Sodium-induced Conformational Changes in the Glucose Transporter of Intestinal Brush Borders*

Using brush-border membranes prepared from rab- bit small intestine by Ca2+ precipitation and KSCN treatment, we have studied the kinetics and confor- mational changes of the glucose carrier. Na+ behaves as a competitive activator of glucose transport under zero-trans conditions. Phenyl isothiocyanate and fluorescein isothiocyanate (FITC) inhibit Na+-dependent transport in an irreversible but substrate-protectable manner. Vesicles pretreated with phenyl isothiocyanate in the presence of substrates were then selectively labeled at the glucose carrier with FITC. Competition experiments with Na+ and phlorizin or glucose indi- cated that FITC binds to the glucose site on the carrier. Carrier-bound FITC displays a saturable quenching of fluorescence in the presence of Na'. The of the Na+-specific quench is 25 mM, which is similar to the apparent Km for Na+ activation of glucose transport. Two tyrosine group-specific reagents, N-acetylimida-zole and tetranitromethane, inhibit glucose uptake and fluorescent quenching in a Na+-protectable fashion. FITC labeled a 75-kilodalton peptide on sodium dode-cy1 sulfate-polyacrylamide gel electrophoresis in a sub- strate-sensitive manner. We conclude that Na+ binds to the glucose symporter of intestinal brush borders, a 75-kilodalton peptide, and this induces a rapid conformation change in the transporter which increases its affinity for D-glucose.

Two tyrosine group-specific reagents, N-acetylimidazole and tetranitromethane, inhibit glucose uptake and fluorescent quenching in a Na+-protectable fashion. FITC labeled a 75-kilodalton peptide on sodium dode-cy1 sulfate-polyacrylamide gel electrophoresis in a substrate-sensitive manner. We conclude that Na+ binds to the glucose symporter of intestinal brush borders, a 75-kilodalton peptide, and this induces a rapid conformation change in the transporter which increases its affinity for D-glucose.
Entry of glucose into intestinal and renal epithelial cells occurs as the result of coupling to Na' transport down its electrochemical potential gradient by a cotransporter in the brush-border membrane. Kinetic studies of the cotransporter indicate that the carrier has an Ordered Bi Bi reaction mechanism (1-3) with Na' the obligate first substrate (2). This is consistent with an increase in the apparent carrier affinity (K,) for glucose with Na' (2). Information concerning the molecular events involved in Na' alteration of carrier glucose affinity is incomplete, although we have recently demonstrated a Na+-dependent fluorescence quench of fluorescein isothiocyanate bound to the glucose carrier (4).
In this paper, we extend our earlier findings with FITC' and present results which confirm that FITC binds to the glucose site and that the Na+-induced fluorescence quenching * This work was supported by Grants AM 19567 and NS 09666 from the United States Public Health Service. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
is the result of interaction with the Na' site on the glucose carrier. We also demonstrate that the Na+-dependent fluorescence quenching is dependent on a sulfhydryl group somewhat removed from the Na' and glucose-active sites.

EXPERIMENTAL PROCEDURES
Preparation of Membranes-Ca'+-precipitated brush-border membrane vesicles were isolated from rabbit small intestine by the method described by Stevens et al. (5). The Ca2+-precipitated vesicles were further purified by a KSCN procedure (4) and were stored at -180 "C until use. Briefly, the Ca'+-precipitated vesicles were exposed to 0.6 M KSCN for 10 min on ice, then diluted 20-fold with 0.3 M mannitol + 10 mM HEPES/Tris, pH 7.5, and centrifuged for 10 min at 6,000 X g. These supernatants were centrifuged for 30 min at 38,000 X g, and the resulting pellets were harvested. The pellets were washed and resuspended in 300 mM mannitol + 10 mM HEPES/Tris, pH 7.5, and pelleted again. This was repeated twice. The membranes were routinely 7o-fold enriched in alkaline phosphatase and 130-fold enriched in y-glutamyltranspeptidase and sucrase compared to the initial homogenate (4).
