Early mitogenic stimulation of metabolic flux through phosphoribosyl pyrophosphate into nucleotides in Swiss 3T3 cells and requirement of external Magnesium for the response

EARLY MITOGENIC STIMULATION OF METABOLIC FLUX THROUGH PHOSPHORIBOSYL PYROPHOSPHATE INTO NUCLEOTIDES IN SWISS 3T3 CELLS AND REQUIREMENT OF EXTERNAL MAGNESIUM FOR THE RESPONSE M. TATIBANA, S. ISHIJIMA, K. KITA, T. 1SHIZUKA and N. SUZUKI Department of Biochemistry, Chiba University School of Medicine, Inohana, Chiba 280, Japan

INTRODUCTION

5-Phosphoribosyl 1-pyrophosphate (PRPP) provides the ribose phosphate moiety to the de novo and salvage synthesis of ribonucleotides. It also serves as a regulator for amidophosphoribosyltransferase (1, 2) and carbamoylphosphate synthetase II (3, 4), key enzymes of purine and pyrimidine de novo synthesis. Purine salvage reactions are also limited by this substrate (1). PRPP synthetase (EC2.7.6.1) has been purified from Salmonella typhimurium (5), human erythrocytes (6), rat liver (7), and Escherichia coli (8). The enzyme activity in various mammalian tissues including tumors was determined (9). ADP and many other nucleotides inhibit the enzyme (for a review, see Ref. 10) and the aggregation states also affect the activity (11). However, findings of the regulatory properties previously reported are not always in agreement. The known control factors do not appear to fully account for the in situ regulation of the enzyme (12). We previously studied protein diet-induced stimulation of purine and pyrimidine biosynthesis in mouse liver, to define the activator function of PRPP in situ (13); the hepatic level of PRPP showed a remarkable increase after protein diet ingestion. However, the known control factors for PRPP synthetase, such as concentrations of ATP, ADP, inorganic phosphate and ribose 5-phosphate, did not show significant changes (unpublished results). We thought that something important remains unknown in the in situ regulation of this enzyme. Toward elucidation of the question, we have taken three approaches, i.e., enzyme studies, genetic analysis, and cellular studies, although it is not possible at present to integrate these studies. 147

148

M. TATIBANA, et aL

We purified PRPP synthetase from rat liver and showed that it is composed of several proteins, of which the 34 kDa species is the catalytic subunit (14). Oligonucleotide probes were synthesized based on the partial amino acid sequences of that subunit and were used for screening a cDNA library from rat Yoshida sarcoma cells. Unexpectedly, two distinct clones were obtained and ensuing studies revealed the presence of at least two isoforms of the enzyme subunit (PRS I and II), encoded by different genes (PRPS 1 and 2) in mammalian tissues (15). Both clones encode 317 amino acids and the deduced amino acid sequences of the two differ only by 13 residues. More recently, we found tissue-differential expression of the two genes and also the presence of an additional gene (PRPS 3), which is specifically expressed in the testis (16). Thus, the structure of mammalian PRPP synthetase is more complex than previously believed, although any functional difference between these isoforms remains to be elucidated. There were two pioneer research groups for studies of the regulation of PRPP synthesis in cultured cells. Buchanan et al. (17) showed that in mouse Swiss 3T6 quiescent fibroblasts in culture, addition of fresh serum activated purine and pyrimidine de novo biosynthesis within 30 min as an early response to growth stimulation. Becker etal. (18) reported that various mitogenic stimuli increased the uptake of guanosine and purine bases into Swiss 3T3 cells within 1 hr. These authors presumed that the increased synthesis of 5-phosphoribosyl 1-pyrophosphate (PRPP) may precede or underlie these activations, though little direct evidence was presented. We previously studied glucagon-induced elevation of PRPP concentration in isolated rat hepatocytes (19), in parallel with the work on mouse liver in vivo. However, the hormonal response of rat hepatocytes was rather variable in our hands, and we changed to use the Swiss 3T3 cell system. In the present work we prepared and used [ribosyl-14C]inosine for an assay of the metabolic flux through PRPP into nucleotides within the cells (20). Epidermal growth factor (EGF) plus insulin, as well as some other growth factors, did increase the flux in Swiss 3T3 cells within 1 hr (20) and the response required the presence of external Mg 2÷.

MATERIALS

AND METHODS

Materials

Mouse EGF was obtained from Toyobo Co., bovine pituitary fibroblast growth factor from Takara Shuzo Co., and bombesin, insulin, snake venom (Naja naja), and hypoxanthine from Sigma. Radioactive materials were obtained either from Amersham or from New England Nuclear (20). Na214CO 3 was prepared from Bal4CO3 . Purine-nucleoside phosphorylase (EC 2.4.2.1, calf spleen) was obtained from Boehringer Mannheim. PRPP

MITOGENIC STIMULATION OF PRPP SYNTHESIS

149

synthetase was highly purified from rat liver (Kita, K. et al., manuscript in preparation). Other materials were of the highest grade from commercial sources.

Preparation of [Ribosyl-14C]Inosine [Ribosyl-14C]inosine was prepared from [U-14C]guanosine (0.5 raM, 280 mCi/mmol) and hypoxanthine (10 mM) by an exchange reaction of purine-nucleoside phosphorylase (0.8 unit) as described in (20). The yield of radioactivity in [ribosyl-14C]inosine at 4 hr was 35%. Paper chromatography with a solvent of 5% Na2HPO 4 saturated with isopentyl alcohol was used for purification of the product. The purified inosine showed a characteristic UV spectrum and contained no radioactivity in the base moiety.

Cell and Cell Culture Swiss 3T3 cells were kindly provided by Dr. Y. Takai (Kobe University School of Medicine, Kobe, Japan). The ceils were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% (v/v) fetal calf serum under a humidified atmosphere of 5% CO2/95% air in a subconfluent state with subcultures every 3 or 4 days (21). For experiments, the cells were cultured initially with 7 x 104 cells per 35mm dish in 2 ml of DMEM containing 10% fetal serum. The cells were fed after 2 days with fresh medium containing 10% fetal serum and were used after 6-7 days, by which time the cells were confluent and quiescent. Swiss 3T6 cells (ATCC CCL96) were obtained from Japanese cancer Resources Bank and maintained in DMEM supplemented 10% (v/v) calf serum. Resting monolayers were prepared as described in (17). Human fibroblasts, established from an embryo in our laboratory, were maintained in Eagle's minimum essential medium supplemented with 10% (v/v) calf serum. One or two days before the experiment, the cells were transferred to fresh medium containing 0.5% serum.

Mitogenic Stimulation and Radiolabeling of Cells Quiescent cells were washed twice with DMEM and incubated generally for 45-60 min at 37°C in 1 ml of DMEM containing growth factors or their combinations. The medium was removed and the cells were washed once with DMEM and incubated with a radioactive compound in 0.5 ml of DMEM at 37°C for 30 min. The amount of tracers used for one dish was 0.025/~Ci (0.5/zM) of [ribosyl-~4C]inosine, 0.15/zCi (20/xM) of [8-14C]adenine, 0.4/xCi (18/.~M) of [1-14C]glycine or 1.5/xCi (50 p.M) of [14C]formate. A minimal concentration of the labeled inosine was used so as not to interfere with intracellular ribose metabolism.

