QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.106.101360v1 317/3/1219    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jackson, E. K. Articles by Dubey, R. K. Search for Related Content PubMed PubMed Citation Articles by Jackson, E. K. Articles by Dubey, R. K.

GASTROINTESTINAL, HEPATIC, PULMONARY, AND RENAL

cAMP-Adenosine Pathway in the Proximal Tubule

Center for Clinical Pharmacology (E.K.J., L.C.Z., M.Z., D.G.G., C.Z., R.K.D.) and Departments of Medicine (E.K.J., L.C.Z., M.Z., D.G.G., C.Z., R.K.D.) and Pharmacology (E.K.J.), University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; and Clinic for Endocrinology, Department of Obstetrics and Gynecology, University Hospital Zurich, Zurich, Switzerland (R.K.D.)

Received January 12, 2006; accepted March 8, 2006.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
The "extracellular cAMP-adenosine pathway" refers to the conversion of cAMP to AMP by ecto-phosphodiesterase, followed by metabolism of AMP to adenosine by ecto-5'-nucleotidase, with all the steps occurring in the extracellular compartment. This study investigated whether the extracellular cAMP-adenosine pathway exists in proximal tubules. Freshly isolated proximal tubules rapidly converted basolaterally administered cAMP to AMP and adenosine. Proximal tubular cells in culture (first passage) rapidly converted apically administered cAMP to AMP and adenosine. In both freshly isolated proximal tubules and cultured proximal tubular cells, conversion of cAMP to AMP and adenosine was affected by a broad-spectrum phosphodiesterase inhibitor (3-isobutyl-1-methylxanthine), an ecto-phosphodiesterase inhibitor (1,3-dipropyl-8-p-sulfophenylxanthine), and a blocker of ecto-5'-nucleotidase (,-methyleneadenosine-5'-diphosphate) in a manner consistent with exogenous cAMP being processed by the extracellular cAMP-adenosine pathway. In cultured proximal tubular cells, but not freshly isolated proximal tubules, stimulation of adenylyl cyclase increased extracellular concentrations of cAMP, AMP, and adenosine plus inosine, and these changes were also modulated by the inhibitors in a manner consistent with the extracellular cAMP-adenosine pathway. Conversion of renal interstitial (basolateral) cAMP and AMP to adenosine in vivo was shown by microdialysis coupled with ion trap mass spectrometry. Western blot analysis showed A₁, A_2A, and A₃ receptors on both apical and basolateral proximal tubular membranes, with A₁ and A_2A receptors more highly expressed on basolateral compared with apical membranes. We conclude that cAMP that reaches either the apical or basolateral membranes of proximal tubular cells is converted in part to adenosine that has ready access to adenosine receptors.

A₁ receptor activation inhibits renin release (Jackson, 1991). Recent experiments with A₁ receptor knockout mice confirm a role for adenosine in mediating the reduction of renin release in response to increased sodium chloride delivery to the macula densa (Kim et al., 2005) or increased intrarenal perfusion pressure (Schweda et al., 2005).

In proximal tubular epithelial cells, adenosine via A₁ receptors mediates enhancement of electrolyte transport. Stimulation of A₁ receptors in cultured epithelial cells that express a proximal phenotype enhances Na⁺-glucose and Na⁺-phosphate symport (Coulson et al., 1991). Stimulation of A₁ receptors in microperfused proximal tubules increases basolateral Na⁺-3HCO ^–₃ symport (Takeda et al., 1993), and selective antagonism of A₁ receptors decreases Na⁺ -dependent phosphate transport in proximal tubular cells (Cai et al., 1994, 1995).

Because adenosine modulates processes in the renal cortex, it is important to understand the mechanisms of adenosine biosynthesis by cortical elements. Recently, we showed the ability of collecting duct cells to generate adenosine via the extracellular cAMP-adenosine pathway (Jackson et al., 2003). This pathway is defined as cAMP efflux from cells during activation of adenylyl cyclase, followed by the extracellular conversion of cAMP to adenosine by the serial actions of ecto-phosphodiesterase and ecto-5'-nucleotidase (Jackson and Dubey, 2001). This mechanism of extracellular adenosine production provides hormonal control of adenosine levels in the cell-surface biophase in which adenosine receptors reside. Tight coupling of the site of adenosine production to the site of adenosine receptors permits a low-capacity mechanism of adenosine biosynthesis to have a large impact on adenosine receptor activation (Jackson and Dubey, 2001). It is conceivable that the extracellular cAMP-adenosine pathway in proximal tubules is also a significant source of adenosine, both for proximal tubular epithelial cells and for other cellular elements in the renal cortex to which adenosine produced by the basolateral aspect of proximal tubules could diffuse. Egress of cAMP into the tubular lumen (across apical membrane) or interstitial space (across basolateral membrane) during activation of adenylyl cyclase could result in adenosine formation, provided that proximal tubular epithelial cells transport cAMP and express ecto-phosphodiesterase and ecto-5'-nucleotidase. Moreover, delivery of cAMP from distal sites (e.g., extrarenal via the bloodstream) could provide cAMP to the renal cortex for conversion to adenosine.

To test the hypothesis that the extracellular cAMP-adenosine pathway exists in proximal tubular epithelial cells, we examined the ability of freshly isolated proximal tubules and proximal tubular epithelial cells in culture to convert exogenous cAMP to AMP and adenosine. We also investigated whether stimulation of adenylyl cyclase increases extracellular levels of cAMP, AMP, and adenosine. Furthermore, we examined whether 3-isobutyl-1-methylxanthine (IBMX; a phosphodiesterase inhibitor that penetrates cell membranes) (Beavo and Reifsnyder, 1990), ,-methyleneadenosine-5'diphosphate (AMPCP; an ecto-5'-nucleotidase inhibitor) (Zimmermann, 1992), and 1,3-dipropyl-8-p-sulfophenylxanthine (DPSPX; a xanthine that inhibits only ecto-phosphodiesterase because it is restricted to the extracellular compartment and cannot gain access to intracellular phosphodiesterase) (Tofovic et al., 1991) alter the conversion of exogenous and endogenous cAMP to AMP and adenosine. To determine whether adenosine produced by the extracellular cAMP-adenosine pathway would have access to adenosine receptors, we characterized the relative expression in the proximal tubular basolateral versus proximal tubular apical membranes of all four adenosine receptor subtypes (A₁, A_2A, A_2B, and A₃). Finally, to determine whether cAMP delivered to the renal interstitial compartment could be converted to adenosine, we developed a method, using ion trap mass spectrometry, to measure purines with high sensitivity and specificity and measured AMP and adenosine levels in renal cortical dialysate during the targeted delivery of cAMP and AMP to the renal cortical interstitial compartment by microdialysis.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Isolation of Proximal Tubules. Kidneys were harvested from adult male rats, and proximal tubules were rapidly isolated using a Percoll solution-centrifugation method described previously in detail by us (Jackson et al., 2003). Western blot studies established that freshly isolated proximal tubules did not express bumetanide-sensitive cotransporter type 1 (marker for thick ascending limb), thiazide-sensitive cotransporter (marker for distal tubules), or aquaporin-2 (marker for collecting duct) but did express aquaporin-1, sodium bicarbonate cotransporter type 1 (NBC-1), and sodium hydrogen exchanger type 3 (NHE-3) (markers for proximal tubules).

