Fig 1.
Western blot of whole cell (WC) homogenates and apical membrane fractions from two immortalized hRPTC lines grown in Transwell membranes.
Arrows on the left-hand side of the blots indicate the molecular sizes, while the arrows on the right-hand side indicate the molecular sizes of the bands of the proteins of interest. The blot was probed for NBCe2 along with the following membrane markers: CD-13 (APN microvilli marker), Na+,K+/ATPase (basolateral membrane marker), and NBCe1 (basolateral membrane marker). The results demonstrate that two isoforms of NBCe2 exist in the apical membrane of polarized RPTCs grown on Transwells. The control western blots demonstrate that αNa+,K+/ATPase (NKA) and NBCe1 only stain weakly in the apical fraction, probably due to some basolateral contamination inherent in this kind of preparation. WC homogenates demonstrate the presence of NBCe2, NBCe1, CD-13, and Na+,K+/ATPase, as expected.
Fig 2.
Immunofluorescence localization of NBCe2 in human renal proximal tubule cells (hRPTCs) carrying wild-type (WT) or homozygous variant (HV) SLC4A5 imaged on Petri dishes using confocal microscopy.
CD-13, a specific membrane-bound ectopeptidase present in RPT but not in other nephron segments, in the membrane is stained green using a more photostable fluor (CD-13-Alexa 488 antibody) (A, D, G, and J), NBCe2 is stained red (B, E, H, and K), nucleus is stained blue in all the images, including merged images (C, F, I, and L). With monensin treatment (10 μmol/L 24 hr) which increases intracellular sodium (↑Na+), NBCe2 expression gets more diffuse similar to the apical membrane stain (E and F, K and L), especially for HV cells. Panels C, F, I, and L show merged images of NBCe2 and CD-13; there is increased NBCe2 on the surface in HV cells treated with monensin (K and L). The scale bar in panel C, F, I, and L = 10 μm.
Fig 3.
Representative TIRFM images of hRPTCs growing on GEM™ 3D microcarriers.
NBCe2 is labeled with a red Alexa 568 fluorescent dye. WT hRPTCs were incubated in vehicle (VEH) (A) and imaged at 70 nm into the plasma membrane to determine intramembranous presence of NBCe2. After 24 h of exposure to high intracellular sodium there was no significant increase in membranous localization of NBCe2 as determined by counting punctate fluorescent spots (B) (quantified in Fig 4). In HV hRPTCs NBCe2 is also expressed in the membrane in VEH treated hRPTCs to a similar level as WT VEH (C) but is increased when the HV hRPTCs were exposed to high intracellular sodium (D)(quantified in Fig 4).
Fig 4.
Colocalization analysis of NBCe2 and CD-13 in hRPTC.
The number of punctate dots per cell was measured using the Fuji program in both confocal and TIRFM images. We counted punctate dots depicting NBCe2 in hRPTCs grown on Petri dishes (conventional cell culture) using a confocal microscope (A) as well as on cells grown as spheroids on novel 3D cell culture hydrogel spheres (GEM™) using our total internal reflectance microscope (TIRFM) (B) TIRFM examines only 70 nm at the surface of the apical membrane. Conventional 2D and the novel 3D cell culture demonstrates significantly increased NBCe2 on the apical (CD-13 positive) cell membrane under high salt conditions (↑Na+) for SLC4A5 homozygous variant (HV) cells. (2D Petri dish, P<0.05, two-way ANOVA; 3D GEM™, P<0.05 two-way ANOVA Holm-Sidak test).
Fig 5.
Intracellular sodium-induced NBCe2 protein expression in hRPTCs carrying Wild-Type (WT) or homozygous variant (HV) SLC4A5.
A) Increasing extracellular sodium from 120 to170 mmol/L for 24 h increases intracellular sodium from 6.1±0.3 to 11.3±1.1 mmol/L and increases NBCe2 protein in HV but not WT hRPTCs (N = 4, *P<0.01 vs 120 mmol/L sodium, two-way ANOVA, Holm-Sidak test). B) Monensin (10 μmol/L, 24h), which increases intracellular sodium (↑Na+) by 6.7±2.3 mmol/L)), slightly increases NBCe2 protein in all four WT cell lines (each cell line from a different individual) but markedly increases NBCe2 expression in all six HV cell lines (each cell line from a different individual) (N = 8–12, *P<0.05 vehicle (VEH) vs monensin (↑Na+) with each cell line).
Fig 6.
NBCe1 and NBCe2 protein and mRNA expression in hRPTCs carrying wild-type (WT) or homozygous variant (HV) SLC4A5.
