Effective countermeasure against poisoning byorganophosphorus insecticides and nerve agentsEdson X. Albuquerque*†, Edna F. R. Pereira*, Yasco Aracava*, William P. Fawcett*, Maristela Oliveira*,William R. Randall*, Tracey A. Hamilton‡, Robert K. Kan‡, James A. Romano, Jr.§, and Michael Adler¶

*Department of Pharmacology and Experimental Therapeutics, University of Maryland School of Medicine, 655 West Baltimore Street, Baltimore, MD 21201;‡Comparative Pathology Branch and ¶Neurobehavioral Toxicology Branch, U.S. Army Medical Research Institute of Chemical Defense, Aberdeen ProvingGround, MD 21010; and §U.S. Army Medical Research and Materiel Command, Fort Detrick, MD 21702

Communicated by John W. Daly, National Institutes of Health, Bethesda, MD, June 28, 2006 (received for review May 24, 2006)

The nerve agents soman, sarin, VX, and tabun are deadly organ-ophosphorus (OP) compounds chemically related to OP insecti-cides. Most of their acute toxicity results from the irreversibleinhibition of acetylcholinesterase (AChE), the enzyme that inacti-vates the neurotransmitter acetylcholine. The limitations of avail-able therapies against OP poisoning are well recognized, and moreeffective antidotes are needed. Here, we demonstrate that galan-tamine, a reversible and centrally acting AChE inhibitor approvedfor treatment of mild to moderate Alzheimer’s disease, protectsguinea pigs from the acute toxicity of lethal doses of the nerveagents soman and sarin, and of paraoxon, the active metabolite ofthe insecticide parathion. In combination with atropine, a singledose of galantamine administered before or soon after acuteexposure to lethal doses of soman, sarin, or paraoxon effectivelyand safely counteracted their toxicity. Doses of galantamineneeded to protect guinea pigs fully against the lethality of OPswere well tolerated. In preventing the lethality of nerve agents,galantamine was far more effective than pyridostigmine, a periph-erally acting AChE inhibitor, and it was less toxic than huperzine,a centrally acting AChE inhibitor. Thus, a galantamine-based ther-apy emerges as an effective and safe countermeasure against OPpoisoning.

galantamine � guinea pig � pyridostigmine � soman � sarin

The organophosphorus (OP) compounds soman, sarin, VX,and tabun, referred to as nerve agents, are among the most

lethal chemical weapons ever developed (1). Some of them wereused with catastrophic results in wars and also in terrorist attacksin Japan in the 1990s (2). The majority of insecticides are alsoOPs, and intoxication with these compounds represents a majorpublic-health concern worldwide (3, 4). The possibility of furtherterrorist attacks with nerve agents and the escalating use of OPinsecticides underscore the urgent need to develop effective andsafe antidotes against OP poisoning.

The acute toxicity of OPs results primarily from their actionas irreversible inhibitors of acetylcholinesterase (AChE) (5). Inthe periphery, acetylcholine accumulation leads to persistentmuscarinic receptor stimulation that triggers a syndrome whosesymptoms include miosis, profuse secretions, bradycardia, bron-choconstriction, hypotension, and diarrhea. It also leads tooverstimulation followed by desensitization of nicotinic recep-tors, causing severe skeletal muscle fasciculations and subse-quent weakness. Central nervous system-related effects includeanxiety, restlessness, confusion, ataxia, tremors, seizures, car-diorespiratory paralysis, and coma.

Current therapeutic strategies to decrease OP toxicity includeatropine to reduce the muscarinic syndrome, oximes to reacti-vate OP-inhibited AChE, and benzodiazepines to control OP-triggered seizures (5). The limitations of these treatments arewell recognized (4), and alternative therapies have been sought.Among these therapies are phosphotriesterases and butyrylcho-linesterase (BuChE), enzymes that act as OP scavengers (6, 7).However, potential adverse immunological reactions and the

difficulty in delivering these large molecules systemically maylimit progress in this field.

Pyridostigmine bromide, a quaternary carbamate that doesnot cross the blood–brain barrier appreciably and that reversiblyinhibits AChE and BuChE with similar potencies, has beenapproved for use among military personnel who are under threatof exposure to nerve agents. Pretreatment with pyridostigmineprevents OP-induced irreversible AChE inhibition in the pe-riphery, and it increases survival of animals acutely exposed tolethal doses of nerve agents, provided that atropine and oximesare administered promptly after an OP exposure (5, 8, 9). Whenused acutely before an OP exposure, reversible inhibitors ofAChE that are capable of crossing the blood–brain barrier,including physostigmine, tacrine, and huperzine A (hereafterreferred to as huperzine), afford better protection than pyri-dostigmine against OP toxicity, but generally this protectionoccurs at doses that produce significant incapacitation andcentral nervous system impairment (10–13).

Galantamine, a drug approved for treatment of mild to moderateAlzheimer’s disease (14), has properties appropriate for an anti-dotal therapy against OP poisoning. Briefly, galantamine (i) is areversible AChE inhibitor that crosses the blood–brain barrier (14);(ii) has anticonvulsant properties (15, 16); and (iii) prevents neu-rodegeneration (17–19), a hallmark of OP poisoning (20). Thus, weused guinea pigs, the best nonprimate model to predict the effec-tiveness of antidotal therapies for OP poisoning in humans (21), totest the hypothesis that galantamine may be an effective and safecountermeasure against OP intoxication. Because of their lowlevels of circulating carboxylesterases, guinea pigs, like nonhumanprimates, are considerably more sensitive to OPs, and they respondbetter than do rats or mice to antidotal therapies consisting ofpretreatment with reversible AChE inhibitors and posttreatmentwith atropine (21).

