridization. We present physical and genetic mapping data here showing that the MAG1 gene is located on chromosome V-R, proximal to and about 10 kilobase pairs away from the SPT15 gene coding for the yeast TATA-binding protein TFIID.

Evidence for Darwinian selection of the 2-micron plasmid STB locus in Saccharomyces cerevisiae

The 2-microns plasmid of industrial and laboratory strains of Saccharomyces cerevisiae exists as two main polymorphic forms designated type I and type II. Polymorphism is restricted to the 3200-bp right unique region where types I and II show approximately 10% nucleotide divergence in trans-acting REP1 and RAF loci and 30% divergence in the cis-acting STB locus. In addition, the cis-acting STB plasmid partition locus of type II plasmids varies in sequence and copy number of a 125-bp repeat. We devised chimeric and 2-microns plasmid stability experiments to evaluate the effect of STB polymorphism on plasmid fitness in amphiploid industrial and haploid laboratory strains. Reciprocal experiments of type-II STB chimeric plasmids in type-I bakers' yeast or a type-I chimeric plasmid in type-II distillers', wine, or haploid strains showed similar partition efficiencies. However, chimeric and 2-microns plasmids carrying a 250-bp STB from a type-II haploid strain had reduced fitness in a type-II industrial wine strain. These results in conjunction with molecular analyses of 2-microns-like and 2-microns plasmids indicates the coevolution of STB with trans-acting plasmid and host-cell factors.

A common element involved in transcriptional regulation of two DNA alkylation repair genes (MAG and MGT1) of Saccharomyces cerevisiae

The Saccharomyces cerevisiae MAG gene encodes a 3-methyladenine DNA glycosylase that protects cells from killing by alkylating agents. MAG mRNA levels are induced not only by alkylating agents but also by DNA-damaging agents that do not produce alkylated DNA. We constructed a MAG-lacZ gene fusion to help identify the cis-acting promoter elements involved in regulating MAG expression. Deletion analysis defined the presence of one upstream activating sequence and one upstream repressing sequence (URS) and suggested the presence of a second URS. One of the MAG URS elements matches a decamer consensus sequence present in the promoters of 11 other S. cerevisiae DNA repair and metabolism genes, including the MGT1 gene, which encodes an O6-methylguanine DNA repair methyltransferase. Two proteins of 26 and 39 kDa bind specifically to the MAG and MGT1 URS elements. We suggest that the URS-binding proteins may play an important role in the coordinate regulation of these S. cerevisiae DNA repair genes.