Health Articles

 

Hkust institutional repository

International Journal of Neuropsychopharmacology (2011), 14, 1247–1256. f CINP 2011 From understanding synaptic plasticity to thedevelopment of cognitive enhancers Zelda H. Cheung and Nancy Y. Ip Division of Life Science, State Key Laboratory of Molecular Neuroscience and Molecular Neuroscience Center,Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China Accumulating evidence reveals that synaptic dysfunction precedes neuronal loss in neurodegenerativediseases such as Alzheimer's disease. Intriguingly, synaptic abnormality is also implicated in a myriad ofpsychiatric disorders including depression. In particular, alterations in spine density and morphologyhave been associated with aberrant synaptic activity in these diseased brains. Understanding the mol-ecular mechanisms underlying the regulation of spine morphogenesis, synaptic function and plasticityunder physiological and pathological conditions will therefore provide critical insights for the develop-ment of potential therapeutic agents against these diseases. Here we summarize existing knowledge onsome of the molecular players in synaptic plasticity, and highlight how these findings from basic neuro-scientific research aid in the identification of novel drug leads for the development of therapeutics.
Received 26 August 2010 ; Reviewed 1 October 2010 ; Revised 5 November 2010 ; Accepted 19 November 2010 ;First published online 6 January 2011 Key words : Alzheimer's disease, BDNF, dendritic spines, depression, synaptic plasticity.
elucidating the pathophysiological mechanisms ofneurodegenerative diseases have focused on identify- Since the first depiction of neurons by Ramon y Cajal ing the signalling pathways that mediate the selective more than a century ago, remarkable progress has degeneration of the susceptible neuronal populations been made to unveil the mystery behind the physi- in these diseases. Interestingly, mounting evidence in- ology, functioning and communication by these un- dicates that loss of synapses precedes actual neuronal ique cells. With the commencement of the molecular death in Alzheimer's disease (AD) (Selkoe, 2002). In and genetic era, knowledge on the molecular pathways addition, advances in unravelling the molecular me- implicated in the control of neuronal survival, synaptic chanisms of depression point unexpectedly to deregu- transmission and synaptic plasticity have exploded.
lation in synaptic plasticity (Krishnan & Nestler, 2008 ; As much as these findings reveal how the neuronal Pittenger & Duman, 2008). It therefore appears that circuitry mediates daily physiological function of dysfunction in synaptic transmission and plasticity the nervous system, they also provide essential infor- may present a common pathogenic mechanism across mation for deciphering the molecular pathophysiology a broad spectrum of neurological disorders. This reali- of various neurological disorders. For example, zation underscores the significance of in-depth in- aberrant synaptic transmission has been implicated vestigations into the molecular control of synapse in psychiatric disorders such as schizophrenia and de- function and plasticity, as these findings are likely to pression, whereby the level of neurotransmitters is generate critical new insights into the future develop- abnormally elevated or reduced, affecting neuro- ment of therapeutics that may be applicable to multiple transmission. On the other hand, studies aimed at disorders harbouring synaptic failures. In this brief re-view, we summarize the current understanding of themolecular mechanisms underlying the control of syn- Address for Correspondence : Professor N. Y. Ip, Division of Life aptic function and plasticity in AD and depression, as Science, Hong Kong University of Science and Technology, examples of neurodegenerative disease and psychi- Clear Water Bay, Hong Kong, China.
atric disorder. How this knowledge may be utilized to Tel. : 852-2358-7304 Fax : 852-2358-2765Email : [email protected] identify novel drug targets will also be discussed.
Z. H. Cheung and N. Y. Ip Aberrant synaptic function and plasticity as a harbouring various mutations in APP, which lead to common pathophysiological mechanism in AD and elevated Ab generation, exhibit impaired synaptic transmission and LTP (Parsons et al. 2007 ; Selkoe,2002 ; Wasling et al. 2009). Interestingly, loss of post- Aberrant synaptic function synaptic marker PSD-95 around Ab plaques in an AD is the leading cause of dementia worldwide and APP transgenic mouse is paralleled by reduction in accounts for more than 50 % of dementia cases. Post- pre-synaptic boutons, suggesting that synapse loss mortem brains of AD patients are characterized by the involves both pre-synaptic and post-synaptic elements presence of extracellular aggregates composed pre- (Spires et al. 2005). Ab treatment in vitro and in vivo dominantly of b-amyloid (also known as Ab) and intracellular neurofibrillary tangles of hyperphos- mediated signals and LTP (Hsieh et al. 2006 ; Selkoe, phorylated tau. Ab generation has been postulated as 2002 ; Wasling et al. 2009), and reduces surface ex- the main culprit in the aetiopathology of AD, as sup- pression of NMDAR subunit NR1 by promoting its ported by its abundant presence in senile plaques and internalization (Kurup et al. 2010 ; Snyder et al. 2005).
the observation that essentially all missense mutations This is accompanied by diminished NMDAR current identified in familial cases of AD result in elevated and the downstream activation of transcription factor production of Ab (Hardy & Selkoe, 2002). Ab is gen- CREB (Kurup et al. 2010 ; Snyder et al. 2005). In ad- erated from sequential cleavage of a transmembrane dition, AMPAR removal was found to underlie protein called amyloid precursor protein (APP) by Ab-stimulated synaptic depression and spine loss b-secretase and c-secretase (Wasling et al. 2009). Since (Hsieh et al. 2006). Taken together, these observations AD is a neurodegenerative disease in nature, earlier indicate that Ab can directly impair synaptic functions efforts have been directed at understanding the mol- and plasticity in AD brains by interfering with gluta- ecular mechanisms by which Ab generation is regu- matergic synapses.
