Transepithelial pressure pulses induce nucleotide release in polarized MDCK cells
http://www.100md.com
《美国生理学杂志》
Clinical Institute, Institute of Physiology, and The Water and Salt Research Center, University of Aarhus, Aarhus, Denmark
ABSTRACT
The release of nucleotides is involved in mechanosensation in various epithelial cells. Intriguingly, kidney epithelial cells are absolutely dependent on the primary cilium to sense changes in apical laminar flow. During fluid passage, the renal epithelial cells are subjected to various mechanical stimuli in addition to changes in the laminar flow rate. In the distal part of the collecting duct, the epithelial cells are exposed to pressure changes and possibly distension during papillary contractions. The aim of the present study was to determine whether nucleotide release contributes to mechanosensation in kidney epithelial cells, thereby establishing whether pressure changes are sufficient to produce nucleotide-mediated responses. Madin-Darby canine kidney (MDCK) cells grown on permeable supports were mounted in a closed double perfusion chamber on an inverted microscope. The intracellular Ca2+ concentration ([Ca2+]i) was monitored with the Ca2+-sensitive fluorescence probe fluo 4. Transepithelial pressure pulses of 30–80 mmHg produced a transient increase in [Ca2+]i of MDCK cells. This response is independent of the primary cilium, since it is readily observed in immature cells that do not yet express primary cilia. The amplitudes of the pressure-induced Ca2+ transients varied with the applied chamber pressure in a quantity-dependent manner. The ATPase apyrase and the P2Y antagonist suramin significantly reduced the pressure-induced Ca2+ transients. Applying apyrase or suramin to both sides of the preparation simultaneously nearly abolished the pressure-induced Ca2+ response. In conclusion, these observations suggest that rapid pressure changes induce both apical and basolateral nucleotide release that contribute to mechanosensation in kidney epithelial cells.
ATP; P2Y; calcium; mechanosensation
DURING FLUID PASSAGE, THE renal tubule epithelial cells are subjected to various types of mechanical stress, including changes in laminar fluid flow rates. In kidney epithelial cell lines, a change in fluid flow rates is sensed by the primary cilium (17, 20, 21). The primary cilium is a nonmotile elongated structure with a 9+0 organization of the microtubules pairs, which protrudes from the centriole and extends into the lumen of the kidney tubules. The primary cilium is present in all kidney epithelial cells except for the intercalated cells of the collecting duct (1). Its strategic position and bending properties as a function of perfusate flow rates led to the hypothesis that it may serve as a sensor of tubular flow (23). This hypothesis was substantiated by a study directly demonstrating that bending of the primary cilium either with a micropipette or increament of the perfusate flow rate induced an elevation of intracellular Ca2+ concentration ([Ca2+]i) in Madin-Darby canine kidney (MDCK) cells. Subsequently, the primary cilium was shown to be essential for the MDCK cells' ability to sense flow, since the deciliated cells were completely irresponsive to changes in fluid flow rate (21). Recent studies of perfused rabbit collecting duct indicate that increase of perfusate flow also stimulate [Ca2+]i elevations in the native tissue (12).
An interesting discovery merged into this entity when polycystin 1 and 2, the two products of the genes defective in autosomal-dominant polycystic kidney disease, were localized to the primary cilium of human and mice kidney epithelial cells (18, 27). Intriguingly, the presence of intact polycystin 1 and 2 appears to be a prerequisite for cilium-dependent, flow-stimulated elevation of [Ca2+]i (17). One current model for flow sensing is as follows: mechanical stress applied to the primary cilium results in conformational changes in polycystin 1 that further interact with the TRP channel, polycystin 2, to allow Ca2+ influx either in the shaft or in the base of the primary cilium (17). This Ca2+ influx induces Ca2+ release from intracellular stores either by itself or synergistically with yet undefined signal-transduction intermediates (20).
The fluid passage in the nephron and collecting duct, however, cannot fully be described by simple laminar flow. Papillary contractions create a distinct flow pattern in the medullary kidney structures. During contractions, the loop of Henle and the collecting duct in the papilla are compressed, and between the contractions the lumen of the tubules is distended as a result of boluses of urine passing (22). The collapse of the tubular lumen must be a consequence of the interstitial pressure increasing above the hydrostatic pressure in the kidney lumen. Thus these papillary contractions potentially submit the kidney epithelium to mechanical stress that is different from changes in laminar flow rates.
Here, we used MDCK cells as a simple model for the collecting duct. These cells are known to release ATP in response to mechanical stress such as stirring of the bath solution (8). In addition, MDCK cells functionally express different P2 receptors (19, 28) and thus should be able to respond to a potential mechanically induced nucleotide release in an autocrine and paracrine manner. The present study shows that transepithelial pressure gradients induce Ca2+ transients dependent on basolateral and apical nucleotide release in polarized MDCK cells. We speculate that nucleotide release might contribute to mechanosensation during fluid passage in kidney tubules.
METHODS
Cell culture. Wild-type (WT) MDCK cells (passages 54–70 from the American Type Culture Collection, Rockville, MD) were grown to confluence on 25-mm-diameter coverslips in DMEM with 10% fetal bovine serum (GIBCO, Grand Island, NY) and 2 mM glutamine, but without riboflavin or antibiotics as previously described (20).
Solutions. The perfusion solution had the following composition (in mM): 137 Na+, 5.3 K+, 1.8 Ca2+, 0.8 Mg2+, 126.9 Cl–, 0.8 SO42–, 14 HEPES, 5.6 glucose, and 5 probenecid, pH 7.4 (37°C, 300 mosmol/l). The Ca2+-free solution had the following composition (in mM): 139 Na+, 5.3 K+, 0.8 Mg2+, 125.3 Cl–, 0.8 SO42–, 1 EGTA, 14 HEPES, 5.6 glucose, and 5 probenecid, pH 7.4 (37°C, 300 mosmol/l). Sources of chemicals were fluo 4-AM (Molecular Probes, Eugene, OR) and EGTA, probenecid, apyrase (grade 1), and suramin (Sigma, St. Louis, MO).
Microscopy and perfusion. MDCK cell monolayers, grown on Anopore filters (Nunc, Roskilde, Denmark) with a pore size of 0.2 μm, were viewed in a perfusion chamber at 37°C on the stage of an inverted microscope (TE-2000, Nikon, BBT-LifeScience) equipped with differential interference contrast (DIC) combined with low-light-level fluorescence provided via a xenon lamp and monochromator (Visitech International, Sunderland, UK). Imaging was performed with a long-distance Plan Fluo x20, 0.45 normal aperture (Nikon, Copenhagen, Denmark), an intensified SVGA CCD camera, and imaging software (Quanticell 2000/Image Pro, Visitech). The cellular fluorescence was sampled at a rate of 0.67 Hz, and measurements were initiated 50 s before changes in the chamber pressure. The custom-made double-sided cell chamber is a closed perfusion system, which, in contrast to the chambers used to study mechanical stimulation of the primary cilium (see Ref. 20), allows the buildup of pressure gradients over the cell layer. The cell chamber consists of two symmetrical compartments separated by an Anopore filter on which the MDCK cells were grown (see Fig. 1A). The inner dimensions of the two compartments in this slit-shaped chamber were 6 (length) x 1 (width) x 2 (height) mm. Solutions were perfused at constant flow rates of 1.7 μl/s, which corresponds to a bulk flow velocity of 800 μm/s. To build up a pressure gradient over the epithelium (short pressure pulses), it was necessary to discontinue the flow on the particular side to which the pressure was to be applied by blocking the outflow line. The pressure pulse was induced by rapid compression of the inflow line with a metal clamp (see Fig. 1A). The opposite compartment was continuously perfused unless stated otherwise.
Intracellular Ca2+ measurements by fluo 4. The cells were incubated for 30 min with the Ca2+-sensitive probe fluo 4-AM (5 mM) at 37°C and washed twice to remove excess probe. Then, they were placed in the perfusion chamber and allowed at least a 20-min deesterification period. Fluo 4 was excited at 488 nm, and emission was detected above 520 nm. The fluorescence intensity was expressed relative to the baseline value, chosen as the mean of five intensity observations before the experimental manipulation. All solutions contained 5 mM probenecid to inhibit extrusion of the dye, and the experiments were carried out at 37°C, pH 7.4.
