Apparent inhibition of active non-electrolyte transport by an increased sodium permeability of the plasma membrane. Mechanism of action of p-chloromercuribenzene sulfonate.

Sodium-dependent amino acid and sugar uptake by intact cells and tissues is reduced by p-chloromercuribenzene sulfonate. This inhibition is believed to be specific for neutral amino acids as the uptake of other amino acids and sugars is affected to a lesser extent. Since p-chloromercuribenzene sulfonate is known to increase the sodium conductance of biological membranes, the inhibition of amino acid transport by jejunal brush border membrane vesicles was examined to determine if the inhibition was at the carrier level or involved the electrochemical sodium gradient. The uptake rate of L-valine is the same in control and treated vesicles. However, the sodium gradient energized overshoot of L-valine is eliminated by pretreatment with p-chloromercuribenzene sulfonate. Since the overshoot of D-glucose is also eliminated, the effect of p-chloromercuribenzene sulfonate is due to an increased dissipation of the sodium gradient which energizes sodium-dependent solute transport. Treatment of membrane vesicles with reagent grade p-chloromercuribenzene sulfonate abolishes the activity of sucrase and reduces the intravesicular space accessible to D-glucose or L-valine. The former effect is due to contaminating mercury. The latter effect, and the effect on the sodium gradient-driven overshoot, are also caused by purified inhibitor and can be reversed by subsequent treatment of the vesicles with dithiothreitol. These effects are due to the sulfhydryl actions of this compound.

Sodium-dependent amino acid and sugar uptake by intact cells and tissues is reduced by p-chloromercuribenzene sulfonate. This inhibition is believed to be specific for neutral amino acids as the uptake of other amino acids and sugars is affected to a lesser extent. Since p-chloromercuribenzene sulfonate is known to increase the sodium conductance of biological membranes, the inhibition of amino acid transport by jejunal brush border membrane vesicles was examined to determine if the inhibition was at the carrier level or involved the electrochemical sodium gradient. The uptake rate of L-valine is the same in control and treated vesicles.
However, the sodium gradient energized overshoot of L-valine is eliminated by pretreatment with p-chloromercuribenzene sulfonate.
Since the overshoot of D-glucose is also eliminated, the effect of p-chloromercuribenzene sulfonate is due to an increased dissipation of the sodium gradient which energizes sodium-dependent solute transport. Treatment of membrane vesicles with reagent grade p-chloromercuribenzene sulfonate abolishes the activity of sucrase and reduces the intravesicular space accessible to D-glucose or L-valine. The former effect is due to contaminating mercury. The latter effect, and the effect on the sodium gradient-driven overshoot, are also caused by purified inhibitor and can be reversed by subsequent treatment of the vesicles with dithiothreitol.
These effects are due to the sulfhydryl actions of this compound. Sulfhydryl reagents have long been known to influence ion, water, and non-electrolyte uptake by various cells and tissues (l-6).
Many sulfhydryl reagents, such as N-ethylmaleimide or  2,3,5,7,9). As the inhibition bypCMBS of amino acid uptake by mouse embryo cells and by the rabbit small intestinal epithelial cells appeared to be specific for certain amino acids, e.g. phenylalanine and alanine, it was concluded that in these cells pCMBS was a specific inhibitor of the sodium-dependent neutral amino acid transport system (10-12).
The objective of this study was to examine more closely the effects of pCMBS on neutral amino acid transport using isolated intestinal brush border membrane vesicles. The use of plasma membrane vesicles is an accepted method of studying amino acid transport (13,14) and enables transport to be studied under a variety of conditions not feasible with intact cells or tissues (15). The results indicate that pCMBS does not specifically interact with the neutral amino acid carrier, but increases the sodium conductance of the membrane. The latter effect of pCMBS is well known (2, 3); however, the implications relating to non-electrolyte transport are not widely understood or well documented. The increased sodium conductance provides a pathway for uncoupled sodium flow which lowers the driving force for all solutes absorbed via sodium-coupled transport systems. The presence of a sodium conductance pathway thus results in an apparent inhibition of transport energized by an electrochemical sodium gradient. A preliminary report of these results has been published ( 16).

