- Email: [email protected]
Metabolic changes induced in rat hippocampal slices by norepinephrine
Brain Research, 202 (1980) 387-399 ~t?)Elsevier/North-Holland Biomedical Press
387
METABOLIC CHANGES INDUCED 1N RAT HIPPOCAMPAL SLICES BY NOREPINEPHRINE
MENAHEM SEGAL, DAFNA BAR SAGIE and AVRAHAM MAYEVSKY
Department of Isotope Research, The Weizmann Institute of Science, Rehovot, and (D.B.S. and A.M.) Department of Life Sciences, Bar-llan University, Ramat-Gan (Israel) (Accepted J u n e 19th, 1980)
Key words: norepinephrine - - c A M P - - p u m p - - h i p p o c a m p a l slice
SUMMARY
The oxidative metabolic activity of restricted regions of hippocampal slices was assessed by a continuous measurement of the fluorescence of intramitochondrial nicotinamide-adenine dinucleotide (NADH). A large increase in NADH fluorescence was triggered by substituting the oxygen supply to the slice by nitrogen gas. A large and transient increase in NADH fluorescence was also produced by superfusion of the the slice with a high (50 raM) potassium-containing medium. Addition of norepinephrine (NE) to the superfusion medium caused a propranolol-inhibited increase in NADH fluorescence. Furthermore, ouabain, which inhibits the Na-K pump, blocked the effects of NE. An analog of cyclic adenosine monophosphate (cAMP), 8-bromo cAMP, mimicked the effect of NE. Finally, effects of NE could still be produced in a kainic acid-treated hippocampus, where most neurons were previously destroyed by the drug. It is suggested that NE activates a Na-K-ATPase pump, that this effect might be mediated by cAMP, and that these interrelations may underly the physiological action of NE in the brain.
INTRODUCTION
Extensive research in the past decade was directed towards an understanding of the mechanism of action of the inhibitory neurotransmitter, norepinephrine (NE), in the brain. Several hypotheses were proposed including the activation of a Na-K pump *6,1s,2°,2z, the activation of a cyclic adenosine monophosphate (cAMP) generating system2a,z~ and inactivation of Ca 2+ currents as. The resolution among the hypotheses was not possible until recently due, in part, to the lack of a proper in vitro test system where physiological and biochemical changes produced by NE can be
388 correlated. The hippocampus contains a well-characterized noradrenergic inputa~,4'l, it, and, when maintained in vitro, preserves its structural organization4s. Recent mtra~ cellular studies of effects of NE 4a in rat hippocampal slice have supported the hypothesis that NE may activate a Na~ K-ATPase (the pump). As the activation of N a - K - A T P a s e is known to markedly alter the rate of oxidative energy metabolism ~2,53, an attempt was made to verify the latter hypothesis by examining the effect of NE on oxidative metabolism. The rate of oxidative metabolic activity of hippocampal slices was measured by the fluorometric technique of Chance et ale' modified for changes in light-scattering of the tissue t~,24. This technique provides a continuous measurement of the oxidation--reduction state of intramflochondriai nicotinamide-adenine dinucleotide (NADH)I Changes in the redox state of NADH, as monitored in vitro by various optical approaches~,:, H,3° were shown to be related to metabolic processes associated with energy demand changes. METHODS Adult male Wistar rats (250-350 g) of a local breeding colony were used. Rats were decapitated, their brains rapidly removed and placed in cold Ringer solution, Some experiments were performed in 6-bydroxy dopamine- 16-OHDAI treated rats. These rats were injected twice with 400 #g 6-OHDA intracisternally 1-3 weeks before decapitation 44. Another group of rats was injected with kainic acid (KAI I--3 weeks before decapitation, Stereotaxic injections were made into dorsal and ventral hippocampus in chloralhydrate-anesthetized rats. Each h ippocampus was rejected with a total of I #g/#l KA solution. The right hippocampus was dissected out and sliced into 350 # m slices with a McIIwain tissue chopper. The slices were collected with a brush into cold Ringer and transferred into the recording chamber 42 where they were placed on a filter paper and superfused continuously with a Ringer solution. The perfusion rate remained constant through changes m media and was kept at 1.5-2 ml rain. The low volume of the chamber ( ~ 1.0 ml) allowed replacement of the medium within 1 rain. The level of the fluid was adjusted so that a thi n layer of fluid covered the slices. The chamber was superfused with humidified 95 % Oe/5 % CO2 gas mixture at a rate of 0.4 l/rain and kept at 34 ~ 0.5 °C with a regulated heating element. The Ringer solution contained (in mM) NaCI 124. KC1 5. NaHCO3 26. KH2PO4 1.25, CaCl 2. MgSO4 2, glucose 10. The pH was adjusted to 7.4 and the solution was saturated with the 95 '% 02/5/°,~ CO2 gas mixture before use. The following drugs were prepared in the oxygenated Ringer solution just before use: L-arterenol-HC1 (NE 0.1 mM, Sigma) prepared in 0.01 ~ ascorbateL to prevent oxidation. 8-bromo-cyclic adenosine monophosphate (cAMP, 0.1 raM" Boehringer Mannheim GmbH), ouabain octah)drate (50/~m, Sigma) propranolol (0.1 mM, Abic). A tungsten monopolar stimulating microelectrode with a tip diameter of 20-50 #m was occasionally placed in stratum radiatum, to activate the Schaffer-collateral-commissural system. A recording micropipette was placed in the pyramidal layer some 2-3 mm away The recording of responses to the stimulation could thus serve to estimate the viability of the tissue. All electrodes were placed under visual control using a Nikon stereomicroscope.
389 The fluorometric technique employed to monitor changes in N A D H redox level at the surface of the hippocampal slice has previously been described 9,1°,32. Fig. 1 presents the measuring system. The excitation light (366 nm) from a 100 W Hg arc was conducted via a Y-shaped light guide (Schott, Mainz, G.F.R.) into a circular area of measurement (1 mm diameter of active light fibers). This light guide was placed on the selected region of the slice, usually on the pyramidal layer of region CAl. NADH, when excited by 366 nm light, fluoresces in a broad band with a peak at 460 nm, whereas the oxidized form of this coenzyme (NAD') does not; therefore, the measurement of changes in NADH fluorescence intensity could be utilized to infer changes in the rate of oxidative metabolism. The fluorescence signal is mainly derived from intramitochondrial NADH with a negligible contribution from cytoplasmic NADH 13,24. Reflected excitation light at 366 nm was also measured because this value varies due to changes in cell volume4, ~9. Changes in the intensities of 366 nm light and 460 nm fluorescence, emitted by the tissue, were monitored by separate photomultiplier tubes. A corrected fluorescence derived by 1:1 electronic subtraction of the reflection signal from the fluorescence signal was also measured. The measurement of corrected fluorescence compensates for non-specific changes of UV and fluorescence absorption and thus is considered as a more accurate indicator of the NADH redox level2~,24. Changes in the optical signal were referred to as percentage of full-scale values with zero being the value when no light entered the photomultiplier and 100 °;i being the value of the optical signal recorded initially from the slice. Slices were allowed to superfuse with the Ringer solution for a period of 15-20 rain before initial monitoring. The fluorescence, reflectance and corrected fluorescence signals were displayed continuously on a Grass (model 7) polygraph. Experiments were performed in the dark to prevent any possible interference with the light signals measured from the tissue. Before experimental recordings were made, optical signals were allowed to reach steady-state level. RESULTS Experiments were performed with 22 slices taken from 13 rats. The slices maintained their viability for the duration of the experiments (usually 4-6 h); electrical stimulation of the Schaffer-co|lateral-commissural system produced a typical 3-10 mV population spike in the stratum pyramidal (Fig. 1). In most cases, the magnitude of the population spikes remained constant throughout the recording session. Macroscopic observation at the end of the experiment did not reveal any tissue damage due to the placement of the optical measuring probe on the surface of the slice. Oxygen deprivation was produced by replacing the oxygen mixture present in the Ringer solution and in the chamber, by nitrogen gas (Fig. l, bottom). Typically, this substitution could produce a large increase in fluorescence which developed over a period of 1-2 rain. This change was followed by a smaller reduction in reflectance of the tissue, to yield a larger (net) corrected fluorescence (CF) increase (7.5 ~ 0.6 ~, n 13, Fig. 2). When N2-saturated medium was replaced by the normal medium within 1 2 rain, the CF signal returned to the initial level in 10-20 min; otherwise, the following anoxic damage to the tissue was irreversible.
