Characterization of an essential disulfide bond associated with the active site of the renal brush-border membrane d-glucose transporter

In a previous report (J. Biol. Chem. 258 (1983) 3565–3570) we have demonstrated that the disulfide-reducing agent dithiothreitol has two effects on the sodium-dependent outer cortical brush border membrane d-glucose transporter; the first results in a reversible increase in the affinity of the transporter for the non-transported competitive inhibitor phlorizin, while the second results in a partially reversible loss of phlorizin binding and glucose-transport activity. Evidence was presented that both of these effects are the result of the reduction of disulfide bonds on the transport molecule. In the present paper we extend our observations on the inactivation of the transporter by dithiothreitol. We provide evidence here (i) that the inactivation of the transporter by dithiothreitol is independent of the effect of the reducing agent on the affinity of the transporter, (ii) that this inactivation process is first-order in dithiothreitol and thus presumably due to the reduction of a single disulfide bond essential to the functioning of the transporter. (iii) that it is the reduction of this disulfide bond and not some subsequent conformational or other change in the transporter which results in its inactivation, (iv) that phlorizin and substrates of the transporter provide protection against inactivation by dithiothreitol and that the degree of protection provided correlates well with the known specificity and phlorizin-binding properties of the transporter, and (iv) that the reactivity of the transporter with dithiothreitol is pH-dependent, decreasing with increasing pH over the pH range 6.5–8.5. We conclude that this site of action of dithiothreitol is a single essential disulfide bond intimately associated with the glucose-binding site on the transport molecule.     

Decreased Na+-gradient-dependent d-glucose transport in brush-border membrane vesicles from rabbits with experimental fanconi syndrome

The effect of anhydro-4-epitetracycline on sodium gradient-dependent d-glucose transport of rabbit renal brush-border membrane vesicles was studied. The purity of isolated brush-border membrane vesicles as judged by enzyme activities was not different between normal control and anhydro-4-epitetracycline-administered rabbits. There was no difference in estimate of intravesicular volume, either. When NaCl was used for sodium gradient, the overshoot of d-glucose uptake into brush-border membrane vesicles isolated from anhydro-4-epitetracycline-treated rabbits was significantly smaller than that of normal control rabbits. In the cases of NaSCN or Na₂SO₄, the former was also smaller than the latter, but not significantly so. To avoid the possible effect of membrane potential on d-glucose uptake, the voltage-clamp method was applied. Even in the voltage-clamped condition, the overshoot of d-glucose uptake into vesicles from anhydro-4-epitetracycline-treated rabbits was decreased compared to that of normal rabbits. In vitro incubation of brush-border membrane vesicles with 20 mM anhydro-4-epitetracycline caused no alteration in sodium gradient-dependent d-glucose uptake. Our results demonstrate that there exists a disorder in sodium gradient-dependent d-glucose uptake of renal brush-border membrane in anhydro-4-epitetracycline-treated rabbits, and suggest that this disorder is one of the underlying mechanisms of experimental Fanconi syndrome.     

Temperature dependence of d-glucose transport in reconstituted liposomes

Sodium-dependent d-glucose uptake into proteoliposomes reconstituted from dimyristoylphosphatidylcholine (DMPC) and hog kidney brush border membrane extract is strongly affected by temperature and the physical state of the membranes. This dependence is defined by a nonlinear Arrhenius plot with a break point at 23°C, a temperature not significantly different from the phase transition temperature of the pure lipid (24°C). The transport process is characterized by different activation energies: 35.1 kcal/mol below and 5.5 kcal/mol above the transition temperature. The shift in the break point for the d-glucose transport activity from 15°C, in the brush border membranes, to 23°C in the reconstituted system leads us to conclude that the lipids surrounding the sodium/d-glucose cotransport system can exchange readily with the bulk lipid used for reconstitution. The results thus provide no evidence for the presence of an annulus of specific lipids surrounding the transport system.     

