Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Expression, sorting and transport studies for the orphan carrier SLC10A4 in neuronal and non-neuronal cell lines and in Xenopus laevis oocytes

Expression, sorting and transport studies for the orphan carrier SLC10A4 in neuronal and... Background: SLC10A4 belongs to the solute carrier family SLC10 whose founding members are the Na /tauro- cholate co-transporting polypeptide (NTCP, SLC10A1) and the apical sodium-dependent bile acid transporter (ASBT, SLC10A2). These carriers maintain the enterohepatic circulation of bile acids between the liver and the gut. SLC10A4 was identified as a novel member of the SLC10 carrier family with the highest phylogenetic relationship to NTCP. The SLC10A4 protein was detected in synaptic vesicles of cholinergic and monoaminergic neurons of the peripheral and central nervous system, suggesting a transport function for any kind of neurotransmitter. Therefore, in the present study, we performed systematic transport screenings for SLC10A4 and also aimed to identify the vesicular sorting domain of the SLC10A4 protein. Results: We detected a vesicle-like expression pattern of the SLC10A4 protein in the neuronal cell lines SH-SY5Y and CAD. Differentiation of these cells to the neuronal phenotype altered neither SLC10A4 gene expression nor its vesicu- lar expression pattern. Functional transport studies with different neurotransmitters, bile acids and steroid sulfates were performed in SLC10A4-transfected HEK293 cells, SLC10A4-transfected CAD cells and in Xenopus laevis oocytes. For these studies, transport by the dopamine transporter DAT, the serotonin transporter SERT, the choline transporter CHT1, the vesicular monoamine transporter VMAT2, the organic cation transporter Oct1, and NTCP were used as posi- tive control. SLC10A4 failed to show transport activity for dopamine, serotonin, norepinephrine, histamine, acetylcho- line, choline, acetate, aspartate, glutamate, gamma-aminobutyric acid, pregnenolone sulfate, dehydroepiandroster- one sulfate, estrone-3-sulfate, and adenosine triphosphate, at least in the transport assays used. When the C-terminus of SLC10A4 was replaced by the homologous sequence of NTCP, the SLC10A4-NTCP chimeric protein revealed clear plasma membrane expression in CAD and HEK293 cells. But this chimera also did not show any transport activity, even when the N-terminal domain of SLC10A4 was deleted by mutagenesis. Conclusions: Although different kinds of assays were used to screen for transport function, SLC10A4 failed to show transport activity for a series of neurotransmitters and neuromodulators, indicating that SLC10A4 does not seem to represent a typical neurotransmitter transporter such as DAT, SERT, CHT1 or VMAT2. Keywords: SLC10A4, Transport, NTCP, Neurotransmitter, Sorting *Correspondence: joachim.m.geyer@vetmed.uni-giessen.de Institute of Pharmacology and Toxicology, Justus Liebig University of Giessen, Schubertstr. 81, 35392 Giessen, Germany © 2015 Schmidt et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Schmidt et al. BMC Neurosci (2015) 16:35 Page 2 of 14 with an in vitro approach. As the endogenous expression Background of SLC10A4 is exclusively directed to neuronal cells and The orphan carrier SLC10A4 belongs to the solute carrier mast cells [13, 14], neuronal cell cultures were thought to family SLC10, whose founding members are the Na /tau- be the most appropriate for this purpose. Therefore, we rocholate co-transporting polypeptide NTCP (SLC10A1) analyzed SLC10A4 expression in the human neuroblas- and the apical sodium-dependent bile acid transporter toma cell line SH-SY5Y as well as in the mouse cell line ASBT (SLC10A2) [1]. These carriers maintain the enter - CAD (Cath.a-differentiated neuronal cells, originating ohepatic circulation of bile acids by sodium-dependent from the locus coeruleus in the brainstem) with different bile acid reabsorption in the gut via ASBT and sodium- SLC10A4-directed antibodies. dependent bile acid transport into hepatocytes via NTCP SH-SY5Y cells showed a typical neuroblast-like appear- [2]. A further member of this carrier family, the sodium- ance with small, round cell bodies and occasional short dependent organic anion transporter SOAT (SLC10A6), extensions. Under incubation with retinoic acid (RA) is expressed in spermatocytes in man and mouse and has transport activity for sulfo-conjugated steroid hor and the neurotropic factors tumor growth factor beta (TGF-β1) and bone morphogenetic protein 2 (BMP-2), mones [3–5]. The other members of this carrier family, the cells stopped proliferation and developed neurite-like SLC10A3, SLC10A4, SLC10A5, and SLC10A7, have been long extensions, as described previously [19, 20]. Under characterized at the molecular and expression level, but both conditions, SLC10A4 showed a clear vesicle-like still remain orphan carriers [1, 6–12]. expression pattern in the SH-SY5Y cells and was detect- The SLC10A4 transcript was first cloned from rat in our able even along the long neurite-like outgrowths, indi- group in 2004 and showed predominant expression in the central nervous system in man, rat, and mouse [13]. Fur cating sorting of the SLC10A4 protein to the synaptic direction of the differentiated SH-SY5Y cells (Figure  1b). ther expression analyses on the protein level localized At the RNA level, SLC10A4 showed an overall higher SLC10A4 to cholinergic and monoaminergic neurons of expression in the SH-SY5Y cells compared with vesicu- the central and peripheral nervous system as well as to lar acetylcholine transporter (VAChT) and vesicular mast cells [13–15]. In contrast to NTCP, ASBT, and SOAT, monoamine transporter 2 (VMAT2) (data not shown), the SLC10A4 protein is not directed to the plasma mem- but incubation with TGF-β1  +  RA or BMP-2  +  RA did brane, but typically showed a vesicular expression pattern not significantly affect the SLC10A4 mRNA expression in neuronal cells, mast cells, and even in transiently or sta- levels, indicating that SLC10A4 expression is not regu- bly transfected cells lines [6, 13–15]. Based on this expres- lated by the RA, BMP-2, or TGF-β1 triggered signaling sion pattern, it was assumed that SLC10A4 may represent a cascades (Figure  1a). Although transient transfection of novel vesicular carrier for any kind of neurotransmitter or SLC10A4 into SH-SY5Y revealed an identical expression neuromodulator [1, 6, 13–15]. More recently, in a series of pattern compared with the endogenous expression, as very elegant experiments on Slc10a4 knockout mice it was shown for an SLC10A4-RFP construct in Figure  1c, the shown by the Kullander group that these mice are hyper- transfection rate of these cells could not be enhanced sensitive to the psychostimulants amphetamine and tra- above 20% by different transfection methods (lipofection, nylcypromine, and have an altered response to cholinergic non-liposomal transfection, electroporation), meaning stimuli at the neuromuscular junction and in the central that SH-SY5Y cells overexpressing SLC10A4 vs. non- cholinergic system, suggesting that SLC10A4 may contrib- transfected SH-SY5Y cells could not be used for trans- ute to the vesicular storage or release of neurotransmitters port studies. For the same reason, down-regulation of [15–17]. Therefore, in the present study, we performed sys - SLC10A4 expression by transfection of SLC10A4 siRNA tematic transport screenings for SLC10A4 in transfected prior to transport experiments was also not considered. neuronal and HEK293 cell lines as well as in Xenopus lae- Similar to the SH-SY5Y cells, the SLC10A4 protein vis oocytes and also aimed to identify the vesicular sorting was detected in vesicular structures in mouse CAD cells. domain of the SLC10A4 protein. Although we have not These cells were differentiated to a neuronal phenotype by identified a transported substrate for SLC10A4 to date, serum depletion, as previously reported [21]. As shown in recent descriptions of taurocholic acid and lithocholic acid Figure 1e, under these conditions, long neurite-like exten- transport by a thrombin-modified variant of SLC10A4 [18] sions were formed which were highly immunoreactive for encouraged us to present our data to provide a broader basis for further SLC10A4 transport studies. the anti-Slc10a4 antibody. However, even in these cells, RNA expression of SLC10A4 was not significantly affected by neuronal differentiation (Figure  1d). Although CAD Results cells could easily be transfected with different SLC10A4 Endogenous expression of SLC10A4 in neuronal cell lines constructs with transfection rates above 80%, these cells The primary goal of the present study was to identify a significantly detached after the transfection procedure transported substrate for the orphan carrier SLC10A4 Schmidt et al. BMC Neurosci (2015) 16:35 Page 3 of 14 Figure 1 Expression and subcellular localization of SLC10A4 in SH-SY5Y and CAD cells. a Relative SLC10A4 gene expression in SH-SY5Y cells after differentiation with TGF-ß1 + RA or BMP-2 + RA. Values represent mean ± SD of triplicate measurements. b Immunofluorescence analysis of the subcellular expression of the SLC10A4 protein in SH-SY5Y cells. Cells were either untreated (UD), or were differentiated with TGF-ß1 + RA or BMP-2 + RA over 4 days prior to immunolabeling. The SLC10A4 protein was detected with the anti-Slc10a4 1338 C antibody (1:1,000) and the Cy3- labelled anti-rabbit secondary antibody (1:800, red fluorescence) and nuclei were stained with DAPI (blue fluorescence). In all cases, the SLC10A4 protein showed a vesicle-like expression pattern within the perikarya and along the neurite-like cellular protrusions. c Even when a fluorescence- tagged SLC10A4-RFP construct was transiently transfected into SH-SY5Y cells, the SLC10A4-RFP protein showed a clear vesicular sorting pattern. d Relative Slc10a4 gene expression analysis in differentiated (Diff ) and undifferentiated (UD) CAD cells. The values represent mean ± SD of triplicate measurements. e Endogenous expression of the SLC10A4 protein in CAD cells, cultivated in FCS containing medium (UD) or FCS-free medium (Diff ). The SLC10A4 protein was detected with the anti-Slc10a4 1338 C antibody (1:500) and the Cy3-labelled secondary antibody (1:800, red fluo - rescence). For control, the primary anti-Slc10a4 antibody was omitted (control) or the antibody was pre-incubated with the immunizing peptide (peptide blocking). f Immunofluorescence detection of the SLC10A4 protein was performed with different SLC10A4-directed antibodies (green fluorescence): self-generated polyclonal rabbit anti-Slc10a4 1338 C antibody, rabbit anti-SLC10A4 Sigma Prestige antibody, rabbit anti-Slc10a4 Abnova antibody, and rabbit anti-SLC10A4 Abgent antibody. Membrane protein enriched fractions of the CAD cells were also subjected to Western Blot analysis with the same antibodies and revealed specific bands for the SLC10A4 protein at an apparent molecular weight of 30–32 kDa. Prestige rabbit anti-SLC10A4), the 1–27 amino acids of the and then could not be further subjected to standard trans- N-terminus of mouse and rat SLC10A4 protein (Abnova port assays. Therefore, overexpression of SLC10A4 or rabbit anti-Slc10a4), and the 377–406 amino acids of the down-expression by RNAi transfection of these cells was C-terminus of the human SLC10A4 protein (Abgent rab- not practicable for transport screening assays. bit anti-SLC10A4)—a clear vesicle-like expression pat- Localization of the SLC10A4 protein in the central and tern. The mouse SLC10A4 protein consistently showed an peripheral nervous system of the rat, as well as in rat PC12 apparent molecular weight of 30–32  kDa in the Western cells [13, 14], human SH-SY5Y cells, and mouse CAD blot (Figure  1f). This is below the calculated molecular cells (present study), was performed by us with a self- weight of 47 kDa, which has been reported for other mem- generated polyclonal rabbit antibody—1338 C—directed bers of the SLC10 carrier family as well [8]. against the amino acid residues 422–437 (VGTDDLVL- METTQTSL) of the deduced rat Slc10a4 protein sequence In vitro transport studies on SLC10A4 with candidate [GenBank:AAV80706]. In order to further verify SLC10A4 substrates expression in neuronal cell cultures with different antibod - As these neuronal cell lines were difficult to handle for ies, we screened mouse CAD cells by immunofluorescence standard transport experiments, we decided to use trans- and Western blot analysis, and found by all of the used fected human embryonic kidney (HEK) 293 cells as well as antibodies—directed against the 324–437 amino acids SLC10A4 cRNA injected X. laevis oocytes, which are both of the C-terminus of human SLC10A4 protein (Sigma Schmidt et al. BMC Neurosci (2015) 16:35 Page 4 of 14 well-established in vitro cell models for transport measure- In our previous study, we reported co-expression of the ments [3, 22]. We previously reported that, apart from its SLC10A4 protein with VMAT2 in synaptic vesicles of predominant intracellular expression, the SLC10A4 pro- rat brain preparations [14], so we next intended to ana- tein can to a lower extent also be detected in the plasma lyze the SLC10A4 transport in direct comparison with membrane of transfected HEK293 cells [13]. Therefore, VMAT2. In this approach, we used HEK293 cells stably we first used HEK293 cells transiently transfected with the transfected with human SLC10A4 or human VMAT2, human SLC10A4 construct for classical uptake measure- which were permeabilized with digitonin in order to ment in intact cells. As positive controls, the plasma mem- facilitate access of the transport substrate to the site of brane neurotransmitter transporters dopamine transporter carrier expression in intracellular vesicular structures, (DAT), choline transporter (CHT1), and serotonin trans- as described before [23]. As shown in Figure  3, VMAT2 porter (SERT) were also transfected, and empty-vector showed significant transport activity for [ H]seroto- 3 3 transfected cells served as negative controls. Whereas DAT nin, [ H]norepinephrine, and [ H]dopamine which was showed a significant nomifensine-sensitive and sodium- dependent on the presence of adenosine triphosphate dependent uptake of [ H]dopamine, SLC10A4-transfected (ATP) in the transport buffer as well as sensitive to tetra - HEK293 cells tended to accumulate more dopamine com- benazin (TBZ) and carbonylcyanide-p-trifluoromethoxy - pared with the control, but without reaching a level of sig- phenylhydrazon (FCCP). However, in contrast, SLC10A4 nificance (Figure  2). SLC10A4 showed no transport activity showed no transport activity for serotonin, norepi- 3 3 3 for [ H]norepinephrine, [ H]serotonin and [ H]choline nephrine, and dopamine at all in this vesicular trans- at all, whereas DAT, SERT and CHT1 significantly trans - port assay. We also could not demonstrate a transport ported their substrates in a sodium-dependent manner. of acetylcholine, neither for SLC10A4 nor for VAChT Figure 2 Transport measurements in transiently transfected HEK293 cells. HEK293 cells were transiently transfected with the indicated carriers 3 3 3 3 SLC10A4, DAT, CHT1, or SERT, respectively. The uptake of 5 µM [ H]dopamine, 5 µM [ H]norepinephrine, 5 µM [ H]choline, or 5 µM [ H]serotonin was + + + measured over the given time periods in the presence and absence of Na (for CHT1 Na was replaced by Li ). Transport via DAT, CHT1, and SERT was blocked by the specific inhibitors nomifensine (10 µM), hemicholinium-3 (HC-3, 1 µM), and citalopram (2 µM), respectively. Values represent mean ± SD of representative experiments, each with quadruplicate determinations (n = 4). *Significantly different from control with p < 0.001. Significantly different from positive uptake, p < 0.001. Schmidt et al. BMC Neurosci (2015) 16:35 Page 5 of 14 Figure 3 Transport measurements in digitonin permeabilized SLC10A4-HEK293 and VMAT2-HEK293 cells. Prior to transport measurements, stably transfected human SLC10A4-HEK293 cells and human VMAT2-HEK293 cells were pre-incubated with 15 µM digitonin for 10 min for permeabiliza- 3 3 3 tion. Then the uptake of 400 nM [ H]serotonin, [ H]norepinephrine, or [ H]dopamine was measured over 10 min in the presence and