Schema-Dependent Gene Activation and Memory Encoding in ...

Schema-Dependent Gene Activation and Memory Encoding in Neocortex Dorothy Tse, et al. Science 333, 891 (2011); DOI: 10.1126/science.1205274

This copy is for your personal, non-commercial use only.

If you wish to distribute this article to others, you can order high-quality copies for your colleagues, clients, or customers by clicking here.

The following resources related to this article are available online at www.sciencemag.org (this infomation is current as of August 11, 2011 ): Updated information and services, including high-resolution figures, can be found in the online version of this article at: http://www.sciencemag.org/content/333/6044/891.full.html Supporting Online Material can be found at: http://www.sciencemag.org/content/suppl/2011/07/06/science.1205274.DC1.html This article cites 39 articles, 18 of which can be accessed free: http://www.sciencemag.org/content/333/6044/891.full.html#ref-list-1 This article appears in the following subject collections: Neuroscience http://www.sciencemag.org/cgi/collection/neuroscience

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright 2011 by the American Association for the Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.

Downloaded from www.sciencemag.org on August 11, 2011

Permission to republish or repurpose articles or portions of articles can be obtained by following the guidelines here.

REPORTS 27. J. L. Muir, T. J. Bussey, B. J. Everitt, T. W. Robbins, Behav. Brain Res. 82, 31 (1996). 28. V. Parikh, J. Ji, M. W. Decker, M. Sarter, J. Neurosci. 30, 3518 (2010). Acknowledgments: We thank J. Staal and H. Lodder. U.M. and A.B.S. received funding from the EU Seventh Framework Programme “Neurocypres” (HEALTH-F2-2007-202088); H.D.M., from the Netherlands Organization for Scientific Research (917.76.360 and 912.06.148), Neurobasic PharmaPhenomics, and the VU University board.

Schema-Dependent Gene Activation and Memory Encoding in Neocortex Dorothy Tse,1* Tomonori Takeuchi,1* Masaki Kakeyama,2 Yasushi Kajii,3 Hiroyuki Okuno,4 Chiharu Tohyama,2 Haruhiko Bito,4 Richard G. M. Morris1† When new learning occurs against the background of established prior knowledge, relevant new information can be assimilated into a schema and thereby expand the knowledge base. An animal model of this important component of memory consolidation reveals that systems memory consolidation can be very fast. In experiments with rats, we found that the hippocampal-dependent learning of new paired associates is associated with a striking up-regulation of immediate early genes in the prelimbic region of the medial prefrontal cortex, and that pharmacological interventions targeted at that area can prevent both new learning and the recall of remotely and even recently consolidated information. These findings challenge the concept of distinct fast (hippocampal) and slow (cortical) learning systems, and shed new light on the neural mechanisms of memory assimilation into schemas. emory consolidation consists of two processes. Cellular consolidation is mediated by synaptic and signal transduction mechanisms that store newly encoded memory traces on-line (1, 2). Systems consolidation involves a time-limited interaction between the medial temporal lobe and the neocortical areas that eventually store long-term memory traces (3–5). Studies monitoring cerebral glucose use, immediate early gene (IEG) activation, and dendritic spine formation (6–9) indicate that rapid on-line encoding of episodic-like memory in the hippocampus can be followed by temporally graded neural changes in the medial prefrontal (mPFC), orbitofrontal (Orb), and retrosplenial (RSC) cortices. This apparent sequence of events does not preclude the possibility of simultaneous encoding or “tagging” in the hippocampus and cortex (9, 10). Indeed, when systems consolidation occurs in the presence of relevant prior knowledge (11, 12), the “assimilation” of new paired-associate (PA) memories into existing activated cortical

M

1 Centre for Cognitive and Neural Systems, University of Edinburgh, Edinburgh EH8 9JZ, UK. 2Laboratory of Environmental Health Sciences, Center for Disease Biology and Integrative Medicine, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. 3Pharmacology Research Laboratories I, Mitsubishi Tanabe Pharma Corporation, 1000 Kamoshida-cho, Aoba-ku, Yokohama 227-0033, Japan. 4Department of Neurochemistry, University of Tokyo Graduate School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.

*These authors contributed equally to this work. †To whom correspondence should be addressed. E-mail: r.g.m. [email protected]

schemas proceeds very rapidly (13), reflecting an influence of prior knowledge on the rate of consolidation (14). The associative encoding of such PAs requires the hippocampus (13, 15, 16), accompanied by novelty-triggered cellular consolidation (17), but may also involve simultaneous cortical encoding. However, if parallel cortical encoding into a schema occurs, it may be driven solely in a bottom-up manner by the hippocampus or may also reflect the influence of activated prior knowledge already stored in cortex. Study 1 mapped IEG activation in numerous brain areas of rats during both the retrieval of original PAs and the learning of new PAs after extensive prior training of a schema over many weeks (fig. S1). Guided by the retrieval cue of different flavors of food given in the start box of an event arena on each of six daily training trials, the animals learned to recall the location of the appropriate sand well, where they were rewarded by retrieving more of that same flavored food. Once performance reached asymptote over 6 weeks (fig. S2), a critical session of retrieval and new learning was conducted. The 21 trained animals were then divided into three groups (Fig. 1A), to which a group of seven caged control animals (group CC) was added. One group had six trials with the original set of PAs and thus had only to retrieve (group OPA, i.e., original paired-associates). Another group had four successive trials with the original PAs and was then exposed to two new PAs that we had shown (13) could be encoded and successfully assimilated into the existing cortical

www.sciencemag.org

SCIENCE

VOL 333

Supporting Online Material www.sciencemag.org/cgi/content/full/333/6044/888/DC1 Materials and Methods SOM Text Figs. S1 to S11 Tables S1 to S5 Reference (29) 15 April 2011; accepted 30 June 2011 10.1126/science.1207079

schema (group NPA, i.e., new paired-associates). The third group was exposed to six new combinations of flavor and location that constituted a set of six new PAs (group NM, i.e., new map). Although this group was subjected to much greater “novelty,” it was in a manner that should not allow successful cortical assimilation (timeline in Fig. 1A). The performance during that single session reflected these different conditions (Fig. 1B; latency data in fig. S3). After a further interval of 80 min (optimized for IEG signal detection of the neural correlates of the events of trials 5 and 6), the animals were first given a cued-recall test. This showed effective memory for the new PAs in the NPA group but no learning by the NM group (Fig. 1C). Brain sections were then prepared for histochemical analysis of two plasticity-associated IEGs—Zif268 (Egr1) and Arc (activity-regulated cytoskeletonassociated protein) (18, 19). Quantitative blind analysis of entire brain regions revealed a striking learning-associated increase in IEG expression in the prelimbic region (PrL) of the mPFC that was nonmonotonic with respect to the extent of learning-associated novelty (Fig. 1, D and E). IEG expression was highest in the NPA rats for whom activated prior knowledge was relevant to new PA information. Detailed analysis revealed three broad patterns of IEG activation (fig. S4 and tables S1 and S2). First, a group of cortical regions [PrL, anterior cingulate (ACC), and RSC] showed the same nonmonotonic pattern of higher Zif 268 and Arc expression in group NPA as in groups OPA and NM, despite the latter group being exposed to greater novelty (Figs. 1E and 2A); analyses of variance (ANOVAs) based on average values from all three regions showed a significant inverted U-shaped effect (Fig. 2A). Non-mnemonic aspects such as motivation were excluded as contributing factors by analysis of latency rather than choice (see fig. S3). Second, and in contrast, area CA1 of the hippocampus (Fig. 2B) showed a large increase in Arc expression in groups NPA and NM, with a monotonic trend favoring the highest expression in group NM; Zif268 levels were unchanged (see also fig. S7). Third, certain cortical regions showed little absolute change in IEG expression across the trained groups or relative to group CC [including the primary somatosensory “barrel” cortex (Ssp); Fig. 2C]. Barrel cortex was therefore chosen as the control region for study 2. In study 2, we sought to determine whether the significant increase in both Zif268 and Arc

12 AUGUST 2011

Downloaded from www.sciencemag.org on August 11, 2011

21. S. Granon, P. Faure, J. P. Changeux, Proc. Natl. Acad. Sci. U.S.A. 100, 9596 (2003). 22. K. Maruki, Y. Izaki, T. Akema, M. Nomura, Neurosci. Lett. 351, 95 (2003). 23. J. W. Dalley, R. N. Cardinal, T. W. Robbins, Neurosci. Biobehav. Rev. 28, 771 (2004). 24. U. Maskos et al., Nature 436, 103 (2005). 25. M. E. Avale et al., FASEB J. 25, 2145 (2011). 26. D. S. McGehee, L. W. Role, Annu. Rev. Physiol. 57, 521 (1995).

891

REPORTS pothesized that interrupting excitatory neurotransmission in the PrL region displaying the largest IEG elevation in cortex in association with learning would disrupt the retrieval of original PAs. Indeed, inactivation of PrL by 6-cyano-7nitroquinoxaline-2,3-dione (CNQX) prevented successful retrieval, whereas control infusions of CNQX into barrel cortex had no effect (Fig. 3A). Second, we asked whether the same would be seen for new PAs that had been encoded only 24 hours earlier. If schema-dependent encoding occurs in a network involving PrL, and if consolidation occurs rapidly against the background of an existing schema, then disruption of fast transmission might impair memory—even though a hippocampal index trace (20) might still remain. CNQX inactivation of PrL blocked cued

recall of such newly stored PAs (Fig. 3B). Third, in common with previous studies of olfactory, spatial, and PA memory showing that NMDA receptor–dependent activity is not required for the retrieval of memory traces in the hippocampus (15, 21), microinfusion of D(–)-2-amino-5phosphonovaleric acid (D-AP5) into the PrL also had no effect on cued recall of both original and newly stored PAs (Fig. 3, C and D). Despite the hippocampal-dependent nature of new PA encoding, we observed that blockade of either AMPA receptor–mediated fast transmission or NMDA receptor–dependent mechanisms in the PrL at the time of learning impaired consolidation (Fig. 4, A and B). The functional inactivation caused by CNQX would, as shown electrophysiologically (15), have lasted ~1 to 2

Downloaded from www.sciencemag.org on August 11, 2011

in PrL reflects parallel encoding of task-related memory traces in cortex. Bilateral drug infusion cannulae were stereotaxically implanted into PrL and, as a within-subjects cortical control, into barrel cortex. New animals were trained on the initial schema over 6 weeks (PAs 1 to 6; see figs. S9 and S10) and then subjected to an extensive series of within-subjects interventions investigating the impact of blocking a-amino3-hydroxy-5-methyl-isoxazole-4-propionic acid (AMPA) or N-methyl-D-aspartate (NMDA) receptors during both retrieval (Fig. 3) and new learning (see Fig. 4 and table S3 for design). Performance on PAs 1 to 6 was stable throughout 6 months of training and testing (fig. S10), and we assumed that the original schema was fully consolidated in cortex. We first hy-