Peculiarities of the Na+/d-glucose cotransport system in necturus renal tubules

The effects of d-glucose addition to a glucose-free luminal perfusate were investigated in the proximal tubule of Necturus kidney, by electrophysiological techniques. The main findings are: (1) In the presence of sodium, d-glucose produces $\text{10.5}\text{mV}\text{± 1.1}$ (S.E.) depolarization. (2) Phlorizin reduces the magnitude of this response to $\text{2.1 ± 0.1}\text{mV}$. (3) The glucose-evoked depolarization, $\text{ΔV}{}_{\mathrm{<mtext>G</mtext>}}$, does not alter the intracellular K⁺ activity nor is it affected by peritubular addition of ouabain. (4) Isosmotic reduction of Na⁺ concentration in luminal perfusate from 95 to 2 mmol/l (choline or Li⁺ substituting for Na⁺) does not change the magnitude of $\text{ΔV}{}_{\mathrm{<mtext>G</mtext>}}$; complete removal of sodium from the lumen lowers the value of $\text{ΔV}{}_{\mathrm{<mtext>G</mtext>}}$ ($\text{3.2 ± 0.2}\text{mV}$) but the response is not abolished. This observation suggests that the d-glucose carrier of renal tubules in Necturus is poorly specific with regard to the cotransported cation species.     

Phlorizin as a probe of the small-intestinal Na+,d-glucose cotransporter. A model

(1)‘Uptake’ of phlorizin by intestinal brush border membrane vesicles is stimulated, much as that of d-glucose, by the simultaneous presence of Na_out⁺ and $\text{Δψ?0}$. However, phlorizin contrary to d-glucose, fulfills all criteria of a non-translocated ligand (i.e., of a fully competitive inhibitor) of the Na⁺,d-glucose cotransporter. (2) The stoicheiometry of Na⁺/phlorizin binding is 1, as shown by a Hill coefficient of approx. 1 in the Na_out⁺-dependence of phlorizin binding. (3) The preferred order of binding at $\text{Δψ?0}$ is Na⁺ first, phlorizin second (4) The velocity of association of phlorizin to the cotransporter, but not the velocity of its dissociation therefrom, responds to Δψ. These observations while agreeing with the effect of $\text{Δψ?0}$ on the $\text{K}{}_{\mathrm{<mtext>d</mtext>}}$ of phlorizin binding in the steady-state time range, also confirm that the mobile part of the cotransporter bears a negative charge of 1. (5) A model is proposed describing the Na⁺,Δψ-dependent interaction of phlorizin with the cotransporter and agreeing with a more general model of Na⁺,d-glucose cotransport. (6) The $\text{k}{}_{\mathrm{<mtext>on</mtext>}}$, $\text{k}{}_{\mathrm{<mtext>off</mtext>}}$ and $\text{K}{}_{\mathrm{<mtext>d</mtext>}}$ constants of phlorizin interaction with the Na⁺,d-glucose cotransporter are smaller in the kidney than in the small-intestinal brush border membrane, which results in a number of quantitative differences in the overall behaviour of the two systems.     

Mechanism of neomycin stimulation of d-glucose uptake in rabbit intestinal brush border membrane

In order to study the effect of the antibiotic neomycin on the intestinal epithelium, d-glucose was used as a probe molecule and its transport into rabbit brush border membrane vesicles was measured by a rapid filtration method. Treatment of the epithelium with neomycin sulfate prior to the preparation of the brush border membrane enhanced the d-glucose uptake, whereas neutral N-acetylated neomycin did not. This action of neomycin was related to its polycationic character and not to its bactericidal action. No significant difference could be demonstrated between the protein content or disaccharidase-specific activities of the brush border fractions from treated or non-treated intestines. Electrophoretic protein patterns of SDS-solubilized membrane were not significantly different after neomycin treatment. To gain more information on the mechanism involved in the stimulation of d-glucose transport, experiments were conducted on phosphatidyl glycerol artificial membranes and the results compared with those obtained with brush border membrane. At a concentration of 10^?7 M, neomycin decreased the nonactin-induced K⁺ conductance by a factor of approx. 100. The membrane conductance was linearly dependent on the neomycin concentration and the conductance in 10^?2 M KCl was 10 times that in 10^?3 M KCl. The valence of neomycin was estimated, from the slope of these curves, to be between 6 and 4. In contrast, acetylated neomycin had no effect on the nonactin-induced K⁺ membrane conductance. Therefore, the effect of neomycin on artificial membrane is related to its 4 to 6 positive charges. It is proposed that the stimulation of sugar transport in brush border membrane is related to screening of the membrane negative charges by the positively-charged neomycin. Accumulation of anions at the membrane surface then occurs and their diffusion into the intravesicular space would increase the transmembrane potential which, in turn, stimulates the entry of d-glucose.     

