Locus coeruleus and dopaminergic consolidation of everyday memory - PubMed (original) (raw)

. 2016 Sep 15;537(7620):357-362.

doi: 10.1038/nature19325. Epub 2016 Sep 7.

Affiliations

Locus coeruleus and dopaminergic consolidation of everyday memory

Tomonori Takeuchi et al. Nature. 2016.

Abstract

The retention of episodic-like memory is enhanced, in humans and animals, when something novel happens shortly before or after encoding. Using an everyday memory task in mice, we sought the neurons mediating this dopamine-dependent novelty effect, previously thought to originate exclusively from the tyrosine-hydroxylase-expressing (TH+) neurons in the ventral tegmental area. Here we report that neuronal firing in the locus coeruleus is especially sensitive to environmental novelty, locus coeruleus TH+ neurons project more profusely than ventral tegmental area TH+ neurons to the hippocampus, optogenetic activation of locus coeruleus TH+ neurons mimics the novelty effect, and this novelty-associated memory enhancement is unaffected by ventral tegmental area inactivation. Surprisingly, two effects of locus coeruleus TH+ photoactivation are sensitive to hippocampal D1/D5 receptor blockade and resistant to adrenoceptor blockade: memory enhancement and long-lasting potentiation of synaptic transmission in CA1 ex vivo. Thus, locus coeruleus TH+ neurons can mediate post-encoding memory enhancement in a manner consistent with possible co-release of dopamine in the hippocampus.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing financial interests. Readers are welcome to comment on the online version of the paper.

Figures

Figure 1

Figure 1. Novelty exploration after memory encoding enhances memory retention

a, Everyday spatial memory task in event arena. Mice [_n_=13; 100 sessions (Ss)) learned and maintained stable performance (S25–S100: _F_14,168=1.68, _P_>0.05). Non-encoding control session performance(S63, orange arrow) dropped to 59.6% (S61–S65: _F_4,48=3.63, P<0.05; S63: _t_-test vs. chance, _t_12<1). Green circles, 5S average; white circles, 1S average; Pre, pre-training. **b**, Memory persistence (probe tests) declined from memory at 1h to chance at 24h (1h vs. 24h: _t_12=2.94, _P_<0.05; 1h: _t_12=4.44, _P_<0.001; 24h: _t_12=1.42, _P_>0.05). Novelty 30 min after encoding resulted in memory at 24h (No novelty vs. Novelty: _t_12=2.24, P<0.05; Novelty: _t_12=3.17, P<0.01). c, Blockade of hippocampal D1/D5 receptor (SCH) but not β-adrenoceptors (Prop) during novelty abolished 24h memory (_F_2,24=3.83, P<0.05; Vehicle: _t_12=4.38, P<0.001; Prop: _t_12=3.33, P<0.01). HPC, hippocampus; Prop, propranolol; SCH, SCH23390. **P<0.01, ***P<0.001 vs. chance. Dashed lines=chance level. Means±s.e.m.

Figure 2

Figure 2. LC-TH+ neurons show stronger modulation by novelty than VTA-TH+ neurons

a, Viral injection and optetrode implantation. Putative VTA-TH+ and LC-TH+ neurons responded to blue light (blue). b, Behavioural protocol. c, Raster plot of VTA-TH+ (top) and LC-TH+ neurons (bottom) in familiar (left) and novel (right) environments. FR, mean firing rate. d, Firing rates of VTA-TH+ (_n_=15 neurons from 5 mice) and LC-TH+ (_n_=10 neurons from 3 mice) neurons were higher in the novel environment (VTA-TH+: _t_14=4.31, P<0.01; LC-TH+: _t_9=3.45, _P_<0.01). Dashed lines=baseline. **e**, LC-TH+ neurons showed stronger modulation by novelty than VTA-TH+ neurons (Brain Area×Condition interaction, _F_29,667=2.28, _P_<0.001). LC-TH+ but not VTA-TH+ neurons displayed habituation to novelty (Brain Area×Condition×Time interaction, _F_29,667=2.03, _P_<0.01; main effect of Time for LC novel, _F_29,261=1.70, _P_<0.05; for VTA novel, _F_29,406=1.24, _P_>0.05). **P<0.01, paired _t_-test. Means±s.e.m.

Figure 3

Figure 3. TH+ axons in the hippocampus originate from LC-TH+ neurons

a–d, Representative coronal sections show overall distribution of eYFP+ axons from VTA (a) and LC (c) in dorsal hippocampus. Triple immunofluorescence for eYFP (green), TH (red) and NET (blue) shows co-labelling of eYFP+ VTA axons with TH (b, top) but not with NET (b, bottom; arrows), co-labelling of eYFP+ LC axons with TH (d, top) and NET (d, bottom). e, f, Quantification of area occupied by eYFP and TH double-positive axons (e), and the ratio of eYFP and TH double-positive axons relative to all TH+ axons (f) in CA1, CA3, and DG (_n_=9 slices from 3 mice/group). Both measures indicate stronger TH+ projections from LC than from VTA in CA1 (Area: _t_16=7.4, P<0.001; Ratio: _t_16=104.1, P<0.001), CA3 (Area: _t_16=11.7, P<0.001; Ratio: _t_16=59.0, P<0.001) and DG (Area: _t_16=10.8, P<0.001; Ratio: _t_16=76.4, P<0.001). ***P<0.001, paired _t_-test. Means±s.e.m.

