![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
METABOLISM, TRANSPORT, AND PHARMACOGENOMICS
Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, Galveston, Texas (A.M.P., K.M.R.); and GlaxoSmithKline, Biochemical and Analytical Pharmacology, Research Triangle Park, North Carolina (S.A.K., C.D.S.)
Received for publication
August 7, 2007
Accepted
December 5, 2007.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
-methyl-D-glucopyranose transport by phlorizin and a new candidate drug, sergliflozin-A [(2-[4-methoxyphenyl]methyl)phenyl β-D-glucopyranoside], in COS-7 cells expressing hSGLT1 and hSGLT2. Inhibition by phlorizin was competitive, with Ki values of 0.3 µM in hSGLT1 and 39 nM in hSGLT2. Inhibition by sergliflozin-A was also competitive, with Ki values of 1 µM in hSGLT1 and 20 nM in hSGLT2. Phloretin [3-(4-hydroxyphenyl)-1-(2,4,6-trihydroxyphenyl)-1-propanone; the aglucone of phlorizin] was a less potent inhibitor, with IC50 values of 142 µM in hSGLT1 and 25 µM in hSGLT2. Site-directed mutagenesis of residues believed to be in the phlorizin binding site showed that only Cys610 is involved in inhibitor binding in the human transporters. Mutation of Cys610 in hSGLT1 to lysine resulted in an increased IC50 for all inhibitors. In contrast, mutagenesis of the analogous Cys615 in hSGLT2 produced the opposite effect, a decrease in IC50 for phlorizin and sergliflozin-A. The differences in the effects of the mutations between hSGLT1 and hSGLT2 suggest that this cysteine holds key residues in place rather than participating directly in inhibitor binding.
). The low-affinity Na+/glucose cotransporter, SGLT2, is found in the early portion of the renal proximal tubule, and it accounts for the bulk of glucose reabsorption. Mutations in the SLC5A2 gene coding for SGLT2 produce familial renal glucosuria (van den Heuvel et al., 2002; Santer et al., 2003; Magen et al., 2005). The high-affinity Na+/glucose cotransporter, SGLT1, is found in the S3 segment of the proximal tubule, but it is more abundant in the small intestine. Mutations in the SLC5A1 gene encoding SGLT1 produce glucose-galactose malabsorption syndrome in the gastrointestinal tract, with minimal effects on renal glucose reabsorption (Lam et al., 1999). The properties of SGLT1 have been very well characterized, but until now, SGLT2 has been difficult to study because of low expression in heterologous systems (Wright, 2001).
There is recent interest in SGLT2 as a drug target for the treatment of type II diabetes using selective inhibitors based on the structure of the phenylglucoside, phlorizin (Oku et al., 1999; Isaji, 2007). Phlorizin has long been known as an inhibitor of sodium-dependent glucose transport (Ehrenkranz et al., 2005). Early studies showed that diabetic animals treated with phlorizin have an increased excretion of glucose in their urine, which normalizes plasma glucose without producing hypoglycemia (Rossetti et al., 1987). However, phlorizin is metabolized and poorly absorbed in the intestine, and newer compounds such as T-1095 and sergliflozin have been developed to overcome these drawbacks (Oku et al., 1999; Katsuno et al., 2007). In addition, the newer compounds are more selective for SGLT2 than SGLT1, thus minimizing potential side effects associated with the broad tissue distribution of SGLT1. Besides intestine and kidney, SGLT1 is found in other organs and tissues, including blood-brain barrier and heart (Zhou et al., 2003; Elfeber et al., 2004; Tazawa et al., 2005). In diabetic animal models, inhibition of SGLT2 produces glucosuria, followed by a normalization of blood glucose levels and a consequent improvement in insulin resistance and renal damage (Asano et al., 2004; Katsuno et al., 2007).
