- Email: [email protected]
Potassium-induced epileptiform activity in area CA3 varies markedly along the septotemporal axis of the rat hippocampus
Brain Research, 378 (1986) 169-173 Elsevier
169
BRE 21651
Potassium-induced epileptiform activity in area CA3 varies markedly along the septotemporal axis of the rat hippocampus ANDREW C. BRAGDON 1'2, DONALD M. TAYLOR3 and WILKIE A. WILSON1'2'3 IVeterans Administration Medical Center, Durham, NC27705 and Departments of 2Medicine (Neurology) and 3Pharmacology, Duke University Medical Center, Durham, NC27710 (U.S.A.) (Accepted March 18th, 1986) Key words: rat - - hippocampus - - CA3 - - physiology - - septotemporal - - potassium - - bursting - - epilepsy
Hippocampal slices are generally treated as equivalent regardless of their site of origin along the septotemporal axis. In this study, spontaneous epileptiform bursting was induced in area CA3 of rat hippocampal slices by bathing them in 7 mM potassium. The frequency of spontaneous bursting was measured in all viable slices from 12 hippocampi. Burst frequency was found to vary markedly and in a consistent fashion with site of slice origin along the septotemporal axis. Burst frequency was maximal in slices from near the temporal end and declined progressively toward the septal end. This finding was independent of slicing angle. These results demonstrate that site of slice origin along the septotemporal axis is an important confounding variable in in vitro studies of hippocampal neuronal activity. Furthermore, they support the notion that the temporal portion of the hippocampus may be more prone to seizure activity than the septal hippocampus, possibly because of factors intrinsic to the hippocampus.
The in vitro slice technique has been widely used to study the physiology and pharmacology of the hippocampus. Although significant differences between septal (dorsal) and temporal (ventral) hippocampus have been described in numerous studies of hippocampal function 1'4'5'7-9'12'13'15'17'19, hippocampal slices are often treated as equivalent, regardless of their site of origin along the septotemporal hippocampal axis. The tendency to regard the hippocampus as functionally homogeneous is especially prevalent in studies related specifically to epilepsy. This applies not only to in vitro studies of epileptiform activity, but is often implicit in in vivo studies as well, in that hippocampal activity during seizures in rats is usually recorded only from dorsal hippocampus, probably for technical reasons. Having noted an apparent septotemporal difference in epileptiform activity among hippocampal slices, we set out to examine this issue systematically. Spontaneous population bursting was induced by elevation of the extracellular potassium concentration. We chose this model because it mimics ionic condi-
tions found during seizures 16, and because it does not employ unusual interference with excitatory or inhibitory circuits normally present in the slice. A preliminary report has been published 2°. Male, 280-385 g, S p r a q u e - D a w l e y rats were sacrificed by rapid decapitation. Hippocampal slices, nominally 625 p m thick, were cut on a Mcllwain tissue chopper and placed in holding chambers. Each holding chamber was a glass crystalization dish, lined on the bottom with tissue paper (Kimwipes), filled with warmed (25-30 °C) bathing medium (ACSF), continually bubbled with a 95% 02/5% CO 2 gas mixture, and covered with a glass petri dish cover except when adding or removing slices. The A C S F composition in the holding chambers w a s ( i n mM): KC1 3.3, CaCI 2 1.8, MgSO4 1.2, NaCI 120, N a H C O 3 25, NaH2PO 4 1.23, dextrose 10. This potassium concentration is within the range normally found in brain extracellular fluid 16, and does not induce bursting in hippocampal slices. Slices stored at these calcium and potassium concentrations remain 'healthy' in area CA3 with no epileptiform activity for at least 5 h. In order to study variations in physiological activi-
Correspondence: A.C. Bragdon, Building 16, Room 24, Veterans Administration Medical Center, Durham NC 27705, U.S.A. 0006-8993/86/$03.50 ~ 1986 Elsevier Science Publishers B.V. (Biomedical Division)
170
Schaffer collaterals, and a recording electrode (2 M NaCl-filled, 2-10 M ~ , glass micropipette) was placed in s. pyramidale of CA3 (Fig. 1A). Potentials were amplified (WPI dual microprobe DC amplifiers), filtered (low pass, 1 kHz), displayed at high speed on Nicolet digital oscilloscopes, and recorded at 1 mm/s on Gould chart recorders. Slices from 12 hippocampi (6 rats) were studied. Most hippocampi provided 13 or 14 slices (range 11-16). Usually one or two slices at each end of the hippocampus were overtly damaged and unsuitable for recording. Every slice not obviously damaged during preparation was tested for an evoked response. Electrode positions were adjusted until an optimal evoked response was obtained, after which no further stimulation was performed. Spontaneous burst frequency was evaluated in all slices which showed an evoked response that was detectable on the chart recording above the baseline
ty along the entire septotemporal axis of the hippocampus, we took the following additional measures. One holding chamber was used for each hippocampus. In each holding chamber, a plastic divider was wedged tightly in place on top of the tissue paper, creating 21 small compartments. As they were taken from the chopper, individual slices were placed sequentially in separate compartments preserving their original septotemporal order. Slices were always numbered from temporal to septai, independent of the order of slicing. Slices were studied, singly or in pairs, in a submersion chamber. The temperature was maintained at 31-33 °C, and the chamber perfused at a rate of 3 - 4 ml/min with ACSF identical to that described above, except that it contained 7 mM KC1. For each slice, a monopolar stimulating electrode was placed in stratum radiatum of CA3 at about the C A 3 a - C A 3 b junction in the dense white band formed by the
UW;'
.."-.---7.'-.
