HIPPOCAMPUS, VOL. 1, NO. 4, PAGES 399-404, OCTOBER 1991
Electrophysiological Characterization of Associational Pathway Terminating on Dentate Gyrus Granule Cells in the Rat Jonathan W. Bekenstein and Eric W. Lothman Department of Neurology, University of Virginia Health Sciences Center, Charlottesville, VA 22908 U.S.A.
ABSTRACT The functional topography and parameters of excitation and inhibition were determined in the in situ associational pathway of the rat dentate gyrus. The functional topography was found to be consistent with previous anatomical studies. The greatest amplitude population spikes and the strongest paired-pulse inhibition were generated with the stimulating electrode placed in the hilus at least 1.5 mm caudal to the ipsilateral dentate gyrus recording electrode. With this standard electrode configuration, neither long-term potentiation of the population spike nor of the population excitatory postsynaptic potential occurred. Hilar associational pathway activation of dentate gyrus granule cells elicited paired-pulse responses similar to those produced in granule cells by perforant path stimulation. Thus, the associational pathway provides another way to assess dentate granule cell function electrophysiologically. Key words: hippocampus, entorhinal cortex, plasticity, ontogeny, excitation, inhibition
The dentate gyrus of the hippocampal formation is a highly axis of the hippocampal formation (Lynch et al., 1976; laminated structure. Its granule cells are organized in a nar- Amaral and Witter, 1989). This extensive pathway outnumrow layer (the stratum granulosum), and the dendrites of gran- bers the contralateral commissural pathway 3: 1 in number of ule cells are arranged in a separate layer (the stratum mole- fibers and, presumably, synaptic sites that are located within culare) that contains a number of afferent fiber systems. The the inner one-third of the molecular layer (Laatsch and segregation of afferent fibers on distinct regions of granule Cowan, 1967; Gottlieb and Cowan, 1972; McWilliams and cell dendrites makes this an ideal system in which to study Lynch, 1978). Except for the more extensive number of ipsilateral, asthe function of discrete pathways. The study of responses elicited by activation of particular fiber groups can serve as sociational fibers, the associational pathway is frequently a foundation for the examination of changes within these considered to be a nearly identical ipsilateral analogue of the pathways that result from selective lesions of chosen affer- commissural pathway. A laminar study of the electrophysiological properties exhibited by the associational and coments to the granule cells. An associational pathway in the hippocampal formation missural systems has corroborated similar anatomical distriarises from polymorphic hilar neurons of the dentate gyrus. butions of the two systems (Steward et al., 1977). In that Axons from these neurons then terminate on the dendrites of study, areas of focal, maximal negativity were restricted to ipsilateral granule cells in the inner third of the stratum mo- the regions of the stratum moleculare where the respective leculare (Zimmer, 1971; Segal and Landis, 1974; Amaral, terminals are known to lie (Lorente de NO, 1934; Hjorth-Si1978; Swanson et al., 1978; Laurberg and Sorensen, 1981). monsen, 1973; Amaral and Witter, 1989). The associational pathway projects to the same region of the Though the anatomical distribution of the associational-tomolecular layer as does the commissural pathway, which granule cell pathway has been well described (Lorente de NO, arises from the contralateral hilar neurons and travels through 1934; Hjorth-Simonsen, 1973; Amaral and Witter, 1989) little the ventral hippocampal commissure (Laatsch and Cowan, is known about its electrophysiological characteristics. Stew1967; Swanson et al., 1981). It has been reported that at least ard et al. (1977) reported that both the associational and comsome of the hilar neurons have both associational and com- missural systems were capable of demonstrating paired-pulse facilitation of extracellularly recorded population excitatory missural branches (Laurberg and Sorensen, 1981). Anatomical tract-tracing studies have also shown that the postsynaptic potentials (pEPSPs). Wong and Prince (1988) associational pathway extends longitudinally for a consid- have reported some functional features of this system. Barerable distance (greater than 2 mm) along the rostrocaudal tesaghi et al. (1983) and Nunez Filipe et al. (1989) have studied longitudinal projections in the hippocampus, including the Correspondence and reprint requests to Eric W . Lothman, Depart- system under consideration. The present study investigated the monosynaptic, excitament of Neurology-Box 394, University of Virginia Health Sciences tory properties of the associational pathway introduced Center, Charlottesville,VA 22908 U.S.A.
399
400 HZPPOCAMPUS VOL. 1, NO. 4, OCTOBER 1991 above, including its topographical representation. This was accomplished by stimulating hilar cells in a number of rostrocaudal locations and recording extracellularly from dentate granule cells to register pEPSP slopes and population spike (PS) amplitudes. The ability to produce long-term potentiation (LTP) in this pathway was also investigated. Last, paired-pulse inhibition, an example of local circuit function, was examined for its dependence upon stimulus intensity and interpulse interval. The data so obtained validate functional aspects of the associational system inferred from prior anatomical studies. Moreover, they also establish the associational pathway as a convenient experimental means for assessing monosynaptic and local circuit functions of dentate granule cells.
MATERIALS A N D METHODS Sprague-Dawley rats (275-300 g) were anesthetized with urethane (1.2 mg/kg, intraperitoneal). A concentric, bipolar stimulating electrode (SNEX 200, tip diameter 100 pm, shaft diameter 250 pm, intraelectrode distance 750 pm, Rhodes Medical Instruments, Tujunga, CA) was placed in the regio inferiodhilar region (-4.5 mm posterior [AP] and 2.5 mm lateral [ML] to bregma, dorsoventral [DV] - 2.5 to - 3.0 mm from the dura, with the incisor bar at zero). A glass capillary recording electrode (2-6 M a resistance) filled with 2 M NaCl and 5% fast green was placed in the ipsilateral dentate gyrus (-3.0 mm AP and 1.5-2.0 mm ML to bregma, -2.8 to -3.0 mm beneath the dura). The positions of the hilar stimulating and dentate gyrus recording electrodes were determined on the basis of preliminary experiments that studied various relative positions of the electrodes along the rostrocaudal axis (see Results). A second stimulating electrode (bipolar “twist,” constructed from Teflon-coated stainless steel wire, 0.05” in diameter) was placed in the ipsilateral angular bundle (AP - 8.1 mm, ML 4.4 mm, DV 2.7-3.0 mm to dura). Animals were maintained at 37°C with a heating pad controlled by a feedback device and a rectal thermoprobe. Following electrophysiological data collection, recording sites and stimulation sites were marked with established methods (Stringer et al., 1989). Responses in the dentate gyrus were activated monosynaptically by stimulation of the ipsilateral hilar region with a constant voltage source, controlled by a digital timer. Final electrode depths were adjusted to maximize evoked field potentials. These positions were allowed to stabilize for 30 minutes prior to data collection. Analyses of pEPSP slopes (mV1 ms) and PS amplitudes (mV) were performed by an analogto-digital interface with a microcomputer and software by Aitken (1985). Four records were averaged for each particular combination of stimulus parameters. Paired or single stimuli were given every 10 seconds. All measurements were made in the stratum granulosum of dorsal blade of the dentate gyrus. The recording electrodes were positioned in the dentate gyrus with a hydraulic microdrive with l pm resolution, maximizing PS evoked by angular bundle stimulation. Paired-pulse inhibition was measured according to techniques described elsewhere (Kapur et al., 1989; Stringer and Lothman, 1989). Input-output curves of PS amplitude vs. stimulus intensity were constructed for paired stimuli of identical intensity at a fixed interpulse in-
terval. At lower stimulus intensities, the ratio of the second (test) PS amplitude (PS[T]) to the first (conditioning, or PS[C]) was variable, often yielding facilitation. However, as stimulus intensity was increased, inhibition invariably appeared, and above a certain intensity, the ratio reached a constant or “plateau” value. Thus, under these plateau conditions the ratio PS(T)/PS(C) depended only on the interpulse interval. For experiments presented below, the stimulus intensity was set to elicit a PS that was 80% of maximal, a setting that was always above the intensity at which the PS(T)/PS(C) ratio was constant at a fixed interpulse interval. The interpulse intervals were then varied 20-500 ms (20-, 30-, 50-, 70-, loo-, 200-, 300-, and 500-ms intervals). PS(T)/PS(C) ratios less than 1.O indicated inhibition; ratios greater than 1.O indicated potentiation. Stimulus parameters for inducing LTP were similar to those in a previous report (Douglas, 1977). Briefly, a tetanizing stimulus of 8 trains (8 pulses at 333 Hzhrain), 10 seconds apart, was given at a stimulus intensity that produced a population spike of maximal amplitude. Input-output curves (PS amplitude or pEPSP slope plotted vs. stimulus intensity) were obtained prior to and 30 minutes following the tetanizing stimulation.