absence of 5 mM ATP in the transport buffer. In addition to ATP, 2 µM of the potent VMAT2 inhibitor tetrabenazine ( TBZ) or 5 µM of the proton ionophore carbonylcyanide-p-trifluoromethoxyphenylhydrazon (FCCP) were added to the transport buffer as indicated. After 10 min, the cells were washed with ice-cold PBS, lysed, and subjected to scintillation counting. The values represent mean ± SD of one representative experiment (for dopamine, n = 4) or two independent experiments (for serotonin and norepinephrine, n = 8). *Significantly different from control with p < 0.05. Significantly different from positive uptake, p < 0.05. thrombin treatment and thereby transport activity for in permeabilized HEK293 cell lines transfected with the taurocholic acid and lithocholic acid may be activated respective construct (data not shown). [18]. Although these data were obtained in a neuronal cell In previous studies at our institute we learned that the culture of TE671 cells, in the present study, we intended transport rates for a particular carrier may be higher in to reproduce these data on the isolated over-expressed the X. laevis oocytes expression system compared with SLC10A4 protein and used HEK293 cells stably trans transfected cell cultures. Furthermore, we demonstrated - in a previous study that at least the rat SLC10A4 protein fected with SLC10A4 as well as with NTCP for the con- is expressed at the plasma membrane of X. laevis oocytes trol. Both, SLC10A4-HEK293 and NTCP-HEK293 cell [13]. Therefore, we injected SLC10A4 cRNA in X. laevis lines were treated with 1 U/200  µl of thrombin for 3  h 3 3 oocytes and screened for transport activity with a series prior to transport measurements with [ H]DHEAS, [ H] 3 3 of neurotransmitters and further SLC10A4 candidate taurocholic acid, [ H]PREGS, and [ H]lithocholic acid. substrates. For these experiments, water-injected oocytes As expected, NTCP showed significant transport activ - served as negative control and, if appropriate, oocytes ity for DHEAS, taurocholic acid and PREGS, and was not injected with cRNA coding for SERT, DAT, NTCP, or affected by thrombin treatment. In contrast, SLC10A4 mouse organic cation transporter 1 (Oct1) were used as did not transport any of these compounds, even after positive controls. As shown in Figure  4, SERT, DAT, and pre-treatment with thrombin (Figure  5). No transport Oct1 showed significant transport activity for seroto - activity could be observed for lithocholic acid at all, nei- nin, dopamine, and histamine, respectively, but SLC10A4 ther in NTCP nor in the SLC10A4-expressing HEK293 cRNA injected oocytes were not different from the con - cells. Therefore, transport activity of SLC10A4 could not trol. Furthermore, the neurosteroids pregnenolone sulfate be activated by thrombin treatment in the cell model (PREGS) and dehydroepiandrosterone sulfate (DHEAS) as used. well as the bile acid taurocholic acid, were not transported by SLC10A4 in this transport assay, although NTCP Localization of the vesicular sorting domain in the showed significant transport activity for all of them. Apart C‑terminus of SLC10A4 from these candidate substrates, SLC10A4 also showed no A further aim of the present study was to localize the transport activity for aspartate, acetate, choline, gamma- particular domain of SLC10A4 that is responsible for aminobutyric acid (GABA), norepinephrine, glutamate, the vesicular sorting of this protein. This question was acetylcholine, estrone-3-sulfate (E-3-S), lithocholic acid, of particular interest, as mutation of this domain might and ATP in the X. laevis oocytes model (Table 1). redirect the SLC10A4 protein to the plasma mem- Very recently, it was reported that the N-terminal brane, which would facilitate further substrate screen- domain of the SLC10A4 protein could be cleaved by ing. As within the SLC10 carrier family, NTCP, which is Schmidt et al. BMC Neurosci (2015) 16:35 Page 6 of 14 Table 1 Transport studies in Xenopus laevis oocytes SLC10A4 Control Ratio Uptake (pmol/oocyte/60 min) [ H]Histamine (10 µM) 0.29 ± 0.02 0.27 ± 0.01 1.0 [ H]Aspartate (1 µM) 0.22 ± 0.01 0.28 ± 0.01 0.8 [ H]Acetate (25 µM) 8.79 ± 0.51 8.36 ± 0.72 1.0 [ H]Choline (1 µM) 2.68 ± 0.52 2.20 ± 0.19 1.2 [ H]GABA (1 µM) 0.52 ± 0.05 0.47 ± 0.04 1.1 [ H]Norepinephrine (1 µM) 0.16 ± 0.01 0.21 ± 0.02 0.8 [ H]Glutamate (1 µM) 0.014 ± 0.001 0.014 ± 0.002 1.0 Uptake (pmol/oocyte/30 min) [ H]Acetylcholine (10 µM) 43.51 ± 3.62 37.51 ± 3.67 1.2 [ H]Dopamine (10 µM) 1.36 ± 0.06 1.43 ± 0.05 1.0 [ H]PREGS (1 µM) 1.19 ± 0.09 1.06 ± 0.09 1.1 [ H]DHEAS (1 µM) 0.043 ± 0.002 0.043 ± 0.004 1.0 [ H]E-3-S (1 µM) 0.050 ± 0.001 0.060 ± 0.005 0.8 Uptake (pmol/oocyte/10 min) [ H]Lithocholic acid (0.3 µM) 3.68 ± 0.23 4.13 ± 0.07 0.9 [ H]Taurocholic acid (0.3 µM) 0.0158 ± 0.0009 0.0159 ± 0.0007 1.0 [ H]ATP (5 mM) 70.66 ± 9.06 68.84 ± 8.26 1.0 [ S]Adenosine 5`-(γ-thio) 69.76 ± 4.36 67.73 ± 2.53 1.0 triphosphate (5 mM) Xenopus laevis oocytes were injected with human SLC10A4 cRNA or with water (control). Values represent uptake of the indicated compound as mean  ±  SD of at least 10 oocytes per group. Ratio represents uptake into SLC10A4 cRNA- injected oocytes divided by uptake into water-injected oocytes. 75ΔSLC10A4 mutant which was truncated by the first 75  N-terminal amino acids. All constructs were trans- fected in neuronal CAD cells as well as in HEK293 cells and used for immunofluorescence analysis of protein localization (Figure  6b) as well as for transport experi- ments with SLC10A4 candidate substrates (Figure  6c; Figure 4 Transport measurements in Xenopus laevis oocytes. Xenopus laevis oocytes were injected with cRNA coding for human Table  2). As shown in Figure  6b, NTCP showed clear SLC10A4, SERT, DAT, or NTCP as well as mouse Oct1. Uptake of [ H] plasma membrane localization in CAD cells, whereas 3 3 3 3 serotonin, [ H]histamine, [ H]PREGS, [ H]dopamine, [ H]DHEAS, or SLC10A4 showed expression in intracellular vesicles. [ H]taurocholic acid, each at 1 µM, was measured over a time period Interestingly, both chimeras bearing the C-terminus of of 10–60 min as indicated in the presence of sodium chloride in the NTCP (i.e. SLC10A4-CtNTCP and NtNTCP-SLC10A4- transport buffer. SERT, Oct1, and DAT served as controls for the trans- 3 3 3 port of [ H]serotonin, [ H]histamine, and [ H]dopamine, respectively. CtNCTP) were detected at the plasma membrane, NTCP was the reference carrier for PREGS, DHEAS and taurocholic whereas all chimeric proteins with the C-terminus of acid. Afterwards, the oocytes were washed with ice-cold transport SLC10A4 (i.e. NTCP-CtSLC10A4, NtSLC10A4-NTCP- buffer, lysed and subjected to scintillation counting. The values CtSLC10A4) were detected in intracellular vesicles. This represent mean ± SD of one representative experiment with n = 10 sorting pattern was basically the same when these con- oocytes each. *Significantly different from control with p < 0.001. structs were transfected and analyzed in HEK293 cells (data not shown). This indicates that the C-termini of NTCP and SLC10A4 are dominant for the sorting to expressed at the plasma membrane, is the most related the plasma membrane and intracellular vesicles, respec- carrier to SLC10A4, we generated several SLC10A4/ tively. In contrast, truncation of the SLC10A4  N-ter- NTCP chimeric constructs in which the C-terminal minus (75ΔSLC10A4 mutant) as well as transfer of the and N-terminal domains were interchanged between SLC10A4  N-terminus into the NTCP (NtSLC10A4- both carriers (Figure  6a). Furthermore, we generated a NTCP chimera) did not affect the sorting of SLC10A4 Schmidt et al. BMC Neurosci (2015) 16:35 Page 7 of 14 Figure 5 Transport measurements in stably transfected SLC10A4-HEK293 and NTCP-HEK293 cells after thrombin treatment. For the transport 3 3 measurements, one part of the cells was pre-incubated with 1 U/200 µl thrombin over 3 h (+Thrombin), before the uptake of [ H]DHEAS, [ H] 3 3 taurocholic acid, [ H]PREGS, or [ H]lithocholic acid (each at 300 nM) was measured over a time period of 10 min at 37°C. The cells were washed with ice-cold PBS, lysed, and subjected to scintillation counting. The values represent mean ± SD of two independent experiments each with triplicate determinations. *Significantly different from control with p < 0.001; n.s. not significantly different. and NTCP, respectively, indicating that the C-terminus, Discussion but not the N-terminus of the SLC10A4 protein is rel- SLC10A4 is a member of the SLC10 carrier family com- evant for the protein sorting. Apart from localizing monly referred to as the “family of sodium-dependent the sorting domains in the SLC10A4 protein, several bile acid transporters”. The rat Slc10a4 transcript was of the constructs were of particular interest for trans- first cloned in our group in 2004 [GenBank:AY825923] port experiments. (I) When the N-terminal domain [13] and the human SLC10A4 sequence was published of SLC10A4, which was previously supposed to hin- in 2006 [6]. The SLC10A4 protein consists of 437 amino der substrate binding to the SLC10A4 carrier protein acids in man, rat, and mouse, and shows a close phylo- [18], were transferred to the NTCP in order to produce genetic relationship to the sodium-dependent bile acid a NtSLC10A4-NTCP elongated chimera, the trans- transporter NTCP. However, whereas NTCP is specifi - port function for taurocholic acid remained completely cally expressed at the sinusoidal plasma membrane of intact compared with wild-type NTCP (Figure  6c). (II) hepatocytes [2], the SLC10A4 protein was localized in Furthermore, activation of the SLC10A4-mediated tau- cholinergic and monoaminergic neurons of the central rocholic acid transport by thrombin cleavage of the and peripheral nervous system as well as in pheochro- N-terminus was supposed [18]. However, for the N-ter- mocytoma PC12 cells and mast cells [13–15]. Based on minally truncated 75ΔSLC10A4 protein, as well as for its genetic classification, SLC10A4 first was suggested to the NtNTCP-SLC10A4-CtNTCP chimera, still no trans- be a novel carrier for already established substrates of the port activity for taurocholic acid could be observed. (III) SLC10 carrier family, including bile acids and sulfo-con- Just by replacing the SLC10A4 C-terminal 63 amino jugated steroid hormones [1, 6, 13]. However, based on acids by the corresponding C-terminus of NTCP in the its expression pattern, it seems more likely that SLC10A4 SLC10A4-CtNTCP and NtNTCP-SLC10A4-CtNTCP plays a role for the vesicular storage or exocytosis of any chimera, the SLC10A4 protein was directed to the kind of neurotransmitter or mediator in neurons and plasma membrane. Therefore, these chimeras represent mast cells [14]. In previous experiments, the neuroster- an interesting tool for SLC10A4 substrate screening. As oids PREGS and DHEAS were already considered to be the CAD cells detached after transfection of the respec- particularly promising candidate substrates. These neu - tive constructs when they were further processed for the rosteroids can modulate several postsynaptic receptor routine transport assay, the radiolabelled candidate sub- systems and even interfere with the release of multiple strates were added directly into the cell culture medium neurotransmitters including acetylcholine, norepineph- and cell associated radioactivity was analyzed. However, rine, dopamine and serotonin [24–26]. However, until neither SLC10A4-CtNTCP nor NtNTCP-SLC10A4- now, all transport measurements performed with E-3-S, CtNTCP showed transport activity so far for taurocholic DHEAS, and PREGS have so far failed to show transport acid and serotonin (Figure  6c) as well as for DHEAS, function for SLC10A4 [6, 13] (present study). However, a PREGS, E-3-S, dopamine, glutamate, histamine, acetyl- more systematic transport screening also including clas- choline, and choline (Table 2). sical neurotransmitters as candidate substrates has not Schmidt et al. BMC Neurosci (2015) 16:35 Page 8 of 14 Figure 6 Localization and transport function of SLC10A4/NTCP chimeras in CAD cells. a The shown SLC10A4/NTCP chimeric constructs were used. All chimeras were generated based on the full length sequences of SLC10A4 (grey marked transmembrane domains and loops with continuous lines) and NTCP (white transmembrane domains and loops as dotted lines), both with c-terminal V5-tag. Potential glycosylation sites were marked by “Y”. b All constructs were transiently transfected in CAD cells and cellular localization was analyzed by immunofluorescence microscopy using rabbit anti-V5 antibody and donkey Cy3-labelled anti-rabbit secondary antibody. Nuclei were stained with DAPI. Whereas SLC10A4 showed a clear vesicle-like expression pattern, the immunofluorescence signals for NTCP, NtSLC10A4-NTCP, SLC10A4-CtNTCP, and NtNTCP-SLC10A4-CtNTCP were clearly directed to the plasma membrane. When the 75 N-terminal amino acids were deleted in SLC10A4, the 75ΔSLC10A4 protein retained its vesicle-like intracellular expression comparable with full-length SLC10A4. c The SLC10A4/NTCP chimeras were also used for transport studies after 3 3 transient transfection into CAD cells with [ H]taurocholic acid and [ H]serotonin, each at 5 µM. These measurements were performed by incubating the cells for 60 min at 37°C in 250 µl cell medium with 50 µl sodium transport buffer containing the radiolabeled and non-radiolabeled compounds. NTCP and SERT were used as a positive control, and empty-vector transfected cells served as the negative control. After the uptake phase, cells were washed with ice-cold PBS, lysed, and subjected to scintillation counting. Data represent mean ± SD of representative experiments each with quadruplicate determinations. *Significantly different from control with p < 0.01. the case that SLC10A4 has transport function for one of yet been performed. Therefore, in the present study, we the mentioned compounds anyway. In HEK293 cells and aimed to screen for SLC10A4 transport function with X. laevis oocytes, an additional factor might have been different transport assays, including transport in intact missed that is only present in neuronal cells. However, HEK293 cells and neuronal CAD cells, permeabilized overexpression of the SLC10A4 protein or permeabi- HEK293 cells and X. laevis oocytes. Overall, no trans- lization of the cells prior to transport experiments was port activity for SLC10A4 was found for taurocholic not practicable in neuronal cell lines as the cells either acid, lithocholic acid, DHEAS, PREGS, E-3-S, acetylcho- showed low transfection rates (as the SH-SY5Y cells) or line, choline, acetate, glutamate, aspartate, GABA, ATP, promptly detached when the incubation medium was serotonin, histamine, dopamine, and norepinephrine in repeatedly changed (CAD cells). Furthermore, all trans- the respective transport assay. Although we used differ - port assays performed in SLC10A4-transfected CAD ent kinds of transport assays and cell models, it might be Schmidt et al. BMC Neurosci (2015) 16:35 Page 9 of 14 Table 2 Transport studies in HEK293 cells Compound (5 µM) NTCP SLC10A4 NtNTCP‑SLC10A4‑ CtNTCP Uptake Ratio Uptake Ratio Uptake Ratio [ H]E-3-S 276.93 ± 29.12 10.1* 35.23 ± 2.01 1.3 27.91 ± 1.38 1.0 [ H]DHEAS 225.12 ± 16.51 6.9* 38.22 ± 0.90 1.2 31.84 ± 1.38 1.0 [ H]PREGS 1,656.09 ± 201.01 4.7* 366.06 ± 22.23 1.0 383.62 ± 17.38 1.0 DAT SLC10A4 NtNTCP‑SLC10A4‑ CtNTCP [ H]Dopamine 1,578.17 ± 107.47 13.2* 128.61 ± 16.60 1.1 157.92 ± 19.11 1.3 [ H]Glutamate 22.60 ± 2.95 1.4 13.57 ± 1.62 0.8 [ H]Histamine 26.24 ± 1.40 1.3 25.68 ± 0.63 1.3 [ H]Acetyl-choline 352.00 ± 38.76 1.0 307.64 ± 31.41 0.9 [ H]Choline 929.53 ± 30.50 1.1 866.83 ± 23.63 1.0 Values represent uptake in pmol/mg protein/30 min of the indicated compound as mean ± SD of quadruplicate determinations. Ratio represents uptake into carrier- expressing HEK293 cells divided by uptake into mock-transfected HEK293 cells. * Significantly different from control with p < 0.01. cells also failed to show transport activity for SLC10A4. Abe et  al. [18] that the N-terminal domain of SLC10A4 Nevertheless, further investigations on the SLC10A4 might hinder substrate binding to the protein. This sug - transport function should focus on vesicle preparations gestion can not be supported