150

M. TATIBANA,et al.

When Na214CO3 was used as the labeled substrate, the mitogen-treated cells were washed once with medium consisting of 140 mM NaC1, 5 mM KCI, 1.8 mM CaCI2, 0.8 mM MgSO4, 1 mM NazHPO4, 25 mM glucose, and 25 mM N-(2-hydroxyethyl)piperazine-N'-2-ethanesulfonic acid (Hepes)-NaOH (pH 7.2) and incubated with the same medium containing Na214CO3 (2.3/zCi). After radiolabeling, the cells were washed three times with DMEM and extracted with 5% trichloroacetic acid at 4°C.

Measurement of 14C Incorporation into Cellular Nucleotides The cells were extracted with triehloroacetic acid and the extract was treated with diethyl ether to eliminate the acid and neutralized with 0.2 M NH4OH. When [14C]adenine was used as a tracer, the nucleotides were adsorbed to DEAE-cellulose paper and the paper was treated and counted as described in (22). When other tracers were used, the sample was fractionated on a column of Dowex l-X8 (formate form, 0.4 × 0.8 era), major nucleotides were eluted as a group and the charcoal-adsorbed radioactivity in the eluate was counted (20). Results are generally expressed as percent of the control value with no growth factors and represent the means + S.D. of at least 3 independent experiments. Separation of Nucleotides and Nucleosides by H P L C The acid-soluble extracts of the cells, treated with diethyl ether, were subjected to analysis by high performance liquid chromatography (HPLC) on a Waters Partisil 10-SAX radial-compression column (0.8 × 10 cm) according to Reiss et al. (23). In determination of the specific radioactivity of total adenine, guanine, and uracil nucleotides, the nucleotides were first converted to nucleosides by treatment with 50/zg of snake venom, which contains phosphodiesterase (venom exonuclease, EC 3.1.15.1) and 5'-nucleotidase (EC 3.1.3.5), in 150 tzl of 0.2 M NHgHCO3 for 1 hr at 37°C. The nucleosides formed were separated by HPLC on a Waters tzBondapak Cas radial-compression column (0.8 x 10 cm), according to Hartwick et al. (24). Separation of Cellular PRPP and Ribose 5-Phosphate The cells were grown in larger (60mm) dishes. After treatment with and without mitogens, the cells were incubated in DMEM for 15 rain. In experiments to determine the specific radioactivity of ribose 5-phosphate, 0.48 ~Ci of [ribosyl-14C]inosine was added. The content of PRPP of the acid-soluble extract was determined by a modification (13) of the method previously reported (25). For separation of cellular ribose 5-phosphate,

MITOGENIC STIMULATION OF PRPP SYNTHESIS

151

the acid-soluble extracts from two dishes were quickly neutralized, treated with charcoal, and filtered. Ribose 5-phosphate in the filtrate was partially purified by column chromatography on Dowex 1 and DEAE-Toyopearl 650S and was determined by conversion to PRPP (26) in the presence of excess rat liver PRPP synthetase. Other Measurements [3H]Thymidine incorporation into acid-insoluble material was measured as described in (27). Protein was determined according to Bradford (28), using bovine serum albumin as a standard. The activity of PRPP synthetase in cell extracts was determined essentially as described by Fox and Kelley (6) except that the amount of PRPP synthesized was determined according to Kornberg et al. (29).

RESULTS

AND DISCUSSION

Rationale for Selection of [14C-Ribosyl]Inosine as Tracer Our studies required a simple, convenient, and more direct method for evaluation of the metabolic flux through PRPP into nucleotides in the cells. Various tracers have so far been used in studies of nucleotide biosynthesis. Radioactive adenine has often been used to measure "PRPP availability" (22, 30). Conversion of adenine to nucleotides depends on adenine phosphoribosyltransferase activity as well as PRPP generation. The enzyme is, however, subject to strong inhibition by AMP (31) the reaction product, and on the other hand, has a low K m value for adenine (31) so that adenine, when exogenously added, can deplete intracellular PRPP and strongly affect nucleotide metabolism. Therefore, the use of radioactive adenine is not very suitable for studying regulatory mechanisms of PRPP synthesis. Other tracers commonly used are radioactive glycine, formate (for purine synthesis), bicarbonate, aspartate, orotate (for pyrimidine synthesis), nicotinate, nicotinamide (for pyridine nucleotide synthesis), and so on. To measure the flux through PRPP into nucleotides, synthetic rates of purine, pyrimidine, and pyridine nucleotides should be separately determined and then summed. Furthermore, many tracers only provide relative measures, because of low permeability through cell membrane (aspartate and orotate) and of difficulty in measuring specific radioactivity of their intracellular pools (formate and bicarbonate). As the ribose moiety of PRPP is derived from ribose 5-phosphate in cells by the reaction catalyzed by PRPP synthetase (ribose 5-phosphate + ATP - ~ PRPP + AMP), the flux through PRPP into nucleotides is measurable by counting radioactivity incorporated from ribose phosphates

152

M. TATIBANA, et al.

into nucleotides. However, neither ribose nor ribose phosphates may serve as tracers in this work, because fibroblasts appear to lack ribokinase and are practically impermeable to ribose phosphates. Accordingly, we used [ribosyl-t4C]inosine as a tracer. Owing to lack of inosine kinase activity in fibroblasts (32), it is expected that the ribose moiety of inosine is primarily converted to ribose 1-phosphate in the cells, and isomerized to ribose 5phosphate, which is then incorporated into nucleotides through PRPP (Fig. 1). An alternative pathway from ribose 1-phosphate to nucleotides involves

Inosine~ Hypoxanthine~ Guanin~

. Rioose 1-phosphate~----~Uridine~ ~ ~Uracil Cytidine

Guanosine~Ribose

5-phosphate Cytosine

1 J~ Purine Nucleotides

Purine Bases I NAD

Pyrimidine Nucleotides

FIG. 1. Metabolic pathways for incorporation of ribose moiety of inosine into nucleotides in fibroblasts. The pathways (1) and (2) represent de n o v o and salvage purine synthetic pathways, respectively. Broken lines represent pathways considered insignificant in these ceils.

formation of uridine, catalyzed by pyrimidine-nucleoside phosphorylase, and conversion of uridine to UMP. This route plays only a minor role in the utilization of the ribose moiety of inosine, as shown in a later section. Similar activity for PRPP-independent formation of purine nucleotides is considered insignificant (33).