lated, and how Ab deposits lead to death of the Another neurological disorder characterized by ab- neurons. Nonetheless, subsequent evidence indicates errant synaptic function is depression. Also known as that decrease in synapse density precedes actual loss major depressive disorder, depression is a psychiatric of neurons in AD brains. More importantly, synapse disorder characterized by low mood, irritability, an- loss, rather than neuron loss, serves as a more accu- hedonia, difficulty in concentrating and cognitive im- rate correlate with cognitive decline in AD patients pairment. While the cause is essentially unknown, (Selkoe, 2002). An increasing number of studies have earlier studies aimed at elucidating the mechanisms of therefore focused on elucidating the effect of Ab on action of antidepressants suggest that reduced trans- synaptic functions.
mission of two monoamine neurotransmitters in the Several lines of evidence indicate that Ab can di- brain, namely 5-HT (serotonin) and noradrenaline, rectly impair synaptic transmission in AD. Glutamate is the major excitatory neurotransmitter in the brain.
(Krishnan & Nestler, 2008 ; Pittenger & Duman, 2008).
There are two subtypes of ionotropic glutamate re- Interestingly, accumulating evidence suggests that ceptors, namely the AMPA receptors (AMPARs) and impaired synaptic plasticity may also play a role in the NMDA receptors (NMDARs). While AMPARs are depression. Indeed, depressed patients exhibit diffi- directly activated by ligand binding and mediate fast culty in declarative memory (Zakzanis et al. 1998).
excitatory transmission at the synapse, NMDAR acti- Stress, which is a known precipitating and aggravat- vation requires concurrent ligand binding and de- ing factor of depression, has been demonstrated to polarization of the post-synaptic neuron. This unique disrupt hippocampal LTP in experimental animals, in property of NMDARs renders them particularly im- addition to inducing atrophy of hippocampus, a situ- portant in synaptic plasticity. NMDARs function as ation also observed in depressed patients (Calabrese ‘coincidence detectors ', allowing the selective, long- et al. 2009 ; Chen et al. 2010 ; Pittenger & Duman, 2008).
term strengthening of synaptic transmission following Expression of brain-derived neurotrophic factor high-frequency stimulation of the synapse in the (BDNF), a neurotrophic factor that is critical for syn- cellular paradigm of long-term potentiation (LTP).
aptic plasticity, was reduced in the hippocampus of Indeed, activation of NMDARs was demonstrated to post-mortem brains of depressed patients (Karege be essential for the generation of synaptic plasticity in et al. 2005). In addition, antidepressant treatment the brain (Lau & Zukin, 2007). Interestingly, Ab has enhances the expression of BDNF and transcription been observed to selectively affect glutamatergic factor CREB, which have been demonstrated as synapses (Wasling et al. 2009). Transgenic mice pivotal players in long-term synaptic plasticity Developing cognitive enhancers from basic research (Chen et al. 2001 ; Thome et al. 2000). These observa- that post-mortem brains of human patients with tions collectively suggest that alterations in synaptic severe stress and longitudinal depression exhibit transmission and plasticity may contribute to the reduced dendritic spine densities (Soetanto et al. 2010).
pathophysiology of depression.
In agreement with these findings, treatment withantidepressants fluoxetine or imipramine enhancespine density in rats (Ampuero et al. 2010 ; Chen et al.
Spine morphology anomaly 2008a ; Hajszan et al. 2005), implicating structural The majority of the excitatory synapses are located on changes in dendritic spines and synaptic abnormality tiny protrusions on dendrites known as spines. Spines in depressed patients.
are highly dynamic structures, with their size, shapeand density along dendrites under constant and attimes, rapid, regulation. While the more elongated Molecular players in the regulation of synaptic protrusions known as filopodia are regarded as im- functions and spine morphogenesis mature spines and are highly plastic ; stubby, mush- A plethora of signalling pathways has been implicated room-shaped spines are generally mature spines that in the control of synaptic functions. Nonetheless, re- are more stable in nature (Tada & Sheng, 2006).
cent studies have highlighted several molecular play- Interestingly, spine morphogenesis is regulated by ers that show promise in allowing identification of synaptic activity, and has been postulated as the novel targets for developing drugs that can reverse structural basis of synaptic plasticity. LTP, for ex- ample, is associated with an enlargement of spinehead and insertion of AMPARs at synaptic sites ; while Glutamate receptors long-term depression (LTD) involves shrinkage ofspines (Dillon & Goda, 2005 ; Lau & Zukin, 2007 ; Tada Being the receptors for the major excitatory synapses & Sheng, 2006 ; Zhou et al. 2004). Changes in spine in the brain, it is no surprise that glutamate receptors morphology require the coordinated regulation of ac- are pivotal for the control of synaptic strength. The tin cytoskeleton and also gene transcription (Ethell & type and number of glutamate receptors present at a Pasquale, 2005). Given the essential role of spine synapse are under tight dynamic regulation and serve morphogenesis in synaptic plasticity, understanding as one of the most direct determining factors of syn- how spine morphology is altered in various neuro- aptic strength. Rapid changes in the number of gluta- logical diseases will provide essential background mate receptors are mediated by insertion or removal knowledge on how synaptic dysfunction may be re- of surface receptors at the post-synaptic densities.