Fig. 1. Immediate early gene (IEG) activation in cortex during pairedassociate (PA) learning. (A) Design for the critical session shows behavioral procedures across groups over 360 min (white, original PAs; pink, new PAs; blue, trials for which Zif268/Arc protein was measured). Immediately after testing, rats were transcardially perfused and their brains taken for IEG analysis. T1 to T6, trials 1 to 6; recall, cued-recall test. (B) Choice performance differed across groups and trials (Group × Trial interaction, F = 4.52, df = 2/18, P < 0.05); for trials 1 to 4, it was above chance at ~70% for groups OPA and NPA (t test, Ps < 0.01 relative to chance) but at chance for group NM [not significant (ns)]. Performance was high for trials 5 and 6 in group

892

12 AUGUST 2011

VOL 333

OPA (P < 0.01) but at chance for the two new PAs of groups NPA and NM (ns). (C) The cued-recall test showed that the new PAs of group NPA had been learned effectively in one trial (P < 0.01 versus chance). In contrast, group NM failed to learn (ns). (D) IEG expression in PrL (layers I to VI) for both Zif268 and Arc. Scale bars, 100 mm. (E) The nonmonotonic increase of IEG expression in PrL revealed a significantly different pattern of expression across groups (Zif268: F = 6.24, df = 3/20, P < 0.01; Arc: F = 14.13, df = 3/24, P < 0.001), with group NPA differing from group CC (*) and the other trained groups (†) [Ryan-Einot-Gabriel-Welsch Range (REGWR) test]. *,†P < 0.05; **P < 0.01; ***,†††P < 0.001. Data are means T SEM. SCIENCE

www.sciencemag.org

REPORTS mals on a separate runway “gap-crossing” task that requires whisker sensation (22, 23). We established that a larger infusion of CNQX into barrel cortex (2.0 ml) is required to severely disrupt this control task, but this still did not affect new PA learning (fig. S12). Thus, the regional dissociation in cortex is meaningful. These findings indicate that the assimilation of rapidly acquired new PA information into existing cortically based mental schemas is associated with cortical encoding of information during hippocampal-dependent learning, and that this simultaneous encoding is essential for long-term memory. First, when animals learned

two new PAs (group NPA), there was an immediate up-regulation of two IEGs (Zif 268 and Arc) in connected regions of the neocortex previously implicated in remote memory consolidation—PrL, ACC, and RSC (5). We postulate that this likely reflects a top-down influence on IEG activation (14) because the lower elevation of IEG expression in cortex of group NM, which was no greater than that associated with memory retrieval (group OPA), was commensurate with that group being unable to incorporate new PAs into a nonexistent cortical schema. In contrast, the relatively higher Arc expression of group NM in hippocampus

Downloaded from www.sciencemag.org on August 11, 2011

hours, barely extending into the systems consolidation time domain of hours to days. It is therefore most logical to suppose that it interrupted a necessary parallel encoding of traces within a cortical network involving PrL. CNQX infusion into barrel cortex had no effect (Fig. 4A). We were concerned about the extent of a brain region functionally affected by the volumes of drug we used (0.5 ml bilaterally). This was chosen, on the basis of pilot data, to affect a high proportion of the volume of PrL, but it may have been too little to affect much of the barrel cortex that served as our control. As a “positive” control, we therefore trained our ani-

Fig. 2. IEG activation in cortex and hippocampus. (A) We found a similar pattern across a group of connected structures (PrL, ACC, and RSC) that, considered together, showed an overall inverted U-shaped function across groups for both Zif268 (F = 6.16, df = 3/20, P < 0.01) and Arc (F = 13.95, df = 3/24, P < 0.001). The individual region analyses were also significant (Zif268: PrL data in Fig. 1E; ACC, F = 3.83, P < 0.05; RSC, F = 4.44, P < 0.05; Arc: ACC, F = 14.00, P < 0.001; RSC, F = 7.70, P < 0.01). In each of these regions considered in isolation, group NPA showed significantly higher IEG expression than group CC and, particularly for Arc, higher levels than groups OPA and NM (REGWR tests, P values displayed). (B) In the hippocampus, the trend across groups with respect to area CA1 was, in contrast, a monotonic increase as a www.sciencemag.org

SCIENCE

function of novelty (i.e., highest in group NM). For Arc, this was highly significant (F = 19.51, df = 3/24, P < 0.001); no significant group differences were seen for Zif268 (F < 1, ns). Across the PrL, ACC, and RSC considered together, the nonmonotonic pattern differed from the monotonic pattern shown in CA1 (Region × Group interaction for Arc: F = 20.58, df = 3/24, P < 0.001). (C) In barrel cortex (Ssp), no significant differences were seen across groups in Zif268 (F < 1, ns), but the data for Arc showed a paradoxical groups difference (F = 22.34, df = 3/24, P < 0.001), with groups OPA and NM being above groups NPA and CC (Ps < 0.001). The numbers below each picture represent the distance (in mm) from bregma. *P < 0.05 versus group CC; †P < 0.05 versus trained group; **,††P < 0.01; ***,†††P < 0.001. Data are means T SEM. VOL 333

12 AUGUST 2011

893

Fig. 3. Memory retrieval with and without cortical interference. (A) Retrieval of original PAs during bilateral CNQX-induced inactivation of the PrL region of mPFC and barrel cortex. Control infusions of NaCl. Inactivation of PrL selectively impaired retrieval of original PAs (Dig location × Drug × Brain Region interaction: F = 11.42, df = 1/10, P < 0.01, Greenhouse-Geisser). Performance was at chance for CNQX into PrL, and above chance for saline into PrL (P < 0.001) and CNQX/saline into barrel cortex (Ps < 0.01). (B) A similar pattern was apparent for retrieval of new PAs (F = 5.33, df = 1.22/10.97, P < 0.05). (C and D) Infusion of D-AP5 had no impact on retrieval of original PAs (C) or new PAs (D), with the ANOVAs showing no significant interactions between D-AP5 and saline conditions in either case. PT, probe test. **P < 0.01, ***P < 0.001 versus chance. Data are means T SEM.

reflects greater novelty and new map learning (24, 25). The lack of correlation between spatial learning and Zif268 RNA expression has been observed previously (18). Second, temporary interruption of AMPA and NMDA receptors in PrL during new PA learning resulted in a failure of memory tested 24 hours later. This is consistent with the idea that Arc activation is necessary for memory consolidation (26), but we now argue that hippocampal and cortical gene activation events are equally required for schema assimilation. Third, the cued recall of original and new PA information required AMPA, but not NMDA, receptor transmission in PrL at the time of memory trace reactivation. These findings complement similar data from human brainimaging studies concerning interactions among the hippocampus, mPFC, and other cortex regions (27–29). However, they differ in detail from recent observations on “cortical tagging” (9). That study showed an increase of c-Fos expression in Orb 90 min after the encoding of a socially transmitted food preference. Local tetrodotoxin-induced interference of Orb impaired recall 30 days after training but had no effect either 1 or 7 days after training. Thus, a slower temporal gradient of consolidation was seen in these experimentally naïve animals. One theory of systems consolidation contrasts hippocampal fast learning with a slow-learning neocortical system. These differential learning rates were shown in computational modeling to help prevent catastrophic interference of new learning by old information (30). This view does not predict IEG activation in cortex at the time of behavioral learning; it is more in keeping with the delayed up-regulation seen at remote time points in previous studies (5). Nor does it predict that AMPA receptor– and NMDA receptor– dependent encoding in PrL would be required at the time of behavioral learning for memory 24 hours later, or that AMPA receptor activity in PrL would be required for retrieval within

Fig. 4. Memory encoding of new PAs with or without glutamate receptor blockade. (A) Inactivation of PrL with CNQX at encoding impaired memory retrieval 24 hours later; no effect of infusions into barrel cortex was observed [Dig location × Drug × Brain region interaction: F = 13.16, df = 1.13/9.03, P < 0.01, GreenhouseGeisser; t tests showed above-chance digging in the probe test 24 hours after encoding for the control conditions (Ps < 0.01) but not for CNQX into PrL (P > 0.05)]. (B) The same bilateral infusions of D-AP5 that had no impact on retrieval of new PAs (Fig. 3D) did impair acquisition when given at the time of memory encoding (Dig location × Drug interaction: F = 9.86, df = 1.25/9.98, P < 0.01; t tests showed a similar pattern for CNQX, Ps < 0.001 and ns, respectively). PT, probe test. **P < 0.01, ***P < 0.001 versus chance. Data are means T SEM.

894

12 AUGUST 2011

VOL 333

SCIENCE

www.sciencemag.org

Downloaded from www.sciencemag.org on August 11, 2011

REPORTS

REPORTS tween these connected cortical structures would disrupt memory retrieval, as the hippocampal index trace would, on its own, be insufficient for effective retrieval. Connections from the hippocampus to mPFC display long-term potentiation (37), and mPFC interacts with the hippocampus in the acquisition of object-place associations (10, 38). Coherence in the theta-frequency band between mPFC and the hippocampus is observed during workingmemory tasks (39, 40), which our PA task also entails as the animals move about the arena to find locations in space that have been recalled by the flavor cue. Intrinsic dynamical oscillations may also be important for integration of cortical circuit performance (41), for episodic-like memory (42), and during the learning of schemas or their later reactivation (43, 44). Thus, the opportunity to learn about the neural basis of cortical schemas of knowledge is opening up and, as it does so, the use of experienced animals possessing activated cortical networks of prior knowledge points to new ways of thinking about systems consolidation and reconsolidation. References and Notes 1. Y. Dudai, R. G. M. Morris, in Brain, Perception and Memory: Advances in Cognitive Sciences, J. J. Bolhuis, Ed. (Oxford Univ. Press, Oxford, 2000), pp. 149–162. 2. J. D. Sweatt, Mechanisms of Memory (Academic Press/Elsevier, San Diego, CA, 2003). 3. L. R. Squire, Memory and Brain (Oxford Univ. Press, New York, 1987). 4. P. Alvarez, L. R. Squire, Proc. Natl. Acad. Sci. U.S.A. 91, 7041 (1994). 5. P. W. Frankland, B. Bontempi, Nat. Rev. Neurosci. 6, 119 (2005). 6. B. Bontempi, C. Laurent-Demir, C. Destrade, R. Jaffard, Nature 400, 671 (1999). 7. T. Maviel, T. P. Durkin, F. Menzaghi, B. Bontempi, Science 305, 96 (2004). 8. L. Restivo, G. Vetere, B. Bontempi, M. Ammassari-Teule, J. Neurosci. 29, 8206 (2009). 9. E. Lesburguères et al., Science 331, 924 (2011). 10. G. R. I. Barker, E. C. Warburton, J. Neurosci. 28, 2837 (2008). 11. F. C. Bartlett, Remembering (Cambridge Univ. Press, Cambridge, 1932). 12. J. D. Bransford, Human Cognition: Learning, Understanding and Remembering (Wadsworth, Belmont, CA, 1979). 13. D. Tse et al., Science 316, 76 (2007). 14. T. Osada, Y. Adachi, H. M. Kimura, Y. Miyashita, Philos. Trans. R. Soc. London Ser. B 363, 2187 (2008). 15. M. Day, R. Langston, R. G. M. Morris, Nature 424, 205 (2003). 16. I. Bethus, D. Tse, R. G. M. Morris, J. Neurosci. 30, 1610 (2010). 17. J. E. Lisman, A. A. Grace, Neuron 46, 703 (2005). 18. J. F. Guzowski, B. Setlow, E. K. Wagner, J. L. McGaugh, J. Neurosci. 21, 5089 (2001).

