the kidney is a major organ for the elimination of potentially harmful metabolites and xenobiotics. The tubular epithelial cells of the kidney are especially enriched with membrane transporters to facilitate the excretion of waste products, drugs, and toxins (14, 17). Unfortunately, the kidney itself is also a target of chemical insults. Recent studies showed that transporter-mediated intrarenal accumulation can serve as an important mechanism underlying drug- and toxin-induced kidney injury (3, 7, 23, 28).

Organic cation transporters (OCTs) from the solute carrier 22 (SLC22) family are the classical polyspecific transporters involved in renal excretion of small hydrophilic organic cations. Known substrates of the OCTs include endogenous compounds (e.g., biogenic amines), drugs (e.g., metformin, platinium drugs), and toxins [e.g., 1-methyl-4-phenylpyridinium (MPP+), paraquat] (6, 26). In humans, OCT2 is the major isoform expressed in the kidney (6, 8). Localized to the basolateral membrane of the proximal tubule, OCT2 mediates Na⁺-independent, electrogenic uptake of organic cations from blood into the tubular cells. These organic cations are subsequently secreted into the lumen by the apical multidrug and toxin extrusion transporter (MATE1) via an organic cation/proton exchange mechanism (15).

We recently reported the cloning and functional characterization of a novel OCT, the plasma membrane monoamine transporter (PMAT) (4, 5). By gene ontology, PMAT belongs to the equilibrative nucleoside transporter (ENT) family (SLC29) and was alternatively named ENT4 (1, 9). We previously demonstrated that PMAT mainly transports organic cations (e.g., MPP+, biogenic amines, metformin) and shares an overlapping substrate and inhibitor profile with the OCTs (4, 5). In addition, PMAT is also able to transport the purine nucleoside, adenosine (1, 27). PMAT-mediated transport is Na⁺ independent, but can be further enhanced by an acidic environment (1, 27). In humans and rodents, PMAT mRNA transcripts are present in multiple tissues including brain, kidney, heart, and small intestine (1, 5). Recently, using a fusion protein antibody of PMAT, we detected the expression of the 55-kDa protein in human kidney homogenate by Western blot (27). However, this fusion protein antibody failed to specifically react with PMAT in kidney tissue sections. Thus, the precise location of PMAT and its function in the kidney remain unknown.

In the present study, we developed a new peptide antibody toward PMAT and determined the intrarenal localization of PMAT using immunofluorescence microscopy. We also investigated the role of PMAT in puromycin aminonucleoside (PAN)-induced kidney toxicity.

METHODS

Production and purification of PMAT antibody.

Computer-aided analysis of human PMAT identified two highly antigenic sequences corresponding to amino acids 90–103 (TDVDYLHHKYPGTS) and 469–482 (ILAAGKVSPKQREL). These sequences are 100% conserved across human and rodent PMATs, but are divergent with other SLC29 family members. The two peptides were chemically synthesized, conjugated to keyhole limpet hemocyanin, and purified to >95% homogeneity. Polyclonal rabbit antisera were then commercially prepared using standard protocols by Sigma Genosys (Woodlands, TX). The antibody directed toward the amino acids 469–482 showed highest ELISA titer toward the purified antigen and was also highly reactive to PMAT produced in Madin-Darby canine kidney (MDCK) cells. This antibody, designated P469, was affinity purified by chromatography on a column prepared by cross-linking the antigenic peptide to Sepharose 4B (ProSci, Poway, CA). The purified antibody was used in immunolocalization studies. The preimmune sera or peptide preabsorbed antibody were used as controls in immunohistochemistry and immunocytochemistry (ICC) experiments.

PMAT expression in MDCK cells.

PMAT cDNA was obtained and stably expressed in MDCK cells as described previously (4). In brief, PMAT cDNA, isolated from a human kidney cDNA library, was subcloned into the pCDNA3 vector (Invitrogen, Carlsbad, CA) and transfected into MDCK cells by liposome-mediated transfection (Lipofectamine; Invitrogen). A stably transfected cell line was obtained by G418 selection and maintained in MEM with l-glutamine containing 10% FBS and 200 μg/ml of G418.