Analysis of LTP Indices of PS long-term potentiation (PS-LTP) and indices of pEPSP long-term potentiation (pEPSP-LTP) were derived from complete input-output curves (PS amplitude vs. stimulus intensity or pEPSP slope vs. stimulus intensity). These methods were adopted from those typically used for studying LTP use in the perforant path-dentate granule cell system. Thirty minutes following the tetanizing stimulus, another set of input-output curves was collected. The maximum values of PS amplitude and of pEPSP slope were measured from baseline input-output curves. To assess LTP for PS and for pEPSP, an index of PS-LTP and an index of pEPSP-LTP were defined from measurements related to the shift of the posttetanization input-output curves, relative to the pretetanization input-output curves. The index of PS-LTP was defined by the percentage change in the stimulus intensity (SI) necessary to produce a PS of one-half maximal amplitude. This normalized index of PSLTP was defined as:
@Iso baseline
-
SIso poststimulus)/SIso baseline
The index for pEPSP-LTP was chosen as the percentage increase in the maximal pEPSP slope 30 minutes following the tetanizing stimulus. This index was defined as follows: (potentiated pEPSP,,,
- baseline pEPSP,,,)/ baseline EPSP,,,
RESULTS Anatomical studies have indicated that associational fibers project longitudinally along the septotemporal axis of the hippocampal formation (Amaral and Witter, 1989). To examine the functional topography of associational inputs onto granule
DENTATE ASSOCIATIONAL PATHWAY PHYSIOLOGY / Bekenstein and Lothman
Tr-P-
W
a
RECORD
A
a 4
"I
401
I
r 1
6
0
b
41)
C" W
n
3a
In
a
wp
1
00
B
10
20
30
STIMULUS INTENSIW (VOLTS)
1I
40
Fig. 2. Input-output curves for (A) PS amplitude vs. stimulus intensity and for (B) pEPSP slope vs. stimulus intensity from the same animal. Maximal values of pEPSP slope and PS amplitude were derived from such curves. Polarity: negativity down.
B Fig. 1 . Examples of paired-pulse responses with associational input to dentate granule cells. (A) Events obtained with recording electrode fixed in the dentate gyrus at AP -3.0 mm, lateral 2.0 mm. Responses were recorded with the stimulating electrode placed in three locations: a, 1.5 mm anterior to the recording electrode, (AP - 1.5 mm, lateral 2.5 mm); b, at the same rostrocaudal position as the recording electrode (AP - 3.0 mm, lateral 2.5 mm); c, I .5 mm posterior to the recording electrode (AP -4.5, lateral 2.5). Responses are of the maximal amplitude achievable at each stimulating electrode position. *Recording electrode position. (B) Pairedpulse facilitation. Stimulation at point c, but lower intensity, produced submaximal response, consisting of only pEPSP. In all instances (a and b), the interpulse interval was 20 ms; calibration pulse (at beginning of each trace) was 5 mV and of 1 ms duration. Polarity: negativity down.
cells, a number of stimulating and recording electrode positions were tested for their ability to generate pEPSP, to produce PS, to demonstrate LTP, and to demonstrate pairedpulse inhibition. Results are summarized in Figure 1A. In the experiment, the recording electrode was placed at a fixed position in the dentate gyrus - 3.0 mm posterior to bregma, and various stimulating sites were tested by placing the stimulating electrode in selected ipsilateral sites. When the stimulating electrode was placed 1.5 mm rostra1 to the recording electrode, a pEPSP was always seen, but no population spike could be elicited. Furthermore, there was an appreciable facilitation of the pEPSP at an interpulse interval of 20 ms (Fig. 1A-A). When the stimulating electrode was placed in the same an-
teroposterior plane as the recording electrode, the pEPSP were larger, and PS were generated readily (Fig. IA-B). With this electrode configuration, paired-pulse inhibition was only infrequently observed. No facilitation of PS was observed. When the stimulating electrode was placed caudal to the recording electrode, even larger pEPSP and PS were produced. However, by lowering the stimulus intensity it was possible to produce pEPSP without PS. With paired stimuli under these conditions, facilitation was seen (Fig. 1B). These observations demonstrate that evoked responses are not antidromic since they appeared without PS. In addition, the electrode configuration placed the stimulation outside of the projection area of the mossy fibers from the granule cells being monitored by the recording electrode (Amaral and Witter, 1989). The facilitation of the potentials provides additional information that they are pEPSP. In addition, the strength of paired-pulse inhibition reached a maximum when the stimulating electrode was positioned I .5-2.0 mm caudal to the recording microelectrode (Fig. IA-C). Thus, all subsequent electrophysiological studies were standardly performed with the stimulating electrode 1.5 mm caudal to the recording electrode. Input-output curves (PS amplitude vs. stimulus intensity; pEPSP slope vs. stimulus intensity) were generated for each animal (Fig. 2). With the standard electrode positions, excitatory responses were consistent. The maximal population spike amplitude was 7.3 2 0.4 mV (mean & SEM, n = 3, and the maximal pEPSP slope was 6.6 2 1.4 mV/ms (n = 5). To confirm the position of the recording electrode within the dentate gyrus, responses were elicited from the ipsilateral
402
HZPPOCAMPUS VOL. I , NO. 4, OCTOBER 1991 PERFORANT PATH TO DENTATE GYRUS GRANULE CELLS n
0-0
FIRSTPS
E W
0-0
SECOND PS
>
,o-0 0 0’
/ 0-0
A
I
HILAR ASSOCIATIONAL PATHWAY
0’ -0’ v)
a 10 - 20
T
A
30
40
50
60
70
80
90
STIMULUS INTENSITY (VOLTS)
B Fig. 3 . Responses in an animal with the recording electrode in the dentate gyrus produced by (A) ipsilateral perforant pathway activation and (B) ipsilateral associational pathway stimulation. Interpulse interval was 20 ms and the calibration pulse was 5 mV, 1 ms. Polarity: negativity down.
