Interaction of the metal chelator DMPS with OAT1 and OAT3 in intact isolated rabbit renal proximal tubules

【摘要】  2,3-Dimercapto-1-propanesulfonic acid (DMPS) is used clinically to increase urinary excretion of heavy metals, including mercury and arsenic. We used single S2 segments and suspensions of rabbit renal proximal tubules (RPT) to test the interaction of this anionic heavy metal chelator with basolateral transporters OAT1 and OAT3. RTPCR revealed expression of both transporters in single S2 segments. [ 3 H]PAH and 3 H-labeled estrone sulfate ([ 3 H]ES) were used as specific substrates for rbOAT1 and rbOAT3, respectively. PAH and ES were transported into nonperfused single RPT segments with K t values of 67 ± 20 and 3.4 ± 1.2 µM, respectively, and into tubule suspensions with K t values of 58 ± 17 and 7.7 ± 2.1 µM, respectively. Reduced DMPS (DMPSH) inhibited uptake of both substrates into single tubule segments with K app values of 405 ± 49 µM (for [ 3 H]PAH) and 320 ± 66 µM (for [ 3 H]ES). Oxidized DMPS (DMPSS), the prevalent form in the blood, also inhibited uptakes of [ 3 H]PAH ( K app of 766 ± 190 µM) and [ 3 H]ES (696 ± 166 µM). Inward gradients of ES, DMPSH, and DMPSS trans -stimulated the 30-s efflux of preloaded [ 3 H]ES across the basolateral membrane of RPT. Similarly, DMPSH, and PAH itself, trans -stimulated the 15-s efflux of [ 3 H]PAH. In contrast, efflux of [ 3 H]PAH was inhibited by the presence of DMPSS in the bathing medium. These data suggest that, whereas both OAT1 and OAT3 probably transport DMPSH, DMPSS transport may be limited to OAT3. This is the first evidence showing that both OAT1 and OAT3 can transport DMPS across the basolateral membrane of RPT.

【关键词】  organic anion transport p aminohippurate estrone sulfate kidney

THE COMMON ENVIRONMENTAL POLLUTANTS arsenic, lead, and cadmium are known to induce chronic renal disease. Nephrotoxicity of these metals is due to the fact that urinary elimination is a main route of excretion, and renal proximal tubules (RPTs) are especially sensitive because of their reabsorptive activity ( 12, 13, 35, 38 ). Because mercuric ions have an especially high affinity for free sulfhydryl groups on biological molecules, the concentration and metabolism of intracellular and extracellular thiols seem to play an important role in the disposition and toxicity of mercury in various organs, including the kidney ( 34, 36 ). Because of its preferential accumulation of inorganic mercury, the RPT is also the critical site of action for various antidotes of mercury poisoning ( 35, 37 ).

2,3-Dimercapto-1-propanesulfonic acid (DMPS) is an acid molecule with two free sulfhydryl groups that form complexes with many heavy metals, including mercury. DMPS is effective at "clearing" the kidneys of mercury and improves kidney function in patients who have been exposed to heavy metals. However, the mechanisms by which DMPS reduces the renal tubular burden of mercury are still unclear. DMPS entry into renal cells appears to involve interaction with organic anion transport processes. For example, in isolated rat kidneys, net secretion of DMPS is inhibited by PAH and probenecid ( 20 ), which are substrate and inhibitor, respectively, of the "classic" renal organic anion secretory pathway ( 27 ). Recent studies found that DMPS interacts with the human ( 19 ), rabbit ( 3 ), and rat ( 21 ) orthologs of organic anion transporter OAT1. As a demonstrated organic anion/dicarboxylate exchanger (e.g., 32 ), OAT1 displays the "physiological fingerprint" long associated with the active, basolateral transport step of renal organic anion secretion ( 26 ). Indeed, the characteristics of the DMPS interaction with OAT1 were shown to be present in isolated S2 segments of rabbit RPT, including the observation that DMPS can trans -stimulate the efflux of fluorescein, a model substrate for OAT1, from RPT cells ( 3 ).

Because organic anion transport in intact proximal tubules reflects the net activity of all associated transport processes, the focus on OAT1 as the predominant element in active secretion of organic anions may be premature. Three related OATs (OAT2, OAT3, and OAT4) have been found in the kidney ( 29 ). Of these, OAT2 and OAT3, in addition to OAT1, are expressed in the basolateral membrane of proximal tubule cells ( 6 ), although in the rat, OAT2 is found within the apical membrane of the tubules of medullary thick ascending limb of Henle's loop and medullary collecting ducts ( 15, 22, 25 ); OAT4 expression appears to be restricted to the apical membrane of proximal tubule cells ( 2 ). The physiological importance of OAT2 is not yet clear, but, in the human kidney, the very low level of OAT2 mRNA expression (compared with that of OAT1 and OAT3; 25) suggests that this transporter may play a comparatively modest role in the secretion of the broad array of substrates that typifies the classic organic anion transport process. As noted above, the ability of OAT1 to mediate organic anion/dicarboxylate exchange has been used to support the contention that OAT1 may be the major basolateral transporter in active organic anion secretion (e.g., 29 ). However, a recent study by Sweet et al. ( 30 ) demonstrated that OAT3 also acts as an organic anion/dicarboxylate exchanger to mediate organic anion transport. Thus energetically distinguishing OAT1 from OAT3 in organic anion transport may be difficult. We recently demonstrated that, in the rabbit kidney, OAT1 and OAT3 are equally involved in the basolateral accumulation of the mycotoxin ochratoxin A by cortical proximal tubules (Zhang X, Bahn A, Groves CE, and Wright SH, unpublished observations). Similarly, Sweet et al. ( 31 ) showed that renal slices from OAT3 knockout mice showed no mediated uptake of estrone sulfate (ES) or taurocholate and 50% reduction of mediated PAH uptake, thereby implicating OAT3 as a quantitatively significant element in the secretion of the broad array of substrates that previously appeared to be the bastion of OAT1.