Western blotting.

Plasma membrane protein extract from vector- or PMAT-transfected MDCK cells was prepared and subjected to Western blot analysis as previously described (30). To verify that P469 does not react with other SLC29 isoforms, Western blot was carried out in cells overexpressing yellow fluorescence protein-tagged ENT1 (SLC29A1) using MDCK cell lines previously generated in this laboratory (30). To detect PMAT expression in human kidney, commercial blots containing human kidney whole lysates (20 μg/lane) were purchased from Imgenex (San Diego, CA). The blot was blocked with 5% (wt/vol) nonfat dry milk powder (Bio-Rad Laboratories, Hercules, CA) and incubated with P469 primary antibody diluted 1:1,000 in 1% milk. After being washed and incubation with horseradish peroxidase-conjugated goat anti-rabbit IgG (1:20,000 dilution), immunoreactive proteins were detected by chemiluminescence using Supersignal West Pico Reagent (Pierce) followed by autoradiography.

ICC.

Methods for immunostaining of vector- and PMAT-transfected MDCK cells were previously described (27). Briefly, confluent cells grown on borosilicated coverglass (Nalge Nunc International, Naperville, IL) were washed and fixed with 4% (vol/vol) paraformaldehyde. Cells were then permeabilized with 0.2% Triton X-100 in PBS for 10 min and incubated with a blocking buffer (10% FBS, 0.1% Triton X-100 in PBS). Cells were then incubated with the primary P469 antibody (1:200 dilution in blocking buffer) for 1 h, washed three times, and incubated for 30 min with a secondary Alexa Fluor 488-conjugated goat anti-rabbit IgG (1:1,000 in blocking buffer; Molecular Probes, Eugene, OR) in the dark. The cells were then washed three times with PBS containing 0.05% Tween 20 and the cell nuclei were counterstained with Topro-3 (Molecular Probes). The cells were then imaged by a Leica SP1/MP confocal microscope with inverted lens at the University of Washington's Keck Imaging Center.

Localization of PMAT protein in kidney sections by immunofluorescence staining.

Cryosections from normal adult human kidney were obtained from the laboratory of renal pathology, Department of Pathology (University of Washington, Seattle, WA). Normal adult rat kidney slides were purchased from Zyagen (San Diego, CA). Slides fixed in cold acetone were thawed 30–60 s at room temperature, briefly washed with PBST (0.05% Tween 20), and blocked with 10% FBS in PBST for 1 h. Slides were then incubated with the purified P469 antibody (1:500) in 10% FBS in PBST overnight at 4°C. After being washed three times for 30 min with PBST, the sections were stained with Alexa Fluor 488-conjugated goat anti-rabbit IgG (1:1,000 in 10% FBS in PBST; Molecular Probes) in the dark for 1 h, and rewashed three times with PBST. The slides were then mounted with ProLong Gold antifade reagent (Molecular Probes) and subjected to confocal microscopy. For dual-immunofluorescence histochemistry, the following primary antibodies for cellular phenotyping were obtained and used at the indicated dilutions: mouse monoclonal anti-WT1 antibody (1:200; Santa Cruz Biotechnology, Santa Cruz, CA), mouse monoclonal anti α-smooth muscle actin antibody (1:1,000; Dako, Carpinteria, CA), and mouse monoclonal anti-CD31 antibody (1:125; Pharmingen, San Diego, CA). Tissue sections were incubated with diluted primary antibody markers containing either P469 antibody (1:500) or prebleed serum (1:500; control) overnight at 4°C. Tissue sections were washed three times in PBS and then incubated for 60 min at room temperature with a mixture of Alexa Fluor 488-conjugated goat anti-rabbit IgG (1:1,000 dilution) and Alexa Fluor 647-conjugated chicken anti-mouse IgG (1:1,000 dilution; Molecular Probes). The secondary antibodies were tested and showed no cross-species reactivity. The slides were washed three times with PBST, mounted with ProLong Gold antifade reagent, and immunofluorescence was observed and photographed by confocal microscopy using a Leica SP confocal microscope at the University of Washington's Keck Imaging Center. All histological studies were repeated three times using different tissue sections.

