Search

IRAK2, a member of the interleukin-1 receptor-associated kinase (IRAK) family, has been implicated in Toll-like receptor (TLR)-mediated signaling. We generated IRAK2-deficient mice to examine its function in detail. These mice are resistant to lipopolysaccharide-induced septic shock, because of impaired TLR4-mediated induction of pro-inflammatory cytokines and chemokines. Although IRAK2 deficiency did not affect TLR4-mediated NFκB activation, a reduction of lipopolysaccharide (LPS)-mediated mRNA stabilization contributed to the reduced cytokine and chemokine production observed in bone marrow-derived macrophages from IRAK2-deficient mice. Furthermore, the ratios of LPS-induced cytokine and chemokine mRNAs in translation-active (polysomal) versus translation-inactive (free ribosomes) pools were reduced in IRAK2-deficient macrophages compared with wild type macrophages. Importantly, LPS-induced phosphorylation of MKK3/6, MNK1, and eIF4E was significantly reduced in IRAK2-deficient macrophages compared with wild type macrophages. Moreover, LPS stimulation induced an interaction of IRAK2 with TRAF6, MKK3/6, and MK2, implicating a critical role for mitogen-activated protein kinase signaling in LPS-induced IRAK2-mediated post-transcriptional control. These results reveal that IRAK2 is required for LPS-mediated post-transcriptional control of cytokine and chemokine expression, which plays an essential role in TLR4-induced septic shock.

The transcription factor STAT1 plays a pivotal role in signal transduction of type I and II interferons (IFNs). STAT1 activation leads to changes in expression of key regulatory genes encoding caspases and cell cycle inhibitors. Deficient STAT1 expression in human cancer cells and virally mediated inhibition of STAT1 function have been associated with cellular resistance to IFNs and mycobacterial infection in humans. Thus, given the relative importance of STAT1, we isolated and characterized a human STAT1 intronic enhancer region displaying IFN-regulated activity. Functional analyses by transient expression identified a repressor region and type I and II IFN-inducible elements within the STAT1 enhancer sequence. A candidate IRF-E/GAS/IRF-E (IGI) sequence containing GAAANN nucleotide repeats was shown by gel shift assay to bind to IFN regulatory factor-1 (IRF-1), but not to IFN-stimulated gene factor-3 (ISGF-3) or STAT1–3. An additional larger IGI-binding complex containing IRF-1 was identified. Mutation of the GAAANN repeats within the IGI DNA element eliminated IRF-1 binding and the IFN-regulated activity of the STAT1 intronic enhancer region. Transfection of the IFN-resistant MM96 cell line to express increased levels of IRF-1 protein also elevated STAT1, STAT2, and p48/IRF-9 expression and enhanced cellular responsiveness to IFN-β. Reciprocating regulation between IRF-1 and STAT1 genes and encoded proteins indicates that an intracellular amplifier circuit exists controlling cellular responsiveness to the IFNs.

Phosphatidylinositol 3'-kinase (PI3K) and the serine/threonine kinase AKT have critical roles in phosphorylating and transactivating the p65 subunit of nuclear factor kappaB (NF-kappaB) in response to the pro-inflammatory cytokines interleukin-1 (IL-1) and tumor necrosis factor (TNF). Mouse embryo fibroblasts (MEFs) lacking either the alpha or beta subunit of IkappaB kinase (IKK) were deficient in NF-kappaB-dependent transcription following treatment with IL-1 or TNF. However, in contrast to IKKbeta-null MEFs, IKKalpha-null MEFs were not substantially defective in the cytokine-stimulated degradation of Ikappabetaalpha or in the nuclear translocation of NF-kappaB. The IKK complexes from IKKalpha- or IKKbeta-null MEFs were both deficient in PI3K-mediated phosphorylation of the transactivation domain of the p65 subunit of NF-kappaB in response to IL-1 and TNF, and constitutively activated forms of PI3K or AKT did not potentiate cytokine-stimulated activation of NF-kappaB in either IKKalpha- or IKKbeta-null MEFs. Collectively, these data indicate that, in contrast to IKKbeta, which is required for both NF-kappaB liberation and p65 phosphorylation, IKKalpha is required solely for the cytokine-induced phosphorylation and activation of the p65 subunit of NF-kappaB that are mediated by the PI3K/AKT pathway.

Type I interferons (IFNs) are cytokines that play a central role in mediating antiviral, antiproliferative, and immunomodulatory activities in virtually all cells. These activities are entirely dependent on the interaction of IFNs with their particular cell surface receptor. In this report, we identify two specific tyrosine residues located within the cytoplasmic domain of IFNAR2c that are obligatory for IFN-dependent signaling. Various IFNAR2c tyrosine mutants were expressed in a human lung fibroscarcoma cell line lacking IFNAR2c (U5A). Stable clones expressing these mutants were analyzed for their ability to induce STAT1 and STAT2 activation, ISGF3 transcriptional complex formation, gene expression, and cell growth regulation in response to stimulation with type I IFNs. The replacement of all seven cytoplasmic tyrosine residues of IFNAR2c with phenylalanine resulted in a receptor unable to respond to IFN stimulation. Substitution of single tyrosines at amino acid residue 269, 316, 318, 337, or 512 with phenylalanine had no effect on IFN-dependent signaling, suggesting that no single tyrosine is essential for IFN receptor-mediated signaling. In addition, IFNAR2c retaining five proximal tyrosines residues (269, 306, 316, 318, and 337) or either two distal tyrosine residues (411 or 512) continued to be responsive to IFN stimulation. Surprisingly, the presence of only a single tyrosine at either position 337 or 512 was sufficient to restore a complete IFN response. These results indicate that IFN-dependent signaling proceeds through the redundant usage of two tyrosine residues in the cytoplasmic domain of IFNAR2c.

