detection of antibodies against botulinum toxins

Detection of Antibodies Against Botulinum Toxins

Dorothea Sesardic, PhD,* Russell G.A. Jones, PhD, Tong Leung, PhD, Toni Alsop, MSc,and Robert Tierney, BSc

Division of Bacteriology, National Institute for Biological Standards and Control, Potters Bar, Hertfordshire, United Kingdom

Abstract: After immunisation with botulinum vaccine, anti-bodies to multiple epitopes are produced. Only some of thesewill have the capacity to neutralise the toxin activity. In fact,the ability of toxoid vaccine to induce toxin neutralising anti-bodies has provided the basis for the use of therapeutic anti-toxins and immunoglobulins for the prophylaxis and treatmentof diseases caused by bacterial toxins. Increasing indicationsfor the chronic use of botulinum toxin for therapy have inevi-tably resulted in concern for patients becoming unresponsivebecause of the presence of circulating toxin-specific antibodies.Highly sensitive and relevant assays to detect only clinicallyrelevant toxin neutralising antibodies are essential. Althoughimmunoassays often provide the sensitivity, their relevance andspecificity is often questioned. The mouse protection LD50

bioassay is considered most relevant but can often only detect10 mIU/ml of antitoxin. This sensitivity, although sufficient for

confirming protective immunity, is inadequate for patients un-dergoing toxin therapy. An intramuscular paralysis assay im-proves the sensitivity to ca. 1 mIU/ml, and a mouse ex vivodiaphragm assay, with sensitivity of �0.5 mIU/ml, is the mostsensitive functional assay to date for this purpose. Alternativeapproaches for the detection of antibodies to botulinum toxinhave included in vitro endopeptidase activity neutralisation.Unlike any other functional assay, this approach is not relianton serotype-specific antibodies for specificity. Most recentpromising developments are focused on cellular assays utilisingprimary rat embryonic cord cells or more conveniently in vitrodifferentiated established cell lines such as human neuroblas-toma cells. © 2004 Movement Disorder Society

Key words: therapeutic agents; bacterial toxins; antitoxins;antibody; assays

The potent neurotoxins produced by strains of Clos-tridium botulinum act by blocking the release of acetyl-choline from peripheral nerve junctions, resulting in aflaccid paralysis. There are seven currently defined typesor serotypes (A–G) of botulinum toxin (BoNT). Highconcentrations (�50 ng/person) of some of these neuro-toxins cause a life-threatening paralysing disease calledbotulism. Although rare, botulism is caused by the in-gestion of food contaminated with botulinum spores (in-fant or intestinal botulism) or kept under anaerobic con-ditions and contaminated with toxin (food-bornebotulism), spore contamination of traumatized tissue(wound botulism), or may potentially result from its useas a biological weapon (e.g., inhalational botulism). Hu-

man botulism, including infant botulism, is generallyassociated with serotypes A, B, and E toxins, for whicheffective treatment is offered by intravenous or intramus-cular injection of antitoxins. Specific antitoxins typicallyconsist of hyperimmune equine F(ab�)2 preparations con-taining not less than 500 IU/ml of each of the types A andB and 50 IU/ml of type E antitoxin. Heptavalent prepa-rations have also been developed against A–G toxins,1

and human-derived botulinum antitoxin (BIG), with 25IU/ml for type A antitoxin, has been successfully used inan infant botulism trial in California, and recentlygranted FDA approval.1–3

Conversely, because of their specific pharmacologicalaction, botulinum toxins are being increasingly usedlocally (by means of subcutaneous or intramuscularroutes) at very low concentrations for cosmetic purposesand in the treatment of several diseases involving invol-untary muscle spasms, pain, migraine, cerebral palsy,and hyperhydrosis.4,5 Although muscle relaxant doses ofbotulinum toxin are considered to be below the levelsrequired to induce an immune response, toxin-neutralis-

Correspondence to: Dorothea Sesardic, PhD, Division of Bacteriol-ogy, National Institute for Biological Standards and Control, BlancheLane, South Mimms, Hertfordshire EN6 3QG, United Kingdom.E-mail: [email protected]

DOI 10.1002/mds.20021Published online in Wiley InterScience (www.interscience.wiley.

com).