The stereospecific d-glucose transport protein in cholate extracts of human erythrocyte membranes Molecular sieve chromatography and estimation of molecular weight

Cholate extracts of human erythrocyte membranes (Lundahl, P., Acevedo, F., Fröman, G. and Phutrakul, S. (1981) Biochim. Biophys. Acta 644, 101–107) were fractionated by molecular sieve chromatography on Sepharose 6B, and the size and molecular weight of the active d-glucose transporter were estimated. The eluent contained 10 or 12.5 mM cholate, since higher concentrations inactivated the glucose transporter, and lower concentrations resulted in aggregation. The chromatographic distribution of the transport activity was reproducible, but was broader than one would expect for a homogeneous component. In the presence of 20 mM EDTA and 5 mM dithioerythritol, a combination which affords a highly stable transport activity, a molecular weight of $\text{400 000 ± 20 000}$ (Stokes' radius 5.9 nm) was estimated for the smallest active component. This value represents an upper limit, since the molecular weight of a non-spherical component would have been overestimated, and since bound cholate was calculated to represent about 12% of the molecular weight. The activity was completely recovered upon rechromatography. In 10 mM EDTA and 10 mM 2-mercaptoethanol, the estimated molecular weight of the smallest active component was $\text{210 000 ± 15 000}$, and this component was not stable upon rechromatography in 10 mM EDTA and 10 mM 2-mercaptoethanol. In the absence of chelating and reducing agents, cholate extracts from membranes which had been kept for 5 days at 4°C showed three additional active components smaller than 200 000 in molecular weight. Most of the phospholipids eluted later than the active components of molecular weight 400 000 or 210 000, in all experiments. Electrophoretic analysis in dodecyl sulfate of the chromatographic eluents indicates that at least one of the band 3-polypeptides (nomenclature according to Steck, T.L. (1974) J. Cell Biol. 62, 1–19) is a component of the active transporter. This band 3-polypeptide, which we denote 3.3, has an apparent molecular weight of 88 000. The stable transporter of molecular weight 400 000 might be a tetramer of the 3.3-polypeptide. Alternatively, a dimer of the 3.3-polypeptide in complex with lipids might account for this molecular weight. If the 3.3-polypeptide is the transporter subunit and if it binds cytochalasin B with high affinity ($\text{1.8 · 10}{}^5$ sites/cell) the recovered activity per 3.3-polypeptide is around 40% A degradation product of the 3.3-component (possibly a 4.5-component) might account for the unstable active transporter of molecular weight 210 000.     

Na+-dependent,electroneutral l-ascorbate transport across brush border membrane vesicles from guinea pig small intestine

In brush border vesicles from guinea pig small intestine l-ascorbate transport is Na⁺-dependent and electroneutral (in the presence of Na⁺, as shown by its lack of response to either positive or negative Δψ across the membrane).l-Ascorbate transporter has the kinetic characteristics of a mobile carrier (K_m for l-ascorbate, 0.3 mM). d-Isoascorbate (erythorbate) seems to be another, but poorer, substrate of the same transporter.l-Ascorbate transport is subjected to heterologous inhibition by d-glucose.     

Studies of the kinetics of Na+ gradient-coupled glucose transport as found in brush-border membrane vesicles from rabbit jejunum

The Na⁺-dependent d-glucose transport reaction in rabbit jejunal brush-border vesicles was studied. Initial rate data were obtained by fitting a polynomial equation to progress curves at different d-glucose concentrations and extracting the slope of the tangent at zero-time. Kinetic replots of the initial rate values produced biphasic Hofstee patterns indicative of two pathways for transport distinguished by their $\text{K}{}_{\mathrm{<mtext>m</mtext>}}$ values for glucose. Neither was dependent on the presence of a membrane potential. Both were dependent on Na⁺ and both were inhibited by phlorizin. Increasing external sodium was found to elevate the apparent $\text{V}{}_{\mathrm{<mtext>max</mtext>}}$ for both pathways. Internal sodium was inhibitory. Pulsed progress curve analysis indicated that the effect of internal sodium was best characterized as carrier sequestration by a sodium-carrier binary complex. Inhibition by internal sodium was completely reversed by the presence, internally, of d-glucose. The presence of two pathways and the kinetic constants for these pathways do not agree with the conclusions of Hopfer and Groseclose (1980) J. Biol. Chem. 255, 4453–4462). Experiments are presented which bear on the reason for the disagreement.     

