Vesicle-associated membrane protein 2 (VAMP2) but Not VAMP3 mediates cAMP-stimulated trafficking of the renal Na+-K+-2Cl- co-transporter NKCC2 in thick ascending limbs

Abstract

In the kidney, epithelial cells of the thick ascending limb (TAL) reabsorb NaCl via the apical Na(+)/K(+)/2Cl(-) co-transporter NKCC2. Steady-state surface NKCC2 levels in the apical membrane are maintained by a balance between exocytic delivery, endocytosis, and recycling. cAMP is the second messenger of hormones that enhance NaCl absorption. cAMP stimulates NKCC2 exocytic delivery via protein kinase A (PKA), increasing steady-state surface NKCC2. However, the molecular mechanism involved has not been studied. We found that several members of the SNARE family of membrane fusion proteins are expressed in TALs. Here we report that NKCC2 co-immunoprecipitates with VAMP2 in rat TALs, and they co-localize in discrete domains at the apical surface. cAMP stimulation enhanced VAMP2 exocytic delivery to the plasma membrane of renal cells, and stimulation of PKA enhanced VAMP2-NKCC2 co-immunoprecipitation in TALs. In vivo silencing of VAMP2 but not VAMP3 in TALs blunted cAMP-stimulated steady-state surface NKCC2 expression and completely blocked cAMP-stimulated NKCC2 exocytic delivery. VAMP2 was not involved in constitutive NKCC2 delivery. We concluded that VAMP2 but not VAMP3 selectively mediates cAMP-stimulated NKCC2 exocytic delivery and surface expression in TALs. We also demonstrated that cAMP stimulation enhances VAMP2 exocytosis and promotes VAMP2 interaction with NKCC2.

Keywords: Cyclic AMP (cAMP); Epithelial Cell; Epithelial Sodium Transport; Exocytosis; Kidney; Membrane Trafficking; Na-K-Cl Co-transporter 2 (NKCC2); SNARE Proteins.

FIGURE 1.

Expression of VAMPs and target membrane SNAREs in the TAL. A, VAMP isoforms 2, 3, 7, and 8 were found in isolated TALs (n = 4). B, SNAP isoforms 23, 29, and 47 are expressed in isolated TALs. SNAP-25, although highly expressed in brain tissue, was not found in the TAL (n = 3). C, expression of syntaxin isoforms 3, 4, and 7 was observed in TALs (n = 3).

FIGURE 2.

VAMP2 co-immunoprecipitates with NKCC2. A, co-immunoprecipitation between NKCC2 and different VAMPs revealed NKCC2 interaction (direct or indirect) with VAMP2 but not VAMP7 or VAMP8 in rat TAL lysates (200 μg of protein input; n = 4). B, reciprocal immunoprecipitation with an anti-VAMP2 antibody co-precipitated NKCC2 from TAL lysates (100 μg of protein input; n = 4). Control IgG failed to precipitate VAMP2/7/8 or NKCC2.

FIGURE 3.

NKCC2-VAMP2 co-localization at the apical surface of TALs. A, diagram showing the exocytic process that allows immunodetection of surface NKCC2 and surface VAMP2-eGFP at the apical membrane of cultured TAL cells before fixation. B, apical distribution of surface NKCC2 (red) and surface VAMP2 (green) in cultured TALs. Apical surface NKCC2 was distributed in clusters. 45 ± 7% of NKCC2 clusters also contained VAMP2 (co-localizing pixels) (n = 5).

FIGURE 4.

Silencing of VAMP2 and VAMP3 in the renal medulla by in vivo gene transfer. A, Western blot showing decreased VAMP2 and VAMP3 protein expression in NRK-52E cells after transfection with nude siRNAs. GAPDH expression was not affected (n = 3). B, Western blot showing VAMP2 protein expression in NRK-52E cells after transduction with adenoviruses carrying the corresponding shRNA. VAMP2-shRNA decreased VAMP2 expression but not VAMP3, VAMP7, or VAMP8 expression (n = 3). C, decrease in VAMP2 and VAMP3 protein expression after in vivo silencing in rat TALs. VAMP2 silencing decreased VAMP2 expression by 69 ± 7% compared with control TALs from the same rat. VAMP3 silencing decreased VAMP3 expression by 68 ± 10%. Values are expressed as mean percentage of scrambled shRNA, and error bars represent S.E. *, p < 0.01 versus scrambled shRNA (n = 3). D, representative Western blots from TALs isolated from rats injected with adenoviruses carrying VAMP2-shRNAs or VAMP3-shRNAs after 3 days of transduction. VAMP2-shRNA decreased VAMP2 but not VAMP3 expression and vice versa (n = 3).

FIGURE 5.

