Comparative aspects of hippocampal and neocortical long-term potentiation

Journal of Neuroscience Methods, 28 (1989) 101-108

101

Elsevier NSM 00927

Comparative aspects of hippocampal and neocortical long-term potentiation * Timothy J. Teyler Neurobiology Program, Northeastern Ohio Universities College of Medicine, Rootstown, OH 44272 (U.S.A.) (Received 7 November 1988) (Accepted 9 November 1988)

Key words: Long-term potentiation; Hippocampus; Neocortex Long-termpotentiation (LTP) is a candidate for the synaptic alternations underlying memory storage in the mammalian CNS. In this chapter LTP in hippocampus and in visual neocortex are compared. Comparisons of the optimal tetanus parameters revealed that 2-3 trains of high-frequency stimulation (100-400 Hz) delivered within a brief period of time (minutes) results in maximal potentiation in hippocampal synapses. In contrast, the parameters most effective in neocortex were either low-frequency (2 Hz for 60 min) or high-frequency bursts (100 Hz, 100 ms train at 1 / 5 s for 10 min), both of which deliver at least an order of magnitude more afferent activation than that required for hippocampus. Hippocampal population spike potentiation averages 250% and the population excitatory postsynaptic potential (EPSP) potentiation averages 50%. Neocortical LTP also averages about 50%. The expression of LTP requires about 5 rain in CA1 hippocampus, whereas about 30 rain axe required for expression of neocortical potentiation. Both hippocampus and visual neocortex display an enhanced potentiation early in development, with a later stabilization at lower adult levels. Centering at postnatal day 15, hippocampal CA1 displays an LTP magnitude that is over twice that seen at day 60. Neocortical responses display a similar peak at postnatal day 15 and a subsequent adult stab'dization at approximately half of the day 15 maximum. Both tissues first display LTP during the early stages of synapse formation between postnatal days 6-10. The role of the NMDA receptor is implicated in aspects of both hippocampal and neocortical LTP.

Introduction The mechanism(s) by which the mammalian central nervous system registers and stores learned information are currently the subject of considerable attention. Long-term potentiation (LTP) is an enhancement of the response of hippocampal and other mammalian neurons that is widely studied as a mnemonic mechanism because its induction, expression, and duration mimic many properties associated with learning and memory (Swanson et

* First presented at the Second International Conference on Brain and Spinal Cord Slice Preparations, held in Louisville (KY) on 15-17 June 1988. Correspondence: T.J. Teyler, Neurobiology Program, Northeastern Ohio Universities College of Medicine, Rootstown, OH 44272, U.S.A.

al., 1982; Teyler and DiScenna, 1987). Although the hippocampus has been the favored neural structure for studies of LTP since its discovery in the early 1970s, other mammalian CNS structures, including neocortex, display similar synaptic plasticity (Wilson and Racine, 1983; Berry et al., 1988). LTP is defined as a stable, relatively long-lasting increase in the magnitude of a postsynaptic response to a constant afferent volley following brief tetanic stimulation of the same afferents (Fig. 1). LTP can also be observed to occur in conjunction with the learning of a behavioral task in the absence of tetanic stimulation. LTP was initially observed in the intrinsic hippocampal synapses (areas CA1, CA3 and dentate) and more recently has been documented in extrinsic hippocampal synapses, other limbic system synapses,

0165-0270/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)

102

Long-Term Potentiation Potentiated Response

Baseline

Afferent

Tetanus or Learning

-

later-)

Fig. 1. Long-termpotentiation(LTP) is an enduringchange in synapticefficacyseen at synapticjunctions of the mammalianCNS. Shown here as an increase in the size of a hippocampal field potential followingeither electrical tetanization of afferents or a behavioral learningexperience,LTP graduallydecaysback to baseline. medial geniculate nucleus, pyriform cortex, primary visual and somatosensory cortex and cerebellar cortex and deep nuclei (see review by Teyler and DiScenna, 1987). Numerous experiments have shown that LTP also can be produced as a result of a behavioral learning experience. Studies investigating the behavioral induction of LTP have utilized the rabbit N M R preparation, a brightness discrimination task, a tone foot-shock conditioning paradigm, appetitive operant conditioning and an environmental novelty task (Ruth_rick et al., 1982; Berger, 1984; Sharp et al., 1985; Skelton et al., 1987; LoTurco et al., 1988). Treatments which block the induction of LTP by electrical stimulation (administration of the drug APV, see below) also impair the acquisition of a place-learning task shown to be dependent on hippocampus (Morris et al., 1986; Danysz et al., 1988), but do not affect the acquisition of a visual discrimination task which is not dependent upon hippocampal activity (Morris et al., 1986). Results of these experiments, which typically sample the degree of synaptic efficacy throughout the course of behavioral learning, indicate that hippocampal LTP develops in conjunction with the mastery of a behavioral task and can facilitate behavioral learning.