390 SSGNAL OUJ . v
HV '1'
|
SI6NAL OUT
SOURCE
I
v
FLUORESCENCE
1O'/
\-
90~
\
366Etm FILTER
y
tx~llt~lIoN
REFLECTANCE
25.,.] CORRECTED
/
6.25 t I I rain
N2
Fig. 1. Top: a schematic diaglam demonstrating the arrangement of experimental system. Excitation light is conducted via a Y-shaped quartz fibers guide which is placed on the surface of the hippocampal slice. Changes in the intensity of fluorescence and reflectance of the tissue are continuously monitored by separate photomultipliers (PM) and recorded on a Grass polygraph. A stimulatingelectrode (right) and a recording electrode (left) are placed on the slice to record extracellular population response to stimulation of the Schaffer-collateral-commissural path. A specimenrecord of an average response to the stimulation is presented on the left. Viability of the tissue is occasionally measured by recording population responses to stimulation of the slice. Bottom: representative records of changes in reflectance and fluorescence upon superfusion of a slice with a N2 saturated medium for the duration indicated by the horizontal bar. Upward deflection indicates an increase in the optical signal for this and subsequent figures. All optical changes are presented as percent of the full scale values (see text).
A consistant a n d large change in tissue fluorescence was produced by superfusion of slices with a high (50 raM) K + - c o n t a i n i n g medium. This i n d u c e d a large increase (8.22 ± 1.9 %) (Fig. 2) in C F a n d was sometimes a c c o m p a n i e d by a reduction in the reflectance of the tissue (Fig. 3A). Replacement of the high K + m e d i u m by the n o r m a l Ringer caused a r e d u c t i o n of the C F signal back to t h ,~ ~ initial level (Fig. 3A). I n order to assess the p o r t i o n of N A D H response to high K + owing to glycolytic activity, glucose was substituted by pyruvate (10 m M ) in the Ringer solution 6,z0.
39l
-8
F//A ¢///J
2
___
E/../.4
N I//z'g'Z
N7
K÷
NE
NE 6Of' IDANE KA
Fig. 2. Averages of CF changes produced by the various treatments. The groups are (from left): N~, n = 13; high K + medium, n = 10; NE, n = 10; NE of 6-OHDA-treated rats, n = 6; NE of KAtreated rats, n 4. W i t h i n 2 m i n f o l l o w i n g p y r u v a t e s u p e r f u s i o n , a d e c r e a s e ( o x i d a t i o n ) in N A D H
level
(5.5 %) was o b s e r v e d (Fig. 3B). T h e n e w steady-state r e a c h e d a stable level w i t h i n 15 m i n . S u p e r f u s i o n o f the slice w i t h h i g h K -F, 20 m i n a f t e r r e p l a c e m e n t o f g l u c o s e by p y r u v a t e , c a u s e d a s m a l l e r increase in N A D H
A
(Fig. 3C, 4.4 %) in c o m p a r i s o n to the
- -
]min
c r------
~
-
6"25%l NE Fig. 3. A and C: a comparison between responses of reflectance (R), fluorescence (F) and corrected fluorescence (CF) signals to 50 mM K + in glucose-containing medium (A) and in pyruvate-containing medium (C). Note the presence of significant smaller increase in CF in the presence of pyruvate. B; CF changes produced by replacement of glucose by pyruvate. D: CF responses of the same slice (reflectance and fluorescence signal are not presented) to NE in the presence of pyruvate. Time interval between K + superfusion and NE superfusion was 15 rain.
392 increase observed in glucose-containing medium (Fig. 3A. 14.7 74,). The effect of pyruvate both on N A D H level and on N A D H response to high K ' was reversible. and has been reproduced twice in the same slice. NE caused a 4.07 ~j: 1.O", (Fig. 2) increase in CF signals. A similar effect of NE was also observed in the presence of pyruvate (Fig. 3D). Usually there were no significant changes in tissue reflectance, but whenever present, changes in reflectance could not account for the observed changes in the CF s~gnal. One of the possible causes for the NE-induced fluorescence changes ~s the process of uptake of the exogenous N E into noradrenergic terminals in the slice. As this process is known to be energy-dependent, it might result in changes in the energy state of the tissue. This possibility was tested in 6 OHDA-treated rats which are depleted of their noradrenergic terminals 44. The fluorescence changes induced by NE in these rats In -: 4) were similar to those observed in normal rats. The pharmacological nature of the response to NE was tested with propranolol, a beta-adrenergic antagonist. Propranolol completely and reversibly antagonized the effects of N E on N A D H fluorescence in two separate experiments (Fig. 4). The magnitude of the response to N,, exposure (Fig. 4E) indicates that N A D H redox level was not significantly shifted during the experiment. The generality of NE effects was tested by comparing the responses of the various fields of the hippocampus to N E It appears that N E exerted similar effects whenever tested with no systematic differences among the CAI and dentate fields of the H P C (Fig. 5~ The mechanisms underlying the observed effect of NE were tested in several experiments; to test the hypothesis that N E may act via a second messenger, the effects of c A M P on N A D H fluorescence were measured. Indeed. c A M P produced a rise in fluorescence, similar to that produced by N E (3 tests, Fig. 5B).
(•) NE
./ .
.
.
.
® 45'
N E ,I- P R O P
70'
NE
~
NE + PROP 90'
-
~
_
_
6.25 % .
-"
N2
•
Fig, 4. A - D : effects of propranolol, a bcta-adrenergic antagonist, on the N A D H fluorescence changes produced by NE. In this and the following figures only the corrected fluorescence trace is presented as there were no significant changes observed in the reflectance of the tissue. Propranolol (0.! r a M ) antagonized reversibly the effects of NE. This effect was reproducedtwice in the same slice, The drug supcr~ fusion was spaced in time as indicated on the left of each trace. E: a continuous replacement of O~ by
N~ caused a symptomatic rise in NADH fluorescence in the slice.
393
® DG
® DG
© CA I
@ CA I 40' Ouoboin
6.25% NE
K÷
" I rain
Fig. 5. A and B: a comparison of the effects of NE and 8-bromo-cyclicadenosine monophosphate (CAMP) on NADH fluorescence of the dentate gyrus (DG). It appears that CAMP has a slower rise time but also a slower recoverythan NE. A and C: a comparison of the effectsof NE in the dentate gyrus and CA l region of the hippocampus. In both cases the light pipe was centered on the cellular layers, the granular and pyramidal layers respectively. C and D: the effects of ouabain (50/tM) on the responses of CA1 to the superfusion of NE and high K+-containingmedium. A drastic reduction of the response to NE is demonstrated 40' after the onset of superfusion with ouabain-containing medium.
The possibility that NE produces the change in N A D H fluorescence by activating the energy demanding N a - K ATPase ('the pump'), was tested by using ouabain, an inhibitor of the N a - K pump. Upon superfusion of the slice with 50/~M ouabain, a large increase in reflectance signal was observed. However, after the reflectance signal reached a new steady-state, NE effects were drastically reduced (Fig. 5D). Under the same conditions, high extracellular K + ions, which activate the N a - K pump, no longer produced the rise in N A D H fluorescence (Fig. 5D). The tissue was still viable as indicated by the presence of a typical response to N~. Finally, in order to examine the effect of NE on non-neuronal elements, the KAtreated hippocampus was used (Fig. 6). Marked changes have occurred in the hippocampus after KA injection, as revealed by histological examination of injected hippocampi (Fig. 6C and D). ]'here was a complete disappearance of the neurons and a marked elevation in the number of glial cells. In such a preparation NE was still capable of producing an elevation in N A D H fluorescence (Fig. 2; Fig. 6A). This effect, as in the case of the normal hippocampus, was sensitive to the action of ouabain (Fig. 6B). These experiments indicate that not all the effects of NE on N A D H fluorescence are localized in the neurons and that it is likely that some are located in glial elements. DISCUSSION
The present experiments have demonstrated that norepinephrine (NE) could produce a rise in intramitochondrial N A D H , measured by a surface fluorometric technique ~,l°,az in the rat hippocampal slice. The NE-stimulated N A D H rise was blocked by a beta-adrenergic antagonist, propranolol. Although more experiments are needed to clarify the possible contribution of alpha- and beta-adrenergic receptors to
394
NF.
