Am J Physiol Cell Physiol AJP: Endocrinology and Metabolism
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Cell Physiol 290: C952-C963, 2006. First published October 5, 2005; doi:10.1152/ajpcell.00339.2005
0363-6143/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
290/4/C952    most recent
00339.2005v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Olteanu, D.
Right arrow Articles by Schwiebert, E. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Olteanu, D.
Right arrow Articles by Schwiebert, E. M.

MEMBRANE TRANSPORTERS, ION CHANNELS, AND PUMPS

Heightened epithelial Na+ channel-mediated Na+ absorption in a murine polycystic kidney disease model epithelium lacking apical monocilia

Dragos Olteanu,1 Bradley K. Yoder,2 Wen Liu,5 Mandy J. Croyle,2 Elisabeth A. Welty,1 Kelley Rosborough,2 J. Michael Wyss,2 P. Darwin Bell,1,3 Lisa M. Guay-Woodford,2,3,4 Mark O. Bevensee,1 Lisa M. Satlin,5 and Erik M. Schwiebert1,2,3

Departments of 1Physiology and Biophysics and of 2Cell Biology and Divisions of 3Nephrology and Genetics and 4Translational Medicine, Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama; and 5Division of Nephrology, Department of Pediatrics and Medicine, Mount Sinai School of Medicine, New York, New York

Submitted 10 July 2005 ; accepted in final form 29 September 2005


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
The Tg737°rpk autosomal recessive polycystic kidney disease (ARPKD) mouse carries a hypomorphic mutation in the Tg737 gene. Because of the absence of its protein product Polaris, the nonmotile primary monocilium central to the luminal membrane of ductal epithelia, such as the cortical collecting duct (CCD) principal cell (PC), is malformed. Although the functions of the renal monocilium remain elusive, primary monocilia or flagella on neurons act as sensory organelles. Thus we hypothesized that the PC monocilium functions as a cellular sensor. To test this hypothesis, we assessed the contribution of Polaris and cilium structure and function to renal epithelial ion transport electrophysiology. Properties of Tg737°rpk mutant CCD PC clones were compared with clones genetically rescued with wild-type Tg737 cDNA. All cells were grown as polarized cell monolayers with similarly high transepithelial resistance on permeable filter supports. Three- to fourfold elevated transepithelial voltage (Vte) and short-circuit current (Isc) were measured in mutant orpk monolayers vs. rescued controls. Pharmacological and cell biological examination of this enhanced electrical end point in mutant monolayers revealed that epithelial Na+ channels (ENaCs) were upregulated. Amiloride, ENaC-selective amiloride analogs (benzamil and phenamil), and protease inhibitors (aprotinin and leupeptin) attenuated heightened Vte and Isc. Higher concentrations of additional amiloride analogs (ethylisopropylamiloride and dimethylamiloride) also revealed inhibition of Vte. Cell culture requirements and manipulations were also consistent with heightened ENaC expression and function. Together, these data suggest that ENaC expression and/or function are upregulated in the luminal membrane of mutant, cilium-deficient orpk CCD PC monolayers vs. cilium-competent controls. When the genetic lesion causes loss or malformation of the monocilium, ENaC-driven Na+ hyperabsorption may explain the rapid emergence of severe hypertension in a majority of patients with ARPKD.

cilia; hypertension; ion transport; epithelial cells

LITTLE IS KNOWN about the transport properties of collecting tubule-lining renal epithelial cells in patients with autosomal recessive polycystic kidney disease (ARPKD). In this study, we sought to investigate whether ion transport and cell signaling abnormalities exist in ARPKD. In a murine model of ARPKD using the Oak Ridge polycystic kidney (Tg737°rpk) mouse, ductal epithelia from multiple tissues lack a well-formed central monocilium in the apical membrane (52, 63). The central monocilium is a key defining morphological feature of the cortical collecting duct (CCD) principal cell (PC) (32). Malformation of this nonmotile cilium is due to the lack of a key protein, Polaris (encoded by the Tg737 gene). Polaris is normally located at the basal body and base of the cilium at the point of attachment to the luminal or apical membrane and along the cilia axoneme (63). Without functional Polaris serving as a key structural protein for the cilium, a well-formed central monocilium cannot be built.

The primary cilium is a relatively obscure organelle, and its function in the kidney remains largely unknown. In tissues from lower organisms, cilia found on neurons in Caenorhabditis elegans are important in sensory perception (22). The C. elegans genetic homolog for Tg737 is osm-5, a gene involved in C. elegans chemosensation within its flagella (22). In this light, recent data published by Praetorius and Spring (39, 40) suggest that the renal cilium may be a mechanosensitive organelle that initiates a Ca2+ spark and wave in Madin-Darby canine kidney (MDCK) epithelia caused by fluid flow or mechanical bending of the cilium. The cilium also modulates secretory K+ channels in MDCK cells (38). In addition to this mechanosensory role, it is possible that the cilium of the CCD PC might be involved in osmosensation as observed in olfactory cilia or flagella in C. elegans (3, 4, 22, 24). Moreover, original studies of the Tg737°rpk mouse revealed that the urine-to-plasma osmolarity ratio, as well as the urine specific gravity, was much lower in mutant orpk vs. genetically rescued or wild-type mice (60–62). A defect in the urine-concentrating mechanisms and/or hyperabsorption of a key osmole could underlie this difference and be a key contributor to cystic disease.

It is important to note that ARPKD and autosomal dominant polycystic kidney disease (ADPKD) differ greatly with regard to the type of tissue remodeling and its impact on transepithelial salt and water transport and vice versa (11, 18–21). In ADPKD, closed, fluid-filled cysts that are encapsulated by a single monolayer of cystic epithelial cells occur along multiple nephron segments. Cl– and fluid secretion into the cyst, as well as trapped autocrine and paracrine growth factor and autacoid signaling, then contribute detrimentally to cyst volume expansion and growth (11, 18, 19). In ARPKD, collecting duct (CD) segments dilate but never close off or encapsulate (20, 21). As such, unstirred layers of tubular fluid and turbulent flow are likely to exist within the CDs of the kidneys in ARPKD; however, the tubular fluid is never trapped. Ions, solutes, and water are still freely reabsorbed and secreted along diseased ARPKD CDs, and ARPKD nephron tubular fluid does eventually constitute the final urine. These dilated CDs are often referred to as pseudocysts (20, 21). However, this term can be misleading because ARPKD pseudocysts are merely dilated renal tubules rather than fully encapsulated, bona fide cysts observed routinely in ADPKD. The term "pseudocyst" also appears to refer more specifically to such structures in the gastrointestinal tract in general and the pancreas in particular (34).