Pressure. The change in pressure within the chamber was measured by connecting two pressure transducers (Baxter, Irvine, CA) to the cell chamber, one for the apical and one for the basal compartment. The pressure transducer was connected exactly at the outflow from the chamber and thus was situated at the level of the cell chamber at the microscope stand. The data were collected via a Cardiomed-CM-2000 (Medi-Stim, Oslo, Norway). Before each experiment, the pressure was calibrated through the internal calibration routines in the CM-2000, against a solution column of isotonic NaCl.
Statistics. All values are shown as means ± SE. Statistical significance was determined using one-way ANOVA followed by a Tukey-Kramer multiple comparison test. P values <0.05 were considered significant. The number of observations refers to the number of cells analyzed. In each experiment, 18–20 cells were chosen randomly at the first picture in the image sequence, at the baseline.
RESULTS
Pressure changes within a perfusion chamber for mechanical stimulation of polarized epithelia. The purpose of the study was to investigate pressure-mediated mechanical stimulation of polarized kidney epithelial cells. We designed a perfusion chamber that would allow defined mechanical stimulation to epithelia grown on permeable supports. The chamber is schematically reproduced in Fig. 1A. It consists of two symmetrically half-sides separated only by the MDCK cells on the permeable support, with their apical side facing toward the objective lens. This defines an "apical" and a "basolatertal" fluid compartment that can be perfused separately. A stiff mineralic Anopore membrane was selected for support. To minimize a pressure-induced deflection of the membrane, the chamber's slit aperture was designed to be as narrow as possible (1 mm). These properties result in negligible pressure transmission between the apical and basolateralt bath when MDCK cells are grown to confluence (Fig. 1B). Pressure changes were recorded by replacing the chamber's outflow lines with pressure transducers. The two inflow lines to the chamber consisted of stiff polyethylene plastic tubing. Pressure application was performed on the inflow side of the chamber by compressing an interspersed 3-cm-long piece of silicon tubing (internal diameter 3 mm) using a metal clamp. The manual clamping of the silicon tube led to complete closure of the inflow line. The onset of pressure application was rapid (Fig. 1B), and the clamp was removed after 1 s. If the clamp was kept in place, the elevated pressure slowly declined toward baseline values over a period of some 80 s (data not shown).
The top and bottom panels in Fig. 1B display the pressure changes in the apical and basolateral compartment of the flow chamber, respectively, in the presence of confluent MDCK cells. The first three identical clamping events were performed on the basolateral inflow line and show rapid pressure increases, which are observed exclusively on the basolateral side (82.9 ± 3.2 mmHg, n = 60). Only a very minor pressure increase occurred on the contralateral side (1.95 ± 0.15 mmHg, n = 60). Subsequently, we applied three pressure stimuli on the apical side and observed the reciprocal; e.g., pressure nearly exclusively increased on the apical side (80.9 ± 2.9 mmHg, n = 62) but not on the basolateral (2.68 ± 0.19 mmHg, n = 62). The pressure pulses were highly reproducible. This system was then used to investigate the ability of the MDCK cells to respond with [Ca2+]i elevations to rapid pressure pulses.
Pressure pulses induce a Ca2+ transient in confluent MDCK cells. Figure 2, A and B, shows that pressure pulses from either side led to rapid and reversible elevations of [Ca2+]i. The flow was discontinued on the side of pressure pulse application by blocking the outflow, while fluid flow was continuously present on the contralateral side. In Fig. 2A, the apical pressure pulse was repeated three times as indicated by the arrows. The present experiment represents the average fluorescence sampled from 19 randomly selected cells. The 19 cells were chosen under baseline conditions, that is, without knowledge of the fate of one particular cell. It is shown that the amplitude of the [Ca2+]i transients declines with repeated pressure applications. This pattern was the most common, but other constellations could be observed (for example, see Fig. 6B). The highest transient of three repeated pressure pulses was used as the basis for mean calculations (1.79 ± 0.04, n = 52). The mean amplitude of the [Ca2+]i transient in all cells tested is shown in the right panel of Fig. 2A. When intracellular Ca2+ was allowed to return completely to prestimulation levels, the pressure-induced [Ca2+]i transient had the same amplitude as the previous one (see Fig. 6B, right). In this way, repeated trains of pressure pulses can be used to evaluate the effects of various antagonists.
Applying pressure pulses to the basal side of the preparation similarly induced [Ca2+]i transients (Fig. 2B) (2.40 ± 0.09, n = 57). The shape and amplitude of the [Ca2+]i transients in response to basal applied pressure pulses were closely similar to the [Ca2+]i transients following apical pressure pulses.
The next task was to investigate the relationship between the magnitude of the applied pressure pulse and the resulting [Ca2+]i transient in polarized MDCK cells. Choosing metal clamps of different sizes for compression of the inflow lines varied the pressure pulse amplitude. The duration of the compression was kept constant. The width of the clamp (2–16 mm) correlated with the amplitude of the clamping-induced pressure pulse; e.g., a wider clamp will produce larger pressure pulses. By simultaneously measuring the Ca2+ concentration and the chamber pressure, a given pressure gradient can be correlated to its corresponding [Ca2+]i transient. In this manner, it was possible to obtain a pressure-response curve (Fig. 2C) shown here for apical pressure pulses. The induced [Ca2+]i increases were significant even at a small pressure pulse amplitude of 30 mmHg (P < 0.001). Furthermore, the pressure-response curve showed saturation behavior, whereby elevations >100 mmHg had no additional effect.
Pressure pulse-induced [Ca2+]i transient is inhibited by apyrase. The ATPase apyrase was used to test whether the pressure-induced [Ca2+]i responses were related to release of nucleotides. Assuming the pressure pulse leads to nucleotide release followed by auto- and paracrine activation of P2 receptors, an extracellular nucleotide-scavenging system like apyrase should inhibit the observed effect.
It is well established that MDCK cells, including their different subclones, express a variety of different P2 receptors on either side of the epithelium (5, 28). First, we tested whether the purchased batch of apyrase showed the desired ATPase activity. A known amount of ATP was incubated with the enzyme before the experiment, and then the mixture was applied apically to confluent MDCK cells (Fig. 3A). Preincubation with apyrase completely removes the ATP-induced Ca2+ response in MDCK cells, and thus the enzyme shows significant ATPase/ADPase activity.
Ipsilateral application of apyrase (1 U/ml) to confluent MDCK cells significantly reduced the pressure pulse-induced [Ca2+]i response. This is exemplified in Fig. 3B, showing the effect of apcial apyrase on apical pressure pulses. Apyrase reduced the amplitude of the apical pressure pulse to 1.22 ± 0.02 (n = 55). This effect was fully reversible, as washout of apyrase restored the amplitude of the pressure-induced [Ca2+]i response to 2.15 ± 0.14 (n = 38). Similarly, basolateral apyrase greatly inhibited the effect of a basolateral pressure pulse (from 1.93 ± 0.1, n = 19, to 1.23 ± 0.04, n = 38, Fig. 3C). Again, the full response of pressure-induced [Ca2+]i increases could be restored by washing out apyrase (1.74 ± 0.07, n = 0.07). The data strongly indicate that the pressure pulse-induced [Ca2+]i elevation is caused by release of nucleotides (ATP or UTP).
In the following, we consider whether the nucleotides are released exclusively to one side of the epithelium or if the pressure pulses induce a bilateral release. To avoid washout of nucleotides, the sidedness of the nucleotide release was addressed under conditions where the flow was ceased in both the apical and basal chamber. Apyrase was included in the apical bath as indicated, and the pressure pulse was applied basolaterally (Fig. 4A). This procedure reduced the amplitude of the [Ca2+]i transient from 1.89 ± 0.03 (n = 111) to 1.51 ± 0.04 (n = 73, P < 0.001). Assuming that tight junctions are impermeable to nucleotides, this means that basally applied pressure pulses induce apical ATP release. Apyrase added to both sides simultaneously resulted in a change in [Ca2+]i of only 1.14 ± 0.01 (n = 131). In a similar fashion, basal apyrase affects the apically induced pressure response. The results are summarized in Fig. 4C.
Pressure pulse-induced [Ca2+]i transient is inhibited by suramin. Another way to interfere with nucleotide-mediated responses is to use P2Y receptor antagonists. Suramin, a nonselective inhibitor of P2Y receptors, was chosen to investigate the dependency of P2Y receptors on the pressure-induced [Ca2+]i response. Figure 5A shows the dose-dependent inhibition of suramin on the ATP-induced (25 μM) [Ca2+]i transient in confluent MDCK cells. Half-maximal inhibition is seen at a concentration of 30 μM suramin. A concentration of 300 μM results in near-complete inhibition of the apically induced ATP response and was thus chosen to evaluate the effect of apical P2Y receptors on the pressure-induced [Ca2+]i response. Suramin (300 μM) significantly reduced the apical pressure-induced [Ca2+]i increase when applied apically (1.13 ± 0.2, n = 76 with suramin vs. 1.60 ± 0.03, n = 76, Fig. 5B). The inhibitory effect of suramin was partially reversible 1–2 min after the blocker was washed out (1.31 ± 0.02, n = 76).
