Arginine restriction induced by delta-N-(phosphonacetyl)-L-ornithine signals increased expression of HIS3, TRP5, CPA1, and CPA2 in Saccharomyces cerevisiae.

delta-N-(Phosphonacetyl)-L-ornithine (PALO), a transition state analog inhibitor of ornithine transcarbamylase, induced arginine limitation in vivo in Saccharomyces cerevisiae. Arginine restriction caused increased expression of HIS3 and TRP5, measured by the beta-galactosidase activity in strains carrying chromosomally integrated fusions of the promoter regions of each gene with the lacZ gene of Escherichia coli. The increase in beta-galactosidase activity induced by PALO was reversed by the addition of arginine and was dependent on GCN4 protein. These results indicate that PALO, like 3-amino-1,2,4-triazole DL-5-methyltryptophan, can be used to study the effect of limitation of a single amino acid, arginine, on the expression of genes under the general amino acid control regulatory system. Arginine deprivation imposed by PALO also caused increased expression of CPA1 and CPA2, coding respectively for the small and large subunits of arginine-specific carbamyl-phosphate synthetase. The observed increase was GCN4 dependent and was genetically separable from arginine-specific repression of CPA1 mRNA translation. The 5'-flanking regions of CPA1 (reported previously) and CPA2 determined in this study each contained at least two copies of the sequence TGACTC, shown to bind GCN4 protein. The beta-galactosidase activities expressed from CPA1- and CPA2-lacZ fusions integrated into the nuclear DNA of gcn4 mutant strains were five to six times less than in the wild type, when both strains were grown under depressed conditions. The gcn4 mutation reduced basal expression of both CPA1 and CPA2. The addition of arginine to strains containing the CPA1-lacZ fusion further reduced beta-galactosidase activity of the gcn4 mutant, indicating independent regulation of the CPA1 gene by the general amino acid control and by arginine-specific repression. In strains overproducing GCN4 protein, the translational control completely overrode transcriptional activation of CPA1 by general amino acid control.

Carbamyl phosphate synthetase III, an evolutionary intermediate in the transition between glutamine-dependent and ammonia-dependent carbamyl phosphate synthetases.

The amino acid sequence of carbamyl phosphate synthetase (CPS) III from liver of spiny dogfish shark Squalus acanthias was deduced from the nucleotide sequence of its cDNA. Alignment of the derived amino acid sequence of CPS III with sequences of rat and frog CPS I and hamster CPS II reveals a high degree of amino acid identity, indicating that CPS III shares the same common ancestral genes as CPSs I and II. All of the CPSs examined show a high conservation of sequences in the adenine nucleotide binding domains and in residues that have been implicated in catalysis. The active-site cysteine residue required for glutamine-dependent activity by CPS II is preserved in the sequence of CPS III. Nevertheless, analysis of the protein sequences indicates that CPS III is more closely related to CPS I than to CPS II. The structure of CPS III, which is composed of a single polypeptide, is consistent with the view that CPS III evolved by fusion of separate genes coding for the glutaminase and synthetase domains of the enzyme and, like other CPSs, the synthetase domain evolved by duplication and fusion of an ancestral kinase gene. These results, together with the recent finding that frog CPS I retains the active site cysteine residue in the glutaminase domain required for glutamine-dependent activity, indicate that other amino acid substitutions critical for glutamine-dependent activity preceded loss of this catalytic cysteine residue. The results described here together with earlier biochemical evidence support the view that acetylglutamate and glutamine-dependent CPS III found in invertebrates and fish species represents an intermediate in the evolution of ancestral glutamine-dependent CPS II toward the acetylglutamate and ammonia-dependent CPS I of ureotelic terrestrial vertebrates.

Yeast shuttle and integrative vectors with multiple cloning sites suitable for construction of lacZ fusions.