Lateral movement of both AMPA and NMDA re- In addition to directly impairing synaptic trans- ceptors also contributes to the rapid changes in gluta- mission and plasticity by interfering with glutamater- matergic transmission at the synapse (Hanley, 2008 ; gic synapses, Ab has also been demonstrated to affect Lau & Zukin, 2007). Furthermore, AMPA and NMDA spine morphogenesis. Ab treatment has been observed receptors play distinct roles in the induction of LTP.
to reduce dendritic spine density in cultured neurons NMDAR activation during the concurrent presence of and also in transgenic animal models of AD (Knobloch synaptic glutamate and post-synaptic depolarization & Mansuy, 2008 ; Lacor et al. 2007 ; Shankar et al. 2007 ; results in calcium influx through the NMDAR and Spires et al. 2005). In particular, a recent study dem- activation of secondary signalling messengers such as onstrated that Ab produced from axons or dendrites Ca2+/calmodulin-dependent kinase 2 (CaMKII). The lowers spine number and plasticity on nearby den- subsequent AMPAR insertion mediates the increase in drites (Wei et al. 2010). These observations suggest that synaptic strength during LTP induction (Lau & Zukin, loss of dendritic spines may also contribute to the 2007 ; Monti & Contestabile, 2009). The signalling cas- cognitive deficits in AD patients.
cade initiated downstream of NMDAR activation also Changes in spine morphology in depression are not triggers gene transcription, which has been demon- as well documented but evidence in support of this strated to be critical for the late phase of LTP (Kelleher possibility is beginning to emerge. Chronic stress was et al. 2004). Aside from being indispensible for gluta- observed to reduce dendritic complexity (Bloss et al.
matergic transmission and synaptic plasticity, recent 2010 ; Hains et al. 2009 ; Liston et al. 2006), in addition to evidence suggests that AMPARs may also directly decreasing spine density in rats (Chen et al. 2008b, regulate spine morphogenesis. The extracellular 2010 ; Hains et al. 2009 ; Radley et al. 2006 ; Silva-Gomez N-terminal domain of the AMPAR subunit GluR2 was et al. 2003). In addition, a recent study demonstrated found to enhance spine growth (Passafaro et al. 2003 ; Z. H. Cheung and N. Y. Ip Saglietti et al. 2007). Collectively these observations early phase of LTP (Waterhouse & Xu, 2009). TrkB underscore the essential role of glutamate receptors in activation at the synapse also enhances local protein the control of synaptic function and plasticity.
synthesis, an action that is required for the late phase Interestingly, as much as glutamate transmission is of LTP (Bramham, 2008). Indeed, BDNF stimulation required for normal function of the neural circuitry, and depolarization increases trafficking of TrkB excessive activation of glutamate receptors during and BDNF mRNA to the dendrite (Righi et al.
pathological condition can also impair synaptic func- 2000 ; Tongiorgi et al. 1997), further supporting an tion and lead to excitotoxic neuronal loss. Energy fail- involvement of local protein synthesis in synaptic ure or impairment of glutamate transporters in AD plasticity. On the other hand, proBDNF was found brains may contribute to an abnormally high level to facilitate LTD induction (Woo et al. 2005). Interest- of extracellular glutamate, leading to excitotoxicity ingly, cleavage of proBDNF to mature BDNF is el- (Lipton, 2006 ; Parsons et al. 2007). In support of a role evated by high-frequency stimulation, one of the of excitotoxicity in AD, memantine, an uncompetitive stimulation paradigms used to induce LTP in hippo- NMDAR antagonist approved by the FDA for the campal slice preparation (Nagappan et al. 2009). These treatment of AD, was found to reduce neuronal loss observations indicate that while BDNF signalling is and also LTP impairment induced by Ab treatment or crucial for synaptic function and plasticity, its role is in animal models of AD (Lipton, 2006 ; Parsons et al.
complex and can be controlled at multiple levels.
2007 ; Rammes et al. 2008). The mechanism by which Aside from being critical for LTP induction, BDNF an NMDAR antagonist may reverse LTP impairment has also been demonstrated to directly regulate spine has not been completely elucidated, but it appears to morphogenesis and has also been demonstrated to involve reduction of tonic activation of glutamatergic increase spine density (Amaral & Pozzo-Miller, 2007 ; synapses (Lipton, 2006 ; Parsons et al. 2007 ; Rammes Ji et al. 2005). Induction of spine head enlargement et al. 2008). These findings reveal that although by pairing of post-synaptic spike with glutamate un- NMDARs may be a very attractive target for devel- caging was found to require BDNF and local protein opment of therapeutics, identifying compounds that synthesis (Tanaka et al. 2008). Furthermore, when inhibit pathological but not physiological glutamater- dendritic targeting of BDNF mRNA is abolished by gic transmission will be essential for ensuring the expression of a truncated form of the long 3k-UTR feasibility and applicability of these potential drugs.
BDNF mRNA, impairment in spine head enlargementand spine pruning are observed (An et al. 2008). Thesefindings demonstrated that BDNF also plays a role in BDNF/TrkB signalling spine morphogenesis, in part through regulation of BDNF is a member of the neurotrophin family that local protein synthesis (Fig. 1).
has been increasingly implicated in the regulation of Extensive studies have demonstrated suppression synaptic function and plasticity. Action of BDNF is of BDNF signalling in AD and depression. Both ex- mediated predominantly by receptor tyrosine kinase pression of BDNF and TrkB are reduced in AD brains TrkB, although it also binds with low affinity to p75.
