The neurotrophin nerve growth factor (NGF) and its high-affinity receptor TrkA play a major role in the survival of basal forebrain cholinergic neurons, which are selectively vulnerable during the progression of AD (Counts et al., 2004). Degeneration of these cholinergic forebrain neurons correlates with the cognitive decline seen in people with this disease. Therefore, the development of drugs that mimic NGF or its signal transducing TrkA receptor are currently under investigation (Massa et al., 2003; Peleshok and Saragovi, 2006) as novel treatments to slow the insidious degeneration of the cholinergic basal forebrain neurons during the progression of AD. In this light, Jang and coworkers used a high-throughput screening assay to identify small-molecule agonists for the TrkA receptor. These investigators found a potent compound, gambogic amide, which selectively binds to TrkA on a different region than does the NGF molecule. Gambogic amide...  Read more

The neurotrophin nerve growth factor (NGF) and its high-affinity receptor TrkA play a major role in the survival of basal forebrain cholinergic neurons, which are selectively vulnerable during the progression of AD (Counts et al., 2004). Degeneration of these cholinergic forebrain neurons correlates with the cognitive decline seen in people with this disease. Therefore, the development of drugs that mimic NGF or its signal transducing TrkA receptor are currently under investigation (Massa et al., 2003; Peleshok and Saragovi, 2006) as novel treatments to slow the insidious degeneration of the cholinergic basal forebrain neurons during the progression of AD. In this light, Jang and coworkers used a high-throughput screening assay to identify small-molecule agonists for the TrkA receptor. These investigators found a potent compound, gambogic amide, which selectively binds to TrkA on a different region than does the NGF molecule. Gambogic amide exhibits many of the neurotrophic actions of NGF, including TrkA dimerization, tyrosine phosphorylation, neuronal survival, and the activation of downstream signaling cascades (e.g., Akt and MAPK).

These findings provide evidence that a compound derived from gamboges, a resin exuded from the Garcinia hanburyi tree in Southeast Asia, could expand the currently available treatment strategies aimed at rescuing the degeneration of cholinergic basal forebrain neurons in AD. These observations come at time when data from a phase 1 clinical trial, in which cells genetically engineered to produce NGF were injected into the cholinergic basal forebrain of patients with mild AD, provided evidence that this treatment reduced cognitive decline by 36-51 percent (Tuszynski et al., 2005; Tuszynski, 2007; Ceregene pipeline). Moreover, a brain autopsy from an AD subject in this trial demonstrated robust “trophic” (i.e., growth) responses to NGF.

If gambogic amide is also found to reverse cognitive decline in animal models of AD, this compound could conceivably be used in combination with NGF or alone as a treatment for AD. In addition, this substance will have to undergo extensive efficacy trials in primates, as well. Many questions still remain concerning the biology of gambogic amide. Considering that there is a decrease in TrkA and a stability of the low-affinity NGF receptor during the progression of AD (Counts et al., 2004), it is crucial to determine whether gambogic amide binds to the low-affinity NGF receptor, which possesses the ability to induce cellular apoptosis in conjunction with proNGF in AD (Lee et al., 2001; Peng et al., 2004). The fact that gambogic amide also prevents neuronal cell death and decreases infarct volume in a middle cerebral artery occlusion model expands its treatment potential to the area of stroke, making it even more clinically relevant. As with all novel findings with a clinical imperative, more questions arise than are answered by the initial studies.

View all comments by Elliott Mufson

This paper describes the discovery of gambogic amide (GA-amide), a compound that selectively binds to the cytoplasmic juxtamembrane domain of TrkA with an apparent Kd of ~75nM, that at nanomolar concentrations activates TrkA and its downstream signaling components AKT and ERK, that prevents death of cultured hippocampal cells, and achieves the NGF-like effect of promoting neurite outgrowth of PC12 cells. This compound is described as a “selective agonist for TrkA,” and the possibility is presented that it might therefore be useful for the treatment of neurodegenerative diseases including Alzheimer disease (AD). The identification of a non-peptide, small molecule capable of selectively promoting TrkA activation at nanomolar concentrations, in the absence of NGF supplementation, is an important advance. This progress can be examined from three important perspectives: previous work developing TrkA agonists; the “small molecule” profile of GA-amide; and application of TrkA agonists in AD.

