- Email: [email protected]
Nanomolar concentrations of lead inhibit glutamatergic and GABAergic transmission in hippocampal neurons
Brain Research 826 Ž1999. 22–34
Research report
Nanomolar concentrations of lead inhibit glutamatergic and GABAergic transmission in hippocampal neurons Maria F.M. Braga b
a,b
, Edna F.R. Pereira a , Edson X. Albuquerque
a,b,)
a Department of Pharmacology and Experimental Therapeutics, UniÕersity of Maryland School of Medicine, Baltimore, MD 21201, USA Department of Basic and Clinical Pharmacology, Institute of Biomedical Sciences, Center of Health Sciences, Federal UniÕersity of Rio de Janeiro, Rio de Janeiro, RJ 21944, Brazil
Accepted 26 January 1999
Abstract To investigate whether lead ŽPb 2q . affects the tetrodotoxin ŽTTX.-sensitive release of neurotransmitters, the whole-cell mode of the patch-clamp technique was applied to cultured hippocampal neurons. Pb 2q ŽG 10 nM. reversibly blocked the TTX-sensitive release of glutamate and g-aminobutyric acid ŽGABA., as evidenced by the reduction of the amplitude and frequency of glutamate- and GABA-mediated postsynaptic currents ŽPSCs. evoked by spontaneous neuronal firing. This effect of Pb 2q, which occurred 2–3 s after exposure of the neurons to Pb 2q-containing external solution, was not related to changes in Naq-channel activity, and was quantified by measurements of changes in the amplitude of PSCs evoked when a 50-ms, 5-V stimulus was applied via a bipolar electrode to a neuron synaptically connected to the neuron under study. With an IC 50 of approximately 68 nM, Pb 2q blocked the evoked release of glutamate and GABA. This effect was most likely mediated by Pb 2q ’s actions on extracellular targets, because there was a very short delay Ž- 3 s. for its onset, and it could be completely reversed by the chelator ethylene diaminetetraacetic acid ŽEDTA.. Given that Pb 2q-induced blockade of evoked transmitter release could be reversed by 4-aminopyridine, it is suggested that the effect on release was mediated via the binding of Pb 2q to voltage-gated Ca2q channels. Thus, it is most likely that the neurotoxic effects of Pb 2q in the mammalian brain involve a decrease of the TTX-sensitive, Ca2q-dependent release of neurotransmitters. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Hippocampus; Pb 2q; Electrophysiology; GABA; Glutamate; Transmitter release
1. Introduction The poor understanding of the mechanisms underlying the neurotoxic effects of lead ŽPb 2q ., a pollutant commonly found in the environment, has hindered the development of efficacious therapeutic means by which Pb 2q-induced neurotoxicity could be prevented andror treated. The higher sensitivity of the developing brain to the toxic effects of Pb 2q could be due in part to cellular processes that occur during early stages of development and that are particularly sensitive to the actions of this environmental pollutant w1,2,10x. Thus, considering that the degree of neuronal activity is an important factor for the establishment of synaptic connections in the central nervous system ŽCNS. w47x, it has been suggested that, by interfering with )
Corresponding author. Department of Pharmacology and Experimental Therapeutics, University of Maryland School of Medicine, 655 West Baltimore Street, Baltimore, MD 21201, USA. Fax: q1-410-706-3991; E-mail: [email protected]
neuronal excitability at early stages of development, Pb 2q could alter synaptic plasticity, thereby causing impairment of learning and memory Žsee Ref. w12x.. Although numerous studies have investigated the effects of Pb 2q in transmitter release processes in the peripheral and central nervous systems w8,22,34,35,43,49,50x, very few have addressed the effects of Pb 2q on specific release processes w20,21,32x. A recent electrophysiological study has demonstrated that at submicromolar concentrations Pb 2q increases specifically the tetrodotoxin ŽTTX.-insensitive release of neurotransmitters from hippocampal neurons, an effect that is apparently mediated by the interaction of the heavy metal with targets within the presynaptic terminals w12x. Given that the establishment of functional synaptic connections in the CNS is dependent on the level of overall neuronal activity w29,53x, in the present study, by means of the patch-clamp technique, we have investigated the effects of Pb 2q on the action potential-dependent release of glutamate and GABA from hippocampal neurons in culture.
0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 1 1 9 4 - 4
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
The results presented herein demonstrate that Pb 2q blocks the action potential-dependent release of both GABA and glutamate with an IC 50 of approximately 68 nM, and that the magnitude of the effect of Pb 2q on evoked transmitter release is the same regardless of the stage of neuronal development in vitro. Our findings also indicate that Pb 2q-induced blockade of this transmitter release process Ži. is partially reversible upon washing of the neurons with nominally Pb 2q-free solution, Žii. requires the use of the chelator ethylene diaminetetraacetic acid ŽEDTA. to be fully reversed, and Žiii. can be accounted for by the blockade of voltage-gated Ca2q channels. The discovery that at very low concentrations ŽG 10 nM. Pb 2q reduces the action potential-dependent release
23
of neurotransmitters from hippocampal neurons strongly suggests that this may be one of the mechanisms underlying the cognitive deficits observed in patients, particularly children intoxicated with this heavy metal.
2. Materials and methods 2.1. Cultures of hippocampal neurons Cultures of hippocampal neurons were prepared by a procedure similar to that published elsewhere w3x. Neurons cultured for 10 to 30 days were used in this study.
Fig. 1. Characterization of spontaneously occurring EPSCs and IPSCs. Glutamate-mediated EPSCs or GABA-mediated IPSCs were recorded from hippocampal neurons continuously perfused with external solution containing the GABA A receptor blocker picrotoxin Ž100 mM. or the AMPA receptor antagonist CNQX Ž10 mM., respectively. Samples of spontaneously occurring EPSCs and IPSCs recorded at various membrane potentials are shown on the right. Notice that EPSCs reverse at approximately 0 mV, whereas IPSCs reverse between y10 and y20 mV. Note also that at membrane potentials more negative than y40 mV, EPSCs and IPSCs can be accompanied by fast current transients ŽU ., which were clipped from many sample recordings. The inset shows, in a different time and amplitude scale, the fast current transient accompanying the EPSC recorded at y60 mV and seen on the right side of the inset. Histograms on the left depict the distribution of the decay-time constants of EPSCs and IPSCs recorded from hippocampal neurons at q20 mV.
24
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
2.2. Postsynaptic currents (PSCs) PSCs were recorded from cultured hippocampal neurons according to the standard patch-clamp technique w26x through an LM-EPC-7 amplifier ŽList Electronics, Heidelberg, Germany.. The signals were filtered at 3 kHz, stored on VCR tapes, digitized at 50 ms and analyzed on an IBM-compatible microcomputer. A specially designed bipolar electrode was used to stimulate a neuron synaptically connected to the neuron under study. Application of a 50-ms, supramaximal stimulus Ž; 5 V. via this bipolar electrode, which was positioned approximately 100 mm from the presynaptic neuron, consistently resulted in the release of neurotransmitters and consequent activation of inhibitory ŽGABA-mediated. or excitatory Žglutamatemediated. PSCs. The bath solution used to perfuse the neurons at a rate of 2–4 mlrmin had the following composition Žin mM.: NaCl, 165; KCl, 5; CaCl 2 , 2;
glucose, 10; and N-w2-hydroxyethylxpiperazine-N X-w2ethane sulfonic acidx ŽHEPES., 5 ŽpH s 7.3 adjusted with NaOH; 340 mOsm.. The solution used to fill the patch pipettes had the following composition Žin mM.: CsCl, 160; Cs-ethylene-glycol bisŽb-amino-ethyl ether.-N, N X-tetraacetic acid, 10; and HEPES, 10 ŽpH s 7.3 adjusted with CsOH; 340 mOsm.. The patch microelectrodes were pulled from borosilicate capillary glass ŽWorld Precision Instruments, New Haven, CT. and when filled with the pipette solution had resistances in the range of 2–5 M V. All the experiments were performed at room temperature Ž22– 258C.. The test solutions were applied to the surface of the hippocampal neurons through an array of parallel glass tubes Ž400-mm i.d.. glued together and assembled on a Narishige micromanipulator. The time constant for solution exchange was 20 ms. These tubes were positioned at about 50–100 mm away from the neuron under study. Each tube was connected to a reservoir filled with the test
Fig. 2. Effects of different concentrations of Pb 2q on spontaneously occurring EPSCs and IPSCs. Samples of EPSCs and IPSCs recorded before, during, and after perfusion of hippocampal neurons with external solution containing Pb 2q Ž0.03, 0.1 or 1 mM.. Notice that both the amplitude and the frequency of the currents are reduced by Pb 2q, and that this effect is concentration dependent, being more intense with higher Pb 2q concentrations. ‘W’ indicates washing of the neurons with nominally Pb 2q-free external solution. Membrane potential, y50 mV.
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
solution and gravity fed at a rate of 1.2 mlrmin. A computer-controlled stepping motor was used to move the barrels, so that a single neuron could be exposed to different test solutions. 2.3. Drugs and toxins Dimethylsulfoxide, EDTA, PbCl 2 , picrotoxin, and TTX were purchased from Sigma ŽSt. Louis, MO.. CNQX was obtained from Research Biochemical International ŽNatick, MA. and from Sigma. A 250-mM stock solution of picrotoxin was made in dimethylsulfoxide, and dilutions were made in external solution. NaOH was used to dissolve CNQX Žthe 10 mM stock of CNQX had 12.2 mM NaOH.. All other chemicals were dissolved in double-distilled water. Solutions containing Pb 2q were prepared just before application to the neurons. 2.4. Data and statistical analysis The peak amplitude, 10–90% rise time, and decay-time constants of the PSCs were analyzed using a suite of SCAN-CDR programs w23x and also the pCLAMP program ŽAxon Instruments, CA.. Results are expressed as mean " S.E.M. The significance of differences was determined with Student’s t-test. To calculate IC 50 s and Hill coefficients, a four-parameter logistic function was used to fit the concentration–response curve by a non-linear leastsquares method: Effect s Ea q Ž Ei y Ea .rw1 q Ž KrD . h x, where: K s IC 50 , h s Hill coefficient, Ea s effect at zero drug concentration, Ei s effect at infinite drug concentration, and D s drug concentration.
25
with a time constant of 17.0 " 15.8 ms at q20 mV, had a reversal potential close to 0 mV, and were sensitive to inhibition by the AMPA receptor antagonist CNQX Ž10 mM. ŽFig. 1.. The CNQX-sensitive, rapidly decaying currents were mediated by glutamate released from spontaneously firing neurons synaptically connected to the neurons under study and are herein referred to as excitatory postsynaptic currents ŽEPSCs.. At membrane potentials more negative than y40 mV, fast current transients that represented back-propagating action potentials accompanied the EPSCs and the IPSCs Žsee Fig. 1.. These fast current transients were the consequence of depolarization induced by the neurotransmitters released from the presynaptic neurons onto unclamped areas of the neuron under study. It should be emphasized that fast current transients accompanying IPSCs were the result of depolarization caused by activation of GABA A receptors in unclamped neuronal areas where the membrane potential is more negative than the IPSC reversal potential. Fast current transients were not observed at membrane potentials more positive than y40 mV Žsee Fig. 1., because the inactivation of the Naq channels in the clamped areas blocked the propagation of the action potentials to the recording site.