www.sciencemag.org

SCIENCE

VOL 333

19. M. W. Jones et al., Nat. Neurosci. 4, 289 (2001). 20. T. J. Teyler, P. DiScenna, Behav. Neurosci. 100, 147 (1986). 21. U. Staubli, O. Thibault, M. DiLorenzo, G. Lynch, Behav. Neurosci. 103, 54 (1989). 22. K. A. Hutson, R. B. Masterton, J. Neurophysiol. 56, 1196 (1986). 23. J. A. Harris, R. S. Petersen, M. E. Diamond, Proc. Natl. Acad. Sci. U.S.A. 96, 7587 (1999). 24. A. Vazdarjanova, B. L. McNaughton, C. A. Barnes, P. F. Worley, J. F. Guzowski, J. Neurosci. 22, 10067 (2002). 25. T. Miyashita, S. Kubik, N. Haghighi, O. Steward, J. F. Guzowski, J. Neurosci. 29, 898 (2009). 26. J. F. Guzowski et al., J. Neurosci. 20, 3993 (2000). 27. M. T. R. van Kesteren, G. Fernández, D. G. Norris, E. J. Hermans, Proc. Natl. Acad. Sci. U.S.A. 107, 7550 (2010). 28. M. T. R. van Kesteren, M. Rijpkema, D. J. Ruiter, G. Fernández, J. Neurosci. 30, 15888 (2010). 29. B. A. Kuhl, A. T. Shah, S. DuBrow, A. D. Wagner, Nat. Neurosci. 13, 501 (2010). 30. J. L. McClelland, B. L. McNaughton, R. C. O’Reilly, Psychol. Rev. 102, 419 (1995). 31. L. Nadel, M. Moscovitch, Curr. Opin. Neurobiol. 7, 217 (1997). 32. G. Winocur, M. Moscovitch, S. Fogel, R. S. Rosenbaum, M. Sekeres, Nat. Neurosci. 8, 273 (2005). 33. D. Griffiths, A. Dickinson, N. Clayton, Trends Cogn. Sci. 3, 74 (1999). 34. R. G. M. Morris, Eur. J. Neurosci. 23, 2829 (2006). 35. Y. Dudai, Annu. Rev. Psychol. 55, 51 (2004). 36. R. G. M. Morris et al., Neuron 50, 479 (2006). 37. S. Laroche, S. Davis, T. M. Jay, Hippocampus 10, 438 (2000). 38. R. W. Komorowski, J. R. Manns, H. Eichenbaum, J. Neurosci. 29, 9918 (2009). 39. M. W. Jones, M. A. Wilson, PLoS Biol. 3, e402 (2005). 40. K. Benchenane et al., Neuron 66, 921 (2010). 41. V. S. Sohal, F. Zhang, O. Yizhar, K. Deisseroth, Nature 459, 698 (2009). 42. M. E. Hasselmo, L. M. Giocomo, M. P. Brandon, M. Yoshida, Behav. Brain Res. 215, 261 (2010). 43. A. Sirota et al., Neuron 60, 683 (2008). 44. A. B. L. Tort, R. W. Komorowski, J. R. Manns, N. J. Kopell, H. Eichenbaum, Proc. Natl. Acad. Sci. U.S.A. 106, 20942 (2009). Acknowledgments: Supported by grants from the Mitsubishi Tanabe Pharmaceutical Company, the UK Medical Research Council (R.G.M.M.), the Uehara and Astellas Foundations (T.T.), SRPBS-Ministry of Education, Culture, Sports, Science and Technology (MEXT) (C.T.), CRESTJapan Science and Technology Agency (H.O., H.B.), and grants-in-aid from MEXT (M.K., H.O., H.B.) and Ministry of Health, Labour and Welfare of Japan (M.K., H.B.). We thank B. Bontempi, S. Alaux-Cantin, M. Ginger, G. Poirier, P. Spooner, and J. Tulloch for technical advice, and S.-H. Wang for scientific discussion.

Downloaded from www.sciencemag.org on August 11, 2011

24 hours of hippocampal-dependent learning. It also predicts that the rapid rate of consolidation that we observed might have caused at least a mild disruption of memory retrieval for original PAs, but original PAs continued to be recalled well. A second idea, multiple trace theory (31), supposes that multiple traces of single-trial episodic events are encoded in both cortex and hippocampus. Our observation of parallel encoding of new PAs in a cortical network involving PrL, ACC, and RSC fits with this idea; however, multiple trace theory predicts that cortical traces may be sufficient for later retrieval but should not be necessary, given that hippocampal traces would be available. Multiple trace theory might accommodate assimilation into a schema as a form of “semanticization”; in that case, assimilation would exclusively involve cortical learning (32). However, this account would have to accommodate our previous observations from lesion and pharmacological studies (13, 16) that the hippocampus remains essential for future learning of new PAs. Once a schema is acquired, assimilation of new information in our protocol may indeed represent an instance of hippocampaldependent “semantic” rather than “episodic-like” learning within a single trial; our cued-recall test does not involve a temporal component (33). Both the standard hypothesis and our schema hypothesis of systems consolidation (3, 4, 34) require simultaneous cortical encoding or “cortical tagging” (9). Models of reconsolidation involving memory updating are also relevant (29, 35, 36). Disconnected items of detailed information can be encoded in parallel in the neocortex at the same time as an index trace of paired association is encoded in the hippocampus (20), but with immediate IEG activation in cortex regulated in part by the relevance of the new information being processed in the hippocampus to an existing cortical schema. According to this version of parallel encoding, systems consolidation may be partly regulated in a top-down manner (14, 29), enabling relevant new PAs to be assimilated into a schema. Recent data on effective connectivity between cortical regions in humans are in keeping with this view (28). The necessity for NMDA receptor–dependent mechanisms in specific areas of cortex (e.g., PrL) is also consistent with neural plasticity in cortex being vital at the time of learning. Our hypothesis also correctly predicts that after rapid consolidation, blockade of neural activity be-

Supporting Online Material www.sciencemag.org/cgi/content/full/science.1205274/DC1 Materials and Methods Figs. S1 to S12 Tables S1 to S3 References 7 March 2011; accepted 21 June 2011 Published online 7 July 2011; 10.1126/science.1205274

12 AUGUST 2011

895

www.sciencemag.org/cgi/content/full/science.1205274/DC1

Supporting Online Material for Schema-Dependent Gene Activation and Memory Encoding in Neocortex Dorothy Tse, Tomonori Takeuchi, Masaki Kakeyama, Yasushi Kajii, Hiroyuki Okuno, Chiharu Tohyama, Haruhiko Bito, Richard G. M. Morris*

*To whom correspondence should be addressed. E-mail: [email protected]

Published 7 July 2011 on Science Express DOI: 10.1126/science.1205274

This PDF file includes: Materials and Methods Figs. S1 to S12 Tables S1 to S3 References

1.

General Materials and Methods

1.1 Subjects and housing The subjects were adult male Lister-hooded rats, aged 8 to 10 weeks at the start of experimentation and weighing 230 to 250 g. They were housed in groups of 3 to 4 rats per cage (Study 1) or, because of indwelling drug-infusion cannulae, in single cages (Study 2). They had free access to water at all times and were maintained at 90% of their free-feeding weight during experiments. Experiments were conducted on a 12 hr (on)/12 hr (off) light cycle, with training during the light phase (7 am-7 pm). A total of 39 rats were used (Study 1, n = 28; Study 2, n = 11). All procedures were compliant with the UK Animals (Scientific Procedures) Act 1986 and with the European Communities Council Directive of 24 November 1986 (86/609/EEC) legislation governing the maintenance of laboratory animals and their use in scientific experiments.

1.2 Apparatus An ‘event arena’ (Fig. S1A) is a square shaped open field ‘maze’ in which ‘events’ happen and, specifically, event-location associations can be studied. We used two event arenas, made of Plexiglas, in which rats were trained to find flavored food. They were placed in adjacent laboratory rooms containing a number of prominent and distinctive cues. The apparatus used has been described previously (S1).

2.

Study 1: Immediate early gene mapping

2.1 Training protocol Twenty-eight rats were used in this study, 7 of which were allocated to a caged control group (Group CC; these rats were handled daily for the first 3-4 days upon arrival, during weekly cage-cleaning, and brought from the animal house to the control room for every session together with the trained groups). Group CC stayed in the home cage at all times and never entered the event arena. The reason for this group was to compare the baseline IEG expression of brain regions with that of the other trained groups. The remainder of the rats (n = 21) underwent behavioral tasks: habituation, pre-training, original paired-associates (PAs) schema training (Sessions 1-17) and critical training session (Session 18).

2

2.1.1 Shaping, habituation and pre-training On the day before habituation began, rats were shaped to dig for food in sand-wells in their home cages. On habituation day 1, the rats were put into the arena containing no sand-wells for 10 min to explore the arena and intramaze cues. For habituation day 2-6, they received one daily habituation trial. In this trial, the animals were put in a start box and given a 0.5 g ‘cue’ control food pellet (non-flavored) to eat. After they had finished eating the pellet (typically around 30 s), the rats were allowed to explore the arena. They started from a different start box in each habituation day. Upon entering the arena, they were to search and dig for control food pellets (0.5 g) in the sand-well in the center location of the event arena. On habituation day 2, one pellet was placed on top of the sand-well. The rats explored the arena until they collected the pellet that, being quite large, was typically taken back to the start box. On habituation day 3, two pellets were provided, one pellet was placed on top of the sand-well and the other buried in the middle of the sand-well. On habituation day 4, three pellets were provided; one pellet was placed on top of the sand-well, two pellets buried in the middle of the sand-well. On habituation day 5, one pellet was placed in the middle of the sand-well, two pellets were buried at the bottom of the sand-well. And on habituation day 6, three pellets were at the bottom of the sand-well. By the end of habituation, all rats were running quickly into the arena, collecting pellets and returning to the start box to eat each pellet. Before commencing the full original schema training protocol, the animals were pretrained to learn the PAs of different flavors specific to different locations. It consisted of 2 sessions, each with 3 of the original 6 PAs introduced in each session. 2.1.2 Acquisition of original schema of 6 flavor location paired-associates (Sessions 1-17) Training protocol: The key feature of the main protocol was the concurrent training of 6 flavor-place PAs [PA = flavor n (Fn) at location n (Ln)] in each session. Each PA was presented to each rat for 1 trial/session: PA1 (strawberry at L1); PA2 (banana at L2); PA3 (chocolate at L3); PA4 (apple and cinnamon at L4); PA5 (marshmallow at L5) and PA6 (bacon at L6) (Fig. S1B). On any trial, all 6 sand-wells were accessible, but only one contained the appropriate flavored reward, with the other 5 containing only the sand and masking flavors mixture. The location of the rewarded sand-well changed from trial to trial. A training trial began with the rat receiving a 0.5 g ‘cue’ flavored pellet in one of the start boxes. After a period of 30 s in the start box during which the rats ate this cue pellet, the door was remotely opened. Initially, the rats would run into the arena, and explore at one or more sand-wells until finding the correct location containing 3 food pellets of the same flavor (0.5 3