Movement DisordersVol. 19, Suppl. 8, 2004, pp. S85–S91© 2004 Movement Disorder Society

S85

ing antibodies are often observed in patients receivingrepeated “higher dose” (�100 LD50 U per injectioncycle) toxin therapy over a long period of time.6 Becausethis condition renders further therapeutic injections inef-fective, it is important to identify the levels of toxin-neutralising antibodies (�1 m IU/ml) in such patients orideally be able to predict nonresponders at an early stage.Assays based on immunodetection methods have beenshown to detect antibodies that do not correlate to toxinneutralization,7–10 so these assays may not be suitable forpredicting clinical responses.

Recent data suggest that the frequency of botulinumtoxin therapy is likely to increase. Type A botulinumtoxin light (L) chain is being developed to target specificnociceptive afferent neurons, and available data indicateit as a promising new treatment in acute and chronic painmanagement.11,12 The dosing strategy of these novelanalgesics remains unknown; however, if “large” re-peated doses are required, neutralising antibodies mayalso be formed. Assays capable of monitoring these newspecific activities or differentiating L-chain fromH-chain effects with endopeptidase or cell binding/inter-nalisation activities, respectively, are, therefore, re-quired.

After immunisation with vaccines against botulinumtoxins (toxoid vaccine), antibodies against multipleepitopes are produced. Only some of these will be neu-tralising. The ability to induce toxin neutralising antibod-ies has provided the basis for the use of therapeuticantitoxins and immunoglobulins for the prophylaxis andtreatment of diseases caused by bacterial toxins. Protec-tion of workers by vaccination also requires monitoring.A primary series of multicomponent toxoid vaccine in-duces less than 0.08 to 2.00 IU/ml of neutralisation titresto type A toxin and less than 0.02 to 0.15 IU/ml to typeB.7 After the booster doses, titres rise to 0.12 to 8.9IU/ml (average 1.6 IU/ml), 0.07 to 0.80 IU/ml (average0.5 IU/ml), and 0.4 to 3.2 IU/ml (average 1.4 IU/ml), fortypes A, B, and E, respectively.13 Much higher neutral-ising titres were reported, however, in earlier studies bySiegel.8 Neutralisation titres of more than 5 IU/ml (5.74–51.60 IU/ml) for type A, more than 0.75 IU/ml (0.75–18.20 IU/ml) for type B, and more than 0.61 IU/ml(0.61–10.00 IU/ml) for type E toxin were reported in 25individuals, after multiple immunisations with pentava-lent vaccine.8 Due to problems in vaccinating the generalpopulation, and the highly variable protection induced,the use of antitoxin remains the best specific defenceagainst the use of botulinum toxin as a biological weapon.14

Sensitive and relevant assay systems are essential tomeasure the neutralising capacity of antibodies to botu-linum toxins. Such assay systems are also required to

determine the potency and stability of therapeutic anti-toxins, to confirm the protection after vaccination, and topredict clinical nonresponders to toxin therapy. Thisstudy reviews the current methods used to measure anddetect antibodies against botulinum toxins.

IN VIVO MOUSE MODELS OFANTITOXIN ACTIVITY

The neutralising potency of therapeutic botulinum an-titoxins is currently assayed exclusively using an in vivomouse lethality bioassay with a known sensitivity ofapproximately 5 to 10mIU per ml.15 The assay is basedon the traditional LD50 assay16 where the number of micesurviving, typically after 96 hours after the intraperito-neal (IP) injection of a fixed lethal dose of toxin, pre-mixed with different amounts of antitoxin, is scored. Oneinternational unit (IU or U) of antitoxin is defined as theamount of antibody that has the ability to specificallyneutralise 10,000 mouse IP LD50 doses of type A, B, orF botulinum toxins, or 1,000 LD50 doses of type Etoxin.17 Because large interlaboratory variations are ex-perienced with bioassays, the use of stable internationalreference standards of defined activity (calibrated inIU/ml or vial) are essential for antitoxin control. Largenumbers of animals, therefore, are used and due to thelethal end point substantial suffering incurred which hasencouraged the development of various alternative assaysystems.