Figure 4

Figure 4. Optogenetic activation of LC-TH+ neurons enhances memory persistence

a, Viral injection, and optic fibre and drug cannulae implantations. Th-Cre mice were injected with a Cre-inducible ChR2-eYFP AAV (ChR2+, _n_=8) or a control eYFP AAV (ChR2−, _n_=6) into LC and VTA. b, Optogenetic burst protocol used in the event arena experiment. c, Design for the optogenetic mimicry experiment. d, Left, LC-TH+ neuron photostimulation (LC-ON) 30 min after encoding enhanced 24-h memory in ChR2+ animals but not in ChR2− controls (Group×Condition interaction, _F_1,12=5.66, P<0.05; ChR2+ in LC-ON vs. chance: _t_7=4.38, P<0.01). Right, VTA-TH+ neuron photostimulation (VTA-ON) caused a trend for enhanced memory that did not differ between groups (Group×Condition interaction, _F_1,12=0.33, _P_=0.58; ChR2+ in VTA-ON: _t_7=2.22, _P_=0.062; ChR2− in VTA-ON: _t_5=1.55, _P_=0.18). e, Design for the optogenetic LC activation experiment with pharmacological interventions. f, Blockade of hippocampal D1/D5 receptor (SCH) but not β-adrenoceptors (Prop) during LC-TH+ neuron photostimulation abolished the effect of LC photostimulation on memory persistence in ChR2+ mice (Group effect, _F_1,12=5.01, P<0.05; Condition effect in ChR2+, _F_3,21=3.18, P<0.05; in ChR2−: _F_3,15<1). ChR2+ mice showed good memory with post-encoding LC-ON in presence of vehicle or Prop, but not in presence of SCH or without light stimulation (Orthogonal comparisons, _F_3,21=9.23, P<0.01; LC-ON with Vehicle in ChR2+: _t_7=3.01, P<0.05; LC-ON with Prop in ChR2+: _t_7=2.41, P<0.05). Prop, propranolol; SCH, SCH23390. *P<0.05, **P<0.01 vs. chance. Dashed lines=chance level. Means±s.e.m.

Figure 5

Figure 5. Optogenetic activation of LC-TH+ axons enhances hippocampal synaptic function

a, Hippocampal slice physiology. Orange line=hippocampal slice plane. b, Left, potentiation of Schaffer collateral (SC)-evoked EPSCs from CA1 pyramidal neurons after strong optogenetic activation (blue) of hippocampal LC-TH+ axons (Light-ON, _n_=5) is unaffected by adrenoceptor antagonists (Light-ON with Praz/Prop, _n_=4) but blocked by D1/D5 receptor antagonist (Light-ON with SCH, _n_=5) (Conditions×Time interaction, _F_7.8, 42.8=2.50, P<0.05, Greenhouse-Geisser correction). Middle, exemplar EPSCs from CA1 pyramidal neurons. Dashed lines=baseline EPSCs; Continuous=EPSCs 30–35 min after the optogenetic stimulation onset. Right, mean EPSCs 35 min after the optogenetic stimulation onset, showing effect of D1/D5 receptor antagonist (SCH) but no effect of adrenoceptor antagonists (Praz/Prop) (_F_2,11=6.38, _P_<0.05). **c**, Left, fEPSP responses to weak theta-burst stimulation (arrow) with or without optogenetic activation of hippocampal LC-TH+ axons (blue). No synaptic potentiation without theta-burst [Light-ON (no LTP), _n_=6], but with it, an increase in synaptic strength lasting > 45 min (Light-OFF + LTP, _n_=6) that was significantly enhanced by a weak physiologically relevant optogenetic stimulation of LC-TH+ axons (Light-ON + LTP, _n_=11). SCH blocked optogenetic enhancement of LTP (Light-ON + LTP with SCH, _n_=5) (Conditions×Time interaction, F19.6/156.7=2.93, P<0.001). Middle, fEPSPs: baseline (dashed lines) and 40–45 min after theta-burst stimulation (continuous lines). Right, mean fEPSP slopes 45 min after theta-burst stimulation, shows blockade of optogenetic augmentation of LTP by SCH (_F_3,24=16.99, P<0.001). Praz, prazosin; Prop, propranolol; SCH, SCH23390 or SCH39166 (see Methods). ns, not significant. *P<0.05, ***P<0.001, Tukey HSD test. Means±s.e.m.

Figure 6

Figure 6. Pharmacological inhibition of VTA has no impact on the novelty effect

a, Microinfusion of lidocaine (Lid) into VTA blocks multi-unit activity (example trace and population data; _n_=8 traces/4 mice) (Pre vs. Lid: _t_7=8.42, P<0.001; Pre vs. Post: _t_7=1.42, _P_>0.05). Grey shading represents±s.e.m. Magenta box, novelty period. b, Lidocaine into the VTA before novelty had no effect on memory enhancement (_n_=15 mice) (Vehicle vs. Lid: _t_14<1, _P_>0.05; Vehicle vs. chance: _t_14=2.95, P<0.05; Lid: _t_14=2.19, P<0.05). c, Systemic injection of α2-adrenergic receptor agonist clonidine before novelty abolishes the novelty effect (_F_2,28=7.70, P<0.01; Novelty with Vehicle: _t_14=4.62, P<0.001). ns, not significant. *P<0.05, ***P<0.001. Dashed lines=chance level. Means±s.e.m.

Comment in

Similar articles

Cited by

References

    1. Ebbinghaus H. Memory: A Contribution to Experimental Psychology. Dover, New York: 1885. - PMC - PubMed
    1. Wixted JT. The psychology and neuroscience of forgetting. Annu Rev Psychol. 2004;55:235–269. - PubMed
    1. Marr D. Simple memory: a theory for archicortex. Philos Trans R Soc Lond B Biol Sci. 1971;262:23–81. - PubMed
    1. Morris RGM. Elements of a neurobiological theory of hippocampal function: the role of synaptic plasticity, synaptic tagging and schemas. Eur J Neurosci. 2006;23:2829–2846. - PubMed
    1. Squire LR. Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans. Psychol Rev. 1992;99:195–231. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources