Extracellular ATP is an autocrine/paracrine factor that regulates renal function. Transient receptor potential vanilloid (TRPV) 4 is a cation channel that mediates release of autocrine/paracrine factors by acting as an osmosensor. The renal medulla, and therefore the thick ascending limb, is exposed to osmotic stress. We hypothesize that reduced osmolality stimulates ATP release from the thick ascending limb via transient receptor potential vanilloid (TRPV) 4 activation. We measured ATP release by medullary thick ascending limb suspensions after reducing bath osmolality from 350 to 323 mosmol/kgH2O, using the luciferin-luciferase assay. Decreasing osmolality stimulated ATP release compared with control (38.9 ± 7.2 vs. 2.4 ± 1.0 pmol/mg protein; n = 6, P < 0.01). To examine the role of TRPV4, we used 1) Ca-free solutions, 2) a TRPV4 inhibitor, 3) small interfering (si) RNA against TRPV4, and 4) a TRPV4 activator. Removal of Ca completely blocked osmolality-induced ATP release (42.2 ± 5.9 vs. 2.6 ± 1.5 pmol/mg protein; n = 6, P < 0.01). In the presence of the TRPV4-selective inhibitor ruthenium red, osmolality-induced ATP release was blocked by 73% (56.4 ± 19.9 vs. 8.8 ± 2.3 pmol/mg protein; n = 6; P < 0.03). In vivo treatment of thick ascending limbs with siRNA against TRPV4 decreased osmolality-induced ATP release by 62% (31.5 ± 3.4 vs. 12.4 ± 1.1 pmol/mg protein; n = 6; P < 0.01), while reducing TRPV4 expression by 74% compared with the nontreated kidney. Treatment with scrambled siRNA did not affect TRPV4 expression and/or osmolality-induced ATP release. Finally, in the absence of changes in osmolality, the specific TRPV4 agonist 4α-PDD increased ATP release (3.6 ± 0.9 vs. 25.4 ± 7.4 pmol/mg protein; n = 6; P < 0.04). We concluded that decreases in osmolality stimulate ATP release by thick ascending limbs and this effect is mediated by TRPV4 activation.
- cell swelling, calcium channels, hyposmolality
extracellular adenosine-triphosphate (ATP) is an autocrine/paracrine factor involved in the regulation of kidney function. Extracellular ATP, via activation of P2 receptors, contributes to the regulation of renal hemodynamics by regulating afferent arterial diameter (21) and Na and water excretion by regulating tubular reabsorption along the nephron (18). However, little is known about the mechanisms of ATP release in the kidney.
The transient receptor potential vanilloid (TRPV) 4 is a divalent cation-permeant channel activated by chemical and/or physical stimuli such as cell swelling (1). In epithelial cells, TRPV4 is primarily activated by cell swelling caused by hypotonicity; therefore, it acts as an osmosensor (24). TRPV4 activation has been shown to stimulate the release of autocrine/paracrine regulators, including ATP (11, 12). Although TRPV4 is expressed in several segments of the nephron, its role in hypotonicity-induced ATP release in the kidney remains unknown.
Under physiological conditions, the renal medulla is constantly subject to changes in osmolality ranging from 300 to 1,200 mosmol/kgH2O. Factors such as hydration (25), salt consumption (14), medullary blood flow (6), and loop diuretics (2) modify medullary osmolality. Therefore, cells of the kidney medulla are susceptible to osmotic stress. One of the major components of the renal medulla is the thick ascending limb, which reabsorbs 20–30% of the total NaCl load. Extracellular ATP is thought to be an important regulator of thick ascending limb NaCl reabsorption (18). Additionally, ATP released from the thick ascending limb could alter collecting duct transport (5) and vasa recta function (4). However, little is known about the stimuli that trigger ATP release from the thick ascending limb. We hypothesize that hypotonic stimuli activate TRPV4, stimulating ATP release by the thick ascending limb.
This study was approved by the Henry Ford Hospital Institutional Animal Care and Use Committee. All studies were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Male Sprague-Dawley rats, weighing 150–200 g (Charles River Breeding Laboratories, Wilmington, MA), were fed a diet containing 0.21% Na and 1.1% K (Purina-Mills, Richmond, IN) for at least 7 days before anesthesia.
Medullary thick ascending limb suspensions.
