Characterization of ST14A Cells for Studying Modulation of Voltage-Gated Calcium Channels

In medium spiny neurons (MSNs) of the striatum, dopamine D2 receptors (D2Rs) specifically inhibit the Cav1.3 subtype of L-type Ca2+ channels (LTCs). MSNs are heterogeneous in their expression of dopamine receptors making the study of D2R pathways difficult in primary neurons. Here, we employed the ST14A cell line, derived from embryonic striatum and characterized to have properties of MSNs, to study Cav1.3 current and its modulation by neurotransmitters. Round, undifferentiated ST14A cells exhibited little to no endogenous Ca2+ current while differentiated ST14A cells expressed endogenous Ca2+ current. Transfection with LTC subunits produced functional Cav1.3 current from round cells, providing a homogeneous model system compared to native MSNs for studying D2R pathways. However, neither endogenous nor recombinant Cav1.3 current was modulated by the D2R agonist quinpirole. We confirmed D2R expression in ST14A cells and also detected D1Rs, D4Rs, D5Rs, Gq, calcineurin and phospholipase A2 using RT-PCR and/or Western blot analysis. Phospholipase C β-1 (PLCβ-1) expression was not detected by Western blot analysis which may account for the lack of LTC modulation by D2Rs. These findings raise caution about the assumption that the presence of G-protein coupled receptors in cell lines indicates the presence of complete signaling cascades. However, exogenous arachidonic acid inhibited recombinant Cav1.3 current indicating that channels expressed in ST14A cells are capable of modulation since they respond to a known signaling molecule downstream of D2Rs. Thus, ST14A cells provide a MSN-like cell line for studying channel modulation and signaling pathways that do not involve activation of PLCβ-1.


Introduction
Two classes of L-type Ca 2+ channel (LTC) α 1 subunits are expressed in the brain: α 1C (Ca V 1.2) and α 1D (Ca V 1.3) [1] with highest expression in cerebral cortex and striatum [2]. While differing in biophysical properties and pharmacological sensitivities, both LTCs contribute to permissive temperature of 33°C in Dulbecco's modified Eagle's medium (DMEM; Gibco) supplemented with 0.11 g/L sodium bicarbonate (Sigma, St. Louis, MO), 0.29 g/L L-glutamine, 3.9 g/liter HEPES, 100 units/ml penicillin-streptomycin, and 10% fetal bovine serum (FBS; Invitrogen, Carlsbac, CA). Cells were transferred to 37°C to promote differentiation and used 1-2 days later. The A9L cell line, derived from A9 L cells co-transfected with the human D 2 R and obtained from the American Type Culture Collection (ATCC, Manassas, VA), were propagated at 37°C in DMEM with 4 mM L-glutamine adjusted to contain 1.5 g/L sodium bicarbonate and 4.5 g/L glucose containing 10% FBS. HEK 293 cells were propagated at 37°C in DMEM/F12 containing 10% FBS. All cells were maintained in a temperature-controlled humidified incubator at 5% CO 2 and passaged once flasks became 80-90% confluent. SCG, striatum and cortex of 1 to 4-day old or adult Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) were isolated following CO 2 exposure and decapitation using a protocol (protocol # 822) approved by the Institutional Animal Care and Use Committee (IACUC) of University of Massachusetts Medical School. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All efforts were made to minimize animal suffering. The Institutional Animal Care and Use Committee (IACUC) of University of Massachusetts Medical School specifically approved animal use for this study.