In the present study, we characterized the interaction of DMPS with two physiologically distinct basolateral organic anion transport processes as expressed in intact isolated rabbit RPT. We took advantage of the observation that the rabbit orthologs of OAT1 and OAT3 (unlike their counterparts in humans, rats, and mice) clearly distinguish between PAH and ES; whereas PAH is effectively restricted to OAT1, ES is restricted to OAT3 (Zhang X, Bahn A, Groves CE, and Wright SH, unpublished observations). In single isolated S2 segments of rabbit RPT, we found evidence for the coexpression of two distinct organic anion transport pathways. The first process supported PAH transport and was comparatively insensitive to ES, and this process is suggested to reflect activity of the rabbit ortholog of OAT1. The second supported ES transport and was virtually insensitive to PAH. This process is suggested to reflect OAT3 activity. Reduced (DMPSH) and oxidized DMPS (DMPSS) were found to inhibit both PAH and ES transport. In addition, based on the trans effects of extracellular substrates, we conclude that DMPSH is a transported substrate of both OAT1 and OAT3. However, transport of DMPSS may be restricted to OAT3.

METHODS

Materials

[ 3 H]PAH (4.54 Ci/mmol) and [ 3 H]ES (43.5 Ci/mmol) were purchased from PerkinElmer Life Science Products (Boston, MA). DMPS and 5,5-dithiobis(2-nitrobenzoic acid) (Ellman's reagent) were purchased from Sigma (St. Louis, MO). RT-PCR reagents were purchased from Roche Molecular System (Alameda, CA). All other chemicals were purchased from standard sources and were generally the highest purity available.

Preparation of Isolated Tubules

New Zealand White rabbits (1.5 kg, Harlan, Indianapolis, IN) were killed by intravenous injection of pentobarbital sodium. The kidneys were flushed via the renal artery with an ice-chilled HEPES-sucrose buffer containing 250 mM sucrose and 10 mM HEPES, adjusted to pH 7.4 with Tris base, bubbled with 100% O 2 before use. They were then gently removed and sliced transversely using a single-edge razor. A kidney slice was transferred to the lid of a plastic petri dish on ice, which contained the standard solution used for dissecting and bathing tubules (in mM: 110 NaCl, 25 NaHCO 3, 5 KCl, 2 NaH 2 PO 4, 1 MgSO 4, 1.8 CaCl 2, 10 Na-acetate, 8.3 D -glucose, 5 L -alanine, 0.9 glycine, 1.5 lactate, 1 malate, and 1 sodium citrate). This standard solution was aerated continuously with 95% O 2 -5% CO 2 to maintain the pH at 7.4. The osmolality of the solutions averaged 290 mosmol/kgH 2 O.

S2 segments of proximal tubules were individually dissected from the cortical zone without the aid of enzymatic agents as described by others ( 5 ). All dissections were performed at 4°C, whereas all experiments were performed at 37°C. Transport rates were normalized to tubule surface area based on tubule lengths and diameters as determined from photomicrographs taken of each tubule before the experiment.

Preparation of Suspensions of Rabbit RPT

Suspensions of rabbit RPT were isolated and purified from New Zealand White rabbits (1.5 kg) as described elsewhere ( 17, 28 ). The final tubule pellet was resuspended at a protein concentration of 2 mg/ml in the medium used for dissection of single S2 segments. Tubular protein was measured using a Bio-Rad protein assay (Hercules, CA) with a -globulin standard.

Measurement of Transport in Nonperfused Single Isolated RPT

These experiments were performed in a manner similar to that used previously ( 7, 8, 16 ). Briefly, tubules were maintained oxygenated (95% O 2 -5% CO 2 ), transferred to oil-covered wells in a temperature-controlled chamber containing bicarbonate-buffer solution at 4°C to prevent evaporation until the start of each experiment, and photographed through a dissecting microscope equipped with a digital image capture system (Snappy, Play). Five minutes before the experiment, the bathing medium was warmed to 37°C. The tubules were then individually transferred to the oil-covered incubation medium at 37°C containing labeled substrate and appropriate test agents. After 15 s-5 min, uptake was stopped by transferring each tubule into 1 N NaOH for extraction. Accumulated labeled substrate was determined by liquid scintillation counting. Control and experimental uptakes were determined alternately and sequentially in tubules from the same kidney.

Measurement of Transport in Suspensions of Rabbit RPT

Tubule suspensions (2 mg/ml) were preincubated in Erlenmeyer flasks for 15 min at 37°C and gassed with 95% O 2 -5% CO 2. An aliquot of tubule suspension (0.5 ml) was transferred to a 15-ml tube containing 0.5 ml of incubation medium with 0.13 µM [ 3 H]PAH or 5 nM [ 3 H]ES and varying concentrations of unlabeled PAH or ES. After 30 s, 5 ml of ice-cold DMEM-F-12 medium (Sigma) were added to stop the uptake, and the tubules were pelleted. The rinse was repeated, the final pellet was dissolved in 0.5 N NaOH/1% SDS, and aliquots were taken for counting radioactivity.

Determination of PAH and ES Effluxes from Cells Across the Basolateral Membrane of Intact Nonperfused Tubules

The effluxes of [ 3 H]PAH and [ 3 H]ES from the cells across the basolateral membrane under various conditions were determined as described in detail previously ( 11 ) with modifications. Briefly, S2 segments of tubules were incubated at 37°C in standard solution containing 5 µM [ 3 H]PAH or 0.15 µM [ 3 H]ES for 15 min so that a steady state was obtained. At the end of the incubation period, each tubule was briefly dipped in a rinse bath of buffer only and then transferred through a 0.2-ml efflux bath for 15 or 30 s to determine PAH or ES effluxes, respectively, from the cells across the peritubular membrane. Standard solution was used as the control efflux bath, and an identical solution with the addition of test agents was used as the experimental efflux bath. At the end of experiment, the tubules were extracted in 1 N NaOH for at least 30 min to measure the concentration of residual [ 3 H]PAH or [ 3 H]ES in the cell, and each efflux bath medium was counted to measure the efflux of [ 3 H]PAH or [ 3 H]ES. The intracellular concentration of radioactivity in each tubule at the beginning of the efflux period was calculated from the effluxed and residual radioactivity. The efflux was standardized by dividing the amount of radioactivity in the efflux bath by the total intracellular concentration at that time to give an efflux coefficient (cm/s) ( 9 ). In each experiment, tubules from the same rabbit kidney were used for efflux into both control and experimental media. The incubations for all tubules were carried out simultaneously. All experiments were performed at 37°C.

