Role of uptake transporters OAT4, OATP2A1, and OATP1A2 in human placental bio-disposition of pravastatin

Abstract

Pravastatin is currently under evaluation for prevention of preeclampsia. Factors contributing to placental disposition of pravastatin are important in assessment of potential undesirable fetal effects. The purpose of this study was to identify the uptake transporters that contribute to the placental disposition of pravastatin. Our data revealed the expression of organic anion transporting polypeptide 1A2 (OATP1A2) and OATP2A1 in the apical, and OATP2B1 and OATP5A1 in the basolateral membranes of the placenta, while organic anion transporter 4 (OAT4) exhibited higher expression in basolateral membrane but was detected in both membranes. Preloading placental membrane vesicles with glutarate increased the uptake of pravastatin suggesting involvement of glutarate-dependent transporters such as OAT4. In the HEK293 cells overexpressing individual uptake transporters, OATP2A1, OATP1A2 and OAT4 were determined to accept pravastatin as a substrate at physiological pH, while the uptake of pravastatin by OATP2B1 (known to interact with pravastatin at acidic pH) and OATP5A1 was not detected at pH 7.4. These findings led us to propose that OATP1A2 and OATP2A1 are responsible for the placental uptake of pravastatin from the maternal circulation, while OAT4 mediates the passage of the drug across placental basolateral membrane in the fetal-to-maternal direction.

INTRODUCTION

Preeclampsia complicates 3–8% of pregnancies worldwide and remains a major cause of maternal and neonatal morbidity and mortality. Preclinical evidence indicated that statins, specifically pravastatin, reverse or prevent various pathophysiological changes associated with preeclampsia.^1–5

The safety and efficacy of pravastatin in prevention and management of preeclampsia in pregnant patients with high-risk for this condition was demonstrated in a pilot double-blind, placebo-controlled, randomized multicenter study conducted by the national Institute of Child health and development (NICHD) network of Obstetrics Fetal Pharmacology research center (OPRC) (Clinicaltrials.gov identifier NCT01717586)⁶ which prompted the launch of a Phase III clinical trial (ClinicalTrials.gov Identifier: NCT03944512).

Human placenta represents a functional barrier between the maternal and fetal circulations and is composed of a maternal-facing apical membrane (AM), fetal-facing basolateral membrane (BM) and fetal endothelial cells. The hydrophilic properties of pravastatin would preclude its transfer across biological membranes, including placenta, by passive diffusion. However, the ex vivo placental perfusion experiments showed that 18 ± 4% of the drug cross the placenta from the maternal to fetal direction.⁷ Furthermore, the placental transfer of pravastatin was asymmetric favoring fetal-to-maternal transport suggesting the involvement of placental transporters.⁷ Recent report from our laboratory indicated that BCRP and MRP1 are the major efflux transporters involved in the placental transport of pravastatin.⁸

The organic anion transporters (OATs) and organic anion transporting polypeptides (OATPs) are the members of two membrane protein families of uptake transporters that belong to the superfamily of solute carrier transporters (SLC).⁹ Pravastatin is present as an anion at physiological or weakly acidic pH (pKa_acid of pravastatin is 4.2), thus it could serve as a suitable substrate for the OATs and OATPs.¹⁰ Indeed, previous studies reported that pravastatin is a substrate of the SLCO/SLC21 family members OATP1B1 (SLCO1B1)^11–13, OATP1B3 (SLCO1B3)^11,12,14, OATP1A2 (SLCO1A2)^15,16, OATP2B1 (SLCO2B1)^15,17–19, and of the SLC22 family members OAT3 (SLC22A8) and OAT4 (SLC22A11)²⁰, while some of these transporters are unlikely to transport pravastatin at neutral pH.^15,17–19 While OATP1B1 and OATP1B3 are key transporters in the hepatic uptake of pravastatin in vivo, OAT3 and OAT4 are involved in the drug renal uptake in vivo. The involvement of OATP1B1 and OATP1A2 in the intestinal absorption of pravastatin in vivo was refuted recently based on the new evidence regarding their intestinal expression.^21,22

Among the transporters that are known to accept pravastatin as a substrate, OATP2B1 and OAT4 were detected in the basolateral surface of the syncytiotrophoblast, ^23,24 while the expression of OATP1A2 was detected in the apical membrane (AM) of the placenta.²⁵ Other members of OATP subfamily of transporters that are expressed in human placenta could also contribute to the bio-disposition of pravastatin. Phillips et al, 2014 reported placental expression of OATP2A1 detected by means of immunohistochemistry, although specific localization of OATP2A1 in the placental tissue was not defined.²⁶ Immunohistochemistry revealed the expression of OATP4A1 in the apical surface of the placental syncytiotrophoblast²⁷, while the presence of OATP3A1 in the placenta was detected on mRNA level.²⁸

The placental membrane transporters can either prevent or promote the transfer of the drugs across the placental barrier depending on their function (uptake or efflux) and localization in either AM or BM of syncytiotrophoblast. Therefore, the aim of this work was to determine the uptake transporters participating in the placental disposition of pravastatin by using human placental membrane vesicles and human embryonic kidney (HEK) 293 cells overexpressing select uptake transporters.

MATERIALS AND METHODS

Chemicals and reagents

[³H]Pravastatin calcium salt (specific activity, 20 Ci/mmol), [³H]Estrone sulfate (E3S) (specific activity, 50 Ci/mmol), and [³H]Prostaglandin-E2 (PE2) (specific activity, 167.9 Ci/mmol) were purchased from American Radiolabeled Chemicals (St. Louis, MO, USA). Unlabeled pravastatin sodium salt and pravastatin-d₃ sodium were purchased from Toronto Research Chemicals Inc. (Toronto, Canada). All other chemicals and reagents were purchased from Fisher Scientific (Pittsburgh, USA) or Sigma-Aldrich (St. Louis, MO, USA), unless stated otherwise, and were of the highest grade available.

