Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 597 (2), 499-519

NHA2 Promotes Cyst Development in an in Vitro Model of Polycystic Kidney Disease

Affiliations

NHA2 Promotes Cyst Development in an in Vitro Model of Polycystic Kidney Disease

Hari Prasad et al. J Physiol.

Abstract

Key points: Significant and selective up-regulation of the Na+ /H+ exchanger NHA2 (SLC9B2) was observed in cysts of patients with autosomal dominant polycystic kidney disease. Using the MDCK cell model of cystogenesis, it was found that NHA2 increases cyst size. Silencing or pharmacological inhibition of NHA2 inhibits cyst formation in vitro. Polycystin-1 represses NHA2 expression via Ca2+ /NFAT signalling whereas the dominant negative membrane-anchored C-terminal fragment (PC1-MAT) increased NHA2 levels. Drugs (caffeine, theophylline) and hormones (vasopressin, aldosterone) known to exacerbate cysts elicit NHA2 expression. Taken together, the findings reveal NHA2 as a potential new player in salt and water homeostasis in the kidney and in the pathogenesis of polycystic kidney disease.

Abstract: Autosomal dominant polycystic kidney disease (ADPKD) is caused by mutations in PKD1 and PKD2 encoding polycystin-1 (PC1) and polycystin-2 (PC2), respectively. The molecular pathways linking polycystins to cyst development in ADPKD are still unclear. Intracystic fluid secretion via ion transporters and channels plays a crucial role in cyst expansion in ADPKD. Unexpectedly, we observed significant and selective up-regulation of NHA2, a member of the SLC9B family of Na+ /H+ exchangers, that correlated with cyst size and disease severity in ADPKD patients. Using three-dimensional cultures of MDCK cells to model cystogenesis in vitro, we showed that ectopic expression of NHA2 is causal to increased cyst size. Induction of PC1 in MDCK cells inhibited NHA2 expression with concordant inhibition of Ca2+ influx through store-dependent and -independent pathways, whereas reciprocal activation of Ca2+ influx by the dominant negative membrane-anchored C-terminal tail fragment of PC1 elevated NHA2. We showed that NHA2 is a target of Ca2+ /NFAT signalling and is transcriptionally induced by methylxanthine drugs such as caffeine and theophylline, which are contraindicated in ADPKD patients. Finally, we observed robust induction of NHA2 by vasopressin, which is physiologically consistent with increased levels of circulating vasopressin and up-regulation of vasopressin V2 receptors in ADPKD. Our findings have mechanistic implications on the emerging use of vasopressin V2 receptor antagonists such as tolvaptan as safe and effective therapy for polycystic kidney disease and reveal a potential new regulator of transepithelial salt and water transport in the kidney.

Keywords: Ca2+ signaling; NFAT; Na+/H+ exchanger; polycystin; vasopressin.

