Placental Expression of Bile Acid Transporters in Intrahepatic Cholestasis of Pregnancy

Intrahepatic cholestasis of pregnancy (ICP) is a pregnancy-related condition characterized by increased maternal circulating bile acids (BAs) having adverse fetal effects. We investigated whether the human placenta expresses specific regulation patterns to prevent fetal exposition to harmful amounts of BAs during ICP. Using real-time quantitative PCR, we screened placentae from healthy pregnancies (n = 12) and corresponding trophoblast cells (n = 3) for the expression of 21 solute carriers and ATP-binding cassette transporter proteins, all acknowledged as BA- and/or cholestasis-related genes. The placental gene expression pattern was compared between healthy women and ICP patients (n = 12 each). Placental SLCO3A1 (OATP3A1) gene expression was significantly altered in ICP compared with controls. The other 20 genes, including SLC10A2 (ASBT) and EPHX1 (EPOX, mEH) reported for the first time in trophoblasts, were comparably abundant in healthy and ICP placentae. ABCG5 was undetectable in all placentae. Placental SLC10A2 (ASBT), SLCO4A1 (OATP4A1), and ABCC2 mRNA levels were positively correlated with BA concentrations in ICP. Placental SLC10A2 (ASBT) mRNA was also correlated with maternal body mass index. We conclude that at the transcriptional level only a limited response of BA transport systems is found under ICP conditions. However, the extent of the transcriptional response may also depend on the severity of the ICP condition and the magnitude by which the maternal BA levels are increased.


Introduction
Intrahepatic cholestasis of pregnancy (ICP) is a pregnancy-specific liver disorder affecting women worldwide, characterized by the onset of pruritus and elevation of serum bile acid (BA) concentrations. Abnormal metabolic profile including elevated cholesterolemia and maternal comorbidity, such as gestational diabetes, have been also reported [1][2][3]. Depending on the severity of the ICP, the occurring fetal adverse outcomes may include spontaneous preterm labor, fetal distress, and stillbirth [4][5][6]. Numerous factors (e.g., genetic, hormonal, and environmental conditions) are thought to be implicated in the pathogenesis of ICP. Among genetic factors, the mutations in the genes coding familial intrahepatic cholestasis protein-1 (FIC1, also named ATPase phospholipid transporting 8B1 (ATP8B1)), bile salt excretory protein (BSEP, also known as ATP-binding cassette (ABC) subfamily B member 11 (ABCB11)), and multi-drug-resistance protein 3 (MDR3, also named ABCB4) and altered activities of multi-drug-resistance-related protein 2 (MRP2, also named ABCC2) have been reported. Nonetheless, the underlying mechanism of ICP is not precisely known so far.
There is accumulating evidence linking the toxicity of BAs to adverse fetal and maternal outcomes. Therefore, the BA equilibrium within the maternal-fetal pool, which necessarily depends on a balanced BA transport across the placental barrier, is critical. In healthy pregnancies, the concentration of BAs is higher in the fetal than in the maternal circulation [7]. Therefore, vectorial transfer of BAs across the placenta mainly occurs from fetus to mother [8]. The transport from fetus to trophoblast is primarily mediated by anion/BA exchangers, whereas the transport from trophoblast to mother especially occurs via ABC transporter proteins, comparable to BA uptake and efflux in hepatocytes. In ICP, the transplacental gradient for BAs is reversed. Consequently, the net transport of BAs is directed towards the fetus, rather than being transported to the maternal side [9]. Surprisingly, fetal BAs are raised to a lesser extent than maternal BAs, implying a protective mechanism that limits BA uptake into the fetal circulation and/or enhances ATP-dependent carriers that transport against concentration gradients towards the maternal circulation [10]. Since the placenta is a vital organ, which plays a key role in fetal protection, it could be assumed that it would prevent fetal exposure to greater amounts of endobiotic toxic compounds, such as BAs. Although the exploration of human placental BA transport systems is clinically relevant, experimental studies reporting the placental gene expression of BA transporters and carriers are scarce or even lacking. This is particularly noticeable for transporters such as the solute carrier (SLC) family 10 member 2 (SLC10A2, also known as apical sodium-dependent bile acid transporter (ASBT)), epoxide hydrolase 1 (EPHX1, known in hepatic tissue as epoxide hydrolase (EPOX)), solute carrier organic anion transport protein 3A1 (SLCO3A), ATP8B1, ABCB11, and ABCB4 in health and disease conditions. An overview of currently known SLC and ABC transporters associated with BA transport is presented in Table 1. Based on these premises, we carried out the current investigations on placental tissues and trophoblast cells obtained from healthy and ICP pregnancies. We hypothesized that (i) the expression of BA-and cholestasis-related transporters (Table 1) is altered in placentae from ICP patients as a protective response to higher maternal serum BA concentrations, and (ii) correlations exist between placental mRNA levels of BA transport proteins and maternal clinical data. Our objectives were to (i) determine in human placental tissues and trophoblast cells obtained from healthy pregnancies mRNA levels of 21 candidate solute carriers and ABC transporters, whose cellular localization and functional role in BA transport are already well established in hepatocytes and enterocytes; (ii) assess the effect of ICP (i.e., BA "overload") on the placental mRNA levels of BA transport proteins; (iii) examine the association between placentally expressed BA-and cholestasis-related transport proteins and selected maternal clinical parameters as well as baby sex; and (iv) summarize the currently available knowledge on the BA transport machinery in human placenta.