Medullary thick ascending limb suspensions were prepared as described previously (27). Kidneys were perfused retrograde via the abdominal aorta with 40 ml of solution A (Table 1). Then, kidneys were removed and the inner stripe of the outer medulla was dissected from coronal slices of the kidney, minced, and incubated at 37°C for 30 min in solution A plus 0.1% collagenase A. During this time, the suspension was agitated and gassed with 100% O2 every 5 min. The tissue was centrifuged at 93 g for 2 min, resuspended in cold solution A, and stirred on ice for 30 min. The suspension was then filtered through a 250-μm nylon mesh and spun again for 2 min. The pellet was washed and resuspended in 1 ml cold solution A. In some experiments, tubules were resuspended in a Ca-free solution (solution C; Table 1).
Measurements of ATP release.
ATP was measured using an ATP bioluminescence assay (luciferin-luciferase; Sigma-Aldrich). Aliquots of thick ascending limb suspensions were incubated at 37°C in solution A (350 mosmol/kgH2O) saturated with oxygen in the dark for 5 min. Then, we performed the baseline measurement. After control measurements, bath osmolality (luminal and basolateral) was reduced by adding 100 μl of solution B (290 mosmol/kgH2O; Table 1). In Ca-free experiments, bath osmolality was reduced by adding 100 μl of solution D (290 mosmol/kgH2O; Table 1). Luminescence was measured for an additional 5 min. The area under the curve obtained from peak intensity after change in osmolality was used to calculate ATP release. Total measurements of ATP release were achieved by mechanical cell lysis and addition of 0.0005% Triton X. Calibration curves were performed daily. At the end of the experiment, the sample was recovered and lysed, and total protein was measured by the Coomassie blue method.
Small interfering RNA-expressing adenovirus.
Recombinant replication-deficient adenoviruses encoding the small interfering (si) RNA sequence for TRPV4 and control nonsilencing siRNA under the control of the H1 mouse RNA polymerase promoter were constructed by ViraQuest as described previously (22). A 63-base annealed oligomer containing the siRNA sequence against TRPV4 [sense: r(CCA ACA UGA AGG UCU GUA A)dTdT, the hairpin loop linker, and antisense: r(UUA CAG ACC UUC AUG UGG G)dTdG] was inserted into a specific siRNA shuttle vector (pVQAd-MIGTHY), which contained an H1 mouse RNA polymerase promoter, a cloning site for insertion of a heterologous gene, and a polyadenylation signal flanked by adenoviral sequences 5′ and 3′. The adenoviral shuttle plasmids were transfected into the permissive HEK 293 host cell line along with adenoviral DNA lacking the E1 region. Virus isolates were plaque-purified and propagated in HEK 293 cells, isolated, concentrated, and titered. The final titer of the adenovirus was 7 × 1011 particles/ml.
In vivo knockdown of TRPV4.
We performed in vivo adenoviral-mediated transduction of thick ascending limbs as previously described (22). Briefly, rats were anesthetized with ketamine (60 mg/kg ip) and xylaxine (20 mg/kg ip) before surgery. The left kidney was exposed via a left flank incision, the fatty tissue around the renal pole was removed, and both the renal artery and vein were clamped. Then, four 20-μl virus injections of 7 × 1011 particles/ml were made using a custom-made 30-gauge needle attached to polyethylene-10 tubing connected to a syringe pump (Harvard Apparatus, Holliston, MA) set at 20 μl/min. The needle was inserted perpendicularly to the renal capsule, parallel to the medullary rays and directed toward the medulla. Injections were made along the longitudinal axis of the kidney. Each injection point was separated by ∼2.5 mm. To avoid bleeding and leakage of the virus, the needle remained in place for 30 s after infusion was complete. The clamp from the renal artery was released before 8 min. After renal blood flow was reestablished, the kidney was returned to the abdominal cavity and the incision was sutured. Previously, we have shown that at least 80% of thick ascending limb cells can be transduced using this technique (22).
Measurements of TRPV4 expression.
The same amount of protein was loaded on SDS-polyacrylamide gels and proteins separated by electrophoresis. TRPV4 levels were detected by Western blotting using a TRPV4 monoclonal antibody (BIOMOL, Plymouth Meeting, PA) as previously described (27). Band intensities were quantified by densitometry.
Determination of protein content.
Total protein content was determined using Coomassie Plus reagent (Pierce, Rockford, IL), based on Bradford's colorimetric method.
Data are reported as means ± SE. Differences in means were analyzed using a Student's t- or unpaired t-test. Statistical analysis was performed by the Department of Biostatistics and Epidemiology of Henry Ford Hospital.