Previous experiments have suggested that induction of the β-R1 gene by interferon (IFN)-β required transcription factor ISGF-3 (IFN-stimulated gene factor-3) and an additional component. We now provide evidence that nuclear factor-κB (NF-κB) can serve as this component. Site-directed mutagenesis of an NF-κB binding site in the β-R1 promoter or over-expression of an IκBα super-repressor abrogated IFN-β-mediated induction of a β-R1 promoter-reporter. IFN-β treatment did not augment abundance of NF-κB but did lead to phosphorylation of the p65 NF-κB subunit. It is proposed that IFN-β-mediated enhancement of the transactivation competence of NF-κB components is required for inducible transcription of the β-R1 promoter. These results provide a novel insight into the role of NF-κB in the transcriptional response to IFN-β.

p53 represses the transcription of cdc2 and cyclin B1, causing loss of Cdc2 activity and G(2) arrest. Here we show that the region -22 to -2 of the cdc2 promoter called the R box is required for repression by p53 but not for basal promoter activity. The R box confers p53-dependent repression on heterologous promoters and binds to p130/E2F4 in response to overexpression of p53. R box-dependent repression requires p21/waf1, and overexpression of p21/waf1 also represses the cdc2 promoter. These observations suggest that p53 represses the cdc2 promoter by inducing p21/waf1, which inhibits cyclin-dependent kinase activity, enhancing the binding of p130 and E2F4, which together bind to and repress the cdc2 promoter.

The LMP2 gene, which encodes a protein required for efficient presentation of viral antigens, requires both unphosphorylated Stat1 and IRF1 for basal expression. LMP2 expression is down-regulated by the adenovirus protein E1A, which binds to Stat1 and CBP/p300, and by the mutant E1A protein RG2, which binds to Stat1 but not to CBP/p300, but not by the mutant protein Delta2-36, which does not bind to either Stat1 or CBP/p300. Stat1 and IRF1 associate in untreated cells and bind as a complex to the overlapping ICS-2/GAS element of the LMP2 promoter. E1A interferes with the formation of this complex by occupying domains of Stat1 that bind to IRF1. These results reveal how adenovirus infection attenuates LMP2 expression, thereby interfering with the presentation of viral antigens.

We report preliminary characterization of a gene designated β-R1, which is selectively expressed in response to interferon β (IFN-β) compared with IFN-α. In human astrocytoma cells, β-R1 was induced to an equivalent extent by 10 IU/mL IFN-β or 2500 IU/mL IFN-α2. To address the mechanism of this differential response, we analyzed induction of the β-R1 gene in fibrosarcoma cells and derivative mutant cells lacking components required for signaling by type I IFNs. β-R1 was readily induced by IFN-β in the parental 2fTGH cell line, but not by recombinant IFN-α2, IFN-α Con1, or a mixture of IFN-α subtypes. IFN-α8 induced β-R1 weakly. β-R1 was not induced by IFN-β in mutant cell lines U2A, U3A, U4A, and U6A, which lack, respectively, p48, STAT1, JAK1, and STAT2. U5A cells, which lack the Ifnar 2.2 component of the IFN-α and -β receptor, also failed to express β-R1. U1A cells are partially responsive to IFN-β and IFN-α8 but lacked β-R1 expression, indicating that TYK2 protein is essential for induction of this gene. Taken together, these results suggest that the expression of β-R1 in response to type I IFN requires IFN-stimulated gene factor 3 plus an additional component, which is more efficiently formed on induction by IFN-β compared with IFN-α.

Carbamyl-P synthetase (EC 2.7.2.9), aspartate transcarbamylase (EC 2.1.3.2), and dihydro-orotase (EC 3.5.2.3), the first three enzymes of the de novo pathway for synthesis of pyrimidine nucleotides, have been co-purified as a single oligomeric protein from a mutant line of hamster cells selected for its ability to resist N-(phosphonacetyl)-L-aspartate (PALA), a potent and specific inhibitor of aspartate transcarbamylase. All three enzymes overaccum,late in the mutant cells (Kempe, T.D., Swyryd, E.A., Bruist, M., and Stark, G.R. (1976) Cell 9, 541-550) and the oligomer represents nearly 10% of the total cellular protein. Tens of milligrams of oligomer have been purified to homogeneity by a simple and rapid procedure, with recovery of about 50% of all three activities. The pure protein contains only one size of polypeptide, Mr approximately 200,000, as revealed by electrophoresis in danaturing gels. All three enzyme activities are associated with this polypeptide, indicating that it is multifunctional. Further evidence for a multifunctional protein is provided by titration of the oligomer with radioactive PALA, which reveals that the number of PALA binding sites approximately equals the number of polypeptide chains. The isolated multifunctional protein is a mixture of trimers and hexamers.