Glucose Transport-Glucose uptake was measured by a rapid quench, rapid filtration method as described by Stevens et al. (5). Na+-dependent glucose uptake (nanomoles mg" s-I) is defined as the uptake seen with 100 mM cis-NaC1 minus uptake seen with 100 mM cis-KC1. All uptakes were measured at 3 s a t room temperature (22 'C) using 100 pg of vesicle protein.
Phlorizin Binding-[3H]Phlorizin binding was performed according to the method of Toggenburger et al. (6) at room temperature and pH 7.5. Na-dependent phlorizin binding is defined as the total phlorizin bound in the presence of 100 mM NaCl minus the phlorizin bound in the presence of 100 mM KCl. Preliminary experiments indicated that phlorizin binding attained a steady state within 3 s, and all results presented were obtained from 3-s incubations with a phlorizin concentration (20 @I ) well above the KD. PITC and FITC Treatment-Vesicles to be studied by fluorescence were pretreated with 2 mM phenyl isothiocyanate in the presence of 100 mM NaCl and 10 mM glucose, and 50 mM Tris-C1, pH 9.2, + 2 mM EDTA for 30 min at room temperature. Exposure of vesicles to FITC or PITC at basic pH favors binding to the t-amino group of lysine. The reaction was stopped by the addition of a 10-fold excess of ice-cold 50 mM Tris-C1, pH 9.2, + 2 mM EDTA and centrifuged for 30 min at 38,000 X g. The pellets were resuspended in a minimum volume of 300 mM mannitol + 10 mM HEPES/Tris, pH 7.5.
Vesicles were then exposed to 50 ,.IM FITC (dissolved in dimethylformamide) in 50 mM Tris-CI, pH 9.2, and 2 mM EDTA for 15 min at 22 "C in the dark. The reaction was stopped by the addition of a 20-fold excess of ice-cold 50 mM Tris-CI, pH 9.2, and 2 mM EDTA, and the mixture was centrifuged for 30 min at 38,000 X g. The pellets were resuspended in 300 mM mannitol and 10 mM HEPES/Tris, pH 7.5.
Studies examining the ability of substrates or inhibitors to protect against FITC inhibition of transport or FITC binding were performed with vesicles pretreated with PITC and substrates. The appropriate concentration of substrates was added during exposure to FITC. Phlorizin, pCMBS, and N-acetylimidazole were added to vesicles 30-60 min prior to FITC treatment in the presence or absence of NaC1. Tetranitromethane was added to vesicles 30 min prior to the addition of FITC in 300 mM mannitol and 10 mM phosphate buffer, pH 8.8. All treatments were performed at 22 "C. For transport studies, reversal of phlorizin inhibition was accomplished by washing, pelleting, and resuspending the pellets twice in 300 mM mannitol and 10 mM HEPES/Tris, pH 7.5. For reversal of pCMBS inhibition, 1 mM dithiothreitol was added to the wash.
Fluorescence-Fluorescence experiments were performed on an Aminco-SLM SPF 500 spectrofluorometer at room temperature set in the ratio mode. The excitation and emission wavelengths were 495 and 525 nm, with slit widths of 2 nm. All fluorescence quench results were corrected for dilution by addition of an equal volume of either buffer or KC1 to the reference cuvette containing an equal amount of substrate-protected FITC-labeled vesicles and are reported as uncorrected fluorescence emission spectra.
SDS-Polyacrylamide Gel Electrophoresis-SDS-polyacrylamide gel electrophoresis was performed on 10-15% linear gradient slab gels or 10% slab gels according to the method of Laemmli (7). Samples were solubilized in 1% SDS in 10 mM Tris-C1, pH 6.8 + 0.4% P-mercaptoethanol. The suspension was then centrifuged for 120 min at 60,000 X g, and the supernatant was collected. Following electrophoresis, 0.3-cm slices of the FITC-labeled tracks were mashed and suspended in 2.5% SDS and 10 mM Tris-C1, pH 9.2, + 1 mM EDTA overnight with agitation in the dark at room temperature. The acrylamide was removed by centrifugation and the supernatants collected. FITC in the slices was determined by fluorescence, and protein was determined after Coomassie Blue-staining by absorbance measurements at 550 nm. Molecular weight standards obtained from Bio-Rad were run in parallel.