perforant path fibers by stimulating the angular bundle (Fig. 3) with a second stimulating electrode. Long-term potentiation did not occur with hilar stimulation. Thirty minutes following the application of a tetanizing stimulus to the hilar region of an intensity that produced a granule cell PS of maximal amplitude, the curves for both pEPSP and PS vs. stimulus intensity were unchanged from baseline. LTP also was not observed with any combination of positions of hilar stimulating and dentate gyms recording electrodes nor with other (ISO- and 200-Hz) tetanizing frequencies (data not shown). However, with electrodes in the standard configuration, even when LTP in the associational pathway was not produced, PS-LTP and pEPSP-LTP of perforant path-granule cell synapses could be demonstrated at the same recording site. There also was n o effect of the production of LTP in the perforant path upon the input-output curves for associational pathway activation of granule cells (data not shown). Paired-pulse responses within the associational system were first examined for dependence upon stimulation intensity (Fig. 4). Paired-pulse responses were measured as the ratios of the population spike amplitude elicited by a single stimulus (PS[T]) following a single stimulus of identical intensity (PS[C]). At a fixed interpulse interval of 20 ms, facilitation was typically observed with hilar stimulus intensities that produced a small conditioning PS (PS[C]). Stronger stimulus intensities produced inhibition (ratios < I .O). The ratios of the test PS to the conditioning PS remained constant after a certain stimulus intensity was reached, indicating that PPI was independent of stimulus intensity under defined conditions. Paired-pulse inhibition was also studied for its dependence upon the interpulse interval (IPI). With the stimulus intensity set so that the ratio of PS amplitudes depended only on the interpulse interval, paired-pulse inhibition was maximal at an IPI of 20 ms (Table 1). Paired-pulse inhibition was evident
A
A\*-A - ~ - A - A - A - A - A I
0
10
B
20
30
40
50
60
70
80
3
STIMULUS INTENSITY (VOLTS)
Fig. 4. Relationship of paired-pulse response to stimulus intensity. (A) Population spike amplitude (ordinate) for stimuli of increasing intensity (abscissa). First (conditioning) population spike (open circles) and second (test) population spike (filled circles) evoked by paired-pulse stimuli are given. (B) Ratios of test population spike to conditioning population spike (ordinate) plotted vs. stimulus intensity. Note pairedpulse inhibition (ratio > 1) for lower intensity and paired-pulse inhibition (ratio < 1 ) for higher intensity.
with interstimulus intervals as long as SO ms. At IPIs of 100 and 200 ms, facilitation of the second PS occurred. A second phase of paired-pulse inhibition occurred at IPIs between 300 and SO0 ms.
DISCUSSION The functional properties of the ipsilateral associational pathway of the hippocampal formation have not been studied to any degree. Steward et al. (1977) reported the ability of this system to demonstrate paired-pulse facilitation of the pEPSP at submaximal stimulus intensities. In that study, the exact position of the associational pathway stimulating electrode was not described, except that it was slightly caudal to the recording electrode. In the same study it was reported that heterosynaptic potentiation between commissural, associational, or entorhinal afferents did not occur. No data
DENTATE ASSOCIATIONAL PATHWAY PHYSIOLOGY / Bekenstein and Lothman
403
Table 1. PS(T)/PS(C)Ratios of Paired-pulse Responses From Associational Fiber Activation of Dentate Granule Cells N
5
20 ms
.21
2
.I1
30 ms
.31
?
.07
50 ms
.81
?
.I0
70 ms
1.11 f .05
were presented for the ability of the associational pathway to produce PS or to undergo LTP, either by exhibiting potentiated PS or potentiated pEPSP slopes. The present study confirms electrophysiologically the topography of the associational pathway as defined anatomically (Blackstad, 1956; Zimmer, 1971; Berger et al., 1980; Amaral and Witter, 1989). The greatest amplitude PS were generated when the stimulating electrode was at least I .5mm caudal to the recording electrode. This electrode configuration also generated the strongest paired-pulse inhibition. Thus, the present electrophysiological data validate the recent conceptualization of the hippocampal formation as a three-dimensional information processor along its longitudinal axis as well as in distinct lamellae (Amaral and Witter, 1989). This concept was advanced on the basis of anatomical work. By consistently placing the electrodes in a particular arrangement as described above, robust, monosynaptic responses that produced PS were elicited. These responses permitted an investigation of PS-LTP and of pEPSP-LTP. Neither form of LTP could be demonstrated in this pathway. The absence of LTP in the associational pathway might have been due to the small number of afferent fibers activated by electrical stimulation (in comparison to perforant path fibers activated by angular bundle stimulation). The associational system also seems to be similar to the commissural pathway with respect to the number of GABAergic projection neurons that arise in the hilus, as determined from anatomical studies (Seress and Ribak, 1983). Those associational, GABAergic projection neurons could serve to directly inhibit dentate granule cells by feedforward inhibition and result in prevention of LTP. This possibility has been suggested for the commissural pathway to the dentate, since a blockade of inhibition allowed the expression of LTP in the dentate commissural system (Steward et al., 1990). Commissural stimulation has also been shown to prevent LTP and the associated increase in glutamate release in the perforant pathgranule cell synapses (Lynch et al., 1989). A number of observations confirmed the separation of the associational pathway from the perforant path fibers. PS-LTP and pEPSP-LTP of the perforant path-granule cell synapses were demonstrated at the same recording site when LTP in the associational pathway was not produced. In addition, the production of LTP in the perforant path had no effect upon the input-output curves for associational pathway activation of granule cells. Other tetanizing parameters might produce LTP within the associational pathway. However, stimulus parameters that routinely produce LTP in the perforant pathgranule cell synapses do not do so in the associational pathway, pointing to a distinction in this afferent system. Paired-pulse inhibition in the associational pathway was robust. While the methods were biased to record feedback inhibition, feedforward activation of many inhibitory neurons in the hilar region could not be excluded (Buzsaki, 1984;
100 ms
1.18
?
.05
200 ms
1.06 ? .I3
300 ms
.85
?