Suramin can also affect the pressure-induced response when applied basolaterally. The effect of suramin on pressure-induced [Ca2+]i responses is summerized in Fig. 5C. Consistently, suramin was able to reduce the pressure pulses by ipsi- and contralateral application. These data are consistent with the interpretation that the pressure pulse-induced [Ca2+]i transients require P2 receptor activation.
Pressure-induced Ca2+ response is independent of the primary cilium. MDCK cells do not present primary cilia until several days after they reached confluence (26). The flow-induced Ca2+ signal in MDCK cells is known to be absolutely dependent on the primary cilium (20, 21). We tested whether the pressure-induced Ca2+ changes could be induced in nonconfluent MDCK cells. The MDCK cells were 95% confluent on Anopore filters and placed in the double-perfusion chamber. Apical pressure pulses induce a very similar change in the intracellular Ca2+ response, as seen in the confluent polarized MDCK cells (1.92 ± 0 05, n = 129), and the changes therefore appear independent of the presence of a primary cilium (Fig. 6A). Apyrase was used to confirm the origin of the pressure-induced Ca2+ response in nonconfluent cells. Apyrase (1 U/ml) significantly reduced the pressure-induced Ca2+ response in nonconfluent cells from 1.76 ± 0.07, n = 75, to 1.21 ± 0.03, n = 57 (Fig. 6B), as would be expected from the observations in the polarized MDCK cells.
The independency of the primary cilium was confirmed in mature MDCK cells where the primary cilium was removed with chloral hydrate. The cells were treated with chloral hydrate (4 mM) for 96 h and allowed 24 h of recovery in normal medium. Apical pressure pulses resulted in a [Ca2+]i increase of 2.02 ± 0.06 (n = 133). The effect of chloral hydrate was verified by immunocytochemistry with antibody against bovine -tubulin as described elsewhere (21). The cells did not respond to changes in apical perfusate flow rates (data not shown).
DISCUSSION
Sensing flow by MDCK cells. MDCK cells originate from connecting tubules or collecting ducts of the canine kidney and are able to respond to changes in apical perfusate flow rate with a Ca2+ transient (20). The cells are absolutely dependent on the primary cilium to sense changes in perfusate flow rate (21). Mechanical stress applied to the cilium induces an intracellular Ca2+ response that depends on extracellular Ca2+ but also involves release from inositol 1,4,5-trisphosphate-sensitive Ca2+ stores. A recent study has suggested that the initial Ca2+ influx may occur through polycystin 2 in kidney epithelial cells (17), and it is speculated that sensory function of the primary cilium is important in the pathogenesis of polycystic kidney disease. However, the native collecting ducts are not only subjected to changes in laminar fluid flow rate as fluid passes along the kidney tubules. Papillary contractions result in the passage of fluid boluses rather than a constant laminar flow (22); thus other physical stimuli had to be considered in addition to changes in laminar fluid flow when mechanosensation in the kidney epithelial cells was addressed. We designed a double-perfusion flow chamber that allowed the investigation of rapid changes in transepithelial pressure using MDCK cells as a model for the collecting duct.
Mechanosensation and nucleotide release. Many forms of mechanical stimulations are considered adequate stimuli for ATP release in various cell types (6, 11, 14, 16, 24). Nucleotide release has been shown to be essential for the ability to sense fluid shear stress in endothelial cells (14, 15). It has been emphasized that collected wall stress rather than the rate of fluid flow itself induces the release of nucleotides (25). A similar mechanism might contribute to mechanosensation in the collecting duct.
Indeed, MDCK-D1 cells release ATP on vigorous stirring of the bath (8), and in addition, express the P2Y1, P2Y2, and P2Y11 receptors (19). Agonists applied both apically and basolaterally resulted in increases in [Ca2+]i in MDCK-C7 cells (5). MDCK-WT cells respond with intracellular Ca2+ increases on both apical and basolateral ATP application (data not shown) and thus possess the apparatus for ATP-mediated mechanosensation.
Nucleotide release from MDCK cells. The presented data show that mature, polarized MDCK-WT cells exhibit a nucleotide-dependent Ca2+ response as a result of short pressure pulses applied in a double-sided closed perfusion chamber. When cells are grown to confluence on stiff, permeate filters, a short-lived pressure gradient of 80 mmHg can be produced by compression of the inflow line to the chamber. The pressure pulse produces a transient increase in [Ca2+]i. The amplitude of the Ca2+ transient is dependent on the size of the pressure implemented over the epithelium. A pressure difference of 30 mmHg is sufficient to raise the Ca2+ concentration over baseline. The pressure-induced Ca2+ response is similar regardless of from which side the pressure pulse is applied. There was no indication that the applied mechanical stimulation damaged the cells. The transmitted light images obtained simultaneously with the fluorescence measurements did not show any obvious changes in cell structure. Futhermore, it was indeed possible to restimulate the cells to a full amplitude after [Ca2+]i was allowed to return to baseline, either by applying yet another mechanical pulse or by adding ATP (100 μM) to the perfusate (data not shown). Thus we feel safe to conclude that the applied pressure pulse did not induce general cell lysis.
The pressure-induced [Ca2+]i transients are almost completely abolished by the ATPase apyrase and the nonselective P2Y receptor antagonist suramin. The nucleotide release appears to occur on both sides of the epithelium, since apical apyrase reduces the Ca2+ response to basal pressure pulses and vice versa. The response is almost completely abolished when either apyrase or suramin is present on both sides of the epithelium simultaneously. This is consistent with the findings in polarized bronchial epithelia (7). Interestingly, in renal epithelial A6 cells, hyperosmotic cell swelling seems primary to release ATP into the lateral intracellular spaces (9). However, the constant flow at the apical surface of their system might have washed away any nucleotides released apically.
Pressure pulse contraflow-induced Ca2+ response in MDCK cells. Rapid pressure pulses will create short-lived waves of increases in fluid flow in the cell chamber. As mentioned above, sensing of fluid flow has recently been shown to be absolutely dependent on the primary cilium in kidney epithelial cells (17, 20, 21). However, the pressure-induced Ca2+ response described here can be distinguished from the cilium-dependent, flow-induced Ca2+ response previously described for MDCK cells. Immature nonconfluent MDCK cells do not present primary cilia (20, 26). Neither immature MDCK cells nor cells from which the primary cilium has been chemically removed are able to respond to changes in perfusate flow (21). However, the pressure-induced Ca2+ response can be provoked in immature, nonconfluent and chloral hydrate-treated MDCK cells equally well, which therefore does not reflect a change in perfusate flow rate. In addition, both apically and basolaterally induced pressure changes results in a very similar Ca2+ response, and basal pressure increases are not likely to changes the apical flow pattern. Finally, the pressure-induced, nucleotide-dependent Ca2+ response can be repeated with close to the same amplitude if the intracellular Ca2+ concentration is allowed to return to baseline. The cilium-dependent flow response in MDCK cells is known to be refractory to additional stimulation. First, after 25 min, a second flow stimulus results in a Ca2+ signal, with an amplitude equal to the first stimulus (20). Therefore, the pressure-induced Ca2+ response does not require a primary cilium and is most likely the result of nucleotide release produced by stress to the plasma membrane.
Flow and mechanical stimulation in the intact kidney. As previously mentioned, fluid passage in the kidney tubules afflicts epithelial cells with different types of mechanical stimulation. The epithelial cells of the loop of Henle and the collecting duct are potentially subjected to changes in fluid flow and stretch/changes in pressure and osmolarity. On the basis of the cell culture experiments, one would assume that the primary cilium in the intact collecting duct could be the sensor of changes in laminar fluid flow. A mathematical model based on observed flow-induced changes in [Ca2+]i in isolated, perfused tubules is consistent with this view (12). However, renal papillary smooth muscle cells produce contractions of the papilla with a frequency of 2–3/min in humans (2). These contractions repeatedly result in complete collapse of the lumen of loops of Henle, the vasa recta, and the medullary collecting duct. When a bolus of urine passes, the tubules are distended to allow fluid passage (22). Thus the epithelial cells in the involved tubules are exposed to changes in both fluid flow and pressure. To our knowledge, the magnitude of pressure changes in the kidney tubles and the interstitium during papillary contractions has not been reported. The vasa recta have been shown to close when the interstitial pressure exceeds the intravascular hydrostatic pressure by 4 mmHg (13). Since the hydrostatic pressure in the vasa recta has been measured to be 13.8 mmHg in rats (4), it is reasonable to assume that the hydrostatic pressure in the papilla during papillary contractions reaches a value of at least 16–18 mmHg. It has been shown for the proximal tubule that, at least under certain conditions, transepithelial pressure gradients can be built up in the kidney (3). Simultaneous pressure measurements in proximal tubules and the adjacent capillaries show that under tubular occlusion, the transepithelial pressure gradients can amount to 30 cmH2O (3). Pressure changes of 30 mmHg did produce small changes in the [Ca2+]i in the closed perfusion chamber, and a threefold higher pressure was needed to obtain a full nucleotide-dependent Ca2+ response.