Electrophysiology
Whole-cell currents were recorded at room temperature (RT, 20-24°C) with an Axon 200B patch clamp amplifier (Molecular Devices, Sunnyvale, CA). Currents were filtered at 1-5 kHz and digitized at 5 times the filter cut-off frequency of the 4-pole Bessel filter of the amplifier. Electrodes were pulled from borosilicate glass capillary tubes and each electrode was fire-polished to~1 m to give the pipette a resistance of 2-3 M. The pipette solution consisted of (in mM): 125 Cs-Aspartate, 10 HEPES, 0.1 BAPTA, 5 MgCl 2 , 4 ATP and 0.4 GTP brought to pH 7.50 with CsOH. High resistance seals were established in Ca 2+ Tyrode's consisting of (in mM): 5 CaCl 2 , 145 NaCl, 5.4 KCl, and 10 HEPES brought to pH 7.50 with NaOH. Once a seal was established and the membrane ruptured, the Tyrode's solution was exchanged for an external bath solution consisting of (in mM): 125 NMG-Aspartate, 20 Ba 2+ , 10 HEPES brought to pH 7.50 with CsOH.
FPL 64176 and nimodipine were prepared as stock solutions in 100% ethanol and stored at -20°C. AA (5,8,11,14-eicosatetraenoic acid; NuCheck Prep, Elysian, MN) and oleic acid (NuCheck Prep) were dissolved in 100% ethanol and stored under nitrogen as stock solutions at -70°C. ω-conotoxin GVIA (Bachem, Torrence, CA) and quinpirole were prepared as stock solutions in double distilled water and stored at -70°C. Oxotremorine-M (Tocris, Ellisville, MO) was prepared fresh daily by making a 10 mM stock in double distilled water. Working dilutions were made fresh daily by diluting stock solutions at least 1:1,000 with external bath solution. For ethanol prepared stocks, the final ethanol concentration was less than 0.1%. Bovine serum albumin (BSA; fraction V, heat shock, fatty acid ultra-free; Roche Applied Science, Indianapolis, IN) was added directly to the bath solution for a final concentration of 1 mg/ml. All chemicals were purchased from Sigma unless otherwise noted.
Data were acquired using Signal 2.14 software (Cambridge Electronic Design, Cambridge, England) and stored for later analysis on a personal computer. Linear leak and capacitive currents were subtracted from all traces. Data are presented as the mean ± s.e.m. Significance was determined statistically using a two-tailed paired or unpaired t-test, or a one-way ANOVA. Analysis programs include Signal (Cambridge Electronic Design), Excel (Microsoft, Redmond, WA) and Origin (OriginLab, Northampton, MA). Transfection ST14A cells were transfected by lipofectamine (Invitrogen, Carlsbad, CA) with a 1:1:1 molar ratio of Ca V 1.3, β 2a , and α 2 δ subunits [30]. Constructs for Ca V 1.3b (+exon11, Δexon32, +-exon42a; GenBank accession #AF370009), and α 2 δ-1 (GenBank accession #AF286488) were a gift from Dr. Diane Lipscombe (Brown University) and the construct for Ca V β 2a (GenBank accession #M80545) was a gift from Dr. Edward Perez-Reyes (University of Virginia). For all transfections, 0.4 μg of DNA was used per well of a 12-well plate. Prior to transfection, cells were washed with DMEM. The DNA mixture was then added dropwise to each well, gently swirled then incubated for 1-3 h at 37°C in a 5% CO 2 incubator. Supplemented media, without antibiotics, was returned to the wells to bring the volume up to 1 ml (normal growth medium volume). Cells were washed with full media 2 and 4 h later and assayed for transient gene expression after 24-72 h.

Reverse Transcriptase-Polymerase Chain Reaction
Homogenized tissue samples (50-100 mg) or confluent cells in a 100 mm dish were lysed in 1 ml TRIzol Reagent (Invitrogen). RNA was separated from DNA by phenol-chloroform phase separation. RNA was precipitated with isopropyl alcohol and washed with 75% ethanol. The RNA pellet was dried and resuspended in RNase-free water. RNA samples had an A 260 /A 280 ratio between 1.6 and 1.8 and were treated with DNase to eliminate contamination with genomic DNA. For reverse transcription, cDNA was synthesized from the mRNA by adding 1 μl 10X buffer RT, 1 μl dNTP Mix (5 mM each dNTP), 1 μl Oligo-dT primer (0.5 mg/ml, Promega, Madison, WI), 0.125 μl RNase Inhibitor (40 U/μl, Promega), 0.5 μl Omniscript Reverse Transcriptase (4 U/μl) and RNase-free water for a total volume of 10 μl (all reagents from QIAGEN, Valencia, CA, unless otherwise noted). The mixture was incubated at 37°C for 1 h. The mixture was then heated at 93°C for 5 min and then placed on ice to inactivate the transcriptase. PCR amplification was then performed with a Techgene thermal cycler (Techne Inc, Burlington, NJ) with thin walled PCR tubes.