Although there is no detailed structural information for any of the Na+/glucose cotransporters, there is evidence that phlorizin interacts with SGLT1 at two distinct sites: the sugar moiety binds to the substrate binding site, and the aromatic rings of the aglucone or phloretin moiety bind to a second site (Hirayama et al., 2001). The aglucone binding site in SGLT1 is in a functionally important location because occupation of this site produces transport inhibition (Hirayama et al., 2001). The aglucone binding site has been hypothesized to include polar residues located in a large loop between transmembrane (TM) helices 13 and 14 (Novakova et al., 2001; Raja et al., 2003). Site-directed mutagenesis of amino acids 604 to 610 of the rabbit SGLT1 has identified residues, including Cys608, that affect IC50 for phlorizin without changing substrate affinity (Novakova et al., 2001). There have been no such studies in SGLT2, which contains Cys615 at the same position. In the present study, we characterized the interaction of human (h) SGLT2, mouse (m) SGLT2, and hSGLT1 with phlorizin and a new phlorizin-based compound, sergliflozin-A. Sergliflozin-A inhibited SGLT2 with high apparent affinity and at least a 50-fold difference in selectivity between hSGLT2 and hSGLT1. Furthermore, we also found that Cys610 of hSGLT1 and Cys615 of hSGLT2 may be important as indirect determinants of inhibitor binding but are not likely to interact directly with inhibitors.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-Methyl-D-glucopyranose (AMG) was obtained from American Radiolabeled Chemicals (St. Louis, MO) or GE Healthcare (Little Chalfont, Buckinghamshire, UK).
|
Cell Culture. COS-7 cells from the American Type Culture Collection were cultured in Dulbecco's modified Eagle's medium, high glucose, pyridoxine HCl, and 25 mM HEPES (Invitrogen, Carlsbad, CA) supplemented with 10% heat-inactivated fetal calf serum (Hyclone Laboratories, Logan, UT), 100 units/ml penicillin, and 100 µg/ml streptomycin at 37°C in 5% CO2. Cells were plated on 24-well white-sided Visiplate TC plates (GE Healthcare) for direct scintillation counting. For transient transfections, 1.2 x 105 cells were plated per well, and each well of cells was transfected with 1.8 µl of Fugene 6 (Roche Applied Science, Indianapolis, IN) and 0.6 µg of plasmid DNA (9:3 ratio) (Pajor and Randolph, 2005). The two mutants with low activity (hSGLT1-C610K and mSGLT2-N173A) were transfected and assayed in six-well plates, as described previously (Pajor and Randolph, 2005).
Transport Assays. Transport assays were carried out 48 h after transfections as described previously except that all steps were done at 37°C (Pajor and Valmonte, 1996). The sodium buffer contained 140 mM NaCl, 2 mM KCl, 1 mM MgCl2, 1 mM CaCl2, and 10 mM HEPES, pH adjusted to 7.4 with 1 M Tris. Choline buffer contained 140 mM choline chloride in place of NaCl. For the assays, each well was washed twice with 1 ml of choline buffer, then incubated with 0.25 ml of sodium buffer containing [14C]-methyl-D-glucopyranoside for 10 min (hSGLT1 and mutants) or 30 min (hSGLT2 and mutants). Inhibitors were added to the transport solutions from dimethyl sulfoxide stocks, with a final volume of dimethyl sulfoxide less than 1%. Phloretin inhibition experiments included a 10-min preincubation with inhibitor, which reduced variability in the results. The plates were rocked in a heated incubator during the incubation period. The uptakes were stopped and extracellular radioactivity removed with four 1-ml washes of choline buffer. After the last wash was removed, each well of cells was dissolved in 1 ml of scintillation cocktail, OptiPhase Supermix (GE Healthcare). The plates were sealed with plastic plate covers and counted directly in a Microbeta Trilux 1450 plate scintillation counter (GE Healthcare). For all experiments, uptakes in vector-transfected cells were subtracted from uptakes in SGLT plasmid-transfected cells (for mouse SGLT2, the vector was pCMV-SPORT6; for the others, the vector was pcDNA3.1). There was no difference in the background counts obtained with each vector. For calculation of Ki values from Dixon plots, inhibition of transport activity was measured at multiple inhibitor and substrate concentrations. The data were analyzed by linear regression, and the x-axis value corresponding to the intersection of the lines was taken as –Ki (Segel, 1975). To determine IC50 values (the inhibitor concentration resulting in 50% inhibition), the transport of 100 µM [14C]AMG was measured in the presence of increasing concentrations of inhibitor and the resulting activity (v%, as a percentage of control) was analyzed by nonlinear regression using v% = IC50/([I] + IC50) x 100, where [I] represents the inhibitor concentration. Statistical analysis with Student's t test or with one-way analysis of variance followed by Dunnett's test was done using the Sigma Stat program (SPSS Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
; Ikari and Suketa, 2002), although a more recent study using stably transfected CHO-K1 cells reports much higher expression (Castaneda and Kinne, 2005). In preliminary experiments, we tested different cell lines and assay conditions. For example, the measurable transport activity was much increased by doing the assays at 37°C instead of room temperature. We also compared the functional expression of hSGLT2 in several cell lines: COS-7, HRPE, and CHO-K1. There was little detectable transport activity in transiently transfected CHO-K1 cells and high background transport of AMG in HRPE cells (data not shown). The COS-7 cells had the lowest background transport and the highest expression of functional transport. As shown in Fig. 2, although the highest uptake activity was measured in cells expressing hSGLT1, there was considerable transport activity of hSGLT2 and mSGLT2 above background. AMG transport in hSGLT2 was completely sodium-dependent; replacing sodium by choline resulted in a decrease in transport activity to the background levels seen in vector-transfected cells (data not shown). Time courses of uptakes were linear through 15 min for hSGLT1 and 45 min for hSGLT2 (data not shown). Therefore, time points of 10 min were used for hSGLT1 and 30 min for hSGLT2 in all subsequent assays.