The observation that a scoring method could be usedwith a low toxin challenge dose by means of subcutane-ous, intravenous, or intramuscular routes to increase thesensitivity for toxin16,18,19 and antitoxin detection20 wasthe first step toward the development of a range ofnonlethal assays.18–21

Various nonlethal (local paralysis) assays have sincebeen developed for botulinum toxins, which are on av-erage ca. 10 times more sensitive than the original le-thality methods, such as the regional chemodenerva-tion,22,23 flaccid paralysis,24 and digit abductionmethods.25,26 These assays use the highest dose of toxinsfailing to induce systemic toxicity (0.2–1.0 LD50/dosefor toxins A to F, as summarised in Table 1 for flaccidparalysis) and can provide information on the duration ofaction for each toxin serotype. When a fixed dose oftoxin, which fails to induce signs of systemic toxicity, ispremixed with a range of antitoxin dilutions, a sensitiveantitoxin assay is achieved (Table 1). For example, theinhibition of type A or B toxin-induced flaccid paralysis,by a corresponding reference antitoxin, is dose depen-dent (Fig. 1A and B). The detection limits for A to Fantitoxins using this assay are summarised in Table 1. Aslittle as 1 mU/ml of A, B, and C1 antitoxin is detected,

S86 D. SESARDIC ET AL.

Movement Disorders, Vol. 19, Suppl. 8, 2004

whereas for E and F antitoxin, detection limits are 5 and10 mU/ml.

IN VITRO FUNCTIONAL MODELS

During the past decade, several in vitro assays havebeen developed to detect and quantify botulinum tox-ins.27 Many of the adopted strategies were based on theimmunodetection of toxins and these are generally notsuitable to measure biological activity.28,29 This is be-cause immunoassays have a limited ability to distinguishbetween biologically active and inactive toxin. Becauseof such limitations, immunoassays could never serve asacceptable alternatives for the potency, safety, or stabil-ity testing of biologicals derived from bacterial toxins,for which it is absolutely essential to quantify the activecomponent.30

As the lethal effects of botulinum toxin result fromrespiratory paralysis, the use of an isolated tissue respon-sible for respiration seems a logical choice for an alter-native. The mouse phrenic nerve hemidiaphragm assay isbecoming established as a clear alternative to in vivoassays for research purposes10,31,32 and may eventuallybecome accepted for batch release testing. Like in vivoassays, it requires the fully functional toxin and is themost sensitive functional model for detecting neutralis-ing antibodies (�0.5 mU/ml). The mouse diaphragm

assay is also the method of choice used to detect the lowlevels of neutralising antibody in nonresponsive patientsof botulinum toxin therapy.10,33

Clostridia neurotoxins potently and specifically inhibitrelease in certain cell types and established cell lines,34,35

providing the opportunity for in vitro cell based assaysfor antitoxins. However, to date, cell-based approacheshave not provided sufficient sensitivity.36 Cellular assaysusing fully differentiated cell lines such as human neu-roblastoma cells,37, although highly attractive, have yetto be developed for this application. Studies with primarycell cultures such as rat embryonic spinal cord cells havebeen used and have substantially increased the sensitivityto provide an assay system capable of detecting 2 to 10mU/ml of antitoxin against type A toxin.38 This assay iscapable of assessing antisera that neutralise the endopep-tidase, translocation, and cell binding functions of thetoxin with sensitivity better than the mouse lethalityassay.

Over the past 10 years, there have been importantadvances in understanding the mode of action of severalbacterial toxins. Of these, the most important were thediscovery that the L-chains of neurotoxins possess zinc-dependent endopeptidase activity and confirmation thatneurotoxins act on nerve terminals by means of an en-

TABLE 1. In vivo bioassays for botulinum toxins and antitoxins based on inhibition of local muscular paralysis

Toxin A B C1 D E F

Paralysisa 0.2 1.0 0.2 0.8 1.0 0.2LD50/dose (dose range) (0.05–0.2) (0.1–1.0) (0.05–0.2) (0.2–0.8) (0.1–1.0) (0.02–0.2)

Antitoxin 1 1 1 100 5 10Detection Limit mU/ml (dose range) (0.4–12) (0.1–13) (0.7–20) (60–180) (0.7–20) (0.3–30)

aHighest dose of toxin inducing paralysis without systemic effects.

