The phenomenon that we now call long-term potentiation (LTP) was first observed in 1966 [1]. Using electrophysiological methods, I was studying the perforant path input to the dentate gyrus in anaesthetised rabbits for my PhD in Per Andersen’s laboratory at the University of Oslo. Stimulating the perforant path with repetitive trains of stimuli at 10-20Hz, I, like others before me [2], saw a marked potentiation of monosynaptic granule cell responses during the stimulation. But I also saw that the potentiation could last for hours after the last stimulus train.Thus, brief stimulation of afferent axons had caused a persistent increase in the efficiency of synaptic transmission, a sign of synaptic plasticity that others had looked for in vain [3]. The observation led to a separate project with Tim Bliss when Bliss came to Oslo in 1968 as a postdoc to work with Andersen. Bliss was primarily interested in mechanisms that might underlie learning and memory and had heard from Andersen about my results. Over the next year, and in between much else, we did the first systematic study of LTP using the perforant input to the dentate gyrus.

Some properties of LTP

This study [4] revealed some basic features of LTP. First, bursts of presynaptic impulses led to a persistent increase in the efficacy of synaptic transmission. Second, the increase built up gradually with repeated bursts until saturation. Third, the potentiation was specific to tetanised cells as nearby control pathways were unaffected. Fourth, the population spike, representing number of discharging granule cells,was often larger than expected from the potentiated field EPSP, suggesting an additional increase in postsynaptic excitability, later termed EPSP-to-spike (E-S) potentiation.Other basic features were shown later,for example cooperativity and associativity, meaning, respectively, that for LTP to occur, an input must be above a certain strength or if weak, that a stronger input to the same neuron must be active at about the same time [5]. Choice of induction protocol is important.So-called theta burst stimulation (TBS), consisting of high frequency bursts of stimuli, often only four stimuli per burst repeated at theta frequency (~5Hz), is often more effective than conventional tetanic stimulations. A protocol that induces LTP in the adult animal may cause long-term depression (LTD) at an early developmental stage [6] and some protocols induce LTD rather than LTP.

The duration of LTP is critical for a role in learning and memory. Today LTP is divided into two main forms,an early form (E-LTP) lasting no more than a few hours and a late form (L-LTP) that is gene expression and protein synthesis dependent and lasts perhaps a life time [7]. In addition, there is a short-term form (STP) lasting up to about one hour.

For several years, most neuroscientists showed little interest in LTP. Interest then exploded after three important discoveries; that LTP could be induced in slices of the hippocampus maintained in vitro [8] and that it depended on activation of NMDA receptors [9] and an increase in intracellular levels of Ca2+ [10].

Mechanisms underlying LTP

At most synapses LTP is induced and expressed postsynaptically. NMDA receptors (NMDARs) are required for the induction of LTP,AMPA receptors (AMPARs),PKMζ (an isoform of PKC), and brain-derived neurotrophic factor (BDNF) for the expression of many forms of LTP, particularly L-LTP.

NMDARs are located in the postsynaptic membrane on dendritic spines. They are opened by glutamate released from active excitatory presynaptic terminals if, and only if, the postsynaptic membrane is depolarised above a threshold at about the same time. Above this threshold, the altered electric field across the postsynaptic membrane causes ejection of Mg2+ ions from their blocking position in the NMDAR channel. Ca2+ enters the cell and activates several postsynaptic signaling pathways with the end result that more AMPA receptors (AMPARs) become inserted into the postsynaptic membrane and the synapse becomes more efficient (potentiated).

AMPARs sit in the same postsynaptic membrane as the NMDARs. They are immediately opened by glutamate from the presynaptic terminal,ensuring fast transmission independently of NMDARs. Therefore, blocking NMDARs may not affect ‘normal’ fast transmission. While an open NMDAR can add to the EPSP generated by opened AMPARs, the essential function of the NMDAR is to control the efficiency of the synapse, that is its plasticity, by controlling Ca2+ influx and, consequently, the number or conductance of AMPARs in the postsynaptic membrane.

