INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

IT IS NOW WELL ESTABLISHED that the forestomachs are the main and most important sites of Mg²⁺ absorption in ruminants (34). Many in vivo and in vitro experiments have been performed on Mg²⁺ transport and the mechanisms of impaired Mg²⁺ absorption by high ruminal K⁺ concentrations (3, 22, 24). The obtained data have revealed that net Mg²⁺ absorption from the rumen occurs against an electrochemical gradient (3, 24) and is mainly mediated by an active transcellular pathway (26). The mechanisms of this cellular Mg²⁺ transport, that is, luminal uptake and basolateral extrusion, are not well understood. So far, investigations at the tissue level suggest that two different transport mechanisms exist for the apical uptake of Mg²⁺. Part of the unidirectional mucosal-to-serosal Mg²⁺ flux (J ^Mg_ms) is potential difference (PD) dependent and K⁺ sensitive and may represent passive Mg²⁺ uptake by a channel (22, 31). Another somewhat greater part (62% of J ^Mg_ms) is electrically silent, dependent on luminal short chain fatty acids (SCFA), CO₂/HCO₃, and Cl, and may in part result from apical Mg²⁺/H⁺ exchange (23). The existence of a Mg²⁺/H⁺ exchange as a second PD-independent Mg²⁺ uptake mechanism has been derived indirectly from several observations. Feeding high levels of easily fermentable carbohydrates to ruminants increases Mg²⁺ availability (10, 13). Such a diet leads to an alteration of microbial activity and composition in the rumen contents. Among other parameters, the concentrations of SCFA and CO₂/HCO₃ are increased; these are the major end products of the microbial digestion of carbohydrates in the ruminal lumen and stimulate Mg²⁺ net absorption in vivo (10, 27) and in vitro (23). In vitro experiments with isolated sheep rumen epithelium have revealed that this increase in net Mg²⁺ absorption is entirely attributable to a stimulation of J ^Mg_ms, which is specifically reduced by removal of SCFA, CO₂/HCO₃, and Cl (23). Because the stimulating effect of SCFA on J ^Mg_ms depends on their lipid solubility (acetate < propionate < butyrate) and because the carbonic anhydrase inhibitor ethoxyzolamide reduces J ^Mg_ms in SCFA-free buffer, it has been suggested that the PD-independent stimulation of ruminal Mg²⁺ transport depends on permeant anions that supply substrates for Mg²⁺/H⁺ and Cl/HCO₃ exchange mechanisms in the apical membrane of the ruminal epithelium (23). SCFA and CO₂/HCO₃ are the predominant anions in the ruminal fluid and are readily absorbed by the stratified epithelium of the rumen (7, 28). Therefore, it has been hypothesized that the supply of H⁺ for this exchange comes in part from the intracellular dissociation of SCFA that are absorbed in their nonionized form by diffusion and from the intracellular hydration of CO₂, produced in the lumen by microbial fermentation and in the mucosa by SCFA catabolism (23). In such a model, the effect of Cl withdrawal could be explained by a reduction of the Cl/HCO₃ exchange activity that is present in the ruminal epithelium (25).

Only a few data are available regarding the basolateral Mg²⁺ efflux from the epithelial cells. The significant uphill electrical gradient (basolateral PD = 50-70 mV) for the basolateral exit of Mg²⁺ would suggest the participation of an energy-dependent mechanism. Because inhibition of the Na⁺-K⁺-ATPase by ouabain reduces the net movement of Mg²⁺ by 90% (26), and because of the correlation between net Mg²⁺ and net Na⁺ efflux from the rumen (3), it is assumed that Mg²⁺ efflux takes place via nonelectrogenic Na⁺/Mg²⁺ exchange, by utilization of the electrochemical gradient for Na⁺ (generated by Na⁺-K⁺-ATPase).