FIG. 1. Dose–response curves of international reference A and B antitoxin standards in a mouse muscular paralysis assay. A: Botulinum neurotoxinserotype A (BoNT/A; 0.2 LD50/dose. B: Botulinum neurotoxin serotype B (BoNT/B; 1 LD50/dose). At 24 hours (squares) and 48 hours (triangles).

ANTIBODIES AGAINST BOTULINUM TOXINS S87


zymatic process.39 Both botulinum and tetanus neurotox-ins target one or more of a set of proteins essential forneurotransmitter release from the synaptic vesicle.40 Aseach neurotoxin cleaves a specific bond within theirtarget protein, it is a highly selective mechanism ofaction. Several approaches for detecting the enzymaticaction of BoNTs have been investigated.41–45 ForBoNT/A, the most sensitive in vitro endopeptidase assaywas achieved by using targeted antibodies for immuno-detection of the toxin cleaved substrate.46,47 Using arecombinant peptide of SNAP-25 (synaptosomal-associ-ated protein of Mr � 25 kDa; amino acid residues134–206), it has been shown that the sensitivity of the invitro endopeptidase assay is better than that of the mouseassay in detecting active BoNT/A in therapeutic prepa-rations. Furthermore, since developing the endopeptidaseassay, more than 100 batches of therapeutic BoNT/Ahave been tested in parallel with the in vivo bioassay forbatch release testing and comparable potency estimatesobtained.48,49

As a result of these advances, a new generation of invitro assays capable of selectively detecting antibodies tothe catalytic domains of neurotoxins have also beendeveloped (Table 2). These may provide an alternative toin vivo assays for monitoring the potency of therapeuticantitoxins. It is unlikely that all of the antibodies, invaccinated individuals, that are responsible for the pro-tective response in vivo will be detected; however, theseassays provide specificity not shared with existing invivo methods, as different neurotoxins cannot be distin-guished in vivo, in the absence of toxin-specific anti-serum.

We, therefore, have investigated the utility of the toxinendopeptidase assay for the in vitro detection of toxinneutralising antibodies and have demonstrated thatBoNT/A in vitro endopeptidase activity is specificallyinhibited only by type A toxin antiserum but not byantiserum to other clostridial serotypes (Fig. 2).

Results indicate that in vitro inhibition of enzymaticactivity can provide comparable information to the invivo bioassay for quantification of toxin neutralisingantibodies,2 with sensitivity in the range of 5 to 50mIU/ml, depending on the antitoxin serotype (Table 2).

In more recent studies, the inhibition of in vitro endo-peptidase activity was able to provide information sim-ilar to that of an in vivo assay with respect to the specificneutralising activity of a new botulinum type A antitoxin,in units/ml, for both IgG and material processed toF(ab�)2, but with a potency ratio (IU/mg protein) forIgG:F(ab�)2 of 4:1 in both assays. Large losses of neu-tralising capacity can represent the use of poor F(ab�)2

processing methods, but recent advances in processingprocedures, however, should dramatically improve theseratios.50

Whilst the biological activity of botulinum neurotox-ins depends upon the integrity of all three of the func-tional domains, neutralising antitoxin activity can bedirected against a single functional epitope,32 such as thereceptor binding domain,51 or the toxin L-chain contain-ing the enzymatic activity.52 A protective response tobotulinum toxin has been reported after immunisationwith only the toxin portion containing the endopeptidaseactivity and lacking the H-chain binding domain31 andlikewise a protective response is also produced by im-munising with purified H-chain.53,54 A monoclonal anti-

TABLE 2. In vitro bioassays for botulinum toxins and antitoxins based on the inhibition of endopeptidase activity

Toxin A B C1 E F

Activitya 150 20 20 150 100LD50/ml (dose range) (0.4–200) (0.1–50) (0.1–50) (1.6–200) (1–100)

Antitoxin 10 10 5 20 50Detection limit mU/ml (standard curve range) (0.8–100) (2–100) (0.2–100) (2–100) (8–1000)

aToxin dose used in endopetidase assay for neutralization studies.