Enzyme Assays-Brush-border marker enzymes were routinely used to monitor vesicle purification. Alkaline phosphatase was assayed according to the method of Mircheff and Wright (8); sucrase was measured by the method of Dahlqvist (9), and y-glutamyltranspeptidase according to the method of Naftalin et al. (10). Protein was measured using the Bio-Rad protein assay with y-globulin as the protein standard.
Materials-FITC was purchased from Molecular Probes, Inc., Junction City, OR. N-Acetylimidazole and PITC were purchased from Sigma. pCMBS and phlorizin were purchased from Calbiochem-Behring. Electrophoresis chemicals were purchased from Bio-Rad. PITC and FITC were dissolved in dimethylformamide which was purchased from Aldrich. All other chemicals were reagent grade or better and purchased from American Scientific Products, Los Angeles, CA.

RESULTS
Brush-border membrane vesicles prepared by a divalent metal precipitation procedure ( 5 ) are routinely 30-fold enriched in alkaline phosphatase and 40-fold enriched in sucrase above that found in the total homogenate. Our adaptation of the KSCN treatment procedure of Hopfer et al. (11) resulted in a 3-4-fold further enrichment of the brush-border enzyme markers alkaline phosphatase, sucrase, and glutamyltranspeptidase, as well as an increase in the Na+-dependent transport velocity (4) and an increase in the Na+-dependent phlorizin binding. Following KSCN treatment, Na+-dependent phlorizin binding increased 9-fold, and the maximum velocity of Na+-dependent glucose uptake increased 11-fold (Table I).
Kinetics-Thiocyanate-treated brush-border membrane vesicles contain a saturable Na+-dependent carrier and a Na+-TABLE I Comparison of Caz+-precipitated brush-border membranes with and without KSCN treatment 100-pg vesicles were assayed for Na+-dependent glucose uptake and Na+-dependent phlorizin binding as described under "Experimental Procedures," and Figs. 1 and 5. All data are given as the mean 2 S.E. with the number of estimates in parentheses.  independent glucose uptake system. The apparent affinity of the Na+-dependent uptake (K,) was 85 p~, similar to the K , of Ca2+-precipitated vesicles (2). The Vmax of Na+-dependent uptake was 10 times that previously reported (0.1 nmol/mg/ s) for Ca2+-precipitated vesicles. Fig. 1 is a Woolf-Augustinsson-Hofstee plot of the effect of Na+ on carrier glucose affinity at three sodium concentrations. As the Na+ concentration increased from 25 to 100 mM, the glucose K, decreased from 200 to 79 FM. Increasing the glucose concentration from 40 to 500 p~ did not affect the Na' affinity: the apparent for Na+ activation of glucose uptake was 34 mM at 40 p M glucose and 36 mM at 500 FM glucose (data not shown). Hill plots gave a slope of 1, indicating a single class of noninteracting Na' sites.