g each) as the cue. These were typically carried back in the start box by the rat, pellet by pellet, and eaten in turn. After the rat returned to the start box with the 3rd pellet, the door was closed. A short period of eating time was allowed before the rat was carried back to its home cage. Later in training, as PA learning developed, the animals were more cautious in leaving the start box, presumably because they were attempting to recall the correct location in the arena to which to run. As all rats were trained consecutively, the inter-trial interval for an individual rat between successive flavor-place pairings was circa 30 min. The various possible sequences of different PAs across 6 trials within a session were carefully counterbalanced across rats and sessions. The start box locations were pseudo-randomly assigned (North, South, East or West) across training sessions. Measurement protocol: During each trial, experimenters recorded the number of errors the rats made before approaching the correct sand-well and the time from when the rat left the start box until it reached the correct sand-well. We recorded an ‘error’ only when a rat placed its front paw on or into incorrect sand-well(s), and left a clear impression in the sand of having done so. If a rat ran past or merely sniffed quickly in the vicinity of an incorrect sandwell, this was not considered as an error – and with experience could easily be judged. In rare cases, it was difficult to tell from the video monitors whether or not the rats had made an error as defined above. In this case, the experimenters entered the room at the end of a trial, and then they checked carefully if there were any traces of digging, that is, whether the sand had been displaced around the sand-well(s). The chance level of errors for 5 incorrect sand-wells is 2.5. Using the number of errors, the performance index score is calculated using 100 – [100 × (errors/5)] – i.e. 50% at 2.5 errors. The performance index during acquisition of original PAs is shown in Fig. S2A.

2.1.3 Cued-recall probe trials (Sessions 2, 9 and 16) Protocol: To examine cued-recall memory, 3 non-rewarded probe tests (PTs 1-3) were conducted. During these tests, all 6 sand-wells were open as usual and the rats could dig in any of them, but none contained any food pellets as reward. The rats were cued with a single 4

flavor as usual, and then allowed into the arena for a total of 120 s. Although not rewarded during the 120 s probe test time, the rats were given 3 pellets (correct flavor) in the correct location at the end of each probe test to limit extinction. If rats did not dig during the 120 s, the experimenter waited for 180 s before terminating the trial. Measurement protocol: Digging time at each sand-well was measured semi-automatically using custom built software (developed in LabVIEW by P. Spooner). The experimenter recorded the time rats spent digging at each of 6 sand-wells, and the relative proportion of time at the cued (correct) and non-cued (incorrect) sand-wells was calculated. The cued-recall probe trials during acquisition are shown in Fig. S2B. 2.1.4 Critical training session (Session 18) Training of PAs (Trials 1-6): Once the rats had learned 6 PAs in the schema paradigm, they were divided into 3 groups, with each group having 7 rats: Group original paired-associates (OPA), Group new paired-associates (NPA) and Group new map (NM). Group CC remained in their home-cages as per normal: • The Group OPA training consisted of 6 original PAs. • The Group NPA training consisted of 4 original PAs and 2 new PAs. • The Groups NM training consisted of 6 new PAs. The critical session consisted of 6 trials (Trials 1-6), each trial with a different PA, followed by a cued-recall test after an interval of 80 min (Fig. 1A). For the first 4 trials, both Groups OPA and NPA were presented with original PAs 2-5 and the Group NM was presented with new PAs 9-12. The inter-trial interval was 30 min. To map the IEG expression induced on Trials 5 and 6 of the training without major influence from Trials 1-4, all rats were placed in the home cage for 180 min after Trial 4 and they were sacrificed 90 min after Trials 5 (Fig. 1A). Previous studies have provided evidence that Zif268 and Arc protein expression in the cortex was elevated between 0.5 hr and 2 hr after a novel event (the highest protein level was seen around 1-1.5 hr), returning to the baseline level at time points ranging from 3 to 6 hr (S2, S3). On Trials 5 and 6, Group OPA was presented with original PAs 1 and 6. The Groups NPA and NM were presented with new PAs 7 and 8. The inter-trial interval was 5 min. The animals were then placed in their home cages for 80 min until the following cued-recall test.

5

The performance index and latency to dig at the correct sand-wells on Trials 1-4 and Trials 56 in the critical session are shown in Fig. 1B and Fig. S3 respectively. Eighty minutes after Trial 6, the cued-recall probe test was performed to see if the rats remembered the PAs with which they had been presented on Trial 5. The results of the cuedrecall test are shown in Fig. 1C.

2.2 Immunohistochemistry 2.2.1 Tissue preparation Ninety minutes following the training of Trial 5 in the critical training session, rats were deeply anesthetised with sodium pentobarbital (1.4 ml/kg of body weight, i.p.) and perfused transcardially with 0.1 M phosphate-buffered saline (PBS), pH 7.4 followed by freshly prepared 4% paraformaldehyde in PBS. After excision from the skull, brains were further immersed overnight in the same fixative and then transferred to 50-ml tubes containing 30% sucrose in PBS and kept at 4°C until they sank to the bottom. Coronal brain sections (40 µm) were serially cut using a freezing sliding microtome and divided into 6 interleaved sets. Each of these sets included sections at 240 µm (40 µm × 6) intervals and collected into an antifreeze solution (50% ethylene glycol and 50 % glycerol) and maintained at –20°C for later processing. 2.2.2 Immunoperoxidase staining Two cohorts of rats were used in this experiment. A single experimental cohort of four rats was processed together, sections from each trained rat in the same container as those of its caged control to minimize the impact of any immunohistochemical variation. All immunohistochemical incubations were done at room temperature in a free-floating state. Zif268 labeling: Sections were first rinsed with PBS (2 × 10 min) to remove the antifreeze solution. Sections were then treated with 0.3% hydrogen peroxide in PBS for 30 min to reduce endogenous peroxidase activity. After washing with PBS (3 × 10 min), sections were incubated with rabbit anti-Zif268 polyclonal antibody (1:3000 dilution) in a blocking solution (2% normal goat serum, 0.2% Triton X-100 and 0.1% bovine serum albumin, all dissolved in PBS) for overnight. Sections were then rinsed in PBS (4 × 10 min), after which they were incubated with biotinylated goat anti-rabbit secondary antibody (1:2000 dilution) in the blocking solution for 2 hr. Sections were once again washed with PBS (4 × 10 min) and 6

incubated in avidin and biotinylated horseradish peroxidase (HRP) complex solution for 1 hr. Sections were then rinsed in PBS (4 × 10 min). The peroxidase reaction end-product was visualised with 3, 3’-diaminobenzidine (DAB) as chromogen until suitable staining developed. Finally, immunolabelled sections were washed in PBS (3 × 10 min), mounted on slides, dehydrated through a graded series of alcohols and coverslipped. Arc labelling: A rabbit anti-Arc polyclonal antibody (OP-1, 1:2000 dilution) was raised against recombinant glutathione S-transferase (GST)-fused full-length Arc (S4, S5). The immunostaining protocol for Arc was essentially same as Zif268 labelling, except that tyramide signal amplification (TSA) system was used according to the manufacturers protocol. After incubation with secondary antibody, the sections were incubated with HRPconjugated streptavidin for 30 min, and visualised with DAB as chromogen. 2.2.3 Regions of interest The regions analysed are depicted in Fig. S4. The hippocampus (CA1, CA3 and dentate gyrus (DG)) was of interest because learning of PAs in the schema task was dependent on the hippocampus (S1). The entire extent of the target dorsal hippocampal region was represented by the coronal sections from –2.8 to –3.3 from bregma (S6). Prelimbic (PrL), infralimbic, anterior cingulate, and both the anterior and posterior retrosplenial cortices were of interest because previous studies have shown their importance in either recent or remote memories (S7, S8). Insular cortex was of interest because it is the primary taste cortex (S9), the lateral entorhinal and orbitofrontal cortices were of interest because these cortices were related to the information processing for smell (S10, S11). Somatosensory cortex (Ssp, barrel cortex) was included as a control region because it is not known to have a specific role in spatial, taste or flavor memory (S12). For the cortical regions of interest that were analysed, all 6 cortical layers were included. In CA1/CA3, the pyramidal cells and the stratum oriens layers were included; while in DG, the molecular and the granule cell layers were included. Representative images of both Zif268 and Arc in PrL and CA1 are shown in Fig. 1D and Fig. S5, respectively. 2.2.4 IEG counts For analyses of both Zif268 and Arc counts, quantitative analyses of entire brain regions were conducted ‘blind’ with respect to assignment of rats to group. Two distinct methods were used in this study to estimate the number of Zif268 and Arc positive cells due to differential 7

subcellular distribution of 2 IEGs in neurons (S2, S13). As Zif268 protein is exclusively expressed in nuclei (S2), low magnification (10× objective), quantitative assessment of background staining and objective threshold setting for Zif268 counting were applied. In contrast, Arc protein is expressed in both cell body and dendrites (S13), therefore, in order to distinguish the cytoplasmic Arc staining from dendritic staining, a high magnification (100× objective) and z-stack method for Arc counting were applied. Zif268 counts: Images were taken with a 10× objective on a microscope equipped with a digital camera. The regions of interest for each section were montaged using Image-Pro Plus. The counts of Zif268 stained nuclei that were above a threshold calculated from background labelling were carried out using the same software. After converting to a grayscale image, the region of interest was drawn manually. In order to measure the background labelling within the region of interest on each section, 10 square background boxes (50 × 50 µm each) were drawn on layer 1 for cortical areas, and on the stratum oriens layer for CA1/CA3 and molecular layer for DG. The software analyzed the optical density (OD) of every pixel in the background boxes and calculated the mean OD and standard deviation of background boxes. Using these values, a threshold (T) was then calculated according to the formula [T = mean OD of the background + (standard deviation of the background × 3.75)] and the number of Zif268 stained nuclei that were above this threshold was counted taking into account the size of the Zif268 stained nuclei (> 7 µm in diameter) and applying a watershed separation method in each section. Arc counts: Arc positive cells were defined and counted using a computer-driven microscope regulated in the x, y and z axes. After regions of interest were traced at a 6.4× objective lens by using an image analysis software, Arc positive cells were defined manually under 100× objective lens as the cell bodies staining positively for Arc. Because this microscopic system can semi-automatically scan all over the traced area (i.e.; region of interest) by using 100× objective lens, we counted all Arc positive cells in the traced area of the section in a high magnification. 2.2.5 Data analyses for IEG counts Raw counts: For each brain region analysed, the number of immunoreactive neurons of Zif268 and Arc were counted using 3 consecutive sections within an interleaved set (i.e. 240 µm apart from each other) from each brain hemisphere. This means that in each region there are 6 data points per rat. The ‘raw cell count’ for each section was calculated by dividing the 8