PAN sensitivity assay.

The cytotoxicity of PAN toward PMAT- and pcDNA3-transfected MDCK cells was measured by MTT assay. Cells were seeded in MEM with 10% FBS on 96-well plates at a density of 5,000 cells/well. After ∼48-h incubation (∼40–50% confluence), cells were changed to fresh growth medium containing PAN at various concentrations. For the protection experiment, cells were incubated in medium containing 250 μM PAN with or without the PMAT inhibitor decynium-22 (2 μM). After a total of 72-h incubation in a 95% O₂ incubator at 37°C, cells were washed and the plates were developed with thiazolyl tetrazolium blue according to the manufacturer's instructions (Sigma, St. Louis, MO). The IC₅₀ values were determined by fitting the cell growth data to the following model using nonlinear regression (WinNonLin version 3.2): S = S_max − [S_max − S₀] × [C_γ/(C_γ + IC_50γ)], where S is the cell survival expressed as percentage of the optical density to untreated control cells, S_max is the maximal cell survival, S₀ is the lowest residual cell survival at the high drug concentration, C is PAN concentration, γ is the Hill coefficient, and IC₅₀ is the PAN concentration leading to half-maximal cell survival. Five to six determinations were carried out within each experiment, and four independent experiments were performed.

Cellular uptake of PAN in MDCK cells.

Uptake assays in vector- or PMAT-transfected cells were performed as previously described (27). In brief, cells were plated in 12-well plates, grown to confluence, washed, and preincubated for 10∼15 min at 37°C in Krebs-Ringer-Henseleit (KRH) buffer at pH 7.4. Transport assays were performed at 37°C for 1 min by incubating cells in KRH buffer containing 100 μM PAN at pH 6.6 [buffered by 2-(N-morpholino)ethanesulfonic acid] or pH 7.4 (buffered by HEPES). Uptake was terminated by washing cells three times with ice-cold KRH buffer. Ten nanograms of 2-chloro-2′-deoxyadenosine (internal standard) were then added to each well, and cells were lysed with 1 ml of ice-cold acetonitrile/H₂O (vol/vol, 2:1). After sonication for 1 min, 25 μl of cell lysate from each well were taken for BCA assay to determine the protein content. The remaining cell lysate in each well was centrifuged and the supernatants were collected, diluted 1:3 into water, and stored at −80°C for PAN analysis by liquid chromatography/mass spectrometry (LC/MS/MS). Cellular PAN concentrations were normalized to protein content in the corresponding well.

Cellular PAN quantification.

The cellular content of PAN was quantified on a Waters Aquity UPLC system coupled with a Micromass Premiere-XE mass spectrometer (Waters, Milford, MA). Briefly, 10 μl of each sample were separated on a Synergi Hydro 50 × 2.1-mm, 4-μm RP 80 column (Phenomenex, CA) using gradient elution with mobile phase consisting of 10 mM formic acid and methanol. The flow rate was set at 0.3 ml/min. The mass spectrum was operated in the positive APCI-ES MRM mode. The transitions monitored were m/z 295>164 for PAN and m/z 286>170 for 2-chloro-2′-deoxyadenosine (internal standard). Compound content in each sample was determined using a standard curve prepared with known concentrations of the PAN.

RESULTS

Polyclonal peptide antibody P469 specifically reacted with PMAT.