We report yeast/Escherichia coli shuttle vectors suitable for fusing yeast promoter and coding sequences to the lacZ gene of E. coli. The vectors contain a region of multiple unique restriction sites including EcoRI, KpnI, SmaI, BamHI, XbaI, SalI, PstI, SphI and HindIII. The region with the unique cloning sites has been introduced in both orientations with respect to lacZ and occurs proximal to the eighth codon of the gene. All the restriction sites have been phased to three different reading frames. Two series of vectors have been constructed. The first series (YEp) has two origins of replication (ori), i.e., of the yeast 2 mu circle and of the ColE1 plasmid of E. coli, and can therefore replicate autonomously in both organisms. These shuttle vectors also have the ApR gene of E. coli and either the yeast LEU2 or URA3 genes to allow for selection of both E. coli and yeast transformants. The second series of vectors (YIp) are identical in all respects to the YEp vectors except that they lack the 2 mu ori. The YIp vectors can be used to integrate lacZ fusions into yeast chromosomal DNA. None of the vectors express beta-galactosidase (beta Gal) in yeast or E. coli in the absence of inserted yeast promoter sequences. The 5'-nontranslated sequences and parts of the coding sequences of various yeast genes have been cloned into representative lacZ fusion vectors. In-frame gene fusions can be detected by beta Gal activity when either yeast or E. coli clones are plated on media containing XGal indicator. Quantitative determinations of promoter activity were made by colorimetric assay of beta Gal activity in whole cells. Fusion of the yeast CYC1 gene to lacZ in one of the vectors allowed detection of regulated expression of this gene when cells were grown under conditions of catabolite repression or derepression.

Detection of an enzyme bound gamma-glutamyl acyl ester of carbamyl phosphate synthetase of Escherichia coli.

E. coli carbamyl phosphate synthetase binds 0.2-0.4 mol equivalents of glutamine in an acid resistant form. The bound material is quantitatively released as glutamate by weak base hydrolysis and as a mixture of 12% glutamate, 10% gamma-glutamylhydroxamate, and 70% pyrrollidonecarboxylic acid by hydrolysis with hydroxylamine. These results provide direct evidence for a gamma-glutamyl acyl ester on the enzyme. The absence of the acyl ester in a mutant carbamyl phosphate synthetase with a Cys269-->Ser substitution in the glutaminase subunit further suggests that the covalent intermediate is a thioester of Cys269. Under equilibrium conditions, the Cys269Ser mutant enzyme binds glutamine with a Kd of 7 +/- 1 microM, indicating that Cys269 is essential for acyl ester formation but not for binding of glutamine.

Catalytically active monomer and dimer forms of rat liver carbamoyl-phosphate synthetase.

Purified carbamoyl-phosphate synthetase of rat liver is shown to exist in a state of rapid, reversible monomer-dimer equilibrium. The allosteric activator N-acetyl-L-glutamate displaces the equilibrium toward monomer formation. This effect is observed over a range of initial protein concentrations of 0.02-5 mg/mL. Measurements of Stokes radii by analytical gel chromatography indicate that at concentrations less than 0.1 mg/mL at 25 degrees C in the presence of all the substrates the enzyme exists as a monomer of 160000 molecular weight. A gel chromatographic method was developed to identify the active form of carbamoyl-phosphate synthetase. On the basis of analysis of the ADP boundary formed during gel chromatography, the monomer is established to be catalytically active. Active enzyme centrifugation studies confirm that the monomer is a reactive species and suggest that the dimer also functions catalytically. Under the conditions of the usual enzyme assay, carbamoyl-phosphate synthetase is mainly in the monomer form. Activation by acetylglutamate can occur at the level of the monomer and is not coupled to dissociation since the enzyme dissociates at low concentrations even in the absence of acetylglutamate. The stoichiometry of the association is observed directly in the electron microscope. The dimensions of the negatively stained particles of the enzyme in the presence or absence of substrates correspond to monomers and dimers, assuming the molecule to be a prolate ellipse. The number of monomers observed in the presence of substrate represents 86% of the total number of enzyme molecules. The average molecular weight calculated from the numbers of particles seen in negatively stained specimens of carbamoyl-phosphate synthetase is 182000. Electron microscope studies provide independent evidence for monomer--dimer interactions and show that under the conditions examined the enzyme is mainly in the monomer form.

The carB gene of Escherichia coli: a duplicated gene coding for the large subunit of carbamoyl-phosphate synthetase.

Previous genetic and biochemical studies indicate that the carB gene of Escherichia coli codes for the large subunit of carbamoyl-phosphate synthetase (EC 6.3.5.5). We have determined the nucleotide sequence of a 4-kilobase-pair cloned fragment of E. coli DNA with genetic determinants for carB. The DNA sequence is a 3,219-nucleotide-long reading frame. The polypeptide encoded by this reading frame has been verified to be the large subunit of carbamoyl-phosphate synthetase. The gene product is similar to the large subunit in its molecular weight, amino acid composition and amino-terminal residue, and carboxyl-terminal sequence. The amino acid sequence derived from the nucleotide sequence shows a highly significant homology between the amino- and carboxyl-terminal halves of the protein. We propose that the carB gene was formed by an internal duplication of a smaller ancestral gene.

