Saccharomyces cerevisiae RNase H(35) functions in RNA primer removal during lagging-strand DNA synthesis, most efficiently in cooperation with Rad27 nuclease.

Correct removal of RNA primers of Okazaki fragments during lagging-strand DNA synthesis is a critical process for the maintenance of genome integrity. Disturbance of this process has severe mutagenic consequences and could contribute to the development of cancer. The role of the mammalian nucleases RNase HI and FEN-1 in RNA primer removal has been substantiated by several studies. Recently, RNase H(35), the Saccharomyces cerevisiae homologue of mammalian RNase HI, was identified and its possible role in DNA replication was proposed (P. Frank, C. Braunshofer-Reiter, and U. Wintersberger, FEBS Lett. 421:23-26, 1998). This led to the possibility of moving to the genetically powerful yeast system for studying the homologues of RNase HI and FEN-1, i.e., RNase H(35) and Rad27p, respectively. In this study, we have biochemically defined the substrate specificities and the cooperative as well as independent cleavage mechanisms of S. cerevisiae RNase H(35) and Rad27 nuclease by using Okazaki fragment model substrates. We have also determined the additive and compensatory pathological effects of gene deletion and overexpression of these two enzymes. Furthermore, the mutagenic consequences of the nuclease deficiencies have been analyzed. Based on our findings, we suggest that three alternative RNA primer removal pathways of different efficiencies involve RNase H(35) and Rad27 nucleases in yeast.

The actin-related protein Act3p of Saccharomyces cerevisiae is located in the nucleus.

Actin-related proteins, a group of protein families that exhibit about 50% sequence identity among each other and to conventional actin, have been found in a variety of eukaryotic organisms. In the budding yeast Saccharomyces cerevisiae, genes for one conventional actin (ACT1) and for three actin-related proteins (ACT2, ACT3, and ACT5) are known. ACT3, which we recently discovered, is an essential gene coding for a polypeptide of 489 amino acids (Act3p), with a calculated molecular mass of 54.8 kDa. Besides its homology to conventional actin, Act3p possesses a domain exhibiting weak similarity to the chromosomal protein HMG-14 as well as a potential nuclear localization signal. An antiserum prepared against a specific segment of the ACT3 gene product recognizes a polypeptide band of approximately 55 kDa in yeast extract. Indirect immunofluorescence experiments with this antiserum revealed that Act3p is located in the nucleus. Nuclear staining was observed in all cells regardless of the stage of the cell cycle. Independently, immunoblotting experiments with subcellular fractions showed that Act3p is indeed highly enriched in the nuclear fraction. We suggest that Act3p is an essential constituent of yeast chromatin.

The nuclear actin-related protein of Saccharomyces cerevisiae, Act3p/Arp4, interacts with core histones.

Act3p/Arp4, an essential actin-related protein of Saccharomyces cerevisiae located within the nucleus, is, according to genetic data, involved in transcriptional regulation. In addition to the basal core structure of the actin family members, which is responsible for ATPase activity, Act3p possesses two insertions, insertions I and II, the latter of which is predicted to form a loop-like structure protruding from beyond the surface of the molecule. Because Act3p is a constituent of chromatin but itself does not bind to DNA, we hypothesized that insertion II might be responsible for an Act3p-specific function through its interaction with some other chromatin protein. Far Western blot and two-hybrid analyses revealed the ability of insertion II to bind to each of the core histones, although with somewhat different affinities. Together with our finding of coimmunoprecipitation of Act3p with histone H2A, this suggests the in vivo existence of a protein complex required for correct expression of particular genes. We also show that a conditional act3 mutation affects chromatin structure of an episomal DNA molecule, indicating that the putative Act3p complex may be involved in the establishment, remodeling, or maintenance of chromatin structures.

Three ribonucleases H and a reverse transcriptase from the yeast, Saccharomyces cerevisiae.

From the yeast, Saccharomyces cerevisiae, three proteins exhibiting ribonuclease H activity were isolated. These proteins differ in molecular weights and enzymatic properties. The two smaller ones, RNAase H(55) and RNAase H(42) are immunologically and structurally related to each other. Neither reacts with antibodies against the largest one, RNAase H(70). Highly purified preparations of RNAase H(70) contain two polypeptides (Mr 70,000 and 160,000) and display reverse transcriptase activity. Deletion of part of the gene for the 160 kDa polypeptide results in mutants possessing about twice the amount of DNA as do wild-type cells. DNA polymerase stimulating activity resides in the 70,000 polypeptide. The processivity of yeast DNA polymerase A(I) does not change in presence of that protein. Possible functions of RNAases H are discussed.

Ribonucleases H of retroviral and cellular origin.