The existence of the proposed transport proteins for apical Mg²⁺/H⁺ and basolateral Na⁺/Mg²⁺ exchange has not as yet been shown directly. Therefore, the aim of the present study has been to obtain preliminary information about the PD-independent component of Mg²⁺ transport and possible transport proteins involved at the cellular level. To this purpose, we have performed experiments with isolated ruminal epithelial cells (REC). With the aid of the fluorescent probes mag-fura 2, sodium-binding benzofuran isophthalate (SBFI), and 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein (BCECF), we have measured the free intracellular Mg²⁺ concentration ([Mg²⁺]_i), the intracellular Na⁺ concentration ([Na⁺]_i), and the intracellular pH (pH_i) of REC under basal conditions, after stimulation with butyrate and HCO₃, and after changing the transmembrane chemical gradients for Mg²⁺, H⁺, and Na⁺.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. Medium 199, trypsin, glutamine, antibiotics (gentamycin, nystatin, and kanamycin), and FCS were purchased from Sigma (St. Louis, MO). Dulbecco's PBS (DPBS) and collagen were obtained from Biochrom (Berlin, Germany). Mag-fura 2-AM, SBFI-AM, BCECF-AM, and pluronic acid were from Molecular Probes (Eugene, OR). All other chemicals were purchased from Sigma.

Cell culture. Primary cultures of REC were prepared as described by Galfi et al. (12). Briefly, REC were isolated by fractional trypsination and grown in medium 199 containing 10% FCS, 1.36 mM glutamine, 20 mM HEPES, and antibiotics (50 mg/l gentamycin and 100 mg/l kanamycin) in an atmosphere of humidified air-5% CO₂ at 38°C. Experiments were performed between 6 and 12 days after seeding.

Solutions. The control solution was the NaCl solution (in mM: 145 NaCl, 5 KCl, 1 CaCl₂, 2 MgCl₂, 10 HEPES, and 5 glucose, pH 7.4). In HCO₃-buffered solutions 20 mM NaCl was replaced by NaHCO₃, and in butyrate-containing solutions a further 20 mM NaCl was replaced by sodium butyrate. All HCO₃-containing solutions were preequilibrated with 5% CO₂ and 95% air. In Na⁺-free solutions, NaCl was replaced by N-methyl-D-glucamine (NMDG)-Cl. The high-K⁺ solution contained (in mM) 15 NaCl, 135 KCl, 1 CaCl₂, 2 MgCl₂, 10 HEPES, and 5 glucose, pH 7.4. When the Mg²⁺ content of any solution was increased, an appropriate amount of NaCl, KCl, or NMDG-Cl was replaced to maintain osmolarity.

Measurement of cytoplasmic Mg²⁺, Na⁺, and pH by spectrofluorometry. Cells were loaded with 5 µM mag-fura 2-AM, 10 µM SBFI-AM, or 0.5 µM BCECF-AM for the determination of [Mg²⁺]_i, [Na⁺]_i, and pH_i, respectively. Cells were subsequently washed twice in DPBS. REC were incubated for a further 30 min to allow for complete deesterification and washed twice before measurement of fluorescence. Intracellular ion concentrations were determined by measuring the fluorescence of the probe-loaded REC in the LS-50 B spectrofluorometer (Perkin-Elmer), by using the fast filter accessory, which allowed fluorescence to be measured at 20-ms intervals, with excitation for mag-fura 2 and SBFI at 340 and 380 nm and for BCECF at 440 and 480 nm, and with emission at 515 nm. All measurements were made at 37°C in a 3-ml cuvette containing 2 ml cell suspension (10% cytocrit) under stirring. The measurements with HCO₃-containing solutions were done after the cell suspensions were preequilibrated with 5% CO₂ and 95% air. During the experiments the cuvette was tightly closed with a plastic cap to prevent CO₂ leakage.