PITC and FITC Inhibition of Transport- Fig. 2a demonstrates that PITC inhibits Na+-dependent glucose uptake. Incubation for 30 min results in inhibition of Na+-dependent glucose uptake with no effect on Na+-independent uptake. Inhibition is dependent on the concentration of PITC and reaches a maximum of 65% at 2 mM. The addition of Na+ and D-glucose during exposure to PITC protects the Na+dependent glucose uptake. Fig. 2b shows the Na+-dependent uptake following exposure for 30 min at room temperature to 2 mM PITC and substrates. In the presence of 10 mM glucose and 100 mM NaC1, PITC inhibition is less than 2% of that seen in the absence of Na+ and glucose. Substitution of K' for Na+ or L-glucose for D-ghCoSe results in no protection of transport. Similar results were obtained with FITC. Kinetics of Na+-dependent glucose uptake as a function of cis-Na. Vesicles (100 pg) in 300 mM mannitol and 10 mM Hepes/Tris, pH 7.5, were added to uptake buffer which included 25, 50, or 100 mM NaCI, 10 mM Hepes/Tri, pH 7.5, 50 pM ~- [ 6 -% ] glucose. The D-glucose concentration was varied between 25 PM and 10 mM, and the osmolarity was maintained with mannitol. Na+independent uptake was determined by substitution of KC1 for NaCl in the uptake buffer and has been subtracted out. Uptake was terminated following a 3-s incubation. The results (means of duplicates) are presented as a Woolf-Augustinsson-Hofstee plot where uptakes (J) are plotted against uptake/glucose concentrations. The plot is representative of three separate experiments. Isothiocyanate inhibition of glucose uptake. a, vesicles (500 pg) were treated with varied concentrations of PITC in 50 mM Tris-C1, pH 9.2, and 2 mM EDTA for 30 min at 22 "C. Unreacted PITC was removed by centrifugation, and Na+-dependent glucose uptake (25 pM) was determined as described under "Experimental Procedures" and plotted against PITC concentration. Results are the means of two samples, and the plot is representative of three separate experiments. b, vesicles (500 pm) were treated with 2 mM PITC in 50 mM Tris-C1, pH 9.2, and 2 mM EDTA in the presence of 100 mM NaCl or 100 mM KC1 and 0-10 mM D-or L-glucose for 30 min at room temperature. Substrates and unreacted PITC were removed by centrifugation, and Na+-dependent glucose uptake (25 p~) was determined under "Experimental Procedures." Results are the means of two samples, and the plot is representative of three separate experiments. Uptake is plotted against the glucose concentrations in the preincubation media. h L j I

FIG. 3. Effect of PITC pretreatment on FITC inhibition of
Na+-dependent glucose uptake and FITC binding. Vesicles (500 pg) were treated with 50 p~ FITC or 2 mM PITC in the presence or absence of 10 mM glucose and 100 mM NaCl in 50 mM Tris-C1, pH 9.2, and 2 mM EDTA at 22 "C for 15 min (FITC) or 30 min (PITC). Unreacted isothiocyanate derivative was removed hy centrifugation, and a portion of the PITC-treated membranes was subsequently exposed to FITC for 15 min in the dark at 22 "C. Na+-dependent glucose uptake (25 p M ) was assayed at 22  Identification of FITC-binding Site-Phlorizin andpCMBS were examined for their ability to protect against FITC inhi-bition of Na+-dependent glucose uptake and ability to decrease carrier-specific FITC binding. The isothiocyanate derivatives are irreversible inhibitors, as judged by the lack of reversal upon repetitive washing (not shown). In contrast, both pCMBS and phlorizin inhibition are reversible (12, 13). Fig. 4 shows that phlorizin, whose interaction with the carrier is thought to be at the glucose site (6), gives 95% protection in a Na+-dependent manner against FITC inhibition of transport and reduces FITC binding by 1.2 nmol. In contrast, pCMBS affords no protection against FITC inhibition of transport and does not significantly affect FITC binding to the vesicles. This result is in agreement with Semenza (14; who reported that the essential sulfhydryl group cannot bt protected by substrates. An estimate of the number of FITC sites related to the carrier glucose site may be determined bs competition between phlorizin and FITC. Fig. 5 illustrates the results of these experiments in which vesicles were titrated with various concentrations of the irreversible inhibitor FITC and then Na+-dependent phlorizin binding was determined. The results indicate that 242 pmol of phlorizin/mg of protein bind to the carrier in a FITC-sensitive manner. The FITCsensitive phlorizin binding is 98% of the Na+-dependent phlorizin binding. The plot of phlorizin bound versus FITC bound (Fig. 6) is a straight line, the slope of which gives the stoichiometry between phlorizin and FITC binding. Since only Na+-dependent phlorizin binding is thought to reflect specific carrier binding, the Na+-dependent phlorizin binding was used. The plot is a straight line with a slope of 2.7 -+ 0.3 which suggests about three FITC sites/phlorizin-binding site. This is in contrast to the effect of substrates on FITC inhibition of transport which results in a slope of 1 (Fig. 6b).
These results indicate three FITC sites/phlorizin site, and approximately one glucose transport site/FITC site.