Study Participants
The clinical characteristics of pregnant women enrolled in ICP and control groups are summarized in Table 2. ICP patients had a comparable body mass index (BMI) to healthy pregnant controls. Maternal circulating BA levels were monitored only in case of serious suspicions of ICP. Thus, corresponding data are available only for women diagnosed to be positive for ICP. Data are expressed as mean ± SD. ICP, intrahepatic cholestasis of pregnancy; BMI, body mass index; n.a., not analyzed.

Expression of Selected Solute Carriers and ATP-Dependent Transporter Genes with Affinity for BA-and Cholestasis-Related Molecules in Control Placentae and Trophoblast Cells
The primers listed in Table 3 (see Section 5), used for the amplification of corresponding genes in placental tissues and trophoblast cells, were validated on positive control tissues and cells (liver and hepatocyte cell line). They amplified the expected products in positive controls (data not shown). Table 3. Primers used for gene amplification.
As illustrated in Table 4, in human placental tissues, except ABCG5, whose mRNA transcripts were not found, the remaining 20 BA-and cholestasis-related transport genes were detected by qPCR and categorized as either expressed (defined in our studies as Ct values < 35) or only marginally expressed (defined as Ct values > 35). Among the expressed genes, some exhibited, however, an inconsistent expression pattern. This was the case for SLC10A1, SLC51A, SLCO1B1, and SLCO1B3 since the corresponding mRNA transcripts were detected only in 3/12, 2/12, 2/12, and 9/12 control specimens, respectively. The same applies for ABCC1 and ABCC4 expressions, as they were only detected in 5/12 and 6/12 control tissues, respectively ( Table 4). The mRNA expression of ABCB11 was marginal (Table 4). We did not find any sex-specific expression profiles of the transport proteins tested.
The gene expression profile of the investigated BA-and cholestasis-related transport proteins did not fully correspond between primary trophoblast cells and control placental tissues (Table 4). This is illustrated, for instance, by the dissimilar gene expression of SLCO1A2 in placental tissue compared with trophoblast cells. In our samples, ABCG5 expression was detected neither in control placental tissues nor in primary trophoblast cells.

Comparison of Transporter Expression in Patients and Healthy Controls
In a next step, only the 13 BA-and cholestasis-related transport proteins, which were unequivocally detected in all heathy placentae, were compared with ICP placental tissues. Interestingly, we found that solely SLCO3A1 mRNA expression was differentially expressed (p = 0.0177) in ICP placentae as compared with controls ( Figure 1).