Statistical analysis. If not otherwise stated, data are presented as means ± SE. Significance was determined by Student's t-test or Tukey's ANOVA as appropriate. Correlations between variables were tested by calculating Pearson's product moment correlation coefficients. P < 0.05 was considered significant. All statistical calculations were performed with SigmaStat (Jandel Scientific).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of butyrate and HCO₃ on pH_i and [Mg²⁺]_i. SCFA and CO₂/HCO₃ stimulate Mg²⁺ net absorption across rumen epithelium in vivo and in vitro. Therefore, our initial goal was to define the acute cellular response to SCFA and/or HCO₃ exposure in terms of both [Mg²⁺]_i and pH_i. Mag-fura 2-loaded and BCECF-loaded REC in suspension were equilibrated in control solution without butyrate and HCO₃, in Na⁺ medium with 20 mM butyrate and 20 mM HCO₃, or in Na⁺ medium with only 20 mM HCO₃, and [Mg²⁺]_i and pH_i were measured over a 10-min period. Extracellular pH (pH_e) was 7.4, and extracellular [Mg²⁺] ([Mg²⁺]_e) was 2 mM in all solutions. Cells in control solution had a mean resting pH_i of 6.83 ± 0.1. On exposure to medium with 20 mM HCO₃ or with 20 mM HCO₃ and 20 mM butyrate, pH_i dropped to 6.67 ± 0.2 and 6.58 ± 0.13, respectively, and then recovered after 10 min to 6.87 ± 0.2 and 6.76 ± 0.12 (Fig. 1). The resting [Mg²⁺]_i, determined at the beginning of the experiments, was not significantly different in experiments performed with HCO₃ and HCO₃/butyrate media (0.64 ± 0.09 mM and 0.7 ± 0.26 mM) compared with that in HCO₃- and butyrate-free control solution (0.56 ± 0.14 mM), but [Mg²⁺]_i starting levels tended to be higher with butyrate and/or HCO₃ in the bath solution. In all media, an increase of [Mg²⁺]_i was observed. The increase of [Mg²⁺]_i was higher (Fig. 2) in media with HCO₃ (27.7 ± 5.0 µM/min) or HCO₃ and butyrate (29.0 ± 10.6 µM/min) compared with control medium (15.0 ± 1.0 µM/min).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Effect of HCO3 and butyrate on intracellular pH (pHi) of ruminal epithelial cells (REC). Measurements were made after a 5-min preincubation in either control solution (Na+ medium without butyrate or HCO3) or medium with 20 mM butyrate and/or HCO3. Values are means ± SE of 4 single experiments. * P < 0.05 for pHi starting level of REC in HCO3/butyrate medium vs. control.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Effect of HCO3 and butyrate on intracellular Mg2+ concentration ([Mg2+]i) of REC. Measurements were made after a 5-min preincubation in either control solution (Na+ medium without butyrate or HCO3) or medium with 20 mM butyrate and/or HCO3. Magnitude of mean [Mg2+]i change ([Mg2+]i) in these media is given. Values are means ± SE of 4 single experiments. * P < 0.05 vs. control.

Experimental evidence of the Na⁺/H⁺ antiporter. It has been observed previously that SCFA stimulate the transepithelial Na⁺ absorption from the ruminal fluid and that this effect is sensitive to high doses (1 mM) of the diuretic amiloride (11, 25). It is thought that acidification caused by nonionic diffusion of SCFA activates Na⁺/H⁺ exchange in the luminal membrane of rumen epithelium. To demonstrate the presence and activity of Na⁺/H⁺ exchange, we have measured pH_i under several conditions known to inhibit this antiporter: 1) after addition of amiloride or ethylisopropylamiloride, which are recognized inhibitors of the antiporter (20); 2) after substituting extracellular Na⁺ with NMDG, thus reversing the transmembrane Na⁺ gradient, which is the driving force for Na⁺/H⁺ exchange; and 3) after addition of 10⁴ M 8-bromo-cAMP, which decreases the activity of the antiporter through cAMP-dependent protein kinase (5, 35). The results of these experiments are presented in Fig. 3A. We have also shown by control experiments that pH_i recovery after butyrate exposure is dependent on extracellular Na⁺ concentration ([Na⁺]_e) (Fig. 3B). Therefore, it is likely that part of the effects of butyrate and/or HCO₃ exposure on the pH_i of REC (initial decrease and recovery to near control level) are attributable to Na⁺/H⁺ exchange and that this activity can mask or overlap Mg²⁺/H⁺ exchange.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   A: effect of various experimental conditions known to inhibit Na+/H+ antiport on pHi in REC. Magnitude of mean pHi change (pHi) is given for each condition. In all cases, pHi fell significantly (* P < 0.05). Bars represent means ± SE, and number of single experiments is shown in parentheses. EIPA, ethylisopropylamiloride. B: original tracing of a measurement of pHi in REC. There is no pHi recovery in cells exposed to butyrate if Na+/H+ exchanger is completely blocked by removal of extracellular Na+ [replaced by N-methyl-D-glucamine (NMDG)-Cl]. [Na+], Na+ concentration. Trace is representative of 6 independent experiments.