Handling of DMPS

Solutions with DMPSH were prepared immediately before experiments, and oxidation of the free sulfhydryls within DMPS occurred at a rate of <6%/60 min (the typical time course of a transport experiment) under the conditions used in these studies. Solutions of DMPSS were obtained by bubbling a DMPS-containing solution in 2 mM phosphate buffer (Na 2 HPO 4 ), adjusted to pH 9.0 with Na 3 PO 4 ·2 H 2 O, with 100% O 2 at room temperature for at least 24 h. Using Ellman's reagent (see below), we detected no free thiol groups remaining after this period. Therefore, 90% of the material was oxidized.

Measurement of Free Thiol Groups

Ellman's reagent was used to determine free thiol groups ( 14 ). Briefly, Ellman's reagent (in 1 mM phosphate buffer, pH 7.4) and the DMPS solution were combined, and the absorption was measured at 412 nm after the color had developed completely. Concentrations were calculated by comparison with standards containing known concentrations of reduced glutathione.

RT-PCR Protocol Using Grouped Kidney Segments

Tubule preparation and digestion. The kidney was perfused via the renal artery with cold sterile HEPES-sucrose buffer, pH 7.4. The kidney slice was gently removed and dissected in HEPES-sucrose buffer on ice using sterile conditions. Twelve to fifteen S2 segments were used for each RT reaction. A 2-µl aliquot of tubules suspended in HEPES-sucrose buffer was transferred into a microfuge tube containing 4 µl of 2% Triton mix (composed of 44.5 µl sterile water, 2 µl RNase inhibitor, 2.5 µl 0.1 M DTT, and 1.2 µl sterile Triton X-100). The digested tubules were kept at room temperature for 5 min and then quick-frozen in liquid nitrogen.

RT. Oligo(dT) 12-18 primers, random primer (3 µg/ml), 10 mM dNTP mix, and sterile water were added to the digested tubules. After heating at 65°C for 5 min, 5 x first-strand buffer, 0.1 M DTT, and RNase inhibitor were added to each tube (RT and negative RT samples) and incubated at 42°C for 2 min. Superscript II (SSII; for RT) or sterile water (for negative RT) was mixed in each tube and incubated at 42°C for 50 min. The reaction was then inactivated by heating at 70°C for 15 min. RNase H was added to each tube, and the samples were incubated at 37°C for 20 min to remove RNA complementary to the cDNA. The resulting cDNA was used as a template for amplification via PCR.

PCR. cDNA from RT product or negative RT product was added to sterile water, (10 x ) PCR buffer, 50 mM MgCl 2, 10 mM dNTP mix, and Taq DNA Polymerase. For amplification of cDNA, the forward and reverse primers for OAT1 or OAT3 were added to tubes containing either RT or negative RT products. Plasmid cDNA for OAT1 and OAT3 was also run for PCR to use as positive control for OAT1 and OAT3 amplification. Each tube was run on the thermal cycler by using a standard PCR program (94°C for 3 min followed by 34 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 3 min, final elongation for 10 min at 72°C, and held at 4°C). Amplified products were visualized with ethidium bromide on agarose gels.

Statistical Analysis

Data are summarized as means ± SE. Tubules from different rabbits were used for each experiment. The level of significance for differences between means was determined with Student's t -test for paired values. In all analyses, differences were considered statistically significant when P < 0.05.

RESULTS

Time Course of ES Uptake in Single Nonperfused S2 Segments of Rabbit RPT

The uptake of [ 3 H]ES by isolated individual S2 segments of RPT increased with time and approached a steady state after 5 min ( Fig. 1 ). Sixty-second incubations of single tubules with ES were chosen for subsequent kinetic analyses, as this time point was the earliest point at which satisfactory radioactive counting could be obtained over the entire range of substrate concentrations used.

Fig. 1. Time course of 3 H-labeled estrone sulfate (ES) uptake into isolated single S2 segments of rabbit renal proximal tubule (RPT). Each point represents the mean ± SE of triplicate measurements from 4 separate experiments.

Kinetics of ES Uptake Across the Basolateral Membrane of Single Nonperfused S2 Segments and Suspensions of Rabbit RPT

The kinetics of peritubular ES uptake were examined to evaluate the physiological characteristics of the transport of this substrate by rabbit RPT. Figure 2 A shows the kinetic profile of ES transport in single nonperfused RPT S2 segments. Increasing concentrations of unlabeled ES progressively inhibited the tubular accumulation [ 3 H]ES. The kinetics of ES uptake were adequately described by the following form of the Michaelis-Menten equation ( 24 )

where J is the rate of [ 3 H]ES transport from a concentration of labeled substrate equal to [ * T]; J max is the maximum rate of mediated ES transport; K t is the ES concentration that results in half-maximal transport (Michaelis constant); is the concentration of unlabeled ES in the transport reaction; C is a constant that represents the component of total ES uptake that is not saturated (over the range of substrate concentrations tested) and presumably reflects the combined influence of diffusive flux, nonspecific binding and/or incomplete rinsing of the tubules. In five separate experiments, the J max for ES uptake was 3.9 ± 1.8 pmol·min -1 ·mm -2, with a K t of 3.4 ± 1.2 µM. The kinetics of peritubular ES uptake were also examined in suspensions of rabbit RPT to evaluate the physiological characteristics of ES transport in native tubules comprising a population primarily of both S1 and S2 segments ( Fig. 2 B ). For kinetic analysis in suspensions, 30-s uptakes were measured because this time was the shortest point at which adequate radioactivity could be accumulated over the range of substrate concentrations employed. As with single tubules, unlabeled ES reduced the peritubular uptake of [ 3 H]ES in RPT suspensions as described by the kinetics of competitive inhibition. Kinetic analysis of ES uptake in RPT suspensions ( n = 5) generated a J max of 1.2 ± 0.6 nmol·mg -1 ·min -1 and a K t of 7.4 ± 1.1 µM. Interestingly, converting the J max for single S2 segments to uptake per milligram dry weight (0.6 ± 0.3 nmol·mg -1 ·min -1 ) reveals that maximal transport of ES in single tubules is similar to the value measured in RPT suspensions.

Fig. 2. A : kinetics of [ 3 H]ES uptake into isolated single S2 segments of rabbit RPT. Uptakes were measured for 1 min. Each point represents the mean ± SE of triplicate measurements from 5 separate experiments. B : kinetics of [ 3 H]ES uptake into isolated suspended rabbit RPT. Uptakes were measured at 30 s. The reported maximum rate of mediated ES transport ( J max ) and ES concentration that results in half-maximal transport ( K t ) represents the mean of duplicate measurements at each concentration for 5 separate experiments. However, each point shown represents the mean ± SE for duplicate measurements from a single representative experiment to yield a J max of 0.89 nmol·mg -1 ·min -1 and K t of 7.5 µM. The solid lines fit to the data were calculated and kinetic parameters were derived with a nonlinear regression algorithm (Sigma Plot, Jandel Scientific).