cA'
'"'-
\
i,,i, ,.,~e,,s.tt..
A
NEC
/
/"
,
/
I
li i,
//
"
./
-
x
/
,
,
C
~
/
/
0
Fig. 1. A: drawing of a temporal hippocampal slice showing typical placements of stimulating and recording electrodes. B: photograph of a right hippocampus, extended as for slicing, with the septal end to the left and temporal end to the right. C and D: tracings of the photograph in B, with lines showing the two slicing orientations used. In C, temporal slices are generally perpendicular, while septal slices are distinctly angled; in D, the situation is reversed.
171 A
noise. Each slice was allowed to equilibrate to the chamber temperature and potassium concentration until spontaneous bursting was stable (at least 15 min; steady-state burst frequency was usually reached in 5-10 min). Spontaneous bursts were then recorded for at least 5 min; if cycles of bursting were observed, longer recordings (usually 10-15 minutes) were made so that the mean frequency could be estimated unbiased by the burst cycle. The bursts in each record were later counted by hand and expressed as bursts per minute. The first 8 hippocampi were positioned so that the slicing angle was perpendicular to the long axis at the temporal end of the hippocampus (Fig. 1C), and were sliced starting at the temporal end. Slices were studied sequentially from the most temporal to the most septal in half of the hippocampi, and from septal to temporal in the other half. The order in which they were studied was found to have no effect on the outcome. Satisfactory evoked, responses were obtained from 85 of 115 slices; 15 slices were overtly damaged, 12 other slices had no evoked response, and two were lost accidently. Electrical stimulation in 7 mM potassium evoked population bursts in area CA3 consisting of a series of population spikes riding a positive wave. In the absence of stimulation, nearly all slices with an evoked response exhibited spontaneous bursts of similar morphology (Fig. 2A).
20 msec
B
lllhIIUlIIIIIhIIIIIIYlIUI lIII
lFiTlTrTriHIlrlJTJIT_lTlfTTIlTllY{TJll]ll]lJ]JlJJJ. 12
ill Ll
II
II]
Fig. 2. Spontaneous bursting in area CA3. A: a typical spontaneous burst recorded from s. pyramidale of area CA3. Calibration bars: 1 mV, 20 ms. B: 1 rain records from slices 3, 7 and 12 of one hippocampus. Vertical calibration same as in A.
14C
70
120
60
z I0C
50
B
t
~E
40
~ 60 40,
:
. - To o oo | 7
o
50
oO
2o o
2C ,
Tlmporol
l
i
•
o o ¢ ~
,
i
I
•
i
I0
~ 9
~
(x~ee
q
9 SIIptOI
i
Temporal
i
i
i
i
i
i
i
,
Septal
SLICE
Fig. 3. Burst frequency vs original slice location along the septotemporal axis. A: individual burst frequencies of all viable slices. Open circles indicate slices from hippocampi cut perpendicular at the temporal end of the hippocampus; filled circles indicate slices from hippocampi cut perpendicular at the septal end. A single outlying value is seen at slice three. B: mean (+ S.E.M.) burst frequency for each slice location. For slice three, these were calculated both without (closed circle) and with (open circle) the outlier, showing its effect to be minor.