(Schindowski et al. 2008). Interestingly, it was recently BDNF is synthesized as proBDNF, which has long demonstrated that BDNF deprivation enhances Ab been regarded simply as the unprocessed form of generation in hippocampal neurons (Matrone et al.
mature BDNF. Nonetheless, recent evidence reveals 2008). In agreement with this observation, BDNF was that proBDNF is also secreted, and exhibits high- found to reduce Ab production through activation of affinity binding to p75 (Lu et al. 2005). While the Sorting protein-related receptor with A-type repeats neurotrophins were initially identified based on their (SORLA), whose expression is reduced in sporadic AD ability to maintain neuronal survival, BDNF was later (Rohe et al. 2009). These observations reveal that BDNF identified as being particularly critical for synaptic signalling may be important for reducing Ab gener- plasticity. Mice lacking BDNF exhibit impaired LTP ation in AD brains. On the other hand, while an earlier induction (Korte et al. 1995 ; Patterson et al. 1996).
observation of reduced BDNF level in the hippocam- In addition, BDNF was found to be secreted at the pus of depressed patients has triggered considerable synapse in an activity-dependent manner. This local interest in the hypothesis that BDNF deficiency may elevation in BDNF results in TrkB activation at the underlie the pathophysiology of depression, recent synapses, and the subsequent initiation of down- data are more controversial. For example, reduction of stream signalling cascade modulates synaptic proteins BDNF in the hippocampus is accompanied by elev- to regulate the efficiency of synaptic transmission.
ated BDNF level in the nucleus accumbens (Krishnan These modulations were found to be critical for the & Nestler, 2008). In addition, while a reduction in

Developing cognitive enhancers from basic research Local protein CREB-dependent Spine size and number No post-synaptic electrical activity Post-synaptic electrical activity Fig. 1. BDNF/TrkB signalling as a key regulator of synaptic plasticity. Synaptic activity increases the local release of BDNFat the synapse, which leads to dendritic targeting of BDNF and TrkB mRNA. Activation of TrkB then enhances expression ofsynaptic proteins through CREB-dependent transcription and local protein translation, resulting in enhanced membraneinsertion of AMPARs at synapses. The local protein synthesis by BDNF also promotes actin polymerization, contributing tostructural changes of dendritic spines and LTP formation. On the contrary, in the absence of post-synaptic activity, localBDNF/TrkB signalling and spine growth are both limited.
BDNF level fails to induce depression, BDNF is re- of NR2B by calpain. This is accompanied by improved quired for the efficacy of antidepressants (Calabrese spatial learning in Cdk5 conditional knockout mice et al. 2009 ; Krishnan & Nestler, 2008). Taken together (Hawasli et al. 2007). These observations implicate these observations reveal that while BDNF is probably Cdk5 in the regulation of glutamate transmission and involved in the pathophysiology of depression, its ac- synaptic plasticity. Furthermore, Cdk5 has also been tion is likely region-specific and additional studies will demonstrated to regulate spine morphogenesis. Cdk5 be required to delineate its precise involvement.
was found to be required for ephrinA1-induced spineretraction through regulation of the downstream acti-vation of RhoA (Fu et al. 2007). In addition, Cdk5 Cyclin-dependent kinase 5 (Cdk5) phosphorylates WAVE1 to inhibit actin polymeriz- Cdk5 is a predominantly neural-specific serine/thre- ation. This is associated with reduction of stubby- onine kinase that has recently been implicated in the shaped spines (Kim et al. 2006). These observations regulation of synaptic function and plasticity (Cheung reveal that Cdk5 is pivotal for the regulation of spine et al. 2006 ; Lai & Ip, 2009). Cdk5 is activated upon binding to its activator p35 or p39. Earlier studies Cdk5 has long been implicated in the pathophy- demonstrated that Cdk5 and its activators are ex- siology of AD. In particular, cleavage of p35 into a p25 pressed at the synapse (Cheung et al. 2006). Evidence fragment, which results in prolonged activation of in support of a role of Cdk5 in the regulation of syn- Cdk5, is associated with neuronal loss in various aptic transmission came from the observation that models of neurodegenerative diseases (Cheung et al.
Cdk5 was found to directly phosphorylate NMDAR 2006). Indeed, Cdk5 was initially identified as a tau subunit NR2A, and inhibition of this phosphorylation kinase (Kobayashi et al. 1993). In addition, neuronal reduces NMDA-evoked current (Li et al. 2001). In ad- loss and impairment in spatial learning are evident dition, a recent study revealed that inhibition of Cdk5 in a mouse model with prolonged expression of p25 activity reduces activity-dependent internalization of (Fischer et al. 2005). Interestingly, expression of NMDARs by modulating phosphorylation of the b-secretase BACE1 is also elevated in this strain of NMDAR subunit NR2B by Src (Zhang et al. 2008).
mouse, leading to elevated production of Ab (Wen Interestingly, Cdk5 was found to mediate degradation et al. 2008). These studies collectively indicate that Z. H. Cheung and N. Y. Ip aberrant activation of Cdk5 may contribute to thepathophysiology of AD.
Explore disease mechanisms Impaired BDNF/TrkB signalling A role of Cdk5 in depression is not as solidly dem- Cdk5 activity deregulation onstrated. Nonetheless, Cdk5 has been observed to Aberrant NMDA receptors activation regulate dopamine signalling, the dysfunction ofwhich has been associated with the pathophysiologyof depression (Rakofsky et al. 2009). Cdk5 was dem- Identify disease-related proteins – molecular targets onstrated to phosphorylate DARPP-32, an importantplayer in dopamine signalling. Cdk5-mediated phos- phorylation of DARPP-32 reduces dopamine-inducedactivation of PKA, thereby attenuating dopaminergicsignalling (Benavides & Bibb, 2004). These observa-tions suggest that Cdk5 may, through its modulation Discovery of Drug Leads of dopaminergic signalling, contribute to monoamine imbalance in depressed patients. Further studies will be required to address the precise role of Cdk5 indepression.