Fluorescence-Bound FITC undergoes a fluorescence quench upon addition of Na+ but not Rb', Cs+, K' , choline+, or glucose. A typical response is shown in Fig. 7 . Addition of 50 m M NaCl to 100 pg of membranes labeled with FITC results in an immediate 19% quench of the FITC fluorescence.
Addition of 50 mM K+ has no effect on the observed fluorescence. The KO., for Na+-induced fluorescence quenching is 25

Effect of pCMBS and phlorizin on FITC-induced inhibition of Na+-dependent glucose uptake and FITC binding.
Vesicles (500 Fg) were pretreated with PITC in the presence of 100 mM NaCl and 10 mM glucose as described in Fig. 2. Following removal of substrates and unreacted PITC by centrifugation, the pCMBS for 30 min at 22 "C and then 50 PM FITC in 50 mM Tris-C1, vesicles were exposed to varying concentrations of phlorizin and pH 9.2, and 2 mM EDTA for 15 min in the dark at 22 "C. Unreacted FITC was removed by centrifugation. Vesicles were washed twice with 300 mM mannitol and 10 D M Hepes/Tris, pH 7.5, with 1 mM dithiothreitol (pCMBS-treated vesicles). Transport and FITC binding were determined as described in Fig. 3. Open squares indicate Na+-dependent glucose uptake or FITC binding in the presence of pCMBS, while solid circles are those in the presence of phlorizin. Each data point is the mean of duplicate samples, and each experiment is representative of three separate experiments. mM as compared to 34 mM (average of three determinations) for glucose uptake. FITC bound in the presence of substrates does not undergo a fluorescence quenching regardless of the ion added.
To examine the site of Na+ interaction with FITC-bound vesicles, two reagents, N-acetylimidazole and tetranitromethane, were examined for their effect on Na+-induced FITC fluorescence quenching and Na-dependent glucose uptake. Fig. 8 indicates that both of these reagents inhibit Na+dependent glucose uptake. N-Acetylimidazole inhibits transport with an apparent of approximately 50 p M , and inhibition is sensitive to the presence of Na+ during exposure to the reagent. Maximum inhibition is 68% (average of four determinations). Sodium at 100 mM protects 98% against inhibition by N-acetylimidazole with an apparent K0.5 of 18 mM, similar to the response seen for Na+ quenching of FITC fluorescence (Fig. 9). The addition of glucose during exposure  Fig. 3 with 2 mM PITC and substrates followed by varying concentrations of FITC. Phlorizin binding was then determined as described in Fig. 5. FITC bound was determined as in Fig. 3. b, vesicles were pretreated with PITC in the presence of substrates as described in Fig. 3. Following removal of substrates and unreacted PITC, vesicles were exposed to varied amounts of FITC in the presence or absence of substrates for 15 min at room temperature. Following removal of substrates and unreacted FITC by centrifugation, Na+-dependent glucose uptake was determined, and FITC was determined by absorbance at 490 nm. FITC bound is the difference between the total FITC and the substrate-protectable FITC bound.
The results are means f S.E. of a single experiment performed in triplicate, and each is representative of the results in three separate experiments.
to N-acetylimidazole does not decrease inhibition in the absence of Na+ or increase protection in its presence. This is consistent with N-acetylimidazole acting at the carrier Na+ site. The other tyrosine group-specific reagent, tetranitromethane, inhibits transport with a KO.5 of 10 p~. Inhibition is not dithiothreitol-sensitive, nor is it sensitive to dilution and washing, indicating that the reagent binds irreversibly to the carrier. Na+ also protects against tetranitromethane inhibition of transport (not shown).
The effect of N-acetylimidazole on the Na+-induced FITC fluorescence quenching is shown in Fig. 10. N-Acetylimidazole inhibits FITC fluorescence quenching with a of 35 p M similar to its apparent for inhibition of transport (35 pM). Inclusion of 100 mM Na during exposure to N-acetylimidazole pr0tect.s the Na+-induced fluorescence quenching in agreement with Na+'s effect on N-acetylimidazole inhibition of Na+-dependent glucose uptake. Addition of glucose during Na+ protection against N-acetylimidazole inhibition does not significantly affect inhibition of transport or Na+-induced FITC fluorescence quenching. The effect of phlorizin and pCMBS on Na+-induced FITC fluorescence quenching is shown in Fig. 11. Both reagents inhibit transport with KO., values of 25 p~ for pCMBS (see also Ref. 12) and 1 p~ for phlorizin (15). In contrast, only pCMBS affects the Na+dependent FITC fluorescence quenching.
