Impaired spatial learning after saturation of long-term potentiation

Science

Volumn 281, Issue 5385, 1998, Pages 2038-2042

  Moser, Edvard I ^a   Krobert, Kurt A ^a   Moser, May Britt ^a   Morris, Richard G M ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

ANIMAL BEHAVIOR; ARTICLE; ELECTROSTIMULATION; HIPPOCAMPUS POTENTIAL; LEARNING; LONG TERM POTENTIATION; MAZE TEST; NERVE CELL PLASTICITY; NONHUMAN; PRIORITY JOURNAL; RAT; SPATIAL ORIENTATION;

ANIMALS; DENTATE GYRUS; ELECTRIC STIMULATION; ELECTRODES, IMPLANTED; EVOKED POTENTIALS; EXCITATORY POSTSYNAPTIC POTENTIALS; HIPPOCAMPUS; LONG-TERM POTENTIATION; MALE; MAZE LEARNING; PERFORANT PATHWAY; RATS; SYNAPSES; TETANY;

EID: 0032566584     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.281.5385.2038     Document Type: Article

References (41)

1. B. L. McNaughton, in Neurobiology of the Hippocampus, W. Siefert, Ed. (Academic Press, London, 1983), pp. 233-252; G. Lynch and M. Baudry, Science 224, 1057 (1984); T. J. Teyler and P. DiScenna, Annu. Rev. Neurosci. 10, 131 (1987); T. V. P. Bliss and G. L. Collingridge, Nature 361, 31 (1993).
2. B. L. McNaughton, in Neurobiology of the Hippocampus, W. Siefert, Ed. (Academic Press, London, 1983), pp. 233-252; G. Lynch and M. Baudry, Science 224, 1057 (1984); T. J. Teyler and P. DiScenna, Annu. Rev. Neurosci. 10, 131 (1987); T. V. P. Bliss and G. L. Collingridge, Nature 361, 31 (1993).
3. B. L. McNaughton, in Neurobiology of the Hippocampus, W. Siefert, Ed. (Academic Press, London, 1983), pp. 233-252; G. Lynch and M. Baudry, Science 224, 1057 (1984); T. J. Teyler and P. DiScenna, Annu. Rev. Neurosci. 10, 131 (1987); T. V. P. Bliss and G. L. Collingridge, Nature 361, 31 (1993).
4. B. L. McNaughton, in Neurobiology of the Hippocampus, W. Siefert, Ed. (Academic Press, London, 1983), pp. 233-252; G. Lynch and M. Baudry, Science 224, 1057 (1984); T. J. Teyler and P. DiScenna, Annu. Rev. Neurosci. 10, 131 (1987); T. V. P. Bliss and G. L. Collingridge, Nature 361, 31 (1993).
5. B. L. McNaughton and R. G. M. Morris, Trends Neurosci. 10, 408 (1987).
6. We view saturation of an intrinsic pathway as a neural state in which the pathway cannot be further potentiated in the intact and awake animal. It is unlikely that any method of physiological stimulation will potentiate all synapses of the perforant path to their maximum values. Many synapses are thought to be silent [D. Liao, N. A. Hessler, R. Malinow, Nature 375, 400 (1995); J. T. Isaac, R. A. Nicoll, R. C. Malenka, Neuron 15, 427 (1995)]. Their whole-scale potentiation would render a brain region susceptible to hyperexcitability or even seizures, a state that may ordinarily be prevented by intrinsic inhibitory activity or rapid depotentiation.
7. We view saturation of an intrinsic pathway as a neural state in which the pathway cannot be further potentiated in the intact and awake animal. It is unlikely that any method of physiological stimulation will potentiate all synapses of the perforant path to their maximum values. Many synapses are thought to be silent [D. Liao, N. A. Hessler, R. Malinow, Nature 375, 400 (1995); J. T. Isaac, R. A. Nicoll, R. C. Malenka, Neuron 15, 427 (1995)]. Their whole-scale potentiation would render a brain region susceptible to hyperexcitability or even seizures, a state that may ordinarily be prevented by intrinsic inhibitory activity or rapid depotentiation.
