Arginine-specific repression in Saccharomyces cerevisiae: kinetic data on ARG1 and ARG3 mRNA transcription and stability support a transcriptional control mechanism.

A specific repression mechanism regulates arginine biosynthesis in Saccharomyces cerevisiae. The involvement of regulatory proteins displaying DNA-binding features and the location of an operator region between the TATA box and the transcription start of the structural gene ARG3 suggest that this mechanism operates at the level of transcription. A posttranscriptional mechanism has, however, been proposed to account for the conspicuous lack of proportionality between ARG3 mRNA steady-state levels (as determined by Northern [RNA] assays; F. Messenguy and E. Dubois, Mol. Gen. Genet. 189:148-156, 1983) and the cognate enzyme activities. In this work, we have analyzed the time course of the incorporation of radioactive precursors into ARG1 and ARG3 mRNAs and the kinetics of their decay under different regulatory statuses. The results (expressed in terms of relative mRNA levels, relative transcription rates, and mRNA half-lives) give the picture expected from a purely transcriptional control. A similar analysis of expression of the gene CPA1, for which a translational regulation by arginine has been clearly demonstrated (M. Werner, A. Feller, F. Messenguy, and A. Piérard, Cell 49:805-813, 1987), indicates that this gene is also partly regulated at the transcriptional level by the ARGR repressor system. Moreover, the half-life of CPA1 mRNA is reduced twofold in the presence of excess arginine; we suggest that this could be inherent in the mechanism of translational regulation of CPA1.

Characterization of the DNA target site for the yeast ARGR regulatory complex, a sequence able to mediate repression or induction by arginine.

We have determined the sequences and positions of the cis elements required for proper functioning of the ARG3 promoter and proper arginine-specific control. A TATA box located 100 nucleotides upstream of the transcription start was shown to be essential for ARG3 transcription. Two sequences involved in normal arginine-mediated repression lie immediately downstream of the TATA box: an essential one (arginine box 1 [AB1]) and a secondary one (arginine box 2 [AB2]). AB1 was defined by saturation mutagenesis and is an asymmetrical sequence. A stringently required CGPu motif in AB1 is conserved in all known target sites of C6 zinc cluster DNA-binding proteins, leading us to propose that AB1 is the binding site of ARGRII, another member of the C6 family. The palindromic AB2 sequence is suggested, on the basis of published data, to be the binding site of ARGRI, possibly in heterodimerization with MCM1. AB2 and AB1 correspond respectively to the 5' and 3' halves of two adjacent similar sequences of 29 bp that appear to constitute tandem operators. Indeed, mutations increasing the similarity of the other halves with AB1 and AB2 cause hyperrepression. To mediate repression, the operator must be located close to the transcription initiation region. It remains functional if the TATA box is moved downstream of it but becomes inoperative in repression when displaced to a far-upstream position where it mediates an arginine and ARGR-dependent induction of gene expression. The ability of the ARG3 operator to act either as an operator or as an upstream activator sequence, depending on its location, and the functional organization of the anabolic and catabolic arginine genes suggest a simple model for arginine regulation in which an activator complex can turn into a repressor when able to interfere sterically with the process of transcription initiation.

General amino acid control and specific arginine repression in Saccharomyces cerevisiae: physical study of the bifunctional regulatory region of the ARG3 gene.

To characterize further the regulatory mechanism modulating the expression of the Saccharomyces cerevisiae ARG3 gene, i.e., the specific repression by arginine and the general amino acid control, we analyzed by deletion the region upstream of that gene, determined the nucleotide sequence of operator-constitutive-like mutations affecting the specific regulation, and examined the behavior of an ARG3-galK fusion engineered at the initiating codon of ARG3. Similarly to what was observed in previous studies on the HIS3 and HIS4 genes, our data show that the general regulation acts as a positive control and that a sequence containing the nucleotide TGACTC, between positions -364 and -282 upstream of the transcription start, functions as a regulatory target site. This sequence contains the most proximal of the two TGACTC boxes identified in front of ARG3. While the general control appears to modulate transcription efficiency, the specific repression by arginine displays a posttranscriptional component (F. Messenguy and E. Dubois, Mol. Gen. Genet. 189:148-156, 1983). Our deletion and gene fusion analyses confirm that the specific and general controls operate independently of each other and assign the site responsible for arginine-specific repression to between positions -170 and +22. In keeping with this assignment, the two operator-constitutive-like mutations were localized at positions -80 and -46, respectively, and thus in a region which is not transcribed. We discuss a hypothesis accounting for the involvement of untranscribed DNA in a posttranscriptional control.

