Characterization of pyrimidine metabolism in the ...

Characterization of pyrimidine metabolism in the cellular slime mold, Dictyostelium discoideum MELINDAE. WALES,MARYG. MANN-DEAN, AND JAMESR. WILD' Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Nanjing University of Posts and Telecommunications on 06/06/13 For personal use only.

Department of Biochemistry and Biophysics, Texas A&M University System, College Station, irX 77843-2128, U.S.A. Received August 16, 1988 Accepted December 9, 1988 WALES,M. E., MANN-DEAN, M. G., and WILD,J. R. 1989. Characterization of pyrimidine metabolism in the cellular slime mold, Dictyostelium discoideum. Can. J . Microbiol. 35: 432-438. The arginine-independent, de novo biosynthetic pathway of pyrimidines in Dictyostelium discoideum is initiated by a class I1 carbamoyl-phosphate synthetase (EC 6.3.5.5) specific for pyrimidine biosynthesis which utilized L-glutamine as its N donor and was partially inhibited by both UTP and CTP. The second step in the de novo pathway was provided by an unregulated aspartate transcarbamoylase (EC 2.1.3.2) which primarily appeared as a multimeric enzyme of 105 kilodaltons. The next enzyme, dihydroorotase (EC 3.5.2.3), was approximately 90 - 100 kilodaltons. Although the early enzymatic activities of the pyrimidine pathway appeared to reside in independent protein complexes, various unstable molecular species were observed. These structural variants may represent proteolytic fragments of a multienzyme complex. In addition to de novo synthesis, the amoeba demonstrated the capacity for salvage utilization of uracil, uridine, and cytidine. Upon starvation on a solid substratum, axenically grown amoebas began a concerted developmental program accompanied by a restructuring of nucleotide metabolism. The absolute levels of the ribonucleotide pools droppedby 98% within 30 h; however, both the adenylate energy charge and the GTPIATP ratios were maintained for 50 h after the initiation of development. The maintenance of these metabolic energy parameters required the tight cell-cell contact necessary for development, and the capacity for pyrimidine metabolism was maintained throughout developmental morphogenesis. Key words: aspartate transcarbamoylase, carbamoyl phosphate synthetase, development, pyrimidine biosynthesis. M. G., et WILD,J. R. 1989. Characterization of pyrimidine metabolism in the cellular slime WALES,M. E., MANN-DEAN, mold, Dictyostelium discoideum. Can. J . Microbiol. 35 : 432-438. La voie de synthbse de novo des pyrimidines, qui est indkpendante de l'arginine chez Dictyostelium discoideum, est initite par une classe I1 de carbamoyl phosphate synthktase (EC 6.3.5.5) spkcifique pour la synthkse des pyrimidines; cette classe d'enzyme utilise la L-glutamine comme donneur de N et est partiellement inhibCe par I'UTP et par le CTP. La seconde Ctape de la voie de novo est assuke par une aspartase transcarbamoylase (EC 2.1.3.2) non r6gulCe qui, i prime abord, semble &tre une enzyme multimkrique de 105 kilodaltons. L'enzyme suivante est la dihydroorotase (EC 3.5.2.3) d'une masse molCculaire approximative de 90- 100 kilodaltons. Bien que les activitks enzymatiques du dCbut de la voie de synthbse des pyrimidines semblent relever de complexes proteiques indkpendants, diverses esptces de masse molCculaire instable ont kt6 observkes. Ces variants structurels pourraient repksenter les fragments protkolytiques d'un complexe multienzymatique. En plus de la synthbse de novo, les amibes se sont avCkes capables de rkcuptrer et d'utiliser l'uracile, l'uridine, et la cytidine. Lorsque mises 2 la ditte sur substrat solide, les amibes en croissance axtnique ont amorck un programme de dkveloppement concertk, accompagnk d'une restructuration du mCtabolisme des nuclkotides. Les niveaux absolus des pools de ribonuclkosides se sont abaissCs de 98 % en dedans de 30 h; toutefois, la charge CnergCtique des adknylates et les rapports GTPIATP ont CtC maintenus durant 50 h aprbs l'initiation du dCveloppement. Le maintien de ces paramktres d'knergie mktabolique requiert le contact cellule-cellule Ctroit ntcessaire au dkveloppement, et la capacitk de mCtaboliser les pyrimidines est maintenue tout au long du dCveloppement morphogCnique. Mots clis : aspartate transcarbamoylase, carbamoyl phosphate synthktase, dkveloppement, biosynth'ese des pyrimidines. [Traduit par la revue]