Isolation of Proximal Tubular Basolateral and Apical Membranes. Isolation of apical and basolateral membrane from kidney cortex was based on the methods described by Kinsella et al. (1979) and Biber et al. (1981). Kidneys were harvested from adult rats and placed in ice-cold phosphate-buffered saline. From this point forward, all the isolation procedures were performed at 4°C. The cortex was removed, minced into small pieces, and washed with ice-cold phosphate-buffered saline. The tissue was transferred to 8% sucrose with protease inhibitors (aprotinin, leupeptin, and pepstatin, each at 10 ng/ml, and 100 µM phenylmethanesulfonyl fluoride) (Sigma, St. Louis, MO). The dispersed cortex was homogenized with 10 strokes in a Dounce glass homogenizer (Fisher, Pittsburgh, PA) with a tight-fitting pestle. The homogenate was centrifuged at 1000g for 10 min, and the supernatant (A) was collected. The pellet was reprocessed as above, and its supernatant (B) was collected. Supernatants A and B were combined and centrifuged at 9500g for 10 min, and the supernatant (C) and soft lighter portion of the pellet were centrifuged at 47,000g for 20 min in a Beckman (Fullerton, CA) Ti 90 rotor. The soft, lighter upper portion of the pellet was harvested and suspended in medium A (25 mM HEPES, 100 mM mannitol, 2 mM CaCl₂, 1 mM MgCl₂, and 1 mM MnCl₂, adjusted to pH 7.0 with Tris) to the final protein concentration of 6 to 8 mg/ml. The suspension was left in an ice bath for 1 h and centrifuged at 1400g for 12 min. The supernatant (D) was saved and contained the apical membranes. The pellet was reprocessed with medium A three times by suspending and centrifuging to reduce apical membrane contamination of the basolateral membranes in the pellet. Next, the pellet containing basolateral membranes was layered over a discontinuous sucrose gradient composed of 8% sucrose solution and 31% sucrose solution and spun at 90,000g for 60 min in a Beckman Ti 90 rotor. The 8 to 31% sucrose interface portion was harvested as basolateral membrane. Both apical and basolateral membrane portions were diluted in 25 mM HEPES buffer and spun down at 47,000g for 20 min. The pellets were dissolved in a lysis buffer (50 mM Tris with pH at 7.0, 2% SDS, and 10% glycerol) for Western blot analysis.

Western Blot Analysis. Electrophoresis, Western blot analysis, and immunostaining were performed as described previously (Jackson et al., 2002). Anti-bumetanide-sensitive cotransporter type 1, anti-thiazide-sensitive cotransporter, anti-aquaporin-1, and anti-aquaporin-2 antibodies were provided by Dr. Mark Knepper (National Institutes of Health, Bethesda, MD); anti-NBC-1, anti-NHE-3, anti-A_2A-receptor, anti-A_2B-receptor, and anti-A₃-receptor antibodies were obtained from Chemicon (Temecula, CA); and anti-A₁-receptor and anti--actin antibodies were obtained from Sigma.

Culture of Proximal Tubular Epithelial Cells. Proximal tubular epithelial cells in culture were obtained from freshly isolated proximal tubules using a method described previously by us for collecting duct cells (Jackson et al., 2003). All the experiments were performed in proximal tubular cells in first passage.

Studies in Freshly Isolated Proximal Tubules. Freshly isolated proximal tubules (approximately 0.2 mg) were incubated in capped tubes in 0.3 ml of L-15 medium in a shaking water bath at 37°C without or with exogenous cAMP at the indicated concentrations in the absence and presence of IBMX (1 mM), AMPCP (0.1 mM), or DPSPX (1 mM). The concentrations of various inhibitors were based on our previous published experience with inhibition of exogenous cAMP metabolism in cultured cardiac fibroblasts and vascular smooth muscle cells (Dubey et al., 1996, 2000). In addition, some proximal tubules were treated with stimulators of adenylyl cyclase. Adenylyl cyclase was stimulated by treating the cells with a mixture of forskolin (10 µM; direct activator of adenylyl cyclase), parathyroid hormone (1 µM), isoproterenol (1 µM; -adrenoceptor agonist), and fenoldopam (10 µM; agonist of dopamine type 1 receptors). Because we did not know which receptors were both expressed and coupled to stimulate adenylyl cyclase, we used the strategy of adding several agonists of adenylyl cyclase simultaneously. Just before adding the proximal tubules to the L-15 medium, the L-15 medium was bubbled with 100% oxygen. Just after adding the isolated proximal tubules, the headspace of the capped tubes was flushed with 100% oxygen. After the indicated time of incubation, the isolated proximal tubules were centrifuged, and the supernatant was collected, transferred immediately into ice-cold tubes, and frozen at –70°C until purine levels were measured. The pellet was solubilized in 0.5 N NaOH, and protein content was measured by the copper bicinchoninic acid method.