8. B. L. McNaughton, C. A. Barnes, G. Rao, J. Baldwin, M. Rasmussen, J. Neurosci. 6, 563 (1986); C. A. Castro, L. H. Silbert, B. L. McNaughton, C. A. Barnes, Nature 342, 545 (1989).
9. B. L. McNaughton, C. A. Barnes, G. Rao, J. Baldwin, M. Rasmussen, J. Neurosci. 6, 563 (1986); C. A. Castro, L. H. Silbert, B. L. McNaughton, C. A. Barnes, Nature 342, 545 (1989).
10. D. L. Korol, T. W. Abel, L. T. Church, C. A. Barnes, B. L. McNaughton, Hippocampus 3, 127 (1993); K. J. Jeffery and R. G. M. Morris, ibid., p. 133; R. J. Sutherland, H. C. Dringenberg, J. M. Hoesing, ibid., p. 141; G. B. Robinson, ibid. 2, 389 (1992).
11. D. L. Korol, T. W. Abel, L. T. Church, C. A. Barnes, B. L. McNaughton, Hippocampus 3, 127 (1993); K. J. Jeffery and R. G. M. Morris, ibid., p. 133; R. J. Sutherland, H. C. Dringenberg, J. M. Hoesing, ibid., p. 141; G. B. Robinson, ibid. 2, 389 (1992).
12. D. L. Korol, T. W. Abel, L. T. Church, C. A. Barnes, B. L. McNaughton, Hippocampus 3, 127 (1993); K. J. Jeffery and R. G. M. Morris, ibid., p. 133; R. J. Sutherland, H. C. Dringenberg, J. M. Hoesing, ibid., p. 141; G. B. Robinson, ibid. 2, 389 (1992).
13. D. L. Korol, T. W. Abel, L. T. Church, C. A. Barnes, B. L. McNaughton, Hippocampus 3, 127 (1993); K. J. Jeffery and R. G. M. Morris, ibid., p. 133; R. J. Sutherland, H. C. Dringenberg, J. M. Hoesing, ibid., p. 141; G. B. Robinson, ibid. 2, 389 (1992).
14. D. P. Cain, E. L. Hargreaves, F. Boon, Z. Dennison, ibid. 3, 153 (1993).
15. C. A. Barnes et al., J. Neurosci. 14, 5793 (1994).
16. T. W. Berger, Science 224, 627 (1984).
17. T. V. P. Bliss and G. Richter-Levin, Hippocampus 3, 123 (1993).
18. Forty-three naïve male Long-Evans rats (300 to 450 g), which were housed in pairs, were on two occasions anesthetized with Equithesin, a mixture containing chloral hydrate and pentobarbital (1.0 ml per 250 g of body weight). During the first surgical session, all rats received complete unilateral hippocampal lesions by the injection of ibotenic acid (Biosearch Technologies, San Rafael, CA) at 14 sites [modified from L. E. Jarrard, J. Neurosci. Methods 29, 251 (1989)]. Ibotenic acid was dissolved in phosphate-buffered saline (pH 7.4) at 10 mg/ml and injected with a 1-μl Hamilton syringe that was mounted to the stereotaxic frame.
19. Two weeks after the induction of the lesions, three bipolar stimulation electrodes (SNEX 100; Rhodes Medical, Woodland Hills, CA) were implanted in the angular bundle of the intact hemisphere 7.0 mm behind and 3.0, 4.0, and 5.0 mm, respectively, lateral to the bregma. A stainless steel recording electrode was placed in the dentate hilus or granule cell layer (3.8 mm behind and 2.4 mm lateral to the bregma). Electrode leads and contacts were encased in dental acrylic, and the animal was allowed 2 weeks for recovery.
20. E. I. Moser, M.-B. Moser, P. Andersen, J. Neurosci. 13, 3916 (1993); M.-B. Moser, E. I. Moser, E. Forrest, P. Andersen, R. G. M. Morris, Proc. Natl. Acad. Sci. U.S.A. 92, 9697 (1995).
