The Use of Small Molecules to Investigate Molecular Mechanisms and

May 3, 2017 - Clostridium botulinum and the disease it causes, botulism, have been known for centuries. This bacterium has been studied for 100 years,...
0 downloads 0 Views 782KB Size
REVIEW

The Use of Small Molecules to Investigate Molecular Mechanisms and Therapeutic Targets for Treatment of Botulinum Neurotoxin A Intoxication Tobin J. Dickerson* and Kim D. Janda*

Departments of Chemistry and Immunology, The Skaggs Institute for Chemical Biology, and Worm Institute for Research and Medicine (WIRM), The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037

C

lostridium botulinum and the disease it causes, botulism, have been known for centuries. This bacterium has been studied for ⬎100 years, with a number of publications dating back to the early 19th century. Case studies were reported as far back as 1815; however, it was not until van Ermengem isolated an anaerobic bacillus from contaminated meat in 1897 that the causative agent of botulism was discovered (1). Cultivation of the bacillus and subsequent introduction into animals led to the development of the symptoms of botulism. The discovery that the organism produced a toxin occurred shortly after the identification of the bacillus; however, it was only in the final decade of the 20th century that a full picture of the toxin’s structure, mechanism of action, and target substrates emerged (2). Structurally, C. botulinum is a rod-shaped, Grampositive, sporulating anaerobic bacillus that is widely distributed in the environment (3). Neurotoxins produced by C. botulinum are some of the most potent naturally occurring compounds known; the lethal dose for humans is ⬃1 ng kg⫺1 of body weight (4). Their exquisite toxicity, coupled with their highly specific mechanism of action, renders the botulinum neurotoxins (BoNTs) both highly dangerous and yet quite useful to medical science (5). BoNTs are typically associated with food poisoning, although they also are seen as a result of wound infections, inhalation, or as a colonizing infection in the intestinal tract of infants (6). BoNTs became a common public-health threat only after the advent of food preservation in the 19th century. Modern food-preparation practices have rendered botulism a rare occurrence from commercially prepared foods, www.acschemicalbiology.org

A B S T R A C T Botulinum neurotoxins (BoNTs) are agents responsible for botulism, a disease characterized by peripheral neuromuscular blockade and subsequent flaccid paralysis. The potent paralytic ability of these toxins has resulted in their use as a therapeutic; however, BoNTs are also classified by the Centers for Disease Control and Prevention as one of the six highest-risk threat agents of bioterrorism. Consequently, a thorough understanding of the molecular mechanism of BoNT toxicity is crucial before effective inhibitors and, ultimately, an approved drug can be developed. In this article, we systematically detail BoNT intoxication by examining each of the discrete steps in this process. Additionally, rationally designed strategies for combating the toxicity of the most potent BoNT serotype are evaluated.

*Corresponding authors, [email protected], [email protected].

Received for review April 25, 2006 and accepted May 26, 2006. Published online July 21, 2006 10.1021/cb600179d CCC: $33.50 © 2006 by American Chemical Society

VOL.1 NO.6 • ACS CHEMICAL BIOLOGY

359

Neuromuscular junction Axon terminal

Normal transmitter release

Action of botulinum neurotoxins

Neuron Light chain

Heavy chain

Muscle cell Synaptic vesicle

ease of production and transport, and need for Botulinum neurotoxin prolonged intensive cleaves care, BoNTs are classiSNARE fied by the Centers for proteins Botulinum SNARE neurotoxin Disease Control and proteins Synaptobrevin receptor Prevention (CDC) as Syntaxin SNAP 25 one of the six highestrisk threat agents for Botulinum Acetylcholine neurotoxin bioterrorism (Category Synaptic released cleft A agents). Yet, food poisoning Acetylcholine and weapons of mass receptor Muscle cell destruction are only two scenarios in which BoNTs play a role. Figure 1. Mechanism of action of BoNTs. (Adapted with permission from Rowland, L. P. (2002) New Engl. J. Med. 347, 382. Copyright 2002 Massachusetts Medical Society). Shown are the individual stages of BoNT intoxication, includDespite being extremely ing cell surface recognition, vesicle internalization, translocation of the LC protease into the cytosol, and proteolytic poisonous, BoNT is a cleavage of one of the proteins of the SNARE complex. These steps lead to inhibition of neurotransmitter-containing highly effective theravesicle release. BoNT/B, -D, -F, and -G cleave proteins of the VAMP family (blue), and BoNT/A, -C, and -E cleave SNAPpeutic agent and valu25 (yellow). BoNT/C can also cleave syntaxin (purple). able research tool (9). although a small but significant number of cases occur At the beginning of the 20th century, it was observed that injecting BoNT could paralyze individual muscle annually from eating canned foods. groups without giving the recipient botulism (10). Botulism is characterized by generalized muscle Indeed, preparations of BoNT serotype A (BoNT/A) have weakness, and in more severe cases, it results in been approved by the U.S. Food and Drug Administraimpaired respiratory function and autonomic dysfunction for use in treating strabismus, blepharospasm, and tion. In most severe cases, this leads to respiratory failure and death. One of the most fascinating aspects of hemifacial spasm (11). The use of BoNT has also been BoNT intoxication is that host death does not result from extended to cover a wide variety of disorders, including those that do not have a neuromuscular basis (12), such target cell death and subsequent accumulated tissue destruction. Rather, death of the host results from a sec- as axillary hyperhidrosis (excessive sweating), myofascial pain and tension, migraine headaches, and multiple ondary event (e.g., respiratory failure) that depends on sclerosis. In addition, polypeptide fragments derived toxin-induced inactivation of neurotransmitter release from the toxin are being evaluated as potential carrier from otherwise viable nerve cells. As previously stated, BoNT poisoning can occur acci- molecules in the construction of oral and inhalation vacdentally via food consumption; however, incident rates cines (13). Molecular Mechanism of BoNT and Neurotransare low. The major concern is its malicious use, especially in bioterrorism and biowarfare. During World War mitter Release. BoNT has seven serologically distinct II, the extremely high potency of BoNTs induced both the serotypes (A–G); these proteins have a molecular weight of ⬃150,000 kDa (14). BoNTs are synthesized as singleAllied and Axis powers to evaluate these proteins as polypeptide chains, and cleavage by intra- or extracellupotential biological warfare agents (7). This work forlar proteases converts them into dimers consisting of a mally ended with the signing of the 1972 Biological 100-kDa heavy chain (HC) coupled to a 50-kDa light Weapons Convention. However, recent events in the Middle East and Asia have confirmed the weaponization chain (LC) by one or more disulfide bonds. The twoof this toxin by the former Soviet Union and by the Iraqi chain molecule is the active form of the toxin that military before and during the 1991 Gulf War; the Japa- poisons cholinergic transmission. Each serotype is produced as the primary toxin by a specific strain of bactenese cult Aum Shinrikyo tried to use BoNT for bioterrorism (8). Because of their extreme potency and lethality, rium and, although they share a high degree of homol360