Clinical materials

Placentas from uncomplicated term pregnancies between 37–0/7 and 39–6/7 weeks of gestation were obtained immediately following vaginal or abdominal deliveries from the Labor and Delivery Ward of the John Sealy Hospital, the teaching hospital of the University of Texas Medical Branch, Galveston, Texas, in accordance with a protocol approved by the Institutional Review Board. Placentas from patients with history of systemic diseases, maternal infection, and drug or alcohol abuse during pregnancy were excluded from this study.

Isolation and characterization of human placental AM and BM vesicles

The vesicles from AM and BM of the human placenta were prepared and characterized as described in details in our previous reports.^8,29 Both placental AM and BM vesicles are 80% right-side-out and carry out the carrier-mediated influx of the substrates that involves membrane uptake transporters.⁸ The rest 20% of the vesicles represent the inside-out vesicles, in which the intracellular side of the transporters is facing outward.⁸ Since the uptake buffer used in the current study was devoid of adenosine triphosphate (ATP), the contribution of ABC efflux transporters such as BCRP and MRP1 in the vesicular transport of pravastatin was negligible under the current experimental condition.

Western blot analysis

Immunoblot analysis was performed using BM and AM placental vesicles following the established protocol.⁸ The following primary antibodies were utilized: rabbit polyclonal anti-OATP2B1 (83532), OATP5A1 (108288), and OATP3A1 (84315) diluted 1:200, and rabbit monoclonal anti-Cytokeratin 7 (ab181598) diluted 1:30, and CD-31 (ab134168) diluted 1:200 (Abcam, Cambridge, MA, USA); rabbit polyclonal anti-OATP2A1 (PGT12-A) and anti-OAT4 (OAT41-A) diluted 1:50 (Alpha Diagnostics Intl., Inc, San Antonio, TX, USA); rabbit polyclonal anti-OATP4A1 (sc-134866), 1:200, and OATP1A2 (sc-365007), 1:100, and mouse monoclonal antibodies raised against placental alkaline phosphatase (PLAP),1:200 (sc-47691) (Santa Cruz Biotechnology, Inc (Dallas, TX, USA). The corresponding secondary antibodies were either goat anti-rabbit or anti-mouse Ig CY5 (1:5,000 to 1:20,000 dilution). The signal was detected using Typhoon FLA 9000 (GE Healthcare, Piscataway, NJ).

Glutarate-stimulated uptake of pravastatin by placental AM and BM vesicles

30 μg of AM or BM vesicles were resuspended in the uptake buffer (pH 7.4) containing 100 mM NaCl, 35 mM sucrose and 25 mM HEPES-Tris, with the addition of 10mM glutarate, and incubated at 4°C for 1 h for glutarate preload following the protocol established previously.³⁰ No glutarate was added to the control samples. After 10 min preincubation at 37°C, the uptake was initiated by addition of the uptake buffer containing 1 μM [³H]Pravastatin at 37°C. The reaction was terminated after 1 min by addition of 3 mL of ice cold uptake buffer followed by rapid filtration through a Whatman glass fiber filter strip (pore size 0.7 μm; Whatman, Clifton, NJ, USA) using a Brandel Cell Harvester (Brandel, Gaithersburg, MD, USA). The filters were submerged in the scintillation cocktail, and radioactivity measured as described in Samples Analysis section. The incubation time and protein amount were selected to measure the uptake within the linear range and ensure the optimal glutarate-dependent accumulation of pravastatin in the vesicles in relation to the background uptake, respectively. The concentration-dependence of glutarate-dependent uptake of pravastatin into the placental BM vesicles was tested using 0.5 μM, 1 μM and 2 μM [³H]Pravastatin. The uptake experiments were also conducted at 4°C to confirm the involvement of carrier-mediated transport. To measure the binding of pravastatin to the filters, the vesicles were omitted from the reaction mixtures.

Cell culture

The human embryonic kidney HEK293 cell lines overexpressing OATP2A1, OATP3A1, OATP4A1, OATP5A1 and the control cells (HEK293-VC, empty vector: pcDNA3.1(+)-G418) were provided by Dr Jörg König.^31,32 Tetracycline-inducible OAT4-Expressing cell line (T-REx-OAT4–293) were a gift from Dr Masatoshi Tomi.³³ Corning® TransportoCells™ HEK293 transiently expressing OATP2B1 and OATP1A2 and the respective control cells were purchased from Corning Life Sciences through Fisher Scientific.

The cell cultures were maintained at 37°C, 95% relative humidity and 5% CO₂. HEK293 overexpressing OATP2A1, OATP3A1, OATP4A1, and OATP5A1 and HEK293-VC cells were cultured in minimum essential medium (MEM) containing 10% fetal bovine serum (FBS), 800 μg/mL geneticin, 100 U/mL penicillin, and 100 μg/mL streptomycin.^31,32 T-REx-OAT4–293 cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin, 15 μg/mL blasticidin and 400 μg/mL hygromycin B.³¹ The cells were routinely subcultured using TripLE dissociating reagent. The expression of OAT4 was induced by culturing the cells with 1 μg/mL tetracycline.³³ HEK293 cells overexpressing OATP2B1 and OATP1A2 and the respective control cells were cultivated in DMEM supplemented with MEM nonessential amino acids and 10% FBS as recommended by the manufacturer.