Identification of novel drug leads that target Fig. 2. From basic research to drug discovery. Basic research regulators of synaptic plasticity focused on understanding the pathophysiologicalmechanism of neurological diseases will enable identification Basic research focused on understanding the patho- of molecular players that may serve as drug targets.
physiology of various diseases is crucial for the de- Subsequent screening of compounds that selectively target velopment of therapeutic interventions. Explicating these molecules of interest will provide important basis for the signalling pathways that are affected in the dis- the development of novel therapeutics.
order will enable the identification of proteins thatmay be targeted to ameliorate symptoms or delay trials have been performed with various NMDAR an- disease progression. Through the uncovering of novel tagonists but since physiological level of glutamate molecular players, drug leads that directly target transmission is pivotal for the normal function of the these molecules can then be developed and tested brain, many of the antagonists tested resulted in major as potential therapeutics against the disease (Fig. 2).
side-effects and the studies were discontinued.
This approach has served as one of the most important Memantine, currently the only NMDAR antagonist strategies for effective design and development approved for the treatment of AD, is unique as a non- of drugs and treatment against various disorders.
competitive, relatively low-affinity, open-channel an- Indeed, advances in understanding the molecular tagonist that exhibits strong voltage-dependence and mechanisms implicated in the pathophysiology of fast off-rate (Lipton, 2006 ; Monti & Contestabile, 2009 ; neurological disorders such as AD and depression Rammes et al. 2008). This special property of meman- have led to the development of the existing treatments tine allows it to preferentially bind to NMDARs that for these diseases. For example, the observed re- are open for a prolonged period, without interfering duction in cholinergic transmission in post-mortem with the normal physiological function of the receptor AD brains prompted the development of AChE (Lipton, 2006 ; Monti & Contestabile, 2009 ; Rammes inhibitors as therapeutic agents for AD (Monti & et al. 2008). These features of memantine explain its Contestabile, 2009). However, recent evidence has re- clinical tolerance and suitability for treatment of ex- vealed that inhibition of nicotinic acetylcholine re- ceptors reduces Ab production and Ab-induced spine The emerging involvement of aberrant synaptic loss (Wei et al. 2010), raising the possibility that acti- function and plasticity in AD and depression provided vation of nicotinic acetylcholine receptors by increas- important insights on identifying new molecular ing acetylcholine level via AChE inhibition may be players as potential drug targets. In particular, modu- detrimental. On the other hand, the induction of ex- lators of NMDARs that facilitate synaptic trans- citotoxic death following Ab treatment and evidence mission or LTP induction could potentially function of excitotoxicity in AD brains led to the exploration of as cognitive enhancers to limit cognitive decline in NMDAR antagonists as potential neuroprotective AD and depression. Interestingly, studies aimed at agents against neuronal loss (Lipton, 2006). Many elucidating the mechanisms of action of memantine Developing cognitive enhancers from basic research reveal that excessive NMDAR activation may also pathophysiological machinery of various neurological impair synaptic plasticity (Parsons et al. 2007), sug- disorders, it is interesting to note that studies aimed at gesting that other NMDAR antagonists may be devel- explicating the mechanism of action of existing drugs oped as cognitive enhancers to combat decline in have also provided remarkable insights. For example, cognitive functions. In light of the successful devel- recent studies have revealed that neurogenesis in- opment of memantine as AD therapeutics, it is im- duced by antidepressants is critical for their thera- portant to select for NMDAR antagonists that preserve peutic effect (Krishnan & Nestler, 2008 ; Pittenger & physiological function of NMDARs. Similarly, devel- Duman, 2008). Interestingly, memantine has also been opment of positive modulators targeting AMPAR recently observed to induce adult neurogenesis in trafficking and function may also prove to be the hippocampus (Maekawa et al. 2009). In light of the beneficial to ameliorating synaptic deficit in these emerging involvement of neurogenesis in learning disorders. On the other hand, the critical involvement and memory (Deng et al. 2010), it will be important to of BDNF in synaptic plasticity indicates that TrkB explore if other modulators of neurogenesis may also agonists may also help to reverse synaptic deficits in be developed as cognitive enhancers.
AD and certain brain regions in depressed patients.
Direct infusion of trophic factors such as NGF as therapeutics for neurodegenerative diseases have longattracted interest based on their neuroprotective effect.
We thank Ka-Chun Lok for his excellent help in Nonetheless, inefficient crossing of the blood–brain preparing the figures and Dr Amy Fu for critical barrier and the non-specific effect of these trophic reading of the manuscript. The study of N. Y. Ip and factors have limited their applicability as therapeutics.
