The phosphate transfer system of Haseltine et al., consisting of a ribosomal wash obtained from a stringent strain of Escherichia coli, washed ribosomes, GTP, and ATP, was used to prepare large quantities of guanosine tetra-and pentaphosphates, the magic spot compounds MS I and MS II of Cashel and Gallant. In our hands, the Haseltine et al. system yielded predominantly guanosine tetraphosphate, ppGpp. This system was used exclusively in the described experiments, with ATP labeled with 32p in the A-and ey-positions as donor. The #-label was found to produce a ppGpp and the y-label a ppGp*. Furthermore the stringency effect, the translocation with which this laboratory has been concerned for some time (4). In the present report we have exploited the availability of a relatively easy method (2) of obtaining sizeable quantities of the "magic spot" guanosine polyphosphates for analysis of the mechanism of the transfer reaction. By obtaining transfer products from ATP marked at various positions, it has been possible to show: (i) that synthesis from ATP is due to pyrophosphoryl transfer, and (ii) that this transfer yields the 3'-pyrophosphoryl derivative of guanosine 5'-polyphosphates. MATERIALS The stringent strain used for preparation of the enzyme system was E. coli K-19 (strepr, thr-, leu-, B1-, galt, ac-, Fe); it was kindly supplied by Dr. Norton Zinder and checked for stringency by Dr. Peter Model. The organism was grown in bulk at the Oak Ridge National Laboratory, with the most valuable cooperation of Dr. Novelli's group in the Biology Division. There it was processed through to ribosomes, from which the enzyme system was prepared in this laboratory. The NH4Cl-washed ribosomes and 0.5 M NH4Cl ribosomal wash were prepared as described (2).Yeast inorganic pyrophosphatase (840 units/mg) and 3'-nucleotidase from Rye Grass (14 units/mg) were purchased from Sigma. Carrier-free 82p, [a-_2P] Since the aim of the present study was to explore the transfer mechanism of phosphate from ATP to GTP and/or GDP, we set up the system as described (2) and followed the course of reaction using GTP and ATP as the reactants. As shown in Fig. 1, the presently used system includes ribosomes and EF-G, and the GTP is rapidly converted to GDP. The figure shows that, in the beginning, with mostly GTP present, the reaction starts with relatively small amounts of guanosine 306

The degradation of guanosine 5'-diphosphate,3'-diphosphate (ppGpp) by the "crude" ribosomal fraction of Escherichia coli CP 78 (rel+, spoT+) was demonstrated and characterized. When the 3'-pyrophosphoryl group of ppGpp was hydrolyzed, the primary degradation product was 5'-GDP. PhospboryIation of ppGpp to guanosine 5'-triphosphate,3'-diphosphate (pppGpp) prior to degradation was not necessary. The degradation process required Mn2+ and was inhibited by EDTA. Levallorphan, an inhibitor of in vivo ppGpp degradation, also inhibited ppGpp degradation by the crude ribosome. Thiostrepton and tetracycline did not have any inhibitory effect, indicating that the reaction is not a reversal of pyrophosphorylation catalyzed by the stringent factor/ribosome complex.Crude ribosome fractions from E. coli NF161 and NF162, both spoT, contained little degrading activity, but similar fractions of E. coli CP79, a reL4-and spoT+ strain, contained ppGpp degrading activity.Guanosine 5'-diphosphate,3'-diphosphate (ppGpp) is a pleotropic effector that regulates various metabolic pathways (1) as well as the transcription of certain operons during nutritional stress (2) in bacterial cells. Bacteria have therefore developed a sensitive balance of biosynthesis and degradation in controlling the cellular concentration of ppGpp (3). The biosynthesis of this important nucleotide during amino acid deprivation was found to be a ribosome-dependent process (4), the basic mechanism of which is that the stringent factor, ATP: GTP(GDP) 3'-pyrophosphotransferase, is activated by a ribosome-mRNA complex when the aminoacyl-tRNA site of the ribosome is occupied by a codon-specified, uncharged tRNA. A nonribosome-dependent synthesis of ppGpp has also been found (5, 6).The degradation process of ppGpp has not been well defined due to the lack of an in vitro cell-free system. In vivo studies have shown that it is very rapid (7,8) and is controlled by the availability of an energy source (9), but studies with spoTmutant, which has a 10-fold slower degradation rate, have led to conflicting theories as to its degradation pathway (10-13). I report here on the characterization of an in vitro cell-free ppGpp degradation system. METHODS Materials. Escherichia coli CP78 (arg-, his-, leu -, thr, thi-, relA+, spoT+), CP79 (arg-, his-, leu-, thr-, thi-, relA-, spoT+), NF161 (met-, arg-, relA+, spoT-), and NF162 (met , arg-, relA , spoT-), kindly supplied by N. Fiil of the University of Copenhagen, were grown in a yeast extract/phosphate medium (14).[3H]ppGpp (8.1 Ci/mmol) was prepared as described (15).[3H]GDP and [3H]GTP were purchased from New England Nuclear. Levallorphan tartrate was a gift from W. Scott, Hoffman-LaRoche, Nutley, NJ. Boric acid gel (particle size 0.1-0.4 mm) was obtained from Aldrich Chemical Co. and was pretreated with acetone as described (16). Nucleotides other than ppGpp were obtained from P-L Biochemicals.Preparation of Crude Ribosomdl Fraction. E. coli CP78 cells from the late logarithmic phase were harvested and washed once with buf...