PCR primers for dopamine receptors D 1 -D 5 (D 5 formerly referred to as D 1b ) were sequences previously published [31]. Qiagen for GAPDH), 2 to 4 mM Mg 2+ (Brinkmann) and distilled water to a final volume of 12.5 μl. High (4 mM) Mg 2+ concentrations were necessary to amplify dopamine receptor mRNAs. The protocol for dopamine receptor amplification was 94°C for 1 min, 58°C for 1 min, and 72°C for 3 min for 20 cycles [31]. For the D 3 R, an annealing temperature of 62°C was also tried. For GAD amplification, the protocol was 94°C for 3 min, 53°C for 1 min, 52°C for 1 min, 51°C for 1 min, and 70°C for 2 min for 30 cycles [32]. A 2 μl aliquot was used as a template for a second round of amplification for thirty cycles. PCR products and a 100 bp DNA ladder (Invitrogen; bright band 600 bp or Promega, Madison, WI; bright band 500 bp) were separated by electrophoresis in 2% agarose gels stained with ethidium bromide. Bands excised from gels were sequenced by the UMASS Medical School Nucleic Acid Facility and compared to published sequences (GenBank accession number): D 1 (M35077), D 2 (M36831), D 4 (M84009), D 5 (M69118), GAD65 (M74826), and GAD67 (M81883).

Results
To determine if ST14A cells express endogenous Ca 2+ current, we measured whole-cell currents from cells grown at 33 or 37°C. Two populations of cells were distinguished based on morphology: round or differentiated, with the definition of having neuron-like projections. From a holding potential of -90 mV, round cells exhibited zero to little endogenous peak or tail current measured at +10 mV or -40 mV respectively, regardless of the temperature at which the cells were grown (Fig 1A). Differentiated cells had significantly more endogenous current than round cells at both potentials ( Fig 1B). Application of the LTC agonist, FPL 64176 (FPL; 1 μM) enhanced endogenous currents at both potentials ( Fig 1A and 1B), indicating that at least a component of the endogenous current was LTC. Since round cells showed little to no LTC current, we transfected the ST14A cells with LTC channel subunits, Ca V 1.3, β 2a , and α 2 δ-1, along with green fluorescent protein (GFP) and recorded whole-cell currents from green fluorescing round cells (Fig 1C). Peak and tail currents from round transfected cells were significantly larger than those recorded from round untransfected cells (Fig 1A and 1B). FPL significantly enhanced recombinant current at -40 mV. Fig 1D shows representative individual traces of round endogenous or round recombinant current before and after application of FPL. The recombinant current shows little to no inactivation with 20 mM Ba 2+ as the charge carrier [35], typical of LTC current coexpressed with the accessory subunit, β 2a [36].
Since Ca V 1.3 activates at more negative voltages compared to Ca V 1.2 [30], we measured peak current at -10 mV. Recombinant current recorded at -10 mV was larger than at +10 mV Currents were recorded from round (undifferentiated, no processes) or differentiated (having processes) endogenous ST14A cells. Note: round cells from ST14A cells grown at either 33°C or 37°C exhibited little to no current and were pooled. Currents were recorded from round ST14A cells grown at 33°C, transfected and grown at 37°C for at least 24 hours before recording. Each set of cells were recorded in the absence (white bars) and presence of 1 μM FPL (black bars). Summary of peak Ca 2+ currents measured from a holding potential of -90 mV to a test potential of +10 mV (A) and then to a tail potential of -40 mV (B).