It should be noted that to limit one's consideration only to an understanding of brain mechanisms underlying the encoding of experience in the adult nervous system is to ignore a profoundly important aspect of brain/behavioral experience --namely, the development of perceptual processes that enable immature organisms to form templates with which to understand the meaning of experience. Examples include the well-studied phenomena of critical periods for visual and auditory plasticity (Rakic and Goldman-Rakic, 1982), which are known to be important in sculpting the structure and function of the nervous system with respect to the environment in which the animal is developing. These development plasticities represent an important form of information storage in the nervous system, even though they are not usually considered as learning and memory. The differences lie in the nature of the induction (usually formed during a critical period of development), the enduring nature of the storage (often for a lifetime), and the apparent lack of what is traditionally defined as motivation and reinforcement in their establishment. LTP thus appears to be widespread phenomenon, however, it has not been established that these different examples of enhanced synaptic ac-

103

tivity all represent the same phenomenon. In many experiments the parameters required to initiate LTP (as well as the phenomenon of LTP itself) are different than in the hippocampus. Whereas it is entirely possible that nature would conserve neuronal mechanisms to be employed in plastic modifications of synapses, there is as yet little hard evidence to support such a contention. Thus, it is possible that different forms of plasticity may be represented in different brain regions. This chapter will examine the similarities and differences in two tissues of the rodent brain--area CA1 of the hippocampus and visual neocortex. Such a comparison is useful not only in terms of eventually determining if the underlying mechanism of LTP is common to all plastic synapses, but also in terms of understanding the respective roles of hippocampus and neocortex in memory functions (Teyler and DiScenna, 1986).

NMDA receptors Synaptic transmission is required for LTP induction. This has been demonstrated with the use of neurotransmitter antagonists and manipulation of Ca 2+ concentration. Given the central role of calcium in synaptic processes, several laboratories have studied the role of calcium-calmodulin systems in LTP. Dunwiddie et al. (1978) and Wigstrom et al. (1979) showed that LTP was blocked under low calcium conditions, indicating that calcium plays a critical role in LTP production. The experimental paradigm involves delivering the tetanic stimulus in a low calcium environment (all experiments done in vitro) and testing for LTP later in a normal calcium environment. These observations were extended by the experiments of Baimbridge and Miller (1981), who demonstrated that the uptake and retention of labeled calcium is increased after LTP. Kuhnt et al. (1985) demonstrated a large increase in dendritic calcium deposits with the induction of LTP. Since glutamate is believed to be the neurotransmitter at the principal trisynaptic junctions of the hippocampus, Dunwiddie and colleagues (1978) tetanized perforant path fibers while bathing the dentate gyrus slice in APB (a glutamate

NMDACao.~~~ u

Fig. 2. The NMDA receptor (a subtype of glutamate receptor) has been implicated in the cellular mechanism of LTP. Norreally blocked by physiological concentrations of Mg 2+ (and thus not involved in normal synaptic transmission), the NMDA receptor is activated by depolarization (as induced by tetanic presynaptic stimulation) resulting in the opening of Ca 2+ channels. The increased Ca 2+ conductance is believed to be responsible for the triggering of the enduring aspects of LTP, perhaps acting through Ca 2+ sequestering organelles such as the endoplasmic reticulum. LTP can be induced without tetanic presynaptic stimulation by concurrent postsynaptic depolarization and synaptic activation or glutamate application.