Early work with synthetic peptides modeled against specific domains of NGF encouraged the...  Read more

This paper describes the discovery of gambogic amide (GA-amide), a compound that selectively binds to the cytoplasmic juxtamembrane domain of TrkA with an apparent Kd of ~75nM, that at nanomolar concentrations activates TrkA and its downstream signaling components AKT and ERK, that prevents death of cultured hippocampal cells, and achieves the NGF-like effect of promoting neurite outgrowth of PC12 cells. This compound is described as a “selective agonist for TrkA,” and the possibility is presented that it might therefore be useful for the treatment of neurodegenerative diseases including Alzheimer disease (AD). The identification of a non-peptide, small molecule capable of selectively promoting TrkA activation at nanomolar concentrations, in the absence of NGF supplementation, is an important advance. This progress can be examined from three important perspectives: previous work developing TrkA agonists; the “small molecule” profile of GA-amide; and application of TrkA agonists in AD.

Early work with synthetic peptides modeled against specific domains of NGF encouraged the view that small molecules might be found that target the extracellular domain of TrkA to achieve its activation (1,2). Further development of small molecule peptidomimetic TrkA ligands has led to application of such compounds in animal disease models (3). Crossing the threshold to non-peptide-related compounds capable of activating TrkA was described recently by Lin et al. (4). These investigators screened combinatorial libraries and identified an asterriquinone (1H5) and a mono-indolyl-quinone (5E5) that activate TrkA, presumably through interaction at an intracellular site, and prevent PC12 cell death at low micomolar concentrations. The present Jang et al. study offers the advances of finding a non-peptide compound that is active at low nanomolar concentrations (a druggable range), the verification of its site of action at the TrkA cytoplasmic juxtamembrane domain, and the relative selectivity for TrkA.

In terms of small molecule development, critical points to consider include target specificity, pharmacological mechanisms, and the potential for druggable derivatives. GA-amide binds to the cytoplasmic juxtamembrane domain of TrkA rather than its extracellular binding site, and hence is not a typical competitive agonist acting at the extracellular ligand binding site. The absence of a high degree of conservation between Trk juxtamembrane regions and the Western blot data showing preferential activation of TrkA in transfected HEK293 cells supports selectivity to TrkA; however, additional studies will be required to establish the degree of this selectivity. Such studies might include determining if GA-amide activates TrkB in the hippocampal cultures, use of neurons with downregulated TrkA, application of GA-amide to other tyrosine kinases, and large-scale receptor screens such as the CEREP panel. Receptor selectivity has been an important issue in attempts to develop small molecule activators of the insulin tyrosine kinase receptor which, like TrkA, undergoes dimerization and autophosphorylation upon protein ligand binding. Zhang et al. (5) identified a compound (L-783,281) that activates the insulin receptor via interaction with its intracellular enzymatic domain. Although initially described as selective for the insulin receptor, it was subsequently found to activate other tyrosine kinases, including TrkA (6).

