Phorbol Esters Induce Differentiation in a Pre-B-lymphocyte Cell Line by Enhancing Na+/H+ Exchange*

The murine pre-B-lymphocyte cell line 70Z/3 may be induced to differentiate to a surface immunoglobu-lin-positive phenotype by the polyclonal B-cell mitogen, lipopolysaccharide. This is accomplished via ac- tivation of an amiloride-sensitive Na+-uptake system (Rosoff, P. M., and Cantley, L. C. (1983) Acad. Sei. U. S. A. 80,7547-7550). Here we show that the active tumor promoter 12-0-tetradecanoylphorbol 13-acetate (TPA) also induces surface IgM expression in 70Z/3 cells. TPA also appears to work by activating a plasma membrane Na+/H+ exchange system. A significant increase in cellular Na+ was detected within 10 min after TPA addition and by 2 h an 80% increase was observed. Amiloride blocked both the induction of surface IgM and the increase in cellular Na+. Further evidence that both TPA and lipopolysaccharide activate Na+/H+ exchange was provided by measurements of intracellular pH with carboxyfluorescein. Both TPA and lipopolysaccharide caused a 0.15 unit increase in cytoplasmic pH within 5 min after addition to the cells at 37 "C. The pH change required high extracellular Na* and was inhibited by amiloride. These data suggest a mechanism by which phorbol esters affect cellular growth and differentiation.

For a variety of cell lines, such as Chinese hamster fibroblasts ( l ) , Swiss 3T3 cells (2), MDCK cells (3), and the human epidermoid carcinoma line A431 (4), addition of growth factors results in a rapid elevation of cellular Na+ content. This increase in [Na+]i appears to be due to activation of an amiloride-sensitive, plasma-membrane Na+/H+ exchange system: an efflux of H+ in exchange for Naf leads to cytoplasmic alkalinization. It is unclear whether or not the change in intracellular pH or [Na+Ii (or both) serves as the signal for proliferation. For immature, differentiating cells, a similar mechanism may be operative, although little evidence to date has been offered to support this idea.
The phorbol ester tumor promoters have also been shown to cause an increase in [Na+Ii in quiescent 3T3 fibroblasts ( 5 ) . These agents are known to have mitogenic activity in a variety of different cell systems (6) and presumably mediate such effects by binding to and activating a Ca2+ and phospho-lipid-dependent protein kinase (7). These results suggest that the Na+/H+ exchange system may be directly or indirectly regulated by this kinase.
We have previously reported that treatment of the immature pre-B-lymphocyte cell line, 70Z/3 with the polyclonal Bcell mitogen LPS,' results in a rapid influx of Na' through an amiloride-sensitive uptake system (8,9). LPS also induces differentiation from a surface IgM negative to a surface IgM positive phenotype (8)(9)(10). Other agents which act to raise [Na+Ii, such as ouabain or the Na' ionophore, monensin, also cause differentiation in 702 /3 cells (8,9).
In this report we show that the phorbol ester, 12-0-tetradecanoylphorbol 13-acetate not only induces differentiation in 70Z/3 cells, but does so in a manner similar to LPS, namely by increasing [Na+Ii. In addition we also demonstrate that both LPS and TPA do this by enhancing Na'/H+ exchange, thus leading to an increase in pHi. Both the TPA-induced differentiation and pH change are inhibited by amiloride. These data suggest a possible mechanism for the observed biological effects of phorbol esters.

MATERIALS AND METHODS
Cell Line-70Z/3.B3 cells were used for all experiments. They are a subclone of the parent line supplied by Dr. C. Paige (Basel Institute of Immunology, Basel, Switzerland). Cells were maintained in RPMI 1640 medium supplemented with 15% FCS (GIBCO) and 10 mM Na pyruvate, 10 mM HEPES buffer, 50 p M 2-mercaptoethanol, 100 units/ ml of penicillin G, and 100 pg/ml of streptomycin sulfate. They were cultured a t 37 "C in humidified air, 5% CO,.