PKMζ operates in postsynaptic dendritic spines. When inhibited, L-LTP and several forms of natural memory are erased, even when the inhibitor is briefly applied days, or even weeks or months after the remembered event. Molecular mechanisms underlying this remarkable phenomenon have now been clarified in some detail [11]. Briefly, it requires transport from the nucleus to dendritic spines of a mRNA encoding constitutively active PKMζ, which remains untranslated in the spines until induction of LTP by NMDAR activation. Local translation then occurs, which together with other signaling events result in persistent PKMζ activity, insertion of AMPA receptors into the postsynaptic membrane, and potentiation of the synapse. Thus, long-term maintenance of LTP and memory requires continuously active PKMζ. Blocking PKMζ activity even briefly, interrupts local feedback circuits that are necessary for persistent PKMζ activity. The synapse reverts to a basal un-potentiated state and LTP and associated memories are erased, leaving affected pathways otherwise intact and ready for new induction of LTP and learning.

BDNF. LTP-producing activity causes release of BDNF from presynaptic terminals.BDNF then binds to TrkB receptors on dendritic spines which is necessary in many cases for persistent activation of PKMζ and L-LTP expression.

Presynaptic mechanisms. One exception to postsynaptic induction and expression is the dentate granule cell-CA3 pyramidal cell synapse, where LTP results from increased transmitter release independently of NMDAR activation [12]. Whether increased glutamate release contributes to postsynaptically expressed LTP at other synapses has been much debated. If so, a retrograde signal from post- to presynaptic structures would be required. Candidate signaling molecules exist, for example arachidonic acid and nitrous oxide,but unequivocal evidence has been difficult to obtain. Nevertheless, evidence for increased presynaptic release exists [13,14], but perhaps only for E-LTP, which is not blocked by inhibition of postsynaptic PKMζ.

Structural changes at synapses undergoing LTP. Insertion of AMPARs into the the postsynaptic membrane implies structural changes and such changes have now been demonstrated [15]. Within minutes of LTP induction some postsynaptic spines on hippocampal CA1 pyramidal cells of mature rats start growing displaying increased levels of mRNAs and polyribosomes that could assist in the translation of PKMζ-mRNA into constitutively active PKMζ.

Is LTP necessary for learning?

From recent evidence the answer very likely is yes,despite contrary arguments [16]. In one set of experiments [17] a conditioned eye-blink response was elicited by combining a tone (CS) with electrical stimulation of the supraorbital nerve (UC). In the hippocampus, one electrode was implanted to stimulate Schaffer collaterals, another electrode to record stimulus evoked monosynaptic field EPSPs of CA1 pyramidal cells, and, finally, a cannula to infuse ZIP, an inhibitor of PKMζ, into CA1. During learning (paired CS and UC stimulations) the eye-blink response to CS alone gradually increased. Remarkably, the field EPSP evoked by single constant stimuli to the Schaffer collaterals also gradually increased,as if the collaterals had been tetanised to induce LTP. Infusion of active ZIP immediately suppressed both the naturally induced LTP and the learned conditioned response, whereas an inactive scrambled ZIP did not.Stimulation-induced LTP before learning also disrupted subsequent natural learning, presumably because the synapses needed for such learning had become saturated and could not be potentiated any further. However, after erasure by ZIP, both LTP and natural learning could again be obtained.

In other experiments [18], a constant repetitive light stimulus over many days evoked a gradually increasing field response, called stimulus specific response potentiation (SRP),a natural analog of LTP, in lamina [4] of the visual cortex of mice.Furthermore, tetanic stimulation of thalamocortical fibres to the same cortical region induced LTP there. The authors then show that stimulation-induced LTP suppressed subsequent naturally-induced SRP, since the relevant synapses also in this case likely had been saturated. And again local infusion of ZIP erased both LTP and natural SRP. Interestingly, LTP potentiated the response to different types of visual stimuli,whereas SRP potentiated only the response to the visual stimulus used for training. Apparently, artificial LTP is indiscriminate (global) in that it affects synapses belonging to many different circuits, whereas natural SRP is specific in that it affects only those synapses that belong to the circuit(s) formed to serve the new responsiveness.