Introduction Dictyosteliurn discoideurn, a eukalyotic myxameoba belonging to the class Acrasieae, is a free-living amoeba that lacks a flagellated stage and aggregates to form fruiting bodies without syncytial formation. The single-cell amoebas grow vegetatively using bacteria or axenic media for nutrition. When the food source is depleted, or the cells are starved for several specifically required amino acids, the amoebas will initiate a developmental program. As the result of starvation, the cells within given association zones gather into aggregates of approximately lo5 cells. Cyclic AMP (CAMP)mediates cell cell communication during aggregation (Konijn et al. 1968), and the CAMP signal response system has been characterized in some detail (Devroetes 1982; Orlow et al. 1981; Franke and Kessin 1981). After aggregation, multicellular aggregates undergo a series of morphological changes leading to the for'Author to whom correspondence should be addressed. Printed In Canada 1 lmprimt au Canada

mation of mature fruiting bodies containing a mass of spore cells supported by a shaft of vacuolated stalk cells. During the process of differentiation, morphological changes are accompanied by many biochemical and molecular changes (Sussman and Brackenbury 1976; Loomis 1980). Once the exogenous source of nutrients is removed, all subsequent activity must be supported by endogenous reserves; thus, it is not surprising that the total dry weight of D. discoideurn decreased by 50% during development (White and Sussman 1961). The rate of oxygen consumption decreased during this reorganization (Liddel and Wright 1961) and was coincident with a general mitochondrial deterioration (George et al. 1972). Total protein content was reduced by approximately 50%, while total RNA content decreased by 60% (White and Sussman 1961). Studies on developmentally regulated mRNAs have demonstrated that the major change in the pattern of gene expression occurred during late aggregation and that the number of mRNA species did not significantly

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Nanjing University of Posts and Telecommunications on 06/06/13 For personal use only.

WALES ET AL.

change during the first 6 h of differentiation. Post-aggregation cells contained 7000 discrete mRNA species, one-third of which appeared after aggregation (Blumberg and Lodish 1980). For these reasons and as a result of direct mRNA population measurements, it is apparent that the synthesis of RNA occurs throughout development and that purine and pyrimidine nucleotides must be available for nucleic acid biosynthesis throughout the asexual life cycle. Dictyostelium could obtain those nucleotides from three sources: (i) de novo synthesis, (ii) RNA degradation and salvage, and (iii) uptake of exogenous precursors. The capacity for de novo purine and pyrimidine biosynthesis was implied by the absence of a nutritional requirement for either nitrogenous base in defined medium (Franke and Kessin 1977). In addition to this physiological evidence, it has been possible to complement Ura- strains of yeast with genomic fragments from D. discoideum (Jacquet et al. 1982; Boy-Marcotte et al. 1984). However, the nature of pyrimidine nucleotide formation could involve the classic de novo biosynthesis utilizing carbamoyl-phosphate synthetase (CPSase) and aspartate transcarbamoylase (ATCase), or salvage and reutilization of arginine via arginine deiminase (EC 3.5.3.6) and catabolic ornithine transcarbamoylse (OTCase; EC 2.1.3.3). Individual amoebas are capable of catabolizing their own RNA (White and Sussman 1961) during differentiation, a process that might serve as an endogenous source of nucleotides via salvage and reutilization of nucleosides and bases. This study characterizes the early enzymatic steps of the de novo pyrimidine biosynthetic pathway in the amoeba of D. discoideum and evaluates endogenous nucleotide pool levels during development.

Materials and methods Materials All nucleotides, grade A sodium salts, and other biochemicals were obtained from Calbiochem, La Jolla, CA. Tri-n-octylamine was purchased from Sigma Chemical Co., St. Louis, MO. Freon 113 was purchased from Matheson Gas Products, La Porte, TX. Bacteriological peptone and yeast extract were obtained from Oxoid Limited, England. All other complex media and supplements were obtained from DIFCO Laboratories, Detroit, MI. Culture The axenic strain Dictyostelium discoideum AX-3 was maintained as spores in buffer at -70°C or dessicated on silica gel at O°C. The wild-type strain Escherichia coli B/r was occasionally used as a restorative food source for the Dictyostelium amoeba. The amoeboid cultures used for enzymatic assays were grown in modified HL-5 medium or the defined minimal medium of Franke and Kessin (1977). Preparation of enzyme extracts and enzymatic assays Cell-free extracts for all enzyme assays were prepared as described previously for bacterial cultures (Foltermann et al. 1984). CPSase was very unstable in cell-free extracts and its assay required the immediate disruption of fresh cell cultures in 60% DMSO and 10% glycerol in 100 rnM triethanolamine buffer (pH 8.0, 1 mM benzamidine, 1 mM DTT). Activity was radiochemically determined using the method of Ingraham and Abdelal (1978). The effector responses were determined at ATP concentrations that are subsaturating in other systems (5 mM). The activity of ATCase was determined by monitoring the enzymatic production of carbamoyl aspartate using the method of Gerhart and Pardee (1962) with the modifications of the color development procedure of Prescott and Jones (1969). Omithine transcarbamoylase (OTCase) activity was determined by measuring the formation of citrulline at 28°C according to the method of Archi-

