Adam C. Munhall

Evaluation of levodopa dose and magnitude of dopamine depletion as risk factors for levodopa-induced dyskinesia in a rat model of Parkinson's disease.

Levodopa dose and severity of Parkinsons disease (PD) are recognized risk factors for levodopa-induced dyskinesia (LID) in humans. The purpose of the present study was to evaluate the ability of these variables to predict severity of LID in a rat model of PD. Varied concentrations of 6-hydroxy-dopamine were injected into the midbrain to produce wide ranges of dopamine depletion in striatum. Three weeks later, rats were given daily injections of levodopa (2–10 mg/kg i.p.) plus benserazide (12.5 mg/kg i.p.) for 15 days. Abnormal involuntary movements (AIMs) were measured for limb, axial, orolingual, and rotatory movements. Dose-response analysis for total AIM scores yielded a levodopa ED50 value of 3.2 mg/kg on treatment day 15. There were strong interrelated correlations between individual AIM categories (ρ > 0.7) and for each AIM category in regard to total AIM score (ρ > 0.7). In rats that received levodopa doses that were greater than the ED50, rates of amphetamine-induced rotation were significantly correlated with total AIM scores (ρ = 0.413). However, of those rotating >5 times/min, 34% had relatively low AIM scores (<8). Likewise, there was a significant correlation between percentages of tyrosine hydroxylase (TH) loss and total AIM scores (ρ = 0.388). However, in those rats that had >85% TH loss, 30% had AIM scores <8. Our results show that given an adequate dose and magnitude of striatal dopamine depletion, levodopa produces dyskinesia with a continuous spectrum of severity. Although levodopa dose and level of dopamine depletion are significant risk factors for LID, we conclude that other factors must contribute to LID susceptibility.

Calcium‐dependent subthreshold oscillations determine bursting activity induced by N‐methyl‐d‐aspartate in rat subthalamic neurons in vitro

We used whole‐cell patch recordings in current clamp to investigate the ionic dependence of burst firing induced by N‐methyl‐d‐aspartate (NMDA) in neurons of the subthalamic nucleus (STN) in slices of rat brain. NMDA (20 µm) converted single‐spike firing to burst firing in 87% of STN neurons tested. NMDA‐induced bursting was blocked by AP5 (50 µm), and was not mimicked by the non‐NMDA receptor agonist AMPA (0.6 µm). Tetrodotoxin (1 µm) converted bursts to oscillations of membrane potential, which were most robust when oscillations ranged between −50 and −70 mV. The NMDA bursts were blocked by an elevated extracellular concentration of Mg2+, but superfusate containing no added Mg2+ either reduced or increased burst firing, depending upon the amount of intracellular current injection. Block of K+ conductances by apamin and tetraethylammonium prolonged burst duration, but iberiotoxin had no effect. NMDA‐induced burst firing and membrane oscillations were completely blocked by superfusate containing no added Ca2+, and they were significantly reduced when patch pipettes contained BAPTA. Selective antagonists for T‐type (mibefradil, 10 µm), L‐type (nifedipine, 3 µm), and N‐type (ω‐conotoxin GVIA, 1 µm) Ca2+ channels had no effect on NMDA burst firing. Superfusate containing a low concentration of Na+ (20 mm) completely abolished NMDA‐induced burst firing. Flufenamic acid (10 µm), which blocks current mediated by Ca2+‐activated nonselective cation channels (ICAN), reversibly abolished NMDA‐depended bursting. These results are consistent with the hypothesis that NMDA‐induced burst firing in STN neurons requires activation of either an ICAN or a Na+–Ca2+ exchanger.