It is likely, however, that the pressure gradients across the epithelium during papillary contractions are larger than the differences in hydrostatic pressure alone. Hyaluronan, an unbranched polysaccharide, present in the interstitium of the inner medulla, is compressible during papillary contractions. The potential energy stored in the compressed hyaluronan will, as the compression ceases, result in negative tissue pressure (for a review, see Ref. 10). It has been speculated that the negative tissue pressure thereby generated is quite substantial and important for the urine concentration process (10).
The present data suggest that kidney epithelial cells are able to release nucleotides to both sides of the epithelium with a transepithelial pressure change as the sole stimulus. Further studies in perfused kidney tubules and measurements of the pressure gradient in the intact kidney are necessary to resolve whether pressure-induced nucleotide release is critical for mechanosensation in the kidney. It will be crucial to resolve whether the phenomenon is important under normal physiological conditions in the kidney or is solely pertinent for high-pressure situations, such as ureteral obstruction. Bearing this in mind, we believe that pressure-induced nucleotide release should be considered when mechanosensation in the renal epithelium is addressed.
ACKNOWLEDGMENTS
We thank the following foundations for support: The Danish Medical Research Foundation, Grundforskningsfonden, Nyreforeningens Forskningsfond, The Aarhus University Research Foundation, Eva and Henry Frnkels Mindefond, and The A. P. Mller Foundation for the Advancement of Medical Science.
FOOTNOTES
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.
REFERENCES
Andrews PM and Porter KR. A scanning electron microscopic study of the nephron. Am J Anat 140: 81–115, 1974.
Bjrk L and Nylen O. Cineradiographic investigations of contraction in the normal upper urinary tract in man. Acta Radiol Diagn (Stockh) 12: 25–33, 1972.
Brenner BM, Troy JL, and Daugharty TM. Pressure in cortical structures of the rat kidney. Am J Physiol 222: 246–251, 1972.
Burnett JC Jr, Haas JA, and Larson MS. Renal interstitial pressure in mineralocorticoid escape. Am J Physiol Renal Fluid Electrolyte Physiol 249: F396–F399, 1985.
Gordjani N, Nitschke R, Greger R, and Leipziger J. Capacitative Ca2+ entry (CCE) induced by luminal and basolateral ATP in polarised MDCK-C7 cells is restricted to the basolateral membrane. Cell Calcium 22: 121–128, 1997.
Grygorczyk R and Hanrahan JW. CFTR-independent ATP release from epithelial cells triggered by mechanical stimuli. Am J Physiol Cell Physiol 272: C1058–C1066, 1997.
Homolya L, Steinberg TH, and Boucher RC. Cell to cell communication in response to mechanical stress via bilateral release of ATP and UTP in polarized epithelia. J Cell Biol 150: 1349–1360, 2000.
Insel PA, Ostrom RS, Zambon AC, Hughes RJ, Balboa MA, Shehnaz D, Gregorian C, Torres B, Firestein BL, Xing M, and Post SR. P2Y Receptors of MDCK cells: epithelial cell regulation by extracellular nucleotides. Clin Expr Pharm Physiol 28: 351–354, 2001.
Jans D, Srinivas SP, Waelkens E, Segal A, Lariviere E, Simaels J, and Van Driessche W. Hypotonic treatment evokes biphasic ATP release across the basolateral membrane of cultured renal epithelia (A6). J Physiol 545: 543–555, 2002.
Knepper MA, Saidel GM, Hascall VC, and Dwyer T. Concentration of solutes in the renal inner medulla: interstitial hyaluronan as a mechano-osmotic transducer. Am J Physiol Renal Physiol 284: F433–F446, 2003.
Lazarowski ER and Harden TK. Quantitation of extracellular UTP using a sensitive enzymatic assay. Br J Pharmacol 127: 1272–1278, 1999.
Liu W, Xu S, Woda C, Kim P, Weinbaum S, and Satlin LM. Effect of flow and stretch on the [Ca2+]i response of principal and intercalated cells in cortical collecting duct. Am J Physiol Renal Physiol 285: F998–F1012, 2003.
MacPhee PJ and Michel CC. Subatmospheric closing pressures in individual microvessels of rats and frogs. J Physiol 484: 183–187, 1995.
Milner P, Bodin P, Loesch A, and Burnstock G. Rapid release of endothelin and ATP from isolated aortic endothelial cells exposed to increased flow. Biochem Biophys Res Commun 170: 649–656, 1990.
Milner P, Bodin P, Loesch A, and Burnstock G. Increased shear stress leads to differential release of endothelin and ATP from isolated endothelial cells from 4- and 12-month-old male rabbit aorta. J Vasc Res 29: 420–425, 1992.
Nakamura F and Strittmatter SM. P2Y1 purinergic receptors in sensory neurons: contribution to touch-induced impulse generation. Proc Natl Acad Sci USA 93: 10465–10470, 1996.
Nauli SM, Alenghat FJ, Luo Y, Williams E, Vassilev P, Li X, Elia AE, Lu W, Brown EM, Quinn SJ, Ingber DE, and Zhou J. Polycystins 1 and 2 mediate mechanosensation in the primary cilium of kidney cells. Nat Genet 33: 129–137, 2003.
Pazour GJ, San Agustin JT, Follit JA, Rosenbaum JL, and Witman GB. Polycystin-2 localizes to the kidney cilia and the ciliary level is elevated in ORPK mice with polycystic kidney disease. Curr Biol 12: R378–R380, 2002.
Post SR, Rump LC, Zambon A, Hughes RJ, Buda MD, Jacobson JP, Kao CC, and Insel PA. ATP activates cAMP production via multiple purinergic receptors in MDCK-D1 epithelial cells. Blockade of an autocrine/paracrine pathway to define receptor preference of an agonist. J Biol Chem 273: 23093–23097, 1998.
Praetorius HA and Spring KR. Bending the MDCK cell primary cilium increases intracellular calcium. J Membr Biol 184: 71–79, 2001.
Praetorius HA and Spring KR. Removal of the MDCK cell primary cilium abolishes flow sensing. J Membr Biol 191: 69–76, 2003.
Schmidt-Nielsen B and Graves B. Changes in fluid compartments in hamster renal papilla due to peristalsis in the pelvic wall. Kidney Int 22: 613–625, 1982.
Schwartz EA, Leonard ML, Bizios R, and Bowser SS. Analysis and modeling of the primary cilium bending response to fluid shear. Am J Physiol Renal Physiol 272: F132–F138, 1997.
Wang Y, Roman R, Lidofsky SD, and Fitz JG. Autocrine signaling through ATP release represents a novel mechanism for cell volume regulation. Proc Natl Acad Sci USA 93: 12020–12025, 1996.
Yamamoto K, Sokabe T, Ohura N, Nakatsuka H, Kamiya A, and Ando J. Endogenously released ATP mediates shear stress-induced Ca2+ influx into pulmonary artery endothelial cells. Am J Physiol Heart Circ Physiol 285: H793–H803, 2003.
Yoder BK, Tousson A, Millican L, Wu JH, Bugg CE Jr, Schafer JA, and Balkovetz DF. Polaris, a protein disrupted in orpk mutant mice, is required for assembly of renal cilium. Am J Physiol Renal Physiol 282: F541–F552, 2002.
Yoder BK, Hou X, and Guay-Woodford LM. The polycystic kidney disease proteins, polycystin-1, polycystin-2, polaris, and cystin, are co-localized in renal cilia. J Am Soc Nephrol 13: 2508–2516, 2002.