Fate of Ribose Moiety of lnosine in Cells When the cells, either treated with mitogens or not treated, were incubated with [ribosyl-laC]inosine, the radioactivity was incorporated into the nucleotide fraction reaching a maximum at about 20 min and remained at that level for a further 20 rain at least (Fig. 2). The radioactivity in the cells at 30 min was about 8% of the amount added to the medium; 7% was in the acid-soluble and 1% in the acid-insoluble fractions. The radioactivity in nucleotides occupied 34-55% of that in the total acid-soluble fraction. Conversion to 14CO2 during the period amounted to only 0.2-0.3% of the total 14C added. Based on the results in Figure 2, cells were incubated with the tracer for 30 min in all further experiments, unless stated otherwise.

MITOGENIC STIMULATION OF PRPP SYNTHESIS I

60

I

I

153

I

Total Acid-soluble

i,° 2(~

10

20

30

40

Period of Incubation with Tracer (rain)

FIG. 2. Time course of 14Cincorporation from [ribosyl-]4C]inosineinto nucleotides and nucleic acids in Swiss3T3 cells, treated with mitogens or not treated. The cellswere incubated with (11, O, &) and without ([], O, A) 10 ng/ml of EGF plus 100 ng/ml of insulin for 45 min, followed by washing and incubation with 0.025 p.Ci (0.5 ~*M)of [ribosyl-14C]inosinefor indicated times and the radioactivities in the total acid-soluble ( I , []), nucleotide (Q, O), and acid-insoluble (&, A) fractions (nucleic acids) were determined. From Ref. (20).

Stimulation of ~4C Incorporation into Nucleotides by EGF plus Insulin Prior treatment of the cells with E G F plus insulin increased the rate and extent of radioactivity incorporation into the nucleotides (charcoal-adsorbed fraction) (Fig. 2). Effects of exposure time on the t4C incorporation are shown in Figure 3. A maximal stimulation occurred after 45--60 rain exposure, whereas no stimulation was observed after 3 hr exposure. E G F and insulin had a synergistic effect on the radioactivity incorporation into nucleotides, in agreement with the report by Becker et al. (18). Insulin alone (100 ng/ml) induced a 25% increase and E G F alone (10 ng/ml) only a 10-15% increase. When the two were added together, the incorporation increased by 90%. A higher concentration of E G F (100 ng/ml) was far less effective. When cellular synthesis of D N A in response to growth factors was measured by [aH]thymidine uptake into the acid-insoluble fraction after 22 hr, it showed a similar dependency on E G F concentrations in the presence of insulin, in accord with an earlier report (34). To further define the effects of E G F and insulin, the nucleotides in the acid-soluble extracts were separated by H P L C and their specific JAER 28~F*

154

M. TATIBANA, et al. I

I

I

0

]3o °-

,.- '~ 2 0 O

m'. _c

0

I

I

I

1

2

3

Exposure Time (hr) FIG. 3. Effects of prior exposure to EGF plus insulin for various periods on ]4C incorporation into nucleotides. The cells were incubated for indicated times with (0) and without (©) 10 ng/ml of EGF plus 100 ng/ml of insulin and further processed for determination of radioactivity in nucleotides as in Figure 2. Incubation with [ribosyl-14C]inosine was for 30 min. From Ref. (20).

radioactivities were determined. EGF plus insulin induced a 1.6- to 2.1-fold increase in specific radioactivity of ADP, ATP, GTP, and UTP (Table 1). The results ensured the validity of the routine method using charcoal for measurement of radioactivity incorporation into nucleotides.

Evidence for Increase in Metabolic Flux through PRPP into Ribonucleotides We further explored the nature of the increase of specific radioactivity of the nucleotides. Such increase may result from an increased cellular permeability to the tracer, a decreased flux from nucleotides to nucleic acids or an increased flux through the PRPP-independent pathway. These possibilities were examined. First, EGF plus insulin increased TABLE 1. INCREASE IN SPECIFIC RADIOACTIVITY OF ISOLATED NUCLEOTIDES BY EGF PLUS INSULIN Addition None EGF + insulin

ADP

Specific radioactivity ATP GTP

UTP

280 ___38 440 _+ 38

dpndnmol 211 + 8 108 + 19 372 _+ 56 225 + 38

223 + 35 352 _+ 61

The cells were stimulated by 10 ng/ml of EGF in the presence of 100 ng/ml of insulin for 45 min at 37°C and then incubated with [ribosyl-laClinosine as in Figure 2. Cellular extracts were prepared and nucleotides were separated by HPLC. Each value represents the mean + S.D. of 3 independent determinations. From Ref. (20).

MITOGENIC STIMULATIONOF PRPP SYNTHESIS

155

the radioactivity incorporation into nucleotides by 1.9 + 0.2-fold (n=24), while the total radioactivity of the acid-soluble fraction, which may provide a measure of tracer permeation into the cells, showed only a small increase (1.2 ___0.2-fold, cf. Fig. 1). We next determined the specific radioactivity of intracellular free ribose 5-phosphate. It decreased slightly in the stimulated cells; the value was 620 + 114 dpm/nmol vs 752 + 140 dpm/nmol in the unstimulated cells (n=3) under conditions described in Table 1. The above results indicate that the increase in the radioactivity incorporation into nucleotides was not a mere result of enhanced tracer permeation into the ceils. The second possibility is unlikely, since the radioactivity of the acid-insoluble fraction, representing nucleic acids, slightly increased in the stimulated cells (Fig. 2). To examine the third possibility, we studied [2-14C]uracil incorporation into nucleotides, as a measure of the activity of PRPP-independent pyrimidine nucleotide f o r m a t i ~ (Fig. 1). When the cells were incubated with 1.25/zCi of [14C]uracil (24 nmol) for 30 min, the radioactivity incorporated into nucleotides was 3100 and 3200 dpm/mg of protein in the unstimulated and EGF plus insulin-stimulated cells, respectively. As the cellular contents of free uracil can be considered low (35), dilution of radioactive uracil within the cells was disregarded in calculation. The flux of this pathway, thus estimated, was only 1.7% and 0.8%, respectively, of the calculated amount of ribose phosphate incorporated into the total uracil nucleotides (the calculation is presented later). Furthermore, the [laC]uracil incorporation was not increased by EGF plus insulin (1.0 + 0.2-fold increase, n=3). These results indicate that the PRPP-independent pathway for uracil nucleotide synthesis contributed little to the increase in the ribose incorporation into nucleotides. Because the 14C incorporation from [14C-ribosyl]inosine into CTP was less than 10% of that into UTP (data not shown), only the uracil incorporation was studied here. Based on the above observations we concluded that the increased incorporation of radioactivity from [ribosyl-laC]inosine into nucleotides represents a stimulated metabolic flux through PRPP into nucleotides. Calculation of Flux of Ribose 5-Phosphate into Free Nucleotides in Unstimulated and Stimulated Cells Calculation of the flux requires the information on the total amounts of all free nucleotides including nucleoside mono-, di- and triphosphates. For this purpose, nucleotides in cellular extracts were enzymatically converted to nucleosides, and the adenosine, guanosine, and uridine thus formed were isolated by HPLC, The adenosine was derived from adenosine phosphates as well as from adenosine-containing dinucleotides such as pyridine and flavin