The term “small molecule” is generally used for compounds with a MW < 500, a size typical of most, but not all, useful drugs. For a first look at a compound of interest, one can apply the four Lipinski criteria (“rule of fives”) which predict oral bioavailability (7). Favorable features include: MW < 500 (GA-amide = 628, and is therefore not a “small molecule” by conventional terminology); cLog p < 5.0 (GA-amide = 9.4); hydrogen bond donors < 5 (GA-amide = 3); and hydrogen bond acceptors < 10 (GA-amide = 7). Thus, GA-amide has only two out of four favorable Lipinski features. In addition, calculation of predicted blood-brain barrier penetration (8) indicates that GA-amide will exhibit poor penetration. Of note, the blood-brain barrier is breached in both of the models tested in the Jang et al. report, kainic acid toxicity and stroke. These characteristics raise the likelihood that a major program in medicinal chemistry would be required to create derivatives useful for testing in CNS applications involving oral delivery and an intact blood-brain barrier. Creating these features, while at the same time maintaining potency in the low nanomolar range and selectivity or specificity for TrkA, will likely constitute a considerable challenge.

A third consideration is the extent to which TrkA agonists might be useful in AD. Given the role of NGF and TrkA in the promotion of pain signaling, and the pain side effects found with NGF delivered to human CNS and peripheral routes along with the current positive clinical trials in treating pain conditions with strategies that block NGF-induced TrkA activation (9), a clinical scenario involving chronic peripheral administration of a TrkA activator seems unworkable. In addition to upregulating pain transmission, peripheral administration of NGF in animals has been found to lead to hypertrophy of sympathetic ganglia, excessive vascular sympathetic innervation (10), and concerns regarding induced hypertension. Moreover, TrkA activation might play a role in promoting tumor growth and metastasis (11). These peripheral obstacles, along with current trials involving the implantation of NGF-secreting fibroblasts in the basal forebrain region of AD patients with the goal of improving function of cholinergic neurons, point to the alternative possibility of local CNS application of TrkA small molecule activators. Cholinergic neurites, with functions likely relevant to cognition, extend to targets throughout the cortex and hippocampus. Application to these extensive target regions would be difficult, particularly on the scale of the human brain. Another option is direct administration to the basal forebrain, but issues of reaching the majority of relevant cholinergic neurons and the practical issues of the hardware or advanced polymers required for focal and chronic delivery remain.

Perhaps the most interesting issue to consider is the role of TrkA activation in AD. Given that most neuronal populations affected in AD express little or no TrkA, any beneficial effect of activating this receptor would be expected to be primarily limited to counteracting the progressive atrophy and loss of function of basal forebrain cholinergic neurons. However, a key question is whether TrkA activation can inhibit one or more of the underlying deleterious mechanisms causing loss of synaptic function and neuronal degeneration in AD. TrkA-induced activation of PI3K/AKT and ERK signaling has the potential to inhibit Aβ-induced activation of stress kinases, GSK3β and JNK (12,13,14). On the other hand, NGF acting through TrkA can exacerbate Aβ-induced toxicity (15) and ERK activation might contribute to tau hyperphosphorylation and caspase activation (16). In addition, TrkA signaling has been shown to induce γ-secretase cleavage of the p75 receptor resulting in increased levels of the p75 intracellular domain (17), a mechanism potentially contributing to degeneration in AD (18).

The Jang et al. work introduces an important new compound with potent effects for selective activation of TrkA. This compound, and its derivatives, will contribute elucidating TrkA signaling mechanisms. Their application in mouse models relevant to AD will provide important new insights regarding TrkA-mediated therapeutic potential in AD.

References:
1. LeSauteur L, Wei L, Gibbs BF and Saragovi HU. Small peptide mimetics of nerve growth factor bind TrkA receptors and affect biological responses. J Biol Chem 270:6564-6569, 1995. Abstract

2. Xie Y, Tisi MA, Yeo TT and Longo FM. NGF loop 4 dimeric mimetics activate ERK, prevent neuronal death and promote neurite outgrowth. J Biol Chem 275:29868-29874, 2000. Abstract

3. Peleshok J and Saragovi HU. Functional mimetics of neurotrophins and their receptors. Biochem Soc Trans 34:612-617, 2006. Abstract

4. Lin B, Pirrung MC, Deng L, Li Z, Liu Y and Webster NG. Neuroprotection by small molecule activators of the nerve growth factor receptor. J Pharm Exp Ther 322:59-69, 2007 (Epub Apr 2007). Abstract