|
). As expected, Dixon plots of inhibition of hSGLT1 and hSGLT2 crossed above the x-axis (Fig. 3), indicative of competitive inhibition (Segel, 1975). The Ki for phlorizin in hSGLT1 was 0.12 µM in the experiment shown in Fig. 3A, and in three experiments, the mean Ki was 0.31 ± 0.10 µM (mean ± S.E.M.). As shown in Fig. 3B, hSGLT2 was more sensitive than hSGLT1 to inhibition by phlorizin, with a Ki of 25 nM in the experiment shown in Fig. 3B and a mean Ki of 39 ± 14 nM (n = 3). The mouse SGLT2 also was sensitive to inhibition by phlorizin with a mean Ki of 65 ± 5 nM (mean ± range, n = 2) (data not shown).
|
). As shown in Fig. 4, sergliflozin-A inhibited both hSGLT1 and hSGLT2, also as a competitive inhibitor. In hSGLT1, the Ki of the experiment shown in Fig. 4A was 1.10 µM, and in four experiments, the mean Ki was 0.96 ± 0.23 µM. The sensitivity of hSGLT2 to sergliflozin-A was at least 50-fold greater than that of hSGLT1. In Fig. 4B, the Ki for sergliflozin-A in hSGLT2 was 9 nM, and in six experiments, the mean Ki was 18 ± 4 nM. The mouse SGLT2 had a mean Ki for sergliflozin-A of 24 ± 2 nM (data not shown), similar to that of the human ortholog.
|
SGLT Mutants. Previous studies have suggested that Cys608 and Tyr604 in the rabbit SGLT1 may interact with phlorizin because replacement of these amino acids with lysine produced increases in IC50 for phlorizin without affecting substrate binding (Novakova et al., 2001). These residues are located in a loop between transmembrane helices 13 and 14, also referred to as Loop 13. The topological orientation of this loop is still in question, although there is some evidence that it may be a re-entrant loop rather than completely intracellular (Gagnon et al., 2005). Therefore, we investigated whether the residues identified in these studies as potential phlorizin binding sites in rbSGLT1 are involved in phlorizin and sergliflozin-A binding in the human SGLT1 and SGLT2 transporters. Sequence alignments showed that Cys608 in rbSGLT1 is conserved in all of the SGLT transporters, at positions 610 (hSGLT1), 615 (hSGLT2), and 614 (mSGLT2). The Tyr604 in rbSGLT1 is found at position 602 in human SGLT1, but it is changed to leucine in SGLT2 (Leu611 in hSGLT2; Leu610 in mSGLT2). Therefore, we substituted both lysine and leucine or tyrosine at the position equivalent to Tyr604. A separate study showed that Asp176 at the outer surface of transmembrane helix 5 in rabbit SGLT1 also affects phlorizin binding independently of sugar affinity (Panayotova-Heiermann et al., 1994). The human and mouse SGLT2 orthologs both have asparagine at the equivalent position, Asn173. Therefore, we made the N173A mutant in first human and later mouse SGLT2 to examine whether this residue could explain differences in apparent affinity for phlorizin. As shown in Fig. 5, most of the mutant transporters had similar transport activity as the wild type, with the exception of hSGLT1-C610K and hSGLT2-N173A, which had undetectable activity. Because the human N173A mutant was inactive, we also made the N173A mutant in the mouse SGLT2 background, and this mutant had low but measurable activity. It was not possible to determine whether the low activity is due to an inactive protein on the plasma membrane or a lack of protein at the cell surface because of a lack of antibodies available for measurement of cell surface protein expression of SGLT2. We tested the same commercial antibodies (S1010-87E; US-Biological, Swampscott, MA) reported in a previous study (Castaneda and Kinne, 2005) but found no signal with the cloned hSGLT2 even at higher antibody concentrations, which could reflect variability between antibody lots or differences in protein abundance.