A) NBCe2 protein. NBCe2 protein is increased following an increase in intracellular sodium by 6.7±2.3 mmol/L for 24 h (monensin, 10 μmol/L) (↑Na+) in HV but not WT SLC4A5 hRPTCs (N = 12, P<0.0001 vs vehicle (VEH), two-way ANOVA, Holm-Sidak test). B) NBCe1 protein. NBCe1 protein is decreased (HV<WT) following an increase in intracellular sodium (monensin, 10 μmol/L, 24 h) (↑Na+) in WT SLC4A5 hRPTCs (N = 36, *P<0.001 vs VEH, two-way ANOVA, Holm-Sidak test) and HV hRPTCs (N = 48, *P<0.05 vs VEH and #P<0.001 vs WT monensin (↑Na+), two-way ANOVA, Holm-Sidak test). C) SLC4A5 mRNA. SLC4A5 mRNA is increased following an increase in intracellular sodium (monensin, 1 μmol/L, 24 h) (↑Na+) in HV but not WT SLC4A5 hRPTCs (N = 9, *P<0.01 vs. others, two-way ANOVA, Holm-Sidak test). D) SLC4A4 mRNA. SLC4A4 mRNA is decreased following an increase in intracellular sodium (monensin, 1 μmol/L, 24 h) (↑Na+) only in WT SLC4A5 hRPTCs (N = 9, *P<0.05 vs. VEH, one-tailed t-test); the apparent decrease in HV is not statistically significant.
Fig 7.
NBCe1 protein expression regulation by D1-like (D1R and D5R) dopamine receptor agonist in hRPTCs carrying wild-type (WT) SLC4A5.
NBCe1 protein expression is decreased following an increase in intracellular sodium (↑Na+) (10 μmol monensin/L, 24 h) (N = 4, *P<0.001 vs VEH, one-way ANOVA, Tukey’s test) in WT SLC4A5 hRPTC. The combination of monensin (↑Na+) and D1R/D5R agonist SKF38393 (SKF, 10 μmol/L, 24 h) further decreases NBCe1 protein (N = 4, **P<0.05 vs monensin (↑Na+), #P<0.001 vs SKF, one-way ANOVA, Tukey’s test) that is blocked by the D1-like receptor antagonist LE300 (10 μmol/l, 24 h), which by itself has no effect. NBCe1 protein is not affected by SKF or LE300 in the absence of monensin.
Fig 8.
HNF4A expression and binding in hRPTCs carrying wild-type (WT) or homozygous variant (HV) SLC4A5.
A) HNF4A protein. HNF4A protein is increased following treatment with monensin (10 μmol/L, 24 h) which increases intracellular sodium (↑Na+) in both WT and HV SLC4A5 hRPTCs (N = 8, *P<0.05 vs WT VEH; N = 12, **P<0.0001 vs HV VEH, two-way ANOVA, Holm-Sidak test). B) HNF4A binding at SLC4A5 SNP site using ChIP. Approximately 15 million hRPTCs carrying either WT or HV SLC4A5 were cross-linked and immunoprecipitated with an HNF4A antibody. Magnetic protein A/G was added to capture HNF4A antibody (sc-6556) and any corresponding protein-DNA complex. Samples were then eluted, uncross-linked, and purified for DNA. RT-PCR, using primers flanking the SLC4A5 SNP site, indicates increased binding of HNF4A to SLC4A5 in HV hRPTCs. (N = 3, *P<0.05, t-test) C) In vitro oligonucleotide binding assay. C-myc-tagged HNF4A protein was added to a solution of double-stranded oligonucleotides labeled with biotin. Oligonucleotides consisted of DNA sequences of WT or HV SLC4A5 alleles. After incubation for 30 min, streptavidin647 was added to label the oligonucleotides; anti-c-Myc antibody was then added followed by magnetic protein A/G to capture HNF4A-oligonucleotide complexes. The samples were washed and read on a microplate reader. HNF4A binding is increased in HV relative to WT sequence (t-test), in agreement with the ChIP data (Fig 8B). D) HNF4A Expression in WT and HV hRPTCs. V5 Protein in empty vector (control WT, HV) and V5 epitope-tagged HNF4A transfection in 3 WT (WT HNF4A) and 3 HT (HV HNF4A) hRPTC cell lines were measured by in-cell western. V5 protein expression is similar in empty vector-transfected and V5 epitope-tagged HNF4A transfected cells. (N = 9, *P<0.001 vs vector controls, one-way ANOVA, Holm-Sidak test) E) Total HNF4A expression in empty vector- and HNF4A-transfected WT and HV hRPTCs. Empty vector control (VEH, WT, HV) and V5 epitope-tagged HNF4A transfected cells (WT HNF4A, HV HNF4A) are equally responsive to monensin (↑Na+) treatment. (*P<0.001 vs. WT VEH or HV VEH; #P<0.001 vs others (two-way ANOVA, Holm-Sidak test). F) NBCe2 expression in WT and HV hRPTCs transfected with empty vector or V5 epitope-tagged HNF4A. An increase in HNF4A expression leads to increased NBCe2 expression in HV but not WT hRPTCs (**P<0.01HV vs HV HNF4A). The monensin-induced increase in intracellular sodium (↑Na+) increases NBCe2 expression in HV (***P<0.001 VEH HV vs monensin (↑Na+) HV) and HV HNF4A (#P<0.001 VEH HV HNF4A vs monensin (↑Na+) HV HNF4A) hRPTCs but to a greater extent in HV HNF4A than VEH HV ($P<0.001). The HNF4A blocker, BI6015, prevents the increase in NBCe2 expression in HV HNF4A cells (&P<0.001, BI6015 HV HNF4A), and in monensin-treated HV cells (↑Na+), without (*P<0.05, BI605 HV) or with HNF4A overexpression (&& P<0.05, Bl6015 HV HNF4A). N = 9 in each experiment. All comparisons were made using two-way ANOVA, Holm-Sidak test.
Fig 9.
Ion transport assays in cultured hRPTCs carrying wild-type (WT) or homozygous variant (HV) SLC4A5.
A) Sodium accumulation assay. Intracellular sodium was measured in SBFI-loaded hRPTCs in response to monensin (MON, 10 μmol/L, 30 min). Monensin increases F340/F380 SBFI ratio to a greater extent in HV than WT SLC4A5 hRPTCs (N = 18, *P<0.005, t-test). B) NHE3 protein expression. Basal NHE3 protein expression is greater in HV than WT SLC4A5 hRPTCs (N = 48, *P<0.02, t-test). C) Transcellular sodium transport in polarized hRPTCs grown in Transwells™. Six HV SLC4A5 hRPTC lines derived from 6 different subjects and 4 WT SLC4A5 hRPTC lines derived from 4 different subjects were grown to confluence on Transwell™ membranes and until a stable trans-epithelial electrical resistance was achieved. Total sodium transport from the upper chamber (luminal) to the lower chamber (basolateral) was measured at 2 h as the amount of sodium measured by atomic absorption of samples taken from the lower chamber. Total sodium transport is greater in HV than WT SLC4A5 hRPTCs (N = 6–8, *P<0.01, t-test). D) Cl-/HCO- exchanger activity. The rate of pH recovery was measured after an alkaline load (CO2/HCO3- removal). Six HV SLC4A5 hRPTC cell lines derived from 6 different subjects (N = 4 experiments per subject) were compared with 4 WT SLC4A5 hRPTC lines derived from 4 different subjects (N = 4 experiments per subject). HV SLC4A5 hRPTCs have enhanced Cl-/HCO3- exchanger activity compared with WT SLC4A5 hRPTCs; rate of pH recovery is faster in hRPTCs carrying HV than WT SLC4A5 (*P<0.05, t-test).
Fig 10.
Bicarbonate-dependent pH recovery rate in hRPTCs isolated from urine of salt-sensitive (SS) and homozygous variant (HV) (for the SLC4A5 gene) or salt-resistant (SR) and wild-type (WT) (for the SLC4A5 gene) individuals.
A) Bicarbonate-dependent pH recovery rate. hRPTCs were isolated from the urine of four SS individuals who are HV for SLC4A5 or from four SR individuals who are WT for SLC4A5. The hRPTCs were cultured with monensin (10 μmol/L/24 h) to increase intracellular sodium (↑Na+), acidified, and the bicarbonate-dependent pH recovery rate was measured using BCECF in media with EIPA (10 μM) to inhibit NHE3 activity. The ratiometric data obtained by confocal microscopy are greater in hRPTCs from SS (HV) than SR (WT) individuals (N = 4 per group, *P<0.05 vs WT, t-test). B) Representative recording of bicarbonate pH recovery rate assay. Exfoliated urinary RPTCs were isolated from two study individuals, one HV-SS and one WT-SR. The sodium bicarbonate recovery rate assay was carried out as in A and imaged by ratiometric spinning disk confocal microscopy. The cells were excited at 485 nm, then 440 nm light and the fluorescence emission ratio at 525 nm was recorded every 10 s. Buffers containing sodium at 120 mmol/L and bicarbonate at 25 mmol/L were rapidly perfused at the fourth time point and imaged for 10 min. Because this assay was performed with EIPA to inhibit NHE3 activity, the increase in BCECF is due to increased bicarbonate-dependent pH recovery rate in the HV-SS individual (*P<0.05, #P<0.01 vs SR, N = 3, multiple t-test). Because the monensin-induced increase in sodium (↑Na+) downregulated NBCe1 protein to a greater extent in WT than HV SLC4A5 hRPTCs (Fig 6B), but upregulated NBCe2 protein in HV but not WT SLCC4A5 hRPTCs, the increased bicarbonate-dependent pH recovery in SS with HV SLC4A5 is most likely due to NBCe2 (Fig 6A).