5. Zhang B, Salituro G, Szalkowski D et al. Discovery of a small molecule insulin mimetic with antidiabetic activity in mice. Science 284:974-977, 1999. Abstract

6. Wilkie N, Wingrove PB, Bilsland JG, Young L, Harper SJ, Hefti F, Ellis S and Pollack SJ. The non-peptidyl fungal metabolite L-783,281 activates TRK neurotrophin receptors. J Neurochem 87:1135-1145, 2001. Abstract

7. Lipinski CA. Drug-like properties and the causes of poor solubility and permeability. J Pharm Tox Meth 44:235-249, 2000. Abstract

8. Fu XC, Chen CX, Liang WQ, Yu QS. Predicting blood-brain barrier penetration of drugs by polar molecular surface area and molecular volume. Acta Pharmacol Sin 22:663-668, 2001. Abstract

9. Hefti FF, Rosenthal A, Walicke PA, Wyatt S, Vergara G, Shelton DL and Davies AM. Novel class of pain drugs based on antagonism of NGF. Trends Pharmacol Sci 27:85-91, 2006. Abstract

10. Zettler C, Head RJ and Rush RA. Chronic nerve growth factor treatment of normotensive rats. Brain Res 538:251-262, 1991. Abstract

11. Kruttgen A, Schneider I and Weis J. The dark side of the NGF family: neurotrophins in neoplasias. Brain Pathol 16:304-310, 2006. Abstract

12. Cross DA, Alessi DR, Cohen P, Andjelkovich M and Hemmings BA. Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature 387:785-789, 1995. Abstract

13. Wei W, Wang X and Kusiak JW. Signaling events in amyloid β-peptide-induced neuronal death and insulin growth factor I protection. J Biol Chem 277:17649, 2002. Abstract

14. Park H-S, Kim MS, Huh SH, Park J, Chung J, Kang SS and Choi EJ. Akt (protein kinase B) negatively regulates SEK1 by means of protein phosphorylation. J Biol Chem 277: 2573-2578, 2002. Abstract

15. Rabizadeh A, Bitler CM, Butcher LL and Bredesen DE. Expression of the low-affinity NGF receptor enhances β-amyloid peptide toxicity. Proc Natl Acad Sci USA 91:10703-10706, 1994. Abstract

16. Chong YH, Shin YJ, Lee EO, Kayed R, Glabe CG and Tenner AJ. ERK1/2 activation mediates Abeta oligomer-induced neurotoxicity via caspase-3 activation and tau cleavage in rat organotypic hippocampal slice cultures. J Biol Chem 281:20315-20325, 2006. Abstract

17. Urra S, Escudero CA, Ramos P, Lisbona F, Allende E, Covarrubias P, Parraguez JI, Zampieri N, Chao MV, Annaert W, Bronfman FC. TrkA receptor activation by nerve growth factor induces shedding of the p75 neurotrophin receptor followed by endosomal gamma-secretase-mediated release of the p75 intracellular domain. J Biol Chem 282:7606-7615, 2007. Abstract

18. Podlesniy P, Kichev A, Pedraza C et al. Pro-NGF from Alzheimer's disease and normal human brain displays distinctive abilities to induce processing and nuclear translocation of intracellular domain of p75NTR and apoptosis. Am J Pathol 169:119-31, 2006. Abstract