172 Spontaneous burst frequency showed a marked departure from uniformity along the septotemporal axis. Bursting was consistently fastest in temporal slices, with a distinct maximum at slice number 3. Burst frequency then declined progressively through the midhippocampal slices, and was slowest in septal slices. One minute records of spontaneous bursting from slices 3, 7 and 12 of one hippocampus are shown in Fig. 2B. The spontaneous burst frequencies of all viable slices from these 8 hippocampi are shown in Fig. 3A (open circles). These data suggested that the factors which determine burst frequency in area CA3 vary along the septotemporal axis of the hippocampus. Alternatively, these data might have resulted from either: (1) the order in which the slices were cut, with the first slices cut bursting faster; or (2) the angle at which the slices were cut. When isolated and extended for slicing, the rat hippocampus is shaped like a boomerang with temporal and septal limbs, of roughly equal length, joined at an oblique angle (Fig. 1B). Slicing angle is fairly uniform within each limb. In the experiments described above, each hippocampus was oriented on the tissue chopper so that the slices from the temporal limb were cut roughly perpendicular to the long axis of that limb (Fig. 1C). As a result, the remaining slices, from the septal limb, were cut at an angle to the septal axis. Some authors have suggested that recurrent inhibition may be better preserved in angled slices3'18, although this has never actually been demonstrated. Thus, the faster burst rate in the temporal slices could have resulted from the greater disruption of recurrent inhibitory pathways in these slices. We performed the following experiments to address these two possibilities directly. Slices were prepared from four additional hippocampi. Each was positioned on the chopper so that septal slices were cut perpendicular to the septal axis and temporal slices were cut at an angle to the temporal axis (Fig. 1D). Each hippocampus was sliced starting at the septal end. The order in which slices were studied was balanced as before. Statisfactory evoked responses were obtained from 47 of 53 slices. The relationship between burst frequency and slice number (Fig. 3A, closed circles) was the same as that seen in the previous group of slices. As the results in the two groups appeared identi-
cal, the data were merged. The mean (+ S.E.M.) burst frequency at each slice location is shown in Fig. 3B. At slice three, mean and standard error were calculated both without (closed circle) and with (open circle) the one outlying value, and demonstrate that the distinct peak at this location is unaffected by exclusion of the outlier. The present results demonstrate that at least one form of neuronal activity, epileptiform bursting of area CA3, varies markedly along the septotemporal axis of the hippocampus in a consistent fashion. These results have two important implications. First, they demonstrate that site of slice origin along the septotemporal axis is an important confounding variable in the design and interpretation of in vitro studies of hippocampal neuronal activity. This variable must be controlled for when assigning slices from naive animals to control and experimental groups, as well as when comparing neuronal activity in slices from control and experimental animals, e.g. between control and kindled animals. Second, these results reinforce the notion that the temporal portion of the hippocampus may participate more intensely in the development and propagation of seizures in vivo than the more commonly studied septal hippocampus, and suggest that increased attention be directed toward temporal hippocampus in both in vivo and in vitro studies of epileptic activity. This notion, that temporal hippocampus is especially prone to seizure activity, is derived from a number of in vivo studies of limbic seizures. In limbic seizures induced by systemic kainic acid, metabolic and electrical mapping using 2-deoxyglucose and multiple depth electrodes showed that the brain region affected earliest and at the lowest dose was the hippocampus, with ventral (temporal) hippocampus substantially more affected than dorsal (septal) hippocampus 1°. In limbic seizures induced by prolonged chemical or electrical stimulation of medial entorhinal cortex, increased metabolic activity was again greater in ventral than in dorsal hippocampus 2. In the kindling model of limbic epilepsy, rats were shown to kindle more rapidly with ventral than with dorsal hippocampal stimulation 14. Finally, hippocampal damage resulting from prolonged status epilepticus in previously kindled animals was more extensive in ventral than dorsal hippocampus 1~. Thus, both our
173 study and these in vivo studies suggest the t e m p o r a l hippocampus m a y play a m o r e active role than septal hippocampus in some forms of limbic epilepsy. The mechanisms underlying this t e m p o r a l predominance are unknown. The responsible factors could be extrinsic to the h i p p o c a m p u s , e.g. qualitative or quantitative s e p t o t e m p o r a l differences in afferent pathways transmitting seizure activity to the hippocampus. Alternatively, s e p t o t e m p o r a l differences in intrinsic neuronal elements could account for the results of these studies. Some support for intrinsic circuits is provided by the kainic acid study cited above 1°. A t the lowest dose of kainic acid which induced seizures, the pattern of increased metabolic activity was not only greater t e m p o r a l l y than septally, in the hippocampus, but was most striking in area C A 3 , and next
most in CA1. Activity in entorhinal cortex, dentate gyrus and medial septum r e m a i n e d at control levels. O u r results d e m o n s t r a t e that elements which could account for these s e p t o t e m p o r a l differences are preserved within the h i p p o c a m p a l slice. This is further s u p p o r t e d by a recent study, c o m p l e m e n t a r y to our own, which found that stimulus-evoked epileptiform potentials in area CA1 were m o r e complex in t e m p o ral than in septal h i p p o c a m p a l slices 6. H o w e v e r , this still leaves open the question of w h e t h e r these elements are afferent terminals and their transmitters or whether they are local neuronal circuits within the hippocampus.