Our results demonstrate that OP toxicity and lethality arecounteracted when galantamine is administered before or soonafter the acute exposure of atropine-treated guinea pigs to thenerve agents soman and sarin, or to paraoxon, the activemetabolite of the OP insecticide parathion. We also provideevidence that the antidotal therapy consisting of galantamineand atropine is more effective and less toxic than alternativetreatments.

ResultsCombination Therapy Consisting of Galantamine and Atropine Effec-tively Prevents the Acute Toxicity of Lethal Doses of Soman, Sarin, andParaoxon: Comparison with Pyridostigmine and Huperzine. Clearsigns of cholinergic hyperexcitation, including miosis, increased

Conflict of interest statement: No conflicts declared.

Freely available online through the PNAS open access option.

Abbreviations: AChE, acetylcholinesterase; BuChE, butyrylcholinesterase; FJ-B, Fluoro-JadeB; OP, organophosphorus.

†To whom correspondence should be addressed. E-mail: ealbuque@umaryland.edu.

© 2006 by The National Academy of Sciences of the USA

13220–13225 � PNAS � August 29, 2006 � vol. 103 � no. 35 www.pnas.org�cgi�doi�10.1073�pnas.0605370103

Dow

nloa

ded

by g

uest

on

Aug

ust 1

2, 2

020

chewing, hypersalivation, muscle fasciculations, difficulty inbreathing, and loss of motor coordination, were evident at 5–15min after the s.c. injection of 1.5� LD50 soman (42 �g�kg of bodyweight) or sarin (63 �g�kg) in prepubertal male guinea pigs.Although an i.m. injection of atropine (6–16 mg�kg) immedi-ately after the OP challenge attenuated the muscarinic signs, allanimals showed tremors and intense convulsions within 15–30min after the challenge. Atropine-treated, OP-challenged guineapigs were euthanized when they developed life-threateningsymptoms, and at 24 h after the exposure to nerve agents, only11% of the animals (7 of 65) remained alive.

All guinea pigs that were pretreated with 5–12 mg�kggalantamine�HBr (hereafter referred to as galantamine) andposttreated with 10 mg�kg atropine survived the s.c. injection of1.5� LD50 soman or sarin, with no toxic signs either before orafter the OP exposure. The ED50 values of galantamine for 24-hsurvival of animals exposed to soman or sarin were 1.82 � 0.37or 2.2 � 0.50 mg�kg, respectively (Fig. 1 A and B). The optimaldosage of galantamine changed as the OP levels increased. Forexample, in animals posttreated with 10 mg�kg atropine, theED50 for galantamine to prevent the lethality of 2.0� LD50soman was 5.1 � 0.66 mg�kg (mean � SEM; n � 8–10 animalsper group), with 100% 24-h survival being achieved with �8mg�kg (Fig. 1 A). Effective doses of galantamine were welltolerated; only animals that received 16–20 mg�kg galantamineshowed mild adverse symptoms, which lasted 10–15 min andincluded increased chewing, hypersalivation, fasciculations, andtremors.

Muscarinic blockade by atropine contributed to the antidotaleffectiveness. Regardless of whether animals were pretreated with5 or 8 mg�kg galantamine, 50% reduction of the lethality of thosenerve agents was achieved with similar doses of atropine (mean �SEM � 5.7 � 0.47 mg�kg and 5.2 � 0.13 mg�kg, respectively).

However, a synergistic interaction occurred between galantamineand �6 mg�kg atropine; increasing the dose of galantamine from5 to 8 mg�kg decreased the dose of atropine needed to protect theanimals from the toxicity of soman (Fig. 1C). Doses of galantamineand atropine required to treat OP intoxication may be optimized byusing response-surface methods (22).

The acute toxicity of paraoxon was also effectively counter-acted by therapy consisting of galantamine and atropine. Allguinea pigs treated with atropine (10 mg�kg, i.m.) immediatelyafter their exposure to �1.8 mg�kg paraoxon developed life-threatening symptoms and were euthanized. In contrast, allatropine-treated animals survived with no signs of toxicity whenthey received galantamine (8 mg�kg, i.m.) 30 min before theirexposure to 2 mg�kg paraoxon (Fig. 1D). Further, galantamine�atropine-treated animals that survived the challenge with 3mg�kg paraoxon (Fig. 1D) displayed only brief, mild signs ofintoxication that included increased chewing and slight tremors.

The effectiveness of the antidotal therapy consisting of ga-lantamine�atropine surpassed that of a combination of pyri-dostigmine and atropine in preventing acute OP toxicity. Only afraction of animals pretreated with pyridostigmine (26–65 �g�kg) and posttreated with 10 mg�kg atropine survived the chal-lenge with 1.5� LD50 soman (Fig. 1E). The effectiveness of thistherapy increased as the dose of pyridostigmine was raised to 52�g�kg (Fig. 1E). Increasing the dose of pyridostigmine to 65�g�kg, however, decreased the effectiveness of the treatment,most likely because the potential benefit of increasing theprotection of AChE from the actions of OPs is counteracted andeventually outweighed by the simultaneous pyridostigmine-induced inhibition of BuChE, an enzyme that serves as anendogenous scavenger of OPs (6).

