Increased T-type Ca2+ channel activity as a determinant of cellular toxicity in neuronal cell lines expressing polyglutamine-expanded human androgen receptors

We have analyzed Ca2+ currents in two neuroblastoma-motor neuron hybrid cell lines that expressed normal or glutamine-expanded human androgen receptors (polyGln-expanded AR) either transiently or stably. The cell lines express a unique,
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  23  Molecular and Cellular Biochemistry  203:  23–31, 2000.© 2000  K luwer Academic Publishers. Printed in the Netherlands. Increased T-type Ca 2+  channel activity as adeterminant of cellular toxicity in neuronal celllines expressing polyglutamine-expanded humanandrogen receptors Adrian Sculptoreanu, 1, 3  Hanan Abramovici, 1  Abdullah A.R. Abdullah, 1, 4 Anna Bibikova, 1  Valerie Panet-Raymond, 1, 4  Dov Frankel, 1, 6  HymanM. Schipper, 1, 6  Leonard Pinsky 1, 2, 4, 5  and Mark A. Trifiro 1, 5 1  Lady Davis Institute for Medical Research, SMBD-Jewish General Hospital; Departments of 2  Human Genetics, 3  Experimental Medicine and Surgery, 4  Biology, 5  Medicine, 6  Neurology and Neurosurgery, McGill University, Montreal,Canada Received 5 May 1999; accepted 8 July 1999 Abstract We have analyzed Ca 2+  currents in two neuroblastoma-motor neuron hybrid cell lines that expressed normal or glutamine-expanded human androgen receptors (polyGln-expanded AR) either transiently or stably. The cell lines express a unique, low-threshold, transient type of Ca 2+  current that is not affected by L-type Ca 2+  channel blocker (PN 200-110), N-type Ca 2+  channel blocker ( ω -conotoxin GVIA) or P-type Ca 2+  channel blocker (Agatoxin IVA) but is blocked by either Cd 2+  or Ni 2+ . This pharmacological profile most closely resembles that of T-type Ca 2+  channels [1–3]. Exposure to androgen had no effect oncontrol cell lines or cells transfected with normal AR but significantly changed the steady-state activation in cells transfectedwith expanded AR. The observed negative shift in steady-state activation results in a large increase in the T-type Ca 2+  channelwindow current. We suggest that Ca 2+  overload due to abnormal voltage-dependence of transient Ca 2+  channel activation maycontribute to motor neuron toxicity in spinobulbar muscular atrophy (SBMA). This hypothesis is supported by the additionalfinding that, at concentrations that selectively block T-type Ca 2+  channel currents, Ni 2+  significantly reduced cell death in celllines transfected with polyGln-expanded AR. (Mol Cell Biochem 203 : 23–31, 2000)  Key words : T-type Ca 2+  channel, polyglutamine-expanded androgen receptor, CAG trinucleotide repeats, spinobulbar muscular atrophy, apoptosis, motorneuron, cell lines, neuroblastoma  Abbreviations : AGA – agatoxin; AR – androgen receptor; AR-20, AR-44 – transiently transfected androgen receptors; AR24,AR65 – stably transfected androgen receptors; ω CTX – conotoxin; DRG – dorsal root ganglia neurons; Mb – Mibolerone; MN – motor neuron; poly Gln – poly glutamine; SBMA – spinobulbar muscular atrophy  Address for offprints : A. Sculptoreanu, University of Pittsburgh, School of Medicine, Department of Pharmacology, E1304 Biomedical Science Tower,Pittsburgh, PA 15261, USA  24 Introduction Eight different hereditary neuronopathies are caused by theexpansion of translated CAG trinucleotide repeats in genesencoding a diverse group of proteins [4]. The members of thisgroup of central nervous system degenerative diseases aredistinguished by selective vulnerability of specific sets of neurons to the noxious effects of the polyglutamine(polyGln)-expanded proteins produced by these mutant genes[5]. The diseases are Huntington disease, spinocerebellar ataxias (SCA) types 1, 2, 6 and 7, Machado-Joseph disease(SCA3), dentatorubral-pallidoluysian atrophy, and spinobulbar muscular atrophy (SBMA) [6 – 9]. Only two of the proteins encoded by these genes are known: the androgen receptor atthe X-linked locus for SBMA [10], and the α 