The renal proximal tubule reabsorbs 90% of the filtered glucose load through the Na-coupled glucose transporter SGLT2, and specific inhibitors of SGLT2 are now available to patients with diabetes to increase urinary glucose excretion. Using expression cloning, we identified an accessory protein, 17 kDa membrane-associated protein (MAP17), that increased SGLT2 activity in RNA-injected Xenopus oocytes by two orders of magnitude. Significant stimulation of SGLT2 activity also occurred in opossum kidney cells cotransfected with SGLT2 and MAP17. Notably, transfection with MAP17 did not change the quantity of SGLT2 protein at the cell surface in either cell type. To confirm the physiologic relevance of the MAP17-SGLT2 interaction, we studied a cohort of 60 individuals with familial renal glucosuria. One patient without any identifiable mutation in the SGLT2 coding gene (SLC5A2) displayed homozygosity for a splicing mutation (c.176+1G>A) in the MAP17 coding gene (PDZK1IP1). In the proximal tubule and in other tissues, MAP17 is known to interact with PDZK1, a scaffolding protein linked to other transporters, including Na/H exchanger 3, and to signaling pathways, such as the A-kinase anchor protein 2/protein kinase A pathway. Thus, these results provide the basis for a more thorough characterization of SGLT2 which would include the possible effects of its inhibition on colocalized renal transporters.

Although water permeation across cell membranes occurs through several types of membrane proteins, the only permeation mechanism resolved at atomic scale is that through aquaporins. Crystallization of the Vibrio parahaemolyticus sodium-galactose transporter (vSGLT) allows investigation of putative water permeation pathways through both vSGLT and the homologous human Na-glucose cotransporter (hSGLT1) using computational methods. Grand canonical Monte Carlo and molecular dynamics simulations were used to stably insert water molecules in both proteins, showing the presence of a water-filled pathway composed of ∼100 water molecules. This provides a structural basis for passive water permeation that is difficult to reconcile with the water cotransport hypothesis. Potential-of-mean-force calculations of water going through the crystal structure of vSGLT shows a single barrier of 7.7 kCal mol(-1), in agreement with previously published experimental data for cotransporters of the SGLT family. Electrophysiological and volumetric experiments performed on hSGLT1-expressing Xenopus oocytes showed that the passive permeation pathway exists in different conformational states. In particular, experimental conditions that aim to mimic the conformation of the crystal structure displayed passive water permeability. These results provide groundwork for understanding the structural basis of cotransporter water permeability.

The Na+/glucose cotransporter (SGLT1) is the archetype of membrane proteins that use the electrochemical Na+ gradient to drive uphill transport of a substrate. The crystal structure recently obtained for vSGLT strongly suggests that SGLT1 adopts the inverted repeat fold of the LeuT structural family for which several crystal structures are now available. What is largely missing is an accurate view of the rates at which SGLT1 transits between its different conformational states. In the present study, we used simulated annealing to analyze a large set of steady-state and pre–steady-state currents measured for human SGLT1 at different membrane potentials, and in the presence of different Na+ and α-methyl-d-glucose (αMG) concentrations. The simplest kinetic model that could accurately reproduce the time course of the measured currents (down to the 2 ms time range) is a seven-state model (C1 to C7) where the binding of the two Na+ ions (C4→C5) is highly cooperative. In the forward direction (Na+/glucose influx), the model is characterized by two slow, electroneutral conformational changes (59 and 100 s−1) which represent reorientation of the free and of the fully loaded carrier between inside-facing and outside-facing conformations. From the inward-facing (C1) to the outward-facing Na-bound configuration (C5), 1.3 negative elementary charges are moved outward. Although extracellular glucose binding (C5→C6) is electroneutral, the next step (C6→C7) carries 0.7 positive charges inside the cell. Alignment of the seven-state model with a generalized model suggested by the structural data of the LeuT fold family suggests that electrogenic steps are associated with the movement of the so-called thin gates on each side of the substrate binding site. To our knowledge, this is the first model that can quantitatively describe the behavior of SGLT1 down to the 2 ms time domain. The model is highly symmetrical and in good agreement with the structural information obtained from the LeuT structural family.

The accuracy of the Microenvironment Modulated_Screened Coulomb Potential (MM_SCP) 1 to predict pKa values for titratable residues in biological macromolecules is analyzed. This self-consistent approach calculates electrostatic energetics using a modified form of the SCP-continuum solvent model 2 , which is modulated by the hydrophobicity of the local environment around the titratable group. The analysis has been applied to 345 ionizable groups (Asp, Glu, His, Cys, Tyr, Lys, N-and C-termini) in 59 proteins belonging to multiple structural classes. Among these residues, 82% were predicted with errors < 1 unit), although half of residues with large shifts were correctly predicted (error <1 unit). Further analysis showed that the MM_SCP accounts for the pKa shifts of ionizable groups in hydrophobic microenvironment 3 and treats interaction energies very well. Most errors originate from steric clashes or improper handling of certain interaction pairs (e.g., ionic H-bonds). A feature in the MM_SCP was used to identify steric clashes between the titratable residue and other parts of the sequence; after local minimization the calculated pKa improved by half a unit in more than half the cases. Other methods to improve the predictions will be discussed, such as optimizing the H-bond network, or using refined reference pKa values 4 which improves the root mean square deviations by at least 0.1 unit for 25 proteins.

No abstract