Z. H. Cheung was supported in part by the Research It is therefore important to identify TrkB agonists that Grants Council of Hong Kong (HKUST 6431/06M, can cross the blood–brain barrier, in addition to de- 661109, 661309 and 1/06C) and the Area of Excellence veloping methods that would allow precise delivery of Scheme of the University Grants Committee (AoE/ the agonists. Interestingly, a recent study reported the B-15/01) and the Hong Kong Jockey Club. N. Y. Ip and identification of a small molecule BDNF mimetic that Z. H. Cheung were Croucher Foundation Senior reduces neurotoxin-induced neuronal loss in vitro, in Research Fellow and Croucher Foundation Fellow, addition to restoring motor learning following trau- matic brain injury (Massa et al. 2010). This observationfurther supports the potential therapeutic efficacy of Statement of Interest TrkB agonists in ameliorating synaptic deficits inneurodegenerative diseases. Cdk5 inhibitors present yet another potential drug target for the developmentof cognitive enhancer in light of its involvement in AD pathology. In particular, since Cdk5 conditionalknockout mice exhibit enhanced spatial learning Amaral MD, Pozzo-Miller L (2007). TRPC3 channels are necessary for brain-derived neurotrophic factor to (Hawasli et al. 2007), the development of Cdk5 in- activate a nonselective cationic current and to induce hibitors as therapeutics for AD will not only alleviate dendritic spine formation. Journal of Neuroscience 27, neuronal loss, but will also enable amelioration of disease progression by targeting synaptic dysfunction Ampuero E, Rubio FJ, Falcon R, Sandoval M, et al. (2010).
early on (Fig. 2). Collectively, these studies have Chronic fluoxetine treatment induces structural plasticity identified several new molecular players that may be and selective changes in glutamate receptor subunits in targeted to reverse synaptic and cognitive deficits in the rat cerebral cortex. Neuroscience 169, 98–108.
AD, and other disorders where synaptic impairment An JJ, Gharami K, Liao GY, Woo NH, et al. (2008). Distinct may be implicated. Interestingly, there has also been role of long 3k UTR BDNF mRNA in spine morphology an increasing demand for cognitive enhancers as a and synaptic plasticity in hippocampal neurons. Cell 134, lifestyle drug for healthy individuals. While the ethical Benavides DR, Bibb JA (2004). Role of Cdk5 in drug abuse issues related to the consumption of cognitive en- and plasticity. Annals of the New York Academy of Sciences hancers remain a debated topic, it is plausible that 1025, 335–344.
these cognitive enhancers may also benefit healthy Bloss EB, Janssen WG, McEwen BS, Morrison JH (2010).
Interactive effects of stress and aging on structural Finally, while novel drug targets are identified plasticity in the prefrontal cortex. Journal of Neuroscience from basic research focused on elucidating the 30, 6726–6731.
Z. H. Cheung and N. Y. Ip Bramham CR (2008). Local protein synthesis, actin dynamics, Hardy J, Selkoe DJ (2002). The amyloid hypothesis of and LTP consolidation. Current Opinion in Neurobiology Alzheimer's disease : progress and problems on the road 18, 524–531.
to therapeutics. Science 297, 353–356.
Calabrese F, Molteni R, Racagni G, Riva MA (2009).
Hawasli AH, Benavides DR, Nguyen C, Kansy JW, et al.
Neuronal plasticity : a link between stress and mood (2007). Cyclin-dependent kinase 5 governs learning and disorders. Psychoneuroendocrinology 34 (Suppl. 1), synaptic plasticity via control of NMDAR degradation.
Nature Neuroscience 10, 880–886.
Chen AC, Shirayama Y, Shin KH, Neve RL, et al.
Hsieh H, Boehm J, Sato C, Iwatsubo T, et al. (2006).
(2001). Expression of the cAMP response element AMPAR removal underlies Abeta-induced synaptic binding protein (CREB) in hippocampus produces depression and dendritic spine loss. Neuron 52, 831–843.
an antidepressant effect. Biological Psychiatry 49, Ji Y, Pang PT, Feng L, Lu B (2005). Cyclic AMP controls BDNF-induced TrkB phosphorylation and dendritic spine Chen F, Madsen TM, Wegener G, Nyengaard JR (2008a).
formation in mature hippocampal neurons. Nature Changes in rat hippocampal CA1 synapses following Neuroscience 8, 164–172.
imipramine treatment. Hippocampus 18, 631–639.
Karege F, Vaudan G, Schwald M, Perroud N, et al. (2005).
Chen Y, Dube CM, Rice CJ, Baram TZ (2008b). Rapid loss Neurotrophin levels in postmortem brains of suicide of dendritic spines after stress involves derangement of victims and the effects of antemortem diagnosis and spine dynamics by corticotropin-releasing hormone.
psychotropic drugs. Molecular Brain Research 136, 29–37.
Journal of Neuroscience 28, 2903–2911.
Kelleher IIIrd RJ, Govindarajan A, Tonegawa S (2004).
Chen Y, Rex CS, Rice CJ, Dube CM, et al. (2010). Correlated Translational regulatory mechanisms in persistent forms memory defects and hippocampal dendritic spine loss of synaptic plasticity. Neuron 44, 59–73.
after acute stress involve corticotropin-releasing hormone Kim Y, Sung JY, Ceglia I, Lee KW, et al. (2006).
signaling. Proceedings of the National Academy of Sciences Phosphorylation of WAVE1 regulates actin polymerization USA 107, 13123–13128.
and dendritic spine morphology. Nature 442, 814–817.
Cheung ZH, Fu AK, Ip NY (2006). Synaptic roles of Knobloch M, Mansuy IM (2008). Dendritic spine loss and Cdk5 : implications in higher cognitive functions and synaptic alterations in Alzheimer's disease. Molecular neurodegenerative diseases. Neuron 50, 13–18.
Neurobiology 37, 73–82.