number of immunoreactive neurons by the area of the region of interest. These counts were then averaged within each rat and then within each group (Groups OPA, NPA, NM and CC). Table S1 tabulates the raw Zif268 and Arc cell counts in all brain regions in all groups. Normalised counts: For each brain region, raw counts of each section for Groups OPA, NPA and NM were normalised to the mean raw count of the members of Group CC within a cohort (2 cohort of rats were used in this experiment and sections were stained at different time points). The counts of each section were then averaged within each rat and then within each group and expressed as a percentage. Table S2 tabulates the normalised Zif268 and Arc cell counts in all brain regions in all groups. Layer-specific analysis in PrL: Separate counts of Zif268-positive cells were made in the superficial (layers II/III) and deep (layers V/VI) layers of PrL. The ratio of the measured width of layers II/III to layers V/VI in Nissl stained sections were in golden ratio (0.618), we applied this ratio to a re-counting of Zif268 expression in PrL. Fig. S6 shows the pattern of Zif268 expression with respect to superficial versus deep layers of PrL. 2.2.6 Densitometric analysis The Zif268-immunostaining reaction product in area CA1 of the hippocampus was also quantified using densitometric methods. Mean OD per Zif268-positive cell: Following the Zif268 cell counting as described above, the mean OD values for each Zif268-positive cell that was above the threshold were measured and the mean OD value of 10 background boxes in each section was subtracted. The mean resultant OD values from 6 sections were combined to give the mean OD values for each animal, and were then normalized to the averaged mean OD value of the members of Group CC within a cohort. Normalized mean OD values were used to obtain the frequency distributions for each animal and the averaged frequency distribution within each group (Groups OPA, NPA, NM and CC) was calculated. Fig. S7, A and B show the median values and cumulative frequency distribution curves for relative Zif268 expression level per cell in CA1 in all groups, respectively. Integral OD of Zif268-immunostaining reaction products: After both a region of interest (the pyramidal cells and the stratum oriens layers in CA1) and a background region (the stratum oriens layer in CA1) were manually traced on each grayscale image, the pixel ODs were measured using Image-Pro Plus. We fitted a Gaussian distribution to the pixel OD data 9

originating from both the region of interest and background region and calculated a coefficient ‘C’ by dividing the peak of the Gaussian distribution for the region of interest by that for the background region using a custom built software (developed in LabVIEW by P. Spooner). Integral OD of Zif268-immunostaining reaction products was calculated by summing all pixel ODs after subtracting ODs originating from the background region multiplied the coefficient ‘C’ and by dividing the area of the region of interest (mm2). Raw integral ODs of each section for Groups OPA, NPA and NM were normalized to the mean raw integral OD of the members of Group CC within a cohort. The normalized integral ODs of each section were then averaged within each rat and then within each group and expressed as a percentage. Fig. S7C shows the normalized integral OD values in CA1 in all groups.

2.3 Statistical analyses 2.3.1 Behavioral Several measures of performance were assessed. These were: the performance index and latency before digging at the correct sand-well during training, and time spent digging in each sand-well during probe trials. Statistical significance was determined by repeated-measures ANOVAs, one-way ANOVA and, where appropriate, one-sample t-tests. 2.3.2 Cell counting A mixed factorial ANOVA was used to determine differences between regions and groups in Zif268 and Arc cell counting. A one-way or a two-way ANOVA, where appropriate, was used to determine differences between groups within region(s). Simple main effects were analysed further with the Ryan-Einot-Gabriel Welsch Range (REGWR) post-hoc test that takes account of multiple comparisons. 2.3.3 Densitometric analysis A 2-sample Kolmogorov-Smirnov test with Bonferroni correction was used to determine differences between groups in cumulative frequency distribution curves of mean OD. A Kruskal-Wallis test followed by Dunn’s multiple comparison was used to determine the differences between groups in median values of mean OD. A one-way ANOVA was used to determine differences between groups of integral OD.

10

3.

Study 2: Pharmacological intervention

3.1 Surgery Rats were anaesthetised with 5% isoflurane and positioned in a stereotaxic frame. The top surface of the animals’ heads were shaved and cleaned with disinfectant wipes. Incisions were made and the skin and underlying tissues were retracted to expose the skull. Holes were drilled into the skull and small stainless steel screws inserted at 6 points over the skull. Small holes were then drilled into the skull over the cannulae target sites, guide cannulae (26 gauge; PrL: bilateral cannulae; Ssp: single cannulae) were implanted bilaterally into (1) PrL [coordinates relative to skull at Bregma: AP = 3.0 mm; ML = 0.7 mm; DV (from dura) = –1.9 mm], and (2) Ssp [AP = –1.8 mm; ML = 5.0 mm; DV (from dura) = –0.9 mm]. Dental cement was then sculpted around the guide cannulae, covering the screws. Dummy cannulae (33 gauge, 0.5 mm protrusion from the end of guide cannulae) were inserted into the implanted guide cannulae to prevent infection or blockages. The analgesia carprofen (0.08 ml/kg body weight) was administered by subcutaneous injection at the end of the surgical procedure. All rats were allowed a recovery period of at least 7 days in order for them to regain their pre-surgery weights before food restriction and behavioral testing commenced.

3.2 Drugs and microinfusions 3.2.1 Drugs Sterile 0.9% saline (NaCl) was used as infusion vehicles and for dissolving drugs. Drug concentrations for infusions were: 5.9 mg/ml (30 mM) of the competitive NMDA receptor antagonist D-(-)-2-Amino-5-phosphonopentanoic acid (D-AP5) and 0.89 mg/ml (3 mM) of the competitive AMPA/kainate receptor antagonist 6-Cyano-7-nitroquinoxaline-2,3-dione (CNQX). The pH of the drug solutions was adjusted to 7.2 by the addition of 1 M NaOH (for D-AP5) or of concentrated phosphoric acid (for CNQX). Drug solutions were prepared in larger quantities, divided into 50 µl aliquots, and kept frozen at –20°C until use.

11

3.2.2 Microinfusions For bilateral infusions, rats were restrained manually and infusions into both hemispheres were performed simultaneously. The dummy cannulae were replaced by infusion cannulae (33 gauge) protruded 0.5 mm from the ends of the guide cannulae within the brain, and were connected to microsyringes on a microinfusion pump via flexible polyvinyl chloride tubing. D-AP5, CNQX and NaCl (0.5 µl/hemisphere unless otherwise specified) were infused at a rate of 0.4 µl/min over 1 to 2 min, after which the infusion cannulae were left in place for a further 1 min. The dummy cannulae were then placed back into the guide cannulae.

3.3 Histology All rats were terminally anesthetised with sodium pentobarbital (1.4 ml/kg, i.p.) and then perfused intracardinally with 0.9% saline, followed by 4% formalin. The brains were removed and stored in 4% formalin for several days. Coronal sections (30 µm) were cut using a cryostat with one in every 5 sections recovered for histological analysis. These sections were mounted on slides, stained with cresyl-violet, and coverslipped. The sections were examined under a light microscope with 10× magnification to verify cannulae placements. For each brain, the infusion site was plotted by determining the deepest point at which tissue damage was evident and marking this location on the appropriate coronal sections from rat brain atlas (S6) (Fig. S8).

3.4 Paired-associates task in the event arena The full experimental design of this study is shown chronologically in Table S3. 3.4.1 Training of original PAs and probe tests (Sessions 1-16) The original geometric arrangement of flavor place paired-associates (PAs 1 to 6) is shown in Fig. S9. Each PA was presented for 1 trial/session [PA1 = F1 (strawberry) at L1, PA2 = F2 (pina colada) at L2, PA3 = F3 (chocolate) at L3, PA4 = F4 (very berry) at L4, PA5 = F5 (marshmallow) at L5, PA6 = F6 (bacon) at L6]. The acquisition of these original 6 PAs is shown in Fig. S10A. To examine cued-recall memory, three non-rewarded probe tests (PTs 1-3) were scheduled on Sessions 2, 9 and 16 (Fig. S10B). To exclude the possibility that an olfactory cue in the correct sand-well guided performance on training sessions, we conducted a single non-cued session of 6 trials (Fig.

12

S10C, Session 43) in which the daily protocol was unchanged excepting that no cue flavors were given in the start box. Once the rats had learned 6 original PAs (schema), we investigated the impact of blocking AMPA or NMDA receptor function during memory retrieval and new learning. Druginfusions targeted at these receptors occurred either (1) on days after any new learning and shortly before memory retrieval tests, or (2) before new encoding sessions with retrieval tests conducted 24 hr later. Standard training sessions on PAs 1-6 were interpolated between these tests (Fig. S10C), with the logical order (Table S3) in which the data is reported in the main text differing, for practical reasons, from that in which the experiments were conducted. The experiment was a within-subject, repeated measures study. 3.4.2 Effects of CNQX and D-AP5 infusions in PrL and barrel cortices on the retrieval of original and new PAs Impact of CNQX on retrieval of original PAs (Sessions 17-22): The rats were presented with 2 sessions (17 and 18) of original PAs. Then, 24 hr later (Session 19), CNQX or NaCl was infused 20 min before probe tests (PTs 4 and 5) in one brain region (either PrL or barrel cortex). These probe tests were 3 hr apart within the same session. After 2 further sessions (20 and 21) of original training, probe trials (PTs 6 and 7) were repeated to counterbalance the different conditions across the rats. Fig. 3A shows the effect of CNQX on the percentage dig time during retrieval of original PAs. Impact of CNQX on retrieval of new PAs (Sessions 23-31): The rats were then given further training on the original PAs for 2 sessions (23 and 24). Then, on Session 25, the sand-wells for PA1 and PA6 were closed, and replaced by another 2 new PAs at neighboring locations: PA7 (apple) and PA8 (paprika) (Fig. S9). Rats were trained for a total of 6 trials: 1 trial for each of the 2 new PAs and the 4 trials for the remaining PAs of the original schema (i.e. PAs 2-5). Twenty-four hr later, on Session 26, drugs (CNQX or NaCl) were infused 20 min before the probe trials (PTs 8-9). These probe trials were designed to test the rats’ memory for the new PAs that had been encoded 24 hr earlier. The performance measures were designed to contrast digging time spent at the new cued location, with that for both the new non-cued location and the original non-cued locations.