Two c-myb proteins differing by their aminotermini exhibit different transcriptional transactivation activities (yeast/reporter-effector system).

We assayed in the yeast S. cerevisiae the transcriptional transactivation activity of the c-myb products encoded by a normal thymus cDNA and of an aminoterminally truncated version of it (minus 58 amino acids) corresponding to the cDNAs isolated from lymphoma and leukemia cells from different origins. Both proto-oncogene products were expressed under the control of the galactose inducible GAL10 promoter. The reporter system used to monitor the transactivation potential of the myb products consisted of a CYCl-lacZ gene fusion in which the UASCYC signals were replaced by one or multiple copies of the myb recognition element (mRE). As shown by Northern blot analyses and by primer extension experiments both c-myb products increase the level of beta-galactosidase transcription. Interestingly, the c-myb product corresponding to lymphoma cDNAs stimulates transcription four to five times more efficiently than does the normal thymic c-myb product.

The arginine repressor of Escherichia coli K-12 makes direct contacts to minor and major groove determinants of the operators.

In order to gain further insight into the molecular mechanism of arginine-dependent operator recognition by the hexameric Escherichia coli arginine repressor we have probed protein-DNA interactions in vitro and in vivo. We have extensively applied the chemical modification-protection and premodification-interference approach to two operators, the natural operator overlapping the P2 promoter of the carAB operon and a fully symmetrical consensus sequence. Backbone contacts were revealed by hydroxyl radical footprinting and phosphate ethylation interference. Base-specific contacts to purines and pyrimidines were revealed by methylation protection and premodification interference, KMnO4 and NH2OH.HCl-specific modification of thymine and cytosine residues, base-removal (depurination and depyrimidation), and base substitution (uracil and inosine). Additional information on the groove specificity of repressor binding was obtained by small ligand binding interference (distamycin and methyl green). In vivo, we measured the effects on the repressibility of 24 single base-pair substitutions obtained by saturation mutagenesis of half an Arg box in the carAB operator. The results of these experiments point to the conclusion that a hexameric arginine repressor molecule covers four turns of the helix, makes base-specific contacts to at least one guanine (G4 or G4') and two thymine (T3, T13', or T3', T13) residues in each one of four consecutive major grooves on one face of the helix and with four A-T/T-A base-pairs, comprising the adenine residues A9, 9', 12, 12' and the thymine residues T10, 10', 11, 11', in the two outermost minor grooves of the operator, on the very same face of the DNA molecule. The hydrophobic 5-methyl groups of four thymine residues (T3, 3', 13, 13') in each Arg box contribute to major groove-specific recognition via hydrophobic and/or van der Waals interactions. The importance of minor groove contacts was further supported by the drastic effect of distamycin binding interference. In vivo, the most pronounced drops in repressibility were occasioned by mutations at positions 10 (A-->G or C), 11 (T-->A or G) and 12 (A-->G, T or C).

On the role of the Shine-Dalgarno sequence in determining the efficiency of translation initiation at a weak start codon in the car operon of Escherichia coli K12.

Translation of the carA gene is efficiently initiated at the intrinsically weak UUG codon. A single nucleotide substitution changing the Shine-Dalgarno box of carA (GGAGG) into the sequence TGAGG reduces translation of carA sevenfold. This result supports the view that extensive complementarity between the Shine-Dalgarno sequence and 16S RNA contributes significantly to the efficiency of translation when the latter starts at a weak initiation triplet.

carP, a novel gene regulating the transcription of the carbamoylphosphate synthetase operon of Escherichia coli.

