The 1.6 A crystal structure of E. coli argininosuccinate synthetase suggests a conformational change during catalysis.

BACKGROUND: Argininosuccinate synthetase (AS) is the rate-limiting enzyme of both the urea and arginine-citrulline cycles. In mammals, deficiency of AS leads to citrullinemia, a debilitating and often fatal autosomal recessive urea cycle disorder, whereas its overexpression for sustained nitric oxide production via the arginine-citrulline cycle leads to the potentially fatal hypotension associated with septic and cytokine-induced circulatory shock. RESULTS: The crystal structure of E. coli AS (EAS) has been determined by the use of selenomethionine incorporation and MAD phasing. The structure has been refined at 1.6 A resolution in the absence of its substrates and at 2.0 A in the presence of aspartate and citrulline (EAS*CIT+ASP). Each monomer of this tetrameric protein has two structural domains: a nucleotide binding domain similar to that of the "N-type" ATP pyrophosphatase class of enzymes, and a novel catalytic/multimerization domain. The EAS*CIT+ASP structure clearly describes the binding of citrulline at the cleft between the two domains and of aspartate to a loop of the nucleotide binding domain, whereas homology modeling with the N-type ATP pyrophosphatases has provided the location of ATP binding. CONCLUSIONS: The first three-dimensional structures of AS are reported. The fold of the nucleotide binding domain confirms AS as the fourth structurally defined member of the N-type ATP pyrophosphatases. The structures identify catalytically important residues and suggest the requirement for a conformational change during the catalytic cycle. Sequence similarity between the bacterial and human enzymes has been used for providing insight into the structural and functional effects of observed clinical mutations.     

Citrulline metabolism in normal and citrullinemic human lymphocyte lines

Citrullinemia is one of the five aminoacidurias associated with the Krebs-Henseleit urea cycle. A long-term lymphocyte line (UM-21) derived from a patient with this disease and nine of ten clones of this line were found to have no activity for the enzyme argininosuccinate synthetase (AS), as demonstrated by their inability to grow in medium in which citrulline had been substituted for arginine, by their inability to incorporate arginine-C14 derived from citrulline-C14 into cellular protein, and by direct enzyme assay. One clone had normal or nearly normal argininosuccinate synthetase activity, as demonstrated by the same criteria. Nutritional "variants" able to grow logarithmically in medium containing citrulline were isolated from UM-21 and three clones. The apparent Kms of AS for citrulline in UM-21, the ten clones, the variant lines, and a normal line were measured and fell into three groups: AS in UM-21 and nine clones had no measurable apparent Km for citrulline; AS in the variant cells had apparent Kms for citrulline of approximately 20 mM; and AS in the normal cell line and one clone had apparent Kms for citrulline of 0.2 mM. The data suggest that the defect in the citrullinemic cell lines is due to a mutation in the structural gene coding for argininosuccinate synthetase.     

Structures of argininosuccinate synthetase in enzyme-ATP substrates and enzyme-AMP product forms: stereochemistry of the catalytic reaction

Argininosuccinate synthetase reversibly catalyzes the ATP-dependent condensation of a citrulline with an aspartate to give argininosuccinate. The structures of the enzyme from Thermus thermophilus HB8 complexed with intact ATP and substrates (citrulline and aspartate) and with AMP and product (argininosuccinate) have been determined at 2.1- and 2.0-A resolution, respectively. The enzyme does not show the ATP-induced domain rotation observed in the enzyme from Escherichia coli. In the enzyme-substrate complex, the reaction sites of ATP and the bound substrates are adjacent and are sufficiently close for the reaction to proceed without the large conformational change at the domain level. The mobility of the triphosphate group in ATP and the side chain of citrulline play an important role in the catalytic action. The protonated amino group of the bound aspartate interacts with the alpha-phosphate of ATP and the ureido group of citrulline, thus stimulating the adenylation of citrulline. The enzyme-product complex explains how the citrullyl-AMP intermediate is bound to the active site. The stereochemistry of the catalysis of the enzyme is clarified on the basis of the structures of tAsS (argininosuccinate synthetase from T. thermophilus HB8) complexes.     

