Smart Drugs: Special Topics in The Neurobiology of Learning/Memory I

Smart Drugs & LTP: Enhancement of Long-term Potentiation Through Actions on AMPA-Receptor Initiation and CREB Consolidation

Unlike the previous installment, this post presumes a great deal of background knowledge on the part of its readers. My intended audience has taken a class in general biology, neuroscience, or physiology at the college level and are familiar with terms like synapse, action potential (depolarization, ion channel, electrochemical gradient), neurotransmission (axons, dendrites, receptors, ligands, inhibition (IPSP), excitation (EPSP)), neurotransmitters (especially glutamate, GABA, acetylcholine, dopamine, serotonin, epinephrine), and gene transcription/translation. Anyone who's not conversant in biochemistry may want to take a minute and bone up on this stuff, perhaps by reading my first post or by just skimming the links above.

Drug-mediated cognitive enhancement has become a topic of great interest among researchers and laypeople alike, but still precious little is known about the neurobiology underlying our cognitive processes. Over the past two million years—a paltry interval in evolutionary time—growth of the human brain has wildly outpaced that of our closest relatives. But it is also true that these expanded cortical areas are undergirded by neural circuitry that we share with our primate, and indeed our reptilian, ancestors. Since complex phenotypes never arise de novo, it is unlikely that human “brain plans” are in any way optimized for cognition; rather, this growth seems to be in line with an inherited simian blueprint.¹ All this is to suggest that presently, our brains fall far short of maximizing these recent specializations that confer human uniqueness. If this assumption is a safe one, then it follows that the current state of our cognition leaves much room for improvement.

The foregoing discussion presents an unsettling scenario fraught with both scientific and ethical dilemmata. Still, our understanding of learning and memory can be greatly enriched by a consideration of these performance enhancing drugs and their effects, both in the medically compromised and in the neurotypical. However, this post does not address the wider world of nootropic substances; my focus is limited to the effects compounds known to faciliate LTP. As long-term potentiation (LTP) is considered to be the major cellular mechanism underlying learning and memory, I seek to examine how this process can be exogenously enhanced by pharmacological tinkering at the two crucial phases of the process: the initiation of memory formation during early long-term potentiation and memory consolidation during late long-term potentiation.

LTP OVERVIEW

Before considering the affects of a given substance on long-term potentiation, it would be prudent to give both an overview of the neuromolecular correlates of these phenomena and due homage to those who discovered them. Donald O. Hebb was a Canadian psychologist whose pioneering efforts in the field earned him the moniker “father of neural networks.”² He postulated that the efficacy of synaptic communication was dependent on activity at the synapse; that strong, repeated activation of these connections can result in lasting structural and functional changes. In his words, “when an axon of cell A is near enough to excite cell B and repeatedly or persistently takes part in firing it, some growth process or metabolic change takes place in one or both cells such that A's efficiency, as one of the cells firing B, is increased.”² His idea that modifications in neural circuits were the mechanisms by which information is stored and retrieved in the brain was borne out in the research of Terje Lømo, who in 1966 discovered in the rabbit hippocampus what would come to be known as long-term potentiation.³ Studies exploring the nature of this effect, along with work done by Eric Kandel on habituation and sensitization of neural circuits in Aplysia californica,⁴ served as a springboard for recent discoveries in the molecular basis of memory and gave the model of synaptic plasticity its modern form.