433

bald (1944) as modified by Prescott and Jones (1969). Dihydroorotase (DHOase) activity was assayed in the reverse direction by monitoring the production of carbamoyl aspartate using the methods of Prescott and Jones (1969). Arginine deiminase activity was determined by measuring the formation of citrulline at 30°C according to the method of Archibald (1944) as modified by Prescott and Jones (1969). Argininosuccinate synthetase (EC 6.3.4.5) activity was assayed by the procedures of Raushel and Seiglie (1983). The method of Lowry et al. (1951) was used to estimate protein concentrations. Molecular mass estimations The molecular mass in kilodaltons (kDa) of individual enzymes was estimated using ascending flow Sephadex G-200 column chromatography. Ribonuclease A (13.7 kDa), chymotrypsinogen A (25 kDa), ovalbumin (45 kDa), aldolase (158 kDa), and fenitin (400 kDa) were used for calibration. Developmental cycle of Dictyostelium discoideum and ribonucleotide analysis The developmental cycle of Dictyostelium discoideum was initiated and sustained by nutrient deprivation on supportive filters as described by Sussman (1966). At designated times, developing cellular aggregates were collected by submerging the supportive filter pads in 1 mL 6% TCA on ice to give a final concentration of 5 X lo7 cells/mL. After a 30-min incubation, the liquid suspension was centrifuged at 2000 rpm for 5 min (1500 x g) to remove the cellular debris. The acid was extracted with 500 mM tri-n-octylamine in freon according to the procedures of Chen et al. (1977). After thorough mixing, the phases were allowed to separate and the aqueous layer was decanted. The samples were filtered through Millipore GS filters (0.45 pm pore size) and ribonucleotide pools were quantitated by HPLC. Nucleotide pool samples were obtained from starved suspension cultures by removing 500-pL samples and adding 500 pL of cold (4°C) 12% TCA. After 30 min at 4"C, the ribonucleotides were extracted as above. An alternative procedure, involving direct cellular disruption without centrifugation, was employed to extract nucleotides from amoeba grown in axenic medium. The vegetative cells were layered into a Costar tissue culture flask (no. 3150; 75 cm2) at a final density of 3 x lo5 cells/cm2. The flasks were incubated for 10 min to allow cell adhesion and the medium was carefully removed after rotating the flasks upside down. The endogenous nucleotide pools were extracted by adding 6 % TCA directly to the adhered cells. High pressure liquid chromatography with a Whatman Partisil 10SAX column was used to separate the nucleotide pools. Resolution of the nucleotides was achieved using a linear solvent gradient (20 min) from 5.9 rnM (NH4),P04 (pH 2.7) to 50% 750 mM (NH4),P04 (pH 3.8), followed by a 10-min gradient from 50 to 100% 750 mM (NH,),P04, all at a flow rate of 2.0 mL/min. Standard solutions were analyzed in order to determine retention times, 2541280 nm absorption ratios, and quantitative values.

Results Characterization of the early enzymes of pyrimidine biosynthesis CPSase was a very fragile enzyme in cell-free extracts. Several different isolation conditions, including ammonium sulfate fractionation, acetone extraction, the presence of proteolytic inhibitors, and chromatographic separations, failed to stabilize its activity. However, it was possible to measure catalytic activity immediately following cellular disruption in DMSO-glycerol buffer. While L-glutamine and ammonia have been reported as substrates for other CPSases, L-glutamine was the only acceptable donor for this enzyme. Both UTP and CTP (2 mM) caused slight inhibition of the enzyme (24 and 27%, respectively) whereas 10 mM N-acetyl-L-glutamate inhibited 63%. Ornithine and UMP did not have any effect (data not shown). It should be mentioned that due to the


434

CAN. J. MICROBIOL. VOL. 35, 1989

observed with ascending Sephadex G-200 column chromatography. ATCase II had a molecular mass of 105 kDa while the larger species (ATCase I) was approximately 220 kDa. Enzymatic assays were performed as described in Materials and methods. its extreme lability, the saturating concentration of Mg-ATP was not determined for this enzyme. Effector responses were evaluated at ATP levels which were subsaturating in other systems. Unlike CPSase, ATCase activity could be readily assayed in cell-free extracts. The enzyme possessed hyperbolic kinetics with a Km of 4.5 and 0.4 mM for aspartate and carbamoyl phosphate, respectively. ATCase had no response to 2—5 mM effector concentrations of the mono-, di-, and tri-phosphate nucleotides of adenine, guanine, uracil, and cytosine, nor to orotidine-5'-monophosphate, at 5 mM aspartate. Multiple molecular weight forms of ATCase were observed with ascending G-200 column chromatography (Fig. 1). The consistently detectable species ATCase II had a molecular mass of 105 kDa. A larger aggregate (approximately 220 kDa, ATCase I) has been observed under conditions where small volumes of cells, which could be processed rapidly, were used. The molecular mass of DHOase was estimated to be approximately 90-100 kDa by G-200 exclusion chromatography (Fig. 2). ATCase II and DHOase overlapped upon G-200 chromatography from partially purified extracts (protamine sulfate precipitation followed by a 4 0 - 6 0 % acetone cut), but they did not appear to be totally coincident. The ATCase I species was never observed under the conditions used to monitor the DHOase peak, so their relative coincidence could not be determined. Separation and characterization of the acid-soluble nucleotide pools There has been extensive discussion in the literature about

FIG. 2. Molecular mass separation of ATCase n ( O ; 105 kDa) and DHOase ( • ; 90 kDa) from actively growing D. discoideum amoebas. Enzymatic assays were performed as described in Materials and methods. TABLE 1. Comparison of adenosine 5'-triphosphate pool concentrations as influenced by different cell harvesting techniques Harvesting technique

jtmol ATP/g dry wt.