antagonist). No LTP was later observed after washout of the APB, suggesting that glutamate neurotransmission is necessary for LTP. Three subvarieties of the glutamate receptor are known to exist: kainate, quisqualate and N-methyl-Daspartate (NMDA). Recently, there has been considerable interest in the activity of the NMDA receptor (Collingridge & Bliss, 1987). A specific N M D A antagonist, D-2-amino-5-phosphonovalerate (APV) does not affect normal transmission (which appears to be gated through kainate/ quisqualate receptors), but it does block the experimental induction of LTP (CA1, slice preparation; Coliingridge et al., 1983; Harris et al., 1984), as well as behavioral LTP (Morris et al., 1986). During normal activity, NMDA receptors (Fig. 2) appear to be blocked by a voltage dependent, Mg2+-mediated blockade (Nowak et al., 1984). The depolarization produced by the tetanic stimulus alleviates the Mg 2+ block and opens the NMDA receptor channds. The NMDA channel has a high conductance to Ca 2÷, resulting in a

104

a: T E T A N U S - I N D U C E D

POTENTIATION b:

A.

, mv [ , 20 ms

CALCIUM-INDUCED

POTENTIATION

[ 4 mV 20

ms

B.

B.

F. ~ C.

H.

D. E.

J.

F.

K.

Fig. 3. a: an intracellular recording from a CA1 cell showing LTP induced by afferent tetanus. Left (single records): A, baseline; B,

15 s posttetanus; C, 15 rain posttetanus; D, 30 rain posttetanus. Right (averages): E, baseline; F, 15-19 rain posttetanus; G, 27-33 rain posttetanus; H, E-G superimposed, b: an intracellular recording from CA1 cell showing LTP induced by 4 mM Ca 2+ and 6.25 mM K +. Left (single records): A, baseline; B, during high Ca2+/K÷; C, 6.5 rain post; D, 18 min post; E, 28.5 min post; F, A, C and E superimposed. Right (averages): G, baseline; H, 1-10 rain post; I, 11-20 rain post; J, 21-30 min post; K, G, I and J superimposed. (From Grover and Teyler, 1988.) build-up of intracellular Ca 2+ concentration. This Ca 2÷ may then activate: (1) Ca2÷-dependent K ÷ or C1- conductances ( M a c D e r m o t t and Dale, 1987), (2) calpain, a Ca2+-dependent protease which u n m a s k s / a c t i v a t e s a quiescent population of glutamate receptors (Lynch and Baudry, 1984), and (3) Ca2+-activated second messenger systems (Collingridge and Bliss, 1987), or (4) mechanisms ultimately responsible for synaptogenesis (Chang and Greenough, 1984). L y n c h et al. (1983) showed that intracellular injection of the calcium chelator, E G T A , into area CA1 pyramidal cells (hippocampal slice) blocks the appearance of LTP. A n interesting result is that of Turner et al. (1982) and G r o v e r and Teyler (1988), who demonstrated that incubating a hip-

p o c a m p a l slice for 5 - 1 0 min in medium-containing elevated calcium levels (4 m M ) and potassium (6.25 m M ) is sufficient to produce L T P that is identical in m a n y ways to that p r o d u c e d by afferent activation (Fig. 3). Considering the ubiquitous role of calcium in neuronal regulation it is not surprising that manipulations of calcium levels would influence LTP, but these results suggest that there is an additional role of calcium in L T P b e y o n d that of n o r m a l synaptic transmission. Considerably less is k n o w n about the nature of the N M D A r e c e p t o r / c h a n n e l in neocortex. Artola and Singer (1987) demonstrated that visual cortical slices f r o m adult rats displayed L T P through the activation of the N M D A receptor. Interestingly, L T P expression was not observed unless

Fig. 4. Top: the postnatal development of long-term potentiation in hippocampal area CA1. LTP, measured as an enhanced population spike response expressed as a percent of pretetanus baseline (100%), first appears at postnatal day 5 and peaks at day 15. Bottom: the postnatal development of long-term potentiation in cortical area 17. Recordings were made at supragranulax and infragranular areas before and after tetanic stimulation of the subjacent white matter. LTP, measured as an enhanced field potential expressed as a percent of pretetanus baseline (100~), first appears at postnatal day 6 and peaks at days 16-20 in both areas. Calibration: 0.5 mV, 10 ms. (From Harris and Teyler, 1984.)

