Dual pathways for organic anion secretion in renal proximal tubule

462 D.S. MILLER AND J.B. PRITCHARDTHE JOURNAL OF EXPERIMENTAL ZOOLOGY 279:462–470 (1997)

© 1997 WILEY-LISS, INC. †This article is a US Governmentwork and, as such, is in the public domain in the United States ofAmerica.

JEZ 859

Dual Pathways for Organic Anion Secretion inRenal Proximal Tubule

DAVID S. MILLER* AND JOHN B. PRITCHARDLaboratory of Pharmacology and Chemistry, National Institute ofEnvironmental Health Sciences, National Institutes of Health, ResearchTriangle Park, North Carolina 27709

ABSTRACT Transport on the “classical” organic anion system in renal proximal tubule is spe-cific, active, Na-dependent, and ouabain sensitive. Here we review recent studies using intactteleost proximal tubules and laser scanning confocal microscopy which show that the secretion oflarge organic anions, such as, fluorescein-methotrexate (FL-MTX, Mw 923 Da) is handled by aseparate and distinct organic anion transport system. In contrast to the classical system, FL-MTXuptake into cells and secretion into the tubular lumen was ouabain insensitive and largely Na-independent. KCN did not affect cellular uptake but abolished secretion into the lumen. PAH andprobenecid, potent inhibitors of transport on the classical system, were weak inhibitors of FL-MTX transport. Uptake and secretion of FL-MTX were inhibited by micromolar concentrations ofother organic anions (MTX, folate, bromocresol green, bromosulfonphthalein). FL-MTX secretioninto the lumen was inhibited by leukotriene C4 and cyclosporine A, neither of which affected trans-port of the model substrate for the classical system, fluorescein. Thus, FL-MTX secretion is spe-cific, but largely Na-independent and ouabain-insensitive. Both the basolateral and luminal stepsin FL-MTX transport differ from those associated with fluorescein and P-aminohippurate secre-tion. J. Exp. Zool. 279:462�470, 1997. © 1997 Wiley-Liss, Inc.†

Renal organic anion transport mechanismsThe vertebrate renal proximal tubule actively

removes a large number of potentially toxic, nega-tively charged waste products of metabolism,drugs, environmental pollutants, and drug andpollutant metabolites from blood and concentratesthem in urine (reviewed in Pritchard and Miller,’93, ’96). Because of their important role inxenobiotic excretion, their ability to transport awide variety of compounds and their remark-able potency, the cellular mechanisms drivingorganic anion secretion have been of interestfor decades. The prototypic substrates for the“classical” organic anion system are P-amino-hippurate (PAH) and fluorescein (FL). Experi-ments from this laboratory and others havedemonstrated that the same basic cellularmechanisms drive organic anion excretion in allanimals studied from fish, birds, and snakes tomammals, even invertebrates.

Three elements comprise the transport system(Fig. 1). First, studies with isolated basolateralmembrane (BLM) vesicles and with intact renaltubules from a number of species have shown thatorganic anion uptake into the cell is driven byindirect coupling to the Na gradient through Na-

divalent organic anion cotransport and divalentorganic anion/monovalent organic anion exchange.Recent work firmly establishes α-KG as the physi-ological counter-ion that directly drives concen-trative organic anion uptake at the basolateralmembrane. It is this first step at the basolateralmembrane, which is indirectly coupled to the Nagradient, that powers overall secretion (Millerand Pritchard, ’91). Second, imaging studiesfrom our laboratories have demonstrated thatorganic anions are specifically sequestered inintracellular vesicles and that vectorial, micro-tubule-dependent vesicle movement (trans-cytosis) could contribute to net secretion (Milleret al., ’93; Miller and Pritchard, ’94). Finally,brush border membrane (BBM) vesicle studiesindicate that the most likely mechanism medi-ating transport from cell to lumen is facilitateddiffusion, driven by the electrical potential dif-ference (PD) across the lumenal membrane(Pritchard and Miller, ’93). However, such a

*Correspondence to: Dr. David S. Miller, LPC, NIH/NIEHS, P.O.Box 12233, Research Triangle Park, NC 27709. E-mail: [email protected]

RENAL ORGANIC ANION TRANSPORT SYSTEMS 463

mechanism has yet to be demonstrated in in-tact tubules and our attempts to do that withkillifish proximal tubules led us to concludethat PD-driven facilitated diffusion was not amajor contributor to the cell to lumen flux ofFL (Miller et al., ’96). In our view, the processresponsible for cell to lumen transport in in-tact proximal tubules remains to be established.Indeed, the recent data of Ullrich et al. (thisvolume) suggest that urate/organic anion ex-change could play a role in luminal efflux.