Studies in Proximal Tubular Epithelial Cells in Culture. Monolayers of proximal tubular cells in culture were grown in 12-well culture plates. On the day of the experiment, the proximal tubular cells were washed with phosphate-buffered saline and then incubated (37°C under 5% carbon dioxide in cell incubator) with various treatments. Some proximal tubular cells were incubated for 0, 5, or 30 min with 0, 3, 10, or 30 µM cAMP in phosphate-buffered saline. In cells treated with 0, 3, 10, or 30 µM cAMP for "0" min, the phosphate-buffered saline containing cAMP was added to the cells and then quickly removed. Other proximal tubular cells were treated with 30 µM cAMP for 30 min in the presence and absence of IBMX, DPSPX, and AMPCP. In addition, some proximal tubular cells were treated for 30 min with stimulators of adenylyl cyclase (see above) and in the presence or absence of IBMX, DPSPX, and AMPCP. At the end of the specified incubation time, the conditioned phosphate-buffered saline was collected, transferred immediately into ice-cold tubes, and frozen at –70°C until purine levels were measured. At the end of the experiment, cells were solubilized in 0.5 N NaOH, and protein content was measured by the copper bicinchoninic acid method.

Analysis of Purines in Studies with Freshly Isolated Proximal Tubules and Proximal Tubular Cells in Culture. In in vitro experiments using exogenous cAMP, purines were measured by high-pressure liquid chromatography using UV absorption as described previously (Jackson et al., 1996). In experiments with adenylyl cyclase activators, cAMP and adenosine concentrations were measured by high-pressure liquid chromatography using fluorescence detection (higher sensitivity relative to UV absorption method) as described previously (Jackson et al., 1996), and AMP was measured by high-pressure liquid chromatography using UV absorption.

In Vivo Microdialysis Experiments. Adult male rats were housed in the University of Pittsburgh Animal Care Facility for at least 1 week before the experiments. Animals were fed Pro Lab RMH 3000 rodent diet containing 0.26% sodium and 0.82% potassium and water ad libitum. On the day of the experiment, each animal was anesthetized with Inactin (100 mg/kg, i.p.) and placed on a Deltaphase isothermal pad. Body temperature was monitored with a rectal temperature probe and maintained at 37°C by adjusting a heat lamp positioned above the rat. A short section of polyethylene-240 tubing was placed in the trachea to facilitate respiration. A polyethylene-50 cannula was inserted into the left jugular vein, and 0.9% saline was infused at 50 µl/min to maintain adequate extracellular fluid volume. A polyethylene-50 catheter was inserted into the left carotid artery for measurement of mean arterial blood pressure and heart rate (to ensure hemodynamic stability) via a digital blood pressure analyzer (model BPA; Digi-Med, Louisville, KY). A microdialysis probe (CMA/20 microdialysis probe, 4 mm; Bioanalytical Systems, Inc., West Lafayette, IN) was placed in the renal cortex of both the left and right kidneys. Microdialysis probes had a membrane outer diameter of 0.5 mm and a 20,000-Da membrane cutoff. Using a Braintree Scientific infusion pump (model BSP 99; Braintree, MA), the microdialysis probe was perfused at 2 µl/min (1 ml of 0.9% saline plus 10 units of heparin). Animals were allowed to rest for 2 h while perfusion of the probe was continued because our previous experience indicated that purine levels are elevated immediately after inserting microdialysis probes and 2 h is required for levels of purines to decrease to baseline. Probe positioning was confirmed by visualization of the probe tip in the kidney at the end of the experiment. After the 2-h rest period, dialysate was collected for 30 min. Next, either cAMP or AMP (final concentration of 30 µM) was added to the perfusate of either the right or left kidney. Whether the right kidney received cAMP and the left AMP, or vice versa, was determined randomly. Fifteen minutes after adding cAMP or AMP, another 30-min dialysate sample was collected.

Analysis of Purines in Dialysate Samples. Measuring purine levels in samples from in vivo studies is challenging because of the low levels of analytes and the presence of potentially interfering substances. To ensure the accuracy of the results, we developed a novel assay for measuring purines using ion trap mass spectrometry. Because this is the first presentation of this assay, the assay is described in detail below.

The assay was developed using a ThermoFinnigan high-pressure liquid chromatographic system coupled to a ThermoFinnigan LCQ Duo ion trap mass spectrometer equipped with an electrospray ionization source (Thermo Electron Corporation, Waltham, MA). The analytical column was a C-18 Eclipse Zorbax XDB column (4.6 mm x 150 mm). The mobile phase flow rate was 0.5 ml/min. From 0 to 1.5 min, the mobile phase was 100% water. The mobile phase composition was then changed (linear gradient) to 90% water plus 10% methanol over 0.5 min and then kept constant at this composition for an additional 8 min. The mobile phase was acidified with 0.1% formic acid. The mass spectrometer was operated in the positive ion mode. Nitrogen was used both as sheath and auxiliary gas at a pressure optimized automatically for each of the analytes during tuning of each analyte. The spray voltage and the heated capillary temperatures were set at 5 kV and 270°C, respectively. The analytes were monitored using single ion monitoring: for AMP, mass-to-charge ratio (m/z) = 348; for hypoxanthine, m/z = 137; for adenosine and adenine 9--D arabinofuranoside (internal standard), m/z = 268; and for inosine, m/z = 291. Samples were injected directly into the analytical system without previous sample preparation. For the first 5 min of the sample run, flow was diverted from the source to a waste bottle. This allowed for the removal of salts present in the sample and prevented blocking of the bore of the heated capillary by the nonvolatile salts. AMP, adenosine, inosine, hypoxanthine, and internal standard were prepared in phosphate-buffered saline. Dilutions were made in phosphate-buffered saline.

The assay was validated for AMP, adenosine, inosine, and hypoxanthine. A standard curve was generated in duplicate using the following concentrations: 0.5, 1, 2, 5, 10, 20, and 30 pg/µl. Twenty-seven quality control samples (QC) were processed along with the standard curve samples (nine QC each at the low, medium, and high range of the standard curve; i.e., 3, 12, and 25 pg/µl, respectively). This was repeated three times on different days and each time with freshly prepared solutions. From the QC, the intra-assay coefficient and interassay coefficient of variation (CV) were calculated.