by data from the present from SLC10A4-deleted neuronal cell cultures or may use study. It is already known that thrombin starting from 1 liposomal reconstitution of the SLC10A4 protein. U/200  µl has a dose-dependent effect on cell viability of Very recently, it was suggested by Abe et  al. [18] that neuronal cell cultures [27]. Furthermore, it was reported SLC10A4 may represent a protease-activated trans- that lithocholic acid selectively kills neuroblastoma cells porter for the bile acids taurocholic acid and lithocholic by triggering the extrinsic and intrinsic apoptotic death acid. They used the human medulloblastoma cell line pathways via binding to the surface [28]. Therefore, it TE671 in which they localized the SLC10A4 protein by cannot be excluded that the suggested uptake of tauro- immunofluorescence and Western Blot analysis with a cholic acid and lithocholic acid may just have resulted commercial anti-SLC10A4 antibody and showed that from the combined surface effects of the bile acids and pre-incubation of the cells with 1U/200  µl of thrombin thrombin, but not from carrier-mediated uptake into the for 3  h increased the cellular uptake of taurocholic acid cells via a thrombin-modified SLC10A4 variant. and lithocholic acid [18]. However, appropriate controls During the completion of this study, the Kullander demonstrating specificity of the antibody were not pro - group published a series of very elegant experiments on vided and the authors failed to seriously show saturable Slc10a4 knockout mice, which provide further intrigu- transport kinetics for this uptake. Furthermore, bioin- ing indications of the molecular transport function of formatics analysis (Expasy Peptide Cutter, http://www. SLC10A4. In the first study by Zelano et  al. [16], they web.expasy.org/peptide_cutter/) indicates that the mouse analyzed whether the absence of SLC10A4 may have an and rat SLC10A4 proteins can be cleaved by thrombin impact on the function of the central cholinergic system. at amino acid position 87, but not the human SLC10A4 Injection of the cholinergic agonist pilocarpin induced protein. Nevertheless, in the present study, these experi- status epilepticus earlier and more often in the Slc10a4 ments were repeated by using HEK293 cells overexpress- knockout mice compared with the wild type mice, sug- ing SLC10A4, providing a better controllable system, but gesting that SLC10A4 may suppress epileptiform activ- did not show any transport activity of SLC10A4 for tau- ity. They concluded that absence of the SLC10A4 protein rocholic acid after treatment with thrombin. Due to these results in cholinergic hypersensitivity of the knockout negative data and based on the bioinformatics prediction mice [16]. In a second study by Patra et al. [17], the role we did not further analyze whether the human SLC10A4 of SLC10A4 on the structure and function of the neuro- protein can really be cleaved by thrombin in our experi- muscular junction was analyzed. Although there were no mental setup. Furthermore, when the large N-terminus abnormalities detectable at the macrostructure level, the of SLC10A4 was deleted by mutagenesis in the NtNTCP- Slc10a4 knockout mice showed misshapen neuro-mus- SLC10A4-CtNTCP and 75ΔSLC10A4 constructs, no cular junctions with an increased number of isolated ace- transport activity for taurocholic acid was detected. tylcholine receptor clusters and a decreased number of This is of particular interest, because it was suggested by endplate branches. Nevertheless, the knockout mice had Schmidt et al. BMC Neurosci (2015) 16:35 Page 10 of 14 normal motor behavior. Electrophysiological measure- The outstanding role of SLC10A4 within the SLC10 ments on nerve-muscle preparations then showed that family is not only based on its inability to transport bile the Slc10a4 knockout mice had decreased spontaneous acids or steroid sulfates. In contrast to NTCP, ASBT and endplate potential amplitudes, which could be explained SOAT, which all are sorted to the plasma membrane, by lower acetylcholine release at the endplates without SLC10A4 has a typical intracellular vesicle-like expres- nerve stimulation. On the other hand, after repeated sion pattern in neurons, mast cells and neuronal cell stimulation, the Slc10a4 knockout mice revealed an lines. This specific expression pattern of SLC10A4 was enlarged pool of readily releasable vesicles at the neuro- also observed in the present study in untreated and dif- muscular junction. However, for the interpretation of ferentiated CAD and SH-SY5Y cells. To more closely these data, it has to be considered that the knockout mice analyze the sorting domains of SLC10A4, a series of may have altered gene-expression in order to compensate NTCP/SLC10A4 chimeras were generated and analyzed for the loss of SLC10A4, as shown by an up-regulation in HEK293 and CAD cells. These experiments clearly of the nicotinic acetylcholine receptor subunits alpha1 showed that the cytoplasmic C-terminus of SLC10A4 and delta in the Slc10a4 knockout mice. The authors must contain dominant signals for its vesicular sorting. came to the conclusion that the loss of SLC10A4 may Previous studies with rat NTCP demonstrated that the result in reduced vesicular filling with the neurotrans - tyrosine-based sorting motifs YXXØ within its C-ter- 307 321 mitter acetylcholine at the neuromuscular junction [17]. minus (i.e. Y -E-K-I and Y -K-A-A) are involved in In a further study by Larhammar et  al. [15], the Slc10a4 membrane delivery [29]. Although the C-terminus of knockout mice were not different from their wild-type SLC10A4 also contains these potential tyrosine-based 406 419 littermates in a series of behavior tests, but showed a sorting motifs (Y -K–K-L und Y -G-T-V), SLC10A4 is hypoactive phenotype. Furthermore, the Slc10a4 knock- obviously not directed to the plasma membrane. In the out mice were hypersensitive to the psychostimulants same way, all chimeric constructs with the C-terminus amphetamine and tranylcypromine. In the central nerv- of SLC10A4 (i.e. NTCP-CtSLC10A4 and NtSLC10A4- ous system, the Slc10a4 knockout mice had reduced lev- NTCP-CtSLC10A4) showed a vesicular expression pat- els of dopamine, serotonin, norepinephrine and choline. tern and failed to transport typical NTCP substrates such Although no evidence for the direct transport of these as taurocholic acid and DHEAS. In contrast, both chi- neurotransmitters by SLC10A4 has been provided, syn- meras bearing the C-terminus of NTCP (i.e. SLC10A4- aptic vesicles from transgenic mice overexpressing the CtNTCP and NtNTCP-SLC10A4-CtNCTP) were SLC10A4 protein showed an increased uptake of [ H] detected at the plasma membrane. Further mutagenesis dopamine, which was probably due to higher synaptic experiments now have to localize the vesicular sorting vesicle acidification. Based on these data, the authors motifs of SLC10A4 at the amino acid level. speculate that SLC10A4 may transport any hitherto unknown organic anion compound, which would allow Conclusions the vesicular accumulation of higher amounts of protons, SLC10A4 is expressed in vesicular structures not only which then would increase the vesicular storage of neu- in neurons of the central and peripheral nervous sys- rotransmitters [15]. This conclusion is in agreement with tem, but also in neuronal cell lines such as SH-SY5Y and data from the present study, in which SLC10A4 showed CAD. Although different kinds of assays were applied to no transport activity for a series of neurotransmitters. screen for a transport function, SLC10A4 failed to show However, even suggested organic anion modulators such transport activity for dopamine, serotonin, norepineph- as ATP or DHEAS were not transported by SLC10A4, rine, histamine, acetylcholine, choline, acetate, aspar- meaning that the function of SLC10A4 as solute carrier is tate, glutamate, GABA, PREGS, DHEAS, E-3-S, and ATP, still a matter of speculation. indicating that SLC10A4 does not seem to be a typical Based on their findings, the Kullander group suggested neurotransmitter transporter. When the C-terminus of renaming SLC10A4 as “vesicular aminergic-associated SLC10A4 was replaced by the homologous sequence of transporter” (VAAT). However, as the naming of the NTCP, the SLC10A4-CtNTCP chimera revealed clear carriers of the SLC10 carrier family until now has only plasma membrane expression in CAD and HEK293 cells. been based on their molecular transport function, i.e. Vice versa, the C-terminus of SLC10A4 directed NTCP Na /taurocholate co-transporting polypeptide NTCP to intracellular vesicles, indicating that the sorting motifs (SLC10A1), apical sodium-dependent bile acid trans- of both carriers seem to be localized in the C-terminus. porter ASBT (SLC10A2), and sodium-dependent organic Further SLC10A4 transport studies should involve vesi- anion transporter SOAT (SLC10A6), we strongly suggest cle preparations from SLC10A4-deleted neuronal cell not renaming SLC10A4 before its molecular transport cultures or may use liposomal reconstitution of the function has been elucidated. SLC10A4 protein. Until then, the functional properties Schmidt et al. BMC Neurosci (2015) 16:35 Page 11 of 14 of the SLC10A4 orphan carrier protein still remain per cm in 24-well plates on poly-l-lysine-coated glass unknown. coverslips and cultured to a confluence of approxi - mately 40%. For immunofluorescence experiments, Methods the cells were washed with phosphate buffered saline Materials, chemicals, and radiochemicals (PBS, containing 137  mM NaCl, 2.7  mM KCl, 1.5  mM All of the chemicals, unless otherwise stated, were from KH PO , and 7.3  mM Na HPO , at pH 7.4) and fixed 2 4 2 4 Sigma-Aldrich (Taufkirchen, Germany). Citalopram was with 2% paraformaldehyde (PFA, Roth, Karlsruhe, Ger- purchased from Biotrend (Cologne, Germany) and col- many) in PBS for 15  min at 4°C. Then, cells were again lagenase D was from Serva (Heidelberg, Germany). [ H] washed with PBS and incubated with 20  mM glycine in DHEAS (70.5  Ci/mmol), [ H]E-3-S (45.6  Ci/mmol), PBS for 5  min. For permeabilization, cells were incu- 3 3 [ H]aspartate (11.3  Ci/mmol), [ H]GABA (76.2  Ci/ bated for 5 min with PBT buffer (0.2% Triton X-100 and 3 3 mmol), [ H]histamine (13.4  Ci/mmol), [ H]choline 20  mM Glycine in PBS). The non-specific binding sites chloride (66.7  Ci/mmol), [ H]norepinephrine (56.6  Ci/ were blocked with 1% bovine serum albumin (BSA) plus 3 3 mmol), [ H]serotonin (28.25  Ci/mmol), [ H]dopamine 4% goat serum (Sigma-Aldrich) in PBS for 30  min at 3 3 (38.7 Ci/mmol), [ H]glutamate (49.6 Ci/mmol), [ H]ATP room temperature. Then, the cells were incubated with (30.9  Ci/mmol), [ S]Adenosine 5′-(γ-thio) triphosphate the primary antibodies, rabbit anti-Slc10a4 (antibody (12.5  mCi/mmol) and [ H]acetylcholine iodide (99.7  Ci/ 1338 C, 1:1,000 dilution, see [13, 14] ), Sigma Prestige mmol) were purchased from PerkinElmer Life Sciences rabbit anti-SLC10A4 (1:500 dilution) [Sigma-Aldrich 3 3 (Boston, MA, USA). [ H]PREGS (20  Ci/mmol), [ H] Cat# HPA028835 RRID:AB_10603025], Abnova rab- taurocholic acid (10.0  Ci/mmol), [ H]acetate (150  mCi/ bit anti-Slc10a4 (1:500 dilution) [Abnova Corporation mmol), and [ H]lithocholic acid (50  Ci/mmol) were Cat# PAB14855 RRID:AB_10696081], or Abgent rabbit obtained from American Radiolabeled Chemicals (St. anti-SLC10A4 (1:500 dilution) [Abgent Cat# AP10250b Louis, MO, USA). RRID:AB_10821224], or with mouse anti-V5 monoclo- nal antibody (1:5,000 dilution) [Invitrogen Cat# R96025 Culture and differentiation of neuronal cell lines RRID:AB_159313] in blocking solution overnight at 4°C. Human neuroblastoma SH-SY5Y cells (obtained from The next day, cells were washed with PBS and incubated DSMZ Braunschweig, Germany) were maintained at with the fluorophore-labeled secondary antibody Cy3- 37°C in RPMI medium (Gibco, Karlsruhe, Germany) con- conjugated goat anti-rabbit IgG (1:800 dilution) [Jackson taining 10% fetal calf serum (FCS, Sigma-Aldrich) and 1% ImmunoResearch Cat# 111-165-003 RRID:AB_2338000] penicillin/streptomycin (P/S, containing 100 U/ml peni- or Alexa fluor 488-labelled goat-anti mouse (1:800 dilu - cillin and 100  mg/ml streptomycin). For differentiation, tion) [Molecular Probes (Invitrogen) Cat# A11001 the cultures at an approximate confluence of 40% were RRID:AB_141367] in blocking solution for 60  min at placed in serum-free RPMI medium supplemented with room temperature. After several washing steps with PBS, either 10 µM all-trans RA (Sigma-Aldrich) plus 10 ng/ml the cells were covered with a DAPI/methanol solution BMP-2 (BioCat, Heidelberg, Germany) or 10 ng/ml TGF- containing 1 µg/ml DAPI (Roche, Mannheim, Germany) β1 (Invitrogen, Karlsruhe, Germany) and were grown and incubated for 5  min at room temperature. The cells for 5  days. As a control, cells were grown in serum-free were rinsed with methanol, air dried and mounted onto RPMI medium without any supplements. Mouse cat- slides with ProLong Gold Antifade (Invitrogen) mount- echolaminergic CAD cells (obtained from European ing medium. Collection of Cell Cultures ECACC, Health Protection Agency, UK) were maintained at 37°C in Dulbecco’s Real‑time quantitative PCR analysis modified Eagle’s/Ham’s F-12 (1:1) medium (DMEM/F12) Relative expression analysis for SLC10A4/Slc10a4 was containing 8% FCS, 4  mM  l-glutamine and 1% P/S at performed with ABI PRISM 7300 technology (Applied 37°C. For cell differentiation, 4  ×  10 cells were seeded Biosystems, Darmstadt, Germany). RNA was isolated in 6-well or 24-well plates on poly-l-lysine-coated glass from human SH-SY5Y and mouse CAD cells with TriRea- coverslips and grown in serum-free medium for 24-72 h. gent (Sigma-Aldrich) and cDNA was reverse-transcribed Under these conditions, the cultures stopped proliferat- using the SuperScript III First Strand Synthesis Sys- ing and large neurite-like cell extensions appeared. tem (Invitrogen). PCR amplification was achieved with the TaqMan Gene Expression Assays Hs00293728 for Immunofluorescence analysis of SH‑SY5Y, CAD, human SLC10A4 and Mm00557788 for mouse Slc10a4 and HEK293 cells (Applied Biosystems). The expression data of human For immunofluorescence analysis, SH-SY5Y, CAD, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH, HEK293 cells were seeded at a density of 5  ×  10 cells Hs99999905) and mouse beta-actin (Mm00607939) were Schmidt et al. BMC Neurosci (2015) 16:35 Page 12 of 14 used as an endogenous control. Triplicate determina- with the primers listed in Table  3. All amplicons were tions were performed in a 96-well optical plate for each sequence-verified by DNA sequencing according to target, using 5  µl cDNA, 1.25  µl TaqMan Gene Expres- the GenBank Accession Nos. given in Table  3 and were sion Assay, 12.5  µl TaqMan Universal PCR Master Mix cloned into the pcDNA5/FRT/V5-His TOPO vector (Life (Applied Biosystems) and 6.25  µl water in each 25  µl Technologies) via T/A cloning. reaction. The plates were heated for 10 min at 95°C, and 40 cycles of 15  s at 95°C and 60  s at 60°C were applied. Generation of the SLC10A4/NTCP chimeras The relative expression (ΔC ) of each target was calcu- For generation of the SLC10A4/NTCP chimeric con- lated by subtracting the signal threshold cycle (C ) of the structs, the full-length SLC10A4-cDNA5 and NTCP- endogenous control from the C value of the target. pcDNA5 clones with C-terminal V5-tag were used. BmtI and HindIII restriction sites were introduced in the N-ter- Western blot analysis mini (amino acid positions A6 in the NTCP and A78 in Protein contents of CAD cell lysates were determined the SLC10A4) and C-termini (amino acid positions E296 with the BCA Protein Assay Kit (Novagen, Darmstadt, in the NTCP and E374 in the SLC10A4), respectively, Germany). Samples of 20  µg protein were mixed with by site-directed mutagenesis, as previously described in Laemmli Sample Buffer (Sigma-Aldrich), separated detail [8]. Then, by double digestion and re-ligation of the on a 12% SDS polyacrylamide gel and transferred to a appropriate fragments, the following chimeric constructs Hybond-ECL nitrocellulose membrane (Amersham Bio- were generated and sequence-verified by direct DNA sciences, Freiburg, Germany). The blotted membranes sequencing: SLC10A4-CtNTCP, NtNTCP-SLC10A4- were