.I4
500 ms
.76
2
.I6
Kapur et al., 1989). Paired-pulse inhibition in the dentate associational pathway is similar in its strength and in duration (IPI up to 70 ms) to paired-pulse inhibition in the CAI region elicited via stimulation of the contralateral CA3 region (Kapur et al., 1989). In contrast, paired-pulse inhibition of dentate gyrus granule cells from perforant path stimulation is of comparable strength but of shorter duration (Stringer and Lothman, 1989). Additionally, hilar associational pathway activation elicited similar paired-pulse responses to those produced in granule cells by perforant path stimulation at other IPI. Thus, the same pattern of a period of inhibition (20-70 ms IPIs), facilitation (100-200 ms IPIs), and inhibition (300-500 ms IPIs) is produced in granule cells by two distinct pathways . The associational pathway as described here will serve as another means for electrophysiological assessment of dentate granule cell function along with the commissural and perforant pathways. The present studies performed in vivo provide a groundwork for study in this pathway under other physiological conditions, such as following deafferentation of the dentate gyrus or during epileptiform activity. The in vivo nature of the present study characterizing of this system should be noted, since the longitudinal excursion of the associational fibers necessarily dictates their disruption in hippocampal slice preparations.
ACKNOWLEDGMENTS The authors thank Rose Powell for assistance in preparing the manuscript. Supported in part by UPS Grants NS21617 and NS25605.
References Aitken, P. J. (1985) Kainic acid and penicillin: Differential effects of excitatory and inhibitory interactions in the CAI region of the hip-
pocampal slice. Brain Res. 325:261-269. Amaral, D. G. (1978) A Golgi study of cell types in the hilar region of the hippocampus in the rat. J. Comp. Neurol. 182:851-914. Amaral, D. G . , and M. P. Witter (1989) The three-dimensional organization of the hippocampal formation: A review of anatomical data. Neuroscience 33:571-591. Bartesaghi, R., T. Gessi, and L. Sperti (1983) Intralamellar transfer of impulses in the hippocampal formation. Exp. Neurol. 82550567. Berger, T. W., S. Semple-Rowland, and J. L. Basset (1980) Hippocampal polymorph neurons are the cells of origin for ipsilateral association and commissural afferents to the dentate gyrus. Brain Res. 215:329-336. Blackstad, T. W . (1956)Commissural connections of the hippocampal region in the rat, with special reference to their mode of termination. J . Comp. Neurol. 105:417-537. Buzsaki, G. (1984) Feed-forward inhibition in the hippocampal formation. Progr. Neurobiol. 22:131-153. Douglass, R. M. (1977) Long-lasting synaptic potentiation in the rat dentate gyrus following brief high frequency stimulation. Brain Res. 126:361-365.
404
HZPPOCAMPUS VOL. 1, NO. 4, OCTOBER 1991
Gottlieb, D. I., and W. M. Cowan (1972) Evidence for a temporal factor in the occupation of available synaptic sites during the development of the dentate gyrus. Brain Res. 41:452-456. Hjorth-Simonsen, A. (1973) Some intrinsic connections of the hippocampus in the rat: An experimental analysis. J. Comp. Neurol. 147:145-162. Kapur, J., J. L. Stringer, and E . W. Lothman (1989) Evidence that repetitive seizures in the hippocampus cause a lasting reduction of GABAergic inhibition. J. Neurophysiol. 61 :417-426. Laatsch, R. H., and W. M. Cowan (1967) Electron microscopic studies of the dentate gyrus of the rat: 11. Degeneration of commissural afferents. J. Comp. Neurol. 130:241-262. Laurberg, S . , and K . E. Sorensen (1981) Associational and commissural collaterals of neurons in the hippocampal formation (hilus fasciae dentatae and subfield CA3). Brain Res. 212:287-300. Lorente de N6, R . (1934) Studies on the structure of the cerebral cortex: 11. Continuation of the study of the ammonic system. J. Psychol. Neurol. 46:113-177. Lynch, G., S. Deadwyler, and C. Cotman (1973) Postlesion axonal growth produces permanent functional connections. Science 180: 1364-1366. Lynch, G . , C. Gall, G. Rose, and C. Cotman (1976) Changes in the distribution of the dentate gyrus associational system following unilateral or bilateral entorhinal lesions in the adult rat. Brain Res. 110:57-71. Lynch, M. A,, M. L . Errington, and T. V. P. Bliss (1989) The increase in [3H]glutamaterelease associated with long-term potentiation in the dentate gyrus is blocked by commissural stimulation. Neurosci. Lett. 103:191-196. McWilliams, R., and G. Lynch (1978) Terminal proliferation and synaptogenesis following partial deafferentation: The reinnervation of the inner molecular layer of the dentate gyms following removal of its commissural afferents. J . Comp. Neurol. 180:581-616. Nunez Filipe, C., J. P. P i n , V . M. Ferrandez de Lima, and F. H. Lopes da Silva (1989) Propagation of afterdischarges along the septo-temporal axis of the rat hippocampus: A quantitative analysis. Electroencephalogr. Clin. Neurophysiol. 73: 172- 178.
Segal, M., and S. Landis (1974) Afferents to the hippocampus of the rat studied with the method of retrograde transport of horseradish peroxidase. Brain Res. 78:l-15. Seress, L., and C. E. Ribak (1983) GABAergic cells in the dentate gyrus appear to he local circuit and projection neurons. Exp. Brain Res. 50:173-182. Steward, O., W. F. White, and C . W. Cotman (1977) Potentiation of the excitatory synaptic action of commissural, associational and entorhinal afferents to dentate granule cells. Brain Res. 134:551560. Steward, O., R. Tomasulo, and W. B. Levy (1990) Blockade of inhibition in a pathway with dual excitatory and inhibitory action unmasks a capability for LTP that is otherwise not expressed. Brain Res. 516:292-300. Stringer, J . L., and E. W. Lothman (1989) Repetitive seizures cause an increase in paired-pulse inhibition in the dentate gyrus. Neurosci. Lett. 105:91-95. Stringer, J. L., J. M. Williamson, and E. W. Lothman (1989) Induction of paroxysmal discharges in the dentate gyrus: Frequency dependence and relationship to afterdischarge production. J . Neurophysiol. 62: 126-135. Swanson, L. W., J. M. Wyss, and W. M. Cowan (1978) An autoradiographic study of the organization of intrahippocampal association pathways in the rat. J . Comp. Neurol. 181:681-716. Swanson, L. W., P. E. Sawchenko, and W. M. Cowan (1981) Evidence for collateral projections by neurons in ammon’s horn, the dentate gyrus, and the subiculum: A multiple retrograde labeling study in the rat. J. Neurosci. 1548-559. West, J . R., S. Deadwyler,C. W. Cotman,andG. Lynch(l975)Timedependent changes in commissural field potentials in the dentate gyrus following lesions of the entorhinal cortex in adult rats. Brain Res. 97:2 15-233. Wong, B. Y . , and D. A. Prince (19881 Hilar neurons: Gatekeepers of the dentate gyrus. SOC.Neurosci. Abstr. 14:246. Zimmer, J . (1971) Ipsilateral afferents to the cornmissural zone of the fascia dentata, demonstrated in decommissurated rats by silver impregnation. J . Comp. Neurol. 142:393-416.