The liquid chromatographic method afforded baseline separation of all four compounds and the internal standard. The retention times of AMP, adenosine, inosine, hypoxanthine, and internal standard (adenine 9--D arabinofuranoside) were 6.57, 7.02, 7.92, 8.76, and 9.47 min, respectively. No interfering peaks were detected in nonspiked (blank) samples. All analytes were detected at the lowest concentrations used (0.5 pg/µl), and all the compounds produced a linear relationship in the range used (0.5–30 pg/µl). The assay system could detect <25 pg injected on column of each of the analytes. The correlation coefficients (R²) of the standard curves used in the validation are shown in Table 1. The standard curves were reproducible in the three independent experiments. The R² values for all the standard curves were 0.99. Table 1 also shows the regression equations for the standard curves for each analyte and for each experiment. The calculated slopes indicate that the assay detected adenosine and hypoxanthine with the highest sensitivity, followed by inosine and AMP.

Table 2 summarizes the intra-assay CV from QC for the three independent validation experiments for each of the compounds. The lowest concentration QC for AMP had CV ranging from 9.37 to 12.5%, the medium concentration QC from 4.3 to 12.2%, and the highest concentration QC from 6.77 to 8.65%. For hypoxanthine, the range for the low concentration QC was 7.52 to 16.6%, for medium concentration QC from 3.97 to 7.05%, and for high concentration QC from 4.12 to 7.99%. For adenosine, the range for low concentration QC was 6.29 to 11.3%, for medium concentration QC from 4.17 to 5.51%, and for high concentration QC from 4.59 to 5.32%. For inosine, the range for low concentration QC was 12.82 to 15.6%, for medium concentration QC from 4.58 to 12.2%, and for high concentration QC from 5.38 to 10%.

Table 3 shows the interassay CV. For AMP, CV were 11.7, 10.4, and 9.25% for the low-, medium-, and high-concentration QC, respectively. For hypoxanthine, the CV were 12, 15.7, and 6.8% for the low, medium, and high concentration QC, respectively. For adenosine, the CV were 8.73, 4.92, and 5.73% for the low, medium, and high concentration QC, respectively. For inosine, the CV for the low, medium, and high concentration QC were 14.7, 9.12, and 8.28%, respectively.

The analytes were very stable as indicated by QC samples run 24 h after being at room temperature in the autosampler. For adenosine, the amounts calculated did not deviate more than 2.4, 4.9, and 10.0% from the average of all the low, medium, and high QC concentrations, and had CV of 2.8, 2.2, and 4.6% for the low, medium, and high concentrations, respectively. For inosine, the deviations from the average of the QC were 10.3, 5.6, and 12.9%, with CV of 17.1, 12.5, and 6.3% for the low, medium, and high concentrations, respectively. For AMP, the deviations from the average of the QC were 5.4, 7.8, and 1.95%, with CV of 9.3, 14.9, and 8.4% for the low, medium, and high concentrations, respectively. Finally, for hypoxanthine, deviations were 9.7, 0.24, and 0.38%, with CV of 2.8, 6.4, and 8% for the low, medium, and high concentrations, respectively.

Statistical Analysis. Data were analyzed by one-factor or two- factor analysis of variance, followed by Fisher's least significant difference (LSD) test or paired Student's t test as appropriate. The criterion of significance was P < 0.05. Results are presented as mean ± S.E.M.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Studies in Freshly Isolated Proximal Tubules. Figure 1 illustrates the time-dependent (0–60 min) conversion of cAMP (30 µM) to AMP (Fig. 1A), adenosine (Fig. 1B), inosine (Fig. 1C), hypoxanthine (Fig. 1D), and xanthine (Fig. 1E) in proximal tubules. Addition of cAMP to proximal tubules caused a significant increase in the extracellular levels of AMP (P < 0.001), adenosine (P < 0.001), inosine (P < 0.001), hypoxanthine (P = 0.002), and xanthine (P < 0.001) (all comparisons with time control; i.e., proximal tubules incubated for the same period but without added cAMP). Extracellular levels of AMP and adenosine reached a plateau in approximately 5 min; however, extracellular levels of inosine, hypoxanthine, and xanthine increased over time of incubation.

View larger version (29K): [in this window] [in a new window]   Fig. 1. Line graph showing the time-dependent metabolism of cAMP to AMP (A), adenosine (B), inosine (C), hypoxanthine (D), and xanthine (E) in freshly isolated proximal tubules. Tubules were incubated for the indicated times with 30 µM cAMP. Values are mean ± S.E.M. of five to six different preparations, and a indicates P < 0.05 by Fisher's LSD test compared with basal (B).

View larger version (24K): [in this window] [in a new window]   Fig. 2. Line graph showing the concentration-dependent metabolism of cAMP to AMP (A), adenosine (B), inosine (C), hypoxanthine (D), and xanthine (E) in freshly isolated proximal tubules. Tubules were incubated with the indicated concentrations of cAMP for 20 min. B indicates basal levels. Values are mean ± S.E.M. of five different preparations, and a indicates P < 0.05 by Fisher's LSD test compared with basal (B).

View larger version (22K): [in this window] [in a new window]   Fig. 3. Freshly isolated proximal tubules were incubated with cAMP (30 µM) for 20 min with and without IBMX (1 mM), DPSPX (1 mM), or AMPCP (0.1 mM), and the amounts of AMP (A), adenosine (B), and inosine (C) were determined. Values are mean ± S.E.M. of five preparations, and a and b indicate P < 0.05 by Fisher's LSD test compared with cAMP alone and cAMP + AMPCP, respectively.

Studies in Proximal Tubular Epithelial Cells in Culture. Figure 4 illustrates conversion in proximal tubular cells in culture of exogenous cAMP to AMP (Fig. 4A) and adenosine (Fig. 4B) as a function of time (0, 5, and 30 min) and cAMP concentration (0, 3, 10, and 30 µM). Addition of increasing concentrations of cAMP to proximal tubular cells and incubating cells for increasing durations with cAMP caused corresponding increases in the extracellular levels of AMP and adenosine.

View larger version (29K): [in this window] [in a new window]   Fig. 4. Line graph showing the time-dependent and concentration-dependent metabolism of cAMP to AMP (A) and adenosine (B) in proximal tubular cells in culture (first passage). Values are mean ± S.E.M. for three preparations, and a and b indicate P < 0.05 by Fisher's LSD test compared with zero cAMP group at corresponding time and zero time within the given concentration group, respectively.

View larger version (25K): [in this window] [in a new window]   Fig. 5. Proximal tubular cells in culture (first passage) were incubated with cAMP (30 µM) for 30 min with and without IBMX (1 mM), DPSPX (1 mM), or AMPCP (0.1 mM), and the amounts of AMP (A) and adenosine (B) were determined. Values are mean ± S.E.M. for three preparations, and a and b indicate P < 0.05 by Fisher's LSD test compared with cAMP alone and cAMP + AMPCP, respectively.