To study whether hyposmolality increases ATP release, we measured ATP release from thick ascending limb suspensions after decreasing bath osmolality. We found that after the osmolality of the bath was decreased from 350 to 323 mosmol/kgH2O (a 9% decrease in osmolality) by adding 100 μl of a hypotonic solution (290 mosmol/kgH2O), thick ascending limbs released 38.9 ± 7.2 pmol ATP/mg protein. In contrast, after the addition of an isotonic solution (350 mosmol/kgH2O), thick ascending limbs released only 2.4 ± 1.0 pmol ATP/mg protein (P < 0.01 vs. hypotonicity; n = 6; Fig. 1). These data indicate that decreases in osmolality stimulate ATP release from the thick ascending limb.
In other cell types, ATP release is dependent on extracellular Ca (23). To test whether that was the case for medullary thick ascending limbs, we measured ATP release from thick ascending limbs after bath osmolality was decreased from 350 to 323 mosmol/kgH2O in a Ca-free solution. We found that in the presence of Ca, after bath osmolality was decreased from 350 to 323 mosmol/kgH2O, thick ascending limbs released 42.2 ± 5.9 pmol ATP/mg protein. In contrast, in the absence of extracellular Ca, after bath osmolality was decreased from 350 to 323 mosmol/kgH2O, thick ascending limbs released only 2.6 ± 1.5 pmol ATP/mg protein (P < 0.02 vs. control; n = 6; Fig. 2). To further demonstrate the role of Ca in hypotonicity-induced ATP release by thick ascending limbs, we next measured ATP release from thick ascending limbs after bath osmolality was decreased in the presence and absence of the intracellular Ca chelator BAPTA-AM (50 μM). In the absence of BAPTA, after bath osmolality was decreased from 350 to 323 mosmol/kgH2O, thick ascending limbs released 37.9 ± 1.9 pmol ATP/mg protein. In contrast, after 15-min preincubation with BAPTA, with the decreasing of bath osmolality from 350 to 323 mosmol/kgH2O thick ascending limbs released only 7.1 ± 2.9 pmol ATP/mg protein (P < 0.01 vs. control; n = 5), a 81% inhibition. Taken together, these data indicate that both extracellular and intracellular Ca are required for osmolality-induced ATP release by the thick ascending limb.
TRPV4 is Ca-permeable channel involved in ATP release from other epithelial cells (11). To test whether TRPV4 was involved in hypotonicity-induced ATP release in the thick ascending limb, we measured ATP release from thick ascending limbs after bath osmolality was decreased in the presence and absence of the TRPV4 inhibitor ruthenium red (15 μM). In the absence of ruthenium red, after bath osmolality was decreased from 350 to 323 mosmol/kgH2O, thick ascending limbs released 56.4 ± 19.9 pmol ATP/mg protein. In contrast, after 15-min preincubation with ruthenium red, with the decreasing of bath osmolality from 350 to 323 mosmol/kgH2O thick ascending limbs released only 8.8 ± 2.3 pmol ATP/mg protein (P < 0.03 vs. control; n = 6; Fig. 3), a 73% inhibition. To show that these results were not due to changes in total ATP pools, we measured total ATP in the presence and absence of ruthenium red. Fifteen-minute preincubation with ruthenium red did not affect the total pool of ATP in our preparation [12.3 ± 2.0 vs. 13.9 ± 1.5 nmol ATP/mg protein; n = 5; not significant (NS)]. These data indicate that pharmacological inhibition of TRPV4 blocks ATP release from the thick ascending limb.
Ruthenium red is a nonspecific TRPV4 inhibitor. To show that the effects of ruthenium red are due to inhibition of TRPV4, we used in vivo unilateral TRPV4 silencing. Viruses expressing siRNA sequences against TRPV4 (AdvTRPV4siRNA) and a scramble sequence (AdvSCRsiRNA) were injected only in the outer medulla of the left kidney while the right kidney served as a control. We found that after decreasing osmolality of the solution bathing thick ascending limbs from control contralateral kidneys from 350 to 323 mosmol/kgH2O, tubules released 31.5 ± 3.4 pmol ATP/mg protein. In contrast, after decreasing osmolality of the solution bathing thick ascending limbs from AdvTRPV4siRNA-injected kidneys from 350 to 323 mosmol/kgH2O, tubules released only 12.4 ± 1.1 pmol ATP/mg protein (Fig. 4A; P < 0.01 vs. control; n = 7), a 62% reduction. AdvTRPV4siRNA injections significantly reduced TRPV4 expression compared with control contralateral kidneys by 72.1 ± 8.7% (Fig. 4B; P < 0.01 vs. control; n = 7). In contrast, injections of AdvSCRsiRNA did not significantly affect hypotonicity-induced ATP release compared with the control contralateral kidneys (Fig. 5A; 4.3 ± 5.9% vs. control; NS; n = 6). AdvSCRsiRNA injections did not significantly affect TRPV4 expression compared with control contralateral kidneys (Fig. 5B; 3.7 ± 5.6% vs. control; NS; n = 6). Taken together, these data indicate that decreases in TRPV4 levels inhibit ATP release from the thick ascending limb.