105 15 days

7 days

60 days

PrEtatilNrtus

Posttatanus

HIPPOCAMPAL LTP PERCENT OF BASELINE 20001 1800 1600 l,lO0 1200 IO00 BOO 600 4O0 200

P'77~ 0

I-2

5-6

3-4

7-8

15

6O-63

AGE (OAYS)

6 days

15 days

60 days

Suw'agranu)ar F~-etatanus

Posttatanus

Infragrsnu)aP Pretetanus

Posttetanus

CORTICAL LTP PERCENT OF BASELINE 22O

[]S°ora,r.ool.r

200 lO0 m 160 14O 120, I00 80

0-5

6-10

IS-15

16-20 AGE (DAYS)

21-25

26-30

60-90

106

GABAergic inhibition was suppressed with bicuculine. That neocortical NMDA receptors are probably involved in other forms of plasticity was demonstrated by Cline et al. (1987), who demonstrated that APV disrupts eye-specific topographic connections in the eye of tadpole, and by Tsumo et al. (1987), who showed that APV blocks visual responses of cortical neurons from cat. APV was much more effective in this effect during the critical period of postnatal development than later in life. LTP, too, is much more evident during the developmental critical period than either before or after as Perkins and Teyler (1988) showed in visual cortical slices taken from rat. Recent studies by Han and Teyler (1988) show that NMDA-receptor induced LTP is limited to layers III-IV and VI in rodent visual cortical slices studied with current source density analysis during the critical period for visual plasticity (postnatal days 15-20). Perkins and Teyler (1988) demonstrated that the LTP recorded at superficial layers of visual cortex was present only during the critical period, whereas the LTP seen at deep layers was also evidenced in adult animals. These results may reflect differential regulation of NMDA-receptors of various cortical laminae, although the work of Han and Teyler (1988) showing the concurrent laminar distribution of NMDA and fl-adrenergic synapses, both capable of evidencing LTP, makes such a conclusion premature. Thus both hippocampus and neocortex possess NMDA-dependent LTP, although neocortex, if not hippocampus as well, possesses other plastic synapses in addition.

LTP magnitude The magnitude of LTP in the two tissues is comparable when differences in cellular density are taken into consideration (Teyler et al., 1988). Hippocampal population spike potentiation averages 250% (baseline = 100) and the population EPSP potentiation averages 50%. Neocortical LTP, measured as the amplitude of the population synaptic response to stimulation of afferents coursing in white matter off-line to the cortical module from which recordings were made, also averages

between 50 and 100% (Berry et al., 1988; Han and Teyler, 1988; Perkins and Teyler, 1988).

Critical period LTP, measured as an enhanced population spike response at 20 min posttetanus, first occurred at postnatal day 5 (P5) and was seen consistently by P7-8 in hippocampal area CA1 (Fig. 4; Harris and Teyler, 1984). The magnitude of LTP at P15 was considerably greater than that observed earlier or at adult ages. LTP in infragranular layers of cortex (layers IV-VI), measured here as an enhanced polysynaptic evoked response, first occurred at P6 and was seen consistently by P l l - 2 0 (Perkins and Teyler, 1988). The magnitude of LTP of the infragranular response was much greater at P16-20 (215% of baseline) than at older ages (130% at P60-90). LTP of the field potential in the supragranular layer (layers I-III) also first occurred at P6 and was seen consistently by Pl1-15. In this layer, the magnitude of LTP was greatest between P16-20 (180%), with a subsequent decline in adults (120%). These studies show that both hippocampal area CA1, and infragranular and supragranular layers of the visual and cortex have developmental periods of peak sensitivity to tetanic stimulation. Both regions show little LTP prior to P5, a peak magnitude around P15, and a subsequent decline to adult levels. The period when maximal hippocampal and visual cortical LTP can be elicited corresponds to a time when many behaviors are first experienced by developing pups. Young rats are first experiencing patterned visual stimuli during this time, which coincides with the maximal period of visual plasticity during development (Teyler et al., 1988). Thus, hippocampal and visual cortical LTP develop at a time when the animal is learning much about its environment. In the infragranular layers of the neocortex, the re-emergence of LTP in the adult animal, following levels of near baseline potentiation between the ages of 21 and 30 days, may reflect the activation of different populations of afferents or targets

107 at these two times. Alternatively, the plastic properties of the synapses may undergo modulation by unknown factors during this period. The identity of the cortical synapses undergoing plastic modification has been addressed by Han and Teyler (1988), using a quantitative current source density (CSD) analysis in conjunction with the application of specific blockers of N M D A receptors (APV) and fl-adrenergic (timolol) receptors. During the critical period for rat visual cortex plasticity studied in vitro, they demonstrated that ,both populations of receptors are represented in supragranular (layers III-IV) and infragranular (layer VI) neocortex. Thus the re-emergence of LTP at deep layers of neocortex in adults may represent either the modulation of the N M D A (or adrenergic) receptor, or the activation of an initially inactive receptor.