1A  subunit of voltage-dependent calcium channel at a locus on 19p for SCA6 [11].SBMA is an adult-onset motor neuronopathy, associatedwith mild androgen insensitivity, caused by an expansion of the CAG repeat in the AR gene from a normal polymorphicrange of 9–36 to a pathogenic range of 38–62 CAG repeats[12]. Since subjects with complete androgen insensitivity dueto a complete gene deletion do not develop SBMA, the polyGln-expanded AR must become neuronotoxic by a gain,not a loss, of function [13]. However, androgens and the AR are well known to be selectively motor neuronotrophic [14– 18]. Therefore, it is possible that a polyGln-expanded AR alsoloses a function that is necessary, or contributory, but notsufficient to be motor neuronopathic. This loss of functionmay, or may not, be the one that is responsible for the mildandrogen insensitivity component of SBMA. Provocatively,there is reason to believe that a combination of gained andlost functions also applies to the α 1A  subunit of a voltage-dependent calcium channel (P-type) that is involved in SCA6[11, 19]. A large body of data strongly suggests that polyGln-expanded portions of the parental proteins are the proximate pathogenic agents in this group of diseases. In this light, itis relevant to record recent evidence that polyGln expansionalters the nature, or the extent, to which the mutant AR iscleaved by proteases [20, 21] including one or more pro- apoptotic cysteine proteases, also known as caspases [22,23]. The polyGln-expanded fragments generated therebyare believed to be motor neuronopathic, at least in part, bytheir tendency to aggregate, noncovalently [24] or cov-alently [25, 26] with each other, or with other proteins.These aggregates appear to form inclusions in, or around,the nucleus of selected neurons [5, 27]. It remains to be determined, however, whether such neuronal inclusions arethe cause, or a consequence, of the neuronal toxicity [28].To help make this determination and to define correlativedysfunctions, we tested the hypothesis that a polyGln-expanded AR (or an expanded fragment thereof) gains aharmful property that alters the behavior of voltage-dep-endent Ca 2+  channels. Materials and methods  DNA transfection studies  NSC-34 cells, a hybrid cell line generated by fusion of mouseneuroblastoma cells and embryonic spinal cord motor neurons [29] were lipofected with either pcDNA3-hAR-20or pcDNA3-hAR-44 using lipofectamine according tomanufacturer’s instructions (GIBCO/BRL). Electrophysio-logical experiments were done within the first week of transfection. Five nM Mibolerone (a synthetic unmeta- bolizable androgen) was used to test the effect of androgen.We also studied spinal cord motor neuron-neuroblastoma(MN) hybrid cells stably transfected with normal (AR24-1)or polyGln-expanded AR (AR65-8) [30]. These cells wereused for electrophysiological and neuronotoxicity studies.They were compared with parental MN hybrid cells that weremock-transfected only with the neomycin-resistance gene.In some experiments, these stably transfected cells weresubjected to differentiating conditions as described by Brooks et al  . [31], for the exception of collagen which was used tocoat the surface of the culture walls. Under these conditions,the cells developed neurite projections.  Electrophysiological measurements For whole-cell recordings, Na + , K  +  and Cl  –   channel currentswere suppressed. The pipette (intracellular) solution consistedof (in mM): N-methyl-D-glucamine (130), EGTA (free acid)(20), BAPTA (5), HEPES (10), Mg(OH) 2  (6), Ca(OH) 2 (4), Mg-ATP (0.3), pH buffered to 7.3 with methanesulfonic acid. Theexternal solution contained (in mM): Ca(OH) 2  (20), tetrae-thylammonium-OH (55), Trizma-base (65), 4-aminopyridine(5), HEPES (10), pH buffered to 7.4 with methanesulfonicacid. Whole-cell currents were recorded using an Axopatch200A patch clamp amplifier. Currents were digitized after filtering at 2 kHz, with capacitative currents nulled and upto 80% of the series resistance compensated. Pulse generation,current recording, and data analysis were performed usingsoftware based on the Fastlab system (Indec Systems,Sunnyvale, CA, USA). The pure (+) stereoisomer of thedihydropyridine antagonist PN 200-110 (isradipine) and SDZ202–791 were generous gifts from Sandoz Canada (presently Novartis). Stock solutions, 10 mM in concentration, were prepared in 98% ethanol and stored at –20°C. The finalconcentrations of solvent were < 0.1% and had no effect onthe calcium channel currents. All experiments were performedat room temperature (20–23°C).  