The studies instrumental in defining themechanisms mediating organic anion transportutilized as substrates small organic anions,such as the PAH, FL, and 2,4-dichlorophenoxy-acetic acid (molecular wt 200–350 Da). However,the proximal tubule handles a wide range ofanionic compounds, some quite large. It hasbeen generally assumed that these larger com-pounds are handled by the same Na-dependentmechanism as smaller organic anions, since 1)high concentrations of PAH and probenecid re-duce transport of the larger anions, and 2) manylarge organic anions are themselves potent in-hibitors of FL and PAH transport. Here we re-view recent data from experiments with intactproximal tubules indicating that a second, Na-independent transport system is responsible forthe secretion of larger organic anions.

Studying renal transport usingconfocal microscopy

In these experiments, the transport of threefluorescent organic anions was measured in iso-lated killifish proximal tubules using confocal mi-croscopy and quantitative image analysis. Thestructures of the fluorescent compounds that wereused are shown in Figure 2. At one extreme is FL(Mw 332 Da). In every respect it appears to behandled by the proximal tubule in the same wayas PAH, the model substrate for the classicalorganic anion transport system (Miller and Prit-chard, ’91, ’94). At the other extreme is the fluo-rescein derivative of methotrexate (FL-MTX; Mw923 Da). Intermediate in size is sulforhodamine101 (Texas red free acid; MW 606 Da).

The work was carried out using isolated proxi-mal tubules from a marine teleost, killifish (Fun-dulus heteroclitus). As discussed previously, renaltissue from teleost fish offers several advantagesfor the study of secretory transport mechanisms(Pritchard and Miller, ’91). These include: easy tu-bule isolation, long viability, a luminal compart-ment closed off from the bathing solution andpotent, uphill organic anion secretion into the lu-men. With these tubules, fluorescent substratesand the modern techniques of confocal microscopyand quantitative image analysis we are able toboth visualize and measure in living tissue all thesteps in renal secretion.

Fig. 1. Mechanisms that drive the classical organic aniontransport system in renal proximal tubule (with fluorescein,FL) as substrate; reviewed in Pritchard and Miller, ’93, ’96).These include 1) Na-α-ketoglutarate (α-KG) cotransport, and

2) α-ketoglutarate-organic anion exchange at the basolateralmembrane, and 3) facilitated diffusion at the lumenal (brushborder) membrane. Accumulation of FL in vesicular, intra-cellular compartments is also shown.

464 D.S. MILLER AND J.B. PRITCHARD

Fluorescein secretionFigure 3A shows a representative confocal slice

through a proximal tubule that was incubated tosteady state in a medium with 1 µM FL. The gen-eral pattern of fluorescence in the image revealsthe two uphill steps in transport: from bath tocell and from cell to lumen. Figure 4A summa-rizes FL transport data for tubules exposed to in-hibitors of transport. These data were obtainedfrom confocal images by measuring the mean fluo-rescence of the cellular and lumenal compart-ments (Miller et al., ’96; Masereeuw et al., ’96).In control tubules, average lumenal fluorescenceis about five times cellular fluorescence (Fig. 4A).Consistent with Na-driven transport, FL uptakeand secretion are nearly abolished when tubulesare incubated in Na-free medium or when the tu-bules are exposed to 100 µM ouabain, a concen-tration of cardiac glycosode that completelyinhibits Na,K-ATPase in teleost tubule homoge-nates (Miller, ’81). Transport is also abolished by1 mM PAH or 0.5 mM probenecid (Fig. 4A). Not

shown here are additional results demonstratingthat both uptake and secretion are stimulated by10–50 µM glutarate (Miller and Pritchard, ’91, ’94)and that both uptake and secretion are abolishedby metabolic poisons (Miller, unpublished data).These are the expected results for a substrate se-creted by the classical, Na-dependent organic an-ion transport system shown in Figure 1.

FL-MTX secretionWhen tubules are incubated in medium with 2

µM FL-MTX, strong secretion of the large fluo-rescent organic anion is found. In fact, images ofthe FL-MTX-loaded tubules do not appear to bevery different from those of FL-loaded tubules(Fig. 3B). Both exhibit lumenal and cellular accu-mulation of the respective dyes with fluorescenceintensity in lumen > cell. Both dyes also exhibitaccumulation in intracellular vesicles, but FL-MTX fills an additional cellular compartment notseen with FL. High magnification micrographs ofthe basal pole of the cells reveal FL-MTX accu-mulation associated with the basolateral mem-brane, suggesting binding to this cellular structure(Masereeuw et al., ’96).