21. E. I. Moser, M.-B. Moser, P. Andersen, J. Neurosci. 13, 3916 (1993); M.-B. Moser, E. I. Moser, E. Forrest, P. Andersen, R. G. M. Morris, Proc. Natl. Acad. Sci. U.S.A. 92, 9697 (1995).
22. Two weeks after implantation, evoked waveforms were recorded in the dentate gyrus at 1.5-hour intersession intervals in response to perforant path stimulation. Recording started 5 min after the rat had been placed in a dark, enclosed recording chamber. Waveforms were sampled in the dentate gyrus in response to constant square-wave pulses (100 μs, 0.1 Hz) delivered to the perforant path at three intensities that were adjusted to give population spikes of 0, 1, and 3 mV, respectively (80 to 1000 μA). The slope of the fEPSP was measured as the amplitude difference at two fixed latencies in the middle of the rising phase of the potential. After the third through seventh recording sessions, the rats received either HF stimulation (n = 17), LF stimulation (n = 12), or no stimulation at all (n = 14). Pilot experiments failed to show more saturation (less residual LTP) in rats receiving LTP across several days than in rats receiving a single day of massed stimulation, possibly because of slowly developing homeostatic changes in synaptic weights in populations undergoing substantial up-or down-regulation of synaptic transmission [G. G. Turrigiano, K. R. Leslie, N. S. Desai, L. C. Rutherford, S. B. Nelson, Nature 391, 892 (1998)]. Thus, a massed stimulation protocol was adopted, with the HF-stimulated rats receiving a total of five episodes of tetanic stimulation at 1.5-hour intervals between the four stimulation sites of the cross-bundle stimulation electrodes. This 1.5-hour interval was used to obtain an optimal saturation of LTP induction in the stimulated pathways, taking into account the fact that LTP does not preclude the further induction of a potentiation during a late phase of previously induced LTP [U. Frey, K. Schollmeier, K. G. Reymann, T. Seidenbecher, Neuroscience 67, 799 (1995); U. Frey and R. G. M. Morris, Nature 385, 533 (1997)]. The tetanic current was passed between pairs of the four poles of stimulation sites (a, b, c, and d) of the two cross-bundle stimulation electrodes (Fig. 1A, left). The choice of anode and cathode was systematically altered between tetanization episodes, subject to the constraint that anode and cathode were always on opposite sides of the angular bundle. In each episode, eight pulse trains (each consisting of eight stimuli at 400 Hz) were first passed at 2-s intervals between two of the poles (for example, a and c); 1 min later, a similar train was given at the opposite polarity. Then, after another 1-min interval, the whole sequence was repeated with the two other poles (for example, b and d). The choice of the anode and cathode pairs was as follows: ac and bd (episode 1), ad and be (episode 2), bd and ac (episode 3), and bc and ad (episode 4). The fifth episode was a repetition of the first (ac and bd). Electroencephalogram epochs were recorded at 4-s intervals for 1 min after each tetanization (all combinations of stimulation across the bundle) in a subset of eight tetanized animals, and the samples were screened carefully for afterdischarges. Control rats (also lesioned and implanted) also received eight pulses at 2-s intervals, which were repeated twice within each stimulation episode. In both groups, the intensity was adjusted to evoke fEPSPs at 80 to 90% of the maximum obtained with these electrodes (500 to 2000 μA and 100-μs pulse width). Nonstimulated rats also received a unilateral hippocampal lesion, and 9 rats out of 14 were implanted.