156

M. TATIBANA, et aL

dinucleotides. The amounts of nucleosides obtained from the unstimulated and EGF plus insulin-stimulated cells showed no significant differences. However, the radioactivity incorporated from [ribosyl-14C]inosine was increased by EGF plus insulin (1.8 + 0.2-, 2.4 + 0.4-, and 1.9 + 0.1-fold increase for adenine, guanine, and uracil nucleotides, respectively, n=4). Based on the above results and the specific radioactivity of ribose 5-phosphate, the flux into adenine, guanine, and uracil nucleotides was calculated (Table 2). EGF plus insulin induced 2.3- to 2.5-fold increases in the flux. Because of analytical limitations, the flux into cytosine and pyridine nucleotides was not determined, and no correction was made for degradation of nucleotides and efflux from nucleotides to nucleic acids. Therefore, the values presented here only provide minimal estimates for the flux from ribose 5-phosphate into nucleotides.

Incorporation of Other Tracers Stimulation of the nucleotide synthesis was also shown using other tracers. There was a parallel increase in the amounts of aaC-labeled formate, glycine, HCO3-, and adenine incorporated into total nucleotides. Prior incubation with EGF plus insulin for 45 rain resulted in 1.8 + 0.2-, 1.7 + 0.2-, 3.0 + 0.2-, and 2.0 _+ 0.3-fold increases in the incorporation of those labeled compounds, respectively (n=3 or 4). Effects of Other Growth-promoting Factors and Response of Swiss 3 T6 Cells and Human Fetal Fibroblasts The 14C incorporation into nucleotides was also stimulated both by bombesin and melittin, in combination with insulin, and by fibroblast growth factor alone (1.6- to 1.9-fold increase, Table 3). The effect of bombesin or melittin alone was not significant, as was the case with EGF alone as discussed above. These mitogens are known to induce DNA synthesis in Swiss 3T3 cells; the response of nucleotide synthesis is intimately related to DNA synthesis or cellular proliferation. It is interesting to note that transmembrane signaling events induced by these mitogens are believed to be different (Ref. 36 for a review). Bombesin (37, 38) and fibroblast growth factor (39) induce the phospholipase C-mediated hydrolysis of phosphoinositides, which subsequently elicit the activation of protein kinase C and Ca z+ mobilization. In contrast, EGF and insulin neither induce phosphoinositide hydrolysis nor protein kinase C activation in this cell line (37, 40, 41). Melittin is reported to induce rapid stimulation ofNa ÷ influx and Na-K pump activity (42). In spite of such differences in the proposed early events, either one could enhance ribonucleotide syntheses within 1 hr in combination with insulin. This raises a possibility that the mitogens may stimulate an unknown common signaling

No EGF+Ins.

No EGF+Ins.

No EGF+Ins.

Guanine nucleotides

Uracil nucleotides

Ribose 5-P

dpm/mol

0.53 + 0.15 0.37 + 0.10

5.78 + 0.33 5.70 + 0.39

2.53 + 0.19 2.43 + 0.17

752 + 140" 620 + 114"

203 + 36 383 + 28

109 + 20 228 + ~7

208 + 31 388 + 53

nmol/mg protein 15.6 + 0.22 15.4 + 0.22

Specific radioactivity n= 4

Cellular content n=4

1.56 3.52

0.37 0.89

4.31 9.64

nmol/30 min/mg protein

Ribosome 5-P incorporated (flux)

The cells were incubated with E G F plus insulin and then with the tracer as in Figure 2. Cellular extracts were treated with snake venom to convert all nucleotides to nucleosides, which were analyzed by HPLC. The flux into each group of nucleotides (F) was calculated by: F = (cellular content of the group of nucleotides) x (specific radioactivity of the nucleotides)/(specific radioactivity of free ribose 5-phosphate). Ins.: insulin; ribose 5-P: ribose 5-phosphate. From Ref. (20). *In experiments to determine the specific radioactivity of ribose 5-phosphate a 3.2-fold higher concentration of [14C]inosine was used. The results were corrected for this. n=3.

No EGF+Ins.

Adenine nucleotides

Compound

Mitogenic treatment of cells

T A B L E 2. I N C R E A S E IN F L U X O F R I B O S E 5 - P H O S P H A T E INTO T O T A L A D E N I N E , G U A N I N E , A N D U R A C I L N U C L E O T I D E S BY E G F PLUS INSULIN

2:

Ct~

2: O

t-

-9 ©

158

M. TATIBANA, et al.

TABLE 3. STIMULATION BY VARIOUS MITOGENS OF 14C INCORPORATION FROM [RIBOSYL-14C]INOSINEINTO NUCLEOTIDES IN SWISS 3T3 CELLS AND SERUM ACTIVATION OF OTHER CELL LINES Cell type

Addition

Swiss 3T3

EGF (10 ng/ml) + Insulin (100 ng/ml) Bombesin(3 riM) + Insulin (100 ng/ml) Melittin (300 ng/ml) + Insulin (100 ng/ml) FGF (3 ng/ml) 10% serum 10% serum

Swiss 3T6 Human fetal fibroblast

Exposure time

14C-Ribose incorporation

(rain) 45

(% control) 201 + 24 (n=26)

45

190 + 21 (n=15)

45

157 + 14 (n=4)

30 30 60

174 + 24 (n=6) 150 + 24 (n=3) 198 + 88 (n=6)

Resting Swiss3T3, Swiss3T6, and human fetal fibroblasts were incubated with the indicated additions. The subsequent procedures were as described in Figure 1. The radioactivity in nucleotide fractions was counted. Results are expressed as percent of the control value and represent the means _+ S.D. of the indicated number of experiments. From Ref. (20). pathway that leads to e n h a n c e m e n t of nucleotide syntheses and also of D N A synthesis, in parallel with or after the known events. Quiescent Swiss 3T6 cells and human fetal fibroblasts also showed similar responses of nucleotide synthesis (Table 3).

Effects of EGF plus Insulin on Cellular PRPP Content and PRPP Synthetase Activity E G F plus insulin induced a slight decrease in the cellular contents of PRPP. The contents of P R P P (pmol/106 cells; n = 4 ) were 93 + 20 and 84 + 21 in the unstimulated and stimulated Swiss 3T3 cells, respectively. Activities of P R P P synthetase in extracts of the unstimulated and stimulated cells, assayed under optimal conditions, showed no detectable difference (4.6 + 0.2 and 4.6 + 0.3 nmol P R P P formation/min/mg of protein, respectively, n = 2 ) . The latter result is consistent with previous observations on mouse fibroblasts (17, 18). P R P P is formed solely by P R P P synthetase, and kinetic properties of the enzyme have been studied in vitro (5-8, Ref. 10 for a review). H o w e v e r , mechanisms for the regulation of in situ P R P P synthesis are poorly understood. In the present case, we could not find significant changes in the cellular contents of known regulators including free ribose 5-phosphate (Table 2), A T P , A D P , and other major nucleotides (detailed data not shown), or in P R P P synthetase activity in extracts of the stimulated cells. It is likely that other important factors are involved in the regulation of P R P P synthesis in situ. We should also consider the regulation of purine and pyrimidine

159

MITOGENIC STIMULATIONOF PRPP SYNTHESIS

nucleotide synthesis. IntraceUular levels of P R P P and free nucleotides are critical control factors for the regulation. However, in Swiss 3T3 cells treated with mitogens, nucleotide syntheses were apparently stimulated whereas the levels of P R P P and major nucleotides remained unchanged. One may assume involvement of another important factor in this stimulation. The subsequent studies are relevant to the question.