|
hSGLT1 Mutants. The hSGLT1 mutants were tested for their sensitivity to phlorizin, sergliflozin-A, and the aglucone of phlorizin, phloretin (Table 1). The IC50 values were calculated from inhibition curves using 100 µM [14C]AMG and increasing concentrations of inhibitor. Because the substrate concentration used in these experiments is lower than the reported Km of 0.7 to 0.9 mM for AMG in hSGLT1 (Hirayama et al., 1996; Gagnon et al., 2005) and approximately 1.6 mM for hSGLT2 (Kanai et al., 1994), the IC50 values provide an estimate of the Ki. The wild-type hSGLT1 IC50 of 0.17 µM for phlorizin (Table 1) compares well with the mean Ki of 0.31 µM determined from Dixon plots. The two mutants at position 606 had no change in IC50 values compared with the wild-type hSGLT1, but the C610K mutant of hSGLT1 had significantly larger IC50 values for phlorizin, sergliflozin-A, and phloretin (Table 1). It is interesting to note that the IC50 value for phlorizin was the most affected by the mutation, by 10-fold, compared with only approximately 5-fold for sergliflozin-A and approximately 2-fold for phloretin, the aglucone of phlorizin (Fig. 1). We measured a mean IC50 value for phloretin in wild-type hSGLT1 of approximately 140 µM (Table 1). For comparison, a published value for phloretin Ki is 50 µM, from two-electrode voltage-clamp studies with hSGLT1 in Xenopus oocytes voltage-clamped to –150 mV (Hirayama et al., 2001). The kinetic constants for hSGLT1 mutants were not significantly affected by the mutations at positions 606 and 610 (Table 2). This result is similar to the previous study with rbSGLT1, which reported no change in apparent Km in the C608K mutant (Novakova et al., 2001). The Km value for AMG transport by hSGLT1 expressed in COS-7 cells is 1.8 mM, compared with previous reports of 0.7 to 0.9 mM (Hirayama et al., 1996; Gagnon et al., 2005). However, the present study measured radiotracer uptakes in nonvoltage-clamped COS-7 cells, whereas the previous studies examined substrate-dependent currents in Xenopus oocytes clamped to –50 or –150 mV. Because the substrate Km in SGLT1 increases as membrane potential becomes more positive (Hirayama et al., 1996), it is not unexpected to find a higher apparent Km in transfected cells.
|
|
hSGLT2 Mutants. Similar to the hSGLT1 mutants, mutations at position 611 in hSGLT2 did not affect the apparent affinity for phlorizin, sergliflozin-A, or phloretin (Table 1). In contrast to the hSGLT1-C610K mutant, which exhibited increased IC50 for inhibitors, the hSGLT2-C615K mutant had a significant 2-fold decrease in IC50 for both phlorizin and sergliflozin-A compared with the wild type. The IC50 value for phloretin in the wild-type hSGLT2 was approximately 25 µM (Table 1). There were no significant differences in phloretin IC50 values in any of the hSGLT2 mutants, and there was considerable variation between experiments with L611Y. The kinetic values for AMG in the hSGLT2 mutants were all quite similar to the wild-type value of 4.8 mM (Table 2). This Km for hSGLT2 is higher than the previous estimated value of 1.8 mM, obtained using Xenopus oocytes with an expression level of only 2-fold above background (Kanai et al., 1994), but compares well with the Km of 6 mM measured in human renal brush border membrane vesicles (Turner and Silverman, 1977).
mSGLT2 Mutant. The IC50 values of inhibitors interacting with the wild-type mouse SGLT2 were very similar to the human SGLT2 values (Table 1). The N173A mutant of mSGLT2 had a significant decrease in IC50 for phlorizin and sergliflozin-A, whereas the IC50 for phloretin was not affected by the mutation. Because of low and variable expression, it was difficult to determine the kinetic constants in the N173A mutant, although it appears that the Km is higher than in wild type. In our initial experiment using AMG concentrations up to 20 mM, the estimated Km for AMG was 32 mM (data not shown). In a more detailed experiment with AMG concentrations up to 75 mM, the Km for AMG was 17.9 mM (Table 2). In comparison, this mutation in rabbit SGLT1 produces an increased IC50 for phlorizin with no change in sugar affinity (Panayotova-Heiermann et al., 1994).