Effects of Na⁺ withdrawal on [Mg²⁺]_i, [Na⁺]_i, and pH_i. To avoid Na⁺/H⁺ exchange and other pH-regulating mechanisms, further experiments were performed with Na⁺- and HCO₃-free media. Figure 4 shows the intracellular H⁺ concentration ([H⁺]_i) and [Na⁺]_i of REC before and after removal of extracellular Na⁺. As [Na⁺]_e was reduced to zero, pH_i fell from 6.86 ± 0.32 to 6.4 ± 0.14 and [Na⁺]_i fell from 18.95 ± 3.9 to 10.3 ± 4.7 mM. Under these conditions, the Na⁺/H⁺ exchanger is not capable of extruding protons, and, as a result, [H⁺]_i increases about three times (Fig. 4). If there is a Mg²⁺/2H⁺ exchanger in the cell membrane of REC, increasing [Mg²⁺]_e should decrease [H⁺]_i and increase [Mg²⁺]_i under these experimental conditions. The results are shown in Fig. 5. As expected, [Mg²⁺]_i increased from 0.87 ± 0.19 to 1.15 ± 0.45 mM when [Mg²⁺]_e was stepwise increased from 0 to 2.5, 5.0, and 7.5 mM; on the contrary, pH_i shifted toward more acidic pH values and fell from 6.42 ± 0.2 to 6.24 ± 0.2 during exposure to high [Mg²⁺]_e. Recovery of pH_i did not occur before Na⁺ was added to the bath.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Intracellular concentrations of H+ and Na+ ([H+]i and [Na+]i) of REC before and after omission of Na+ from extracellular solution. NaCl was isosmotically replaced by NMDG-Cl. Number of single experiments is shown in parentheses. ** P < 0.01 vs. control (solution with a [Na+] of 145 mM).

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Cytoplasmic pH and Mg2+ responses to addition of extracellular Mg2+ in REC exposed to Na+-free NMDG medium. Under these experimental conditions, H+ gradient was outwardly directed (pHi << extracellular pH), and we simultaneously generated an inwardly directed Mg2+ gradient {extracellular [Mg2+] ([Mg2+]e) >> [Mg2+]i} by increasing the [Mg2+] in the medium from 0 to 2.5, 5.0, and 7.5 mM in a stepwise manner. A: records from a typical experiment are shown. B: means ± SE of results from 6 different experiments are shown; pHi and [Mg2+]i are shown plotted against time. There was an inverse relationship between [Mg2+]i and pHi in all experiments (r = 0.98).

Interrelationship between [Na⁺]_i and [Mg²⁺]_i. As shown above, the [Mg²⁺]_i of REC increased continuously when the cells were exposed to Na⁺-free media and [Mg²⁺]_e was stepwise increased from 0 mM to 2.5, 5.0, and 7.5 mM. At the same time, [Na⁺]_i fell from 10.2 ± 4.8 to 8.4 ± 4.4 mM. An example for this interrelationship is given by the original recordings presented in Fig. 6. After readdition of Na⁺ to the extracellular solution, [Na⁺]_i increased and the [Mg²⁺]_i increase was blocked completely (Fig. 6). Figure 7 shows in summary that the increase in [Mg²⁺]_i and the decrease in [Na⁺]_i are substantially reduced if the [Na⁺]_e is increased. In Na⁺-free media, [Mg²⁺]_i increased by 114 ± 31, 189 ± 47, and 248 ± 64 µM, and [Na⁺]_i decreased by 1.3 ± 0.3, 1.6 ± 0.2, and 1.9 ± 0.3 mM at [Mg²⁺]_e of 2.5, 5.0, and 7.5 mM, respectively. In media containing 145 mM Na⁺, a rise in [Mg²⁺]_i and a fall in [Na⁺]_i was not detectable until the external Mg²⁺ concentration was 7.5 mM. Compared with Na⁺-free media, there were only slight changes in [Mg²⁺]_i and [Na⁺]_i; [Mg²⁺]_i rose by 33 ± 24 and 87 ± 9 µM and [Na⁺]_i fell by 0.2 and 0.9 mM in Na⁺ media with 5 and 7.5 mM Mg²⁺, respectively.