25  Determination of cell toxicity For cell toxicity measurements, the monolayers were stainedwith 10 µg/ml ethidium monoazide bromide (EMA; Molecular Probes) in PBS for 10 min at room temperature under UV lightexposure. The cells were subsequently washed twice in PBSand fixed in 4% paraformaldehyde for 20 min at room tem- perature. Unlike standard ethidium bromide, EMA covalently binds to the DNA of dead cells thereby preventing dye leakage post-fixation [32]. Total cell numbers were determined in 25 ×  fields under phase microscopy. Dead cells were identified by the presence of bright red nuclear staining under epi-fluorescence microscopy (Leitz Diaplan Photomicroscope)using a rhodamine filter. The appearance of shrunken, con-densed nuclei and the presence of small round apoptotic-like bodies constituted morphological evidence of cell death inthese cultures. Eight sister cultures were evaluated for eachexperimental group. A total of 140 cells were assessed per culture. The extent of cell death was expressed as the ratio of EMA-positive cells to total cells in each monolayer. Statisticalanalysis was performed using Student’s unpaired t  -test (2-tailed) with p < 0.05 indicating significance. Results  Electrophysiological characterization of motor neuron-likehybrid cell lines lacking androgen receptors, or stablyexpressing normal or polyGln-expanded AR We characterized the Ca 2+  channel currents in the untrans-fected NSC-34 neuroblastoma-motor neuron hybrid cell line[29], and in a similar line [30] stably transfected with normal (AR24) or polyGln-expanded AR (AR65; [31]).The parameters studied were voltage-dependent activationand inactivation, as well as the effects of specific Ca 2+  blockers and agonists. These effects of Ca 2+  blockers andagonists were compared with those on dorsal root ganglianeurons that are known to express all major neuronal-typeCa 2+  channel currents ([33], Figs 1 and 2, right panels).The Ca 2+  currents in these cell lines were unresponsive toL-type Ca 2+  channel dihydropyridine agonist (SDZ 202–791,Fig. 1B) and antagonist (PN 200-110, Figs 1D and 2B), N-type Ca 2+  channel antagonist ω -conotoxin GVIA (CTX, Fig.1B) or P-type channel blocker Agatoxin IVA (Fig. 2B).However, Cd 2+  (Fig. 2B) and Ni 2+  (Fig. 2D) effectively blocked the Ca 2+  currents in the hybrid cell lines at 50–200µM, and their effect was independent of androgen exposureor extent of cell differentiation (results not shown). Figure2D shows the effect of Ni 2+  in a AR65 MN cell line. Similar effects of Ni 2+  were observed in NSC cells (not shown). Thesedata suggest that the only type of Ca 2+ -channel currentexpressed in the untransfected NSC-34 cell line and in bothstably transfected cell lines (AR24; AR65) most closelyresembles the T-type Ca 2+  channel [2, 34, 35].  Effect of polyGln-expanded AR on Ca 2+  channels intransiently transfected NSC-34 cells We tested the effect of androgen on Ca 2+  channel currentsin mock-transfected NSC-34 cells, and in their counterpartstransfected with AR-20 or AR-44. In all three situations(Fig. 3, A–C), their currents activated around –50 mV duringdepolarization, their normalized peak current had a maximumaround –20 mV, and their steady-state inactivation wasunchanged by the addition of the androgen mibolerone (Mb,Fig. 3, D–F). In contrast, addition of Mb resulted in asignificant shift in activation toward negative voltages inthose cells transfected with AR-44 (Figs 3C and 3F), but notin those mock-transfected or expressing AR-20.  