VOL.1 NO.6 • 359–369 • 2006

Acetylcholine

DICKERSON AND JANDA

www.acschemicalbiology.org

REVIEW ogy, they differ in their toxicity and molecular site of action. BoNT intoxication (as summarized in Figure 1) occurs through a multistep process involving each of the toxin functional domains and can be described as the outcome of discrete stages (15, 16). Binding to the Target Cell and Internalization. BoNTs bind to cholinergic nerve terminals by their HC domains and are subsequently internalized by receptor-mediated endocytosis. The identity of the putative receptor used by BoNT/A has recently been reported to be the synaptic vesicle protein SV2 (17, 18). Additionally, a second component of cellular recognition is thought to be due to low-affinity interactions between the toxin and the gangliosides (19). A double-receptor model has been proposed whereby BoNTs also bind to both of these receptors before internalization can occur (19). In such a scenario, within the first step, the toxin associates with the cell membrane and the gangliosides in a low-affinity complex. Next, this complex migrates laterally until it interacts with a high-affinity binding site; the latter interaction then allows subsequent events, such as receptormediated endocytosis. Toxin binding to receptors also appears to be serotype-specific (20, 21). The precise identities of the receptors for each of the BoNTs have remained elusive. However, this area of research remains highly studied because the elucidation of these receptors may lead to novel therapeutics for the treatment of botulism. Receptor-Mediated Endocytosis. Most researchers presume that the process of BoNT receptor-mediated endocytosis is basically identical to that of most ligands that are internalized into cells (16). In fact, this may be the case; however, it has been noted that a retrieval phase of the vesicle recycling mechanism may also be plausible (22). Nerves that exocytose have a vigorous and well-developed mechanism for membrane retrieval (23); such a recycling mechanism could be a viable route by which the toxin can enter nerve cells. Synaptotagmin II, a protein that serves as a receptor for BoNT/B, shows that vesicles are involved in this process. Synaptotagmin has an exposed domain in the lumen of vesicles. Thus, in the cycle, exocytosis would place synaptotagmin on the exterior of the nerve cell for a brief time, wherein toxin binding to its receptor(s) would occur and both synaptotagmin and BoNT would be internalized during membrane retrieval. Labeled derivatives of synaptotagmin antibodies have been used to monitor www.acschemicalbiology.org

membrane retrieval and reformation of intraneuronal vesicles in order to support this theory (24). Translocation. It has been proposed that a pH-dependent structural rearrangement of the toxin inside an acidic compartment within the cell allows for toxin entry into the cytosol, a process common for several other bacterial toxins (25). Thus, the substrates for BoNTs are in the cytosol, and the LC protease must escape the endosome. Translocation is believed to take place wherein buried endosomal domains are exposed as the pH decreases (16). These domains then facilitate penetration of the lipid bilayer in a way that promotes translocation of the active region to the cytosol. This mechanism has been investigated via the pretreatment of neuromuscular junctions with chloroquine, a small molecule that can effectively and specifically raise the endosomal pH (26). This approach represents the first nonpeptidic approach for BoNT antagonism by preventing toxic escape from the endosome. The notion that BoNT is internalized by pH-induced translocaton is now widely accepted, but the exact nature of the membrane penetration remains unclear. Studies have been conducted that measured the change in resistance of artificial membranes as a function of the location of the toxin in an effort to clarify this mechanism (27). More recently, a new perspective has taken shape in which it has been proposed that the HC of BoNT can act both as a channel and as a chaperone (28). Substantiation of this hyposthesis is given by the fact that BoNT/A and -E form ion channels in phospholipid bilayers and PC12 cell membranes under conditions similar to those believed to exist in vivo. Inhibition of Neurotransmitter Release. SNARE proteins are involved in the fusion of synaptic vesicles with the plasma membrane; thus, the action of BoNT is to prevent exocytosis (29). At a more specific level, cleavage of SNARE proteins by BoNT inhibits the release of acetylcholine at the neuromuscular junction; this leads to inhibition of neurotransmission (15, 16). Cleavage of individual SNARE proteins does not prevent SNARE complex formation but results in a nonfunctional complex in which the coupling between Ca2⫹ influx and fusion is disrupted (30). The role of Ca2⫹ is fundamental to the process of BoNT-dependent inhibition of neurotransmitter release, because increasing the Ca2⫹ concentration in the synaptic terminal partially reverses the effect of BoNT/A (31). VOL.1 NO.6 • 359–369 • 2006