Transport studies

For the cellular uptake of pravastatin, [³H]Pravastatin (0.65 μM - 0.75 μM), or unlabeled pravastatin, or a mixture of radiolabeled and unlabeled drug were used. The uptake into T-REx-OAT4–293 cells was carried out following the protocol described previously in details.³³ In brief, 100,000 cells/cm² were seeded in the 12- or 24-well plates coated with poly-D-lysine and grown for 24 h, after which the medium was replaced with the fresh medium supplemented with 1 μg/mL tetracycline 24 h prior to the experiment; no tetracycline was added to the medium of control wells. The uptake reaction was carried out using extracellular fluid (ECF) buffer (pH 7.4) containing 122 mM NaCl, 25 mM NaHCO₃, 3 mM KCl, 1.4 mM CaCl₂, 2 mM MgSO₄, 0.4 mM K₂HPO₄, 10 mM D-glucose, 10 mM HEPES, with or without addition of 5 mM glutarate in the preincubation step. For the chloride-free uptake condition, the chloride in the buffer was replaced with equimolar gluconate. The time-dependence of OAT4-mediated uptake was determined in the absence or presence of extracellular chloride.³³

To study the OAT4-medited efflux, the T-REx-OAT4–293 cells (with or without tetracycline pretreatment) were first preloaded with 4 μM of unlabeled pravastatin in the chloride-free ECF buffer by incubation for 30 s at 37 °C. Then the cells were incubated with either chloride-free or chloride-containing ECF buffer for 15 s, 30 s, and 60 s, or with chloride-free buffer containing 100 μM of E3S for 30 s. The cellular retention of pravastatin at each time point was calculated as % of the preloaded (accumulated by the cells via the initial uptake).

The uptake experiments with HEK293-OATP1A2, HEK293-OATP2B1 and the respective control cells were performed following the manufacturers protocol using Hank’s balanced salt solution (HBSS) containing 0.5mM HEPES (pH 7.4). For OATP1A2-dependent uptake, the experiments were also conducted using HBSS-MES (pH 6) as recommended by the manufacturer. The HEK293-OATP2A1 and HEK293-OATP5A1 and their respective HEK293-VC (control) were seeded on the 12-well plates coated with poly-D-lysine at the density 143,000 cells/cm² and used for the experiments the next day using HBSS-HEPES (pH 7.4) following the assay protocol for the HEK293-OATP2B1 and HEK293-OATP1A2 cells.

The uptake and efflux reactions were terminated at designated times by washing the wells three times with cold corresponding uptake buffer, the cells were lysed with deionized water at room temperature, and lysates were processed as described below. The probe substrates were used as follows: [³H]E3S for OAT4, OATP1A2 and OATP2B1, [³H]PE2 for OATP2A1, and [³H]PE2, [³H]E3S and sodium fluorescein for OATP5A1.^34,35

Samples analysis

The protein concentration of the cellular lysate in each well was determined using Bradford protein assay reagent (Bio-Rad laboratories, Hercules, CA, USA) using bovine serum albumin as a standard. The amount of radiolabelled substrate was determined by Tri-Carb 4910 TR 110V Liquid Scintillation Counter (PerkinElmer, Waltham, MA, USA). Sodium fluorescein was determined by measuring fluorescence of the cellular lysate at λ_ex=485 nm and λ_em=528 nm using an FLX800 microplate reader (BioTek Instruments, Inc., Winooski, VT, USA). Data acquisition was performed using Gen5™ software (BioTek Instruments, Inc.). The unlabeled pravastatin was quantified using liquid chromatography-mass spectrometry method (LC-MS/MS) following the method reported previously with modifications (Supplementary Methods).³⁶

For the uptake into AM and BM placental vesicles, the uptake data were normalized to the protein amount, and the net glutarate-dependent uptake was calculated as a difference in the radiolabaled substrate quantified in the glutarate-preloaded vesicles and non-preloaded (control). For the cellular uptake studies, the uptake data were normalized to the protein determined for each well, and the transporter-mediated uptake was calculated as a difference in the uptake activity between the transporter-expressing cells and the respective control cells.

Kinetic analysis of OAT4-, OATP1A2- and OATP2A1-mediated transport of pravastatin and the effect of known inhibitors

The kinetic parameters for saturable uptake of pravastatin (equivalent of the Michaelis-Menten constant (K_t) and maximal uptake velocity (V_max) were derived using SigmaPlot version 14.5 (Systat Software, Inc., San Jose, CA USA) by fitting the data into the Michaelis-Menten-type non-linear curve. The elimination rate constant (K_e) was derived from the exponential regression using SPSS software (IBM SPSS Statistics for Macintosh, Version 25.0. Armonk, NY: IBM Corp). The influence of the following inhibitors on the uptake of 2 μM pravastatin was determined: diclofenac (20 μM and 200 μM) for OATP2A1, and E3S (100 μM) for OAT4.^33,34 For each condition, the inhibitors were added to the uptake buffer containing pravastatin. OATP1A2 was not tested with inhibitors because the expression system was commercially developed and validated. The effect of inhibitor was calculated as percentage of the uptake from the control experiments in the absence of inhibitors (100% uptake).

Statistical analysis

Statistical analysis was performed using SPSS software (IBM SPSS Statistics for Macintosh, Version 25.0. Armonk, NY: IBM Corp). Unpaired T-test or Mann-Whitney U test were used for the comparison between two groups. One-way ANOVA followed by Dunnett’s post hoc test was utilized for multiple comparisons. Results were considered significant if P < 0.05. All data are presented as mean ± SEM, unless indicated otherwise.

RESULTS

Uptake of pravastatin by placental membrane vesicles

In the absence of preloaded glutarate, there was no difference in the uptake of 1 μM [³H]Pravastatin between the placental BM and AM vesicles at 37°C (Fig 1A). Preloading with glutarate increased the uptake of 1 μM [³H]Pravastatin into BM vesicles (P < 0.05, Fig 1A), resulting in the net glutarate-stimulated uptake of 320 ± 42 fmol/mg protein*min⁻¹. On the other hand, the glutarate-stimulated uptake of pravastatin into AM vesicles was observed to be insignificant (Fig 1A).

