Intracellular pH of Stimulated Thymocytes Measured with a New Fluorescent Indicator*

A new fluorescent intracellular pH indicator is de- scribed (“quene l”) which is related to the tetracarbox-ylate Ca2+ indicator based on the quinoline fluorophor (“quin 2”). Quene 1 has excitation and emission maxima at 390 and 530 nm, respectively, and shows a 30-fold increase in fluorescence between pH 5 and 9 with a pK, of 7.3. The fluorescence is insensitive to Ca2+ and M 8 + at free concentrations up to lo-‘ M and to the propor-tions of Na+ and K+ at total concentrations of Na+ and K+ from 100 to 200 m ~ . The indicator is loaded into thymocytes using the tetraacetoxymethyl ester deriv- ative which is hydrolyzed in the cells to give the tetra-carboxylate anion. Intracellular pH can be measured at intracellular quene 1 concentrations of approximately 0.1 m~ and quene 1 does not perturb glycolysis or the ATP level in resting cells at concentrations up to 0.8 m ~ . The intracellular pH of mouse thymocytes indi- cated by quene 1 is 7.15 f 0.04 and it is insensitive to the concentration of Ca2+ or M$+ in the extracellular medium. The intracellular pH decreased when the pH of the medium was lowered by addition of HCl, but was insensitive to NaOH at extracellular pH values up to 8.0. Rapid transient changes in intracellular pH are induced by m C 1 , NaC02CH3, or HC03-/C02. The thymocytes showed no early changes (within 30 min) in

medium. The intracellular pH decreased when the pH of the medium was lowered by addition of HCl, but was insensitive to NaOH at extracellular pH values up to 8.0. Rapid transient changes in intracellular pH are induced by m C 1 , NaC02CH3, or HC03-/C02. The thymocytes showed no early changes (within 30 min) in intracellular pH in response to mitogenic concentrations of lectins or 4/3-phorbol-12-myristate-13-acetate.
An increase in pHi,' following an increase in the [Ca], may occur during the stimulation of cell growth in a variety of cells. Within 2 min after fertilization of the sea urchin egg, a large transient increase in [Cali is followed by an increase in pHi (1,2). In fibroblasts, an early increase in pHi of about 0.2 unit has been reported in response to a synergistic mixture of growth factors, and the effect has been attributed to the stimulation of Naf/H' exchange (3). From other studies, it * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" i n accordance with 18 U.S.C. Section 1734 solely to indicate this fact. has been suggested that stimulation of the Na+ flux in fibroblasts may be due to a prior increase in [Cali (4). In lymphocytes, prolonged changes in pHi in response to the mitogenic lectin ConA have recently been described (5), with maximum increases jr pHi in mouse splenocytes at 6 and 40 h and a minimum pHi, close to the value in quiescent cells, at 12 h. Using the fluorescent indicator quin 2, the [Cali in thymocytes has been shown to increase by about 2-fold within 30 s of the addition of mitogenic concentrations of ConA (6). These observations suggest that there may be a common sequence of changes in [Cali and pHi when different types of cells enter the cell cycle. Are these ionic changes obligatory for entry into the cell cycle and are they causally related to each other?
The current techniques available for the measurement of pHi in thymocytes and other small cells are unable to resolve the time course of any changes in pHi which may occur within 30 s of stimulation. We have synthesized a fluorescent pH indicator related to quin 2, designed to respond rapidly to changes in normal pHi values and to have negligible affinity for Ca2+ at normal levels of [Cali. We report the characterization of this new probe ("quene l"), the measurement of pHi in mouse thymocytes, and the effect of varying external pH and cation concentrations on pHi. We have also examined the effects of several mitogens on thymocytes to determine whether there are any changes in pHi correlated with the early changes reported in [Cali.

