Figure 1.
Aquaporin-7 (AQP7) and -10 (AQP10) mRNA expression in human subcutaneous adipose tissue (AT) and in isolated adipocytes (A).
Upper panels, mRNA levels were measured by real-time RT-PCR relative to the β-actin internal standard (see Materials and Methods section) and the values obtained were reported as ΔCt. Bars represent the mean ± SEM of at least 4 different experiments each from different RNA extracts. *P<0.05 versus A (Student’s t test). Lower panels, Gel electrophoresis of the PCR products. Specific PCR products for AQP7 (139 bp band), AQP10 (115 bp band) and β-actin (146 bp) were observed in both AT and A.
Figure 2.
Aquaporin-7 (A), and -10 (B) protein expression in human adipocytes plasma membranes.
Blots representative of four were shown. Lanes were loaded with 35 µg of proteins, probed with anti-AQP7 rabbit polyclonal antibody (A left) and processed as described in Materials and Methods. The same blots were stripped and re-probed with anti-AQP10 rabbit polyclonal antibody (B left), and anti-β-actin antibody as housekeeping (C). A major band of about 34 kDa and two bands of about 30 kDa (monomer) and 60 kDa (dimer) were observed when the blots were probed with anti-AQP7 and anti-AQP10 antibodies respectively. No bands were detected when preadsorbed anti-aquaporin-7 or -10 antibodies were used (A and B right). AM, adipocytes plasma membranes. D, duodenal crude homogenate.
Figure 3.
Representative immunofluorescence confocal microscopical images of aquaporin-7 and -10 localization in human subcutaneous adipose tissue.
Green labeling indicates the presence of aquaporin-7 (A) and -10 (B) at the plasma membrane and at the cytoplasm of the human adipocytes. No or faint staining was observed when anti-aquaporin-7 and anti-aquaporin-10 preadsorbed antibodies were used (C-D). Nuclei were counterstained by DAPI (blue). Colocalization of aquaporin-7 or -10 and the endothelial cell marker CD34 in human subcutaneous adipose tissue (E-F) was performed as detailed in “Materials and methods”. Green labeling indicated the presence of aquaporin-7 (E) or -10 (F), red labeling the vessels, while nuclei were counterstained by DAPI (blue). Merged images showed strong colocalization signal of aquaporin-7 and CD34 (yellow labelling) in the capillary, even though aquaporin-7 was also expressed in the adipocytes (E). On the contrary, aquaporin 10 and CD34 did not colocalized (F).
Figure 4.
Effect of insulin and isoproterenol on subcellular localization of aquaporin-7 (AQP7) and -10 (AQP10) in human cultured differentiated adipocytes.
The aquaporin localization was detected in basal conditions (control: A, B), after insulin (C, D) or isoproterenol (E, F) stimulation. Control adipocytes show an intracellular AQP7 and AQP10 green labeling, particularly evident around small lipid droplets. Green labeling indicated the presence of AQP-7-10, while nuclei were counterstained by DAPI (blue). Insulin treatment increased the AQP7 and AQP10 staining around the lipid droplets while that of isoproterenol reduced the lipid droplets labeling, greatly increasing the plasma membrane staining. Negative controls gave a faint or negligible signal (H). Accumulation of lipid droplets in adipocytes was demonstrated by Oil Red O staining (G).
Figure 5.
Water and glycerol permeability of human adipocyte plasma membrane vesicles.
(A) Representative light scattering curves were obtained by exposing the isolated adipocytes to a 150 mOsm osmotic gradient in two different conditions: normal untreated cells (Control) and cells treated for 15 min with 0.5 M DMSO (DMSO). (B) Bars represent water permeability of isolated adipocytes, expressed as relative k. Values are means ± SEM of at least 15 single shots for each of five different adipocyte preparations. *, P<0.05 vs Control (Student t test for pair data). (C) Representative light scattering curves were obtained by exposing the adipocytes plasma membrane vesicles to a 150 mOsm osmotic gradient in three different conditions: normal untreated vesicles (Control), vesicles treated for 10 min with 1 mM HgCl2, vesicles treated with 1 mM HgCl2 followed by 15 min treatment with 15 mM β-mercaptoethanol (β-ME). (D) Bars represent water permeability of adipocyte plasma membrane vesicles, expressed as relative k. Values are means ± SEM of at least 8 single shots for each of five different preparations. *, P<0.05 vs Control and β-ME (repeated measure ANOVA, followed by Newman-Keuls’s Q test). (E) Representative light scattering curves were obtained by exposing the adipocytes plasma membrane vesicles to a 150 mM inwardly directed gradient of glycerol. The initial increase in light scattering results from osmotic water efflux caused by vesicle shrinkage (before the dashed line), while the subsequent slower decrease is caused by glycerol entry (after the dashed line). (F) Water (Pf) and glycerol (Pgly) permeability coefficients were calculated as described in Materials and Methods.
Figure 6.
Aquaporin-10 (AQP10) silencing in human differentiated adipocytes.
AQP10 short interfering RNA (siRNA) and scrambled siRNA (Ctr) were transfected in differentiated adipocytes as described in Materials and methods. A, AQP10 mRNA levels were measured by real-time RT-PCR relative to the β-actin internal standard and the values obtained were reported as fold change (see Materials and Methods section). Bars represent the mean ± SEM of at least 4 different experiments each from different RNA extracts. AQP10 transcript was reduced in silenced differentiated adipocytes compared to controls; *P<0.006 versus Ctr (Student’s t test). B, Western blot and densitometry demonstrate that undifferentiated adipose stem cells (ASC) had no AQP10 expression, while AQP10 protein was reduced in silenced differentiated adipocytes (siRNA) compared to controls (scrambled; Ctr)(*, P<0.003; Student’s t test). Blots representative of three were shown (B, lower pannel). The same blots were stripped and re-probed with anti-β-actin antibody. Bands of the expected molecular weights were shown and acquired with the Image Master VDS (GE Healthcare Life Sciences, Italy). Densitometric analysis of the bands was performed by Total Lab V 1.11 computer program (GE Healthcare Life Sciences, Italy) and the results were normalized to the corresponding β-actin (B, upper panel). C, Water (Pf) and glycerol (Pgly) permeability coefficients were calculated as described in Materials and Methods. Both Pf and Pgly were significantly reduced in siRNA compared to controls by 46% and 51%, respectively (*, P<0.002; Student’s t test).