blocked with blocking solution containing 5% CtNTCP, NtSLC10A4-NTCP, NtSLC10A4-NTCP- ECL-blocking agent (Amersham Biosciences) in TBS-T CtSLC10A4, and NTCP-CtSLC10A4. For generation of (137  mM NaCl, 10  mM Tris, pH 8.0, 0.05% Tween-20), the N-terminally truncated 75ΔSLC10A4 construct, the followed by overnight exposure to antigen-specific pri - SLC10A4 open reading frame starting from amino acid mary antibodies at 4°C in the same buffer. After several 75 was PCR-amplified and cloned, whereby the codon for washing steps in TBS-T, the membranes were probed Gly75 was replaced by an artificial start codon ATG. with the appropriate horseradish-peroxidase-labeled secondary antibodies in TBS-T for 60 min at room tem- Transfection of HEK293 and CAD cells perature. Signals were developed using the Roti-Lumin HEK293 cells were grown in DMEM supplemented with ECL Detection Kit (Roth) and visualized by exposure to 10% FCS, 4  mM glutamine and 1% P/S. HEK293 and Hyperfilm ECL (Amersham Biosciences). CAD cells were grown at 37°C in 5% CO . For transient transfection, both cell lines were seeded in 24-well plates Cloning of the reference carriers at a density of 1.4–2.0 ×  10 cells per well and transfec- Full length transcripts covering the whole open reading tion was performed by using the Lipofectamine 2000 frame for human SLC10A4, DAT, CHT1, SERT, VMAT2, reagent according to the manufacturer’s instructions VAChT, NTCP, and mouse Oct1 were amplified by PCR (Invitrogen). For the establishment of stably transfected Table 3 Primers used for full-length carrier cloning Carrier Forward primer Reverse primer GenBank Accession Nos. Human SLC10A1 (Na /taurocholate co- 5′-tct cta gag gat gga ggc cca caa c-3′ 5′-ggc tgt gca agg gga gca gtc-3′ [GenBank:NM_003049] transporting polypeptide, NTCP) Human SLC10A4 [GenBank:NM_152679] 5′-acc gac ggg cag aac gac-3′ 5′-gag aga agt ctg agc ggt ttc-3′ Human SLC6A3 (dopamine transporter, 5′-ctc cca gtg tgc cca tga gta aga g-3′ 5′-cac ctt gag cca gtg gcg gag-3′ [GenBank:NM_001044] DAT ) Human SLC5A7 (choline transporter, 5′-aaa aat ggc ttt cca tgt gga agg-3′ 5′-ctg taa att atc ttc agt ccc ag-3′ [GenBank:NM_021815] CHT1) Human SLC6A4 (serotonin transporter, 5′-agg atg gag acg acg ccc ttg aat tc-3′ 5′-cac agc att caa gcg gat gtc ccc a-3′ [GenBank:NM_001045] SERT ) Human SLC18A2 (vesicular monoamine 5′-gcc atg gcc ctg agc gag ctg-3′ 5′-gtc act ttc aga ttc ttc atc ttc acc tat [GenBank:NM_003054] transporter, VMAT2) c-3′ Human SLC18A3 (vesicular acetylcho- [GenBank:NM_003055] 5′-cgg aag agc atc ggg gtg-3′ 5′-gct gcg ggt gta gta gta g-3′ line transporter, VAChT ) Mouse Slc22a1 (organic cation trans- 5′-att tca agc cac cgc agt tc-3′ 5′-ggt atg tgg gga ttt gcc t-3′ [GenBank:NM_009202] porter, Oct1) Schmidt et al. BMC Neurosci (2015) 16:35 Page 13 of 14 cells, Flp-In HEK293 cells were used (Invitrogen), as cells were permeabilized with 15 µM digitonin in PB for described previously [3]. Briefly, Flp-In HEK293 cells 15 min at 37°C, as reported [23]. The buffer was replaced were seeded in 6-well plates coated with poly-d-lysine at by 250 µl of PB containing the radiolabeled and non-radi- a density of 1 ×  10 cells per well and grown to 60–80% olabeled compounds. After incubation for the indicated confluence in antibiotic free medium. Then, the cells time at 37°C, the uptake was stopped, as described above. were transfected with 1  µg of the respective pcDNA5/ FRT/V5-His vector construct plus 7 µg of pOG44 vector. Transport studies in Xenopus laevis oocytes For the selection of positive clones, hygromycin B was The pcDNA5/FRT/V5-His constructs containing the used at a concentration of 150 µg/ml. cloned cDNAs were linearized with KpnI. After phenol/ chloroform extraction, the mMESSAGEmMACHINE Transport studies in HEK293 and CAD cells Kit (Ambion, Life Technologies) was used to generate a For transport studies, 24-well plates were coated with capped cRNA and the Poly(A) Tailing Kit (Ambion, Life poly-d-lysine for better attachment of the cells. Stably Technologies) was used to add a poly(A) tail to the RNA or transiently transfected cells were plated at a density transcripts. Afterwards, the cRNA was purified with the of 1.25  ×  10 cells per well and grown under standard MEGAclear Kit (Ambion, Life Technologies) according medium for 48–72  h. Before starting the transport to the manufacturer’s protocol. Oocytes were obtained 2+ experiments, cells were washed three times with PBS and from female X. laevis frogs and incubated in Ca -free pre-incubated with sodium transport buffer (142.9  mM OR-2 solution (82.5 mM NaCl, 5 mM HEPES–NaOH, pH NaCl, 4.7  mM KCl, 1.2  mM MgSO , 1.2  mM KH PO , 7.6, 2.5 mM KCl, 1 mM MgCl and 1 mM Na HPO ) sup- 4 2 4 2 2 4 1.8  mM CaCl , and 20  mM HEPES, adjusted to pH 7.4). plemented with 0.4  mg/ml collagenase type D (Serva) at When transport assays were performed in sodium-free 18°C overnight. Then, oocytes were further incubated in transport buffer, sodium chloride was substituted with modified Barth’s solution (88 mM NaCl, 15 mM HEPES– equimolar concentrations of choline chloride and for the NaOH, pH 7.6, 2.4 mM NaHCO , 1 mM KCl, 0.3 mM Ca transport assay with the substrate [ H]choline chloride (NO ) , 0.41 mM CaCl and 0.82 mM MgSO ) containing 3 2 2 4 it was substituted with equimolar concentrations of lith- 50 µg/ml gentamicin. Defolliculated oocytes were selected ium chloride. For transport studies with [ H]serotonin, and microinjected with 4.6 ng (46 nl) cRNA encoding for the uptake buffer contained 100  µM ascorbic acid and the carrier protein or with a corresponding volume of 100  µM pargyline, to prevent the degradation of seroto- water. After 3 days of culture in modified Barth’s medium, nin. Transport studies were performed by incubating the uptake of the indicated radiolabelled substrates was cells with 250 µl transport buffer containing the radiola - assessed at 25°C in transport buffer containing 100  mM beled and non-radiolabeled compounds for the indicated NaCl, 2 mM KCl, 1 mM CaCl , 1 mM MgCl , and 10 mM 2 2 time at 37°C. CAD cells could not be washed before start- HEPES-Tris, pH 7.5. After washing with the same buffer, ing the transport experiments because these cells easily each individual oocyte was dissolved in 500  µl of 10% detached after repeated handling. Therefore, these cells SDS. The radioactivity was counted after the addition of were incubated in 250 µl fresh medium plus 50 µl trans- 4 ml scintillation fluid in a liquid scintillation counter. port buffer containing the radiolabeled and non-radi - olabeled compounds. Uptake studies were stopped by Statistics removing the transport buffer and washing the cells five Columns are shown as mean  ±  SD. Prism software times with ice-cold PBS. Afterwards, cells were lysed in (GraphPad Software Inc., San Diego, CA, USA) was used 1 N NaOH with 0.1% SDS and the cell-associated radio- for data presentation and statistical analysis. Statisti- activity was measured by liquid scintillation counting. cal significance of two groups was analyzed by Student’s The protein content was determined using aliquots of the t test. Statistical analysis of more than two groups was lysed cells with BSA as the standard [30]. performed by one-way analysis of variance (ANOVA) fol- lowed by Bonferroni post hoc testing. Transport studies in permeabilized HEK293 cells Stably transfected HEK293 cells expressing SLC10A4, Abbreviations VMAT2 or VAChT were seeded in 24-well plates and ASBT: Apical sodium-dependent bile acid transporter; ATP: adenosine triphos- were grown until confluence. Then, cells were rinsed phate; BMP-2: bone morphogenetic protein 2; BSA: bovine serum albumin; CHT1: choline transporter; DAT: dopamine transporter; DHEAS: dehydroepi- three times with potassium-rich buffer (PB) contain - androsterone sulfate; E-3-S: estrone-3-sulfate; FCCP: carbonylcyanid-p-trifluo - ing 110  mM potassium tartrate, 5  mM glucose, 0.2% romethoxyphenylhydrazon; FCS: fetal calf serum; GABA: gamma-aminobutyric bovine serum albumin, 200  µM CaCl 1  mM ascorbic acid; HC-3: hemicholinium-3; HEK293: human embryonic kidneys 293 cells; 2, NTCP: Na /taurocholate co-transporting polypeptide; Oct1: organic cation acid, 10 µM pargyline and 20 mM piperazine-N,N′-bis(2- transporter 1; PB: potassium-rich buffer; PBS: phosphate buffered saline; PFA: ethanesulfonic acid) (PIPES), adjusted to pH 6.8. The paraformaldehyde; PREGS: pregnenolone sulfate; RA: retinoic acid; RFP: red Schmidt et al. BMC Neurosci (2015) 16:35 Page 14 of 14 fluorescent protein; SERT: serotonin transporter; SLC10: solute carrier family 10. Bijsmans ITGW, Bouwmeester RAM, Geyer J, Faber KN, van de Graaf 10; SOAT: sodium-dependent organic anion transporter; TGF-β: tumor growth SFJ (2012) Homo- and hetero-dimeric architecture of the human liver factor beta; VAChT: vesicular acetylcholine transporter; VMAT2: vesicular mono- Na -dependent taurocholate co-transporting protein. Biochem J amine transporter. 441(3):1007–1015 11. Jiang L, Alber J, Wang J, Du W, Yang X, Li X et al (2012) The Candida Author’s contributions albicans plasma membrane protein Rch1p, a member of the vertebrate 2+ SS carried out cloning of the carriers SLC10A4, DAT, SERT, CHT1, VMAT2, and SLC10 carrier family, is a novel regulator of cytosolic Ca homoeostasis. VAChT, performed the transport experiments in HEK293 cells and Xenopus Biochem J 444(3):497–502 laevis oocytes, was involved in drafting of the manuscript, and performed data 12. Alber J, Jiang L, Geyer J (2013) CaRch1p does not functionally interact 2+ presentation and statistical analysis. MM generated the SLC10A4/NTCP chi- with the high-affinity Ca influx system (HACS) of Candida albicans. meric constructs, and analyzed them by immunofluorescence and transport Yeast 30(11):449–457 studies. SB carried out immunofluorescence, Western blot, and qPCR analysis 13. Geyer J, Fernandes CF, Döring B, Burger S, Godoy JR, Rafalzik S et al (2008) on the neuronal cell lines, and was involved in drafting of the manuscript. JG Cloning and molecular characterization of the orphan carrier protein conceived the study, performed data interpretation, and drafted and finalized Slc10a4: expression in cholinergic neurons of the rat central nervous the manuscript. All authors read and approved the final manuscript. system. Neuroscience 152:990–1005 14. Burger S, Döring B, Hardt M, Beuerlein K, Gerstberger R, Geyer J (2011) Co- expression studies of the orphan carrier protein Slc10a4 and the vesicular Acknowledgements carriers VAChT and VMAT2 in the rat central and peripheral nervous We thank Anita Neubauer, Klaus Schuh, and Regina Leidolf for their excellent system. Neuroscience 193:109–121 technical support and would like to acknowledge Dr. Barbara Döring and Dina 15. Larhammar M, Patra K, Blunder M, Emilsson L, Peuckert C, Arvidsson E Kleinlützum for providing clones of NTCP and Oct1, respectively. This research et al (2014) SLC10A4 Is a vesicular amine-associated transporter modulat- was supported in part by the German Research Foundation DFG (Grant GE ing dopamine homeostasis. Biol Psychiatr 77:526–536 1921/3-1) and in part by Doktor Robert Pfleger-Stiftung (Bamberg, Germany). 16. Zelano J, Mikulovic S, Patra K, Kühnemund M, Larhammar M, Emils- son L et al (2013) The synaptic protein encoded by the gene Slc10A4 Compliance with ethical guidelines suppresses epileptiform activity and regulates sensitivity to cholinergic chemoconvulsants. Exp Neurol 239:73–81 Competing interests 17. Patra K, Lyons DJ, Bauer P, Hilscher MM, Sharma S, Leão RN et al (2014) A The authors declare that they have no competing interests. role for solute carrier family 10 member 4, or vesicular aminergic-associ- ated transporter, in structural remodelling and transmitter release at the Received: 28 January 2015 Accepted: 5 June 2015 mouse neuromuscular junction. Eur J Neurosci 41:316–327 18. Abe T, Kanemitu Y, Nakasone M, Kawahata I, Yamakuni T, Nakajima A et al (2013) SLC10A4 is a protease-activated transporter that transports bile acids. J Biochem 154(1):93–101 19. Gómez-Santos C, Ambrosio S, Ventura F, Ferrer I, Reiriz J (2002) TGF-β1 increases tyrosine hydroxylase expression by a mechanism blocked by References BMP-2 in human neuroblastoma SH-SY5Y cells. Brain Res 958(1):152–160 1. Geyer J, Wilke T, Petzinger E (2006) The solute carrier family SLC10: 20. Kohr D, Tschernatsch M, Schmitz K, Singh P, Kaps M, Schäfer KH et al more than a family of bile acid transporters regarding function and (2009) Autoantibodies in complex regional pain syndrome bind phylogenetic relationships. Naunyn Schmiedebergs Arch Pharmacol to a differentiation-dependent neuronal surface autoantigen. Pain 372(6):413–431 143(3):246–251 2. Döring B, Lütteke T, Geyer J, Petzinger E (2012) The SLC10 carrier family: 21. Qi Y, Wang JK, McMillian M, Chikaraishi DM (1997) Characterization of a transport functions and molecular structure. Curr Top Membr 70:105–168 CNS cell line, CAD, in which morphological differentiation is initiated by 3. Geyer J, Döring B, Meerkamp K, Ugele B, Bakhiya N, Fernandes serum deprivation. J Neurosci 17(4):1217–1225 CF et al (2007) Cloning and functional characterization of human 22. Geyer J, Godoy JR, Petzinger E (2004) Identification of a sodium-depend- sodium-dependent organic anion transporter (SLC10A6). J Biol Chem ent organic anion transporter from rat adrenal gland. Biochem Biophys 282(27):19728–19741 Res Commun. 316(2):300–306 4. Fietz D, Bakhaus K, Wapelhorst B, Grosser G, Günther S, Alber J et al (2013) 23. Erickson JD, Eiden LE, Hoffman BJ (1992) Expression cloning of a Membrane transporters for sulfated steroids in the human testis-cellular reserpine-sensitive vesicular monoamine transporter. Proc Natl Acad Sci localization, expression pattern and functional analysis. PLoS One USA 89(22):10993–10997 8(5):e62638 24. Rupprecht R (1997) The neuropsychopharmacological potential of neuro- 5. Grosser G, Fietz D, Günther S, Bakhaus K, Schweigmann H, Ugele B et al active steroids. J Psychiatr Res 31(3):297–314 (2013) Cloning and functional characterization of the mouse sodium- 25. Reddy DS (2003) Pharmacology of endogenous neuroactive steroids. Crit dependent organic anion transporter Soat (Slc10a6). J Steroid Biochem Rev Neurobiol 15(3–4):197–234 Mol Biol 138:90–99 26. Zheng P (2009) Neuroactive steroid regulation of neurotransmitter 6. Splinter PL, Lazaridis KN, Dawson PA, LaRusso NF (2006) Cloning and release in the CNS: action, mechanism and possible significance. Prog expression of SLC10A4, a putative organic anion transport protein. World Neurobiol 89(2):134–152 J Gastroenterol 12(42):6797–6805 27. Jiang Y, Wu J, Hua Y, Keep RF, Xiang J, Hoff JT et al (2002) Thrombin-recep - 7. Fernandes CF, Godoy JR, Döring B, Cavalcanti MC, Bergmann M, Petz- tor activation and thrombin-induced brain tolerance. J Cereb Blood Flow inger E et al (2007) The novel putative bile acid transporter SLC10A5 is Metab 22(4):404–410 highly expressed in liver and kidney. Biochem Biophys Res Commun 28. Goldberg AA, Beach A, Davies GF, Harkness TAA, Leblanc A, Titorenko VI 361(1):26–32 (2011) Lithocholic bile acid selectively kills neuroblastoma cells, while 8. Godoy JR, Fernandes C, Döring B, Beuerlein K, Petzinger E, Geyer J (2007) sparing normal neuronal cells. Oncotarget 2(10):761–782 Molecular and phylogenetic characterization of a novel putative mem- 29. Sun AQ, Arrese MA, Zeng L, Swaby I, Zhou MM, Suchy FJ (2001) The rat brane transporter (SLC10A7), conserved in vertebrates and bacteria. Eur J liver Na /bile acid cotransporter. Importance of the cytoplasmic tail to Cell Biol 86(8):445–460 function and plasma membrane targeting. J Biol Chem 276(9):6825–6833 9. Visser WE, Wong WS, van Mullem AAA, Friesema ECH, Geyer J, Visser TJ 30. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measure- (2010) Study of the transport of thyroid hormone by transporters of the ment with the Folin phenol reagent. J Biol Chem 193(1):265–275 SLC10 family. Mol Cell Endocrinol 315(1–2):138–145 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Neuroscience Springer Journals

Expression, sorting and transport studies for the orphan carrier SLC10A4 in neuronal and non-neuronal cell lines and in Xenopus laevis oocytes

Loading next page...