View all comments by Frank Longo
View all comments by Stephen Massa

Comment by Carolyn Tyler and Howard Federoff
NGF-TrkA signaling has been shown to be necessary for the survival and normal physiology of basal forebrain cholinergic neurons (BFCNs) (Sofroniew et al., 2001). NGF has been intensely studied as a potential therapeutic for Alzheimer disease (AD), as the BFCNs have been shown to be particularly susceptible to degeneration in AD patients and the pathophysiology of these cells is believed to cause some of the debilitating cognitive decline observed in AD (Mufson et al., 2003). NGF therapy, however, has presented some challenging technical issues due to the complexity of neurotrophin biology and limited access to the CNS. Current potential NGF therapies involve invasive administration to specific sites within the CNS and an investigational gene delivery approach, in which the NGF gene is delivered by a virus vector administered to the brain neurosurgically. Both direct NGF protein and gene delivery are complex and difficult to regulate dose. To circumvent these problems, small-molecule NGF mimetics and TrkA agonists have been...  Read more

Comment by Carolyn Tyler and Howard Federoff
NGF-TrkA signaling has been shown to be necessary for the survival and normal physiology of basal forebrain cholinergic neurons (BFCNs) (Sofroniew et al., 2001). NGF has been intensely studied as a potential therapeutic for Alzheimer disease (AD), as the BFCNs have been shown to be particularly susceptible to degeneration in AD patients and the pathophysiology of these cells is believed to cause some of the debilitating cognitive decline observed in AD (Mufson et al., 2003). NGF therapy, however, has presented some challenging technical issues due to the complexity of neurotrophin biology and limited access to the CNS. Current potential NGF therapies involve invasive administration to specific sites within the CNS and an investigational gene delivery approach, in which the NGF gene is delivered by a virus vector administered to the brain neurosurgically. Both direct NGF protein and gene delivery are complex and difficult to regulate dose. To circumvent these problems, small-molecule NGF mimetics and TrkA agonists have been pursued to allow for an easier route of administration as well as more dosage control (Saragovi, 2000; Massa, 2003; Longo, 2005). In this paper, Jang et al. have identified a new and promising TrkA agonist, gambogic amide, that binds to the cytoplasmic juxtamembrane domain of TrkA with high affinity. Using cultured hippocampal neurons, the authors demonstrate that binding of gambogic amide to TrkA elicits the same molecular and cellular responses as NGF, including the induction of receptor phosphorylation, activation of downstream signaling cascades such as MAP-kinase and Akt, and subsequent neurite outgrowth and inhibition of apoptosis.

The authors report several interesting observations regarding this agonist. First, as previously mentioned, gambogic amide binds independently of the NGF binding site on TrkA (TrkA-d5), but still induces similar receptor activation. This is an important insight into the nature of TrkA signaling, which has also been demonstrated previously. The activation of TrkA signaling through a separate cytoplasmic domain by gambogic amide illustrates the potential for the development of small-molecule TrkA agonists that bind independently of the TrkA-d5 domain.

The authors also report that the juxtamembrane domain is one of the few regions not highly homologous between the Trk family of receptors. This could allow gambogic amide to bind specifically to TrkA, without the simultaneous activation of TrkB and C, which are also expressed in the CNS. This is a very important point for AD therapy, because TrkA receptor expression is highly restricted in the CNS, while TrkB expression is widespread (Kordower et al., 1988; Muragaki et al., 1995). The specificity of gambogic amide for TrkA over TrkB reduces any “off-target” effects that may complicate a systemically administered treatment.

Another observation in this paper is the ability of gambogic amide to protect TrkA-expressing cells from apoptosis due to excitotoxicity, even though it is a derivative of gambogic acid (GA), which has been previously demonstrated to be an apoptosis-inducing ligand for the transferrin receptor (TfR) (Kasibhatla et al., 2005). It is unclear if gambogic amide can also induce TfR-dependent apoptosis, as it possesses the tricyclic ring and α, β unsaturated ketone that is critical to GA activity, but the authors demonstrate that GA is able to protect cells in a TrkA-dependent manner. They also show that both gambogic amide and GA can protect hippocampal neurons from apoptosis during oxygen-glucose deprivation. This indicates that in the presence of a high level of TrkA expression, the default action of gambogic amide and GA is to prevent apoptosis, but in cells lacking TrkA no protective effect of these compounds is observed. Furthermore, because GA has been previously demonstrated to induce apoptosis in TfR-expressing cells, more investigation is warranted into the dynamics between TrkA and TfR expression and the effect on apoptosis. This would be a critical point for in vivo applications of gambogic amide, as TrkA and TfR are both expressed systemically and it is unclear what effect the interaction of the two signaling pathways would have on a variety of cell populations.