We previously developed a polyclonal fusion-protein antibody toward the NH₂ terminus of PMAT. While highly reactive in Western blot, the fusion protein antibody failed to detect PMAT expression in kidney tissue sections due to high background staining (27). To determine the cellular localization of PMAT in the kidney, a new antibody, P469, was developed against the 14-amino acid sequence (ILAAGKVSPKQREL) composing the last intracellular loop of human PMAT (Fig. 1A). The peptide sequence is completely conserved among human, rat, and mouse PMAT proteins. The purified P469 antibody specifically reacted with PMAT protein in Western blot on membrane extract from PMAT-transfected MDCK cells (Fig. 1B), whereas no band was observed in vector-transfected cells. In addition, P469 did not react with the prototype ENT1 (SLC29A1; Supplemental Fig. 1; the online version of this article contains supplemental data). A strong, single band with expected molecular weight of PMAT (∼55 kDa) was observed using lysates from human kidney homogenate (Fig. 1C). The specificity of this antibody was further tested in ICC using vector- and PMAT-transfected MDCK cells. As expected, there was no specific staining in vector-transfected MDCK cells (Fig. 1D). In contrast, PMAT-expressing cells showed strong immunofluorescence on their plasma membranes (Fig. 1E). This specific staining disappeared when the primary antibody is substituted with prebleed rabbit serum (Fig. 1F).

PMAT immunoreactivity was predominantly found in the glomerulus.

To determine the intrarenal localization of PMAT, cryopreserved tissue sections from adult human and rat were subjected to immunostaining using P469 antibody and confocal fluorescence microscopy (Fig. 2). Strong and specific fluorescence labeling of the glomeruli was observed in rat kidney sections (Fig. 2, A and B). There was essentially no staining of renal tubules or vasculatures. No fluorescence signal was seen in sections incubated with prebleed serum and secondary antibodies (Fig. 2C). Similar patterns showing predominant glomerular localization were also observed on cryosections of adult human kidney (Fig. 2, D and E).

PMAT was specifically localized in glomerular visceral epithelial cells.

The morphology of the PMAT-positive cells in the glomeruli from single immunofluorescence staining (Fig. 2) is most similar to that of podocytes, which are highly specialized glomerular visceral epithelial cells that play a central role in the formation of the filtration barrier of the glomerulus (20, 21). To discern the cellular expression of PMAT in the glomeruli, double immunofluorescence labeling was performed on human kidney cryosections using established cell type-specific markers for podocytes, mesangial cells, and glomerular endothelial cells–the three major cell types that make up the glomerulus. No overlap of specific PMAT staining was observed with α-smooth muscle actin (α-SMA), a cell body marker for mesangial cells (13), or CD31, a cell surface marker for endothelial cells (24) (Fig. 3, A and B). In contrast, the majority of the PMAT-positive cells were colabeled with the Wilms’ tumor suppressor (WT1; Fig. 3, C and D), a nuclear protein and a well-established marker for podocytes (10, 20). At higher resolution, PMAT labeling was clearly seen in cell body and foot processes of the podocyte, while WT1 staining was restricted to the nucleus of the same cell (Fig. 3D). Similar results were obtained when the studies were performed on rat kidney cryosections (data not shown). These data demonstrated that PMAT is specifically localized to podocytes in the kidney.

PMAT-expressing cells exhibited enhanced sensitivity to PAN.