FIGURE 1.

Glutarate-stimulated uptake of pravastatin by AM and BM vesicles prepared from human placenta. The uptake of 1μM [³H]Pravastatin was determined using pooled membrane vesicles prepared from 14 term placentas obtained from uncomplicated pregnancies. (A) The uptake of pravastatin into glutarate-preloaded and non-preloaded membrane vesicles. (B) The effect of temperature and substrate concentration on the net (glutarate-dependent) uptake of [³H]Pravastatin into BM vesicles. The net uptake is reported as the difference between the uptake into glutarate-preloaded and non-preloaded BM vesicles. Data presented as mean ± SEM (n=8–9 (A) and n=2–9 (B), each point in duplicates or triplicates). AM, apical membrane; BM, basolateral membrane

The uptake of [³H]Pravastatin by the placental vesicles in the absence of preloaded glutarate at 4°C did not differ from that at 37°C (data not shown), suggesting that the vesicular uptake is not carrier-mediated without the driving force of glutarate gradient. While we did not detect any glutarate-stimulated uptake of pravastatin into the BM vesicles at the pravastatin concentrations below 1 μM at 37°C, a two-fold increase of the substrate concentration from 1 μM to 2 μM lead to a proportional increase of the glutarate-stimulated uptake (Fig 1B). Meanwhile, the net uptake of the drug at 4°C remained negligible in the range of the concentrations tested (FIG 1B). Further escalation of pravastatin concentration in the reaction mixture was restricted by a high content of organic solvent in the substrate stock solution precluding kinetic analysis of the uptake.

Expression of OATP and OAT4 proteins in the placental membrane vesicles

Immunoblot analysis showed the expression of OATP1A2 and OATP2A1 in AM vesicles, OATP5A1 and OATP2B1 in BM vesicles. The expression of OAT4 was detected, but the band intensity of OAT4 was higher in BM than in AM (Fig 2). Cytokeratin-7 (marker for epithelial cells) and PLAP (marker for the placental trophoblast) were expressed in both BM and AM. For Cytokeratin-7, the immunostaining revealed two bands of 42 and 52 kDA, which was consistent with the antibody description and reflected different isoforms of this protein. CD-31 (marker for endothelial cells) was expressed in the BM, but not AM confirming the absence of cross-contamination between the preparations. We confirmed the tetracycline-induced expression of OAT4 in the T-REx-293-OAT4 cells, and the expression of OATP1A2, OATP2B1, OATP2A1 and OATP5A1 in the corresponding cell lines (Supplementary Fig S1).

FIGURE 2.

Expression of OATP1A2, OATP4A1, OATP5A1, OATP3A1, OATP2A1, OATP2B1, and OAT4 proteins in vesicles prepared from AM and BM of human placenta. Lane 1 - positive control (cellular homogenate from HeLa cells, for OATP1A2 and OATP2A1; from HEK293 cells, for OATP4A1, OATP5A1, and OATP3A1; Corning® TransportoCells™ HEK293-OATP2B1, for OATP2B1), lanes 2–5- vesicles prepared from AM of human placenta (n=4), lanes 6–9- vesicles prepared from BM/villi of human placenta (n=4). Both AM and BM preparations showed immunoreactivity towards cytokeratin-7 (CK-7) and placental alkaline phosphatase (PLAP), while BM preparations were immunoreactive towards CD-31. AM, apical membrane; BM, basolateral membrane

The expression of OATP4A1 and OATP3A1 were detected in both BM and AM vesicles (Fig 2). The multiple bands of OATP4A1 revealed in the HeLa cells (positive control) and BM vesicles probably represent different glycosylated states of OATP4A1 (Fig 2). However, it should be noted that the band pattern for OATP3A1 and OATP 4A1 was similar in the corresponding transporter-expressing cells and the control (empty vector) cells (Supplementary Fig S1).

Transport studies using cell lines overexpressing OAT4, OATP2A1, OATP1A2, OATP2B1 and OATP5A1

While the mechanism of OATP-mediated uptake of substrates is not completely elucidated, several studies suggest the involvement of electroneutral exchangers that could be transporter-specific.^10,37 Moreover, there is a great overlap in the inhibitors specificity to different OATPs and OATs.^10,34 Taken together, these factors limited the use of the placental membrane vesicles as an experimental model in assessment of the contribution of each individual placental uptake transporter to the influx of pravastatin. Hence, cell lines overexpressing the uptake transporters were used as an in vitro model to study the interaction of pravastatin with each of the transporter. Per the guideline of the International Transporter Consortium, the uptake of a substrate into the transfected cells has to be higher than of the uptake into the mock-transfected cells (control), with the statistical significance attained, and it has to be inhibitable by a known inhibitor of the transporter.³⁸

The uptake of pravastatin into the cells expressing OAT4, OATP2A1, and OATP1A2 exceeded that of the corresponding control cells at pH 7.4 (Table 1), showing that pravastatin is a substrate of OATP1A2 and OATP2A1 and confirming previous reports on the ability of OAT4 to accept pravastatin as a substrate.²⁰ We confirmed the functional uptake by these transporters using the respective probe substrates (data not shown).

Table 1.

Pravastatin uptake by OAT4, OATP2A1, OATP1A2, OATP2B1 and OATP5A1 proteins overexpressed in the HEK293 cells.