Kinetics of PAH Uptake Across the Basolateral Membrane of Isolated RPT S2 Segments and Suspensions of Rabbit RPT

For kinetic analyses, incubations for PAH uptake were performed at 15 and 30 s for single tubules and tubule suspensions, respectively (as used in our previous studies) ( 10 ). Increasing concentrations of unlabeled PAH progressively reduced the uptake of [ 3 H]PAH ( Fig. 3, A and B ), a relationship that was described by the kinetics of competitive inhibition using the above equation. The J max for PAH uptake into nonperfused single S2 segments was 47.2 ± 12.7 pmol·min -1 ·mm -2 ( n = 5) (or 7.7 ± 2.1 nmol·mg -1 ·min -1 ), whereas the J max for PAH uptake by tubule suspensions, which represents PAH accumulation by S1 and S2 segments, was approximately sixfold lower (1.4 ± 0.5 nmol·mg -1 ·min -1 ). In contrast, the K t for PAH uptake by S2 segments ( K t = 66.9 ± 20.5 µM) was similar to that measured for tubule suspensions ( K t = 57.6 ± 17 µM; n = 5). The affinities for PAH uptake in both single S2 segments and tubule suspensions are also similar to those reported previously by our laboratory ( 11 ). As emphasized below, this uptake probably reflected activity of OAT1.

Fig. 3. A : kinetics of [ 3 H]PAH uptake into isolated single S2 segments of rabbit RPT. Uptakes were measured at 15 s. Each point represents the mean ± SE of triplicate measurements from 5 separate experiments. B : kinetics of [ 3 H]PAH uptake into isolated suspended rabbit RPT. Uptakes were measured at 30 s. The reported J max and K t represent the mean of duplicate measurements at each concentration for 5 separate experiments. However, each point shown represents the mean ± SE for duplicate measurements from a single representative experiment to yield a J max of 0.755 nmol·mg -1 ·min -1 and K t of 51.5 µM. The sold lines fit to the data were calculated and kinetic parameters were derived with a nonlinear regression algorithm (Sigma Plot).

Reciprocal Inhibitory Effects of PAH and ES

ES has been shown to interact selectively with OAT3, over OAT1, (e.g., 31 ). Figure 4 A indicates that, in single rabbit RPT, ES has a comparatively weak inhibitory interaction with basolateral PAH uptake. The concentration of ES that blocked 50% of basolateral PAH uptake (IC 50 ) was calculated using the relationship ( 39 )

in which J is the rate of uptake of labeled substrate; J 0 is the rate of labeled substrate uptake in the absence of inhibitor; is the concentration of inhibitor; and n is the Hill coefficient. The IC 50 of this interaction, determined in six experiments with different rabbits, was 91.1 ± 33.2 µM in single S2 segments. This value is 20-fold larger than the K t for basolateral ES uptake noted above. In suspensions of RPT, the IC 50 (171 ± 56 µM) for ES inhibition of PAH uptake is similar to that measured in S2 segments. Consequently, the inhibition of PAH transport by ES must reflect an interaction of ES with a pathway distinct from the one that mediates ES uptake. In other words, basolateral PAH uptake in rabbit RPT probably reflects a relatively exclusive interaction with OAT1, with little or no interaction with the OAT3 transporters that supported high-affinity ES transport. Indeed, in experiments with suspensions of RPT, the comparatively weak inhibition of PAH uptake produced by ES was not competitive but, instead, displayed a "mixed-type" profile (i.e., influencing both J max and apparent K t; data not shown).

Fig. 4. A : kinetics of ES inhibition of [ 3 H]PAH uptake into isolated single S2 segments of rabbit RPT. Increasing concentrations of unlabeled ES inhibited the 15-s uptake of [ 3 H]PAH. Each point represents the mean ± SE of triplicate measurements from 6 separate experiments. B : kinetics of PAH inhibition of [ 3 H]ES uptake into isolated single S2 segments of rabbit RPT. Increasing concentrations of unlabeled PAH inhibited the 1-min uptake of [ 3 H]ES. Each point represents the mean ± SE of triplicate measurements from 4 separate experiments. The solid lines fit to the data were derived from the kinetic parameters determined in separate experiments calculated using Eq. 2 and a nonlinear regression algorithm (Sigma Plot).

When expressed in heterologous expression systems, the rabbit orthologs of OAT1 and OAT3 display marked preferences for PAH and ES, respectively (Zhang X, Bahn A, Groves CE, and Wright SH, unpublished observations), results consistent with the present observations on the reciprocal inhibitory interactions of PAH and ES with basolateral transport of these compounds ( Fig. 4 B ). In four experiments with S2 segments, PAH inhibited [ 3 H]ES uptake with an IC 50 of 3 mM, which was 50-fold higher than the K t for PAH uptake into isolated S2 segments of rabbit RPT. Thus the ES transport process inhibited by PAH is not likely mediated by OAT1. Although the observed inhibition of ES transport by PAH could reflect an interaction with OAT3, this inhibition did not appear to reflect strict competition; the highest concentrations 50-60% of total [ 3 H]ES uptake and, as noted above, that interaction was not with OAT1. It may well be that large concentrations of PAH exert an indirect effect(s) on ES transport.

Interaction of DMPS with Basolateral PAH and ES Transport in Rabbit RPT

DMPS can exist in either the DMPSH or DMPSS form, with the latter comprising some 80% of the DMPS in the blood within 30 min of its introduction to the circulatory system ( 23 ). Figure 5 shows the effect of DMPSH and DMPSS on the uptake of [ 3 H]PAH into single nonperfused RPT. Evidence discussed below showed that DMPSH served as a substrate of OAT1. Consequently, the inhibitory interaction shown in Fig. 5 probably reflected competition between DMPSH and PAH for a common binding site, a relationship described by rearranging Eq. 1 ( 16 )

where J app is defined as ( K i / K t ) J max and K app is an apparent K i for the test agent that is defined as K i (1 + [T * ]/ K t ). When [T * ] is << K t, K app K i. The K app for inhibition of PAH uptake by DMPSH and DMPSS was 406 ± 49.2 and 766 ± 190 µM, respectively ( Fig. 5, A and B ). Calculation of the K app for DMPSS was complicated by the fact that in saline solution 90% of the DMPSS in our transport buffers was in a cyclic dimeric form ( 19 ), the test solutions presumably included both cis and trans configurations of the dimer. As a result, the molecular form (and therefore molar concentration) of DMPSS in our solutions was not precisely defined. In view of this caveat, the concentration of DMPSS was expressed as the concentration of DMPSH present before oxidation ("DMPSH equivalent concentration"). While there was ambiguity concerning the exact molar concentration of oxidized DMPS in our solutions, DMPSS was clearly an effective inhibitor of [ 3 H]PAH uptake in single nonperfused RPT ( Fig. 5 B ). These data demonstrated that both forms of DMPS interact with OAT1 in native rabbit RPT.