The carAB operon, encoding carbamoylphosphate synthetase (CPSase; EC 6.3.5.5) is transcribed from two tandem promoters. The upstream promoter (P1) is controlled by pyrimidines and the downstream promoter (P2) is controlled by arginine. We have isolated a new type of constitutive mutation (carP) that specifically affects the control of the pyrimidine-sensitive promoter but does not appear to influence other genes of the pyrimidine pathway. The carP mutation acts in trans and is dominant, which suggests that the carP product is an activator of car transcription. The downstream promoter P2, which is repressed by arginine, overlaps two operator modules characteristic of the arginine regulon. We have isolated two operator-constitutive mutations that specifically affect P2; both map in the upstream ARG box at a strongly conserved position.

Molecular interactions in the control region of the carAB operon encoding Escherichia coli carbamoylphosphate synthetase.

The control region of the carAB operon, encoding carbamoylphosphate synthetase, comprises two tandem promoters (P1, upstream and P2, downstream) located 67 base-pairs apart and repressed respectively by pyrimidines and arginine. RNA polymerase and pure arginine repressor bind to the P2 region in mutually exclusive ways. Repressor protects the two adjacent palindromic ARG boxes overlapping P2 against DNase I. Binding of RNA polymerase to P1 is abnormal; the region protected against DNase I is shifted upstream by about 20 nucleotides with respect to the position expected from the transcription startpoint. This pattern is not due to interference with polymerase binding at P2, since it is observed also in the presence of repressor and on an isolated P1 region. Binding of RNA polymerase is relatively weak and heparin-sensitive suggesting that, in vivo, an ancillary factor is required to promote the formation of an open complex. S1 nuclease mapping experiments show that the simultaneous presence of pyrimidines and arginine represses the downstream arginine-specific promoter (P2) more efficiently than arginine alone. This effect is not due to a direct regulatory interaction between pyrimidines and P2, since it is not observed when P1 is inactivated by insertion mutations or partial deletion. It has been shown that transcription initiated at P1 can proceed even when arginine represses P2. We therefore suggest that P2 operator-arginine repressor complex is destabilized by RNA polymerase binding at P1 or transcription from P1. We describe a novel technique to select for expression-down mutants in a lac fusion context.

Pyrimidine regulation of the Escherichia coli and Salmonella typhimurium carAB operons: CarP and integration host factor (IHF) modulate the methylation status of a GATC site present in the control region.

By measuring the protection against Dam methylase modification of a GATC sequence located 106 bp upstream of the startpoint of promoter P1 in the control region of the carAB operon (encoding carbamoylphosphate synthetase) we have obtained evidence for a direct correlation between the degree of in vivo occupancy of a specific regulatory target site and the repressibility of the P1 promoter by pyrimidine residues. A high uridine nucleotide pool as well as binding of the carP (alias xerB/pepA) gene product and of the integration host factor (IHF) to the carAB control region are prerequisites to observe this in vivo protection. Purified CarP binds in vitro to the carAB control region and protects against DNase I two approximately 25 bp long stretches, one of which is located just downstream of the GATC sequence. Mutations in this site strongly impair the pyrimidine regulation of the P1 promoter and the interference with Dam methylase modification. These processes are also strongly impaired in the absence of integration host factor and in mutants affected in the IHF site located some 200 bp upstream of this Dam methylase modification site. IHF therefore exerts at least part of its antagonistic effects on P1, i.e. increased expression in minimal medium but increased repression in the presence of pyrimidine residues, indirectly by influencing the formation or the stability of a particular protein-DNA complex. Furthermore, we demonstrate that the distance separating the IHF and Dam methylase target sites is crucial for the in vivo protection and for pyrimidine-mediated regulation of the promoter expression. Mutations altering this distance result in severe reductions of the degree of in vivo protection and, concomitantly, of the repressibility by pyrimidine residues of promoter P1 activity in a way indicative of the formation of a complex nucleoprotein structure. Since neither IHF nor CarP require pyrimidine residues to bind to the carAB control region, at least not in vitro, it is tempting to suggest that IHF and CarP-induced bending and looping provide changes in DNA topology that are required for assembling a specific pyrimidine-dependent nucleoprotein complex that modulates P1 activity.

carP, involved in pyrimidine regulation of the Escherichia coli carbamoylphosphate synthetase operon encodes a sequence-specific DNA-binding protein identical to XerB and PepA, also required for resolution of ColEI multimers.