VAMP2 silencing decreases cAMP-stimulated steady-state surface NKCC2 expression in rat TALs. A, TALs were treated with forskolin (20 μm) plus IBMX (0.5 mm) for 30 min at 37 °C to stimulate intracellular cAMP production. Steady-state surface NKCC2 was measured by surface biotinylation in adenovirus-transduced TALs. VAMP2 silencing decreased cAMP-stimulated steady-state surface NKCC2, but VAMP3 silencing did not (scrambled, 79 ± 7% stimulation; VAMP2-shRNA, 45 ± 6% stimulation; VAMP3-shRNA, 76 ± 9% stimulation; n = 7). B, VAMP2 silencing did not affect baseline constitutive steady-state surface NKCC2 (scrambled, 100%; VAMP2-shRNA, 91 ± 8%; n = 7). C, VAMP2 silencing did not affect total NKCC2 protein expression (scrambled, 100%; VAMP2-shRNA, 86 ± 7%; n = 7). Values are expressed as mean percentage of scrambled shRNA. Error bars, S.E.; *, p < 0.05 versus scrambled shRNA.

FIGURE 6.

VAMP2 silencing blocks cAMP-stimulated NKCC2 exocytic delivery in rat TALs. Exocytic delivery of NKCC2 was measured as biotinylated NKCC2 at the TAL surface after masking with NHS-acetate (representative blot in the top panel). Constitutive NKCC2 exocytic delivery at 30 min was not affected by VAMP2 silencing (scrambled, 49 ± 9%; VAMP2-shRNA, 55 ± 13%), but cAMP-stimulated NKCC2 exocytic delivery was completely blocked (scrambled, 93 ± 16%; VAMP2-shRNA, 51 ± 14%). Values are expressed as mean percentage of NHS-acetate-masked fraction. Error bars, S.E.; *, p < 0.05 versus vehicle/scrambled shRNA (n = 5).

FIGURE 7.

VAMP2 silencing does not affect NKCC2 phosphorylation at Ser-126 in TALs. NKCC2 phosphorylation at Ser-126 (pSer-126) was measured with an anti-phospho antibody. Baseline NKCC2 phosphorylation at Ser-126 was undetectable and was not affected by VAMP2 silencing. Stimulation with 100 μm db-cAMP for 20 min enhanced NKCC2 Ser-126 phosphorylation 69 ± 14-fold in control TALs and 73 ± 24-fold in VAMP2-shRNA-transduced TALs. The degree of stimulation by db-cAMP was not statistically significant. Values are expressed as mean -fold increase from vehicle/scrambled shRNA, and error bars represent S.E. *, p < 0.05 versus vehicle (n = 4).

FIGURE 8.

cAMP stimulates exocytic delivery of VAMP2 to the TAL cell surface. A, primary cultures of TALs were transfected with VAMP2-eGFP, and the time course of exocytic delivery was measured as biotinylated VAMP2-eGFP at the cell surface after masking with NHS-acetate. Biotinylated VAMP2-eGFP signal decreased on average by 78 ± 1% after masking with NHS-acetate (time 0), and it gradually recovered after 20 and 30 min at 37 °C. B, representative Western blot showing increase in VAMP2-eGFP exocytic delivery after stimulation of cAMP with forskolin (20 μm) plus IBMX (0.5 mm) for 30 min. C, cAMP stimulation enhanced the rate of VAMP2-eGFP delivery to the cell surface from 4.8 ± 0.9 to 8.7 ± 1.4 arbitrary units (A.U.)/min. Arbitrary units are defined as percentage of the NHS-acetate-masked fraction. Error bars, S.E.; *, p < 0.05 versus vehicle (n = 4).

FIGURE 9.

cAMP stimulates exocytosis of individual VAMP2-carrying vesicles in renal epithelial cells. A, representative time course of a real-time exocytic event in LLC-PK1 cells. Cultured cells were transfected with VAMP2-eGFP, and transient exocytic events were measured in real time by TIRF microscopy. B, representative trace of the change in fluorescence over time within a region of interest where an exocytic event shorter than 20 s occurred. a.u., arbitrary units. C, representative trace of the change in fluorescence over time within a region of interest where a VAMP2-eGFP vesicle remained in close proximity to the membrane for a time lapse of at least 65 s before disappearing. D, quantification of the type of events described in B. Stimulation with cAMP increased the number of VAMP2-eGFP exocytic events from 10 ± 4 to 87 ± 8 events/min/area. Values are expressed as the number of exocytic events/min/area. Error bars, S.E.; *, p < 0.01 versus control (n = 3).

FIGURE 10.

Stimulation of PKA enhances VAMP2-NKCC2 interaction in TALs. Incubation of TAL suspensions with a 1 mm concentration of the PKA-selective agonist N⁶Bnz-cAMP for 30 min enhanced NKCC2-VAMP2 interaction by 108 ± 12%, as measured by co-immunoprecipitation. Treatment with N⁶Bnz-cAMP did not affect NKCC2 protein levels (top). Values are expressed as mean percentage of vehicle, and error bars represent S.E. *, p < 0.01 versus vehicle (n = 5).