Long-term potentiation is the name given to the discovery that a brief trains of high-frequency stimuli to monosynaptic excitatory pathways in the hippocampus cause a sustained increase in the efficiency of synaptic transmission.⁵ This effect persists for hours in ex vivo hippocampal slices and for weeks in the hippocampus of living mammals. LTP is commonly divided into two phases—early LTP and late LTP—each consisting of three processes: induction, maintenance, and expression.⁶ In such a monosynaptic excitatory pathway, a stimulus causes presynaptic release of glutamate onto the postsynaptic cell membrane where it binds to α-amino-3-hydroxy-5-methyl-4-isoxazolepropionicacid (AMPA) receptors. This triggers the influx of sodium cations into the postsynaptic cell, effecting an excitatory postsynaptic potential through depolarization. High frequency stimulation of this sort results in EPSP summation and greater depolarization of the postsynaptic cell. If sufficient, this depolarization causes the ejection of Mg2+ from the ion channel of N-methyl-D-aspartate (NMDA) receptors, effectively unblocking them. Now, if these receptors are also bound by glutamate released from the presynaptic cell, they become active and allow Ca2+ to flow into the postsynaptic cell. _ This rapid rise in intracellular calcium concentration is the pivotal step for subsequent signaling cascades_; it activates several enzymes that mediate early LTP induction such as Ca2+/calmodulin-dependent protein kinases II(CaMKII), protein kinase C (PKC), and to a lesser extent proteinkinase A (PKA) and Mitogen-activated protein kinase K (MAPK).⁶ All kinases are enzymes that affect the state of a molecule (changing its activity, reactivity, or binding ability) by simply sticking a phosphate group on it; they serve many important functions in the regulation of complex cellular processes. Through persistent activation of these kinases (particularly CaMKII and PKC), existing AMPA receptors are phosphorylated (activated) and additional AMPA receptors are inserted into the postsynaptic membrane, both of which increase postsynaptic response to released glutamate so future excitatory stimuli generate larger postsynaptic potentials. In addition, CaMKII may lead to the synthesis of a retrograde messenger that acts to increase presynaptic neurotransmitter vesicle number, probability of vesicle release, or both.⁶ The process described above accounts for the few hours of LTP observed in ex vivo hippocampal tissue. Functional modification of the circuit must rely on changes in protein synthesis or alterations in the rate of synthesis and degradation of proteins already present.

The late phase of LTP is induced by changes gene transcription and protein synthesis brought about by the persistent activation of protein kinases activated during early LTP, such as MAPK.¹⁶ This process is necessary for memory formation; it has been show in many studies that  _inhibition of protein synthesis disrupts late LTP ⁷ PKA activation and calcium influx converge on CRTC1, a potent transcription factor for cAMP response element binding protein(CREB). Through phosphorylation-dependent activation this molecule affects the transcription of many genes, including genes encoding other transcription factors⁸ and genes involved in synaptic plasticity.⁹ CREB-mediated transcriptional activity is important in habituation and sensitization as well.¹⁰ and is a good candidate for mediating the molecular switch to long-term memory. Research continues to suggest that CREB plays an important role in memory formation and retrieval.

This brief description of LTP belies its complexity and diversity; in truth, much about it remains to be discovered and the list of potential modulators (molecules that can alter LTP but are not essential for it) is ever-growing. For instance, beta-adrenergicreceptor agonists, nitric oxide synthase, and estradiol have all been proposed to have an effect on LTP.⁶ What follows will be limited to a consideration of the known effects of certain drugs on AMPA receptors and CREB-mediated transcriptional activity in NMDA receptor-dependent hippocampal LTP.

SUBSTANCES THAT AFFECT LTP

To achieve an enhancement of memory through a direct effect on LTP, a drug can act at either the early phase or the late phase of the process described above. In the case of the early phase, the ionotropic glutamatergic receptors are obvious targets for these drugs. A class of pyrrolidine-derived compounds known as racetamsbind to modulator sites on the AMPA receptor, including the cyclothiazide site, and have been found to have a positive effect on memory.¹¹ Piracetam is the most well-known racetam and was the first of this class of molecules to be discovered.¹² In cultured neurons, it enhances the Ca2+ influx produced by the AMPA receptor but not that produced by the NMDA receptor.¹³. In electrophysiological studies it increases the peak amplitude of the ion current generated through AMPA receptors, reduces it rate of decay, and increases the maximal density of low-affinity binding sites for AMPA in the postsynaptic membrane.¹³ It also increases muscarinic cholinergic receptor density in the frontal cortex of mice and has the general effect of activating the cholinergic system.¹⁴ Given the dual action of the these compounds, choline and piracetam administered together have been shown to substantially improve memory in dementia patients.¹⁵ Indeed, a reduction in the activity of cholinergic neurons is a well-known feature of Alzheimer's disease, and several modern treatments for mild to moderate Alzheimer's are based on increasing the concentration of acetylcholine in the brain. Piracetam alone has been shown to effectively overcome amnesia induced by scopolamine, diazepam, and electroconvulsive shock through actions on the hippocampus.¹⁶ Aniracetam, a second member of the racetam family, has been known to improve cognitive functions impaired in rodents by experimental procedures since the early 1980s.¹⁷ It facilitates LTP in the same way as piracetam, slowing entry of AMPA receptors into a desensitized state and increasing excitatory synaptic strength,¹⁸ but it also seems to enhance cortical GABA-mediated inhibition.¹⁹ Aniracetam facilitates LTP in hippocampal tissue,²⁰ has improved performance of rhesus monkeys in delayed match-to-sampletasks,²¹ and reverses learning impairment in rodents.²²