1 Brazier, M.A.B., Regional activities within the human hippocampus and hippocampal gyrus, Exp. Neurol., 26 (1970) 354-368. 2 Collins, R.C., Tearse, R.G. and Lothman, E.W., Functional anatomy of limbic seizures: focal discharges from medial entorhinal cortex in rat, Brain Research, 280 (1983) 25-40. 3 Dingledine, R. and Langmoen, I.A., Conductance changes and inhibitory actions of hippocampal recurrent IPSPs, Brain Research, 185 (1980) 277-287. 4 Elul, R., Regional differences in the hippocampus of the cat. I. Specific discharge patterns of the dorsal and ventral hippocampus and their role in generalized seizures, Electroencephalogr. Clin. Neurophysiol., 16 (1964) 470-488. 5 Fried, P.A., Limbic system lesions in rats: differential effects in an approach-avoidance task, J. Comp. Physiol. Psychol., 74 (1971) 349-353. 6 Gilbert, M.E., Racine, R.J. and Smith, G.K., Epileptiform burst responses in ventral vs dorsal hippocampal slices, Soc. Neurosci. Abstr., 11 (1985) 1229. 7 Hughes, K.R., Dorsal and ventral hippocampus lesions and maze learning: influence of preoperative environment, Can. J. Psychol., 19 (1965) 325-332. 8 Koreli, A., Influences of dorsal and ventral hippocampus on hypothalamic self-stimulation, Physiol. Behav., 19 (1977) 713-717. 9 Lanier, L.P. and Isaacson, R.L., Activity changes related to the location of lesions in the hippocampus, Behav. Biol., 13 (1975) 59-69. 10 Lothman, E.W. and Collins, R.C., Kainic acid induced limbic seizures: Metabolic, behavioral, electroencephalographic and neuropathological correlates, Brain Research, 218 (1981) 299-318. ll Mclntyre, D.C., Edson, N. and Nathanson, D., A new
model of partial status epilepticus based on kindling, Brain Research, 250 (1982) 53-63. 12 Myhrer, T., Locomotor, avoidance, and maze behavior in rats with selective disruption of hippocampal output, J. Comp. Physiol. Psychol., 89 (1975) 759-777. 13 Nadel, L., Dorsal and ventral hippocampal lesions and behavior, Physiol. Behav., 3 (1968) 891-900. 14 Racine, R., Rose, P.A. and Burnham, W.M., Afterdischarge thresholds and kindling rates in dorsal and ventral hippocampus and dentate gyrus, Can. J. Neurol. Sci., 4 (1977) 273-278. 15 Siegel, A. and Flynn, J.P., Differential effects of electrical stimulation and lesions of the hippocampus and adjacent regions upon attack behavior in cats, Brain Research, 7 (1968) 252-267. 16 Somjen, G.G., Extracellular potassium in the mammalian central nervous system, Annu. Rev. Physiol., 41 (1979) 159-177. 17 Stevens, R. and Cowey, A., Effects of dorsal and ventral hippocampal lesions on spontaneous alternation, learned alternation and probability learning in rats, Brain Research, 52 (1973) 203-224. 18 Struble, R.G., Desmond, N.L. and Levy, W.B., Anatomical evidence for interlamellar inhibition in the fascia dentata, Brain Research, 152 (1978) 580-585. 19 Watson, R.E., Edinger, H.M. and Siegel, A., A [14C]2-deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. III. The hippocampal formation, Brain Res. Rev., 5 (1983) 133-176. 20 Wilson, W.A., Bragdon, A.C. and Taylor, D.M., Potassium-induced epileptiform activity in hippocampal area CA3 varies markedly along the septo-temporal axis, Soc. Neurosci. Abstr., 11 (1985) 1324.