Herein we report the electrophysiological analysis of mutant Tg737°rpk CCD PC cells with defects in the formation of the central cilium. We compared results obtained from mutant clones with those from clones that were genetically rescued by expression of the wild-type Tg737 cDNA, which was previously shown to correct the renal pathology in Tg737°rpk mutant mice in vivo (60–62). This panel of cells is an excellent tool with which to study the role of Polaris and the structural and functional integrity of the monocilium in ductal epithelial cell function in a well-polarized cell monolayer. Our results reveal a markedly elevated transepithelial voltage (Vte) and short-circuit current (Isc) in the mutant orpk monolayers, despite comparable transepithelial resistance (Rte). Although we first suspected enhanced Cl– secretion on the basis of our ADPKD studies (47), pharmacological and cell biological examination of this enhanced ion transport revealed that epithelial Na+ channels (ENaCs) were upregulated. Manipulation of defined cell culture conditions also suggested primary ENaC involvement. As such, we hypothesized that Polaris and the central monocilium normally act to limit apical ENaC-mediated, Na+-absorptive pathways. When this cilium-mediated inhibition is lost, Na+ transport becomes dysregulated and hyperabsorption results, producing an underlying primary etiology of early-onset severe hypertension in the majority of patients with ARPKD.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Generation of CCD PC clones from mutant and genetically rescued orpk mice by genetic cross with the ImmortoMouse. The details of tubule dissection, clonal selection, and characterization of the cilia defect in the original mutant (mutant 1) and genetically rescued CCD PC clones (rescued 1 and 2) were published in detail previously (63). The details of genetic rescue of the orpk mouse were also published previously (60–62). These studies led to the generation of clones that are excellent tools with which to study the contributions of Polaris and nonmotile central cilia to ductal epithelial cell function. The present work was designed to establish similar cell models in ductal epithelia from other tissues.

In this study, we isolated two additional mutant clones (mutant 2 and mutant 3). We developed these mutants using clonal dilution of a mixed mutant CCD cell population that was expanded from the 94D CCD individually dissected from the kidney of an animal that was generated by cross-breeding an orpkTg737 mouse with an H-2Kb-ts-A58 transgenic mouse (the ImmortoMouse). We also expanded and studied this original mixed mutant 94D CCD cell population grown in the PC-defined medium described above. In addition, we studied a third genetically rescued clone from a separate, individually dissected CCD, 94E, referred to hereinafter as the genetically rescued clone rescued B2. Mutant 1, rescued 1, rescued 2, and rescued B2 were grown under G418 selection, in which mutant 1 was transfected with pcDNA3.1 empty vector without Tg737 cDNA because the three rescued clones stably express the Tg737 cDNA. The mixed mutant 94D CCD cell population, mutant 2, and mutant 3 are not selected in G418.

Cell culture and seeding of filter supports for transepithelial electrophysiology. Mixed mutant CCD cells and mutant and genetically rescued CD clones were grown initially under nonpolarized conditions and at a permissive temperature of 33°C in collagen-coated tissue culture flasks. The medium used for this initial culture was a defined CD medium. Its recipe included DMEM-Ham's F-12 medium supplemented with 10% FBS, 10 U/ml IFN-{gamma}, 1.3 µg/l sodium selenite, 1.3 µg/l triiodothyronine, 5 mg/l insulin, 5 mg/l transferrin, 5 µM dexamethasone, 2.5 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin in a humidified 5% CO2 incubator. These conditions are known to upregulate ENaC expression. No EGF was present at any time during culture, because EGF is known to inhibit ENaC (15, 49) and stimulate an aberrantly hyperactive EGF receptor signaling cascade in polycystic kidney disease cells (41, 54). For this and other reasons, EGF was excluded from the defined and supplemented medium. G418 (200 µg/ml) was present in the medium throughout 33°C culture to select continually the clones bearing pcDNA3.1 vector without or with the Tg737 cDNA cassette. G418 was left out of the medium used for the mixed uncloned CCD cells and the clones mutant 2 and mutant 3. Cells were then lifted in minimal trypsin-EDTA solution and seeded onto 6.5-mm-diameter Costar Transwell filter supports (0.45-µm-diameter pore size polycarbonate filters) coated with diluted Vitrogen 100 solution (1:15 dilution with Ca2+- and Mg2+-free PBS) at a density more than adequate to cover the entire filter at day 0 (i.e., the seeding day). Cells were then grown in the same defined CD medium at 39°C, but without IFN-{gamma}. All cells were grown at similar passage numbers and were passaged a minimum of 10 times without any change in morphological or functional phenotypes. All monolayers were fed 1 day before the experiment.

Transepithelial electrophysiology. Open-circuit measurements of Rte and Vte using a Millipore Voltohmmeter were obtained on day 2 after seeding with Ag+-AgCl chopstick electrodes. Vte readings were stable in these mutant and rescued model renal epithelia, because the Rte values achieved using these clones grown as monolayers were high (near or ≥20,000 {Omega}·cm2 after 4–6 days on filters). All such measurements were performed while the cells were in the defined CD medium. All monolayers were fed 1 day before the experiment. When Rte and Vte reached stable values (between days 4 and 6), open-circuit measurements were performed in CD medium in unstirred conditions before and at defined time points after application of inhibitors or agonists.

Isc analysis in Ussing chambers. Recordings of Isc were performed as described previously (47) using a homemade system designed to accommodate 6.5-mm-diameter Transwell filter supports. All monolayers were fed 1 day before the experiment. Because the defined CD medium was found to be critical to maintaining at least part of the large Isc and Vte in mutant and rescued monolayers, these recordings were also performed in the CD medium bubbled with 5% CO2 in a circulating system warmed to 37°C. The only difference between these recordings and open-circuit readings was that the medium was devoid of FBS in the Ussing chamber. A voltage pulse was injected periodically by the amplifier during most recordings to gauge resistance (see RESULTS), and inhibitors were added only after a stable baseline was achieved.

Data analysis and statistics. Isc data were converted to microamperes per square centimeter of filter support surface area by multiplying that area (0.33 cm2) by 3. Data are expressed as means ± SE. Differences before and after application of a drug were assessed using a paired Student's t-test, with P < 0.05 used as the significance level. Differences between mutant and rescued or wild-type monolayers were assessed using ANOVA, with P < 0.05 used as the significance level.

Materials. Hormone supplements for the defined CD medium, amiloride and amiloride analogs, and protease and protease inhibitors were obtained from Sigma. Vitrogen 100 was obtained from Cohesion, and filter supports were purchased from Costar through Fisher Scientific.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Vte is elevated in mutant orpk CCD PC monolayers vs. genetically rescued CCD PC monolayers, despite similar Rte. Rte was measured on day 2 for each monolayer after seeding at confluence on day 0 to allow the cells to attach to the filter support and to initiate monolayer formation. After day 2, Rte was assessed frequently until a plateau was measured for both Rte and Vte. This plateau was reached routinely on day 5 and remained stable on days 6 and 7. In general, and despite equivalent Rte values, mutant monolayers demonstrated three- to sixfold augmented Vte vs. rescued monolayers. The overall summary of all Vte measurements for these mutant and genetically rescued CCD PC models is provided in Fig. 1. Together, and despite equivalent Rte for both mutant and genetically rescued monolayers that approached or exceeded 20 k{Omega}·cm2, mutant orpk CCD PC monolayers had profoundly upregulated Vte as a result of heightened activity in one or more ionic conductances.


Figure 1
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 1. Transepithelial voltage (Vte) is enhanced markedly in a mutant Oak Ridge polycystic kidney (orpk) mouse carrying hypomorphic mutation in the Tg737 gene (Tg737°rpk) in which collecting duct (CD) principal cell (PC) clones were grown as well-polarized monolayers. Data summary of all measurements of Vte measured using Millipore Voltohmmeter chopstick electrodes is shown. Data derived from mutant clones grown as monolayers are shown in open bars. Data from genetically rescued clones are shown in filled bars. Measurements were performed on day 5 or 6 after seeding of cells on filter supports at day 0. Transepithelial resistance (Rte) of these monolayers was routinely 18–20 k{Omega}·cm2 and often was >20 k{Omega}·cm2, which is the highest value the Millipore Voltohmmeter could measure. n = 24 experiments for each clone of mixed mutant monolayers, and n = 48–72 for all other clones; *P < 0.05 (ANOVA or unpaired Student's t-test as appropriate).