Transporter	Pravastatin, μM	Experimental condition	Uptake, pmol/mg protein/min
			Transporter-expressing cells	Control	Net (transporter-mediated)
OAT4	4		4.09 ± 0.46*	1.24 ± 0.20	2.86 ± 0.27
	4	Cl-free	8.66 ± 0.80*, #	1.51 ± 0.24	7.15 ± 0.59#
	4	GA	7.04 ± 0.97*, #	1.87 ± 0.25#	5.17 ± 0.84#
OATP2A1	1		0.46 ± 0.06*	0.26 ± 0.06	0.27 ± 0.03
OATP1A2	0.75a		1.73 ± 0.31*	0.66 ± 0.15	1.06 ± 0.32
OATP2B1	0.75a		0.71 ± 0.13	0.61 ± 0.13	0.10 ± 0.16
OATP5A1	10		1.85 ± 0.18	1.62 ± 0.18	0.23 ± 0.01

The uptake assays were conducted using HEK293 cells stably overexpressing OATP2A1 and OATP5A1, HEK293 cells transiently overexpressing OATP1A2 and OATP2B1, and tetracycline-inducible T-REx-OAT4-293 cells. Pravastatin or [³H]Pravastatin were utilized. The uptake was conducted at 37°C for 30 s (OAT4, OATP1A2) or 1 min (OATP2A1, OATP2B1, OATP5A1) at neutral pH in the presence of extracellular chloride and without glutarate preload, unless noted otherwise in the experimental condition. The transporter-mediated uptake was calculated by subtracting the uptake by corresponding control cells from that of transporter-expressing cells. Data presented as mean ± SEM, n=3 experiments, each data point in duplicates or triplicates.

^*P<0.05, statistical significance attained in comparison of the uptake into the transporter-expressing cells with the uptake into the corresponding control cells

^#P<0.05, statistical significance attained for the uptake of 4 μM of pravastatin in the absence of extracellular chloride and with glutarate preloading in the T-REx-OAT4-293 cells.

GA, glutaric acid preload; Cl-free, chloride-free uptake buffer

^a[³H]Pravastatin was used, specific activity 20 Ci/mmol

Glutarate preload and the use of chloride-free uptake buffer stimulated the uptake of 4 μM pravastatin into OAT4-expressing cells resulting in the increase of OAT4-dependent uptake (Table 1). Further uptake experiments with T-REx-OAT4–293 cells were conducted with glutarate preload using chloride-containing uptake buffer (unless specified).

Consistent with the previous reports^15,17–19, we did not detect any OATP2B1-mediated uptake of pravastatin at pH 7.4 (Table 1), while the uptake of the probe substrate E3S by OATP2B1–expressing cells was several folds higher than that of the control cells (data not shown). Neither pravastatin, nor the probe substrates, namely [³H]E3S, [³H]PE2, and sodium fluorescein were found to interact with OATP5A1 (Table 1).

Fore further OAT4 and OATP1A2 experiments, 30 s reaction duration was used for the uptake to be within the linear range (Supplemental Fig S2&S3). The OATP2A1-mediated uptake of pravastatin increased over a period ranging from 10 s to 5 min, however the increase was not linear (data not shown), therefore 1 min reaction duration was used in further experiments and considered the initial uptake phase.

The OATP2A1 inhibitor diclofenac (20 μM) reduced the uptake of 2 μM pravastatin into the OATP2A1-expressing cells (P<0.05), which resulted in a 50% decrease of the net, OATP2A1-dependent, uptake (Supplementary Fig S4). Further inhibition by higher concentrations (200 μM) of diclofenac did not result in a statistically significant decrease of the uptake of pravastatin (Supplementary Fig S4). The use of E3S (200 μM), an OAT4 prototypical substrate, resulted in a complete inhibition of OAT4-mediated uptake of 2 μM pravastatin (P<0.05, Supplementary Fig S5)

Kinetics of the OAT4, OATP2A1 and OATP1A2-mediated uptake of pravastatin

The uptakes of pravastatin by OAT4, OATP2A1 and OATP1A2 were concentration-dependent and saturable. OAT4-mediated uptake was fitted into Michaelis-Menten curve and exhibited biphasic profile as was evident form the shape of the Eadie-Hofstee plot (Fig 3, Table 2). The apparent K_t and V_max values for the lower concentration range were several folds lower than the kinetic parameters estimated for the higher concentration range (Table 2) suggesting the presence of low- and high-affinity binding sites responsible for the interaction of OAT4 with pravastatin. The data on the concentration-dependent OATP2A1- and OATP1A2-mediated uptake of pravastatin were also fitted into the Michaelis-Menten curve (Fig 4&5, Table 2), since the alternative kinetic models did not result in a superior goodness of fit (judged by either R² or Akaike information criterion). Noteworthy, the Eadie-Hofstee plot for the OATP1A2-dependent uptake at extracellular pH 6.0 illustrated a better fit of the data into a typical hyperbolic kinetics (Supplementary Fig S6).

FIGURE 3.

Concentration dependence of OAT4-mediated uptake of pravastatin. The uptake of pravastatin by T-REx-OAT4–293 cells was measured at 37°C in the presence of chloride for 30 s. OAT4-mediated uptake of pravastatin was calculated by subtracting the uptake into tetracycline-untreated (control) cells from that of the tetracycline-treated (OAT4-expressing) cells. The data were fitted into the Michaelis-Menten curve (main). Eadie-Hofstee plot is shown (inset). Each point represents the mean ± SEM (n=4 experiments, each point in duplicates).

TABLE 2.

Apparent kinetic parameters for OATP2A1-, OAT4- and OATP1A2-mediated uptake of pravastatin at physiological pH in vitro.