3. Results 3.1. Pharmacological and kinetic characterization of PSCs mediated by glutamate and GABA released by spontaneous firing of hippocampal neurons in culture In the absence of the Naq-channel blocker TTX, two types of pharmacologically and kinetically distinct PSCs could be recorded from hippocampal neurons in culture. One type consisted of currents that decayed with a time constant of 53.5 " 16.1 ms at q20 mV, reversed at a membrane potential between y10 and y20 mV, and were sensitive to blockade by the GABA A receptor antagonist picrotoxin Ž100 mM. ŽFig. 1.. Considering the composition of the physiological solutions used in this study, the calculated Nernst potential for Cly is approximately y13 mV, a value that is within the range of the experimentally estimated IPSC reversal potential. These currents, which were mediated by GABA released from spontaneously firing neurons connected to the neuron under study, are herein referred to as inhibitory postsynaptic currents ŽIPSCs.. The other type consisted of currents that decayed
Fig. 3. Pb 2q does not affect fast current transients induced by step depolarization of hippocampal neurons. In the continuous presence of the Ca2q-channel blocker Cd 2q Ž10 mM., fast current transients were elicited by step depolarization of hippocampal neurons from y50 to q40 mV. Perfusion of these neurons with external solution containing Pb 2q Ž100 nM. affected neither the amplitude nor the decay phase of the current transients. Traces represent the average of 20–25 recordings obtained from each of three neurons prior to and during their exposure to Pb 2q.
26
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
3.2. Pb 2 q blocks the TTX-sensitiÕe release of glutamate and GABA from spontaneously firing hippocampal neurons in culture Following 2–3 s perfusion of the hippocampal neurons with Pb 2q Ž10 nM–1 mM.-containing external solution, there was a substantial reduction of the frequency and amplitude of both IPSCs and EPSCs ŽFig. 2.. The fact that there was a very short latency Ž- 3 s. for the onset of the effects of Pb 2q on spontaneously occurring EPSCs and IPSCs suggests that Ži. such effects are mediated by the interactions of this heavy metal with an extracellular site, and Žii. there is a rapid equilibrium between Pb 2q and its site of action. The blockade by Pb 2q of EPSCs and IPSCs was concentration dependent, and at 1 mM Pb 2q completely blocked these currents ŽFig. 2.. This effect of Pb 2q, which had the same magnitude regardless of the number of days
the neurons were in culture, was due to changes in transmitter release, because evidence has been provided that at the concentrations tested, this heavy metal does not affect the activity of the postsynaptic GABA A and AMPA receptors w4,12,52x. 3.3. Pb 2 q-induced reduction of the frequency and amplitude of spontaneously occurring EPSCs and IPSCs is not due to changes in Na q-channel actiÕity To ascertain whether Pb 2q could block Naq channels, and, consequently, the action potentials that trigger the release of GABA and glutamate, current transients evoked by depolarizing steps were recorded in the presence and in the absence of Pb 2q. To abolish the Ca2q component from these transients, the neurons were continuously perfused with external solution containing the Ca2q-channel blocker Cd 2q Ž10 mM..
Fig. 4. Characterization of field stimulation-induced EPSCs and IPSCs. Glutamate-mediated EPSCs or GABA-mediated IPSCs were recorded from hippocampal neurons continuously perfused with external solution containing the GABA A receptor blocker picrotoxin Ž100 mM. or the AMPA receptor antagonist CNQX Ž10 mM., respectively. Right traces: Samples of evoked EPSCs and IPSCs recorded at various membrane potentials. Notice that evoked EPSCs reverse at approximately 0 mV, whereas evoked IPSCs reverse at about y10 mV. Left traces: Samples of evoked EPSCs Žtop three traces. and evoked IPSCs Žbottom three traces. recorded before, during, and after perfusion of the hippocampal neurons with Cd 2q-containing external solution. The blocking effect of Cd 2q on the evoked EPSCs and IPSCs confirms the Ca2q dependence of the release process. Left graphs: Plots of the amplitude of the evoked EPSCs and IPSCs vs. recording time indicates that there is no rundown of these currents during the time of the experiments. In one neuron, EPSCs were evoked every 10 s, and the amplitudes of all currents recorded during 5 min were averaged and plotted against the recording time. Results for the IPSCs were obtained from another neuron. In each graph, each point is the average of 30 events.
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
27
Pb 2q ŽG 100 nM. affected neither the amplitude nor the decay and rising phases of current transients triggered by depolarizing steps of q50 mV applied to neurons held at y70 mV ŽFig. 3.. This result suggests that Pb 2q-induced changes in TTX-sensitive transmitter release may not be due to alterations in Naq-channel activity. It also suggests that back-propagating action potentials Žviewed as fast current transients in neurons held at potentials more negative than y40 mV. were not detected in the presence of Pb 2q Žsee Fig. 2. because of the blockade of transmitter release rather than because of a direct action of the heavy metal on the Naq-channel activity. The possibility that Naq channels present in the nerve terminals have a higher sensitivity to Pb 2q cannot be disregarded at this point. 3.4. Pharmacological and kinetic characterization of the release of glutamate and GABA eÕoked by field stimulation of hippocampal neurons in culture Application of a 50-ms, 5-V pulse via an electroplated bipolar platinum electrode to a neuron synaptically connected to the neuron under study always resulted in the activation of a PSC that was mediated by either glutamate or GABA. In each experiment, the evoked PSCs had variable amplitudes throughout the recording time, but did not show any sign of rundown ŽFig. 4.. In agreement with the concept that the PSCs were the consequence of evoked release of neurotransmitters was the finding that these currents could not be detected in the presence of the Ca2q channel blocker Cd 2q Ž10 mM . ŽFig. 4.. Field stimulation-evoked EPSCs and IPSCs could be distinguished from one another on the basis of one of the following criteria: Ži. EPSCs were sensitive to blockade by CNQX Ž10 mM. and IPSCs were sensitive to inhibition by picrotoxin Ž100 mM.; Žii. the decay-time constants of the EPSCs and IPSCs were 12.0 " 7.6 and 48.0 " 12.7 ms, respectively ŽFig. 4.; and Žiii. the zero-current potential for EPSCs and IPSCs were 0 and y20 mV, respectively ŽFig. 4.. To study the effects of Pb 2q on the evoked release of glutamate and GABA, EPSCs and IPSCs were pharmacologically isolated; evoked EPSCs were recorded from neurons that were perfused with picrotoxin Ž100 mM. and evoked IPSCs were recorded from neurons that were perfused with CNQX Ž10 mM.. In addition, quantification of the effects of Pb 2q on evoked transmitter release was made possible by carrying out all the experiments at positive membrane potentials so that fast current transients representing back-propagating action potentials would not contaminate the PSCs. 3.5. Pb 2 q at low nanomolar concentrations reduces the eÕoked release of glutamate from cultured hippocampal neurons Following 2–3 s perfusion of the hippocampal neurons with external solution containing Pb 2q Ž10 nM–10 mM., the amplitude of evoked EPSCs was reduced ŽFig. 5A..
Fig. 5. Concentration-dependent effect of Pb 2q on evoked EPSCs. Top traces: Samples of evoked EPSCs recorded before, during, and after exposure of two different hippocampal neurons to external solution containing Pb 2q Ž30 or 100 nM.. Membrane potential, q40 mV. Bottom graph: The amplitude of evoked EPSCs recorded in the presence of Pb 2q is expressed as a percentage of the amplitude of the currents recorded in the absence of the heavy metal and is plotted against the test concentrations of Pb 2q. Fit of the data points indicates that the IC 50 for Pb 2q-induced blockade of evoked EPSCs is approximately 68 nM. Each point and bar represents the mean"S.E. of results obtained from at least four neurons ŽU p- 0.05; UU p- 0.01..
This effect of Pb 2q, which was concentration dependent ŽFig. 5B., was only partially reversible upon washing of the neurons with nominally Pb 2q-free external solution ŽFig. 5A.. To determine the IC 50 for Pb 2q in decreasing the amplitude of the evoked EPSCs, single neurons were exposed a single time to a given concentration of Pb 2q. A four-parameter logistic equation was used to fit the con-
28
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
centration–response relationship Žsee Material and Methods ., and the IC 50 and Hill coefficient derived from the best fitting were 67.6 " 2.3 mM and 0.81 " 0.02, respectively ŽFig. 5B.. The ability of Pb 2q to decrease the amplitude of evoked EPSCs was certainly due to changes in the evoked release of glutamate, because, as stated above, the postsynaptic AMPA receptors are insensitive to the tested concentrations of Pb 2q. In agreement with this concept was the finding that Pb 2q at 100 nM, a concentration that reduces by more than 50% the amplitude of the EPSCs, had no effect on the rise time and decay-time constant of these currents ŽFig. 6.. 3.6. Pb 2 q at low nanomolar concentrations reduces the eÕoked release of GABA from cultured hippocampal neurons Pb 2q Ž10 nM–10 mM. also reduced the amplitude of evoked IPSCs ŽFig. 7. without affecting their kinetics of activation and inactivation ŽFig. 8.. This effect of Pb 2q was concentration dependent and was detected 2–3 s after the neurons were exposed to this heavy metal. When the neurons were superfused with nominally-Pb 2q free external solution, a partial recovery of the amplitude of the evoked IPSCs was observed Žsee Fig. 7.. The IC 50 and Hill coefficient estimated from the best fit of the concentration–response curve were 66.07 " 8.16 mM and 0.80 " 0.08, respectively. The fact that at the concentrations tested Pb 2q does not affect the activity of GABA A receptors w4,12,52x and the finding that Pb 2q did not alter the rise
time and the decay-time constants of the evoked IPSCs ŽFig. 8., indicates that the Pb 2q-induced decrease in the IPSC amplitude is attributable to a reduction of evoked release of GABA. 3.7. Full reÕersal of Pb 2 q-induced block in eÕoked transmitter release requires the perfusion of the neurons with EDTA In an attempt to determine whether Pb 2q-induced blockade of field stimulation-evoked transmitter release is mediated by the interaction of this heavy metal with an intracellular or an extracellular target, we analyzed the ability of the chelator EDTA to reverse the effect of Pb 2q. As described above, at 100 nM Pb 2q reduces by more than 50% the mean amplitude of evoked EPSCs, and some of this effect remains even at 30 min after washout of the neurons with nominally Pb 2q-free external solution. At this point, perfusion of the neurons with EDTA Ž200 mM.-containing external solution results in the full recovery of the amplitude of the evoked EPSCs ŽFig. 9.. The same was observed with the evoked IPSCs. Given that EDTA is a positively charged compound, it would not permeate the membrane of the neurons. Thus, it is feasible to hypothesize that EDTA, by chelating Pb 2q extracellularly, removes it from its site of action. 3.8. ReÕersal by the K q-channel blocker 4-aminopyridine of Pb 2 q-induced blockade of eÕoked transmitter release After blocking with Pb 2q either partially or completely the field stimulation-evoked release of GABA and gluta-
Fig. 6. Pb 2q has no effect on the rise time and on decay-time constant of evoked EPSCs. Left histograms: Distribution of the rise times and decay-time constants of evoked EPSCs recorded before and during a 20-min exposure of a hippocampal neuron to external solution containing Pb 2q Ž100 nM.. Right graph: The rise time and the decay-time constants of evoked EPSCs recorded in the absence of Pb 2q were defined as control and taken as 100%. Then, the rise time and the decay-time constants of currents recorded in the presence of Pb 2q were calculated as a percentage of control values. Each vertical bar and error bar represents the mean " S.E. of results obtained from five neurons.
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
29
4. Discussion 4.1. Pb 2 q inhibits the action potential-triggered, Ca 2 qdependent release of neurotransmitters from hippocampal neurons
Fig. 7. Concentration-dependent effect of Pb 2q on evoked IPSCs. Top traces: Samples of evoked IPSCs recorded before, during, and after exposure of two different hippocampal neurons to external solution containing Pb 2q Ž30 or 100 nM.. Membrane potential, q40 mV. Bottom graph: The amplitude of evoked IPSCs recorded in the presence of Pb 2q is expressed as a percentage of the amplitude of the currents recorded in the absence of the heavy metal and is plotted against the test concentrations of Pb 2q. Fit of the data points indicates that the IC 50 for Pb 2q-induced blockade of evoked IPSCs is approximately 66 nM. Each point and bar represents the mean"S.E. of results obtained from at least five neurons ŽUU indicates p- 0.01..
mate, perfusion of the neurons with external solution containing both Pb 2q and the Kq-channel blocker 4-aminopyridine Ž10 mM. resulted in the recovery of the amplitude of the evoked EPSCs and IPSCs. This finding suggested that Pb 2q did not deplete the pool of vesicles available for release. In the absence of changes on voltage-gated Naq channel activity, the inhibition by Pb 2q of evoked transmitter release, being overcome by the 4-aminopyridine-induced prolongation of action potentials, could be explained by an interaction of the heavy metal with voltage-gated Ca2q channels.