 
The large Rte achieved by these clones grown as monolayers, as well as the upregulated Vte in the mutant orpk cell monolayers vs. counterpart control monolayers, necessitated verification and constant and reproducible comparison between mutant and genetically rescued clones. As such, we performed open-circuit measurements of Rte and Vte in mutant (M) vs. rescued (R) monolayers using the Millipore open-circuit assay (Fig. 2). The first data set showed the development of Vte as the monolayers matured from day 4 to day 7 to day 10 (Fig. 2). Subsequent data sets were measured on day 5 and/or day 6, when Vte and Rte were most stable. In each data set, mutant and rescued cell monolayers were compared routinely. In all data sets and for at least 10 passages, mutant monolayers lacking a well-formed apical central cilium (defined as cilium deficient) demonstrated augmented Vte vs. genetically rescued monolayers (controls defined as cilium competent). Figure 2 also demonstrates the difference in morphological phenotype of the central monocilium for mutant and genetically rescued PC clones. Examples of this phenotype were published previously (63); however, these additional, more recently gathered data shown in better-quality images are provided for fuller presentation. It should be noted that cilia length in the rescued monolayers was 2–3 µm, whereas the malformed cilium in mutant monolayers was <0.5 µm in length. It also should be noted that these data represent maximal monocilium length in mature monolayers. In contrast, MDCK cell monolayers express monocilia that grow longer and longer as the monolayers age (39). This was not the case in our mouse PCs derived from individually dissected CCDs. Together, these data show a reproducible and upregulated ionic conductance in the apical membrane of mutant orpk mixed CCD cells and in CCD PC clones grown as monolayers and deficient in an apical central monocilium.


Figure 2
View larger version (51K):
[in this window]
[in a new window]
  		Fig. 2. Heightened Vte is a consistent and reproducible phenotype in mutant Tg737°rpk CD PC clones grown as well-polarized monolayers and studied in parallel with counterpart controls. Data shown in this figure are based on the subsets of mutant and genetically rescued clones measured and compared over time in each data set. Multiple blinded investigators assessed these monolayers unprompted regarding the monolayers' identity. n = 6–12 experiments for each data set; *P < 0.05 (ANOVA or unpaired Student's t-test as appropriate). Rte of these monolayers was routinely 18–20 k{Omega}·cm2 and often was >20 k{Omega}·cm2, which is the highest value that the Millipore Voltohmmeter can measure. This upregulated Vte phenotype [and also short-circuit current (Isc) phenotype; see below] remained consistent for at least 10 passages after clonal selection and cryopreservation. Inset, more recent analysis of monocilium morphology in the mutant 1 clone vs. rescued 2 clone. Malformed cilium is shown in the mutant monolayers (94D pcDNA3.1), whereas a well-formed cilium is shown in the rescued 2 monolayers (94D Tg737-2). This newer data set was obtained with cilium-specific tubulin antibodies to echo data already published regarding these cell models and to show the morphological difference with specificity to apical central monocilia in these PC clones. However, these data have been published already (63), and we did not want to recount them here. Inset is intended only as a reminder of the morphological difference. Mutant 1 clone was compared with the rescued 2 clone below in additional electrical experiments. **P < 0.01 by ANOVA.

 
Elevated Vte and Isc in mutant monolayers is inhibited in a dose-dependent manner by amiloride, an inhibitor of Na+ transport. On the basis of our experience with ADPKD epithelial cell models (47), we first tested the hypothesis that secretory Cl– current was upregulated, causing the enhanced Vte and Isc observed in mutant orpk monolayers. Experiments with apical Cl– channel blockers (DIDS, diphenylamine-2-carboxylate, and glibenclamide) as well as basolateral Na+-K+-2Cl– cotransporters (bumetanide) were without effect (data not shown). This is not surprising, because CCD PCs are not thought to be capable of much or any Cl– secretion under basal conditions (29, 32, 33, 45).

Therefore, because PCs are primarily Na+-absorbing, K+-secreting cells within the CCD under normal conditions, we postulated that Na+ absorption across the apical membrane of mutant orpk ARPKD PC monolayers might be upregulated. Figure 3 shows the sensitivity of Vte in mutant and rescued cell clones grown as monolayers to apical application of 50 µM amiloride. Both the greatly enhanced Vte in mixed mutant CCD PC cells from the orpk mouse and mutant orpk PC clones and the reduced but significant Vte in the genetically rescued clones were highly sensitive to amiloride. Apical application of amiloride to mutant and rescued cells inhibited the Vte virtually completely within 5 min (Fig. 3). Together, these data further confirm that these cell models derived originally from individually dissected CCD segments are Na+-absorbing PCs and that a Na+-absorptive transport mechanism appeared to be upregulated in mutant vs. rescued orpk PC monolayers.


Figure 3
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 3. Heightened Vte in mutant Tg737°rpk CD PC monolayers and measurable voltage in genetically rescued Tg737°rpk monolayers was inhibited in a dose-dependent manner by amiloride. The effect of 50 µM amiloride on Vte was examined in both panels of mutant and genetically rescued clones grown as monolayers. All epithelial cell models are Na+-absorbing PCs. This maximal dose of amiloride virtually abolished Vte in all mutant and rescued clones. The mutant monolayers had 3- to 6-fold upregulated Vte. n = 12 experiments for each set of data. Number of experiments for summary data regarding Isc is shown at center. *P < 0.005 by ANOVA.

 
To examine this Na+ transport mechanism more deeply and to confirm that it was electrogenic, we performed electrophysiological analysis of Isc in mutant vs. rescued orpk PC monolayers mounted in Ussing chambers. Despite equivalent Rte (as indicated by the voltage pulses causing brief current deflections at fixed times on the raw traces), amiloride-sensitive Isc was upregulated approximately fourfold in mutant monolayers vs. rescued monolayers (Fig. 4). Basal Isc in both mutant and rescued monolayers was inhibited markedly by 10 µM amiloride (Fig. 4). In correlative Vte measurements, the IC50 for amiloride was ~1 µM, whereas maximal inhibition was achieved at 10 µM or greater (Fig. 4). These initial data suggested that ENaCs may be upregulated in this murine model of ARPKD; however, the small residual Vte and Isc resistant to amiloride suggest that other Na+ transport pathways might be present.


Figure 4
View larger version (20K):
[in this window]
[in a new window]
  		Fig. 4. Heightened transepithelial Isc in mutant Tg737°rpk CD PC monolayers and measurable current in genetically rescued Tg737°rpk monolayers is inhibited in a dose-dependent manner by amiloride. Original Isc traces are shown at left. Inhibition with 10 µM amiloride is shown for mutant and rescued monolayers in these raw traces. Note the significantly larger amount of Isc in the mutant monolayer despite equivalent resistance. The spikes in the trace reflect the current deflections generated by a voltage pulse injected by the amplifier to monitor resistance. Ten micromolar amiloride was used to fully block the Na+ currents during Isc recordings. At center, summary Isc data are provided as measured before and after addition of amiloride. Isc is expressed as µA·cm2. Number of experiments from which summary data for Isc were derived is shown at center. At right, open-circuit Vte data summary with full dose-response assessment with amiloride is shown. n = 6–12 experiments for each clone in each data set. IC50 for amiloride inhibition of Vte was determined to be ~1 µM. *P < 0.05 for Isc data shown at center; *P < 0.01 or lower for open-circuit data shown at right. n = 12–24 experiments is shown for Isc data for amiloride dose-response data set of Vte. Paired Student's t-test was used to assess amiloride inhibition.