Cell Preparation and Loading with
Quene I-Thymocytes were prepared from 4-6-week-old BALB/c mice by teasing the thymus into RPMI 1640 medium buffered with 10 m~ Hepes, pH 7.3. The cells were centrifuged (500 X g, 5 min) and resuspended at 107/ml in a standard medium of the inorganic salts of RPMI 1640 without phenol red (103 m~ NaCl, 5.6 mM NazHP04, 5.36 m KCl, 0.41 mM MgS04. 7&0,0.42 m~ Ca (N03)~.4H20) supplemented with 11 mM glucose and buffered with 10 m~ Hepes, pH 7.3. The cells were incubated for 30 min at 37 "C with 1 PM [3H]quene 1 acetoxymethyl ester (540 Ci/mole) added from a stock solution in Me2530 (final concentration c 0.1% v/v) before washing by centrifugation (500 X g, 5 min) and suspending the cells in standard medium. After a further 30 min at 37 "C, the cells were washed as before and resuspended in standard medium at 5 X IO6/& for fluorescence measurements. Cell aliquots (2.0 ml) were equilibrated in 1-cm path length cuvettes for 10 min at 37 "C in a Perkin Elmer Model 44B spectrofluorimeter before measurements of pHi (excitation 390 nm, emission 530 n m ) . C o d , PHA, and WGA were obtained from Miles Yeda and were added from stock solutions in standard medium. PMA, TTFB (Sigma), and valinomycin (Serva International) were added in Me2SO (final Me2SO concentration less than 0.1%). The intracellular [3H]quene 1 concentration was estimated by counting 3H in an aliquot of the cells, assuming an aqueous cell volume of 104 pm3 determined from the [3H]H20 volume.

Synthesis of Quene I-8-[bis(ethoxycarbonylmethyl)amino]-6methoxy-2-[trans-2[bis(ethoxycarbonyl methyl)amino]styryl]quino-
line was prepared by condensation of 6-methoxy-8-nitroquinoline with 0-nitrobenzaldehyde followed by reduction to the diamine with stannous chloride. Alkylation of the bis amine with ethyl bromoacetate and hydrolysis of the ester gave quene 1 (Structure 1). Full details of the chemical synthesis of quene 1 and other new quinolinebased indicators will be published elsewhere. The preparation of quene 1 and [3H]quene 1 tetraacetoxymethyl esters were by the methods described previously for quin 2 (7). ester are shown in Fig. 1. On hydrolysis to quene 1, the emission maximum shifts from 490 to 530 nm and is independent of pH. The fluorescence intensity of quene 1 increases by more than 30-fold from pH 5 to 9 and an apparent pK, of 7.30 was derived from the titration shown in Fig. 1 (inset). Quene 1 in total concentrations of sodium and/or potassium chloride from 100 to 200 m~ gave the same fluorescence intensity at constant pH and, therefore, the indicated pHi is unaffected by any changes which may occur in Na+ or K' concentrations in the cells. Both Ca2+ and M$' ions quench quene 1 fluorescence at pH 7.1 at concentrations above M (Fig. 2). This concentration of Ca2+ is much higher than normal [Cali values (-10" M), but free intracellular Mg2' concentrations of up to 1 m~ will slightly reduce intracellular quene 1 fluorescence by an amount equivalent to an offset for pHi calibration of up to +0.05 unit. Measurement ofpH, in Thymocytes-The measurement of pHi is illustrated in Fig. 3. The fluorescence from thymocytes loaded with quene 1 is calibrated by the addition of 0.05% Triton X-100 in 0.5 m~ EGTA + 0.5 mM EDTA to lyse the cells and release quene 1 into medium containing free Ca2' and M$+ concentrations well below the levels which affect quene 1 fluorescence (Fig. 2). Serial additions of 1 M Tris and 0.5 M HC1 solutions calibrate the fluorescence of the released quene 1 as a function of pH. Complete release of quene 1 from the cells was demonstrated by the addition of 0.1 mM MnC12 which quenches quene 1 fluorescence immediately by more than 99% to give a background signal which is the same as from cells without quene 1. The emission maximum of quene 1 was the same whether loaded inside the cells or added externally to unloaded cells. Addition of Triton X-100 to unloaded cells in the presence of external quene 1 did not affect the fluorescence intensity under the conditions defined in the legend to Fig. 1; therefore, the calibration is not affected by changes in light scattering or autofluorescence of the cells on addition of Triton X-100.