In the present study, we demonstrate that at low nanomolar concentrations Pb 2q inhibits the action potential-dependent release of GABA and glutamate from hippocampal neurons, as evidenced by the fact that this heavy metal reduces the amplitude and frequency of spontaneously occurring EPSCs and IPSCs as well as the amplitude of field stimulation-evoked EPSCs and IPSCs. Of interest, much higher Žmicromolar. concentrations of Pb 2q are required to block nerve-evoked ACh release at the neuromuscular junction w7,21,32x, and there has been considerable controversy regarding the effects of this heavy metal on the action potential-dependent release of transmitters in the CNS. In some neurochemical studies carried out in brain synaptosomes, Pb 2q ŽG 1 mM. was shown to inhibit Kq-induced release of transmitters w33,46x. In others, this heavy metal was shown to increase or to have no effect on the Kq-induced release of transmitters w41,45x. These conflicting results have been attributed to methodological problems associated with the synaptosomes w34x. The purity of the preparation, the timing between exposure of the synaptosomes to high Kq and to Pb 2q, and the time the preparation was exposed to Pb 2q could contribute to such conflicting findings, because Ži. neuronal organelles can serve as a sink for Pb 2q w44x; Žii. the effectiveness of heavy metals in blocking depolarization-dependent Ca2q influx can be affected by the duration of the depolarizing stimulus w36,37x; and Žiii. a prolonged exposure of the neuronal preparation to Pb 2q results in an increase of spontaneous release w12,34,35x that can mask or outweigh a possible inhibitory effect of the heavy metal on the evoked release of transmitters. Furthermore, in some instances, substances that alter nerve-evoked transmitter release do not affect Kq-induced release of neurotransmitters w13x. Thus, a direct analysis of specific release processes became necessary to elucidate the actions of this heavy metal in CNS synapses and to determine the effective concentrations at which Pb 2q alters such processes. Electrophysiologically, at least two different transmitter release processes can be studied in cultured neurons or in neurons in brain slices. One process is not dependent on the propagation of action potentials, and is, therefore, insensitive to blockade by TTX. The other process, which is Ca2q dependent, is sensitive to blockade by TTX, because it also depends on the propagation of action potentials. The TTX-sensitive, Ca2q-dependent release of neurotransmitters can be triggered by the spontaneous, asynchronous firing of presynaptic neurons or by field stimulation of presynaptic neurons. Taking into account
30
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
Fig. 8. Pb 2q has no effect on the rise time and on decay-time constant of evoked IPSCs. Left histograms: Distribution of the rise times and decay-time constants of evoked IPSCs recorded before and during exposure of a hippocampal neuron to external solution containing Pb 2q Ž100 nM.. Right graph: The rise time and the decay-time constants of evoked IPSCs recorded in the absence of Pb 2q were defined as control and taken as 100%. Then, the rise time and the decay-time constants of currents recorded in the presence of Pb 2q were calculated as percentage of control values. Each vertical bar and error bar represents the mean " S.E. of results obtained from five neurons.
that these processes can be studied independently, it becomes possible to determine with more certainty the effects and the effective concentrations of a given substance on a specific release process. Thus, although our present findings are in agreement with previous reports that Pb 2q blocks the action potential-dependent transmitter release in the CNS w33–35,41,46x, they demonstrate for the first time that the effective concentrations of this heavy metal in blocking this release process are in the nanomolar range, i.e., well below those detected in the neurochemical studies. 4.2. Sites of action of Pb 2 q in inhibiting the action potential-triggered, Ca 2 q-dependent release of GABA and glutamate from hippocampal neurons The finding that there was a very short latency for the onset of the Pb 2q-induced inhibition of evoked and spontaneously occurring EPSCs and IPSCs indicated an interaction of Pb 2q with extracellular targets. In accordance with this concept was the fact that the effect of Pb 2q on evoked transmitter release was fully reversed by the chelator EDTA, which, being positively charged, poorly crosses the neuronal membrane. Recent studies have indicated that Pb 2q-sensitive receptors such as the nicotinic receptors and the NMDA receptors w4,25,28,38,52x, when present on presynaptic neurons, control the action potential-dependent release of transmitters w5,40x. However, Pb 2q-induced inhibition of transmitter release was observed even in the
presence of specific antagonists of these receptors, and, therefore, could not be accounted for by the interaction of the heavy metal with such receptors. The concentrations at which Pb 2q blocked GABA and glutamate release were very similar, suggesting that the heavy metal was probably acting via a mechanism that is common to the release of neurotransmitters. Such mechanisms could be related, albeit not restricted, to activation of Naq, Kq, and Ca2q channels. At the concentrations that affected transmitter release, Pb 2q did not alter the Naqchannel activity in hippocampal neurons. In fact, previous studies carried out in human neuroblastoma cells had demonstrated that up to 10 mM Pb 2q has no significant effects on either Naq- or Kq-channel currents w42x. Thus, the finding that the Kq-channel blocker 4-aminopyridine reversed the Pb 2q-induced reduction of evoked transmitter release led to the hypothesis that this effect of Pb 2q is mediated by its ability to block Ca2q channels ŽFig. 10.. In rat hippocampal neurons and in several other preparations, Pb 2q blocks both L- and N-type Ca2q channels with IC 50 s of approximately 30 and 80 nM, respectively, and both of these channels are apparently involved in controlling transmitter release from CNS neurons w9,14– 17,24x. Therefore, the finding that Pb 2q inhibited the evoked release of GABA and glutamate with an IC 50 of about 68 nM suggests an action of this heavy metal on N-type Ca2q channels. Further supporting the concept that Pb 2q inhibits the evoked release of neurotransmitters by blocking Ca2q channels is the fact that full reversal of
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
31
Fig. 9. Full reversal of Pb 2q-induced blockade of evoked EPSCs requires treatment of the neurons with EDTA. Left traces: Samples of evoked EPSCs recorded from a hippocampal neuron that was subjected to the following consecutive treatments: Ži. perfusion with external solution Žcontrol., Žii. perfusion with Pb 2q Ž100 nM.-containing external solution, Žiii. perfusion with nominally Pb 2q-free external solution ŽWashout., and Živ. perfusion with nominally Pb 2q-free external solution containing EDTA Ž200 mM.. Graph: The amplitude of the evoked EPSCs recorded under the control condition was taken as 100% and used to normalize the amplitude of the currents evoked under each of the other experimental conditions. Each vertical bar and error bar represents the mean " S.E. of results obtained from four neurons. Notice that a full reversal of Pb 2q-induced blockade of evoked EPSCs requires the perfusion of the neurons with EDTA. ŽU p - 0.05; UU p - 0.01..
Pb 2q-induced blockade of Ca2q channels requires treatment of the neurons with chelators w11x, and the same treatment was needed to fully reverse the effects of Pb 2q on evoked transmitter release Žsee Fig. 9.. 4.3. Toxicological releÕance of the actions of Pb 2 q in transmitter release from hippocampal neurons The toxicological relevance of our current findings that Pb 2q at an extracellular concentration as low as 10 nM blocks the action potential triggered-, Ca2q-dependent release of both GABA and glutamate is emphasized by the fact that concentrations of Pb 2q ranging from 25–100 nM have been reported to be present in the cerebrospinal fluid of humans not known to be occupationally exposed to Pb 2q w18x. There are several means by which alterations in transmitter release can account for the impairment in cognitive functions that have been reported in patients, particularly in children chronically intoxicated with Pb 2q Že.g., Ref. w39x.. First, Pb 2q-induced changes of synaptic activity could alter synaptic organization in the hippocampus, and,
consequently, impair normal brain development. In fact, blockade of Ca2q-dependent transmitter release has been shown to delay neuronal differentiation w53x, and both spontaneously generated and experience-dependent neuronal activity are critical for the establishment of functional cortical circuits during development and throughout life w29x. Thus, the dual effect of Pb 2q on the different transmitter release processes, i.e., the enhancement of TTX-insensitive spontaneous synaptic activity w12x and the reduction of TTX-sensitive, action potential-dependent synaptic activity Žthis paper., could account for the highly deleterious effects of this heavy metal in the developing brain. Secondly, long-term potentiation ŽLTP., a phenomenon considered as the neuronal substrate for memory and learning Žreviewed in Ref. w48x., is impaired by Pb 2q w6,19,30x. The toxicological relevance of Pb 2q-induced blockade of LTP was initially questioned, because in vitro such an effect was detected only in the presence of micromolar concentrations of the heavy metal w6,19x. The debate was settled after the demonstration that LTP is in fact impaired in hippocampal slices taken from rats chronically exposed to low levels of Pb 2q w30x. Attempts to determine
32
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
Fig. 10. The Kq-channel blocker 4-aminopyridine reverses Pb 2q-induced blockade of evoked EPSCs and IPSCs. Left traces: Samples of evoked EPSCs Žtop three traces. and IPSCs Žbottom three traces. recorded from two different hippocampal neurons subjected to the following consecutive treatments: Ži. perfusion with external solution ŽControl., Žii. perfusion with external solution containing Pb 2q Ž100 nM., and Žiii. perfusion with external solution containing the admixture of Pb 2q Ž100 nM. plus 4-aminopyridine Ž10 mM.. Membrane potential, q40 mV. Right graphs: Amplitude of evoked EPSCs and IPSCs recorded under control condition was taken as 100%, and the amplitude of currents recorded in the presence of Pb 2q and Pb 2q plus 4-aminopyridine is presented as a percentage of control. Each vertical bar and error bar represents the mean" S.E. of results obtained from three neurons. Notice that 4-aminopyridine fully reverses Pb 2q-induced blockade of the evoked EPSCs and IPSCs ŽUU p - 0.01..