This work was s u p p o r t e d by the V e t e r a n s A d m i n istration, and N I H Grants G M 47105 and NS 17771.
169
BRE 21651
Potassium-induced epileptiform activity in area CA3 varies markedly along the septotemporal axis of the rat hippocampus ANDREW C. BRAGDON 1'2, DONALD M. TAYLOR3 and WILKIE A. WILSON1'2'3 IVeterans Administration Medical Center, Durham, NC27705 and Departments of 2Medicine (Neurology) and 3Pharmacology, Duke University Medical Center, Durham, NC27710 (U.S.A.) (Accepted March 18th, 1986) Key words: rat - - hippocampus - - CA3 - - physiology - - septotemporal - - potassium - - bursting - - epilepsy
Hippocampal slices are generally treated as equivalent regardless of their site of origin along the septotemporal axis. In this study, spontaneous epileptiform bursting was induced in area CA3 of rat hippocampal slices by bathing them in 7 mM potassium. The frequency of spontaneous bursting was measured in all viable slices from 12 hippocampi. Burst frequency was found to vary markedly and in a consistent fashion with site of slice origin along the septotemporal axis. Burst frequency was maximal in slices from near the temporal end and declined progressively toward the septal end. This finding was independent of slicing angle. These results demonstrate that site of slice origin along the septotemporal axis is an important confounding variable in in vitro studies of hippocampal neuronal activity. Furthermore, they support the notion that the temporal portion of the hippocampus may be more prone to seizure activity than the septal hippocampus, possibly because of factors intrinsic to the hippocampus.
The in vitro slice technique has been widely used to study the physiology and pharmacology of the hippocampus. Although significant differences between septal (dorsal) and temporal (ventral) hippocampus have been described in numerous studies of hippocampal function 1'4'5'7-9'12'13'15'17'19, hippocampal slices are often treated as equivalent, regardless of their site of origin along the septotemporal hippocampal axis. The tendency to regard the hippocampus as functionally homogeneous is especially prevalent in studies related specifically to epilepsy. This applies not only to in vitro studies of epileptiform activity, but is often implicit in in vivo studies as well, in that hippocampal activity during seizures in rats is usually recorded only from dorsal hippocampus, probably for technical reasons. Having noted an apparent septotemporal difference in epileptiform activity among hippocampal slices, we set out to examine this issue systematically. Spontaneous population bursting was induced by elevation of the extracellular potassium concentration. We chose this model because it mimics ionic condi-
tions found during seizures 16, and because it does not employ unusual interference with excitatory or inhibitory circuits normally present in the slice. A preliminary report has been published 2°. Male, 280-385 g, S p r a q u e - D a w l e y rats were sacrificed by rapid decapitation. Hippocampal slices, nominally 625 p m thick, were cut on a Mcllwain tissue chopper and placed in holding chambers. Each holding chamber was a glass crystalization dish, lined on the bottom with tissue paper (Kimwipes), filled with warmed (25-30 °C) bathing medium (ACSF), continually bubbled with a 95% 02/5% CO 2 gas mixture, and covered with a glass petri dish cover except when adding or removing slices. The A C S F composition in the holding chambers w a s ( i n mM): KC1 3.3, CaCI 2 1.8, MgSO4 1.2, NaCI 120, N a H C O 3 25, NaH2PO 4 1.23, dextrose 10. This potassium concentration is within the range normally found in brain extracellular fluid 16, and does not induce bursting in hippocampal slices. Slices stored at these calcium and potassium concentrations remain 'healthy' in area CA3 with no epileptiform activity for at least 5 h. In order to study variations in physiological activi-
Correspondence: A.C. Bragdon, Building 16, Room 24, Veterans Administration Medical Center, Durham NC 27705, U.S.A. 0006-8993/86/$03.50 ~ 1986 Elsevier Science Publishers B.V. (Biomedical Division)
170
Schaffer collaterals, and a recording electrode (2 M NaCl-filled, 2-10 M ~ , glass micropipette) was placed in s. pyramidale of CA3 (Fig. 1A). Potentials were amplified (WPI dual microprobe DC amplifiers), filtered (low pass, 1 kHz), displayed at high speed on Nicolet digital oscilloscopes, and recorded at 1 mm/s on Gould chart recorders. Slices from 12 hippocampi (6 rats) were studied. Most hippocampi provided 13 or 14 slices (range 11-16). Usually one or two slices at each end of the hippocampus were overtly damaged and unsuitable for recording. Every slice not obviously damaged during preparation was tested for an evoked response. Electrode positions were adjusted until an optimal evoked response was obtained, after which no further stimulation was performed. Spontaneous burst frequency was evaluated in all slices which showed an evoked response that was detectable on the chart recording above the baseline
ty along the entire septotemporal axis of the hippocampus, we took the following additional measures. One holding chamber was used for each hippocampus. In each holding chamber, a plastic divider was wedged tightly in place on top of the tissue paper, creating 21 small compartments. As they were taken from the chopper, individual slices were placed sequentially in separate compartments preserving their original septotemporal order. Slices were always numbered from temporal to septai, independent of the order of slicing. Slices were studied, singly or in pairs, in a submersion chamber. The temperature was maintained at 31-33 °C, and the chamber perfused at a rate of 3 - 4 ml/min with ACSF identical to that described above, except that it contained 7 mM KC1. For each slice, a monopolar stimulating electrode was placed in stratum radiatum of CA3 at about the C A 3 a - C A 3 b junction in the dense white band formed by the
UW;'
.."-.---7.'-.