Electrophysiological Characterization of Associational Pathway Terminating on Dentate Gyrus Granule Cells in the Rat Jonathan W. Bekenstein and Eric W. Lothman Department of Neurology, University of Virginia Health Sciences Center, Charlottesville, VA 22908 U.S.A.
ABSTRACT The functional topography and parameters of excitation and inhibition were determined in the in situ associational pathway of the rat dentate gyrus. The functional topography was found to be consistent with previous anatomical studies. The greatest amplitude population spikes and the strongest paired-pulse inhibition were generated with the stimulating electrode placed in the hilus at least 1.5 mm caudal to the ipsilateral dentate gyrus recording electrode. With this standard electrode configuration, neither long-term potentiation of the population spike nor of the population excitatory postsynaptic potential occurred. Hilar associational pathway activation of dentate gyrus granule cells elicited paired-pulse responses similar to those produced in granule cells by perforant path stimulation. Thus, the associational pathway provides another way to assess dentate granule cell function electrophysiologically. Key words: hippocampus, entorhinal cortex, plasticity, ontogeny, excitation, inhibition
The dentate gyrus of the hippocampal formation is a highly axis of the hippocampal formation (Lynch et al., 1976; laminated structure. Its granule cells are organized in a nar- Amaral and Witter, 1989). This extensive pathway outnumrow layer (the stratum granulosum), and the dendrites of gran- bers the contralateral commissural pathway 3: 1 in number of ule cells are arranged in a separate layer (the stratum mole- fibers and, presumably, synaptic sites that are located within culare) that contains a number of afferent fiber systems. The the inner one-third of the molecular layer (Laatsch and segregation of afferent fibers on distinct regions of granule Cowan, 1967; Gottlieb and Cowan, 1972; McWilliams and cell dendrites makes this an ideal system in which to study Lynch, 1978). Except for the more extensive number of ipsilateral, asthe function of discrete pathways. The study of responses elicited by activation of particular fiber groups can serve as sociational fibers, the associational pathway is frequently a foundation for the examination of changes within these considered to be a nearly identical ipsilateral analogue of the pathways that result from selective lesions of chosen affer- commissural pathway. A laminar study of the electrophysiological properties exhibited by the associational and coments to the granule cells. An associational pathway in the hippocampal formation missural systems has corroborated similar anatomical distriarises from polymorphic hilar neurons of the dentate gyrus. butions of the two systems (Steward et al., 1977). In that Axons from these neurons then terminate on the dendrites of study, areas of focal, maximal negativity were restricted to ipsilateral granule cells in the inner third of the stratum mo- the regions of the stratum moleculare where the respective leculare (Zimmer, 1971; Segal and Landis, 1974; Amaral, terminals are known to lie (Lorente de NO, 1934; Hjorth-Si1978; Swanson et al., 1978; Laurberg and Sorensen, 1981). monsen, 1973; Amaral and Witter, 1989). The associational pathway projects to the same region of the Though the anatomical distribution of the associational-tomolecular layer as does the commissural pathway, which granule cell pathway has been well described (Lorente de NO, arises from the contralateral hilar neurons and travels through 1934; Hjorth-Simonsen, 1973; Amaral and Witter, 1989) little the ventral hippocampal commissure (Laatsch and Cowan, is known about its electrophysiological characteristics. Stew1967; Swanson et al., 1981). It has been reported that at least ard et al. (1977) reported that both the associational and comsome of the hilar neurons have both associational and com- missural systems were capable of demonstrating paired-pulse facilitation of extracellularly recorded population excitatory missural branches (Laurberg and Sorensen, 1981). Anatomical tract-tracing studies have also shown that the postsynaptic potentials (pEPSPs). Wong and Prince (1988) associational pathway extends longitudinally for a consid- have reported some functional features of this system. Barerable distance (greater than 2 mm) along the rostrocaudal tesaghi et al. (1983) and Nunez Filipe et al. (1989) have studied longitudinal projections in the hippocampus, including the Correspondence and reprint requests to Eric W . Lothman, Depart- system under consideration. The present study investigated the monosynaptic, excitament of Neurology-Box 394, University of Virginia Health Sciences tory properties of the associational pathway introduced Center, Charlottesville,VA 22908 U.S.A.
399
400 HZPPOCAMPUS VOL. 1, NO. 4, OCTOBER 1991 above, including its topographical representation. This was accomplished by stimulating hilar cells in a number of rostrocaudal locations and recording extracellularly from dentate granule cells to register pEPSP slopes and population spike (PS) amplitudes. The ability to produce long-term potentiation (LTP) in this pathway was also investigated. Last, paired-pulse inhibition, an example of local circuit function, was examined for its dependence upon stimulus intensity and interpulse interval. The data so obtained validate functional aspects of the associational system inferred from prior anatomical studies. Moreover, they also establish the associational pathway as a convenient experimental means for assessing monosynaptic and local circuit functions of dentate granule cells.
MATERIALS A N D METHODS Sprague-Dawley rats (275-300 g) were anesthetized with urethane (1.2 mg/kg, intraperitoneal). A concentric, bipolar stimulating electrode (SNEX 200, tip diameter 100 pm, shaft diameter 250 pm, intraelectrode distance 750 pm, Rhodes Medical Instruments, Tujunga, CA) was placed in the regio inferiodhilar region (-4.5 mm posterior [AP] and 2.5 mm lateral [ML] to bregma, dorsoventral [DV] - 2.5 to - 3.0 mm from the dura, with the incisor bar at zero). A glass capillary recording electrode (2-6 M a resistance) filled with 2 M NaCl and 5% fast green was placed in the ipsilateral dentate gyrus (-3.0 mm AP and 1.5-2.0 mm ML to bregma, -2.8 to -3.0 mm beneath the dura). The positions of the hilar stimulating and dentate gyrus recording electrodes were determined on the basis of preliminary experiments that studied various relative positions of the electrodes along the rostrocaudal axis (see Results). A second stimulating electrode (bipolar “twist,” constructed from Teflon-coated stainless steel wire, 0.05” in diameter) was placed in the ipsilateral angular bundle (AP - 8.1 mm, ML 4.4 mm, DV 2.7-3.0 mm to dura). Animals were maintained at 37°C with a heating pad controlled by a feedback device and a rectal thermoprobe. Following electrophysiological data collection, recording sites and stimulation sites were marked with established methods (Stringer et al., 1989). Responses in the dentate gyrus were activated monosynaptically by stimulation of the ipsilateral hilar region with a constant voltage source, controlled by a digital timer. Final electrode depths were adjusted to maximize evoked field potentials. These positions were allowed to stabilize for 30 minutes prior to data collection. Analyses of pEPSP slopes (mV1 ms) and PS amplitudes (mV) were performed by an analogto-digital interface with a microcomputer and software by Aitken (1985). Four records were averaged for each particular combination of stimulus parameters. Paired or single stimuli were given every 10 seconds. All measurements were made in the stratum granulosum of dorsal blade of the dentate gyrus. The recording electrodes were positioned in the dentate gyrus with a hydraulic microdrive with l pm resolution, maximizing PS evoked by angular bundle stimulation. Paired-pulse inhibition was measured according to techniques described elsewhere (Kapur et al., 1989; Stringer and Lothman, 1989). Input-output curves of PS amplitude vs. stimulus intensity were constructed for paired stimuli of identical intensity at a fixed interpulse in-
terval. At lower stimulus intensities, the ratio of the second (test) PS amplitude (PS[T]) to the first (conditioning, or PS[C]) was variable, often yielding facilitation. However, as stimulus intensity was increased, inhibition invariably appeared, and above a certain intensity, the ratio reached a constant or “plateau” value. Thus, under these plateau conditions the ratio PS(T)/PS(C) depended only on the interpulse interval. For experiments presented below, the stimulus intensity was set to elicit a PS that was 80% of maximal, a setting that was always above the intensity at which the PS(T)/PS(C) ratio was constant at a fixed interpulse interval. The interpulse intervals were then varied 20-500 ms (20-, 30-, 50-, 70-, loo-, 200-, 300-, and 500-ms intervals). PS(T)/PS(C) ratios less than 1.O indicated inhibition; ratios greater than 1.O indicated potentiation. Stimulus parameters for inducing LTP were similar to those in a previous report (Douglas, 1977). Briefly, a tetanizing stimulus of 8 trains (8 pulses at 333 Hzhrain), 10 seconds apart, was given at a stimulus intensity that produced a population spike of maximal amplitude. Input-output curves (PS amplitude or pEPSP slope plotted vs. stimulus intensity) were obtained prior to and 30 minutes following the tetanizing stimulation.