The stereospecific d-glucose transport activity of cholate extracts from human erythrocyte membranes

The glucose transport protein of human erythrocyte membranes was solubilized with cholate to facilitate rapid reconstitution and direct glucose transport measurements. This may simplify the isolation of the native glucose transporter. In most experiments the membranes were prepared from fresh blood within 8 h, frozen in liquid nitrogen and stored at ?70°C to minimize proteolytic degradation. Solubilization with 25 mM cholate in the presence of 200 mM NaCl at pH 8.4 for 12 min at room temperature gave a high d-glucose transport activity. The solubilized mixture contained 20% of the total membrane protein, only 6% of the polypeptides of molecular weight around 90 000, 23% of the polypeptides of molecular weight around 55 000, 30% of the phospholipids and at least 6% of the stereospecific d-glucose transport activity. At cholate concentrations up to 22 mM the ratio of solubilized phospholipids to cholate increased steeply, concomitant with an increase in solubilized activity. Above 30 mM cholate the activity diminished. At 4°C the activity of the extrac decreased rapidly within the first day and slowly during the next few days. The initial changes seem to have produced a fairly stable, but not native form or fragment of the transporter. When 20 mM EDTA and 5 mM dithioerythritol were included in the solubilization mixture a high activity was preserved for about one day.     

The static head method for determining the charge stoichiometry of coupled transport systems Applications to the sodium-coupled d-glucose transporters of the renal proximal tubule

The static head method for determining the charge stoichiometry (the number of moles of charge translocated per mole of substrate) of a coupled transport system is presented. The method involves establishing experimental conditions under which a membrane potential exactly balances the thermodynamic driving force of a known substrate gradient. The charge stoichiometry can then be calculated from thermodynamic principles. In contrast to the usual steady-state method for determining charge stoichiometry in cell suspensions and vesicle preparations, the static head method is applicable to systems which are not capable of maintaining a constant membrane potential over time. The charge stoichiometries of two renal sodium coupled d-glucose transporters previously identified in brush-border membrane vesicle preparations from the outer cortex (early proximal tubule) and outer medulla (late proximal tubule) are determined. The charge stoichiometries of these transporters are in good agreement with their sodium/glucose coupling ratios arguing against the possibility that glucose transport is coupled to ions other than sodium in these membranes.     

Identification of the d-glucose-inhibitable cytochalasin B binding site as the glucose transporter in rat diaphragm plasma and microsomal membranes

[³H]Cytochalasin B binding and its competitive inhibition by d-glucose have been used to identify the glucose transporter in plasma and microsomal membranes prepared from intact rat diaphragm. Scatchard plot analysis of [³H]cytochalasin B binding yields a binding site with a dissociation constant of roughly 110 nM. Since the inhibition constant of cytochalasin B for d-glucose uptake by diaphragm plasma membranes is similar to this value, this site is identified as the glucose transporter. Plasma membranes prepared from diaphragms bind approx. 17 pmol of cytochalasin B/mg of membrane protein to the d-glucose-inhibitable site. If 280 nM (40 000 μunits/ml) insulin is present during incubation, cytochalasin B binding is increased roughly 2-fold without alteration in the dissociation constant of this site. In addition, membranes in the microsomal fraction contain 21 pmol of d-glucose-inhibitable cytochalasin B binding sites/mg of membrane protein. In the presence of insulin during incubation the number of these sites in the microsomal fraction is decreased to 9 pmol/mg of membrane protein. These results suggest that rat diaphragm contain glucose transporters with characteristics identical to those observed for the rat adipose cell glucose transporter. In addition, insulin stimulates glucose transport in rat diaphragm through a translocation of functionally identical glucose transporters from an intracellular membrane pool to the plasma membrane without an alteration in the characteristics of these sites.     