the mechanism by which Pb 2q blocks LTP have led to the conclusion that this effect is not mediated by the direct action of the heavy metal on NMDA receptors w27x. Considering that LTP is apparently associated with both a postsynaptic and a presynaptic mechanism, and that en-
hancement of presynaptic transmitter release accounts in part for the expression of this phenomenon w31,51x, we suggest that Pb 2q-induced blockade of the action potential-dependent release of glutamate from hippocampal neurons could underlie the blockade by this heavy
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
metal of LTP, and, consequently, its ability to impair cognitive functions. In conclusion, the present study demonstrates that at low nanomolar concentrations Pb 2q inhibits the action potential-dependent release of neurotransmitters from hippocampal neurons, primarily by apparently blocking presynaptic Ca2q channels, thus improving our understanding of the mechanisms underlying the neurotoxic effects of this heavy metal. Further, the finding that full reversal of the Pb 2q-induced blockade of evoked transmitter release requires the treatment of the neurons with EDTA indicates that centrally-acting chelators may be therapeutically useful for the treatment andror prevention of some of the toxic effects of this heavy metal. Acknowledgements The superb technical assistance of Ms. Mabel Zelle, Ms. Barbara Marrow, and Mr. Benjamin Cumming is gratefully acknowledged. This study was supported by USPHS grant ES05730 and by PRONEX Žfrom Brazil. Žfor EXA., by a CNPq fellowship from Brazil Žfor MFMB., and by a minority fellowship under USPHS grant ES05730 Žfor EFRP.. References w1x D.P. Alfano, T.L. Petit, Behavioral effects of postnatal lead exposure: possible relationship to hippocampal dysfunction, Behav. Neural Biol. 32 Ž1981. 319–333. w2x D.P. Alfano, T.L. Petit, Neonatal lead exposure alters the dendritic development of hippocampus dentate granule cells, Exp. Neurol. 75 Ž1982. 275–288. w3x M. Alkondon, E.X. Albuquerque, Diversity of nicotinic acetylcholine receptors in rat hippocampal neurons: I. Pharmacological and functional evidence for distinct structural subtypes, J. Pharmacol. Exp. Ther. 265 Ž1993. 1455–1473. w4x M. Alkondon, A.C.S. Costa, V. Radhakrishnan, R.S. Aronstam, E.X. Albuquerque, Selective blockade of NMDA-activated channel currents may be implicated in learning deficits caused by lead, FEBS Lett. 261 Ž1990. 124–130. w5x M. Alkondon, E.F.R. Pereira, C.T.F. Barbosa, E.X. Albuquerque, Neuronal nicotinic acetylcholine receptor activation modulates gaminobutyric acid release from CA1 neurons of rat hippocampal slices, J. Pharmacol. Exp. Ther. 283 Ž1997. 1396–1411. w6x L. Altmann, K. Sveinsson, H. Wiegand, Long-term potentiation in rat hippocampal slices is impaired following lead perfusion, Neurosci. Lett. 128 Ž1991. 109–112. w7x W.D. Atchison, T. Narahashi, Mechanism of action of lead on neuromuscular junctions, NeuroToxicology 5 Ž1984. 267–282. w8x G. Audesirk, Effects of lead exposure on the physiology of neurons, Prog. Neurobiol. 24 Ž1985. 199–231. w9x G. Audesirk, T. Audesirk, The effects of inorganic lead on voltagesensitive calcium channels differ among cell types and among channel subtypes, NeuroToxicology 14 Ž1993. 137–147. w10x D. Bellinger, K.N. Dietrich, Low-level lead exposure and cognitive function in children, Pediatr. Ann. 23 Ž1994. 600–605. w11x J. Bernal, J.-H. Lee, L.L. Cribbs, E. Perez-Reyer, Full reversal of Pb 2q block of L-type Ca2q channels requires treatment with heavy metal antidotes, J. Pharmacol. Exp. Ther. 282 Ž1997. 172–180.
33
w12x M.F.M. Braga, E.F.R. Pereira, M. Marchioro, E.X. Albuquerque, Lead increases tetrodotoxin-insensitve spontaneous release of glutamate and GABA from hippocampal neurons, Brain Res. Ž1998. submitted. w13x P.L. Buckle, I. Spence, The actions of adenosine and some analogues on evoked and potassium stimulated release at skeletal and autonomic neuromuscular junctions, Naunyn-Schmiedeberg’s Arch. Pharmacol. 319 Ž1982. 130–135. w14x D. Busselberg, M.L. Evans, H. Rahmann, D.O. Carpenter, Lead ¨ inhibits the voltage-activated calcium currents of Aplysia neurons, Toxicol. Lett. 52 Ž1990. 51–57. w15x D. Busselberg, M.L. Evans, H. Rahmann, D.O. Carpenter, Lead and ¨ zinc block a voltage-activated calcium channel of Aplysia neurons, J. Neurophysiol. 65 Ž1991. 786–795. w16x D. Busselberg, M.L. Evans, H.L. Haas, D.O. Carpenter, Blockade of ¨ mammalian and invertebrate calcium channels by lead, NeuroToxicology 14 Ž1993. 249–258. w17x D. Busselberg, B. Platt, D. Michael, H.L. Haas, D.O. Carpenter, ¨ Mammalian voltage-activated calcium channel currents are blocked by Pb 2q, Zn2q and Al 3q, J. Neurophysiol. 71 Ž1994. 1491–1497. w18x A. Cavalleri, C. Minoia, M. Ceroni, M. Poloni, Lead in cerebrospinal fluid and its relationship to plasma lead in humans, J. Appl. Toxicol. 4 Ž1984. 63–65. w19x D.O. Carpenter, M.R. Matthews, P.J. Parsons, N. Hori, Long-term potentiation in the piriform cortex is blocked by lead, Cell. Mol. Neurobiol. 14 Ž1994. 723–733. w20x G.P. Cooper, R.S. Manalis, Influence of heavy metals on synaptic transmission: a review, NeuroToxicology 4 Ž1983. 69–84. w21x G.P. Cooper, R.S. Manalis, Interactions of lead and cadmium on acetylcholine release at the frog neuromuscular junction, Toxicol. Appl. Pharmacol. 74 Ž1984. 411–416. w22x G.P. Cooper, J.B. Suszkiw, R.S. Manalis, in: T. Narahashi ŽEd.., Cellular and Molecular Neurotoxicology, Presynaptic Effects of Heavy Metals, Raven Press, New York, 1984, pp. 1–21. w23x J. Dempster, The computer analysis of eletrophysiological data, in: P.J. Frazer ŽEd.., Microcomputers in Physiology: A Pratical Approach, Oxford IRl Press, Oxford, 1989, pp. 51–93. w24x M.L. Evans, D. Busselberg, D.O. Carpenter, Pb 2q blocks calcium ¨ currents of cultured dorsal root ganglion cells, Neurosci. Lett. 129 Ž1991. 103–106. w25x T.R. Guilarte, R.C. Miceli, Age-dependent effects of lead on w3 HxMK-801 binding to the NMDA receptor-gated ionophore: in vitro and in vivo studies, Neurosci. Lett. 148 Ž1992. 27–30. w26x O.P. Hamill, A. Marty, E. Neher, B. Sakmann, F.J. Sigworth, Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches, Pflug. Arch. 391 Ž1981. 85–100. w27x N. Hori, D. Busselberg, M.R. Matthews, P.J. Parsons, D.O. Carpen¨ ter, Lead blocks LTP by an action not at NMDA receptors, Exp. Neurol. 119 Ž1993. 192–197. w28x K. Ishihara, M. Alkondon, J.G. Montes, E.X. Albuquerque, Nicotinic responses in acutely dissociated rat hippocampal neurons and the selective blockade of fast-desensitizing nicotinic currents by lead, J. Pharmacol. Exp. Ther. 273 Ž1995. 1471–1482. w29x L.C. Katz, C.J. Shatz, Synaptic activity and the construction of cortical circuits, Science ŽWashington, DC. 274 Ž1996. 1133–1138. w30x S.M. Lasley, J. Polan-Curtin, D.L. Armstrong, Chronic exposure to environmental levels of lead impairs in vivo induction of long-term potentiation in rat hippocampal dentate, Brain Res. 614 Ž1993. 347–351. w31x A. Malgaroli, A.E. Ting, B. Wendland, A. Bergamaschi, A. Villa, R.W. Tsien, R.H. Scheller, Presynaptic component of long-term potentiation visualized at individual hippocampal synapses, Science ŽWashington, DC. 268 Ž1995. 1624–1628. w32x R.S. Manalis, G.P. Cooper, Presynaptic and postsynaptic effects of lead at the frog neuromuscular junction, Nature ŽLond.. 243 Ž1973. 354–355.
34
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
w33x D.J. Minnema, R.D. Greenland, I.A. Michaelson, Effect of in vitro inorganic lead on dopamine release from superfused rat striatal synaptosomes, Toxicol. Appl. Pharmacol. 84 Ž1986. 400–411. w34x D.J. Minnema, I.A. Michaelson, Differential effects of inorganic lead and d-aminolevulinic acid in vitro on synaptosomal g-aminobutyric acid release, Toxicol. Appl. Pharmacol. 86 Ž1986. 437–447. w35x D.J. Minnema, I.A. Michaelson, G.P. Cooper, Calcium efflux and neurotransmitter release from rat hippocampal synaptosomes exposed to lead, Toxicol. Appl. Pharmacol. 92 Ž1988. 351–357. w36x D.A. Nachshen, M.P. Blaustein, Some properties of potassiumstimulated calcium influx in presynaptic nerve endings, J. Gen. Physiol. 76 Ž1980. 709–728. w37x D.A. Nachshen, M.P. Blaustein, Influx of calcium, strontium, and barium in presynaptic nerve endings, J. Gen. Physiol. 79 Ž1982. 1065–1087. w38x K. Nagata, C.S. Huang, J.H. Songand, T. Narahashi, Lead modulation of the neuronal nicotinic acetylcholine receptor in PC12 cells, Brain Res. 754 Ž1997. 21–27. w39x H.L. Needleman, A. Schell, D. Bellinger, A. Leviton, E.N. Allred, The long-term effects of exposure to low doses of lead in childhood, N. Engl. J. Med. 322 Ž1990. 83–88. w40x K. Ohta, N. Arake, M. Shibata, S. Komatsumoto, K. Shimazu, Y. Fukuchi, Presynaptic ionotropic glutamate receptors modulate in vivo release and metabolism of striatal dopamine, noradrenaline, and 5-hydroxytryptamine: Involvement of both NMDA and AMPArkainate subtypes, Neurosci. Res. 21 Ž1994. 83–89. w41x P.B. Ramsay, M.R. Krigman, P. Morell, Developmental studies of the uptake of choline, GABA and dopamine by crude synaptosomal preparations after in vivo and in vitro lead treatment, Brain Res. 187 Ž1980. 383–402. w42x E. Reuveny, T. Narahashi, Potent blocking action of lead on voltage-activated calcium channels in human neuroblastoma cells SH-SY5Y, Brain Res. 545 Ž1991. 312–314.
w43x Z. Shao, J.B. Suszkiw, Ca2q-surrogate action of Pb 2q on acetylcholine release from rat brain synaptosomes, J. Neurochem. 56 Ž1991. 568–574. w44x E.K. Silbergeld, H.S. Adler, J.L. Costa, Subcellular localization of lead in synaptosomes, Res. Commun. Chem. Pathol. Pharmacol. 17 Ž1977. 715–725. w45x E.K. Silbergeld, R.E. Hruska, L.P. Miller, N. Eng, Effects of lead in vivo and in vitro on GABAergic neurochemistry, J. Neurochem. 34 Ž1980. 1712–1718. w46x J. Suszkiw, G. Toth, M. Murawsky, G. Cooper, Effects of Pb 2q and Cd 2q on acetylcholine release and Ca2q movements in synaptosomes subcellular fractions from rat brain and torpedo electric organ, Brain Res. 323 Ž1984. 31–46. w47x T. Sutula, X.X. He, J. Cavazos, G. Scott, Synaptic reorganization in the hippocampus induced by abnormal functional activity, Science ŽWashington, DC. 239 Ž1988. 1147–1150. w48x T.J. Teyler, P. Discenna, Long-term potentiation, Annu. Rev. Neurosci. 10 Ž1987. 131–161. w49x J.L. Tomsig, J.B. Suszkiw, Pb 2q-induced secretion from bovine chromaffin cells: fura-2 as probe for Pb 2q, Am. J. Physiol. 259 Ž1990. C762–C768. w50x J.L. Tomsig, J.B. Suszkiw, Intracellular mechanism of Pb 2q-induced norepinephrine release from bovine chromaffin cells, Am. J. Physiol. 265 Ž1993. C1630–C1636. w51x R.W. Tsien, R. Malinow, Changes in presynaptic function during long-term potentiation, Ann. New York Acad. Sci. 635 Ž1991. 208–220. w52x H. Ujihara, E.X. Albuquerque, Developmental change of the inhibition by lead of NMDA-activated currents in cultured hippocampal neurons, J. Pharmacol. Exp. Ther. 263 Ž1992. 868–875. w53x H. Xie, L. Ziskind-Conhaim, Blocking Ca2q-dependent synaptic release delays motoneuron differentiation in the rat spinal cord, J. Neurosci. 15 Ž1995. 5900–5911.