The carP gene involved in pyrimidine-specific regulation of the upstream P1 promoter of the Escherichia coli carAB operon has been cloned in vivo on a mini-Mu replicon, sequenced and shown to be identical to the xerB (pepA) gene encoding aminopeptidase A, a protein also involved in the Xer-mediated site-specific recombination at ColEI cer. The trans-dominant allele carP6 was cloned as well and shown to bear a single G-->A transition that converts the TGG codon (Trp473) into a TAG amber stop codon. The truncated mutant protein, missing the 31 C-terminal amino acid residues, was shown to be partially active; in the multicopy state the carP6 allele can restore pyrimidine repressibility of the carAB promoter P1. The trans-dominant character of the single copy carP6 allele was found to be suppressed in the presence of multiple copies of the wild-type gene. The carP (pepA) control region was sequenced and transcription shown to be initiated at three promoters, the most upstream one of which was shown to be subject to negative autoregulation. The aminopeptidase activity of CarP (PepA) was found to be dispensable for its role in pyrimidine-mediated repression of carAB transcription. CarP (PepA) was shown to be a sequence-specific DNA-binding protein that does not require, at least not in vitro, any pyrimidine cofactor to bind to the DNA. Mobility-shift and DNase I footprinting experiments have revealed a specific binding of purified CarP (PepA) to two sites in each one of the control regions of the E. coli and Salmonella typhimurium carAB operons and to a single site in the carP (pepA) control region. We propose that integration host factor and CarP/PepA-induced structural modifications in the carAB control region cause conformational changes required to assemble a pyrimidine-specific nucleo-protein regulatory complex.

Arginine regulon of Escherichia coli K-12. A study of repressor-operator interactions and of in vitro binding affinities versus in vivo repression.

The 12 genes which in E. coli K-12 constitute the arginine regulon are organized in nine transcriptional units all of which contain in their 5' non-coding region two 18 bp partially conserved imperfect palindromes (ARG boxes) which are the target sites for binding of the repressor, a hexameric protein. In vitro binding experiments with purified repressor (a gift from W. K. Maas) were performed on the operator sites of four genes, argA, argD, argF, argG, and of two operons, carAb and the bipolar argECBH cluster. A compilation of results obtained by DNase I and hydroxyl radical footprinting clearly indicates that in each case the repressor binds symmetrically to four helical turns covering adjacent pairs of boxes separated by 3 bp, but to one face of the DNA only. Methylation protection experiments bring to light major base contacts with four highly conserved G residues symmetrically distributed in four consecutive major grooves. Symmetrical contacts in the minor groove with A residues have also been identified. Stoichiometry experiments suggest that a single hexameric repressor molecule binds to a pair of adjacent ARG boxes. Although the wild-type operator consists of a pair of adjacent ARG boxes separated by 3 bp (except argR where there are only 2 bp), repressor can bind to a single box but with a greatly reduced affinity. Therefore, adjacent boxes behave co-operatively with respect to the Arg repressor binding, in the sense that the presence of one box largely stimulates the binding of the properly located second box. The optimal distance separating two boxes is 3 bp, but one bp more or less does not abolish this stimulation effect. However, it is completely abolished by the introduction of two or more additional bp unless a full helical turn is introduced. Large variations in the in vivo repression response between individual arginine genes or a wild-type gene and cognate Oc type mutants are not reflected by similar differences in the in vitro binding results where only small differences are observed. The significance of this lack of correlation is discussed.

pyrH-encoded UMP-kinase directly participates in pyrimidine-specific modulation of promoter activity in Escherichia coli.