Zegarra-Moran O, Romeo G, and Galietta LJ. Regulation of transepithelial ion transport by two different purinoceptors in the apical membrane of canine kidney (MDCK) cells. Br J Pharmacol 114: 1052–1056, 1995.(H. A. Praetorius, J. Frki)
ABSTRACT
The release of nucleotides is involved in mechanosensation in various epithelial cells. Intriguingly, kidney epithelial cells are absolutely dependent on the primary cilium to sense changes in apical laminar flow. During fluid passage, the renal epithelial cells are subjected to various mechanical stimuli in addition to changes in the laminar flow rate. In the distal part of the collecting duct, the epithelial cells are exposed to pressure changes and possibly distension during papillary contractions. The aim of the present study was to determine whether nucleotide release contributes to mechanosensation in kidney epithelial cells, thereby establishing whether pressure changes are sufficient to produce nucleotide-mediated responses. Madin-Darby canine kidney (MDCK) cells grown on permeable supports were mounted in a closed double perfusion chamber on an inverted microscope. The intracellular Ca2+ concentration ([Ca2+]i) was monitored with the Ca2+-sensitive fluorescence probe fluo 4. Transepithelial pressure pulses of 30–80 mmHg produced a transient increase in [Ca2+]i of MDCK cells. This response is independent of the primary cilium, since it is readily observed in immature cells that do not yet express primary cilia. The amplitudes of the pressure-induced Ca2+ transients varied with the applied chamber pressure in a quantity-dependent manner. The ATPase apyrase and the P2Y antagonist suramin significantly reduced the pressure-induced Ca2+ transients. Applying apyrase or suramin to both sides of the preparation simultaneously nearly abolished the pressure-induced Ca2+ response. In conclusion, these observations suggest that rapid pressure changes induce both apical and basolateral nucleotide release that contribute to mechanosensation in kidney epithelial cells.
ATP; P2Y; calcium; mechanosensation
DURING FLUID PASSAGE, THE renal tubule epithelial cells are subjected to various types of mechanical stress, including changes in laminar fluid flow rates. In kidney epithelial cell lines, a change in fluid flow rates is sensed by the primary cilium (17, 20, 21). The primary cilium is a nonmotile elongated structure with a 9+0 organization of the microtubules pairs, which protrudes from the centriole and extends into the lumen of the kidney tubules. The primary cilium is present in all kidney epithelial cells except for the intercalated cells of the collecting duct (1). Its strategic position and bending properties as a function of perfusate flow rates led to the hypothesis that it may serve as a sensor of tubular flow (23). This hypothesis was substantiated by a study directly demonstrating that bending of the primary cilium either with a micropipette or increament of the perfusate flow rate induced an elevation of intracellular Ca2+ concentration ([Ca2+]i) in Madin-Darby canine kidney (MDCK) cells. Subsequently, the primary cilium was shown to be essential for the MDCK cells' ability to sense flow, since the deciliated cells were completely irresponsive to changes in fluid flow rate (21). Recent studies of perfused rabbit collecting duct indicate that increase of perfusate flow also stimulate [Ca2+]i elevations in the native tissue (12).
An interesting discovery merged into this entity when polycystin 1 and 2, the two products of the genes defective in autosomal-dominant polycystic kidney disease, were localized to the primary cilium of human and mice kidney epithelial cells (18, 27). Intriguingly, the presence of intact polycystin 1 and 2 appears to be a prerequisite for cilium-dependent, flow-stimulated elevation of [Ca2+]i (17). One current model for flow sensing is as follows: mechanical stress applied to the primary cilium results in conformational changes in polycystin 1 that further interact with the TRP channel, polycystin 2, to allow Ca2+ influx either in the shaft or in the base of the primary cilium (17). This Ca2+ influx induces Ca2+ release from intracellular stores either by itself or synergistically with yet undefined signal-transduction intermediates (20).
The fluid passage in the nephron and collecting duct, however, cannot fully be described by simple laminar flow. Papillary contractions create a distinct flow pattern in the medullary kidney structures. During contractions, the loop of Henle and the collecting duct in the papilla are compressed, and between the contractions the lumen of the tubules is distended as a result of boluses of urine passing (22). The collapse of the tubular lumen must be a consequence of the interstitial pressure increasing above the hydrostatic pressure in the kidney lumen. Thus these papillary contractions potentially submit the kidney epithelium to mechanical stress that is different from changes in laminar flow rates.
Here, we used MDCK cells as a simple model for the collecting duct. These cells are known to release ATP in response to mechanical stress such as stirring of the bath solution (8). In addition, MDCK cells functionally express different P2 receptors (19, 28) and thus should be able to respond to a potential mechanically induced nucleotide release in an autocrine and paracrine manner. The present study shows that transepithelial pressure gradients induce Ca2+ transients dependent on basolateral and apical nucleotide release in polarized MDCK cells. We speculate that nucleotide release might contribute to mechanosensation during fluid passage in kidney tubules.
METHODS
Cell culture. Wild-type (WT) MDCK cells (passages 54–70 from the American Type Culture Collection, Rockville, MD) were grown to confluence on 25-mm-diameter coverslips in DMEM with 10% fetal bovine serum (GIBCO, Grand Island, NY) and 2 mM glutamine, but without riboflavin or antibiotics as previously described (20).
Solutions. The perfusion solution had the following composition (in mM): 137 Na+, 5.3 K+, 1.8 Ca2+, 0.8 Mg2+, 126.9 Cl–, 0.8 SO42–, 14 HEPES, 5.6 glucose, and 5 probenecid, pH 7.4 (37°C, 300 mosmol/l). The Ca2+-free solution had the following composition (in mM): 139 Na+, 5.3 K+, 0.8 Mg2+, 125.3 Cl–, 0.8 SO42–, 1 EGTA, 14 HEPES, 5.6 glucose, and 5 probenecid, pH 7.4 (37°C, 300 mosmol/l). Sources of chemicals were fluo 4-AM (Molecular Probes, Eugene, OR) and EGTA, probenecid, apyrase (grade 1), and suramin (Sigma, St. Louis, MO).
Microscopy and perfusion. MDCK cell monolayers, grown on Anopore filters (Nunc, Roskilde, Denmark) with a pore size of 0.2 μm, were viewed in a perfusion chamber at 37°C on the stage of an inverted microscope (TE-2000, Nikon, BBT-LifeScience) equipped with differential interference contrast (DIC) combined with low-light-level fluorescence provided via a xenon lamp and monochromator (Visitech International, Sunderland, UK). Imaging was performed with a long-distance Plan Fluo x20, 0.45 normal aperture (Nikon, Copenhagen, Denmark), an intensified SVGA CCD camera, and imaging software (Quanticell 2000/Image Pro, Visitech). The cellular fluorescence was sampled at a rate of 0.67 Hz, and measurements were initiated 50 s before changes in the chamber pressure. The custom-made double-sided cell chamber is a closed perfusion system, which, in contrast to the chambers used to study mechanical stimulation of the primary cilium (see Ref. 20), allows the buildup of pressure gradients over the cell layer. The cell chamber consists of two symmetrical compartments separated by an Anopore filter on which the MDCK cells were grown (see Fig. 1A). The inner dimensions of the two compartments in this slit-shaped chamber were 6 (length) x 1 (width) x 2 (height) mm. Solutions were perfused at constant flow rates of 1.7 μl/s, which corresponds to a bulk flow velocity of 800 μm/s. To build up a pressure gradient over the epithelium (short pressure pulses), it was necessary to discontinue the flow on the particular side to which the pressure was to be applied by blocking the outflow line. The pressure pulse was induced by rapid compression of the inflow line with a metal clamp (see Fig. 1A). The opposite compartment was continuously perfused unless stated otherwise.
Intracellular Ca2+ measurements by fluo 4. The cells were incubated for 30 min with the Ca2+-sensitive probe fluo 4-AM (5 mM) at 37°C and washed twice to remove excess probe. Then, they were placed in the perfusion chamber and allowed at least a 20-min deesterification period. Fluo 4 was excited at 488 nm, and emission was detected above 520 nm. The fluorescence intensity was expressed relative to the baseline value, chosen as the mean of five intensity observations before the experimental manipulation. All solutions contained 5 mM probenecid to inhibit extrusion of the dye, and the experiments were carried out at 37°C, pH 7.4.
Pressure. The change in pressure within the chamber was measured by connecting two pressure transducers (Baxter, Irvine, CA) to the cell chamber, one for the apical and one for the basal compartment. The pressure transducer was connected exactly at the outflow from the chamber and thus was situated at the level of the cell chamber at the microscope stand. The data were collected via a Cardiomed-CM-2000 (Medi-Stim, Oslo, Norway). Before each experiment, the pressure was calibrated through the internal calibration routines in the CM-2000, against a solution column of isotonic NaCl.