361

Potassium channel blockers

Translocation antagonists

NH2

OMe O O

HO NH2

NH4Cl

H OH

CH3NH2

O

O

NH2

O

O

OMe

N

Concanamycin A

N Br

O

OH

OH

3,4-DAP

N

OH

MeO

NH

H

OMe



O

O

O

H

O

H

O

HO

produce adequate antibody titers. Additionally, the CDC also distributes two HO OMe N OH H H OH equine antitoxins for O O Et O treatment of adult OH O O O O MeO H HO HO H O N MeO botulism. Although O HO OH equine antibodies CO H Neostigmine Bafilomycin A1 Monensin are broadly effective, they can cause Figure 2. Chemical structures of various compounds used to block BoNT intoxication by inhibiting either potassium adverse reactions, channels or toxin translocation out of endosomes. such as serum sickness and anaphylaxis (34). Clearly, a pharmacological Details defining the cleavage of the SNARE proteins on a molecular level (Figure 2) can be traced back to the intervention, especially one that would be effective after BoNT LC, which functions as a zinc-dependent metallo- BoNT internalization, is highly desirable. More recently, protease (32). A highly conserved segment of 20 amino the orphan drug Human Botulism Immune Globulin (BIG) has been developed as a therapeutic antitoxin that acids located within these LCs displays the common neutralizes BoNT. Notably, this is the first licensed zinc endopeptidase motif His-Glu-Xaa-Xaa-His. Each of product for the treatment of patients suffering from botuthe seven serotypes of BoNT cleaves one of the three lism. Although this antibody approach has proven effiSNARE proteins [synaptobrevin (vesicle-associated cacy in the treatment of infant botulism (35), antibodies membrane protein, VAMP), SNAP-25 (synaptosomalassociated protein of 25 kDa), and syntaxin], which are are not a small-molecule therapeutic approach and thus necessary for vesicle fusion and acetylcholine release. It are beyond the scope of this article (36, 37). Potassium Channel Blockers. Potassium channels has been firmly established that VAMP is the target for play crucial physiological roles in almost all types of BoNT/B, -D, -F, and -G. The target for BoNT/A and -E is cells in all organisms (38). In brief, potassium channels SNAP-25, whereas BoNT/C can cleave both SNAP-25 form a remarkably diverse group of ion channel strucand syntaxin. tures; the first type of potassium channel to be Therapeutic Strategies described was the classic voltage-activated channel for Treating BoNT IntoxicaKEYWORDS found in the squid. Other types of potassium channels tion. Because of their excepBotulinum neurotoxin (BoNT): One of the seven related proteins (A–G) secreted by Clostridium include hyperpolarizing voltages that are modulated by tional potency and ease of botulinum that lead to botulism. These toxins production, BoNTs are formi- intracellular metabolites and second messengers, tranare among the most toxic species known, and sient outward currents or A-currents, and large-conducdable biothreat agents. In certain preparations are commercially called Botox. The molecule is composed of two particular, BoNT/A is consid- tance Ca2⫹-sensitive, K⫹ channels, or BKCa channels, in chains, termed heavy and light chains. The ered the deadliest serotype, which neurons are characterized by calcium-activated heavy chain is critical in toxin recognition of and as such, we have potassium currents that contribute to re-polarization the target cellular surface and translocation out of endosomal vesicles into the cytoplasm, focused our discussion of and firing. BKCa channels exist as a complex of two difwhereas the light chain is a zinc-dependent therapeutic approaches on ferent subunits, the pore-forming ␣ subunit and a ␤ metalloprotease that cleaves specific SNARE this specific toxin. Currently, regulatory subunit. BKCa channels are generally proteins within the cell. Botulism: The bacterium C. botulinum causes no approved pharmacologibelieved to be tetraethylammonium (TEA)-sensitive, and this disease, which results in peripheral nerve cal treatments exist for BoNT they are therapeutic targets for BoNT poisoning. paralysis and can ultimately lead to death. intoxication. Although an The best-known K⫹ channel blocker for botulism poiThe causative agent is BoNT, a proteinaceous toxin. effective vaccine is available soning is 3,4-diaminopyridine (3,4-DAP) (Figure 2) (39). Cholinergic transmission: A mechanism of for immunoprophylaxis (33), 3,4-DAP is highly effective in antagonizing muscle neurotransmission that employs acetylcholine the development of protecparalysis after BoNT/A exposure in vitro and is the least as a neurotransmitter. Botulism interferes with this process by preventing the release of tion is slow, and multiple toxic of the currently available K⫹ blockers. After a rat neurotransmitter at synapses, thus leading to inoculations and annual diaphragm muscle was paralyzed by BoNT/A exposure, flaccid paralysis. boosters are required to 3,4-DAP induced a rapid and pronounced increase of Cl

TEA

N

Quinacrine

HO2C

H

H

H

Nigericin

HO

OH

2

362

VOL.1 NO.6 • 359–369 • 2006

DICKERSON AND JANDA

www.acschemicalbiology.org

REVIEW twitch tensions (39). 3,4-DAP continued to work, and compared with the control, little or no decrease was seen for ⱖ8 h after its addition. Combining in vivo and in vitro recording techniques, Adler and co-workers (40) studied the actions of 3,4-DAP in the rat extensor digitorum longus muscle after local inhibition of neuromuscular transmission by BoNT. The results showed that 3,4DAP markedly potentiated twitch tensions in BoNT/Aintoxicated muscle. The sensitivity of the extensor digitorum longus muscle to 3,4-DAP did not diminish with time or with repeated application. Because a major obstacle to the clinical use of 3,4-DAP is its brief duration, other investigations have attempted to deliver the compound via osmotic minipump infusion (41). It was also found that, for muscle function to be maintained, the drug must be delivered continuously for the entire period of BoNT intoxication. 3,4-DAP is an impressive compound for BoNT treatment; however, several problems have been associated with its use. First, its efficacy is primarily limited to BoNT/A. Second, while it seems to be very beneficial for increasing muscle strength, it only provides limited improvement for respiratory muscles, and no spontaneous ventilation has been seen. Third, it has toxicity issues, and seizures have been noted with its use, mainly from its penetration across the blood–brain barrier. To surmount some of these difficulties, researchers combined 3,4-DAP with neostigmine and TEA bromide (39). Researchers had hoped that combining 3,4-DAP and TEA, both of which are K⫹ channel blockers, would reduce the dose of 3,4-DAP needed and its toxic effects. Furthermore, this combination would have produced a greater Ca2⫹ influx than was attainable with either inhibitor alone. Accordingly, the increased Ca2⫹ entry should have resulted in an enhanced acetylcholine release. However, inhibition rather than enhancement was observed; this suggests that the postsynaptic inhibitory action of TEA on the nicotinic ion channel was sufficient to counteract any beneficial action of TEA on transmitter release. Although TEA was unsuccessful, this concept of “combinatorial” channel blockers is still believed to be valid for the treatment of BoNT poisoning. The second combination therapy of neostigmine and 3,4-DAP was considered viable, because neostigmine would increase the persistence of acetylcholine by inhibiting the activity of acetylcholinesterase in peripheral tissues. However, this combination was insufficient to restore neuromuscular transmission. www.acschemicalbiology.org

The efficacy of potassium channel blockers in antagonizing the action of BoNT/A is generally attributed to their ability to enhance the influx of Ca2⫹ as a result of inhibiting voltage-dependent K⫹ currents. Substances that can increase intracellular Ca2⫹ can partially overcome the paralysis due to BoNT poisoning. Continued efforts to optimize agents such as 3,4-DAP to increase their efficacy and reduce their toxicity are thus considered of significant interest. However, debate exists about the viability of this approach in the treatment of BoNT intoxication; for a potassium channel blocker to be effective, it must be administered over a period of weeks or months. Antagonists of Toxin Binding to Target Cells. The initial step in the mechanism of BoNT poisoning is the binding of the toxin to the cellular membranes of target neurons. Thus, blocking the interaction between BoNT and the cognate cellular receptor can inhibit nerve paralysis. Monoclonal antibodies have been investigated in this role and have been reviewed elsewhere (36, 37). Two approaches can be envisioned to accomplish the goal of antagonizing toxin–cell interactions: molecules that can coat the toxin with a small molecule and interfere with its ability to interact with a cell, and molecules that bind to the cellular receptor, thus blocking toxin binding. In the former case, polysiaylated gangliosides such as GT1b were observed ⬎30 years ago to be potential receptors for BoNT/A (42, 43) and more recently were shown to inhibit BoNT/A binding to synaptosomes (44) and to quench BoNT/A fluorescence (45). In the instance of molecules that compete with BoNT for the cellular binding site, lectins from two sources, one derived from animals and the other from plants (Limax flavus and Triticum vulgaris, respectively, both of which possess affinity for sialic acid), could serve as competitive antagonists of all BoNT serotypes as well as tetanus toxins (46). In the context of BoNT/A, the amount of time necessary to cause neuromuscular paralysis in mouse phrenic-nerve hemidiaphragm preparations was increased from 78 to 128 min. Importantly, this study was the first to report on a small molecule that could potentiate the activity of all BoNT serotypes. Antagonists of pH-Dependent BoNT Translocation. The lethal effects of BoNT involve the inhibition of synaptic transmission at the skeletal neuromuscular junction. Despite variations in the SNARE protein targets of different BoNT serotypes, a common point in the action of each of these toxins is the encapsulation of the holoVOL.1 NO.6 • 359–369 • 2006