TRPV4 can be chemically activated by the phorbol ester 4α-phorbol-12,13-didecanoate (4α-PDD) (29). Thus we next tested whether activation of TRPV4 by 4α-PDD (1 μM) in isosmotic conditions stimulated ATP release from thick ascending limbs. After treatment of tubules with 4α-PDD, in the absence of changes in osmolality, thick ascending limbs released 25.4 ± 7.4 pmol ATP/mg protein. In contrast, after treatment with the vehicle for 4α-PDD, thick ascending limbs released only 3.6 ± 0.9 pmol ATP/mg protein (Fig. 6; P < 0.04 vs. 4α-PDD; n = 6). These data indicate that pharmacological activation of TRPV4 stimulates ATP release in the thick ascending limb. Taken together, these data indicate that TRPV4 mediates hypotonicity-induced ATP release in the thick ascending limb.
Decreased osmolality in the thick ascending limb causes cell swelling. This activates K-Cl cotransport, initiating a regulatory volume decrease (7). To test whether cell swelling or regulatory volume decrease stimulates ATP release, we performed the experiments in the presence of the K-Cl cotransporter inhibitor 2-n-butyl-6,7-dichloro-2-cyclopentyl-2,3-dihydro-1-oxo-1H-inden-5-yl-oxy acetic acid (DIOA; 100 μM, Sigma-Aldrich). This inhibitor has been previously used to block regulatory volume decreases (3, 28). We found that in the absence of DIOA, after a decrease in bath osmolality to 323 mosmol/kgH2O, thick ascending limbs released 34.8 ± 5.1 pmol ATP/mg protein. Similarly, in the presence of DIOA, after a decrease in bath osmolality to 323 mosmol/kgH2O, thick ascending limbs released 35.5 ± 4.9 pmol ATP/mg protein (NS vs. control; n = 5). These data indicate that cell swelling, rather than a regulatory volume decrease, stimulates ATP release by the thick ascending limb.
Decreases in osmolality have been shown to activate ATP release in other cell types. Thick ascending limb cells are subject to osmotic stress. Consequently, we studied whether decreases in osmolality stimulated ATP release from the thick ascending limb. We found that a 9% decrease in osmolality (from 350 to 323 mosmol/kgH2O) increased ATP release by 16-fold. We chose to start with a value of 350 mosmol/kgH2O because we had previously shown that the normal osmolality of the renal medulla in rats was around this value (14).
ATP release caused by hypotonicity could be due to cell swelling (8) or cell damage. To preserve the integrity of the cells, we chose to add a hypotonic solution instead of a solution with osmolality close to zero, because the latter before mixing could cause cell lysis. Also, we chose to decrease bath osmolality by only 27 mosmol/kgH2O because similar decreases did not cause cell damage in thick ascending limb cells (13). In addition, the ATP released from our preparations is <1% of the total pool of ATP; thus our results are not likely due to ATP release caused by cell damage.
In other cells, ATP release is an extracellular Ca-mediated process. To test whether extracellular Ca was implicated in osmolality-induced ATP release by the thick ascending limbs, we first measured osmolality-induced ATP release in the absence of extracellular Ca. We found that Ca-free solutions reduced ATP release by 72%. To demonstrate that an increase in intracellular Ca is required for osmolality-induced ATP release, we then measured osmolality-induced ATP release in the presence of the intracellular Ca chelator BAPTA. We found that BAPTA blocked ATP release by 81%.
ATP release in other epithelial cells is mediated by the Ca-permeable channel TRPV4 (1). Thus we measured osmolality-induced ATP release in the presence of the TRPV4 antagonist ruthenium red. Ruthenium red blocked hypotonicity-induced ATP release by 65%. Because ruthenium red is a nonspecific TRPV4 inhibitor, we inserted a siRNA sequence against TRPV4 into an adenoviral vector and performed in vivo knockdown of TRPV4. We found that in adenoviral-treated tubules, osmolality-induced ATP release decreased by 62%, while TRPV4 expression was reduced by 72%. Thus inhibition of ATP release was correlated with a decrease in TRPV4 expression.