13

Three further sessions (27-29) of original training were given to the rats, followed by 2 further new PAs (PA9 = almond, PA10 = cinnamon) training (Fig. S9). Probe trials (PTs 10 and 11) were scheduled 24 hr after the new PAs training sessions. Fig. 3B shows the effect of CNQX on the percentage dig time during retrieval of new PAs. Impact of D-AP5 on retrieval of original PAs (Sessions 54-58): On Sessions 54-57, rats were given training on the original PAs. Twenty-four hr later, on Session 58, drugs (D-AP5 or NaCl) were infused into PrL 20 min before the probe tests (20 and 21). Fig. 3C shows the effect of D-AP5 on the percentage dig time during retrieval of original PAs. Impact of D-AP5 on retrieval of new PAs (Sessions 49-53): On Sessions 49-51, rats were given training on the original PAs. On Session 52, two new PAs were introduced: PA17 (orange) and PA18 (chocolate and apple) (Fig. S9). Twenty-four hr later, on Session 53, drugs (D-AP5 or NaCl) were infused into PrL 20 min before the probe trials (18 and 19). Fig. 3D shows the effect of D-AP5 on the percentage dig time during retrieval of new PAs. Additional performance measures - Absolute dig time: The absolute dig time is shown for various probe tests in Fig. S11, A, B, D and E. This measure enabled us to examine differences in absolute dig time between different drug conditions in different regions during the 120 s probe tests. Impact of CNQX on animals’ performance - Control experiment 1: We conducted a control experiment to rule out the possibility that targeted pharmacological inactivation by CNQX into PrL interfered with motor, motivational or procedural processes. This involved training rats to search and dig for control food pellets (non-flavored) in the sand-well in the center location of the event arena for 1 trial. The rationale for this was that no spatial memory was required for this task – the animals had only to run into the arena to the center and dig. Twenty-four hr later, drugs (CNQX or NaCl) were infused into PrL 20 min before the probe trial. Fig. S11C shows the absolute dig time between the 2 drug conditions (CNQX and NaCl) during the 120 s probe test with a single sand-well in the center of the event arena.

14

3.4.3 Effects of CNQX and D-AP5 infusions in PrL at encoding of new PAs The aim of this aspect of the main PA experiment was to investigate whether AMPA and NMDA receptors in PrL were involved in the encoding of new PAs. This was achieved by infusing the CNQX and D-AP5 into PrL prior to encoding. Infusions were carried out 20 min before the start of encoding trials and memory was tested in probe trials 24 hr later. Impact of CNQX on encoding of new PAs (Sessions 32-41): The rats were given further training on the original PAs for 3 sessions (32-34). On Session 35, the sand-wells for PA1 and PA6 were replaced by another 2 new PAs at neighboring locations: PA11 (cherry) and PA12 (ginger) (Fig. S9). Rats were trained for a total of 6 trials - 1 trial for each of the 2 new PAs (Trials 2 and 6) and the 4 trials for the original PAs (i.e. PAs 2-5). Drugs (CNQX or NaCl) were infused 20 min before Trial 2 and again before Trial 6 (there was 5 hr between Trials 2 and 6) in one brain region (either PrL or barrel cortex). Twenty-four hr later, on Session 36, two probe trials (PTs 12 and 13) were scheduled. Three further sessions (37-39) of original training were given, and then 2 further new PAs (PA13 = rum, PA14 = butter) were introduced (Fig. S9). Probe trials (PTs 14 and 15) were then scheduled 24 hr after the new PAs training sessions. Fig. 4A shows the effect of CNQX administered 24 hr earlier (during encoding) on the percentage dig time. Impact of D-AP5 on encoding of new PAs (Sessions 45-48 and 59-63): On Sessions 45-46, rats were given training on the original PAs. On Session 47, 2 new PAs were introduced: PA15 (grape) and PA16 (banana and cherry) (Fig. S9). Rats were trained for a total of 6 trials presented with 2 new PAs and 4 original PAs. Drugs (D-AP5 or NaCl) were infused 20 min before Trial 2 and again before Trial 6 in PrL. Twenty-four hr later, on Session 48, two probe trials (PTs 16 and 17) were conducted. This sequence was repeated again during Sessions 5963. Three further sessions (59-61) of original training were given to the rats, and then 2 further new PAs (PA19 = banana, PA20 = apple and cinnamon; Fig. S9) training. Probe trials (22 and 23) were scheduled 24 hr after the new PAs training sessions. Fig. 4B shows the effect of D-AP5 administered 24 hr earlier (during encoding) on the percentage dig time.

3.5 Gap-crossing task – Control experiment 2 3.5.1 Apparatus The apparatus was a modified version of that used by S14, S15. The apparatus consisted of two individually moveable identical platforms (the starting platform and the goal platform) 15

made of transparent Plexiglas (Fig. S12A). The two platforms (11 cm wide × 30 cm long) with 10-cm-high U-shaped walls were elevated by 34 cm. The walls were covered in black adhesive cardboard. Velcro tape was placed at the edges of the platforms to indicate the platform edge to the rats. A 50 cm long ruler was used to measure the distance between the platforms in each trial. A cup containing approximately 10 g of food pellets was available as reward at the goal platform. Each pellet in the cup was cut into half, and each half weighed approximately 0.2 g. To prevent the use of visual information, experiments were conducted under dim red light (wavelength 620–750 nm), invisible to the rats (S15). An infared camera connected to a DVD recorder was placed above the gap to observe and record the rats’ movements in the dark. 3.5.2 Behavioral training Habituation: The purpose of the habituation sessions were to allow the rats to become accustomed to exploring the platforms and getting food reward after they had successfully moved from the starting platform to the goal platform. Normal illumination (100 lux) was present in habituation sessions. On habituation session 1 and 2, rats were placed on the starting platform and the gap distance between the 2 platforms was 0 cm. Rats were free to explore both platforms. A cup of pellets was available at the end of the goal platform. The cup was kept full to encourage the rats to cross readily to the goal platform. Rats were allowed 10 s to eat the pellets, and then picked up and carried to their home cage. Each rat was given 3 trials to obtain rewards from the goal platform. Between trials, the platforms were wiped with ethanol to remove odors, which may have affected animal’s performance. Main training: On Session 1, a 2 cm gap was introduced on the first trial and gap length was increased in 2 cm increments up to 6 cm following 3 successful trials; Session 2 was carried out similarly with gap sizes from 6-10 cm. This was conducted under normal lighting conditions. Sessions 3 followed the same protocol but with gap sizes going from 10-16 cm. From session 3 on, training was carried out under red light. On Sessions 4 and 5, distances of between 12 and 25 cm were imposed in pseudo random order. This was designed to determine the maximum distance that each individual rat would cross. To ensure that rats were obtaining information using only their whiskers, a 30 cm gap trial (a catch trial in which the gap was too wide to cross) was introduced pseudo randomly at least three times per session. By the end of Session 5, the maximum gap distance that each rat would cross was determined. On Session 6, 4 gap distances were introduced pseudo randomly 4 times per 16

session: 8 cm, 14 cm, the maximum distance for individual rat (typically 16-18 cm) and 30 cm (“a gap-too-far”). During each trial of the gap-crossing experiment, the latency for each rat to cross the gap between 2 platforms was measured. First, we recorded the time when a rat placed its front paws on the edge (the Velcro) of the starting platform. Second, we measured the time when the rat placed its front paws on the edge (the Velcro) of the goal platform. The latency for a rat to cross the gap was calculated by taking the difference between these 2 time measurements. Impact of CNQX during the gap-crossing task: On Sessions 7-10, 20 min before the gapcrossing task, rats were injected with CNQX or NaCl (Session 7-8, 0.5 µl in each; Session 910, 2.0 µl in each) in the barrel cortex. There were 16 trials in total, with the four gap distances introduced in pseudorandom order as in training. The first trial of each different gap distance was considered an ‘information’ trial, and the latency data measured on the remaining 3 trials per distance: 8, 14, maximum distance for individual rat (16-18 cm) and 30 cm. Fig. S12A shows the mean latencies for the rats to cross the gap at different distances with a 2.0 µl infusion of CNQX.

3.6 Return to PAs task in event arena Impact of CNQX (2.0 µl) on the encoding of new PAs in the barrel cortex (Sessions 64-68): Following our finding that it was necessary to increase the infusion volume to 2.0 µl to observe a deficit, we then retrained the animals in the event arena to examine whether a same volume (2.0 µl) of CNQX infused into the barrel cortex would affect the encoding of new PAs. After 3 sessions (64-66) of standard training on original PAs, 2 new PAs were introduced: PA21 (lemon) and PA22 (vanilla and brandy) (Fig. S9). Twenty-four hr later, on Session 68, drugs (CNQX or NaCl, 2.0 µl in each) were infused into the barrel cortex 20 min before the probe trials (24 and 25). Fig. S12B presents percentage dig time in the PTs 24 hr after encoding of new PAs with 2.0 µl of CNQX into the barrel cortex.

3.7 Statistical analyses Several measures of performance were assessed. These were: the performance index during training, and both absolute and percentage time spent digging in sand-well(s) during probe

17

trials. Statistical significance was determined by repeated-measures ANOVAs (2 or 3 way ANOVA), paired t-tests and, where appropriate, one-sample t-tests.

18

A

B

Start box Door PA1

PA2 PA3 PA4

PA5

PA6

Fig. S1. (A) Photograph of an event arena. (B) The spatial arrangement of original 6 PAs of the standard schema.

19

*** *

50 40 30

2

4

6

8 10 12 Sessions

14

16

18

Probe trials during acquisition 70

0.5

60

1.0

50

1.5 2.0 2.5

% dig time

60

6

No. of errors

70

4

Critical session + PT4

80

5

0

2

3

PT3

90

1

PT2

100

Performance index

B

Acquisition of original paired-associates

PT1

A

1 5

4

30

3.0

10

3.5

0

6

*

40 20

orig. cued orig. non-cued

2

3

***

ns

PT1

PT2

PT3

Fig. S2. Performance of training in Study 1. (A) Acquisition of the original 6 PAs (schema). Training on the standard ‘schema’ in the event arena revealed a gradually improving ‘performance index’ across sessions (F = 20.32, df = 6.75/134.92, P < 0.001). Performance rose to a stable level of 70% with minimal variability (from Session 5 onwards, performance was above chance, t-test). (B) Three cued-recall probe trials for the acquisition of the original schema. These probe trials revealed, a graded learning of the original PAs from Sessions 1-16 (F = 10.95, df = 1.78/35.50, P < 0.001). T-test results indicated that digging at the cued location was above chance in PT2 (P < 0.05) and PT3 (P < 0.001). Dashed lines indicate the chance level. ns, nonsignificant; *P < 0.05; ***P < 0.001 versus chance. Means ± 1 S.E.M.