Synthesis of phlorizin derivatives and their inhibitory effect on the renal sodium/d-glucose cotransport system

To characterize further the Na⁺/d-glucose cotransport system in renal brush border membranes, phlorizin - a potent inhibitor of d-glucose transport - has been chemically modified without affecting the d-glucose moiety or changing the side groups that are essential for the binding of phlorizin to the Na⁺/d-glucose cotransport system. One series of chemical modifications involved the preparation of 3-nitrophlorizin and the subsequent catalytic reduction of the nitro compound to 3-aminophlorizin. From 3-aminophlorizin, 3-bromoacetamido-, 3-dansyl- and 3-azidophlorizin have been synthesized. In another approach, 3′-mercuryphlorizin was obtained by reaction of phlorizin with Hg(II) acetate. The phlorizin derivatives inhibit sodium-dependent but not sodium-independent d-glucose uptake by hog renal brush border membrane vesicles in the following order of potency: 3′-mercuryphlorizin = phlorizin > 3-aminophlorizin > 3-bromoacetamidophlorizin > 3-azidophlorizin > 3-nitrophlorizin > 3-dansylphlorizin. 3-Bromoacetamidophlorizin - a potential affinity label - also inhibits sodium-dependent but not sodium-independent phlorizin binding to brush border membranes. In addition, sodium-dependent phosphate and sodium-dependent alanine uptake are not affected by 3-bromoacetamidophlorizin. The results described above indicate that specific modifications of the phlorizin molecule at the A-ring or B-ring are possible that yield phlorizin derivatives with a high affinity and high specificity for the renal Na⁺/d-glucose cotransport system. Such compounds should be useful in future studies using affinity labeling (3-bromoacetamido- and 3-azidophlorizin) or fluorescent probes (3-dansylphlorizin).     

Small-intestinal Na+/d-glucose cotransport. Inactivation of sugar transport and phlorizin binding by thiol-group and amino-group reagents

It has previously been shown that mercurials acting from the cytoplasmic side or from within the hydrophobic part of the membrane inactivate the small intestinal Na⁺/d-glucose cotransporter by blocking essential SH-groups (Klip, A., Grinstein, S. and Semenza, G. (1979) Biochim. Biophys. Acta 558, 233–245). Another (set of) sulfhydryl(s) which are critical for phlorizin binding and sugar transport function and which may lie on the luminal side of the brush border membrane, can be blocked by DTNB and 4,4′-dithiopyridine but not by $\text{N-}\text{ethylmaleimide}$. In addition, modification of amino groups by fluorescamine, reductive methylation and (under certain conditions) DIDS also lead to inactivation of the carrier's binding and transport functions. No evidence was obtained that any of the above groups is directly involved in the binding of either Na⁺/d-glucose or phlorizin, since none of these compounds prevented inactivation of the cotransporter.     

Specific d-glucose transport in sarcolemma vesicles

The sarcolemmal fraction prepared from rat skeletal muscle consists of osmotically active vesicles that accumulate d-glucose in preference to l-glucose, apparently by facilitated diffusion into intravesicular space. Stereospecific d-glucose uptake by these vesicles is a saturable process, inhibited by phloridzin, by cytochalasin B, and by certain sugars, and enhanced by counterflow. An additional leak pathway permits entry of both d- and l-glucose into the vesicles.Stereospecific d-glucose transport by sarcolemmal vesicles is enhanced to a small extent by insulin, provided the hormone is administered prior to cell disruption. In membranes prepared from insulin-pretreated muscle, Ca²⁺ produces a small further enhancement. Local anesthetics preferentially inhibit stereospecific d-glucose transport. Apparent uptake of both d- and l-glucose is greater when vesicles are suspended in salt solutions rather than sucrose, an effect attributed to increased functional vesicular volume.     