To further demonstrate that TRPV4 is involved in ATP release from the thick ascending limb, we measured ATP release after adding the specific TRPV4 agonist 4α-PDD to the bath in isotonic conditions. We found that in the absence of changes in osmolality, pharmacological activation of TRPV4 stimulated ATP release by the thick ascending limb by sixfold. Taken together, these data indicate that reductions in osmolality stimulate ATP release by the thick ascending limb and this is mediated by TRPV4.
Osmolality-induced ATP release could be due to cell swelling or a volume-regulatory decrease. To test which activates TRPV4, we blocked the regulatory decrease with the K-Cl cotransport inhibitor DIOA and measured ATP release after decreases in osmolality. We found that in the presence of DIOA, decreases in bath osmolality stimulated ATP released to the same extent as in the absence of DIOA. These data suggest that ATP release is stimulated by cell swelling rather than by a regulatory volume decrease.
Our results showing ATP release is stimulated by cell swelling rather than regulatory volume decrease are similar to those found for other cell types. For instance, in hepatoma cell lines, hypotonicity-induced cell swelling enhances ATP release and the subsequent stimulation of P2 receptors. Activation of P2 receptors in these cells stimulates Cl efflux via K-Cl cotransport. Efflux of Cl contributes to the regulatory volume decrease by favoring water loss and cell volume recovery (8).
Decreases in osmolality may occur on a regular base in the renal medulla. Also, several factors could contribute to decreasing osmolality in the renal medulla. Medullary thick ascending limb cells are subject to decreases in interstitial osmolality after reduced salt reabsorption (14), increased medullary blood flow (6), use of loop diuretics (2), or hydration (25). Therefore, ATP release by the thick ascending limb could be part of an adaptation mechanism to changes in osmolality.
This is the first report demonstrating that decreasing osmolality by 9% in thick ascending limb tubule suspensions results in ATP release and this is mediated by TRPV4 activation. Our findings are supported by the fact that TRPV4 is involved in osmotic sensation in several types of epithelial cells. Renal epithelial cells, such as collecting duct cells, express TRPV4, and it serves as an osmosensor (30). Similarly, in airway epithelium, TRPV4 is involved in cell volume regulation after decreases in osmolality (1). In addition, mice lacking TRPV4 cannot maintain plasma osmolality at a constant value (19).
The present study shows the first description of the role of TRPV4 in mediating ATP release from a specific nephron segment. Our data are supported by similar reports in different epithelial cell models. Bladder epithelial cells derived from TRPV4 knockout mice have decreased ATP release following mechanical stimuli. This results in impaired bladder voiding in these animals (11). Also, biliary epithelial cells have been shown to detect changes in osmolality via TRPV4 activation. In these cells, stimulation of TRPV4 by decreasing osmolality or activation of TRPV4 with the specific agonist 4α-PDD leads to ATP release (12).
In the kidney, the effects of ATP release involve regulation of renal microvasculature and tubular function. Release of ATP mediates tubuloglomerular feedback. Macula densa cells release ATP from the basolateral side following increases in apical NaCl (17). Also, in vitro and in vivo data show that ATP release is involved in afferent arterial autoregulation (16). In addition, interstitial ATP levels have been shown to correlate with renal arterial pressure (20), supporting the hypothesis that ATP release contributes to adjustment of renal hemodynamics. ATP also regulates NaCl reabsorption in the kidney. In several nephron segments, extracellular ATP inhibits NaCl absorption (9, 18). We have reported that extracellular ATP stimulates nitric oxide (NO) production (26). Because we have previously shown that NO inhibits NaCl absorption (15) in the thick ascending limb, the ATP released after decreases in osmolality could inhibit NaCl absorption in this segment by stimulating NO, or affecting collecting duct Na and water transport (5), as well as vasa recta function (4).
In conclusion, we found that decreases in osmolality stimulate ATP release by the thick ascending limb, and this is mediated by the osmosensor TRPV4.
This work was supported in part by grants from the National Heart, Lung, and Blood Institute (HL-28982, HL-70985) to J. L. Garvin and from the American Heart Association-Greater Midwest (0615718Z) to G. B. Silva.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society