N-(phosphonacetyl)-l-aspartate (PALA) is an analog of the activated complex for the reaction catalyzed by aspartate transcarbamylase and has previously been shown to be a potent and specific inhibitor of the enzymes from Escherichia coli and mouse spleen. PALA is now shown to be a potent inhibitor of de novo pyrimidine nucleotide biosynthesis in five lines of mammalian cells in culture. When the concentration of PALA in the medium is 10-4 m, simian virus 40-transformed hamster cells do not divide and, within about 2 days, have detached from the culture dishes and appear dead. The toxicity of 10-4 m PALA is prevented completely if 10-4 m uridine is present in the medium, showing that PALA inhibits uridylic acid biosynthesis specifically. The same effect can be observed with the other cell lines. When the external concentration of radioactive PALA is in the range 10-4 to 10-5 m, its concentration in the cytosol is approximately equal to the external concentration. Most of the PALA recovered from cells remains undegraded. In crude cell extracts, the Ki for PALA is about 1 x 10-9 m at 37° and pH 7.5. Addition of 1 x 10-7 m PALA in vitro to an extract of transformed hamster cells inhibits by 60% the incorporation of radioactivity from CO2 into UMP and CMP; approximately 10-6 m PALA is required to give comparable inhibition of incorporation from carbamyl-P. Higher concentrations give virtually complete inhibition in both cases. These results suggest that PALA may find clinical use in suppressing cellular proliferation. A small number of the virus-transformed hamster cells become resistant to 10-4 m PALA in culture and retain resistance after they are grown for many generations in the absence of the drug. Resistance is accompanied by a large increase in aspartate transcarbamylase activity.

We describe a new, rapid, and convenient procedure for purifying 2',5'-oligoadenylate synthetases, employing precipitation with ammonium sulfate, fractionation by gel filtration, rapid binding to poly(I) X poly(C) cellulose, and elution with 0.35 M KCl. Unlike previously published methods, the procedure does not require sedimentation of the enzyme at 200,000 X g. Therefore, it is more general and more likely to succeed with synthetases extracted from a variety of cells or tissues, or from different subcellular fractions. We have purified the enzymes from two sources to apparent homogeneity, about 2500-fold from the cytoplasm of HeLa cells in 40% yield and more than 400,000-fold from the cytoplasm of rabbit reticulocytes in 25% yield. The specific activity of the HeLa enzyme is about 4 times higher than reported previously. The physical and functional properties of the pure enzymes are very similar to those reported by others for preparations of 2',5'-oligoadenylate synthetase from rabbit reticulocytes, mouse L cells, and human HeLa cells. A new affinity matrix was prepared by linking periodate-oxidized poly(I) X poly(C) to a hydrazide derivative of finely divided cellulose. Poly(I) X poly(C) cellulose binds about twice as much synthetase as the corresponding amount of poly(I) X poly(C) paper and activates the bound enzyme about three times better.

N-(phosphonacetyl)-l-aspartate (PALA), a transition state analogue for aspartate transcarbamylase, is an effective inhibitor of the enzyme in a number of tissues when administered to mice. After a single injection of 1 mg of [3H]PALA (1 µCi per mg), PALA concentrations reach a peak of 1.5 x 10-6 m within 2 hours in spleen, and remain at a constant level of 5 x 10-7 m for at least 72 hours in both spleen and liver. PALA concentrations in jejunum are maintained at 10-7 m for at least 72 hours following a single subcutaneous injection. More than 90% of the tissue radioactivity can be recovered as unchanged PALA. The drug is equally effective when given either orally or by subcutaneous injection. Using the activity of aspartate transcarbamylase as a measure of the effective concentration of PALA in the tissue, a single oral dose of as little as 100 µg gives maximum inhibition of the enzyme. Aspartate transcarbamylase activity is maximally inhibited within 15 min of administration and 72 hours later the enzyme activity in spleen is still substantially inhibited. Addition of 1 x 10-7 m PALA in vitro to an extract of spleen inhibits by 80% the incorporation of radioactivity from CO2 into UMP and CMP; approximately 10-6 m PALA is required to give comparable inhibition of incorporation from carbamyl-P. Mice injected with 1 mg of PALA 1 hour before injection with isoproterenol incorporated [14C]thymidine into the DNA of their submaxillary glands at a rate no greater than that of control animals, whereas incorporation by mice injected with isoproterenol alone was markedly stimulated, indicating that PALA effectively blocks the supply of pyrimidine nucleotides required for DNA synthesis in this instance of drug-stimulated proliferation.

Some preparations of both native aspartate transcarbamylase from Escherichia coli and catalytic subunit have fewer tight binding sites per oligomer for carbamyl-P than the number of catalytic peptide chains. In contrast, the number of sites for the tight-binding inhibitor N-(phosphonacetyl)-L-aspartate does equal the number of catalytic chains in each case. Binding of the labile carbamyl-P was determined using rapid gel filtration, with conversion to stable carbamyl-L-aspartate during collection. Native enzyme (six catalytic chains) obtained from cells grown under the conditions of J.C. Gerhart and H. Holoubek (J. Biol. Chem. (1967) 242, 2886-2892) has 5.4 tight sites for carbamyl-P at pH 8.0 (KD = 9.9 muM), whereas native enzyme from cells grown with higher concentrations of glucose, uracil, and histidine (to yield more enzyme per unit volume of culture) has only 1.9 tight sites at pH 8.0 (KD = 4.6 muM) and only 2.3 tight sites at pH 7.0 (KD = 2.6 muM). At pH 8.0, catalytic subunit (three catalytic chains) obtained from the former native enzyme has 2.2 tight sites for carbamyl-P (KD = 2.4 muM) and the number of sites is 2.3 in the presence of 35 mM succinate, whereas catalytic subunit obtained from the latter native enzyme has 1.8 tight sites (KD = 3.6 muM) in the absence of succinate and 2.3 tight sites in its presence. The number of tight binding sites is also less than the number of subunit peptide chains in 19F nuclear magnetic resonance experiments performed with catalytic subunit and two fluorinated analogs of carbamyl-P at comparable concentrations of analogs and active sites. A model is proposed in which incomplete removal of formylmethionine from the NH2 termini of the enzyme under conditions of extreme depression affects affinity for ligands.