The carAB operon of the enterics Escherichia coli K-12 and Salmonella typhimurium LT2, encoding the sole carbamoylphosphate synthetase (CPSase) of these organisms, is transcribed from two promoters in tandem, carP1 upstream and carP2 downstream, repressed respectively by pyrimidines and arginine. We present evidence that the pyrH gene product (the hexameric UMP-kinase) directly participates in the pyrimidine-specific control of carP1 activity. Indeed, we have isolated in E. coli a particular type of pyrH mutation (pyrH41) that retains a quasi-normal UMP-kinase activity, but yet is impaired in the pyrimidine-specific repression of the P1 promoter of the carAB operon of E. coli and of S. typhimurium. Moreover, the pyrimidine-dependent inhibition of in vivo Dam methylase modification of adenine -106 upstream of the carP1 promoter is altered in this pyrH mutant. The recessive pyrH41 allele bears a single C-G to A-T transversion that converts alanine 94 into glutamic acid (A94E). Although overexpression of pyrH41 results in UMP-kinase levels far above that of a wild-type strain, pyrimidine-specific repression of the carAB operon is not restored under these conditions. Similarly, overexpression of the UMP-CMP-kinase gene of Dictyostelium discoideum in the pyrH41 mutant does not restore pyrimidine-mediated control of carP1 promoter activity, in spite of the elevated UMP-kinase activity measured in such transformants. These results indicate that besides its catalytic function in the de novo pyrimidine biosynthesis, E. coli UMP-kinase fulfils an additional, but previously unrecognized role in the regulation of the carAB operon. UMP-kinase might function as the real sensor of the internal pyrimidine nucleotide pool and act in concert with the integration host factor (IHF) and aminopeptidase A (PepA alias CarP and XerB) in the elaboration of the complex nucleoprotein structure required for pyrimidine-specific repression of carP1 promoter activity.

Structure-function relationship in allosteric aspartate carbamoyltransferase from Escherichia coli. I. Primary structure of a pyrI gene encoding a modified regulatory subunit.

In a previous article, we have identified a lambda bacteriophage directing the synthesis of a modified aspartate carbamoyltransferase lacking substrate-co-operative interactions and insensitive to the feedback inhibitor CTP. These abnormal properties were ascribed to a mutation in the gene pyrI encoding the regulatory polypeptide chain of the enzyme. We now report the sequence of the mutated pyrI and show that, during the generation of this pyrBI-bearing phage, six codons from lambda DNA have been substituted for the eight terminal codons of the wild-type gene. A model is presented for the formation of this modified pyrI gene during the integrative recombination of the parental lambda phage with the Escherichia coli chromosome. An accompanying paper emphasizes the importance of the carboxy-terminal end of the regulatory chain for the homotropic and heterotropic interactions of aspartate carbamoyltransferase.

Structure-function relationship in allosteric aspartate carbamoyltransferase from Escherichia coli. II. Involvement of the C-terminal region of the regulatory chain in homotropic and heterotropic interactions.

The modified aspartate transcarbamylase (ATCase) encoded by the transducing phage described by Cunin et al. has been purified to homogeneity. In this altered form of enzyme (pAR5-ATCase) the last eight amino acids of the C-terminal end of the regulatory chains are replaced by a sequence of six amino acids coded for by the lambda DNA. This modification has very informative consequences on the allosteric properties of ATCase. pAR5-ATCase lacks the homotropic co-operative interactions between the catalytic sites for aspartate binding and is "frozen" in the R state. In addition, this altered form of enzyme is insensitive to the physiological feedback inhibitor CTP, in spite of the fact that this nucleotide binds normally to the regulatory sites. Conversely, pAR5-ATCase is fully sensitive to the activator ATP. However, this activation is limited to the extent of the previously described "primary effect" as expected from an ATCase form "frozen" in the R state. These results emphasize the importance of the three-dimensional structure of the C-terminal region of the regulatory chains for both homotropic and heterotropic interactions. In addition, they indicate that the primary effects of CTP and ATP involve different features of the regulatory chain-catalytic chain interaction area.

Mutational analysis of Escherichia coli PepA, a multifunctional DNA-binding aminopeptidase.