Carrier Structural Studies-FITC-labeled KSCN brushborder membrane vesicles were examined on 10-15% linear gradient SDS-polyacrylamide slab gels according to the method of Laemmli (7). Fig. 12 shows the Coomassie Bluestaining pattern at the top of the figure and the FITC-binding pattern at the bottom. The solid line is the FITC-binding pattern following PITC pretreatment in the presence of substrates followed by FITC treatment. The dashed line is substrate-protected vesicles. Only one polypeptide band is protected by the presence of substrates during exposure to FITC, the 75,000-dalton band. In five experiments on three preparations, the molecular mass of the FITC-binding polypeptide band was 75 f 2 kDa.

DISCUSSION
Na+-dependent organic solute transporters have been shown to increase their affinity for their second substrate on exposure to Na+. Na+-induced decreases in carrier K,,, for organic solute have been reported for glucose (2), succinate (16), and proline (17). We have examined the mechanism for this increase in carrier glucose affinity using a fluorescent group specific reagent, FITC.
FITC has been demonstrated in a number of enzyme systems (18)(19)(20)(21)(22) to be a highly sensitive reporter of changes in protein conformation. In these studies, relatively pure enzyme preparations were employed, while the glucose carrier is only Effect of N-acetylimidazole and tetranitromethane on Na+-dependent glucose uptake. a, vesicles were treated with varied concentrations of N-acetylimidazole in 10 mM potassium phosphate buffer, pH 7.5, for 60 min at 22 "C. Unreacted reagent was removed by centrifugation and Na'-dependent glucose uptake (25 PM) in the presence of 100 mM in NaCl determined as described in Fig. 1. Results are means of duplicates, and the plot is representative of five experiments. b, vesicles (500 pg) were treated with varied concentrations of tetranitromethane in 10 mM potassium phosphate buffer, pH 8.8, for 30 min at 22 "C. Unreacted tetranitromethane was removed by centrifugation, the vesicles were washed with 300 mM mannitol, and 10 mM Hepes/Tris, pH 7.5, with and without 1 mM dithiothreitol (DTT). Na+-dependent glucose (25 p~) uptake was determined as described in Fig. 1 with 100 Fig. 7 were subsequently treated with varied concentrations of N-acetylimidazole in the presence and absence of 100 mM NaCl in 10 mM potassium phosphate buffer, pH 7.5, for 60 min at room temperature. Membranes were recovered by centrifugation and were resuspended in 300 mM mannitol + 10 mM HEPES/Tris, pH 7.5.
The FITC fluorescence quenching by Na+ was determined as described in Fig. 7. Results are means of two estimates, and the plot is representative of three separate experiments. a minor component of brush-border membranes (0.1-0.3% (12)). To overcome this problem, we used brush-border membranes enriched in carrier and selectively labeled the carrier by pretreatment of the membranes with a nonfluorescent FITC analogue, PITC, in the presence of substrates.
Carrier appeared to be selectively purified by an adaptation of the KSCN procedure of Hopfer (ll), as previously described (4). As judged by the increase in V,,, of Na+-dependent glucose uptake and Na+-dependent phlorizin binding (Table  I) FIG. 11. Effect of N-acetylimidazole, phlorizin, andpCMBS on the Na+-dependent FITC fluorescence quenching. Vesicles (500 pg) were treated with PITC in the presence of substrates, followed by FITC as described in Fig. 3. Following removal of unreacted FITC by centrifugation, vesicles were treated with variable concentrations of N-acetylimidazole in 10 mM potassium phosphate buffer, pH 7.5, for 60 min at 22 "C or with variable concentrations of pCMBS or phlorizin in 10 mM Hepes/Tris, pH 7.5, for 30 min at room temperature. Fluorescence quench by 50 mM Na+ was then measured as described in Fig. 7. Results are expressed as means of 2 S.E. of four estimates.