Conclusions The comparisons between neocortical and hippocampal L T P are limited by differences in experimental parameters and our incomplete understanding of the involved neocortical circuitry and synaptic elements. However, LTP in both systems share numerous properties, and share systems that appear to be blocked by APV, suggesting that the N M D A receptor-channel complex is involved in both loci. These results suggest that similar phenomena may be operative in hippocampus and visual neocortex, but that the anatomical and physiological complexity of neocortex is mirrored in the complexity of plastic phenomena observed there.

Acknowledgement LTP induction/ expression To examine the induction and expression of LTP in the two areas data were obtained from brain slices of hippocampus and visual neocortex (area Ocl) of rats between birth and 60-90 days of age (Harris and Teyler, 1988; Berry et al., 1988). Comparison of the optimal tetanus parameters revealed that 2-3 trains of high-frequency stimulation (100-400 Hz) delivered within a brief period of time (minutes) results in maximal potentiation in hippocampal synapses. In contrast, the parameters most effective in neocortex were either low-frequency (2 Hz for 60 min) or higher frequency bursts (100 Hz, 100 ms train a t - 1 / 5 s for 10 min), both of which deliver at least an order of magnitude more afferent activation than that required for hippocampus. The expression of LTP requires about 5 rain in CA1 hippocampus, whereas up to 60 min are required for expression of neocortical potentiation. The effectiveness of such different tetanus parameters in neocortex suggests the concurrent activation of different plastic synapses possessing different induction requirements. Based on the hippocampal data, the cortical N M D A receptor system may be most responsive to the higher frequency tetanus, whereas the lower frequency tetanus may be optional for the fl-adrenergic synapses.

Supported by grants from NIH, O N R and EPA.

References Artola, A. and Singer, W. (1987) Long-term potentiation and NMDA receptors in rat visual cortex, Nature (Lond.), 330: 649-652. Bambridge, K.G. and Miller, J.J. (1981) Calcium uptake and retention during long-termpotentiation of neuronal activity in the rat hippocampal slice preparation, Brain Res., 221: 229-305. Berger, T. (1984) Long-term potentiation of hippocampal synaptic transmission affects rate of behavioral learning, Science, 224: 627-630. Chang, F.-L.F. and Greenough, W.T. (1984) Transient and enduring morphologicalcorrelates of synaptic activity and efficacy in the rat hippocampal slice, Brain Res., 309: 35-46. Cline, H.T., Debski, E.A. and Constantine-Paton, M. (1987) N-Methyl-D-aspartate receptor antagonist desegregates eye-specific stripes, Proc. Natl. Acad. Sci. U.S.A., 84: 4342-4345. Collingridge, G. and Bliss, T.V.P. (1987) NMDA receptors-in their role in long-term potentiation, Trends Neurosci., 10: 288-293. Collingridge, G.L., Kehl, S.J. and "McLennan, H. (1983) Excitator amino acids in synaptic transmission in the Schaffercommissural pathway of the rat hippocampus, J. Physiol., 334: 33-46. Danysz, W., Wroblewsld, J.T. and Costa, E. (1988) Learning impairment in rats by N-methyl-D-aspartate receptor antagonists, Neuropharmacology,6: 653-656.