cA'
'"'-
\
i,,i, ,.,~e,,s.tt..
A
NEC
/
/"
,
/
I
li i,
//
"
./
-
x
/
,
,
C
~
/
/
0
Fig. 1. A: drawing of a temporal hippocampal slice showing typical placements of stimulating and recording electrodes. B: photograph of a right hippocampus, extended as for slicing, with the septal end to the left and temporal end to the right. C and D: tracings of the photograph in B, with lines showing the two slicing orientations used. In C, temporal slices are generally perpendicular, while septal slices are distinctly angled; in D, the situation is reversed.
171 A
noise. Each slice was allowed to equilibrate to the chamber temperature and potassium concentration until spontaneous bursting was stable (at least 15 min; steady-state burst frequency was usually reached in 5-10 min). Spontaneous bursts were then recorded for at least 5 min; if cycles of bursting were observed, longer recordings (usually 10-15 minutes) were made so that the mean frequency could be estimated unbiased by the burst cycle. The bursts in each record were later counted by hand and expressed as bursts per minute. The first 8 hippocampi were positioned so that the slicing angle was perpendicular to the long axis at the temporal end of the hippocampus (Fig. 1C), and were sliced starting at the temporal end. Slices were studied sequentially from the most temporal to the most septal in half of the hippocampi, and from septal to temporal in the other half. The order in which they were studied was found to have no effect on the outcome. Satisfactory evoked, responses were obtained from 85 of 115 slices; 15 slices were overtly damaged, 12 other slices had no evoked response, and two were lost accidently. Electrical stimulation in 7 mM potassium evoked population bursts in area CA3 consisting of a series of population spikes riding a positive wave. In the absence of stimulation, nearly all slices with an evoked response exhibited spontaneous bursts of similar morphology (Fig. 2A).
20 msec
B
lllhIIUlIIIIIhIIIIIIYlIUI lIII
lFiTlTrTriHIlrlJTJIT_lTlfTTIlTllY{TJll]ll]lJ]JlJJJ. 12
ill Ll
II
II]
Fig. 2. Spontaneous bursting in area CA3. A: a typical spontaneous burst recorded from s. pyramidale of area CA3. Calibration bars: 1 mV, 20 ms. B: 1 rain records from slices 3, 7 and 12 of one hippocampus. Vertical calibration same as in A.
14C
70
120
60
z I0C
50
B
t
~E
40
~ 60 40,
:
. - To o oo | 7
o
50
oO
2o o
2C ,
Tlmporol
l
i
•
o o ¢ ~
,
i
I
•
i
I0
~ 9
~
(x~ee
q
9 SIIptOI
i
Temporal
i
i
i
i
i
i
i
,
Septal
SLICE
Fig. 3. Burst frequency vs original slice location along the septotemporal axis. A: individual burst frequencies of all viable slices. Open circles indicate slices from hippocampi cut perpendicular at the temporal end of the hippocampus; filled circles indicate slices from hippocampi cut perpendicular at the septal end. A single outlying value is seen at slice three. B: mean (+ S.E.M.) burst frequency for each slice location. For slice three, these were calculated both without (closed circle) and with (open circle) the outlier, showing its effect to be minor.