Effects of Divalent Ionophore A23187, EGTA, and Furosemide on EGF and Insulin-induced Stimulation of Metabolic Flux through PRPP In order to characterize the intracellular signaling pathways leading to this response, we examined effects of several inhibitors and agents on the t4C incorporation into nucleotides (43). The divalent cation ionophore A23187 increased the activity to an extent similar to that seen with E G F and insulin (Table 4). The ionophore could not further increase the incorporation stimulated by E G F and insulin. The results suggest that mobilization of a divalent cation(s) is involved in this E G F and insulin-induced response. Since EGF-induced Ca 2÷ mobilization from the external medium has been observed in Swiss 3T3 cells (44, 45), we examined whether chelation of Ca 2+ of external medium has any effect on the response to E G F and insulin. The addition of [ethylene bis(oxyethylene-nitrilo)]tetraacetate ( E G T A ) at 3 mM, a sufficient concentration to chelate Ca 2+ of the external medium, did not affect the response, thus excluding participation of exogenous Ca 2+ . We then examined the effects of furosemide on the response and observed a partial inhibition (Table 4). The drug is reported to inhibit Mg 2÷ uptake in Mg2+-depleted Yoshida ascites tumor cells (46). TABLE 4. EFFECTS OF THE DIVALENT CATION IONOPHORE A23187, EGTA, AND FUROSEMIDE ON EGF AND INSULIN-INDUCED STIMULATIONOF lac INCORPORATION FROM [RIBOSYL-IaC]INOSINE INTO NUCLEOTIDES Addition

14CIncorporation into nucleotides unstimulated stimulatedwith EGF and insulin % of control

None A23187 (100nM) EGTA(3mM) Furosemide (lmM)

100 185 ± 29t 109 ± 5 91 ± 6

188 ± 176 ± 183 ± 131 ±

21" 32t 25t 18

The cells were incubated for 45 min (or 60 min in experiments using A23187) with the indicated agent in the presence and absence of EGF (10 ng/ml) and insulin (100 ng/ml), then incubated with the tracer for 30 min, and further processed as described in Figure 2. Medium in the second incubation with the tracer was free of these agents. Values are means + S.D. of 3 to 5 independent experiments. From Ref. (43). *p<0.01 compared with control. tp<0.05 compared with control.

160

M. T A T I B A N A , et al.

Requirement of External Mg2÷lon forthe Mitogen-induced Response The above results suggested involvement of another major cation, Mg 2÷, in the response. Use of EDTA might be considered, instead of EGTA, to chelate Mg 2+ in medium. However, the agent detached the cells from the dish during incubation. Therefore, we examined the effects of omission of Ca 2÷ or Mg 2÷ from the external medium (Table 5). Consistent with the above experiments using EGTA, omission of Ca 2+ had no significant effects on the response to EGF and insulin. In contrast, omission of Mg 2+ abolished the stimulation. Notable is the finding that omission of Mg 2÷ did not significantly affect the basal level of 14C incorporation in the unstimulated cells, suggesting that depletion of intracellular Mg 2÷ may be limited in extent under the conditions. MgSO4 could substitute for MgCI 2 in maintaining the cellular capacity for producing the response (data not shown). The lack of the cellular response in medium devoid of added Mg 2÷ was readily recovered in fresh Mg2+-containing medium. This indicated that the Mg 2÷ deprivation did not extensively alter the cellular activities. We also examined the effects of omission of Mg 2+ on specific radioactivity of nucleotides isolated by HPLC. Upon omission of Mg 2÷ from medium, the specific radioactivity of ADP, ATP, GTP, and UTP decreased by 0.5- to 0.6-fold (Table 6). Under these conditions, the decrease in total radioactivity of the acid-soluble fractions was small (0.9 __+0.2-fold), indicating that the decrease in the specific radioactivity of nucleotides was not ascribable to decreased availability of the tracer, but to decreased synthesis of the nucleotides. TABLE 5. EFFECTS OF OMISSION OF E X T E R N A L Mg2+ O R Ca 2÷ ON EGF AND INSULIN-INDUCED STIMULATION OF J4C I N C O R P O R A T I O N FROM [RIBOSYL-14C]INOSINE INTO NUCLEOTIDES

Medium

14C Incorporation into nucleotides unstimulated stimulated with EGF and insulin % of control

(A) MgCI2 added ( n = l l ) 0.8 mM 0 (B) CaCI2 added (n=3) 1.8 mM 0

100 93 ___ 13

188 + 23* 89 + 16

100 95--+6

180 + 19t 1 6 3 + 15t

The cells were incubated with or without EGF and insulin for 45 min in the medium. In experiments (A), MgCI2 was none or added at 0.8 mM with CaCI 2 fixed at 1.8 mM, and in experiments (B), CaCI 2 was none or added at 1.8 mM with MgCI2 fixed at 0.8 mM, Then, the cells were incubated with the tracer in normal DMEM containing 0.8 mM Mg2+, and further processed as in Figure 2. Values are means + S.D. of the indicated number of independent experiments. From Ref. (43). *p<0.001 compared with control. tp
MITOGENIC STIMULATION OF PRPP SYNTHESIS

161

TABLE 6. SPECIFIC RADIOACTIVITY OF NUCLEOTIDES OBTAINED FROM THE CELLS TREATED WITH EGF AND INSULIN IN THE PRESENCE OR ABSENCE OF Mg2+ IN MEDIUM

Nucleotide ADP ATP GTP UTP

Specificradioactivity MgCI2 added (raM) 0.8 0 430 + 37 364 + 43 225 + 31 364 _+51