 
/lp/springer-journals/expression-sorting-and-transport-studies-for-the-orphan-carrier-Yzb5uff08i

References (30)

Publisher
Springer Journals
Copyright
Copyright © 2015 by Schmidt et al.
Subject
Biomedicine; Neurosciences; Neurobiology; Animal Models
eISSN
1471-2202
DOI
10.1186/s12868-015-0174-2
pmid
26084360
Publisher site
See Article on Publisher Site

Abstract

Background: SLC10A4 belongs to the solute carrier family SLC10 whose founding members are the Na /tauro- cholate co-transporting polypeptide (NTCP, SLC10A1) and the apical sodium-dependent bile acid transporter (ASBT, SLC10A2). These carriers maintain the enterohepatic circulation of bile acids between the liver and the gut. SLC10A4 was identified as a novel member of the SLC10 carrier family with the highest phylogenetic relationship to NTCP. The SLC10A4 protein was detected in synaptic vesicles of cholinergic and monoaminergic neurons of the peripheral and central nervous system, suggesting a transport function for any kind of neurotransmitter. Therefore, in the present study, we performed systematic transport screenings for SLC10A4 and also aimed to identify the vesicular sorting domain of the SLC10A4 protein. Results: We detected a vesicle-like expression pattern of the SLC10A4 protein in the neuronal cell lines SH-SY5Y and CAD. Differentiation of these cells to the neuronal phenotype altered neither SLC10A4 gene expression nor its vesicu- lar expression pattern. Functional transport studies with different neurotransmitters, bile acids and steroid sulfates were performed in SLC10A4-transfected HEK293 cells, SLC10A4-transfected CAD cells and in Xenopus laevis oocytes. For these studies, transport by the dopamine transporter DAT, the serotonin transporter SERT, the choline transporter CHT1, the vesicular monoamine transporter VMAT2, the organic cation transporter Oct1, and NTCP were used as posi- tive control. SLC10A4 failed to show transport activity for dopamine, serotonin, norepinephrine, histamine, acetylcho- line, choline, acetate, aspartate, glutamate, gamma-aminobutyric acid, pregnenolone sulfate, dehydroepiandroster- one sulfate, estrone-3-sulfate, and adenosine triphosphate, at least in the transport assays used. When the C-terminus of SLC10A4 was replaced by the homologous sequence of NTCP, the SLC10A4-NTCP chimeric protein revealed clear plasma membrane expression in CAD and HEK293 cells. But this chimera also did not show any transport activity, even when the N-terminal domain of SLC10A4 was deleted by mutagenesis. Conclusions: Although different kinds of assays were used to screen for transport function, SLC10A4 failed to show transport activity for a series of neurotransmitters and neuromodulators, indicating that SLC10A4 does not seem to represent a typical neurotransmitter transporter such as DAT, SERT, CHT1 or VMAT2. Keywords: SLC10A4, Transport, NTCP, Neurotransmitter, Sorting *Correspondence: joachim.m.geyer@vetmed.uni-giessen.de Institute of Pharmacology and Toxicology, Justus Liebig University of Giessen, Schubertstr. 81, 35392 Giessen, Germany © 2015 Schmidt et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Schmidt et al. BMC Neurosci (2015) 16:35 Page 2 of 14 with an in vitro approach. As the endogenous expression Background of SLC10A4 is exclusively directed to neuronal cells and The orphan carrier SLC10A4 belongs to the solute carrier mast cells [13, 14], neuronal cell cultures were thought to family SLC10, whose founding members are the Na /tau- be the most appropriate for this purpose. Therefore, we rocholate co-transporting polypeptide NTCP (SLC10A1) analyzed SLC10A4 expression in the human neuroblas- and the apical sodium-dependent bile acid transporter toma cell line SH-SY5Y as well as in the mouse cell line ASBT (SLC10A2) [1]. These carriers maintain the enter - CAD (Cath.a-differentiated neuronal cells, originating ohepatic circulation of bile acids by sodium-dependent from the locus coeruleus in the brainstem) with different bile acid reabsorption in the gut via ASBT and sodium- SLC10A4-directed antibodies. dependent bile acid transport into hepatocytes via NTCP SH-SY5Y cells showed a typical neuroblast-like appear- [2]. A further member of this carrier family, the sodium- ance with small, round cell bodies and occasional short dependent organic anion transporter SOAT (SLC10A6), extensions. Under incubation with retinoic acid (RA) is expressed in spermatocytes in man and mouse and has transport activity for sulfo-conjugated steroid hor and the neurotropic factors tumor growth factor beta (TGF-β1) and bone morphogenetic protein 2 (BMP-2), mones [3–5]. The other members of this carrier family, the cells stopped proliferation and developed neurite-like SLC10A3, SLC10A4, SLC10A5, and SLC10A7, have been long extensions, as described previously [19, 20]. Under characterized at the molecular and expression level, but both conditions, SLC10A4 showed a clear vesicle-like still remain orphan carriers [1, 6–12]. expression pattern in the SH-SY5Y cells and was detect- The SLC10A4 transcript was first cloned from rat in our able even along the long neurite-like outgrowths, indi- group in 2004 and showed predominant expression in the central nervous system in man, rat, and mouse [13]. Fur cating sorting of the SLC10A4 protein to the synaptic direction of the differentiated SH-SY5Y cells (Figure  1b). ther expression analyses on the protein level localized At the RNA level, SLC10A4 showed an overall higher SLC10A4 to cholinergic and monoaminergic neurons of expression in the SH-SY5Y cells compared with vesicu- the central and peripheral nervous system as well as to lar acetylcholine transporter (VAChT) and vesicular mast cells [13–15]. In contrast to NTCP, ASBT, and SOAT, monoamine transporter 2 (VMAT2) (data not shown), the SLC10A4 protein is not directed to the plasma mem- but incubation with TGF-β1  +  RA or BMP-2  +  RA did brane, but typically showed a vesicular expression pattern not significantly affect the SLC10A4 mRNA expression in neuronal cells, mast cells, and even in transiently or sta- levels, indicating that SLC10A4 expression is not regu- bly transfected cells lines [6, 13–15]. Based on this expres- lated by the RA, BMP-2, or TGF-β1 triggered signaling sion pattern, it was assumed that SLC10A4 may represent a cascades (Figure  1a). Although transient transfection of novel vesicular carrier for any kind of neurotransmitter or SLC10A4 into SH-SY5Y revealed an identical expression neuromodulator [1, 6, 13–15]. More recently, in a series of pattern compared with the endogenous expression, as very elegant experiments on Slc10a4 knockout mice it was shown for an SLC10A4-RFP construct in Figure  1c, the shown by the Kullander group that these mice are hyper- transfection rate of these cells could not be enhanced sensitive to the psychostimulants amphetamine and tra- above 20% by different transfection methods (lipofection, nylcypromine, and have an altered response to cholinergic non-liposomal transfection, electroporation), meaning stimuli at the neuromuscular junction and in the central that SH-SY5Y cells overexpressing SLC10A4 vs. non- cholinergic system, suggesting that SLC10A4 may contrib- transfected SH-SY5Y cells could not be used for trans- ute to the vesicular storage or release of neurotransmitters port studies. For the same reason, down-regulation of [15–17]. Therefore, in the present study, we performed sys - SLC10A4 expression by transfection of SLC10A4 siRNA tematic transport screenings for SLC10A4 in transfected prior to transport experiments was also not considered. neuronal and HEK293 cell lines as well as in Xenopus lae- Similar to the SH-SY5Y cells, the SLC10A4 protein vis oocytes and also aimed to identify the vesicular sorting was detected in vesicular structures in mouse CAD cells. domain of the SLC10A4 protein. Although we have not These cells were differentiated to a neuronal phenotype by identified a transported substrate for SLC10A4 to date, serum depletion, as previously reported [21]. As shown in recent descriptions of taurocholic acid and lithocholic acid Figure 1e, under these conditions, long neurite-like exten- transport by a thrombin-modified variant of SLC10A4 [18] sions were formed which were highly immunoreactive for encouraged us to present our data to provide a broader basis for further SLC10A4 transport studies. the anti-Slc10a4 antibody. However, even in these cells, RNA expression of SLC10A4 was not significantly affected by neuronal differentiation (Figure  1d). Although CAD Results cells could easily be transfected with different SLC10A4 Endogenous expression of SLC10A4 in neuronal cell lines constructs with transfection rates above 80%, these cells The primary goal of the present study was to identify a significantly detached after the transfection procedure transported substrate for the orphan carrier SLC10A4 Schmidt et al. BMC Neurosci (2015) 16:35 Page 3 of 14 Figure 1 Expression and subcellular localization of SLC10A4 in SH-SY5Y and CAD cells. a Relative SLC10A4 gene expression in SH-SY5Y cells after differentiation with TGF-ß1 + RA or BMP-2 + RA. Values represent mean ± SD of triplicate measurements. b Immunofluorescence analysis of the subcellular expression of the SLC10A4 protein in SH-SY5Y cells. Cells were either untreated (UD), or were differentiated with TGF-ß1 + RA or BMP-2 + RA over 4 days prior to immunolabeling. The SLC10A4 protein was detected with the anti-Slc10a4 1338 C antibody (1:1,000) and the Cy3- labelled anti-rabbit secondary antibody (1:800, red fluorescence) and nuclei were stained with DAPI (blue fluorescence). In all cases, the SLC10A4 protein showed a vesicle-like expression pattern within the perikarya and along the neurite-like cellular protrusions. c Even when a fluorescence- tagged SLC10A4-RFP construct was transiently transfected into SH-SY5Y cells, the SLC10A4-RFP protein showed a clear vesicular sorting pattern. d Relative Slc10a4 gene expression analysis in differentiated (Diff ) and undifferentiated (UD) CAD cells. The values represent mean ± SD of triplicate measurements. e Endogenous expression of the SLC10A4 protein in CAD cells, cultivated in FCS containing medium (UD) or FCS-free medium (Diff ). The SLC10A4 protein was detected with the anti-Slc10a4 1338 C antibody (1:500) and the Cy3-labelled secondary antibody (1:800, red fluo - rescence). For control, the primary anti-Slc10a4 antibody was omitted (control) or the antibody was pre-incubated with the immunizing peptide (peptide blocking). f Immunofluorescence detection of the SLC10A4 protein was performed with different SLC10A4-directed antibodies (green fluorescence): self-generated polyclonal rabbit anti-Slc10a4 1338 C antibody, rabbit anti-SLC10A4 Sigma Prestige antibody, rabbit anti-Slc10a4 Abnova antibody, and rabbit anti-SLC10A4 Abgent antibody. Membrane protein enriched fractions of the CAD cells were also subjected to Western Blot analysis with the same antibodies and revealed specific bands for the SLC10A4 protein at an apparent molecular weight of 30–32 kDa. Prestige rabbit anti-SLC10A4), the 1–27 amino acids of the and then could not be further subjected to standard trans- N-terminus of mouse and rat SLC10A4 protein (Abnova port assays. Therefore, overexpression of SLC10A4 or rabbit anti-Slc10a4), and the 377–406 amino acids of the down-expression by RNAi transfection of these cells was C-terminus of the human SLC10A4 protein (Abgent rab- not practicable for transport screening assays. bit anti-SLC10A4)—a clear vesicle-like expression pat- Localization of the SLC10A4 protein in the central and tern. The mouse SLC10A4 protein consistently showed an peripheral nervous system of the rat, as well as in rat PC12 apparent molecular weight of 30–32  kDa in the Western cells [13, 14], human SH-SY5Y cells, and mouse CAD blot (Figure  1f). This is below the calculated molecular cells (present study), was performed by us with a self- weight of 47 kDa, which has been reported for other mem- generated polyclonal rabbit antibody—1338 C—directed bers of the SLC10 carrier family as well [8]. against the amino acid residues 422–437 (VGTDDLVL- METTQTSL) of the deduced rat Slc10a4 protein sequence In vitro transport studies on SLC10A4 with candidate [GenBank:AAV80706]. In order to further verify SLC10A4 substrates expression in neuronal cell cultures with different antibod - As these neuronal cell lines were difficult to handle for ies, we screened mouse CAD cells by immunofluorescence standard transport experiments, we decided to use trans- and Western blot analysis, and found by all of the used fected human embryonic kidney (HEK) 293 cells as well as antibodies—directed against the 324–437 amino acids SLC10A4 cRNA injected X. laevis oocytes, which are both of the C-terminus of human SLC10A4 protein (Sigma Schmidt et al. BMC Neurosci (2015) 16:35 Page 4 of 14 well-established in vitro cell models for transport measure- In our previous study, we reported co-expression of the ments [3, 22]. We previously reported that, apart from its SLC10A4 protein with VMAT2 in synaptic vesicles of predominant intracellular expression, the SLC10A4 pro- rat brain preparations [14], so we next intended to ana- tein can to a lower extent also be detected in the plasma lyze the SLC10A4 transport in direct comparison with membrane of transfected HEK293 cells [13]. Therefore, VMAT2. In this approach, we used HEK293 cells stably we first used HEK293 cells transiently transfected with the transfected with human SLC10A4 or human VMAT2, human SLC10A4 construct for classical uptake measure- which were permeabilized with digitonin in order to ment in intact cells. As positive controls, the plasma mem- facilitate access of the transport substrate to the site of brane neurotransmitter transporters dopamine transporter carrier expression in intracellular vesicular structures, (DAT), choline transporter (CHT1), and serotonin trans- as described before [23]. As shown in Figure  3, VMAT2 porter (SERT) were also transfected, and empty-vector showed significant transport activity for [ H]seroto- 3 3 transfected cells served as negative controls. Whereas DAT nin, [ H]norepinephrine, and [ H]dopamine which was showed a significant nomifensine-sensitive and sodium- dependent on the presence of adenosine triphosphate dependent uptake of [ H]dopamine, SLC10A4-transfected (ATP) in the transport buffer as well as sensitive to tetra - HEK293 cells tended to accumulate more dopamine com- benazin (TBZ) and carbonylcyanide-p-trifluoromethoxy - pared with the control, but without reaching a level of sig- phenylhydrazon (FCCP). However, in contrast, SLC10A4 nificance (Figure  2). SLC10A4 showed no transport activity showed no transport activity for serotonin, norepi- 3 3 3 for [ H]norepinephrine, [ H]serotonin and [ H]choline nephrine, and dopamine at all in this vesicular trans- at all, whereas DAT, SERT and CHT1 significantly trans - port assay. We also could not demonstrate a transport ported their substrates in a sodium-dependent manner. of acetylcholine, neither for SLC10A4 nor for VAChT Figure 2 Transport measurements in transiently transfected HEK293 cells. HEK293 cells were transiently transfected with the indicated carriers 3 3 3 3 SLC10A4, DAT, CHT1, or SERT, respectively. The uptake of 5 µM [ H]dopamine, 5 µM [ H]norepinephrine, 5 µM [ H]choline, or 5 µM [ H]serotonin was + + + measured over the given time periods in the presence and absence of Na (for CHT1 Na was replaced by Li ). Transport via DAT, CHT1, and SERT was blocked by the specific inhibitors nomifensine (10 µM), hemicholinium-3 (HC-3, 1 µM), and citalopram (2 µM), respectively. Values represent mean ± SD of representative experiments, each with quadruplicate determinations (n = 4). *Significantly different from control with p < 0.001. Significantly different from positive uptake, p < 0.001. Schmidt et al. BMC Neurosci (2015) 16:35 Page 5 of 14 Figure 3 Transport measurements in digitonin permeabilized SLC10A4-HEK293 and VMAT2-HEK293 cells. Prior to transport measurements, stably transfected human SLC10A4-HEK293 cells and human VMAT2-HEK293 cells were pre-incubated with 15 µM digitonin for 10 min for permeabiliza- 3 3 3 tion. Then the uptake of 400 nM [ H]serotonin, [ H]norepinephrine, or [ H]dopamine was measured over 10 min in the presence and absence of 5 mM ATP in the transport buffer. In addition to ATP, 2 µM of the potent VMAT2 inhibitor tetrabenazine ( TBZ) or 5 µM of the proton ionophore carbonylcyanide-p-trifluoromethoxyphenylhydrazon (FCCP) were added to the transport buffer as indicated. After 10 min, the cells were washed with ice-cold PBS, lysed, and subjected to scintillation counting. The values represent mean ± SD of one representative experiment (for dopamine, n = 4) or two independent experiments (for serotonin and norepinephrine, n = 8). *Significantly different from control with p < 0.05. Significantly different from positive uptake, p < 0.05. thrombin treatment and thereby transport activity for in permeabilized HEK293 cell lines transfected with the taurocholic acid and lithocholic acid may be activated respective construct (data not shown). [18]. Although these data were obtained in a neuronal cell In previous studies at our institute we learned that the culture of TE671 cells, in the present study, we intended transport rates for a particular carrier may be higher in to reproduce these data on the isolated over-expressed the X. laevis oocytes expression system compared with SLC10A4 protein and used HEK293 cells stably trans transfected cell cultures. Furthermore, we demonstrated - in a previous study that at least the rat SLC10A4 protein fected with SLC10A4 as well as with NTCP for the con- is expressed at the plasma membrane of X. laevis oocytes trol. Both, SLC10A4-HEK293 and NTCP-HEK293 cell [13]. Therefore, we injected SLC10A4 cRNA in X. laevis lines were treated with 1 U/200  µl of thrombin for 3  h 3 3 oocytes and screened for transport activity with a series prior to transport measurements with [ H]DHEAS, [ H] 3 3 of neurotransmitters and further SLC10A4 candidate taurocholic acid, [ H]PREGS, and [ H]lithocholic acid. substrates. For these experiments, water-injected oocytes As expected, NTCP showed significant transport activ - served as negative control and, if appropriate, oocytes ity for DHEAS, taurocholic acid and PREGS, and was not injected with cRNA coding for SERT, DAT, NTCP, or affected by thrombin treatment. In contrast, SLC10A4 mouse organic cation transporter 1 (Oct1) were used as did not transport any of these compounds, even after positive controls. As shown in Figure  4, SERT, DAT, and pre-treatment with thrombin (Figure  5). No transport Oct1 showed significant transport activity for seroto - activity could be observed for lithocholic acid at all, nei- nin, dopamine, and histamine, respectively, but SLC10A4 ther in NTCP nor in the SLC10A4-expressing HEK293 cRNA injected oocytes were not different from the con - cells. Therefore, transport activity of SLC10A4 could not trol. Furthermore, the neurosteroids pregnenolone sulfate be activated by thrombin treatment in the cell model (PREGS) and dehydroepiandrosterone sulfate (DHEAS) as used. well as the bile acid taurocholic acid, were not transported by SLC10A4 in this transport assay, although NTCP Localization of the vesicular sorting domain in the showed significant transport activity for all of them. Apart C‑terminus of SLC10A4 from these candidate substrates, SLC10A4 also showed no A further aim of the present study was to localize the transport activity for aspartate, acetate, choline, gamma- particular domain of SLC10A4 that is responsible for aminobutyric acid (GABA), norepinephrine, glutamate, the vesicular sorting of this protein. This question was acetylcholine, estrone-3-sulfate (E-3-S), lithocholic acid, of particular interest, as mutation of this domain might and ATP in the X. laevis oocytes model (Table 1). redirect the SLC10A4 protein to the plasma mem- Very recently, it was reported that the N-terminal brane, which would facilitate further substrate screen- domain of the SLC10A4 protein could be cleaved by ing. As within the SLC10 carrier family, NTCP, which is Schmidt et al. BMC Neurosci (2015) 16:35 Page 6 of 14 Table 1 Transport studies in Xenopus laevis oocytes SLC10A4 Control Ratio Uptake (pmol/oocyte/60 min) [ H]Histamine (10 µM) 0.29 ± 0.02 0.27 ± 0.01 1.0 [ H]Aspartate (1 µM) 0.22 ± 0.01 0.28 ± 0.01 0.8 [ H]Acetate (25 µM) 8.79 ± 0.51 8.36 ± 0.72 1.0 [ H]Choline (1 µM) 2.68 ± 0.52 2.20 ± 0.19 1.2 [ H]GABA (1 µM) 0.52 ± 0.05 0.47 ± 0.04 1.1 [ H]Norepinephrine (1 µM) 0.16 ± 0.01 0.21 ± 0.02 0.8 [ H]Glutamate (1 µM) 0.014 ± 0.001 0.014 ± 0.002 1.0 Uptake (pmol/oocyte/30 min) [ H]Acetylcholine (10 µM) 43.51 ± 3.62 37.51 ± 3.67 1.2 [ H]Dopamine (10 µM) 1.36 ± 0.06 1.43 ± 0.05 1.0 [ H]PREGS (1 µM) 1.19 ± 0.09 1.06 ± 0.09 1.1 [ H]DHEAS (1 µM) 0.043 ± 0.002 0.043 ± 0.004 1.0 [ H]E-3-S (1 µM) 0.050 ± 0.001 0.060 ± 0.005 0.8 Uptake (pmol/oocyte/10 min) [ H]Lithocholic acid (0.3 µM) 3.68 ± 0.23 4.13 ± 0.07 0.9 [ H]Taurocholic acid (0.3 µM) 0.0158 ± 0.0009 0.0159 ± 0.0007 1.0 [ H]ATP (5 mM) 70.66 ± 9.06 68.84 ± 8.26 1.0 [ S]Adenosine 5`-(γ-thio) 69.76 ± 4.36 67.73 ± 2.53 1.0 triphosphate (5 mM) Xenopus laevis oocytes were injected with human SLC10A4 cRNA or with water (control). Values represent uptake of the indicated compound as mean  ±  SD of at least 10 oocytes per group. Ratio represents uptake into SLC10A4 cRNA- injected oocytes divided by uptake into water-injected oocytes. 75ΔSLC10A4 mutant which was truncated by the first 75  N-terminal amino acids. All constructs were trans- fected in neuronal CAD cells as well as in HEK293 cells and used for immunofluorescence analysis of protein localization (Figure  6b) as well as for transport experi- ments with SLC10A4 candidate substrates (Figure  6c; Figure 4 Transport measurements in Xenopus laevis oocytes. Xenopus laevis oocytes were injected with cRNA coding for human Table  2). As shown in Figure  6b, NTCP showed clear SLC10A4, SERT, DAT, or NTCP as well as mouse Oct1. Uptake of [ H] plasma membrane localization in CAD cells, whereas 3 3 3 3 serotonin, [ H]histamine, [ H]PREGS, [ H]dopamine, [ H]DHEAS, or SLC10A4 showed expression in intracellular vesicles. [ H]taurocholic acid, each at 1 µM, was measured over a time period Interestingly, both chimeras bearing the C-terminus of of 10–60 min as indicated in the presence of sodium chloride in the NTCP (i.e. SLC10A4-CtNTCP and NtNTCP-SLC10A4- transport buffer. SERT, Oct1, and DAT served as controls for the trans- 3 3 3 port of [ H]serotonin, [ H]histamine, and [ H]dopamine, respectively. CtNCTP) were detected at the plasma membrane, NTCP was the reference carrier for PREGS, DHEAS and taurocholic whereas all chimeric proteins with the C-terminus of acid. Afterwards, the oocytes were washed with ice-cold transport SLC10A4 (i.e. NTCP-CtSLC10A4, NtSLC10A4-NTCP- buffer, lysed and subjected to scintillation counting. The values CtSLC10A4) were detected in intracellular vesicles. This represent mean ± SD of one representative experiment with n = 10 sorting pattern was basically the same when these con- oocytes each. *Significantly different from control with p < 0.001. structs were transfected and analyzed in HEK293 cells (data not shown). This indicates that the C-termini of NTCP and SLC10A4 are dominant for the sorting to expressed at the plasma membrane, is the most related the plasma membrane and intracellular vesicles, respec- carrier to SLC10A4, we generated several SLC10A4/ tively. In contrast, truncation of the SLC10A4  N-ter- NTCP chimeric constructs in which the C-terminal minus (75ΔSLC10A4 mutant) as well as transfer of the and N-terminal domains were interchanged between SLC10A4  N-terminus into the NTCP (NtSLC10A4- both carriers (Figure  6a). Furthermore, we generated a NTCP chimera) did not affect the sorting of SLC10A4 Schmidt et al. BMC Neurosci (2015) 16:35 Page 7 of 14 Figure 5 Transport measurements in stably transfected SLC10A4-HEK293 and NTCP-HEK293 cells after thrombin treatment. For the transport 3 3 measurements, one part of the cells was pre-incubated with 1 U/200 µl thrombin over 3 h (+Thrombin), before the uptake of [ H]DHEAS, [ H] 3 3 taurocholic acid, [ H]PREGS, or [ H]lithocholic acid (each at 300 nM) was measured over a time period of 10 min at 37°C. The cells were washed with ice-cold PBS, lysed, and subjected to scintillation counting. The values represent mean ± SD of two independent experiments each with triplicate determinations. *Significantly different from control with p < 0.001; n.s. not significantly different. and NTCP, respectively, indicating that the C-terminus, Discussion but not the N-terminus of the SLC10A4 protein is rel- SLC10A4 is a member of the SLC10 carrier family com- evant for the protein sorting. Apart from localizing monly referred to as the “family of sodium-dependent the sorting domains in the SLC10A4 protein, several bile acid transporters”. The rat Slc10a4 transcript was of the constructs were of particular interest for trans- first cloned in our group in 2004 [GenBank:AY825923] port experiments. (I) When the N-terminal domain [13] and the human SLC10A4 sequence was published of SLC10A4, which was previously supposed to hin- in 2006 [6]. The SLC10A4 protein consists of 437 amino der substrate binding to the SLC10A4 carrier protein acids in man, rat, and mouse, and shows a close phylo- [18], were transferred to the NTCP in order to produce genetic relationship to the sodium-dependent bile acid a NtSLC10A4-NTCP elongated chimera, the trans- transporter NTCP. However, whereas NTCP is specifi - port function for taurocholic acid remained completely cally expressed at the sinusoidal plasma membrane of intact compared with wild-type NTCP (Figure  6c). (II) hepatocytes [2], the SLC10A4 protein was localized in Furthermore, activation of the SLC10A4-mediated tau- cholinergic and monoaminergic neurons of the central rocholic acid transport by thrombin cleavage of the and peripheral nervous system as well as in pheochro- N-terminus was supposed [18]. However, for the N-ter- mocytoma PC12 cells and mast cells [13–15]. Based on minally truncated 75ΔSLC10A4 protein, as well as for its genetic classification, SLC10A4 first was suggested to the NtNTCP-SLC10A4-CtNTCP chimera, still no trans- be a novel carrier for already established substrates of the port activity for taurocholic acid could be observed. (III) SLC10 carrier family, including bile acids and sulfo-con- Just by replacing the SLC10A4 C-terminal 63 amino jugated steroid hormones [1, 6, 13]. However, based on acids by the corresponding C-terminus of NTCP in the its expression pattern, it seems more likely that SLC10A4 SLC10A4-CtNTCP and NtNTCP-SLC10A4-CtNTCP plays a role for the vesicular storage or exocytosis of any chimera, the SLC10A4 protein was directed to the kind of neurotransmitter or mediator in neurons and plasma membrane. Therefore, these chimeras represent mast cells [14]. In previous experiments, the neuroster- an interesting tool for SLC10A4 substrate screening. As oids PREGS and DHEAS were already considered to be the CAD cells detached after transfection of the respec- particularly promising candidate substrates. These neu - tive constructs when they were further processed for the rosteroids can modulate several postsynaptic receptor routine transport assay, the radiolabelled candidate sub- systems and even interfere with the release of multiple strates were added directly into the cell culture medium neurotransmitters including acetylcholine, norepineph- and cell associated radioactivity was analyzed. However, rine, dopamine and serotonin [24–26]. However, until neither SLC10A4-CtNTCP nor NtNTCP-SLC10A4- now, all transport measurements performed with E-3-S, CtNTCP showed transport activity so far for taurocholic DHEAS, and PREGS have so far failed to show transport acid and serotonin (Figure  6c) as well as for DHEAS, function for SLC10A4 [6, 13] (present study). However, a PREGS, E-3-S, dopamine, glutamate, histamine, acetyl- more systematic transport screening also including clas- choline, and choline (Table 2). sical neurotransmitters as candidate substrates has not Schmidt et al. BMC Neurosci (2015) 16:35 Page 8 of 14 Figure 6 Localization and transport function of SLC10A4/NTCP chimeras in CAD cells. a The shown SLC10A4/NTCP chimeric constructs were used. All chimeras were generated based on the full length sequences of SLC10A4 (grey marked transmembrane domains and loops with continuous lines) and NTCP (white transmembrane domains and loops as dotted lines), both with c-terminal V5-tag. Potential glycosylation sites were marked by “Y”. b All constructs were transiently transfected in CAD cells and cellular localization was analyzed by immunofluorescence microscopy using rabbit anti-V5 antibody and donkey Cy3-labelled anti-rabbit secondary antibody. Nuclei were stained with DAPI. Whereas SLC10A4 showed a clear vesicle-like expression pattern, the immunofluorescence signals for NTCP, NtSLC10A4-NTCP, SLC10A4-CtNTCP, and NtNTCP-SLC10A4-CtNTCP were clearly directed to the plasma membrane. When the 75 N-terminal amino acids were deleted in SLC10A4, the 75ΔSLC10A4 protein retained its vesicle-like intracellular expression comparable with full-length SLC10A4. c The SLC10A4/NTCP chimeras were also used for transport studies after 3 3 transient transfection into CAD cells with [ H]taurocholic acid and [ H]serotonin, each at 5 µM. These measurements were performed by incubating the cells for 60 min at 37°C in 250 µl cell medium with 50 µl sodium transport buffer containing the radiolabeled and non-radiolabeled compounds. NTCP and SERT were used as a positive control, and empty-vector transfected cells served as the negative control. After the uptake phase, cells were washed with ice-cold PBS, lysed, and subjected to scintillation counting. Data represent mean ± SD of representative experiments each with quadruplicate determinations. *Significantly different from control with p < 0.01. the case that SLC10A4 has transport function for one of yet been performed. Therefore, in the present study, we the mentioned compounds anyway. In HEK293 cells and aimed to screen for SLC10A4 transport function with X. laevis oocytes, an additional factor might have been different transport assays, including transport in intact missed that is only present in neuronal cells. However, HEK293 cells and neuronal CAD cells, permeabilized overexpression of the SLC10A4 protein or permeabi- HEK293 cells and X. laevis oocytes. Overall, no trans- lization of the cells prior to transport experiments was port activity for SLC10A4 was found for taurocholic not practicable in neuronal cell lines as the cells either acid, lithocholic acid, DHEAS, PREGS, E-3-S, acetylcho- showed low transfection rates (as the SH-SY5Y cells) or line, choline, acetate, glutamate, aspartate, GABA, ATP, promptly detached when the incubation medium was serotonin, histamine, dopamine, and norepinephrine in repeatedly changed (CAD cells). Furthermore, all trans- the respective transport assay. Although we used differ - port assays performed in SLC10A4-transfected CAD ent kinds of transport assays and cell models, it might be Schmidt et al. BMC Neurosci (2015) 16:35 Page 9 of 14 Table 2 Transport studies in HEK293 cells Compound (5 µM) NTCP SLC10A4 NtNTCP‑SLC10A4‑ CtNTCP Uptake Ratio Uptake Ratio Uptake Ratio [ H]E-3-S 276.93 ± 29.12 10.1* 35.23 ± 2.01 1.3 27.91 ± 1.38 1.0 [ H]DHEAS 225.12 ± 16.51 6.9* 38.22 ± 0.90 1.2 31.84 ± 1.38 1.0 [ H]PREGS 1,656.09 ± 201.01 4.7* 366.06 ± 22.23 1.0 383.62 ± 17.38 1.0 DAT SLC10A4 NtNTCP‑SLC10A4‑ CtNTCP [ H]Dopamine 1,578.17 ± 107.47 13.2* 128.61 ± 16.60 1.1 157.92 ± 19.11 1.3 [ H]Glutamate 22.60 ± 2.95 1.4 13.57 ± 1.62 0.8 [ H]Histamine 26.24 ± 1.40 1.3 25.68 ± 0.63 1.3 [ H]Acetyl-choline 352.00 ± 38.76 1.0 307.64 ± 31.41 0.9 [ H]Choline 929.53 ± 30.50 1.1 866.83 ± 23.63 1.0 Values represent uptake in pmol/mg protein/30 min of the indicated compound as mean ± SD of quadruplicate determinations. Ratio represents uptake into carrier- expressing HEK293 cells divided by uptake into mock-transfected HEK293 cells. * Significantly different from control with p < 0.01. cells also failed to show transport activity for SLC10A4. Abe et  al. [18] that the N-terminal domain of SLC10A4 Nevertheless, further investigations on the SLC10A4 might hinder substrate binding