RESULTS
Sodium-dependent amino acid or glucose uptake by intact cells or membrane vesicles is driven by an electrochemical sodium gradient (13,14,23,(27)(28)(29)(30)(31)(32)(33). Pretreatment of brush border membranes with as little as 0.1 mM pCMBS results in an inhibition of concentrative non-electrolyte transport energized by a sodium thiocyanate or sodium chloride gradient. This is illustrated in Fig. 1 by a lack of the "overshooting" part of the amino acid or glucose uptake. These observations are similar to those of pCMBS effects with intact epithelia (10).
To determine whether the inhibition of energized glucose and valine transport is due to an inhibition of the carriers or to a faster dissipation of the energy, sodium-dependent solute transport was measured under equilibrating conditions in the absence of any driving force (see "Experimental Procedures"). The results in Table I  inhibit the rate of L-valine uptake by rat or rabbit brush border membrane vesicles. Similar results were obtained with D-glucose (Table I). In both control and pCMBS-pretreated membranes, L-valine was transported by a carrier-mediated process as shown by the inhibitory effect of 70 mM L-methionine (Table II). A similar argument can be made for D-glUCOSe since 0.27 mM phlorizin inhibited the rate of D-glucose uptake by more than 65% in pCMBS-pretreated rat membranes. Carrier activity can also be assessed by countertransport or tracer exchange experiments.
As shown by the open triangles in Fig. 2, A and B, preloading of membrane vesicles with 40 mM L-valine can be used to drive the transport of labeled Dglucose or labeled L-valine against their respective concentration gradients (the latter in terms of isotope concentration).
This ability is expressed in Fig. 2, A and B as "overshooting" uptake. In contrast, loading the vesicles with fructose, which is transported by a sodium-independent system (34), does not stimulate L-valine uptake ( Fig. 2A, open circles). Pretreatment of membranes with pCMBS abolishes the "overshoot" for D-glucose but not for labeled L-valine (Fig. 2, A and B, filled triangles).
The differential effects of pCMBS on Dglucose and L-valine uptake can be explained on the basis of the different nature of the coupling between L-valine efflux and the uptake of labeled D-glUCOSe and labeled L-valine. Countertransport between neutral amino acids and D-glUCOS2 has been shown in intestinal membranes to result from the sodium and electrical charge movements coupled to non-electrolyte fluxes and the requirement for overall electroneutrality (14). This type of interaction is sensitive to the presence of other parallel conductance pathways in the membrane, especially those for sodium. For example, the overshoot of labeled D-glucose driven by unlabeled L-valine efflux is abolished by monactin (Fig. 2B, open squares and Ref. 14). In contrast, the overshooting tracer exchange in Fig. 2A is due to carriermediated coupling of unlabeled and labeled L-valine fluxes  (35). These results (Figs. 1 and 2) demonstrate the lack of a direct inhibitory effect of pCMBS on the amino acid carrier and are consistent with a pCMBS-dependent increase in the sodium conductance of the membrane. Since the amino acid and glucose transport systems are not inhibited by pCMBS, the inhibition of energized non-electrolyte transport is probably due to a faster dissipation of any electrochemical sodium gradient in the pCMBS-pretreated membranes. The observed inhibition pattern ofpCMBS is the same as that produced by the addition of monactin, an ionophore which increases the cation conductance of membranes (14, 32, 36). Therefore, the results in Table I and Figs. 1 and 2 strongly suggest that pretreatment of membranes with pCMBS increases the sodium conductance of the small intestinal brush border membrane. In agreement with this conclusion is the finding that the sodium chloride permeability is increased by pCMBS pretreatment ( Table I). The effect of pCMBS on energized solute transport across the brush border membrane is qualitatively similar for both rat and rabbit membranes (Fig. l), although the rat membranes are more sensitive than the rabbit membranes (Fig. 3).