20

Latency during encoding 100

Latency (s)

80

OPA NPA NM

60 40 20 0

T1-4

T5-6

Fig. S3. Trial latency to dig at the correct sand-well on the critical session 18. An ANOVA of these latency scores revealed no overall interaction between Groups OPA, NPA and NM in Trials (T)1-4 and T5-6 (F = 2.19, df = 2/18, ns). There are no significant differences between Groups OPA, NPA and NM in both T1-T4 and T5-T6 (Fs < 1, ns). Means ± 1 S.E.M.

21

3.2

1.2

0.7 ACC

PrL IL

Ins

Orb

–1.4

–2.8

–5.6

aRSC CA1 Ssp DG

pRSC

CA3

LEnt

Fig. S4. Areas examined for Zif268 and Arc expression. Blue regions indicate the areas were selected for measurement of Zif268 and Arc expression. The numbers above each picture represent the distance (in mm) of the section from bregma (S6). IEG counts for the retrosplenial cortex (RSC) was the average normalised cell count of the anterior retrosplenial (aRSC) and posterior retrosplenial (pRSC) cortices. ACC, anterior cingulate cortex; DG, dentate gyrus; IL, infralimbic cortex; Ins, insular cortex; LEnt, lateral entorhinal cortex; Orb, orbitofrontal cortex; PrL, prelimbic cortex; Ssp, somatosensory (barrel) cortex.

22

IEG expression in CA1

NPA

NM

CC

Arc

Zif268

OPA

Fig. S5. IEG expression in area CA1 of the dorsal hippocampus for both Zif268 and Arc. Scale bars, 50 µm.

23

Superficial

Zif268 counts (% CC)

250 200 150

Deep

OPA NPA NM CC

†

**

**

Superficial

Deep

*

100 50

Fig. S6. A layer-specific analysis of the pattern of Zif268 expression in PrL. There was no differential pattern for Zif268 expression with respect to superficial versus deep layers of PrL (Layer x Group interaction, F < 1, ns). Scale bar, 100 µm. *P < 0.05, **P < 0.01 versus Group CC (REGWR test); †P < 0.05 versus trained group. Means ± 1 S.E.M.

24

80

OPA NPA NM CC

60 40 20 0 0.1 1.0 10.0 Relative Zif268 expression per cell

B 150 100 50 0

OPA NPA NM CC

C 150

Integral OD (% CC)

Cumulative %

100

Mean OD per cell (% CC)

A

100 50 0

Fig. S7. Densitometric analysis of CA1. (A and B) Cumulative frequency curves (A) and median (B) of mean OD per Zif268-positive cell for different groups. There was no significant difference in the cumulative frequency distributions (Kolmogorov-Smirnov test with Bonferroni correction, ns) or in median OD values (Kruskal-Wallis test, ns) between groups. (C) The integral OD (per 1 mm2) of Zif268 immunostaining reaction products for different groups. There was no significant difference in the integral OD between groups (F = 1.37, ns).

25

A

PrL

Ssp

B

Image C

4.2

D

–0.4 –0.8

3.7

–0.9 –1.4

Fig. S8. Histology of cannula positions. (A and B) Cresyl violet sections showing example of representative cannulae tracks in PrL (A) and barrel cortices (B) in each hemisphere of the brain. Scale bars, 1 mm. (C and D) Schematic representations to indicate the location of the cannula tips in PrL (C) and barrel (D) cortices. The numbers above each picture represent the distance (in mm) of the section from bregma (S6).

26

PA7 PA17 PA11 PA1

PA9

PA2

PA19 PA15 PA13 PA21 PA3

PA4 PA22 PA14 PA16 PA20 PA10 PA6 PA12

PA5

PA18 PA8

Fig. S9. Spatial arrangement of the PAs in the event arena in Study 2. White circles (PAs 1-6) are original PAs (a schema) and pink circles (PA7 to PA22) are new PAs. On each occasion of the new PAs trials, there were always two original PAs (e.g. PA1 and PA6) were removed and two new PAs were introduced (eg. PA7 and PA8), and upon return to the main training, there was a return to the PAs1-6 arrangement.

27

% dig time

ns

**

PT2

PT3

40 30 20

ns

10

3.5

0

PT1 D-AP5 (0.5 μl)

CNQX (2 μl) 0 0.5 1.0 No. fo errors

PT 24&25

PT 22&23

PT 20&21

PT 18&19

Encoding Retrieval Retrieval Encoding Encoding of new PAs of new PAs of orig. PAs of new PAs of new PAs

PT 16&17

Encoding of new PAs

PT 14&15

CNQX (0.5 μl)

Retrieval of new PAs

16

PT 12&13

Retrieval of orig. PAs

9 Sessions

PT 10&11

12

PT 8&9

80

PT 3

3.0

80

50

2.0

40

orig. cued orig. non-cued

60

1.5

2.5

PT 6&7

Performance index

90

**

90 70

1.0

50

PT 4&5

100

**

***

0.5 No. of errors

60

***

30

C

PT 2

80 Surgery

Performance index

90

70

B

0 PT 1

100

Non-cued

A

70

1.5

60

2.0

50

2.5

40

3.0

30

17 19

22

26

31

35 36

40 41 43 Sessions

47 48

53

58

62

63

67 68

3.5

Fig. S10. Performance of training and testing in Study 2. (A) Acquisition of original 6 PAs. The performance index scores showed that performance improved across all training sessions and a repeated measures ANOVA showed that this was significant (F = 19.22, df = 5.67/56.68, P < 0.001). From Session 5 onwards, the performance index score was between 70% and 80%, and which is significantly different from chance (t-test, Ps < 0.01). (B) Three cued-recall probe trials for the acquisition of original schema. These probe trials revealed a graded learning of the original PAs from Sessions 1-16 (F = 4.17, df = 1.46/14.61, P < 0.05). T-test results indicated that digging at the cued location was above chance in PT3 (P < 0.01 versus chance). (C) Performance during pharmacological intervention. On Session 43, removing cue flavors from the start box results in performance dropping to 60% and then returning to 80% correct on a succeeding normal session (see S1). Dashed lines indicate the chance level. ns, nonsignificant; **P < 0.01; ***P < 0.001. Means ± 1 S.E.M.

28

A 30 25

CNQX

Dig time (s)

20

NaCl CNQX

***

B30

Retrieval of new PAs

ns

25 ns

ns

20

Control experiment

C30 25 20

15

15

15

10

10

10

5

5

5

0

D 30 25 Dig time (s)

20

D-AP5

Retrieval of original PAs

15

PrL

PT 4-7

NaCl D-AP5

ns

Ssp

0

PrL

25 20 ns

15 10

5

5 PrL PT 20-21

0

E 30

10

0

PrL Ssp PT 8-11

ns

0

PrL PT 18-19

Fig. S11. Impact of CNQX or D-AP5 infusions on rats’ performance in the event arena. (A) Infusion of CNQX into PrL resulted in a significantly lower absolute dig time than the rats infused with NaCl into PrL during retrieval of original PAs (t-test; t = 5.49, df = 10, P < 0.001). In contrast, infusion of CNQX in Ssp had no impact on the performance of the retrieval of original PAs, with paired t-test showing no significant different in drugs conditions (ns). (B) Infusion of CNQX in both PrL and Ssp had no impact on the performance of the retrieval of new PAs, with paired t-test showing no significant different in drugs conditions in PrL and Ssp (ns, in each). (C) Effects of CNQX infusions in PrL prior to the probe trial with a single sand-well in the center of the event arena. There were no differences in absolute dig time between the 2 conditions (ns). This result suggested that infusion CNQX into PrL did not affect motor, motivational or procedural brain process. (D and E) There were no differences in absolute dig time between D-AP5 and NaCl infusions into PrL during retrieval of original PAs (D) and of new PAs (E) (ns in each). ns, nonsignificant; ***P < 0.001. Means ± 1 S.E.M.

29

A Gap-crossing task Starting point

Gap distance

Starting platform

Reward

Goal platform

New PAs encoding

Probe test

0h

24 h new cued new non-cued orig. non-cued

90

Ruler

NaCl CNQX (2 μl)

†

80 70 60

**

***

% dig time

Latencies (s)

15

B Encoding of new PAs

10

50 40 30

5

20 10

0

0

8 cm

14 cm Distances

Max

NaCl CNQX (2 μl) Ssp PT 24-25

Fig. S12. Control experiments for the extent of drug used in barrel cortex in Study 2. (A) Gap-crossing test. Bilateral infusion of 0.5 μl of CNQX was insufficient to significantly slow down this task, even at the maximum distance that rats could reach across using their whiskers - though a trend towards slowing was apparent (data not shown). Hence, a larger volume (2.0 μl) was infused into the barrel cortex to address this concern. CNQX infusion into the barrel cortex with the higher volume of 2.0 μl bilaterally caused the predicted slowing of gap-crossing on the runway as the distance that rats could reach across using their whiskers increased (Drug × Distance interaction, F = 5.95, df = 1.02/6.11, P < 0.05), which an orthogonal comparison of CNQX and NaCl at the maximum distance revealed to be significant (F = 13.03, df = 1/6, P < 0.05). (B) This same infusion volume of CNQX into the barrel cortex had no impact on memory encoding of new PAs (Drug × Dig location interaction, ns). Dashed line indicates the chance level. †P < 0.05 versus NaCl group. **P < 0.01, ***P < 0.001 versus chance. Means ± 1 S.E.M.