IntracellularpH and
In most experiments, the quene 1 fluorescence from cells loaded in the standard medium was stable for many minutes. Addition of 0.1 m~ MnC12 to aliquots of cells at increasing times after the cells had been washed by centrifugation (500 X g, 5 m i n ) to remove any external quene 1 indicated that leakage of quene 1 was insignificant over 30 min in stable preparations (cO.l%/min). Centrifugation of the loaded cells caused variable leakage of quene 1, detected from the immediate decrease in fluorescence from the resuspended cells on adding 0.1 m~ MnCL In some preparations, the fluorescence intensity from cells loaded with quene 1 decreased continuously for at least 20 min and there was also enhanced leakage of quene 1 from the cells (up to 5% in 10 min). We have been unable to determine why this occurs in some preparations, but noted that the downward drift in fluorescence could usually be stopped and the fluorescence intensity stabilized by the addition of 10 m~ NaHC03/C02 (see below).
The estimated intracellular pH was 7.15 -+ 0.04 (S.D., n = 9) when the extracellular pH was 7.3 throughout the stages of cell preparation, loading with quene 1, and assay. This pHi value is uncorrected for the small quenching effect of a free intracellular Mg+ concentration of approximately 0.8 mM (12), equivalent to an offset of approximately t0.04 unit estimated from the data in Figs. 1 and 2. It should be noted that the technique is very sensitive to relative changes in pHi (kO.01 unit) in contrast to the reproducibility of absolute pHi measurements. The replacement of Na' by K' in the medium and the subsequent addition of up to 10 m~ NaCl had no effect on quene 1 fluorescence. However, when 10 m~ NaCl was added to cells in medium in which Na' was replaced by  choline (choline medium), there was a small increase in fluorescence equivalent to an increase in pHi of approximately 0.02 unit.
Lowering the external Ca2+ and MgZf concentration in the medium to M by the addition. of 0.5 m~ EGTA + 0.5 m~ EDTA had no immediate effect on quene 1 fluorescence from the cells. The intracellular pH, therefore, appears to be insensitive to the extracellular concentrations of Ca2+ and Mg2+ for at least several minutes.
In previous studies (7), we have found that intracellular quin 2 causes increased lactate production at intracellular concentrations between 0.1 and 1 m, and depletes cellular ATP levels at concentrations above 0.8 mM. In contrast, quene 1 had no significant effect on lactate production or the ATP level (<lo% decrease) at intracellular concentrations up to 0.8 m~ (Fig. 4). It is therefore unlikely that the intracellular as described under "Materials and Methods" for fluorescence measurements and resuspended in standard medium at 5 X lo6 cells/ ml at 37 "C. ATP content (B) and lactate output (0) were measured after incubation for 1 h at 37 "C as described previously (7). In control cell samples without quene 1, the ATP content was 500 fmol/cell    serial additions of 0.5 M NaOH to the medium had no effect on quene 1 fluorescence in the cells over at least 10 min (Fig.   sa). Centrifugation of the cells in medium at pH 7.3 and resuspension at pH 5 8.0 also had no effect on the indicated pHi. In contrast, additions of 0.5 M HC1 caused a slow decrease in pHi over several minutes to a new stable level (Fig. 5 b ) . Subsequent additions of 0.5 M NaOH did not affect the quene 1 fluorescence, indicating that the additions of HC1 did not release quene 1 from the cells. Centrifugation of the cells in medium at pH 7.3 and resuspension at pH 7.3 to 6.8 caused a decrease in pHi similar to that due to the addition of HC1. The response to additions of HC1, therefore, is not attributable to damage to the cells by transient exposures to low pH during mixing. The pattern of responses to additions of HC1 or NaOH were similar in medium containing 10 m~ HCOs-/COz and, therefore, is not due to the absence of physiological buffer in the medium. The pHi can be made to respond rapidly and reversibly to changes in external pH by Tris (or NaOH) and HC1 by treating the cells with the mitochondrial uncoupler TTFB and 1 p~ valinomycin in a high K' medium containing 103 m~ KC1 and 5.6 m~ NaCl instead of the standard Na+, K+ concentrations (Fig. 5c).
Effect of Weak Bases and Acids-Addition of 5 m~ NH&1 to the medium at constant pH (7.27 f 0.02) caused a rapid transient increase in pHi (Fig. 5d). Further additions of 5 m~ NKCl caused progressively smaller responses. Addition of 10 mM sodium acetate to the medium caused transient decreases in pHi and similar effects were observed on addition of 10 m~ NaHC03/C02 from a stock solution of 1 M NaHC03 saturated with Con (Fig. 5, e and f). From these data, the buffering capacity of the cell is estimated as equivalent to approximately 3 fmol of base/cell compared with 0.01 to 0.1 fmol of quene 1/ cell (=0.1 to 1 mM intracellular quene 1) (8).