The safety of the antidotal therapy consisting of galantamine�atropine was greater than that of a combination of huperzine andatropine. Approximately 80% of the animals challenged s.c. with1.5� LD50 soman survived if they were pretreated with 100–200�g�kg huperzine and posttreated with 10 mg�kg atropine; theminimum dose of huperzine needed to provide 100% survival ofsoman-challenged, atropine-treated guinea pigs was 300 �g�kg(Fig. 1E). However, at doses �300 �g�kg, huperzine triggeredtransient, albeit incapacitating side effects that included profusesecretions, muscle fasciculations, abnormal gait, tremors, andrespiratory distress. The stereotypic behavior of animals treatedwith huperzine was quantitatively analyzed in an open-fieldarena, as described below.

Galantamine Maintains Long-Term Survival of OP-Challenged, Atro-pine-Treated Guinea Pigs and Has No Significant Effect on GrossBehavior of the Animals: Comparison with Huperzine. Even thoughall animals survived the first 24 h after the soman challenge whenthey were pretreated with 6 mg�kg galantamine and posttreatedwith 6 mg�kg atropine, only 80% of them remained alive afterthe 3rd day post-OP exposure (Fig. 2A). In contrast, during theentire observation period, survival remained at 100% in animalspretreated with 5–8 mg�kg galantamine and posttreated with 10mg�kg atropine. Increasing the dose of atropine to 16 mg�kgreduced the acute and long-term efficacy of doses of galan-tamine �8 mg�kg (Fig. 2 A). Thus, 10 mg�kg atropine ensuredthe highest long-term effectiveness of galantamine against thetoxicity of 1.5� LD50 soman.

Within 1 week after a single i.m. injection of saline, galan-tamine (8 mg�kg), or atropine (10 mg�kg), guinea pigs gainedweight at similar rates, i.e., 2.51 � 0.11% per day, 2.30 � 0.05%per day, and 2.37 � 0.03% per day (Fig. 2B). In contrast, guineapigs that received a single i.m. injection of huperzine (300 �g�kg)gained weight at a rate of 1.72 � 0.17% per day (Fig. 2B), whichis significantly slower than that measured for saline-injectedanimals (P � 0.01 compared with saline-injected animals ac-cording to ANOVA followed by Dunnett’s post hoc test).

Fig. 1. Pretreatment with galantamine prevents the acute toxicity of lethaldoses of OPs: Comparison with pyridostigmine and huperzine. In all experi-ments, guinea pigs received an i.m. injection of selected doses of galantamine,pyridostigmine, or huperzine followed 30 min later by a single s.c. injection of1.5� LD50 (42 �g�kg) or 2.0� LD50 (56 �g�kg) soman, 1.5� LD50 sarin (56�g�kg), or the indicated doses of paraoxon. At 1 min after the OP challenge,all animals received atropine (1–10 mg�kg, i.m.). (A–C) Dose–response rela-tionships for galantamine or atropine to maintain 24-h survival of animalschallenged with nerve agents. (D) Dose–response relationship for paraoxon-induced decrease in 24-h survival of atropine-treated guinea pigs that werepretreated with saline or galantamine. (E) Effects of increasing doses ofpyridostigmine or huperzine in maintaining 24-h survival of soman-challenged, atropine-treated animals. Each group had 8–12 animals. Percentsurvival represents the percent of animals that were kept alive because theypresented no life-threatening symptoms.

Albuquerque et al. PNAS � August 29, 2006 � vol. 103 � no. 35 � 13221

PHA

RMA

COLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 1

2, 2

020

Although galantamine�atropine-treated, soman-challenged an-imals lost, on average, 10% of their body weight at 24 h after theOP exposure (Fig. 2B), their rate of weight gain during theremaining recovery period (2.72 � 0.26% per day; mean � SEM;n � 5 animals) was not significantly different from that ofsaline-treated animals that were not challenged with soman.Galantamine�atropine was equally effective in maintaining therates of weight gain of guinea pigs challenged with 1.5� LD50sarin or 3 mg�kg paraoxon at 2.53 � 0.20% per day or 2.66 �0.21% per day (mean � SEM; n � 3–5 animals per group),respectively. The acute toxicity of huperzine was not reflected inthe rates of weight gain of animals that survived the OPchallenge when treated with huperzine�atropine (Fig. 2B).

In an attempt to quantify potential untoward behavioraleffects of the doses of galantamine and huperzine needed toprevent acute OP poisoning, the overall ambulatory activity ofguinea pigs was examined in an open-field arena. Previousstudies reported that other centrally acting AChE inhibitors,including physostigmine, decrease locomotor activity and ste-reotypic behavior of rodents in the open field (23). Further,inhibition of the NMDA type of glutamate receptors, a mech-

anism that appears to contribute to the effectiveness of huper-zine in preventing OP toxicity (24), is known to increase stereo-typy in rodents (25).

Each guinea pig, immediately after receiving an i.m. injectionof saline, galantamine (8 mg�kg), or huperzine (300 �g�kg), wasplaced in an open-field arena equipped with infrared sensors. Atthe dose tested, galantamine had no significant effect on theoverall locomotor activity of guinea pigs (Fig. 2C). However,huperzine increased the locomotor activity of the animals, andthis effect became significant at 30 min after the treatment (Fig.2C). At this time, a distinct pattern of locomotor stereotypy,including repetitive routes of locomotion in the open-field arena,was also significantly higher in huperzine- than in saline-treatedanimals (Fig. 2C). These effects of huperzine resemble those ofthe NMDA receptor antagonists ketamine, phencyclidine, anddizolcipine (25).