363

Another possible therapy for the inhibition of BoNT neurotoxicity is to limit the metallopeptidase activity of the BoNT light chains.

toxin in an endocytotic vesicle that then undergoes acidification. This acidification is needed for BoNT to induce muscle failure; agents that inhibit acidification delay the onset of the paralysis in vitro and lengthen the time the toxin is susceptible to neutralization with antisera (47). Thus, endocytotic vesicle acidification is a logical point at which to inhibit the activity of BoNT serotypes. Simpson (48) has reported that ammonium chloride and methylamine produce concentration- and timedependent antagonism of the onset of neuromuscular blockade caused by BoNT/A–C. These amines exerted their effects only when they were added before the toxin was introduced or ⬃10–20 min later. At concentrations that produce antagonism of BoNT-induced paralysis onset, these amines did not inactivate toxin molecules nor did they produce irreversible changes in tissue function. Presumably, the amines did not inhibit BoNT from binding to its receptor(s) and did not reverse neuromuscular blockade; rather, they acted solely to antagonize internalization of the toxins. Acidification of endocytotic vesicles depends on a vesicular H⫹-ATPase that acts as a proton pump to accumulate protons from the cytoplasm into the lumen of a vesicle. In 1994, Simpson (49) reported that bafilomycin A1, an inhibitor of this ATPase, is a universal antagKEYWORDS Clostridium botulinum: A rod-shaped, sporeonist of BoNTs (Figure 2). forming bacterium that is widely distributed in This compound produced a the environment. The neurotoxin secreted by concentration-dependent this bacterium, BoNT, is among the most toxic species known. blockage of neuromuscular Ganglioside: A molecule that is composed of a transmission without affectlipid domain and a carbohydrate domain that ing nerve- or muscle-action contains one or more sialic acid moieties. These compounds are embedded within the potentials. Application of plasma membrane of numerous cell types proton ionophores can also and are critically important in signal transdeplete this pH gradient duction events. Lectin: Receptor proteins that specifically bind to without affecting ATP hydrocarbohydrates. Lectins are found in both lysis. Sheridan (50) preplants and animals and have diverse roles in sented evidence that two each. They play a large role in the normal function of the immune system. monocarboxylate polyether SNARE protein: Acronym for “soluble Nionophores, nigericin and ethylmaleimide sensitive factor attachment monensin (Figure 2), increase receptor”. These proteins make up a large superfamily whose primary function is to membrane permeability to mediate vesicle fusion in mammalian and H⫹ and K⫹, or H⫹, Na⫹, and yeast cells. K⫹, respectively, and block Synaptosome: The isolated synapse of a neuron purified by homogenization of nerve tissue endosomal acidification by commonly used in assays addressing BoNT acting as proton shunts to paralysis. neutralize pH gradients. 364

VOL.1 NO.6 • 359–369 • 2006

DICKERSON AND JANDA

Nanomolar concentrations of both compounds were able to block BoNT effects at neuromuscular junctions up to 3-fold times more than unprotected muscles. Unfortunately, higher concentrations of these antibiotics also blocked synapses. More recently, concanamycin A (Con A) has also been examined as an inhibitor of endosomal acidification (Figure 2) (51). Interestingly, Con A prevented SNAP-25 cleavage in pretreated cultures or those treated up to 15 min after toxin exposure, whereas the addition of the compound 40 min later was not protective. Groups have also examined several clinically used antimalarial drugs (aminoquinolines) for their effectiveness in antagonizing BoNT/A-induced neuromuscular blockade (52). These studies concluded that these compounds may produce their protective efficacy through the blockade of endosomal acidification. Alternatively, it has been suggested that, because these drugs block channel formation, they may be a key element in LC release from the endosome. Lastly, Adler and co-workers (53) have investigated drug combinations for additive or synergistic effects. Here, quinacrine and the metal chelator N,N,N=,N=-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN), when added together, protected up to 100% more than either drug alone. Inhibition of the BoNT Metalloprotease LC. As detailed earlier, BoNT impairs neuronal exocytosis through specific proteolysis of the SNARE proteins. SNARE assembly into a low-energy ternary complex is thought to be intimately involved with membrane fusion and thus neurotransmitter release (54). The site-specific SNARE hydrolysis is catalyzed by the BoNT LC, which comprises a unique group of endopeptidases. Thus, another possible therapy for the inhibition of BoNT neurotoxicity is to limit the metallopeptidase activity of the BoNT LC. Current research efforts to identify BoNT protease inhibitors have concentrated only on BoNT/A and -B, primarily because of their potent toxicity and extended duration. Two approaches can be envisioned for inhibitor development, peptide sequences/peptidomimetics based upon the native SNARE protein substrate or small organic molecules that specifically bind to the toxin and inactivate it. Several effective inhibitors based upon peptide scaffolds have been reported (55–57); however, these molecules are unlikely to become leads for new pharmaceuticals because of their short in vivo lifetimes. Initial efforts in this area focused on the use of www.acschemicalbiology.org