 
Cell culture requirements are consistent with ENaC expression and function. Analysis of the initial cell culture findings and responses to assay modifications also provided insight into the identity of the Na+ transport mechanisms responsible for this pathophysiological phenotype. Vte was reduced to 0 within 24–48 h in mutant and rescued monolayers when the monolayers were moved from their defined CD medium into a standard FBS-containing medium without hormone supplements. Figure 5 shows these data regarding the mutant 1 clone grown to well-polarized monolayers because of the upregulated Vte signal. The upregulated Vte also was lost transiently (during a period of 5–12 h) when the apical medium was aspirated and replaced (Fig. 5). In contrast, replacement of the basolateral medium fueled Vte within 30 min (Fig. 5). The difference in electrophysiological phenotype was most dramatic in the unstirred, open-circuit measurements of Vte. Curiously, in standard Ringer solution in Ussing chambers, upregulated Vte and Isc were lost completely during the 15-min lag during setup and stabilization of the Vte and Isc under clamped conditions. In fact, Isc decayed over time while we waited for the basal Isc to stabilize in these renal cell monolayers. This loss in Isc (vs. calculated Isc from open-circuit measurements) was partially prevented by performing Ussing chamber experiments in defined CD medium without FBS or in an OptiMEM-1 serum-free medium or Ringer solution supplemented with insulin (at the same concentration used for the defined culture medium) (Fig. 5). Importantly, Ringer solution supplemented with insulin that lacked Na+ (replaced by NMDG) could not support Isc (Fig. 5). Together, these data suggest that amiloride-sensitive Vte and Isc are sensitive to cell culture conditions and are mediated by electrogenic Na+ transport mechanisms, likely ENaCs, in a polarized cell monolayer system.


Figure 5
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 5. Cell culture conditions and dynamics of electrical measurements affect Vte and Isc in mutant and rescued Tg737°rpk CD PC monolayers profoundly. Only data from mutant monolayers are shown, because the electrical signal was larger; however, these maneuvers also affected voltage and current in rescued monolayers to a similar degree (data not shown). A: removal of the defined PC medium with 5% FBS and hormone supplements into another basal medium containing 5% FBS but without hormones caused a complete loss of amiloride-sensitive Vte. B: feeding the apical side of the monolayers with new defined PC medium caused a transient loss in amiloride-sensitive Vte that returned after 7 h. C: feeding the basolateral side of the monolayers with fresh PC medium potentiated the Vte. D: transfer of the monolayers from open-circuit conditions in which the Vte and Rte were measured (and Isc was calculated) into the circulating Ussing chamber led to complete loss of amiloride-sensitive Isc in Ringer solution of various formulations. This loss was partially prevented by performing the Ussing chamber Isc recordings in the defined PC medium (without FBS) or in OptiMEM-I medium or Ringer solution with insulin. Ringer solution that was devoid of Na+ but also contained supplemental insulin failed to support Isc. ANOVA was used to compare different experimental conditions as appropriate in A and D. A paired Student's t-test was used to compare data in B and C. n = 12–24 experiments for each data set; *P < 0.05.

 
These data also show that hormonal supplements (dexamethasone and insulin in particular) support overall ENaC expression and function in renal epithelia. However, these data also suggest that the renal epithelium itself secretes stimulatory factors that support ENaC activity acutely and independently of the defined culture medium. One possible candidate for this acute stimulation of ENaC is a secreted protease that cleaves the extracellular domain of ENaCs to keep ENaC open. A protease or another factor may be removed by changing the medium or may be washed away and/or diluted when exposed to the circulating conditions of the Ussing chamber. It should be noted emphatically that these properties do not affect the overall difference in activity observed in cilium-deficient vs. cilium-rescued monolayers. Rather, these observations and factors merely support the idea that ENaC activity is being measured in these studies and that the data are consistent with the hormone requirements, properties, and regulation of ENaC described in the literature (6, 28). These data led us to adopt a primary focus on ENaC.

ENaC-selective amiloride analogs inhibit the elevated Vte and Isc in mutant monolayers. Because amiloride at higher doses can inhibit multiple Na+ transporters as well as multiple Na+ and cation channels that might be involved in Na+ absorption, we tested ENaC-selective amiloride analogs. Further proof of ENaC involvement in this phenotype was provided by inhibition of the majority of the upregulated Vte and Isc with apical benzamil (Fig. 6). Phenamil was also effective at nanomolar concentrations, with an IC50 of ~200 nM (data not shown). These amiloride analogs are more selective for ENaC than other Na+ channels or transporters. The IC50 for benzamil was ~50 nM (Fig. 6), consistent with a more potent effect of benzamil than of amiloride known to exist for ENaCs (6, 28). Together, these data led us to focus on ENaC as a mediator of electrogenic Na+ hyperabsorption in mutant models of ARPKD grown as well-polarized monolayers and lacking apical central cilia.


Figure 6
View larger version (20K):
[in this window]
[in a new window]
  		Fig. 6. Epithelial Na+ channels (ENaC)-selective amiloride analog benzamil inhibits Vte and Isc markedly but not completely in Tg737°rpk CD PC monolayers. Original Isc traces are shown at left, summary Isc data are shown at center, and open-circuit Vte data summary for mutant monolayers assessing benzamil IC50 is shown at right. Data presentation here is similar to that shown in Fig. 4. Amiloride inhibits more than just Na+ channels involved in Na+ absorption. As such, we tested an expanded panel of amiloride analogs. Benzamil at 10 µM inhibited the hyperactive absorptive Na+ current in the mutant monolayers as well as the low level of current in the rescued monolayer. As in Fig. 3 with regard to amiloride, 10 µM benzamil was the concentration used to fully block Na+ current in Isc recordings. Number of experiments for Isc summary data is shown at center. At right, IC50 for benzamil inhibition of the Vte was determined to be ~50 nM. *P < 0.05 Isc data at center vs. open-circuit data at right. N is shown for Isc data; n = 12–24 experiments for the Vte data. A paired Student's t-test was used to assess benzamil inhibition.

 
As internal controls for this amiloride analog pharmacology, we also tested ethylisopropylamiloride (EIPA) and dimethylamiloride (DMA), analogs that are less selective for ENaCs and more selective for Na+/H+ exchangers (NHEs) at low micromolar concentrations (37). To our initial surprise, EIPA inhibited Vte with an IC50 of ~2 µM (Fig. 7). DMA inhibited Vte as well with an IC50 of ~20 µM (Fig. 7). It should be noted, however, that apical application of ouabain (100 µM) was without effect, ruling out a role for mislocalized Na+-K+-ATPase pumps in Na+ hyperabsorption (data not shown). Although these data suggest a possible parallel role for NHEs in Na+ absorption, the activity of such exchangers should not affect Vte, owing to their universally shared 1:1 stoichiometry of cation exchange (37). Moreover, further examination of the ENaC literature showed a rank order potency of benzamil > amiloride > DMA for ENaC was reported previously (12). However, in our studies, we also evaluated the effects of EIPA, which is a key highly selective amiloride analog at lower doses for NHE vs. ENaC. Although the rank-order potency in our study was similar to that described above, it also was more comprehensive (benzamil > amiloride ≥ EIPA > DMA). Taken together, we conclude that ENaC upregulation likely underlies the upregulated Vte and Isc observed in mutant cilium-deficient cells. Nevertheless, we cannot rule out that parallel and upregulated NHE activity may stimulate ENaCs through effects on external apical pH (27, 58), intracellular pH (13), or both.