233 + 194 + 129 + 205 +

28 32 19 31

The cells were treated with EGF and insulin for 45 min in Mga+-containingor Mg2+-freemedium and then incubated with the tracer as described in Figure 2. Nucleotides in the acid-soluble extracts were separated by HPLC. Values are means + S.D. of 3 independent experiments. The above observations show that the increase in the metabolic flux through PRPP, in Swiss 3T3 cells induced by E G F and insulin, required external Mg 2+. One may next ask what steps in signal transduction require Mg a÷. It is not likely that external Mg 2÷ is required for the extracellular binding of E G F and insulin to their receptors, since another rapid response to E G F and insulin in this cell line, stimulation of glycolysis, occurred irrespective of whether Mg 2+ was added to medium or not (47, our unpublished results). The simplest hypothesis is that entry of the external Mg a+ into the cells is stimulated in an early response to E G F and insulin, via activation of an Mg a+ transport system or an Mg 2+ channel(s), and then the cytoplasmic free Mg z÷ concentration is increased. In murine lymphoma cells (48) and rabbit thymocytes (49), the tumor promotor tetradecanoyl phorbol acetate induces a significant magnesium uptake within 1-1.5 hr. The synthesis of PRPP and nucleotides could be stimulated directly by an increased intracellular free Mg e+ concentration. Some key enzymes in nucleotide synthesis including PRPP synthetase (10), carbamoyl phosphate synthetase II (EC 6.3.5.5) (50), and amidophosphoribosyltransferase (EC 2.4.2.14) (2), require Mg 2+ for their activities, not only for formation of a "true substrate" such as MgATP 2-, but as a protein ligand. Association constants for Mg binding to the latter sites are relatively low so that a sufficiently high concentration of free Mg 2+ is required for full activation of such enzymes. Intracellular free Mg e÷ concentration [0.3-3.0 mM (51)] is in a regulatory range for those enzymes (2, 50). We also should consider other Mg2+-requiring enzymes. For example, activities in vitro of both EGF- and insulin-stimulated tyrosine-specific protein kinases, which may play important roles in signal transduction, depend on concentrations of Mn 2+ or Mg 2+ (52-54).

162

M. TATIBANA, et al.

While potential importance of magnesium in the regulation of cellular metabolism and growth is conceivable from the many effects of this metal on in vitro systems, experimental evidence to support the notion is rather limited. Only a few investigators suggested that magnesium serves as a regulator in the cell cycle (55), protein synthesis (56), and glycolysis (57). Grubbs et al. (58) reported that in resting lymphocytes, newly transported Mg 2+ from external medium exchanges with only about 3% of the total cytoplasmic Mg 2÷ within 2 hr, whereas in proliferating cells it exchanges extensively with the cytoplasmic Mg 2+. The observations suggest that compartmentation of Mg 2+ as well as intracellular, physicochemical microenvironments vary greatly with proliferative states of the cells. It is interesting to know how activities of free Mg2+-requiring enzymes are regulated in situ under such conditions. Effects of Some lnhibitors on Stimulation of Nucleotide Synthesis

We previously showed that colchicine inhibits enhancement of PRPP levels in the liver of mouse fed on protein-rich diet and also in the isolated rat hepatocytes stimulated by glucagon (19). Effects of colchicine were thus examined here, along with cycloheximide, an inhibitor of protein synthesis, and p-bromophenacyl bromide, a stimulator of phopholipase A 2 (EC 3.1.1.4). Colchicine (1 /zM) inhibited by 63% the EOF plus insulin-induced increase in the 14C incorporation into nucleotides. However, the total radioactivity in the acid-soluble fraction was also decreased by 37%, suggesting that transport of the tracer into the cells was inhibited. It is possible that the decrease in the radioactivity of nucleotides was only apparent partly at least. A similar inhibition was seen with cycloheximide (60/zg/ml), which almost nullified the stimulation and also decreased by 33% the total radioactivity of the acid-soluble fraction. The latter agent may have some direct effects on the signaling pathway for the stimulation of nucleotide synthesis. Further characterization of the effects of the inhibitors remains to be done. Buchanan el al. (17) previously observed a complete suppression of serum activation of PRPP synthesis and of the early steps of purine synthesis in Swiss 3T6 cells. p-Bromophenacyl bromide (10 txM) had no significant effect on the stimulation by EGF and insulin.

SUMMARY 5-Phosphoribosyl 1-pyrophosphate (PRPP) is a common precursor for the synthesis of all nucleotides and also serves as a critical regulator for the synthesis. In spite of a number of studies in vitro on mammalian

MITOGENIC STIMULATIONOF PRPP SYNTHESIS

163

PRPP synthetase, our understanding of the regulation of PRPP synthesis in situ is very limited. Various mitogens are known to activate purine and pyrimidine de novo biosynthesis and purine base phosphoribosylation as an early response in quiescent mouse fibroblasts. We aimed at elucidation of the underlying mechanism for the possible increase in PRPP synthesis in mitogen-stimulated mouse fibroblasts in culture. In order to quantitatively follow metabolic flux through PRPP into nucleotides, [ribosyl-14C]inosine was enzymatically prepared and used as a tracer to preferentially label intracellular ribose phosphate. The radioactivity incorporation into cellular nucleotides was measured. Evidence supported the validity of the method. Prior exposure of quiescent Swiss 3T3 cells in culture to epidermal growth factor (EGF) plus insulin for 45--60 min enhanced approximately 2-fold the radioactivity incorporation from [ribosyl-lnC]inosine into nucleotides, without increasing the specific radioactivity of intracellular free ribose 5-phosphate. [14C]Uracil incorporation into nucleotides, a measure for PRPP-independent ribose phosphate utilization for nucleotide synthesis, was not increased. These and other results indicate that EGF plus insulin stimulates the metabolic flux through PRPP. A similar stimulation was induced by bombesin and melittin in combination with insulin and by fibroblast growth factor alone. Quiescent Swiss 3T6 ceils and human fetal fibroblasts showed a similar stimulation of nucleotide synthesis in response to exposure to serum. For characterization of intracellular signaling pathways, we examined effects of several inhibitors and agents on the stimulation. The divalent cation ionophore A23187 mimicked the response to EGF and insulin in Swiss 3T3 cells, thereby suggesting involvement of divalent cation mobilization in this increase. The effect of the ionophore was not additive to that of the growth factors. Omission of Ca 2+ from the incubation medium did not affect the response to EGF and insulin, whereas the omission of Mg 2÷ did abolish the response. Furosemide, an inhibitor of Mg 2÷ influx, partially inhibited the stimulated synthesis of nucleotides. Thus, the entry of external Mg 2+ into the cells may play a critical role in this signal transduction. These results provided an important access to elucidation of the intracellular mechanisms for the mitogen-induced increase in PRPP and nucleotide syntheses.

ACKNOWLEDGEMENTS The present investigations were supported in part by grants from the Ministries of Education, Science and Culture (61440030) and of Health and Welfare, Japan. We are grateful to Dr. Y. Takai, Kobe University School of Medicine, for providing a stock culture of Swiss 3T3 cells.