The stimulatory effect of alloxan diabetes on citrulline formation in rabbit liver mitochondria

The effect of alloxan diabetes on citrulline formation from NH₄Cl and bicarbonate was studied in rabbit liver mitochondria incubated with glutamate or succinate as respiratory substrate, as well as with exogenous ATP in the presence of uncoupler and oligomycin. In contrast to ornithine transcarbamoylase, the activity of carbamoyl-phosphate synthetase (ammonia) was higher in mitochondria from diabetic animals than in those from normal ones. In diabetic rabbits the rates of citrulline synthesis were stimulated under all conditions studied. In contrast, levels of N-acetyglutamate, an activator of carbamoyl-phosphate synthetase (ammonia), were significantly increased only in the presence of glutamate, while the highest rates of citrulline formation occurred in uncoupled mitochondria incubated with exogenous ATP as energy source. Treatment of animals with alloxan resulted in an increase of both the intramitochondiral ATP level and the rate of adenine nucleotide translocation across the mitochondrial membrane. The results indicate that the stimulation of citrulline formation in liver mitochondria of diabetic rabbits is mainly due to an increase in carbamoyl-phosphate synthetase (ammonia) activity and an elevation of content of intramitochondrial ATP, a substrate of this enzyme.     

Effect of different dietary proteins on plasma and liver fatty acid compositions in growing rats

The activities of key glutamine and urea cycle enzymes were assayed in liver homogenates from control and chronically acidotic rats and compared with citrulline and urea productions by isolated mitochondria and intact liver slices, respectively. Glutamine-dependent urea and citrulline synthesis were increased significantly in isolated mitochondria and in liver slices; the activities of carbamoyl phosphate synthetase and arginase were unchanged and increased, respectively. Glutamine was not a precursor in the carbamoyl phosphate synthetase system, suggesting that the glutamine effect is an indirect one and that glutamine requires prior hydrolysis. Increased mitochondrial citrulline synthesis was associated with enhanced oxygen consumption, suggesting glutamine acts both as a nitrogen and fuel source. Hepatic phosphate-dependent glutaminase was elevated by chronic acidosis. The results indicate that the acidosis-induced reduction in ureagenesis and reversal from glutamine uptake to release observed in vivo are not reflections of corresponding changes in the hepatic enzyme content. Rather, when available, glutamine readily supports ureagenesis, suggesting a close coupling of hepatic glutaminase flux with citrulline synthesis.     

Regulation of argininosuccinate synthetase level by corticosteroid and pancreatic hormones during perinatal period

The urea cycle takes place in the hepatocyte of ureothelic animals. The conversion of ammonia into urea involves five reactions. The first 2 take place in the matrix of the mitochondria, the last 2 occur in the cytosol. Argininosuccinate synthetase (AS) is the third reaction of the urea cycle. It catalyses the condensation of citrulline and aspartate into arginonosuccinate. We have previously reported that rat AS activity was present in the cytosol and the outer membrane of the mitochondria. We have shown that, at the activity level, the colocation of AS was changing during fetal and neonatal development and was under the control of corticosteroid and pancreatic hormones. However, an unresolved issue was whether both AS had the same specific activity and that their location was changing during ontogenesis or that the specific activities of mitochondrial and cytosolic enzymes were different and/or modified during this period. In the present report, we compared the compartmentalization of AS activity and protein level in the fetus, the new-born and the adult rat and the role of corticosteroid and pancreatic hormones. Specific activities of both AS remained unchanged during ontogenesis. Glucocorticoids induced an increase in mitochondrial AS while glucagon appeared to induce a concomitant decrease in the level of mitochondrial AS and an increase in cytosolic AS.     

Nutritional influences on the distribution of the urea cycle: intermediates in isolated hepatocytes

After the urea cycle was proposed, considerable efforts were put forth to identify critical intermediates. This was then followed by studies of dietary and nutritional control of urea cycle enzyme activity and allosteric effectors of urea cycle enzymes. Correlation of urea cycle enzyme activity with isolated cell experiments indicated conditions where enzyme activity would be rate limiting. At physiological levels of ammonia the activation of carbamoyl-phosphate synthetase (EC 6.3.4.16) by N-acetylglutamate (NAG) is important. Various levels of NAG corresponded well with changes in the rate of citrulline and urea synthesis. Arginine was found to be an allosteric activator of N-acetylglutamate synthetase (EC 2.3.1.1). Therefore, it was possible that the rate of carbamoyl phosphate synthesis was dependent on the level of urea cycle intermediates, particularly arginine. Evidence for arginine in the regulation of NAG synthesis is not as clear as for NAG on carbamoyl phosphate synthetase I. The concentration of hepatic arginine is not necessarily an indication of the mitochondrial concentration. Only mitochondrial arginine stimulates the N-acetylglutamate synthetase. Recent studies indicate that the mitochondrial concentration of arginine is higher than the cytosolic concentration and is well above the Ka for N-acetylglutamate synthetase. Therefore, it appears that changes in arginine concentration are not physiologically important in regulating levels of NAG. However, it is possible that responses to the effector may vary with time after eating, and it may be this responsiveness that controls the level of NAG and thereby urea synthesis.     

Relative argininosuccinate synthetase mRNA levels and gene copy number in canavanine-resistant lymphoblasts

Mutants resistant to the arginine analogue, canavanine, have been isolated from two normal lymphoblast lines, MGL8B2 and MGL33. These mutants constitutively express up to 200-fold higher amounts of structurally normal argininosuccinate synthetase, the urea cycle enzyme that converts citrulline to argininosuccinate. Relative levels of argininosuccinate synthetase mRNA were compared among normal and canavanine-resistant lines using in vitro translation of poly(adenylic acid) RNA and blot hybridization of total cytoplasmic RNA to an argininosuccinate synthetase cDNA. Both of these approaches indicated that the canavanine-resistant lines contain increased steady-state levels of synthetase-specifc mRNA relative to their sensitive parents and that these were roughly correlated with levels of enzyme activity. Blot hybridization of Eco RI-digested genomic DNA preparations revealed no detectable differences in argininosuccinate synthetase structural gene copy number between normal and canavanine-resistant lymphoblasts, demonstrating that the canavanine-resistant phenotype is not caused by gene amplification.     

Identification of fumonisin B1 as an inhibitor of argininosuccinate synthetase using fumonisin affinity chromatography and in vitro kinetic studies

Fumonisin B1, a fungal mycotoxin that grows on corn and other agricultural products, alters sphingolipid metabolism by inhibiting ceramide synthase. The precise mechanism of fumonisin B1 toxicity has not been completely elucidated; however, a central feature in the cytotoxicity is alteration of sphingolipid metabolism through interruption of de novo ceramide synthesis. An affinity column consisting of fumonisin B1 covalently bound to an HPLC column matrix was used to isolate a rat liver protein that consistently bound to the column. The protein was identified as argininosuccinate synthetase by protein sequencing. The enzyme-catalyzed formation of argininosuccinic acid from citrulline and aspartate by recombinant human and rat liver argininosuccinate synthetase was inhibited by fumonisin B1. Fumonisin B1 showed mixed inhibition against citrulline, aspartate, and ATP to the enzyme. Fumonisin B1 had a Ki' of approximately 6 mM with the recombinant human argininosuccinate synthase and a Ki' of 35 mM with a crude preparation of enzyme prepared from rat liver. Neither tricarballylic acid nor hydrolyzed fumonisin B1 inhibited recombinant human argininosuccinate synthetase. This is the first demonstration of fumonisin B1 inhibition of argininosuccinate synthethase, a urea cycle enzyme, which adds to the list of enzymes that are inhibited in vitro by fumonisin B1 (ceramide synthase, protein serine/threonine phosphatase). The extent of the inhibition of argininosuccinate synthetase in cells, and the possible role of this enzyme inhibition in the cellular toxicity of FB1, remains to be established.     