Although fluorescence distribution patterns forthe two fluorescent organic anions appear simi-lar in control tubules, there are substantial dif-ferences in the changes to those patterns elicitedby inhibitors of transport. Figure 4B shows datawhich bear on the energetics of FL-MTX trans-port. As with FL, control tubules incubated in me-dium with 2 µM FL-MTX exhibited lumen to cellfluorescence ratios of 3–5. In contrast to FL, re-moving Na from the bath only reduced cellularand lumenal fluorescence by 30–40%; ouabain hadno significant effects. Also, unlike FL whereglutartate or α-KG stimulates uptake and secre-tion (Miller and Pritchard, ’91), glutarate had noeffect on FL-MTX transport (Fig. 4B). Cyanide,however, abolished FL-MTX transport into the lu-men, but did not reduce cellular fluorescence.These data demonstrate that FL-MTX transportis much less sensitive to Na-replacement and oua-bain than FL transport and that only the lume-nal step in FL-MTX transport is affected whenmetabolism is inhibited.

FL-MTX transport is sensitive to inhibition byother organic anions (Fig. 5). High concentrationsof PAH and probenecid reduce both cellular andlumenal fluorescence, but only partially. Thus, FL-MTX transport is much less sensitive to these in-hibitors of the classical organic anion system thanFL transport. Methotrexate and folate, structural

Fig. 2. Structures of the fluorescent organic anions usedto probe transport mechanisms in renal proximal tubule.


Fig. 3. Fluorescein (1 µM; Left) and fluorescein methotr-exate (2 µM; Right) secretion in killifish renal proximal tu-bules. Shown are confocal images of tubules incubated tosteady state with the indicated fluorescent substrates. To ap-

preciate the potency of the transport systems involved, com-pare the fluorescence intensities of the medium, cells and tu-bular lumens. Scale bars = 15 µm.

analogues of FL-MTX also inhibit. The larger or-ganic anions, BCG and BSP, are potent inhibitorsof cell to lumen FL-MTX transport, but of the two,only BCG reduces cellular accumulation. Togetherwith the results on Na dependence and cyanidesensitivity, these data indicate that both cellularaccumulation and cell to lumen transport of FL-MTX are specific, but that the specificity charac-teristics look very different from that seen for theclassical organic anion transport system.

Differences between FL and FL-MTX transportare even more dramatic when we consider the ef-fects of drugs known to inhibit specifically an

organic anion transporting ATPase on the canali-cular membrane of hepatocytes. This ATPase,called the canalicular multispecific organic aniontransporter (cMOAT; Oude Elferink and Jansen,’94), is particularly sensitive to inhibition byleukotriene C4 (LTC4), exhibiting a Ki and Km fortransport of about 0.3 µM (Ishikawa et al., ’90).It is also sensitive to inhibition by the immuno-suppressive cyclic polypeptide, cyclosporin A(CSA), which exhibits a Ki of 3 µM (Bohme et al.,’93). Figure 6 shows the effects of these drugs onthe transport of FL-MTX and FL in killifish tu-bules. With FL-MTX as substrate, both LTC4 and


Fig. 4. Comparison of fluorescein (FL; A) and fluoresceinmethotrexate (FL-MTX; B) secretion in killifish renal proxi-mal tubules. Tubules were incubated in medium with 1 µMFL or 2 µM FL-MTX and the indicated additions for 30 min.For FL, Na-replacement (with N-methylglucamine), 0.1 mMouabain, 1 mM P-aminohippurate (PAH), and 0.5 mM

probenecid (PROB) significantly reduced the cellular and lu-menal fluorescence (P < 0.01). For FL-MTX, Na-replacementreduced cellular and lumenal fluorescence significantly (P <0.05); KCN reduced only lumenal fluorescence significantly(P < 0.01). Data from Masereeuw et al., ’96.

CSA substantially reduce lumenal fluorescence,but have no effects on cellular fluorescence. In con-trast, with FL as substrate, LTC4 and CSA arewithout effect. This latter result indicates thatLTC4 and CSA do not affect transport on the clas-sical organic anion system. It also argues thatLTC4 and CSA do not inhibit secretion of FL-MTXinto the lumen by indirect means, such as open-ing tight junctions or inhibiting cellular metabo-lism, since we would have expect FL secretion tobe reduced by increased junctional permeabilityor by disrupted metabolism.