Structural state of the Na+ /d-glucose cotransporter in calf kidney brush-border membranes Target size analysis of Na+-dependent phlorizin binding and Na+-dependent d-glucose transport

Target sizes of the renal sodium-d-glucose cotransport system in brush-border membranes of calf kidney cortex were estimated by radiation inactivation. In brush-border vesicles irradiated at ?50°C with 1.5 MeV electron beams, sodium-dependent phlorizin binding, and Na⁺-dependent d-glucose tracer exchange decreased exponentially with increasing doses of radiation (0.4–4.4 Mrad). Inactivation of phlorizin binding was due to a reduction in the number of high-affinity phlorizin binding sites but not in their affinity. The molecular weight of the Na⁺-dependent phlorizin binding unit was estimated to be 230 000 ± 38 000. From the tracer exchange experiments a molecular weight of 345 000 ± 24 500 was calculated for the d-glucose transport unit. The validity of these target size measurements was established by concomitant measurements of two brush-border enzymes, alkaline phosphatase and γ-glutamyltransferase, whose target sizes were found to be 68 570 ± 2670 and 73 500 ± 2270, respectively. These findings provide further evidence for the assumption that the sodium-d-glucose cotransport system is a multimeric structure, in which distinct complexes are responsible for phlorizin binding and d-glucose translocation.     

Characteristics of dipeptide transport in normal and papain-treated brush border membrane vesicles from mouse intestine I. Uptake of glycyl-l-phenylalanine

Papain treatment of isolated brush border membrane vesicles was carried out to study peptide transport in the absence of hydrolytic events associated with the brush border membrane. Such a treatment allowed a 70% decrease in the activity of membrane-associated oligopeptidases and the study of peptide transport in the complete absence of free amino acids up to 1 min of incubation. A comparison between the time course curves of glycyl-l-phenylalanine uptake by normal and papain-treated vesicles showed that the overshoots seen in the presence of Na⁺ and K⁺ gradients (extravesicular intravesicular) when using normal vesicles were no longer evident after papain treatment. This result, together with the demonstration of uptake into an osmotically reactive intravesicular space and the analysis of uptake of free phenylalanine, allowed the coclusion that peptide transport was the result of two complementary mechanisms, uptake of free amino acids following hydrolysis by the membrane-bound oligopeptidases, and intact peptide transport down a concentration gradient by a non-Na⁺ (and non-K⁺)-dependent process. These results also showed the non-involvement of γ-glutamyltransferase and the γ-glutamyl cycle in peptide absorption. A linear relationship has been established between initial dipeptide uptake and glycyl-l-phenylalanine concentration for the intact peptide transport process. However, this process can be inhibited to various extents by other di- and tripeptides but the inhibition never exceeded 43%. These results are consistent with both passive and facilitated diffusion mechanisms of intact peptide transport, the latter occuring by either a low affinity-high capacity or a high affinity-low capacity system.     

Effects of various conditions on the alkaline degradation of d-fructose and d-glucose

Dilute solutions of d-fructose and d-glucose undergo alkaline degradation, and, at temperatures in the range of 30–70°, almost two moles of alkali are consumed per mole of the carbohydrate. The degradation is partly guided by the dielectric constant of the medium; such additives as acetone and urea have specific effects where the reactions are not essentially guided by the medium dielectric. Acetone and urea presumably form complexes with the carbohydrates; this is revealed for the former by the formation of a dark red solution having a spectral band at 320 nm, like that observed earlier in the presence of ethylenediamine.     

Nature of the uptake of d-galactose, d-glucose and α-methyl-d-glucoside by Saccharomyces cerevisiae

Both glucose-grown baker's yeast after induction and galactose-grown yeast appear to take up d-galactose by a system not requiring phosphorylation and only up to a diffusion equilibrium, as shown by pulse labelling, sampling at very short intervals and chromatographic analysis of extracts. Part of the sugar taken up is transformed into trehalose which is present in substantially greater amounts in cells than the transported sugar itself. The effect of 2,4-dinitrophenol and of iodoacetamide, as well as the nature of the efflux of sugars from preloaded cells, support the results. d-Glucose and α-methylglucoside are also taken up without phosphorylation.     

The effect of areneboronic acids on the alkaline conversion of d-glucose into d-fructose

Benzeneboronic acid, 4-methoxybenzeneboronic acid, 3-nitrobenzeneboronic acid, and sulphonated benzeneboronic acid have been used to displace the pseudo-equilibria established in aqueous alkali between d-glucose, d-fructose, and d-mannose to give greatly increased yields of d-fructose. The effect of reaction temperature, pH, overall concentration, and molar ratio of acid:sugar on the yield of d-fructose has been investigated by using an automated assay for d-fructose.