Analysis of LTP Indices of PS long-term potentiation (PS-LTP) and indices of pEPSP long-term potentiation (pEPSP-LTP) were derived from complete input-output curves (PS amplitude vs. stimulus intensity or pEPSP slope vs. stimulus intensity). These methods were adopted from those typically used for studying LTP use in the perforant path-dentate granule cell system. Thirty minutes following the tetanizing stimulus, another set of input-output curves was collected. The maximum values of PS amplitude and of pEPSP slope were measured from baseline input-output curves. To assess LTP for PS and for pEPSP, an index of PS-LTP and an index of pEPSP-LTP were defined from measurements related to the shift of the posttetanization input-output curves, relative to the pretetanization input-output curves. The index of PS-LTP was defined by the percentage change in the stimulus intensity (SI) necessary to produce a PS of one-half maximal amplitude. This normalized index of PSLTP was defined as:
@Iso baseline
-
SIso poststimulus)/SIso baseline
The index for pEPSP-LTP was chosen as the percentage increase in the maximal pEPSP slope 30 minutes following the tetanizing stimulus. This index was defined as follows: (potentiated pEPSP,,,
- baseline pEPSP,,,)/ baseline EPSP,,,
RESULTS Anatomical studies have indicated that associational fibers project longitudinally along the septotemporal axis of the hippocampal formation (Amaral and Witter, 1989). To examine the functional topography of associational inputs onto granule
DENTATE ASSOCIATIONAL PATHWAY PHYSIOLOGY / Bekenstein and Lothman
Tr-P-
W
a
RECORD
A
a 4
"I
401
I
r 1
6
0
b
41)
C" W
n
3a
In
a
wp
1
00
B
10
20
30
STIMULUS INTENSIW (VOLTS)
1I
40
Fig. 2. Input-output curves for (A) PS amplitude vs. stimulus intensity and for (B) pEPSP slope vs. stimulus intensity from the same animal. Maximal values of pEPSP slope and PS amplitude were derived from such curves. Polarity: negativity down.
B Fig. 1 . Examples of paired-pulse responses with associational input to dentate granule cells. (A) Events obtained with recording electrode fixed in the dentate gyrus at AP -3.0 mm, lateral 2.0 mm. Responses were recorded with the stimulating electrode placed in three locations: a, 1.5 mm anterior to the recording electrode, (AP - 1.5 mm, lateral 2.5 mm); b, at the same rostrocaudal position as the recording electrode (AP - 3.0 mm, lateral 2.5 mm); c, I .5 mm posterior to the recording electrode (AP -4.5, lateral 2.5). Responses are of the maximal amplitude achievable at each stimulating electrode position. *Recording electrode position. (B) Pairedpulse facilitation. Stimulation at point c, but lower intensity, produced submaximal response, consisting of only pEPSP. In all instances (a and b), the interpulse interval was 20 ms; calibration pulse (at beginning of each trace) was 5 mV and of 1 ms duration. Polarity: negativity down.
cells, a number of stimulating and recording electrode positions were tested for their ability to generate pEPSP, to produce PS, to demonstrate LTP, and to demonstrate pairedpulse inhibition. Results are summarized in Figure 1A. In the experiment, the recording electrode was placed at a fixed position in the dentate gyrus - 3.0 mm posterior to bregma, and various stimulating sites were tested by placing the stimulating electrode in selected ipsilateral sites. When the stimulating electrode was placed 1.5 mm rostra1 to the recording electrode, a pEPSP was always seen, but no population spike could be elicited. Furthermore, there was an appreciable facilitation of the pEPSP at an interpulse interval of 20 ms (Fig. 1A-A). When the stimulating electrode was placed in the same an-
teroposterior plane as the recording electrode, the pEPSP were larger, and PS were generated readily (Fig. IA-B). With this electrode configuration, paired-pulse inhibition was only infrequently observed. No facilitation of PS was observed. When the stimulating electrode was placed caudal to the recording electrode, even larger pEPSP and PS were produced. However, by lowering the stimulus intensity it was possible to produce pEPSP without PS. With paired stimuli under these conditions, facilitation was seen (Fig. 1B). These observations demonstrate that evoked responses are not antidromic since they appeared without PS. In addition, the electrode configuration placed the stimulation outside of the projection area of the mossy fibers from the granule cells being monitored by the recording electrode (Amaral and Witter, 1989). The facilitation of the potentials provides additional information that they are pEPSP. In addition, the strength of paired-pulse inhibition reached a maximum when the stimulating electrode was positioned I .5-2.0 mm caudal to the recording microelectrode (Fig. IA-C). Thus, all subsequent electrophysiological studies were standardly performed with the stimulating electrode 1.5 mm caudal to the recording electrode. Input-output curves (PS amplitude vs. stimulus intensity; pEPSP slope vs. stimulus intensity) were generated for each animal (Fig. 2). With the standard electrode positions, excitatory responses were consistent. The maximal population spike amplitude was 7.3 2 0.4 mV (mean & SEM, n = 3, and the maximal pEPSP slope was 6.6 2 1.4 mV/ms (n = 5). To confirm the position of the recording electrode within the dentate gyrus, responses were elicited from the ipsilateral
402
HZPPOCAMPUS VOL. I , NO. 4, OCTOBER 1991 PERFORANT PATH TO DENTATE GYRUS GRANULE CELLS n
0-0
FIRSTPS
E W
0-0
SECOND PS
>
,o-0 0 0’
/ 0-0
A
I
HILAR ASSOCIATIONAL PATHWAY
0’ -0’ v)
a 10 - 20
T
A
30
40
50
60
70
80
90
STIMULUS INTENSITY (VOLTS)
B Fig. 3 . Responses in an animal with the recording electrode in the dentate gyrus produced by (A) ipsilateral perforant pathway activation and (B) ipsilateral associational pathway stimulation. Interpulse interval was 20 ms and the calibration pulse was 5 mV, 1 ms. Polarity: negativity down.