Fig. 6. The action of NE in kainic acid-(KA) treated hippocampus. A : the response to NE i n the hippocampus which was injected with KA 3 weeks before the experiment. B: a reduction in the response to NE after preincubation with 50/~M ouabain. C and D: a comparison of normal (C) and KA- (D) treated hippocampus. Magnification 250 ,. Note the absence of neurons and the large increase in glial cells in the KA-treated hippocampus. the observed fluorescence changes, the antagonistic action of propranolol does suggest that N E produces these changes by activating m e m b r a n o u s adrenergic receptors. The effects o f N E and an elevated extracellular K -~ on N A D H fluorescence were blocked by ouabain. These observations suggest that N E activates an energy-dependent N a - K - A T P a s e much like high K + does. The activation of this ' p u m p ' m a y underly the inhibitory action of N E observed in the hippocampus 43 and etsewhere2L The slice preparation presents some advantages in comparison with the intact brain for the metabolic studies presented above. The advantage gained through perfusion of the tissue is that kinetic considerations are simplified by a continual provision o f fresh media to the tissue which prevents accumulation o f metabolic products in the incubation media. The hemoglobin-free perfused hippocampat slice provides a useful preparation for fluorometric studies; h e m o d y n a m i c artifacts, such as changes in hemoglobin oxygenation and shifts in blood volume which interfere with
395 the in vivo fluorescence measurement, are absent. However, upon exposure of the tissue to anoxia (Fig. 1), as well as to elevated concentrations o f K + (Fig. 3), changes in light scattering, as measured by the reflectance signal, did occur. As both anoxia 3~,52 and high K +I9 depolarize cell membrane, the observed decrease in reflectance probably reflects an increase in cell volume which follows membrane depolarization zg. In order to eliminate the interference of reflectance changes with the fluorescence measurement, we used the correction technique suggested by J6bsis et M. 24 and Harbig eta]. 21. This technique was shown to compensate appropriately for reflectance changes in the slice preparation a0. Thus, the larger portion of the observed corrected fluorescence responses reflect the changes in fluorescence, i.e. in NADH redox state. An increase in the concentration of extracellular potassium in nervous tissue has been shown to result in activation of Na-K-ATPase in both in vivo2S,49 and in vitro studies a,47. Consequently, elevation of K ~ concentration to which brain slices are exposed in the incubation media produce a stimulatory effect on the rate of energy metabolism 6,17,:~4. The stimulation of Na-K-ATPase produced by the exposure of hippocampa] slice to 50 mM K + was characterized by a transient reduction (corrected fluorescence increase) of N A D H (Fig. 3A). This response is quite surprising in view of the usual oxidation of N A D H reported to occur in vivo when oxidative energy metabolism is enhanced. It also differs from the biphasic response of N A D H (transient oxidation followed by a net reduction) to high K + described previously in brain slices 6,7,~4. The difference between the previous observations and ours can be discussed in view of the metabolic conditions of the slice. The oxygen supply to the slice is probably deficient due to limitations in diffusion distances 46. Thus, some portion of the measurement area is probably somewhat hypoxic. This condition might result in the reductive response of N A D H to high K +. Similar response of N A D H (reduction cycle) to cortical spreading depression has been observed in the ischemic intact brain3L It is also possible that due to the presumed hypoxic state of the slice the initial N A D H oxidation phase observed in vitro by others upon exposure of the slice to high K + does not appear in our measurements. The observed rise in N A D H in the presence of high K + can also reflect the substrate-linked association of N A D H to metabolic control reactions of glycolysis. Activation of glycolysis by high K T M may account for the increase of NADH~ by making the entry of substrate into the mitochondria a rate-limiting factorS,iV,24. This activation of glycolysis might be, in addition, supported by the hypoxic conditions which enhance glycolytic rate 15 and by the continual removal of lactic acid formed by the tissue. The contribution of glycolytic activity to the observed rise in N A D H level was assessed by substitution of glucose by pyruvate as substrate. Pyruvate was described to participate differently from glucose in the oxidative metabolism of the slice6,S,30,3a. The observed oxidation of N A D H upon superfusion of the slice with pyruvate (Fig. 3B) indicates that some portion of the N A D H fluorescence level is associated with glycolytic activity. However, the reductive response of N A D H to high K + (Fig. 3C) still observed in the presence ofpyruvate suggests that both enhancement ofglycolysis and hypoxic conditions are responsible for elevation of N A D H level in the presence of
396 high K +. The combination of metabolic control in the hippocampal slice, which is operative with N A D H response, is probably involved in the observed NE-induced rise in N A D H level. The qualitatively similar N A D H changes produced by high K~ and NE, as well as the ability of ouabain to block these responses, support the assumption that N E activates a N a - K - A T P a s e . Our observations are consonant with several previous reports. N[5 has been shown to activate N a - K - A T P a s e :~8 more than M g - A T P a s e TM. and this activation is blocked by ouabaina, TM. These studies were done mainly in homogenates of brain tissue, a condition required for direct measurements of ATPase activity. Indirectly, Heinemann et al. 22 have shown that in the braim NE reduced extracellular K concentration, indicating an activation o f a N a - K pump. Also. ouabain has been shown to antagonize effects of N E applied iontophoretically ,~4. In contrast, measurement of energy indices in the brain have not yielded a coherent picture. A b r a h a m et al. ~ have described a minor decrease in 2-deoxy-glucose uptake in mice cortex produced by stimulation of the nucleus locus coeruleus (LC]. Lesions of the LC caused changes in N A D H fluorescence in cortex26, 4°. but these effects are largely caused by blood flow changes in the cortex of lesioned and stimulated brains 26.~7. Concerning the inhibitory action of NE. the hypothesis that NE activates a N a - K "pump' stands in a traditional apparent conflict with the hypothesis that NE activates a cAMP-generating system, cAMP. in turn. enhances phosphor~lation of specific membrane proteins which produce the observed changes in membrane potential 5. A large body of evidence indicates that. indeed. NE causes stimulation of c A M P formationa,av, ~°. The physiological outcome of this newly formed cAMP is less clear. The present study suggests that the two hypotheses on the inhibitory mode of action of NE are perhaps complementary rather than contrasting; it is possible that N E activates c A M P formation which, in turn. activates a N a - K pump. This hypothesis is supported by previous observations on the effects of c A M P on the N a - K p u m p 22,25,5~ and by our recent observations including an intracellular recording of" effects of NE and cyclic A M P in the hippocampus 43. A similar mechanism has been proposed for the beta-adrenerglc receptor in smooth muscle 41 The observations that a hippocampal slice, devoid of its neurons as a result of kainic acid injection, is still capable of producing large NE-induced changes in N A D H is somewhat surprising. Yet. glial cells do possess a N a - K - A T P a s e ~°.36 and are capable of responding to N E by large increases m cyclic A M P (Segal, in preparation). Moreover, activation of glial N a - K - A T P a s e might be instrumental in modifying neuronal membrane potential leading to a common end product .... hyperpolarizauon of the neurons and cessation of spontaneous activity. It is unlikely that the changes in N A D H observed in the normal hippocampus are produced solely by glial cells since there was a large increase in the number of glial cells in the KA-treated hippocampus. yet the fluorometric response was smaller than in normal rats. Altogether the present work demonstrates that the measuring of energy metabolism in a slice preparation can yield vital information for further understanding of the effects and mechanisms of
397
action of neurotransmitter substances. Furthermore, the experimental results advance the hypothesis that NE activates, perhaps via a cyclic AMP generating system, the Na-K-ATPase which produces the observed effects of NE on spontaneous cellular discharges.