Statistics. All values are shown as means ± SE. Statistical significance was determined using one-way ANOVA followed by a Tukey-Kramer multiple comparison test. P values <0.05 were considered significant. The number of observations refers to the number of cells analyzed. In each experiment, 18–20 cells were chosen randomly at the first picture in the image sequence, at the baseline.
RESULTS
Pressure changes within a perfusion chamber for mechanical stimulation of polarized epithelia. The purpose of the study was to investigate pressure-mediated mechanical stimulation of polarized kidney epithelial cells. We designed a perfusion chamber that would allow defined mechanical stimulation to epithelia grown on permeable supports. The chamber is schematically reproduced in Fig. 1A. It consists of two symmetrically half-sides separated only by the MDCK cells on the permeable support, with their apical side facing toward the objective lens. This defines an "apical" and a "basolatertal" fluid compartment that can be perfused separately. A stiff mineralic Anopore membrane was selected for support. To minimize a pressure-induced deflection of the membrane, the chamber's slit aperture was designed to be as narrow as possible (1 mm). These properties result in negligible pressure transmission between the apical and basolateralt bath when MDCK cells are grown to confluence (Fig. 1B). Pressure changes were recorded by replacing the chamber's outflow lines with pressure transducers. The two inflow lines to the chamber consisted of stiff polyethylene plastic tubing. Pressure application was performed on the inflow side of the chamber by compressing an interspersed 3-cm-long piece of silicon tubing (internal diameter 3 mm) using a metal clamp. The manual clamping of the silicon tube led to complete closure of the inflow line. The onset of pressure application was rapid (Fig. 1B), and the clamp was removed after 1 s. If the clamp was kept in place, the elevated pressure slowly declined toward baseline values over a period of some 80 s (data not shown).
The top and bottom panels in Fig. 1B display the pressure changes in the apical and basolateral compartment of the flow chamber, respectively, in the presence of confluent MDCK cells. The first three identical clamping events were performed on the basolateral inflow line and show rapid pressure increases, which are observed exclusively on the basolateral side (82.9 ± 3.2 mmHg, n = 60). Only a very minor pressure increase occurred on the contralateral side (1.95 ± 0.15 mmHg, n = 60). Subsequently, we applied three pressure stimuli on the apical side and observed the reciprocal; e.g., pressure nearly exclusively increased on the apical side (80.9 ± 2.9 mmHg, n = 62) but not on the basolateral (2.68 ± 0.19 mmHg, n = 62). The pressure pulses were highly reproducible. This system was then used to investigate the ability of the MDCK cells to respond with [Ca2+]i elevations to rapid pressure pulses.
Pressure pulses induce a Ca2+ transient in confluent MDCK cells. Figure 2, A and B, shows that pressure pulses from either side led to rapid and reversible elevations of [Ca2+]i. The flow was discontinued on the side of pressure pulse application by blocking the outflow, while fluid flow was continuously present on the contralateral side. In Fig. 2A, the apical pressure pulse was repeated three times as indicated by the arrows. The present experiment represents the average fluorescence sampled from 19 randomly selected cells. The 19 cells were chosen under baseline conditions, that is, without knowledge of the fate of one particular cell. It is shown that the amplitude of the [Ca2+]i transients declines with repeated pressure applications. This pattern was the most common, but other constellations could be observed (for example, see Fig. 6B). The highest transient of three repeated pressure pulses was used as the basis for mean calculations (1.79 ± 0.04, n = 52). The mean amplitude of the [Ca2+]i transient in all cells tested is shown in the right panel of Fig. 2A. When intracellular Ca2+ was allowed to return completely to prestimulation levels, the pressure-induced [Ca2+]i transient had the same amplitude as the previous one (see Fig. 6B, right). In this way, repeated trains of pressure pulses can be used to evaluate the effects of various antagonists.
Applying pressure pulses to the basal side of the preparation similarly induced [Ca2+]i transients (Fig. 2B) (2.40 ± 0.09, n = 57). The shape and amplitude of the [Ca2+]i transients in response to basal applied pressure pulses were closely similar to the [Ca2+]i transients following apical pressure pulses.
The next task was to investigate the relationship between the magnitude of the applied pressure pulse and the resulting [Ca2+]i transient in polarized MDCK cells. Choosing metal clamps of different sizes for compression of the inflow lines varied the pressure pulse amplitude. The duration of the compression was kept constant. The width of the clamp (2–16 mm) correlated with the amplitude of the clamping-induced pressure pulse; e.g., a wider clamp will produce larger pressure pulses. By simultaneously measuring the Ca2+ concentration and the chamber pressure, a given pressure gradient can be correlated to its corresponding [Ca2+]i transient. In this manner, it was possible to obtain a pressure-response curve (Fig. 2C) shown here for apical pressure pulses. The induced [Ca2+]i increases were significant even at a small pressure pulse amplitude of 30 mmHg (P < 0.001). Furthermore, the pressure-response curve showed saturation behavior, whereby elevations >100 mmHg had no additional effect.
Pressure pulse-induced [Ca2+]i transient is inhibited by apyrase. The ATPase apyrase was used to test whether the pressure-induced [Ca2+]i responses were related to release of nucleotides. Assuming the pressure pulse leads to nucleotide release followed by auto- and paracrine activation of P2 receptors, an extracellular nucleotide-scavenging system like apyrase should inhibit the observed effect.
It is well established that MDCK cells, including their different subclones, express a variety of different P2 receptors on either side of the epithelium (5, 28). First, we tested whether the purchased batch of apyrase showed the desired ATPase activity. A known amount of ATP was incubated with the enzyme before the experiment, and then the mixture was applied apically to confluent MDCK cells (Fig. 3A). Preincubation with apyrase completely removes the ATP-induced Ca2+ response in MDCK cells, and thus the enzyme shows significant ATPase/ADPase activity.
Ipsilateral application of apyrase (1 U/ml) to confluent MDCK cells significantly reduced the pressure pulse-induced [Ca2+]i response. This is exemplified in Fig. 3B, showing the effect of apcial apyrase on apical pressure pulses. Apyrase reduced the amplitude of the apical pressure pulse to 1.22 ± 0.02 (n = 55). This effect was fully reversible, as washout of apyrase restored the amplitude of the pressure-induced [Ca2+]i response to 2.15 ± 0.14 (n = 38). Similarly, basolateral apyrase greatly inhibited the effect of a basolateral pressure pulse (from 1.93 ± 0.1, n = 19, to 1.23 ± 0.04, n = 38, Fig. 3C). Again, the full response of pressure-induced [Ca2+]i increases could be restored by washing out apyrase (1.74 ± 0.07, n = 0.07). The data strongly indicate that the pressure pulse-induced [Ca2+]i elevation is caused by release of nucleotides (ATP or UTP).
In the following, we consider whether the nucleotides are released exclusively to one side of the epithelium or if the pressure pulses induce a bilateral release. To avoid washout of nucleotides, the sidedness of the nucleotide release was addressed under conditions where the flow was ceased in both the apical and basal chamber. Apyrase was included in the apical bath as indicated, and the pressure pulse was applied basolaterally (Fig. 4A). This procedure reduced the amplitude of the [Ca2+]i transient from 1.89 ± 0.03 (n = 111) to 1.51 ± 0.04 (n = 73, P < 0.001). Assuming that tight junctions are impermeable to nucleotides, this means that basally applied pressure pulses induce apical ATP release. Apyrase added to both sides simultaneously resulted in a change in [Ca2+]i of only 1.14 ± 0.01 (n = 131). In a similar fashion, basal apyrase affects the apically induced pressure response. The results are summarized in Fig. 4C.
Pressure pulse-induced [Ca2+]i transient is inhibited by suramin. Another way to interfere with nucleotide-mediated responses is to use P2Y receptor antagonists. Suramin, a nonselective inhibitor of P2Y receptors, was chosen to investigate the dependency of P2Y receptors on the pressure-induced [Ca2+]i response. Figure 5A shows the dose-dependent inhibition of suramin on the ATP-induced (25 μM) [Ca2+]i transient in confluent MDCK cells. Half-maximal inhibition is seen at a concentration of 30 μM suramin. A concentration of 300 μM results in near-complete inhibition of the apically induced ATP response and was thus chosen to evaluate the effect of apical P2Y receptors on the pressure-induced [Ca2+]i response. Suramin (300 μM) significantly reduced the apical pressure-induced [Ca2+]i increase when applied apically (1.13 ± 0.2, n = 76 with suramin vs. 1.60 ± 0.03, n = 76, Fig. 5B). The inhibitory effect of suramin was partially reversible 1–2 min after the blocker was washed out (1.31 ± 0.02, n = 76).