perforant path fibers by stimulating the angular bundle (Fig. 3) with a second stimulating electrode. Long-term potentiation did not occur with hilar stimulation. Thirty minutes following the application of a tetanizing stimulus to the hilar region of an intensity that produced a granule cell PS of maximal amplitude, the curves for both pEPSP and PS vs. stimulus intensity were unchanged from baseline. LTP also was not observed with any combination of positions of hilar stimulating and dentate gyms recording electrodes nor with other (ISO- and 200-Hz) tetanizing frequencies (data not shown). However, with electrodes in the standard configuration, even when LTP in the associational pathway was not produced, PS-LTP and pEPSP-LTP of perforant path-granule cell synapses could be demonstrated at the same recording site. There also was n o effect of the production of LTP in the perforant path upon the input-output curves for associational pathway activation of granule cells (data not shown). Paired-pulse responses within the associational system were first examined for dependence upon stimulation intensity (Fig. 4). Paired-pulse responses were measured as the ratios of the population spike amplitude elicited by a single stimulus (PS[T]) following a single stimulus of identical intensity (PS[C]). At a fixed interpulse interval of 20 ms, facilitation was typically observed with hilar stimulus intensities that produced a small conditioning PS (PS[C]). Stronger stimulus intensities produced inhibition (ratios < I .O). The ratios of the test PS to the conditioning PS remained constant after a certain stimulus intensity was reached, indicating that PPI was independent of stimulus intensity under defined conditions. Paired-pulse inhibition was also studied for its dependence upon the interpulse interval (IPI). With the stimulus intensity set so that the ratio of PS amplitudes depended only on the interpulse interval, paired-pulse inhibition was maximal at an IPI of 20 ms (Table 1). Paired-pulse inhibition was evident
A
A\*-A - ~ - A - A - A - A - A I
0
10
B
20
30
40
50
60
70
80
3
STIMULUS INTENSITY (VOLTS)
Fig. 4. Relationship of paired-pulse response to stimulus intensity. (A) Population spike amplitude (ordinate) for stimuli of increasing intensity (abscissa). First (conditioning) population spike (open circles) and second (test) population spike (filled circles) evoked by paired-pulse stimuli are given. (B) Ratios of test population spike to conditioning population spike (ordinate) plotted vs. stimulus intensity. Note pairedpulse inhibition (ratio > 1) for lower intensity and paired-pulse inhibition (ratio < 1 ) for higher intensity.
with interstimulus intervals as long as SO ms. At IPIs of 100 and 200 ms, facilitation of the second PS occurred. A second phase of paired-pulse inhibition occurred at IPIs between 300 and SO0 ms.
DISCUSSION The functional properties of the ipsilateral associational pathway of the hippocampal formation have not been studied to any degree. Steward et al. (1977) reported the ability of this system to demonstrate paired-pulse facilitation of the pEPSP at submaximal stimulus intensities. In that study, the exact position of the associational pathway stimulating electrode was not described, except that it was slightly caudal to the recording electrode. In the same study it was reported that heterosynaptic potentiation between commissural, associational, or entorhinal afferents did not occur. No data
DENTATE ASSOCIATIONAL PATHWAY PHYSIOLOGY / Bekenstein and Lothman
403
Table 1. PS(T)/PS(C)Ratios of Paired-pulse Responses From Associational Fiber Activation of Dentate Granule Cells N
5
20 ms
.21
2
.I1
30 ms
.31
?
.07
50 ms
.81
?
.I0
70 ms
1.11 f .05
were presented for the ability of the associational pathway to produce PS or to undergo LTP, either by exhibiting potentiated PS or potentiated pEPSP slopes. The present study confirms electrophysiologically the topography of the associational pathway as defined anatomically (Blackstad, 1956; Zimmer, 1971; Berger et al., 1980; Amaral and Witter, 1989). The greatest amplitude PS were generated when the stimulating electrode was at least I .5mm caudal to the recording electrode. This electrode configuration also generated the strongest paired-pulse inhibition. Thus, the present electrophysiological data validate the recent conceptualization of the hippocampal formation as a three-dimensional information processor along its longitudinal axis as well as in distinct lamellae (Amaral and Witter, 1989). This concept was advanced on the basis of anatomical work. By consistently placing the electrodes in a particular arrangement as described above, robust, monosynaptic responses that produced PS were elicited. These responses permitted an investigation of PS-LTP and of pEPSP-LTP. Neither form of LTP could be demonstrated in this pathway. The absence of LTP in the associational pathway might have been due to the small number of afferent fibers activated by electrical stimulation (in comparison to perforant path fibers activated by angular bundle stimulation). The associational system also seems to be similar to the commissural pathway with respect to the number of GABAergic projection neurons that arise in the hilus, as determined from anatomical studies (Seress and Ribak, 1983). Those associational, GABAergic projection neurons could serve to directly inhibit dentate granule cells by feedforward inhibition and result in prevention of LTP. This possibility has been suggested for the commissural pathway to the dentate, since a blockade of inhibition allowed the expression of LTP in the dentate commissural system (Steward et al., 1990). Commissural stimulation has also been shown to prevent LTP and the associated increase in glutamate release in the perforant pathgranule cell synapses (Lynch et al., 1989). A number of observations confirmed the separation of the associational pathway from the perforant path fibers. PS-LTP and pEPSP-LTP of the perforant path-granule cell synapses were demonstrated at the same recording site when LTP in the associational pathway was not produced. In addition, the production of LTP in the perforant path had no effect upon the input-output curves for associational pathway activation of granule cells. Other tetanizing parameters might produce LTP within the associational pathway. However, stimulus parameters that routinely produce LTP in the perforant pathgranule cell synapses do not do so in the associational pathway, pointing to a distinction in this afferent system. Paired-pulse inhibition in the associational pathway was robust. While the methods were biased to record feedback inhibition, feedforward activation of many inhibitory neurons in the hilar region could not be excluded (Buzsaki, 1984;
100 ms
1.18
?
.05
200 ms
1.06 ? .I3
300 ms
.85
?