FIG. 2. Serotype-specific inhibition profile of in vitro endopeptidaseactivities. Reference A, B, E, and F antitoxin standards were premixedwith botulinum toxins before in vitro A (hatched bars) and B (shadedbars) endopeptidase assays.



body with partial ability to block BoNT/A activity invivo has been mapped to the toxin L-chain containing theenzymatic activity of the toxin that clearly neutralises theendopeptidase activity in vitro.52,55 But conversely, anti–H-chain monoclonal antibodies also neutralise the tox-in’s in vivo effects, and when three such antibodiesagainst different regions of the H-chain are combined,they are significantly more potent than the sum of theindividual antibodies.55,56 Ganglioside (GT1b) and syn-aptotagmin receptor binding assays have also been de-veloped; however, these assays lack the required sensi-tivity and have only been used for toxin binding studiesto date.57,58

IN VITRO NONFUNCTIONALIMMUNOASSAYS

Immunoassays such as Western/dot blotting, hapten-labelled elution, radioimmunoassay (immunoprecipita-tion), and enzyme-linked immunosorbent assays arewidely used for the detection of toxin specific antibodiesin serum samples both from vaccinated individuals, aspredictors of population immunity, and from vaccinatedanimals, as indicators of vaccine immunogenicity.9,59–65

These assays vary considerably in their sensitivity (Table3) and specificity, and it is not possible to compareassays in the absence of common reference reagents.Immunoassays also cannot distinguish between neutral-ising and nonneutralising antibodies,9 although for manyantigens, where the neutralising epitopes may also beimmunodominant, good correlation between immunode-tection and protective potency can be observed.66

The radioimmuno (immunoprecipitation) assay, likethe mouse diaphragm assay, has been shown to be capa-ble of detecting antibody in patients nonresponsive tobotulinum toxin therapy.62,63 The assay is simpler, faster,cheaper, and avoids the use of animals but unfortunatelyrequires radioisotope handling and is less sensitive thanthe mouse diaphragm assay. Further work is required toconfirm the clinical significance of low levels of neutral-ising antibody in patients still responsive to toxin ther-

apy, as these low levels may provide early indicators offuture therapy failure.63

CONCLUSIONS

With recent world events, the need for antitoxins tocombat the bioweapons threat has been highlighted alongwith their more traditional uses in treating food andinfant botulism poisoning. There currently is a highdemand for highly potent botulinum antitoxins and sub-sequently an increased interest in relevant testing meth-ods. Monitoring patient sera for neutralising antibodies isalso on the increase due to the expanding clinical use ofbotulinum toxins and their re-targeted counterparts.

Functional methods for the detection of botulinumantitoxins are essential for potency testing of therapeuticformulations. However, nonfunctional or surrogatemethods may be suitable for some applications, subjectto validation. But irrespective of the test method, appro-priate reference standards are essential to ensure consis-tency. A human reference standard (NIBSC 00/500) hasbeen prepared recently from volunteers immunized witha pentavalent (A–E) botulinum toxoid vaccine whichcontains a set amount of stable neutralising activity pervial (130–200 mIU and 60–100 mIU against type A andB toxins, respectively). This preparation should ideallybe adopted as a reference standard for human serologyassays.

Acknowledgment: We thank Defence, Science, and Tech-nology Laboratory (DSTL), Ministry of Defence, United King-dom, for supporting this work.

REFERENCES

1. Hibbs RG, Weber JT, Corwin A, Allos BM, Abd el Rehim MS,Sharkawy SE, et al. Experience with the use of an investigationalF(ab�)2 heptavalent botulism immune globulin of equine originduring an outbreak of type E botulism in Egypt. Clin Infect Dis1996;23:337–340.

2. Sesardic D, Martin H, Tierney R, Bigalke H. An in vitro assay fortesting of neutralising antibodies to botulinum toxin. In: Balls M,van Zeller A-M, Halder ME, editors. Progress in the reduction,refinement and replacement of animal experimentation. Amster-dam: Elsevier; 2000. p 1001–1008.

3. Arnon SS. Clinical trial of human botulism immune globulin. In:DasGupta BR, editor. Botulinum and tetanus neurotoxins: neuro-transmission and biomedical aspects. New York: Plenum Press;1993. p 477–482.

4. Jankovic J, Brin MF. Therapeutic uses of botulinum toxin. N EnglJ Med 1991;324:1186–1194.

5. Brin MF. Interventional neurology: treatment of neurological con-ditions with local injection of botulinum toxin. Arch Neurobiol(Madr ) 1991;54(Suppl. 3):7–23.