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
), there is very little information on the functional properties of this protein primarily because it has low activity when expressed in heterologous systems. In previous studies, transport activities of up to 2-fold above background have been reported (Kanai et al., 1994; Ikari and Suketa, 2002). A more recent study has shown that stably transfected CHO-K1 cells exhibit increased expression of hSGLT2 (Castaneda and Kinne, 2005), but transiently transfected cells have advantages for rapid screening of mutants. In the present study, we report transient expression of approximately 8-fold above background for human SGLT2 and almost 20-fold for mouse SGLT2, which allows a more detailed functional characterization.
The Na+/glucose cotransporters can be inhibited with high affinity by a number of low-molecular weight inhibitors. Previous estimates of the IC50 for phlorizin in hSGLT2 range between 1 and 2 µM (Kanai et al., 1994; Ikari and Suketa, 2002), but these were not detailed studies, and the values may be incorrect because of problems associated with low transport activity. We found that the Ki for phlorizin in hSGLT2 was approximately 40 nM, and the mechanism of inhibition was competitive, similar to a recent report of a Ki of 19 nM in CHO-K1 cells stably transfected with hSGLT2 (Katsuno et al., 2007). In the present study, the Ki for phlorizin in hSGLT1 was approximately 0.3 µM, which compares well with previous reports of approximately 0.15 to 0.2 µM (Hirayama et al., 1996; Katsuno et al., 2007). Another study of stably transfected CHO-K1 cells reported phlorizin Ki values of 19 and 12 µM for hSGLT1 and hSGLT2, respectively (Castaneda and Kinne, 2005), but there may be problems such as unstirred layers associated with the assay method. Interestingly, we found that the properties of mouse SGLT2 were very similar to those of the human ortholog, consistent with the 90% sequence identity between the two proteins.
Sergliflozin-A seems to be a promising compound for the treatment of Type II diabetes. The inhibition of SGLT2 reduces glucose reabsorption in the kidney and thus increases urinary glucose excretion, with subsequent normalization of plasma glucose levels (Katsuno et al., 2007). In transiently transfected COS-7 cells in the present study, sergliflozin-A acted as a competitive inhibitor of hSGLT2 with a Ki of approximately 20 nM. The increased apparent affinity of sergliflozin-A compared with phlorizin is probably a result of a stronger interaction between the aglucone moiety and the hSGLT2 protein. More importantly, the selectivity of sergliflozin-A for hSGLT2 is at least 50-fold higher than for hSGLT1. We found a Ki for sergliflozin-A in hSGLT1 of 
1 µM. The Ki values reported for sergliflozin-A in hSGLT1 and hSGLT2 in stably transfected CHO-K1 cells are 0.7 µM and 2.4 nM, respectively (Katsuno et al., 2007). The high specificity of sergliflozin-A for hSGLT2 over hSGLT1 should help to minimize unwanted side effects. An earlier phlorizin-based SGLT inhibitor, T-1095, is also absorbed well in the intestine and converted to an active form, but its affinity and selectivity are very similar to that of phlorizin (Oku et al., 1999).
The mechanism of glucose transporter inhibition by phlorizin and other phenylglucoside inhibitors has been shown to involve binding of the glucose moiety to the glucose binding site, which positions the aglucone or phloretin moiety in a vestibule that is accessible from the outside (Lostao et al., 1994; Díez-Sampedro et al., 2000; Hirayama et al., 2001). The aglucone moiety of the inhibitor appears to interact with hydrophobic or aromatic residues on SGLT1 with at least five main contact points (Hirayama et al., 2001). Phlorizin binding occurs in the outward facing conformation of the transporter, after the sodium binding has occurred, presumably as the result of Na+-induced conformational changes resulting in exposure of residue side chains for binding interactions (Quick et al., 2003).