Figure 1: A much simplified scheme of transmission at unpotentiated (A) and potentiated (B) synapses.
A. A weak afferent input releases presynaptic glutamate, which binds to and
opens postsynaptic AMPAR-channels to evoke EPSPs and fast transmission.
Glutamate also binds to NMDARs but Mg2+ ions block their channels.

B. A strong afferent input or a weak input that coincides with a strong input elsewhere on the cell depolarises the postsynaptic membrane sufficiently to force
Mg2+ ions out and allow glutamate to open the channel. Ca2+ ions enter and
cause activation of CaMKII, PKA, and PKMζ (green) and other protein kinases.
Kinase activity potentiates AMPARdependent EPSPs by increasing AMPA
channel conductance and causing insertion of more AMPARs into the postsynaptic membrane. Such insertion requires PKMζ, translated from PKMζmRNA at local ribosomes. The process is enhanced by PKA-dependent signaling to the nucleus, up-regulation of PKMζmRNA expression, and transport of PKMζ-mRNA to the spine, where translated PKMζ maintains L-LTP by becoming autonomously and persistently active. Co-release of BDNF from presynaptic terminals ‘tags’ the synapse for potentiation by binding to TrkB receptors, a necessary event for the formation of constitutively active
PKMζ. Induction of L-LTP also induces expression of postsynaptic BDNF, which
together with PKMζ serve as plasticity related proteins (PRPs) to promote the protein synthesis and structural changes underlying L-LTP, which includes insertion of AMPARS. Such PRPs will be ‘captured’ at tagged synapses only and thereby restricted to the synapses engaged in a particular learning or memory task. Other inputs converge and release modulatory transmitters such as dopamine and acetylcholine (ACh) depending on sensory inputs related to novelty, attention, motivation, reward, or punishment. Under their influence, E-LTP induced by a weak input can be transformed into L-LTP. Not shown in this figure is the E-LTP that fails to persist because the input is either too weak or is unassisted by coincident modulatory or other inputs. Early-LTP lasts at most a few hours, is independent of protein synthesis, is not blocked by inhibition of PKMζ and therefore independent of insertion of AMPARs into the postsynaptic membrane. Instead, it appears to be primarily accounted for by increased presynaptic transmitter release.

Unlike declarative memories (memories of events, places, what has been seen or heard),procedural learning,like learning to ride a bicycle, is independent of the hippocampus.Instead,it requires the motor cortex where synapses of horizontally running fibres display LTP as rats learn new motor skills [19]. Importantly, both types of long-term memories are erased when ZIP, the inhibitor of PKMζ, is infused into either the hippocampus or the motor cortex [20].

Results such as these and the lack of convincing examples of learning without LTP in a vast literature make it seem obvious that LTP in one of its many forms is necessary for learning.

LTP in humans

Mammals are evolutionarily related, have similarly organised brains, and must be able to learn, remember, and acquire skills to survive in unpredictable and changing environments. Hence, one would expect LTP to be at least as important in humans as in other mammals. Recent work on humans confirms this expectation (see review 21. Examples are (1) LTP, similar to that observed in animal models, can be induced in slices from parts of the temporal lobe or hippocampus removed during surgery for epilepsy [22], (2) transcranial magnetic stimulation (TMS) when paired with peripheral nerve stimulation, induces increased motor cortex responsiveness indicative of underlying LTP, [23] (3) electrical stimulation of peripheral nociceptive fibres that induce LTP in ascending pain pathways in animal models also induce LTP-like responses in humans as indicated by subjective pain scores [24].

Modulators of LTP

Dopamine is one of many substances that nerve terminals release to modulate the functions of neuronal networks. Sensory inputs related to reward, punishment, or novelty can drive bursts of impulse activity in dopaminergic neurons in the brainstem and thus enhance dopamine release in target structures. In the hippocampus, infusion of dopamine receptor antagonists or genetic deletions of dopamine receptors impair not only long-term memories of locations related to reward or punishment but also the production of L-LTP. Conversely, dopamine agonists rescue both LTP and natural learning after dopamine depletion (see review 25). Hence, LTP appears necessary for this type of learning to take place.