REFERENCES 1 Abraham, W. C., Delanoy, R. L., Dunn, A. J. and Zornetzer, S. F., Locus coeruleus stimulation decreases deoxyglucose uptake in ipsilateral mouse cerebra[ cortex, Brain Research, 172 ([979) 387 392. 2 Ahmed, K., Judah, J. D. and Schoelfield, P. G., Interaction of sodium and potassium with a cationdependent adenosine triphosphatase system from rat brain, Biochim. biophys. Acta (Amst.), 120 (1966) 351-360. 3 Akera, T., Membrane adenosinetriphosphatase: a digitalis receptor?, Science, 198 (1977) 569-574. 4 Aubert, X., Chance, B. and Keynes, R. D., Optical studies of biochemical events in the electric organ of electrophorus, Proc. roy. Soc. B, 160 (1964) 211-245. 5 Bloom, F. E., The role of cyclic nucleotides in central synaptic function, Rev. Physiol. Biochem. Pharmacol., 74 (1975) 1 103. 6 Bull, R. J. and Lutkenhoff, S. D., Early changes in respiration aerobic glycolysis and cellular NAD(P)H in slices of rat cerebral cortex exposed to elevated concentrations of potassium, J. Neurochem., 21 ([973) 913 922. 7 Bull, R. J. and Cummins, J. T., Influence of potassium on the steady state redox potential of the electron transport chain in slices of rat cerebral cortex and the effect of ouabain, J. Neurochem., 21 (1973) 923 927. 8 Bull, R. J., Cytochrome redox potential dependence on substrate in rat cerebral cortex slices: importance of cytoplasmic NAD(P)H and potassium, J. Neurochem., 26 (1976) 149-156. 9 Chance, B., Cohen, M., J6bsis, F. F. and Schoener, B., lntracellular oxidation reduction states in v ivo, Science, 137 (1962) 499-508. 10 Chance, B., Legallais, V., Sorge, J. and Graham, N., A versatile time-sharing multichannel spectrophotometer, reflectometer and fluorometer, Analyt. Biochern., 66 (1975) 498-517. 11 Chance, B., Oshino, N., Sugano, T. and Mayevsky, A., Basic principles of tissue oxygen determination from mitochondria signals. In H. I. Bicher and D. F. Bruley (Eds.), Int. Symp. on Oxygen Transport to Tissue, Advanc. Exp. Med. Biol. Vol. 37,4, Plenum Press, New York, 1973, pp. 277 292. 12 Chance, B. and Williams, G. R., Respiratory enzymes in oxidative phosphorylation. I. Kinetics of oxygen utilization, J. biol. Chem., 217 (1955) 383 393. 13 Chapman, J. B., Fluorometric studies of oxidative metabolism in isolated pupillary muscle of the rabbit, J. gen. Ph.vsiol., 59 (1972) 135-154. 14 Cummins, J. T. and Bull, R., Spectrophotometric measurements of metabolic responses in isolated rat brain cortex, Biochim. biophys. A cta (A rest.), 253 ([ 971 ) 29-38. 15 Dahl, N. A. and Balfour, W. M., Prolonged anoxic survival due to anoxia pre-exposure: brain ATP, lactate and pyruvate, Amer. J. PhysioL, 220 (1964) 1182 1186. 16 Disaiah, D. and Ho, I. K., Kinetics of catecholamirle sensitive Na+-K + ATPase activity in mouse brain synaptosomes, Biochem. Pharmacol., 26 (1977) 2029-2035. 17 Elliott, K. A. C. and Bilodeau, E., The influence of potassium on respiration and glycolysis by brain slices, Biochem. J., 84 (1962) 421 428. 18 Ewart, R. W. and Logan, J. G., A possible role for (Na + K +) ATPase in monoaminergic synaptic transmitter mechanisms, J. Physiol. (Lond.), 284 (1978) 132P. 19 Gibson, 1. M. and Mcllwain, H , Continuous recording of changes in membrane potential in mammalian cerebral tissue in vitro; recovery after depolarization by added substances, J. Physiol. (Loml.), 176 (1965) 261-283. 20 Grisar, T., Frere, J. M. and Franck, G., Effect of K + ions on kinetic properties of the (Na K)ATPase of bulk isolated glial cells, perikarya and synaptosomes from rabbit brain cortex, Brain Research, 165 (1979) 87-103. 21 Harbig, K., Change, B., Kov~ich, A. G. B. and Reivich, M., In vivo measurement of pyridine nucleotide fluorescence from cat brain cortex, J. appl. Physiol., 31 (1971) 93-96. 22 H einemann, U., Lux, H. D. and Zander, K. J., Effects of norepinephrine and DB-c-AMP on active
398
23
24 25 26
27
28 29 30 31 32
33 34 35 36 37 38
39 40
41 42 43 44 45 46 47
uptake of K + in the cerebral cortex of cats. In J. Kelly and R. W. Ryall (Eds.). lontophoresis and Transmitter Mechanisms in the Mammalian Nervous System, Elsevier/North-Holland Biomedical Press, Amsterdam, 1978. Hoffer, B. J., Siggins. G. R., Oliver, A. P. and Bloom, F. E., Activation of the pathway l¥om locus coeruleus to rat cerebellar purkinje neurons: pharmacological evidence of noradrenergic central inhibition, J. Pharmaeol. exp. Ther.. 184 (1973) 553 -569. J6bsis, F. F., O'Connor, M. J.. Vitale. A. and Verman, H.. lntracellular redox changes in functioning cerebral cortex. I. Metabolic effects ofepileptiform activity, J. Neurophysiol.. 34 (I 971 ) 735 -749. Kometiani, P., Kometiani, Z. and Mikeladze. D.. 3'5'-AMP-dependent protein kinase and membrane ATPase of the nerve cell, Progr. Neurobiol.. 1 I (1978) 223 247. LaManna, J. C., Light, A. 1.. Harik. S. E. and Rosenthal, M.. Locus coeruteus lesions: cffects on cerebral oxidative metabolism in vivo: II. The coupling between oxygen delivery and energy demand. Neurosei. Abstr.. 5 (1979) 340. LaManna. J. C., Sylvia, A. L., Martel, D. and Rosenthal, M., Fluorometric monitoring of the effects of adrenergic agents on oxidative metabolism in intact cerebral cortex. Neurophar,naeol~zgy, 15 (1975) 17-24. Lewis. D. V. and Schuette, W H., N A D H fluorescence and [K +] changes during hippocampal electrical stimulation, J. Neurophysiol., 38 (1975) 405 417. Lipton, P.. Effects of membrane depolarization on light scattering by cerebral slices. J, PhysioL (Lond.). 231 (1973~ 365-383. Lipton, P., Effect of membrane depolarization on nicotinamide nucleotide fluorescence m brain slices, Biochem. J., 136 (1973) 999-1009. Mayevsky, A., lschemia in the brain: the effects of carotid artery ligation and decapitation on the energy state of the awake and anesthetized rat. Brain Research, 140 (1978) 217-230. Mayevsky, A., Shedding light on the awake brain. In P. L. Dutton, J. Leigh and A. Searna rEds.), Frontiers in Bioenergetics: From Electrons to Tissues. Vol. 2. Academic Press. New York, 1978. pp. 1467-1476. Mayevsky, A., Crowe, W. and Mela. L.. The interrelation between brain oxidative metabolism and extracellular potassium in the unanesthetized gerbil, Neurol. Res., 1 (I 980) 213-225. MclIwain, H., Biochemistry and the Central Nervous System (3rd edn.), Little and Brown. Boston, 1966, pp. 61--62: 89. Mcllwain, H. and Buchelard, H. S.. Biochemistry trod the Central Nervous System. (4th ed.~. Williams and Wilkins, Baltimore. 1971, pp. 17--20: 70-72; 80-82. Orkand, P. M., Bracho, H. and Orkand, R. K.. Glial metabolism: alteration by potassium levels comparable to those during neural activity, Brain Research, 55 (1973) 467-471. Palmer, E C., Increased cyclic A M P response to norepinephrine in the rat brain following 6-hydroxydopamine, Neuropharmacology, 11 (1972) 145-149. Phillis, J. W.. An involvement of calcium and Na, K ATPase in the inhibitory actions of various compounds on central neurons. In R Huxtable and A Barbeau (Eds.). Taurine. Raven Press. New York. t976, 209-223 Pickel, V., Segal, M. and Bloom, F. E., An autoradiographic study of the efferent pathways of the nucleus locus coeruleus, J. comp. Neurol., 155 (1974) 15-41. Rosenthal, M., Harik, S. I., LaManna, J. C. and Light, A. !.. Locus coeruleus : effects on cerebral oxidative metabolism in vivo. I. cytochrome oxidase in the steady state, Neurosci. Abstr.. 5 (1979) 350. Schwid, C. R., Honeyman, T. W. and Fay, F. S., Mechanism of beta adrenergic relaxation of smooth muscle, Nature rLond.), 277 (1979) 32-36. Segal, M., The action ofserotonin in the rat hippocampat slice, J. Physiol. /Lond. ). 303 (1980) 423 439. Segal, M., The action of norepinephrine in rat hippocampus : intraceUular studies in the slice preparation, Brain Research, in press. Segal, M. and Bloom, F. E., Norepinephrine in the rat hippocampus, I. [ontophoretic studies. Brain Research, 72 (1974) 79-97. Segal, M. and Bloom, F. E., The action of norepinephrine in the rat hippocampus. II. Activation of the input pathway, Brain Research, 72 (1974) 99-114. Siesj~5, B. K., Brain Energy Metabolism, John Wiley, New York, 1978, p. 133. Skou. J. C., The relationship of the (Na~-K ~) activated enzyme system to transport of sodium and potassium across the cell membrane, Bioenergetlcs, 4 (1972) 203-232.
399 48 Skrede, K. K. and Westgaard, R. H., The transverse hippocampal slice: a well-defined cortical structure maintained in vitro, Brain Research, 35 (1971) 589-593. 49 Somjen, G. G., Rosenthal, M., Cordingley, G., LaManna, J. and Lothman, E., Potassium, neuroglia and oxidative metabolism in central gray matter, Fed. Proc., 35 (1976) 1266-1271. 50 Sutherland, E. W. and Robison, G. A., Metabolic effects of catecholamines: the role of cyclic 3'5"-AMP in responses to catecholamines and other hormones, Pharmaeol. Rev., 18 (1966) 145-161. 51 Tria, E., Luly, P., Tomasi, V., Trevisani, A. and Barnabei, O., Modulation by cyclic AMP in vitro of liver plasma membrane (Na-K)-ATPase and protein kinases, Biochim. biophys. Acta (Amst.), 343 (1974) 297-306. 52 Vysko6il, F., Kri2, N. and Buret, J., Potassium selective microelectrodes used for measuring the extracellular brain potassium during spreading depression and anoxic depolarization in rats, Brain Research, 39 (1972) 255-259. 53 Whittam, R., The molecular mechanism of active transport. In G. C. Quarton, T. Melnechuk and F. O. Schmitt (Eds.), The Neurosciences, Rockefeller University Press, New York, 1967, pp. 313326. 54 Yarbrough, G. G., Ouabain antagonism of noradrenaline inhibitions of cerebellar purkinje cells and dopamine inhibitions of caudate neurons, Neuropharmaeology, 15 (1976) 335-338.