( Fig 1A) and this increase was also reflected in the tail current amplitude at -40 mV (Fig 1B).
The summary of these results shows that round ST14A cells have little endogenous LTC current but are capable of being transfected with LTCs and expressing functional LTC currents. Therefore, the round cells represent a population of ST14A cells that we used to study recombinant LTC function in isolation from other types of native Ca 2+ channels.
We characterized recombinant Ca V 1.3 current further to determine whether transfected channel activity in ST14A cells exhibits biophysical and pharmacological properties of Ca V 1.3 observed in oocytes and HEK 293 cells [30]. First, we measured peak current across a range of voltages to show that channels open at relatively negative voltages compared to Ca V 1.2 [30]. Indeed, in 20 mM Ba 2+ , Ca V 1.3 currents activated at a test potential of -60 mV and peaked at -10 mV to 0 mV (Fig 2A). Recombinant current in ST14A cells exhibited a Ca V 1.3 LTC pharmacological profile. Fig 2B shows a time course of the effects of Ca 2+ channel ligands on Ca V 1.3 current. The current was insensitive to the N-type Ca 2+ channel antagonist, ω--conotoxin GVIA (1 μM; CTX) but was inhibited in a concentration-dependent manner by the LTC antagonist, nimodipine (0.1-3.0 μM; NIM). The inhibition produced by 1.0 μM NIM (51.8 ± 4%) is characteristic of the low-voltage sensitivity of Ca V 1.3 compared to Ca V 1.2, which would be fully blocked by 1.0 μM NIM [30]. Inhibition fully reversed by washing with bath solution. After wash out, channels remained sensitive to FPL. The antagonist data are summarized in the bar graph in Fig 2C. These findings show that recombinant current in ST14A cells displayed both current-voltage and pharmacological profiles specific to Ca V 1.3 channels.
To determine whether D 2 R activation by the agonist quinpirole (Quin) inhibits Ca V 1.3 currents in ST14A cells, we recorded recombinant currents in the presence of FPL to enhance current amplitude. At a concentration of 10 μM, Quin had no significant effect on peak or tail current amplitude over time, (Fig 3A left) or in individual traces (Fig 3A right). To determine if only recombinant Ca V 1.3 current was insensitive to D 2 R activation, we tested whether endogenous ST14A current from differentiated cells could undergo modulation. Again 10 μM Quin had no significant effect on endogenous peak or long-lasting tail current amplitude. Fig 3B  shows representative current traces in the presence of 1 μM FPL from a range of voltages before and after Quin. Since a majority of MSNs express muscarinic M 1 receptors (M 1 Rs) as well as dopamine receptors [37], we tested whether activation of this receptor would inhibit endogenous current. The muscarinic agonist oxotremorine-M (Oxo-M; 10 μM) had no effect on endogenous peak or long-lasting tail current amplitude. Fig 3C shows representative current traces in the presence of FPL from a range of voltages (-60 mV to -10 mV) tested before and after Oxo-M. Fig 3D summarizes the effect of Quin and Oxo-M on peak (left) and long-lasting tail current (right) from recombinant Ca V 1.3 versus endogenous ST14A current. These results suggest that the D 2 R and M 1 R signaling pathways, which inhibit LTC current in MSNs, are not intact in ST14A cells. However, application of dopamine or Quin, increases CREB Endogenous current from differentiated cells was significantly larger than endogenous current from round cells ( †; p < 0.05); Transfected current was significantly larger than endogenous current from round cells ( ‡; p < 0.05); FPL significantly increased differentiated endogenous and transfected tail current (*; p < 0.05), n = 7-15. (C) Transfected ST14A cells expressed GFP throughout the cell soma and in a small percentage of differentiated cells, in the processes. Images (20X magnification) were captured~24 hours post-transfection. The transfection rate for these cells was~50%. (D) Top: Protocol for eliciting currents from a holding potential of -90 mV to the test potential of +10 mV for 100 ms before repolarizing to -40 ms. Individual traces from round endogenous (middle) or round transfected (bottom) cells. Dashed lines indicate where peak and tail current were measured 65 ms after depolarization and 15 ms after repolarization, respectively. In the presence of LTC agonist, 1 μM FPL, both the peak and long-lasting tail current increase (gray trace) and display slowed activation and deactivation kinetics, characteristic of FPL-induced L-current.  phosphorylation in ST14A cells, indicating that D 2 Rs do couple to intact signaling cascades such as the adenylyl cyclase and MAPK pathways [29].