Figures

Figure 1
Figure 1. NHA2 is up‐regulated in polycystic kidney disease
A–C, expression profiling of plasma membrane SLC9A (NHE) genes in ADPKD patient cysts of different sizes showed no change (NHE1 and NHE3) or modest repression (NHE2). D, NHA2 (SLC9B2) expression was increased relative to normal kidney tissue and correlated with cyst size and disease severity. (normal: n = 3; minimal cyst: n = 5; small cyst: n = 5; medium cyst: n = 5; large cyst: n = 3.) Student's t test: NS P > 0.05, * P < 0.05, ** P < 0.01, *** P < 0.001. E, representative confocal microscopic images of MDCK cells stably expressing NHA2–GFP (green) showing colocalization with basolateral marker E‐cadherin (red) in DAPI (blue) stained cells, as seen in the merged image (yellow). Bottom row is higher magnification of boxed region from top row. F, quantification of colocalization. Pearson's correlation coefficient = 0.66 ± 0.11; Manders’ coefficient = 0.64 ± 0.17; n = 25. Scale bar: 10 μm. Error bars are S.E.
Figure 2
Figure 2. NHA2 promotes cyst development in vitro
A, schematic depiction of in vitro MDCK cell model of cystogenesis in extracellular matrix (Matrigel). B, representative image of a MDCK cyst with polarized, single‐layer, thinned epithelium surrounding a fluid‐filled lumen (left; brightfield) with surface expression of NHA2–GFP (right; fluorescence). C, NHA2+ MDCK cells formed larger cysts (right) relative to control (left). D, quantification of cyst size on day 10 of culture showed significant cyst expansion from NHA2+ cells relative to control (Control: 100 ± 66.4, n = 363; NHA2+: 232.3 ± 213.7, n = 367; Student's t test, **** P < 0.0001). E, percentage of large cysts is significantly higher from NHA2+ cells as compared to control. Classification of cysts in three subclasses according to their relative diameter (small/medium/large) showed significantly elevated percentage of cysts with large size in NHA2+ and reduced percentage of cysts with small and medium sizes (χ2 test, **** P < 0.0001). F and G, representative images of MDCK cysts documenting mutilayered and multiloculated cysts in NHA2+ MDCK cells (left; brightfield) with surface expression of NHA2–GFP (right; fluorescence) that was not seen in cysts derived from control MDCK cells. Scale bars: 10 μm. All error bars are S.E.
Figure 3
Figure 3. Polycystin 1 downregulates Ca2+ influx and NHA2 expression
A, hypothesis for PC1‐mediated inhibition of Ca2+ influx through PC2 and NHA2 expression in MDCK cells. B, western blotting of MDCK cell lysates in the absence or presence of tetracycline for 17 h to induce PC1 expression, in biological triplicates, probed with antibody to PC1 (top), NHA2 (middle) and β‐actin (bottom). C, densitometric quantification of western blot shown in B. NHA2 protein expression was normalized to actin controls and was significantly down‐regulated by ∼2.9‐fold (n = 3; Student's t test, * P < 0.05) in MDCK cells with PC1 induction. D, quantitative PCR (qPCR) showing significant down‐regulation of NHA2 mRNA (n = 3; Student's t test, ** P < 0.01) and no change in PC2 mRNA (n = 3; Student's t test, NS P > 0.05) with PC1 induction. E and F, representative Fura‐2 fluorescence ratio traces (E) and quantitation (F) showing significant and proportionate reduction in store‐independent calcium entry (SICE) upon induction of PC1 for 24 h (n = 3; Student's t test, * P < 0.05) and 48 h (n = 3; Student's t test, *** P < 0.001) in MDCK cells. G, representative Fura‐2 fluorescence ratio traces showing reduction in store‐operated calcium entry (SOCE) following thapsigargin‐mediated release of store Ca2+ in MDCK cells with PC1 induction for 48 h. H, representative Fura‐2 fluorescence ratio traces showing efficacy of Ca2+ ionophore ionomycin to enhance cytosolic Ca2+ levels. I, qPCR showing significant and dose‐dependent increase in NHA2 expression with ionomycin treatment to levels similar to MDCK cells without PC1 induction. Error bars are S.D.
Figure 4
Figure 4. NFAT and Ca2+‐mediated regulation of NHA2 expression
A–C, gene expression changes in response to Ca2+ influx following treatment with ionomycin and phorbol myristate in human primary T lymphoblasts obtained from the GSE50971 dataset. A, NHA2 showed significant up‐regulation (∼2.4‐fold higher; Student's t test, * P < 0.05). B, NHE1 expression showed no change (Student's t test, NS P > 0.05). C, NHE2 expression was lower, but non‐significant (Student's t test, NS P > 0.05). D, NFATc1 deletion resulted in profound, >95% down‐regulation of NHA2 expression (Student's t test, ** P < 0.01) and no change in related NHE1 isoform (Student's t test, NS P > 0.05) during osteoclast differentiation in vitro. E and F, down‐regulation of well‐known NFAT target genes, CLCN7 and MMP9, is shown for comparison. G–I, gene expression changes were determined from the GSE57468 dataset. In this experiment, mouse bone marrow‐derived osteoclast precursors were treated with RANKL for up to 3 days to differentiate into osteoclasts. G, robust, time‐dependent up‐regulation of NHA2 levels in response to NFAT‐mediated osteoclast differentiation in vitro (∼10‐fold on day 1; ∼50‐fold on day 2–3; Student's t test, *** P < 0.001). H and I, significant up‐regulation of well‐known NFAT target genes, CLCN7 (∼1.3‐fold on day 1, ∼3.3‐fold on day 2, ∼5.8‐fold on day 3; Student's t test, * P < 0.05, *** P < 0.001) and MMP9 (∼69‐fold on day 1, ∼249‐fold on day 2, ∼270‐fold on day 3; Student's t test, *** P < 0.001). Error bars are S.D.