PAN is a well-established nephrotoxin that specifically damages podocytes in the glomerulus (16, 19). PAN-induced nephrosis in rats has been used extensively as an experimental model to study the fundamental processes involved in proteinuria and other glomerular diseases (16). PAN (3′-amino-3′-deoxy-N⁶,N⁶-dimethyladenosine) is structurally similar to the purine nucleoside adenosine, but the 3′ NH₂ group of PAN (calculated pKa = 13.3) carries a positive charge as it is protonated at physiologic pH. We previously showed that PMAT transports structurally diverse organic cations (4). Adenosine, a purine nucleoside, is also transported by PMAT (1, 27). Because PAN is an adenosine analog and exists in the cationic form at physiologic pH, we hypothesized that PMAT transports PAN and PMAT-mediated cellular accumulation of PAN contributes to its specific toxicity toward podocytes. To test this hypothesis, we first performed MTT assays to determine the dose-response curves of PAN cytotoxicity in vector- and PMAT-transfected MDCK cells. The IC₅₀ values for PMAT-expressing and vector-transfected cells were 48.9 ± 2.8 and 122.1 ± 14.5 μM, respectively, suggesting expression of PMAT-enhanced cell sensitivity to PAN (Fig. 4A). At 250 μM, PAN was toxic to both PMAT-expressing and vector-transfected cells. However, PMAT-expressing cells were four times more sensitive to PAN toxicity than vector-transfected cells (Fig. 4B). Decynium 22 (2 μM), a potent PMAT inhibitor (K_i= 0.1 μM) (4), completely reversed PAN toxicity in PMAT-expressing cells, making them about eightfold more resistant than cells without inhibitor treatment (Fig. 4B). Decynium 22 also exhibited some protective effect on vector-transfected cells, which was likely due to inhibition of some endogenous transporters in MDCK cells that also transport PAN. Treatment with decynium 22 alone (2 μM) did not have any effect on cell growth and viability (data not shown). Thus, these data suggest that PMAT transports PAN, and PMAT-mediated cellular uptake of PAN is an important determinant of its cytotoxicity.

PAN is transported by PMAT in a pH-dependent manner.

To confirm whether PAN is indeed a PMAT substrate, we developed a LC/MS/MS method to directly measure the cellular accumulation of PAN in PMAT-expressing and control cells. Because we and others (1, 27) showed that PMAT activity is stimulated by acidic pH, we performed uptake assays at both pH 7.4 and pH 6.6. A short incubation time (1 min) was used to minimize intracellular metabolism of PAN. At a substrate concentration of 100 μM, a 1.7- to 2.3-fold increase in PAN uptake was observed in PMAT-expressing cells compared with vector-transfected cells at both pH 6.6 and 7.4. PAN uptake in PMAT-expressing cells was fourfold higher at pH 6.6 than that at pH 7.4. These data suggest that PAN is transported by PMAT, and as seen with many other PMAT substrates (e.g., MPP+, adenosine, metformin) (1, 27, 31), PMAT-mediated PAN transport is stimulated by lower extracellular pH (Fig. 5).

DISCUSSION

PMAT is a new OCT first cloned and characterized in our laboratory (5). We previously showed that PMAT functions as a Na⁺-independent polyspecific transporter whose substrate specificity is remarkably similar to the OCTs (4). While PMAT does not typically interact with other nucleosides and nucleoside analogs, adenosine is recognized and transported by PMAT, probably in the protonated form (1, 30). In the present study, we determined the intrarenal cellular localization of PMAT using a newly developed peptide antibody. We also demonstrated that PMAT transports PAN and may contribute to PAN-induced kidney toxicity.

Using an affinity-purified polyclonal antibody developed toward the 14-amino acid residues in the last intracellular loop of PMAT, we localized PMAT protein to the glomerulus of human and rat kidneys by confocal immunofluorescence microscopy (Fig. 2). Using dual-color labeling with established cellular markers, we found that PMAT expression is specifically restricted to podocytes (Fig. 3). There is no significant staining in mesangial or glomerular endothelial cells (Fig. 3). The specific expression of PMAT in podocytes is rather unexpected. Originally, we suspected a tubular expression for PMAT, since other renal OCTs, including OCT2 and MATE1, are primarily localized to tubular epithelial cells (6, 12, 15). The lack of PMAT expression in the nephron tubules suggests that this transporter plays a different role in the kidney and is unlikely to be involved in tubular transport of organic solutes.