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Effect of variation in extracellular [Na+] ([Na+]e) and [Mg2+]e on [Mg2+]i and [Na+]i in REC. Representative original recordings for [Mg2+]i (top) and [Na+]i (bottom) are shown. After reversing Na+ gradient by exposing cells to Na+-free NMDG solution, [Mg2+]e was increased in a stepwise manner from 0 to 2.5, 5.0, and 7.5 mM. Subsequently, Na+ was readded to bath solution. Tracings are representative of 6 separate experiments.

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Changes of [Mg2+]i and [Na+]i of REC as a function of [Mg2+]e and [Na+]e. Cytosolic Mg2+ and Na+ were measured in Na+-free NMDG medium or in control solution with a [Na+] of 145 mM. [Mg2+]e was increased stepwise from 0 mM to 2.5, 5.0, and 7.5 mM. [Mg2+]i or magnitude of mean [Na+]i change ([Na+]i) after a time period of 100 s is given. Values are means ± SE (shown if larger than symbol size) of 6 experiments.

View larger version (31K): [in this window] [in a new window]   Fig. 8.   Increase of [Mg2+]i in relation to [Mg2+]e. Cytosolic Mg2+ was measured in Na+-free NMDG medium with various [Mg2+] [0 (control), 2.5, 5.0, and 7.5 mM]. Number of single experiments is shown in parentheses. * P < 0.05 vs. control (nominally Mg2+- and Na+-free medium).

View larger version (25K): [in this window] [in a new window]   Fig. 9.   Influence of imipramine and quinidine on the [Mg2+]i change in Na+-free solution with an [Mg2+]e of 5 mM. Number of single experiments is shown in parentheses. ** P < 0.01 vs. control.

Influence of K⁺-rich medium on [Mg²⁺]_i and pH_i. Because it is known that the membrane potential (E_m) acts as a driving force for Mg²⁺ uptake, and since removal of Na⁺ may have hyperpolarized the E_m of REC, E_m was short circuited by increasing the external K⁺ concentration in the next series of experiments.

View larger version (25K): [in this window] [in a new window]   Fig. 10.   [Mg2+]i response to an increase in extracellular K+ concentration ([K+]e). REC were exposed to either control solution (145 mM Na+ and 5 mM K+) or high-K+ medium (15 mM Na+ and 135 mM K+). Values are means ± SE of 4 single experiments. * P < 0.05 vs. control.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

[Mg²⁺]_i under control conditions. The basal [Mg²⁺]_i of REC (0.56 ± 0.14 mM) is within the range of 0.5-1.0 mM that has been reported for other cell types (6, 30, 32).

Modulation of [Mg²⁺]_i and pH_i by HCO₃ and butyrate. The finding that butyrate and HCO₃ increase [Mg²⁺]_i of REC compared with the control value is in agreement with results from in vivo and in vitro studies at the tissue level (23, 27). The in vivo results have been explained partly on the basis of the stimulating effects of SCFA and CO₂ on the blood flow in the rumen wall (27, 33), but such an effect can be excluded under in vitro conditions. This is also true for changes in pH_e, which is known to decrease after supplementation of readily fermentable carbohydrates, thereby increasing Mg²⁺ solubility. Exposure of REC to butyrate- and/or HCO₃-containing solutions leads to an initial decrease of pH_i accompanied by an increase of [Mg²⁺]_i that is independent of the presence of extracellular Mg²⁺ and attributable to the mobilization of Mg²⁺ from intracellular buffering systems. The pH_i-dependent intracellular redistribution of Mg²⁺ is responsible for the higher [Mg²⁺]_i starting levels of REC exposed to butyrate and/or HCO₃. The magnitude of the initial changes in pH_i and hence the evoked increase in [Mg²⁺]_i are reduced when the pH-buffering capacity of the cytosol is increased, as in HCO₃-buffered solutions without butyrate. If mobilization from intracellular stores is the only source of the observed elevation of [Mg²⁺]_i, then it would be expected that [Mg²⁺]_i decreases to control levels on pH_i recovery. This is not the case, and, in addition, it has been shown that 1) with the same level of acidification (by removal of extracellular Na⁺), the increase in [Mg²⁺]_i is significantly higher in Mg²⁺-containing solutions compared with Mg²⁺-free media, 2) exposure of cells to Na⁺ media with 2 mM Mg²⁺ and HCO₃ alone or with both butyrate and HCO₃ increases [Mg²⁺]_i to the same extent (Fig. 2), although the initial pH_i decrease is more pronounced in media with butyrate, and 3) significant differences in the [Mg²⁺]_i of REC exposed to HCO₃ and butyrate compared with those suspended in control solution has been found only in media containing Mg²⁺. Therefore, the modulation of [Mg²⁺]_i by butyrate and HCO₃ is probably associated with the movement of Mg²⁺ across the cell membrane. This raises the question as to what constitutes the transport mechanism.