Fig 11.
Bicarbonate-dependent pH recovery assays in cultured hRPTCs carrying Wild-Type (WT) or homozygous variant (HV) SLC4A5.
A) Bicarbonate-dependent pH recovery: pH recovery was measured in WT and HV SLC4A5 hRPTCs expressing empty vector (vector control, VC), overexpressing (OE) SLC4A5 (4A5OE), knock-down (KD) SLC4A5 (4A5KD), and knock-down (KD) SLC4A4 (4A5KD). Bicarbonate-dependent pH recovery is faster in SLC4A5 OE (4A5OE) (*P<0.001 and &P<0.01) and slower in SLC4A5 KD (4A5KD) (**P<0.001 and &&P<0.001), relative to VC in both WT and HV SLC4A5 hRPTCs. By contrast, bicarbonate-dependent pH recovery is not altered in SLC4A4 KD, relative to VC in either WT or HV SLC4A hRPTCs. Monensin (10 μmol/L, 24 h) treatment that increases intracellular sodium (↑Na+) increases pH recovery rate in HV but not WT SLC4A5 hRPTCs (#P<0.001). In SLC4A5 OE (4A5OE) hRPTCs, pH recovery rate is increased by monensin (10 μmol/L/24 h) (↑Na+) in both WT (##P<0.001) and HV SLC4A5 ($ $P<0.001) but to a greater extent in the latter than in the former ($P<0.001). Knockdown of SLC4A5 (4A5KD) prevents the stimulatory effect of monensin on pH recovery rate in both WT and HV SLC4A5. Knockdown of SLC4A4 (4A4KD) does not affect the increased pH recovery in monensin-treated (↑Na+) HV SLC4A5 hRPTCs (%P<0.05 HV 4A4KD vs WT 4A4KD). B) Increased bicarbonate-dependent pH recovery rate in monensin-treated HV SLC4A5 hRPTCs is blocked by HNF4A inhibitors. Bicarbonate-dependent pH recovery was measured in vehicle (VEH) or monensin (10 μmol/L, 24 h)-treated (↑Na+) HV SLC4A5 hRPTC in the presence of HNF4A inhibitors BIM5078 or BI6015. These inhibitors have no effect when added alone, but either inhibitor completely blocks (n = 4 #P<0.05 vs monensin HV, two-way ANOVA, Holm-Sidak test) the monensin-stimulated (↑Na+) (n = 12 *P<0.05, two way ANOVA, Holm-Sidak test) increase in bicarbonate-dependent pH recovery rate in HV hRPTCs.
Fig 12.
The effect of BI6015 on the expression of Na+,K+/ATPase (NKA) and the putative anion transporter type 1 (PAT-1).
Compared to vehicle control (VEH), neither NKA nor PAT-1 is affected by the addition of the HNF4A inhibitor BI6015. α tubulin was used as a protein loading control.
Fig 13.
Effect of HNF4A inhibitors on basal expression of NHE3, PAT1 and Na+,K+/ATPase in hRPTCs.
A) NHE3 protein expression Basal NHE3 protein expression is greater in HV than WT SLC4A5 hRPTCs (N = 3, *P<0.01, t-test). Pharmacological inhibition of HNF4A with BIM5078 does not significantly inhibit basal NHE3 protein expression. B) Sodium influx assay. Intracellular sodium was measured in SBFI-loaded hRPTCs in response to ouabain (100 μmol/L, 30 min). Ouabain increases F340/F380 SBFI ratio, converted to mM sodium to a greater extent in HV than WT SLC4A5 hRPTCs (N = 3, *P<0.01, t-test). HNF4A inhibitor (BIM5078, 10 μM, 24 h) inhibits the increased sodium influx in HV hRPTCs (N = 3, #P<0.01 vs HV VEH, two-way ANOVA, Holm-Sidak test) but not in WT hRPTCs.