172 Spontaneous burst frequency showed a marked departure from uniformity along the septotemporal axis. Bursting was consistently fastest in temporal slices, with a distinct maximum at slice number 3. Burst frequency then declined progressively through the midhippocampal slices, and was slowest in septal slices. One minute records of spontaneous bursting from slices 3, 7 and 12 of one hippocampus are shown in Fig. 2B. The spontaneous burst frequencies of all viable slices from these 8 hippocampi are shown in Fig. 3A (open circles). These data suggested that the factors which determine burst frequency in area CA3 vary along the septotemporal axis of the hippocampus. Alternatively, these data might have resulted from either: (1) the order in which the slices were cut, with the first slices cut bursting faster; or (2) the angle at which the slices were cut. When isolated and extended for slicing, the rat hippocampus is shaped like a boomerang with temporal and septal limbs, of roughly equal length, joined at an oblique angle (Fig. 1B). Slicing angle is fairly uniform within each limb. In the experiments described above, each hippocampus was oriented on the tissue chopper so that the slices from the temporal limb were cut roughly perpendicular to the long axis of that limb (Fig. 1C). As a result, the remaining slices, from the septal limb, were cut at an angle to the septal axis. Some authors have suggested that recurrent inhibition may be better preserved in angled slices3'18, although this has never actually been demonstrated. Thus, the faster burst rate in the temporal slices could have resulted from the greater disruption of recurrent inhibitory pathways in these slices. We performed the following experiments to address these two possibilities directly. Slices were prepared from four additional hippocampi. Each was positioned on the chopper so that septal slices were cut perpendicular to the septal axis and temporal slices were cut at an angle to the temporal axis (Fig. 1D). Each hippocampus was sliced starting at the septal end. The order in which slices were studied was balanced as before. Statisfactory evoked responses were obtained from 47 of 53 slices. The relationship between burst frequency and slice number (Fig. 3A, closed circles) was the same as that seen in the previous group of slices. As the results in the two groups appeared identi-
cal, the data were merged. The mean (+ S.E.M.) burst frequency at each slice location is shown in Fig. 3B. At slice three, mean and standard error were calculated both without (closed circle) and with (open circle) the one outlying value, and demonstrate that the distinct peak at this location is unaffected by exclusion of the outlier. The present results demonstrate that at least one form of neuronal activity, epileptiform bursting of area CA3, varies markedly along the septotemporal axis of the hippocampus in a consistent fashion. These results have two important implications. First, they demonstrate that site of slice origin along the septotemporal axis is an important confounding variable in the design and interpretation of in vitro studies of hippocampal neuronal activity. This variable must be controlled for when assigning slices from naive animals to control and experimental groups, as well as when comparing neuronal activity in slices from control and experimental animals, e.g. between control and kindled animals. Second, these results reinforce the notion that the temporal portion of the hippocampus may participate more intensely in the development and propagation of seizures in vivo than the more commonly studied septal hippocampus, and suggest that increased attention be directed toward temporal hippocampus in both in vivo and in vitro studies of epileptic activity. This notion, that temporal hippocampus is especially prone to seizure activity, is derived from a number of in vivo studies of limbic seizures. In limbic seizures induced by systemic kainic acid, metabolic and electrical mapping using 2-deoxyglucose and multiple depth electrodes showed that the brain region affected earliest and at the lowest dose was the hippocampus, with ventral (temporal) hippocampus substantially more affected than dorsal (septal) hippocampus 1°. In limbic seizures induced by prolonged chemical or electrical stimulation of medial entorhinal cortex, increased metabolic activity was again greater in ventral than in dorsal hippocampus 2. In the kindling model of limbic epilepsy, rats were shown to kindle more rapidly with ventral than with dorsal hippocampal stimulation 14. Finally, hippocampal damage resulting from prolonged status epilepticus in previously kindled animals was more extensive in ventral than dorsal hippocampus 1~. Thus, both our
173 study and these in vivo studies suggest the t e m p o r a l hippocampus m a y play a m o r e active role than septal hippocampus in some forms of limbic epilepsy. The mechanisms underlying this t e m p o r a l predominance are unknown. The responsible factors could be extrinsic to the h i p p o c a m p u s , e.g. qualitative or quantitative s e p t o t e m p o r a l differences in afferent pathways transmitting seizure activity to the hippocampus. Alternatively, s e p t o t e m p o r a l differences in intrinsic neuronal elements could account for the results of these studies. Some support for intrinsic circuits is provided by the kainic acid study cited above 1°. A t the lowest dose of kainic acid which induced seizures, the pattern of increased metabolic activity was not only greater t e m p o r a l l y than septally, in the hippocampus, but was most striking in area C A 3 , and next
most in CA1. Activity in entorhinal cortex, dentate gyrus and medial septum r e m a i n e d at control levels. O u r results d e m o n s t r a t e that elements which could account for these s e p t o t e m p o r a l differences are preserved within the h i p p o c a m p a l slice. This is further s u p p o r t e d by a recent study, c o m p l e m e n t a r y to our own, which found that stimulus-evoked epileptiform potentials in area CA1 were m o r e complex in t e m p o ral than in septal h i p p o c a m p a l slices 6. H o w e v e r , this still leaves open the question of w h e t h e r these elements are afferent terminals and their transmitters or whether they are local neuronal circuits within the hippocampus.