Sulforhodamine 101 secretionIf FL and FL-MTX, two fluorescent organic an-

ions with very different molecular weights are se-creted by different mechanisms, would an organicanion of intermediate size be handled by bothmechanisms? Figure 7 summarizes initial experi-ments with sulforhodamine 101 as substrate. Incontrol tubules incubated in medium with 5 µM

sulforhodamine 101, lumenal fluorescence exceedscellular by a factor of 3–5, a pattern similar tothat seen with FL and FL-MTX. Ouabain partiallyreduces sulforhodamine 101 fluorescence in bothregions of the tubule as does PAH; probenecid isa potent inhibitor of sulforhodamine 101 trans-port. LTC4 and CSA partially inhibit sulforho-damine 101 transport from cell to lumen, buthave no effect on cellular accumulation. Thus,with regard to ouabain, probenecid, LTC4, andCSA, the inhibition pattern seen for sulfor-hodamine 101 is intermediate between thatseen with FL and FL-MTX.

Two renal transport systems fororganic anions

Figure 8 summarizes our current knowledge ofthe mechanisms that drive organic anion secre-tion in proximal tubule. The top cell shows theelements of the classical system as presented ear-lier and the bottom cell the elements of what ap-


pears to be a newly discovered, separate sys-tem for larger organic anions. These are shownin two cells for convenience; there is no evidencethat would lead us to believe that both mecha-nisms are not present in the same cells. Thetwo systems are similar in that they providetwo uphill steps for organic anion secretion andthat substrates for both are compartmentalizedwithin renal cells. They differ in several impor-tant respects: First, the classical system isstrongly Na dependent and ouabain sensitive,but the new system is not. Second, uptake andsecretion on the classical system is abolishedby cyanide, but CN only blocks the lumenal stepin the new system. Third, the two systems dif-fer in their sensitivities to PAH and probenecid,which are potent inhibitors of transport on theclassical system (complete inhibition of 1 µMFL transport with 100–200 µM; Miller et al.,’96 and Miller, unpublished data), but weak

inhibitors of the new system (only partial inhi-bition with 0.5–1 mM). Finally, LTC4 andCSA, inhibitors of certain drug-transporting AT-Pases, do not affect transport on the classicalsystem, but nearly block the lumenal step ofthe new system.

Let us for a moment consider the implicationsof these results. They argue that, as in liver, therenal proximal tubule possesses both Na-depen-dent and Na-independent transport systems fororganic anions. They suggest that size is one fac-tor that determines the partitioning of substratesbetween the two renal pathways. However, thereappears to be substantial overlap in the specifici-ties of the two systems, since, sulforhodamine 101,an organic anion, intermediate in size between FLand FL-MTX, is handled by both.

Finally, what do we know about the individualmechanisms involved in the secretion of largeorganic anions in proximal tubule? Confocal mi-

Fig. 5. Inhibition of FL-MTX transport by organic anions.Tubules were incubated for 30 min in medium containing 2µM FL-MTX and the indicated additions. All treatments re-duced lumenal fluorescence significantly (P < 0.01); PAH,probenecid (PROB), and folate reduced cellular fluorescence

significantly (P < 0.01); bromocresol green (BCG) reduced cel-lular fluorescence significantly (P < 0.05); bromosulphon-phthalein (BSP) did not reduce cellular fluorescence. Datafrom Masereeuw et al., ’96.


Fig. 6. Effects of cysteinyl leukotriene C4 (LTC4) andcyclosporin A (CSA) on the transport of (A) 2 µM FL-MTXand (B) 1 µM FL. Tubules were incubated for 30 min in me-dium containing the fluorescent substrate and the indicated

additions. For FL-MTX, LTC4, and CSA significantly reducedlumenal fluorescence (P < 0.01). None of the treatments sig-nificantly altered FL fluorescence. Data from Masereeuw etal., ’96.

crographs show that FL-MTX accumulates inproximal tubule cells to levels that exceed thoseof the medium. Cellular accumulation is only par-tially Na-dependent, but is reduced by several or-ganic anions suggesting that uptake is specific.Cellular accumulation is not reduced by CN. Thissuggests that the possibility of a large bindingcomponent in total cellular uptake, a suggestionconsistent with our high magnification imagesshowing that FL-MTX accumulates at the baso-lateral membrane and on or in intracellularvesicles (Masereeuw et al., ’96).

