S-Adenosyhnethionine Synthetase-Encoding Gene Conferring Resistance to the Methionine Toxic Analogue Ethionine

eth-1: a thermosensitive allele of the Neurospora C T U S S ~ Sadenosylmethionine (AdoMet) synthetase gene that confers ethionine resistance, has been cloned and sequenced. Replacement of an aspartic amino acid residue (D48 + N4*), perfectly conserved in prokaryotic, fungal and higher eukaryotic AdoMet synthetases, was found responsible for both thermosensitivity and ethionine resistance conferred by eth1'. Gene fusion constructs, designed to overexpress eth-1' in vivo, render transformant cells resistant to ethionine. Dominance of ethionine resistance was further demonstrated in eth-l+/eth-l' partial diploids carrying identical gene doses of both alleles. Heterozygous eth-l+/eth-I' cells have, at the same time, both the thermotolerance conferred by eth-1' and the ethionine-resistant phenotype conferred by eth1'. AdoMet levels and AdoMet synthetase activities were dramatically decreased in heterozygous eth-l+/ eth-1' cells. We propose that this negative effect exerted by eth-1' results from the in vivo formation of heteromeric eth-l+/eth-I' AdoMet synthetase molecules. E THIONINE is an analogue of methionine that is toxic to cells (ALIX 1982). Mechanisms underlying ethionine toxicity have been mainly explored in simple cellular systems by the isolation of mutant organisms resistant to this drug. Particularly informative have been the studies made on Eth' (ethionine-resistant) mutants of the enteric bacteria Escherichia coliand Salmonella thyphimurium (LAWRENCE et al. 1968; GREENE et al. 1970; HOL LOWAY et al. 1970; HOBSON and SMITH 1973; HAFNER et al. 1977; SAINT-GIRONS et al. 1988), the yeast Saccharomyces cmeuisiae (CHEREST et al. 1969,1973; THOMAS and SURDINKERJAN 1991), the filamentous fungus Neurospora crassa (METZENBERG et al. 1964; KERR and FLAVIN 1970; BURTON and METZENBERG 1975) and the plant Arabidopsis thalzana (INABA et al. 1994). Mutations to ethionine resistance occur mainly in the gene encoding Sadenosyl-Lmethionine synthetase (ATP:L-methionine Sadenosyltransferase, EC 2.5.1.6; CANTONI 1953). This enzyme catalyzes the conversion of L-methionine and ATP to S adenosylmethionine (AdoMet), which is the major cellular donor of methyl groups in a variety of biochemical pathways (TABOR and TABOR 1984). The eth-1' mutant of N. crassa has been exhaustively analyzed at the genetic and biochemical levels (METThis work is dedicated to the memory of Dr. Federico A. CUmiW. Corresponding author: Alberto L. Rosa, Departamento de Quimica BiolBgica, Facultad de Ciencias Quimicas, Univenidad Nacional de CBrdoba, Ciudad Univenitaria, 5016 CBrdoba, Argentina. E-mail: [email protected] Universitk de Paris-Sud, Bat. 400, 91405 Orsay Cedex, France. Genetics 144: 145.51462 (December, 1996) Present address; Institute de GCnetique et Microbiologie, C.N.R.S., ZENBERG et al. 1964; KAPw and METZENBERC 1965; METZENBERG and PARSON 1966; BURTON and METZENBERC 1975; JACOBSON et al. 1977). It was recently proved ( ~ ~ A U T I N O et al. 1996a) that this locus encodes AdoMet synthetase, as first suggested by KERR and FLAVIN (1970). A phenotypic feature associated with this mutation, which appears to be crucial for the mechanism underlying ethionine resistance, is the overproduction of methionine (KAFTY and METZENBERC 1965). It has been proposed that AdoMet synthetase mutants overproduce methionine by the additive effect of both a lower rate of conversion of methionine to AdoMet and a derepression of some enzymes involved in the methionine biosynthetic pathway normally repressed by AdoMet (KAPW and METZENBERG 1965; GREENE et al. 1970; KERR and FLAVIN 1970; BOERJAN et al. 1994). There is strong evidence coming from Eth' mutants of the yeast S. cermzsiae that a variety of additional reasons may be involved in conferring resistance to ethionine (ALIX 1982; SHIOMI et al. 1991; THOMAS and SURDIN-KERJAN 1991; HEILAND and HILL 1993). In this work we report the molecular cloning and sequence of the N. crassa eth-1' allele. In addition, we show that ethionine resistance conferred by eth-1' is dominant. Analyses of heterozygous strains suggest