Neuroscience Letters, 138 (1992) 9-13
9
© 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
NSL 08513
Comparison of carbon monoxide and nitrogen induced effects on synaptic transmission in the rat hippocampal slice David J. Doolette and David I.B. Kerr Department of Anaesthesia and Intensive Care, University of Adelaide, Adelaide (Australia (Received 20 November 1991; Accepted 10 January 1992)
Key words: Carbon monoxide; Hypoxia; Field potential; Synaptic transmission; Hippocampus A comparison has been made of the effects of carbon monoxide (CO) or nitrogen (N2) exposure on synaptic transmission in the hippocampal slice. CAI field potentials, evoked by Schaffer collateral stimulation, were unaffected by superfusion of slices with artificial cerebral spinal fluid (ACSF) equilibrated with either 15% CO or 15% N2 for 120 min. However, superfusion with hypoxic ACSF equilibrated with either 85% CO or 85% N2 caused a rapid depression of synaptic transmission. Reperfusion with control ACSF following 30 min hypoxia led to recovery of evoked responses and a slight hyperexcitability. In the hippocampal slice, synaptic transmission, as. assessed by input/output curves, was not different during or following hypoxia induced by exposure to CO or N2. In the short term, CO is not toxic.
Carbon monoxide (CO) is a commonly encountered toxic gas. Severe poisoning may either be fatal or may lead to neurological dysfunctions including coma or seizures, often followed by delayed neuropsychiatric disorders that may occur days of weeks after apparent recovery [29]. Such CO toxicity is associated with delayed neuronal degeneration which resembles ischaemic cell damage [17, 29]. This occurs in brain structures vulnerable to hypoxic/ischaemic insults, particularly the hippocampus and striatum, but also typically in the globus pallidus which is less commonly affected by hypoxia [17, 20, 22]. CO toxicity is generally thought to be a result of tissue hypoxia due to failure of oxygen delivery as a result of the greater affinity of haemoglobin (Hb) for CO than for oxygen (O2), although such uncomplicated hypoxia is considered unlikely to cause neuronal damage [6]. In whole animal studies, the prevalence and extent of symptoms of CO poisoning often correlate poorly with carboxy-haemoglobin (COHb) levels in the blood [36], and there is evidence to suggest that CO has direct toxic effects, independent of COHb induced hypoxia [14, 25, 28, 32]. A possible cellular target for CO in producing direct toxic effects would be cytochrome a3, part of the terminal electron carrier complex in the mitochondrial Correspondence: D. Doolette, Department of Aneasthesia and Intensive Care, Royal Adelaide Hospital, Adelaide SA 5000, Australia. Fax: (61) (8) 232 3283
respiratory chain. This complex is known to bind CO [7], and the resulting block in respiration would cause histotoxic hypoxia [6]. The consequences of hypoxia and ischaemia for hippocampal neurons are well known, including an immediate loss of electrical activity, an early functional recovery, and a delayed neuronal degeneration which can be demonstrated in vivo [26, 30] and in vitro [33, 34]. However, the short term consequences of CO poisoning in the hippocampus have not been thoroughly investigated. The hippocampal slice is a convenient model for studying the early functional changes caused by hypoxia. Evoked field potentials recorded in the CA1 are rapidly depressed by hypoxic hypoxia [34] or histotoxic hypoxia [1]. At standard low incubation temperatures, synaptic transmission recovers fully upon reoxygenation following long periods of hypoxia [34], although at physiological temperature hypoxia may cause acute toxicity, manifest as only a partial recovery of field potentials [16]. The present study investigates the suspected toxicity of CO in the hippocampus. Hippocampal slices have been exposed t o N 2 under conditions compatible with full recovery of synaptic evoked potentials, and compared to slices similarly exposed to CO, to establish whether this aspect of neuronal function is differently affected by the two gases. In this in vitro, Hb-free preparation, tissue p02 and pCO are dependent on diffusion alone, and are independent of any CO effects on oxygen transport. Aspects of this work have appeared in abstract form [9].