Transporter	Kt, μM	Vmax, pmol/mg protein*min−1
OAT4	30.5 ± 75.0	42.0 ± 93.4
	295 ± 99.0	345 ± 54.8
OATP2A1	188±109	73.8 ± 19.2
OATP1A2	1.60 ± 0.97	4.08 ± 0.72

The uptake of pravastatin by OAT4, OATP2A1 and OATP1A2 was measured at physiological pH at 37°C using and tetracycline-inducible T-REx-OAT4-293 cells, HEK293 cells transiently overexpressing OATP1A2, and HEK293 cells stably overexpressing OATP2A1 and their respective control cells. The transporter-mediated uptake was calculated by subtracting the uptake by corresponding control cells from that of transporter-expressing cells. The kinetic parameters were estimated by non-linear curve fitting using SigmaPlot version 14.5 (Systat Software, Inc., San Jose, CA USA) and presented as mean ± SEM. The concentration-dependence of OAT4-, OATP2A1- and OATP1A2-mediated uptake of pravastatin is shown in Figures 3, 4 and 5, respectively. K_t, Michaelis constant equivalent; V_max, maximum velocity.

FIGURE 4.

Concentration dependence of OATP2A1-mediated uptake of pravastatin. The uptake of pravastatin by HEK293-OATP2A1 and HEK293-CV (control) cells was measured at 37°C in the presence of chloride for 1 min. OATP2A1-mediated uptake of pravastatin was calculated by subtracting the uptake into control cells from that of OATP2A1-expressing cells. Data were fitted into the Michaelis-Menten curve (main). Eadie-Hofstee plot is shown (insets). Each point represents the mean ± SEM (n=5 experiments, each point in triplicates).

FIGURE 5.

Concentration dependence of OATP1A2-mediated pravastatin uptake. The uptake of pravastatin by HEK293-OATP1A2 and HEK293-CV (control) cells was measured at 37°C in the presence of chloride for 30 s. OATP1A2-mediated uptake of pravastatin was calculated by subtracting the uptake into control cells from that of OATP1A2-expressing cells. Data were fitted into the Michaelis-Menten curve (main). Eadie-Hofstee plot is shown (inset). Each point represents the mean ± SEM (n=2 experiments, each point in duplicates).

The apparent K_t and V_max values for the OATP1A2-mediated uptake of pravastatin at pH 6.0 were 23.6 ± 2.87 μM and 361 ± 29.3 pmol/mg protein*min⁻¹, respectively. The greater V_max/K_t at pH 6.0 confirmed the stimulating effect of pH on OATP1A2-dependent uptake of pravastatin.¹⁵

OAT4-mediated efflux of pravastatin

OAT4 operates as an asymmetric exchanger³⁹, therefore we also examined whether OAT4 is able to eliminate pravastatin from the cells. First, we utilized chloride as a counter-ion for the exchange transport as chloride’s physiological gradient is directed inwardly (120 mM extracellular and 5 mM intracellular).⁴⁰ Second, we utilized E3S, which represents a fetal-derived steroidal metabolic product and is transported into the syncytiotrophoblast by OATP2B1 and OAT4 localized in the placental basal membrane.^41,42 The efflux of pravastatin would inversely correlate with the levels of remaining intracellular pravastatin.

The efflux of pravastatin from OAT4-expressing cells was increased by an inwardly directed chloride gradient at higher concentration of extracellular chloride compared to an outwardly directed chloride gradient in a chloride-free experimental condition (no extracellular chloride and 5 mM intracellular chloride) at 60 s (29 ± 1.4% of remaining preloaded pravastatin vs 39 ± 2%, (P<0.05) (Fig 6A). In the chloride-free condition, the estimated K_e of the pravastatin efflux from the OAT4-expressing and control cells were 2.1 ± 0.5 (10⁻²s⁻¹) and 3.6 ± 2.1 (10⁻²s⁻¹), respectively. In the presence of extracellular chloride, the estimated K_e values were 2.6 ± 0.6 (10⁻²s⁻¹) and 3.0 ± 1.0 (10⁻²s⁻¹), respectively. While the direction of chloride gradient had no effect on pravastatin efflux by either of OAT4-expressing and control cells, the apparent K_e values were slightly higher for the drug efflux from the control cells than from OAT4-expressing cells (statistical significance not attained). Using E3S (100 μM) as a counter ion resulted in the decreased efflux of pravastatin from OAT4-expressing cells, while pravastatin retention was similar in the OAT4-expressing and control cells with or without E3S (Fig 6B).

FIGURE 6.

The effect of extracellular chloride (A) and E3S (B) on the efflux of pravastatin by OAT4. (A) The efflux of preloaded pravastatin from tetracycline-treated (circle) and untreated (square) T-REx-OAT4–293 cells in the presence (closed symbols) or absence (open symbols) of 128 mM chloride. *P<0.05, significant difference between the pravastatin retained by OAT4-expressing cells in the absence and presence of chloride. (B) The effect of extracellular E3S (100 μM), on the efflux of preloaded pravastatin from tetracycline-treated and untreated T-REx-OAT4–293 cells in a chloride-free condition. The remaining pravastatin was quantified after 30 s of efflux in the absence (open bars) or presence (closed bars) of extracellular E3S. *P<0.05 Data are expressed as % of preloaded pravastatin remaining in the cells and shown as mean ± SEM (n=3 experiments, each data point in duplicates).

DISCUSSION

The goal of this investigation was to determine the role of uptake membrane transporters in the biodisposition of pravastatin across human placenta. One of the most studied uptake transporter in the placenta is OAT4, a member of SLC22 family of transporters, which was shown to take part in the transport of xenobiotics and endogenous substrates across the trophoblast tissue.^30,33 While previous reports indicated that OAT4 is expressed in the BM of placental syncytiotrophoblast,²⁴ our western blot analysis utilizing placental subcellular membrane preparations revealed its expression in both AM and BM. As the source of antibodies was the same in both studies, the discrepancy in the OAT4 placental localization could be explained by the difference in immunoreaction between the immunohistochemistry and western blot analysis.