Differentiation Assay-All differentiation experiments were conducted with 70Z/3 cells at a concentration of 1-2 X 106/ml and LPS at 10 pg/ml (diluted from a stock solution of 2 mg/ml in phosphatebuffered saline): this concentration had previously been determined to induce maximal surface IgM expression at 24 h (8). Cell viability was determined by exclusion of trypan blue. Surface IgM positive cells were counted by direct immunofluorescence microscopy using a Zeiss fluorescence microscope with epiillumination as previously described (8,9). quantity of *,NaCl was added to 70Z/3 cells suspended in RPMI 1640, Isotope Flux Measurements-For assays of Na" uptake, a trace 15% FCS (pH 7.4) at 2 X IO6 cells/ml and allowed to equilibrate for 15 min at 37 "C (isotopic equilibrium of "NaCl is complete in less ' The abbreviations used are: LPS, lipopolysaccharide; TPA, 12-0tetradecanoylphorbol 13-acetate; [Na+], intracellular Na+ concentration; pH,, intracellular pH, HEPES, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid, MES, 2-(N-morpho1ino)ethanesulfonic acid; FCS, fetal calf serum. than 10 min in these cells (see Ref. 8). At various times after the addition of the test inducing agent, quadruplicate 0.5-ml aliquots were removed and rapidly added to 1.5-ml microfuge tubes containing 0.3 ml of ice-cold phosphate-buffered saline and 0.3 ml(1:I) (v/v) dinonyl phtha1ate:silicone oil. The tubes were centrifuged for 30 s in a Beckman microfuge after which the fluid was drained. Radioactivity was determined in the pellet as described (8,11). ffiRb+ uptake (as a measure of K+ flux) was determined by measuring the rate of ouabain-sensitive and -resistant ffiRb+ influx in TPAor LPS-treated 70Z/3 cells. 50 nM TPA or 10 pg/ml of LPS was added to 70Z/3 cells (suspended at 2 X lo6 cells/ml in RPMI 1640, 15% FCS) 45 min prior to the addition of trace ffiRbC1. When used, ouabain was added to a 1 mM final concentration (from a 10 mM stock in RPMI 1640, 15% FCS) at t = -1 min. =Rb+ (1 pCi/sample) was added at t = 0 min. All experiments using TPA were carried out in subdued light.
Intracellular p H Measurements-Determination of pH, was made by a modification of the method of Thomas et al. (12) by utilizing the pH-dependent fluorescence of carboxyfluorescein. The permeant ester, 5,6-carboxyfluorescein diacetate was loaded into 5 X lo6 702/3 cells/ml at a concentration of 50 p M in a buffer containing 145 mM NaCI, 5 mM KCl, 20 mM MES buffer, 1 mM KH2P04, 2 mM M,SO4, 1 mM CaC12, 0.1 mM EDTA, and 10 mM glucose (pH 6.3). This mixture was incubated in the dark at 37 "C for 35-40 min. At the end of the Ioading period, the cells were washed three times with incubation buffer (130 mM NaCl, 4 mM KC1, 25 mM HEPES buffer, 1.0 mM NaH2P04, 1.8 mM CaCl,, 0.8 mM MgCl2, and 10 mM glucose, pH 7.4) and incubated a further 10 min at 37 "C. The cells were washed once more prior to each experiment. The cells were then suspended at 5 X lo5 cells/ml in the incubation buffer and absorption spectra were obtained from X4w to nm in a Kontron Uvikon 810 dual beam spectrophotometer. Base-line spectra for the appropriate buffer containing untreated cells were subtracted. The ratio of absorption at X4so,r6s nm was compared to a standard calibration curve to obtain pH,: h465 ,,m is an absorption isosbestic point for carboxyfluorescein above pH 6 (12). A standard calibration curve was generated by loading cells with 5,6-~arboxyfluorescein diacetate, then washing and resuspending them in a solution containing 130 mM KCl, 1 mM MgC12, and one of the following buffers: 25 mM MES (pH 6.3), 25 mM MES (pH 6.8), 25 mM HEPES (pH 7.1), 25 mM HEPES (pH 7.5), or 25 mM Tris base (pH 8.2). Cells were permeabilized with 5 pg/ml of nigericin and X490/465 nm absorbance ratios taken. For experiments in which pHi was determined with variable [Na'],, the standard incubation buffer (pH 7.4) was used but with a change in the ratio of NaCl to choline C1 (while maintaining approximate isotonicity). The ratios used were (a) 146 mM NaCl, 0 mM choline CI (

Phorbol Esters
In.duce 70213 Cell Differentiation-When 70Z/3 cells were exposed to low concentrations of the potent tumor promoter TPA for 24 h and tested for the presence of surface IgM, 74% of the cells differentiated to the surface IgM positive phenotype (Table I). LPS induced a similar percentage of 70213 cells. The effects of TPA were detected at concentrations as low as 10 nM, a dose comparable@ that reported to induce differentiation in HL-60 cells (13). The biologically less active phorbol ester, .la-phorbol 12,13-didecanoate, failed to induce expression of surface IgM in 70213 cells. Table I also shows that amiloride inhibits both LPS and TPA induction of differentiation in these cells. These results imply that similar Na+ uptake mechanisms may be important to the mode of action of both agents.