10 Outbred male Sprague-Dawley rats aged 8 10 weeks were decapitated and the brain rapidly removed into ice cold, oxygenated ACSF. Four h u n d r e d / t m transverse slices were cut from the middle quarter of the hippocampus with a vibrating blade, and submerged in a recording chamber where they were held between two layers of nylon mesh and superfused with artifical cerebral spinal fluid (ACSF) at a rate of 2.5 ml min -1. The slices were allowed to recover at room temperature for 60 90 min, after which time ACSF temperature was maintained at 30°C. After recording control field potentials, the slices were exposed to test ACSF, then allowed to recover for 30 min before field potentials were again recorded in control ACSF. Indirect lighting was used during experiments. Control and hypoxic ACSF were switched by a 3-way Luer lock valve. Each slice used in this study was prepared from a separate animal. ACSF, composition in mM, CaCI 127, KCI 3.5, NaH2PO4 1.25, MgSO4 1.2 or 0, NaHCO3 28 and Dglucose 10, was made in purified water. Control ACSF was bubbled with carbogen (95% 02/5% CO2) through a 10/~m pore size stainless steel high-performance liquid chromatography (HPLC) filter at a flow rate of 500 ml min-~ for 45 min at 30°C. ACSF with various dissolved gas compositions was produced by bubbling in the same manner with a gas mixture of 5% C02, the required percentage of 02, and the balance being either N: or CO. The gases were mixed through a triple rotameter bank, and 02 and CO, partial pressures were monitored by an online polarographic 02 meter and an UV CO2 monitor, respectively, pO2 of the ACSF was measured by a blood gas machine, and dissolved CO by headspace gas chromatography. Micropipettes for extracellular recording were filled with ACSF and the tips broken to give a resistance of 1 2 M~2. These recording electrodes were positioned in the stratum pyramidale and in the stratum radiatum of the CA1 region of the hippocampus, and field potentials were recorded in response to stimulation through a tungsten monopolar stimulating electrode positioned in the adjacent stratum radiatum. Field potential recordings were digitized and stored for off-line processing by an AT microcomputer-based data acquisition and analysis program. Six field potentials were averaged at each of a range of stimulus strengths, from the onset of measurable postsynaptic potentials to supramaximal stimulus, and measurements were made of prevolley amplitude, rate of rise of field excitatory postsynaptic potential (fEPSP), and population spike amplitude. In the case of multiple population spikes the first population spike was measured. The prevolley and the population spike are the compound action potentials in the afferent fiber pathway and the postsynaptic pyramidal
8
140 - ~ C A R B O N
< > Fz
120
MONOXIDE
] 140 i
120
[
100
i
< kd kd
0 t .
.
.
. . . . . . PREV fEPSP vs vs STIM PREV A B
.
. . . PS vs PREV C
.
. . . PS vs fEPSP D
.
100
0
MAX PS AMPL E
Fig. 1. H i s t o g r a m c o m p a r i n g evoked responses f o l l o w i n g 120 m i n sup e r f u s i o n w i t h artifical cerebral spinal f l u i d ( A C S F ) e q u i l i b r a t e d w i t h
either 15% CO or 15% N 2 in the hippocampal slice. Postexposure values of the 4 input/output curves, and for m a x i m u m population spike amplitude are expressed as a percentage of control recordings made before perfusion with CO or N> mean values _+ S.E.M. CO (n=5), N 2 (n=4). No
significant difference was seen between measurements following exposure to CO or N2 (unpaired Student's t-test, P Posthypoxic values of the 4 input/output curves, and for maximum population spike amplitude are expressed as a percentage of the same recordings made before perfusion with hypoxic ACSF, mean values _+ S.E.M. (n=5). No significant difference was seen between any posthypoxic measurements comparing CO with N2 (unpaired Student's t-test, P 5% CO2, balance N2, identical results are obtained using a balance of CO. Bars = 1 ms and 1 mV, negative down. Stimulus artifacts have been edited for clarity.