Research report
Nanomolar concentrations of lead inhibit glutamatergic and GABAergic transmission in hippocampal neurons Maria F.M. Braga b
a,b
, Edna F.R. Pereira a , Edson X. Albuquerque
a,b,)
a Department of Pharmacology and Experimental Therapeutics, UniÕersity of Maryland School of Medicine, Baltimore, MD 21201, USA Department of Basic and Clinical Pharmacology, Institute of Biomedical Sciences, Center of Health Sciences, Federal UniÕersity of Rio de Janeiro, Rio de Janeiro, RJ 21944, Brazil
Accepted 26 January 1999
Abstract To investigate whether lead ŽPb 2q . affects the tetrodotoxin ŽTTX.-sensitive release of neurotransmitters, the whole-cell mode of the patch-clamp technique was applied to cultured hippocampal neurons. Pb 2q ŽG 10 nM. reversibly blocked the TTX-sensitive release of glutamate and g-aminobutyric acid ŽGABA., as evidenced by the reduction of the amplitude and frequency of glutamate- and GABA-mediated postsynaptic currents ŽPSCs. evoked by spontaneous neuronal firing. This effect of Pb 2q, which occurred 2–3 s after exposure of the neurons to Pb 2q-containing external solution, was not related to changes in Naq-channel activity, and was quantified by measurements of changes in the amplitude of PSCs evoked when a 50-ms, 5-V stimulus was applied via a bipolar electrode to a neuron synaptically connected to the neuron under study. With an IC 50 of approximately 68 nM, Pb 2q blocked the evoked release of glutamate and GABA. This effect was most likely mediated by Pb 2q ’s actions on extracellular targets, because there was a very short delay Ž- 3 s. for its onset, and it could be completely reversed by the chelator ethylene diaminetetraacetic acid ŽEDTA.. Given that Pb 2q-induced blockade of evoked transmitter release could be reversed by 4-aminopyridine, it is suggested that the effect on release was mediated via the binding of Pb 2q to voltage-gated Ca2q channels. Thus, it is most likely that the neurotoxic effects of Pb 2q in the mammalian brain involve a decrease of the TTX-sensitive, Ca2q-dependent release of neurotransmitters. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Hippocampus; Pb 2q; Electrophysiology; GABA; Glutamate; Transmitter release
1. Introduction The poor understanding of the mechanisms underlying the neurotoxic effects of lead ŽPb 2q ., a pollutant commonly found in the environment, has hindered the development of efficacious therapeutic means by which Pb 2q-induced neurotoxicity could be prevented andror treated. The higher sensitivity of the developing brain to the toxic effects of Pb 2q could be due in part to cellular processes that occur during early stages of development and that are particularly sensitive to the actions of this environmental pollutant w1,2,10x. Thus, considering that the degree of neuronal activity is an important factor for the establishment of synaptic connections in the central nervous system ŽCNS. w47x, it has been suggested that, by interfering with )
Corresponding author. Department of Pharmacology and Experimental Therapeutics, University of Maryland School of Medicine, 655 West Baltimore Street, Baltimore, MD 21201, USA. Fax: q1-410-706-3991; E-mail: [email protected]
neuronal excitability at early stages of development, Pb 2q could alter synaptic plasticity, thereby causing impairment of learning and memory Žsee Ref. w12x.. Although numerous studies have investigated the effects of Pb 2q in transmitter release processes in the peripheral and central nervous systems w8,22,34,35,43,49,50x, very few have addressed the effects of Pb 2q on specific release processes w20,21,32x. A recent electrophysiological study has demonstrated that at submicromolar concentrations Pb 2q increases specifically the tetrodotoxin ŽTTX.-insensitive release of neurotransmitters from hippocampal neurons, an effect that is apparently mediated by the interaction of the heavy metal with targets within the presynaptic terminals w12x. Given that the establishment of functional synaptic connections in the CNS is dependent on the level of overall neuronal activity w29,53x, in the present study, by means of the patch-clamp technique, we have investigated the effects of Pb 2q on the action potential-dependent release of glutamate and GABA from hippocampal neurons in culture.
0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 1 1 9 4 - 4
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
The results presented herein demonstrate that Pb 2q blocks the action potential-dependent release of both GABA and glutamate with an IC 50 of approximately 68 nM, and that the magnitude of the effect of Pb 2q on evoked transmitter release is the same regardless of the stage of neuronal development in vitro. Our findings also indicate that Pb 2q-induced blockade of this transmitter release process Ži. is partially reversible upon washing of the neurons with nominally Pb 2q-free solution, Žii. requires the use of the chelator ethylene diaminetetraacetic acid ŽEDTA. to be fully reversed, and Žiii. can be accounted for by the blockade of voltage-gated Ca2q channels. The discovery that at very low concentrations ŽG 10 nM. Pb 2q reduces the action potential-dependent release
23
of neurotransmitters from hippocampal neurons strongly suggests that this may be one of the mechanisms underlying the cognitive deficits observed in patients, particularly children intoxicated with this heavy metal.
2. Materials and methods 2.1. Cultures of hippocampal neurons Cultures of hippocampal neurons were prepared by a procedure similar to that published elsewhere w3x. Neurons cultured for 10 to 30 days were used in this study.
Fig. 1. Characterization of spontaneously occurring EPSCs and IPSCs. Glutamate-mediated EPSCs or GABA-mediated IPSCs were recorded from hippocampal neurons continuously perfused with external solution containing the GABA A receptor blocker picrotoxin Ž100 mM. or the AMPA receptor antagonist CNQX Ž10 mM., respectively. Samples of spontaneously occurring EPSCs and IPSCs recorded at various membrane potentials are shown on the right. Notice that EPSCs reverse at approximately 0 mV, whereas IPSCs reverse between y10 and y20 mV. Note also that at membrane potentials more negative than y40 mV, EPSCs and IPSCs can be accompanied by fast current transients ŽU ., which were clipped from many sample recordings. The inset shows, in a different time and amplitude scale, the fast current transient accompanying the EPSC recorded at y60 mV and seen on the right side of the inset. Histograms on the left depict the distribution of the decay-time constants of EPSCs and IPSCs recorded from hippocampal neurons at q20 mV.
24
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
2.2. Postsynaptic currents (PSCs) PSCs were recorded from cultured hippocampal neurons according to the standard patch-clamp technique w26x through an LM-EPC-7 amplifier ŽList Electronics, Heidelberg, Germany.. The signals were filtered at 3 kHz, stored on VCR tapes, digitized at 50 ms and analyzed on an IBM-compatible microcomputer. A specially designed bipolar electrode was used to stimulate a neuron synaptically connected to the neuron under study. Application of a 50-ms, supramaximal stimulus Ž; 5 V. via this bipolar electrode, which was positioned approximately 100 mm from the presynaptic neuron, consistently resulted in the release of neurotransmitters and consequent activation of inhibitory ŽGABA-mediated. or excitatory Žglutamatemediated. PSCs. The bath solution used to perfuse the neurons at a rate of 2–4 mlrmin had the following composition Žin mM.: NaCl, 165; KCl, 5; CaCl 2 , 2;
glucose, 10; and N-w2-hydroxyethylxpiperazine-N X-w2ethane sulfonic acidx ŽHEPES., 5 ŽpH s 7.3 adjusted with NaOH; 340 mOsm.. The solution used to fill the patch pipettes had the following composition Žin mM.: CsCl, 160; Cs-ethylene-glycol bisŽb-amino-ethyl ether.-N, N X-tetraacetic acid, 10; and HEPES, 10 ŽpH s 7.3 adjusted with CsOH; 340 mOsm.. The patch microelectrodes were pulled from borosilicate capillary glass ŽWorld Precision Instruments, New Haven, CT. and when filled with the pipette solution had resistances in the range of 2–5 M V. All the experiments were performed at room temperature Ž22– 258C.. The test solutions were applied to the surface of the hippocampal neurons through an array of parallel glass tubes Ž400-mm i.d.. glued together and assembled on a Narishige micromanipulator. The time constant for solution exchange was 20 ms. These tubes were positioned at about 50–100 mm away from the neuron under study. Each tube was connected to a reservoir filled with the test
Fig. 2. Effects of different concentrations of Pb 2q on spontaneously occurring EPSCs and IPSCs. Samples of EPSCs and IPSCs recorded before, during, and after perfusion of hippocampal neurons with external solution containing Pb 2q Ž0.03, 0.1 or 1 mM.. Notice that both the amplitude and the frequency of the currents are reduced by Pb 2q, and that this effect is concentration dependent, being more intense with higher Pb 2q concentrations. ‘W’ indicates washing of the neurons with nominally Pb 2q-free external solution. Membrane potential, y50 mV.
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
solution and gravity fed at a rate of 1.2 mlrmin. A computer-controlled stepping motor was used to move the barrels, so that a single neuron could be exposed to different test solutions. 2.3. Drugs and toxins Dimethylsulfoxide, EDTA, PbCl 2 , picrotoxin, and TTX were purchased from Sigma ŽSt. Louis, MO.. CNQX was obtained from Research Biochemical International ŽNatick, MA. and from Sigma. A 250-mM stock solution of picrotoxin was made in dimethylsulfoxide, and dilutions were made in external solution. NaOH was used to dissolve CNQX Žthe 10 mM stock of CNQX had 12.2 mM NaOH.. All other chemicals were dissolved in double-distilled water. Solutions containing Pb 2q were prepared just before application to the neurons. 2.4. Data and statistical analysis The peak amplitude, 10–90% rise time, and decay-time constants of the PSCs were analyzed using a suite of SCAN-CDR programs w23x and also the pCLAMP program ŽAxon Instruments, CA.. Results are expressed as mean " S.E.M. The significance of differences was determined with Student’s t-test. To calculate IC 50 s and Hill coefficients, a four-parameter logistic function was used to fit the concentration–response curve by a non-linear leastsquares method: Effect s Ea q Ž Ei y Ea .rw1 q Ž KrD . h x, where: K s IC 50 , h s Hill coefficient, Ea s effect at zero drug concentration, Ei s effect at infinite drug concentration, and D s drug concentration.
25
with a time constant of 17.0 " 15.8 ms at q20 mV, had a reversal potential close to 0 mV, and were sensitive to inhibition by the AMPA receptor antagonist CNQX Ž10 mM. ŽFig. 1.. The CNQX-sensitive, rapidly decaying currents were mediated by glutamate released from spontaneously firing neurons synaptically connected to the neurons under study and are herein referred to as excitatory postsynaptic currents ŽEPSCs.. At membrane potentials more negative than y40 mV, fast current transients that represented back-propagating action potentials accompanied the EPSCs and the IPSCs Žsee Fig. 1.. These fast current transients were the consequence of depolarization induced by the neurotransmitters released from the presynaptic neurons onto unclamped areas of the neuron under study. It should be emphasized that fast current transients accompanying IPSCs were the result of depolarization caused by activation of GABA A receptors in unclamped neuronal areas where the membrane potential is more negative than the IPSC reversal potential. Fast current transients were not observed at membrane potentials more positive than y40 mV Žsee Fig. 1., because the inactivation of the Naq channels in the clamped areas blocked the propagation of the action potentials to the recording site.