A non-selective D 2 R-like agonist, Quin also activates D 3 and D 4 Rs [38]. Subsequently, D 3 and D 4 Rs modulate immediate early gene expression [39,40] raising the question of the identity of dopamine receptors in ST14A cells. Using RT-PCR, we examined dopamine receptor (D 1 , D 2 , D 3 , D 4 , and D 5 ) mRNA content in ST14A cells, striatum (positive control for all dopamine receptors), cortex, A9L cells (D 2 R positive control; see Materials and Methods) or HEK  293 cells (negative control for all dopamine receptors). PCR products were detected for long and short splice variants of the D 2 R in ST14A cells (n = 4/11 and 2/11, respectively, striatum (n = 1/2 and 2/2, respectively), cortex (n = 1/2 for both) and A9L cells (n = 2/5 and 2/5, respectively) as shown in Fig 4A. Additionally, D 1 R (n = 2/10), D 4 R (7/10), D 5 R (4/10) (data not shown) mRNAs were also detected in ST14A cells, regardless of the temperature at which cells were grown. D 1 R, D 4 R and D 5 R mRNAs were detected in striatum whereas D 1 R and D 4 R mRNAs were detected in the cortex (data not shown). D 4 R mRNA was also detected in A9L cells (n = 3/4). Experiments for D 3 R mRNA expression in striatum, cortex, and ST14A cell samples resulted in a smear despite several attempts to adjust the protocol. HEK 293 cells showed no expression of any of the dopamine receptors tested. These results show that ST14A cells express mRNA for more than one D 2 -like receptor; this finding could account for the previously reported D 2 R-like changes in pCREB [29].
Since MSNs are GABAergic, a defining characteristeric of ST14A cells being MSN-like would be expression of glutamic acid decarboxylase (GAD), the enzyme that catalyzes GABA synthesis from glutamate. Using RT-PCR, we measured whether ST14A cells express GAD2 and GAD1, two genes which encode GAD with molecular weights of 65 and 67 kDa, respectively. The primers used against GAD1 detect both embryonic and adult splice forms of GAD67. All three bands corresponding to GAD65 and embryonic and adult GAD67 were detected in ST14A cells (Fig 4B). Alternate lanes in which RNA samples were not reverse transcribed served as controls for genomic DNA contamination and yielded no product. Using striatal and cortical tissues as positive controls for the GAD genes, and in A9L cells, we detected all three forms of GAD as well. In contrast, no bands for GAD65 and only faint bands for GAD67 were detected in HEK 293 cells. The presence of GAD in ST14A cells further confirms that this cell line exhibits GABAergic characteristics of MSNs.
After confirming ST14A cells exhibit a similar mRNA expression profile for dopamine receptors and GAD as MSNs, we examined whether D 2 Rs and downstream signaling molecules were expressed in ST14A cells. Using Western blot analysis, D 2 R and M 1 R expression was confirmed in ST14A cells. D 2 R antibodies recognized a band at 50 kDa, the expected molecular weight of D 2 Rs (Fig 5A, top panel) for striatum and cortex (positive controls). A second band at 75 kDa, absent in the A9L cell line, most likely represents a glycosylated form of the receptor [41], further supporting the idea that the ST14A cells are similar to neurons in these brain regions regarding post-translational modification of proteins. Fig 5A (middle panel) shows that ST14A cells, but not A9L cells, express M 1 Rs, displayed as a 50 kDa band on Western blots. This antibody recognized M 1 R expression in SCG, striatum, and cortex [34]. Expression of G q α (Fig 5A, bottom panel), which couples to M 1 Rs, was also detected in the cell lines and tissue samples at the predicted molecular weight of 42 kDa. These results show that ST14A cells express D 2 R, M 1 R and G q α proteins.