Dopamine receptor supersensitivity in rat subthalamus after 6-hydroxydopamine lesions

The subthalamic nucleus (STN) receives direct dopaminergic innervation from the substantia nigra pars compacta, but the importance of this input in the pathophysiology of parkinsonism remains to be determined. We used whole‐cell patch‐clamp recordings in brain slices to study presynaptic dopaminergic modulation of synaptic inputs to the STN in unilateral 6‐hydroxydopamine (6‐OHDA)‐lesioned rats. Here, we report that dopamine was more potent for inhibiting GABA IPSCs and glutamate EPSCs in the STN ipsilateral to the lesion, and was less potent for suppressing IPSCs and EPSCs in the STN contralateral to the lesion, compared with the effects of dopamine in control STN. Dopamine reduced IPSCs with an IC50 value of 20.9 ± 3.6 µm in control STN, whereas IC50 values were 0.83 ± 0.15 and 55.1 ± 11.1 µm in STN ipsilateral and contralateral to 6‐OHDA lesions, respectively. Dopamine also inhibited EPSCs with an IC50 value of 12.8 ± 2.8 µm in control STN, whereas IC50 values were 4.5 ± 0.9 and 41.6 ± 9.8 µm in STN ipsilateral and contralateral to 6‐OHDA lesions, respectively. Results with paired stimuli to evoke EPSCs and IPSCs suggest that endogenous dopamine acts presynaptically to inhibit transmitter release in the STN. These results show that chronic dopamine denervation significantly alters the regulation of synaptic input to the STN. Our results also suggest that the STN may be an important target for levodopa therapy in Parkinsons disease.

Dopamine-mediated actions of ephedrine in the rat substantia nigra

Although ephedrine is a centrally active stimulant, its effect on midbrain dopamine neurons is not known. To study the effect of ephedrine on dopamine-containing cells, current-clamp microelectrode recordings were made from substantia nigra pars compacta (SNC) neurons in horizontal brain slice preparations. Ephedrine (100-1000 microM) slowed spontaneous firing and produced a modest concentration-dependent hyperpolarization of membrane potential (EC50 279 microM), with a concomitant net decrease in membrane resistance. These effects were blocked by the D2-like dopamine antagonist sulpiride (1 microM). Electrically evoked inhibitory synaptic potentials mediated by GABAB receptors were reduced 28% by ephedrine. However, ephedrine did not reduce fast synaptic potentials mediated by GABAA or ionotropic glutamate receptors. Inhibition of the GABAB response appeared to be mediated by a postsynaptic mechanism because ephedrine also reduced baclofen-induced hyperpolarization by 28%. Both ephedrine-induced hyperpolarization and inhibition of baclofen-induced hyperpolarization were abolished when slices were superfused with the tyrosine hydroxylase inhibitor alpha-methyl-para-tyrosine (AMPT). Despite perfusion with AMPT, the ability of ephedrine to cause hyperpolarization was restored after perfusing the slice with dopamine (30 microM). Taken together, these results suggest that ephedrine causes hyperpolarization and suppresses GABAB receptor-mediated effects by releasing endogenous dopamine. However, the high concentrations required to observe these effects in vitro suggest that biologically relevant central effects of ephedrine are more likely to be mediated either by non-dopamine systems, such as those involving noradrenaline, or by dopamine systems outside the SNC.

NMDA enhances a depolarization-activated inward current in subthalamic neurons.

Previous studies have shown that N-methyl-D-aspartate (NMDA) receptor stimulation evokes Ca2+- and Na+-dependent burst firing in subthalamic nucleus (STN) neurons. Using whole-cell patch pipettes to record currents under voltage-clamp, we identified a time-dependent depolarization-activated inward current (DIC) that may underlie NMDA-induced burst firing in STN neurons in rat brain slices. Continuous superfusion with NMDA (20 microM) elicited a marked TTX-insensitive inward current when the membrane was depolarized to the level of -70 or -50 mV, from a holding potential of -100 mV. This current had a long duration, and its peak amplitude occurred at a test potential of -60 mV. DIC could not be evoked using the non-NMDA receptor agonist D,L-alpha-amino-3-hydroxy-5-methylisoxalone-4-propionic acid (AMPA). DIC was blocked by either intracellular BAPTA or by removal of extracellular Ca2+, but selective blockers of T-type (mibefradil), L-type (nifedipine) and N-type (omega-conotoxin GVIA) Ca2+ channels did not. Perfusing slices with a low extracellular concentration of sodium abolished the NMDA-induced DIC, implying that both Ca2+ and Na+ are necessary for the expression of DIC. Transient receptor potential (TRP) channel blockers flufenamic acid and SKF96365 severely reduced DIC amplitude, whereas NMDA-gated currents were either increased or were unchanged. These results suggest that the activation of NMDA receptors enhances a Ca2+-activated non-selective cation current that may be mediated by a member of the TRP channel family in STN neurons.