Effects of Mitogens-In many experiments, mitogenic and supramitogenic concentrations of C o d (1 pg/ml and 3 pg/ml) or PHA (1 p g / d and 2 pg/ml) had no effect on quene 1 fluorescence in the first 5 min after addition of the mitogens to the cells, irrespective of the medium used or the intracellular concentration of quene 1 preparations in which quene 1 fluorescence was stable. The fluorescence intensity and emission maximum were the same for quene 1 both within cells and in media which simulated the intracellular concentrations of cations. No significant effects of quene 1 on ATP level or lactate production which might affect pHi were detected at intracellular concentrations below 0.8 m~. This observation is consistent with previous data which indicated that metabolic stimulation by intracellular cation chelators is correlated with their affinity for Ca2+ (or other M"+ ions). Quene 1 has an affinity for Ca2+ of pKcaz+ = 2.70 compared with 7.05 for quin 2 (7). An adverse feature of quene 1 as a pH indicator is that it is necessary to correct the pHi calibration for the concentration of free Mg2+ in the cell by about +0.05 pH unit for thymocytes, and changes in intracellular Mg2+ levels of greater than 2-fold could interfere with pHi measurement by quene 1. In the present experiments, however, external Mg'+ concentrations from to M had no effect on quene 1 fluorescence over several minutes, or on the pattern of responses to acids and bases shown in Fig. 5. It should be possible to design derivatives of quene 1 with PKM~Z+ well below 2, which would be insensitive to normal intracellular free M g + concentrations.
The indicated pHi value of 7.15 k 0.04 is close to the value of 7.18 obtained from the uptake of DM0 into mouse spleen lymphocytes (5). However, the two probes may report on different intracellular environments. The distribution of DM0 gives some weighted, average intracellular measure of pH including the intra-organelle environment (11) whereas, using image intensified microscopy, quene 1 appears to be uniformly distributed throughout the cells with no observable concentration into organelles. Furthermore, quene 1 is not accumulated by isolated thymocyte mitochondria, but we cannot exclude the possibility that there may be significant uptake into other organelles (e.g. lysosomes or endoplasmic reticulum). In a recent paper, an intracellular indicator based on fluorescein was used to measure pHi in pig lymphocytes (12).
The pHi value was approximately 7.0, which is lowel than that reported here using quene 1, but the general pattern of responses obtained with strong bases and weak acids and bases was similar to that described here.
The insensitivity of pHi to the concentrations of Ca2+ and Mg2+ in the extracellular medium suggests that neither of these ions are directly coupled to the systems which regulate pHi. The pHi is normally stable without added HC03-/C02, but the stabilizing effect of low concentrations of HC03-/C02 on preparations in which the pH, drifts downwards may indicate a physiological role for bicarbonate in pH, regulation in these cells, as found in other systems (13). This observation and the small increase in pHi in response to Na' in the choline medium require further evaluation.
The sensitivity of the indicator to changes in pHi is demonstrated by the rapid transient responses to additions of NH,', CH3COO-and HC03-/C02 ions, and the titrations of pHi by Tris and HC1 in the presence of TTFB and valinomycin in high K+ medium. In normal medium, the cells are insensitive to increase in external pH by strong bases, but respond to HC1 with a slow decrease in pHi. This decrease in pHi is not due to damage to the mechanisms of pHi regulation since similar effects are observed when the cells are resuspended in medium at lower pH with or without HC03-/C02. Treatment of the cells with HC1, Tris, or NaOH under the conditions described did not affect the [Ca], in resting cells indicated by quin 2 or the increase in [Cali in response to C o d , confiiing that the functions of the plasma membrane proteins are not generally degraded by additions of acid or base.
The absence of any effect on pHi of the mitogens ConA and PHA which increase [Cali, or of PMA which decreases [Cali, indicates that the free concentrations of Ca2+ and H+ in the cytoplasm can be altered independently in the thymocyte. Furthermore, any pH, changes which may be necessary for, or result from, mitogenic stimulation occur at least 30 min after the early increase in [Cali observed within 30 s of the addition of ConA. Using quene 1, we have observed no change in pHi over the fist 12 h after stimulation of thymocytes by ConA, in marked contrast to the data reported for mouse splenocytes using DM0 to measure pHi (5).