Figure 7
View larger version (20K):
[in this window]
[in a new window]
  		Fig. 7. Additional amiloride analogs ethylisopropylamiloride (EIPA) and dimethylamiloride (DMA) also inhibit upregulated Vte in mutant monolayers. Additive inhibition with different pairs of analogs is shown. A: dose-response relationship for EIPA to estimate IC50 for inhibition of upregulated Vte in mutant monolayers. EIPA IC50 was ~2 µM. B: similar dose-response experiment with the more Na+/H+ exchanger (NHE)-selective analog DMA. DMA IC50 was ~20 µM. C and D: use of IC50 doses for the 4 amiloride analogs used in this study to document additive inhibition of upregulated Vte. These data provide evidence for upregulated ENaC-mediated Na+ hyperabsorption in cilium-deficient Tg737°rpk CD PC monolayers, although we cannot rule out that parallel NHE activity might modulate ENaC function in this epithelial model. *P < 0.05 by ANOVA.

 
Proteases influence upregulate ENaC-mediated Isc or Vte in the mutant monolayers. In light of a primary role for ENaCs in upregulated amiloride-sensitive Vte and Isc in cilium-deficient PC cells, we tested whether endogenous secreted or membrane-bound proteases were active in the regulation of ENaCs in this model epithelium. Such proteases are thought to recognize Kunitz-like domains in the cysteine-rich regions of the extracellular domain of ENaCs and to activate ENaCs in epithelia specifically through direct protease-dependent cleavage (8, 10, 25, 51, 53, 56). We also used protease inhibitors as another diagnostic cell biological tool to assess ENaCs in this dysregulated ion transport phenotype in the mutant monolayers. Aprotinin is a protease inhibitor used widely in ENaC studies (8, 10, 25, 51, 53, 56). Although aprotinin lowered the current only slightly in the rescued monolayers, it had a significant inhibitory effect on Ussing chamber Isc in mutant monolayers, which was further inhibited by benzamil (Fig. 8). If left on the apical side of the mutant monolayers longer in open-circuit experiments (10 min vs. 5 min in the Ussing chamber), the inhibitory effect of aprotinin was more robust (Fig. 8). Vte was virtually abolished in the mutant Tg737°rpk monolayers by 1 and 10 µg/ml aprotinin after 30 min (Fig. 8). The inhibitory effect was diluted by reducing the protease inhibitor concentration to nanograms per milliliter levels. Leupeptin, another broad-specificity serine protease inhibitor, also inhibited the Vte driven by ENaCs (data not shown). Protease inhibitors attenuate ENaC activity in the mutant orpk cell monolayers. We speculate that secreted proteases may be responsible for keeping ENaCs active at the membrane. They may be washed away during a medium change, upon transfer from unstirred to stirred conditions, or upon exposure of the monolayer to a 20-fold larger volume of medium bathing its apical surface in the Ussing chamber (10 ml) vs. in an apical cup of a filter support (0.5 ml). Other locally released factors also may be necessary to gate or to maintain ENaC activity and cannot be discounted. Nevertheless, these data provide another piece of diagnostic evidence implicating ENaCs in this enhanced Isc and Vte phenotype that is in agreement with similar work performed by multiple laboratories in human primary and immortalized airway epithelial cell monolayers (8, 51, 53).


Figure 8
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 8. Protease inhibitors attenuate Vte and Isc in mutant Tg737°rpk CD PC monolayers, suggesting ENaC hyperactivity. Endogenous secreted or membrane-bound proteases are thought to recognize Kunitz-like domains in cysteine-rich regions of extracellular domain of ENaC and activate ENaCs specifically. As such, aprotinin, a protease inhibitor used widely in these ENaC protease studies, was also tested for its inhibitory effect. Although aprotinin lowered the current in the rescued monolayer, the protease inhibitor had a significant inhibitory effect on Ussing chamber Isc in mutant monolayers that was inhibited further by benzamil. Number of experiments is shown for Isc summary data at center. When left on the apical side of the monolayer for longer periods in open-circuit experiments (15–30 min vs. 5 min in Ussing chamber), the inhibitory effect of aprotinin was more robust. n = 6–12 experiments for each clone in each data set. *P < 0.05 (paired Student's t-test as appropriate).

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
We hypothesized that ion transport across the apical membrane of PCs lacking well-formed apical monocilia might be abnormal on the basis of evidence implicating nonmotile cilia and flagella as sensory organelles in neurons (3, 4, 22, 24). Although the apical central monocilium might be a flow sensor in renal tubules with higher flow rates, such as the proximal tubule, its role in the CD may be different in a region of lower tubular fluid flow. We thus speculated that the apical central monocilium could be an osmosensor or chemosensor for the CCD PCs and, if lost, that it might affect ductal epithelial ion transport in the renal CD and in tubules and ducts from extrarenal tissues.

On the basis of the aforementioned data, we have formed working hypotheses concerning the cellular and molecular mechanisms of ENaC upregulation in the face of the loss of a fully formed apical central monocilium. These postulates are most germane to this mouse model of ARPKD. However, they also are in accord with findings in a human ARPKD cell model (42) and with early-onset hypertension in the majority of patients with ARPKD (20). First, ENaC proteins might be localized normally on or near the monocilium in a limited and highly localized pattern. When the cilium is lost, the PC may overexpress ENaC in the apical membrane in a compensatory manner. Data published by Rohatgi et al. (42) on the basis of human ARPKD cell models documented a modest yet significant upregulation in {alpha}-ENaC mRNA and protein. Second, we hypothesized that an inhibitory signal that normally originates from the monocilium (e.g., Ca2+ spark and wave, phospholipids, cilium-specific protein kinase) tonically inhibits ENaC-mediated Na+ absorption. Autocrine purinergic signaling could drive one or more of these inhibitory signals as could other ligands, such as growth factors and hormones whose receptors localize to the luminal membrane of ductal epithelia. This inhibitory signal is lost when the monocilium is lost, leading to the upregulation of ENaC activity. Early data suggest that ATP release is impaired in mutant, cilium-deficient cell monolayers vs. cilium-competent controls (Fintha A, Hanson EL, Olteanu D, Bell PD, and Schwiebert EM, unpublished observations). Third, we postulated that the loss of the apical central monocilium causes a redistribution of ENaC from a primarily intracellular location to the plasma membrane. An effect on the microfilament- or microtubule-based cytoskeleton due to the absence of the monocilium could affect ENaC protein distribution in the apical pole of the polarized CCD PCs (50, 64). ENaC localization and/or function are also affected by actin (50, 64). Fourth, our amiloride analog pharmacological data do not discount the possibility of parallel upregulation of NHEs. We speculate that the loss of a well-formed monocilium might cause upregulation of NHEs and either acidification of the tubular fluid bathing the apical membrane or alkalinization of the cytosol in a cilium-deficient cell and thus might potentiate ENaC activity (13, 27, 58). Early data suggest that NHE activity may also be present and/or upregulated in both the apical and basolateral membranes in vitro and in vivo in orpk cilium-deficient cell monolayers and isolated perfused tubules, respectively (Olteanu D, Liu W, Bevensee M, Satlin LM, Yoder BK, and Schwiebert EM, unpublished observations). We are keeping an open mind regarding all possible models described above and are cognizant that a combination of our working hypotheses and postulates may actually contribute to the observed Na+ hyperabsorption phenotype.