In addition to the effect of pCMBS on energized transport, pCMBS pretreatment at higher concentrations (21 IDM) decreases the intravesicular space accessible to L-valine and Dglucose (Fig. 4, D-glucose results not shown). The space in control and pCMBS-treated membranes is the same for both  spherical shape) dropped to 72% of the control vesicles (from 0.0022 pm" to 0.0016 pm") after treatment with 1 mM &MBS. However, this cannot account for the change in the equilibrium space, as for the same membrane preparation the glucose accessible volume of the pCMBS-pretreated vesicles was 29% of the control vesicles. Since dithiothreitol at 1 IIIM reversed the pCMBS effect (Fig. 6) changes in vesicle shape, such as from spherical to oblate spheroid, are the most likely explanation for the equilibrium effect. The loss of the sodium gradient-energized overshoot (Fig. 1)  L-valine and n-glucose and does not change from 15 to 60 min after the addition of the solute (results not shown). The rat membranes are more sensitive to this equilibrium space effect ofpCMBS than the rabbit membranes (Fig. 4). The observed decrease in equilibrium space per mg of protein can arise from fragmentation of vesicles, changes in vesicle shape to yield higher surface area to volume ratios (e.g. sphere to ellipsoid at fixed surface area), or conversion of membrane vesicles to membrane sheets. A small decrease in the average vesicle size was detected in electron micrographs of negatively stained vesicles (Fig. 5). The calculated mean volume (assuming effect which reduces the equilibration space because the concentrations required for the two effects are very different (see Figs. 3 and 4). For example, pretreatment of rat membranes with 0.1 mM pCMBS abolishes the sodium gradient-driven L-valine overshoot (Fig. 3) while the equilibration space is reduced by only 20% (Fig. 4).
In preliminary experiments, the sucrase activity of brush border membranes treated with pCMBS was abolished with the rat enzyme more sensitive than the rabbit enzyme (   FIG. 7. Effect of pCMBS on the sucrase activity of brush border membrane vesicles. Purified membranes were treated with pCMBS and washed, and the sucrase activity was measured as described in "Experimental Procedures." The control membranes had a specific activity of 1.44 units/mg of protein for the rat and 0.80 units/mg of protein for the rabbit. Untreated pCMBS, 0, A; purified pCMBS, 0, A, rabbit membranes, A, A; rat membranes, 0,O. 7). Sucrase does not require a sulfhydryl group for activity; therefore, we suspected that the inhibition was due to the presence of trace amounts of inorganic mercury in the pCMBS, as sucrase is very sensitive to mercury (37). Subsequent analysis demonstrated that -0.5% (by weight) of the pCMBS was mercury. Contrary to previous results (l), pCMBS is not unstable in solution, the mercury content of pCMBS solutions was the same as that of the anhydrous salt (results not shown). Solutions of pCMBS freed of mercury (see "Experimental Procedures") did not inhibit sucrase (Fig.  7); however, the equilibration space for L-valine was still reduced (Fig. 4, open symbols) and the sodium thiocyanate driven L-valine or D-ghCOSe overshoot was still abolished (Fig.  1). The latter two effects are evidently not due to mercury contamination of the pCMBS. The effects of pCMBS on the equilibration space for Lvaline and D-ghCOSe can be completely reversed by treatment with dithiothreitol ( Fig. 6; L-valine not shown). The sodium thiocyanate-driven D-glucose overshoot was partially restored by dithiothreitol (Fig. 6). Both effects are therefore due to the sulfhydryl blocking action of pCMBS.

DISCUSSION
The failure of high concentrations of pCMBS to inhibit the rate of nonenergized L-valine or D-glucose uptake by brush border membrane vesicles indicates thatpCMBS has no direct effect on the carriers (Table 1). Both solutes are transported by carrier-mediated systems as shown by sensitivity of the Lvaline uptake to L-methionine (Table II and Ref. 13), and by sensitivity of D-glucose uptake to phlorizin (23, 35). These results conflict with reports that pCMBS inhibits the uptake of glucose analogues (12, 38) and amino acids via sodiumdependent transport systems (10)(11)(12)38). In all of the latter studies, solute uptake was measured into intact cells or tissues, therefore, other effects of pCMBS, such as a decrease in the energy supply for transport may have produced the apparent inhibition of transport (16). It should be stressed that our results on the sodium-dependent glucose transport do not have any bearing on the sodium-independent glucose transport system for which inhibition by sulfhydryl reagents is well established (1, 39).