Deng W, Aimone JB, Gage FH (2010). New neurons and new Kobayashi S, Ishiguro K, Omori A, Takamatsu M, et al.
memories : how does adult hippocampal neurogenesis (1993). A cdc2-related kinase PSSALRE/cdk5 is affect learning and memory? Nature Reviews Neuroscience homologous with the 30 kDa subunit of tau protein 11, 339–350.
kinase II, a proline-directed protein kinase associated Dillon C, Goda Y (2005). The actin cytoskeleton : integrating with microtubule. FEBS Letters 335, 171–175.
form and function at the synapse. Annual Review of Korte M, Carroll P, Wolf E, Brem G, et al. (1995).
Neuroscience 28, 25–55.
Hippocampal long-term potentiation is impaired in mice Ethell IM, Pasquale EB (2005). Molecular mechanisms of lacking brain-derived neurotrophic factor. Proceedings of dendritic spine development and remodeling. Progress the National Academy of Sciences USA 92, 8856–8860.
in Neurobiology 75, 161–205.
Krishnan V, Nestler EJ (2008). The molecular neurobiology Fischer A, Sananbenesi F, Pang PT, Lu B, et al. (2005).
of depression. Nature 455, 894–902.
Opposing roles of transient and prolonged expression of Kurup P, Zhang Y, Xu J, Venkitaramani DV, et al. (2010).
p25 in synaptic plasticity and hippocampus-dependent Abeta-mediated NMDA receptor endocytosis in memory. Neuron 48, 825–838.
Alzheimer's disease involves ubiquitination of the Fu WY, Chen Y, Sahin M, Zhao XS, et al. (2007). Cdk5 tyrosine phosphatase STEP61. Journal of Neuroscience 30, regulates EphA4-mediated dendritic spine retraction through an ephexin1-dependent mechanism. Nature Lacor PN, Buniel MC, Furlow PW, Clemente AS, et al.
Neuroscience 10, 67–76.
(2007). Abeta oligomer-induced aberrations in synapse Hains AB, Vu MA, Maciejewski PK, van Dyck CH, et al.
composition, shape, and density provide a molecular basis (2009). Inhibition of protein kinase C signaling protects for loss of connectivity in Alzheimer's disease. Journal of prefrontal cortex dendritic spines and cognition from the Neuroscience 27, 796–807.
effects of chronic stress. Proceedings of the National Academy Lai KO, Ip NY (2009). Recent advances in understanding of Sciences USA 106, 17957–17962.
the roles of Cdk5 in synaptic plasticity. Biochimica et Hajszan T, MacLusky NJ, Leranth C (2005). Short-term Biophysica Acta 1792, 741–745.
treatment with the antidepressant fluoxetine triggers Lau CG, Zukin RS (2007). NMDA receptor trafficking in pyramidal dendritic spine synapse formation in rat synaptic plasticity and neuropsychiatric disorders. Nature hippocampus. European Journal of Neuroscience 21, Reviews Neuroscience 8, 413–426.
Li BS, Sun MK, Zhang L, Takahashi S, et al. (2001).
Hanley JG (2008). AMPA receptor trafficking pathways and Regulation of NMDA receptors by cyclin-dependent links to dendritic spine morphogenesis. Cellular Adhesion kinase-5. Proceedings of the National Academy of Sciences and Migration 2, 276–282.
USA 98, 12742–12747.
Developing cognitive enhancers from basic research Lipton SA (2006). Paradigm shift in neuroprotection by neurons through a phosphatidylinositol-3 kinase- NMDA receptor blockade : memantine and beyond.
dependent pathway. Journal of Neuroscience 20, 3165–3174.
Nature Reviews Drug Discovery 5, 160–170.
Rohe M, Synowitz M, Glass R, Paul SM, et al. (2009). Brain- Liston C, Miller MM, Goldwater DS, Radley JJ, et al.
derived neurotrophic factor reduces amyloidogenic (2006). Stress-induced alterations in prefrontal cortical processing through control of SORLA gene expression.
dendritic morphology predict selective impairments in Journal of Neuroscience 29, 15472–15478.
perceptual attentional set-shifting. Journal of Neuroscience Saglietti L, Dequidt C, Kamieniarz K, Rousset MC, et al.
26, 7870–7874.
(2007). Extracellular interactions between GluR2 and Lu B, Pang PT, Woo NH (2005). The yin and yang of N-cadherin in spine regulation. Neuron 54, 461–477.
neurotrophin action. Nature Reviews Neuroscience 6, Schindowski K, Belarbi K, Buee L (2008). Neurotrophic factors in Alzheimer's disease : role of axonal transport.
Maekawa M, Namba T, Suzuki E, Yuasa S, et al. (2009).
Genes, Brain and Behavior 7 (Suppl. 1), 43–56.
NMDA receptor antagonist memantine promotes cell Selkoe DJ (2002). Alzheimer's disease is a synaptic failure.
proliferation and production of mature granule neurons Science 298, 789–791.
in the adult hippocampus. Neuroscience Research 63, Shankar GM, Bloodgood BL, Townsend M, Walsh DM, et al. (2007). Natural oligomers of the Alzheimer Massa SM, Yang T, Xie Y, Shi J, et al. (2010). Small molecule amyloid-beta protein induce reversible synapse loss BDNF mimetics activate TrkB signaling and prevent by modulating an NMDA-type glutamate receptor- neuronal degeneration in rodents. Journal of Clinical dependent signaling pathway. Journal of Neuroscience 27, investigation 120, 1774–1785.
Matrone C, Ciotti MT, Mercanti D, Marolda R, et al. (2008).
Silva-Gomez AB, Rojas D, Juarez I, Flores G (2003).