Fig. 5. A : kinetics of reduced 2,3-dimercapto-1-propanesulfonic acid [DMPS (DMPSH)] inhibition of [ 3 H]PAH uptake into isolated single S2 segments of rabbit RPT. Increasing concentrations of DMPSH inhibited the 15-s uptake of [ 3 H]PAH. Each point represents the mean ± SE of triplicate measurements from 8 separate experiments. B : kinetics of oxidized DMPS (DMPSS) inhibition of [ 3 H]PAH uptake into isolated single S2 segments of rabbit RPT. Increasing concentrations of DMPSS inhibited the 15-s uptake of [ 3 H]PAH. Each point represents the mean ± SE of triplicate measurements from 6 separate experiments. The solid lines fit to the data were calculated and kinetic parameters were derived with a nonlinear regression algorithm (Sigma Plot).

As shown earlier, ES transport in rabbit RPT is likely to be largely restricted to OAT3. Figure 6 shows that DMPSH and DMPSS interact with OAT3, as shown by their inhibition of ES transport into rabbit RPT. In seven separate experiments using S2 segments, the K app for inhibition of PAH uptake by DMPSH was 321 ± 66.5 µM ( Fig. 6 A ). For DMPSS, the K app was 696 ± 167 µM ( n = 6; Fig. 6 B ).

Fig. 6. A : kinetics of DMPSH inhibition of [ 3 H]ES uptake into isolated single S2 segments of rabbit RPT. Increasing concentrations of DMPSH inhibited the 1-min uptake of [ 3 H]ES. Each point represents the mean ± SE of triplicate measurements from 7 separate experiments. B : kinetics of DMPSS inhibition of [ 3 H]ES uptake into isolated single S2 segments of rabbit RPT. Increasing concentrations of DMPSS inhibited the 1-min uptake of [ 3 H]ES. Each point represents the mean ± SE of triplicate measurements from 6 separate experiments. The sold lines fit to the data were calculated and kinetic parameters were derived with a nonlinear regression algorithm (Sigma Plot).

Effects of Unlabeled PAH and DMPSH and DMPSS on Efflux of [ 3 H]PAH and [ 3 H]ES Across the Basolateral Membrane of Isolated Tubules

Despite the observed inhibitory interaction between DMPS and the transporters for PAH and ES in rabbit RPT, independent evidence was sought that the two species of DMPS serve as transported substrates for the two carriers. Isolated tubules were preloaded with radiolabeled substrate and then placed in a bath containing a comparatively large concentration of an unlabeled test compound to determine whether an inwardly directed trans gradient stimulated efflux of the preloaded labeled compound. Figure 7 shows the effect on the [ 3 H]PAH efflux coefficient caused by imposing inwardly directed gradients of unlabeled PAH (1 mM), unlabeled DMPSH (2.5 mM), and unlabeled DMPSS (2.5 mM). Consistent with previous observations ( 11 ), the inwardly directed gradient of PAH trans -stimulated PAH efflux, in the present case by 25%. The inwardly directed gradient of DMPSH also resulted in a small (8%) but significant ( P < 0.05) stimulation of PAH efflux. The gradient of DMPSS, however, did not stimulate PAH efflux. Instead, the inwardly directed DMPSS gradient decreased PAH efflux by 25% ( P < 0.05). Whereas these data are consistent with the conclusion that DMPSH is transported by OAT1, the oxidized form of the chelator was either not transported or turnover of the DMPSS-loaded exchanger occurred at a suffi-ciently low rate (compared with control) that PAH efflux was slowed.

Fig. 7. Effect of inwardly directed (i.e., trans ) gradients of PAH (1 mM), DMPSH (2.5 mM), or DMPSS (2.5 mM) on efflux of [ 3 H]PAH from isolated single S2 segments of rabbit RPT. S2 segments were preloaded with [ 3 H]PAH for 15 min before measurement of 15-s efflux. Data are expressed as means ± SE of triplicate measurements from 5 separate experiments. * P < 0.05 vs. control.

Figure 8 shows the effect of the inwardly directed gradients on the efflux of [ 3 H]ES from preloaded rabbit RPT. The gradient of unlabeled ES (250 µM) trans -stimulated [ 3 H]ES efflux by almost 70%, supporting the contention ( 30 ) that OAT3 can operate as an anion exchanger. Similarly, [ 3 H]ES efflux was significantly stimulated by inwardly directed gradients of DMPSH and DMPSS (50 and 42%, respectively). Thus OAT3 appears to support the transport of both species of DMPS.

Fig. 8. Effect of inwardly directed (i.e., trans ) gradients of ES (250 µM), DMPSH (2.5 mM), or DMPSS (2.5 mM) on efflux of [ 3 H]ES from isolated single S2 segments of rabbit RPT. S2 segments were preloaded with [ 3 H]ES for 15 min before measurement of 30-s efflux. Data are expressed as means ± SE of triplicate measurements from 5 separate experiments. * P < 0.05 vs. control.

Expression of OAT1 and OAT3 mRNA in S2 Segments of Rabbit RPT

RT-PCR was used to amplify sequences specific for the rabbit orthologs of OAT1 and OAT3 mRNA obtained from isolated S2 segments of rabbit RPT. In Figure 9, a representative gel of amplified OAT1 and OAT3 products shows that both orthologs were routinely found in isolated S2 segments. Amplified fragments were excised from a representative gel and sequenced, confirming the identity of the products. These data indicate that both of these transporters are expressed in the S2 segment and support the contention that the transport of PAH and ES reflects activity of OAT1 and OAT3, respectively.