Transport of large organic anions from cell totubular lumen is clearly specific (inhibitable byother organic anions), concentrative, and energydependent (abolished by CN). It occurs by a pro-cess that is not shared with FL but that is blockedby LTC4 and CSA, both inhibitors of cMOAT, ahepatic ATPase that transports anionic drugs andmetabolites (Bohme et al., ’93; Oude Elferink andFig. 7. Effects of Na-replacement, ouabain, P-amino-

hippurate (PAH), probenecid (PROB), cyclosporin A (CSA),and cysteinyl leukotriene C4 (LTC4) on the transport ofsulforhodamine 101 in killifish renal proximal tubules. Tu-bules were incubated for 30 min in medium containing 5 µMsulforhodamine 101 and the indicated additions. Ouabain,PAH, and probenecid reduced lumenal and cellular fluores-

cence significantly (P < 0.01); lumenal fluorescence was re-duced significantly by LTC4 (P < 0.05). Data from Masereeuwet al. (’96).


Jansen, ’94; Ishikawa et al., ’90). For this reasonwe speculate that the lumenal step in FL-MTXsecretion is mediated by a related drug transport-ing ATPase, perhaps a renal form of cMOAT,which has yet to be identified. If this speculationproves to be correct, this transporter will be thesecond xenobiotic transporting ATPase on the lu-menal membrane of renal proximal tubule cells,the first being the multidrug resistance trans-porter or p-glycoprotein, which handles large or-ganic cations and some neutral compounds (Fordand Hait, ’90; Schramm et al., ’95). These ATPaseswould then function as parallel excretory pumpsfor larger organic anions and organic cations, com-pounds that might not be handled well by the clas-sical organic anion and organic cation systems,respectively.

Fig. 8. Two transport systems for organic anions in renalproximal tubule. The top cells shows the classical system;the bottom cell shows additional 4) basolateral and 5) lume-nal steps proposed for fluorescein-methotrexate (FL-MTX)

transport. The ATP-dependence of step 5 has yet to be dem-onstrated. Accumulation of both substrates in intracellularcompartments is also shown.

LITERATURE CITEDBohme, M., M. Buchler, M. Miller, and D.F. Keppler (1993)

Differential inhibition by cyclosporins of primary-active ATP-dependent transporters in the hepatocyte canalicular mem-brane. FEBS Lett., 333:193–196.

Ford, J.M., and W.N. Hait (1990) Pharmacology of drugs thatalter multidrug resistance in cancer. Pharmacol. Rev.,42:155–199.

Ishikawa, T., M. Muller, C. Klunemann, T. Schaub, and D.Keppler (1990) ATP-dependent primary active transport ofcysteinyl leukotrienes across liver canalicular membrane.Role of the ATP-dependent transport system for glutathioneS-conjugates. J. Biol. Chem., 265:19279–19286.

Masereeuw, R., F.G.M. Russel, and D.S. Miller (1997) Mul-tiple pathways of organic anion secretion in renal proximaltubule revealed by confocal microscopy. Am. J. Physiol.,271:F1173–F1182.

Miller, D.S. (1981) Heavy metal inhibition of p-amino-hippurate transport in flounder renal tissue: Sites of HgCl2action. J. Pharmacol. Exp. Therap., 219:428–434.


Miller, D.S., S. Letcher, and D.M. Barnes (1996) Fluorescentimaging study of organic anion transport from renal proxi-mal tubule cell to lumen. Am. J. Physiol., 271:F508–F520.

Miller, D.S., and J.B. Pritchard (1991) Indirect coupling oforganic anion secretion to sodium in teleost (Paralichthyslethostigma) renal tubules. Am. J. Physiol., 261:R1470–R1477.

Miller, D.S., and J.B. Pritchard (1994) Nocodazole inhibitionof organic anion secretion in teleost renal proximal tubules.Am. J. Physiol., 267:R695–R704.

Miller, D.S., D.E. Stewart, and J.B. Pritchard (1993) Intra-cellular compartmentation of organic anions within renalcells. Am. J. Physiol., 264:R882–R890.

Oude Elferink, R.P.J., and P.L.M. Jansen (1994) The role of

the canalicular multispecific organic anion transporter in thedisposal of endo- and xenobiotics. Pharmac. Ther., 64:77–97.

Pritchard, J.B., and D.S. Miller (1991) Comparative insightsinto the mechanisms of renal organic anion and cation se-cretion. Am. J. Physiol., 261:R1329–R1340.

Pritchard, J.B., and D.S. Miller (1993) Mechanisms mediat-ing renal secretion of organic anions and cations. Physiol.Rev., 73:765–796.

Pritchard, J.B., and D.S. Miller (1996) Renal secretion of or-ganic anions and cations. Kidney Intl., 49:1649–1654.

Schramm, U., G. Fricker, R. Wenger, and D.S. Miller (1995)p-Glycoprotein-mediated secretion of a fluorescent cyclo-sporin analogue by teleost renal proximal tubules. Am. J.Physiol., 268:F46–F52.

Documents

Dual pathways for organic anion secretion in renal proximal tubule