Fig. 1 .L- and N- type Ca 2+  channel agonists and antagonists are ineffectivein the NSC-34 cell line. Calcium channel current was recorded in the whole-cell patch clamp configuration from NSC-34 cells and dorsal root ganglia(DRG) neurons. Only control currents and steady-state effect of drugs onCa 2+  currents are shown. Numbers represent the sequence in which thedrugs were added. (A) Test pulse protocol used to stimulate Ca 2+  channels;a square pulse from –80 mV to either –20 mV (B, C) or 0 mV (D, E) wasrepeated every 10 sec. (B, D) Ca 2+  currents in NSC-34 cells wereunresponsive to L- and N-type Ca 2+  channel blockers (PN, PN200-110;CTX, ω -conotoxin GVIA) or L-type channel dihydropyridine (DHP) agonist(SDZ 202–791). (C, E) Comparative results obtained in DRG neurons.  26Analysis of tail currents at the end of depolarization pulsessuggests a biphasic voltage dependency of activation with ahalf-activation voltage of the first Boltzmann fit (smoothcurves, V 0.5 ) before androgen addition at about –30 mV (–27to –30 mV, Fig. 4, A–C). The V 0.5  was unchanged by exposureto androgen in the mock-transfected NSC-34 cells (Fig. 4A),or cells transfected with normal AR (Fig. 4B). However,androgen induced a marked negative shift in the V 0.5  of theBoltzmann fit (from –30 to –44 mV, Fig. 4C) in NSC-34 cellstransfected with polyGln-expanded AR. The androgen-induced negative shift in the voltage-dependent activation of Ca 2+  channel currents in the NSC-34 cells transfected with polyGln-expanded AR more than doubles the size of thewindow current (Fig. 3F). Note that the 20 mM Ca 2+  presentin these experiments would lead to a positive shift in thecurrent-voltage relationship of Ca 2+  channel currents at least10 mV in magnitude. Therefore the changes in windowcurrents reported here (negative shift) could lead to activationof Ca 2+  channels at the membrane potential of motor neurons(–70 to –60 mV).  Effect of differentiation on cell lines stably transfected with normal or polyGln-expanded AR Undifferentiated MN cells whether mock-transfected or stably transfected with AR24 or AR65 had Ca 2+  channel  Fig. 2 .P-Type Ca 2+  channel antagonists are also ineffective in the NSC-34cell line but T-type Ca 2+  channel blockers are effective. (D) shows the effectof Ni 2+  in a AR65 MN cell line. Similar effects of Ni 2+  were observed in NSCcells (not shown). Calcium channel currents were recorded in the whole-cell patch clamp configuration from NSC-34 cells and dorsal root ganglia (DRG)neurons. Only control currents and steady-state effect of drugs on Ca 2+  currentsare shown. Numbers represent the sequence in which the drugs were added.(A): Test pulse protocol used to stimulate Ca 2+  channels; a square pulse from –80 mV to 0 mV was repeated every 10 sec; (B, D): Ca 2+  currents in NSC-34cells were unresponsive to a combination of P-type Ca 2+  channel blocker,agatoxin IVA (AGA) and PN 200-110 (B-2), or to AGA alone (not shown).However, the inorganic Ca 2+  channel blockers, Cd 2+  and Ni 2+ , effectively blocked the Ca 2+  channel currents in NSC-34 cells; (C, E): Comparative resultsobtained in DRG neurons with AGA and Ni 2+ .  Fig. 3 .Expression of polyGln-expanded AR (AR-44) changes the voltagedependency of peak Ca 2+  channel currents. (Left panels) Current-voltagerelationship of peak Ca 2+  currents in NSC-34 cell lines before (emptysymbols) and after exposure to androgen (filled symbols). Square test pulses50 msec in duration were applied in increments of 10 mV from a holding potential of –80 mV. Peak currents were measured, normalized to individualcell membrane capacitances and plotted against step voltages. (Right panels)Simultaneous current-voltage relationship and steady-state inactivation. A pre-pulse 800 msec long to increasing voltages (pre-pulse), was followed by a 5 msec interval at –60 mV and a test pulse, 100 msec in duration, to0 mV. Peak currents during the pre-pulse (current-voltage relationship)and test-pulse (steady-state inactivation) were measured, normalized toindividual cell membrane capacitances and plotted against step voltages.