.I4
500 ms
.76
2
.I6
Kapur et al., 1989). Paired-pulse inhibition in the dentate associational pathway is similar in its strength and in duration (IPI up to 70 ms) to paired-pulse inhibition in the CAI region elicited via stimulation of the contralateral CA3 region (Kapur et al., 1989). In contrast, paired-pulse inhibition of dentate gyrus granule cells from perforant path stimulation is of comparable strength but of shorter duration (Stringer and Lothman, 1989). Additionally, hilar associational pathway activation elicited similar paired-pulse responses to those produced in granule cells by perforant path stimulation at other IPI. Thus, the same pattern of a period of inhibition (20-70 ms IPIs), facilitation (100-200 ms IPIs), and inhibition (300-500 ms IPIs) is produced in granule cells by two distinct pathways . The associational pathway as described here will serve as another means for electrophysiological assessment of dentate granule cell function along with the commissural and perforant pathways. The present studies performed in vivo provide a groundwork for study in this pathway under other physiological conditions, such as following deafferentation of the dentate gyrus or during epileptiform activity. The in vivo nature of the present study characterizing of this system should be noted, since the longitudinal excursion of the associational fibers necessarily dictates their disruption in hippocampal slice preparations.
ACKNOWLEDGMENTS The authors thank Rose Powell for assistance in preparing the manuscript. Supported in part by UPS Grants NS21617 and NS25605.
References Aitken, P. J. (1985) Kainic acid and penicillin: Differential effects of excitatory and inhibitory interactions in the CAI region of the hip-
pocampal slice. Brain Res. 325:261-269. Amaral, D. G. (1978) A Golgi study of cell types in the hilar region of the hippocampus in the rat. J. Comp. Neurol. 182:851-914. Amaral, D. G . , and M. P. Witter (1989) The three-dimensional organization of the hippocampal formation: A review of anatomical data. Neuroscience 33:571-591. Bartesaghi, R., T. Gessi, and L. Sperti (1983) Intralamellar transfer of impulses in the hippocampal formation. Exp. Neurol. 82550567. Berger, T. W., S. Semple-Rowland, and J. L. Basset (1980) Hippocampal polymorph neurons are the cells of origin for ipsilateral association and commissural afferents to the dentate gyrus. Brain Res. 215:329-336. Blackstad, T. W . (1956)Commissural connections of the hippocampal region in the rat, with special reference to their mode of termination. J . Comp. Neurol. 105:417-537. Buzsaki, G. (1984) Feed-forward inhibition in the hippocampal formation. Progr. Neurobiol. 22:131-153. Douglass, R. M. (1977) Long-lasting synaptic potentiation in the rat dentate gyrus following brief high frequency stimulation. Brain Res. 126:361-365.
404
HZPPOCAMPUS VOL. 1, NO. 4, OCTOBER 1991
Gottlieb, D. I., and W. M. Cowan (1972) Evidence for a temporal factor in the occupation of available synaptic sites during the development of the dentate gyrus. Brain Res. 41:452-456. Hjorth-Simonsen, A. (1973) Some intrinsic connections of the hippocampus in the rat: An experimental analysis. J. Comp. Neurol. 147:145-162. Kapur, J., J. L. Stringer, and E . W. Lothman (1989) Evidence that repetitive seizures in the hippocampus cause a lasting reduction of GABAergic inhibition. J. Neurophysiol. 61 :417-426. Laatsch, R. H., and W. M. Cowan (1967) Electron microscopic studies of the dentate gyrus of the rat: 11. Degeneration of commissural afferents. J. Comp. Neurol. 130:241-262. Laurberg, S . , and K . E. Sorensen (1981) Associational and commissural collaterals of neurons in the hippocampal formation (hilus fasciae dentatae and subfield CA3). Brain Res. 212:287-300. Lorente de N6, R . (1934) Studies on the structure of the cerebral cortex: 11. Continuation of the study of the ammonic system. J. Psychol. Neurol. 46:113-177. Lynch, G., S. Deadwyler, and C. Cotman (1973) Postlesion axonal growth produces permanent functional connections. Science 180: 1364-1366. Lynch, G . , C. Gall, G. Rose, and C. Cotman (1976) Changes in the distribution of the dentate gyrus associational system following unilateral or bilateral entorhinal lesions in the adult rat. Brain Res. 110:57-71. Lynch, M. A,, M. L . Errington, and T. V. P. Bliss (1989) The increase in [3H]glutamaterelease associated with long-term potentiation in the dentate gyrus is blocked by commissural stimulation. Neurosci. Lett. 103:191-196. McWilliams, R., and G. Lynch (1978) Terminal proliferation and synaptogenesis following partial deafferentation: The reinnervation of the inner molecular layer of the dentate gyms following removal of its commissural afferents. J . Comp. Neurol. 180:581-616. Nunez Filipe, C., J. P. P i n , V . M. Ferrandez de Lima, and F. H. Lopes da Silva (1989) Propagation of afterdischarges along the septo-temporal axis of the rat hippocampus: A quantitative analysis. Electroencephalogr. Clin. Neurophysiol. 73: 172- 178.
Segal, M., and S. Landis (1974) Afferents to the hippocampus of the rat studied with the method of retrograde transport of horseradish peroxidase. Brain Res. 78:l-15. Seress, L., and C. E. Ribak (1983) GABAergic cells in the dentate gyrus appear to he local circuit and projection neurons. Exp. Brain Res. 50:173-182. Steward, O., W. F. White, and C . W. Cotman (1977) Potentiation of the excitatory synaptic action of commissural, associational and entorhinal afferents to dentate granule cells. Brain Res. 134:551560. Steward, O., R. Tomasulo, and W. B. Levy (1990) Blockade of inhibition in a pathway with dual excitatory and inhibitory action unmasks a capability for LTP that is otherwise not expressed. Brain Res. 516:292-300. Stringer, J . L., and E. W. Lothman (1989) Repetitive seizures cause an increase in paired-pulse inhibition in the dentate gyrus. Neurosci. Lett. 105:91-95. Stringer, J. L., J. M. Williamson, and E. W. Lothman (1989) Induction of paroxysmal discharges in the dentate gyrus: Frequency dependence and relationship to afterdischarge production. J . Neurophysiol. 62: 126-135. Swanson, L. W., J. M. Wyss, and W. M. Cowan (1978) An autoradiographic study of the organization of intrahippocampal association pathways in the rat. J . Comp. Neurol. 181:681-716. Swanson, L. W., P. E. Sawchenko, and W. M. Cowan (1981) Evidence for collateral projections by neurons in ammon’s horn, the dentate gyrus, and the subiculum: A multiple retrograde labeling study in the rat. J. Neurosci. 1548-559. West, J . R., S. Deadwyler,C. W. Cotman,andG. Lynch(l975)Timedependent changes in commissural field potentials in the dentate gyrus following lesions of the entorhinal cortex in adult rats. Brain Res. 97:2 15-233. Wong, B. Y . , and D. A. Prince (19881 Hilar neurons: Gatekeepers of the dentate gyrus. SOC.Neurosci. Abstr. 14:246. Zimmer, J . (1971) Ipsilateral afferents to the cornmissural zone of the fascia dentata, demonstrated in decommissurated rats by silver impregnation. J . Comp. Neurol. 142:393-416.