Figure 5
Figure 5. NFAT mediates Ca2+‐dependent NHA2 expression
A, NFATc1 expression profiling of ADPKD patient cysts of different sizes revealed up‐regulation of NFAT expression relative to normal kidney tissue that correlated with cyst size and disease severity. B, proposed model for PC1‐mediated modulation of Ca2+ homeostasis and regulation of NHA2 expression through calcineurin (CaN) and NFAT signalling. C, schematic representation of full‐length PC1 (top) and membrane‐anchored C‐terminal tail fragment of PC1 (PC1‐MAT) (bottom) containing the C‐terminal tail fragment of PC1 linked to the signal peptide (SP) of CD16 and transmembrane domain of CD7. D, Luciferase assay in HEK293 cells to determine NFAT activation using a 3× NFAT binding sequence that drives a firefly luciferase reporter gene and is measured luminometrically. Renilla luciferase driven by a constitutively active SV40 promoter was used to normalize for variations in both cell number and the transfection efficiency. Expression of full‐length PC1 resulted in reduction (∼4‐fold lower; n = 3; Student's t test, **** P < 0.0001) and of PC1‐MAT resulted in increase (∼5‐fold higher; n = 3; Student's t test, **** P < 0.0001) in NFAT reporter activity, relative to empty vector transfection control. E and F, representative micrographs (E, scale bar= 10 μm) and quantification using Manders’ coefficient (F) determining fractional colocalization of NFAT–GFP (green) with DAPI (blue). Colocalization is evident in the merge and orthogonal slices (Z) as cyan puncta. In vector transfected cells, NFAT–GFP is predominantly localized in the cytoplasm. Note prominent overlap between NFAT–GFP and DAPI, consistent with increased nuclear translocation, in cells expressing PC1‐MAT, similar to cells expressing constitutively active NFAT–GFP (CA‐NFAT) (n = 40/each condition; Student's t test, **** P < 0.0001). G and H, representative Fura‐2 fluorescence ratio traces (G) and quantification (H) showing ∼2.4‐fold increase in store‐independent calcium entry (SICE) in cells transfected with PC1‐MAT (n = 3; Student's t test, * P < 0.05), relative to empty vector transfection. Note a higher baseline with PC1‐MAT, relative to the empty vector control, suggesting higher basal Ca2+ levels with PC1‐MAT expression. I and J, qPCR showing ∼2.6‐fold higher NHA2 expression in cells expressing PC1‐MAT (n = 3; Student's t test, *** P < 0.001) and ∼25% lower NHA2 levels in cells expressing full‐length PC1 (n = 3; Student's t test, ** P < 0.01), relative to empty vector transfection. Error bars S.D.
Figure 6
Figure 6. Drug and hormonal regulation of NHA2
A, expression profiling of NHA2 expression (y‐axis) obtained from an unbiased bioinformatics analysis of 1078 microarray studies (x‐axis), as described in Methods. Note that highest up‐regulation of NHA2 (≥3‐fold) was observed in response to methyl xanthine drugs: caffeine (7.5 mm; 24 h) and theophylline (10 mm; 24 h). Four drugs that resulted in maximal down‐regulation of NHA2 are valproic acid, dexamethasone, cisplatin and doxorubicin. B and C, bar graphs of NHA2 expression derived from microarray analysis of caffeine (1.5 and 7.5 mm; B) and theophylline (2 and 10 mm; C) treatments showing dose‐ and duration‐ (8 h and 24 h) dependent effects. D, representative Fura‐2 fluorescence ratio traces showing significant increase in cytoplasmic Ca2+ with caffeine (top) and theophylline (bottom) treatment in MDCK cells. E and F, qPCR analysis to validate bioinformatic studies documenting significant, dose‐dependent increase in NHA2 expression with caffeine (E) and theophylline (F), relative to vehicle controls in HEK293 cells. G, hypothesis for vasopressin‐mediated regulation of NHA2 expression in renal epithelial cells. Vasopressin stimulation of the V2R receptors results in accumulation of cAMP and activation of PKA. Like caffeine and theophylline, vasopressin‐driven PKA activation stimulates Ca2+ release via the ryanodine receptor channel from the endoplasmic reticulum (ER). Increased cytosolic Ca2+ would in turn increase NHA2 expression. H, representative Fura‐2 fluorescence ratio traces showing significant increase in cytoplasmic Ca2+ with vasopressin treatment in MDCK cells, relative