3. Results 3.1. Pharmacological and kinetic characterization of PSCs mediated by glutamate and GABA released by spontaneous firing of hippocampal neurons in culture In the absence of the Naq-channel blocker TTX, two types of pharmacologically and kinetically distinct PSCs could be recorded from hippocampal neurons in culture. One type consisted of currents that decayed with a time constant of 53.5 " 16.1 ms at q20 mV, reversed at a membrane potential between y10 and y20 mV, and were sensitive to blockade by the GABA A receptor antagonist picrotoxin Ž100 mM. ŽFig. 1.. Considering the composition of the physiological solutions used in this study, the calculated Nernst potential for Cly is approximately y13 mV, a value that is within the range of the experimentally estimated IPSC reversal potential. These currents, which were mediated by GABA released from spontaneously firing neurons connected to the neuron under study, are herein referred to as inhibitory postsynaptic currents ŽIPSCs.. The other type consisted of currents that decayed
Fig. 3. Pb 2q does not affect fast current transients induced by step depolarization of hippocampal neurons. In the continuous presence of the Ca2q-channel blocker Cd 2q Ž10 mM., fast current transients were elicited by step depolarization of hippocampal neurons from y50 to q40 mV. Perfusion of these neurons with external solution containing Pb 2q Ž100 nM. affected neither the amplitude nor the decay phase of the current transients. Traces represent the average of 20–25 recordings obtained from each of three neurons prior to and during their exposure to Pb 2q.
26
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
3.2. Pb 2 q blocks the TTX-sensitiÕe release of glutamate and GABA from spontaneously firing hippocampal neurons in culture Following 2–3 s perfusion of the hippocampal neurons with Pb 2q Ž10 nM–1 mM.-containing external solution, there was a substantial reduction of the frequency and amplitude of both IPSCs and EPSCs ŽFig. 2.. The fact that there was a very short latency Ž- 3 s. for the onset of the effects of Pb 2q on spontaneously occurring EPSCs and IPSCs suggests that Ži. such effects are mediated by the interactions of this heavy metal with an extracellular site, and Žii. there is a rapid equilibrium between Pb 2q and its site of action. The blockade by Pb 2q of EPSCs and IPSCs was concentration dependent, and at 1 mM Pb 2q completely blocked these currents ŽFig. 2.. This effect of Pb 2q, which had the same magnitude regardless of the number of days
the neurons were in culture, was due to changes in transmitter release, because evidence has been provided that at the concentrations tested, this heavy metal does not affect the activity of the postsynaptic GABA A and AMPA receptors w4,12,52x. 3.3. Pb 2 q-induced reduction of the frequency and amplitude of spontaneously occurring EPSCs and IPSCs is not due to changes in Na q-channel actiÕity To ascertain whether Pb 2q could block Naq channels, and, consequently, the action potentials that trigger the release of GABA and glutamate, current transients evoked by depolarizing steps were recorded in the presence and in the absence of Pb 2q. To abolish the Ca2q component from these transients, the neurons were continuously perfused with external solution containing the Ca2q-channel blocker Cd 2q Ž10 mM..
Fig. 4. Characterization of field stimulation-induced EPSCs and IPSCs. Glutamate-mediated EPSCs or GABA-mediated IPSCs were recorded from hippocampal neurons continuously perfused with external solution containing the GABA A receptor blocker picrotoxin Ž100 mM. or the AMPA receptor antagonist CNQX Ž10 mM., respectively. Right traces: Samples of evoked EPSCs and IPSCs recorded at various membrane potentials. Notice that evoked EPSCs reverse at approximately 0 mV, whereas evoked IPSCs reverse at about y10 mV. Left traces: Samples of evoked EPSCs Žtop three traces. and evoked IPSCs Žbottom three traces. recorded before, during, and after perfusion of the hippocampal neurons with Cd 2q-containing external solution. The blocking effect of Cd 2q on the evoked EPSCs and IPSCs confirms the Ca2q dependence of the release process. Left graphs: Plots of the amplitude of the evoked EPSCs and IPSCs vs. recording time indicates that there is no rundown of these currents during the time of the experiments. In one neuron, EPSCs were evoked every 10 s, and the amplitudes of all currents recorded during 5 min were averaged and plotted against the recording time. Results for the IPSCs were obtained from another neuron. In each graph, each point is the average of 30 events.
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
27
Pb 2q ŽG 100 nM. affected neither the amplitude nor the decay and rising phases of current transients triggered by depolarizing steps of q50 mV applied to neurons held at y70 mV ŽFig. 3.. This result suggests that Pb 2q-induced changes in TTX-sensitive transmitter release may not be due to alterations in Naq-channel activity. It also suggests that back-propagating action potentials Žviewed as fast current transients in neurons held at potentials more negative than y40 mV. were not detected in the presence of Pb 2q Žsee Fig. 2. because of the blockade of transmitter release rather than because of a direct action of the heavy metal on the Naq-channel activity. The possibility that Naq channels present in the nerve terminals have a higher sensitivity to Pb 2q cannot be disregarded at this point. 3.4. Pharmacological and kinetic characterization of the release of glutamate and GABA eÕoked by field stimulation of hippocampal neurons in culture Application of a 50-ms, 5-V pulse via an electroplated bipolar platinum electrode to a neuron synaptically connected to the neuron under study always resulted in the activation of a PSC that was mediated by either glutamate or GABA. In each experiment, the evoked PSCs had variable amplitudes throughout the recording time, but did not show any sign of rundown ŽFig. 4.. In agreement with the concept that the PSCs were the consequence of evoked release of neurotransmitters was the finding that these currents could not be detected in the presence of the Ca2q channel blocker Cd 2q Ž10 mM . ŽFig. 4.. Field stimulation-evoked EPSCs and IPSCs could be distinguished from one another on the basis of one of the following criteria: Ži. EPSCs were sensitive to blockade by CNQX Ž10 mM. and IPSCs were sensitive to inhibition by picrotoxin Ž100 mM.; Žii. the decay-time constants of the EPSCs and IPSCs were 12.0 " 7.6 and 48.0 " 12.7 ms, respectively ŽFig. 4.; and Žiii. the zero-current potential for EPSCs and IPSCs were 0 and y20 mV, respectively ŽFig. 4.. To study the effects of Pb 2q on the evoked release of glutamate and GABA, EPSCs and IPSCs were pharmacologically isolated; evoked EPSCs were recorded from neurons that were perfused with picrotoxin Ž100 mM. and evoked IPSCs were recorded from neurons that were perfused with CNQX Ž10 mM.. In addition, quantification of the effects of Pb 2q on evoked transmitter release was made possible by carrying out all the experiments at positive membrane potentials so that fast current transients representing back-propagating action potentials would not contaminate the PSCs. 3.5. Pb 2 q at low nanomolar concentrations reduces the eÕoked release of glutamate from cultured hippocampal neurons Following 2–3 s perfusion of the hippocampal neurons with external solution containing Pb 2q Ž10 nM–10 mM., the amplitude of evoked EPSCs was reduced ŽFig. 5A..
Fig. 5. Concentration-dependent effect of Pb 2q on evoked EPSCs. Top traces: Samples of evoked EPSCs recorded before, during, and after exposure of two different hippocampal neurons to external solution containing Pb 2q Ž30 or 100 nM.. Membrane potential, q40 mV. Bottom graph: The amplitude of evoked EPSCs recorded in the presence of Pb 2q is expressed as a percentage of the amplitude of the currents recorded in the absence of the heavy metal and is plotted against the test concentrations of Pb 2q. Fit of the data points indicates that the IC 50 for Pb 2q-induced blockade of evoked EPSCs is approximately 68 nM. Each point and bar represents the mean"S.E. of results obtained from at least four neurons ŽU p- 0.05; UU p- 0.01..
This effect of Pb 2q, which was concentration dependent ŽFig. 5B., was only partially reversible upon washing of the neurons with nominally Pb 2q-free external solution ŽFig. 5A.. To determine the IC 50 for Pb 2q in decreasing the amplitude of the evoked EPSCs, single neurons were exposed a single time to a given concentration of Pb 2q. A four-parameter logistic equation was used to fit the con-
28
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
centration–response relationship Žsee Material and Methods ., and the IC 50 and Hill coefficient derived from the best fitting were 67.6 " 2.3 mM and 0.81 " 0.02, respectively ŽFig. 5B.. The ability of Pb 2q to decrease the amplitude of evoked EPSCs was certainly due to changes in the evoked release of glutamate, because, as stated above, the postsynaptic AMPA receptors are insensitive to the tested concentrations of Pb 2q. In agreement with this concept was the finding that Pb 2q at 100 nM, a concentration that reduces by more than 50% the amplitude of the EPSCs, had no effect on the rise time and decay-time constant of these currents ŽFig. 6.. 3.6. Pb 2 q at low nanomolar concentrations reduces the eÕoked release of GABA from cultured hippocampal neurons Pb 2q Ž10 nM–10 mM. also reduced the amplitude of evoked IPSCs ŽFig. 7. without affecting their kinetics of activation and inactivation ŽFig. 8.. This effect of Pb 2q was concentration dependent and was detected 2–3 s after the neurons were exposed to this heavy metal. When the neurons were superfused with nominally-Pb 2q free external solution, a partial recovery of the amplitude of the evoked IPSCs was observed Žsee Fig. 7.. The IC 50 and Hill coefficient estimated from the best fit of the concentration–response curve were 66.07 " 8.16 mM and 0.80 " 0.08, respectively. The fact that at the concentrations tested Pb 2q does not affect the activity of GABA A receptors w4,12,52x and the finding that Pb 2q did not alter the rise
time and the decay-time constants of the evoked IPSCs ŽFig. 8., indicates that the Pb 2q-induced decrease in the IPSC amplitude is attributable to a reduction of evoked release of GABA. 3.7. Full reÕersal of Pb 2 q-induced block in eÕoked transmitter release requires the perfusion of the neurons with EDTA In an attempt to determine whether Pb 2q-induced blockade of field stimulation-evoked transmitter release is mediated by the interaction of this heavy metal with an intracellular or an extracellular target, we analyzed the ability of the chelator EDTA to reverse the effect of Pb 2q. As described above, at 100 nM Pb 2q reduces by more than 50% the mean amplitude of evoked EPSCs, and some of this effect remains even at 30 min after washout of the neurons with nominally Pb 2q-free external solution. At this point, perfusion of the neurons with EDTA Ž200 mM.-containing external solution results in the full recovery of the amplitude of the evoked EPSCs ŽFig. 9.. The same was observed with the evoked IPSCs. Given that EDTA is a positively charged compound, it would not permeate the membrane of the neurons. Thus, it is feasible to hypothesize that EDTA, by chelating Pb 2q extracellularly, removes it from its site of action. 3.8. ReÕersal by the K q-channel blocker 4-aminopyridine of Pb 2 q-induced blockade of eÕoked transmitter release After blocking with Pb 2q either partially or completely the field stimulation-evoked release of GABA and gluta-
Fig. 6. Pb 2q has no effect on the rise time and on decay-time constant of evoked EPSCs. Left histograms: Distribution of the rise times and decay-time constants of evoked EPSCs recorded before and during a 20-min exposure of a hippocampal neuron to external solution containing Pb 2q Ž100 nM.. Right graph: The rise time and the decay-time constants of evoked EPSCs recorded in the absence of Pb 2q were defined as control and taken as 100%. Then, the rise time and the decay-time constants of currents recorded in the presence of Pb 2q were calculated as a percentage of control values. Each vertical bar and error bar represents the mean " S.E. of results obtained from five neurons.
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
29
4. Discussion 4.1. Pb 2 q inhibits the action potential-triggered, Ca 2 qdependent release of neurotransmitters from hippocampal neurons
Fig. 7. Concentration-dependent effect of Pb 2q on evoked IPSCs. Top traces: Samples of evoked IPSCs recorded before, during, and after exposure of two different hippocampal neurons to external solution containing Pb 2q Ž30 or 100 nM.. Membrane potential, q40 mV. Bottom graph: The amplitude of evoked IPSCs recorded in the presence of Pb 2q is expressed as a percentage of the amplitude of the currents recorded in the absence of the heavy metal and is plotted against the test concentrations of Pb 2q. Fit of the data points indicates that the IC 50 for Pb 2q-induced blockade of evoked IPSCs is approximately 66 nM. Each point and bar represents the mean"S.E. of results obtained from at least five neurons ŽUU indicates p- 0.01..
mate, perfusion of the neurons with external solution containing both Pb 2q and the Kq-channel blocker 4-aminopyridine Ž10 mM. resulted in the recovery of the amplitude of the evoked EPSCs and IPSCs. This finding suggested that Pb 2q did not deplete the pool of vesicles available for release. In the absence of changes on voltage-gated Naq channel activity, the inhibition by Pb 2q of evoked transmitter release, being overcome by the 4-aminopyridine-induced prolongation of action potentials, could be explained by an interaction of the heavy metal with voltage-gated Ca2q channels.