Galantamine Can Be Safely Used Pre- or Posttreatment to CounteractAcute OP Toxicity: Therapeutic Windows of Time. An effectiveantidotal therapy should afford long-lasting protection for firstresponders who will attend a population acutely exposed to toxiclevels of OPs. Thus, experiments were designed to determinehow long before an exposure to OPs an acute pretreatment withgalantamine would remain effective in preventing their toxicity.All atropine-treated guinea pigs that received 8 mg�kg galan-tamine up to 1 h before soman survived with no signs of toxicity(Fig. 3A). As the interval between the injections of galantamineand soman increased beyond 1 h, the survival decreased (Fig.3A). Increasing the dose of galantamine to 10 mg�kg prolongedthe time within which the antidotal therapy remained effective(Fig. 3A).

Considering the difficulty of predicting when a person will beexposed to toxic levels of OPs under battlefield conditions, in thecase of a terrorist attack, or during handling of insecticides,experiments were also designed to determine whether posttreat-ment with galantamine could effectively counteract the acutetoxicity of OPs. All animals treated with 8 or 10 mg�kg galan-tamine up to 5 min after the soman challenge survived (Fig. 3B)with no signs of intoxication; the rate of weight gain and grossbehavior of these animals were indistinguishable from those ofsaline-treated animals that were not exposed to soman. Galan-tamine was no longer effective when given 10 min after 1.5�LD50 soman. Posttreatment with galantamine�atropine alsoprevented the acute toxicity of supralethal doses of paraoxon(Fig. 3C). The therapeutic window of time within which post-treatment with galantamine remained effective in sustaining100% survival of the animals decreased as the dose of the OPincreased (Fig. 3C).

Fig. 2. Long-term effectiveness and acute toxicity of different antidotaltherapies against OP poisoning. (A) Seven-day survival of guinea pigs treatedwith galantamine at 30 min before and atropine at 1 min after their challengewith 1.5� LD50 soman. Each group had 8–12 animals. (B) Seven-day follow-upof the weight of animals subjected to different treatments. Weights areexpressed as percent of the weights measured 1 h before a treatment. Controlgroups consist of animals that received a single i.m. injection of atropine,galantamine, huperzine, or saline. The soman�atropine groups consist ofanimals treated with galantamine or huperzine at 30 min before and atropineat 1 min after soman (n � 5–8 animals per treatment). (C) Graphs of theaverage total distance traveled and stereotypy of guinea pigs in an open-fieldarena at the indicated times after they received an i.m. injection of saline,galantamine, or huperzine (n � 6 animals per treatment). In B and C, resultsare presented as the mean � SEM. Asterisks indicate that results from huper-zine- and saline-treated animals are significantly different at P � 0.05 (ANOVAfollowed by Dunnett’s post hoc test).

Fig. 3. Efficacy of galantamine as a pre- or posttreatment for OP poisoningis dose- and time-dependent. (A) Twenty-four-hour survival of animals thatreceived a single i.m. injection of 8 or 10 mg�kg galantamine at 1, 2, 3, 4, or5 h before the s.c. injection of 1.5� LD50 soman that was followed 1 min laterby an i.m. injection of 10 mg�kg atropine. (B and C) Twenty-four-hour survivalof animals that received a single i.m. injection of specific doses of galantamineat different times after their challenge with 1.5� LD50 soman or 2–3 mg�kgparaoxon, respectively. Each group had 8–10 animals.

13222 � www.pnas.org�cgi�doi�10.1073�pnas.0605370103 Albuquerque et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 1

2, 2

020

No Signs of Neurotoxicity Were Observed in the Brains of Atropine-Treated Guinea Pigs That Received Galantamine Before or After theChallenge with Soman. Neurodegeneration in three areas of thebrain, the pyriform cortex, the amygdala, and the hippocampus,is characteristic of OP intoxication. The components of theantidotal therapy regimen, by themselves, were not neurotoxic.No signs of brain damage were detected at 24 h after an i.m.injection of saline (Fig. 4A), 8 mg�kg galantamine (Fig. 4B), or10 mg�kg atropine (data not shown). Galantamine is a criticalcomponent of the antidotal therapy regimen because atropinealone was unable to prevent the well described neuronal deathtriggered by 1.5� LD50 soman (Fig. 4C). Large numbers ofshrunken neurons that were labeled with Fluoro-Jade B (FJ-B),an anionic fluorescein derivative that binds with high affinity todegenerating cells, were consistently seen in the hippocampus,amygdala, and pyriform cortex of atropine-treated guinea pigsthat survived for 24 h after the challenge with 1.5� LD50 soman(Fig. 4C). In contrast, staining with FJ-B was rarely seen in brainsections of soman-challenged, atropine-treated animals thatwere given 8 mg�kg galantamine at 30 min before or 5 min afterthe OP (Fig. 4 D and E). Further, the edema observed in thehippocampus and the marked parenchymal spongy state of theamygdala and pyriform cortex of soman-exposed, atropine-treated animals were absent in animals that received galan-tamine 30 min before or 5 min after the nerve agent (Fig. 4 C–E).