Escherichia coli PepA is a hexameric aminopeptidase that is also endowed with a DNA-binding activity that functions in transcription control and plasmid dimer resolution. To gain further insight into the functioning of PepA, mutants were selected on the basis of reduced repressibility of a genomic carA-lacZ fusion and studied for the various cellular processes requiring PepA, i.e. repression of the carAB operon, autoregulation, resolution of ColE1 multimers, and peptide proteolysis. The methylation status of the carAB control region was analysed in several pepA mutants and purified proteins were assayed in vitro for car operator DNA binding. This study provides a critical test of predictions advanced on the basis of the structural analysis of PepA and demonstrates the importance for DNA binding of several secondary structural elements in the N-terminal domain and near the very C terminus. By analysis of single amino acid substitutions, we could distinguish the mode of PepA action in car regulation from its action in plasmid resolution. We demonstrate that mere binding of PepA to the car control region is not sufficient to explain its role in pyrimidine-specific regulation; protein-protein interactions appear to play an important role in transcriptional repression. The multifunctional character of PepA and of an increasing number of transcriptional regulators that combine catalytic and regulatory properties, of which several participate in the metabolism of arginine and of the pyrimidines, suggests that enzymes and DNA (RNA) binding proteins fulfilling an essential primeval function may have been recruited in evolution to fulfil an additional regulatory task.

Heterotropic interactions in Escherichia coli aspartate transcarbamylase. Subunit interfaces involved in CTP inhibition and ATP activation.

In Escherichia coli aspartate transcarbamylase, each regulatory chain is involved in two kinds of interfaces with the catalytic chains, one with the neighbour catalytic chain which belongs to the same half of the molecule (R1-C1 type of interaction), the other one with a catalytic chain belonging to the other half of the molecule (R1-C4 type of interaction). In the present work, site-directed mutagenesis was used to investigate the involvement of the C-terminal region of the regulatory chain in the process of feed-back inhibition by CTP. Removal of the two last C-terminal residues of the regulatory chains is sufficient to abolish entirely the sensitivity of the enzyme to CTP. Thus, it appears that the contact between this region and the 240s loop of the catalytic chain (R1-C4 type of interaction) is essential for the transmission of the regulatory signal which results from CTP binding to the regulatory site. None of the modifications made in the R1-C4 interface altered the sensitivity of the enzyme to the activator ATP, suggesting that the effect of this nucleotide rather involves the R1-C1 type of interface. These results are in agreement with the previously proposed interpretation that CTP and ATP do not simply act in inverse ways on the same equilibrium.

Co-operative interactions between the catalytic sites in Escherichia coli aspartate transcarbamylase. Role of the C-terminal region of the regulatory chains.

In aspartate transcarbamylase (ATCase) each regulatory chain interacts with two catalytic chains each one belonging to a different trimeric catalytic subunit (R1-C1 and R1-C4 types of interactions as defined in Fig. 1). In order to investigate the interchain contacts that are involved in the co-operative interactions between the catalytic sites, a series of modified forms of the enzyme was prepared by site-directed mutagenesis. The amino acid replacements were devised on the basis of the previously described properties of an altered form of ATCase (pAR5-ATCase) which lacks the homotropic co-operative interactions between the catalytic sites. The results obtained (enzyme kinetics, bisubstrate analog influence and pH studies) show that the R1-C4 interaction is essential for the establishment of the enzyme conformation that has a low affinity for aspartate (T state), and consequently for the existence of co-operativity between the catalytic sites. This interaction involves the 236-250 region of the aspartate binding domain of the catalytic chain (240s loop) and the 143-149 region of the regulatory chain which comprises helix H3'.

Mutational analysis of the thermostable arginine repressor from Bacillus stearothermophilus: dissecting residues involved in DNA binding properties.