Comparison of Transporter Expression in Patients and Healthy Controls
In a next step, only the 13 BA-and cholestasis-related transport proteins, which were unequivocally detected in all heathy placentae, were compared with ICP placental tissues. Interestingly, we found that solely SLCO3A1 mRNA expression was differentially expressed (p = 0.0177) in ICP placentae as compared with controls ( Figure 1). The remaining tested genes were unaltered (p > 0.05) by the ICP condition (Table 5). There were no sex-specific expression patterns found.

Correlation between Placental BA Transport Proteins and Clinical Parameters
We found significant positive relationships between placental SLC10A2 mRNA levels and maternal BMI values ( Table 6), independently of the maternal health status. In patients, SLC10A2 mRNA levels were correlated with circulating BA levels (Table 6), whereas SLCO4A1 and ABCC2 mRNA levels were positively correlated with maternal serum BA concentrations (Table 6). The remaining tested genes were unaltered (p > 0.05) by the ICP condition (Table 5). There were no sex-specific expression patterns found.

Correlation between Placental BA Transport Proteins and Clinical Parameters
We found significant positive relationships between placental SLC10A2 mRNA levels and maternal BMI values ( Table 6), independently of the maternal health status. In patients, SLC10A2 mRNA levels were correlated with circulating BA levels (Table 6), whereas SLCO4A1 and ABCC2 mRNA levels were positively correlated with maternal serum BA concentrations (Table 6).

Screening of Transporters in Placental Tissues and Trophoblast Cells
The present study describes the mRNA expression profile of important BA solute carriers and ABC transporters in placental tissues/cells and discusses their potential relevance as protective mechanisms preventing fetal exposure to excessive harmful BAs.
One of the main findings of the study is that, for the first time, the gene expression of SLC10A2 is described in human placental tissue and trophoblast cells. SLC10A2/ASBT is a sodium-dependent transporter that exerts a crucial function in the enterohepatic circulation. It enables, at the apical membrane of enterocytes, the uptake of BAs from the intestinal lumen [11]. Previous findings have established that SLC10A2/ASBT abundance in the intestine is inversely correlated with maternal BA concentrations [12]. This is consistent with a role of this gene in preventing excess absorption of BAs into the portal circulation. By assuming that the (apical) localization of SLC10A2/ASBT is conserved in the trophoblast, the unexpected positive relationship of placental SLC10A2 mRNA and maternal BA levels found in the present study appears intriguing. This finding does not argue for a role of placental SLC10A2/ASBT as a feto-protective mechanism against the deleterious effect of elevated maternal serum BA concentrations in ICP. Indeed, the herein observed positive correlation would suggest a parallel increase in placental SLC10A2 expression with augmentation of maternal serum BA concentrations.
Nonetheless, the interpretation of the mentioned relationship requires some caution. Maternal blood samples (used for BA measurements during diagnosis) and placental tissue (for SLC10A2 mRNA analysis after birth) have not been collected at identical time points. Hence, due to limitations based on our ethical approval, maternal BA levels at delivery (i.e., at the time of placenta tissue collection) could not be monitored. Thus, it is not certain whether the detected placental SLC10A2 mRNA expression fully reflects its gene expression at the time of blood sampling.
Next, we identified mRNA transcripts of SLC51B in human placental tissue and in primary trophoblasts isolated from term control placentae. This finding is valuable since the available literature concerning trophoblast cells has reported so far only the gene expression of SLC51A/OST-α [13]. Considering the identification of SLC51A and SLC51B mRNA isoforms in the current study, it is likely that SLC51A/OST-α and SLC51B/OST-β are important in modulating BA fluxes across the placental barrier, similar to their role in other tissue/cell types [14,15].
An additional new finding in this study is based on the detection of mRNA expression of EPHX1 (also called EPOX or mEH in hepatic tissue; see Table 1) in trophoblast cells. This result complements investigations by Coller et al., who, studying human placental tissue, described the presence of EPHX1/EPOX only in placental blood vessels and Hofbauer cells [16]. The discrepancy between these studies may be explained by the difference of the sensitivity of the methods employed. We applied the highly sensitive real-time quantitative PCR using cDNA from well-characterized isolated trophoblast cells, whereas Coller et al. used an immunostaining technique. Considering that EPHX1/EPOX operates as a sodium-dependent BA transporter in other mammalian cells [17,18], the identification of its mRNA in placental tissues and trophoblast cells may suggest a similar function in the human placenta. However, the lack of a significant correlation between the placental EPHX1 gene expression and maternal BA concentrations could indicate a minor role of EPHX1/EPOX in controlling BA fluxes across the human placenta.
Considering that pregnant women with ICP are also prone to other metabolic features, especially dyslipidemia [2], ABCG5 mRNA expression was determined. Surprisingly, we did not detect ABCG5 mRNA expression, neither in our human placental tissues nor in the isolated trophoblast cells. These data are in contrast to findings in rats, where Abcg5 and Abcg8 mRNAs were detected [19]. Nonetheless, we did not find literature data reporting the expression of ABCG5/STSL in human placenta, implying that the placental expression of this membrane protein could be species specific.