In light of the fourth postulate described in the immediately preceding paragraph, extracellular H+ concentration ([H+]e) is known to affect brain Na+ channels, acid-sensing ion channels, and the topologically related P2X purinergic receptor channels (1, 5, 35, 46, 48). Human {delta}-ENaC is stimulated by extracellular protons (acidic pH, EC50 5.0) and has been shown to be expressed in human kidney and elsewhere (58). This effect on {delta}-ENaC, when coexpressed with beta-ENaC and {gamma}-ENaC in Xenopus oocytes, was also confirmed by Ji and Benos (27) and is conferred by degenerin domains within {delta}-ENaC. A mouse ortholog has not been found but is likely to emerge in the near term. {delta}-ENaC, as part of a novel or additional heteromultimeric ENaC in renal CD, may be critical in this disease paradigm and may sense and be upregulated by the NHE-dependent acidification of the tubular fluid. To our knowledge, the effect of extracellular pH on the other ENaC subunits has not been addressed. Alternatively, profound intracellular alkalinization due to upregulated NHE activity in the apical and basolateral membranes of PC monolayers may potentiate ENaCs. Chalfant et al. (13) showed, in Xenopus oocytes and planar lipid bilayers, that intracellular alkalinization stimulated various ENaC heteromultimers (due specifically to the presence of {alpha}-ENaC) and that intracellular acidification did the opposite. We are currently pursuing these early observations regarding NHE function and its effects on ENaC activity in our native mouse CCD PC model epithelia. We also cannot discount the possible upregulation of additional types of cation channels that are sensitive to amiloride, such as cyclic nucleotide-gated nonselective cation channels expressed in CD epithelial cell models (30, 31, 42).

It is important to draw key distinctions between ADPKD and ARPKD. We and others have proposed that ion, solute, and water secretion into encapsulated ADPKD cysts contributes and is detrimental to the expansion of cyst volume and size over time (11, 18, 19, 47). Particular attention has been paid to Cl– secretion via CFTR and other Cl– channels (11, 18, 19, 47). Curiously, cystic epithelia in both forms of the disease, ARPKD and ADPKD, either never fully differentiate or revert to an undifferentiated phenotype. In this sense, cystic cells from ARPKD and ADPKD share this trait. As such, any and all ENaC subunits may be expressed in the mutant monolayers in a way that may differ from controls. It is likely, however, that ENaC subunits are exclusively apical in both mutant and control monolayers. This fact is universally accepted in all absorptive epithelia (6). The opposite is true in highly secretory epithelia such as the choroid plexus of the brain and ciliary process of the eye (see below; see also Ref. 9). However, their relative abundance and/or function may be upregulated at the mRNA, protein, and/or cell surface level. We have designed RT-PCR primers and gathered antibodies to all relevant ENaC subunits to perform a thorough analysis of expression and localization of these Na+ transport proteins. That study is being conducted separately, is beyond the scope of the present study, and will require collaboration with multiple laboratories.

Despite this similarity in undifferentiated cell phenotypes, it is important to restate that ARPKD and ADPKD are different diseases in terms of how the architecture of the kidney is remodeled (11, 18–21). The type of remodeling has profound implications for salt and water movement and the role of a kidney affected by polycystic kidney disease with regard to salt and water balance. Encapsulated cysts arise mainly in ADPKD, in which such drastic remodeling occurs throughout the nephron. Tubules along the entire nephron fully pinch off and cause the formation of fluid-filled cysts surrounded by single-cell monolayers of cystic epithelial cells. Tubules dilate profoundly only in ARPKD. This phenomenon occurs mainly along the CD and more rarely in the proximal tubule. Importantly, ARPKD dilated tubules do not pinch off to form encapsulated cysts but merely create an unstirred environment within the renal parenchyma. Thus, on the basis of our results, Na+ is likely hyperabsorbed in ARPKD. Interestingly, early characterization of the orpk mouse revealed a low urine-to-plasma osmolality ratio (60–62). This finding may be due to a lack of ability to concentrate the urine and/or to hyperabsorption of a key osmole such as Na+, or both. Similar data have been published showing hyperabsorption of 22Na+ in polarized monolayers of human ARPKD cells vs. normal age-matched controls (42). Data from our laboratory also suggest that Na+-absorptive mechanisms, in addition to ENaC, may contribute to Na+ hyperabsorption, owing to a high IC50 for amiloride in these studies (42). Tubular fluid collected from patients with ARPKD with dilated ducts that resembled pseudocysts and were destined to be part of the final urine at the time of nephrectomy before or at the time of transplant showed extremely low Na+ concentration ([Na+]), suggesting avid renal Na+ absorption (43). These ducts demonstrated low [Na+] (2–7 mM vs. average <5 mM) and K+ concentration ([K+]) greater than that observed in plasma (8–9 mM). An important final point, however, is that hypertension in ARPKD is likely caused by a lesion that creates a primary or secondary defect in a Na+ transport mechanism. This defect would be a gain-of-function lesion, and we hypothesize that this is likely the case with ENaC upregulation in the renal CD in ARPKD.

Although ENaC hyperactivity and/or upregulation is consistent with human ARPKD and the early onset of hypertension (20), hypertension is also present in ADPKD well before the onset of renal disease and insufficiency. In clinical studies of ADPKD, hypertension is thought to reflect the distortion and/or destruction of the normal renal parenchyma by the progressive expansion of cyst size, which activates the renin-angiotensin system. However, is this the sole cause? In ADPKD cysts, Na+ may not be excreted, because it fails to enter the urinary bladder. This effect would be only a minor contributing factor, because cysts arise from only 5 to 10% of nephron segments throughout the kidney. As such, could Na+ hyperabsorption also occur in ADPKD and contribute to hypertension that arises before presentation of ADPKD kidney disease? Hypertension tends to arise before significant kidney and extrarenal tissue dysfunction in ADPKD.

In this light, importantly, the [Na+] in cyst fluids in ADPKD has been documented. Interestingly, [Na+] in cyst fluids from the kidney and elsewhere in ADPKD is heterogeneous. Gardner and Grantham (17) found a range of [Na+] levels from 3.1 to 150 meq/l in 12 cyst fluids. They stated that the cyst fluids that had lower [Na+] were likely of distal origin because of this and other attributes (17). Proximal cysts likely had isotonic [Na+] with respect to plasma or interstitium because of isotonic reabsorption that occurs in proximal tubule segments (17). Eckardt et al. (16) measured [Na+] when assessing erythropoietin production and expression in cysts. Heterogeneity was similar to that found in the study mentioned above, and the researchers concluded that cysts of distal nephron origin had low [Na+] of 20–40 mM (some <20 mM), whereas cysts from proximal segments had high [Na+] isotonic to plasma or interstitium (16). A third study, by Hurley et al. (26), compared cyst fluid content in kidney cysts in ADPKD vs. breast cysts in another disease. There were kidney and breast cysts with low and high [Na+] content. Interestingly, cysts with low amounts of [Na+] and other ions had greater concentrations of amino acids. Cysts with high [Na+] and other ions had low amino acid concentrations. Both low- and high-[Na+] cysts in kidney and breast were isotonic to plasma or interstitium because of amino acids. It is difficult to know that cysts of distal nephron origin have low [Na+] because of hyperabsorption or because the distal tubule and collecting tubular fluid have low [Na+] to begin with, because most of it is reabsorbed from the glomerular filtrate by this point in the nephron. However, these data and the preemergence of hypertension before renal decline in ADPKD suggest that Na+ transport pathways should also be assessed in ADPKD cystic epithelia vs. controls.