23. Two weeks after implantation, evoked waveforms were recorded in the dentate gyrus at 1.5-hour intersession intervals in response to perforant path stimulation. Recording started 5 min after the rat had been placed in a dark, enclosed recording chamber. Waveforms were sampled in the dentate gyrus in response to constant square-wave pulses (100 μs, 0.1 Hz) delivered to the perforant path at three intensities that were adjusted to give population spikes of 0, 1, and 3 mV, respectively (80 to 1000 μA). The slope of the fEPSP was measured as the amplitude difference at two fixed latencies in the middle of the rising phase of the potential. After the third through seventh recording sessions, the rats received either HF stimulation (n = 17), LF stimulation (n = 12), or no stimulation at all (n = 14). Pilot experiments failed to show more saturation (less residual LTP) in rats receiving LTP across several days than in rats receiving a single day of massed stimulation, possibly because of slowly developing homeostatic changes in synaptic weights in populations undergoing substantial up-or down-regulation of synaptic transmission [G. G. Turrigiano, K. R. Leslie, N. S. Desai, L. C. Rutherford, S. B. Nelson, Nature 391, 892 (1998)]. Thus, a massed stimulation protocol was adopted, with the HF-stimulated rats receiving a total of five episodes of tetanic stimulation at 1.5-hour intervals between the four stimulation sites of the cross-bundle stimulation electrodes. This 1.5-hour interval was used to obtain an optimal saturation of LTP induction in the stimulated pathways, taking into account the fact that LTP does not preclude the further induction of a potentiation during a late phase of previously induced LTP [U. Frey, K. Schollmeier, K. G. Reymann, T. Seidenbecher, Neuroscience 67, 799 (1995); U. Frey and R. G. M. Morris, Nature 385, 533 (1997)]. The tetanic current was passed between pairs of the four poles of stimulation sites (a, b, c, and d) of the two cross-bundle stimulation electrodes (Fig. 1A, left). The choice of anode and cathode was systematically altered between tetanization episodes, subject to the constraint that anode and cathode were always on opposite sides of the angular bundle. In each episode, eight pulse trains (each consisting of eight stimuli at 400 Hz) were first passed at 2-s intervals between two of the poles (for example, a and c); 1 min later, a similar train was given at the opposite polarity. Then, after another 1-min interval, the whole sequence was repeated with the two other poles (for example, b and d). The choice of the anode and cathode pairs was as follows: ac and bd (episode 1), ad and be (episode 2), bd and ac (episode 3), and bc and ad (episode 4). The fifth episode was a repetition of the first (ac and bd). Electroencephalogram epochs were recorded at 4-s intervals for 1 min after each tetanization (all combinations of stimulation across the bundle) in a subset of eight tetanized animals, and the samples were screened carefully for afterdischarges. Control rats (also lesioned and implanted) also received eight pulses at 2-s intervals, which were repeated twice within each stimulation episode. In both groups, the intensity was adjusted to evoke fEPSPs at 80 to 90% of the maximum obtained with these electrodes (500 to 2000 μA and 100-μs pulse width). Nonstimulated rats also received a unilateral hippocampal lesion, and 9 rats out of 14 were implanted.
24. Two weeks after implantation, evoked waveforms were recorded in the dentate gyrus at 1.5-hour intersession intervals in response to perforant path stimulation. Recording started 5 min after the rat had been placed in a dark, enclosed recording chamber. Waveforms were sampled in the dentate gyrus in response to constant square-wave pulses (100 μs, 0.1 Hz) delivered to the perforant path at three intensities that were adjusted to give population spikes of 0, 1, and 3 mV, respectively (80 to 1000 μA). The slope of the fEPSP was measured as the amplitude difference at two fixed latencies in the middle of the rising phase of the potential. After the third through seventh recording sessions, the rats received either HF stimulation (n = 17), LF stimulation (n = 12), or no stimulation at all (n = 14). Pilot experiments failed to show more saturation (less residual LTP) in rats receiving LTP across several days than in rats receiving a single day of massed stimulation, possibly because of slowly developing homeostatic changes in synaptic weights in populations undergoing substantial up-or down-regulation