A membrane-bound preparation of adenylyl cyclase was isolated from homogenates of Sacchromyces fragilis. In organisms growing in a maximally aerated culture with glucose or lactate, activity of the cyclase increased threeto fourfold when growth entered late logarithmic and stationary phases. At submaximal aeration, the lactate-grown cells had a seven-to eightfold greater cyclase activity as compared c V^Vyclic 3',5'-adenosine monophosphate* 1 *has been shown to control a variety of functions in prokaryotes as well as in eukaryotes (Pastan and Perlman, 1970;Robinson et al., 1971). This control includes catabolite repression in Escherichia coli, differentiation in slime mold, and various hormonal responses in mammalian systems. Little is known, however, of the function(s) of this nucleotide in yeast cells. In vitro, cAMP was shown (Chance and Schoener, 1964;Cheung, 1966) to affect oscillatory cycles of NAD+ reduction and oxidation in cell-free extracts. Measurement of the intracellular concentration of cAMP in Saccharomyces carlbergensis showed an increase in the level of cAMP when yeast was derepressed from glucose repression (Van Wijk and Konijn, 1970). A cAMP phosphodiesterase seems to have a regulating function on the level of intracellular cAMP (Speziali and Van Wijk, 1971). In the preceding paper (Sy and Richter, 1972), we reported on the isolation of a protein that specifically binds cAMP. This binding protein could not be related to a protein kinase, and so far no function has been established for it. To gather more information about the functioning of cAMP in yeast, we turned our attention to its metabolism. We report here that the yeast Saccharomyces fragilis contains a membrane-bound adenylyl cyclase with an activity strongly influenced by growth conditions which were also found to modify the intracellular cAMP content. Experimental SectionGrowth conditions were the same as reported in the preceding paper (Sy and Richter, 1972).Preparation of Adenylyl Cyclase from S. fragilis. The cells were suspended in two volumes of buffer containing 20 mM Tris-HCl (pH 7.4), 5 mM MgCL, 0.25 mM EDTA, and 1 mM dithiothreitol (buffer A), and disrupted by passing them through a French press cell at 15,000 psi. In later experiments, 20 mxt potassium phosphate buffer (pH 7.0

A factor in the ribosomal wash of stringent strains of E. coli was identified by Haseltine et al. as a complement to the ribosomal system for the synthesis of the magic spot compounds of Cashel and Gallant. This factor has been found, in the absence of ribosomes, to catalyze the enzymatic pyrophosphoryl transfer from ATP to GTP or GDP in the formation of magic spot I and magic spot II, the guanosine tetra-and pentaphosphates (ppGpp and pppGpp), respectively. The enzyme, which normally requires the presence of the ribosome-tRNA-mRNA complex for activity, catalyzes a very slow synthesis that is stimulated tenfold by 20% methanol. The temperature optimum of the methanol-stimulated system is 25-30°a nd activity is drastically depressed at 370, presumably by inactivation. Catalysis is linear with enzyme concentration and with time for the first 3 hr; during this period 25% of the added GTP is converted. The nonribosomal system is distinguished from the ribosomal system by having a lower Mg++ and a higher NH4+ optimum. The two systems differ in their response to antibiotics: thiostrepton strongly inhibits the ribosomal system but has no effect on the nonribosomal system.In continuing experiments on the enzymatic mechanism of the pyrophosphate transfer from ATP to GTP or GDP (1) with the 0.5 M NH4Cl wash obtained from ribosomes of stringent strains of Escherichia coli (2), we have attempted to localize the factor on one of the ribosomal subunits but were frustrated by the observation that, on separation of the 50S and 30S parts, the factor was released into the supernatant. We then tried, unsuccessfully, to see if, by supplying the wash factor to either subunit, any ppGpp or pppGpp was formed. During these experiments a slow production of both, which was remarkably stimulated by methanol, was observed in the absence of ribosomes or subunits. We will report here on results of a closer study of this nonribosomal synthesis. MATERIALS AND METHODSE. coli K-19 (stringent) ribosomes, NH4Cl-washed ribosomes, and crude NH4Cl wash were obtained as described (1). Unless otherwise stated, the NH4Cl wash preparation used was a 0-35% (NH4)2S04 cut of the crude NH4Cl wash, which was resuspended in 10 mM Tris HCl (pH 7.8)-10% glycerol-1 mM dithiothreitol (buffer A) and dialyzed overnight against the same buffer (Fraction I). From the supernatant of the Abbreviations: ppGpp, guanosine 5'-diphosphate-3'-diphosphate; pppGpp, guanosine 5'-triphosphate-3'-diphosphate. These compounds are also known as MS I and MS II, i.e., magic spot compounds, by the terminology of Cashel et al. (3,4

A cyclic 3 ' 3 '-adenosine monophosphate (CAMP) binding protein was isolated from several yeast strains. The protein interacted specifically with the 3 ',5 '-phosphodiester ring. At pH 7.4, the estimated binding constant of cAMP for