1 Brazier, M.A.B., Regional activities within the human hippocampus and hippocampal gyrus, Exp. Neurol., 26 (1970) 354-368. 2 Collins, R.C., Tearse, R.G. and Lothman, E.W., Functional anatomy of limbic seizures: focal discharges from medial entorhinal cortex in rat, Brain Research, 280 (1983) 25-40. 3 Dingledine, R. and Langmoen, I.A., Conductance changes and inhibitory actions of hippocampal recurrent IPSPs, Brain Research, 185 (1980) 277-287. 4 Elul, R., Regional differences in the hippocampus of the cat. I. Specific discharge patterns of the dorsal and ventral hippocampus and their role in generalized seizures, Electroencephalogr. Clin. Neurophysiol., 16 (1964) 470-488. 5 Fried, P.A., Limbic system lesions in rats: differential effects in an approach-avoidance task, J. Comp. Physiol. Psychol., 74 (1971) 349-353. 6 Gilbert, M.E., Racine, R.J. and Smith, G.K., Epileptiform burst responses in ventral vs dorsal hippocampal slices, Soc. Neurosci. Abstr., 11 (1985) 1229. 7 Hughes, K.R., Dorsal and ventral hippocampus lesions and maze learning: influence of preoperative environment, Can. J. Psychol., 19 (1965) 325-332. 8 Koreli, A., Influences of dorsal and ventral hippocampus on hypothalamic self-stimulation, Physiol. Behav., 19 (1977) 713-717. 9 Lanier, L.P. and Isaacson, R.L., Activity changes related to the location of lesions in the hippocampus, Behav. Biol., 13 (1975) 59-69. 10 Lothman, E.W. and Collins, R.C., Kainic acid induced limbic seizures: Metabolic, behavioral, electroencephalographic and neuropathological correlates, Brain Research, 218 (1981) 299-318. ll Mclntyre, D.C., Edson, N. and Nathanson, D., A new
model of partial status epilepticus based on kindling, Brain Research, 250 (1982) 53-63. 12 Myhrer, T., Locomotor, avoidance, and maze behavior in rats with selective disruption of hippocampal output, J. Comp. Physiol. Psychol., 89 (1975) 759-777. 13 Nadel, L., Dorsal and ventral hippocampal lesions and behavior, Physiol. Behav., 3 (1968) 891-900. 14 Racine, R., Rose, P.A. and Burnham, W.M., Afterdischarge thresholds and kindling rates in dorsal and ventral hippocampus and dentate gyrus, Can. J. Neurol. Sci., 4 (1977) 273-278. 15 Siegel, A. and Flynn, J.P., Differential effects of electrical stimulation and lesions of the hippocampus and adjacent regions upon attack behavior in cats, Brain Research, 7 (1968) 252-267. 16 Somjen, G.G., Extracellular potassium in the mammalian central nervous system, Annu. Rev. Physiol., 41 (1979) 159-177. 17 Stevens, R. and Cowey, A., Effects of dorsal and ventral hippocampal lesions on spontaneous alternation, learned alternation and probability learning in rats, Brain Research, 52 (1973) 203-224. 18 Struble, R.G., Desmond, N.L. and Levy, W.B., Anatomical evidence for interlamellar inhibition in the fascia dentata, Brain Research, 152 (1978) 580-585. 19 Watson, R.E., Edinger, H.M. and Siegel, A., A [14C]2-deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. III. The hippocampal formation, Brain Res. Rev., 5 (1983) 133-176. 20 Wilson, W.A., Bragdon, A.C. and Taylor, D.M., Potassium-induced epileptiform activity in hippocampal area CA3 varies markedly along the septo-temporal axis, Soc. Neurosci. Abstr., 11 (1985) 1324.
This work was s u p p o r t e d by the V e t e r a n s A d m i n istration, and N I H Grants G M 47105 and NS 17771.