Acetylcholine is another modulator affecting cognitive functions and LTP. Recent work shows that stimulation of cholinergic fibres from the septum facilitates LTP at hippocampal Schaffer collateral-CA1 pyramidal cell synapses by activating acetylcholine receptors on CA1 postsynaptic spines [26,27].The mechanisms are complex involving muscarinic or nicotinic receptors with effects that depend on the temporal order and time intervals of cholinergic input and Schaffer collateral activation. Interestingly, phasic ACh release in CA1 of anaesthetised rats is highly correlated with spontaneous or induced theta oscillations, suggesting that coincident ACh release and theta oscillations promote neural plasticity and thereby learning and memory [28].

Insulin-like growth factor II (IGF-II) has recently emerged as an enhancer of LTP and memory retention [29]. It is highly expressed in the hippocampus and its expression decreases with age. In rats trained to remember fear-conditioning stimuli or to avoid places where they receive foot shocks, IGF-II expression in hippocampus is markedly but transiently enhanced during the 2nd day after the training. Bilateral hippocampal injection of anti-sense oligonucleotides blocks both the enhanced IGF-II expression and the memory of the shock, whereas injection of recombinant IGF-II enhances the memory and ensures its maintenance when tested three weeks later.In addition,IGF-II transforms E-LTP induced by weak high frequency stimulation into L-LTP.

Understanding how modulators of synaptic plasticity affect LTP should help in linking subjective states, such as thinking, motivation, expectation, anxiety, fear, and addiction, to concrete brain processes, and in finding remedies when such states become abnormal.

Clinical implications of maladaptive LTP

Pain memories. LTP is likely to contribute to the chronic pain states that can develop after acute injuries to peripheral nerves or tissues. In the spinal cord, both high and low frequency stimulation of nociceptive fibres in peripheral nerves produce LTP at the synapses that these fibres form with 2nd order sensory neurons in the superficial dorsal horn.30 Experimental nerve damage or tissue inflammation evokes similar afferent impulse patterns in nociceptive fibres and have similar long-lasting effects at the same synapses. Such central changes may not only amplify primary hyperalgesia but also be responsible for secondary hyperalgesia, allodynia and spontaneous pain [31]. Spinal LTP is NMDA receptordependent, and relies on insertion of new AMPA receptors in the postsynaptic membrane. Moreover, inhibition of PKMζ in the spinal cord erases both the L-LTP and the persistent pain evoked by peripheral injury [32]. The anterior cingulate cortex (ACC) is a major cortical target for impulses in nociceptive fibres. Here also NMDA and AMPA receptor-dependent LTP occurs at glutamatergic synapses in animal models of hyperpathic pain. And again, local inhibition of PKMζ markedly reduces both pain behaviour and LTP.

Stress memories. Stress affects cognitive and emotional behaviours, as expected from the expression of stress hormones and their receptors in neurons of the hippocampus and amygdala, two regions that play important roles in such behaviours and display prominent LTPs. But the response to stress is opposite in dorsal and ventral regions. In dorsal CA1, where spatial and cognitive functions and efferent projections to the cortex predominate, stress or corticosteroids markedly suppress LTP and impair cognitive functions. In ventral CA1, on the other hand, where emotional functions and connections with the amygdala and hypothalamus
predominate, stress and corticosteroids enhance LTP and potentiate emotional responses [33]. Early-life stress is of particular interest because it affects brain development and can have adverse effects later in adult life and on offsprings. Rat pups exposed to one postnatal week of disrupted maternal care display enhanced corticotropin releasing hormone (CRH) expression in hippocampal neurons together with dendritic atrophy, impaired LTP, and poor memory performance as adults [34]. Remarkably, all these effects are counteracted by CRH receptor (CRHR) antagonists applied after the early-stress period, evidently by acting directly on CRHRs in the hippocampus. Mice with conditionally knocked out CRHR1s in glutamatergic forebrain neurons (cortex, hippocampus, amygdala) display reduced anxiety in behavioural tests and reduced LTP in the basolateral amygdala [35].