Significant evidence exists for upregulation of ENaC not only in human CD epithelia and mouse ARPKD model epithelia but also clinical and mouse model evidence in other tissues besides the kidney that has hints of Na+ transport dysregulation defects in ARPKD. In a significant percentage of patients with ARPKD from multiple cohorts, profound pulmonary hypoplasia has been noted that causes respiratory insufficiency at birth and/or chronic lung disease that leads to significant mortality (20). In salt and water transport physiology, Na+ absorption is absent during fetal lung development. Rather, secretion of an acidic and Cl–-rich fluid is solely active. This allows for fetal lung fluid to fill the developing lung and airways and allow optimal growth factor function (7). Any Na+ absorption, regardless of whether it is hyperactive during this time in fetal lung development, could hamper normal development and lead to hypoplasia or incomplete branching of the airways and lung (7). There are also reports of low amniotic fluid in humans during pregnancy (14). Low fluid volume in the amniotic sac may suggest Na+ and water hyperabsorption. Similarly, Na+ hyperabsorption in gastrointestinal (GI) tissues such as pancreas and liver may hamper optimal secretion of anions such as HCO3– and Cl– that water follows (2, 36, 44). Such an impairment of secretion due to the counteractive movement of Na+ absorption (which is normally less prominent in these two particular GI tissues) could be detrimental. Finally, the choroid plexus is a highly secretory organ in which Cl– and Na+ are secreted in parallel to fuel cerebrospinal fluid (CSF) secretion at high rates (9). In this epithelium, ENaC is on the basolateral membrane as the entry step for Na+ secretion, whereas the Na+-K+-ATPase pump actively secretes Na+ into the ventricles of the brain to help form CSF. Na+ hypersecretion, with parallel movement of excessive Cl– and water, could explain, at least in part, hydrocephalus in the orpk animal (9).

Having stated all of the above, we do not intend to imply that Na+ hyperabsorption is a universal phenotype in all mouse models of ARPKD. It is also not universal in the human condition. These and other polycystic kidney disease mouse models have been generated by deletional, insertional, and chemical mutagenesis. The structural ciliary phenotype is dramatic in orpk mice; however, it is less dramatic or not present in other ARPKD mouse models such as BALB/c-bpk/bpk (bpk) and C57BL-6Jcpk/cpk (cpk) mice. It should be noted that Veizis et al. (55) showed reduced ENaC-mediated Na+ absorption in the bpk mouse model in mixed CD cell monolayers. This finding could be due to the profound inhibitory effects of EGF on ENaC function and abnormal EGF signaling in cystic epithelia (15, 41, 49), given that these monolayers were established in the presence of EGF (55). Despite the overnight removal of EGF, EGF effects likely remained (55). Nevertheless, protein products of genes mapped in humans and specific mouse models display shared localization in cilia (23, 59). As such, and given the similar phenotype of the Na+ hyperabsorptive cyst-lining cell in orpk and human ARPKD CDs, further investigation is warranted in this area of polycystic kidney disease.

Why would the nonmotile monocilium central to the apical membrane of PCs influence ENaC subunits or NHE subtypes? In the proximal tubule or in any duct with high flow rates, the apical central monocilium is ideally positioned to be a flow sensor. Flow has indeed been shown to influence Na+ and K+ transport in the CCD (45, 57), and the luminal monocilium may play a critical role in transducing this mechanical signal. It is likely, however, in ducts and tubules with lower flow rates such as the CD of the kidney, the bile duct, the pancreatic duct, and elsewhere, that the role of the nonmotile monocilium may be different, not unlike that of an osmosensor or a chemosensor. It is intriguing to speculate that this monocilium sensor may sense the degree of diuresis vs. antidiuresis or the balance between natriuresis and antinatriuresis. Each PC may need a way to sample the tubular fluid osmolality and transduce that information to the ion transport pathways within the same cell. The same may be true for tubular fluid pH to sense states of acidosis vs. alkalosis. Further studies are in progress to address the possible cellular mechanism or mechanisms that connect the cilium to ENaCs and other solute and water transport pathways. In this light, it would be intriguing to evaluate a possible connection between the functions of the polycystins on cilia and ENaC activity as well as ENaC activity in conditional knockouts of the cilium, other than Polaris, induced by affecting ciliary proteins critical to cilia structure and function. Indeed, the similar Na+ hyperabsorptive states reported in human ARPKD (42) and in our orpk mouse cell models suggest that a hypertensive state is a general feature of ARPKD. Regardless of the cellular and molecular mechanisms to be elucidated, upregulated ENaC-mediated hyperabsorption of Na+ may be a primary underlying cause of profound hypertension in a majority of patients with ARPKD and should be contemplated as a target for clinical treatment.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Grants R01 DK-67343 (to E. M. Schwiebert) and R01 DK-55007 (to B. K. Yoder), as well as by NIDDK Grants DK-71007 (to P. D. Bell) and P01 DK-62345 (to L. M. Satlin), and by fellowship grants from the Polycystic Kidney Disease Research Foundation and the Revson Foundation (to W. Liu).


   FOOTNOTES
 

Address for reprint requests and other correspondence: E. M. Schwiebert, Depts. of Physiology and Biophysics and of Cell Biology, Univ. of Alabama at Birmingham, MCLM 740, 1918 University Blvd., Birmingham, AL 35294-0005 (e-mail: eschwiebert{at}physiology.uab.edu)

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
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
1. Adams CM, Snyder PM, and Welsh MJ. Paradoxical stimulation of a DEG/ENaC channel by amiloride. J Biol Chem 274: 15500–15504, 1999.[Abstract/Free Full Text]

2. Ahn YJ, Brooker DR, Kosari F, Harte BJ, Li J, Mackler SA, and Kleyman TR. Cloning and functional expression of the mouse epithelial sodium channel. Am J Physiol Renal Physiol 277: F121–F129, 1999.[Abstract/Free Full Text]

3. Bargmann CI, Hartwieg E, and Horvitz HR. Odorant-selective genes and neurons mediate olfaction in C. elegans. Cell 74: 515–527, 1993.[CrossRef][ISI][Medline]

4. Bargmann CI and Horvitz HR. Chemosensory neurons with overlapping functions direct chemotaxis to multiple chemicals in C. elegans. Neuron 7: 729–742, 1991.[CrossRef][ISI][Medline]

5. Baron A, Waldmann R, and Lazdunski M. ASIC-like, proton-activated currents in rat hippocampal neurons. J Physiol 539: 485–494, 2002.[Abstract/Free Full Text]

6. Benos DJ and Stanton BA. Functional domains within the degenerin/epithelial sodium channel (Deg/ENaC) superfamily of ion channels. J Physiol 520: 631–644, 1999.[Abstract/Free Full Text]

7. Bland RD. Fetal lung liquid and its removal near birth. In: The Lung: Scientific Foundations (2nd ed.), edited by Crystal RG, West JB, Barnes PJ, and Weibel ER. Philadelphia, PA: Lippincott Williams & Wilkins, 1997, vol. 2, chapt. 161, p. 2115–2127.