Since D 2 Rs and M 1 Rs both couple to PLCβ-1, we examined whether ST14A cells express PLCβ-1. Protein expression of PLCβ-1 was not detected in either the A9L or ST14A cell lines, but was detected in SCG tissue (Fig 5B). The unanticipated absence of PLCβ-1 may account for the lack of Ca V 1.3 modulation since PLCβ-1 is required for LTC current inhibition by both the D 2 R and M 1 R pathways [10,42]. To determine the presence of molecules downstream of PLCβ-1 reported to participate in LTC modulation by these receptors [10,25], we tested for cPLA 2 and PP2B expression. Fig 5C shows that ST14A cells cultured at 33 and 37°C, as well as A9L cells, express cPLA 2 . cPLA 2 was detected at low expression levels from postnatal day 1 SCG (positive control) as has been reported previously [25]. Fig 5D shows that ST14A cells also express PP2B. Since PLCβ-1 is downstream of both D 2 and M 1 Rs, this result may explain our lack of inhibition of Ca V 1.3 or endogenous LTC current by Quin or Oxo-M in ST14A cells. To circumvent the absence of a key signaling molecule, and determine whether Ca V 1.3 could be modulated, we directly applied exogenous AA (10 μM) to the bath and measured Ca V 1.3 recombinant current over several minutes. In the presence of FPL, AA inhibited Ca V 1.3 peak and long-lasting tail currents by 40 ± 12% and 29 ± 25% respectively after 1 min (n = 4). Fig 6A shows representative sweeps before (FPL) and after AA application (FPL + AA). The large variability in tail current inhibition by AA suggested that voltage may be important for this modulation. However, AA inhibited Ca V 1.3 current at all voltages when tested over a range of test potentials as shown in the I-V plot in Fig 6B. Bovine serum albumin (BSA), which binds free fatty acids [43], reversed inhibition at all test potentials. To show that inhibition was not due to AA competing with FPL, we measured Ca V 1.3 current in the absence of FPL prior to and after application of AA ( Fig 6C) and still observed inhibition that could be reversed after adding BSA. Inhibition of Ca V 1.3 over time by AA (open bars) and recovery by BSA (solid bar) is summarized in Fig 6D. Conversely, oleic acid (10 μM) enhanced current by 7.9 ± 0.7% after 1 minute (Fig 6E; p < 0.001; n = 3), suggesting that inhibition by AA is not simply the result of a nonspecific fatty acid effect. Moreover, inhibition of Ca V 1.3 by AA demonstrates that the transfected Ca V 1.3 LTCs in ST14A cells are capable of modulation.

Discussion
MSNs have a resting membrane potential that oscillates between~-85 mV (during the "down" state) and~-60 mV (during the "up" state) [4,6]. Inhibition of Ca V 1.3 is of particular interest in MSNs because it activates at potentials approximately 25 mV more negative than Ca V 1.2, [30]. Since Ca V 1.3 activates at the low voltage of -60 mV, this channel may open during the "up" state and contribute to reaching threshold for firing an action potential [4]. Moreover, increased D 2 R signaling or inhibition of LTCs in MSNs decreases membrane excitability [10]. MSN activity produces the only output from the striatum and thus the finely tuned regulation of MSN activity is critical for normal motor function. Disruption of MSN regulation results in severe malfunction of the basal ganglia, as seen when MSNs lose dopaminergic input in Parkinson's disease, or when MSNs undergo cell death as seen in Huntington's disease [5,44].