View larger version (17K): [in this window] [in a new window]   Fig. 6. Proximal tubular cells in culture (first passage) were incubated with a mixture of forskolin (10 µM; direct activator of adenylyl cyclase), parathyroid hormone (1 µM), isoproterenol (1 µM; -adrenoceptor agonist), and fenoldopam (10 µM; agonist of dopamine type 1 receptors) for 30 min with and without IBMX (1 mM), DPSPX (1 mM), or AMPCP (0.1 mM), and the amounts of cAMP (n = 3) (A–D) and AMP (n = 5–6) (E–H) were determined. Values are mean ± S.E.M, and a indicates comparison (P < 0.05 by Fisher's LSD test) between no adenylyl cyclase stimulation (–) versus with adenylyl cyclase stimulation (+) within a given treatment group. In addition, b indicates comparison (P < 0.05 by Fisher's LSD test) of the indicated group with all the other groups at the corresponding level of adenylyl cyclase activation. AC, adenylyl cyclase; <DL, less than detection limit.

View larger version (15K): [in this window] [in a new window]   Fig. 7. Proximal tubular cells in culture (first passage) were incubated with a mixture of forskolin (10 µM; direct activator of adenylyl cyclase), parathyroid hormone (1 µM), isoproterenol (1 µM; -adrenoceptor agonist), and fenoldopam (10 µM; agonist of dopamine type 1 receptors) for 30 min without (A) and with either IBMX (1 mM; B) or AMPCP (0.1 mM; C), and the amounts of adenosine + inosine were determined. Values are mean ± S.E.M. for six preparations, and a indicates comparison (P < 0.05 by Fisher's LSD test) between no adenylyl cyclase stimulation (–) versus with adenylyl cyclase stimulation (+) within a given treatment group. In addition, b indicates comparison (P < 0.05 by Fisher's LSD test) with the control group at the corresponding level of adenylyl cyclase activation. AC, adenylyl cyclase.

Expression of Adenosine Receptor Subtypes in Apical Versus Basolateral Proximal Tubular Membranes. As shown in Fig. 8, apical membranes, relative to basolateral membranes, were enriched in NHE-3 (Fig. 8A) but expressed little NBC-1 relative to basolateral membranes (Fig. 8B), a finding consistent with the well accepted distribution of NHE-3 and NBC-1 in apical versus basolateral proximal tubular membranes. As shown in Fig. 9, although both apical and basolateral membranes expressed both A₁ (Fig. 9A) and A_2A (Fig. 9B) receptors, basolateral membranes clearly expressed higher levels of both of these adenosine receptor subtypes. We were unable to identify A_2B receptors in either apical or basolateral proximal tubular membranes (data not shown). Unlike A₁ or A_2A receptors, A₃ receptors were more abundant on apical versus basolateral membranes (Fig. 10).

View larger version (21K): [in this window] [in a new window]   Fig. 8. Expression of NHE-3 (A) and NBC-1 (B) in apical and basolateral membranes isolated from proximal tubules. Top of each panel is Western blot, and bottom is a bar graph showing densitometry results normalized to -actin. Values are mean ± S.E.M. for six preparations, and a indicates comparison (P < 0.05 by Student's t test) with apical.

View larger version (27K): [in this window] [in a new window]   Fig. 9. Expression of A1 receptor (A) and A2A receptor (B) in apical and basolateral membranes isolated from proximal tubules. Top of each panel is Western blot, and bottom is a bar graph showing densitometry results normalized to -actin. Values are mean ± S.E.M. for six preparations, and a indicates comparison (P < 0.05 by Student's t test) with apical.

View larger version (24K): [in this window] [in a new window]   Fig. 10. Expression of A3 receptor in apical and basolateral membranes isolated from proximal tubules. Top is Western blot, and bottom is a bar graph showing densitometry results normalized to -actin. Values are mean ± S.E.M. for six preparations, and a indicates comparison (P < 0.05 by Student's t test) with apical.

View larger version (15K): [in this window] [in a new window]   Fig. 11. Effects of addition of cAMP to the perfusate on microdialysate levels of AMP (A) and adenosine (B). C, effects of addition of AMP to the perfusate on microdialysate levels of adenosine. In addition, a indicates comparison (P < 0.05 by paired Student's t test) between basal and treatment with either cAMP or AMP.

	Discussion

Top Abstract Materials and Methods Results Discussion References

Evidence indicates that the kidney expresses an extracellular cAMP-adenosine pathway. Infusion of cAMP into isolated kidneys increases production of AMP, adenosine, and inosine, and the increases in AMP and adenosine are reduced by IBMX (phosphodiesterase inhibitor) and DPSPX (ectophosphodiesterase inhibitor) (Mi and Jackson, 1995). In addition, AMPCP (ecto-5'-nucleotidase inhibitor) blocks the increases in adenosine, but not AMP, caused by cAMP in isolated kidneys (Mi and Jackson, 1995). Moreover, cultured preglomerular vascular smooth muscle cells (Jackson et al., 1997) and collecting duct cells (Jackson et al., 2003) and freshly isolated preglomerular microvessels (Jackson and Mi, 2000) and collecting ducts (Jackson et al., 2003) express the cAMP-adenosine pathway.

The present study supports the conclusion that renal epithelial cells in proximal tubules also express an extracellular cAMP-adenosine pathway. Addition of exogenous cAMP to isolated proximal tubules or proximal tubular cells in culture causes synthesis of AMP and adenosine. Importantly, the conversion of exogenous cAMP to AMP and adenosine is blocked by inhibition of phosphodiesterase (with IBMX) and ecto-phosphodiesterase (with DPSPX), and inhibition of ecto-5'-nucleotidase (with AMPCP) blocks the conversion of exogenous cAMP to adenosine and enhances the accumulation of AMP.