REVIEW common pharmacophore scaffolds; however, these initial leads have yet to HN OH yield any new compounds of greater HN NH OH potency. An in silico screen of 2.5 million NH compounds has also been conducted HO Cl N HN recently in an effort to isolate BoNT/AO OH selective inhibitors. After extensive structure-guided modification of the Cl N OH O selected scaffolds, an inhibitor with a Ki of Q2-15 Michellamine B 12 ␮M was found (61). This inhibitor was (62% inhibition at 20 µM) (60% inhibition at 20 µM) designed to display competitive kinetics O OH by chelating the active site zinc atom NH NH through a hydroxamic acid moiety. HN NH2 The success of this study validates the N S use of the cationic dummy atom approach H N O OH H2N (CaDA) to develop BoNT inhibitors by molecO ular dynamics simulations. However, a NH2 key conclusion in that article, that effective CaDA-identified inhibitor L-Arginine hydroxamic acid (Ki = 12 µM) (Ki = 60 µM) BoNT inhibition requires a length of 10 atoms, is not supported by our own studies on BoNT metalloprotease inhibitors. O Recently, we have initiated a multifacOH Cl N S eted research program aimed at identifyH O ing novel small-molecule inhibitors of the CO2H O N Cl H BoNT/A LC metalloprotease. Given the presence of a critical zinc ion in the LC 2,4-dichlorocinnamic hydroxamic acid Fmoc-D-Cys(Trt)-OH (Ki = 0.30 ± 0.01 µM) (Ki = 18 ± 2 µM) active site, we speculated that the hydroxamate zinc-binding functionality, when Figure 3. BoNT/A LC metalloprotease inhibitors. coupled to a suitable scaffold to impart heavy-metal chelators that possessed cell permeability specificity, would provide potent BoNT/A inhibitors. Our as potential therapeutic targets. In particular, TPEN has initial studies reported that arginine hydroxamic acid, a been extensively studied for its ability to antagonize the single modified amino acid, could modestly inhibit catalytic action of BoNT in mouse phrenic-nerve hemidi- BoNT/A in a high-throughput FRET-based assay developed for the screening of compound libraries (Ki ⫽ aphragm preparations and in a mouse model (42, 58, 59). However, TPEN does not possess any inherent 60 ␮M) (62). The native SNAP-25 cleavage site is specificity for BoNT over other metalloproteins; this may between Gln197 and Arg198; thus, inhibition by arginpreclude its use in a clinical setting because of various ine derivatives was not entirely unexpected. However, it side effects that could result from chelation of other criti- was surprising that little difference in inhibition effical metals. ciency was observed between the D- and L-isomers of Other studies targeted toward the identification of arginine hydroxamic acid. low-molecular-weight, small-molecule inhibitors of In a further study, the in situ preparation of a library of BoNT/A include screening of the National Cancer Instihydroxamic acids has been described (63). Here, 150 tute Diversity Set and a series of 4-aminoquinolines that commercially available carboxylic acids were converted were originally identified to prolong the time required for to hydroxamic acids in a facile two-step procedure to BoNT/A to block neuromuscular transmission (Figure 3) allow for rapid lead identification. On the basis of the (60). Interestingly, several compounds possessing initial screen, 4-chlorocinnamic hydroxamate displayed ⬎50% inhibition (at 20-␮M concentration) were identi- an IC50 of 15 ␮M and was considered a promising lead fied. From these lead “hits”, modeling studies predicted for further development. A small series of compounds OH

www.acschemicalbiology.org

VOL.1 NO.6 • 359–369 • 2006

365

were synthesized to explore the structure–activity relationships of this lead. Amazingly, substitutions of the chloro substituent were not tolerated, whereas a 2,4dichloro-substituted compound was found to be the most potent nonpeptidic inhibitor of BoNT to date (Ki ⫽ 0.30 ⫾ 0.01 ␮M) (Figure 3). Not surprisingly, in light of the hydroxamate zinc-binding functionality, this compound was found to be a competitive inhibitor of BoNT/A. It is important to note that the inhibitors in this study contradict the previously articulated hypothesis that in order to be an effective inhibitor of BoNT, a small molecule must have a length of ⱖ10 carbon atoms (61). Before any BoNT/A LC inhibitor can advance to animal models of BoNT intoxication, efficacy in cellular models must be shown. We recently reported the first demonstration of a BoNT/A LC inhibitor that shows protection from SNAP-25 cleavage in a cellular model (64). Quite surprisingly, this compound is simply a protected amino acid used in peptide synthesis, Fmoc-D-Cys(Trt)OH (Figure 3). Kinetic analysis of this compound revealed that it also competitively inhibits BoNT/A (Ki ⫽ 18 ␮M). Computational docking studies found that the predicted binding constant for the inhibitor (10 ␮M) was in close agreement with the experimentally determined value. Structurally, this model also revealed that a significant amount of binding energy can be attributed to burying the Fmoc group in a hydrophobic pocket, while the carboxylic acid moiety was positioned in close proximity to several positively charged residues. Interestingly, while competitive inhibition was observed, the docking model predicted no interaction between the active site zinc-binding residues and the inhibitor. In light of the potency of this compound combined with its ready availability, it was next tested in a cellular model of BoNT intoxication. At a concentration of 30 ␮M, almost total protection of SNAP-25 cleavage was observed in Neuro-2a cells, whereas complete protection was seen when the compound was added to cells at a concentration of 60 ␮M. The discovery of potent LC protease inhibitors of the BoNTs could be a crucial step in rescuing nerve activity after toxin internalization. However, to date, research has yet to advance any nonpeptide molecules from enzymatic assays into the corresponding cellular and animal models. Therapeutics with an Undefined Mechanism of Action. Theaflavins are unique active ingredients produced when green tea ferments into black or oolong tea. Nishimura and co-workers (65, 66) have reported how 366

VOL.1 NO.6 • 359–369 • 2006

DICKERSON AND JANDA

O

O

AcO Me

O OH O AcO HO

Me

Me

Me

O OH

H Me

Me

H H Me Me

H Toosendanin

Limonoid OH

O Me

HO

O OH

Me H

OH

OHC

HO Me

H

O

H

H

O Cortisol

Aldosterone

Figure 4. Structure of the limonoid toosendanin. For comparison, the structures of the general limonoid skeleton, as well as representative common steroids, are also shown.

the thearubigin fraction, complex phenols formed during fermentation by polymerization of theaflavins, of Camellia sinensis extract protected against the effects of BoNT/A, -B, and -E in mouse phrenic-nerve diaphragm assays. In a series of studies aimed at elucidating the mechanism of action of thearubigin, the binding of 125Ilabeled BoNT/A, -B, or -E to rat cerebrocortical synaptosomes was inhibited by these polyphenols. The authors suggest that the thearubigin fraction attenuated the effects of BoNT via simple binding to the toxin. Although which molecule or molecules in the thearubigin fraction are responsible for the loss of toxin activity is unknown, such fractions are made up of a plethora of polyphenols, and multiple compounds are likely involved. Limonoids are tetranortriterpenoids with a 4,4,8trimethyl-17-furanylsteroid skeleton derived from euphane or tirucallane triterpenoids, and they generally have an intense, bitter taste (67). Limonoids from the Melia toosendan tree have been found to be effective anthelmintics (68); a major limonoid constituent in M. toosendan is the compound toosendanin (Figure 4), which appears to have multiple modes of action in insects (69). Interest in the application of limonoid natural products in pest management remains high. Although this area of research remains fertile, a series of reports detailing the activity of toosendanin on neurotransmitter release in motor-nerve terminals have emerged over the www.acschemicalbiology.org