Doses of Galantamine That Effectively Prevent OP-Induced Toxicityand Lethality Are Clinically Relevant. To help establish the clinicalrelevance of the doses of galantamine needed to counteract OPpoisoning, plasma and brain concentrations of the drug weredetermined by HPLC at various times after treatment of guineapigs with 8 mg�kg galantamine. This dose was selected because(i) in association with atropine, it afforded full protection againstOP-induced toxicity and lethality, and (ii) it was half of theminimum dose at which galantamine triggered mild side effects.

In guinea pigs, as in humans (26), plasma levels of galantaminedeclined with first-order kinetics. After an i.m. injection of 8mg�kg galantamine, plasma and brain levels of the drug peakedbetween 5 and 30 min and decayed with half-times of 71.7 � 14.4min and 57.8 � 4.31 min, respectively (Fig. 5 A and B). As shownin Fig. 3A, full protection against acute toxicity was achievedwhen 8 mg�kg galantamine was administered to guinea pigs upto 1 h before soman, a time when plasma and brain levels of thedrug were 0.90 � 0.01 �g�ml and 0.80 � 0.04 �g�g, respectively(Fig. 5 A and B). Because the molecular weight of galantamineis 287.4, these findings suggest that the minimal plasma concen-tration of galantamine needed to prevent OP toxicity andlethality is �2.8 �M. Doses of galantamine recommended fortreatment of patients with Alzheimer’s disease are between 8and 24 mg�day (14), and peak plasma concentrations of 0.2–3

Fig. 4. Soman-induced neurodegeneration is not present in the hippocam-pus, pyriform cortex, and amygdala of guinea pigs pre- or posttreated withgalantamine. (A and B) Representative photomicrographs of the hippocampalCA1 field, the pyriform cortex, and the amygadala of guinea pigs that wereeuthanized 24 h after an i.m. injection of saline (A) or 8 mg�kg galantamine(B). No FJ-B-positive neurons were seen in the brains of these animals. (C) Largenumbers of FJ-B-positive neurons were seen in all three index areas of thebrain of a guinea pig that survived for 24 h after the challenge with 1.5� LD50

soman. (D and E) FJ-B-positive neurons were rarely seen in brain sections ofanimals that received galantamine (8 mg�kg, i.m.) at 30 min before (D) or 5min after (E) soman. In C–E, all animals received atropine (10 mg�kg, i.m.) at1 min after the OP, and they were euthanized at 24 h after the OP challenge.Photomicrographs are representative of results obtained from each group,which had five animals.

Fig. 5. Differential sensitivity of brain and blood AChE activities to inhibition bygalantamine in vivo and in vitro. (A and B) Concentrations of galantaminemeasured in blood (A) and brain (B) samples obtained at various times aftertreatment of guinea pigs (n � 4–6 animals per time point) with galantamine (8mg�kg, i.m.) is plotted on a logarithmic scale against time. (C) AChE activitymeasured in samples from saline-treated animals was taken as 1 and used tonormalize the enzyme activity measured in samples obtained at various timesafter treatment of animals with galantamine (8 mg�kg, i.m.). Normalized inhi-bition (1 � normalized activity) is plotted against the time at which samples wereobtained. Asterisks indicate that results from galantamine- and saline-treatedanimals are significantly different at P � 0.001 (***) or � 0.01 (**) (ANOVAfollowed by Dunnett’s post hoc test). In A–C, the first point corresponds to resultsobtained at 5 min after the treatment. (D) Increasing concentrations of galan-tamine were added in vitro to brain homogenates and blood samples obtainedfrom naive animals. AChE activity in untreated samples was taken as 1, and it wasused to normalize activity measured in galantamine-treated samples. The graphof normalized AChE activity vs. galantamine concentrations was fitted with theHill equation. Results are presented as the mean � SEM (n � 4–6 animals pergalantamine concentration).

Albuquerque et al. PNAS � August 29, 2006 � vol. 103 � no. 35 � 13223

PHA

RMA

COLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 1

2, 2

020

�M have been detected in healthy human subjects treated orallyor s.c. with a single dose of 10 mg of galantamine (26, 27). Thus,doses of galantamine needed to prevent OP toxicity generatepeak plasma concentrations similar to those achieved with dosesclinically used to treat Alzheimer’s disease.

In agreement with the concept that galantamine-inducedAChE inhibition is reversible, the degree of AChE inhibition inbrain and blood from galantamine-treated guinea pigs decreasedas the galantamine levels declined in both compartments. Inhi-bition of AChE became negligible at 6 h after the treatment (Fig.5C), when plasma and brain levels of the drug were �0.1 �g�mland 0.1 �g�g, respectively (Fig. 5 A and B). Maximal inhibitionof blood AChE activity was �70% (Fig. 5C), observed at 30 minafter the treatment when the plasma levels of galantamine hadpeaked. The effectiveness of galantamine in patients with Alz-heimer’s disease has been correlated with 40–70% inhibition ofAChE in blood (28).