164

M. TATIBANA, et al. REFERENCES

1. I . H . FOX and W. N. KELLEY, Phosphoribosylpyrophosphate in man: biochemical and clinical significance, Ann. Intern. Med. 74, 424--433 (1971). 2. E. W. HOLMES, Regulation of purine biosynthesis de novo, Vol. 51, pp. 21-41 in Handbook of Experimental Pharmacology (W. N. KELLEY and I. M. WEINER, eds.), Springer-Verlag, Heidelberg (1978). 3. M. T A T I B A N A and K. SHIGESADA, Control of pyrimidine biosynthesis in mammalian tissues: V. Regulation of glutamine-dependent carbamyl phosphate synthetase: activation by 5-phosphoribosyl 1-pyrophosphate and inhibition by uridine triphosphate, J. Biochem. 72, 549-560 (1972). 4. M. TATIBANA, Interrelationship of purine and pyrimidine metabolism, Vol. 51, pp. 125-154 in Handbook of Experimental Pharmacology (W. N. KELLEY and I. M. WEINER, eds.), Springer-Verlag, Heidelberg (1978). 5. R.L. SWITZER and K. J. GIBSON, Phosphoribosylpyrophosphate synthetase (ribose5-phosphate pyrophosphokinase) from Salmonella typhimurium, Methods Enzymol. 51, 3-11 (1978). 6. I. H. FOX and W. N. KELLEY, Human phosphoribosylpyrophosphate synthetase: distribution, purification, and properties, J. Biol. Chem. 246, 5739-5748 (1971). 7. D. G. ROTH, E. SHELTON and T. F. DEUEL, Purification and properties of phosphoribosyl pyrophosphate synthetase from rat liver, J. Biol. Chem. 249, 291-296 (1974). 8. B.H.-JENSEN, K. W. HARLOW, C. J. KING and R. L. SWITZER, Phosphoribosylpyrophospbate synthetase of Escherichia coli : properties of the purified enzyme and primary structure of theprs gene, J. Biol. Chem. 261, 6765-6771 (1986). 9. H. M. B A L O - B A N G A and G. WEBER, Increased 5-phospho-a-D-ribose-l-diphosphate synthetase (ribosephosphate pyrophosphokinase, EC 2.7.6.1) activity in rat hepatomas, Cancer Res. 44, 5004-5009 (1984). 10. M. A. BECKER, Abnormalities of PRPP metabolism leading to an overproduction of uric acid, Vol. 51, pp. 155-183 in Handbook of Experimental Pharmacology (W. N. KELLEY and I. M. WEINER, eds.), Springer-Verlag, Heidelberg (1978). 11. L.J. MEYER and M. A. BECKER, Human erythrocyte phosphoribosylpyrophosphate synthetase, J. Biol. Chem. 252, 3919-3925 (1977). 12. R. B. PILZ, R. C. WILLIS and G. R. BOSS, The influence of ribose 5-phosphate availability on purine synthesis of cultured human lymphoblasts and mitogen-stimulated lymphocytes, J. Biol. Chem. 259, 2927-2935 (1984). 13. T, CHIKENJI, T. A S A I and M. TATIBANA, Protein-diet-induced elevation of 5-phosphoribosyl 1-diphosphate concentrations in mouse liver associated with increased syntheses of various nucleotides and the possible involvement of glucagon, Biochim. Biophys. Acta 802, 274-281 (1984). 14. K. KITA, T. ISHIZUKA, S. ISHIJIMA and M. TATIBANA, Rat liver phosphoribosylpyrophosphate synthetase: existence as heterogeneous aggregates and identification of catalytic subunit, Abstracts for the 6th International Symposium on Human Purine and Pyrimidine Metabolism in Pediatr. Res. 24, 123 (1988). 15. M. TAIRA, S. ISHIJIMA, K. KITA, K. Y A M A D A , T. IIZASA and M. TATIBANA, Nucleotide and deduced amino acid sequences of two distinct cDNAs for rat phosphoribosylpyrophosphate synthetase, J. Biol. Chem. 262, 14867-14870 (1987). 16. M. TAIRA, T. IIZASA, K. YAMADA, H. SHIMADA and M. TATIBANA, Tissue-differential expression of two distinct genes for phosphoribosyl pyrophosphate synthetase and existence of the testis-specific transcript, Biochim. Biophys. Acta, in press. 17. J . M . BUCHANAN, M. L. SMITH and R. J. SMITH, Regulation of nucleotide and pentose synthesis in resting and stimulated 3T6 fibroblasts, Advan. Enzyme Regul. 20, 135-152 (1981). 18. M. A. BECKER, P. DICKER and E. ROZENGURT, Mitogenic enhancement of purine base phosphoribosylation in Swiss mouse 3T3 cells, Am. J. Physiol. 244 (Cell Physiol. 13), C288-C296 (1983).

MITOGENIC STIMULATION OF PRPP SYNTHESIS

165

19. T. HISATA, N. KATSUFUJI and M. TATIBANA, Glucagon or cyclic AMP-stimulated synthesis of 5-phosphoribosyl 1-pyrophosphate in isolated hepatocytes and inhibition by antimicrotubular drugs, Biochem. Biophys. Res. Commun. 81,704-709 (1978). 20. S. ISHIJIMA, K. KITA, N. KINOSHITA, T. ISHIZUKA, N. SUZUKI and M. TATIBANA, Evidence for early mitogenic stimulation of metabolic flux through phosphoribosyl pyrophosphate into nucleotides in Swiss 3T3 cells, J. Biochem. 104, 730-734 (1988). 21. E. R O Z E N G U R T and L. A. HEPPEL, Serum rapidly stimulates ouabain-sensitive 86Rb+ influx in quiescent 3T3 cells, Proc. Natl. Acad. Sci. U.S.A. 72, 4492-4495 (1975). 22. S.D. SKAPER, R. C. WILLIS and J. E. SEEGMILLER, Intracellular 5-phosphoribosyl1-pyrophosphate: decreased availability during glutamine limitation, Science 193,587-588 (1976). 23. P. D. REISS, P. F. ZUURENDONK and R. L. VEECH, Measurement of tissue purine, pyrimidine, and other nucleotides by radial compression high-performance liquid chromatography, Anal. Biochem. 140, 162-171 (1984). 24. R. A. HARTWlCK, S. P. ASSENZA and P. R. BROWN, Identification and quantitation of nucleosides, bases and other UV-absorbing compounds in serum, using reversed-phase high-performance liquid chromatography, J. Chromatogr. 186, 647-658 (1979). 25. T. HISATA, An accurate method for estimating 5-phosphoribosyl 1-pyrophosphate in animal tissues with the use of acid extraction, Anal. Biochem. 68,448-457 (1975). 26. M. A. BECKER, Patterns of phosphoribosylpyrophosphate and ribose-5-phosphate concentration and generation in fibroblasts from patients with gout and purine overproduction, J. Clin. Invest. 57,308-318 (1976). 27. P. DICKER and E. ROZENGURT, Phorbol esters and vasopressin stimulate DNA synthesis by a common mechanism, Nature 287, 607-612 (1980). 28. M.M. BRADFORD, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72, 248-254 (1976). 29. A. KORNBERG, I. LIEBERMAN and E. S. SIMMS, Enzymatic synthesis and properties of 5-phosphoribosylpyrophosphate, J. Biol. Chem. 215,389--402 (1955). 30. J. F. HENDERSON and M. K. Y. KHOO, Availability of 5-phosphoribosyl 1-pyrophosphate for ribonucleotide synthesis in Ehrlich ascites tumor cells in vitro, J. Biol. Chem. 240, 2358--2362 (1965). 31. M. HORI and J. F. HENDERSON, Kinetic studies of adenine phosphoribosyltransferase, J. Biol. Chem. 241, 3404-3408 (1966). 32. T. FRIEDMANN, J. E. SEEGMILLER and J. H. SUBAK-SHARPE, Evidence against the existence of guanosine and inosine kinases in human fibroblasts in tissue culture, Exptl Cell Res. 56, 425--429 (1969). 33. J . F . HENDERSON and A. R. P. PATERSON, Purine ribonucleotide synthesis from purine bases and ribonucleosides, pp. 124-135 in Nucleotide Metabolism, Academic Press, New York (1973). 34. S. T A L H A and L. HAREL, Early effect of growth factors (EGF + insulin) upon ATP turnover in 3T3 cells, Exptl Cell Res. 149, 471-481 (1983). 35. E. NISSINEN, Analysis of purine and pyrimidine bases, ribonucleosides, and ribonucleotides by high-pressure liquid chromatography, Anal. Biochem. 106, 497-505 (1980). 36. E. ROZENGURT, Early signals in the mitogenic response, Science 234, 161-166 (1986). 37. J. P. HESLOP, D. M. BLAKELEY, K. D. BROWN, R. F. IRVINE and M. J. BERRIDGE, Effects of bombesin and insulin on inositol (1,4,5)triphosphate and inositol (1,3,4)triphosphate formation in Swiss 3T3 cells, Cell 47,703-709 (1986). 38. I. ZACHARY, J. W. SINNETT-SMITH and E, ROZENGURT, Early events elicited by bombesin and structurally related peptides in quiescent Swiss 3T3 cells: I. Activation of protein kinase C and inhibition of epidermal growth factor binding, J. Cell Biol. 102, 2211-2222 (1986). 39. K. KAIBUCHI, T. TSUDA, A. KIKUCHI, T. TANIMOTO, T. YAMASHITA