of synaptic transmission [G. G. Turrigiano, K. R. Leslie, N. S. Desai, L. C. Rutherford, S. B. Nelson, Nature 391, 892 (1998)]. Thus, a massed stimulation protocol was adopted, with the HF-stimulated rats receiving a total of five episodes of tetanic stimulation at 1.5-hour intervals between the four stimulation sites of the cross-bundle stimulation electrodes. This 1.5-hour interval was used to obtain an optimal saturation of LTP induction in the stimulated pathways, taking into account the fact that LTP does not preclude the further induction of a potentiation during a late phase of previously induced LTP [U. Frey, K. Schollmeier, K. G. Reymann, T. Seidenbecher, Neuroscience 67, 799 (1995); U. Frey and R. G. M. Morris, Nature 385, 533 (1997)]. The tetanic current was passed between pairs of the four poles of stimulation sites (a, b, c, and d) of the two cross-bundle stimulation electrodes (Fig. 1A, left). The choice of anode and cathode was systematically altered between tetanization episodes, subject to the constraint that anode and cathode were always on opposite sides of the angular bundle. In each episode, eight pulse trains (each consisting of eight stimuli at 400 Hz) were first passed at 2-s intervals between two of the poles (for example, a and c); 1 min later, a similar train was given at the opposite polarity. Then, after another 1-min interval, the whole sequence was repeated with the two other poles (for example, b and d). The choice of the anode and cathode pairs was as follows: ac and bd (episode 1), ad and be (episode 2), bd and ac (episode 3), and bc and ad (episode 4). The fifth episode was a repetition of the first (ac and bd). Electroencephalogram epochs were recorded at 4-s intervals for 1 min after each tetanization (all combinations of stimulation across the bundle) in a subset of eight tetanized animals, and the samples were screened carefully for afterdischarges. Control rats (also lesioned and implanted) also received eight pulses at 2-s intervals, which were repeated twice within each stimulation episode. In both groups, the intensity was adjusted to evoke fEPSPs at 80 to 90% of the maximum obtained with these electrodes (500 to 2000 μA and 100-μs pulse width). Nonstimulated rats also received a unilateral hippocampal lesion, and 9 rats out of 14 were implanted.
25. T. J. Shors, T. B. Seib, S. Levine, R. F. Thompson, Science 244, 224 (1989); L. Xu, R. Anwyl, M. J. Rowan, Nature 387, 497 (1997).
26. T. J. Shors, T. B. Seib, S. Levine, R. F. Thompson, Science 244, 224 (1989); L. Xu, R. Anwyl, M. J. Rowan, Nature 387, 497 (1997).
27. E. Moser, I. Mathiesen, P. Andersen, Science 259, 1324 (1993).
28. Four of these animals showed <10% on the test for residual LTP (18).
29. Behavioral testing was conducted in a water maze, a 198-cm-diameter pool with a featureless white surface, filled to a depth of 40 cm with water at 25° ± 2°C [R. G. M. Morris, J. Neurosci. Methods 11, 47 (1984)]. Latex liquid was added to make the water opaque. A pneumatic escape platform (11 cm in diameter) was located at a fixed position midway between the center and the periphery of the pool. The platform could be moved vertically between an upper available position (1.5 cm below the water surface) and a lower unavailable position (22 cm below the water level) by remote control. Behavioral training started 7 hours after tetanic stimulation was completed. The rats were trained hourly in blocks of two trials, which were separated by 15 s (a total of 10 blocks, corresponding to trials 1 through 20), and were released from one out of eight equally spaced start positions along the perimeter of the pool in a pseudorandom predetermined order. If the rat failed to find the platform within 120 s, the rat was guided onto it. The rat was always left on the platform for 30 s. The position of the black head of the swimming rat was identified and stored at 10 Hz by a video tracking system (VP200, HVS Image, Hampton, UK; Watermaze Software, Edinburgh, UK). Probe tests (with the platform initially unavailable) were conducted on the first trial of blocks 1, 6, and 8 to assess the spatial precision of the search behavior. The platform was kept on the bottom of the pool for the first 40 s and then raised. A final transfer test with the platform submerged for 60 s was conducted at the end of training (called block 11, although consisting of only one trial). On probe trials during training, the latency to cross the platform location was substituted for the actual latency to climb the platform.