that dominant ethionine resistance could be mediated by the in vivo formation of heteromeric (eth-l+/eth-l') AdoMet synthetase molecules. MATERIALS AND METHODS Strains, growth conditions and crosses: N. crassa strains ethI T A and a (FGSC #1212 and #1220), wild type 74OR23-1A 1456 J. L. Barra, M. R. Mautino and A. L. Rosa (FGSC #987) and 740R8-la (FGSC #988), T(I+ ZZ)39311 to1 tqb-4, A (FGSC #2985) were obtained from the Fungal Genetics Stock Center (FGSC). Standard N. crassa methodologies were used according to DAVIS and DE SEWS (1970). Growth rates were determined in race tubes (RYAN et al. 1943). Construction of partially duplicated strains, Dp39311 (eth-l+/eth1') to1 trp-4 A/a , was done essentially as described (PERKINS 1972). The tol mutation (PERKINS et al. 1982) was included to suppress the vegetative incompatibility associated with the mating type alleles A and a. methionine (Sigma) was added to solid and liquid media at 50-250 pg/ml, when indicated. L-tryptophan was added at 250 pg/ml. DNA manipulations and protoplast transformation: Total N. crassa DNA was purified and subjected to Southern blot analysis as described (HAEDO et al. 1992). Plasmid DNA was purified and analyzed using standard procedures (SAMBROOK et al. 1989). Protoplasts were prepared and transformed following the protocol of VOLLMER and YANOFSKY (1986). Cotransformations experiments were carried out using 1 pg of pBT6 DNA (ORBACH et al. 1986; MAUTINO et al. 1993). Benomyl at 0.75 pg/ml was included in the bottom agar to select for transformant cells. Probes were oligolabelled according to the method of FEINBERG and VOGELSTEIN (1983). [cu-~'P]dATP (3000 Ci/mmol) was from NEN-DuPont. PCR amplification, molecular cloning and DNA sequencing of the e h Z ' allele: Degenerate oligonucleotides d l / d ? and PCR were used to amplify a N. crassa genomic fragment corresponding to almost all of the open reading frame of the AdoMet synthetase encoded by the &-I' allele. The sequence of these oligonucleotides, which were designed for the cloning of the AdoMet synthetase gene of A. immersus (MAUTINO et al. 1996b), are 5'CGGGATCCGTIGGWGAIGGTCAYCCI3' and 5'CCCTCGAGAACTTSAIAGYCTTIGGCTT3'. PCR reactions were performed with Taq polymerase (Perkin Elmer) using 100 ng of genomic DNA from the mutant eth-1'. Reactions were carried out for 35 cycles at 94" for 45 sec, 60" for 30 sec, and 72" for 1 min 20 sec, in a 50-pl reaction volume. An amplified fragment of 1150 bp, identical in size to that obtained when using wild-type DNA as the template, was gel purified (Prep-A-Gene; BioRad), digested with BamHI and cloned into the BamHI-EcoRV sites of pBKS( +) (Stratagene). Unidirectional deletions of several independent clones were made with Exonuclease I11 (HENIKOFF 1987), after digestion with the enzymes HindIII and ApaI, which do not cut into the insert fragment. Deletion clones were sequenced by the chain termination method (SANGER et al. 1977) using the reverse sequencing primer and Sequenase (USB). Multiple sequence alignment was performed with the program PILEUP and PRE?TyBOX of the GCG sequence analysis package (DEVEREUX et al. 1984). The nucleotide sequence reported in this work has been submitted to GenBank with accession number U36761. Construction of the pETH expression plasmids carrying the mutant AdoMet synthetase (eth-Z') gene fused to heterologous promoters: Two alternative constructs, pETH-N and pETHA, carrying eth-1' coding sequences under the control of the gene promoters from either the N. crassa cfp (pyruvate decarboxylase) (ALVAREZ et al. 1993; E. TEMPORINI and A. ROSA, unpublished results) or the A. nidulans trpC (STABEN et al. 1989) genes, respectively, were made from plasmids pSGT-N and pSGT-A that were previously designed to overexpress the wild-type allele of the AdoMet synthetase gene (MAUTINO et al. 1996a). To construct pETH-N and pETH-A, a 315-bp BstEII restriction fragment from the wild-type AdoMet synthetase gene was replaced by the corresponding fragment of the eth1' allele that includes a G + A transition in position +142. Measurements of AdoMet synthetase activity and cellular AdoMet contents: AdoMet synthetase was partially purified 30