Mg²⁺ as activator of uridine phosphorylation in coordination with other cellular responses to growth factors

Abstract

The divalent cation ionophore A23187 facilitates the manipulation of intracellular Mg²⁺ without increasing the general permeability of the cell. The uptake of uridine into cells is limited by its rate of intracellular phosphorylation that increases within minutes after the addition of growth factors. In the experiments described here, the rate of uridine uptake in ionophore-treated cells stimulated by either serum or insulin depended on the extracellular and intracellular concentrations of Mg²⁺ and was independent of the extracellular Ca²⁺ concentration. In very high concentrations of Mg²⁺ (50 mM), ionophore-treated cells take up uridine as fast, in the absence of growth factors as in their presence, demonstrating that Mg²⁺ can replace the growth factor requirement for the stimulation of uridine uptake. In contrast, thymidine uptake, which also is limited by its rate of intracellular phosphorylation, showed no early response to either growth factors or Mg²⁺ concentration, which is consistent with the 10-fold lower Mg²⁺ requirement of thymidine kinase compared with uridine kinase. The feedback inhibition of uridine kinase by UTP and CTP in cell-free extracts was alleviated by increased Mg²⁺ concentration. The results support the thesis that the increased uptake of uridine in cells treated with growth factors is determined by a membrane-induced increase in intracellular free Mg²⁺. Such increase would also accelerate the rate of translation-initiation and other coordinate responses that, unlike increased uridine uptake, are essential for cell proliferation. The rate of uridine uptake is suggested as a direct indicator of free cytosolic Mg²⁺ that drives the shift from quiescence to proliferation.

Treatment of quiescent cell cultures with growth factors elicits a coordinate response that includes increased rates of uptake of various substrates, acceleration of intermediary metabolism, and increased synthesis of protein and RNA that precede the onset of DNA synthesis (1). All of the early responses to growth factors that have been tested depend on the concentration of Mg²⁺ in the cells as produced by varying the concentration of Mg²⁺ in the medium (1-4). The most critical of the early responses for accelerating progress through G1 to S is the increased rate of protein synthesis (5, 6). The rate of protein synthesis in living cells and in cell-free systems is increased by raising the concentration of Mg²⁺ to a certain level but is inhibited at still higher concentrations (4, 7). Cells self-adjust within a few hours from an externally imposed, abnormally high concentration of intracellular Mg²⁺, which inhibits protein synthesis, to a somewhat lower, stimulatory concentration (4). This autoregulation of intracellular Mg²⁺, with its accompanying rise in protein synthesis, is followed by a correspondingly delayed onset of DNA synthesis and attests to the physiological nature of proliferation control by Mg²⁺. The complexity of the regulation of protein synthesis in cells makes it uncertain how much of the Mg²⁺-dependent effect is exerted by directly influencing the binding of ribosomal subunits to mRNA, or by activating one or more protein kinases in mitogenic pathways that regulate the initiation of translation, or both (8, 9).

A much simpler component of the coordinate response is the increased uptake of uridine (10, 11). The rate-limiting step in the uptake of uridine is its trapping through phosphorylation by uridine kinase, using MgATP^2- as a cosubstrate (12-14). Thus, uridine uptake by intact cells is a direct measure of the activity of the uridine kinase enzyme. The activity of uridine kinase can also be assayed in cell-free extracts, where the concentrations of MgATP^2- and uridine can be controlled. The trapping of uridine, which increases within a few minutes of the application of growth factors to quiescent cells, is tightly linked to the extracellular and intracellular Mg²⁺ concentrations (15-17). The increased uptake of uridine is gratuitous for the coordinate response and progression through G1 into S, as indicated by the absence of uridine in any of the media commonly used for cell culture. However, understanding the regulation of this simple reaction could shed light on what is controlling the more complex and more essential elements of the coordinate response such as protein synthesis.