(-75,000 daltons) that approximately 2.5% of the membrane protein is glucose carrier. This is similar to the enrichment reported following papain digestion, deoxycholate extraction, and alkali extraction (23). The selective enrichment of the glucose transporter suggests that this polypeptide is more firmly anchored to the membrane lipid than the marker enzymes.
Selective FITC binding to the carrier was accomplished by pretreatment of the vesicles in the presence of substrate with a nonfluorescent irreversible, but substrate-protectable, FITC analogue, PITC. Fig. 3 shows a 90% reduction in FITC labeling of membranes following PITC pretreatment. Pretreatment with PITC in the presence of substrates does not affect Na'/D-ghcose uptake, nor does it affect subsequent FITC inhibition. Maximum inhibition by both isothiocyanate derivatives is approximately 65% which is somewhat larger than that reported by Weber and Semenza (24). This may reflect the increased purification of carrier in KSCN-treated vesicles, or the higher pH of the reaction medium. PITC reacts with unprotonated amino groups and so even greater inhibition may be seen a t higher pH values.
The site of FITC interaction with the carrier was identified to be at, or near, the glucose site. Glucose in the presence of Na' completely protects against inhibition of Na+-dependent glucose uptake by both isothiocyanate derivatives with an apparent for glucose of 50 p~, similar to the apparent K,,, for transport of 85 p M in the presence of 100 mM Na' . Phlorizin, which has been identified as a competitive inhibitor of glucose uptake with an apparent of 8 PM in rabbit small intestine brush-border membrane vesicles (25), protects against FITC binding and inhibition of Na+-dependent glucose uptake with a KO, of 15 PM. The reduction in FITC binding by Na' and phlorizin is similar to the reduction in binding seen with Na' and glucose. These results indicate that FITC binds near to or at the carrier glucose-binding site.
The slope of FITC binding uersus phlorizin binding is approximately 3 (Fig. 6), indicating that FITC competes for the same sites which bind phlorizin in the presence of Na'. At this juncture, we have no definitive explanation for the stoichiometry between FITC and phlorizin binding. Weber and Semenza (24) have suggested that a lysine is at or near the glucose-binding site on the basis of inhibition of glucose uptake by lysine group-specific reagents. FITC is also thought to react with e N H 2 groups of lysine in a number of enzyme systems. Fig. 12 indicates that FITC binds specifically to a 75kilodalton polypeptide band on SDS-reducing gels. The pres-ence of substrates during exposure to FITC specifically reduces the amount of FITC bound to this polypeptide 60% without significantly affecting FITC binding to any other polypeptide. At present, it is not known if this band contains more than one polypeptide. In light of the results with 4azidophlorizin (26) and with a monoclonal antibody against the glucose carrier (27), we feel that this polypeptide band is a good candidate for the carrier. The results of Malathi and Preiser (28) suggest that the transport competent carrier is a dimer composed of two 85,000-dalton monomers, but we have no direct information bearing on this point. Other reports have suggested molecular masses of 50,000 to 360,000 daltons (29). The identity of the Na'-binding site is a t present unknown.
The fluorescence of FITC bound to the carrier-enriched vesicles is quenched in a specific saturable fashion by Na' (Ref. 4 and Fig. 4). Li' , Cs' , Rb' , K' and choline' have no effect on FITC fluorescence. The Na' quench is saturable with a KO, of 25 mM which is similar to the K o . 5 for Na+dependent transport (34 mM, Fig. 1 and Table 11). The quench of FITC fluorescence by Na+ indicates a rapid change in the carrier conformation at the glucose site. The inhibition of transport and the Na'-induced FITC fluorescence quench by N-acetylimidazole confirms that FITC fluorescence is indeed monitoring glucose carrier conformation. The tyrosine group-specific reagent N-acetylimidazole (30) inhibits both transport and Na' quench with similar Ko.5 values (50 and 35 pM, respectively), while Na' in the absence of glucose protects N-acetylimidazole inhibition of transport and the Na' quench with values of 18 and