The proposed mechanism of OAT4-mediated uptake involves an antiport of a counter-ion, such as hydroxyl and dicarboxilate ions (e.g. a α-ketoglutarate).^10,43 While sodium-proton exchanger maintains the outwardly directed gradient of hydroxyl ions, the α-ketoglutarate gradient is supported by the secondary active sodium-dicarboxylate co-transporter, for which the sodium gradient is maintained by the primary active Na⁺/K⁺ATPase.¹⁰ Moreover, a recent study suggested that OAT4-mediated uptake of substrates by placenta is coupled with the efflux of glutamate.⁴⁴ The reuptake of glutamate by placenta is mediated by sodium-dependent excitatory amino acid transporter, which maintains the outwardly directed gradient of glutamate with 4 mM intracellular glutamate concentration against its 0.05 mM concentration in plasma.⁴⁴

According to previous reports, glutarate trans-stimulated the uptake of 16α-hydroxydehydroepiandrosterone and olmesartan into placental BM vesicles resulting in a 56% and 33% increase of the uptake, respectively, which was attributed to the involvement of OAT4 in the transport of these substrates.^30,33 In our study, we observed a modest trans-stimulatory effect of glutarate on the uptake of pravastatin into the placental BM vesicles by 18% (Fig 1A). These results suggest the role of OAT4 in the placental uptake of pravastatin from the fetal circulation and are in agreement with the placental localization of OAT4 in the BM (Fig 2). It has to be noted that the contribution of dicarboxilate-dependent transporter(s) other than OAT4 cannot be ruled out. Meanwhile, despite the detected expression of OAT4 in the AM of the placenta (Fig 2), the trans-stimulating effect of glutarate on the uptake of pravastatin by placental AM vesicles was not evident under current experimental condition.

The OAT4-mediated uptake of pravastatin into the tetracycline inducible T-REx-293-OAT4 cells exhibited biphasic saturation kinetics suggesting the presence of more than one binding site, and low and high affinity transport activity (Fig 3). Nakagomi-Hagihara et al, 2007 utilized S2 cells expressing OAT4 and reported a single component saturation kinetics for pravastatin with the K_t = 257 μM²⁰, which was similar to the value determined in our work for the low affinity K_t = 295 ± 99 μM (Table 2). This discrepancy could be attributed to the different ranges of pravastatin concentrations used in that and our studies, although we could not rule out the innate differences between the expression systems.⁴³ The steady state plasma concentrations of pravastatin were in nanomolar range in pregnant women taking 10 mg daily dose of pravastatin in the course of a pilot clinical trial⁶, suggesting that high affinity binding site of OAT4 is likely involved in the OAT4-mediated uptake of the drug by placenta.

OAT4 is proposed to operate in an asymmetric manner because of its different substrate recognition characteristics on the intra- and extracellular sides.³⁹ Our data showed that glutarate (which intracellular concentration is higher than extracellular in physiological state)¹⁰ could serve as a counter-ion for OAT4-mediated uptake of pravastatin by the placenta (Table 1). Meanwhile, several studies reported that chloride, which physiological gradient is directed inwardly⁴⁰, could serve as a counterpart for the OAT4-mediated efflux.^33,39 The observed increase of the OAT4-dependent cellular uptake of pravastatin in the chloride-free condition in vitro (Table 1) could occur due to decrease (or abolishment) of OAT4-mediated efflux of pravastatin when the chloride gradient was directed outwardly (no extracellular chloride and 5 mM intracellular chloride). However, the chloride-driven efflux of pravastatin via OAT4 was not evident as the apparent K_e values were similar in the presence and absence of extracellular chloride. Of note, the apparent rates of the drug efflux were slightly higher in control cells than in the OAT4-expressing cells. It is possible that the extruded pravastatin undergoes immediate OAT4-mediated re-uptake, which could as well be facilitated by outwardly directed chloride gradient in the chloride-free experimental condition (Fig 6A). With the use of E3S as a counter-ion for the OAT4-mediated efflux of pravastatin, the retention of pravastatin by OAT4-expressing cells was increased (Fig 6B). A recent study suggested that the transport of E3S by OAT4 involves binding to the cavity on the carrier and partitioning the substrate into the membrane, which is different from the conventional translocation of substrates into the cytosol.⁴⁶ As such, the mechanism of OAT4-mediated uptake of E3S could be incompatible with the efflux cycle of pravastatin. It is also possible that the immediate OAT4-mediated re-uptake of the extruded pravastatin was inhibited by E3S. Taken together, our results suggest that OAT4-mediated efflux of pravastatin is unlikely to occur to any great extent, and if OAT4 is able to mediate the efflux of pravastatin from the intracellular space, the counterpart anion other than chloride and E3S should be involved.

The transport activity of OATPs is independent of ATP, extracellular sodium and chloride, and may be accompanied by extrusion of bicarbonate, which acts as a physiological counter-ion of OATPs.³⁷ Previous report showed that OATP1A2 accepts pravastatin as a substrate at pH 5.5, but not at physiological pH.¹⁵ However, using the same expression system, we observed a two-fold increase in the uptake of the drug by OATP1A2-expressing cells compared to the control cells at neutral extracellular pH (Table 1). Overall, our results had shown that uptake of pravastatin by OATP1A2 occurs at physiological pH in vitro, which suggests that this transporter could contribute to the placental uptake of pravastatin. Cao et al, 2019 reported a decrease of maternal arterial blood pH in cases of pregnancy-induced hypertension⁴⁷; and gestational hypertension can be complicated with preeclampsia in up to 50% of the affected women.⁴⁸ Thus, a shift in the pH microenvironment of the maternal-placental interface towards acidification in preeclampsia⁴⁹ can affect the activity of OATP1A2-mediated uptake of the drug, as the function of this transporter is sensitive to pH.¹⁵