Support for a role of Mg²⁺ as the primary regulator of uridine trapping comes from the observation that its V_max, but not its K_m, is similarly altered by applying growth factors, and by altering the intracellular Mg²⁺ concentration (15). Further support for a regulatory role of Mg²⁺ in uridine uptake comes from its comparison with thymidine uptake, which is also limited by phosphorylation (18) but is not an early response to growth factors (17, 19). The K_m of uridine kinase for Mg²⁺ in cell-free extracts is 10 times higher than that of thymidine kinase. It suggests that thymidine kinase is saturated for MgATP^2-, even in quiescent cells, whereas uridine kinase is not saturated in such cells and could be activated by any increase in Mg²⁺ induced by growth factors. That Mg²⁺, rather than ATP^4-, is the regulator of uridine trapping is consistent with the failure of ATP to increase in growth-stimulated cells (17, 20-22), whereas total Mg²⁺ (23, 24) and free Mg²⁺ (25, 26) do increase in a sustained manner after treatment with growth factors.

There are, however, some problems in a simple interpretation of the published effects on uridine uptake/trapping caused by varying the Mg²⁺ concentration in cells. To achieve a wide range of Mg²⁺ concentrations in cells, the Ca²⁺ concentration of the medium was reduced to increase the permeability of the cells to Mg²⁺ (16, 17, 27). Severe reduction in the Ca²⁺ supply by itself decreases the trapping of uridine. Ca²⁺ restriction also increases the general permeability of cells, resulting in a large increase in cellular Na⁺ and a large decrease in cellular K⁺ (16). Although the inhibitory effects of Ca²⁺ deprivation on uridine trapping are overcome by raising the Mg²⁺ concentration of the cell, it was desirable to remove the complications of Ca²⁺ deprivation by using a method for allowing freer exchange of Mg²⁺ between medium and cell without increasing general permeability of the cell or disturbing its Ca²⁺ content. We therefore studied the effects of the Mg²⁺ and Ca²⁺ ionophore A23187 (28) on general permeability of cells, variation in their Mg²⁺ concentrations, and its control of uridine trapping. We also examined the feedback inhibition of uridine kinase by UTP and CTP in cell-free extracts and the modulating effect of Mg²⁺ on that inhibition. The results confirm the regulatory role of Mg²⁺ in uridine trapping and, thereby, support a primary role for Mg²⁺ as the coordinating regulator of the other early direct responses of cells to stimulation by growth factors.

Materials and Methods

BALB/c3T3 cells were maintained in MEM with 10% calf serum. Cells were grown to confluence on 60-mm polystyrene tissue culture dishes and then switched to MEM with 1% serum overnight before use in experiments. MEM was prepared for experimental manipulations without added Ca²⁺ or Mg²⁺. Serum was dialyzed against physiological saline that was free of Ca²⁺ and Mg²⁺. The medium with 10% dialyzed serum contained 0.015-0.020 mM of contaminating Ca²⁺ and Mg²⁺ as determined by atomic absorption spectrophotometry. Insulin and the divalent cation ionophore A23187 were purchased from Calbiochem.

Labeling of cultures was performed by addition of [³H]uridine (New England Nuclear; 13.3 Ci/mmol; 1 Ci = 37 GBq) or [³H]thymidine (New England Nuclear; 20 Ci/mmol) for 10 min. Label was washed off by washing three times with ice-cold Tris-buffered saline, and acid-soluble material was extracted for 10 min with cold 5% trichloroacetic acid. This material was sampled for scintillation counting. The remaining material on each dish was dissolved in 0.1 M NaOH and used to measure protein content by the method of Lowry (29). All labeling was performed on duplicate cultures.

Cultures were measured for their Mg²⁺ contents in the following way. Dishes were washed four times with ice-cold 150 mM NaCl and then scraped into distilled water. This material was sonicated and diluted for measurement by atomic absorption spectrophotometry. Samples to be read were made 15 mM La³⁺, 4 mM Cs⁺, and 100 mM HCl to minimize chemical and ionization interference.

Cell-free extracts were prepared by sonicating cells in a solution containing 10 mM Na₂HPO₄ and 150 mM KCl at 4°C. Each reaction mixture consisted of 0.4 ml of total volume that included a given amount of cell extract, 10 mM Na₂HPO₄, 150 mM KCl, 2 μCi/ml [³H]uridine, various concentrations of UTP or CTP, and either 2 or 0.3 mM MgCl₂ and 2 mM ATP. Each reaction was carried out at 37°C and began with the addition of cell extract. After 30 min, each assay was boiled for 2 min and spotted onto 2.4-cm discs of Whatman DE81 paper. The discs were dried and washed in two successive 10-ml washes of 1 mM ammonium formate, pH 3.5. The discs were then counted for retained counts that represented the phosphorylated species. The assay was linear for up to 40 min and linear with the amount of extract.