Novakova et al. (2001) have proposed that phlorizin binds nonspecifically to a hydrophobic pocket formed by the last intracellular loop of SGLT1, particularly amino acids 604 to 610. Several findings support this assertion. For example, replacements of polar amino acids in this region by lysine alter the IC50 for phlorizin without much change in Km for AMG (Novakova et al., 2001). The hydrophobicity and overall structure of the binding pocket seem to be more important than the identity of individual amino acids because replacement of these polar residues by nonpolar amino acids, such as alanine (Novakova et al., 2001) or cysteine (Gagnon et al., 2005), does not have an effect on apparent binding affinity. Furthermore, a purified peptide containing amino acids 604 to 610 binds both phlorizin and phloretin but does not interact with glucose (Raja et al., 2003; Xia et al., 2003). However, there is some evidence that does not support the hypothesis. For example, a truncated rbSGLT1 consisting of TM 10 to 14, called C5, transports glucose but is not inhibited by phlorizin (Panayotova-Heiermann et al., 1997, 1999). Instead, glucose transport in C5 is inhibited by the aglucone, phloretin. C5 contains the loop between TM 13 and 14, including residues 604 to 610, and theoretically should bind phlorizin. Notably in that study, phlorizin binding was not tested directly but indirectly by the inhibition of transport, which leaves open the possibility that the phlorizin binding site in C5 is intact but no longer inhibits transport. The orientation of the putative phlorizin binding site in loop 13 is still not clear because there is contradictory evidence that the loop is intracellular, extracellular, or a re-entrant loop (Turk et al., 1996; Lin et al., 1999; Gagnon et al., 2005). To bind phlorizin, a water-soluble membrane-impermeable compound, the binding site would have to be accessible from the outside of the cell when the transporter is in an outward-facing conformation. Finally, the results of our current study show that Tyr606 in hSGLT1 and Leu611 in hSGLT2 are not involved in inhibitor binding, unlike the results reported for Tyr604 in rbSGLT1 at the equivalent position (Novakova et al., 2001).
The conserved residues, Cys610 in hSGLT1 and Cys615 in hSGLT2, appear to participate in inhibitor binding, similar to the findings with the C608K mutant of rbSGLT1 in the previous study (Novakova et al., 2001). We found that substitution of lysine for Cys610 in hSGLT1 produced increases in IC50 for phlorizin, sergliflozin-A, and phloretin consistent with interaction of all of these compounds with a single binding pocket. The mutation affected phlorizin binding more than sergliflozin-A suggesting differences in amino acids that interact with these compounds. It is interesting to note that the C615K mutation in human SGLT2 produced the opposite effect on inhibitor binding compared with hSGLT1, namely an increased apparent affinity for phlorizin and sergliflozin-A. Although it was somewhat surprising that mutation of these conserved cysteines would have opposite effects, it is possible that the function of these residues is to position key amino acids that bind directly with the inhibitors. The differences in effect of cysteine mutations could be due to differences in the key amino acids in hSGLT1 and hSGLT2.
The N173A mutation also produced opposite effects in SGLT2 compared with the previous study in SGLT1 (Panayotova-Heiermann et al., 1994). In SGLT1, mutation of the aspartate residue at position 176 (which corresponds to Asn173 in SGLT2) to asparagine had no effect on function, but replacement with alanine resulted in an increased phlorizin Ki with no effect on sugar binding (Panayotova-Heiermann et al., 1994). This amino acid is likely to be important for the stability or structure of SGLT1 and SGLT2 because mutations at this position resulted in proteins with very low functional activity. In the present study, the N173A mutation in SGLT2 resulted in a decrease in the Ki for both phlorizin and sergliflozin-A, although the change was relatively modest, approximately 2-fold.
In conclusion, we report inhibition of the human Na+/glucose cotransporters, hSGLT1 and hSGLT2, by the inhibitors phloretin, phlorizin and sergliflozin-A. These compounds were more potent inhibitors of hSGLT2 than hSGLT1. Site-directed mutagenesis experiments suggest that Cys610 in hSGLT1, and the analogous Cys615 in hSGLT2 may be important in maintaining the structure of the inhibitor binding site, but these residues are not likely to bind directly to inhibitors.
|
Footnotes |
|---|
ABBREVIATIONS: SGLT2, high-affinity Na+/glucose cotransporter, predominant renal form; SGLT1, human high-affinity Na+/glucose cotransporter, predominantly found in intestine; phlorizin, phloretin-2'-β-glucoside; h, human; m, mouse; sergliflozin-A, (2-[4-methoxyphenyl]methyl)phenyl β-D-glucopyranoside; phloretin, 3-(4-hydroxyphenyl)-1-(2,4,6-trihydroxyphenyl)-1-propanone; TM, transmembrane; AMG, -methyl-D-glucopyranose; CHO, Chinese hamster ovary; T-1095, 3-(benzo[b]furan-5-yl)-2',6'-dihydroxy-4'-methylpropiophenone-2'-O-(6-O-methoxycarbonyl)-β-D-glucopyranoside.