Cloning of a yeast gene coding for arginine-specific carbamoyl-phosphate synthetase.

Several recombinant plasmids containing cpaII, the gene that encodes the large subunit of yeast arginine-specific carbamoyl-phosphate synthetase [carbamoyl-phosphate synthetase (glutamine-hydrolyzing), carbon-dioxide: L-glutamine amido-ligase (ADP-forming, carbamate-phosphorylating), EC 6.3.3.5], have been isolated. The plasmids were selected by transformation of a yeast strain with a mutation in the structural gene of the large subunit of carbamoyl-phosphate synthetase. By using a recombinant pool with inserts of yeast nuclear DNA of 5-20 kilobase pairs, we obtained 13 transformants. Of five transformants studied, three have been found to have stable plasmid inserts. These plasmids could be amplified in Escherichia coli and transferred back into the yeast carbamoyl-phosphate synthetase-deficient strains with concomitant complementation of the nuclear mutation. Plasmids pJL2/T1 and pJL2/T5 contain identical nuclear DNA inserts of 5.9 kilobase pairs. Although the insert of plasmid pJL2/T3 is also 5.9 kilobase pairs long, the sequence overlap with pJL2/T1 and pJL2/T5 is only 4.5 kilobase pairs long. The T3 insert has an orientation in the vector opposite to that of the T1 and T5 inserts. The recombinant plasmids with the yeast cpaII gene fail to cross-hybridize with a cloned fragment of E. coli DNA containing the carA and carB genes for the bacterial carbamoyl-phosphate synthetase.

Reaction of argininosuccinase with bromomesaconic acid: role of an essential lysine in the active site.

We have undertaken studies on bovine liver argininosuccinase (L-argininosuccinate arginine-lyase, EC 4.3.2.1) with the active site-directed reagent bromo[U-14C]mesaconic acid, an analogue of fumaric acid. Reactivity, measured by enzyme inactivation, followed pseudo-first-order kinetics, and the rate increased with reagent concentration. Argininosuccinate completely protected the enzyme against inactivation, but neither arginine nor fumarate was protective. A plot of the degree of inactivation as a function of alkyl groups incorporated was extrapolated to 4 mol per mol of enzyme, or 1 mol per active site. After large-scale alkylation of the enzyme (and digestion with trypsin), two 14C-labeled tryptic peptides were isolated. These were chemically sequenced by the Edman method. The amino acid sequences proved to be identical with regions of the deduced amino acid sequences or argininosuccinases from human and yeast sources [O'Brien, W. E., McInnes, R., Kalumuck, K. & Adcock, M. (1986) Proc. Natl. Acad. Sci. USA 83, 7211-7215; Beacham, I. R., Schweitzer, B. W., Warrick, H. M. & Carbon, J. (1984) Gene 29, 271-279]. The 14C-labeled tryptic peptide in the active site region had the sequence Gly-Leu-Glu-Xaa-Ala-Gly-Leu-Leu-Thr-Lys; Xaa represents an unknown phenylthiohydantoin derivative detected in cycle 4. The corresponding amino acid was identified as lysine-51 on the basis of sequence similarity with human and yeast amino acid sequences in this region. The reaction of the enzyme with the alkylating agent and the specific protection against inactivation by argininosuccinate suggest that this lysine residue has an essential role in the binding of argininosuccinate to the enzyme and, consequently, is essential for catalysis.

Substitution of Glu841 by lysine in the carbamate domain of carbamyl phosphate synthetase alters the catalytic properties of the glutaminase subunit.