Suramin can also affect the pressure-induced response when applied basolaterally. The effect of suramin on pressure-induced [Ca2+]i responses is summerized in Fig. 5C. Consistently, suramin was able to reduce the pressure pulses by ipsi- and contralateral application. These data are consistent with the interpretation that the pressure pulse-induced [Ca2+]i transients require P2 receptor activation.
Pressure-induced Ca2+ response is independent of the primary cilium. MDCK cells do not present primary cilia until several days after they reached confluence (26). The flow-induced Ca2+ signal in MDCK cells is known to be absolutely dependent on the primary cilium (20, 21). We tested whether the pressure-induced Ca2+ changes could be induced in nonconfluent MDCK cells. The MDCK cells were 95% confluent on Anopore filters and placed in the double-perfusion chamber. Apical pressure pulses induce a very similar change in the intracellular Ca2+ response, as seen in the confluent polarized MDCK cells (1.92 ± 0 05, n = 129), and the changes therefore appear independent of the presence of a primary cilium (Fig. 6A). Apyrase was used to confirm the origin of the pressure-induced Ca2+ response in nonconfluent cells. Apyrase (1 U/ml) significantly reduced the pressure-induced Ca2+ response in nonconfluent cells from 1.76 ± 0.07, n = 75, to 1.21 ± 0.03, n = 57 (Fig. 6B), as would be expected from the observations in the polarized MDCK cells.
The independency of the primary cilium was confirmed in mature MDCK cells where the primary cilium was removed with chloral hydrate. The cells were treated with chloral hydrate (4 mM) for 96 h and allowed 24 h of recovery in normal medium. Apical pressure pulses resulted in a [Ca2+]i increase of 2.02 ± 0.06 (n = 133). The effect of chloral hydrate was verified by immunocytochemistry with antibody against bovine -tubulin as described elsewhere (21). The cells did not respond to changes in apical perfusate flow rates (data not shown).
DISCUSSION
Sensing flow by MDCK cells. MDCK cells originate from connecting tubules or collecting ducts of the canine kidney and are able to respond to changes in apical perfusate flow rate with a Ca2+ transient (20). The cells are absolutely dependent on the primary cilium to sense changes in perfusate flow rate (21). Mechanical stress applied to the cilium induces an intracellular Ca2+ response that depends on extracellular Ca2+ but also involves release from inositol 1,4,5-trisphosphate-sensitive Ca2+ stores. A recent study has suggested that the initial Ca2+ influx may occur through polycystin 2 in kidney epithelial cells (17), and it is speculated that sensory function of the primary cilium is important in the pathogenesis of polycystic kidney disease. However, the native collecting ducts are not only subjected to changes in laminar fluid flow rate as fluid passes along the kidney tubules. Papillary contractions result in the passage of fluid boluses rather than a constant laminar flow (22); thus other physical stimuli had to be considered in addition to changes in laminar fluid flow when mechanosensation in the kidney epithelial cells was addressed. We designed a double-perfusion flow chamber that allowed the investigation of rapid changes in transepithelial pressure using MDCK cells as a model for the collecting duct.
Mechanosensation and nucleotide release. Many forms of mechanical stimulations are considered adequate stimuli for ATP release in various cell types (6, 11, 14, 16, 24). Nucleotide release has been shown to be essential for the ability to sense fluid shear stress in endothelial cells (14, 15). It has been emphasized that collected wall stress rather than the rate of fluid flow itself induces the release of nucleotides (25). A similar mechanism might contribute to mechanosensation in the collecting duct.
Indeed, MDCK-D1 cells release ATP on vigorous stirring of the bath (8), and in addition, express the P2Y1, P2Y2, and P2Y11 receptors (19). Agonists applied both apically and basolaterally resulted in increases in [Ca2+]i in MDCK-C7 cells (5). MDCK-WT cells respond with intracellular Ca2+ increases on both apical and basolateral ATP application (data not shown) and thus possess the apparatus for ATP-mediated mechanosensation.
Nucleotide release from MDCK cells. The presented data show that mature, polarized MDCK-WT cells exhibit a nucleotide-dependent Ca2+ response as a result of short pressure pulses applied in a double-sided closed perfusion chamber. When cells are grown to confluence on stiff, permeate filters, a short-lived pressure gradient of 80 mmHg can be produced by compression of the inflow line to the chamber. The pressure pulse produces a transient increase in [Ca2+]i. The amplitude of the Ca2+ transient is dependent on the size of the pressure implemented over the epithelium. A pressure difference of 30 mmHg is sufficient to raise the Ca2+ concentration over baseline. The pressure-induced Ca2+ response is similar regardless of from which side the pressure pulse is applied. There was no indication that the applied mechanical stimulation damaged the cells. The transmitted light images obtained simultaneously with the fluorescence measurements did not show any obvious changes in cell structure. Futhermore, it was indeed possible to restimulate the cells to a full amplitude after [Ca2+]i was allowed to return to baseline, either by applying yet another mechanical pulse or by adding ATP (100 μM) to the perfusate (data not shown). Thus we feel safe to conclude that the applied pressure pulse did not induce general cell lysis.
The pressure-induced [Ca2+]i transients are almost completely abolished by the ATPase apyrase and the nonselective P2Y receptor antagonist suramin. The nucleotide release appears to occur on both sides of the epithelium, since apical apyrase reduces the Ca2+ response to basal pressure pulses and vice versa. The response is almost completely abolished when either apyrase or suramin is present on both sides of the epithelium simultaneously. This is consistent with the findings in polarized bronchial epithelia (7). Interestingly, in renal epithelial A6 cells, hyperosmotic cell swelling seems primary to release ATP into the lateral intracellular spaces (9). However, the constant flow at the apical surface of their system might have washed away any nucleotides released apically.
Pressure pulse contraflow-induced Ca2+ response in MDCK cells. Rapid pressure pulses will create short-lived waves of increases in fluid flow in the cell chamber. As mentioned above, sensing of fluid flow has recently been shown to be absolutely dependent on the primary cilium in kidney epithelial cells (17, 20, 21). However, the pressure-induced Ca2+ response described here can be distinguished from the cilium-dependent, flow-induced Ca2+ response previously described for MDCK cells. Immature nonconfluent MDCK cells do not present primary cilia (20, 26). Neither immature MDCK cells nor cells from which the primary cilium has been chemically removed are able to respond to changes in perfusate flow (21). However, the pressure-induced Ca2+ response can be provoked in immature, nonconfluent and chloral hydrate-treated MDCK cells equally well, which therefore does not reflect a change in perfusate flow rate. In addition, both apically and basolaterally induced pressure changes results in a very similar Ca2+ response, and basal pressure increases are not likely to changes the apical flow pattern. Finally, the pressure-induced, nucleotide-dependent Ca2+ response can be repeated with close to the same amplitude if the intracellular Ca2+ concentration is allowed to return to baseline. The cilium-dependent flow response in MDCK cells is known to be refractory to additional stimulation. First, after 25 min, a second flow stimulus results in a Ca2+ signal, with an amplitude equal to the first stimulus (20). Therefore, the pressure-induced Ca2+ response does not require a primary cilium and is most likely the result of nucleotide release produced by stress to the plasma membrane.
Flow and mechanical stimulation in the intact kidney. As previously mentioned, fluid passage in the kidney tubules afflicts epithelial cells with different types of mechanical stimulation. The epithelial cells of the loop of Henle and the collecting duct are potentially subjected to changes in fluid flow and stretch/changes in pressure and osmolarity. On the basis of the cell culture experiments, one would assume that the primary cilium in the intact collecting duct could be the sensor of changes in laminar fluid flow. A mathematical model based on observed flow-induced changes in [Ca2+]i in isolated, perfused tubules is consistent with this view (12). However, renal papillary smooth muscle cells produce contractions of the papilla with a frequency of 2–3/min in humans (2). These contractions repeatedly result in complete collapse of the lumen of loops of Henle, the vasa recta, and the medullary collecting duct. When a bolus of urine passes, the tubules are distended to allow fluid passage (22). Thus the epithelial cells in the involved tubules are exposed to changes in both fluid flow and pressure. To our knowledge, the magnitude of pressure changes in the kidney tubles and the interstitium during papillary contractions has not been reported. The vasa recta have been shown to close when the interstitial pressure exceeds the intravascular hydrostatic pressure by 4 mmHg (13). Since the hydrostatic pressure in the vasa recta has been measured to be 13.8 mmHg in rats (4), it is reasonable to assume that the hydrostatic pressure in the papilla during papillary contractions reaches a value of at least 16–18 mmHg. It has been shown for the proximal tubule that, at least under certain conditions, transepithelial pressure gradients can be built up in the kidney (3). Simultaneous pressure measurements in proximal tubules and the adjacent capillaries show that under tubular occlusion, the transepithelial pressure gradients can amount to 30 cmH2O (3). Pressure changes of 30 mmHg did produce small changes in the [Ca2+]i in the closed perfusion chamber, and a threefold higher pressure was needed to obtain a full nucleotide-dependent Ca2+ response.