Address correspondence to: Dr. Ana M. Pajor, Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, Galveston, TX 77555. E-mail: ampajor{at}utmb.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Asano T, Anai M, Sakoda H, Fujishiro M, Ono H, Kurihara H, and Uchijima Y (2004) SGLT as a therapeutic target. Drugs Future 29: 461–466.[CrossRef]
Castaneda F and Kinne RK (2005) A 96-well automated method to study inhibitors of human sodium-dependent D-glucose transport. Mol Cell Biochem 280: 91–98.[CrossRef][Medline]
Díez-Sampedro A, Lostao MP, Wright EM, and Hirayama BA (2000) Glycoside binding and translocation in Na+-dependent glucose cotransporters: comparison of SGLT1 and SGLT3. J Membr Biol 176: 111–117.[CrossRef][Medline]
Ehrenkranz JR, Lewis NG, Ronald KC, and Roth J (2005) Phlorizin: a review. Diabetes Metab Res Rev 21: 31–38.[CrossRef][Medline]
Elfeber K, Kohler A, Lutzenburg M, Osswald C, Galla HJ, Witte OW, and Koepsell H (2004) Localization of the Na+-D-glucose cotransporter SGLT1 in the blood-brain barrier. Histochem Cell Biol 121: 201–207.[CrossRef][Medline]
Gagnon DG, Holt A, Bourgeois F, Wallendorff B, Coady MJ, and Lapointe JY (2005) Membrane topology of loop 13–14 of the Na+/glucose cotransporter (SGLT1): a SCAM and fluorescent labelling study. Biochim Biophys Acta 1712: 173–184.[Medline]
Hirayama BA, Diez-Sampedro A, and Wright EM (2001) Common mechanisms of inhibition for the Na+/glucose (hSGLT1) and Na+/Cl–/GABA (hGAT1) cotransporters. Br J Pharmacol 134: 484–495.[CrossRef][Medline]
Hirayama BA, Lostao MP, Panayotova-Heiermann M, Loo DD, Turk E, and Wright EM (1996) Kinetic and Specificity differences between rat, human, and rabbit Na+-glucose cotransporters (SGLT1). Am J Physiol Gastrointest Liver Physiol 270: G919–G926.
Ikari A and Suketa Y (2002) Expression of GFP-tagged low affinity Na+-dependent glucose transporter in Xenopus oocytes and CHO cells. Jpn J Physiol 52: 395–398.[CrossRef][Medline]
Isaji M (2007) Sodium-glucose cotransporter inhibitors for diabetes. Curr Opin Invest Drugs 8: 285–292.[Medline]
Kanai Y, Lee WS, You G, Brown D, and Hediger MA (1994) The human kidney low affinity Na+/glucose cotransporter SGLT2: delineation of the major renal reabsorptive mechanism for D-glucose. J Clin Invest 93: 397–404.[Medline]
Katsuno K, Fujimori Y, Takemura Y, Hiratochi M, Itoh F, Komatsu Y, Fujikura H, and Isaji M (2007) Sergliflozin, a novel selective inhibitor of low affinity sodium glucose cotransporter (SGLT2), validates the critical role of SGLT2 in renal glucose reabsorption and modulates plasma glucose level. J Pharmacol Exp Ther 320: 323–330.
Lam JT, Martin MG, Turk E, Hirayama BA, Bosshard NU, Steinmann B, and Wright EM (1999) Missense mutations in SGLT1 cause glucose-galactose malabsorption by trafficking defects. Biochim Biophys Acta 1453: 297–303.[Medline]
Lin J, Kormanec J, Homerova D, and Kinne RK (1999) Probing transmembrane topology of the high-affinity sodium/glucose cotransporter (SGLT1) with histidine-tagged mutants. J Membr Biol 170: 243–252.[CrossRef][Medline]
Lostao MP, Hirayama BA, Loo DD, and Wright EM (1994) Phenylglucosides and the Na+/glucose cotransporter (SGLT1): analysis of interactions. J Membr Biol 142: 161–170.[Medline]
Magen D, Sprecher E, Zelikovic I, and Skorecki K (2005) A novel missense mutation in SLC5A2 encoding SGLT2 underlies autosomal-recessive renal glucosuria and aminoaciduria. Kidney Int 67: 34–41.[CrossRef][Medline]
Novakova R, Homerova D, Kinne RKH, Kinne-Saffran E, and Lin JT (2001) Identification of a region critically involved in the interaction of phlorizin with the rabbit sodium-D-glucose cotransporter, SGLT1. J Membr Biol 184: 55–60.[CrossRef][Medline]
Oku A, Ueta K, Arakawa K, Ishihara T, Nawano M, Kuronuma Y, Matsumoto M, Saito A, Tsujihara K, Anai M, et al. (1999) T-1095, an inhibitor of renal Na+-glucose cotransporters, may provide a novel approach to treating diabetes. Diabetes 48: 1794–1800.[Abstract]
Pajor AM and Randolph KM (2005) Conformationally sensitive residues in extracellular loop 5 of the Na+/dicarboxylate co-transporter. J Biol Chem 280: 18728–18735.