to the protein. This sug - transport function should focus on vesicle preparations gestion can not be supported by data from the present from SLC10A4-deleted neuronal cell cultures or may use study. It is already known that thrombin starting from 1 liposomal reconstitution of the SLC10A4 protein. U/200  µl has a dose-dependent effect on cell viability of Very recently, it was suggested by Abe et  al. [18] that neuronal cell cultures [27]. Furthermore, it was reported SLC10A4 may represent a protease-activated trans- that lithocholic acid selectively kills neuroblastoma cells porter for the bile acids taurocholic acid and lithocholic by triggering the extrinsic and intrinsic apoptotic death acid. They used the human medulloblastoma cell line pathways via binding to the surface [28]. Therefore, it TE671 in which they localized the SLC10A4 protein by cannot be excluded that the suggested uptake of tauro- immunofluorescence and Western Blot analysis with a cholic acid and lithocholic acid may just have resulted commercial anti-SLC10A4 antibody and showed that from the combined surface effects of the bile acids and pre-incubation of the cells with 1U/200  µl of thrombin thrombin, but not from carrier-mediated uptake into the for 3  h increased the cellular uptake of taurocholic acid cells via a thrombin-modified SLC10A4 variant. and lithocholic acid [18]. However, appropriate controls During the completion of this study, the Kullander demonstrating specificity of the antibody were not pro - group published a series of very elegant experiments on vided and the authors failed to seriously show saturable Slc10a4 knockout mice, which provide further intrigu- transport kinetics for this uptake. Furthermore, bioin- ing indications of the molecular transport function of formatics analysis (Expasy Peptide Cutter, http://www. SLC10A4. In the first study by Zelano et  al. [16], they web.expasy.org/peptide_cutter/) indicates that the mouse analyzed whether the absence of SLC10A4 may have an and rat SLC10A4 proteins can be cleaved by thrombin impact on the function of the central cholinergic system. at amino acid position 87, but not the human SLC10A4 Injection of the cholinergic agonist pilocarpin induced protein. Nevertheless, in the present study, these experi- status epilepticus earlier and more often in the Slc10a4 ments were repeated by using HEK293 cells overexpress- knockout mice compared with the wild type mice, sug- ing SLC10A4, providing a better controllable system, but gesting that SLC10A4 may suppress epileptiform activ- did not show any transport activity of SLC10A4 for tau- ity. They concluded that absence of the SLC10A4 protein rocholic acid after treatment with thrombin. Due to these results in cholinergic hypersensitivity of the knockout negative data and based on the bioinformatics prediction mice [16]. In a second study by Patra et al. [17], the role we did not further analyze whether the human SLC10A4 of SLC10A4 on the structure and function of the neuro- protein can really be cleaved by thrombin in our experi- muscular junction was analyzed. Although there were no mental setup. Furthermore, when the large N-terminus abnormalities detectable at the macrostructure level, the of SLC10A4 was deleted by mutagenesis in the NtNTCP- Slc10a4 knockout mice showed misshapen neuro-mus- SLC10A4-CtNTCP and 75ΔSLC10A4 constructs, no cular junctions with an increased number of isolated ace- transport activity for taurocholic acid was detected. tylcholine receptor clusters and a decreased number of This is of particular interest, because it was suggested by endplate branches. Nevertheless, the knockout mice had Schmidt et al. BMC Neurosci (2015) 16:35 Page 10 of 14 normal motor behavior. Electrophysiological measure- The outstanding role of SLC10A4 within the SLC10 ments on nerve-muscle preparations then showed that family is not only based on its inability to transport bile the Slc10a4 knockout mice had decreased spontaneous acids or steroid sulfates. In contrast to NTCP, ASBT and endplate potential amplitudes, which could be explained SOAT, which all are sorted to the plasma membrane, by lower acetylcholine release at the endplates without SLC10A4 has a typical intracellular vesicle-like expres- nerve stimulation. On the other hand, after repeated sion pattern in neurons, mast cells and neuronal cell stimulation, the Slc10a4 knockout mice revealed an lines. This specific expression pattern of SLC10A4 was enlarged pool of readily releasable vesicles at the neuro- also observed in the present study in untreated and dif- muscular junction. However, for the interpretation of ferentiated CAD and SH-SY5Y cells. To more closely these data, it has to be considered that the knockout mice analyze the sorting domains of SLC10A4, a series of may have altered gene-expression in order to compensate NTCP/SLC10A4 chimeras were generated and analyzed for the loss of SLC10A4, as shown by an up-regulation in HEK293 and CAD cells. These experiments clearly of the nicotinic acetylcholine receptor subunits alpha1 showed that the cytoplasmic C-terminus of SLC10A4 and delta in the Slc10a4 knockout mice. The authors must contain dominant signals for its vesicular sorting. came to the conclusion that the loss of SLC10A4 may Previous studies with rat NTCP demonstrated that the result in reduced vesicular filling with the neurotrans - tyrosine-based sorting motifs YXXØ within its C-ter- 307 321 mitter acetylcholine at the neuromuscular junction [17]. minus (i.e. Y -E-K-I and Y -K-A-A) are involved in In a further study by Larhammar et  al. [15], the Slc10a4 membrane delivery [29]. Although the C-terminus of knockout mice were not different from their wild-type SLC10A4 also contains these potential tyrosine-based 406 419 littermates in a series of behavior tests, but showed a sorting motifs (Y -K–K-L und Y -G-T-V), SLC10A4 is hypoactive phenotype. Furthermore, the Slc10a4 knock- obviously not directed to the plasma membrane. In the out mice were hypersensitive to the psychostimulants same way, all chimeric constructs with the C-terminus amphetamine and tranylcypromine. In the central nerv- of SLC10A4 (i.e. NTCP-CtSLC10A4 and NtSLC10A4- ous system, the Slc10a4 knockout mice had reduced lev- NTCP-CtSLC10A4) showed a vesicular expression pat- els of dopamine, serotonin, norepinephrine and choline. tern and failed to transport typical NTCP substrates such Although no evidence for the direct transport of these as taurocholic acid and DHEAS. In contrast, both chi- neurotransmitters by SLC10A4 has been provided, syn- meras bearing the C-terminus of NTCP (i.e. SLC10A4- aptic vesicles from transgenic mice overexpressing the CtNTCP and NtNTCP-SLC10A4-CtNCTP) were SLC10A4 protein showed an increased uptake of [ H] detected at the plasma membrane. Further mutagenesis dopamine, which was probably due to higher synaptic experiments now have to localize the vesicular sorting vesicle acidification. Based on these data, the authors motifs of SLC10A4 at the amino acid level. speculate that SLC10A4 may transport any hitherto unknown organic anion compound, which would allow Conclusions the vesicular accumulation of higher amounts of protons, SLC10A4 is expressed in vesicular structures not only which then would increase the vesicular storage of neu- in neurons of the central and peripheral nervous sys- rotransmitters [15]. This conclusion is in agreement with tem, but also in neuronal cell lines such as SH-SY5Y and data from the present study, in which SLC10A4 showed CAD. Although different kinds of assays were applied to no transport activity for a series of neurotransmitters. screen for a transport function, SLC10A4 failed to show However, even suggested organic anion modulators such transport activity for dopamine, serotonin, norepineph- as ATP or DHEAS were not transported by SLC10A4, rine, histamine, acetylcholine, choline, acetate, aspar- meaning that the function of SLC10A4 as solute carrier is tate, glutamate, GABA, PREGS, DHEAS, E-3-S, and ATP, still a matter of speculation. indicating that SLC10A4 does not seem to be a typical Based on their findings, the Kullander group suggested neurotransmitter transporter. When the C-terminus of renaming SLC10A4 as “vesicular aminergic-associated SLC10A4 was replaced by the homologous sequence of transporter” (VAAT). However, as the naming of the NTCP, the SLC10A4-CtNTCP chimera revealed clear carriers of the SLC10 carrier family until now has only plasma membrane expression in CAD and HEK293 cells. been based on their molecular transport function, i.e. Vice versa, the C-terminus of SLC10A4 directed NTCP Na /taurocholate co-transporting polypeptide NTCP to intracellular vesicles, indicating that the sorting motifs (SLC10A1), apical sodium-dependent bile acid trans- of both carriers seem to be localized in the C-terminus. porter ASBT (SLC10A2), and sodium-dependent organic Further SLC10A4 transport studies should involve vesi- anion transporter SOAT (SLC10A6), we strongly suggest cle preparations from SLC10A4-deleted neuronal cell not renaming SLC10A4 before its molecular transport cultures or may use liposomal reconstitution of the function has been elucidated. SLC10A4 protein. Until then, the functional properties Schmidt et al. BMC Neurosci (2015) 16:35 Page 11 of 14 of the SLC10A4 orphan carrier protein still remain per cm in 24-well plates on poly-l-lysine-coated glass unknown. coverslips and cultured to a confluence of approxi - mately 40%. For immunofluorescence experiments, Methods the cells were washed with phosphate buffered saline Materials, chemicals, and radiochemicals (PBS, containing 137  mM NaCl, 2.7  mM KCl, 1.5  mM All of the chemicals, unless otherwise stated, were from KH PO , and 7.3  mM Na HPO , at pH 7.4) and fixed 2 4 2 4 Sigma-Aldrich (Taufkirchen, Germany). Citalopram was with 2% paraformaldehyde (PFA, Roth, Karlsruhe, Ger- purchased from Biotrend (Cologne, Germany) and col- many) in PBS for 15  min at 4°C. Then, cells were again lagenase D was from Serva (Heidelberg, Germany). [ H] washed with PBS and incubated with 20  mM glycine in DHEAS (70.5  Ci/mmol), [ H]E-3-S (45.6  Ci/mmol), PBS for 5  min. For permeabilization, cells were incu- 3 3 [ H]aspartate (11.3  Ci/mmol), [ H]GABA (76.2  Ci/ bated for 5 min with PBT buffer (0.2% Triton X-100 and 3 3 mmol), [ H]histamine (13.4  Ci/mmol), [ H]choline 20  mM Glycine in PBS). The non-specific binding sites chloride (66.7  Ci/mmol), [ H]norepinephrine (56.6  Ci/ were blocked with 1% bovine serum albumin (BSA) plus 3 3 mmol), [ H]serotonin (28.25  Ci/mmol), [ H]dopamine 4% goat serum (Sigma-Aldrich) in PBS for 30  min at 3 3 (38.7 Ci/mmol), [ H]glutamate (49.6 Ci/mmol), [ H]ATP room temperature. Then, the cells were incubated with (30.9  Ci/mmol), [ S]Adenosine 5′-(γ-thio) triphosphate the primary antibodies, rabbit anti-Slc10a4 (antibody (12.5  mCi/mmol) and [ H]acetylcholine iodide (99.7  Ci/ 1338 C, 1:1,000 dilution, see [13, 14] ), Sigma Prestige mmol) were purchased from PerkinElmer Life Sciences rabbit anti-SLC10A4 (1:500 dilution) [Sigma-Aldrich 3 3 (Boston, MA, USA). [ H]PREGS (20  Ci/mmol), [ H] Cat# HPA028835 RRID:AB_10603025], Abnova rab- taurocholic acid (10.0  Ci/mmol), [ H]acetate (150  mCi/ bit anti-Slc10a4 (1:500 dilution) [Abnova Corporation mmol), and [ H]lithocholic acid (50  Ci/mmol) were Cat# PAB14855 RRID:AB_10696081], or Abgent rabbit obtained from American Radiolabeled Chemicals (St. anti-SLC10A4 (1:500 dilution) [Abgent Cat# AP10250b Louis, MO, USA). RRID:AB_10821224], or with mouse anti-V5 monoclo- nal antibody (1:5,000 dilution) [Invitrogen Cat# R96025 Culture and differentiation of neuronal cell lines RRID:AB_159313] in blocking solution overnight at 4°C. Human neuroblastoma SH-SY5Y cells (obtained from The next day, cells were washed with PBS and incubated DSMZ Braunschweig, Germany) were maintained at with the fluorophore-labeled secondary antibody Cy3- 37°C in RPMI medium (Gibco, Karlsruhe, Germany) con- conjugated goat anti-rabbit IgG (1:800 dilution) [Jackson taining 10% fetal calf serum (FCS, Sigma-Aldrich) and 1% ImmunoResearch Cat# 111-165-003 RRID:AB_2338000] penicillin/streptomycin (P/S, containing 100 U/ml peni- or Alexa fluor 488-labelled goat-anti mouse (1:800 dilu - cillin and 100  mg/ml streptomycin). For differentiation, tion) [Molecular Probes (Invitrogen) Cat# A11001 the cultures at an approximate confluence of 40% were RRID:AB_141367] in blocking solution for 60  min at placed in serum-free RPMI medium supplemented with room temperature. After several washing steps with PBS, either 10 µM all-trans RA (Sigma-Aldrich) plus 10 ng/ml the cells were covered with a DAPI/methanol solution BMP-2 (BioCat, Heidelberg, Germany) or 10 ng/ml TGF- containing 1 µg/ml DAPI (Roche, Mannheim, Germany) β1 (Invitrogen, Karlsruhe, Germany) and were grown and incubated for 5  min at room temperature. The cells for 5  days. As a control, cells were grown in serum-free were rinsed with methanol, air dried and mounted onto RPMI medium without any supplements. Mouse cat- slides with ProLong Gold Antifade (Invitrogen) mount- echolaminergic CAD cells (obtained from European ing medium. Collection of Cell Cultures ECACC, Health Protection Agency, UK) were maintained at 37°C in Dulbecco’s Real‑time quantitative PCR analysis modified Eagle’s/Ham’s F-12 (1:1) medium (DMEM/F12) Relative expression analysis for SLC10A4/Slc10a4 was containing 8% FCS, 4  mM  l-glutamine and 1% P/S at performed with ABI PRISM 7300 technology (Applied 37°C. For cell differentiation, 4  ×  10 cells were seeded Biosystems, Darmstadt, Germany). RNA was isolated in 6-well or 24-well plates on poly-l-lysine-coated glass from human SH-SY5Y and mouse CAD cells with TriRea- coverslips and grown in serum-free medium for 24-72 h. gent (Sigma-Aldrich) and cDNA was reverse-transcribed Under these conditions, the cultures stopped proliferat- using the SuperScript III First Strand Synthesis Sys- ing and large neurite-like cell extensions appeared. tem (Invitrogen). PCR amplification was achieved with the TaqMan Gene Expression Assays Hs00293728 for Immunofluorescence analysis of SH‑SY5Y, CAD, human SLC10A4 and Mm00557788 for mouse Slc10a4 and HEK293 cells (Applied Biosystems). The expression data of human For immunofluorescence analysis, SH-SY5Y, CAD, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH, HEK293 cells were seeded at a density of 5  ×  10 cells Hs99999905) and mouse beta-actin (Mm00607939) were Schmidt et al. BMC Neurosci (2015) 16:35 Page 12 of 14 used as an endogenous control. Triplicate determina- with the primers listed in Table  3. All amplicons were tions were performed in a 96-well optical plate for each sequence-verified by DNA sequencing according to target, using 5  µl cDNA, 1.25  µl TaqMan Gene Expres- the GenBank Accession Nos. given in Table  3 and were sion Assay, 12.5  µl TaqMan Universal PCR Master Mix cloned into the pcDNA5/FRT/V5-His TOPO vector (Life (Applied Biosystems) and 6.25  µl water in each 25  µl Technologies) via T/A cloning. reaction. The plates were heated for 10 min at 95°C, and 40 cycles of 15  s at 95°C and 60  s at 60°C were applied. Generation of the SLC10A4/NTCP chimeras The relative expression (ΔC ) of each target was calcu- For generation of the SLC10A4/NTCP chimeric con- lated by subtracting the signal threshold cycle (C ) of the structs, the full-length SLC10A4-cDNA5 and NTCP- endogenous control from the C value of the target. pcDNA5 clones with C-terminal V5-tag were used. BmtI and HindIII restriction sites were introduced in the N-ter- Western blot analysis mini (amino acid positions A6 in the NTCP and A78 in Protein contents of CAD cell lysates were determined the SLC10A4) and C-termini (amino acid positions E296 with the BCA Protein Assay Kit (Novagen, Darmstadt, in the NTCP and E374 in the SLC10A4), respectively, Germany). Samples of 20  µg protein were mixed with by site-directed mutagenesis, as previously described in Laemmli Sample Buffer (Sigma-Aldrich), separated detail [8]. Then, by double digestion and re-ligation of the on a 12% SDS polyacrylamide gel and transferred to a appropriate fragments, the following chimeric constructs Hybond-ECL nitrocellulose membrane (Amersham Bio- were generated