Ampakines, close relatives of the racetam family, were the first allosteric modulators of AMPA receptors found to be able to augment excitatory transmission in the brain.²⁰²³ They consist of thiazide derivatives and unlike their parent compound they are able to cross the blood-brain barrier to bind to the cyclothiazide binding site of the AMPA receptor, slowing receptor desensitization and deactivation. Since racetams and ampakines are allosteric modulators, they affect only AMPA receptors activated by endogenous transmitter and thereby restrict their influence to regions that are engaged in brain activity.²⁴ Ampakines have also been found to induce the expression of neurotrophin genes, such as growth factors like BDNF.²⁰ They have also been found to improve delayed recall in aged individuals and to facilitate memory encoding generally. ²⁵[26] In all, racetams and ampakines act similarly to slow deactivation and attenuate desensitization of AMPA receptor currents, increase synaptic responses, and enhance long-term potentiation.

In the late phase of LTP, the transcription factor CREB has been shown to be crucial to memory consolidation—its loss of function results in an impairment of long-term memory, while increases in CREB activity enhance long-term memory; importantly CREB activity does not seem to affect short-term memory.²⁷ CREB-dependent gene expression is mainly regulated through phosphorylation and through chromatin remodeling.²⁷[28] Inhibition of phosphodiesterase (PDE) activity leads to increases in cAMP or cGMP levels, which drives CREB phosphorylation and activation through increases in protein kinase activity . Whether cAMP or cGMP is increased depends on the phosphodiesterase—PDE4 inhibition increases cAMP levels while PDE5 inhibition increases cGMP levels.²⁹ The prototypical PDE4 inhibitor is rolipram, which leads to CREB phosphorylation and CREB-dependent gene transcription though activation of PKA.³⁰ In animal models, rolipram has been shown to facilitate memory formation by increasing CREB phosphorylation.²⁹

Epigenetic chromatin remodeling and modifications of DNA represent central mechanisms for regulation of gene expression during memory formation. In order for gene expression to take place, chromatin must be unpacked to expose DNA regulatory sequences to transcription factors such as CREB.³¹ A primary mechanism for attaining the chromatin state required for transcriptional activity is histone acetylation, which depends on the relative activities of enzymes histone-acetyl transferase (HAT) and histone deacetylase (HDAC). To promote long-term memory related gene expression, CREB requires a coactivator called CREB binding protein (CBP). This coactivator possesses histone acetyl-transferase activity required for transcription.³² CBP histone acetyl-transferase activity is an important component in memory consolidation; truncated CBP protein in transgenic mice significantly reduced late-phase LTP in hippocampal slices. These mice also exhibited behavioral deficits in two hippocampus-dependent tasks: spatial learning in the Morris watermaze and long-term memory for contextual fear conditioning. Corroborating these findings, it has been shown that _elevated levels of histone acetylation through the use of HDAC inhibitors such as sodium butyrate enhances induction of long-term potentiation at Schaffer-collateral synapses in the hippocampus in vitro as well as long-term memory formation in a contextual fear conditioning paradigm³³.

An examination of two well-characterized steps in the process of long-term potentiation has recommended three key processes through which its enhancement could be mediated: AMPA receptor modulation, phosphodiesterase inhibition, and histone deacetylase inhibition. The compounds promoting these processes have the potential to be therapeutically valuable in cases of cognitive impairment. This is illustrative of just how labile the processes underlying our behavioral memory actually are.