Rapid centrifugation Immediate filtration Formaldehyde treatment Direct disruption

6.46 6.56 4.32 22.34

NOTE: Adenosine 5'-triphosphate (ATP) pool levels were determined in the amoeba of D. discoideum from log phase cells grown in HL-5 medium. Pool extractions and analysis were performed as described in Materials and methods

the instability of pools during the time ( 3 - 5 min) required to harvest cells and initiate nucleotide pool extractions. Therefore, the concentration of ATP was determined in amoebas harvested by filtration through membrane filters, by rapid centrifugation at 4°C, by direct formaldehyde fixation of the cells prior to centrifugation, and by direct cellular disruption (Table 1). The ATP levels from the traditional methods of rapid centrifugation or filtration were not significantly different from each other, but were approximately 3.5-fold lower than the levels detected by direct cellular disruption. Vegetative cells of D. discoideum were found to contain 22.3 jtmol ATP / g dry weight by the latter method, while the values reported for various enteric bacteria varied from 3.4 to 6.5 fimol ATP / g dry weight of cells (Knowles 1977). Nucleotide pools in Neurospora, Crithida, and Saccharomyces have been reported to range from 2.4 to 8.9 /unol / g dry weight during logarithmic growth (Jones 1970). Table 2 summarizes a comparison of nucleoside triphosphate

WALES ET AL.

TABLE2. Nucleoside triphosphate pool concentrations (nmol15 x lo7cells) D,discoideum under different growth conditions


HL-5 medium

CTP UTP ATP GTP ECA GTPIATP

Exponential growth

Stationary phase

DM-medium, exponential growth

3.90 6.67 26.27 8.27 0.82 0.32

1.35 6.61 29.70 1.55 ND 0.05

1.26 3.14 17.22 5.57 0.80 0.32

NOTE: DM-medium, defined minimal medium; ND, not determined since AMPIADP pools were too low.

pool levels in D. discoideum amoebas grown under different conditions. The levels of the purine nucleotide pools were always higher than the levels of the pyrimidine nucleotide pools, with the adenylate pools maintaining 3- to 5-fold higher concentrations than the corresponding guanidylate pools. The ATP and UTP pools remained stable in stationary phase cells; in contrast, CTP and GTP pools were reduced by greater than 50%. When the amoebas were grown in a defined minimal medium, the relative concentrations of the triphosphate pools were comparable to those obtained when the cells were grown in the complex HL-5 medium; however, the absolute values were significantly lower. The maintenance of lower pool sizes appeared to be associated with the differences in the growth rate (doubling time of 19 h in minimal defined medium versus 8- 10 h in HL-5).

Effects of exogenous pyrimidines on triphosphate pool levels Dictyostelium amoebas were grown in defined minimal medium in the presence of selected pyrimidine bases and nucleosides (Table 3). The cells were collected during exponential growth and the acid-soluble pools extracted. Cytosine, thymine, and thymidine showed no effect; however, all other compounds tested (cytidine, uracil, and uridine) selectively increased the pyrimidine pools. Growth in cytidine increased the CTP pool from 3.6 % of the total nucleotide pools in minimal defined medium to 9.8 % in minimal defined plus 0.1 mg cytidinelml. A similar effect was seen on the UTP pool while no significant changes were observed in either the ATP or GTP pools. Growth in the presence of either uridine or uracil increased the CTP and UTP pools, although the effect of uridine was only slight.

Nucleotide pool analysis through development of D. discoideum When D. discoideum amoebas were subjected to starvation under conditions that allowed cellular aggregation and differentiation, the nucleotide pools were reoriented (Table 4). ATP pools had decreased by more than 90% to about 0.29 nmol/5 X lo7 cells by 50 h after the initiation of development. The GTP and UTP pools had similar decreases and CTP was not detectable after 30 h. All of the nucleoside triphosphate pools decreased progressively through the first 7 h, with the exception of GTP which experienced a transient increase at 3 h and then decreased. After morphogenic development became firmly established (8- 10 h), all triphosphate pool levels decreased steadily. Although there were dramatic changes in the ATP pools and a gradual decrease in total adenylates, the adenylate energy charge (E.C., = [ATP] + O.S[ADP]/[AMP] + [ADP] + [AMP]) did not significantly decrease until 24 h after the initia-