to vehicle control. I, qPCR data showing fold change in NHA2 transcript levels following treatment with vasopressin for indicated time periods ranging from 2 to 24 h. Note significant and phasic increase in NHA2 expression with vasopressin treatment that peaked (∼6‐fold) at 4–8 h and reached baseline by 24 h. J, qPCR data showing fold change in NHA2 transcript levels following treatment of MDCK cells with aldosterone for indicated time periods ranging from 2 to 24 h. Note significant and phasic increase in NHA2 expression with aldosterone treatment that peaked (∼2.6‐fold) at 8 h and reached baseline by 24 h. K, western blot showing NHA2 expression levels following a 6 h treatment of MDCK cells with vehicle (lane 1), vasopressin (lane 2) or aldosterone (lane 3), with tubulin serving as a loading control (bottom panel). L, lithium sensitivity assay to evaluate functional consequences of the hormonal induction of NHA2. Consistent with increased NHA2 mRNA and protein expression, treatment with either vasopressin or aldosterone for 8 h resulted in increased cell survival in medium supplemented with 90 mm LiCl, relative to untreated control. Cell survival in the presence of LiCl was measured using an MTT assay. Error bars are S.D.
Figure 7
Figure 7. NHA2 promotes hemicyst formation
A, schematic depiction of generation of fluid‐filled hemicysts or domes with MDCK cells. B, representative image of a MDCK hemicyst with fluid accumulation focally beneath the epithelium (white arrow). C, NHA2+ MDCK cells formed larger hemicysts (right) relative to control (left). D, quantification of hemicyst development showed significant expansion in hemicyst area with NHA2 expression relative to control (Control: 100 ± 5.01, n = 100; NHA2+: 275.1 ± 12.75, n = 100; Student's t test, **** P < 0.0001), suggesting an increase in vectorial transport of salt and water with NHA2 expression. Scale bar: 10 μm. Error bars are S.E.
Figure 8
Figure 8. Knockdown or inhibition of NHA2 attenuates cyst development in vitro
A, lentivirally mediated knockdown of NHA2 in NHA2+ MDCK cells, using two different lentiviral shRNA constructs (ShRNA 1 and ShRNA 2) targeting human NHA2, resulted in robust knockdown of GFP‐tagged NHA2. Western blot of total protein was probed using anti‐NHA2 antibody. B, lentivirally mediated knockdown of NHA2 does not alter growth of MDCK cells. Cell growth activity was measured by an MTT assay, normalized to the scrambled control (n = 8; Student's t test, NS P > 0.05). C and D, cytoplasmic pH in NHA2+ cells with vehicle or phloretin treatment was determined, as described under Methods. Calibration curve is shown in C. Inhibition of NHA2 with phloretin did not significantly alter cytoplasmic pH (D) (n = 3; Student's t test, NS P > 0.05). E, lithium sensitivity assay to validate the effect of phloretin. Cell survival in the presence of 100 mm LiCl was measured using MTT. Phloretin treatment and lentiviral NHA2 knockdown conferred significant growth sensitivity to lithium (n = 8; Student's t test, **** P < 0.0001). Note no difference in lithium sensitivity between phloretin treatment with or without NHA2 knockdown (n = 8; Student's t test, NS P > 0.05). F, lentivirally mediated knockdown of NHA2 shown in A resulted in formation of significantly smaller cysts, relative to control. Representative images are shown, and bottom row is higher magnification of boxed region from top row. G, cyst size is significantly reduced following NHA2 depletion relative to control on day 10 of culture (Control: 100.0 ± 5.6; n = 50, NHA2 ShRNA 1: 28.0 ± 1.46; n = 50, NHA2 ShRNA 2: 16.7 ± 0.91; n = 50, Student's t test, **** P < 0.0001). H, phloretin (20–150 μm) causes a dose‐dependent reduction in cyst size. Representative images are shown, with the vehicle treatment on the left. I, quantification of cyst size on day 10 of culture (n = 50/each condition). Scale bar: 25 μm. Error bars are S.E.
Figure 9
Figure 9. Proposed role for NHA2 in PKD
Polycystins 1 and 2 (PC1 and PC2) regulate Ca2+ influx, NFAT signalling and NHA2 expression. Methylxanthine drugs (caffeine and theophylline) activate ryanodine‐sensitive receptor channel (RyR) and release Ca2+ stores from the endoplasmic reticulum (ER) in response to an initial Ca2+ entry through PC2 and other plasma membrane Ca2+ channels, thereby effectively amplifying cytoplasmic Ca2+ and increasing NHA2 expression. Vasopressin stimulation of the V2R receptors results in accumulation of cyclic AMP (cAMP) and activation of protein kinase A (PKA). Like caffeine and theophylline, vasopressin‐driven PKA activation stimulates Ca2+ release via the RyR, leading to an increase NHA2 expression which is associated with cyst expansion.

Comment in

Similar articles

See all similar articles

Cited by 1 PubMed Central articles

Publication types

LinkOut - more resources

Feedback