Results

Effect of Ionophore A23187 on Uridine Uptake in the Absence of Added Ca²⁺ or Mg²⁺. A23187 was applied to cells in various concentrations in media with no addition of Ca²⁺, or no addition of Mg²⁺, to observe the effects of the separate removal from cells of either cation on the uptake of uridine (Fig. 1). In control medium containing serum with physiological concentrations of Ca²⁺ and Mg²⁺, the addition of A23187 had a small stimulatory effect on uridine uptake. In medium with only residual traces of Mg²⁺, concentrations of ionophore of >1 μg/ml had increasingly inhibitory effects on uridine uptake; 5 μg/ml ionophore reduced uptake to a rate almost as low as it was in the absence of serum with normal amounts of both cations. Thus, Mg²⁺ restriction in ionophore-treated cells quantitatively reproduced the inhibitory effect of serum removal on uridine uptake. The virtual absence of Ca²⁺ in medium with no ionophore and 1 mM Mg²⁺ reduced uridine uptake almost 2-fold, but increasing ionophore concentrations increased uptake to a normal rate. Because the inhibitory effects of Ca²⁺ deprivation on uridine uptake can be overcome by raising the extracellular Mg²⁺ concentration (16), the data in Fig. 1 suggest that the ionophore also relieves this inhibition by increasing the availability of Mg²⁺. The results show that (i) the presence of serum in medium with normal Mg²⁺ and Ca²⁺ strongly increases uridine uptake; (ii) effective concentrations of ionophore in Mg²⁺-deficient medium inhibit uridine uptake, presumably by allowing a decrease of cellular Mg²⁺; (iii) Ca²⁺ deprivation has no effect on uridine uptake except as it affects the availability of Mg²⁺; and (iv) that 5 μg/ml is an effective concentration of A23187 to facilitate exchange of Mg²⁺ between medium and cell.

Fig. 1.

Effect of ionophore A23187 on uridine uptake by cells deprived of extracellular Ca²⁺ or Mg²⁺. Cells were washed and incubated in the indicated media for 1 h, followed by addition of fresh medium containing [³H]uridine for 10 min to measure uptake into acid-soluble pools. Each time point is an average of uptake by duplicate cultures.

Ionophore Effect on Mg²⁺ Concentration in Cells. The effect of 5.0 μg/ml A23187 on the intracellular concentration of Mg²⁺ was measured in the presence of various concentrations of Mg²⁺ in the medium (Table 1). In the presence of ionophore, lowering the extracellular Mg²⁺ from 1.0 to 0.02 mM caused a 30% decrease in intracellular Mg²⁺. This is a considerably greater loss of Mg²⁺ than the 10% loss that occurs when cells are deprived of extracellular Mg²⁺ in the absence of ionophore (16). Addition of the ionophore to cells in the presence of physiological 1.0 mM Mg²⁺ increases the Mg²⁺ concentration of the cells by 15% over cells in 1.0 Mg²⁺ without ionophore. It is apparent that the ionophore is permitting greater exchange of Mg²⁺ between cells and medium than occurs in the absence of the ionophore.

Table 1. Mg²⁺ content of cells, [Mg]_i, exposed to ionophore A23187 in various concentrations of extracellular Mg²⁺, [Mg²⁺]₀.

A23187, μg/ml	[Mg2+]0, mM	[Mg]i, μmol/mg protein	[Mg]i (relative)*
0	1.0	0.090	0.85
5	0.02	0.077	0.73
5	0.1	0.098	0.92
5	1.0	0.106	1.0
5	15.0	0.140	1.32

Cells were washed and incubated for 1 h in medium with 5 μg/ml A23187 in 10% dialyzed calf serum, 1.7 mM Ca²⁺, and various concentrations of Mg²⁺. Control cultures received no ionophore. The cells were then washed, and their Mg²⁺ content was determined. Value in italics means arbitrary unit value for comparative purposes.