Mutant Syrian hamster cells resistant to N-(phosphonacetyl)-L-aspartate (PALA), a transition state analog inhibitor of aspartate transcarbamylase, overproduce CAD, a multifunctional protein which catalyzes the first three reactions of de novo UMP biosynthesis. Increased levels of a single mRNA cause the overproduction of CAD in all PALA-resistant mutants examined thus far. A recombinant plasmid containing a 2,3-kilobase insert complementary to the 3'-proximal region of this 7.9-kilobase mRNA has been prepared and used to show that the CAD gene is amplified in each of the 10 PALA-resistant mutants examined. Rates of association of CAD sequences in DNA isolated from PALA-sensitive and PALA-resistant cells with labeled plasmid DNA indicated that the degree of amplification is approximately equal to the degree of overproduction of protein and mRNA in each mutant. The patterns of digestion of these DNAs with restriction enzymes confirmed this result and showed that the lower limit for the size of the amplified unit is 19 kilobases, much larger than the mRNA. A comparison of restriction endonuclease digests of the cloned cDNA with digests of genomic DNA indicated that part of this difference is attributable to intervening sequences in the CAD gene. A 10.2-kilobase RNA which contains CAD sequences is found in cytoplasmic fractions from some PALA-resistant mutants but not in wild type cells. Restriction patterns were analyzed by a new method in which fragments of DNA are transferred from agarose gels to diazo paper with a high efficiency which is independent of size.

Pyridoxal-P reacts specifically with a single lysine residue at the active site of Escherichia coli aspartate transcarbamylase (Greenwell, P., Jewett, S. L., and Stark, G. R. (1973) J. Biol. Chem. 248, 5994-6001). Reduction of the Schiff base with sodium borohydride, succinylation of the remaining lysine residues, and digestion with trypsin result in formation of a single pyridoxyl peptide, which was purified to homogeneity after chromatography on DEAE-cellulose, treatment with alkaline phosphatase, and rechromatography. Amino acid composition and the results of limited sequential degradation showed that this peptide corresponds to residues 62 to 98 in the sequence of Konigsberg and co-workers, and contains 2 residues of lysine (Henderson, L., Roy, D., Martin, D., and Konigsberg, W., personal communication). By similar isolation, a second peptide was obtained from unsuccinylated catalytic subunit, containing only the pyridoxylated lysine, which corresponds to Lys-80. Derivatives of catalytic subunit containing an average of either one, two, or three pyridoxamine-P moieties per trimer have been prepared by reduction. These species, which retain catalytic activity in proportion to their unmodified active sites, were recombined with regulatory subunit to prepare partially modified derivatives of native aspartate transcarbamylase. At pH 8, fluorescence emission bands were observed at 340 nm, due to aromatic amino acids in the protein, and at 395 nm, due to the pyridoxamine-P moiety. Upon excitation at 280 nm energy transfer from protein to pyridoxamine-P was approximately 15%. The properties of the probe were used to study changes accompanying the binding of substrates and inhibitors. The effects of CTP and ATP were small. With the transition state analog N-(phosphonacetyl)-L-aspartate (PALA) or the substrate carbamyl-P, two types of response were observed. Derivatives of catalytic subunit and native enzyme which contain some unmodified sites and hence retain partial catalytic activity gave large increases in fluorescence at 395 nm. However, fully modified inactive derivatives gave much smaller increases. A derivative of native enzyme containing one triply modified and one unmodified catalytic subunit behaved like the other partially modified species. These results indicate that there is communication among the active sites of different catalytic trimers in modified native enzyme, as well as among active sites within the same modified catalytic trimer. The increases in fluorescence result from a red shift of the absorption maximum of the pyridoxamine-P moiety from 315 to 325 nm, which increases the absorbance at the excitation wavelength for fluorescence. At pH 7, the absorption spectrum is already shifted and, consequently, the binding of PALA and carbamyl-P has little effect on the fluorescence. Therefore, the binding of these compounds at pH 8.0 must cause a structural change in the protein, which in turn causes protonation of a group in the modified active sites, altering the spectral properties.