Because cAMP does not penetrate lipid bilayers, the conversion of exogenous cAMP to extracellular adenosine most likely is caused by extracellular metabolism of cAMP. This conclusion is supported by the results with AMPCP and DPSPX because AMPCP inhibits ecto-5'-nucleotidase but not cytosolic 5'-nucleotidase (Zimmermann, 1992), and DPSPX does not penetrate cell membranes (Tofovic et al., 1991). The ability of these inhibitors to attenuate the conversion of exogenous cAMP to extracellular adenosine and its metabolites is strong evidence for an extracellular site of cAMP metabolism. Moreover, the ability of AMPCP to increase extracellular AMP levels is consistent with an extracellular site of cAMP conversion to AMP. In this regard, if cAMP were metabolized to AMP intracellularly, inhibition of ecto-5'-nucleotidase would not prevent the further metabolism of AMP to adenosine and would not cause the accumulation of extracellular AMP.

The source of endogenous cAMP for processing to adenosine by proximal tubules could come from at least two nonexclusive sources. One source of cAMP would be the local egress of cAMP onto the proximal tubular cell surface after activation of proximal tubular adenylyl cyclase (autocrine/paracrine model). Indeed, in proximal tubular cells in culture, but not in freshly isolated proximal tubules, activation of adenylyl cyclase causes accumulation of cAMP into the extracellular medium. Moreover, in proximal tubular cells in culture, when ecto-5'-nucleotidase is inhibited, AMP accumulates in the extracellular medium. The lack of extracellular AMP accumulation in the absence of ecto-5'-nucleotidase inhibition is consistent with the observation that proximal tubular cells express high levels of ecto-5'-nucleotidase (Takahashi et al., 1989), particularly on apical membranes (Vekaria et al., 2006), which would metabolize AMP to downstream purines and thus prevent the accumulation of AMP. It is noteworthy that the increase in extracellular cAMP induced by stimulation of adenylyl cyclase is reduced by IBMX, DPSPX, and AMPCP. This may be because of blockade of adenosine receptors by IBMX and DPSPX and inhibition of adenosine formation by AMPCP, thus reducing A_2A receptor-mediated cAMP production. More important is the observation that stimulation of adenylyl cyclase in cultured proximal tubular cells significantly increases the extracellular accumulation of adenosine plus inosine (metabolite of adenosine) and that this increase is blocked by either inhibition of phosphodiesterase with IBMX or inhibition of ecto-5'-nucleotidase with AMPCP.

Unlike proximal tubular cells in culture, stimulation of adenylyl cyclase in isolated proximal tubules does not increase extracellular levels of cAMP in the bulk medium. This observation is consistent with the fact that cAMP is transported across the apical membrane into the tubular lumen via MRP4 (van Aubel et al., 2002), a direction of cAMP transport that would allow intracellular cAMP to reach the outer surface of the apical membrane. In contrast, cAMP is transported by OAT1 in basolateral membranes (Sekine et al., 1997), but only in the direction from extracellular to intracellular. Thus, stimulation of adenylyl cyclase would be expected to increase extracellular cAMP on apical, but not basolateral, membrane outer surfaces. In studies with freshly isolated proximal tubules, the bulk medium would be influenced nearly entirely by efflux across the basolateral membrane because the ends of the tubular segments are collapsed. In contrast, in proximal tubular cells in culture, the bulk medium would be influenced nearly entirely by efflux across the apical membrane because the basolateral membrane is against the culture plate matrix.

The best explanation for the results with activators of adenylyl cyclase is that in proximal tubules, local stimulation of adenylyl cyclase places cAMP on the extracellular aspect of apical, but not basolateral, membranes. However, this does not imply that the extracellular cAMP-adenosine pathway only functions on apical proximal tubular surfaces. Another source of cAMP for the extracellular cAMP-adenosine pathway is cAMP released into the bloodstream from other organs (e.g., the liver) (Kuster et al., 1973), followed by filtration of cAMP into proximal tubules and delivery of cAMP to the cortical renal interstitial space via peritubular capillaries. Thus, both the basolateral and apical proximal tubular membranes would be bathed in systemically delivered cAMP, which according to the present study should be converted in part to adenosine. In addition, the basolateral proximal tubular membranes may receive cAMP produced by adjacent renal cell types in the renal cortex. Importantly, ecto-5'-nucleotidase is present both on apical membranes in the proximal tubules and in the interstitial compartment (Vekaria et al., 2006).

The present study shows that both A₁ and A_2A receptors are enriched in basolateral membranes relative to apical membranes. Importantly, both A₁ and A_2A receptors are coupled to adenylyl cyclase via G proteins, and in proximal tubules, adenylyl cyclase is located nearly exclusively on basolateral, not apical, membranes (Hanson and Linas, 1995). These concepts would imply that adenosine formed from cAMP within the renal interstitial compartment would have more opportunity to affect tubular function than adenosine formed from cAMP in the luminal compartment.

An important issue is whether cAMP is converted to AMP and adenosine in the renal cortical interstitial compartment of the intact kidney. To address this question, we placed cAMP and AMP into the renal interstitial compartment using a microdialysis probe and measured local production of AMP and adenosine via the same probe. Consistent with our results in isolated proximal tubules and consistent with the enrichment of A₁ and A_2A receptors in proximal tubular basolateral membranes, addition of cAMP to the renal cortical interstitial compartment caused a 3-fold increase in renal cortical interstitial levels of AMP and adenosine, and addition of AMP to the interstitial compartment caused a 15-fold increase in adenosine. These results strongly support the concept that in intact kidneys conversion of cAMP to adenosine via the extracellular cAMP-adenosine pathway provides renal cortical interstitial adenosine, which acts on basolateral adenosine receptors to modulate tubular function. The conversion of cAMP to adenosine in the renal cortical interstitial space probably involves multiple cell types, including proximal tubular cells, but also preglomerular vascular smooth muscle cells as previously shown by us (Jackson and Mi, 2000).

An important question is whether the extracellular cAMP-adenosine pathway influences renal biology. Because adenosine potently affects multiple renal physiological processes, it is likely that adenosine from any source, including the cAMP-adenosine pathway, alters renal biology. In support of this conclusion, our previous studies show that the extracellular cAMP-adenosine pathway inhibits glomerular mesangial cell proliferation and the production of collagen by mesangial cells (Dubey et al., 1997). In the present study, the increase in extracellular cAMP induced by stimulation of adenylyl cyclase in cultured proximal tubular cells was reduced by blocking the extracellular-cAMP pathway with IBMX, DP-SPX, or AMPCP. This finding is consistent with adenosine formed from the pathway activating A_2A receptor-mediated cAMP production so that blocking the pathway reduces cAMP production.