curves, or maximum population spike amplitude following exposure to either N2 (n=4) or CO (n=5) (P
9
© 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
NSL 08513
Comparison of carbon monoxide and nitrogen induced effects on synaptic transmission in the rat hippocampal slice David J. Doolette and David I.B. Kerr Department of Anaesthesia and Intensive Care, University of Adelaide, Adelaide (Australia (Received 20 November 1991; Accepted 10 January 1992)
Key words: Carbon monoxide; Hypoxia; Field potential; Synaptic transmission; Hippocampus A comparison has been made of the effects of carbon monoxide (CO) or nitrogen (N2) exposure on synaptic transmission in the hippocampal slice. CAI field potentials, evoked by Schaffer collateral stimulation, were unaffected by superfusion of slices with artificial cerebral spinal fluid (ACSF) equilibrated with either 15% CO or 15% N2 for 120 min. However, superfusion with hypoxic ACSF equilibrated with either 85% CO or 85% N2 caused a rapid depression of synaptic transmission. Reperfusion with control ACSF following 30 min hypoxia led to recovery of evoked responses and a slight hyperexcitability. In the hippocampal slice, synaptic transmission, as. assessed by input/output curves, was not different during or following hypoxia induced by exposure to CO or N2. In the short term, CO is not toxic.
Carbon monoxide (CO) is a commonly encountered toxic gas. Severe poisoning may either be fatal or may lead to neurological dysfunctions including coma or seizures, often followed by delayed neuropsychiatric disorders that may occur days of weeks after apparent recovery [29]. Such CO toxicity is associated with delayed neuronal degeneration which resembles ischaemic cell damage [17, 29]. This occurs in brain structures vulnerable to hypoxic/ischaemic insults, particularly the hippocampus and striatum, but also typically in the globus pallidus which is less commonly affected by hypoxia [17, 20, 22]. CO toxicity is generally thought to be a result of tissue hypoxia due to failure of oxygen delivery as a result of the greater affinity of haemoglobin (Hb) for CO than for oxygen (O2), although such uncomplicated hypoxia is considered unlikely to cause neuronal damage [6]. In whole animal studies, the prevalence and extent of symptoms of CO poisoning often correlate poorly with carboxy-haemoglobin (COHb) levels in the blood [36], and there is evidence to suggest that CO has direct toxic effects, independent of COHb induced hypoxia [14, 25, 28, 32]. A possible cellular target for CO in producing direct toxic effects would be cytochrome a3, part of the terminal electron carrier complex in the mitochondrial Correspondence: D. Doolette, Department of Aneasthesia and Intensive Care, Royal Adelaide Hospital, Adelaide SA 5000, Australia. Fax: (61) (8) 232 3283
respiratory chain. This complex is known to bind CO [7], and the resulting block in respiration would cause histotoxic hypoxia [6]. The consequences of hypoxia and ischaemia for hippocampal neurons are well known, including an immediate loss of electrical activity, an early functional recovery, and a delayed neuronal degeneration which can be demonstrated in vivo [26, 30] and in vitro [33, 34]. However, the short term consequences of CO poisoning in the hippocampus have not been thoroughly investigated. The hippocampal slice is a convenient model for studying the early functional changes caused by hypoxia. Evoked field potentials recorded in the CA1 are rapidly depressed by hypoxic hypoxia [34] or histotoxic hypoxia [1]. At standard low incubation temperatures, synaptic transmission recovers fully upon reoxygenation following long periods of hypoxia [34], although at physiological temperature hypoxia may cause acute toxicity, manifest as only a partial recovery of field potentials [16]. The present study investigates the suspected toxicity of CO in the hippocampus. Hippocampal slices have been exposed t o N 2 under conditions compatible with full recovery of synaptic evoked potentials, and compared to slices similarly exposed to CO, to establish whether this aspect of neuronal function is differently affected by the two gases. In this in vitro, Hb-free preparation, tissue p02 and pCO are dependent on diffusion alone, and are independent of any CO effects on oxygen transport. Aspects of this work have appeared in abstract form [9].