Mitochondrial uncoupling agents antagonize rotenone actions in rat substantia nigra dopamine neurons.

Mild uncoupling of oxidative phosphorylation has been reported to reduce generation of reactive oxygen species (ROS) and therefore may be neuroprotective. We reported previously that the mitochondrial poison rotenone enhanced currents evoked by N-methyl-D-aspartate (NMDA) by a ROS-dependent mechanism in rat midbrain dopamine neurons. Thus, rotenone, which produces a model of Parkinsons disease in rodents, may increase the risk of dopamine neuron excitotoxicity. The purpose of this study was to test the hypothesis that oxidative phosphorylation uncoupling agents would antagonize the effect of rotenone on NMDA current. We used patch pipettes to record whole-cell currents under voltage-clamp (-60 mV) in substantia nigra dopamine neurons in slices of rat brain. Rotenone, NMDA and uncoupling agents were added to the brain slice superfusate. Inward currents evoked by NMDA (30 μM) more than doubled in amplitude after slices were superfused for 30 min with 100 nM rotenone. Continuous superfusion with the uncoupling agent carbonyl cyanide-p-trifluoromethoxy-phenylhydrazone (1-3 nM) or 2,4-dinitrophenol (100 nM) significantly antagonized and delayed the ability of rotenone to potentiate NMDA currents. Coenzyme Q₁₀ (1-10 nM), which has been reported to facilitate uncoupling protein activity, also antagonized this action of rotenone. These results suggest that mild uncoupling of oxidative phosphorylation may protect dopamine neurons against injury from mitochondrial poisons such as rotenone.

Tyramine excites rat subthalamic neurons in vitro by a dopamine-dependent mechanism

Tyramine, an endogenous ligand for mammalian trace amine-associated receptors, may act as a neuromodulator that regulates neuronal activity in basal ganglia. Using whole-cell patch recordings of subthalamic nucleus (STN) neurons in rat brain slices, we found that bath application of tyramine evoked an inward current in voltage-clamp in over 60% of all STN neurons. The inward current induced by tyramine was mimicked by the D(2)-like dopamine receptor agonist quinpirole, but was only partially blocked by the D(2)-like receptor antagonist sulpiride. In contrast, the D(1)-like receptor agonist SKF38393 evoked no current in STN neurons. Inward current evoked by tyramine was significantly reduced by the catecholamine uptake inhibitor nomifensine, and by exhausting catecholamines in the brain via pretreatment with reserpine. Tyramine also reduced the amplitude of GABA(A) receptor-mediated IPSCs that were evoked by focal electrical stimulation of the slice. Inhibition of IPSCs by tyramine was mimicked by quinpirole and was blocked by sulpiride but not by SCH23390, a D(1) receptor antagonist. Moreover, tyramine-induced inhibition of IPSCs was reduced in slices pretreated with reserpine, and this inhibition could be restored by briefly superfusing the slice with dopamine. These results suggest that tyramine acts as an indirect dopamine agonist in the STN. Although inhibition of IPSCs is mediated by D(2)-like receptors, the dopamine-dependent inward currents evoked by tyramine do not fit a typical dopamine receptor pharmacological profile.