In summary, the proximal tubule expresses an extracellular cAMP-adenosine pathway that provides extracellular adenosine. On the apical membrane, the source of cAMP is probably both cAMP locally released from proximal tubular cells and filtered cAMP; however, on the basolateral membrane, the source of cAMP is probably cAMP delivered from remote sources or produced by other renal cell types in the renal cortex.

	Footnotes

 
This work was supported by National Institutes of Health Grant DK68575.

doi:10.1124/jpet.106.101360.

ABBREVIATIONS: IBMX, 3-isobutyl-1-methylxanthine; AMPCP, ,-methyleneadenosine-5'-diphosphate; DPSPX, 1,3-dipropyl-8-p-sulfophenylxanthine; NBC-1, sodium bicarbonate cotransporter type 1; NHE-3, sodium hydrogen exchanger type 3; QC, quality control sample(s); CV, coefficient(s) of variation; LSD, least significant difference.

Address correspondence to: Edwin K. Jackson, Center for Clinical Pharmacology, University of Pittsburgh School of Medicine, 100 Technology Drive, Suite 450, Pittsburgh, PA 15219. E-mail: edj{at}pitt.edu

	References

Top Abstract Materials and Methods Results Discussion References

 

Agmon Y, Dinour D, and Brezis M (1993) Disparate effects of adenosine A₁- and A₂-receptor antagonists on intrarenal blood flow. Am J Physiol 265: F802–F806.[Medline]

Barber R and Butcher RW (1981) The quantitative relationship between intracellular concentration and egress of cyclic AMP from cultured cells. Mol Pharmacol 19: 38–43.[Abstract/Free Full Text]

Beavo JA and Reifsnyder DH (1990) Primary sequence of cyclic nucleotide phosphodiesterase isozymes and the design of selective inhibitors. Trends Pharmacol Sci 11: 150–155.[CrossRef][Medline]

Biber J, Stieger B, Haase W, and Murer H (1981) A high yield preparation for rat kidney brush border membranes. Different behaviour of lysosomal markers. Biochim Biophys Acta 647: 169–176.[Medline]

Brown R, Ollerstam A, Johansson B, Skott O, Gebre-Medhin S, Fredholm B, and Persson AE (2001) Abolished tubuloglomerular feedback and increased plasma renin in adenosine A₁ receptor-deficient mice. Am J Physiol 281: R1362–R1367.

Cai H, Batuman V, Puschett DB, and Puschett JB (1994) Effect of KW-3902, a novel adenosine A₁ receptor antagonist, on sodium-dependent phosphate and glucose transport by the rat renal proximal tubular cell. Life Sci 55: 839–845.[CrossRef][Medline]

Cai H, Puschett DB, Guan S, Batuman V, and Puschett JB (1995) Phosphate transport inhibition by KW-3902, an adenosine A₁ receptor antagonist, is mediated by cyclic adenosine monophosphate. Am J Kidney Dis 26: 825–830.[Medline]

Coulson R, Johnson RA, Olsson RA, Cooper DR, and Scheinman SJ (1991) Adenosine stimulates phosphate and glucose transport in opossum kidney epithelial cells. Am J Physiol 260: F921–F928.[Medline]

Dubey RK, Gillespie DG, Mi Z, and Jackson EK (1997) Cyclic AMP-adenosine pathway inhibits glomerular mesangial cell growth (Abstract). Hypertension 30: 506.

Dubey RK, Gillespie DG, Mi Z, and Jackson EK (2000) Cardiac fibroblasts express the cAMP-adenosine pathway. Hypertension 36: 337–342.[Abstract/Free Full Text]

Dubey RK, Mi Z, Gillespie DG, and Jackson EK (1996) Cyclic AMP-adenosine pathway inhibits vascular smooth muscle cell growth. Hypertension 28: 765–771.[Abstract/Free Full Text]

Hansen PB and Schnermann J (2003) Vasoconstrictor and vasodilator effects of adenosine in the kidney. Am J Physiol 285: F590–F599.

Hanson AS and Linas SL (1995) Beta-adrenergic receptor function in rat proximal tubule epithelial cells in culture. Am J Physiol 268: F553–F560.[Medline]

Hedqvist P and Fredholm BB (1976) Effects of adenosine on adrenergic neurotransmission; prejunctional inhibition and postjunctional enhancement. Naunyn-Schmiedeberg's Arch Pharmacol 293: 217–223.[CrossRef][Medline]

Hedqvist P, Fredholm BB, and Olundh S (1978) Antagonistic effects of theophylline and adenosine on adrenergic neuroeffector transmission in the rabbit kidney. Circ Res 43: 592–598.[Abstract/Free Full Text]

Holz FG and Steinhausen M (1987) Renovascular effects of adenosine receptor agonists. Renal Physiol 10: 272–282.[Medline]

Jackson EK (1991) Adenosine: a physiological brake on renin release. Annu Rev Pharmacol Toxicol 31: 1–35.[CrossRef][Medline]

Jackson EK (1997) Renal actions of purines, in Purinergic Approaches in Experimental Therapeutics (Jacobson KA and Jarvis MF eds) pp 217–250, Wiley-Liss, New York.

Jackson EK (2001) P1 and P2 receptors in the renal system, in Handbook of Experimental Pharmacology Volume 151/II; Purinergic and Pyrimidinergic Signalling II: Cardiovascular, Respiratory, Immune, Metabolic and Gastrointestinal Tract Function (Abbracchio MP and Williams M eds) pp 33–71, Springer-Verlag, Berlin.

Jackson EK and Dubey RK (2001) Role of the extracellular cAMP-adenosine pathway in renal physiology. Am J Physiol 281: F597–F612.

Jackson EK and Mi Z (2000) Preglomerular microcirculation expresses the cAMP-adenosine pathway. J Pharmacol Exp Ther 295: 23–28.[Abstract/Free Full Text]

Jackson EK, Mi Z, Gillespie DG, and Dubey RK (1997) Metabolism of cAMP to adenosine in the renal vasculature. J Pharmacol Exp Ther 283: 177–182.[Abstract/Free Full Text]

Jackson EK, Mi Z, Koehler MT, Carcillo JA Jr, and Herzer WA (1996) Injured erythrocytes release adenosine deaminase into the circulation. J Pharmacol Exp Ther 279: 1250–1260.[Abstract/Free Full Text]

Jackson EK, Mi Z, Zhu C, and Dubey RK (2003) Adenosine biosynthesis in the collecting duct. J Pharmacol Exp Ther 307: 888–896.[Abstract/Free Full Text]

Jackson EK, Zhu C, and Tofovic SP (2002) Expression of adenosine receptors in the preglomerular microcirculation. Am J Physiol 283: F41–F51.