It is well known that pCMBS increases the cation permeability of the plasma membrane (2, 3, 7, 8); indeed, pCMBS is often used to load cells with sodium (2, 40, 41). Detailed studies have established thatpCMBS-induced sodium uptake occurs by a conductance pathway (3,42,43). Under energized steady state conditions, such as exists with intact cells or tissues, an electrochemical sodium gradient (greatest sodium concentration outside the cell and a negative electrical potential inside the cell) is present across the plasma membrane. This gradient accounts for "active" accumulation of amino acids and glucose which are co-transported with sodium (14, 27-33). In the vesicle system, an increased sodium conductance would reduce or abolish any experimentally established sodium gradient and would result in an inhibition of solute flux, often erroneously equated with an inhibition of the transport system. In an intact viable cell, thepCMBS-induced sodium conductance would allow a dissipative sodium flux, which would constitute an additional "load" on the sodium pump (the (Na' + K')-ATPase) and the metabolic reserves of the cell. The additional sodium flux induced by pCMBS may or may not produce a significant reduction of the electrochemical sodium gradient depending upon the total sodium and charge fluxes across the plasma membrane, and whether the reserve capacity of the sodium pump can handle the additional fluxes. Any change in the electrochemical sodium gradient would be reflected in the sodium-dependent nonelectrolyte uptake rates. Therefore, it is conceivable that pCMBS could produce an apparently specific inhibition of amino acid transport (10-12). If the pCMBS-independent sodium fluxes (e.g. those associated with 30methylglucose uptake in the small intestine) are not high, the additional pCMBS-induced load could be handled by the sodium pump. In contrast, in the presence of already high non-electrolytecoupled sodium fluxes (e.g. in the presence of amino acids or D-galactose), the pCMBS-induced sodium flux could overload the pump, which would be expressed as an inhibition of sodium-dependent non-electrolyte transport. Unfortunately, it is experimentally not possible to circumvent the "load" problem in the intact tissue or cells by using "initial" uptake rates. The depolarization of the plasma membrane potential due to sodium fluxes via any type of conductance pathway occurs within seconds and before non-electrolyte fluxes could be measured experimentally (44,45). The apparent specificity of the pCMBS effect for certain amino acid transport systems in intact cells or tissues (10-12) may simply result from a specific dependency of the kinetic parameters on the electrical field and sodium concentrations.
This dependency may differ for the various sodium-dependent transport systems (46). The effect of pCMBS on the equilibrium space for either D-&COSe or L-vahne is thought to represent a different effect from that causing an increased sodium permeability as much higher concentrations are required (see Figs. 3 and 4). The decreased space is not caused by fragmentation of the vesicles because pCMBS did not sufficiently affect the vesicles' size, estimated from negatively stained images, to account for the large decrease in the intravesicular space ( Fig. 5 and see "Results"), and because the space can be restored to that of the control by treating the vesicles with dithiothreitol ( Fig.   6).
Membrane vesicles probably behave in suspension as loose sacs with the membrane "sides" free to move. The solute accessible intravesicular space may be reduced by pCMBSinduced cross-linking of opposing pieces of membrane, perhaps through residual microfilament attachment sites. The effects ofpCMBS on sodium-energized solute transport and the intravesicular space accessible to solutes are due to pCMBS and not to contaminant mercury. However, the sensitivity of the membrane-bound sucrase to traces of mercury demonstrates that studies involvingpCMBS should be undertaken with considerable care as mercury is a well known inhibitor of many enzymes and transport systems (1, 9, 47, 22. 23.