^*Relative to cells in ionophore with 1.0 mM Mg²⁺ in the medium

Permeability of Mg²⁺-Deprived Cells to [³H]l-Glucose in the Presence of Either Low Ca²⁺ or Ionophore A23187. The effects of Ca²⁺ deprivation versus ionophore A23187 on general permeability of cells was measured by the rate of uptake of l-glucose, the nonphysiological isomer of d-glucose, that enters the cells by simple diffusion. There was a 20-fold increase in uptake of [³H]l-glucose in cells deprived of Ca²⁺ and Mg²⁺ as compared with cells in physiological concentrations of both cations or with cells deprived only of Mg²⁺ in the presence of the ionophore (Table 2). The results indicate that the use of A23187 to allow large decreases to be made in cellular Mg²⁺ spares cells from the general disruption of permeability occasioned by the deprivation of Ca²⁺ and Mg²⁺.

Table 2. Effect of Ca²⁺ deprivation or A23187 addition on uptake of [³H]l-glucose in very low Mg²⁺.

Medium constituents
Ca2+, mM	Mg2+, mM	A23187, μg/ml	[3H]l-glucose (cpm/μg protein)
1.7	1.0	0	0.80
0.025	0.008	0	15.3
1.7	0.008	5	0.92

Three cultures were washed and incubated with each experimental medium, all containing 10% dialyzed calf serum, for 1.5 h, at which time a change was made to media of the same composition containing [³H]l-glucose, 2 μCi/ml for 30 min. The cells were then washed, and the trichloracetic-acid-soluble material was taken up for scintillation counting.

Response of Uridine Uptake to Variations in Ca²⁺ and Mg²⁺ in the Presence of Ionophore A23187. The effects of various concentrations of Mg²⁺ or Ca²⁺ on uridine uptake were studied at a single effective concentration of A23187 (5 μg/ml). The rate of uridine uptake increased monotonically with concentrations of Mg²⁺ in the medium between 0.02 and 0.5 mM (Fig. 2). In contrast, there was no change in uridine uptake, with concentrations of Ca²⁺ between 0.02 and 1.0 mM, although there was a slight drop at 5 mM. The clear-cut distinction between the effects of reducing Mg²⁺ and Ca²⁺ concentrations supports an essential role for Mg²⁺, but not Ca²⁺, in regulating the rate of uridine uptake in cells.

Fig. 2.

Response of uridine uptake to variations in extracellular Ca²⁺ or Mg²⁺ in the presence of ionophore A23187. Cells were washed and incubated for 1 h in media containing 5 μg/ml A23187 and the indicated concentration of Ca²⁺ (in 1 mM Mg²⁺) or Mg²⁺ (in 1.7 mM Ca²⁺), followed by addition of fresh medium containing [³H]uridine for 10 min to measure uptake into acid-soluble pools. Each time point is an average of uptake by duplicate cultures.

Effect of Serum on the Uridine Uptake Response to Mg²⁺ Deprivation in Cells Treated with Ionophore A23187. Previous results indicated that a reduction of Mg²⁺ in Ca²⁺-deprived medium with 10% calf serum reduced the uptake of uridine to the same extent as the omission of serum in physiological concentrations of Mg²⁺ and Ca²⁺ (16). We wished to determine the effect of Mg²⁺ deprivation on uridine uptake in media with and without serum in the presence of various concentrations of A23187. The addition of 10% calf serum to medium containing 1.7 mM Ca²⁺, 0.008 mM Mg²⁺, and no ionophore increased the uptake of uridine >4-fold over medium containing no serum (Fig. 3). Concentrations of A23187 >1.0 μg/ml in low extracellular Mg²⁺ reduced uridine uptake in the serum-containing medium to the same extent as shown in Fig. 1. In contrast, there was no effect of even the highest concentration of A23187 on the already low rate of uridine uptake in the cultures deprived of both serum and Mg²⁺. The results support the concept that mitogenic agents such as serum raise uridine uptake through an increase in free Mg²⁺ and that this effect is prevented by depleting cells of Mg²⁺.

Fig. 3.

Effect of ionophore A23187 on uridine uptake in Mg²⁺-deprived cells in the absence or presence of serum. Cells were washed and incubated for 1 h in medium containing 0.008 mM Mg²⁺ and the indicated concentrations of serum and ionophore A23187, followed by addition of fresh medium containing [³H]uridine for 10 min to measure uptake into acid-soluble pools. Each time point is an average of uptake by duplicate cultures.