Drug memories. The mesocorticolimbic system includes projections from the ventral tegmental area (VT) through nucleus accumbens (NAc) to prefrontal or anterior cingulate cortex. It handles inputs that lead to expectations of reward, memories of rewarding inputs, and associations with environmental cues of forthcoming rewards as in Pavlovian conditioning [36]. Cues that predict rewards (positive reinforcers) induce phasic firing and LTP in glutamatergic synapses onto DA neurons in VT and NAc. Drugs of abuse, such as cocaine, similarly activate the mesocorticolimbic system resulting in AMPA receptormediated LTP. In rats, this LTP lasts up to two weeks after forced injections of cocaine but up to three months after cocaine self-administration, suggesting that voluntary intake further promotes learning and results in a drug memory that facilitates reinstatement of drug-seeking behaviour after periods of abstinence [37]. As in the hippocampus where LTP and learning require continuously active PKMζ, blockage of PKMζ in NAc by ZIP erases memories of places that rats prefer after conditioning with morphine or cocaine injections, an erasure that is prevented by blockage of AMPA receptor endocytosis [38]. Thus, Late-LTP in Nac may contribute to the formation of circuits that store memories of the reward and that can be reactivated later by appropriate cues to cause relapse even after long periods of abstinence. Ethanol, another substance of abuse with serious societal consequences, similarly affects the mesocorticolimbic system by promoting Ca2+ influx in dopamine neurons and LTP of NMDA receptor-mediated transmission at glutamatergic synapses with DA neurons in VT [39].

| Sign up to the ACNR email newsletter here |

Findings such as these are likely to facilitate the development of drugs and procedures for treating chronic pain, stress or drug abuse. By suppressing or blocking LTP in the relevant neuronal circuits one might erase the memories that lead to maladaptive behaviours. Already, local anaesthetics are used with good effect as a supplement to general anaesthesia during major surgeries to block LTP-inducing afferent impulse activity.

Alzheimer’s disease (AD). In this disease, cognitive decline is associated with impaired cholinergic function that is also involved in the production of LTP. βamyloid peptide (Aβ), a toxic break-down product in Alzheimer’s disease, strongly suppresses AChRdependent LTP.40 Accumulations of PKMζ occur in neurofibrillary tangles in limbic and medial temporal lobe structures in autopsy material from demented but not non-demented elderly individuals [41]. Given that LTP production requires active PKMζ at appropriate neuronal sites, this finding further suggests that dysfunctional LTP and memory loss are linked.

Concluding remarks

This review is based on the premise that neuronal circuits and impulse traffic along them determine an individual’s behaviour. Building appropriate circuits then becomes critical for brain development, as does changing circuits throughout life to allow learning, remembering, and acquiring skills. To build and change circuits, synapses must be plastic, and LTP is one means to ensure such plasticity. Any major neurological or psychiatric illness seems likely to reflect faulty circuits in the brain. Hence they are probably caused by or accompanied by altered synaptic plasticity. LTP production is a sign of many dysfunctional states of the brain. Development of drugs that interfere with LTP production, such as inhibitors of PKMζ, stress hormone analogs or their antagonists, and that can ameliorate abnormal behaviour in animal models, point towards helpful therapies in the future. However, the danger of overuse and side effects from targeted and nontargeted tissues, is real, as amply illustrated by drugs already in use. Evidently, the beneficial effects of many psychoactive drugs, such as anti-depressants, are mainly placebo effects [42]. Placebo effects demonstrate that positive expectations and motivations play important roles in the treatment of many brain-related disorders, effects that involve synaptic plasticity, including LTP. Perhaps the best way forward is to combine positive expectation and motivations with training, as in various forms of cognitive therapy, together with judicious use of appropriate drugs to allow new and more appropriate behaviours to be learnt and old and inappropriate ones unlearnt. Such treatment likely targets the relevant circuits better than drugs alone