30

Zif268

Arc

Area

OPA

n NPA

n NM

n CC

n

PrL

204.9 ± 17.5

** 6 254.9 ± 13.8

6 190.9 ± 26.8

6 148.5 ± 10.2

6

6

6

6

IL

79.7 ± 12.3

83.3 ± 5.5

60.6 ± 9.0

83.4 ± 13.2*

n NPA

n CC

n

***,†††,‡‡‡ 7 146.1 ± 11.1 7

n NM 72.8 ± 12.9

7 44.6 ± 4.4

7

6

15.5 ± 1.7

7

16.5 ± 1.3

7

13.4 ± 2.2

7 13.8 ± 0.9

7

Orb

222.1 ± 20.8

* 6 264.3 ± 25.0

6 211.6 ± 31.9

6 151.0 ± 26.5

6

66.9 ± 6.8

7

76.4 ± 6.8

7

66.9 ± 10.5

7 66.9 ± 5.2

7

Ins

143.1 ± 11.0

6 143.0 ± 9.1

6 159.7 ± 28.0

6 135.4 ± 28.4

6

45.0 ± 5.9

7

45.5 ± 5.6

7

43.2 ± 6.1

7 41.0 ± 4.5

7

ACC

272.8 ± 27.4

6 317.1 ± 29.5

6 268.4 ± 41.6

6 202.2 ± 28.6

6

70.5 ± 5.9***

7

90.8 ± 8.2***,‡‡

7

52.9 ± 7.8*

7 29.4 ± 4.8

7

Ssp

210.3 ± 22.6

6 229.2 ± 23.6

6 237.1 ± 41.6

6 191.8 ± 40.2

6

85.7 ± 7.2***

7

28.3 ± 4.6†††,‡‡‡

7

90.2 ± 9.2***

7 36.5 ± 5.2

7

aRSC

107.5 ± 20.8

6 139.1 ± 20.1

5 109.7 ± 29.3

5 100.2 ± 18.1

6

86.0 ± 11.1

***,†††,‡‡‡ 7 7 150.4 ± 13.4

77.2 ± 10.9

7 53.3 ± 7.9

7

CA1

222.7 ± 20.6

6 285.6 ± 26.6

5 294.4 ± 27.3

5 259.2 ± 20.5

6

20.8 ± 4.9

7

89.9 ± 15.6***,†††

***,††† 7 101.3 ± 11.0

7 15.6 ± 2.7

7

CA3

16.2 ± 5.4

6

16.3 ± 4.3

5

13.9 ± 3.7

5

18.0 ± 9.6

6

16.4 ± 4.0

7

36.1 ± 6.3**,††

7

37.4 ± 3.9**,††

7 12.7 ± 2.2

7

DG

16.5 ± 4.4

6

17.0 ± 4.8

5

19.7 ± 7.0

5

18.4 ± 5.1

6

8.7 ± 2.1

7

21.1 ± 4.1**,†

7

14.4 ± 2.7

7

6.9 ± 1.0

7

pRSC

68.7 ± 9.8

* 6 111.7 ± 18.2

6

80.2 ± 22.2

5

49.0 ± 12.1

6

64.2 ± 6.9

7

68.1 ± 3.6

7

61.7 ± 9.9

7 60.0 ± 5.2

7

LEnt

71.4 ± 15.3

6

5

71.1 ± 29.6

5

46.2 ± 10.6

6

49.4 ± 8.2

7

57.8 ± 6.0

7

49.0 ± 5.8

7 45.7 ± 3.6

7

78.5 ± 16.9

66.5 ± 17.9

OPA

Data were expressed as mean raw counts (per 1 mm2) ± SEM. n, numbers of animal. Significantly different from Group CC (REGWR test, *P < 0.05, **P < 0.01, ***P < 0.001). Significantly different from Group OPA (†P < 0.05, ††P < 0.01, †††P < 0.001). Significantly different from Group NM (‡P < 0.05, ‡‡P < 0.01, ‡‡‡P < 0.001).

Table S1. Zif268 and Arc raw cell counts in Groups OPA, NPA, NM and CC.

31

Zif268

Arc

Area

OPA

n NPA

n NM

n CC

n

OPA

n NPA

PrL

137.9 ± 11.6

**,‡ 6 172.8 ± 8.0

6 129.6 ± 18.2

6 100.0 ± 6.7

6

186.5 ± 27.6

***,†††,‡‡‡ 7 345.3 ± 35.5 7 168.8 ± 30.7

IL

146.6 ± 30.7

6 165.0 ± 18.5

6 132.5 ± 41.9

6 100.0 ± 7.1

6

112.4 ± 12.1

7 122.4 ± 10.5

7

99.0 ± 16.4

7 100.0 ± 6.5

7

Orb

148.0 ± 14.3

* 6 179.0 ± 14.0

6 141.9 ± 18.4

6 100.0 ± 16.4

6

100.0 ± 10.1

7 115.7 ± 10.8

7 100.9 ± 16.0

7 100.0 ± 8.0

7

Ins

107.2 ± 10.4

6 103.6 ± 9.3

6 118.3 ± 23.4

6 100.0 ± 21.8

6

109.0 ± 13.2

7 110.5 ± 14.5

7 105.0 ± 15.8

7 100.0 ± 11.2

7

ACC

139.7 ± 16.8

* 6 171.0 ± 13.3

6 141.7 ± 18.2

6 100.0 ± 9.9

6

239.2 ± 18.2**

***,‡‡ 7 317.2 ± 32.3

* 7 185.0 ± 27.8

7 100.0 ± 15.5

7

Ssp

109.9 ± 12.3

6 119.0 ± 13.5

6 123.9 ± 23.3

6 100.0 ± 21.4

6

236.6 ± 20.0*** 7

*** 7 259.3 ± 24.9

7 100.0 ± 12.2

7

aRSC

103.9 ± 17.7

6 139.8 ± 17.6

5 106.6 ± 25.6

5 100.0 ± 16.7

6

161.7 ± 19.6

***,†††,‡‡‡ 7 151.2 ± 20.9 7 303.2 ± 37.4

7 100.0 ± 12.9

7

6 108.3 ± 8.1

5 112.3 ± 9.0

5 100.0 ± 6.7

6

134.5 ± 32.8

***,††† 7 594.6 ± 106.1

***,††† 7 660.3 ± 73.4

7 100.0 ± 16.3

7

6 116.3 ± 31.9

5 124.7 ± 41.9

5 100.0 ± 23.2

6

130.1 ± 32.5

**,†† 7 291.2 ± 51.9

**,†† 7 298.3 ± 31.0

7 100.0 ± 16.4

7

6

5 100.8 ± 30.4

5 100.0 ± 26.9

6

126.4 ± 30.1

**,†† 7 301.1 ± 56.7

7 206.7 ± 37.5

7 100.0 ± 15.2

7

CA1

86.7 ± 7.8

CA3

119.6 ± 25.6

DG

85.3 ± 18.2

86.1 ± 19.9

85.3 ± 17.1†††,‡‡‡

n NM

n CC

n

7 100.0 ± 9.4

7

pRSC

139.7 ± 16.2

**,†,‡ 6 159.4 ± 38.4 6 241.3 ± 20.6

5 100.0 ± 21.5

6

106.6 ± 9.8

7 118.2 ± 8.8

7 106.1 ± 17.3

7 100.0 ± 8.2

7

LEnt

146.6 ± 15.9

6 174.6 ± 15.9

5 100.0 ± 21.7

6

108.4 ± 18.0

7 127.6 ± 13.7

7 108.1 ± 12.5

7 100.0 ± 8.0

7

5 135.8 ± 47.5

Data were expressed as mean ± SEM. n, numbers of animal. Significantly different from Group CC (REGWR test, *P < 0.05, **P < 0.01, ***P < 0.001). Significantly different from Group OPA (†P < 0.05, ††P < 0.01, †††P < 0.001). Significantly different from Group NM (‡P < 0.05, ‡‡P < 0.01, ‡‡‡P < 0.001).

Table S2. Zif268 and Arc normalised cell counts in Groups OPA, NPA, NM and CC.

32

Order

Sessions

Task

1

1-16

Event arena

2

3

4

17-22

23-31

32-41

Event arena

Event arena

Event arena

Experimental condition Original PAs training

Impact of CNQX on retrieval of original PAs

Impact of CNQX on retrieval of new PAs

9

59-63

Event arena

Impact of D-AP5 on encoding of new PAs

11

12

– – 64-68

Event arena

Gap-crossing Event arena

Imapct of CNQX on rats' performance

Gap-crossing task

Impact of CNQX (2 l) on encoding of new PAs

PAs 17-18

PTs 22-23

PAs 19-20

–

–

–

–

PTs 24-25

PAs 21-22

Table S3. Chronological order of behavioral experiments in Study 2.

33

9

4A, S10C

9

4B, S10C

9

PAs 11-14

– PTs 18-19

3A, S10C

PAs 1-6

PAs 1-6

Impact of D-AP5 on retrieval of new PAs

11

10

PTs 20-21

Event arena

10

PAs 1-6

PAs 7-10

Impact of D-AP5 on retrieval of original PAs

49-53

Event arena

PTs 4-7

PTs 8-11

PAs 15-16

7

54-58

S10A, B

PTs 16-17

Non-cued

8

Figure

11

Impact of D-AP5 on encoding of new PAs

Event arena

Event arena

No. of rats

PAs 1-6

PTs 12-15

42-44

45-48

PAs

PTs 1-3

Impact of CNQX on encoding of new PAs

5

6

Probe trials

3B, S10C S10C

9

3D, S10C

9

4B, S10C

9 9

7

7

3C, S10C S11C

S12A

S12B

References S1.

Tse D, Langston RF, Kakeyama M, Bethus I, Spooner PA, Wood ER, Witter MP, Morris RGM (2007) Schemas and memory consolidation. Science 316:76-82.

S2.

Zangenehpour S, Chaudhuri A (2002) Differential induction and decay curves of c-fos and zif268 revealed through dual activity maps. Brain Res Mol Brain Res 109:221-225.

S3.

Ramírez-Amaya V, Vazdarjanova A, Mikhael D, Rosi S, Worley PF, Barnes CA (2005) Spatial exploration-induced Arc mRNA and protein expression: evidence for selective, network-specific reactivation. J Neurosci 25:1761-1768.

S4.

Kawashima T, Okuno H, Nonaka M, Adachi-Morishima A, Kyo N, Okamura M, Takemoto-Kimura S, Worley PF, Bito H (2009) Synaptic activity-responsive element in the Arc/Arg3.1 promoter essential for synapse-to-nucleus signaling in activated neurons. Proc Natl Acad Sci U S A 106:316-321.

S5.

Okuno H, et al., submitted.

S6.

Paxinos G, Watson C (1998) The rat brain in stereotaxic coordinates. San Diego: Academic Press.

S7.

Maviel T, Durkin TP, Menzaghi F, Bontempi B (2004) Sites of neocortical reorganization critical for remote spatial memory. Science 305:96-99.

S8.

Frankland PW, Bontempi B, Talton LE, Kaczmarek L, Silva AJ (2004) The involvement of the anterior cingulate cortex in remote contextual fear memory. Science 304:881-883.

S9.

Verhagen JV, Kadohisa M, Rolls ET (2004) Primate insular/opercular taste cortex: neuronal representations of the viscosity, fat texture, grittiness, temperature, and taste of foods. J Neurophysiol 92:1685-1699.

S10. Haberly LB, Price JL (1977) The axonal projection patterns of the mitral and tufted cells of the olfactory bulb in the rat. Brain Res. 129:152-157. S11. Schoenbaum G, Roesch M (2004) Orbitofrontal cortex, associative learning, and expectancies. Neuron. 47:633-636.

34

S12. Vann SD, Brown MW, Erichsen JT, Aggleton JP (2000) Using fos imaging in the rat to reveal the anatomical extent of the disruptive effects of fornix lesions. J Neurosci 20:8144-8152. S13. Lyford GL, Yamagata K, Kaufmann WE, Barnes CA, Sanders LK, Copeland NG, Gilbert DJ, Jenkins NA, Lanahan AA, Worley PF (1995). Arc, a growth factor and activity-regulated gene, encodes a novel cytoskeleton-associated protein that is enriched in neuronal dendrites. Neuron. 14:433-445. S14. Hutson KA, Masterton RB (1986) The sensory contribution of a single vibrissa's cortical barrel. J Neurophysiol 56:1196-1223. S15. Harris JA, Petersen RS, Diamond ME (1999) Distribution of tactile learning and its neural basis. Proc Natl Acad Sci U S A 96:7587-7591.

35