It is likely, however, that the pressure gradients across the epithelium during papillary contractions are larger than the differences in hydrostatic pressure alone. Hyaluronan, an unbranched polysaccharide, present in the interstitium of the inner medulla, is compressible during papillary contractions. The potential energy stored in the compressed hyaluronan will, as the compression ceases, result in negative tissue pressure (for a review, see Ref. 10). It has been speculated that the negative tissue pressure thereby generated is quite substantial and important for the urine concentration process (10).
The present data suggest that kidney epithelial cells are able to release nucleotides to both sides of the epithelium with a transepithelial pressure change as the sole stimulus. Further studies in perfused kidney tubules and measurements of the pressure gradient in the intact kidney are necessary to resolve whether pressure-induced nucleotide release is critical for mechanosensation in the kidney. It will be crucial to resolve whether the phenomenon is important under normal physiological conditions in the kidney or is solely pertinent for high-pressure situations, such as ureteral obstruction. Bearing this in mind, we believe that pressure-induced nucleotide release should be considered when mechanosensation in the renal epithelium is addressed.
ACKNOWLEDGMENTS
We thank the following foundations for support: The Danish Medical Research Foundation, Grundforskningsfonden, Nyreforeningens Forskningsfond, The Aarhus University Research Foundation, Eva and Henry Frnkels Mindefond, and The A. P. Mller Foundation for the Advancement of Medical Science.
FOOTNOTES
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.
REFERENCES
Andrews PM and Porter KR. A scanning electron microscopic study of the nephron. Am J Anat 140: 81–115, 1974.
Bjrk L and Nylen O. Cineradiographic investigations of contraction in the normal upper urinary tract in man. Acta Radiol Diagn (Stockh) 12: 25–33, 1972.
Brenner BM, Troy JL, and Daugharty TM. Pressure in cortical structures of the rat kidney. Am J Physiol 222: 246–251, 1972.
Burnett JC Jr, Haas JA, and Larson MS. Renal interstitial pressure in mineralocorticoid escape. Am J Physiol Renal Fluid Electrolyte Physiol 249: F396–F399, 1985.
Gordjani N, Nitschke R, Greger R, and Leipziger J. Capacitative Ca2+ entry (CCE) induced by luminal and basolateral ATP in polarised MDCK-C7 cells is restricted to the basolateral membrane. Cell Calcium 22: 121–128, 1997.
Grygorczyk R and Hanrahan JW. CFTR-independent ATP release from epithelial cells triggered by mechanical stimuli. Am J Physiol Cell Physiol 272: C1058–C1066, 1997.
Homolya L, Steinberg TH, and Boucher RC. Cell to cell communication in response to mechanical stress via bilateral release of ATP and UTP in polarized epithelia. J Cell Biol 150: 1349–1360, 2000.
Insel PA, Ostrom RS, Zambon AC, Hughes RJ, Balboa MA, Shehnaz D, Gregorian C, Torres B, Firestein BL, Xing M, and Post SR. P2Y Receptors of MDCK cells: epithelial cell regulation by extracellular nucleotides. Clin Expr Pharm Physiol 28: 351–354, 2001.
Jans D, Srinivas SP, Waelkens E, Segal A, Lariviere E, Simaels J, and Van Driessche W. Hypotonic treatment evokes biphasic ATP release across the basolateral membrane of cultured renal epithelia (A6). J Physiol 545: 543–555, 2002.
Knepper MA, Saidel GM, Hascall VC, and Dwyer T. Concentration of solutes in the renal inner medulla: interstitial hyaluronan as a mechano-osmotic transducer. Am J Physiol Renal Physiol 284: F433–F446, 2003.
Lazarowski ER and Harden TK. Quantitation of extracellular UTP using a sensitive enzymatic assay. Br J Pharmacol 127: 1272–1278, 1999.
Liu W, Xu S, Woda C, Kim P, Weinbaum S, and Satlin LM. Effect of flow and stretch on the [Ca2+]i response of principal and intercalated cells in cortical collecting duct. Am J Physiol Renal Physiol 285: F998–F1012, 2003.
MacPhee PJ and Michel CC. Subatmospheric closing pressures in individual microvessels of rats and frogs. J Physiol 484: 183–187, 1995.
Milner P, Bodin P, Loesch A, and Burnstock G. Rapid release of endothelin and ATP from isolated aortic endothelial cells exposed to increased flow. Biochem Biophys Res Commun 170: 649–656, 1990.
Milner P, Bodin P, Loesch A, and Burnstock G. Increased shear stress leads to differential release of endothelin and ATP from isolated endothelial cells from 4- and 12-month-old male rabbit aorta. J Vasc Res 29: 420–425, 1992.
Nakamura F and Strittmatter SM. P2Y1 purinergic receptors in sensory neurons: contribution to touch-induced impulse generation. Proc Natl Acad Sci USA 93: 10465–10470, 1996.
Nauli SM, Alenghat FJ, Luo Y, Williams E, Vassilev P, Li X, Elia AE, Lu W, Brown EM, Quinn SJ, Ingber DE, and Zhou J. Polycystins 1 and 2 mediate mechanosensation in the primary cilium of kidney cells. Nat Genet 33: 129–137, 2003.
Pazour GJ, San Agustin JT, Follit JA, Rosenbaum JL, and Witman GB. Polycystin-2 localizes to the kidney cilia and the ciliary level is elevated in ORPK mice with polycystic kidney disease. Curr Biol 12: R378–R380, 2002.
Post SR, Rump LC, Zambon A, Hughes RJ, Buda MD, Jacobson JP, Kao CC, and Insel PA. ATP activates cAMP production via multiple purinergic receptors in MDCK-D1 epithelial cells. Blockade of an autocrine/paracrine pathway to define receptor preference of an agonist. J Biol Chem 273: 23093–23097, 1998.
Praetorius HA and Spring KR. Bending the MDCK cell primary cilium increases intracellular calcium. J Membr Biol 184: 71–79, 2001.
Praetorius HA and Spring KR. Removal of the MDCK cell primary cilium abolishes flow sensing. J Membr Biol 191: 69–76, 2003.
Schmidt-Nielsen B and Graves B. Changes in fluid compartments in hamster renal papilla due to peristalsis in the pelvic wall. Kidney Int 22: 613–625, 1982.
Schwartz EA, Leonard ML, Bizios R, and Bowser SS. Analysis and modeling of the primary cilium bending response to fluid shear. Am J Physiol Renal Physiol 272: F132–F138, 1997.
Wang Y, Roman R, Lidofsky SD, and Fitz JG. Autocrine signaling through ATP release represents a novel mechanism for cell volume regulation. Proc Natl Acad Sci USA 93: 12020–12025, 1996.
Yamamoto K, Sokabe T, Ohura N, Nakatsuka H, Kamiya A, and Ando J. Endogenously released ATP mediates shear stress-induced Ca2+ influx into pulmonary artery endothelial cells. Am J Physiol Heart Circ Physiol 285: H793–H803, 2003.
Yoder BK, Tousson A, Millican L, Wu JH, Bugg CE Jr, Schafer JA, and Balkovetz DF. Polaris, a protein disrupted in orpk mutant mice, is required for assembly of renal cilium. Am J Physiol Renal Physiol 282: F541–F552, 2002.
Yoder BK, Hou X, and Guay-Woodford LM. The polycystic kidney disease proteins, polycystin-1, polycystin-2, polaris, and cystin, are co-localized in renal cilia. J Am Soc Nephrol 13: 2508–2516, 2002.
Zegarra-Moran O, Romeo G, and Galietta LJ. Regulation of transepithelial ion transport by two different purinoceptors in the apical membrane of canine kidney (MDCK) cells. Br J Pharmacol 114: 1052–1056, 1995.(H. A. Praetorius, J. Frki)