In previous studies a Glu841-->Lys replacement in the carbamate phosphorylating domain located in the COOH half of the synthetase subunit of Escherichia coli carbamyl phosphate synthetase was shown to reduce overall synthesis of carbamyl phosphate by 4 orders of magnitude with either glutamine or NH3 as nitrogen donor (Guillou et al., 1992). In the present study, the mutant enzyme has been further analyzed for its glutamine hydrolytic activity. The glutaminase activity of the mutant enzyme has the following properties. (1) In the absence of other substrates the turnover number is only marginally different from that of the wild-type complex. (2) The Km for glutamine is 60 times higher than in wild-type complex and three times higher than in the separated glutaminase subunit. (3) In the present study wild-type carbamyl phosphate synthetase has been shown to catalyze glutamine hydrolysis by a mechanism involving an enzyme-bound acyl ester intermediate (gamma-glutamyl thioester). This intermediate is formed and is hydrolyzed with rates consistent with overall glutamine hydrolysis. At physiological concentrations of glutamine (1.2 mM), the steady-state concentration of gamma-glutamyl thioester is 0.3 mol/mol of wild-type enzyme. Under the same conditions, only 0.02 mol of thioester is measured in the mutant enzyme. Maximal accumulation of this covalent intermediate by the mutant enzyme required 10 times higher concentrations of free glutamine. (4) The rate of reaction with 2-amino-4-oxo-5-chloropentanoate, a glutamine analog known to specifically alkylate an active site cysteine residue, is 2 orders of magnitude slower in the mutant.(ABSTRACT TRUNCATED AT 250 WORDS)

Sequence of the small subunit of yeast carbamyl phosphate synthetase and identification of its catalytic domain.

The yeast gene CPA1 coding for the small subunit of arginine-specific carbamyl phosphate synthetase has been cloned by complementation of a cpa 1 mutant with a plasmid library of total yeast chromosomal DNA. Two of the plasmids, pJL113/ST4 and pJL113/ST15, contain DNA inserts in opposite orientations with overlapping sequences of 2.6 kilobases. The nucleotide sequence of a 2.2-kilobase region of the DNA insert carrying the CPA1 gene has been determined. The CPA1 gene has been identified to be 1233 nucleotides long and to code for a polypeptide of 411 amino acids with a calculated molecular weight of 45,358. The amino acid sequence encoded in CPA1 is homologous to the recently determined sequence of the small subunit of Escherichia coli carbamyl phosphate synthetase (Piette, J., Nyunoya, H., Lusty, C.J., Cunin, R., Weyens, G., Crabeel, M., Charlier, D., Glandsdorff, N., and Pierard, A. (1984) Proc. Natl. Acad. Sci. U.S.A. 81, 4134-4138) over the entire length of the polypeptide chain. Comparison of the amino acid sequences of the small subunits of yeast and E. coli carbamyl phosphate synthetases to the sequences of Component II of anthranilate and p-aminobenzoate synthases suggests that these amidotransferases are evolutionarily related. The most highly conserved region of the yeast and E. coli enzymes includes a cysteine residue previously found to be at the active site of Pseudomonas putida anthranilate synthase Component II (Kawamura, M., Keim, P.S., Goto, Y., Zalkin, H., and Heinrikson, R.L. (1978) J. Biol. Chem. 253, 4659-4668). Based on the observed homologies in the primary sequences of the other amidotransferases examined, we propose a 13-amino acid long sequence to be part of the catalytic domain of this class of enzymes.

Alterations in the energetics of the carbamoyl phosphate synthetase reaction by site-directed modification of the essential sulfhydryl group.

The change in reaction energetics of the bicarbonate-dependent ATPase reaction of Escherichia coli carbamoyl phosphate synthetase has been investigated for two site-directed mutations of the essential cysteine in the small subunit. Cysteine 269 has been proposed to facilitate the hydrolysis of glutamine by the formation of a glutamyl-thioester intermediate. The two mutant enzymes, C269S and C269G, along with the isolated large subunit, exhibit a 2-2.6-fold increase in the bicarbonate-dependent ATPase reaction relative to that observed for the wild type enzyme. In the presence of glutamine the overall enhancement is 3.7 and 9.0 for the C269G and C269S mutant enzymes, respectively. Carboxyphosphate is an intermediate in the bicarbonate-dependent ATPase reaction. The cause of the rate enhancements was investigated by measuring the positional isotope exchange rate in [gamma-18O4] ATP relative to the net rate of ATP hydrolysis. This ratio (Vex/Vchem) is a measure of the partitioning of the enzyme-carboxyphosphate-ADP complex. The partitioning ratio for the mutants is identical within experimental error to that observed for the wild type enzyme. This observation is consistent with the conclusion that the ground state for the enzyme-carboxyphosphate-ADP complex in the mutants is destabilized relative to the same complex in the wild type enzyme. If the increase in the absolute rate of ATP hydrolysis was due to a stabilization of the transition state for carboxyphosphate hydrolysis then the positional isotope exchange rate relative to the chemical hydrolysis rate would have been expected to decrease in the mutants.