and sequence-verified by direct DNA sciences, Freiburg, Germany). The blotted membranes sequencing: SLC10A4-CtNTCP, NtNTCP-SLC10A4- were blocked with blocking solution containing 5% CtNTCP, NtSLC10A4-NTCP, NtSLC10A4-NTCP- ECL-blocking agent (Amersham Biosciences) in TBS-T CtSLC10A4, and NTCP-CtSLC10A4. For generation of (137  mM NaCl, 10  mM Tris, pH 8.0, 0.05% Tween-20), the N-terminally truncated 75ΔSLC10A4 construct, the followed by overnight exposure to antigen-specific pri - SLC10A4 open reading frame starting from amino acid mary antibodies at 4°C in the same buffer. After several 75 was PCR-amplified and cloned, whereby the codon for washing steps in TBS-T, the membranes were probed Gly75 was replaced by an artificial start codon ATG. with the appropriate horseradish-peroxidase-labeled secondary antibodies in TBS-T for 60 min at room tem- Transfection of HEK293 and CAD cells perature. Signals were developed using the Roti-Lumin HEK293 cells were grown in DMEM supplemented with ECL Detection Kit (Roth) and visualized by exposure to 10% FCS, 4  mM glutamine and 1% P/S. HEK293 and Hyperfilm ECL (Amersham Biosciences). CAD cells were grown at 37°C in 5% CO . For transient transfection, both cell lines were seeded in 24-well plates Cloning of the reference carriers at a density of 1.4–2.0 ×  10 cells per well and transfec- Full length transcripts covering the whole open reading tion was performed by using the Lipofectamine 2000 frame for human SLC10A4, DAT, CHT1, SERT, VMAT2, reagent according to the manufacturer’s instructions VAChT, NTCP, and mouse Oct1 were amplified by PCR (Invitrogen). For the establishment of stably transfected Table 3 Primers used for full-length carrier cloning Carrier Forward primer Reverse primer GenBank Accession Nos. Human SLC10A1 (Na /taurocholate co- 5′-tct cta gag gat gga ggc cca caa c-3′ 5′-ggc tgt gca agg gga gca gtc-3′ [GenBank:NM_003049] transporting polypeptide, NTCP) Human SLC10A4 [GenBank:NM_152679] 5′-acc gac ggg cag aac gac-3′ 5′-gag aga agt ctg agc ggt ttc-3′ Human SLC6A3 (dopamine transporter, 5′-ctc cca gtg tgc cca tga gta aga g-3′ 5′-cac ctt gag cca gtg gcg gag-3′ [GenBank:NM_001044] DAT ) Human SLC5A7 (choline transporter, 5′-aaa aat ggc ttt cca tgt gga agg-3′ 5′-ctg taa att atc ttc agt ccc ag-3′ [GenBank:NM_021815] CHT1) Human SLC6A4 (serotonin transporter, 5′-agg atg gag acg acg ccc ttg aat tc-3′ 5′-cac agc att caa gcg gat gtc ccc a-3′ [GenBank:NM_001045] SERT ) Human SLC18A2 (vesicular monoamine 5′-gcc atg gcc ctg agc gag ctg-3′ 5′-gtc act ttc aga ttc ttc atc ttc acc tat [GenBank:NM_003054] transporter, VMAT2) c-3′ Human SLC18A3 (vesicular acetylcho- [GenBank:NM_003055] 5′-cgg aag agc atc ggg gtg-3′ 5′-gct gcg ggt gta gta gta g-3′ line transporter, VAChT ) Mouse Slc22a1 (organic cation trans- 5′-att tca agc cac cgc agt tc-3′ 5′-ggt atg tgg gga ttt gcc t-3′ [GenBank:NM_009202] porter, Oct1) Schmidt et al. BMC Neurosci (2015) 16:35 Page 13 of 14 cells, Flp-In HEK293 cells were used (Invitrogen), as cells were permeabilized with 15 µM digitonin in PB for described previously [3]. Briefly, Flp-In HEK293 cells 15 min at 37°C, as reported [23]. The buffer was replaced were seeded in 6-well plates coated with poly-d-lysine at by 250 µl of PB containing the radiolabeled and non-radi- a density of 1 ×  10 cells per well and grown to 60–80% olabeled compounds. After incubation for the indicated confluence in antibiotic free medium. Then, the cells time at 37°C, the uptake was stopped, as described above. were transfected with 1  µg of the respective pcDNA5/ FRT/V5-His vector construct plus 7 µg of pOG44 vector. Transport studies in Xenopus laevis oocytes For the selection of positive clones, hygromycin B was The pcDNA5/FRT/V5-His constructs containing the used at a concentration of 150 µg/ml. cloned cDNAs were linearized with KpnI. After phenol/ chloroform extraction, the mMESSAGEmMACHINE Transport studies in HEK293 and CAD cells Kit (Ambion, Life Technologies) was used to generate a For transport studies, 24-well plates were coated with capped cRNA and the Poly(A) Tailing Kit (Ambion, Life poly-d-lysine for better attachment of the cells. Stably Technologies) was used to add a poly(A) tail to the RNA or transiently transfected cells were plated at a density transcripts. Afterwards, the cRNA was purified with the of 1.25  ×  10 cells per well and grown under standard MEGAclear Kit (Ambion, Life Technologies) according medium for 48–72  h. Before starting the transport to the manufacturer’s protocol. Oocytes were obtained 2+ experiments, cells were washed three times with PBS and from female X. laevis frogs and incubated in Ca -free pre-incubated with sodium transport buffer (142.9  mM OR-2 solution (82.5 mM NaCl, 5 mM HEPES–NaOH, pH NaCl, 4.7  mM KCl, 1.2  mM MgSO , 1.2  mM KH PO , 7.6, 2.5 mM KCl, 1 mM MgCl and 1 mM Na HPO ) sup- 4 2 4 2 2 4 1.8  mM CaCl , and 20  mM HEPES, adjusted to pH 7.4). plemented with 0.4  mg/ml collagenase type D (Serva) at When transport assays were performed in sodium-free 18°C overnight. Then, oocytes were further incubated in transport buffer, sodium chloride was substituted with modified Barth’s solution (88 mM NaCl, 15 mM HEPES– equimolar concentrations of choline chloride and for the NaOH, pH 7.6, 2.4 mM NaHCO , 1 mM KCl, 0.3 mM Ca transport assay with the substrate [ H]choline chloride (NO ) , 0.41 mM CaCl and 0.82 mM MgSO ) containing 3 2 2 4 it was substituted with equimolar concentrations of lith- 50 µg/ml gentamicin. Defolliculated oocytes were selected ium chloride. For transport studies with [ H]serotonin, and microinjected with 4.6 ng (46 nl) cRNA encoding for the uptake buffer contained 100  µM ascorbic acid and the carrier protein or with a corresponding volume of 100  µM pargyline, to prevent the degradation of seroto- water. After 3 days of culture in modified Barth’s medium, nin. Transport studies were performed by incubating the uptake of the indicated radiolabelled substrates was cells with 250 µl transport buffer containing the radiola - assessed at 25°C in transport buffer containing 100  mM beled and non-radiolabeled compounds for the indicated NaCl, 2 mM KCl, 1 mM CaCl , 1 mM MgCl , and 10 mM 2 2 time at 37°C. CAD cells could not be washed before start- HEPES-Tris, pH 7.5. After washing with the same buffer, ing the transport experiments because these cells easily each individual oocyte was dissolved in 500  µl of 10% detached after repeated handling. Therefore, these cells SDS. The radioactivity was counted after the addition of were incubated in 250 µl fresh medium plus 50 µl trans- 4 ml scintillation fluid in a liquid scintillation counter. port buffer containing the radiolabeled and non-radi - olabeled compounds. Uptake studies were stopped by Statistics removing the transport buffer and washing the cells five Columns are shown as mean  ±  SD. Prism software times with ice-cold PBS. Afterwards, cells were lysed in (GraphPad Software Inc., San Diego, CA, USA) was used 1 N NaOH with 0.1% SDS and the cell-associated radio- for data presentation and statistical analysis. Statisti- activity was measured by liquid scintillation counting. cal significance of two groups was analyzed by Student’s The protein content was determined using aliquots of the t test. Statistical analysis of more than two groups was lysed cells with BSA as the standard [30]. performed by one-way analysis of variance (ANOVA) fol- lowed by Bonferroni post hoc testing. Transport studies in permeabilized HEK293 cells Stably transfected HEK293 cells expressing SLC10A4, Abbreviations VMAT2 or VAChT were seeded in 24-well plates and ASBT: Apical sodium-dependent bile acid transporter; ATP: adenosine triphos- were grown until confluence. Then, cells were rinsed phate; BMP-2: bone morphogenetic protein 2; BSA: bovine serum albumin; CHT1: choline transporter; DAT: dopamine transporter; DHEAS: dehydroepi- three times with potassium-rich buffer (PB) contain - androsterone sulfate; E-3-S: estrone-3-sulfate; FCCP: carbonylcyanid-p-trifluo - ing 110  mM potassium tartrate, 5  mM glucose, 0.2% romethoxyphenylhydrazon; FCS: fetal calf serum; GABA: gamma-aminobutyric bovine serum albumin, 200  µM CaCl 1  mM ascorbic acid; HC-3: hemicholinium-3; HEK293: human embryonic kidneys 293 cells; 2, NTCP: Na /taurocholate co-transporting polypeptide; Oct1: organic cation acid, 10 µM pargyline and 20 mM piperazine-N,N′-bis(2- transporter 1; PB: potassium-rich buffer; PBS: phosphate buffered saline; PFA: ethanesulfonic acid) (PIPES), adjusted to pH 6.8. The paraformaldehyde; PREGS: pregnenolone sulfate; RA: retinoic acid; RFP: red Schmidt et al. BMC Neurosci (2015) 16:35 Page 14 of 14 fluorescent protein; SERT: serotonin transporter; SLC10: solute carrier family 10. Bijsmans ITGW, Bouwmeester RAM, Geyer J, Faber KN, van de Graaf 10; SOAT: sodium-dependent organic anion transporter; TGF-β: tumor growth SFJ (2012) Homo- and hetero-dimeric architecture of the human liver factor beta; VAChT: vesicular acetylcholine transporter; VMAT2: vesicular mono- Na -dependent taurocholate co-transporting protein. Biochem J amine transporter. 441(3):1007–1015 11. Jiang L, Alber J, Wang J, Du W, Yang X, Li X et al (2012) The Candida Author’s contributions albicans plasma membrane protein Rch1p, a member of the vertebrate 2+ SS carried out cloning of the carriers SLC10A4, DAT, SERT, CHT1, VMAT2, and SLC10 carrier family, is a novel regulator of cytosolic Ca homoeostasis. VAChT, performed the transport experiments in HEK293 cells and Xenopus Biochem J 444(3):497–502 laevis oocytes, was involved in drafting of the manuscript, and performed data 12. Alber J, Jiang L, Geyer J (2013) CaRch1p does not functionally interact 2+ presentation and statistical analysis. MM generated the SLC10A4/NTCP chi- with the high-affinity Ca influx system (HACS) of Candida albicans. meric constructs, and analyzed them by immunofluorescence and transport Yeast 30(11):449–457 studies. SB carried out immunofluorescence, Western blot, and qPCR analysis 13. Geyer J, Fernandes CF, Döring B, Burger S, Godoy JR, Rafalzik S et al (2008) on the neuronal cell lines, and was involved in drafting of the manuscript. JG Cloning and molecular characterization of the orphan carrier protein conceived the study, performed data interpretation, and drafted and finalized Slc10a4: expression in cholinergic neurons of the rat central nervous the manuscript. All authors read and approved the final manuscript. system. Neuroscience 152:990–1005 14. Burger S, Döring B, Hardt M, Beuerlein K, Gerstberger R, Geyer J (2011) Co- expression studies of the orphan carrier protein Slc10a4 and the vesicular Acknowledgements carriers VAChT and VMAT2 in the rat central and peripheral nervous We thank Anita Neubauer, Klaus Schuh, and Regina Leidolf for their excellent system. Neuroscience 193:109–121 technical support and would like to acknowledge Dr. Barbara Döring and Dina 15. Larhammar M, Patra K, Blunder M, Emilsson L, Peuckert C, Arvidsson E Kleinlützum for providing clones of NTCP and Oct1, respectively. This research et al (2014) SLC10A4 Is a vesicular amine-associated transporter modulat- was supported in part by the German Research Foundation DFG (Grant GE ing dopamine homeostasis. Biol Psychiatr 77:526–536 1921/3-1) and in part by Doktor Robert Pfleger-Stiftung (Bamberg, Germany). 16. Zelano J, Mikulovic S, Patra K, Kühnemund M, Larhammar M, Emils- son L et al (2013) The synaptic protein encoded by the gene Slc10A4 Compliance with ethical guidelines suppresses epileptiform activity and regulates sensitivity to cholinergic chemoconvulsants. Exp Neurol 239:73–81 Competing interests 17. Patra K, Lyons DJ, Bauer P, Hilscher MM, Sharma S, Leão RN et al (2014) A The authors declare that they have no competing interests. role for solute carrier family 10 member 4, or vesicular aminergic-associ- ated transporter, in structural remodelling and transmitter release at the Received: 28 January 2015 Accepted: 5 June 2015 mouse neuromuscular junction. Eur J Neurosci 41:316–327 18. Abe T, Kanemitu Y, Nakasone M, Kawahata I, Yamakuni T, Nakajima A et al (2013) SLC10A4 is a protease-activated transporter that transports bile acids. J Biochem 154(1):93–101 19. Gómez-Santos C, Ambrosio S, Ventura F, Ferrer I, Reiriz J (2002) TGF-β1 increases tyrosine hydroxylase expression by a mechanism blocked by References BMP-2 in human neuroblastoma SH-SY5Y cells. Brain Res 958(1):152–160 1. Geyer J, Wilke T, Petzinger E (2006) The solute carrier family SLC10: 20. Kohr D, Tschernatsch M, Schmitz K, Singh P, Kaps M, Schäfer KH et al more than a family of bile acid transporters regarding function and (2009) Autoantibodies in complex regional pain syndrome bind phylogenetic relationships. Naunyn Schmiedebergs Arch Pharmacol to a differentiation-dependent neuronal surface autoantigen. Pain 372(6):413–431 143(3):246–251 2. Döring B, Lütteke T, Geyer J, Petzinger E (2012) The SLC10 carrier family: 21. Qi Y, Wang JK, McMillian M, Chikaraishi DM (1997) Characterization of a transport functions and molecular structure. Curr Top Membr 70:105–168 CNS cell line, CAD, in which morphological differentiation is initiated by 3. Geyer J, Döring B, Meerkamp K, Ugele B, Bakhiya N, Fernandes serum deprivation. J Neurosci 17(4):1217–1225 CF et al (2007) Cloning and functional characterization of human 22. Geyer J, Godoy JR, Petzinger E (2004) Identification of a sodium-depend- sodium-dependent organic anion transporter (SLC10A6). J Biol Chem ent organic anion transporter from rat adrenal gland. Biochem Biophys 282(27):19728–19741 Res Commun. 316(2):300–306 4. Fietz D, Bakhaus K, Wapelhorst B, Grosser G, Günther S, Alber J et al (2013) 23. Erickson JD, Eiden LE, Hoffman BJ (1992) Expression cloning of a Membrane transporters for sulfated steroids in the human testis-cellular reserpine-sensitive vesicular monoamine transporter. Proc Natl Acad Sci localization, expression pattern and functional analysis. PLoS One USA 89(22):10993–10997 8(5):e62638 24. Rupprecht R (1997) The neuropsychopharmacological potential of neuro- 5. Grosser G, Fietz D, Günther S, Bakhaus K, Schweigmann H, Ugele B et al active steroids. J Psychiatr Res 31(3):297–314 (2013) Cloning and functional characterization of the mouse sodium- 25. Reddy DS (2003) Pharmacology of endogenous neuroactive steroids. Crit dependent organic anion transporter Soat (Slc10a6). J Steroid Biochem Rev Neurobiol 15(3–4):197–234 Mol Biol 138:90–99 26. Zheng P (2009) Neuroactive steroid regulation of neurotransmitter 6. Splinter PL, Lazaridis KN, Dawson PA, LaRusso NF (2006) Cloning and release in the CNS: action, mechanism and possible significance. Prog expression of SLC10A4, a putative organic anion transport protein. World Neurobiol 89(2):134–152 J Gastroenterol 12(42):6797–6805 27. Jiang Y, Wu J, Hua Y, Keep RF, Xiang J, Hoff JT et al (2002) Thrombin-recep - 7. Fernandes CF, Godoy JR, Döring B, Cavalcanti MC, Bergmann M, Petz- tor activation and thrombin-induced brain tolerance. J Cereb Blood Flow inger E et al (2007) The novel putative bile acid transporter SLC10A5 is Metab 22(4):404–410 highly expressed in liver and kidney. Biochem Biophys Res Commun 28. Goldberg AA, Beach A, Davies GF, Harkness TAA, Leblanc A, Titorenko VI 361(1):26–32 (2011) Lithocholic bile acid selectively kills neuroblastoma cells, while 8. Godoy JR, Fernandes C, Döring B, Beuerlein K, Petzinger E, Geyer J (2007) sparing normal neuronal cells. Oncotarget 2(10):761–782 Molecular and phylogenetic characterization of a novel putative mem- 29. Sun AQ, Arrese MA, Zeng L, Swaby I, Zhou MM, Suchy FJ (2001) The rat brane transporter (SLC10A7), conserved in vertebrates and bacteria. Eur J liver Na /bile acid cotransporter. Importance of the cytoplasmic tail to Cell Biol 86(8):445–460 function and plasma membrane targeting. J Biol Chem 276(9):6825–6833 9. Visser WE, Wong WS, van Mullem AAA, Friesema ECH, Geyer J, Visser TJ 30. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measure- (2010) Study of the transport of thyroid hormone by transporters of the ment with the Folin phenol reagent. J Biol Chem 193(1):265–275 SLC10 family. Mol Cell Endocrinol 315(1–2):138–145

Journal

BMC NeuroscienceSpringer Journals

Published: Jun 19, 2015

There are no references for this article.