In the present study, we demonstrate that at low nanomolar concentrations Pb 2q inhibits the action potential-dependent release of GABA and glutamate from hippocampal neurons, as evidenced by the fact that this heavy metal reduces the amplitude and frequency of spontaneously occurring EPSCs and IPSCs as well as the amplitude of field stimulation-evoked EPSCs and IPSCs. Of interest, much higher Žmicromolar. concentrations of Pb 2q are required to block nerve-evoked ACh release at the neuromuscular junction w7,21,32x, and there has been considerable controversy regarding the effects of this heavy metal on the action potential-dependent release of transmitters in the CNS. In some neurochemical studies carried out in brain synaptosomes, Pb 2q ŽG 1 mM. was shown to inhibit Kq-induced release of transmitters w33,46x. In others, this heavy metal was shown to increase or to have no effect on the Kq-induced release of transmitters w41,45x. These conflicting results have been attributed to methodological problems associated with the synaptosomes w34x. The purity of the preparation, the timing between exposure of the synaptosomes to high Kq and to Pb 2q, and the time the preparation was exposed to Pb 2q could contribute to such conflicting findings, because Ži. neuronal organelles can serve as a sink for Pb 2q w44x; Žii. the effectiveness of heavy metals in blocking depolarization-dependent Ca2q influx can be affected by the duration of the depolarizing stimulus w36,37x; and Žiii. a prolonged exposure of the neuronal preparation to Pb 2q results in an increase of spontaneous release w12,34,35x that can mask or outweigh a possible inhibitory effect of the heavy metal on the evoked release of transmitters. Furthermore, in some instances, substances that alter nerve-evoked transmitter release do not affect Kq-induced release of neurotransmitters w13x. Thus, a direct analysis of specific release processes became necessary to elucidate the actions of this heavy metal in CNS synapses and to determine the effective concentrations at which Pb 2q alters such processes. Electrophysiologically, at least two different transmitter release processes can be studied in cultured neurons or in neurons in brain slices. One process is not dependent on the propagation of action potentials, and is, therefore, insensitive to blockade by TTX. The other process, which is Ca2q dependent, is sensitive to blockade by TTX, because it also depends on the propagation of action potentials. The TTX-sensitive, Ca2q-dependent release of neurotransmitters can be triggered by the spontaneous, asynchronous firing of presynaptic neurons or by field stimulation of presynaptic neurons. Taking into account
30
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
Fig. 8. Pb 2q has no effect on the rise time and on decay-time constant of evoked IPSCs. Left histograms: Distribution of the rise times and decay-time constants of evoked IPSCs recorded before and during exposure of a hippocampal neuron to external solution containing Pb 2q Ž100 nM.. Right graph: The rise time and the decay-time constants of evoked IPSCs recorded in the absence of Pb 2q were defined as control and taken as 100%. Then, the rise time and the decay-time constants of currents recorded in the presence of Pb 2q were calculated as percentage of control values. Each vertical bar and error bar represents the mean " S.E. of results obtained from five neurons.
that these processes can be studied independently, it becomes possible to determine with more certainty the effects and the effective concentrations of a given substance on a specific release process. Thus, although our present findings are in agreement with previous reports that Pb 2q blocks the action potential-dependent transmitter release in the CNS w33–35,41,46x, they demonstrate for the first time that the effective concentrations of this heavy metal in blocking this release process are in the nanomolar range, i.e., well below those detected in the neurochemical studies. 4.2. Sites of action of Pb 2 q in inhibiting the action potential-triggered, Ca 2 q-dependent release of GABA and glutamate from hippocampal neurons The finding that there was a very short latency for the onset of the Pb 2q-induced inhibition of evoked and spontaneously occurring EPSCs and IPSCs indicated an interaction of Pb 2q with extracellular targets. In accordance with this concept was the fact that the effect of Pb 2q on evoked transmitter release was fully reversed by the chelator EDTA, which, being positively charged, poorly crosses the neuronal membrane. Recent studies have indicated that Pb 2q-sensitive receptors such as the nicotinic receptors and the NMDA receptors w4,25,28,38,52x, when present on presynaptic neurons, control the action potential-dependent release of transmitters w5,40x. However, Pb 2q-induced inhibition of transmitter release was observed even in the
presence of specific antagonists of these receptors, and, therefore, could not be accounted for by the interaction of the heavy metal with such receptors. The concentrations at which Pb 2q blocked GABA and glutamate release were very similar, suggesting that the heavy metal was probably acting via a mechanism that is common to the release of neurotransmitters. Such mechanisms could be related, albeit not restricted, to activation of Naq, Kq, and Ca2q channels. At the concentrations that affected transmitter release, Pb 2q did not alter the Naqchannel activity in hippocampal neurons. In fact, previous studies carried out in human neuroblastoma cells had demonstrated that up to 10 mM Pb 2q has no significant effects on either Naq- or Kq-channel currents w42x. Thus, the finding that the Kq-channel blocker 4-aminopyridine reversed the Pb 2q-induced reduction of evoked transmitter release led to the hypothesis that this effect of Pb 2q is mediated by its ability to block Ca2q channels ŽFig. 10.. In rat hippocampal neurons and in several other preparations, Pb 2q blocks both L- and N-type Ca2q channels with IC 50 s of approximately 30 and 80 nM, respectively, and both of these channels are apparently involved in controlling transmitter release from CNS neurons w9,14– 17,24x. Therefore, the finding that Pb 2q inhibited the evoked release of GABA and glutamate with an IC 50 of about 68 nM suggests an action of this heavy metal on N-type Ca2q channels. Further supporting the concept that Pb 2q inhibits the evoked release of neurotransmitters by blocking Ca2q channels is the fact that full reversal of
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
31
Fig. 9. Full reversal of Pb 2q-induced blockade of evoked EPSCs requires treatment of the neurons with EDTA. Left traces: Samples of evoked EPSCs recorded from a hippocampal neuron that was subjected to the following consecutive treatments: Ži. perfusion with external solution Žcontrol., Žii. perfusion with Pb 2q Ž100 nM.-containing external solution, Žiii. perfusion with nominally Pb 2q-free external solution ŽWashout., and Živ. perfusion with nominally Pb 2q-free external solution containing EDTA Ž200 mM.. Graph: The amplitude of the evoked EPSCs recorded under the control condition was taken as 100% and used to normalize the amplitude of the currents evoked under each of the other experimental conditions. Each vertical bar and error bar represents the mean " S.E. of results obtained from four neurons. Notice that a full reversal of Pb 2q-induced blockade of evoked EPSCs requires the perfusion of the neurons with EDTA. ŽU p - 0.05; UU p - 0.01..
Pb 2q-induced blockade of Ca2q channels requires treatment of the neurons with chelators w11x, and the same treatment was needed to fully reverse the effects of Pb 2q on evoked transmitter release Žsee Fig. 9.. 4.3. Toxicological releÕance of the actions of Pb 2 q in transmitter release from hippocampal neurons The toxicological relevance of our current findings that Pb 2q at an extracellular concentration as low as 10 nM blocks the action potential triggered-, Ca2q-dependent release of both GABA and glutamate is emphasized by the fact that concentrations of Pb 2q ranging from 25–100 nM have been reported to be present in the cerebrospinal fluid of humans not known to be occupationally exposed to Pb 2q w18x. There are several means by which alterations in transmitter release can account for the impairment in cognitive functions that have been reported in patients, particularly in children chronically intoxicated with Pb 2q Že.g., Ref. w39x.. First, Pb 2q-induced changes of synaptic activity could alter synaptic organization in the hippocampus, and,
consequently, impair normal brain development. In fact, blockade of Ca2q-dependent transmitter release has been shown to delay neuronal differentiation w53x, and both spontaneously generated and experience-dependent neuronal activity are critical for the establishment of functional cortical circuits during development and throughout life w29x. Thus, the dual effect of Pb 2q on the different transmitter release processes, i.e., the enhancement of TTX-insensitive spontaneous synaptic activity w12x and the reduction of TTX-sensitive, action potential-dependent synaptic activity Žthis paper., could account for the highly deleterious effects of this heavy metal in the developing brain. Secondly, long-term potentiation ŽLTP., a phenomenon considered as the neuronal substrate for memory and learning Žreviewed in Ref. w48x., is impaired by Pb 2q w6,19,30x. The toxicological relevance of Pb 2q-induced blockade of LTP was initially questioned, because in vitro such an effect was detected only in the presence of micromolar concentrations of the heavy metal w6,19x. The debate was settled after the demonstration that LTP is in fact impaired in hippocampal slices taken from rats chronically exposed to low levels of Pb 2q w30x. Attempts to determine
32
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
Fig. 10. The Kq-channel blocker 4-aminopyridine reverses Pb 2q-induced blockade of evoked EPSCs and IPSCs. Left traces: Samples of evoked EPSCs Žtop three traces. and IPSCs Žbottom three traces. recorded from two different hippocampal neurons subjected to the following consecutive treatments: Ži. perfusion with external solution ŽControl., Žii. perfusion with external solution containing Pb 2q Ž100 nM., and Žiii. perfusion with external solution containing the admixture of Pb 2q Ž100 nM. plus 4-aminopyridine Ž10 mM.. Membrane potential, q40 mV. Right graphs: Amplitude of evoked EPSCs and IPSCs recorded under control condition was taken as 100%, and the amplitude of currents recorded in the presence of Pb 2q and Pb 2q plus 4-aminopyridine is presented as a percentage of control. Each vertical bar and error bar represents the mean" S.E. of results obtained from three neurons. Notice that 4-aminopyridine fully reverses Pb 2q-induced blockade of the evoked EPSCs and IPSCs ŽUU p - 0.01..