Maximal AChE inhibition in the brains of galantamine-treated animals was significantly different from that observed intheir blood (Fig. 5C). Measured peak concentrations of galan-tamine were 1.6 � 0.13 �g�ml in the plasma and 1.38 � 0.11 �g�gin the brain. These concentrations resulted in �70% and 25%inhibition of AChE in the blood and brain, respectively. Mea-sured peak levels of galantamine in the plasma correspond to5.6 � 0.5 �M. Considering 80% of the brain weight as water,measured peak levels of galantamine in brain tissue wouldcorrespond to 3.8 � 0.3 �M. Based on the concentration–response relationships obtained for galantamine-induced inhi-bition of guinea pig blood and brain AChE in vitro (Fig. 5D), itis estimated that 5.6 �M galantamine would inhibit blood AChEactivity by 68%, and 3.8 �M galantamine would inhibit brainAChE activity by 25%. In vitro, galantamine inhibited guinea pigblood and brain AChE with EC50 values of 1.8 � 0.38 �M and16.9 � 9.8 �M, respectively (mean � SEM; Fig. 5D). In humans,blood AChE activity is also 10-fold more sensitive to inhibitionby galantamine than is brain AChE activity (29).

DiscussionThe present study demonstrates the remarkable potential ofgalantamine to improve antidotal therapy for even the mostdeadly OPs. In combination with atropine, well tolerated, clin-ically relevant doses of galantamine, administered acutely eitherbefore or soon after an exposure to the nerve agents soman andsarin or paraoxon, fully counteract the toxicity and lethality ofthese compounds. Although atropine alone attenuates the mus-carinic syndrome resulting from the exposure of guinea pigs tothe OPs, it does not afford significant protection against theirlethality.

The exact mechanisms that account for the superiority ofgalantamine as a countermeasure against OP poisoning are yetto be fully elucidated. However, it can be postulated that theeffectiveness of galantamine is related both to the higher potencywith which it inhibits AChE compared with BuChE (30), anaction that should help preserve the scavenger capacity ofplasma BuChE for OPs, and to the protection of brain AChEfrom OP-induced irreversible inhibition. The finding that galan-tamine was essential to counteract soman-induced neurodegen-eration in the brain supports the notion that AChE-relatedand�or -unrelated actions of this drug in the central nervoussystem contribute to its effectiveness. Neuronal loss in the brainsof OP-intoxicated animals correlates to some extent with theintensity and duration of OP-triggered seizures (31–33). Yet,neurodegeneration and consequent cognitive impairment in-duced by OPs can be significantly reduced by therapeuticinterventions that, although unable to suppress OP-triggeredseizures, effectively decrease glutamate excitotoxicity (32). Theability of the galantamine-based therapy to prevent OP-inducedconvulsions and the well reported neuroprotective effects of

galantamine against different insults (17–19, 34, 35) may beimportant determinants of the antidotal effectiveness. Becauseno cognitive impairment has been detected in soman-challengedanimals when neuronal loss in their brains remains below acertain threshold (32), galantamine is likely to maintain normalcognitive performance of OP-exposed subjects.

Inhibition of brain AChE by �60–70% has been shown totrigger severe incapacitating effects, including seizures (36).Maximal degrees of inhibition of AChE activities observed inguinea pigs treated with doses of galantamine that effectivelycounteracted OP intoxication were �70% in blood and 25% inbrain. All other centrally acting AChE inhibitors studied to date,including huperzine, acutely prevent OP toxicity when used atdoses that decrease blood AChE activity by �70% (10–13).However, brain AChE activity is inhibited to a similar extent(�70%) by these drugs (13). Therefore, a high degree ofreversible and selective AChE inhibition in the blood appears tobe necessary to counteract the peripheral toxic effects of OPsacutely. A low degree of reversible inhibition of brain AChE maybe sufficient to protect a significant pool of the enzyme fromOP-induced irreversible inhibition, and it may be critical to limitthe occurrence of untoward side effects of centrally actingreversible AChE inhibitors.

Development of effective and safe antidotes against OPtoxicity will help improve the treatment of the victims of aterrorist attack with nerve agents, and it will help reduce themortality associated with OP insecticide poisoning worldwide.The demonstration that an acute galantamine-based therapyeffectively and safely counteracts OP poisoning is, therefore, ofutmost relevance for farm workers and others who handle OPinsecticides, for the general population under threat of OPexposure in terrorist attacks, and for soldiers, who, despite theGeneva Protocol, may be exposed to deadly nerve agents in thecourse of battle.

Materials and MethodsAnimal Care and Treatments. Male albino guinea pigs [Crl(HA)Br;Charles River Laboratories, Wilmington, MA] weighing 350–420 g (5–6 weeks old) were used. Galantamine, pyridostigmine,or huperzine was injected in one hindlimb, and atropine wasinjected in the other. The nerve agents, diluted in sterile saline,and paraoxon, diluted in DMSO, were injected s.c. between theshoulder blades of the animals. All injections (�0.5 ml�kg) wereperformed by using disposable tuberculin syringes with 25- to26-gauge needles. Handling and disposal of nerve agents wereaccording to the rules set forth by the U.S. Army. All conditionsfor animal maintenance conformed to the regulations of theAssociation for Assessment and Accreditation of LaboratoryAnimal Care, complied with the standards of the AnimalWelfare Act, and adhered to the principles of the 1996 Guide forthe Care and Use of Laboratory Animals (37). Atropine sulfate,pyridostigmine bromide, (�)-huperzine A, and paraoxon werepurchased from Sigma–Aldrich (St. Louis, MO). Soman andsarin were obtained from the U.S. Army Medical Research andDevelopment Command (Fort Detrick, MD). Galantamine�HBrwas a generous gift from Alfred Maelicke (Galantos, Mainz,Germany).