Ribonucleases H (RNases H) are enzymes which catalyse the hydrolysis of the RNA-strand of an RNA-DNA hybrid. Retroviral reverse transcriptases possess RNase H activity in addition to their RNA- as well as DNA-dependent DNA-polymerizing activity. These enzymes transcribe the viral single stranded RNA-genome into double stranded DNA, which then can be handled by the host cell like one of its own genes. Various, sometimes highly repeated, sequences related to retroviruses and like these encompassing two separate domains, one of which potentially codes for a DNA polymerizing, the other for an RNase H activity, are found in genomes of uninfected cells. In addition proteins coded for by cellular genes (e.g. from E. coli and from yeast) are known, which exhibit RNase H activity, the biological function of which is not fully understood. In the light of these facts the question of whether retroviral RNases H could be promising targets for antiviral drugs is discussed.

Pedigree analyses of yeast cells recovering from DNA damage allow assignment of lethal events to individual post-treatment generations.

Haploid cells of Saccharomyces cerevisiae were treated with different DNA damaging agents at various doses. A study of the progeny of individual such cells (by pedigree analyses up to the third generation) allowed the assignment of lethal events to distinct post treatment generations. By microscopically inspecting those cells which were not able to form visible colonies we could discriminate between cells dying from immediately effective lethal hits and those generating microcolonies (three to several hundred cells) probably as a consequence of lethal mutation(s). The experimentally obtained numbers of lethal events (which we call apparent lethal fixations) were mathematically transformed into mean probabilities of lethal fixations as taking place in cells of certain post treatment generations. Such analyses give detailed insight into the kinetics of lethality as a consequence of different kinds of DNA damage. For example, X-irradiated cells lost viability mainly by lethal hits (which we call 00-fixations); only at a higher dose also lethal mutations fixed in the cells that were in direct contact with the mutagen (which we call 0-fixations), but not in later generations, occurred. Ethyl methanesulfonate (EMS)-treated cells were hit by 00-fixations in a dose dependent manner; 0-fixations were not detected for any dose of EMS applied; the probability for fixation of lethal mutations was found equally high for cells of the first and second post treatment generation and, unexpectedly, was well above control in the third post-treatment generation. The distribution of all sorts of lethal fixations taken together, which occurred in the EMS-damaged cell families, was not random.(ABSTRACT TRUNCATED AT 250 WORDS)

Isolation of a cDNA clone for human NAD+: protein ADP-ribosyltransferase.

NAD+:Protein ADP-ribosyltransferase (EC 2.4.2.30) (ADPRT) was purified from human placenta by affinity chromatography. With the purified enzyme specific antibodies were raised and partial amino acid sequences were determined. To one of the amino acid sequences corresponding oligonucleotides were synthesized. A sized HeLa lambda gt11 cDNA library was constructed and screened. Positive clones were characterized to be ADPRT specific by immuno- and hybridization techniques. Clone ADPRT-G8 reacted with affinity chromatographically purified specific antibodies and with two specific oligonucleotides. The DNA of this clone detected an mRNA of about 4 kb, sufficient in size to code for the ADPRT with an Mr of 116,000. Partial sequence analysis of this clone confirmed its identity by revealing sequences which code for peptides which were found in cyanogen bromide (CNBr) fragments of the purified enzyme. The ADPRT-G8 clone was characterized with respect to its restriction pattern. The cloned ADPRT cDNA now opens the possibility to investigate the role of this enzyme in control of cellular functions.

On the origins of genetic variants.

Two contrasting mechanisms responsible for the creation of genetic variants are described: one is the manifestation of the limited accuracy of the cellular machinery for DNA replication, the other results from the ability of cells to repair damaged DNA. Replication-dependent variants and those caused by episodical DNA damage enhance the probability that a small fraction of a cell population may survive a sudden (physical or biological) change of environmental conditions. Replication-independent variants arise during persistent but not immediately lethal stress (e.g. starvation) of a non-dividing population. The variants observed under these conditions are of selective advantage because they are able to cope with the particular stress situation. The molecular basis of their creation is a matter of intensive debate.

Yeast ribonuclease H(70) cleaves RNA-DNA junctions.

A specific substrate, M13 DNA:RNA-[32P]DNA, was synthesized to investigate the mode of cleavage of enzymes with RNase H activity. RNase H(70) from Saccharomyces cerevisiae hydrolyzes the phosphodiester bond at the RNA-DNA junction of this substrate, thereby producing a 5'-monophosphate-terminated polydeoxyribonucleotide and 3'-hydroxyl-terminated oligoribonucleotides.

Poly ADP-ribosylation--a cellular emergency reaction?

We propose that the activation of poly(ADP-ribose) synthetase by DNA damage serves to decrease rapidly and transiently the cellular level of NAD (by production therefrom of poly ADP-ribose). The result is a slow-down of energy-requiring reactions, in particular of replicative DNA synthesis giving cells more time to repair the damage. We do not attribute any specific role to poly ADP-ribosylated proteins in this reaction beyond their action as acceptors for poly ADP-ribose.