387
METABOLIC CHANGES INDUCED 1N RAT HIPPOCAMPAL SLICES BY NOREPINEPHRINE
MENAHEM SEGAL, DAFNA BAR SAGIE and AVRAHAM MAYEVSKY
Department of Isotope Research, The Weizmann Institute of Science, Rehovot, and (D.B.S. and A.M.) Department of Life Sciences, Bar-llan University, Ramat-Gan (Israel) (Accepted J u n e 19th, 1980)
Key words: norepinephrine - - c A M P - - p u m p - - h i p p o c a m p a l slice
SUMMARY
The oxidative metabolic activity of restricted regions of hippocampal slices was assessed by a continuous measurement of the fluorescence of intramitochondrial nicotinamide-adenine dinucleotide (NADH). A large increase in NADH fluorescence was triggered by substituting the oxygen supply to the slice by nitrogen gas. A large and transient increase in NADH fluorescence was also produced by superfusion of the the slice with a high (50 raM) potassium-containing medium. Addition of norepinephrine (NE) to the superfusion medium caused a propranolol-inhibited increase in NADH fluorescence. Furthermore, ouabain, which inhibits the Na-K pump, blocked the effects of NE. An analog of cyclic adenosine monophosphate (cAMP), 8-bromo cAMP, mimicked the effect of NE. Finally, effects of NE could still be produced in a kainic acid-treated hippocampus, where most neurons were previously destroyed by the drug. It is suggested that NE activates a Na-K-ATPase pump, that this effect might be mediated by cAMP, and that these interrelations may underly the physiological action of NE in the brain.
INTRODUCTION
Extensive research in the past decade was directed towards an understanding of the mechanism of action of the inhibitory neurotransmitter, norepinephrine (NE), in the brain. Several hypotheses were proposed including the activation of a Na-K pump *6,1s,2°,2z, the activation of a cyclic adenosine monophosphate (cAMP) generating system2a,z~ and inactivation of Ca 2+ currents as. The resolution among the hypotheses was not possible until recently due, in part, to the lack of a proper in vitro test system where physiological and biochemical changes produced by NE can be
388 correlated. The hippocampus contains a well-characterized noradrenergic inputa~,4'l, it, and, when maintained in vitro, preserves its structural organization4s. Recent mtra~ cellular studies of effects of NE 4a in rat hippocampal slice have supported the hypothesis that NE may activate a Na~ K-ATPase (the pump). As the activation of N a - K - A T P a s e is known to markedly alter the rate of oxidative energy metabolism ~2,53, an attempt was made to verify the latter hypothesis by examining the effect of NE on oxidative metabolism. The rate of oxidative metabolic activity of hippocampal slices was measured by the fluorometric technique of Chance et ale' modified for changes in light-scattering of the tissue t~,24. This technique provides a continuous measurement of the oxidation--reduction state of intramflochondriai nicotinamide-adenine dinucleotide (NADH)I Changes in the redox state of NADH, as monitored in vitro by various optical approaches~,:, H,3° were shown to be related to metabolic processes associated with energy demand changes. METHODS Adult male Wistar rats (250-350 g) of a local breeding colony were used. Rats were decapitated, their brains rapidly removed and placed in cold Ringer solution, Some experiments were performed in 6-bydroxy dopamine- 16-OHDAI treated rats. These rats were injected twice with 400 #g 6-OHDA intracisternally 1-3 weeks before decapitation 44. Another group of rats was injected with kainic acid (KAI I--3 weeks before decapitation, Stereotaxic injections were made into dorsal and ventral hippocampus in chloralhydrate-anesthetized rats. Each h ippocampus was rejected with a total of I #g/#l KA solution. The right hippocampus was dissected out and sliced into 350 # m slices with a McIIwain tissue chopper. The slices were collected with a brush into cold Ringer and transferred into the recording chamber 42 where they were placed on a filter paper and superfused continuously with a Ringer solution. The perfusion rate remained constant through changes m media and was kept at 1.5-2 ml rain. The low volume of the chamber ( ~ 1.0 ml) allowed replacement of the medium within 1 rain. The level of the fluid was adjusted so that a thi n layer of fluid covered the slices. The chamber was superfused with humidified 95 % Oe/5 % CO2 gas mixture at a rate of 0.4 l/rain and kept at 34 ~ 0.5 °C with a regulated heating element. The Ringer solution contained (in mM) NaCI 124. KC1 5. NaHCO3 26. KH2PO4 1.25, CaCl 2. MgSO4 2, glucose 10. The pH was adjusted to 7.4 and the solution was saturated with the 95 '% 02/5/°,~ CO2 gas mixture before use. The following drugs were prepared in the oxygenated Ringer solution just before use: L-arterenol-HC1 (NE 0.1 mM, Sigma) prepared in 0.01 ~ ascorbateL to prevent oxidation. 8-bromo-cyclic adenosine monophosphate (cAMP, 0.1 raM" Boehringer Mannheim GmbH), ouabain octah)drate (50/~m, Sigma) propranolol (0.1 mM, Abic). A tungsten monopolar stimulating microelectrode with a tip diameter of 20-50 #m was occasionally placed in stratum radiatum, to activate the Schaffer-collateral-commissural system. A recording micropipette was placed in the pyramidal layer some 2-3 mm away The recording of responses to the stimulation could thus serve to estimate the viability of the tissue. All electrodes were placed under visual control using a Nikon stereomicroscope.
389 The fluorometric technique employed to monitor changes in N A D H redox level at the surface of the hippocampal slice has previously been described 9,1°,32. Fig. 1 presents the measuring system. The excitation light (366 nm) from a 100 W Hg arc was conducted via a Y-shaped light guide (Schott, Mainz, G.F.R.) into a circular area of measurement (1 mm diameter of active light fibers). This light guide was placed on the selected region of the slice, usually on the pyramidal layer of region CAl. NADH, when excited by 366 nm light, fluoresces in a broad band with a peak at 460 nm, whereas the oxidized form of this coenzyme (NAD') does not; therefore, the measurement of changes in NADH fluorescence intensity could be utilized to infer changes in the rate of oxidative metabolism. The fluorescence signal is mainly derived from intramitochondrial NADH with a negligible contribution from cytoplasmic NADH 13,24. Reflected excitation light at 366 nm was also measured because this value varies due to changes in cell volume4, ~9. Changes in the intensities of 366 nm light and 460 nm fluorescence, emitted by the tissue, were monitored by separate photomultiplier tubes. A corrected fluorescence derived by 1:1 electronic subtraction of the reflection signal from the fluorescence signal was also measured. The measurement of corrected fluorescence compensates for non-specific changes of UV and fluorescence absorption and thus is considered as a more accurate indicator of the NADH redox level2~,24. Changes in the optical signal were referred to as percentage of full-scale values with zero being the value when no light entered the photomultiplier and 100 °;i being the value of the optical signal recorded initially from the slice. Slices were allowed to superfuse with the Ringer solution for a period of 15-20 rain before initial monitoring. The fluorescence, reflectance and corrected fluorescence signals were displayed continuously on a Grass (model 7) polygraph. Experiments were performed in the dark to prevent any possible interference with the light signals measured from the tissue. Before experimental recordings were made, optical signals were allowed to reach steady-state level. RESULTS Experiments were performed with 22 slices taken from 13 rats. The slices maintained their viability for the duration of the experiments (usually 4-6 h); electrical stimulation of the Schaffer-co|lateral-commissural system produced a typical 3-10 mV population spike in the stratum pyramidal (Fig. 1). In most cases, the magnitude of the population spikes remained constant throughout the recording session. Macroscopic observation at the end of the experiment did not reveal any tissue damage due to the placement of the optical measuring probe on the surface of the slice. Oxygen deprivation was produced by replacing the oxygen mixture present in the Ringer solution and in the chamber, by nitrogen gas (Fig. l, bottom). Typically, this substitution could produce a large increase in fluorescence which developed over a period of 1-2 rain. This change was followed by a smaller reduction in reflectance of the tissue, to yield a larger (net) corrected fluorescence (CF) increase (7.5 ~ 0.6 ~, n 13, Fig. 2). When N2-saturated medium was replaced by the normal medium within 1 2 rain, the CF signal returned to the initial level in 10-20 min; otherwise, the following anoxic damage to the tissue was irreversible.