2',5'-oligoadenylates can be assayed sensitively in cell extracts by use of an antiserum having maximum specificity for any compound containing the moiety -pA2'pA2'pA-. These compounds reached high concentrations (25-2000 nM) in monkey CV-1 cells after infection with simian virus 40 (SV40) and treatment with human leukocyte interferon. The levels were highest late in infection and increased in parallel with the accumulation of SV40 late messenger RNAs. Alone, neither interferon nor SV40 caused the 2',5'-oligoadenylate concentrations to increase above the levels present in untreated CV-1 cells, 3 nM or less. Analyses by high performance liquid chromatography revealed little or no (p)pp(A2'p)2A or (p)pp(A2'p)3A, and the extracts showed only very low activity in functional assays with ppp(A2'p)nA-dependent nucleases, equivalent to 3 nM ppp(A2'p)3A or less. Some of the 2',5'-oligoadenylates eluted in the positions of the nonphosphorylated "cores," (A2'p)nA, and a substantial fraction was found in several peaks intermediate between ppp(A2'p)3A and cores. The positions of most of these peaks did not change when digestion with alkaline phosphatase was performed before chromatography, indicating that most of the 2',5'-oligoadenylates lack exposed phosphate groups. In contrast to the effects of infection with SV40, addition of poly(I) X poly(C) to interferon-treated CV-1 cells led to accumulation of high levels (up to 3000 nM) of 2',5'-oligoadenylate-5'-di- or triphosphates capable of activating the ppp(A2'p)nA-dependent ribonuclease.

Cells resistant to pyrazofurin and 6-azauridine have been selected from a simian virus 40-transformed Syrian hamster line and from a Chinese hamster lung line. By increasing the concentrations of inhibitors in several steps, mutant cells from both lines have been obtained which resist high concentrations (1 to 5 mM) of the two inhibitors separately or together. Orotidine-5'-phosphate decarboxylase (EC 4.1.1.23), the sixth and last enzyme in UMP biosynthesis, is inhibited by the nucleoside monophosphates derived from pyrazofurin or 6-azauridine. The activity of this enzyme is increased in each resistant cell line tested. Furthermore, there is a parallel increase in each case in the activity of the fifth enzyme of the pathway, orotate phosphoribosyltransferase (EC 2.4.2.10), which is not inhibited by pyrazofurin or 6-azauridine monophosphates, and the amount of increase is up to 67 times the level found in wild type cells. In contrast, the activities of the first three enzymes of UMP biosynthesis remain essentially unchanged in the mutants. Resistant Chinese hamster cells remain sensitive to 5-fluorouridine; this indicates that uridine kinase, the enzyme necessary to convert 6-azauridine to the monophosphate, is still functional.

The interactions of several dicarboxylic acids and monoanions with Escherichia coli aspartate transcarbamylase and with its catalytic subunit have been studied by ultraviolet difference spectroscopy and steady state kinetics, with the following major findings. 1. A variety of dicarboxylic acids compete with carbamyl-P for the active sites of unliganded catalytic subunit, with steric requirements very different from those important for competition with L-aspartate for the subunit/carbamyl-P complex. Competition with carbamyl-P is much reduced if the dicarboxylic acid has a positively charged amino group. Acetate and chloride also compete. 2. At pH 7, equal concentrations of lysine acetate and L-aspartate are equally effective in displacing the transition state analog N-(phosphonacetyl)-L-aspartate (PALA) from the active sites of the concentrations of L-aspartate and lysine acetate is constant, increasing the concentration of L-aspartate does not relieve inhibition of the enzyme by PALA (Collins, K.C., and Stark, G. R. (1971) J. Biol. Chem. 246, 6599-6605). Therefore, the L-aspartate/subunit complex, like the acetate/subunit complex, must be incapable of participating in the catalytic reaction. We conclude that the kinetic mechanism is ordered, in agreement with the recent findings of Wedler and Gasser (Wedler, F.C., and Gasser, F.J. (1974), Arch. Biochem. Biophys. 163, 57-68) and in disagreement with the interpretation of Heyde et al. (Heyde, E., Nagabhushanam, A., And Morrison, J.F. (1973) Biochemistry 12, 4718-4726)...

We have investigated the mechanism of overproduction of the multifunctional protein catalyzing the first three steps of UMP biosynthesis in stable mutants of Syrian hamster cells in culture. The rate of degradation of this protein is unaltered in one mutant cell line which overproduces it by 118-fold. In all mutants tested, the increase in the rate of synthesis of this protein is equal to the increase in its steady state concentration. There is a similar correlation between steady state levels of this protein in vivo and the capacity of polysomal RNA isolated from these cells to direct the synthesis of the protein in vitro. The one mutant cell line studied contains large amounts of a polysomal poly(A)-containing RNA (Mr = 2.7 +/- 0.2 times 10(6)) that is not detected in wild type cells. This large RNA co-sediments in sucrose gradients with the capacity to direct the synthesis of the multifunctional protein in vitro.

We have studied expression of the carbamoyl-P synthetase-aspartate transcarbamoylase-dihydroorotase (CAD) gene in growing or confluent Syrian hamster cells, and also in cells arrested by depriving them of serum or restimulated by adding fresh serum. In contrast to other biosynthetic enzymes such as dihydrofolate reductase, the amount of the CAD enzyme decreased very little when growth was arrested, as judged by the small change in aspartate transcarbamoylase activity. However, the level of CAD mRNA was about 10-fold lower in arrested cells than in growing cells. The rate of transcription of CAD, measured by the nuclear run-off technique, decreased roughly in parallel with the decrease in the steady-state level of the mRNA, suggesting that control of transcription is probably an important element in the overall regulation of CAD gene expression. Expression of a different gene which lies near the 5' end of CAD changed in parallel with that of CAD, but expression of two other genes in the same region was insensitive to the growth state of the cells. Therefore, control of gene expression as a function of growth was specific rather than a reflection of more general changes in chromatin structure in the vicinity of CAD.