The physiological function of PMAT in podocytes is unclear, but may be related to monoamine signaling pathways in the kidney. It is well-recognized that the catecholamine dopamine plays important roles in blood pressure regulation by modulating renal blood flow, glomerular filtration rate, and epithelial sodium transport (22, 29). Dopamine is synthesized in the proximal tubule and is released as a paracrine and autocrine hormone to act on different receptors in the renal tubules, glomeruli, and renal blood vessels (2, 29). Dopamine exerts profound effects on renal hemodynamics in a dose-dependant manner (22, 29). At low concentrations, dopamine activates D1 and D2 receptors and dilates interlobular arteries and both the afferent and efferent arterioles, resulting in an increased renal blood flow and a marked natriuretic response. At high concentrations, however, dopamine induces renal vasoconstriction via activation of the α-adrenergic receptors (22, 29). We previously demonstrated that dopamine is a preferred physiological substrate of PMAT (5). It is possible that PMAT-mediated cellular uptake may be involved in dopamine regulation of renal hemodynamics by influencing extracellular dopamine concentrations at the receptor sites. However, it should be pointed out that dopamine-mediated renal response is very complex, and PMAT may represent only one checkpoint of a very complex and elaborate signaling network. Given the many unknowns of dopamine receptor function in podocytes and possible presence of other functionally redundant transporters, the exact role of PMAT in podocytes and renal physiology awaits further investigation.

Podocytes are highly specialized, terminally differentiated cells that form a critical part of the glomerular filtration barrier. The molecular mechanism leading to the podocyte-specific expression of PMAT in the kidney is unknown. Recently, a number of podocyte-specific transcription factors, such as Pod1, Pax2, WT1, and maf-1, have been identified in developing and mature podocytes (18). These transcription factors are thought to orchestrate the specification and differentiation of the podocyte lineage during kidney development. It is possible that one or more of these transcription factors is responsible for the specific expression of PMAT in the podocyte. Recently, Li et al. (11) showed that PMAT is a direct target of the transcription factor, EWS/WT1, a fusion oncogene consisting of the NH₂-terminal domain of the Ewing's sarcoma gene (EWS) and the COOH-terminal DNA binding domain of the Wilms’ tumor suppressor WT1. WT1 has several functional spliced variants and plays a key role in podocyte lineage determination during kidney development (10, 18). It would be interesting to see whether WT1 and its spliced variants can directly activate the transcription of PMAT.

Podocyte injury is the most common cause of diseases affecting the glomerulus. The clinical signature of podocyte injury is proteinuria, with or without loss of renal function owing to glomerulosclerosis (21, 25). PAN is an experimental toxin that causes severe glomerulopathy and massive proteinuria in animal models. When injected into rats, PAN specifically induces podocyte damage in the kidney, resulting in foot process effacement and focal detachment of these cells from the glomerular basement membrane (16, 19). It is generally believed that overproduction of reactive oxygen species and damage of intracellular macromolecules are responsible for PAN-induced toxicity. However, it is not clear how this positively charged toxin enters podocytes to exert its cytotoxic effect. In this study, we demonstrated that PMAT is specifically expressed in podocytes in human and rat kidneys. Using a gain-of-function cell culture model, we showed that overexpression of PMAT on cell surface enhanced cells’ sensitivity to PAN (Fig. 4). Blocking PMAT activity by decynium 22 protected cells from PAN toxicity (Fig. 4B). Direct uptake studies further demonstrated that PAN is indeed transported by PMAT. Together, these data strongly suggest that PMAT-mediated cellular entry of PAN may represent an early event underlying PAN-induced cytotoxicity. The specific localization of PMAT in podocytes may, in part, explain the specific toxicity of PAN toward these cells.

GRANTS

This work was supported by National Institutes of Health Grant GM-066233 and the Provost's Research Bridge Funding from the University of Washington.

Present address of L. Xia: U. S. Food and Drug Administration, Center for Drug Evaluation and Research, 7520 Standish Place, Rockville, MD 20855.