Joyner WL, Mohama RE, Myers TO, and Gilmore JP (1988) The selective response to adenosine of renal microvessels from hamster explants. Microvasc Res 35: 122–131.[CrossRef][Medline]

Kim SM, Mizel D, Huang YG, Briggs JP, and Schnermann J (2005) Adenosine as a mediator of macula densa-dependent inhibition of renin secretion. Am J Physiol DOI: 10.1152/ajprenal.00367.

Kinsella JL, Holohan PD, Pessah NI, and Ross CR (1979) Isolation of luminal and antiluminal membranes from dog kidney cortex. Biochim Biophys Acta 552: 468–477.[Medline]

Kuster J, Zapf J, and Jakob A (1973) Effects of hormones on cyclic AMP release in perfused rat livers. FEBS Lett 32: 73–77.[CrossRef][Medline]

Mi Z and Jackson EK (1995) Metabolism of exogenous cyclic AMP to adenosine in the rat kidney. J Pharmacol Exp Ther 273: 728–733.[Abstract/Free Full Text]

Nishiyama A, Inscho EW, and Navar LG (2001) Interactions of adenosine A₁ and A_2A receptors on renal microvascular reactivity. Am J Physiol 280: F406–F414.

Schweda F, Segerer F, Castrop H, Schnermann J, and Kurtz A (2005) Blood pressure-dependent inhibition of renin secretion requires A₁ adenosine receptors. Hypertension 46: 780–786.[Abstract/Free Full Text]

Sekine T, Watanabe N, Hosoyamada M, Kanai Y, and Endou H (1997) Expression cloning and characterization of a novel multispecific organic anion transporter. J Biol Chem 272: 18526–18529.[Abstract/Free Full Text]

Sun D, Samuelson LC, Yang T, Huang Y, Paliege A, Saunders T, Briggs J, and Schnermann J (2001) Mediation of tubuloglomerular feedback by adenosine: evidence from mice lacking adenosine 1 receptors. Proc Natl Acad Sci USA 98: 9983–9988.[Abstract/Free Full Text]

Takahashi T, Kakuno K, Yamada H, and Endou H (1989) Intranephron distribution of purine-metabolizing enzymes in rats. Renal Physiol Biochem 12: 287–294.[Medline]

Takeda M, Yoshitomi K, and Imai M (1993) Regulation of Na⁺-3HCO-3 cotransport in rabbit proximal convoluted tubule via adenosine A₁ receptor. Am J Physiol 265: F511–F519.[Medline]

Tofovic SP, Branch KR, Oliver RD, Magee WD, and Jackson EK (1991) Caffeine potentiates vasodilator-induced renin release. J Pharmacol Exp Ther 256: 850–860.[Abstract/Free Full Text]

van Aubel RA, Smeets PH, Peters JG, Bindels RJ, and Russel FG (2002) The MRP4/ABCC4 gene encodes a novel apical organic anion transporter in human kidney proximal tubules: putative efflux pump for urinary cAMP and cGMP. JAm Soc Nephrol 13: 595–603.[Abstract/Free Full Text]

Vekaria RM, Shirley DG, Sevigny J, and Unwin RJ (2006) Immunolocalization of ectonucleotidases along the rat nephron. Am J Physiol 290: F550–F560.

Zimmermann H (1992) 5'-Nucleotidase: molecular structure and functional aspects. Biochem J 285 (Pt 2): 345–365.[Medline]

This article has been cited by other articles:

				J. Ren, Z. Mi, N. A. Stewart, and E. K. Jackson Identification and Quantification of 2',3'-cAMP Release by the Kidney J. Pharmacol. Exp. Ther., March 1, 2009; 328(3): 855 - 865. [Abstract] [Full Text] [PDF]

				J. Ren, Z. Mi, and E. K. Jackson Assessment of Nerve Stimulation-Induced Release of Purines from Mouse Kidneys by Tandem Mass Spectrometry J. Pharmacol. Exp. Ther., June 1, 2008; 325(3): 920 - 926. [Abstract] [Full Text] [PDF]

				E. K. Jackson and Z. Mi Regulation of Renal Ectophosphodiesterase by Protein Kinase C and Sodium Diet J. Pharmacol. Exp. Ther., April 1, 2008; 325(1): 210 - 216. [Abstract] [Full Text] [PDF]

				E. K. Jackson, J. Ren, L. C. Zacharia, and Z. Mi Characterization of Renal Ecto-Phosphodiesterase J. Pharmacol. Exp. Ther., May 1, 2007; 321(2): 810 - 815. [Abstract] [Full Text] [PDF]

				E. K. Jackson, Z. Mi, L. C. Zacharia, S. P. Tofovic, and R. K. Dubey The Pancreatohepatorenal cAMP-Adenosine Mechanism J. Pharmacol. Exp. Ther., May 1, 2007; 321(2): 799 - 809. [Abstract] [Full Text] [PDF]

				H. Funakoshi, L. C. Zacharia, Z. Tang, J. Zhang, L. L. Lee, J. C. Good, D. E. Herrmann, Y. Higuchi, W. J. Koch, E. K. Jackson, et al. A1 Adenosine Receptor Upregulation Accompanies Decreasing Myocardial Adenosine Levels in Mice With Left Ventricular Dysfunction Circulation, May 1, 2007; 115(17): 2307 - 2315. [Abstract] [Full Text] [PDF]

				E. K. Jackson, Z. Mi, and R. K. Dubey The Extracellular cAMP-Adenosine Pathway Significantly Contributes to the in Vivo Production of Adenosine J. Pharmacol. Exp. Ther., January 1, 2007; 320(1): 117 - 123. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.106.101360v1 317/3/1219    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jackson, E. K. Articles by Dubey, R. K. Search for Related Content PubMed PubMed Citation Articles by Jackson, E. K. Articles by Dubey, R. K.