Cloning and characterization of the cDNA encoding human adenylosuccinate synthetase.

Adenylosuccinate synthetase (AS) catalyzes the first committed step in the conversion of IMP to AMP. A cDNA was isolated from a human liver library which encodes a protein of 455 amino acids (M(r) of 49,925). Alignments of human, mouse, Dictyostelium discoideum and E. coli AS sequences identify a number of invariant residues which are likely to be important for structure and/or catalysis. The human AS sequence was also 19% identical to the human urea cycle enzyme, argininosuccinate synthetase (ASS), which catalyzes a chemically similar reaction. Both human liver and HeLa AS mRNA showed signals of 2.3 and 2.8 kb. An unmodified N-terminus is required for function of the human AS enzyme in E. coli mutants lacking the bacterial enzyme. The human cDNA provides a means to assess the possible role of AS abnormalities in unclassified, idiopathic cases of gout.     

Argininosuccinate synthetase activity in cultured human lymphocytes

The activity of argininosuccinate synthetase (E.C. 6.3.4.5), a urea cycle enzyme, was measured in cultured human lymphocytes using a new radioactive assay. Control cells had a maximum specific activity of 15.7±8.7 nmoles per hour per milligram of protein and an apparent K _m for citrulline of 2 × 10^–4 m, whereas cells derived from a patient with citrullinemia had no detectable activity. A nutritional variant, selected out of the citrullinemic lymphocyte population by ability to grow in citrulline, had a maximum specific activity of 10.7±3.8 nmoles/hr/mg and an apparent K _m for citrulline of 2 × 10^–2 m. These measurements confirm the observation that citrullinemia is associated with a defect in argininosuccinate synthetase activity and provide further evidence that citrullinemia is expressed in cultured lymphocytes. The emergence of a nutritional variant with a partial defect in argininosuccinate synthetase enzyme suggests that this citrullinemic patient has a heterogeneous population of cells, some totally defective and others only partially defective in argininosuccinate synthetase. The new activity assay is described in detail.This research was supported by a National Institutes of Health Training Grant (5-TO1-GM-0071) and NIH Program Project Grant (2-PO1-GM-15419).     

Citrulline immunohistochemistry for demonstration of NOS activity in vivo and in vitro.

Nitric oxide (NO), a biomolecule with major cytotoxic potency, is generated by NO synthases (NOS) utilizing l-arginine as substrate and citrulline is formed as a "side product." In brain tissue, citrulline is considered to be produced exclusively by NOS, due to the incomplete urea cycle in the brain. We aimed to characterize NOS activity by citrulline immunostaining in different cell types of the brain under in situ conditions and in slice and culture experiments. NOS-positive neurons and activated microglial cells were the most prominent citrulline-positive structures. Lack of citrulline immunoreaction in neurons of nNOS knockout mice emphasizes the dependency of citrulline positivity on NOS activity, and likewise there was no citrulline staining after application of the NOS inhibitors 7-nitroindazole and NIL. Interestingly, only a portion of NOS-containing neurons costained for citrulline. The inhibition of argininosuccinate synthetase by alpha-methyl-dl-aspartate increased the number of citrulline-positive cells, apparently due to reduction of the turnover rate of citrulline. Cells positive for NOS but negative for citrulline may indicate that the enzyme is either not activated or inhibited by cellular control mechanisms. The fact that not all citrulline-positive cells were NOS positive may be explained by an insufficient detection sensitivity or by disparate sites of citrulline production and recycling. The present results show that citrulline immunocytochemistry offers a viable and convenient means for studying NOS activity at the single-cell level to elicit its posttranslational control under physiological and pathophysiological conditions.     