390 SSGNAL OUJ . v
HV '1'
|
SI6NAL OUT
SOURCE
I
v
FLUORESCENCE
1O'/
\-
90~
\
366Etm FILTER
y
tx~llt~lIoN
REFLECTANCE
25.,.] CORRECTED
/
6.25 t I I rain
N2
Fig. 1. Top: a schematic diaglam demonstrating the arrangement of experimental system. Excitation light is conducted via a Y-shaped quartz fibers guide which is placed on the surface of the hippocampal slice. Changes in the intensity of fluorescence and reflectance of the tissue are continuously monitored by separate photomultipliers (PM) and recorded on a Grass polygraph. A stimulatingelectrode (right) and a recording electrode (left) are placed on the slice to record extracellular population response to stimulation of the Schaffer-collateral-commissural path. A specimenrecord of an average response to the stimulation is presented on the left. Viability of the tissue is occasionally measured by recording population responses to stimulation of the slice. Bottom: representative records of changes in reflectance and fluorescence upon superfusion of a slice with a N2 saturated medium for the duration indicated by the horizontal bar. Upward deflection indicates an increase in the optical signal for this and subsequent figures. All optical changes are presented as percent of the full scale values (see text).
A consistant a n d large change in tissue fluorescence was produced by superfusion of slices with a high (50 raM) K + - c o n t a i n i n g medium. This i n d u c e d a large increase (8.22 ± 1.9 %) (Fig. 2) in C F a n d was sometimes a c c o m p a n i e d by a reduction in the reflectance of the tissue (Fig. 3A). Replacement of the high K + m e d i u m by the n o r m a l Ringer caused a r e d u c t i o n of the C F signal back to t h ,~ ~ initial level (Fig. 3A). I n order to assess the p o r t i o n of N A D H response to high K + owing to glycolytic activity, glucose was substituted by pyruvate (10 m M ) in the Ringer solution 6,z0.
39l
-8
F//A ¢///J
2
___
E/../.4
N I//z'g'Z
N7
K÷
NE
NE 6Of' IDANE KA
Fig. 2. Averages of CF changes produced by the various treatments. The groups are (from left): N~, n = 13; high K + medium, n = 10; NE, n = 10; NE of 6-OHDA-treated rats, n = 6; NE of KAtreated rats, n 4. W i t h i n 2 m i n f o l l o w i n g p y r u v a t e s u p e r f u s i o n , a d e c r e a s e ( o x i d a t i o n ) in N A D H
level
(5.5 %) was o b s e r v e d (Fig. 3B). T h e n e w steady-state r e a c h e d a stable level w i t h i n 15 m i n . S u p e r f u s i o n o f the slice w i t h h i g h K -F, 20 m i n a f t e r r e p l a c e m e n t o f g l u c o s e by p y r u v a t e , c a u s e d a s m a l l e r increase in N A D H
A
(Fig. 3C, 4.4 %) in c o m p a r i s o n to the
- -
]min
c r------
~
-
6"25%l NE Fig. 3. A and C: a comparison between responses of reflectance (R), fluorescence (F) and corrected fluorescence (CF) signals to 50 mM K + in glucose-containing medium (A) and in pyruvate-containing medium (C). Note the presence of significant smaller increase in CF in the presence of pyruvate. B; CF changes produced by replacement of glucose by pyruvate. D: CF responses of the same slice (reflectance and fluorescence signal are not presented) to NE in the presence of pyruvate. Time interval between K + superfusion and NE superfusion was 15 rain.
392 increase observed in glucose-containing medium (Fig. 3A. 14.7 74,). The effect of pyruvate both on N A D H level and on N A D H response to high K ' was reversible. and has been reproduced twice in the same slice. NE caused a 4.07 ~j: 1.O", (Fig. 2) increase in CF signals. A similar effect of NE was also observed in the presence of pyruvate (Fig. 3D). Usually there were no significant changes in tissue reflectance, but whenever present, changes in reflectance could not account for the observed changes in the CF s~gnal. One of the possible causes for the NE-induced fluorescence changes ~s the process of uptake of the exogenous N E into noradrenergic terminals in the slice. As this process is known to be energy-dependent, it might result in changes in the energy state of the tissue. This possibility was tested in 6 OHDA-treated rats which are depleted of their noradrenergic terminals 44. The fluorescence changes induced by NE in these rats In -: 4) were similar to those observed in normal rats. The pharmacological nature of the response to NE was tested with propranolol, a beta-adrenergic antagonist. Propranolol completely and reversibly antagonized the effects of N E on N A D H fluorescence in two separate experiments (Fig. 4). The magnitude of the response to N,, exposure (Fig. 4E) indicates that N A D H redox level was not significantly shifted during the experiment. The generality of NE effects was tested by comparing the responses of the various fields of the hippocampus to N E It appears that N E exerted similar effects whenever tested with no systematic differences among the CAI and dentate fields of the H P C (Fig. 5~ The mechanisms underlying the observed effect of NE were tested in several experiments; to test the hypothesis that N E may act via a second messenger, the effects of c A M P on N A D H fluorescence were measured. Indeed. c A M P produced a rise in fluorescence, similar to that produced by N E (3 tests, Fig. 5B).
(•) NE
./ .
.
.
.
® 45'
N E ,I- P R O P
70'
NE
~
NE + PROP 90'
-
~
_
_
6.25 % .
-"
N2
•
Fig, 4. A - D : effects of propranolol, a bcta-adrenergic antagonist, on the N A D H fluorescence changes produced by NE. In this and the following figures only the corrected fluorescence trace is presented as there were no significant changes observed in the reflectance of the tissue. Propranolol (0.! r a M ) antagonized reversibly the effects of NE. This effect was reproducedtwice in the same slice, The drug supcr~ fusion was spaced in time as indicated on the left of each trace. E: a continuous replacement of O~ by
N~ caused a symptomatic rise in NADH fluorescence in the slice.
393
® DG
® DG
© CA I
@ CA I 40' Ouoboin
6.25% NE
K÷
" I rain
Fig. 5. A and B: a comparison of the effects of NE and 8-bromo-cyclicadenosine monophosphate (CAMP) on NADH fluorescence of the dentate gyrus (DG). It appears that CAMP has a slower rise time but also a slower recoverythan NE. A and C: a comparison of the effectsof NE in the dentate gyrus and CA l region of the hippocampus. In both cases the light pipe was centered on the cellular layers, the granular and pyramidal layers respectively. C and D: the effects of ouabain (50/tM) on the responses of CA1 to the superfusion of NE and high K+-containingmedium. A drastic reduction of the response to NE is demonstrated 40' after the onset of superfusion with ouabain-containing medium.
The possibility that NE produces the change in N A D H fluorescence by activating the energy demanding N a - K ATPase ('the pump'), was tested by using ouabain, an inhibitor of the N a - K pump. Upon superfusion of the slice with 50/~M ouabain, a large increase in reflectance signal was observed. However, after the reflectance signal reached a new steady-state, NE effects were drastically reduced (Fig. 5D). Under the same conditions, high extracellular K + ions, which activate the N a - K pump, no longer produced the rise in N A D H fluorescence (Fig. 5D). The tissue was still viable as indicated by the presence of a typical response to N~. Finally, in order to examine the effect of NE on non-neuronal elements, the KAtreated hippocampus was used (Fig. 6). Marked changes have occurred in the hippocampus after KA injection, as revealed by histological examination of injected hippocampi (Fig. 6C and D). ]'here was a complete disappearance of the neurons and a marked elevation in the number of glial cells. In such a preparation NE was still capable of producing an elevation in N A D H fluorescence (Fig. 2; Fig. 6A). This effect, as in the case of the normal hippocampus, was sensitive to the action of ouabain (Fig. 6B). These experiments indicate that not all the effects of NE on N A D H fluorescence are localized in the neurons and that it is likely that some are located in glial elements. DISCUSSION
The present experiments have demonstrated that norepinephrine (NE) could produce a rise in intramitochondrial N A D H , measured by a surface fluorometric technique ~,l°,az in the rat hippocampal slice. The NE-stimulated N A D H rise was blocked by a beta-adrenergic antagonist, propranolol. Although more experiments are needed to clarify the possible contribution of alpha- and beta-adrenergic receptors to
394
NF.