Pajor AM and Valmonte HG (1996) Expression of the renal Na+/dicarboxylate cotransporter, NaDC-1, in COS-7 Cells. Pflugers Arch 431: 645–651.[Medline]
Panayotova-Heiermann M, Eskandari S, Turk E, Zampighi GA, and Wright EM (1997) Five transmembrane helices form the sugar pathway through the Na+/glucose cotransporter. J Biol Chem 272: 20324–20327.
Panayotova-Heiermann M, Leung DW, Hirayama BA, and Wright EM (1999) Purification and functional reconstitution of a truncated human Na+/glucose cotransporter (SGLT1) expressed in E. coli. FEBS Lett 459: 386–390.[CrossRef][Medline]
Panayotova-Heiermann M, Loo DDF, Lostao MP, and Wright EM (1994) Sodium/D-glucose cotransporter charge movements involve polar residues. J Biol Chem 269: 21016–21020.
Quick M, Tomasevic J, and Wright EM (2003) Functional asymmetry of the human Na+/glucose transporter (hSGLT1) in bacterial membrane vesicles. Biochemistry 42: 9147–9152.[CrossRef][Medline]
Raja MM, Tyagi NK, and Kinne RK (2003) Phlorizin recognition in a c-terminal fragment of SGLT1 studied by tryptophan scanning and affinity labeling. J Biol Chem 278: 49154–49163.
Rossetti L, Smith D, Shulman GI, Papachristou D, and DeFronzo RA (1987) Correction of hyperglycemia with phlorizin normalizes tissue sensitivity to insulin in diabetic rats. J Clin Invest 79: 1510–1515.[Medline]
Santer R, Kinner M, Lassen CL, Schneppenheim R, Eggert P, Bald M, Brodehl J, Daschner M, Ehrich JH, Kemper M, et al. (2003) Molecular analysis of the SGLT2 gene in patients with renal glucosuria. J Am Soc Nephrol 14: 2873–2882.
Segel IH (1975) Enzyme Kinetics, John Wiley and Sons, New York.
Tazawa S, Yamato T, Fujikura H, Hiratochi M, Itoh F, Tomae M, Takemura Y, Maruyama H, Sugiyama T, Wakamatsu A, et al. (2005) SLC5A9/SGLT4, a new Na+-dependent glucose transporter, is an essential transporter for mannose, 1,5-anhydro-D-glucitol, and fructose. Life Sci 76: 1039–1050.[CrossRef][Medline]
Turk E, Kerner CJ, Lostao MP, and Wright EM (1996) Membrane topology of the human Na+/glucose cotransporter SGLT1. J Biol Chem 271: 1925–1934.
Turner RJ and Silverman M (1977) Sugar uptake into brush border vesicles from normal human kidney. Proc Natl Acad Sci U S A 74: 2825–2829.
van den Heuvel LP, Assink K, Willemsen M, and Monnens L (2002) Autosomal recessive renal glucosuria attributable to a mutation in the sodium glucose co-transporter (SGLT2). Hum Genet 111: 544–547.[CrossRef][Medline]
Wells RG, Pajor AM, Kanai Y, Turk E, Wright EM, and Hediger MA (1992) Cloning of a human kidney cDNA with similarity to the sodium-glucose cotransporter. Am J Physiol Renal Physiol 263: F459–F465.
Wright EM (2001) Renal Na+-glucose cotransporters. Am J Physiol Renal Physiol 280: F10–F18.
Xia X, Lin JT, and Kinne RK (2003) Binding of phlorizin to the isolated C-terminal extramembranous loop of the Na+/glucose cotransporter assessed by intrinsic tryptophan fluorescence. Biochemistry 42: 6115–6120.[CrossRef][Medline]
Zhou L, Cryan EV, D'Andrea MR, Belkowski S, Conway BR, and Demarest KT (2003) Human cardiomyocytes express high level of Na+/glucose cotransporter 1 (SGLT1). J Cell Biochem 90: 339–346.[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  A. M. Brooks and S. M Thacker Dapagliflozin for the Treatment of Type 2 Diabetes Ann. Pharmacother., July 1, 2009; 43(7): 1286 - 1293. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