Recently the crystal structure of the DNA-unbound form of the full-length hexameric Bacillus stearothermophilus arginine repressor (ArgR) has been resolved, providing a possible explanation for the mechanism of arginine-mediated repressor-operator DNA recognition. In this study we tested some of these functional predictions by performing site-directed mutagenesis of distinct amino acid residues located in two regions, the N-terminal DNA-binding domain and the C-terminal oligomerization domain of ArgR. A total of 15 mutants were probed for their capacity to repress the expression of the reporter argC - lacZ gene fusion in Escherichia coli cells. Substitutions of highly conserved amino acid residues in the alpha2 and alpha3 helices, located in the winged helix-turn-helix DNA-binding motif, reduced repression. Loss of DNA-binding capacity was confirmed in vitro for the Ser42Pro mutant which showed the most pronounced effect in vivo. In E. coli, the wild-type B. stearothermophilus ArgR molecule behaves as a super-repressor, since recombinant E. coli host cells bearing B. stearothermophilusargR on a multicopy vector did not grow in selective minimal medium devoid of arginine and grew, albeit weakly, when l -arginine was supplied. All mutants affected in the DNA-binding domain lost this super-repressor behaviour. Replacements of conserved leucine residues at positions 87 and/or 94 in the C-terminal domain by other hydrophobic amino acid residues proved neutral or caused either derepression or stronger super-repression. Substitution of Leu87 by phenylalanine was found to increase the DNA-binding affinity and the protein solubility in the context of a double Leu87Phe/Leu94Val mutant. Structural modifications occasioned by the various amino acid substitutions were confirmed by circular dichroism analysis and structure modelling.

Genetic factors affecting recovery of nonpoint mutations in the region of a gene coding for ornithine transcarbamylase: involvement of both the F factor in its chromosomal state and the recA gene.

Mutants of E. coli K12 that overproduce ornithine transcarbamylase can be identified in Car- strains because they permit utilization of citrulline as a carbamyl phosphate source, due to reversal of the normal OTCase reaction; they are called Cut mutants (citrulline utilizers). Hfr strains that carry the F factor adjacent to argF (one of two duplicate genes that code for ornithine transcarbamylase in E. coli K12) yield more Cut mutants than do F+ or F- strains, or Hfr strains in which the F factor is not adjacent to argF. When Hfr strains in which the F factor is integrated adjacent to argF are made recA, they yield few Cut mutants. Many of the Cut mutants recovered from one of the Hfr strains used in the investigation (Hfr P4X) are unstable; the properties of these unstable mutations suggest that they carry aberrations in the region of the argF gene. Thus, the increased yields of Cut mutants probably result from aberrations that occur when the F factor is integrated adjacent to argF. The nature of these aberrations is not yet known. The unstable Cut mutants are to a large extent stabilized by recA; such stabilization is one of the properties of duplications. Other data indicate that the aberrations may be more complex than simple gene duplications; in particular properties of segregants and some recombinants derived from unstable Cut mutants are most easily interpreted by assuming that segregation from, and possibly formation of, the unstable mutants occurs in several stages.

Turn-on of inactive genes by promoter recruitment in Escherichia coli: inverted repeats resulting in artificial divergent operons.

We have characterized two rearrangements consisting of inverted repeats of the argE gene. The promoters (p) of argE and of argCBH face each other over an internal operator. The rearrangements were obtained as reactivations of argE in a strain harboring an argEp deletion on a lambda darg prophage. In both cases the repeat included argE and argCBHp on either side of a unique sequence; the result is a divergent operon in which each copy of argCBHp reads into the adjacent argE repeat. In one case, the pair of repeats adjoins the silent parental gene, forming a triplication (comes from leads to comes from). The other rearrangement consists of a single argE palindrome, but the whole prophage is rearranged into an inverted repeat, analogous to certain lambda dv's. Both structures could be explained by breakage of a replication fork passing argE and by inaccurate rejoining of strands. The lambda dv-like rearrangement would result from breakage at both replication forks of a phage or prophage replicating during transient release of immunity. The triplication would imply breaking of a chromosomal replication fork, formation of a cyclic intermediate by recombination between the daughter duplex molecules and reinsertion into the parental argE gene. Formation of a triplication by replication errors involving appropriate strand switchings and branch migrations can not be excluded however.