Because studying Ca 2+ currents in MSNs has been challenging, we searched for a cell line that might serve as a model system for MSNs. However, remarkably few neuronal cell lines have been developed despite ongoing demand for their use in biophysical studies of ion channels and in high through put systems for therapeutic drug testing. We hypothesized that ST14A cells, compared to HEK 293 cells, would express postsynaptic endogenous D 2 R signaling microdomains closely matching MSNs, thus making ST14A cells useful for studying Ca V 1.3 modulation by D 2 R specific signaling. Therefore we examined ST14A cells to determine whether this cell line exhibits sufficient striatal properties to serve as a suitable model system for studying MSN functioning. We first tested whether ST14A cells express endogenous voltage-gated Ca 2+ current.
We characterized a small, endogenous Ca 2+ current from differentiated ST14A cells that develop neuronal-like processes when grown at 37°C. However, we found that only 5/15 cells had a current amplitude larger than 200 pA even in the presence of the LTC agonist FPL. This small current and the relatively low number of cells expressing a measurable current, is similar to previous findings [29]. K + , Na + and HCN-mediated currents, in addition to the ability to fire action potentials, are also reported in a small percentage of ST14A cells [29,45,46]. However, when present, the Ca 2+ current was within the lower range of current amplitudes recorded in MSNs, and was similar to L-current in MSNs in that the tail current was sensitive to LTC agonists, increasing~6-fold following exposure to FPL. L-current dominates few types of neurons; one of them being MSNs (e.g., [10] [47]). Though the Ca 2+ current in differentiated ST14A cells is small, L-current appears to make up much of the whole-cell Ca 2+ current. Further optimization of culture conditions and/or recording conditions may increase Ca V 1.3 expression and consequently L-current amplitudes.
As an alternative strategy to examining native Ca V 1.3 activity in ST14A cells, we attempted to transiently transfect round ST14A cells, which lack endogenous Ca 2+ currents, with LTC channel subunits and GFP. We didn't know whether ST14A cells would tolerate transfection of the multiple Ca 2+ channel subunits as well as express functional channels. However ST14A cells were transfected successfully with channel subunits and exhibited robust voltage-gated Ca 2+ currents. We characterized the biophysical properties of isolated, recombinant Ca V 1.3 current in a striatal-like background without the complications of primary MSNs, i.e., requiring several pharmacological blockers to silence other ionic or multiple types of Ca 2+ currents. These large currents appeared ideal for testing whether ST14A cells would support Ca V 1.3 current inhibition by the D 2 R agonist.
From the biochemical characterization of ST14A cells as MSN-like due to the expression of dopamine receptors (D 1 , D 2 , D 4 , D 5 ) and GAD65/67 mRNAs as well as expression of D 2 R, M 1 R, G q α, cPLA 2 and PP2B protein we anticipated observing LTC modulation. However, no inhibition of Ca V 1.3 current was observed with Quin or Oxo-M. Since PLCβ-1 expression is so widespread, we tested for its expression in ST14A cells only after finding no LTC modulation by Quin. Lack of PLCβ-1 protein expression was unexpected since several intact signaling cascades are described in the growing literature regarding ST14A cells. Most notably, ST14A cells have functional CREB phosphorylation following D 2 R stimulation and Ras/MAPK, adenylyl cyclase, Wnt and JAK/STAT signaling pathways [29,[48][49][50][51]. Moreover these cells express other enzymes that act on lipids including N-acyl-phosphatidylethanolamine-hydrolyzing phospholipase D (NAPE-PLD), fatty acid amide hydrolase, diacylglyceride lipase, monoacylglycerol lipase (Bari et al., 2013), and cPLA 2 (Fig 3C).