Fig 14.
Intracellular pH recovery assay.
pH recovery assay in WT and HV hRPTCs shows no significant differences between HV hRPTCs (6 HV hRPTCs each from a different individual) and WT hRPTCs (4 WT hRPTCs each from a different individual), repeated 3 times, a total 18 replicates in HV and 12 replicates in WT hRPTCs (Fig 14). These data were transformed mathematically (see Methods) to provide the intrinsic buffering capacity, which demonstrates that there is no significant difference between WT and HV (Fig 15).
Fig 15.
Intracellular intrinsic buffering capacity.
Measurement of intrinsic buffering capacity in WT and HV hRPTCs. hRPTCs were labeled with the pH sensitive ratiometric dye BCECF and pH measured by microplate fluorometry by incubating cells in sodium-, CO2-, and bicarbonate-free HEPES buffer by sequential incubation in decreasing amounts of ammonium chloride. There are no significant differences in intrinsic buffering capacity between WT and HV hRPTCs (N = 8 per cell type).
Fig 16.
Models of ion transport in hRPTCs.
This is a model of the hRPTC with the apical (brush border, facing the lumen) (left hand side) and the basolateral side (right hand side). The principal ion transporters and some of the receptors that regulate them are shown. Starting at 11 o’clock in blue is shown the classic pathway for transporting bicarbonate (HCO3-) into the cell. Filtered NaHCO3 dissociates into Na+ and HCO3. HCO3- in the luminal fluid and H+ secreted into the lumen form H2CO3. Carbonic anhydrase type 4 (CA IV) in the luminal membrane catalyzes the conversion of H2CO3 to H2O and CO2.CO2 diffuses inside the hRPTC where intracellular carbonic anhydrase type 2 (CA II) catalyzes the conversion of CO2 and H2O into H2CO3 which then dissociates into HCO3- and H+. At 9 o’clock is NHE3 which exchanges one Na+ from the lumen with one H+ inside the hRPTC. At 7 o’clock HCO3- Cl- exchanger (PAT1) is depicted which exchanges luminal Cl- with cytoplasmic HCO3-. At 3 o’clock is depicted NBCe1 at the basolateral membrane which electrogenically transports 2–3 Na+ and one HCO3- into the basolateral space. At 4 o’clock is Na+, K+/ATPase which pumps 3 Na+ out of the cell into the blood stream and pumps in 2 K+ inside the cell The topic of this manuscript deals with NBCe2, drawn at 8 o’clock. Under a normal sodium load it plays a minor role in Na+ and HCO3- transport into the hRPTC. There are various plasma membrane receptors that regulate some of these transporters/ exchanger/pumps. The dopamine-1 receptor (D1R) (ten o’clock) when stimulated with dopamine (green box) inhibits (red lines) both NHE3 and Na+, K+/ATPase (without the red line, for simplicity) activities resulting in reduced Na+ reabsorption and increased Na+ excretion. The AT1R (5 o’clock) increases Na+, K+/ATPase activity (green arrow) resulting in increased Na+ reabsorption. An increase in intracellular Na+ increases Na+, K+/ATPase activity (5 o’clock) that is abetted by AT1R (green arrow) resulting in increased Na+ transport from inside the cell to the basolateral space. The D1R and AT1R oppose each other. The D1R inhibits the AT1R, resulting in reduced Na+ transport. We showed that NBCe2 is not affected by stimulation of the D1R or AT1R. Under basal conditions, NBCe1 is more active than NBCe2 (depicted as relatively larger directional transport arrows).
Fig 17.
Model of ion transport in hRPTCs with HNF4A binding to SCL4A5 promoter.
In a hRPTC containing the SNP rs7571842 in SCL4A5 the model changes to what is depicted in Fig 17. Increasing intracellular Na+ concentration with high extracellular Na+ concentration or monensin (9 o’clock) increases NBCe2 mRNA, protein, and activity, while only marginally attenuating the protein and activity of NBCe1 (2 o’clock) in hRPTCs with SNPs in NBCe2. This results in a net increase in Na+ transport into the basolateral space. PAT1 (7 o’clock) activity increases because of an increase in intracellular bicarbonate (7 o’clock). NHE3 (9 o’clock) activity also increases because the increase in NBCe2 activity increases intracellular H+ following the conversion of transported HCO3- to H2CO3 and its dissociation to H+ and HCO3- resulting in a further increase in sodium reabsorption. An increase in HNF4A binding to the rs7571842 SLC4A5 increases its expression.