15N isotope effects in glutamine hydrolysis catalyzed by carbamyl phosphate synthetase: evidence for a tetrahedral intermediate in the mechanism.

15N isotope effects have been measured on the hydrolysis of glutamine catalyzed by carbamyl phosphate synthetase of Escherichia coli. The isotope effect in the amide nitrogen of glutamine is 1. 0217 at 37 degrees C with the wild-type enzyme in the presence of MgATP and HCO(3)(-) (overall reaction taking place). This V/K isotope effect indicates that breakdown of the tetrahedral intermediate formed with Cys 269 to release ammonia is the rate-limiting step in the hydrolysis. A full isotope effect of 1. 0215 is also seen in the partial reaction catalyzed by an E841K mutant enzyme, whose rate of glutamine hydrolysis is not affected by MgATP and HCO(3)(-). With wild-type enzyme in the absence of MgATP and HCO(3)(-), however, the (15)N isotope effect is reduced to 1. 0157. These isotope effects are interpreted in terms of partitioning of the tetrahedral intermediate whose rate of formation is dependent upon a conformation change which closes the active site after glutamine binding and prepares the enzyme for catalysis. An Ordered Uni Bi mechanism for glutamine hydrolysis that is consistent with the isotope effects and with the catalytic properties of the enzyme is proposed.

Location of the binding site for the allosteric activator IMP in the COOH-terminal domain of Escherichia coli carbamyl phosphates synthetase.

Using UV-irradiation we cross-linked IMP, the allosteric activator of E. coli carbamyl phosphate synthetase (a heterodimer of 117.7 and 41.4 kDa subunits), to the large subunit of the enzyme. As in the native enzyme-IMP complex, the cross-linked complex was resistant to attack by trypsin. Thus, IMP is attached to its normal site and induces the normal conformational changes. Limited digestion of the [3H]IMP-labeled enzyme with V8 staphylococcal protease or with trypsin in the presence of SDS, and NH2-terminal sequencing, showed that [3H]IMP is cross-linked to the COOH-terminal 20 kDa domain of the large subunit, downstream of residue 912, supporting the proposal that this domain is specialized in effector binding and regulation.

Ornithine transcarbamylase of rat liver. Kinetic, physical, and chemical properties.

Ornithine transcarbamylase of rat liver has been purified to homogeneity. The purified enzyme of specific activity 870 to 920 focuses as a single protein at pH 7.2. At pH 7.7, the Km for carbamyl phosphate is 0.026 mM, and the Km for ornithine is 0.04 mM. The inhibition constants of a number of amino acids that act as competitive inhibitors of the enzyme are reported. The native enzyme of Mr = 112,000 is composed of three subunits of Mr = 39,600 +/- 1,000. Chemical evidence indicates that the subunits are identical in amino acid composition and amino acid sequence. The amino acid sequence of the NH2-terminal region of ornithine transcarbamylase is Ser-Gln-Val-Gln-Leu-Lys-Gly-Ser-Asp-Leu-Leu-Thr-Leu-Lys-Asn-(Phe)-X-Thr-X-Glu-Ile-Gln-Tyr-Met-.

Catalytic domains of carbamyl phosphate synthetase. Glutamine-hydrolyzing site of Escherichia coli carbamyl phosphate synthetase.