108 Dunwiddie, T., Madison, D. and Lynch, G. (1978) Synaptic transmission is required for initiation of long-term potentiation, Brain Res., 150: 413-417. Grover, L. and Teyler, T.J. (1988) Effect of potassium concentration on calcium induced potentiation in hippocampal CA1 pyramidal cells, Soc. Neurosci. Abstr., A418.9. Han, T. and Teyler, T.J. (1988) Long-term potentiation in the visual cortex of developing rats: in vitro slice preparation, J. Neurosci. Methods, in press. Harris, E.W., Ganong, A.H. and Cotman, C.W. (1984) Longterm potentiation in the hippocampus involves activation of N-methyl-D-aspartate receptors, Brain Res., 323: 132-137. Harris, K.M. and Teyler, T.J. (1984) Developmental onset of long-term potentiation in area CA1 of the rat hippocampus, J. Physiol. (Lond.), 346: 27-48. Kuhnt, U., Mihaly, A. and Joo, F. (1985) Increased binding of calcium in the hippocampus during long-term potentiation, Neurosci. Lett., 53: 149-154. LoTurco, J.M., Coulter, D.A. and Alkon, D.L. (1988) Enhancement of synaptic potentials in rabbit CA1 pyramidal neurons following classical conditioning, Proc. Natl. Acad. Sci. U.S.A., 85: 1672-1676. Lynch, G. and Baudry, M. (1984) The biochemistry of memory: a new and specific hypothesis, Science, 224: 1057-1063. Lynch, G., Larson, J., Kelso, S., Barrionuevo, G. and Schottler, F. (1983) Intracellular injections of EGTA block induction of hippocampal long-term potentiation, Nature (Lond.), 305: 719-721. MacDermott, A.B. and Dale, N. (1987) Receptors, ion channels and synaptic potentials underlying the integrative actions of excitatory amino acids, Trends Neurosci., 10: 280-284. Morris, R.G.M., Anderson, E., Lynch, G. and Baudry, M. (1986) Selective impairment of learning and blockade of long-term potentiation by an N-methyl-D-aspartate receptor antagonist, AP5, Nature (Lond.), 319: 774-776. Nowak, L., Bregestovski, P., Ascher, P., Herbet, A. and Prochiantz, A. (1984) Magnesium gates glutamate-activated channels in mouse central neurones, Nature (Lond.), 307: 462-464. Perkins, T. and Teyler, T.J. (1988) A critical period for long-

term potentiation in the developing rat visual cortex, Brain Res., 439: 222-229. Rakic, P. and Goldman-Rakic, P.S. (1982) Development and modifiability of the cerebral cortex, Neurosci. Res. Progr. Bull., 20: 249-611. Ruthrich, H., Matthies, H. and Ott, T. (1982) Long-term changes in synaptic excitability of hippocampal cell populations as a result of training. In C. Ajmone Marson and H. Matties (Eds.), Neuronal Plasticity and Memory Formation, Raven, New York, pp. 594-598. Sharp 7 P.E., McNaughton, B.L.K. and Barnes, C.A. (1985) Enhancement of hippocampal field potentials in rats exposed to a novel, complex environment, Brain Res., 339: 361-365. Skelton, R.E., Scarth, A.S., Wilkie, D.M., Miller, J.J. and Phillips, A.G. (1987) Long-term increases in dentate granule cell responsivity accompany operant conditioning, J. Neurosci., 7: 3081-3087. Swanson, L.W., Teyler, T.J. and Thompson, R.F. (1982) Mechanisms and functional implications of hippocampal longterm potentiation, Neurosci. Res. Progr. Bull., Vol. 20. Teyler, T.J. and DiScenna, P. (1986) The hippocampal memory indexing theory, Behav. Neurosci., 100: 147-154. Teyler, T.J., Perkins, A.T., IV and Harris, K.H. (1989) The development of long-term potentiation in hippocampus and neocortex, Neuropsychologia, in press. Teyler, T.J. and DiScenna, P. (1987) Long-term potentiation, Annu. Rev. Neurosci., 10: 131-161. Tsumoto, T., Hagihara, K., Sato, H. and Hata, Y. (1987) NMDA receptors in the visual cortex of young kittens are more effective than those of adult cats, Nature (Lond.), 327: 513-514. Turner, R.W., Baimbridge, K.G. and Miller, J.J. (1982) Calcium-induced long-term potentiation in the hippocampus, Neuroscience, 7: 1411-1416. Wigstrom, H., Swann, J.W. and Anderson, P. (1979) Calcium dependency of synaptic long-lasting potentiation in the hippocampal slice, Acta Physiol. Scand., 105: 126-128. Wilson, D.A. and Racine, R. (1983) The postnatal development of post-activation potentiation in the rat neocortex, Dev. Brain Res., 7: 271-276.