Fig. 9. Detection of rabbit organic anion [rbOat1 ( left ) and rbOat3 ( right )] mRNA levels by RT-PCR in isolated single S2 segments of rabbit RPT. mRNA prepared from 12-15 tubule segments was used for first-strand cDNA synthesis. Subsequent PCR amplification was performed with rbOat1- or rbOat3-specific primers using the indicated amount of RT reaction. Negative controls consisted of product obtained from tubule reactions run without added reverse transcriptase (to test for the influence of genomic DNA). PCR products for the negative controls (-RT) and for rbOat1 and rbOat3 (RT) were loaded on gels and visualized with ethidium bromide. Left lanes, base pair ladder (ldr); right lanes, the amplified products obtained when the cDNA-containing plasmid for each transporter was used as a template (tmpl).

DISCUSSION

DMPS has been used clinically to treat intoxication by heavy metals, including mercury and arsenic ( 1 ). It reduces their accumulation in the kidney and increases their urinary excretion ( 1 ). The proximal tubule organic anion secretory pathway is apparently involved in this detoxification process. Experimental findings in an oocyte expression system and isolated S2 segments of rabbit RPT indicate that DMPS acts as an OAT1 substrate for transport into the cell across basolateral membrane of RPT ( 3, 19, 21, 37 ). Once inside the cells, DMPS can combine with the heavy metal, move across the luminal membrane, and be excreted in the urine. The results from the present study using isolated S2 segments of rabbit RPT confirmed that DMPS interacts with and is transported into the cells by OAT1. The data showed that the uptake of PAH, a model substrate for OAT1, was inhibited by DMPS in single S2 segments. The stimulatory effect of reduced DMPS on PAH efflux across the basolateral membrane of intact S2 segments indicated that reduced DMPS not only was bound to but also was transported by OAT1. These results are in close agreement with previous reports involving the human ortholog of OAT1 ( 19 ), the rabbit ortholog of OAT1, and isolated S2 segments of rabbit RPT ( 3 ).

We also used ES to examine the interaction of DMPS with and possible transport by OAT3 in renal tubule cells. ES was transported across the basolateral membrane of single S2 segments and suspensions of RPT (comprising S1 and S2 segments) with a K t of 3.44 and 7.7 µM, respectively. These values are extremely similar to the K t of 4.5 µM for ES uptake mediated by the rabbit ortholog of OAT3 when expressed in Chinese hamster ovary (CHO) cells (Zhang X, Bahn A, Groves CE, and Wright SH, unpublished observations). In addition, rabbit OAT3 has been shown to interact very weakly with PAH, with IC 50 1,000 µM when expressed in CHO or COS-7 cells (Zhang X, Bahn A, Groves CE, and Wright SH, unpublished observations). Thus basolateral uptake of ES in intact rabbit RPT probably reflects activity of OAT3, and not of OAT1, whereas uptake of PAH is probably indicative of OAT1 activity, and not of OAT3. The present results showed that DMPSH inhibited ES uptake into S2 segments of rabbit RPT. The K app values for inhibition of ES and PAH uptakes by DMPSH were quite similar (321 and 406 µM, respectively), indicating that rbOAT1 and rbOAT3 have comparable affinities for DMPS.

Because DMPS is rapidly oxidized within 30 min in saline and blood and the oxidized form exists to a significant extent in blood ( 18, 23 ), we also examined the interaction of DMPSS with the transport of PAH and ES in isolated rabbit RPT. The results showed that PAH uptake was inhibited by DMPSS, which implied that DMPSS could interact with OAT1. These data are consistent with observations from a previous study in human OAT1-expressing Xenopus laevis oocytes ( 19 ). The K app value for DMPSH inhibition of PAH uptake in S2 segments was about twofold lower than that of DMPSS (406 and 766 µM, respectively). A similar result was observed for DMPS inhibition of ES uptake ( K app of 321 and 696 µM for DMPSH and DMPSS, respectively). The difference in these K app values was similar to the relative difference observed for those values obtained in the study of human OAT1 in the oocyte expression system (22.4 µM for DMPSH and 66 µM for DMPSS) ( 19 ). In the present study, well over 90% of DMPS was oxidized after bubbling with 100% O 2 for at least 24 h. However, because a solution of DMPSS so prepared mainly consists of cyclic dimers and some trimers ( 19 ), the actual concentration of DMPSS was probably less than one-half of the starting concentration expressed in DMPS equivalents. Thus the K app value for DMPSS is almost certainly an underestimate of the affinity of this compound for the PAH uptake pathway(s) in intact RPT. This information suggests that OAT1 has a comparable affinity for DMPSH and DMPSS in single S2 segments.

It is important to note that these inhibitory effects do not demonstrate that DMPS is actually transported by either OAT1 or OAT3. Consequently, we examined the influence of trans -oriented DMPS gradients on the flux of PAH and ES as an indirect indicator of DMPS transport. Trans -effects, be they stimulatory or inhibitory, must be interpreted with caution; alternative explanations for such effects do exist (e.g., allosteric influences on transporter function). Nevertheless, the modest, but significant, trans -stimulatory effect of an inwardly directed gradient of reduced DMPS on PAH and ES efflux suggests that DMPSH is transported by OAT1 and OAT3 (Figs. 7 and 8 ). An inwardly directed gradient of oxidized DMPS, in contrast, only stimulated efflux of ES from preloaded tubules. Indeed, the 15-s efflux of PAH was inhibited when DMPSS was present in the bathing medium ( Fig. 7 ), implying that DMPSS interacts with but is not necessarily transported by rabbit OAT1 (or may bind to the transporter and slow down its rate of turnover). This observation, coupled with the fact that trans -stimulatory effects of reduced DMPS on ES efflux were routinely greater than those observed for PAH efflux, suggests that, although both OAT1 and OAT3 may transport DMPS into the cells of rabbit S2 segments of RPT, OAT3 may be the more important route for DMPS entry into RPT.

The apparent affinity of DMPS for PAH transporters in rabbit tubules was much lower than that measured in human OAT1-expressing X. laevis oocytes ( 19 ). It was also signifi-cantly lower than the K app measured for DMPSH inhibition of fluorescein (FL) uptake in COS-7 cells expressing rabbit OAT1 (102 µM) and nonperfused rabbit S2 segments (71 µM) ( 3 ). These differences may result from a number of factors, including differences in the techniques and substrates used in each experiment. The previous studies with COS-7 cells expressing rabbit OAT1 and with rabbit RPT S2 segments examined the transport of FL, a substrate for both rabbit OAT1 and OAT3 (unpublished observations), and measured the accumulation of the fluorescence of FL within the tubule cells. The measurement of FL transport may have given different results from the measurement of [ 3 H]PAH transport in the present study.