Role of Na⁺/H⁺ exchange. It has long been known that feeding diets supplemented with easily fermentable carbohydrates and thus increasing the intraruminal concentration of SCFA and CO₂/HCO₃ leads to a stimulation of fluid and electrolyte absorption (10, 13, 27). As in a variety of other epithelia (1, 29), it is well established for the sheep rumen that SCFA (and HCO₃) stimulate Na⁺ transport via Na⁺/H⁺ exchange in the apical membrane (11, 25). Considering the limited cellular supply of H⁺, recirculation of H⁺ is essential for such a system, and SCFA and CO₂ may serve this function. The existence of a Na⁺/H⁺ exchange mechanism in the cell membrane of ruminal epithelial cells has been confirmed by the results of the present study. The finding that ethylisopropylamiloride, amiloride, and cAMP (Fig. 3A) affect basal pH_i suggests that a Na⁺/H⁺ exchanger is also active in "resting" REC at physiological pH_i (6.83 ± 0.1). The Na⁺ potential (E_Na⁺) is larger than the H⁺ potential (E_H⁺), i.e., +54 mV compared with 35 mV, and sufficient to drive the Na⁺/H⁺ exchange, even under basal conditions. As expected, exposure of REC to HCO₃ and HCO₃/ butyrate media leads to an initial decrease of pH_i, resulting from the uptake of nonionized butyrate and/or CO₂. After entry into the cells, the protonated form of butyrate readily dissociates because of the low acidic dissociation constant (pK_a) value (~4.8) of the SCFA, thereby delivering H⁺ to the cell interior. The intracellular hydration of CO₂ supplies not only H⁺ but also HCO₃, and therefore pH_i decreases to a lesser extent after HCO₃ exposure. However, in both cases, there is a substantial elevation of [H⁺]_i coupled with an increased driving force for H⁺ secretion (and Na⁺ uptake), that is, an E_Na⁺ E_H⁺ of +98.4 mV (HCO₃ medium) and +104.0 mV (HCO₃/butyrate medium) compared with +88.6 mV under basal conditions. The acidification activates Na⁺/H⁺ exchange, and it is therefore not surprising that pH_i recovers to near control levels during the experimental period.

Is there an Mg²⁺/2H⁺ exchange in REC? Because Mg²⁺ absorption from the ruminal fluid is also stimulated by SCFA and CO₂/HCO₃, an attractive hypothesis is that such a model may also apply to electroneutral Mg²⁺ uptake and that a Mg²⁺/H⁺ exchange mechanism is present in the apical membrane of ruminal epithelium. To exclude influences of the Na⁺/H⁺ exchanger, experiments were performed in Na⁺-free NMDG media. Under these conditions, the Na⁺/H⁺ exchanger is not capable of extruding protons (E_Na⁺ E_H⁺ = 135 mV) and could even cause proton influx. In comparison, the driving force for an electroneutral Mg²⁺/2H⁺ exchange (E_Mg²⁺ 2E_H⁺) is +159, +180, and +194 mV for an [Mg²⁺]_e of 2.5, 5.0, or 7.5 mM, respectively. Therefore, a Mg²⁺/2H⁺ exchanger, if present in the cell membrane of REC, should work as a H⁺ -exporting and Mg²⁺-importing mechanism and hence result in pH_i recovery accompanied by an increase of [Mg²⁺]_i. However, the results show that cells take up Mg²⁺ but pH_i remains acidic (Fig. 5). The chemical PDs for Mg²⁺ and H⁺ that must be considered the only driving forces for electroneutral ion exchange are optimal under our experimental conditions (2.5-7.5 mM [Mg²⁺]_e >> 0.99-1.15 mM [Mg²⁺]_i; 6.4-6.2 pH_i << 7.4 pH_e). In the physiological situation the chemical gradient for Mg²⁺ (2-4 mM luminal and about 1 mM intracellular) is small, and with normal, low pH values (5.5-6.5) in the ruminal fluid the H⁺ gradient is inwardly, not outwardly, directed. Together, these data and considerations indicate that Mg²⁺ uptake is not directly coupled to an efflux of H⁺ via the proposed Mg²⁺/H⁺ exchanger. Furthermore, under our experimental conditions, we do not have any evidence for the existence of a Na⁺-independent H⁺ extrusion. But it cannot be precluded from the presented results that CO₂/HCO₃ or SCFA (butyrate) stimulate an additional acid extrusion process across the luminal membrane (e.g., an H⁺ pump or H⁺ conductance) that energizes Mg²⁺ uptake, for example by hyperpolarizing the E_m. Another possible explanation for the increase of [Mg²⁺]_i after exposure to butyrate and HCO₃ is a cotransport of Mg²⁺ with anions (e.g., Cl, SCFA, and HCO₃). Because [Mg²⁺]_i increases to the same extent with HCO₃ alone and with HCO₃ and butyrate in the media, and because it is known that SCFA increase the production and secretion of HCO₃ in vivo (9), a Mg²⁺-2HCO₃ cotransporter seems to be a possible candidate for electroneutral Mg²⁺ uptake in REC. A furosemide- and bumetanide-sensitive electroneutral Mg²⁺-HCO₃ cotransport has been shown by Günther et al. (17) in Yoshida ascites tumor cells. Further experiments are necessary to test this hypothesis in REC.

