Figure 1.
Uptake of the fluorescent glucose analog 2-NBDG is mediated by SGLT2 in KPT2 cells.
KPT2 cells (A) had robust SGLT2 immunolabeling compared to KDT3 cells (B). Increasing doses of (C–F) D-glucose (molar ratios of 1∶0, 1∶1, 1∶50 or 1∶1000 [2-NBDG/D-glucose]), or (G–J) phlorizin (molar ratios of 1∶0, 1∶1, 1∶10 or 1∶50 [2-NBDG/phlorizin]) dose-dependently decreased 2-NBDG fluorescence in KPT2. Scale bar = 20 µm. (K, L). The fluorescence intensity of 2-NBDG in KPT2 cells was significantly decreased with increasing doses of D-glucose (K) or phlorizin (L; **p<0.01).
Figure 2.
SGLT2-mediated uptake of GTTR can be competitively inhibited.
(A–D) Cells were treated with 5 µg/ml GTTR for 20 minutes at 37°C with a dose-range of phlorizin (molar ratios of 1∶0, 1∶5, 1∶10 or 1∶20 [GTTR∶phlorizin]) in DMEM buffer. Scale bar = 20 µm. Increasing doses of (E) phlorizin or (F) D-glucose (molar ratios of 1∶0, 1∶40, 1∶2000 or 1∶40000 [GTTR∶D-glucose]) reduced GTTR fluorescence in KPT2 cells (*p<0.05; **p<0.01). Cell ELISAs demonstrated that (G) GTTR or (H) gentamicin levels in KPT2 cells are decreased by increasing doses of phlorizin.
Figure 3.
SGLT2-mediated uptake of GTTR by KPT2 cells was inhibited by Na+ free buffer.
(A) KPT2 cells were incubated with GTTR for 5 minutes, 10 minutes or 20 minutes at 37°C in Na+ free buffer or Na+ buffer. GTTR fluorescence of KPT2 cell in Na+ buffer for 20 minutes was more intense than in Na+ free buffer (**p<0.01). (B) KPT2 cells were treated with GTTR and phlorizin in DMEM buffer. GTTR uptake by KPT2 cells was also inhibited by phlorizin (100 µg/ml) over time (*p<0.05).
Figure 4.
Heterologous expression of SGLT2 in KDT3 cells increased cellular uptake of GTTR.
(A–C) KDT3-SGLT2 cells with positive SGLT2 immunofluorescence displayed robust GTTR uptake (B, C). (D–F) Empty vector control clones (KPT2-pBabe) showed negligible SGLT2 immunofluorescence (D) and weak, uniform levels of GTTR fluorescence (E, F) compared to (B, C). (H, I) GTTR fluorescence in KDT3-SGLT2 cells in the presence of phlorizin (100 µg/ml) was visibly less intense than in KDT3-SGLT2 cells without phlorizin treatment (B, C). (K, L) GTTR fluorescence in phlorizin-treated KDT3-pBabe cells showed weak levels of GTTR fluorescence as untreated in KDT3-pBabe cells (E, F). Scale bar = 20 µm. (M) Fluorescence intensities of GTTR in KDT3-SGLT2 or KDT3-pBabe cells in the presence or absence of phlorizin (100 µg/ml; **p<0.01).
Figure 5.
Knockdown of SGLT2 reduced gentamicin-induced cytotoxicity.
(A–H) KPT2 cells transfected with siRNA for SGLT2 showed reduced immunoexpression of SGLT2 compared with cells transfected with control siRNA. (A–D) The effect of SGLT2 siRNA began within 1 day of transfection and was most apparent 2 days of transfection. (A–H) The effect SGLT2 siRNA tranfection lasted for at least 5 days. (I) MTT assay on cells (2-days post-transfection) treated with gentamicin for 1, 2 or 3 days revealed greater viability of SGLT2 siRNA-transfected KPT2 cells compared with KPT2 cells treated with control siRNA (**p<0.01).
Figure 6.
SGLT2 immunofluorescence in the kidney and cochlea.