We showed that pravastatin is a substrate for OATP2A1, however the OATP2A1-dependent uptake of pravastatin was only partially inhibited by 20 μM diclofenac, a prototypical inhibitor of OATP2A1, while 200 μM diclofenac exerted no inhibitory effect on the uptake into HEK293-OATP2A1 cells and consequently, on the OATP2A1-dependent uptake (Supplementary Fig S4). Glaeser et al, 2014 reported the lack of effect of 100 μM diclofenac on the OATP2A1-mediated uptake of quercetin in the same expression system and attributed it to differences in the binding sites for the substrate and the inhibitor, although the effect of the inhibitor on the uptake of quercetin into HEK293-OATP2A1 and HEK293-VC cells was not disclosed.³¹ We speculate that 200 μM diclofenac stimulated non-OATP2A1-related paths of the uptake of pravastatin into both HEK293-OATP2A1 and HEK293-VC cells along with inhibiting OATP2A1-mediated uptake resulting in no apparent change in the net uptake. Indirect stimulatory effect of diclofenac on sodium-coupled monocarboxilate transporter 1 in vitro in mammalian expression systems was reported.⁵⁰

Earlier reports showed various kinetic profiles of OATP-mediated uptake of some substrates in vitro. E.g., more than one binding site for E3S transport by OATP1B1 and OATP2B1 was reported.⁴⁵ To our knowledge, our group is the first to report the kinetics of OATP1A2- and OATP2A1-dependent uptake of pravastatin. On the other hand, we did not detect any uptake of pravastatin by OATP2B1 at neutral pH (Table 1), which was consistent with the previous studies.^18,19 Interaction of pravastatin with OATP2B1 at acidic pH has been described previously, and the reported uptake was either fitted into a single component hyperbolic curve or followed biphasic kinetics.^17–19 Thus, OATP2B1 is unlikely involved in the placental bio-disposition of the drug unless a pathological condition such as preeclampsia changes the pH dynamics shifting it towards acidification at the fetal-placental interface.⁴⁹

Our immunoblot analysis revealed no difference in the expression of OATP3A1 and OATP4A1 in the corresponding stably transfected HEK293 cells as compared to the cells transfected with the empty vector (Supplementary Fig S1). Same expression systems were successfully utilized by other groups to study the uptake of simvastatin and mesalazine.^34,51 There is a possibility that in our case the promoter preceding the transporter gene underwent spontaneous epigenetic silencing (or silencing by other means such as mutations). Thus, further studies are needed to establish whether OATP3A1 and OATP4A1 play role in the placental uptake of pravastatin.

It has been shown that OATP5A1 is expressed in plasma membranes of normal and malignant breast tissue as well as hepatic tumor cells.⁵² Our group is the first to report the localization of OATP5A1 in the BM of the syncytiotrophoblast (Fig 2). However, we were unable to confirm functional activity of OATP5A1 in HEK293-OATP5A1 cells: none of the substrates tested, namely pravastatin, E3S, PE2 and fluorescein were found to be transported by OATP5A1. Fluorescein was identified as a general OATP substrate in baculovirus-Sf9 cells³⁵, thus it might not be the optimal probe substrate in expression systems that are based on mammalian cells. On the other hand, Sebastian et al, 2013 reported the lack of interaction of OATP5A1 with a number of prototypical OATP substrates in the Xenopus laevis oocytes expressing hSLCO5A1.⁵²

At the AM of the human placenta, OATP1A2 and OATP2A1 uptake transporters carry out the influx of pravastatin from the maternal circulation to the syncytiotrophoblast. On the other hand, the AM-expressed BCRP and MRP1 efflux transporters extrude pravastatin from the placenta back to the maternal circulation.⁸ Meanwhile, at the BM of the placenta, OAT4 mediates transport of pravastatin from the fetal circulation to the syncytiotrophoblast. While the extent of transplacental passage of pravastatin would depend on the expression and activity of each of the participating transporter, the dense microvilli of the AM are projected to increase the AM surface area by 5–7 folds compared to the BM⁵³ sustaining greater contribution of transport processes that take place at the AM. Indeed, our ex vivo placental perfusion study showed that the clearance index of pravastatin was higher in the fetal-to-maternal direction than in the maternal-to-fetal direction⁷, which could explain limited transplacental transfer of pravastatin to the fetal circulation in vivo.⁶

CONCLUSION

The data presented in this study showed protein expression of OATP1A2 and OATP2A1 in the AM, OATP5A1 in the BM, and OAT4, OATP3A1 and OATP4A1 in both membranes of human syncytiotrophoblast. While our data confirmed the interaction of pravastatin with OAT4 at the BM of human syncytiotrophoblast, we also demonstrated the ability of OATP2A1 and OATP1A2 to mediate the uptake of pravastatin in vitro at physiological pH. Taken together, OATP1A2 and OATP2A1 are likely responsible for the placental uptake of pravastatin from the maternal circulation, while OAT4 mediates the passage of the drug across placental BM in the fetal-to-maternal direction.

ABBREVIATIONS:

AM

apical membrane

ATP

adenosine triphosphate

BM

basolateral membrane

BCRP

breast cancer resistance protein

DMEM

Dulbecco’s modified Eagle’s medium

E3S

estrone-3-sulfate

ECF

extracellular fluid

FBS

fetal bovine serum

HBSS

Hank’s balanced salt solution

HEK293

human embryonic kidney 293 cells

K_t

equivalent of Michaelis-Menten constant

K_e

elimination constant

LC-MS/MS

liquid chromatography-mass spectrometry

MDCK II

Madin-Darby Canine Kidney II cells

MEM

minimum essential medium

MRP1

multidrug resistance protein 1

OAT

organic anion transporter

OATP

organic anion transporting polypeptide

PE2

Prostaglandin-E2

PLAP

placental alkaline phosphatase

SLC

solute carrier transporters

V_max

maximum velocity