Fig. 6. The action of NE in kainic acid-(KA) treated hippocampus. A : the response to NE i n the hippocampus which was injected with KA 3 weeks before the experiment. B: a reduction in the response to NE after preincubation with 50/~M ouabain. C and D: a comparison of normal (C) and KA- (D) treated hippocampus. Magnification 250 ,. Note the absence of neurons and the large increase in glial cells in the KA-treated hippocampus. the observed fluorescence changes, the antagonistic action of propranolol does suggest that N E produces these changes by activating m e m b r a n o u s adrenergic receptors. The effects o f N E and an elevated extracellular K -~ on N A D H fluorescence were blocked by ouabain. These observations suggest that N E activates an energy-dependent N a - K - A T P a s e much like high K + does. The activation of this ' p u m p ' m a y underly the inhibitory action of N E observed in the hippocampus 43 and etsewhere2L The slice preparation presents some advantages in comparison with the intact brain for the metabolic studies presented above. The advantage gained through perfusion of the tissue is that kinetic considerations are simplified by a continual provision o f fresh media to the tissue which prevents accumulation o f metabolic products in the incubation media. The hemoglobin-free perfused hippocampat slice provides a useful preparation for fluorometric studies; h e m o d y n a m i c artifacts, such as changes in hemoglobin oxygenation and shifts in blood volume which interfere with
395 the in vivo fluorescence measurement, are absent. However, upon exposure of the tissue to anoxia (Fig. 1), as well as to elevated concentrations o f K + (Fig. 3), changes in light scattering, as measured by the reflectance signal, did occur. As both anoxia 3~,52 and high K +I9 depolarize cell membrane, the observed decrease in reflectance probably reflects an increase in cell volume which follows membrane depolarization zg. In order to eliminate the interference of reflectance changes with the fluorescence measurement, we used the correction technique suggested by J6bsis et M. 24 and Harbig eta]. 21. This technique was shown to compensate appropriately for reflectance changes in the slice preparation a0. Thus, the larger portion of the observed corrected fluorescence responses reflect the changes in fluorescence, i.e. in NADH redox state. An increase in the concentration of extracellular potassium in nervous tissue has been shown to result in activation of Na-K-ATPase in both in vivo2S,49 and in vitro studies a,47. Consequently, elevation of K ~ concentration to which brain slices are exposed in the incubation media produce a stimulatory effect on the rate of energy metabolism 6,17,:~4. The stimulation of Na-K-ATPase produced by the exposure of hippocampa] slice to 50 mM K + was characterized by a transient reduction (corrected fluorescence increase) of N A D H (Fig. 3A). This response is quite surprising in view of the usual oxidation of N A D H reported to occur in vivo when oxidative energy metabolism is enhanced. It also differs from the biphasic response of N A D H (transient oxidation followed by a net reduction) to high K + described previously in brain slices 6,7,~4. The difference between the previous observations and ours can be discussed in view of the metabolic conditions of the slice. The oxygen supply to the slice is probably deficient due to limitations in diffusion distances 46. Thus, some portion of the measurement area is probably somewhat hypoxic. This condition might result in the reductive response of N A D H to high K +. Similar response of N A D H (reduction cycle) to cortical spreading depression has been observed in the ischemic intact brain3L It is also possible that due to the presumed hypoxic state of the slice the initial N A D H oxidation phase observed in vitro by others upon exposure of the slice to high K + does not appear in our measurements. The observed rise in N A D H in the presence of high K + can also reflect the substrate-linked association of N A D H to metabolic control reactions of glycolysis. Activation of glycolysis by high K T M may account for the increase of NADH~ by making the entry of substrate into the mitochondria a rate-limiting factorS,iV,24. This activation of glycolysis might be, in addition, supported by the hypoxic conditions which enhance glycolytic rate 15 and by the continual removal of lactic acid formed by the tissue. The contribution of glycolytic activity to the observed rise in N A D H level was assessed by substitution of glucose by pyruvate as substrate. Pyruvate was described to participate differently from glucose in the oxidative metabolism of the slice6,S,30,3a. The observed oxidation of N A D H upon superfusion of the slice with pyruvate (Fig. 3B) indicates that some portion of the N A D H fluorescence level is associated with glycolytic activity. However, the reductive response of N A D H to high K + (Fig. 3C) still observed in the presence ofpyruvate suggests that both enhancement ofglycolysis and hypoxic conditions are responsible for elevation of N A D H level in the presence of
396 high K +. The combination of metabolic control in the hippocampal slice, which is operative with N A D H response, is probably involved in the observed NE-induced rise in N A D H level. The qualitatively similar N A D H changes produced by high K~ and NE, as well as the ability of ouabain to block these responses, support the assumption that N E activates a N a - K - A T P a s e . Our observations are consonant with several previous reports. N[5 has been shown to activate N a - K - A T P a s e :~8 more than M g - A T P a s e TM. and this activation is blocked by ouabaina, TM. These studies were done mainly in homogenates of brain tissue, a condition required for direct measurements of ATPase activity. Indirectly, Heinemann et al. 22 have shown that in the braim NE reduced extracellular K concentration, indicating an activation o f a N a - K pump. Also. ouabain has been shown to antagonize effects of N E applied iontophoretically ,~4. In contrast, measurement of energy indices in the brain have not yielded a coherent picture. A b r a h a m et al. ~ have described a minor decrease in 2-deoxy-glucose uptake in mice cortex produced by stimulation of the nucleus locus coeruleus (LC]. Lesions of the LC caused changes in N A D H fluorescence in cortex26, 4°. but these effects are largely caused by blood flow changes in the cortex of lesioned and stimulated brains 26.~7. Concerning the inhibitory action of NE. the hypothesis that NE activates a N a - K "pump' stands in a traditional apparent conflict with the hypothesis that NE activates a cAMP-generating system, cAMP. in turn. enhances phosphor~lation of specific membrane proteins which produce the observed changes in membrane potential 5. A large body of evidence indicates that. indeed. NE causes stimulation of c A M P formationa,av, ~°. The physiological outcome of this newly formed cAMP is less clear. The present study suggests that the two hypotheses on the inhibitory mode of action of NE are perhaps complementary rather than contrasting; it is possible that N E activates c A M P formation which, in turn. activates a N a - K pump. This hypothesis is supported by previous observations on the effects of c A M P on the N a - K p u m p 22,25,5~ and by our recent observations including an intracellular recording of" effects of NE and cyclic A M P in the hippocampus 43. A similar mechanism has been proposed for the beta-adrenerglc receptor in smooth muscle 41 The observations that a hippocampal slice, devoid of its neurons as a result of kainic acid injection, is still capable of producing large NE-induced changes in N A D H is somewhat surprising. Yet. glial cells do possess a N a - K - A T P a s e ~°.36 and are capable of responding to N E by large increases m cyclic A M P (Segal, in preparation). Moreover, activation of glial N a - K - A T P a s e might be instrumental in modifying neuronal membrane potential leading to a common end product .... hyperpolarizauon of the neurons and cessation of spontaneous activity. It is unlikely that the changes in N A D H observed in the normal hippocampus are produced solely by glial cells since there was a large increase in the number of glial cells in the KA-treated hippocampus. yet the fluorometric response was smaller than in normal rats. Altogether the present work demonstrates that the measuring of energy metabolism in a slice preparation can yield vital information for further understanding of the effects and mechanisms of
397
action of neurotransmitter substances. Furthermore, the experimental results advance the hypothesis that NE activates, perhaps via a cyclic AMP generating system, the Na-K-ATPase which produces the observed effects of NE on spontaneous cellular discharges.