References

  1. Lømo T. Frequency potentiation of excitatory synaptic activity in the dentate area of the hippocampal formation. Acta Physiol Scand 1966;68:128.
  2. Andersen P, Holmqvist B, Voorhoeve PE. Entorhinal activation of dentate granule cells. Acta Physiol Scand 1966;66:448-60.
  3. Eccles JC. The Physiology of Synapses. 1964;Berlin, Gøttingen, Heidelberg:Springer Verlag.
  4. Bliss TV, Lomo T. Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J Physiol 1973;232:331-56.
  5. McNaughton BL, Douglas RM, Goddard GV. Synaptic enhancement in fascia dentata:cooperativity among coactive afferents. Brain Res 1978;157:277-93.
  6. Ziburkus J, Dilger EK, Lo FS, Guido W. LTD and LTP at the developing retinogeniculate synapse. J Neurophysiol 2009;102:3082-90.
  7. Abraham WC, Logan B, Greenwood JM, Dragunow M. Induction and experience-dependent consolidation of stable long-term potentiation lasting months in the hippocampus. J Neurosci 2002;22:9626-34.
  8. Schwartzkroin PA, Wester K. Long-lasting facilitation of a synaptic potential following tetanization in the in vitro hippocampal slice. Brain Res 1975;89:107-19.
  9. Collingridge GL, Kehl SJ, McLennan H. Excitatory amino acids in synaptic transmission in the Schaffer collateral-commissural pathway of the rat hippocampus. J Physiol 1983;334:33-46.
  10. Lynch G, Larson J, Kelso S, Barrionuevo G, Schottler F. Intracellular injections of EGTA block induction of hippocampal long-term potentiation. Nature 1983;305:719-21.
  11. Sacktor TC. How does PKMzeta maintain long-term memory? Nat Rev Neurosci 2011;12:9-15.
  12. Weisskopf MG, Nicoll RA. Presynaptic changes during mossy fibre LTP revealed by NMDA receptormediated synaptic responses. Nature 1995;376:256-9.
  13. Errington ML, Galley PT, Bliss TV. Long-term potentiation in the dentate gyrus of the anaesthetized rat is accompanied by an increase in extracellular glutamate:real-time measurements using a novel dialysis electrode. Philos Trans R Soc Lond B Biol Sci 2003;358:675-87.
  14. Enoki R, Hu YL, Hamilton D, Fine A. Expression of long-term plasticity at individual synapses in hippocampus is graded, bidirectional, and mainly presynaptic:optical quantal analysis. Neuron 2009;62:242-53.
  15. Bourne JN, Harris KM. Coordination of size and number of excitatory and inhibitory synapses results in a balanced structural plasticity along mature hippocampal CA1 dendrites during LTP. Hippocampus 2011;21:354-73.
  16. Sanes JR, Lichtman JW. Can molecules explain longterm potentiation? Nat Neurosci 1999;2:597-604.
  17. Madronal N, Gruart A, Sacktor TC, Delgado-Garcia JM. PKMzeta inhibition reverses learning-induced increases in hippocampal synaptic strength and memory during trace eyeblink conditioning. PLoS One 2010;5:e10400.
  18. Cooke SF, Bear MF. Visual experience induces longterm potentiation in the primary visual cortex. J Neurosci 2010;30:16304-13.
  19. Rioult-Pedotti MS, Friedman D, Donoghue JP. Learning-induced LTP in neocortex. Science 2000;290:533-6.
  20. von Kraus LM, Sacktor TC, Francis JT. Erasing sensorimotor memories via PKMzeta inhibition. PLoS One 2010;5:e11125.
  21. Bliss TV, Cooke SF. Long-term potentiation and longterm depression:a clinical perspective. Clinics (Sao Paulo) 2011;66 Suppl 1:3-17.
  22. Beck H, Goussakov IV, Lie A, Helmstaedter C, Elger CE. Synaptic plasticity in the human dentate gyrus. J Neurosci 2000;20:7080-6.
  23. Stefan K, Kunesch E, Cohen LG, Benecke R, Classen J. Induction of plasticity in the human motor cortex by paired associative stimulation. Brain 2000;123 Pt 3:572-84.