Effect of growth hormone on rat liver N-acetyl-L-glutamate

Earlier studies have revealed, upon hypophysectomy, a specific increase in mitochondrial urea cycle enzymes, namely carbamyl phosphate synthetase and ornithine transcarbamylase. Administration of growth hormone to hypophysectomized rats brought these enzyme activities back to normal. Since growth hormone plays a role in the formation of citrulline and ultimately urea, in the present study its effect on the levels of N-acetyl-L-glutamate, an allosteric activator of carbamyl phosphate synthetase has been investigated. A significant increase in N-acetyl-L-glutamate concentration in rat liver on hypophysectomy and its reversal back to normal levels on growth hormone administration was reported. These results suggest that the lack of growth hormone tends to amplify urea production by the liver.     

Citrulline synthesis: Regulation by alterations in the total mitochondrial adenine nucleotide content

The total adenine nucleotide content of rat liver mitochondria was varied in vitro over a wide range in order to investigate a possible relationship between net changes in the total matrix ATP + ADP + AMP content and the overall rate of citrulline synthesis. Isolated mitochondria were specifically depleted of matrix adenine nucleotides by incubating with inorganic pyrophosphate (G. K. Asimakis and J. R. Aprille, 1980, Arch. Biochem. Biophys.203, 307–316); alternatively, matrix adenine nucleotides were increased by incubating mitochondria with 1 mm ATP at 30 °C. No exogenous ATP or ADP was included in the subsequent incubations for the determination of citrulline synthesis. Rates varied from 0.1 to 1.6 μmol citrulline/mg protein/h as a linear function of total adenine nucleotide content in the range 2–15 nmol (ATP + ADP + AMP)/mg protein. Further increases in the matrix ATP + ADP + AMP content caused no further increase in citrulline synthesis rates. Changes in the total adenine nucleotide content were reflected in proportional changes in both the ATP and ADP content of the matrix. The $\text{ATP}\text{ADP}$ ratio did not change significantly. Therefore, the variations in citrulline synthesis were most simply explained as the effect of different concentrations of ATP on the activity of carbamoyl-phosphate synthetase. It was concluded that net changes in the total adenine nucleotide content can contribute to the control of citrulline synthesis. These findings are significant in the context of recent evidence which shows that the matrix adenine nucleotide pool size is under hormonal control.     

RNA interference of argininosuccinate synthetase restores sensitivity to recombinant arginine deiminase (rADI) in resistant cancer cells

Background  

Sensitivity of cancer cells to recombinant arginine deiminase (rADI) depends on expression of argininosuccinate synthetase (AS), a rate-limiting enzyme in synthesis of arginine from citrulline. To understand the efficiency of RNA interfering of AS in sensitizing the resistant cancer cells to rADI, the down regulation of AS transiently and permanently were performed in vitro, respectively.     

Inhibition of urea cycle enzymes by lysine and saccharopine

Crude and purified preparations of argininosuccinate synthetase, argininosuccinate lyase and arginase were subjected to inhibition studies with L-lysine and saccharopine. Saccharopine proved to be the more potent inhibitor of argininosuccinate synthetase and lyase, whereas lysine had more effect on arginase. Similar results were found with pure enzyme and crude preparations. Computer analysis of the results suggested that inhibition of urea cycle enzymes by saccharopine and lysine might have contributed to the high levels of citrulline found in a human patient with saccharopinuria, a defect of saccharopine metabolism, but that this was unlikely to be the sole explanation.     

Crystal structure of argininosuccinate synthetase from Thermus thermophilus HB8. Structural basis for the catalytic action

Argininosuccinate synthetase catalyzes the ATP-dependent condensation of a citrulline with an aspartate to give argininosuccinate. The three-dimensional structures of the enzyme from Thermus thermophilus HB8 in its free form, complexed with intact ATP, and complexed with an ATP analogue (adenylyl imidodiphosphate) and substrate analogues (arginine and succinate) have been determined at 2.3-, 2.3-, and 1.95-A resolution, respectively. The structure is essentially the same as that of the Escherichia coli argininosuccinate synthetase. The small domain has the same fold as that of a new family of "N-type" ATP pyrophosphatases with the P-loop specific for the pyrophosphate of ATP. However, the enzyme shows the P-loop specific for the gamma-phosphate of ATP. The structure of the complex form is quite similar to that of the native one, indicating that no conformational change occurs upon the binding of ATP and the substrate analogues. ATP and the substrate analogues are bound to the active site with their reaction sites close to one another and located in a geometrical orientation favorable to the catalytic action. The reaction mechanism so far proposed seems to be consistent with the locations of ATP and the substrate analogues. The reaction may proceed without the large conformational change of the enzyme proposed for the catalytic process.     