TABLE3. Nucleotide triphosphate pool comparison in the amoebas of D. discoideum grown on exogeneous sources of pyrimidines Pyrimidine (mg/mL) added to DM-medium None Cytosine 0.01 0.1 Cytidine 0.01 0.1 Uracil 0.01 0.1 Uridine 0.01 0.1 Thymine 0.01 0.1 Thymidine 0.01 0.1

% of total nucleoside triphosphate pool as:

CTP

UTP

ATP

GTP

3.6

7.7

63.3

20.6

3.1 2.8

7.3 7.8

67.1 70.0

22.5 19.4

4.4 9.8

15.1 17.3

64.6 60.0

16.0 12.4

5.9 7.2

12.5 16.9

62.1 60.3

19.5 15.6

3.8 4.3

7.7 8.3

60.2 67.3

28.3 20.1

6.1 6.1

12.1 11.O

62.5 64.8

19.4 18.0

3.0 3.1

6.5 5.4

69.5 71.6

20.9 19.9

tion of development (Table 4; Fig. 3). In contrast, under starvation conditions that precluded cell-cell contact (shaking cultures) and morphogenetic development, the ECA was unstable and had dropped to below 0.5 by 10 h, and to less than 0.1 by 50 h after the initiation of starvation. The drop in ECA at 5 h in cells prevented from undergoing morphogenic development can be explained by an increase in CAMP, which cannot be separated from AMP in our gradient chromatography system. The cAMP signal and relay system initiates activity about 2 h into the developmental cycle and reaches maximal activity after about 5 h. Developing cells exposed to pulses of cAMP accumulate extracellular cAMP for up to 14 h as a result of signal amplification. Inhibition of cAMP signaling would prevent the accumulation of extracellular cAMP (Saxe et al. 1988).

Arginine biosynthesis and catabolism When grown in defined minimal medium, D. discoideum had an essential requirement for L-arginine (3.3 mM; Franke and Kessin 1977) indicating that it could not synthesize arginine de novo. However, there could be an abbreviated pathway which would use omithine or citrulline instead of arginine. To evaluate this possibility, enzyme assays and nutritional feeding experiments were performed. Neither OTCase nor argininosucI

CAN. J. MICROBIOL. VOL. 35, 1989


TABLE4. Nucleoside triphosphate pool concentrations (nmol X lo7 cells), EC,, and GTPIATP ratios during the developmental cycle of D. discoideum Time, ha

CTP

UTP

ATP

GTP

1 2 3 4 5 6 7 8 10 12 14 24 30 50

3.3(1.8)b 2.6(0.3) 2.1(0.64) 1.8(0.18) 1.4(0.16) 1.2(0.47) l.O(O.2) 1.3(01) 1.3(0.41) 1.2(0.05) 1.2(0.10) 0.7(0.03) 0.006

4.6(0.9) 3.7(0.31) 3.0(0.51) 2.8(0.14) 2.3(0.31) 1.7(0.4) 1.6(0.13) 2.3(0.31) 2.3(0.67) 2.3(0.67) 2.4(0.04) 0.7(0.04) 0.1 0.06

15.2(2.9) 13.4(1.3) 12.5(2.6) 12.3(1.4) 7.9(2.5) 6.9(1.2) 6.3(0.04) 8.9(1.8) 8.9(3.3) 8.9(3.3) l(1.2) 5.7(1.4) 0.53 0.29

5.6(1.6) 4.1(0.16) 4.9(0.13) 3.1(0.53) 2.8(0.50) 2.3(0.64) 2.1(0.30) 2.7(0.50) 2.7(0.88) 2.7(0.88) 2.9(0.3) 2.0(0.12) 0.15 0.08

ndc

EC,

0.86 0.87 0.85 0.87 0.83 0.83 0.85 0.87 0.87 0.87 0.85 0.81 0.81 0.74

GTPIATP

0.37 0.31 0.39 0.25 0.35 0.33 0.33 0.29 0.30 0.30 0.36 0.34 0.28 0.27

"Afterthe initiation of the developmental cycle. bThe value in parenthesis is the standard deviation of 6 independent determinations at each time point with the exception of 30 and 50 h, which were only evaluated once. 'nd, not dectectable.

HOURS OF DEVELOPMENT FIG. 3. Adenylate energy charge of D. discoideum during the developmental cycle ( A ) and under conditions of vigorous shaking which Adenylate pool levels were determined as described in Materials and methods. The stages of the preclude morphogenic development (0). developmental cycle are indicated at the top of the figure. The time course indicates hours after the initiation of development by removal of the nutrient source.

cinate synthetase activity was detectable. Furthermore, neither citrulline nor omithine (3.3 to 33 mM) supported growth at any concentration tested in defined medium. Both [14C]arginine and [14C]citrullinewere taken up to the same extent by actively growing amoebas as indicated by the counts incorporated into the TCA-soluble fraction (1766 cpm for [14C]arginineversus 1374 cpm for [14C]citrulline,per lo5cells). However, the relative incorporation into protein was significantly different (202 8 19 cpm for [14C]arginineversus 1113 cpm [14C]citrul-

line, per milligram of protein). Both labelled compounds were added to defined medium at a specific activity of 25.5 pCilpmol (1 Ci = 37 GBq). The whole cells were washed in 100 mM KH2P04until free label was at or below background (80 cpm). The TCA-soluble and -insoluble fractions were then extracted and counted. No radioactivity from either source was incorporated into purine or pyrimidine nucleotides separated by HPLC as described previously. In addition, neither catabolic OTCase nor arginine deiminase activity was detectable.