Na⁺-dependent Mg²⁺ uptake. REC showed a pH_i- and K⁺-insensitive increase of [Mg²⁺]_i that seems to be dependent on the Na⁺ gradient across the cell membrane. In cells in nominally Na⁺-free NMDG media, we have demonstrated an elevation of [Mg²⁺]_i accompanied by a decrease in [Na⁺]_i. The magnitude of the observed rise in [Mg²⁺]_i is dependent on the [Mg²⁺]_e, reaching maximal values with 5 mM Mg²⁺ in the medium. Addition of Na⁺ to the bathing solution stops this Mg²⁺ uptake, and, in media containing 145 mM Na⁺, it is difficult to measure any significant change in [Mg²⁺]_i unless [Mg²⁺]_e is 7.5 mM. The interactions between external Na⁺ and [Mg²⁺]_i described here are consistent with the existence of a Na⁺/Mg²⁺ antiporter in REC, a conclusion that is supported by the imipramine and quinidine sensitivity of the Mg²⁺ uptake in Na⁺-free media. A Na⁺/Mg²⁺ antiport has been proposed as the regulator of [Mg²⁺]_i in a variety of cells and is the best-characterized Mg²⁺-transport protein (2, 15). At physiological intra- and extracellular [Na⁺] (145 mM [Na⁺]_e >> 20 mM [Na⁺]_i), a 2Na⁺/Mg²⁺ antiport will obviously mediate Mg²⁺ efflux; this has been confirmed by preliminary experiments with Mg²⁺-loaded (by A-23187) REC, where we have demonstrated a Mg²⁺ efflux in Mg²⁺-free high-Na⁺ media (results not shown). Since J^Mg_ms decreases on reduction of the serosal Na⁺ concentration (21), the exchanger is probably located mainly at the basolateral side of the epithelium. Exposing REC to Na⁺-free media, thus reversing the transmembrane Na⁺ gradient (0.005 mM [Na⁺]_e << 9 mM [Na⁺]_i), changes the driving force (2E_Na⁺ E_Mg²⁺) for an electroneutral 2Na⁺/Mg²⁺ antiport to 430 mV, compared with +83 mV under control conditions. This indicates that the transporter should operate in the reverse mode, thereby increasing the Mg²⁺ influx and Na⁺ efflux. As expected, [Mg²⁺]_i rises (from 0.87 to 1.15 mM) and [Na⁺]_i falls (from 8.9 to 8.36 mM) in the Na⁺-free period. Because of the presence of other transport mechanisms, all of which are active in intact cells, it is impossible to measure the precise stoichiometry of the Na⁺/Mg²⁺ exchanger in these experiments. However, the exchanger is probably PD independent and therefore electroneutral because it also works in high-K⁺ media.