The transport of organic anions across basolateral membrane is the net activity of all associated transporters present. Transporters other than OAT1, such as OAT3 and the sodium-dicarboxylate transporter 3 (NaDC-3), are expressed in the basolateral membrane of the RPT, and these can transport organic anions as well. A recent study demonstrated that DMPS is translocated by flounder NaDC-3 expressed in oocytes ( 4 ). Thus DMPS appears to be able to enter renal cells by at least three distinct basolateral processes.

Like OAT1, OAT3 is expressed in the basolateral membrane of RPT in the human ( 6 ) and rat ( 22 ). In rats, distribution of OAT1 along the tubule is largely limited to the S2 segment ( 22, 33 ), whereas OAT3 shows immunoreactivity in all segments (S1 S2 = S3) ( 22 ). However, in human renal tissue, OAT1 expression is not restricted to the S2 segment; instead, it appears to occur in all segments of the proximal tubule ( 25 ). Thus the distribution of OAT1 may differ from species to species. In human renal tissues, OAT3 mRNA is the most highly expressed member of the organic anion transporter family. The present physiological data suggest that both OAT1 and OAT3 play significant roles in organic anion transport across the basolateral membrane of rabbit RPT. Moreover, the RT-PCR data on S2 segments of rabbit RPT confirm the existence of both OAT1 and OAT3 in these segments.

The maximum plasma concentration of DMPS in humans after a normal intravenous dose is on the order of 70 µM ( 23 ). Although this is 8-10 times lower than the apparent K i for interaction of DMPS (both DMPSH and DMPSS) with OAT1 and OAT3 (in intact rabbit tubules), it is appropriate to acknowledge that the plasma concentration of endogenous substrates, for which these transporters frequently display a higher affinity (e.g., endogenous hippurates, estrone sulfate), is unlikely to exceed the submicromolar level. Thus clinical concentrations of DMPS are not likely to interfere with the interaction of OAT1 and OAT3 with their physiological substrates, and it is equally unlikely that these physiological substrates will interfere with the interaction of DMPS with the secretory opportunities offered by these OATs.

In conclusion, the present data suggest that both OAT1 and OAT3 can translocate DMPSH across basolateral membrane into renal cells, at least in the S2 segment of the rabbit RPT. However, DMPSS, which is the predominant form in the blood, may be transported only by OAT3, and this may be the dominant transporter for DMPS in this tubule segment. This study provides the first evidence that both OAT1 and OAT3 can play roles in DMPS transport into the cells of intact renal tubules.

ACKNOWLEDGMENTS

The authors thank the Ministry of University Affairs of Thailand for support.

GRANTS

This work was supported in part by National Institutes of Health Grants DK-56224, ES-04940, and ES-06694.

【参考文献】
  Aposhian HV, Maiorino RM, Gonzalez-Ramirez D, Zuniga-Charles M, Xu Z, Hurlbut KM, Junco MP, Dart RC, and Aposhian MM. Moblization of heavy metals by newer, therapeutically useful chelating agents. Toxicology 97: 23-38, 1995.

Babu E, Takeda M, Narikawa S, Kobayashi Y, Enomoto A, Tojo A, Cha SH, Sekine T, Sakthisekaran D, and Endou H. Role of human organic anion transporter 4 in the transport of ochratoxin A. Biochim Biophys Acta 1590: 64-75, 2002.

Bahn A, Knabe M, Hagos Y, Rodiger M, Godehardt S, Graber-Neufeld DS, Evans KK, Burckhardt G, and Wright SH. Interaction of the metal chelator 2,3-dimercapto-1-propanesulfonate with the rabbit multispecific organic anion transporter 1 (rbOAT1). Mol Pharmacol 62: 1128-1136, 2002.

Burckhardt BC, Drinkuth B, Menzel C, Konig A, Steffgen J, Wright SH, and Burckhardt G. The renal Na + -dependent dicarboxylate transporter, NaDC-3, translocates dimethyl- and disulfhydryl-compounds and contributes to renal heavy metal detoxification. J Am Soc Nephrol 13: 2628-2638, 2002.

Burg M, Grantham J, Abramov M, and Orloff J. Preparation and study of fragments of single rabbit nephrons. Am J Physiol 210: 1293-1298, 1966.

Cha SH, Sekine T, Fukushima JI, Kanai Y, Kobayashi Y, Goya T, and Endou H. Identification and characterization of human organic anion transporter 3 expressing predominantly in the kidney. Mol Pharmacol 59: 1277-1286, 2001.

Chatsudthipong V and Dantzler WH. PAH/ -KG countertransport stimulates PAH uptake and net secretion in isolated rabbit renal tubules. Am J Physiol Renal Fluid Electrolyte Physiol 263: F384-F391, 1992.

Chatsudthipong V and Jutabha P. Effect of steviol on para-aminohippurate transport by isolated perfused rabbit renal proximal tubule. J Pharmacol Exp Ther 298: 1120-1127, 2001.

Dantzler WH and Brokl O. NMN transport by snake renal tubules: choline effects, countertransport, H + -NMN exchange. Am J Physiol Renal Fluid Electrolyte Physiol 253: F656-F663, 1987.

Dantzler WH, Evans KK, Groves CE, Welborn JR, North J, Stevens JL, and Wright SH. Relation of cysteine conjugate nephrotoxicity to transport by the basolateral organic anion transport system in isolated S2 segments of rabbit proximal renal tubules. J Pharmacol Exp Ther 286: 52-60, 1998.

Dantzler WH, Evans KK, and Wright SH. Kinetics of interactions of para-aminohippurate, probenecid, cysteine conjugates, and N -acetyl cysteine conjugates with basolateral organic anion transporter in isolated rabbit proximal renal tubules. J Pharmacol Exp Ther 272: 663-672, 1995.

Diamond GL and Zalups RK. Understanding renal toxicity of heavy metals. Toxicol Pathol 26: 92-103, 1998.

Dresser MJ, Leabman MK, and Giacomini KM. Transporters involved in the elimination of drugs in the kidney: organic anion transporters and organic cation transporters. J Pharm Sci 90: 397-421, 2001. <a href="/cgi/external_ref?access_num=10.1002/1520-6017(200104)90:4

Ellman GL. Tissue sulfhydryl groups. Arch Biochem Biophys 82: 70-77, 1959.