Histopathological Analyses. Guinea pigs were anesthetized atappropriate times after their treatments and transcardially per-fused with 0.9% saline (70 ml�min) until blood was cleared andsubsequently perfused with 10% formalin. Their brains werethen removed, placed in 10% formalin for no longer than 48 h,dehydrated, and embedded in paraffin. Sections 5 �m thick werecut and then dried in an incubator at 37°C for 12 h before theywere stained with FJ-B (38). After it was mounted, the tissue wasexamined under an epifluorescence microscope with blue (450–490 nm) excitation light and a filter for fluorescein isothiocya-

13224 � www.pnas.org�cgi�doi�10.1073�pnas.0605370103 Albuquerque et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 1

2, 2

020

nate. Photomicrographs were taken with a digital microscopecamera (AxioCam; Zeiss, Jena, Germany).

Analysis of Galantamine Concentrations in the Brain and Plasma ofGuinea Pigs. At various times after treatment with galantamine (8mg�kg, i.m.), animals were anesthetized with CO2. Blood (5–10ml) was collected by cardiopuncture with a plastic heparinizedsystem and kept in dry ice. Immediately after cardiopuncture,the animals were exsanguinated by carotid artery transection.Their brains were removed, superfused with 0.9% saline, andsnap frozen in liquid nitrogen. Frozen blood samples and brainswere kept at �80°C until further processing. Brain and plasmalevels of galantamine were measured by using a modified HPLCmethod (39).

Radiometric Enzymatic Assay. Pulverized brain tissue was mixedwith buffer containing antiproteases (0.5 unit�ml aprotinin, 30�g�ml leupeptin, 1 mg�ml bacitracin, 2 mM benzamidine, and 5mM N-ethylmaleimide) and sonicated for 20 s on ice. Aliquotsof the resulting suspensions and of blood samples were used fordetermination of protein concentration (micro BCA proteinassay; Pierce, Rockford, IL). Measurements of AChE activitywere performed in the presence of the BuChE inhibitor tetrai-sopropyl pyrophosphoramide (1 mM) with a modified two-phaseradiometric assay (40) using 20 pM [3H]acetylcholine iodide[specific activity, 76 Ci�mmol (1 Ci � 37 GBq); PerkinElmerLife Sciences, Boston, MA], which produced �200,000 cpmwhen totally hydrolyzed by eel AChE (2 units).

Behavioral Assays. Locomotor activity and stereotypy of guineapigs were analyzed in an open-field arena equipped with infraredsensors (AccuScan Instruments, Columbus, OH), as describedby June et al. (41). Counts obtained from the total number ofinterruptions of the infrared beams were automatically compiledevery 5 min and processed for measures of total distance traveledand stereotypy.

We thank Dr. Harry L. June and Dr. Jacek Mamczarz (Department ofPsychiatry, University of Maryland School of Medicine) for guidanceand assistance with the behavioral experiments. We are grateful to Dr.Robert Bloch (Department of Physiology, University of MarylandSchool of Medicine), Dr. David Burt (Department of Pharmacology andExperimental Therapeutics, University of Maryland School of Medi-cine), Col. George Korch (U.S. Army Medical Research Institute ofInfectious Diseases, Fort Detrick, MD), Col. Dr. David Moore (StrategicResearch Program Development, Office of the Commander, U.S. ArmyMedical Research Institute of Chemical Defense), and Dr. SusanWonnacott (Department of Biology and Biochemistry, University ofBath, Bath, U.K.) for helpful comments. We are also indebted to Ms.Mabel Zelle for technical and editorial support and to Ms. Christina M.Tompkins and Ms. Denise M. Fath for excellent assistance with thehistological techniques. This work was supported by U.S. Army MedicalResearch and Development Command Contract DAMD-17-95-C-5063,Battelle Scientific Services Contract TCN 03132, U.S. Public HealthService Grant NS25296 from the National Institutes of Health, andNational Institutes of Environmental Health Sciences�National Insti-tutes of Health Training Grant T32 ES07263 (all to E.X.A.). The use ofgalantamine as an antidote against OP poisoning is protected under theInternational Patent Application PCT�US05�33789 filed on September23, 2005.

1. Coupland, R. & Leins, K. R. (2005) Science 308, 1841.2. Romano, J. A., Jr., & King, J. M. (2001) Mil. Med. 166, 21–22.3. Karalliedde, L. & Senanayake, N. (1989) Br. J. Anaesth. 65, 736–750.4. Buckley, N. A., Karalliedde, L., Dawson, A., Senanayake, N. & Eddleston, M.

(2004) J. Toxicol. Clin. Toxicol. 42, 113–116.5. Bajgar, J. (2004) Adv. Clin. Chem. 38, 151–216.6. Doctor, B. P., Raveh, L., Wolfe, A. D., Maxwell, D. M. & Ashani, Y. (1991)

Neurosci. Biobehav. Rev. 15, 123–128.7. Ghanem, E. & Raushel, F. M. (2005) Toxicol. Appl. Pharmacol. 207, 459–470.8. Wetherell, J., Hall, T. & Passingham, S. (2002) Neurotoxicology 23, 341–349.9. Leadbeater, L., Inns, R. H. & Rylands, J. M. (1985) Fundam. Appl. Toxicol. 5,

S225–S231.10. Deshpande, S. S., Viana, G. B., Kauffman, F. C., Rickett, D. L. & Albuquerque,

E. X. (1986) Fundam. Appl. Toxicol. 6, 566–577.11. Grunwald, J., Raveh, L., Doctor, B. P. & Ashani, Y. (1994) Life Sci. 54,

991–997.12. Fricke, R. F., Koplovitz, I., Scharf, B. A., Rockwood, G. A., Olson, C. T.,

Hobson, D. W. & Blank, J. A. (1994) Drug Chem. Toxicol. 17, 15–34.13. Lallement, G., Baille, V., Baubichon, D., Carpentier, P., Collombet, J. M.,

Filliat, P., Foquin, A., Four, E., Masqueliez, C., Testylier, G., et al. (2002)Neurotoxicology 23, 1–5.