TPA Increases Cellular Nu-Since TPA induces differentiation in 70Z/3 cells, we wished to determine if it did so by a mechanism similar to that of LPS, i.e. by causing an increase in [Na+], through an amiloride-sensitive uptake system. We equilibrated cells with 22NaC1 and measured the cellular Na+ content after addition of 50 nM TPA. The results are shown in Fig. 1. As previously observed with LPS, TPA caused a significant increase in total cellular Na+ within 10 min after addition of the drug. The cellular water content, measured by LPS, TPA, 4a-phorbol 12,13-didecanoate, and amiloride were used at 10 pg/ml, 50 nM, 50 nM, and 66 pM, respectively, and were present for 24 h. Surface IgM positive cells were determined at the end of the 24-h incubation period by immunofluorescence microscopy as described under "Materials and Methods." At least 500 cells were examined in each experiment. Cell viability was greater than 90% as judged by exclusion of trypan blue in all experiments. Results represent means of three experiments.
PDDU, 4a-phorbol12,13-didecanoate. [14C]sucrose and 3H20 equilibration (8) was approximately 0.5 f 0.05 pL/cell and was not significantly altered by TPA addition (data not shown). This result is in agreement with previous measurements made in the presence or absence of LPS (8). Thus, the amount of Na+/liter of cell water (concentration) increased with both treatments. After subtracting away the Na+ trapped in the extracellular fluid (approximately 50% of the total Na' in the pellet in control cells; Ref. 8), the cellular Na+ concentration was estimated to be about 28 mM in control cells and about 50 mM in cells treated with TPA for 2 h).
In order to determine whether the elevation in cellular Na+ was due to an increased influx or inhibited efflux, we measured 86Rb' uptake as an indicator of the activity of the Na+/K+-ATPase. The results are shown in Fig. 2 ., TPA. Data points represent =Rb+ (K+) uptake after subtraction of mean of total uptake minus the ouabain- . .
caused stimulation of ouabain-sensitive %Rb+ uptake (127 and 54%, respectively) with little effect on the ouabainresistant component (although TPA caused a slight inhibition of ouabain-resistant %Rb' uptake). These data indicate that Na+/K+-ATPase activity is increased in these cells in the presence of TPA or LPS, presumably in response to an elevated [Na+],. Thus, the increase in cellular Na' must result from enhanced uptake rather than a decrease in Na+/K+ pump activity. An uncoupling of the pump is not excluded by these data (however, see below). If TPA is acting in a manner similar to that of LPS, then its effects should be blocked by amiloride. Both TPA-induced differentiation (Table I) and Na' uptake ( Fig. 3) were inhibited by amiloride. The Kllz of 0.75 mM amiloride for inhibition of TPA-induced Na+ uptake is in close agreement with that observed for inhibition of LPS-stimulated Na+ uptake (8,9).
LPS and TPA Increase Intracellular pH-Further evidence that TPA and LPS activate a Na'/H+ exchange system was provided from measurements of cytoplasmic pH using the pH-sensitive dye, 5,6-carboxyfluorescein. After cellular loading, this dye appears to be diffusely distributed throughout the cytoplasm as observed by fluorescence microscopy. This reagent does not appear to enter lysosomes or mitochondria (12). The results are shown in Fig. 4. Both LPS and TPA rapidly increased the pHi: this change was detectable within 2 min after adding either drug. The elevated pHi was sustained for at least 20 min at 37 "C (measurements taken at time periods longer than 25-30 min were inaccurate due to slow leakage of the dye from the cells). The less active phorbol, 4a-phorbol 12,13-didecanoate, failed to cause an increase in pHi (data not shown). Amiloride (1 mM) blocked the effects of both TPA and LPS. Amiloride alone caused a decrease in pHi as would be expected if it were inhibiting a membrane transport system that normally functions as a homeostatic mechanism for control of pH,.
In order to determine whether the pH change induced by TPA and LPS is due to Na+/H' exchange, we examined the dependence of the TPA/LPS-induced increase in pHi on external [Na+]. Fig. 5 shows the total change in pH, at 10 min as a function of [Na+],. The elevation in pH, was clearly dependent on the presence of external Na' using either drug. These data show that the drug-induced change in cytoplasmic Inhibition of TPA-stimulated aaNa+ uptake in 7021 3 cells by amiloride: effects of increasing extracellular amiloride concentration. 70Z/3 cells were equilibrated with "Na+, then treated with 50 nM TPA. Various concentrations of amiloride were added at the same time as the TPA. Total cellular Na+ content was measured periodically over 120 min after the addition of the phorbol and amiloride as described in the legend to Fig. 1. The average rate of =Na+ uptake over the first 45 min was plotted versus amiloride concentration. Each data point was determined by linear regression. The initial uptake was 70 pmol/106 cells/min in the absence of amiloride; the error is approximately +15% for each point. pH is driven by a high external Na' concentration. Accurate estimate of initial rates of proton fluxes could not be measured by this technique so the ICllZ for extracellular NaCl stimulation does not accurately reflect the affinity of the transport system for Na' . In any event, when extracellular Na' was reduced to 45 mM, thus effectively eliminating the Na' gradient, no TPA or LPS-induced pH change was observed.