437


WALES ET AL.

Growth studies in fluorinated analogues of the pyrimidine bases and nucleosides Dictyostelium amoebas maintained in complex axenic medium were transferred to defined minimal medium 3-5 days before the start of the growth inhibition studies. High concentrations of 5-fluorocytosine (1.0-7.5 mM) did not affect the growth rate, while significant retardation was observed with 5-fluorocytidine, 5-fluorouridine, and 5-fluorouracil at concentrations as low as 0.05 mM (data not shown).

Discussion Dictyostelium discoideum had an absolute requirement for arginine that could not be satisfied by either omithine or citrulline. While citrulline and arginine were taken into the acidsoluble pool equally, citrulline was not incorporated into protein. Other eukaryotic microorganisms (Crithidia) with an absolute arginine requirement are capable of utilizing arginine or citrulline degradation to provide carbamoyl phosphate for pyrimidine biosynthesis (Makoff and Radford 1978). However, it was not possible to demonstrate the catabolism of arginine or citrulline for pyrimidine biosynthesis either directly by metabolic labelling or indirectly by demonstrating the presence of the appropriate enzymes. As a result, pyrimidine biosynthesis in D. discoideum appears to be provided through a single carbamoyl phosphate pool that is specific for pyrimidine biosynthesis and not utilized for arginine biosynthesis. Three different classes of CPSasc have been described for arginine and (or) pyrimidine biosynthesis (Aoki et al. 1980). Arginine-specific CPSase I has been shown to utilize ammonia as the nitrogen donor and absolutely requires N-acetyl-L-glutamate as an enzyme activator. The pyrimidine-specific CPSase 11utilizes L-glutamineas a nitrogen donor, has no requirement for N-acetyl-L-glutamate, is feedback inhibited by a uracil nucleotide, and is activated by 5-phosphoribosyl-1-pyrophosphate. This enzyme is contained in various multienzyme complexes in eukaryotic systems in which it is strongly inhibited by UTP. Most bacteria possess a single CPSase subject to UMP inhibition (class 11) that is not involved in a multienzyme complex. CPSase 111utilizes L-glutamine, requires N-acetyl-Lglutamate, and is localized in the mitochondria. The enzyme from D. discoideum most closely resembled the CPSase I1 enzymes since it required L-glutamine as a substrate, had no requirement for N-acetyl-L-glutamate,and was partially inhibited by both UTP and CTP. ATCase exists in a variety of multimeric aggregates and multifunctional enzyme complexes in various organisms (Table 5; for review see Makoff and Radford 1978; Jones 1980). The simplest organization is a nonregulated trimer reported in Gram positive bacteria (class C ATCase; Bethel1 and Jones 1968). The enteric bacteria possess an allosterically regulated enzyme composed of 2 catalytic trimers associated with 3 regulatory dimers (class B ATCase). In some lower eukaryotes (e.g., Saccharomyces and Neurospora), ATCase and CPSase are expressed as a bifunctional protein complex, whereas in mammals, amphibians, and Drosophila, ATCase, CPSase, and DHOase are synthesized as a trifunctional enzyme (CAD, 800 - 1200 kDa; Makoff and Radford 1978). In higher plants (e.g., Triticum, Pisum, and Phaseolus), ATCase does not appear to participate in either of these enzyme complexes (Yon et al. 1982). The D. discoideum ATCase in these studies appeared to have the general characteristics of plant and class C bacterial ATCases: a molecular mass of 105 kDa, hyper-

TABLE5. Comparison of the first three enzymes in pyrimidine biosynthesis, presenting structural organization and molecular mass (kDa) CPSase

E. ~01i B. subtilis

(YP

(160)

DHOase

c6r6(300)

012

I

a@(100)

,Cg (100) ,

I

S. cerevisiae

\

Mammalian

\

D. discoideum

ATCase

(CAI, (800)

I

(80) I

a2

(761,

? I

(CAD), (600 - 800) D ?