8. Bridges RJ, Newton BB, Pilewski JM, Devor DC, Poll CT, and Hall RL. Na+ transport in normal and CF human bronchial epithelial cells is inhibited by BAY 39-9437. Am J Physiol Lung Cell Mol Physiol 281: L16–L23, 2001.[Abstract/Free Full Text]

9. Brown PD, Davies SL, Speake T, and Millar ID. Molecular mechanisms of cerebrospinal fluid production. Neuroscience 129: 957–970, 2004.[CrossRef][ISI][Medline]

10. Caldwell RA, Boucher RC, and Stutts MJ. Serine protease activation of near-silent epithelial Na+ channels. Am J Physiol Cell Physiol 286: C190–C194, 2004.[Abstract/Free Full Text]

11. Calvet JP and Grantham JJ. The genetics and physiology of polycystic kidney disease. Semin Nephrol 21: 107–123, 2001.[CrossRef][ISI][Medline]

12. Carr MJ, Gover TD, Weinreich D, and Undem BJ. Inhibition of mechanical activation of guinea-pig airway afferent neurons by amiloride analogues. Br J Pharmacol 133: 1255–1262, 2001.[CrossRef][ISI][Medline]

13. Chalfant ML, Denton JS, Berdiev BK, Ismailov II, Benos DJ, and Stanton BA. Intracellular H+ regulates the {alpha}-subunit of ENaC, the epithelial Na+ channel. Am J Physiol Cell Physiol 276: C477–C486, 1999.[Abstract/Free Full Text]

14. D'Alton M, Romero R, Grannum P, DePalma L, Jeanty P, and Hobbins JC. Antenatal diagnosis of renal anomalies with ultrasound: IV. Bilateral multicystic kidney disease. Am J Obstet Gynecol 154: 532–537, 1986.[ISI][Medline]

15. Deachapunya C and O'Grady SM. Epidermal growth factor regulates the transition from basal sodium absorption to anion secretion in cultured endometrial epithelial cells. J Cell Physiol 186: 243–250, 2001.[CrossRef][ISI][Medline]

16. Eckardt KU, Möllmann M, Neumann R, Brunkhorst R, Burger HU, Lonne G, Scholz H, Keusch G, Buchholz B, Frei U, Bauer C, and Kurtz A. Erythropoietin in polycystic kidneys. J Clin Invest 84: 1160–1166, 1989.[ISI][Medline]

17. Gardner KD Jr and Grantham JJ. Milestones in nephrology: composition of fluid in twelve cysts of a polycystic kidney. J Am Soc Nephrol 9: 1965–1970, 1998. [Original citation: Gardner KD Jr. Composition of fluid in twelve cysts of a polycystic kidney. N Engl J Med 281: 985–988, 1969.][Medline]

18. Grantham JJ. Polycystic kidney disease: from the bedside to the gene and back. Curr Opin Nephrol Hypertens 10: 533–542, 2001.[CrossRef][ISI][Medline]

19. Grantham JJ and Wallace DP. Return of the secretory kidney. Am J Physiol Renal Physiol 282: F1–F9, 2002.[Abstract/Free Full Text]

20. Guay-Woodford LM and Desmond RA. Autosomal recessive polycystic kidney disease: the clinical experience in North America. Pediatrics 111: 1072–1080, 2003.[Abstract/Free Full Text]

21. Guay-Woodford LM. Murine models of polycystic kidney disease: molecular and therapeutic insights. Am J Physiol Renal Physiol 285: F1034–F1049, 2003.[Abstract/Free Full Text]

22. Haycraft CJ, Swoboda P, Taulman PD, Thomas JH, and Yoder BK. The C. elegans homolog of the murine cystic kidney disease gene Tg737 functions in a ciliogenic pathway and is disrupted in osm-5 mutant worms. Development 128: 1493–1505, 2001.[Abstract]

23. Hou X, Mrug M, Yoder BK, Lefkowitz EJ, Kremmidiotis G, D'Eustachio P, Beier DR, and Guay-Woodford LM. Cystin, a novel cilia-associated protein, is disrupted in the cpk mouse model of polycystic kidney disease. J Clin Invest 109: 533–540, 2002.[CrossRef][ISI][Medline]

24. Huangfu D, Liu A, Rakeman AS, Murcia NS, Niswander L, and Anderson KV. Hedgehog signalling in the mouse requires intraflagellar transport proteins. Nature 426: 83–87, 2003.[CrossRef][Medline]

25. Hughey RP, Bruns JB, Kinlough CL, Harkleroad KL, Tong Q, Carattino MD, Johnson JP, Stockand JD, and Kleyman TR. Epithelial sodium channels are activated by furin-dependent proteolysis. J Biol Chem 279: 18111–18114, 2004.[Abstract/Free Full Text]

26. Hurley DK, Bandy SM, Glew RH, Morris DM, and Gardner KD Jr. Kidney and breast cysts: a comparative study of fluids. Nephron 77: 304–308, 1997.[ISI][Medline]

27. Ji HL and Benos DJ. Degenerin sites mediate proton activation of {delta}beta{gamma}-epithelial sodium channel. J Biol Chem 279: 26939–26947, 2004.[Abstract/Free Full Text]

28. Kellenberger S and Schild L. Epithelial sodium channel/degenerin family of ion channels: a variety of functions for a shared structure. Physiol Rev 82: 735–767, 2002.[Abstract/Free Full Text]

29. Kizer NL, Lewis B, and Stanton BA. Electrogenic sodium absorption and chloride secretion by an inner medullary collecting duct cell line (mIMCD-K2). Am J Physiol Renal Fluid Electrolyte Physiol 268: F347–F355, 1995.[Abstract/Free Full Text]

30. Light DB, Schwiebert EM, Karlson KH, and Stanton BA. Atrial natriuretic peptide inhibits a cation channel in renal inner medullary collecting duct cells. Science 243: 383–385, 1989.[Abstract/Free Full Text]

31. Light DB, Corbin JD, and Stanton BA. Dual ion channel regulation by cyclic GMP and cyclic GMP-dependent protein kinase. Nature 344: 336–339, 1990.[CrossRef][Medline]

32. Madsen KM and Tisher CC. Structural-functional relationships along the distal nephron. Am J Physiol Renal Fluid Electrolyte Physiol 250: F1–F15, 1986.[Abstract/Free Full Text]

33. McCoy DE, Taylor AL, Kudlow BA, Karlson K, Slattery MJ, Schwiebert LM, Schwiebert EM, and Stanton BA. Nucleotides regulate NaCl transport across mIMCD-K2 cells via P2X and P2Y purinergic receptors. Am J Physiol Renal Physiol 277: F552–F559, 1999.[Abstract/Free Full Text]

34. McFarlane ME. The role of percutaneous drainage in the modern management of pancreatic pseudocysts. Int J Clin Pract 59: 383–384, 2005.[CrossRef][ISI][Medline]

35. North RA. Molecular physiology of P2X receptors. Physiol Rev 82: 1013–1067, 2002.[Abstract/Free Full Text]

36. Novak I and Hansen MR. Where have all the Na+ channels gone? In search of functional ENaC in exocrine pancreas. Biochim Biophys Acta 1566: 162–168, 2002.[Medline]

37. Orlowski J and Grinstein S. Diversity of the mammalian sodium/proton exchanger SLC9 gene family. Pflügers Arch 447: 549–565, 2004.[CrossRef][ISI][Medline]

38. Praetorius HA, Frokiaer J, Nielsen S, and Spring KR. Bending the primary cilium opens Ca2+-sensitive intermediate-conductance K+ channels in MDCK cells. J Membr Biol 191: 193–200, 2003.[CrossRef][ISI][Medline]

39. Praetorius HA and Spring KR. Bending the MDCK cell primary cilium increases intracellular calcium. J Membr Biol 184: 71–79, 2001.[CrossRef][ISI][Medline]

40. Praetorius HA and Spring KR. Removal of the MDCK cell primary cilium abolishes flow sensing. J Membr Biol 191: 69–76, 2003.[CrossRef][ISI]