Fig 1.
FXYD1 is implicated in urinary concentration.
Box and whisker plot of the osmolality of daytime (afternoon) samples of male mouse WT and knockout (KO) urine at baseline and after water deprivation, showing a difference in baseline osmolality but the near-normal ability of the KO to concentrate its urine. The asterisks indicate a P value of <0.0001, 2-tailed Student’s t-test. When calculated as average ± SEM the results were as follows. WT control conditions, 2,019 ± 143, n = 20; KO control conditions 1,161 ± 123, n = 21; Student’s t-test, P < 0.0001. WT after 36 hours water deprivation, 4,224 ± 140, n = 8; KO water deprivation, 3,716 ± 383, n = 8; t-test, P = 0.26. Female mice were also tested in control conditions, and the results for baseline osmolality were WT, 2,169+/-92, n = 7; KO, 1,085+/-108, n = 7, t-test P < 0.001.
Fig 2.
Wild type mice were deprived of water for 36 hours to elicit water conservation mechanisms.
(A) Water deprivation revealed an intracellular pool of FXYD1 that was mostly not co-localized with basolateral Na,K-ATPase in mouse IMCD. Kidney sections from male WT control (a-c) or water deprived mice (d-f) were stained for α1 subunit of Na, K-ATPase (a,d) and FXYD1 (b,e). FXYD1 detection with the PLM-C1 antibody was enhanced in water-deprived mice. Nuclei were co-stained with TO-PRO-3 (blue). Bars, 20 microns. The images were enhanced and sharpened by subjecting the entire field to the high-pass filter in Adobe Photoshop. (B) Membranes from inner medulla from control or water-deprived (WD) animals were tested on blots with antibodies specific to the α1 subunit of Na,K-ATPase, FXYD2a, or FXYD1 (PLM-C2). (+) Canine cardiac sarcolemma was used as a positive blot control for FXYD1. The blot is representative of three independent experiments.
Fig 3.
In wild type IMCD, FXYD1 was phosphorylated at the basal state and dephosphorylated by vasopressin treatment.
(A) Dephosphorylation of FXYD1 by exogenous phosphatase in situ revealed enhancement of detection by PLM-C1 antibody, which is specific for dephospho-FXYD1. Fresh-frozen kidney sections from male wild type mice were treated with vehicle (a) or Lambda protein phosphatase (b), fixed with 4% PLP, and stained with PLM-C1. The data confirm that FXYD1 is phosphorylated at basal conditions. Bars, 50 microns. (B) FXYD1 phosphorylation under basal conditions was detected on blots. Wild type mice were injected with either vehicle (-) or dDAVP (IP, 1 μg/kg). Four hours post-injection membranes from inner medulla were prepared and tested on Western blots with PLM-C2 antibody, also specific for non-phosphorylated protein, and FXYD1 phospho-specific antibodies CP63 and CP689 that recognize phosphorylated Ser63 and Ser68/Thr69 [18], respectively. Vasopressin elicited reciprocal changes in antibody signal.
Fig 4.
Vasopressin induced trafficking of FXYD1 in wild type IMCD in vivo.
Wild type mice were injected with dDAVP (1 μg/kg) and sacrificed at 0 time (a-c), 4 hours (d-f), and 16 hours (g-i) post-injection. Fixed kidneys were sectioned, and slides were stained with antibodies against AQP2 (a,d,g) and PLM-C1 antibodies against FXYD1 (b,e,h). dDAVP stimulated the trafficking of FXYD1 from an intracellular location towards apical membrane. Co-localization of FXYD1 and AQP2 at apical membrane is seen in yellow after 4 hours (f). This figure is representative of 3 independent experiments. Nuclei were labeled with TO-PRO-3. Bars, 10 μm.
Fig 5.
AQP2 acute response to vasopressin differed in slices of inner medulla from wild type and Fxyd1 knockout mice.
Kidney slices from WT (a-c) and KO (d-f) male mice were fixed without treatment (a,d), incubated in HBSS to wash out endogenous VP (b,e), and then treated with dDAVP for 15 min (c,f). Slices were post-fixed with 2% PLP and sectioned. The distribution of AQP2 was monitored by immunofluorescence. The absence of FXYD1 reduced shuttling of AQP2 between intracellular vesicles and subapical spaces and apical membrane. This figure is representative of 3 independent experiments. Nuclei were labeled with Dapi. Bars, 10 μm.
Fig 6.
Washout of dDAVP revealed rapid redistribution of AQP2 in slices of inner medulla from Fxyd1 knockout mice.
Slices from WT (a) and KO mice (b) treated with dDAVP for 15 min as in Fig 5 were washed in HBSS for 60 min. Slices were fixed with PLP, sectioned, and stained for AQP2. The absence of FXYD1 resulted in less juxta-apical and more heterogeneous localization of AQP2 in kidney from KO mice (b). Arrows point to regions with AQP2 in proximity to the basolateral membrane. The images were enhanced and sharpened by subjecting the entire field to the high-pass filter in Adobe Photoshop. Bar, 10 μm.
Fig 7.
AQP2 abundance was reduced in FXYD1 knockout mice.
(A, B) Fxyd1-/- mice had a lower abundance of AQP2 in inner medulla. Inner medulla from WT and KO mice was obtained as lysates and tested on blots with specific antibodies [representative blot in (A)]. Both core (c) and glycosylated (g) species of AQP2 were reduced in KO animals. (B) quantification of results of three experiments. (C, D) Blot quantification of the relative changes in inner medullary AQP2 levels after 2 hour treatment of mice with dDAVP, 3 experiments. Stimulation with dDAVP resulted in a similar increase of AQP2 recovered in pelleted crude membranes in both WT (C) and KO (D). Bars are ± S.E.M. and significance was evaluated by Student’s t-test.
Fig 8.
Flotation experiments revealed a difference in distribution of AQP2 in detergent-resistant membranes (DRM) from WT and knockout mouse.
Membranes from inner medulla from wild type (solid lines in A and B) or FXYD1 KO male mice (dashed lines in A and B) treated with either vehicle (A) or dDAVP (B) for 1 hour were solubilized with 1% Triton, 30 min, 4°C, and separated on a step sucrose gradient of 5, 30, and 40% sucrose. DRM float near the top of the gradient (peaks 1 and 2). The distribution of AQP2 was quantified on Western blots with a cooled-CCD imager and expressed as % of total recovered AQP2 in each fraction. The data are mean ± SEM of four experiments. (A) Under basal conditions, the DRM AQP2 fraction from vehicle-treated knockout mice peaked in peak 1, while WT DRM appeared in peak 2. Roughly half of the AQP2 was in denser membranes (peak 3). (B) Stimulation with dDAVP caused redistribution of DRM AQP2 from the KO to peak 2. (C) Vasopressin-stimulated recruitment of AQP2 from peak 3 to DRM membranes in peak 2 occurred in WT but not in KO. (**) The difference is significant by Student’s t-test, P < 0.005, n = 4.
Table 1.
The daytime defect in urine concentration in Fxyd1-/- mice was compensated over 24 h.