Enomoto A, Takeda M, Shimoda M, Narikawa S, Kobayashi Y, Kobayashi Y, Yamamoto T, Sekine T, Cha SH, Niwa T, and Endou H. Interaction of human organic anion transporters 2 and 4 with organic anion transport inhibitors. J Pharmacol Exp Ther 301: 797-802, 2002.

Groves CE, Evans K, Dantzler WH, and Wright SH. Peritubular organic cation transport in isolated rabbit proximal tubules. Am J Physiol Renal Fluid Electrolyte Physiol 266: F450-F458, 1994.

Groves CE and Schnellmann RG. Suspensions of rabbit renal proximal tubules. In: Methods in Renal Toxicology, edited by Zalups RK and Lash LH. Boca Raton, FL: CRC, 1996, p. 147-162.

Hurlbut KM, Maiorino RM, Mayersohn M, Dart RC, Bruce DC, and Aposhian HV. Determination and metabolism of dithiol chelating agents XVI: pharmacokinetics of 2,3-dimercapto-1-propanesulfonate after intravenous administration in human volunteers. J Pharmacol Exp Ther 268: 662-668, 1994.

Islinger F, Gekle M, and Wright SH. Interaction of 2,3-dimercapto-1-propane sulfonate with the human organic anion transporter hOAT1. J Pharmacol Exp Ther 299: 741-747, 2001.

Klotzbach JM and Diamond GL. Complexing activity and excretion of 2,3-dimercapto-1-propane sulfonate in rat kidney. Am J Physiol Renal Fluid Electrolyte Physiol 254: F871-F878, 1988.

Koh AS, Simmons-Willis TA, Pritchard JB, Grassl SM, and Ballatori N. Identification of a mechanism by which the methylmercury antidotes N -acetylcysteine and dimercaptopropanesulfonate enhance urinary metal excretion: transport by the renal organic anion transporter-1. Mol Pharmacol 62: 921-926, 2002.

Kojima R, Sekine T, Kawachi M, Cha SH, Suzuki Y, and Endou H. Immunolocalization of multispecific organic anion transporters, OAT1, OAT2, and OAT3, in rat kidney. J Am Soc Nephrol 13: 848-857, 2002.

Maiorino RM, Dart RC, Carter DE, and Aposhian HV. Determination and metabolism of dithiol chelating agents. XII. Metabolism and pharmacokinetics of sodium 2,3-dimercaptopropane-1-sulfonate in humans. J Pharmacol Exp Ther 259: 808-814, 1991.

Malo C and Berteloot A. Analysis of kinetic data in transport studies: new insights from kinetic studies of Na + - D -glucose cotransport in human intestinal brush-border membrane vesicles using a fast sampling, rapid filtration apparatus. J Membr Biol 122: 127-141, 1991.

Motohashi H, Sakurai Y, Saito H, Masuda S, Urakami Y, Goto M, Fukatsu A, Ogawa O, and Inui KI. Gene expression levels and immunolocalization of organic anion transporters in the human kidney. J Am Soc Nephrol 13: 866-874, 2002.

Pritchard JB. Luminal and peritubular steps in the renal transport of p -aminohippurate. Biochim Biophys Acta 906: 295-308, 1987.

Pritchard JB and Miller DS. Mechanisms mediating renal secretion of organic anions and cations. Physiol Rev 73: 765-796, 1993.

Rodeheaver DP, Aleo MD, and Schnellmann RG. Differences in enzymatic and mechanical isolated rabbit renal proximal tubules: comparison in long-term incubation. In Vitro Cell Dev Biol 26: 898-904, 1990.

Sekine T, Cha SH, and Endou H. The multispecific organic anion transporter (OAT) family. Pflügers Arch 440: 337-350, 2000.

Sweet DH, Chan LM, Walden R, Yang XP, Miller DS, and Pritchard JB. Organic anion transporter 3 (Slc22a8) is a dicarboxylate exchanger indirectly coupled to the Na + gradient. Am J Physiol Renal Physiol 284: F763-F769, 2003.

Sweet DH, Miller DS, Pritchard JB, Fujiwara Y, Beier DR, and Nigam SK. Impaired organic anion transport in kidney and choroid plexus of organic anion transporter 3 [Oat3 (Slc22a8)] knockout mice. J Biol Chem 277: 26934-26943, 2002.

Sweet DH, Wolff NA, and Pritchard JB. Expression cloning and characterization of ROAT1: the renal basolateral organic anion transporter in rat kidney. J Biol Chem 272: 30088-30095, 1997.

Tojo A, Sekine T, Nakajima N, Hosoyamada M, Kanai Y, Kimura K, and Endou H. Immunohistochemical localization of multispecific renal organic anion transporter 1 in rat kidney. J Am Soc Nephrol 10: 464-471, 1999.

Zalups RK and Barfuss D. Accumulation of inorganic mercury along the renal proximal tubule of the rabbit. Toxicol Appl Pharmacol 106: 245-253, 1990.

Zalups RK and Barfuss DW. Transport and toxicity of methylmercury along the proximal tubule of the rabbit. Toxicol Appl Pharmacol 121: 176-185, 1993.

Zalups RK and Lash LH. Effects of uninephrectomy and mercuric chloride on renal glutathione homeostasis. J Pharmacol Exp Ther 254: 962-970, 1990.

Zalups RK, Parks LD, Cannon VT, and Barfuss DW. Mechanisms of action of 2,3-dimercaptopropane-1-sulfonate and the transport, disposition, and toxicity of inorganic mercury in isolated perfused segments of rabbit proximal tubules. Mol Pharmacol 54: 353-363, 1998.

Zalups RK, Robinson MK, and Barfuss DW. Factors affecting inorganic mercury transport and toxicity in the isolated perfused proximal tubule. J Am Soc Nephrol 2: 866-878, 1991.

Zhang L, Gorset W, Dresser MJ, and Giacomini KM. The interaction of n-tetraalkylammonium compounds with a human organic cation transporter, hOCT1. J Pharmacol Exp Ther 288: 1192-1198, 1999.

作者单位：1 Department of Physiology, Mahidol University, Bangkok 1070 Thailand; and 2 Department of Physiology, University of Arizona, Tucson, Arizona 85724