Inhibition of ureagenesis by valproate in rat hepatocytes. Role of N-acetylglutamate and acetyl-CoA.

Valproate (0.5-5 mM) strongly inhibited urea synthesis in isolated rat hepatocytes incubated with 10 mM-alanine and 3 mM-ornithine. Valproate at the same concentrations markedly decreased concentrations of N-acetylglutamate, an essential activator of carbamoyl-phosphate synthetase I (EC 6.3.4.16), in parallel with the inhibition of urea synthesis by valproate. This compound also lowered the cellular concentration of acetyl-CoA, a substrate of N-acetylglutamate synthase (EC 2.3.1.1); glutamate, aspartate and citrulline were similarly decreased. Valproate in a dose up to 2 mM did not significantly affect the cellular concentration of ATP and had no direct effect on N-acetylglutamate synthesis, carbamoyl-phosphate synthetase I and ornithine transcarbamoylase (EC 2.1.3.3) activities.     

Stereochemical probes of the argininosuccinate synthetase reaction

The stereochemical course of the argininosuccinate synthetase reaction has been determined. The SP isomer of [alpha-17O,alpha-18O,alpha beta-18O]ATP is cleaved to (SP)-[16O,17O,18O]AMP by the action of argininosuccinate synthetase in the presence of citrulline and aspartate. The overall stereochemical transformation is therefore net inversion, and thus the enzyme does not catalyze the formation of an adenylylated enzyme intermediate prior to the synthesis of citrulline adenylate. The RP isomer of adenosine 5'-O-(2-thiotriphosphate) (ATP beta S) is a substrate in the presence of Mg2+, but the SP isomer is a substrate when Cd2+ is used as the activating divalent cation. Therefore, the lambda screw sense configuration of the beta,gamma-bidentate metal--ATP complex is preferred by the enzyme as the actual substrate. No significant discrimination could be detected between the RP and SP isomers of adenosine 5'-O-(1-thiotriphosphate) (ATP alpha S) when Mg2+ or Mn2+ are used as the divalent cation. Argininosuccinate synthetase has been shown to require a free divalent cation for full activity in addition to the metal ion needed to complex the ATP used in the reaction.     

The stimulation of the mitochondrial respiration by citrulline synthesis.

1. The influence of ammonia and ornithine on the oxygen uptake and the formation of citrulline was investigated with isolated rat liver mitochondria. The experiments were performed in a cytosol-like saline medium at 38 degrees C. 2. Under these conditions an increase of the respiration rate by ammonia and ornithine was observed, but a small response to external ADP, only. The missing stimulation by ADP was due to a partial inhibition of the respiratory chain by traces of zinc (approximately 1 microM) present in the medium. This inhibition was only detected at low concentrations of mitochondria. 3. For activation of respiration by ammonia plus ornithine two different processes were responsible: (i) chelation of the inhibiting zinc by ornithine, which could be prevented by EDTA; (ii) ADP production in the matrix space during formation of carbamoyl phosphate, which could be prevented by oligomycin but not by carboxyatractyloside. 4. This stimulus of the carbamoyl phosphate formation and of the equivalent citrulline synthesis on the mitochondrial respiration ran to 12% of that increase caused by phosphorylation of external ADP. The maximum rate of citrulline formation was limited by the activity of carbamoyl phosphate synthetase. 5. Added ADP suppresses the production of citrulline probably by the exchange of extramitochondrial ADP versus intramitochondrial ATP. The data suggest a common adenine nucleotide pool delivering ATP to the adenine nucleotide translocase as well as to the carbamoyl phosphate synthetase.