Summary of the BA Transport Machinery in Human Placenta
Given the expression profiles of BA transport proteins detected in healthy placental tissues and primary trophoblasts in the current study, we suggest the following scheme summarizing the BA transport machinery in human placental tissue (Figure 2). This overview is based on the assumption that the genes' substrate affinity and polarization patterns are conserved and would therefore reflect findings in enterocytes/hepatocytes [14,15,20]. The fetal liver produces BAs as early as 12 weeks of gestation, which are eliminated as waste products by transporting them across the placenta towards the maternal circulation ( Figure 2A). Conversely, maternal-originating BAs are also directed to the fetus through the placenta (Figure 2B,C). The transport proteins involved are located at the plasma membrane of the placental apical (microvillus) and basal layers ( Figure 2C). The expressed (Ct value < 35) and consistently (present in all specimens) detected transport proteins in placental tissue and syncytiotrophoblasts are illustrated with symbols, filled in green and red colors, respectively. Faded colors indicate equivocally expressed transport proteins. The indicated localization of transport proteins in the trophoblast (Figure 2C), the substrate affinity, and the directionality of transport are according to existing literature under physiological conditions.
Notably, for a few transport proteins, such as SLCO1A2/OATP1A2 and SLCO1B3/ OATP1B3, whose mRNA transcripts were absent in trophoblast cells used in this study, their placental expression is still a matter of controversy. Both absence [21] and presence [13,22,23] have been reported.

Comparison of ICP Versus Controls
Considering recently published data, stratifying the severity of ICP and its relationship to hazard risks of the prevalence of adverse perinatal outcomes [3], patients investigated in this study appeared to be "mildly" affected. The mean concentration of serum BAs was around 56 µmol/L, although a considerable interindividual variation was observed within the studied cohort. Of all 13 analyzed genes, solely the SLCO3A1 mRNA expression was significantly altered in ICP compared with controls. It cannot be excluded that the clinical severity of ICP has an impact on the maternal and fetal regulation of the studied genes. Thus, depending on the severity of the disease and the cohort size, the expression of the genes might vary. The downregulation of SLCO3A1 detected in this study per se seems interesting. SLCO3A1/OATP3A1 is an uptake transporter that has transport affinity for BAs, various steroid hormones, and others [21,24]. Among them are also prostaglandins, key regulators of myometrium contraction, which plays an important role, for example, during preterm labor associated with ICP [25].
Next, all subjects enrolled into the ICP cohort were treated with ursodeoxycholic acid (UDCA treatment). Contrary to the finding by Azzaroli et al. [26], who reported the gene-promoting effect of UDCA on ABCC2 expression, we did not observe alterations in the ABCC2 expression in this study. We were unable to determine whether UDCA treatment caused changes in the placental SLCO3A1 mRNA expression or whether the altered expression pattern resulted from the effect of maternal BA. Moreover, further studies aiming to precisely identify the SLCO3A1/OATP3A1 localization (apical versus basal membranes) and functionality (e.g., with placental trophoblast cell-based Transwell ® system or ex vivo dual perfusion of the human placenta) are needed. They will help to draw firm conclusions regarding whether SLCO3A1 gene expression downregulation constitutes a protective measure for the fetus exposed to maternal ICP. Notably, for a few transport proteins, such as SLCO1A2/OATP1A2 and SLCO1B3/OATP1B3, whose mRNA transcripts were absent in trophoblast cells used in this study, their placental expression is still a matter of controversy. Both absence [21] and presence [13,22,23] have been reported.