The absence of PLCβ-1 expression in ST14A cells coincides with the lack of Ca V 1.3 current inhibition by either Quin or Oxo-M, supporting the importance of PLCβ-1 for both D 2 R and M 1 R signal transduction pathways [10,52,53]. Despite PLCβ-1's absence, it was not obvious that no other variant of PLCβ would substitute for the missing PLCβ-1. The absence of PLCβ-1 and its apparent requirement for D 2 R signaling will be of interest to researchers who study MSNs. Interestingly, PLCβ-1 has been implicated in regulating growth and proliferation, where its absence results in uncontrolled cell proliferation [54,55]. Thus a loss of PLCβ-1 may have occurred during the immortalization of ST14A cells.
In these experiments, we used a short splice variant of Ca V 1.3, Ca V 1.3b [30]. Ca V 1.3a, which has a longer C-terminus, was not inhibited by activation of D 2 Rs when expressed in HEK 293 cells [56]; however, Ca V 1.3a has been shown to bind the scaffolding protein Shank found in the postsynaptic density of synapses [57]. This association is necessary for Ca V 1.3 current inhibition by D 2 Rs in primary MSNs [11]. Although the lack of the long C-terminus also could explain the absence of channel modulation by D 2 Rs in ST14A cells, Ca V 1.3b LTCs can be modulated by both IGF-1 and AA. Exposure of SH-SY5Y cells expressing either Ca V 1.3a or Ca V 1.3b to IGF-1 enhances both currents and requires phosphorylation of S1486, a residue shared by both splice variants [58]. We have measured significant Ca V 1.3b inhibition by AA in transiently transfected HEK 293 cells, consistent with a potential transmembrane site of action [27].
Since AA release occurs in the striatum from both neurons and astrocytes [59][60][61] following stimulation of D 2 Rs [21] or M 1 Rs [62], we tested whether bath application of AA modulated LTC in ST14A cells to be certain that Ca V 1.3b could be modulated by molecules downstream of PLCβ-1. We were unsure whether AA would inhibit Ca V 1.3b channels in ST14A cells similarly to Ca V 1.3 channels in HEK 293 cells [27] or native LTCs in SCG neurons [23][24][25] since a wide range of actions have been reported for AA modulation of a variety of Ca 2+ currents by other groups (see review by Roberts-Crowley et al [63]). We found that direct application of AA circumvented the lack of receptor-activated channel modulation in ST14A cells and inhibited Ca V 1.3b recombinant current. This finding demonstrates that Ca V 1.3 channels are capable of being modulated in ST14A cells as we have observed previously in HEK 293 cells [27]. The properties of Ca V 1.3 inhibition by AA in HEK 293 and ST14A cells appear similar. Additionally we have found that Oxo-M inhibits currents from Ca V 1.3b in HEK 293 cells stably transfected with M 1 Rs (unpublished data). Lastly, we have found that activation of D 2 Rs by 10 μM Quin inhibits Ca V 2.2 currents in HEK 293 cells (unpublished data) demonstrating that this agonist protocol should be sufficient to activate D 2 Rs in ST14A cells. Therefore, the lack of current modulation by D 2 Rs or M 1 Rs, reported here, is supported by the absence of PLCβ-1 rather than an inability of Ca V 1.3b to respond to modulation.

Conclusions
ST14A cells are used as a model system for both the study and treatment of Huntington's disease [64][65][66][67][68][69]. Despite the lack of D 2 R-or M 1 R-mediated Ca 2+ channel modulation in ST14A cells reported here, this cell line was useful for elucidating that AA inhibits Ca V 1.3 channels. We hypothesize that the consistency of AA's inhibitory actions on Ca V 1.3 across cell types will be of interest to researchers who study lipid signaling molecules. Moreover, we anticipate the D 2 R signaling cascade could be rescued by transfecting ST14A cells with PLCβ-1. Whether D 2 R signaling would then cause a lipid mediated inhibition of Ca V 1.3b LTCs awaits future studies. Thus, ST14A cells are a valuable tool for studying the biophysical properties of an isolated Ca 2+ current and the modulation of these channels by signaling molecules within the context of a striatal background to aid in understanding neuronal malfunctions of the striatum.