Two different SGLT2 antibodies were used, a rabbit polyclonal IgG to synthetic peptide derived from residues 250–350 of human SGLT2 and a goat polyclonal IgG against a murine peptide sequence within the N-terminal extracellular domain of SGLT2. (A, C) In wild-type mice, SGLT2 was immunolocalized at the apical membranes (arrows) of proximal tubules (p), but not in adjacent glomerular (not shown) or distal tubule (d) regions. (B, D) In Sglt2−/− mice, no immunoexpression for renal SGLT2 was observed with either antibody. (E, F) No labeling above background was observed in cochlear marginal cells of wild-type or Sglt2−/− mice with rabbit antisera for SGLT2. (G, H) Goat antisera for SGLT2 produced labeling patterns in cochlear marginal cells of both wild-type mice and Sglt2−/− mice, suggestive if substantial non-specificity in this cell type. (I, K) In the intra-strial layer of wild-type mice, predominantly composed of both marginal and intermediate cells, both antisera exhibited a punctate labeling pattern not observed in Sglt2−/− mice (J, L). Scale bar = 20 µm. (M) Immunoblotting with the goat antibody for SGLT2 revealed SGLT2 protein expression in wild-type and Sglt2+/− mice, but not Sglt2−/− mice. The ratio of SGLT2 to actin expression in kidney tissues of wild-type and Sglt2+/− mice were significantly higher than that in Sglt2−/− mice. There was no statistical difference in SGLT2 protein expression between wild-type and Sglt2+/− mice. (N) Genotyping demonstrated the absence of wild-type SGLT2 alleles in Sglt2−/− mice.
Figure 7.
Phlorizin decreased renal GTTR uptake, and increased serum drug levels in vivo.
(A) In Sglt2+/− mice, rabbit anti-SGLT2 immunolabeling was predominantly localized at the apical, lumenal region of proximal tubules (p), with negligible labeling in distal tubules (d). (B) GTTR fluorescence was most intense (as saturated puncta) in the apical region of proximal tubules (p), with less intense diffuse labeling in the cytoplasm of these same cells. Very weak and only diffuse GTTR fluorescence was observed in the cytoplasm of distal tubule cells (d). (C) Merged image showing colocalization of SGLT2 (green) and GTTR (red) in proximal tubules. (D–F) When Sglt2+/− mice were pre-treated with phlorizin, significantly reduced GTTR fluorescence was observed in the cytoplasm and apical brush border (arrows) of proximal tubule cells (E) compared to untreated mice (D, F; **p<0.01). (G, I) In Sglt2−/− mice, GTTR fluorescence was diffusely distributed throughout the cytoplasm of proximal tubule cells, with intense fluorescence at the apical brush border. (H, I) Phlorizin had no effect on the uptake, distribution or intensity of GTTR fluorescence in Sglt2−/− proximal tubule cells (**p<0.01). Scale bar = 20 µm. (J, K) In Sglt2+/− mice, phlorizin pre-treatment significantly increased both gentamicin and GTTR serum levels compared to vehicle treated control mice (*p<0.05). In Sglt2−/− mice, phlorizin did not significantly change gentamicin or GTTR serum levels. However, serum levels of gentamicin or GTTR serum level were significantly higher in Sglt2−/− mice than in Sglt2+/− mice in the absence of phlorizin treatment (*p<0.05).
Figure 8.
Loss of SGLT2 function had no effect on cochlear uptake of GTTR or auditory function.
In the stria vascularis, GTTR was localized in marginal (A, E) and intermediate (B, F) cells of Sglt2+/− (A, B) and Sglt2−/− (E, F) mice. The nucleoplasm of marginal and intermediate cell nuclei displayed weak labeling. (C, D, G, H) Phlorizin had no effect on the uptake or distribution of GTTR fluorescence in the stria vascularis of Sglt2+/− or Sglt2−/− mice. Scale bar = 20 µm. (I, J) Wild-type, Sglt2+/− and Sglt2−/− mice, at 6 or 12 weeks of age, displayed no significant differences in ABR thresholds.
Figure 9.
Phlorizin does not ameliorate kanamycin-induced ototoxicity.
Three weeks after dosing, mice treated with kanamycin in DPBS or mice treated with kanamycin plus DMSO (vehicle for phlorizin) had significant ABR threshold shifts at 32 kHz only compared to mice treated with DPBS only (**p<0.01). The kanamycin plus phlorizin group had significantly different threshold shifts at 4, 8 and 16 kHz compared to the kanamycin in DPBS group; and significantly different threshold shifts at 4, 8, 16 and 32 kHz compared DPBS only group (*p<0.05; **p<0.01). However, no significant threshold shifts were observed between kanamycin plus phlorizin (in DMSO) and kanamycin plus DMSO groups.