? (105)

1

? (90)

\ I -

NOTE: Greek letters denote kinds of subunits: @ means the enzyme is a , chain, the sequences of a n d B being heterodimer containing one g and one @ different. Small letters denote chains by function: c, catalytic chain; r, regulatory chain. Capital letters signify superdomain units corresponding to individual enzymatic activities: C, carbamoyl-phosphate synthetase; A, aspartate transcarbamoylase; D, dihydmomtase. Values in parentheses are molecular mass in kilodaltons.

bolic substrate kinetics, and no effector responses. It has been suggested that CPSase, ATCase, and DHOase are part of a multifunctional complex in D. discoideum (Kessin 1988). It is possible that these enzymes are organized into a multienzyme complex in D. discoideum since their molecular mass profiles are similar and multiple molecular mass forms of ATCase have been observed. The physical separation of the enzymes could result from proteolysis between functional superdomains, which would explain the instability of CPSase activity. An alternative route of providing for pyrimidine requirements is to salvage and reutilize RNA degradation products with subseauent nucleotide interconversions. When D. discoideum amoebas were grown in an exogeneous source of pyrimidine nucleosides and bases, it was found that uracil, uridine, and cytidine increased the UTP and CTP pools while cytosine had no effect. This indirect evidence for the uptake and utilization of uracil, uridine, and cytidine was supported by growth sensitivity on 5-fluorinated analogues in which 5-fluorocytosine had no effect on growth rate at low concentrations while the uracil, uridine, and cytidine analogues did. Nucleotide pools have been extensively analyzed in prokaryotes with respect to regulatory control and adenylate energy charge; however, few studies have been oriented toward determining changes in nucleotide pools in a nonsteady-state, developing system. During morphogenic development in D. discoideum, there are major shifts in RNA, DNA, and protein synthesis, which are reflected by changes in nucleotide metabolism and in the levels of the nucleotide pools. The adenylate energy charge (Atkinson 1969) measures-the coupling of energy-producing and energy-utilizing metabolic sequences and is a tightly controlled parameter in metabolically stable cell systems (Knowles 1977). Even when prokaryotic cells are subjected to metabolic stress (such as limitation of energy or nitrogen sources), only a moderate change in the energy charge (0.95 to 0.80) is usually observed. The adenylate energy charge concept has been applied to eukaryotic systems although the measurable levels of adenine nucleotides are only the composite average of compartmentalized endogenous pools. Nevertheless, the ECA values obtained for D. discoideum in this study ( > 0.80 throughout development) correspond with the maintenance of


438

CAN. J. MICROBI(3L.

appropriate homeostasis according to the energy charge theory. Another important parameter is the positive correlation of endogenous GTPIATP pool ratios with general cellular homeostasis o r developmental reorientations (Karl 1978). Both the adenylate energy charges and the GTPIATP ratios in actively growing cells were similar regardless of the growth medium o r the growth rates. The purine nucleotide pool ratios were very stable throughout the normal developmental cycle, while the pyrimidine nucleotide pools exhibited considerable variation in both concentration and ratio. These differences could be related to the proposed multivarious roles of purine nucleotides in regulation, while the pyrimidine pools would fluctuate more specifically in response to cellular demand for nucleic acid biosynthesis. Overall, the levels of the nucleotide pools decreased during development with the most rapid decline occurring in the first 7 h prior to the late aggregation stages, the most active period for metabolic and morphological reorientation. By these evaluations, both the adenylate energy charge and GTPIATP theories of regulation of nucleotide pools are consistent with maintaining the homeostatic regulatory posture of D. discoideum during development. I n summary, D. discoideum utilizes both d e novo biosynthesis and salvage-reutilization of nucleosides and bases for providing pyrimidine nucleotides, and specifically manages its nucleotide metabolism to maintain a n appropriate adenylate energy charge during development, even during a decline in intracellular reserves.

Acknowledgement This research has been supported by the National Institutes of Health (GM-33191). M. 1980. Control of AOKI, T., YA, H., MORI,M., and TATIBANA, pyrimidine biosynthesis in the ascaris ovary: regulatory properties of glutamine-dependent carbamoylphosphate synthetase and copurification of the enzyme with aspartate carbamoyltransferase. Mol. Biochem. Parasitol. 1: 55 -68. ARCHIBALD, R. M. 1944. Determination of citrulline and allantoin and demonstration of citrulline in blood plasma. J. Biol. Chem. 156: 121 - 142. ATKINSON, D. E. 1969. Regulation of enzyme function. Annu. Rev. Microbiol. 23: 47 -68. BETHELL,M. R., and JONES,M. E. 1969. Molecular size and feedback-regulation characteristics of bacterial aspartate transcarbamoylase. Arch. Biochem. Biophys. 134: 352 -365. BLUMBERG, D. D., and LODISH,H. F. 1980. Changes in the messenger RNA population during differentiation of Dictyostelium discoideum. Dev. Biol. 78: 285 -300. BOY-MARCOTTE, E., VILAINE,F., CAMONIS, J., and JACQUET,M. 1984. A DNA sequence from Dictyosteliurn discoideum complements ura3 and ura5 mutations of Saccharomyces cerevisiae. Mol. Gen. Genet. 193: 406-413. CHEN,S. C., BROWN,P. R., and ROSIE,D. M. 1977. Extraction procedures for use prior to high performance liquid chromotography nucleotide analysis using microparticle chemically bonded packings. J. Chromatog. Sci. 15: 218-221. DEVROETES, P. N. 1982. Chemotaxis. In The development of Dictyostelium discoideum. Edited by W . F. Loomis. Academic Press, New York. pp. 117-165. FOLTERMANN, K. F., SHANLEY,M. S., and WILD, J. R. 1984, Assembly of the asparate transcarbamoylase holoenzyme from transcriptionally independent catalytic and regulatory cistrons. J. Bacteriol. 157: 891 -897. FRANKE, J., and KESSIN,R. 1977. A defined minimal medium for axenic strains of Dictyostelium discoideum. Proc. Natl. Acad. Sci. U.S.A. 74: 2157-2161.