DISCUSSION
In this report we have shown that both the tumor-promoting phorbol ester TPA and the polyclonal B-cell mitogen LPS induce differentiation in a murine pre-B-lymphocyte cell line by enhancing amiloride-sensitive Na+/H' exchange. In 70Z/ 3 cells this event is the critical, rate-limiting step for induction to a surface IgM positive phenotype (8,9). The diuretic amiloride blocks both differentiation and TPA (or LPS)induced Na+/H+ exchange in this cell line. The biologically less active phorbol ester 4a-phorbol 12,13-didecanoate induces neither differentiation (Table I) nor the rise in pHi that accompanies it (data not shown).
Recently, Burns and Rozengurt (15) reported that phorbol esters, as well as other growth factors, raised intracellular pH in Swiss 3T3 cells. They had previously shown that these agents also produced a rapid influx of Na' into these cells (5). These data certainly suggest that a Na+/H' antiport mechanism may mediate phorbol ester stimulation in 3T3 cells, as we have reported here for 70Z/3 cells. Our evidence of the extracellular Na+ requirement for the TPA-and LPS-induced pH change further supports this mechanism.
We also found that TPA decreases ouabain-resistant ffiRb+ uptake in 70Z/3 cells (Fig. 2). Similar results have been reported for BALB/c 3T3 pre-adipose cells exposed to phorbol esters (16). The increase in ouabain-sensitive ffiRb' uptake in 70Z/3 cells after treatment with phorbol esters or LPS presumably results from increased activity of the Na+/K+-ATPase in response to the rise in [Na+]i. It is still unclear whether the signal required for surface IgM expression in 70Z/3 cells is an increase in [Na+Ii or pH,, or both. Ouabain induces 70Z/3 cells to express surface IgM at concentrations which elevate cytoplasmic Na+ suggesting that cytoplasmic Na' levels are important for differentiation (8,9,17). However, we found that ouabain also raises pH, by an unknown mechanism (data not shown). Ouabain has also been shown to affect pH regulation in A431 cells (4). Monensin also induces differentiation in these cells presumably by exchanging Na+ for H' , thereby leading to a rise in [Na+], as well as cytoplasmic alkalinization (8). Monensin is also capable of overcoming the amiloride inhibition of differentiation in this system (8,9), further supporting the importance of the Na'/H+ exchange system for the initiation of differentiation. However, since all agents which induce surface IgM expression cause both a rise in cytoplasmic pH and Na', it is still not possible to determine which ionic event is critical to this process.
The evidence that amiloride blocks surface IgM expression by inhibiting Na'/H' exchange should be discussed further. Several reports have suggested that amiloride worked by competitively inhibiting Na' binding to an antiport protein from the outside of the cell (1,4,14). In this paper we show that the K1l2 of amiloride for inhibition of TPA-induced Na' uptake is 0.75 mM whereas only 66 p M amiloride is needed to block differentiation. However, we have previously shown that when 70Z/3 cells are incubated in 66 p~ amiloride at 37 "C the drug is concentrated into the cells to a plateau value of approximately 0.8 mM after 2-3 h and this level is maintained for up to 16 h (9). Since LPS-stimulated Na+ uptake is inhibited after several hours of preincubation in 66 PM amiloride we have suggested that amiloride can also block Na+/ H' exchange from inside the cells (9). These data could explain the apparent discrepancy between the amount of amiloride needed to block Na+ uptake from the outside of the cell immediately after addition of drug and that required to inhibit differentiation over a 24-h period. Although amiloride may be having diverse effects in the cell, the fact that the Na+/H' ionophore, monensin, circumvents the amiloride inhibition of differentiation argues that the important effect is inhibition of the exchange system (8,9).
Perhaps the most intriguing aspects of the data presented here concern the mode of action of the phorbol esters. These agents have a variety of biological effects in addition to their tumor-promoting activity (5)(6)(7)(18)(19)(20)(21)(22)(23)(24)(25)(26). In some cell lines they induce proliferation or differentiation while in others they appear to retard growth. These compounds have also been reported to have a variety of effects on lymphocytes such as stimulation of phospholipid metabolism (27), and activation (28), as well as growth cessation with concomitant stimulation of immunoglobulin secretion in human B-cells (29). Recently it has been shown that phorbol esters bind specifically to, and activate, protein kinase C, the Ca2+-activated, phospholipiddependent protein kinase, by substituting for diacylglycerol (7). Although the physiologically important substrates of this kinase have not been identified, the rapid changes in Na+ and H+ fluxes observed in 70Z/3 cells after treatment with TPA suggest that the Na'/H+ exchange system is directly or indirectly affected by this kinase, the activation of which may be integral in triggering differentiation in this cell line.