  24. Klein T, Magerl W, Hopf HC, Sandkuhler J, Treede RD. Perceptual correlates of nociceptive long-term potentiation and long-term depression in humans. J Neurosci 2004;24:964-71.
  25. Lisman J, Grace AA, Duzel E. A neoHebbian framework for episodic memory;role of dopamine dependent late LTP. Trends Neurosci 2011;34:536- 47.
  26. Buchanan KA, Petrovic MM, Chamberlain SE, Marrion NV, Mellor JR. Facilitation of long-term potentiation by muscarinic M(1) receptors is mediated by inhibition of SK channels. Neuron 2010;68:948-63.
  27. Giessel AJ, Sabatini BL. M1 muscarinic receptors boost synaptic potentials and calcium influx in dendritic spines by inhibiting postsynaptic SK channels. Neuron 2010;68:936-47.
  28. Zhang H, Lin SC, Nicolelis MA. A distinctive subpopulation of medial septal slow-firing neurons promote hippocampal activation and theta oscillations. J Neurophysiol 2011;106:2749-63.
  29. Chen DY, Stern SA, Garcia-Osta A, Saunier-Rebori B, Pollonini G, Bambah-Mukku D, Blitzer RD, Alberini CM. A critical role for IGF-II in memory consolidation and enhancement. Nature 2011;469:491-7.
  30. Ruscheweyh R, Wilder-Smith O, Drdla R, Liu XG, Sandkuhler J. Long-term potentiation in spinal nociceptive pathways as a novel target for pain therapy. Mol Pain 2011;7:20.
  31. Sandkuhler J. Understanding LTP in pain pathways. Mol Pain 2007;3:9.
  32. Asiedu MN, Tillu DV, Melemedjian OK, Shy A, Sanoja R, Bodell B, Ghosh S, Porreca F, Price TJ. Spinal protein kinase M zeta underlies the maintenance mechanism of persistent nociceptive sensitization. J Neurosci 2011;31:6646-53.
  33. Maggio N, Segal M. Striking variations in corticosteroid modulation of long-term potentiation along the septotemporal axis of the hippocampus. J Neurosci 2007;27:5757-65.
  34. Ivy AS, Rex CS, Chen Y, Dube C, Maras PM, Grigoriadis DE, Gall CM, Lynch G, Baram TZ. Hippocampal dysfunction and cognitive impairments provoked by chronic early-life stress involve excessive activation of CRH receptors. J Neurosci
    2010;30:13005-15.
  35. Refojo D, Schweizer M, Kuehne C, Ehrenberg S, Thoeringer C, Vogl AM, Dedic N, Schumacher M, von Wolff G, Avrabos C, Touma C, Engblom D, Schutz G, Nave KA, Eder M, Wotjak CT, Sillaber I, Holsboer F, Wurst W, Deussing JM. Glutamatergic and dopaminergic neurons mediate anxiogenic and anxiolytic effects of CRHR1. Science 2011;333:1903-7.
  36. Bowers MS, Chen BT, Bonci A. AMPA receptor synaptic plasticity induced by psychostimulants: the past, present, and therapeutic future. Neuron 2010;67:11-24.
  37. Chen BT, Bowers MS, Martin M, Hopf FW, Guillory AM, Carelli RM, Chou JK, Bonci A. Cocaine but not natural reward self-administration nor passive cocaine infusion produces persistent LTP in the VTA. Neuron 2008;59:288-97.
  38. Li YQ, Xue YX, He YY, Li FQ, Xue LF, Xu CM, Sacktor TC, Shaham Y, Lu L. Inhibition of PKMzeta in nucleus accumbens core abolishes long-term drug reward memory. J Neurosci 2011;31:5436-46.
  39. Bernier BE, Whitaker LR, Morikawa H. Previous ethanol experience enhances synaptic plasticity of NMDA receptors in the ventral tegmental area. J Neurosci 2011;31:5205-12.
  40. Gu Z, Yakel JL. Timing-dependent septal cholinergic induction of dynamic hippocampal synaptic plasticity. Neuron 2011;71:155-65.
  41. Crary JF, Shao CY, Mirra SS, Hernandez AI, Sacktor TC. Atypical protein kinase C in neurodegenerative disease I:PKMzeta aggregates with limbic neurofibrillary tangles and AMPA receptors in Alzheimer disease. J Neuropathol Exp Neurol 2006;65:319-26.
  42. Angell M. The Epidemic of Mental Illness:Why? The New York Review of Books 2011;June 23:20-22.