REFERENCES 1 Abraham, W. C., Delanoy, R. L., Dunn, A. J. and Zornetzer, S. F., Locus coeruleus stimulation decreases deoxyglucose uptake in ipsilateral mouse cerebra[ cortex, Brain Research, 172 ([979) 387 392. 2 Ahmed, K., Judah, J. D. and Schoelfield, P. G., Interaction of sodium and potassium with a cationdependent adenosine triphosphatase system from rat brain, Biochim. biophys. Acta (Amst.), 120 (1966) 351-360. 3 Akera, T., Membrane adenosinetriphosphatase: a digitalis receptor?, Science, 198 (1977) 569-574. 4 Aubert, X., Chance, B. and Keynes, R. D., Optical studies of biochemical events in the electric organ of electrophorus, Proc. roy. Soc. B, 160 (1964) 211-245. 5 Bloom, F. E., The role of cyclic nucleotides in central synaptic function, Rev. Physiol. Biochem. Pharmacol., 74 (1975) 1 103. 6 Bull, R. J. and Lutkenhoff, S. D., Early changes in respiration aerobic glycolysis and cellular NAD(P)H in slices of rat cerebral cortex exposed to elevated concentrations of potassium, J. Neurochem., 21 ([973) 913 922. 7 Bull, R. J. and Cummins, J. T., Influence of potassium on the steady state redox potential of the electron transport chain in slices of rat cerebral cortex and the effect of ouabain, J. Neurochem., 21 (1973) 923 927. 8 Bull, R. J., Cytochrome redox potential dependence on substrate in rat cerebral cortex slices: importance of cytoplasmic NAD(P)H and potassium, J. Neurochem., 26 (1976) 149-156. 9 Chance, B., Cohen, M., J6bsis, F. F. and Schoener, B., lntracellular oxidation reduction states in v ivo, Science, 137 (1962) 499-508. 10 Chance, B., Legallais, V., Sorge, J. and Graham, N., A versatile time-sharing multichannel spectrophotometer, reflectometer and fluorometer, Analyt. Biochern., 66 (1975) 498-517. 11 Chance, B., Oshino, N., Sugano, T. and Mayevsky, A., Basic principles of tissue oxygen determination from mitochondria signals. In H. I. Bicher and D. F. Bruley (Eds.), Int. Symp. on Oxygen Transport to Tissue, Advanc. Exp. Med. Biol. Vol. 37,4, Plenum Press, New York, 1973, pp. 277 292. 12 Chance, B. and Williams, G. R., Respiratory enzymes in oxidative phosphorylation. I. Kinetics of oxygen utilization, J. biol. Chem., 217 (1955) 383 393. 13 Chapman, J. B., Fluorometric studies of oxidative metabolism in isolated pupillary muscle of the rabbit, J. gen. Ph.vsiol., 59 (1972) 135-154. 14 Cummins, J. T. and Bull, R., Spectrophotometric measurements of metabolic responses in isolated rat brain cortex, Biochim. biophys. A cta (A rest.), 253 ([ 971 ) 29-38. 15 Dahl, N. A. and Balfour, W. M., Prolonged anoxic survival due to anoxia pre-exposure: brain ATP, lactate and pyruvate, Amer. J. PhysioL, 220 (1964) 1182 1186. 16 Disaiah, D. and Ho, I. K., Kinetics of catecholamirle sensitive Na+-K + ATPase activity in mouse brain synaptosomes, Biochem. Pharmacol., 26 (1977) 2029-2035. 17 Elliott, K. A. C. and Bilodeau, E., The influence of potassium on respiration and glycolysis by brain slices, Biochem. J., 84 (1962) 421 428. 18 Ewart, R. W. and Logan, J. G., A possible role for (Na + K +) ATPase in monoaminergic synaptic transmitter mechanisms, J. Physiol. (Lond.), 284 (1978) 132P. 19 Gibson, 1. M. and Mcllwain, H , Continuous recording of changes in membrane potential in mammalian cerebral tissue in vitro; recovery after depolarization by added substances, J. Physiol. (Loml.), 176 (1965) 261-283. 20 Grisar, T., Frere, J. M. and Franck, G., Effect of K + ions on kinetic properties of the (Na K)ATPase of bulk isolated glial cells, perikarya and synaptosomes from rabbit brain cortex, Brain Research, 165 (1979) 87-103. 21 Harbig, K., Change, B., Kov~ich, A. G. B. and Reivich, M., In vivo measurement of pyridine nucleotide fluorescence from cat brain cortex, J. appl. Physiol., 31 (1971) 93-96. 22 H einemann, U., Lux, H. D. and Zander, K. J., Effects of norepinephrine and DB-c-AMP on active
398
23
24 25 26
27
28 29 30 31 32
33 34 35 36 37 38
39 40
41 42 43 44 45 46 47
uptake of K + in the cerebral cortex of cats. In J. Kelly and R. W. Ryall (Eds.). lontophoresis and Transmitter Mechanisms in the Mammalian Nervous System, Elsevier/North-Holland Biomedical Press, Amsterdam, 1978. Hoffer, B. J., Siggins. G. R., Oliver, A. P. and Bloom, F. E., Activation of the pathway l¥om locus coeruleus to rat cerebellar purkinje neurons: pharmacological evidence of noradrenergic central inhibition, J. Pharmaeol. exp. Ther.. 184 (1973) 553 -569. J6bsis, F. F., O'Connor, M. J.. Vitale. A. and Verman, H.. lntracellular redox changes in functioning cerebral cortex. I. Metabolic effects ofepileptiform activity, J. Neurophysiol.. 34 (I 971 ) 735 -749. Kometiani, P., Kometiani, Z. and Mikeladze. D.. 3'5'-AMP-dependent protein kinase and membrane ATPase of the nerve cell, Progr. Neurobiol.. 1 I (1978) 223 247. LaManna, J. C., Light, A. 1.. Harik. S. E. and Rosenthal, M.. Locus coeruteus lesions: cffects on cerebral oxidative metabolism in vivo: II. The coupling between oxygen delivery and energy demand. Neurosei. Abstr.. 5 (1979) 340. LaManna. J. C., Sylvia, A. L., Martel, D. and Rosenthal, M., Fluorometric monitoring of the effects of adrenergic agents on oxidative metabolism in intact cerebral cortex. Neurophar,naeol~zgy, 15 (1975) 17-24. Lewis. D. V. and Schuette, W H., N A D H fluorescence and [K +] changes during hippocampal electrical stimulation, J. Neurophysiol., 38 (1975) 405 417. Lipton, P.. Effects of membrane depolarization on light scattering by cerebral slices. J, PhysioL (Lond.). 231 (1973~ 365-383. Lipton, P., Effect of membrane depolarization on nicotinamide nucleotide fluorescence m brain slices, Biochem. J., 136 (1973) 999-1009. Mayevsky, A., lschemia in the brain: the effects of carotid artery ligation and decapitation on the energy state of the awake and anesthetized rat. Brain Research, 140 (1978) 217-230. Mayevsky, A., Shedding light on the awake brain. In P. L. Dutton, J. Leigh and A. Searna rEds.), Frontiers in Bioenergetics: From Electrons to Tissues. Vol. 2. Academic Press. New York, 1978. pp. 1467-1476. Mayevsky, A., Crowe, W. and Mela. L.. The interrelation between brain oxidative metabolism and extracellular potassium in the unanesthetized gerbil, Neurol. Res., 1 (I 980) 213-225. MclIwain, H., Biochemistry and the Central Nervous System (3rd edn.), Little and Brown. Boston, 1966, pp. 61--62: 89. Mcllwain, H. and Buchelard, H. S.. Biochemistry trod the Central Nervous System. (4th ed.~. Williams and Wilkins, Baltimore. 1971, pp. 17--20: 70-72; 80-82. Orkand, P. M., Bracho, H. and Orkand, R. K.. Glial metabolism: alteration by potassium levels comparable to those during neural activity, Brain Research, 55 (1973) 467-471. Palmer, E C., Increased cyclic A M P response to norepinephrine in the rat brain following 6-hydroxydopamine, Neuropharmacology, 11 (1972) 145-149. Phillis, J. W.. An involvement of calcium and Na, K ATPase in the inhibitory actions of various compounds on central neurons. In R Huxtable and A Barbeau (Eds.). Taurine. Raven Press. New York. t976, 209-223 Pickel, V., Segal, M. and Bloom, F. E., An autoradiographic study of the efferent pathways of the nucleus locus coeruleus, J. comp. Neurol., 155 (1974) 15-41. Rosenthal, M., Harik, S. I., LaManna, J. C. and Light, A. !.. Locus coeruleus : effects on cerebral oxidative metabolism in vivo. I. cytochrome oxidase in the steady state, Neurosci. Abstr.. 5 (1979) 350. Schwid, C. R., Honeyman, T. W. and Fay, F. S., Mechanism of beta adrenergic relaxation of smooth muscle, Nature rLond.), 277 (1979) 32-36. Segal, M., The action ofserotonin in the rat hippocampat slice, J. Physiol. /Lond. ). 303 (1980) 423 439. Segal, M., The action of norepinephrine in rat hippocampus : intraceUular studies in the slice preparation, Brain Research, in press. Segal, M. and Bloom, F. E., Norepinephrine in the rat hippocampus, I. [ontophoretic studies. Brain Research, 72 (1974) 79-97. Segal, M. and Bloom, F. E., The action of norepinephrine in the rat hippocampus. II. Activation of the input pathway, Brain Research, 72 (1974) 99-114. Siesj~5, B. K., Brain Energy Metabolism, John Wiley, New York, 1978, p. 133. Skou. J. C., The relationship of the (Na~-K ~) activated enzyme system to transport of sodium and potassium across the cell membrane, Bioenergetlcs, 4 (1972) 203-232.
399 48 Skrede, K. K. and Westgaard, R. H., The transverse hippocampal slice: a well-defined cortical structure maintained in vitro, Brain Research, 35 (1971) 589-593. 49 Somjen, G. G., Rosenthal, M., Cordingley, G., LaManna, J. and Lothman, E., Potassium, neuroglia and oxidative metabolism in central gray matter, Fed. Proc., 35 (1976) 1266-1271. 50 Sutherland, E. W. and Robison, G. A., Metabolic effects of catecholamines: the role of cyclic 3'5"-AMP in responses to catecholamines and other hormones, Pharmaeol. Rev., 18 (1966) 145-161. 51 Tria, E., Luly, P., Tomasi, V., Trevisani, A. and Barnabei, O., Modulation by cyclic AMP in vitro of liver plasma membrane (Na-K)-ATPase and protein kinases, Biochim. biophys. Acta (Amst.), 343 (1974) 297-306. 52 Vysko6il, F., Kri2, N. and Buret, J., Potassium selective microelectrodes used for measuring the extracellular brain potassium during spreading depression and anoxic depolarization in rats, Brain Research, 39 (1972) 255-259. 53 Whittam, R., The molecular mechanism of active transport. In G. C. Quarton, T. Melnechuk and F. O. Schmitt (Eds.), The Neurosciences, Rockefeller University Press, New York, 1967, pp. 313326. 54 Yarbrough, G. G., Ouabain antagonism of noradrenaline inhibitions of cerebellar purkinje cells and dopamine inhibitions of caudate neurons, Neuropharmaeology, 15 (1976) 335-338.