Effect of Mg²⁺ Concentration on Uridine Uptake in Cells Stimulated by Insulin in the Presence of Ionophore. Serum is a complex mixture of growth factors plus other proteins such as protease inhibitors that help sustain cell proliferation. It is a much more powerful stimulant of the proliferation of BALB/c3T3 cells than purified insulin; in fact, insulin requires supplementation with at least two other hormones to stimulate protein synthesis and proliferation of these cells, and even then it is a weaker stimulant than serum (30). However, the mode of insulin action after combining with its specific receptor is much better understood (31) and is undoubtedly simpler than that of multicomponent serum. Therefore, it was of interest to determine whether insulin would stimulate uridine uptake and how that would respond to Mg²⁺ concentration in the presence of ionophore. In very low Mg²⁺, the addition of insulin had no stimulatory effect on uridine uptake (Fig. 4). In Mg²⁺ concentrations of >1.0 mM, there was a marked increase in uridine uptake in the insulin-treated cells that tended to level off at concentrations of Mg²⁺ above ≈5 mM. In the absence of insulin, the uptake of uridine rose at a slower rate up to 1 mM Mg²⁺ and increased in parallel to the insulin-treated cultures up to ≈5 mM Mg²⁺. When the Mg²⁺ concentration of the medium was raised to 50 mM, the uptake of uridine in the absence of insulin rose to the same level as in its presence. The capacity of 50 mM Mg²⁺ to raise the uridine uptake in the cells without growth factor to the same level as the insulin-treated cells strongly supports a controlling role for Mg²⁺ in the stimulation of uridine uptake by growth factors.

Fig. 4.

Stimulation of uridine uptake by Mg²⁺ in the presence or absence of insulin. Cells were washed and incubated for 1.5 h in the indicated media, all containing 5 μg/ml A23187, followed by addition of fresh medium containing [³H]uridine for 10 min to measure uptake into acid-soluble pools. Each time point is an average of uptake by duplicate cultures.

Effect of Mg²⁺ Concentration on Thymidine Uptake in Cells Stimulated by Insulin in the Presence of Ionophore. As with uridine uptake, the uptake of thymidine is limited by its rate of intracellular phosphorylation (18). Unlike the uptake of uridine, however, the uptake of thymidine is not affected by serum stimulation of cells or by variations in Mg²⁺ that produce large changes in uridine uptake (17). Because the Mg²⁺ variations had been carried out in Ca²⁺-deprived cells with their drastically increased general permeability, a comparison was made of the effects of Mg²⁺ on thymidine and uridine uptake in cells treated with A23187 and stimulated by insulin (Fig. 5). The uptake of uridine in this comparative experiment exhibited the same overall response to Mg²⁺ in the presence and absence of insulin as it had in Fig. 4. In contrast, there was no effect of insulin on the rate of uptake of thymidine, nor was there any effect of Mg²⁺ concentration, either with or without insulin. This result is in accord with the finding that the K_m of thymidine kinase for Mg²⁺ in cell-free extracts is <1/10th that of uridine kinase (17). It suggests that thymidine kinase is saturated with MgATP^2-, even in quiescent cells, so it responds neither to growth factor stimulation nor to increases in cellular Mg²⁺ brought on by addition of Mg²⁺ to the medium.

Fig. 5.

Effect on uridine or thymidine uptake by Mg²⁺ in the presence or absence of insulin. Cells were washed and incubated for 1 h in the indicated media, all containing 5 μg/ml A23187, followed by addition of fresh medium containing [³H]uridine or [³H]thymidine for 10 min to measure uptake into acid-soluble pools. Each time point is an average of uptake by duplicate cultures.

Feedback Inhibition of Uridine Kinase by UTP and CTP and Its Alleviation by Mg²⁺. We have shown that ATP that is not complexed to Mg²⁺ inhibits the phosphorylation of uridine by uridine kinase in cell-free extracts and that inhibition is relieved by Mg²⁺ (17). UTP and CTP are strong feedback inhibitors of uridine kinase (32), which raised the question of whether Mg²⁺ would also relieve their inhibitory activity. The concentration of nucleotide giving half-maximal inhibition decreased 4- and 2.6-fold, respectively, for UTP and CTP, when the Mg²⁺ concentration was lowered from 2 to 0.3 mM (Fig. 6). It is therefore clear that an increase in Mg²⁺ in cells stimulated by growth factors could contribute to uridine trapping by complexing the feedback inhibitors UTP and CTP.