N-(Phosphonacetyl)-l-aspartate (PALA) is a stable analogue of the transition state in the reaction catalyzed by aspartate transcarbamylase from Escherichia coli; it combines in 1 molecule most of the structural features of the two natural substrates, carbamyl phosphate and l-aspartate. The inhibition constant (Ki) for PALA with the catalytic subunit is 2.7 x 10-8m at 28° and pH 7.0 in 0.2 m imidazole acetate; the value of Ki is approximately the same with the native enzyme. PALA binds to aspartate transcarbamylase about 1000 times more tightly than carbamyl phosphate, the substrate bound most tightly. In addition to tight binding, other data indicate that PALA interacts with the enzyme at both the carbamyl phosphate and l-aspartate sites simultaneously and puts the enzyme into the same contracted conformation as do carbamyl phosphate and succinate (an analogue of l-aspartate) acting together. PALA alone or carbamyl phosphate and succinate together produce the same ultraviolet difference spectrum with the catalytic subunit; the binding of PALA alone or carbamyl phosphate and succinate together at some of the catalytic sites in the native enzyme activates the remaining catalytic sites to the same extent. PALA gives competitive inhibition with carbamyl phosphate but not with l-aspartate, indicating an obligatory binding order for the substrates: carbamyl phosphate must bind before l-aspartate for catalysis to occur. Titration of the catalytic subunit with PALA, monitored by ultraviolet difference spectroscopy, indicates that there are 3.07 active sites per 100,000 daltons. The interaction of PALA with the enzyme is consistent with the "compression" model for aspartate transcarbamylase previously proposed (Collins, K.D., and Stark, G.R., J. Biol. Chem., 244, 1869 (1969)).

Erythro-β-hydroxy-l-aspartate is a good substrate for the catalytic subunit of the aspartate transcarbamylase of Escherichia coli, with Km = 24 mm (approximately the same as the Km for l-aspartate) and Vmax = 0.18 mole per min per g (22% of the Vmax for l-aspartate) at pH 8 and 28°. Aminomalonate is also carbamylated by the enzyme (Km approximately 770 mm and Vmax = 0.014 mole per min per g under the same conditions), but threo-β-hydroxy-dl-aspartate, erythro, threo-β-methyl-dl-aspartate, α-methyl-dl-aspartate, l-glutamate, and O-phospho-dl-serine are all inactive, even at high concentrations of enzyme. dl-α-Methylsuccinate, d- and l-tartrate, d-aspartate, the α- and β-methylaspartates, and threo-β-hydroxy-dl-aspartate are all poor inhibitors (Ki greater than 100 mm), whereas mesotartrate (Ki = 2.7 mm at pH 7 and 28°) and d- and l-malate are good competitive inhibitors for l-aspartate. Since maleate is a much better inhibitor than fumarate, we concluded previously that l-aspartate is bound to the aspartate transcarbamylase-carbamyl phosphate complex with its carboxylates oriented approximately cis to one another. Using erythro-β-hydroxy-l-aspartate as an example, four loci can be designated on the enzyme-carbamyl phosphate complex, corresponding to the positions of the α-amino group (A), the β-hydroxyl group (B), the β-hydrogen (C), and the α-hydrogen (D). When the carboxylates are cis, A is cis to B and C is cis to D. A single set of restrictions on these four loci can be used to rationalize the specificity of the enzyme for analogues of l-aspartate as substrates or inhibitors and the previous observation that l-aspartate binds very poorly to the enzyme-carbamyl phosphate complex. The restrictions are (a) that Loci A and B can accept hydroxyl groups without substantial impairment of binding but Loci C and D cannot, and (b) that an amino or methyl group in any of the four loci undergoes an interaction with the enzyme-carbamyl phosphate complex that prevents tight binding. These restrictions lead to the conclusions that mesotartrate binds with the hydroxyl groups in Loci A and B but not in C and D, that l-malate binds with the hydroxyl group in Locus A, and that d-malate binds with the hydroxyl group in Locus B.

A transient nuclear magnetic resonance method has been used to measure the rate at which succinate, a competitive inhibitor of l-aspartate, binds to the aspartate transcarbamylase of Escherichia coli and to its catalytic subunit. In the presence of catalytic subunit and carbamyl phosphate, the relaxation time T2 for succinate is greatly shortened, while the relaxation time T1 is not affected much. In an analogous experiment with inorganic phosphate in place of carbamyl phosphate, T2 is also greatly shortened and, in both instances, the relaxation times are independent of the nuclear magnetic resonance frequency. The expressions for relaxation in the presence of chemical exchange were examined for limiting cases of frequency independence. Two limiting cases could be distinguished by the opposite dependences of T2 on temperature, and it was demonstrated clearly that succinate exchanges slowly with the enzyme-carbamyl phosphate complex, but rapidly with the enzyme phosphate complex. For exchange of succinate with the catalytic subunit-carbamyl phosphate complex, a bimolecular rate constant of 2.3 x 105 m-1 sec-1 at 33° for binding of succinate and an activation energy of 12 kcal per mole for dissociation of succinate from the complex were found. The exchange of succinate with the catalytic subunit-phosphate complex was too rapid to permit measurement of the rate constant by nuclear magnetic resonance methods, but a rotational correlation time of 0.73 x 10-8 sec at 33° was determined for the ternary complex. The results indicate that succinate has relatively free access to the active site of the subunit-phosphate complex and is bound rigidly. The unusually slow bimolecular rate constant for binding of succinate to the subunit-carbamyl phosphate complex may indicate that a conformational change is concerted with binding. A fast bimolecular step, followed by a slow step, cannot be ruled out entirely, but the concentration of the intermediate complex is probably low (less than 2%). For the native enzyme, the rate constant for binding of the last molecule of succinate to the complex with carbamyl phosphate is 1.9 x 105 m-1 sec-1 at 33°, with an activation energy of 8 kcal per mole for dissociation of succinate from the complex. The rotational correlation time for the native enzyme-phosphate-succinate complex is 2.2 x 10-8 sec at 33°. These results, very similar to those obtained with the catalytic subunit, indicate that the regulatory subunits have little influence on the properties of the active sites when the native enzyme is saturated with succinate.