30. The extent of saturation at perforant-path/granule-cell synapses was estimated after the completion of the water maze training by tetanizing the fibers activated by the central electrode. The tetanization consisted of two blocks of eight 400-Hz pulses repeated eight times at 2-s intervals and at the same polarity. There was a 1-min interval between the blocks. The tetanization intensity was adjusted to evoke fEPSPs at 80 to 90% of maximum, as above. After the completion of these tests, the rats were killed with an overdose of Equithesin and perfused intracardially with saline and 4% formaldehyde. The brains were stored in formaldehyde for >1 week. Frozen sections were cut coronally (25 μm) and stained with cresyl violet, and the sections were examined for hippocampal and extrahippocampal damage. Sixteen animals (4 HF, 6 LF, and 6 NS) were excluded because of neocortical or thalamic lesions or because of incomplete hippocampal lesions. The exclusion of these animals did not change the pattern of results. Analyses conducted on the entire data set (n = 43) gave group [f(2,40) = 4.3, P = 0.02] and groups × block [F(18,360) = 2.0, P < 0.005] effects on escape latency and gave a groups × quadrants effect on the probe tests [F(6,120) = 3.3, P < 0.005].
31. The population spike increased 0.61 ± 0.30 mV (HF group) and 1.34 ± 0.68 mV (LF group). The increase in the HF group was not related to fEPSP enhancement.
32. This conclusion is corroborated by a pilot experiment suggesting that the disruption of spatial learning after LTP saturation is reversible. Six animals receiving HF stimulation were impaired when tested subsequently in a delayed-matching task in the water maze [R. G. M. Morris, J. J. Hagan, J. N. P. Rawlins, Q. J. Exp. Psychol. 38B, 365 (1986)]. These animals showed no improvement in escape latency from the first to the second trial (trial 2 latencies were, on average, 6.5 s longer; there was an intertrial interval of 2 hours). One month later, when LTP had decayed, the animals showed clear evidence of learning from trial 1 to trial 2 on the same test. Latencies were 20.9 s shorter in trial 2 than in trial 1. In NS control rats, the difference was 34.0 s.
33. M. F. Yeckel and T. W. Berger, J. Neurosci. 18, 438 (1998).
34. M. Nosten-Bertrand et al., Nature 379, 826 (1996).
35. M. L. Errington, T. V. P. Bliss, R. J. Morris, S. Laroche, S. Davis, ibid. 387, 666 (1997).
36. D. Marr, Proc. R. Soc. London Ser. B 262, 23 (1971); A. Treves and E. T. Rolls, Hippocampus 4, 374 (1994).
37. D. Marr, Proc. R. Soc. London Ser. B 262, 23 (1971); A. Treves and E. T. Rolls, Hippocampus 4, 374 (1994).
38. R. G. M. Morris, E. Anderson, G. S. Lynch, M. Baudry, Nature 319, 774 (1986); A. J. Silva, R. Paylor, J. M. Wehner, S. Tonegawa, Science 257, 206 (1992); J. Z. Tsien, P. T. Huerta, S. Tonegawa, Cell 87, 1327 (1996).
39. R. G. M. Morris, E. Anderson, G. S. Lynch, M. Baudry, Nature 319, 774 (1986); A. J. Silva, R. Paylor, J. M. Wehner, S. Tonegawa, Science 257, 206 (1992); J. Z. Tsien, P. T. Huerta, S. Tonegawa, Cell 87, 1327 (1996).
40. R. G. M. Morris, E. Anderson, G. S. Lynch, M. Baudry, Nature 319, 774 (1986); A. J. Silva, R. Paylor, J. M. Wehner, S. Tonegawa, Science 257, 206 (1992); J. Z. Tsien, P. T. Huerta, S. Tonegawa, Cell 87, 1327 (1996).
41. We thank P. Andersen for helpful advice and R. Pedersen, P. Spooner, A.-K. Amundgàrd, K. Haugen, K. Barmen, and G. Dyb for technical assistance. This work was supported by grants from the Human Frontiers Science Panel, the British Medical Research Council, NSF, and the Norwegian Research Council.