REVIEW past 2 decades. In total, these studies have indicated that toosendanin is a selective presynaptic blocker that inhibits quantal release of acetylcholine (70). In contrast with BoNT, the blocking effect of toosendanin has always been preceded by a facilitator phase that depends on the presence of Ca2⫹ in the medium. It is noteworthy that, during the facilitatory period, the tolerance of the neuromuscular junction toward BoNT was enhanced significantly (71). Further studies investigating the mechanism of toosendanin provided evidence that it causes a decrease in, and ultimately the disappearance of, Ca2⫹ sensitivity, thus, inhibiting the voltage-dependent potassium current in neuroblastoma glioma cells (72). Toosendanin was also shown to induce submicroscopic changes in the neuromuscular junction, namely, a reduction in the number of synaptic vesicles and an increase in the width of synaptic cleft (73). Furthermore, toosendanin inhibits the delayed rectifier potassium channel by intracellular and/or extracellular addition (74). Toosendanin is membrane-permeable; thus, its modulation of ion channels may be due to its ability to span membrane bilayers and alter channel integrity. The relevance of this finding is that BoNT is capable of forming ion channels in artificial bilayers and in PC12 cell membranes (75). In addition, treatment of synaptosomes with toosendanin imparted resistance to BoNT/ A-mediated proteolytic cleavage of SNAP-25. This protective effect did not result from inhibition of the endopeptidase activity; thus, interference must result from disruption of the BoNT LC before it becomes a functional protease (76). Along this line of investigation, BoNT

translocation and channel formation are known to correlate (16). Taken together, the effects of toosendanin on BoNT might be achieved by interfering with the LC translocation. However, the precise mechanism of BoNT antagonism by toosendanin remains unclear. Conclusion and Outlook. BoNTs are one of the deadliest agents of bioterrorism. Consequently, specific pharmaceutical agents are urgently needed to treat BoNT intoxication. Although an investigational pentavalent toxoid and a horse polyclonal serum are currently available from the CDC and a recombinant vaccine is under development, post-exposure vaccination is virtually useless because of the rapid onset of the toxin. A drug for the prevention or treatment of botulism would be exciting, yet no small molecule has advanced to even phase I clinical trials. Future work in this area will likely be driven by several factors, including the development of robust high-throughput assays for the rapid identification of compound efficacy in cellular models and the complete elucidation of the molecular mechanism of BoNT intoxication. Furthermore, by analyzing natural products such as toosendanin that protect against BoNT-induced paralysis, researchers may also uncover novel molecular scaffolds for BoNT inhibitors. In total, this area of research epitomizes the importance of conducting studies at the interface of chemistry and biology; only by having a firm understanding of the chemical and biological processes at play can an efficient route for the development of new therapeutics be charted.

REFERENCES

7. Christopher, G. W., Cieslak, T. J., Pavlin, J. A., and Eitzen, E. M., Jr., (1997) Biological warfare. A historical perspective, JAMA, J. Am. Med. Assoc. 278, 412–417. 8. Zilinskas, R. A. (1997) Iraq’s biological weapons. The past as future? JAMA, J. Am. Med. Assoc. 278, 418–424. 9. Montecucco, C., and Molgo, J. (2005) Botulinal neurotoxins: revival of an old killer, Curr. Opin. Pharmacol. 5, 274–279. 10. Johnson, E. A. (1999) Clostridial toxins as therapeutic agents: benefits of nature’s most toxic proteins, Annu. Rev. Microbiol. 53, 551–575. 11. Shukla, H. D., and Sharma, S. K. (2005) Clostridium botulinum: a bug with beauty and weapon, Crit. Rev. Microbiol. 31, 11–18. 12. Munchau, A., and Bhatia, K. P. (2000) Uses of botulinum toxin injection in medicine today, Br. Med. J. 320, 161–165. 13. Kobayashi, R., Kohda, T., Kataoka, K., Ihara, H., Kozaki, S., Pascual, D. W., Staats, H. F., Kiyono, H., McGhee, J. R., and Fujihashi, K. (2005) A novel neurotoxoid vaccine prevents mucosal botulism, J. Immunol. 174, 2190–2195.

1. van Ermengem, E. (1897) Uber einen neuen anaeroben Bacillus und seine Beziehungen zum Botulismus, Z. Hyg. Infekt. 26, 1–56. 2. Franciosa, G., Aureli, P., and Schechter, R. (2003) Clostridium botulinum, Food Sci. Technol. 125, 61–90. 3. Johnson, E. A., and Bradshaw, M. (2001) Clostridium botulinum and its neurotoxins: a metabolic and cellular perspective, Toxicon 39, 1703–1722. 4. Schantz, E. J., and Johnson, E. A. (1992) Properties and use of botulinum toxin and other microbial neurotoxins in medicine, Microbiol. Rev. 56, 80–99. 5. Turton, K., Chaddock, J. A., and Acharya, K. R. (2002) Botulinum and tetanus neurotoxins: structure, function and therapeutic utility, Trends Biochem. Sci. 27, 552–558. 6. Hatheway, C. L, and Johnson, E. A. (1998) Clostridium. the sporebearing anaerobes, in Topley and Wilson’s Microbiology and Microbial Infections, Vol. 2. Systematic Bacteriology, (Balows, A., and Duerden, B., Eds.) pp 731–782, Arnold Publishers: London.

www.acschemicalbiology.org

Acknowledgment: This work has been supported by the Skaggs Institute of Chemical Biology and the National Institutes of Health (AI066507 and BT010-04).