Spectrophotometric titration studies at pH 8.0 have revealed that pyridoxal 5'-phosphate binds as a Schiff base to the catalytic subunit of Escherichia coli aspartate transcarbamylase with a stoichiometry of 3 pyridoxal phosphates per trimer and a dissociation constant of 0.9 µm. The ultraviolet absorption of the Schiff base near 430 nm is lost upon addition of the inhibitor N-(phosphonacetyl)-l-aspartate (PALA) or the substrate carbamyl-P. In steady state kinetic studies, pyridoxal-P was found to be a competitive inhibitor of carbamyl-P, with an inhibition constant of 4 ± 2 µm. Pyridoxal does not bind to or inhibit the catalytic subunit. These results imply that pyridoxal-P binds at the active site not only through a Schiff base but also through noncovalent, ionic interactions with the phosphate binding region. Reduction of the Schiff base with sodium borohydride yields an N-e-pyridoxyl-l-lysine derivative of the enzyme with 2.9 to 3.0 pyridoxal-P incorporated per catalytic trimer. This reduced derivative is inactive but still binds 3 molecules of PALA per catalytic trimer at pH 8.0. However, the binding of PALA is characterized by at least two dissociation constants in the range 100 to 10 µm. PALA binds much more tightly to unmodified catalytic subunit at pH 8 with a single dissociation constant of about 0.1 µm. Photo-oxidation of the unreduced catalytic subunit-pyridoxal-P Schiff base causes loss of 90% to 95% of the activity and loss of 2.0 ± 0.5 histidine residues per chain. No other amino acid is found to be modified by amino acid analysis. However, 0.5 ± 0.05 molecules per chain of an unknown derivative of pyridoxal-P derivative are covalently bound. Covalently linked dimers and trimers of the catalytic chains are also formed, especially at high protein concentrations. At low concentrations at which dimer and trimer formation is negligible, the covalent incorporation of 0.5 pyridoxal-P per chain does not account for the almost complete loss of enzyme activity. Therefore, the loss of histidine is correlated with loss of activity. These studies indicate that 1 lysine residue and 2 histidine residues are probably near the active site of aspartate transcarbamylase.

A protocol is presented for the quantitative conversion of cysteine and cystine residues in proteins to residues of S-cyanocysteine, using the reagent 2-nitro-5-thiocyanobenzoic acid (Degani, Y., and Patchornik, A. (1971) J. Org. Chem. 36, 2727). Cleavage of the amino peptide bond of the S-cyanocysteine residue is obtained upon exposure to 6 m guanidinium chloride-0.1 m sodium borate, pH 9.0, at 37° for 12 hours, with concomitant formation of 2-iminothiazolidine-4-carboxylyl peptides. Application of the method to several proteins indicates that virtually complete cleavage can be obtained and that there are no significant side reactions due to exposure to alkaline pH. A mechanism for the cleavage reaction is proposed in which specific hydroxide ion catalysis is followed by concerted peptide bond cleavage and ring closure, without the intermediate formation of an acyliminothiazolidine.

14C- and 18O-labeled carbamyl phosphates have been used to study primary kinetic isotope effects in the reaction catalyzed by the catalytic subunit of the aspartate transcarbamylase of Escherichia coli. A novel aspect of the methodology is the use of 14C as a tracer to determine the isotope effect caused by 18O alone, without the need for analysis by mass spectrometry. At optimum concentrations of carbamyl phosphate and l-aspartate and at pH 7.8 (near the pH optimum of the enzyme), the effect of 14C at the carbonyl carbon or 18O at the anhydride oxygen of carbamyl phosphate is very small. One possible explanation, but not the only one, is that kinetic steps in which the bonds to these atoms change are not important in determining the observed rate. Such an interpretation is consistent with a rate-determining conformational change of the enzyme at the pH optimum. At pH 10, where the activity of the enzyme is low, the rate with 14C-labeled carbamyl phosphate is about 95% of the rate with the 12C compound. On the other hand, no substantial effect is seen for 18O in the anhydride position. Taken together, these observations indicate that a step in which the bonding to the carbon atom changes and the bonding to the anhydride oxygen does not change becomes important in the observed rate at pH 10. Such a process might be the formation of a tetrahedral intermediate. The solvent deuterium isotope effect in D2O has also been studied, at optimum concentrations of the substrates. The maximum velocity is essentially the same in D2O and in H2O. Thus, proton transfer does not seem to be involved in a ratedetermining step in the aspartate transcarbamylase reaction. The pH activity profile of the enzyme is shifted up by about 0.8 pH unit in D2O, an effect which can be accounted for approximately quantitatively by the expected change in the pKa of the amino group of l-aspartate.