Conclusions
Data reported in the current study indicate that the human placenta exhibits a limited response at the transcriptional level when the mother suffers from a "moderate" ICP condition. The newly identified gene expression of SLC10A2 and EPHX1 in human placenta tissue and trophoblasts was unaltered in ICP, while SLC10A2 mRNA strongly correlated with both maternal BMI and BA levels. Nonetheless, given the relatively small cohort size of controls and ICP patients in the present study, the reported findings and interpretations may not be generalized unless confirmed in a larger cohort.

Human Placental Tissue and Trophoblast Cells
The study was approved by the ethics institutional review board of the Canton of Bern with an informed consent obtained from each participant prior to giving birth. The study was conducted in accordance with the Declaration of Helsinki. Pregnant women were under obstetrical care at the Department of Obstetrics and Gynecology, University Hospital, Bern, Switzerland. Placentae from healthy controls (n = 12) and ICP (n = 12) pregnancies were obtained after elective caesarean section or from spontaneous delivery between 2009 and 2011 after having obtained informed consent from the pregnant women. Placentae from healthy pregnancies were used. The clinical characteristics of pregnancies, whose placentae were investigated, are summarized in Table 2.
Upon serious suspicion, pregnant women were diagnosed for ICP following the routinely applied procedure at the University Hospital. The criteria of eligibility for the pregnant women's inclusion to the ICP group include, among others, increased serum BA concentration in combination with pruritus. All ICP women were treated with appropriate doses of De-Ursil ® from diagnosis until delivery.
In addition to placental tissues, a cDNA pool of three independent trophoblast cell isolations was also tested. The procedure of trophoblast isolation has been previously described in detail [27].

RNA Extraction and Quantitative RT-PCR
Considering the heterogeneity of the human placenta, we standardized the placental tissue collection procedure for gene expression analysis [28]. Thus, each specimen analyzed was taken from the central area of the placenta. Total RNA was extracted from placental tissues using Trizol ® reagent (Thermo Fisher Scientific, Waltham, MA, USA). The concentrations of the extracted total RNA were calculated by measuring absorbance (A) at 260 nm. Purity was assessed by the A260/280 and A260/230 ratios, measured using a NanoDrop TM 1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Next, 2 µg of total RNA from each sample was reverse-transcribed using the GoScript ™ Reverse Transcriptase System (Promega, Madison, WI, USA) according to the manufacturer's instructions. Real-time PCR was carried out on the ViiA 7 Real-Time PCR Detection System (Bio-Rad Laboratories Inc., Hercules, CA, USA) using an SYBR ® Green PCR master mix detection kit (Promega, Madison, WI, USA). The reaction conditions were as follows: an initial denaturation at 95 • C for 10 min, followed by 40 cycles of 95 • C for 15 s and 60 • C for 60 s, for melting curve temperature was increased from 60 • C to 95 • C with an increment rate of 0.5 • C every 0.05 s. The primers used for PCR amplification of the 21 currently known human BA-and cholestasis-related transport proteins are shown in Table 3. The relative mRNA expression of BA transport proteins was calculated by using the formula 2 −dCt . For each individual sample, the dCt equates the difference between the Ct value of the transport protein of interest and the mean Ct value of the measured references. The latter were β-actin, YWHAZ, GAPDH, and ubiquitin.

Statistical Analysis
The statistical evaluation was performed using GraphPad Prism ® (GraphPad Software Inc., San Diego, CA, USA). All data are shown as mean ± SD. The gene expression data were analyzed for normality of distribution. Differences in the placental mRNA expression of targeted BA transport proteins between healthy controls and ICP patients were analyzed with unpaired t-test. The correlations of clinical data and gene expression of transport proteins were calculated with Pearson's correlation test. The level of statistical significance was set at p ≤ 0.05.