Finally, the authors demonstrate that during two different in-vivo insults, kainic acid excitotoxicity and middle cerebral artery occlusion (MCAO), subcutaneously administered gambogic amide decreases apoptosis and infarct volume in rat brain. However, the extent of blood-brain barrier (BBB) compromise during these insults is unknown, and therefore the permeability of gambogic amide to an intact BBB cannot be inferred. The ability of gambogic amide to efficiently cross into the CNS is obviously an important issue for the development of a systemically administered AD therapy. It has been demonstrated that as AD progresses the BBB is compromised (Zlokovic, 2002), which could allow for gambogic amide access to the diseased brain in the absence of normal permeability, but the treatment of BFCNs would perhaps be most efficacious during early-stage AD before extensive synapse loss and cell death has occurred. Therefore, it needs to be determined how effectively gambogic amide can penetrate the normal CNS. Newer therapeutics, particularly those that may change disease history, are needed for AD. Whether gambogic amide, which appears to function to promote the function and survival of BFCN, will make it to clinical trials and perhaps demonstrate efficacy for AD is unknown. However, discovery of new agents for this and other progressive neurodegenerative diseases is important and should be expanded.

References:
Kasibhatla S, Jessen KA, Maliartchouk S, Wang JY, English NM, Drewe J, Qiu L, Archer SP, Ponce AE, Sirisoma N, Jiang S, Zhang HZ, Gehlsen KR, Cai SX, Green DR, Tseng B. A role for transferrin receptor in triggering apoptosis when targeted with gambogic acid. Proc Natl Acad Sci U S A. 2005 Aug 23;102(34):12095-100. Abstract

Kordower JH, Bartus RT, Bothwell M, Schatteman G, Gash DM. Nerve growth factor receptor immunoreactivity in the nonhuman primate (Cebus apella): distribution, morphology, and colocalization with cholinergic enzymes. J Comp Neurol 1988; 277 (4):465-86. Abstract

Longo FM, Massa SM. Neurotrophin receptor-based strategies for Alzheimer's disease. Curr Alzheimer Res. 2005 Apr;2(2):167-9. Abstract

Massa SM, Xie Y, Longo FM. Alzheimer's therapeutics: neurotrophin domain small molecule mimetics. J Mol Neurosci. 2003;20(3):323-6. Abstract

Mufson EJ, Ginsberg SD, Ikonomovic MD, DeKosky ST. Human cholinergic basal forebrain: chemoanatomy and neurologic dysfunction. J. Chem. Neuroanat. 2003; 26:233-242. Abstract

Muragaki Y, Timothy N, Leight S, Hempstead BL, Chao MV, Trojanowski JQ, Lee VM. Expression of trk receptors in the developing and adult human central and peripheral nervous system. J Comp Neurol. 1995 Jun 5;356(3):387-97. Abstract

Saragovi HU, Gehring K. Development of pharmacological agents for targeting neurotrophins and their receptors.Trends Pharmacol Sci. 2000 Mar;21(3):93-8. Abstract

Sofroniew MV, Howe CL, Mobley WC. Nerve growth factor signaling, neuroprotection, and neural repair. Annu Rev Neurosci 2001; 24:1217-1281. Abstract

Zlokovic BV. Vascular disorder in Alzheimer's disease: role in pathogenesis of dementia and therapeutic targets. Adv Drug Deliv Rev. 2002 Dec 7;54(12):1553-9. Abstract

View all comments by Howard Federoff
View all comments by Carolyn Tyler