VOL.1 NO.6 • 359–369 • 2006

367

14. Oguma, K., Fujinaga, Y., and Inoue, K. (1995) Structure and function of Clostrium botulinum toxins, Microbiol. Immunol. 39, 161–168. 15. Schiavo, G., Matteoli, M., and Montecucco, C. (2000) Neurotoxins affecting neuroexocytosis, Physiol. Rev. 80, 717–766. 16. Simpson, L. L. (2004) Identification of the major steps in Botulinum toxin action, Annu. Rev. Pharmacol. Toxicol. 44, 167–193. 17. Dong, M., Yeh, F., Tepp, W. H., Dean, C., Johnson, E. A., Janz, R., and Chapman, E. R. (2006) SV2 is the protein receptor for botulinum neurotoxin A, Science 312, 592–596. 18. Mahrhold, S., Rummel, A., Bigalke, H., Davletov, B., and Binz, T. (2006) The synaptic vesicle protein 2C mediates the uptake of botulinum neurotoxin A into phrenic nerves, FEBS Lett. 580, 2011–2014. 19. Montecucco, C. (1986) How do tetanus and botulinum toxins bind to neuronal membranes? Trends Biochem. Sci. 11, 314–317. 20. Black, J. D., and Dolly, J. O. (1986) Interaction of 125I-labeled botulinum neurotoxins with nerve terminals. I. Ultrastructural autoradiographic localization and quantitation of distinct membrane acceptors for types A and B on motor nerves, J. Cell Biol. 103, 521–534. 21. Black J. D., and Dolly, J. O. (1986) Interaction of 125I-labeled botulinum neurotoxins with nerve terminals. II. Autoradiographic evidence for its uptake into motor nerves by acceptor-mediated endocytosis, J. Cell Biol. 103, 535–544. 22. Matthews, G. (1996) Neurotransmitter release, Annu. Rev. Neurosci. 19, 219–233. 23. Betz, W. J., and Angleson, J. K. (1998) The synaptic vesicle cycle, Annu. Rev. Physiol. 60, 347–363. 24. Matteoli, M., Takei, K., Perin, M. S., Sudhof, T. C., and De Camilli, P. (1992) Exo-endocytotic recycling of synaptic vesicles in developing processes of cultured hippocampal neurons, J. Cell Biol. 117, 849–861. 25. Sandvig, K. (2003) Transport of toxins across intracellular membranes, in Bacterial Protein Toxins, (Burns, D. L., Barbieri, J. T., and Iglewski, B. H., Eds.) pp 157–172, American Society for Microbiology, Washington, DC. 26. Simpson, L. L. (1982) The interactions between aminoquinoline and presynaptically acting neurotoxins, J. Pharmacol. Exp. Ther. 222, 43–48. 27. Hoch, D. H., Romero-Mira, M., Ehrlich, B. E., Finkelstein, A., DasGupta, B. R., and Simpson, L. L. (1985) Channels formed by botulinum, tetanus, and diphtheria toxins in planar lipid bilayers: relevance to translocation of proteins across membranes, Proc. Natl. Acad. Sci. U.S.A. 82, 1692–1696. 28. Koriazova, L. K., and Montal, M. (2003) Translocation of botulinum neurotoxin light chain protease through the heavy chain channel, Nat. Struct. Biol. 10, 13–18. 29. Bajjalieh, S. M. (1999) Synaptic vesicle docking and fusion, Curr. Opin. Neurobiol. 9, 321–328. 30. Humeau, Y., Doussau, F., Grant, N. J., and Poulain, B. (2000) How botulinum and tetanus neurotoxins block neurotransmitter release, Biochemie 82, 427–446. 31. Meunier, F. A., Schiavo, G., and Molgo, J. (2002) Botulinum neurotoxins: from paralysis to recovery of functional neuromuscular transmission, J. Physiol. (Paris) 96, 105–113. 32. Schiavo, G., Benfenati, F., Poulain, B., Rossetto, O., Polverino de Laureto, P., DasGupta, B. R., and Montecucco, C. (1992) Tetanus and botulinum-B neurotoxins block neurotransmitter release by proteolytic cleavage of synaptobrevin, Nature 359, 832–835. 33. Arnon, S. S., Schechter, R., Inglesby, T. V., Henderson, D. A., Bartlett, J. G., Ascher, M. S., Eitzen, E., Fine, A. D., Hauer, J., Layton, M., Lillibridge, S., Osterholm, M. T., O’Toole, T., Parker, G., Perl, T. M., Russell, P. K., Swerdlow, D. L., and Tonat, K. (2001) Botulinum toxin as a biological weapon: medical and public health management, JAMA, J. Am. Med. Assoc. 285, 1059–1070. 34. Black, R. E., and Gunn, R. A. (1980) Hypersensitivity reactions associated with botulinal antitoxin, Am. J. Med. 69, 567–570.

368

VOL.1 NO.6 • 359–369 • 2006

DICKERSON AND JANDA

35. Arnon, S. S., Schechter, R., Maslanka, S. E., Jewell, N. P., and Hatheway, C. L. (2006) Human botulism immune globulin for the treatment of infant botulism, N. Engl. J. Med. 354, 462–471. 36. Marks, J. D. (2004) Deciphering antibody properties that lead to potent botulinum neurotoxin neutralization, Mov. Disord. 19, S101–S108. 37. Byrne, M. P., and Smith, L. A. (2000) Development of vaccines for the prevention of botulism, Biochemie 82, 955–966. 38. Rudy, B. (1988) Diversity and ubiquity of K channels, Neuroscience 25, 729–749. 39. Adler, M., Scovill, J., Parker, G., Lebeda, F. J., Piotrowski, J., and Deshpande, S. S. (1995) Antagonism of botulinum toxin-induced muscle weakness by 3,4-diaminopyridine in rat phrenic nerve hemidiaphragm preparations, Toxicon 33, 527–537. 40. Adler, M., Macdonald, D. A., Sellin, L. C., and Parker, G. W. (1996) Effect of 3,4-diaminopyridine on rat extensor digitorum longus muscle paralyzed by local injection of botulinum neurotoxin, Toxicon 34, 237–249. 41. Adler, M., Capacio, B., and Deshpande, S. S. (2000) Antagonism of botulinum toxin A-mediated muscle paralysis by 3,4-diaminopyridine delivered via osmotic minipumps, Toxicon 38, 1381–1388. 42. Simpson, L. L., and Rapport, M. M. (1971) The binding of botulinum toxin to membrane lipids: sphingolipids, steroids and fatty acids, J. Neurochem. 18, 1751–1759. 43. Simpson, L. L., and Rapport, M. M. (1971) Ganglioside inactivation of botulinum toxin, J. Neurochem. 18, 1341–1343. 44. Kitamura, M., Iwamori, M., and Nagai, Y. (1980) Interaction between Clostridium botulinum neurotoxin and gangliosides, Biochim. Biophys. Acta 628, 328–335. 45. Kamata, Y., Yoshimoto, M., and Kozaki, S. (1997) Interaction between botulinum neurotoxin type A and ganglioside: ganglioside inactivates the neurotoxin and quenches its tryptophan fluorescence, Toxicon 35, 1337–1340. 46. Bakry, N., Kamata, Y., and Simpson, L. L. (1991) Lectins from Triticum vulgaris andare universal antagonists of botulinum neurotoxin and tetanus toxin, J. Pharmacol. Exp. Ther. 258, 830–836. 47. Simpson, L. L. (1982) The interaction between aminoquinolines and presynaptically acting neurotoxins, J. Pharmacol. Exp. Ther. 222, 43–48. 48. Simpson, L. L. (1983) Ammonium chloride and methylamine hydrochloride antagonize clostridial neurotoxins, J. Pharmacol. Exp. Ther. 225, 546–552. 49. Simpson, L. L., Coffield, J. A., and Bakry, N. (1994) Inhibition of vacuolar adenosine triphosphatase antagonizes the effects of clostridial neurotoxins but not phospholipase A2 neurotoxins, J. Pharmacol. Exp. Ther. 269, 256–262 50. Sheridan, R. E. (1996) Protonophore antagonism of botulinum toxin in mouse muscle, Toxicon 34, 849–855. 51. Keller, J. E., Cai, F., and Neale, E. A. (2004) Uptake of botulinum neurotoxin into cultured neurons, Biochemistry 43, 526–532. 52. Deshpande, S. S., Sheridan, R. E., and Adler, M. (1997) Efficacy of certain quinolines as pharmacological antagonists in botulinum neurotoxin poisoning, Toxicon 35, 433–445. 53. Adler, M., Dineterman, R. E., and Wannemacher, R. W. (1997) Protection by the heavy metal chelator N,N,N=,N=-tetrakis(2pyridylmethyl)ethylenediamine (TPEN) against the lethal action of botulinum neurotoxin A and B, Toxicon 35, 1089–1100. 54. Sollner, T., Whiteheart, S. W., Brunner, M., Erdjument-Bromage, H., Geromanos, S., Tempst, P., and Rothman, J. E. (1993) SNAP receptors implicated in vesicle targeting and fusion, Nature 362, 318–324. 55. Schmidt, J. J., and Stafford, R. G. (2002) A high-affinity competitive inhibitor of type A botulinum neurotoxin protease activity, FEBS Lett. 532, 423–426.