166

40. 41. 42. 43.

44. 45. 46. 47. 48. 49.

50.

51, 52. 53. 54. 55. 56. 57. 58.

M. TATIBANA, et al. and Y. TAKAI, Possible involvement of protein kinase C and calcium ion in growth factor-induced expression of c-myc oncogene in Swiss 3T3 fibroblasts, J. Biol. Chem. 261, 1187-1192 (1986). F. V A R A and E. ROZENGURT, Stimulation of Na+/H+ antiport activity by epidermal growth factor and insulin occurs without activation of protein kinase C, Biochem. Biophys. Res. Commun. 130, 646-653 (1985). T. TSUDA, K. KAIBUCHI, Y. KAWAHARA, H. FUKUZAKI and Y. TAKAI, Induction of protein kinase C activation and Ca 2+ mobilization by fibroblast growth factor in Swiss 3T3 cells, FEBS Lett. 191,205-210 (1985). E. ROZENGURT, T. D. GELEHRTER, A. LEGG and P. PETTICAN, Melittin stimulates Na entry, Na-K pump activity and DNA synthesis in quiescent cultures of mouse cells, Cell 23, 781-788 (1981). S. ISHIJIMA, K. KITA, T. ISHIZUKA, N. SUZUKI and M. TATIBANA, External Mg 2+ is required for signal transduction in early mitogenic stimulation of nucleotide synthesis in Swiss 3T3 cell, Abstracts for 14th International Congress o f Biochemistry (Prague) p. 73 (1988). T . R . HESKETH, J. P. MOORE, J. D. H. MORRIS, M. V. TAYLOR, J. ROGERS, G. A. SMITH and J. C. METCALFE, A common sequence of calcium and pH signals in the mitogenic stimulation of eukaryotic cells, Nature 313, 481-484 (1985). A. PANDIELLA, A. MALGAROLI, J. MELDOLESI and L. M. VICENTINI, EGF raises cytosolic Ca 2+ in A431 and Swiss 3T3 cells by a dual mechanism, Exptl Cell Res. 170, 175-185 (1987). T. GONTHER, J. VORMANN and R. AVERDUNK, Characterization of furosemidesensitive Mg e+ influx in Yoshida ascites tumor cells, FEBS Lett. 197, 297-300 (1986). I. DIAMOND, A. LEGG, J. A. SCHNEIDER and E. ROZENGURT, Glycolysis in quiescent cultures of 3T3 cells, J. Biol. Chem. 253, 866--871 (1978). R. D. GRUBBS and M. E. MAGUIRE, Regulation of magnesium but not calcium transport by phorbol ester, J. Biol. Chem. 261, 12550-12554 (1986). P. CSERMELY, P. FODOR and J. SOMOGYI, The tumor promoter tetradecanoylphorbol-13-acetate elicits the redistribution of heavy metals in subceUular fractions of rabbit thymocytes as measured by plasma emission spectroscopy, Carcinogenesis 8, 1663-1666 (1987). M. TATIBANA and K. SHIGESADA, Control of pyrimidine biosynthesis in mammalian tissues: IV. Requirements of a quantitative assay of glutamine-dependent carbamyl phosphate synthetase and effect of magnesium ion as an essential activator, J. Biochem. 72, 537-547 (1972). T. GIANTHER, Functional compartmentation of intracellular magnesium, Magnesium 5, 53-59 (1986). G. CARPENTER, L. KING, JR. and S. COHEN, Rapid enhancement of protein phosphorylation in A-431 cell membrane preparations by epidermal growth factor, J. Biol. Chem. 254, 4884-4891 (1979). M . F . WHITE, H.-U. HAR1NG, M. KASUGA and C. R. KAHN, Kinetic properties and sites of autophosphorylation of the partially purified insulin receptor from hepatoma cells, J. Biol. Chem. 259, 255-264 (1984). L.J. PIKE, E. A. KUENZEL, J. E. CASNELL1E and E. G. KREBS, A comparison of the insulin- and epidermal growth factor-stimulated protein kinases from human placenta, J. Biol. Chem. 259, 9913--9921 (1984). G. M. WALKER, Magnesium and cell cycle control: an update, Magnesium 5, 9-23 (1986). M. T E R A S A K I and H. RUBIN, Evidence that intracellular magnesium is present in cells at a regulatory concentration for protein synthesis, Proc. Natl. Acad. Sci. U.S.A. 82, 7324--7326 (1985). L. GARFINKEL and D. GARFINKEL, Magnesium regulation of the glycolytic pathway and the enzymes involved, Magnesium 4, 60-72 (1985). R.D. GRUBBS, S. D. COLLINS and M. E. MAGUIRE, Differential compartmentation of magnesium and calcium in murine $49 lymphoma cells, J. Biol. Chem. 259, 12184-12192 (1984).