NMDA alters rotenone toxicity in rat substantia nigra zona compacta and ventral tegmental area dopamine neurons

Previous patch-clamp studies by our laboratory showed that acute exposure to the pesticide rotenone augments inward currents evoked by N-methyl-d-aspartate (NMDA) in substantia nigra zona compacta (SNC) dopamine neurons in slices of rat brain. The present experiments were done to search for histological evidence of increased neurotoxicity produced by combined rotenone and NMDA treatments. In horizontal slices of rat midbrain, we found that a 30 min superfusion with 100 nM rotenone caused significant injury to tyrosine hydroxylase (TH)-positive proximal dendrites in dorsal and ventral regions of the SNC and ventral tegmental area (VTA). Moreover, treatment with 100 μM NMDA potentiated rotenone toxicity. In contrast, treatment with 30 μM NMDA protected against rotenone-induced injury to dendrites in the ventral SNC and ventral VTA. Interestingly, treatment with 30 μM NMDA-alone produced an apparent increase in proximal dendrite scores in ventral SNC and dorsal VTA. We conclude that NMDA has concentration-dependent actions on rotenone toxicity that differ according to regional subtype of dopamine neuron.

AMP kinase regulates ligand-gated K-ATP channels in substantia nigra dopamine neurons

AMP-activated protein kinase (AMPK) is a master enzyme that regulates ATP-sensitive K(+) (K-ATP) channels in pancreatic beta-cells and cardiac myocytes. We used patch pipettes to record currents and potentials to investigate effects of AMPK on K-ATP currents in substantia nigra compacta (SNC) dopamine neurons in slices of rat midbrain. When slices were superfused repeatedly with the K-ATP channel opener diazoxide, we were surprised to find that diazoxide currents gradually increased in magnitude, reaching 300% of the control value 60min after starting whole-cell recording. However, diazoxide current increased significantly more, to 472% of control, when recorded in the presence of the AMPK activator A769662. Moreover, superfusing the slice with the AMPK blocking agent dorsomorphin significantly reduced diazoxide current to 38% of control. Control experiments showed that outward currents evoked by the K-ATP channel opener NN-414 also increased over time, but not currents evoked by the GABAB agonist baclofen. Delaying the application of diazoxide after starting whole-cell recording correlated with augmentation of current. Loose-patch recording showed that diazoxide produced a 34% slowing of spontaneous firing rate that did not intensify with repeated applications of diazoxide. However, superfusion with A769662 significantly augmented the inhibitory effect of diazoxide on firing rate. We conclude that K-ATP channel function is augmented by AMPK, which is activated during the process of making whole-cell recordings. Our results suggest that AMPK and K-ATP interactions may play an important role in regulating dopamine neuronal excitability.

Phosphoinositol metabolism affects AMP kinase-dependent K-ATP currents in rat substantia nigra dopamine neurons

We reported recently that ligand-gated ATP-sensitive K+ (K-ATP) current is potentiated by AMP-activated protein kinase (AMPK) in rat substantia nigra compacta (SNC) dopamine neurons. Because phosphatidylinositol 4,5-bisphosphate (PI(4,5)P2) regulates K-ATP current, we explored the hypothesis that changes in PI(4,5)P2 modify the ability of AMPK to augment K-ATP current. To influence PI(4,5)P2 levels, we superfused brain slices with phospholipase C (PLC) activators and inhibitors while recording whole-cell currents in SNC dopamine neurons. Diazoxide, superfused for 5 min every 20 min, evoked K-ATP currents that, on average, increased from 38 pA at first application to 122 pA at the fourth application, a 220% increase. This enhancement of diazoxide-induced current was AMPK dependent because K-ATP current remained at baseline when slices were superfused with either the AMPK inhibitor dorsomorphin or the upstream kinase inhibitor STO-609. The PLC inhibitor U73122 significantly increased diazoxide current over control values, and this increase was blocked by dorsomorphin. Enhancement of diazoxide-induced current was also completely prevented by the PLC activator m-3M3FBS. Agonists at 5-HT2C and group I metabotropic glutamate receptors, both of which activate PLC, also prevented augmentation of diazoxide-induced current. Finally, inhibition of spike discharges by diazoxide was significantly antagonized by m-3M3FBS. These results suggest that PLC activity significantly influences the inhibitory effect of K-ATP channels by altering PI(4,5)P2 content. Results also suggest that modification of K-ATP current by PLC requires AMPK activity.