10 Outbred male Sprague-Dawley rats aged 8 10 weeks were decapitated and the brain rapidly removed into ice cold, oxygenated ACSF. Four h u n d r e d / t m transverse slices were cut from the middle quarter of the hippocampus with a vibrating blade, and submerged in a recording chamber where they were held between two layers of nylon mesh and superfused with artifical cerebral spinal fluid (ACSF) at a rate of 2.5 ml min -1. The slices were allowed to recover at room temperature for 60 90 min, after which time ACSF temperature was maintained at 30°C. After recording control field potentials, the slices were exposed to test ACSF, then allowed to recover for 30 min before field potentials were again recorded in control ACSF. Indirect lighting was used during experiments. Control and hypoxic ACSF were switched by a 3-way Luer lock valve. Each slice used in this study was prepared from a separate animal. ACSF, composition in mM, CaCI 127, KCI 3.5, NaH2PO4 1.25, MgSO4 1.2 or 0, NaHCO3 28 and Dglucose 10, was made in purified water. Control ACSF was bubbled with carbogen (95% 02/5% CO2) through a 10/~m pore size stainless steel high-performance liquid chromatography (HPLC) filter at a flow rate of 500 ml min-~ for 45 min at 30°C. ACSF with various dissolved gas compositions was produced by bubbling in the same manner with a gas mixture of 5% C02, the required percentage of 02, and the balance being either N: or CO. The gases were mixed through a triple rotameter bank, and 02 and CO, partial pressures were monitored by an online polarographic 02 meter and an UV CO2 monitor, respectively, pO2 of the ACSF was measured by a blood gas machine, and dissolved CO by headspace gas chromatography. Micropipettes for extracellular recording were filled with ACSF and the tips broken to give a resistance of 1 2 M~2. These recording electrodes were positioned in the stratum pyramidale and in the stratum radiatum of the CA1 region of the hippocampus, and field potentials were recorded in response to stimulation through a tungsten monopolar stimulating electrode positioned in the adjacent stratum radiatum. Field potential recordings were digitized and stored for off-line processing by an AT microcomputer-based data acquisition and analysis program. Six field potentials were averaged at each of a range of stimulus strengths, from the onset of measurable postsynaptic potentials to supramaximal stimulus, and measurements were made of prevolley amplitude, rate of rise of field excitatory postsynaptic potential (fEPSP), and population spike amplitude. In the case of multiple population spikes the first population spike was measured. The prevolley and the population spike are the compound action potentials in the afferent fiber pathway and the postsynaptic pyramidal
8
140 - ~ C A R B O N
< > Fz
120
MONOXIDE
] 140 i
120
[
100
i
< kd kd
0 t .
.
.
. . . . . . PREV fEPSP vs vs STIM PREV A B
.
. . . PS vs PREV C
.
. . . PS vs fEPSP D
.
100
0
MAX PS AMPL E
Fig. 1. H i s t o g r a m c o m p a r i n g evoked responses f o l l o w i n g 120 m i n sup e r f u s i o n w i t h artifical cerebral spinal f l u i d ( A C S F ) e q u i l i b r a t e d w i t h
either 15% CO or 15% N 2 in the hippocampal slice. Postexposure values of the 4 input/output curves, and for m a x i m u m population spike amplitude are expressed as a percentage of control recordings made before perfusion with CO or N> mean values _+ S.E.M. CO (n=5), N 2 (n=4). No
significant difference was seen between measurements following exposure to CO or N2 (unpaired Student's t-test, P Posthypoxic values of the 4 input/output curves, and for maximum population spike amplitude are expressed as a percentage of the same recordings made before perfusion with hypoxic ACSF, mean values _+ S.E.M. (n=5). No significant difference was seen between any posthypoxic measurements comparing CO with N2 (unpaired Student's t-test, P 5% CO2, balance N2, identical results are obtained using a balance of CO. Bars = 1 ms and 1 mV, negative down. Stimulus artifacts have been edited for clarity.
curves, or maximum population spike amplitude following exposure to either N2 (n=4) or CO (n=5) (P