Fig. 6.

Inhibition of uridine kinase by its feedback inhibitors CTP and UTP at two different Mg²⁺ concentrations. Cell-free extract was combined with [³H]uridine, 2 mM ATP, and the indicated concentrations of CTP, UTP, and Mg²⁺ for 30 min at 37°C to determine the rate of uridine phosphorylation. Values are percentages of the rate of uridine phosphorylation at 0 mM CTP or UTP.

Discussion

Treatment of BALB/c3T3 mouse cells with the Mg²⁺ and Ca²⁺ ionophore A23187 made it possible to change the intracellular magnesium concentration by varying extracellular Mg²⁺ without altering the general permeability of the cell, as indicated by the unaltered uptake of l-glucose, and without changing the calcium concentration of the cell. The results confirm that the uptake of uridine, which is limited by its phosphorylation, is responsive to small changes in the extracellular concentration of Mg²⁺ and is unaffected by even large changes of extracellular and intracellular Ca²⁺. It is most sensitive to reductions in Mg²⁺ when the ionophore-treated cells are stimulated by either serum or insulin, although unstimulated cells can be brought to a high rate of uridine uptake by increasing extracellular Mg²⁺ 50-fold in the presence of the ionophore. [It is noteworthy that such high concentrations of Mg²⁺ added to confluent cultures in the absence of Ca²⁺ actually inhibit rather than stimulate protein synthesis (4).] Apparently, uridine kinase is not sensitive to inhibition by the same high concentrations of Mg²⁺ that inhibit the initiation of translation and DNA synthesis (4, 16).

The increased sensitivity of uridine uptake to Mg²⁺ reduction in serum-stimulated confluent BALB/c3T3 cells contrasts with the failure of low Mg²⁺ to inhibit uridine uptake in serum-stimulated lines of Nil 8 hamster cells treated with ionophore, although uptake is practically eliminated in these cells by low Mg²⁺ in the absence of serum stimulation (33). The insensitivity of the serum-stimulated Nil 8 cells to low Mg²⁺ could be understood if the cells were transformed, because proliferation of transformed cells has a much lower requirement for Mg²⁺ than that of nontransformed cells (34, 35). In addition, DNA synthesis in sparse, rapidly multiplying BALB/c3T3 cells is much less sensitive to low Mg²⁺ than it is in quiescent, confluent cells (27), and subculture of stationary confluent mammary epithelial cells at a rapidly growing lower population density results in a 7-fold increase in the total magnesium content of the cells (22). These results are consistent with the membrane, magnesium, mitosis (MMM) model of growth regulation that proposes that perturbation of the cell membrane by growth factors, low population density, or neoplastic transformation releases intracellularly bound Mg²⁺, which activates a wide variety of transphosphorylation and other Mg²⁺-dependent reactions, such as uridine trapping and protein synthesis associated with cell proliferation (9).

The MMM model would also explain the apparent increase in affinity of uridine kinase for ATP^4- in serum-stimulated ATP-depleted cells (36), because the true substrate for this reaction is MgATP^2-, which would be increased by the release of Mg²⁺ from membrane-bound sites. Perhaps more significantly, an increase in intracellular Mg²⁺ would reduce the concentration of other forms of ATP that are not complexed with Mg²⁺ such as ATP^4-, KATP^3-, and HATP^3-, which are strongly inhibitory to transphosphorylation reactions (37, 38). As we show here, the feedback inhibition of uridine kinase by UTP and CTP would also be relieved by increased availability of Mg²⁺. The MMM model also explains why there is a sharp increase in uridine uptake in stimulated cells, even though it is not needed for the growth and proliferation of the cells; i.e., the uridine effect is an incidental by-product of the increased availability of Mg²⁺, which drives the acceleration of translation initiation and energy metabolism vital for DNA synthesis and mitosis that follow. The simple relationship between uridine phosphorylation by its Mg²⁺-dependent kinase, and the rate of uridine trapping by cells might serve as a convenient indicator of the availability of free Mg²⁺ in the cytosol during the transition from quiescence to proliferation.