Yeast RNase H(35) is the counterpart of the mammalian RNase HI, and is evolutionarily related to prokaryotic RNase HII.

We cloned the Saccharomyces cerevisiae homologue of mammalian RNase HI, which itself is related to the prokaryotic RNase HII, an enzyme of unknown function and previously described as having minor activity in Escherichia coli. Expression of the corresponding yeast 35 kDa protein (named by us RNase H(35)) in E. coli and immunological analysis proves a close evolutionary relationship to mammalian RNase HI. Deletion of the gene (called RNH35) from the yeast genome leads to an about 75% decrease of RNase H activity in preparations from the mutated, still viable cells. Sequence comparison discriminates this new yeast RNase H from earlier described yeast enzymes, RNase H(70) and RNase HI.

Purification of Saccharomyces cerevisiae RNase H(70) and identification of the corresponding gene.

We purified Saccharomyces cerevisiae RNase H(70) to homogeneity, using an optimized chromatographic purification procedure. Renaturation gel assay assigned RNase H activity to a 70 kDa polypeptide. Sequencing of tryptic peptides identified the open reading frame YGR276c on chromosome VII of the S. cerevisiae genome as the corresponding gene, which encodes a putative polypeptide of molecular mass of 62849. We therefore renamed this gene RNH70. Immunofluorescence microscopy using a RNH70-EGFP fusion construct indicates nuclear localization of RNase H(70). Deletion of RNH70 from the yeast genome did not result in any serious phenotype under the conditions tested. Homology searches revealed striking similarity with a number of eukaryotic proteins and open reading frames, among them the chimpanzee GOR protein, a homolog of a human autoimmune antigen, found to elicit autoimmune response in patients infected with hepatitis C virus.

Adaptive reversions of a frameshift mutation in arrested Saccharomyces cerevisiae cells by simple deletions in mononucleotide repeats.

Adaptive mutations are characterised as the outcome of an as yet unknown mechanism, which allows a few individuals of a cell population to overcome a starvation-induced cell cycle arrest and to proliferate. A release from such a non-lethal growth limitation is accomplished by mutations generated without DNA replication. Originally adaptive mutations were described in Escherichia coli, but more recently also in a simple eukaryote, the budding yeast Saccharomyces cerevisiae. We are studying the adaptive reversion of a frameshift allele which occurs when an auxotrophic yeast strain is starved for the amino acid essential for its proliferation. In this communication, we report on the DNA sequences from the locus concerned. Comparison between sequences from revertant clones which arose several days after growth arrest by starvation and those from revertants produced during proliferation shows significantly different mutation spectra: for replication-dependent revertants nucleotide gains and losses in a variety of sequence contexts are reasonably balanced, whereas for the replication-independent, i.e. adaptive, revertants mainly simple deletions in mononucleotide repeats were observed. These mutations resemble those known to originate from DNA polymerase slippage errors which were miscorrected or had escaped correction by the mismatch repair machinery. Our data present strong evidence for differences in the mechanistic origins of adaptive versus DNA replication-dependent mutations in a eukaryote. Most probably, mutations in non-replicating cells contribute to evolution, and if conserved in mammals, to human carcinogenesis.

DNA damage induced mating type switching in Saccharomyces cerevisiae.

Haploid cells of the yeast Saccharomyces cerevisiae are able to undergo a differentiation-like process: they can switch their mating type between the a and the alpha state. The molecular mechanism of this interconversion of mating types is intrachromosomal gene conversion. It has been shown in a variety of studies that mating type switching in heterothallic strains can be induced by DNA damaging agents, and that different DNA damaging agents differ in the length of incubation after treatment required for induction. Because X-rays induce switching immediately after irradiation and because the DNA double-strand break repair pathway is required for switching, the event initiating heterothallic mating type switching is likely to be a DNA double-strand break. Therefore the assay for heterothallic mating type switching may screen for the induction of DNA double-strand breaks. Several aspects indicating a relationship of mating type switching to mechanisms associated with carcinogenesis are discussed.

After a single treatment with EMS the number of non-colony-forming cells increases for many generations in yeast populations.

The course of lethal events occurring in populations of haploid Saccharomyces cerevisiae after DNA-damaging treatments was studied. After X-irradiation and after incubation with methyl methanesulfonate (MMS) populations recovered according to expectation, if one assumes successive dilution of killed cells by the proliferating survivors. However, populations treated with ethyl methanesulfonate (EMS) for many generations of proliferation contained more inviable cells than expected. This behaviour was not due to EMS or toxic reaction products remaining with the cells after treatment but to residual divisions of lethally mutated cells. In addition the data suggest that lethal fixations may occur in cells originating from later than the first generation after EMS treatment.