1. Steady state kinetics indicates that the binding of substrates by the catalytic subunit of aspartate transcarbamylase is ordered. Product inhibition patterns show that carbamyl phosphate binds first, aspartate binds second, carbamylaspartate dissociates first, and phosphate dissociates second. 2. Nonlinear product inhibition and substrate inhibition indicate that several dead end complexes are formed: aspartate binds to the enzyme-phosphate complex, a 2nd mole of phosphate binds to the enzyme-phosphate complex, a 2nd mole of carbamyl phosphate binds to the enzyme-carbamyl phosphate complex, and carbamylaspartate binds weakly to the free enzyme. 3. The interaction of the catalytic subunit with analogues of both substrates has been studied in an attempt to reveal those structural features of the substrates that are required for binding and function. Acetyl phosphate and N-methylcarbamyl phosphate are the only analogues of carbamyl phosphate tested which are substrates, with maximum velocities at pH 7.8 of 2.4% and 0.03% of the maximum velocity with carbamyl phosphate. Since acetyl phosphate lacks the NH2 group of carbamyl phosphate, this group cannot be essential for function. N,N-Dimethylcarbamyl phosphate is not a substrate, but it is a competitive inhibitor, as is any analogue of carbamyl phosphate with a phosphate or phosphonate dianion, suggesting that at least part of the site for carbamyl phosphate is readily accessible. In fact, with the catalytic subunit, even cytidine triphosphate is a competitive inhibitor. 4. Several dicarboxylic acids were found to be competitive inhibitors of l-aspartate. Succinate and maleate are the strongest inhibitors, malonate and d- and l-malate are good inhibitors, whereas fumarate, glutarate, and d- and l-bromosuccinate are poor inhibitors, and d-aspartate does not inhibit significantly. 5. The pH dependence of the dissociation constant for succinate indicates that a group with pKa 7.1 is required to be positively charged for this inhibitor to bind to the enzyme carbamyl phosphate complex. In contrast, the dissociation constants for carbamyl phosphate and phosphonacetamide vary little between pH 6 and pH 9. 6. Product inhibition by carbamylaspartate when aspartate is varied indicates that the constant for dissociation of aspartate from the central complex is much larger than the dissociation constants for unreactive aspartate analogues such as succinate. This indicates that some of the binding energy of aspartate may be used to facilitate the reaction with carbamyl phosphate. 7. The catalytic subunit is unstable at concentrations below 1 µg per ml. There is an immediate decrease in specific activity with decreasing enzyme concentration and a slower irreversible inactivation during incubation at low concentrations. Both the immediate and the slow losses of activity are prevented by the presence of bovine serum albumin (50 µg per ml) in dilute enzyme solutions. However, because bovine serum albumin forms a complex with the catalytic subunit and alters the apparent kinetic parameters, it was not used in these kinetic studies. 8. At pH 7.8 and 28°, the dissociation constant for carbamyl phosphate is 1.4 x 10-5 m, the Km for aspartate is 0.020 m, and the Vmax is 9.0 x 104 moles per min per 100,000 g of subunit. The dissociation constant for carbamyl phosphate and Km for aspartate vary little between pH 6.4 and pH 8.7, whereas Vmax changes greatly, with an optimum near pH 8. With the use of acetyl phosphate instead of carbamyl phosphate, Km for aspartate is unchanged at pH 7.8, but the dissociation constant for acetyl phosphate is 35-fold higher. With N-methylcarbamyl phosphate, Km for aspartate is 7-fold higher and the dissociation constant for N-methylcarbamyl phosphate is 4-fold higher than for carbamyl phosphate.

Each of the three polypeptide chains of the catalytic subunit (CSU) of aspartate transcarbamylase from Escherichia coli has a single —SH group and no disulfides. Although the —SH group is unreactive toward several S-alkylating agents and symmetrical disulfides, it reacts slowly with 5,5'-dithio-bis(2-nitrobenzoate) (DTNB) and p-hydroxymercuribenzoate. CSU is inactivated by these reagents and also, transiently, by inorganic polysulfides. Kinetic studies of the reaction with DTNB show (a) that DTNB does not bind to CSU as a prerequisite for reaction, (b) that reaction depends on the ionization of a group on the enzyme with apparent pKa = 7.9, probably the —SH group, and (c) that inhibitors of enzymatic activity also inhibit reaction of the —SH group with both DTNB and p-hydroxymercuribenzoate. Dissociation constants for phosphate and succinate determined from inhibition of the DTNB reaction agree with those previously determined by steady state kinetics and ultraviolet difference spectroscopy. Undissociated native aspartate transcarbamylase does not react with DTNB under conditions which lead to complete reaction of CSU. Treatment of the CSU-thionitrobenzoate mixed disulfide with nucleophiles such as cyanide, sulfite, and 2-mercapto-ethanol leads to new S- derivatives of CSU. The S-cyano derivative is fully active, but the S-sulfo and S-2-mercapto-ethanol derivatives have little or no activity. We conclude that the —SH group of CSU is very near the active site but is not required for catalytic activity and discuss possible alternative functions for this group. We also discuss the general use of enzyme-thionitro-benzoate mixed disulfides as intermediates in the preparation of specific derivatives of enzyme —SH groups and the potential use of enzyme thiocyanates, prepared through the thionitrobenzoate disulfides, in specific chemical cleavage of polypeptide chains at cysteine and cystine residues.