www.acschemicalbiology.org

REVIEW 56. Sukonpan, C., Oost, T., Goodnough, M., Tepp, W., Johnson, E. A., and Rich, D. H. (2004) Synthesis of substrates and inhibitors of botulinum neurotoxin type A metalloprotease, J. Pept. Res. 63, 181–193. 57. Yiadom, K. P. A. B., Muhie, S., and Yang, D. C. H. (2005) Peptide inhibitors of botulinum neurotoxin by mRNA display, Biochem. Biophys. Res. Commun. 335, 1247–1253. 58. Simpson, L. L., Coffield, J. A., and Bakry, N. (1993) Chelation of zinc antagonizes the neuromuscular blocking properties of the seven serotypes of botulinum neurotoxin as well as tetanus toxin, J. Pharmacol. Exp. Ther. 267, 720–727. 59. Sheridan, R. E., and Deshpande, S. S. (1995) Interactions between heavy metal chelators and botulinum neurotoxins at the mouse neuromuscular junction, Toxicon 33, 539–549. 60. Burnett, J. C., Schmidt, J. J., Stafford, R. G., Panchal, R. G., Nguyen, T. L., Hermone, A. R., Vennerstrom, J. L., McGrath, C. F., Lane, D. J., Sausville, E. A., Zaharevitz, D. W., Gussio, R., and Bavari, S. (2003) Novel small molecule inhibitors of botulinum neurotoxin A metalloprotease activity, Biochem. Biophys. Res. Commun. 310, 84–93. 61. Park, J. G., Sill, P. C., Makiyi, E. F., Garcia-Sosa, A. T., Millard, C. B., Schmidt, J. J., and Pang, Y.-P. (2006) Serotype-selective, smallmolecule inhibitors of the zinc endopeptidase of botulinum neurotoxin serotype A, Bioorg. Med. Chem. 14, 395–408. 62. Boldt, G. E., Kennedy, J. P., Hixon, M. S., McAllister, L. A., Barbieri, J. T., Tzipori, S., and Janda, K. D. (2006) Synthesis, characterization and development of a high-throughput methodology for the discovery of botulinum neurotoxin A inhibitors, J. Comb. Chem., published online May 19, http://dx.doi.org/10.1021/cc060010h. 63. Boldt, G. E., Kennedy, J. P., and Janda, K. D. (2006) Identification of a potent botulinum neurotoxin A protease inhibitor using in situ lead identification chemistry, Org. Lett. 8, 1729–1732. 64. Boldt, G. E., Eubanks, L. M., and Janda, K. D. (2006) Identification of a botulinum neurotoxin A protease inhibitor displaying efficacy in a cellular model, Chem. Commun., published online May 3, http://dx.doi.org/10.1039/b603099h. 65. Satoh, E., Ishii, T., Shimizu, Y., Sawamura, S., and Nishimura, M. (2001) Black tea extract, thearubigin fraction, counteract the effects of botulinum neurotoxins in mice, Br. J. Pharmacol. 132, 797–798. 66. Satoh, E., Ishii, T., Shimizu, Y., Sawamura, S.-L., and Nishimura, M. (2002) The mechanism underlying the protective effect of the thearubigin fraction of black tea (Camellia sinensis) extract against the neuromuscular blocking action of botulinum neurotoxins, Pharmacol. Toxicol. 90, 199–202. 67. Nakatani, M. (1999) Limonoids from Melia toosendan (Meliaceae) and their antifeedant activity, Heterocycles 50, 595–609. 68. Xie, Y. S., Fields, P. G., Isman, M. B., Chen, W. K., and Zhang, X. (1995) Insecticidal activity of Melia toosendan extracts and toosendanin against three stored-product insects, J. Stored Prod. Res. 31, 259–265. 69. Zhou, J.-B., Minami, Y., Yagi, F., Tadera, K., and Nakatani, M. (1997) Antifeeding limonoids from Melia toosendan, Heterocycles 45, 1781–1786. 70. Shih, Y.-L., Wei, N., Yang, Y., and Wang, Z. (1980) Toosendanina presynaptic neuromuscular blocking agent, Acta Physiologica Sin. 32, 293–297. 71. Shih, Y.-L., and Hsu, K. (1983) Anti-botulismic effect of toosendanin and its facilitatory action on miniature end-plate potentials, Jpn. J. Physiol. 33, 677–680. 72. Li, M.-F., Wu, Y., Wang, Z.-F., and Shi, Y.-L. (2004) Toosendanin, a triterpenoid derivative, increases Ca2⫹ current in NG108-15 cells via L-type channels, Neurosci. Res. 49, 197–203. 73. Wang, Z.-F., and Shi, Y.-L. (2001) Inhibition of large-conductance Ca2⫹-activated K⫹ channels in hippocampal neurons by toosendanin, Neuroscience 104, 41–47.

www.acschemicalbiology.org

74. Wang, Z.-F., and Shi, Y.-L. (2001) Modulation of inward rectifier potassium channel by toosendanin, a presynaptic blocker, Neurosci. Res. 40, 211–215. 75. Tang, M.-Z., Wang, Z.-F., and Shi, Y.-L. (2003) Toosendanin induces outgrowth of neuronal processes and apoptosis in PC12 cells, Neurosci. Res. 45, 225–231. 76. Wang, Z.-F., Zhou, J.-Y., Ren, X.-M., Tang, M.-Z., and Shi, Y.-L. (2003) Antagonism of botulinum toxin type A-induced cleavage of SNAP25 in rat cerebral synaptosome by toosendanin, FEBS Lett. 555, 375–379.

VOL.1 NO.6 • 359–369 • 2006

369