We present evidence that cysteine 269 of the small subunit of Escherichia coli carbamyl phosphate synthetase is essential for the hydrolysis of glutamine. When cysteine 269 is replaced with glycine or with serine by site-directed mutagenesis of the carA gene, the resulting enzymes are unable to catalyze carbamyl phosphate synthesis with glutamine as nitrogen donor. Even though the glycine 269, and particularly the serine 269 enzyme bind significant amounts of glutamine, neither glycine 269 nor serine 269 can hydrolyze glutamine. The mutations at cysteine 269 do not affect carbamyl phosphate synthesis with NH3 as substrate. The NH3-dependent activity of the mutant enzymes was equal to that of wild-type. Measurements of Km indicate that the enzyme uses unionized NH3 rather than ammonium ion as substrate. The apparent Km for NH3 of the wild-type enzyme is calculated to be about 5 mM, independent of pH. The substitution of cysteine 269 with glycine or with serine results in a decrease of the apparent Km value for NH3 from 5 mM with the wild-type to 3.9 mM with the glycine, and 2.9 mM with the serine enzyme. Neither the glycine nor the serine mutation at position 269 affects the ability of the enzyme to catalyze ATP synthesis from ADP and carbamyl phosphate. Allosteric properties of the large subunit are also unaffected. However, substitution of cysteine 269 with glycine or with serine causes an 8- and 18-fold stimulation of HCO-3 -dependent ATPase activity, respectively. The increase in ATPase activity and the decrease in apparent Km for NH3 provide additional evidence for an interaction of the glutamine binding domain of the small subunit with one of the two known ATP sites of the large subunit.

The gene coding for carbamoyl-phosphate synthetase I was formed by fusion of an ancestral glutaminase gene and a synthetase gene.

A near full-length cDNA copy of rat carbamoyl-phosphate synthetase I (EC 6.3.4.16) mRNA has been cloned. The cDNA insert in the recombinant plasmid pHN234 is 5.3 kilobases long. Analysis of the sequence coding for carbamoyl-phosphate synthetase I indicates that the gene has arisen from a fusion of two ancestral genes: one homologous to Escherichia coli carA, coding for a glutaminase subunit, and the second homologous to the carB gene that codes for the synthetase subunit. A short amino acid sequence previously proposed to be part of the active site involved in glutamine amide nitrogen transfer in the E. coli and yeast carbamoyl-phosphate synthetases (EC 6.3.5.5) is also present in the rat enzyme. In the mammalian enzyme, however, the glutaminase domain lacks a cysteine residue previously shown to interact with glutamine. The cysteine is replaced by a serine residue. This substitution could, in part, account for the inability of mammalian carbamoyl-phosphate synthetase I to catalyze the hydrolysis of glutamine to glutamic acid and ammonia.

Membrane and cytosolic components affecting transport of the precursor for ornithine carbamyltransferase into mitochondria.

A higher molecular weight precursor (Mr = 39,000) to the liver mitochondrial matrix enzyme, ornithine carbamyltransferase (Mr = 36,000), is imported and processed by heart mitochondria in vitro in a manner similar to liver mitochondria. In both systems, however, an additional 37-kDa ornithine carbamyltransferase polypeptide appears, but this arises from nonspecific events and, therefore, does not represent a bona fide intermediate in the overall processing sequence. Our experiments demonstrate that the outer mitochondrial membrane of mitochondria contains a protease-sensitive (5 micrograms of trypsin or chymotrypsin/ml, 15 min at 2 degrees C), salt-resistant (1.0 M KCl) protein which is required to maintain import functions. In addition, functional post-translational import requires a component of the reticulocyte lysate (i.e. cytosol) that is used for initially synthesizing precursor enzyme. The component is retained by Sephadex G-25. Import of Sephadex G-25-excluded precursor is restored by fresh reticulocyte lysate but not by a combination of other additives, including Mg2+, K+, ATP, ADP, Pi, succinate, and total translation mixture (minus lysate).

In vivo synthesis of carbamyl phosphate from NH3 by the large subunit of Escherichia coli carbamyl phosphate synthetase.

The cloned carAB operon of Escherichia coli coding for the small and large subunits of carbamyl phosphate synthetase has been used to construct a recombinant plasmid with a 4.16 kilobase ClaI fragment of the car operon that lacks the major promoters, P1 and P2. The plasmid, pHN12, carries a functional carB gene. A mutant E. coli strain lacking both subunits of carbamyl phosphate synthetase when transformed with pHN12 overproduces the large subunit by 200-fold (8-10% of the cellular protein). The elevated levels of the large subunit enable the transformed cells to utilize NH3 but not glutamine as nitrogen donor for carbamyl phosphate synthesis. The large subunit has been purified from the overexpressing strain. The purified native large subunit is capable of synthesizing carbamyl phosphate from ammonia, HCO-3, and ATP. The kinetic properties of the large subunit compared with the holoenzyme indicate that the Michaelis constants of the large subunit for HCO-3 and ATP are modulated by its association with the small glutamine binding subunit.