14. Corey-Bloom, J. (2003) Int. J. Clin. Pract. 57, 219–223.15. Dreyer, R. (1968) Muench. Med. Wochenschr. 110, 1481.16. Losev, N. A. & Tkachenko, E. I. (1986) Biull. Eksp. Biol. Med. 101, 436–438.17. Pereira, E. F. R., Hilmas, C., Santos, M. D., Alkondon, M., Maelicke, A. &

Albuquerque, E. X. (2002) J. Neurobiol. 53, 479–500.18. Arias, E., Ales, E., Gabilan, N. H., Cano-Abad, M. F., Villarroya, M., Garcia,

A. G. & Lopez, M. G. (2004) Neuropharmacology 46, 103–114.19. Kihara, T., Sawada, H., Nakamizo, T., Kanki, R., Yamashita, H., Maelicke, A.

& Shimohama, S. (2004) Biochem. Biophys. Res. Commun. 325, 976–982.20. Shih, T. M., Duniho, S. M. & McDonough, J. H. (2003) Toxicol. Appl.

Pharmacol. 188, 69–80.21. Maxwell, D. M., Brecht, K. M., Lenz, D. E. & O’Neill, B. L. (1988) J. Pharmacol.

Exp. Ther. 246, 986–991.22. Carter, W. H., Jones, D. E. & Carchman, R. A. (1985) Fundam. Appl. Toxicol.

5, S232–S241.23. Silvestre, J. S., Fernandez, A. G. & Palacios, J. M. (1999) Pharmacol. Biochem.

Behav. 64, 1–5.

24. Gordon, R. K., Nigam, S. V., Weitz, J. A., Dave, J. R., Doctor, B. P. & Vedr,H. S. (2001) J. Appl. Toxicol. 21, S47–S51.

25. Koek, W., Woods, J. H. & Winger, G. D. (1988) J. Pharmacol. Exp. Ther. 245,969–974.

26. Bickel, U., Thomsen, T., Weber, W., Fischer, J. P., Bachus, R., Nitz, M. &Kewitz, H. (1991) Clin. Pharmacol. Ther. 50, 420–428.

27. Mihailova, D., Yamboliev, I., Zhivkova, Z., Tencheva, J. & Jovovich, V. (1989)Pharmacology 39, 50–58.

28. Jann, M. W., Shirley, K. L. & Small, G. W. (2002) Clin. Pharmacokinet. 41,719–739.

29. Thomsen, T., Kaden, B., Fischer, J. P., Bickel, U., Barz, H., Gusztony, G.,Cervos-Navarro, J. & Kewitz, H. (1991) Eur. J. Clin. Chem. Clin. Biochem. 29,487–492.

30. Thomsen, T. & Kewitz, H. (1990) Life Sci. 46, 1553–1558.31. McDonough, J. H. & Shih, T. M. (1997) Neurosci. Biobehav. Rev. 21, 559–579.32. Filliat, P., Baubichon, D., Burckhart, M. F., Pernot-Marino, I., Foquin, A.,

Masqueliez, C., Perrichon, C., Carpentier, P. & Lallement, G. (1999) Neuro-toxicology 20, 535–549.

33. Myhrer, T., Andersen, J. M., Nguyen, N. H. & Aas, P. (2005) Neurotoxicology26, 39–48.

34. Nakamizo, T., Kawamata, J., Yamashita, H., Kanki, R., Kihara, T., Sawada, H.,Akaike, A. & Shimohama, S. (2005) Biochem. Biophys. Res. Commun. 330,1285–1289.

35. Capsoni, S., Giannotta, S. & Cattaneo, A. (2002) Proc. Natl. Acad. Sci. USA 99,12432–12437.

36. Tondulli, L. S., Testylier, G., Marino, I. P. & Lallement, G. (1999) J. Neurosci.Res. 58, 464–473.

37. Institute of Laboratory Animal Resources (1996) Guide for the Care and Useof Laboratory Animals (National Academy Press, Washington, DC).

38. Schmued, L. C. & Hopkins, K. J. (2000) Brain Res. 874, 123–130.39. Claessens, H. A., van Thiel, M., Westra, P. & Soeterboek, A. M. (1983)

J. Chromatogr. 275, 345–353.40. Johnson, C. D. & Russell, R. L. (1975) Anal. Biochem. 64, 229–238.41. June, H. L., Duemler, S. E., Greene, T. L., Williams, J. A., Lin, M., Devaraju,

S. L., Chen, S. H., Lewis, M. J. & Murphy, J. M. (1995) J. Pharmacol. Exp. Ther.274, 1105–1112.

Albuquerque et al. PNAS � August 29, 2006 � vol. 103 � no. 35 � 13225

PHA

RMA

COLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 1

2, 2

020