(A) Mock transfection: no Mb, n = 29; with Mb, n = 15. (B) Cells transfectedwith AR-20: no Mb, n = 10; with Mb, n = 5. (C) Cells expressing AR-44:no Mb, n = 12; with Mb, n = 17. Note the marked shift to negative voltages.(D–F) Simultaneous current-voltage relationship of peak Ca 2+  currents andsteady-state inactivation curves (D: no Mb, n = 19; with Mb, n = 6, E: noMb, n = 8; with Mb, n = 6, F: no Mb, n = 6; with Mb, n = 8). Note thatinactivation is unchanged in all cells tested.  27currents very similar to those of NSC-34 cells. For example,their peak current-voltage relationship had a threshold around –50 mV and a maximum around –20 mV (Fig. 5, A–F). These properties were not changed by differentiation or by additionof androgen, except in those cells stably transfected with AR65(Fig. 5F), wherein these two factors acted cumulatively, assummarized in Fig. 6. Toxicity in cell lines expressing polyGln-expanded AR isreversed by a specific T-type Ca 2+  channel blocker  Cells stably transfected with polyGln-expanded AR (AR65,Fig. 7B) showed a marked increase in cell death relative tocells transfected with normal AR (Fig. 7A). To test if theincrease in T-type Ca 2+  channel window current correlateswith cell toxicity, we measured the effect of Ni 2+ , a specificT-type Ca 2+  channel blocker, or the effect of 10 mM K  + , which by depolarizing the membrane potential near 0 mV wouldinactivate most voltage-dependent Ca 2+  entry mechanisms(AR65, Figs 8B and 8C). Fifty µM Ni 2+  blocked about 50%of the Ca 2+  currents. At the half-maximal concentrationrequired for T-type Ca 2+  channel block, Ni 2+  specifically andlargely reversed cell toxicity in cell lines stably transfected  Fig. 4 .Transient expression of polyGln-expanded AR-44 shifts the voltagedependency of Ca 2+  channel currents to more negative values in NSC-34cells. (A–C) Current-voltage relationship of tail Ca 2+  current densities atthe end of a 50 msec test pulse before (empty symbols) and after (filledsymbols) exposure to androgen. (A) mock transfection: no Mb, n = 26;with Mb, n = 27). (B) AR-20: no Mb, n = 7; with Mb, n = 5). (C) AR-44: noMb, n = 13; with Mb, n = 11. Note the marked shift toward negative voltages(from –30 mV to –44 mV) in the AR-44-transfected cells after exposureto androgen. Currents were normalized to individual cell membranecapacitances. Smooth curves were fitted to the data with the sum of twoBoltzmann’s equations of the form: (p1 / (1 + exp(–(V–V1 0.5 )/k1)) + (p2 /(1 + exp(–(V–V2 0.5 )/k2)). The half voltage of activation of the firstBoltzmann (V 0.5 ) is shown as insets.  Fig. 5 .Differentiation increases T-type Ca 2+  channel currents in cells stablytransfected with polyGln-expanded AR65. Ca 2+  currents were measured inwhole-cell patch clamp configuration in ‘mock’ transfected cells, or in thoselines stably transfected with AR24 or AR65 and are plotted as current-voltage relationship of peak Ca 2+  currents. The behavior of Ca 2+  channelcurrents in cell lines without differentiation before (empty symbols) andafter (filled symbols) exposure to androgen. (A) mock transfection: no Mb,n = 10; with Mb, n = 17). (C) AR24: no Mb, n = 15; with Mb, n = 15. (E)AR65: no Mb, n = 22; with Mb, n = 21. The behavior of Ca 2+  channelcurrents in differentiated cells before (empty symbols) and after (filledsymbols) exposure to androgen. (B) mock transfection: no Mb, n = 31;with Mb, n = 27. (D) AR24: no Mb, n = 31; with Mb, n = 27. (F) AR65: noMb, n = 39; with Mb, n = 34. Note the marked increase in Ca 2+  channelcurrent densities after differentiation in AR65-transfected cells, and further increases of the current densities in these cells after exposure to androgen.Currents were normalized to individual cell membrane capacitances. Resultsare shown as mean ± S.E.M.
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!