the mechanism by which Pb 2q blocks LTP have led to the conclusion that this effect is not mediated by the direct action of the heavy metal on NMDA receptors w27x. Considering that LTP is apparently associated with both a postsynaptic and a presynaptic mechanism, and that en-
hancement of presynaptic transmitter release accounts in part for the expression of this phenomenon w31,51x, we suggest that Pb 2q-induced blockade of the action potential-dependent release of glutamate from hippocampal neurons could underlie the blockade by this heavy
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
metal of LTP, and, consequently, its ability to impair cognitive functions. In conclusion, the present study demonstrates that at low nanomolar concentrations Pb 2q inhibits the action potential-dependent release of neurotransmitters from hippocampal neurons, primarily by apparently blocking presynaptic Ca2q channels, thus improving our understanding of the mechanisms underlying the neurotoxic effects of this heavy metal. Further, the finding that full reversal of the Pb 2q-induced blockade of evoked transmitter release requires the treatment of the neurons with EDTA indicates that centrally-acting chelators may be therapeutically useful for the treatment andror prevention of some of the toxic effects of this heavy metal. Acknowledgements The superb technical assistance of Ms. Mabel Zelle, Ms. Barbara Marrow, and Mr. Benjamin Cumming is gratefully acknowledged. This study was supported by USPHS grant ES05730 and by PRONEX Žfrom Brazil. Žfor EXA., by a CNPq fellowship from Brazil Žfor MFMB., and by a minority fellowship under USPHS grant ES05730 Žfor EFRP.. References w1x D.P. Alfano, T.L. Petit, Behavioral effects of postnatal lead exposure: possible relationship to hippocampal dysfunction, Behav. Neural Biol. 32 Ž1981. 319–333. w2x D.P. Alfano, T.L. Petit, Neonatal lead exposure alters the dendritic development of hippocampus dentate granule cells, Exp. Neurol. 75 Ž1982. 275–288. w3x M. Alkondon, E.X. Albuquerque, Diversity of nicotinic acetylcholine receptors in rat hippocampal neurons: I. Pharmacological and functional evidence for distinct structural subtypes, J. Pharmacol. Exp. Ther. 265 Ž1993. 1455–1473. w4x M. Alkondon, A.C.S. Costa, V. Radhakrishnan, R.S. Aronstam, E.X. Albuquerque, Selective blockade of NMDA-activated channel currents may be implicated in learning deficits caused by lead, FEBS Lett. 261 Ž1990. 124–130. w5x M. Alkondon, E.F.R. Pereira, C.T.F. Barbosa, E.X. Albuquerque, Neuronal nicotinic acetylcholine receptor activation modulates gaminobutyric acid release from CA1 neurons of rat hippocampal slices, J. Pharmacol. Exp. Ther. 283 Ž1997. 1396–1411. w6x L. Altmann, K. Sveinsson, H. Wiegand, Long-term potentiation in rat hippocampal slices is impaired following lead perfusion, Neurosci. Lett. 128 Ž1991. 109–112. w7x W.D. Atchison, T. Narahashi, Mechanism of action of lead on neuromuscular junctions, NeuroToxicology 5 Ž1984. 267–282. w8x G. Audesirk, Effects of lead exposure on the physiology of neurons, Prog. Neurobiol. 24 Ž1985. 199–231. w9x G. Audesirk, T. Audesirk, The effects of inorganic lead on voltagesensitive calcium channels differ among cell types and among channel subtypes, NeuroToxicology 14 Ž1993. 137–147. w10x D. Bellinger, K.N. Dietrich, Low-level lead exposure and cognitive function in children, Pediatr. Ann. 23 Ž1994. 600–605. w11x J. Bernal, J.-H. Lee, L.L. Cribbs, E. Perez-Reyer, Full reversal of Pb 2q block of L-type Ca2q channels requires treatment with heavy metal antidotes, J. Pharmacol. Exp. Ther. 282 Ž1997. 172–180.
33
w12x M.F.M. Braga, E.F.R. Pereira, M. Marchioro, E.X. Albuquerque, Lead increases tetrodotoxin-insensitve spontaneous release of glutamate and GABA from hippocampal neurons, Brain Res. Ž1998. submitted. w13x P.L. Buckle, I. Spence, The actions of adenosine and some analogues on evoked and potassium stimulated release at skeletal and autonomic neuromuscular junctions, Naunyn-Schmiedeberg’s Arch. Pharmacol. 319 Ž1982. 130–135. w14x D. Busselberg, M.L. Evans, H. Rahmann, D.O. Carpenter, Lead ¨ inhibits the voltage-activated calcium currents of Aplysia neurons, Toxicol. Lett. 52 Ž1990. 51–57. w15x D. Busselberg, M.L. Evans, H. Rahmann, D.O. Carpenter, Lead and ¨ zinc block a voltage-activated calcium channel of Aplysia neurons, J. Neurophysiol. 65 Ž1991. 786–795. w16x D. Busselberg, M.L. Evans, H.L. Haas, D.O. Carpenter, Blockade of ¨ mammalian and invertebrate calcium channels by lead, NeuroToxicology 14 Ž1993. 249–258. w17x D. Busselberg, B. Platt, D. Michael, H.L. Haas, D.O. Carpenter, ¨ Mammalian voltage-activated calcium channel currents are blocked by Pb 2q, Zn2q and Al 3q, J. Neurophysiol. 71 Ž1994. 1491–1497. w18x A. Cavalleri, C. Minoia, M. Ceroni, M. Poloni, Lead in cerebrospinal fluid and its relationship to plasma lead in humans, J. Appl. Toxicol. 4 Ž1984. 63–65. w19x D.O. Carpenter, M.R. Matthews, P.J. Parsons, N. Hori, Long-term potentiation in the piriform cortex is blocked by lead, Cell. Mol. Neurobiol. 14 Ž1994. 723–733. w20x G.P. Cooper, R.S. Manalis, Influence of heavy metals on synaptic transmission: a review, NeuroToxicology 4 Ž1983. 69–84. w21x G.P. Cooper, R.S. Manalis, Interactions of lead and cadmium on acetylcholine release at the frog neuromuscular junction, Toxicol. Appl. Pharmacol. 74 Ž1984. 411–416. w22x G.P. Cooper, J.B. Suszkiw, R.S. Manalis, in: T. Narahashi ŽEd.., Cellular and Molecular Neurotoxicology, Presynaptic Effects of Heavy Metals, Raven Press, New York, 1984, pp. 1–21. w23x J. Dempster, The computer analysis of eletrophysiological data, in: P.J. Frazer ŽEd.., Microcomputers in Physiology: A Pratical Approach, Oxford IRl Press, Oxford, 1989, pp. 51–93. w24x M.L. Evans, D. Busselberg, D.O. Carpenter, Pb 2q blocks calcium ¨ currents of cultured dorsal root ganglion cells, Neurosci. Lett. 129 Ž1991. 103–106. w25x T.R. Guilarte, R.C. Miceli, Age-dependent effects of lead on w3 HxMK-801 binding to the NMDA receptor-gated ionophore: in vitro and in vivo studies, Neurosci. Lett. 148 Ž1992. 27–30. w26x O.P. Hamill, A. Marty, E. Neher, B. Sakmann, F.J. Sigworth, Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches, Pflug. Arch. 391 Ž1981. 85–100. w27x N. Hori, D. Busselberg, M.R. Matthews, P.J. Parsons, D.O. Carpen¨ ter, Lead blocks LTP by an action not at NMDA receptors, Exp. Neurol. 119 Ž1993. 192–197. w28x K. Ishihara, M. Alkondon, J.G. Montes, E.X. Albuquerque, Nicotinic responses in acutely dissociated rat hippocampal neurons and the selective blockade of fast-desensitizing nicotinic currents by lead, J. Pharmacol. Exp. Ther. 273 Ž1995. 1471–1482. w29x L.C. Katz, C.J. Shatz, Synaptic activity and the construction of cortical circuits, Science ŽWashington, DC. 274 Ž1996. 1133–1138. w30x S.M. Lasley, J. Polan-Curtin, D.L. Armstrong, Chronic exposure to environmental levels of lead impairs in vivo induction of long-term potentiation in rat hippocampal dentate, Brain Res. 614 Ž1993. 347–351. w31x A. Malgaroli, A.E. Ting, B. Wendland, A. Bergamaschi, A. Villa, R.W. Tsien, R.H. Scheller, Presynaptic component of long-term potentiation visualized at individual hippocampal synapses, Science ŽWashington, DC. 268 Ž1995. 1624–1628. w32x R.S. Manalis, G.P. Cooper, Presynaptic and postsynaptic effects of lead at the frog neuromuscular junction, Nature ŽLond.. 243 Ž1973. 354–355.
34
M.F.M. Braga et al.r Brain Research 826 (1999) 22–34
w33x D.J. Minnema, R.D. Greenland, I.A. Michaelson, Effect of in vitro inorganic lead on dopamine release from superfused rat striatal synaptosomes, Toxicol. Appl. Pharmacol. 84 Ž1986. 400–411. w34x D.J. Minnema, I.A. Michaelson, Differential effects of inorganic lead and d-aminolevulinic acid in vitro on synaptosomal g-aminobutyric acid release, Toxicol. Appl. Pharmacol. 86 Ž1986. 437–447. w35x D.J. Minnema, I.A. Michaelson, G.P. Cooper, Calcium efflux and neurotransmitter release from rat hippocampal synaptosomes exposed to lead, Toxicol. Appl. Pharmacol. 92 Ž1988. 351–357. w36x D.A. Nachshen, M.P. Blaustein, Some properties of potassiumstimulated calcium influx in presynaptic nerve endings, J. Gen. Physiol. 76 Ž1980. 709–728. w37x D.A. Nachshen, M.P. Blaustein, Influx of calcium, strontium, and barium in presynaptic nerve endings, J. Gen. Physiol. 79 Ž1982. 1065–1087. w38x K. Nagata, C.S. Huang, J.H. Songand, T. Narahashi, Lead modulation of the neuronal nicotinic acetylcholine receptor in PC12 cells, Brain Res. 754 Ž1997. 21–27. w39x H.L. Needleman, A. Schell, D. Bellinger, A. Leviton, E.N. Allred, The long-term effects of exposure to low doses of lead in childhood, N. Engl. J. Med. 322 Ž1990. 83–88. w40x K. Ohta, N. Arake, M. Shibata, S. Komatsumoto, K. Shimazu, Y. Fukuchi, Presynaptic ionotropic glutamate receptors modulate in vivo release and metabolism of striatal dopamine, noradrenaline, and 5-hydroxytryptamine: Involvement of both NMDA and AMPArkainate subtypes, Neurosci. Res. 21 Ž1994. 83–89. w41x P.B. Ramsay, M.R. Krigman, P. Morell, Developmental studies of the uptake of choline, GABA and dopamine by crude synaptosomal preparations after in vivo and in vitro lead treatment, Brain Res. 187 Ž1980. 383–402. w42x E. Reuveny, T. Narahashi, Potent blocking action of lead on voltage-activated calcium channels in human neuroblastoma cells SH-SY5Y, Brain Res. 545 Ž1991. 312–314.
w43x Z. Shao, J.B. Suszkiw, Ca2q-surrogate action of Pb 2q on acetylcholine release from rat brain synaptosomes, J. Neurochem. 56 Ž1991. 568–574. w44x E.K. Silbergeld, H.S. Adler, J.L. Costa, Subcellular localization of lead in synaptosomes, Res. Commun. Chem. Pathol. Pharmacol. 17 Ž1977. 715–725. w45x E.K. Silbergeld, R.E. Hruska, L.P. Miller, N. Eng, Effects of lead in vivo and in vitro on GABAergic neurochemistry, J. Neurochem. 34 Ž1980. 1712–1718. w46x J. Suszkiw, G. Toth, M. Murawsky, G. Cooper, Effects of Pb 2q and Cd 2q on acetylcholine release and Ca2q movements in synaptosomes subcellular fractions from rat brain and torpedo electric organ, Brain Res. 323 Ž1984. 31–46. w47x T. Sutula, X.X. He, J. Cavazos, G. Scott, Synaptic reorganization in the hippocampus induced by abnormal functional activity, Science ŽWashington, DC. 239 Ž1988. 1147–1150. w48x T.J. Teyler, P. Discenna, Long-term potentiation, Annu. Rev. Neurosci. 10 Ž1987. 131–161. w49x J.L. Tomsig, J.B. Suszkiw, Pb 2q-induced secretion from bovine chromaffin cells: fura-2 as probe for Pb 2q, Am. J. Physiol. 259 Ž1990. C762–C768. w50x J.L. Tomsig, J.B. Suszkiw, Intracellular mechanism of Pb 2q-induced norepinephrine release from bovine chromaffin cells, Am. J. Physiol. 265 Ž1993. C1630–C1636. w51x R.W. Tsien, R. Malinow, Changes in presynaptic function during long-term potentiation, Ann. New York Acad. Sci. 635 Ž1991. 208–220. w52x H. Ujihara, E.X. Albuquerque, Developmental change of the inhibition by lead of NMDA-activated currents in cultured hippocampal neurons, J. Pharmacol. Exp. Ther. 263 Ž1992. 868–875. w53x H. Xie, L. Ziskind-Conhaim, Blocking Ca2q-dependent synaptic release delays motoneuron differentiation in the rat spinal cord, J. Neurosci. 15 Ž1995. 5900–5911.