VOL. 35, 1989

1981. The cAMP phosphodiesterase inhibitory protein of Dictyosteliurn discoideum. Purification and characterization. J. Biol. Chem. 117: 213-218. GEORGE,R. P., HOHL,H. R., and RAPER,K. 1972. Ultrastructural development of stalk-producing cells in Dictyosteliurn discoideum, a cellular slime mould. J. Gen. Microbiol. 70: 447-489. GERHART, J. C., and PARDEE,A. B. 1962. The enzymology of control by feedback inhibition. J. Biol. Chem. 237: 891-896. INGRAHAM, J. L., and ABDELAL,A. T. H. 1978. Carbamoyl phosphate synthase (glutamine): Salmonella. Methods Enzymol. 51: 29-35. JACQUET, M., BOY-MARCOTTE, E., ROSSIER,C., and KESSIN,R. H. 1982. A fragment of Dictyostelium discoideum that complements the ural mutation of Saccharomyces cerevisiae. J. Mol. Appl. Genet. 1: 513-525. JONES,M. E. 1980. Pyrimidine nucleotide biosynthesis in animals: genes, enzymes, and regulation of UMP. Annu. Rev. Microbiol. 49: 253 -279. JONES,P. C. T. 1970. The interaction of light and temperature in determining ATP levels in the myxamoeba of the cellular slime mould Dictyosteliurn discoideum. Cytobios, 6: 89 -94. KARL,D. M. 1978. Occurrence and ecological significance of GTP in the ocean and in microbial cells. Appl. Environ. Microbiol. 36: 349 -355. KESSIN,R. H. 1988. Genetics of early Dictyostelium discoideum development. Microbiol. Rev. 52: 29 -49. KNOWLES,C. J. 1977. Microbial metabolic regulation by adenine nucleotide pools. Symp. Soc. Gen. Microbiol. 27: 241 -283. KONIJN,T. M., BARKLEY, D., CHANG, Y. Y., and BONNER, J. 1968. Cyclic AMP: a naturally occuning acrasin in the cellular slime molds. Am. Nat. 102: 225-233. LIDDEL,G. U., and WRIGHT,B. E. 1961. The effect of glucose on respiration of the differentiating slime mold. Dev. Biol. 3: 265 -276. LOOMIS,W. F. 1980. Genetic analysis of development in Dictyostelium. In The molecular genetics of development. Edited by T. Leighton and W. F. Loomis. Academic Press Inc., Orlando, FA. pp. 179-212. LOWRY,0 . H., ROSENBROUGH, N. J., BARR,A. L., and RANDALL, R. J. 1951. Protein measurements with the folin-phenol reagent. J. Biol. Chem. 193: 265 -275. MAKOFF,A. J., and RADFORD, A. 1978. Genetics and biochemistry of carbamoyl phosphate biosynthesis and its utilization in the pyrimidine biosynthetic pathway. Microbiol. Rev. 42: 307-328. ORLOW,S. J., SHAPIRO, R. I., FRANKE, J., and KESSIN,R. H. 1981. The extracellular cyclic nucleotide phosphodiesterase of Dictyostelium discoideum: purification and characterization. J. Biol. Chem. 256: 7620 -7627. L. M., and JONES,M. E. 1969. Modified methods for the PRESCOTT, determination of carbamyl aspartate. Anal. Biochem. 32: 408419. RAUSHEL,F. M., and SEIGLIE,J. L. 1983. Kinetic mechanism of argininosuccinate synthetase. Arch. Biochem. Biophys. 225: 979985. SAXE,C. L., 111, SAXE,S. S., and KIMMEL,A. R. 1988. Gene expression in Dictyostelium discoideum: the changing cAMP story. ASM News, 54: 293-298. SUSSMAN, M. 1966. Protein synthesis and the temporal control of genetic transcription during slime mold development. Proc. Natl. Acad. Sci. U.S.A. 58: 813-818. SUSSMAN, M., and BRACKENBURY, R. 1976. Biochemical and molecular genetic aspects of cellular slime mold development. Annu. Rev. Plant Physiol. 27: 229-265. WHITE, G. J., and SUSSMAN, M. 1961. Metabolism of major cell components during slime mold morphogenesis. Biochim. Biophys. Acta, 53: 285-293. YON,R., GRAYSON, J. E., CHAWDA, A., and BUTTERWORTH, P. J. 1982. The quaternary structure of wheat-germ aspartate transcarbamoylase. Biochem. J. 203: 413 -417.