NGF and BDNF signaling control amyloidogenic route and Decreased dendritic spine density on prefrontal cortical Abeta production in hippocampal neurons. Proceedings of and hippocampal pyramidal neurons in postweaning the National Academy of Sciences USA 105, 13139–13144.
social isolation rats. Brain Research 983, 128–136.
Monti B, Contestabile A (2009). Memory-enhancing drugs : a Snyder EM, Nong Y, Almeida CG, Paul S, et al. (2005).
molecular perspective. Mini-Reviews in Medicinal Chemistry Regulation of NMDA receptor trafficking by amyloid-beta.
9, 769–781.
Nature Neuroscience 8, 1051–1058.
Nagappan G, Zaitsev E, Senatorov Jr. VV, Yang J, et al.
Soetanto A, Wilson RS, Talbot K, Un A, et al. (2010).
(2009). Control of extracellular cleavage of ProBDNF by Association of anxiety and depression with high frequency neuronal activity. Proceedings of the National microtubule-associated protein 2- and synaptopodin- Academy of Sciences USA 106, 1267–1272.
immunolabeled dendrite and spine densities in Parsons CG, Stoffler A, Danysz W (2007). Memantine : a hippocampal CA3 of older humans. Archives of General NMDA receptor antagonist that improves memory by Psychiatry 67, 448–457.
restoration of homeostasis in the glutamatergic Spires TL, Meyer-Luehmann M, Stern EA, McLean PJ, et al.
system – too little activation is bad, too much is even (2005). Dendritic spine abnormalities in amyloid precursor worse. Neuropharmacology 53, 699–723.
protein transgenic mice demonstrated by gene transfer and Passafaro M, Nakagawa T, Sala C, Sheng M (2003).
intravital multiphoton microscopy. Journal of Neuroscience Induction of dendritic spines by an extracellular domain 25, 7278–7287.
of AMPA receptor subunit GluR2. Nature 424, 677–681.
Tada T, Sheng M (2006). Molecular mechanisms of dendritic Patterson SL, Abel T, Deuel TA, Martin KC, et al. (1996).
spine morphogenesis. Current Opinion in Neurobiology 16, Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout Tanaka J, Horiike Y, Matsuzaki M, Miyazaki T, et al. (2008).
mice. Neuron 16, 1137–1145.
Protein synthesis and neurotrophin-dependent structural Pittenger C, Duman RS (2008). Stress, depression, and plasticity of single dendritic spines. Science 319, 1683–1687.
neuroplasticity : a convergence of mechanisms.
Thome J, Sakai N, Shin K, Steffen C, et al. (2000). cAMP Neuropsychopharmacology 33, 88–109.
response element-mediated gene transcription is Radley JJ, Rocher AB, Miller M, Janssen WG, et al. (2006).
upregulated by chronic antidepressant treatment. Journal of Repeated stress induces dendritic spine loss in the rat Neuroscience 20, 4030–4036.
medial prefrontal cortex. Cerebral Cortex 16, 313–320.
Tongiorgi E, Righi M, Cattaneo A (1997). Activity- Rakofsky JJ, Holtzheimer PE, Nemeroff CB (2009).
dependent dendritic targeting of BDNF and TrkB mRNAs Emerging targets for antidepressant therapies. Current in hippocampal neurons. Journal of Neuroscience 17, Opinion in Chemical Biology 13, 291–302.
Rammes G, Danysz W, Parsons CG (2008).
Wasling P, Daborg J, Riebe I, Andersson M, et al. (2009).
Pharmacodynamics of memantine : an update. Current Synaptic retrogenesis and amyloid-beta in Alzheimer's Neuropharmacology 6, 55–78.
disease. Journal of Alzheimer's Disease 16, 1–14.
Righi M, Tongiorgi E, Cattaneo A (2000). Brain-derived Waterhouse EG, Xu B (2009). New insights into the role of neurotrophic factor (BDNF) induces dendritic targeting brain-derived neurotrophic factor in synaptic plasticity.
of BDNF and tyrosine kinase B mRNAs in hippocampal Molecular and Cellular Neuroscience 42, 81–89.
Z. H. Cheung and N. Y. Ip Wei W, Nguyen LN, Kessels HW, Hagiwara H, et al. (2010).
Zakzanis KK, Leach L, Kaplan E (1998). On the nature Amyloid beta from axons and dendrites reduces local and pattern of neurocognitive function in major spine number and plasticity. Nature Neuroscience 13, depressive disorder. Neuropsychiatry, Neuropsychology, and Behavioral Neurology 11, 111–119.
Wen Y, Yu WH, Maloney B, Bailey J, et al. (2008).
Zhang S, Edelmann L, Liu J, Crandall JE, et al. (2008).
Transcriptional regulation of beta-secretase by p25/cdk5 Cdk5 regulates the phosphorylation of tyrosine 1472 leads to enhanced amyloidogenic processing. Neuron 57, NR2B and the surface expression of NMDA receptors.
Journal of Neuroscience 28, 415–424.
Woo NH, Teng HK, Siao CJ, Chiaruttini C, et al. (2005).
Zhou Q, Homma KJ, Poo MM (2004). Shrinkage of Activation of p75NTR by proBDNF facilitates dendritic spines associated with long-term hippocampal long-term depression. Nature depression of hippocampal synapses. Neuron 44, Neuroscience 8, 1069–1077.

Source: http://repository.ust.hk/ir/bitstream/1783.1-8094/1/S1461145710001537a.pdf