6. Borodic G, Johnson E, Goodnough M, Schantz E. Botulinum toxintherapy, immunologic resistance, and problems with available ma-terials. Neurology 1996;46:26–29.

7. Siegel LS. Human immune response to botulinum pentavalent(ABCDE) toxoid determined by a neutralization test and by anenzyme-linked immunosorbent assay. J Clin Microbiol 1988;26:2351–2356.

TABLE 3. Summary of methods used for the detection ofbotulinum antitoxins

MethodSensitivity(mU/ml)

Mouse lethality (in vivo) 10Endopeptidase activity 10Neuronal cells 10Mouse paralysis (in vivo) 1Mouse diaphragm 0.3Human foot muscle �1 (?)Immunoassays (nonfunctional) 0.3–1.0 (vary)



8. Siegel LS. Evaluation of neutralizing antibodies to type A, B, E,and F botulinum toxins in sera from human recipients of botulinumpentavalent (ABCDE) toxoid. J Clin Microbiol 1989;27:1906–1908.

9. Hanna PA, Jankovic J. Mouse bioassay versus Western blot assayfor botulinum toxin antibodies: correlation with clinical response.Neurology 1998;50:1624–1629.

10. Goschel H, Wohlfarth K, Frevert J, Dengler R, Bigalke H. Botu-linum A toxin therapy: neutralizing and nonneutralizing antibod-ies: therapeutic consequences. Exp Neurol 1997;147:96–102.

11. Duggan MJ, Quinn CP, Chaddock JA, Purkiss JR, Alexander FC,Doward S, et al. Inhibition of release of neurotransmitters from ratdorsal root ganglia by a novel conjugate of a Clostridium botuli-num toxin A endopeptidase fragment and Erythrina cristagallilectin. J Biol Chem 2002;277:34846–34852.

12. Chaddock JA, Duggan MJ, Hall YHJ, Kirby ER, Moulsdale HJ,Purkiss JR, et al. Retargeted clostridial endopeptidases: inhibitionof nociceptive neurotransmitter release in vitro, and antinocicep-tive activity in in vivo models of pain. Naunyn SchmiedebergsArch Pharmacol 2002;365:R15.

13. Torii Y, Tokumaru Y, Kawaguchi S, Izumi N, Maruyama S,Mukamoto M, et al. Production and immunogenic efficacy ofbotulinum tetravalent (A, B, E, F) toxoid. Vaccine 2002;20:2556–2561.

14. Casadevall A. Passive Antibody administration (immediate immu-nity) as a specific defense against biological weapons. EmergInfect Dis 2002;8:833–841.

15. European Pharmacopoeia. 4th ed. 2002. (Botulinum antitoxin2002:0085). Strasbourg: EDQM Council of Europe.

16. Sakaguchi G, Oishi I, Kozaki S, Sakaguchi S, Kitamura M. Mo-lecular structures and biological activities of Clostridium botuli-num toxins. Jpn J Med Sci Biol 1974;27:95–99.

17. Hatheway CH, Snyder JD, Seals JE, Edell TA, Lewis GE Jr.Antitoxin levels in botulism patients treated with trivalent equinebotulism antitoxin to toxin types A, B, and E. J Infect Dis 1984;150:407–412.

18. Sugiyama H, Brenner SL, Dasgupta BR. Detection of Clostridiumbotulinum toxin by local paralysis elicited with intramuscularchallenge. Appl Microbiol 1975;30:420–423.

19. Takahashi M, Kameyama S, Sakaguchi G. Assay in mice for lowlevels of Clostridium botulinum toxin. Int J Food Microbiol 1990;11:271–277.

20. Takahashi M, Komiya T, Kameyama S, Sakaguchi G. Titration ofbotulinum antitoxin of low levels by the score method. Jpn J MedSci Biol 1990;43:163–170.

21. Sakaguchi G, Sakaguchi S, Kondo H. Rapid bioassay for Clostrid-ium botulinum type-E toxins by intravenous injection into mice.Jpn J Med Sci Biol 1968;21:369–378.

22. Pearce LB, Borodic GE, First ER, MacCallum RD. Measurementof botulinum toxin activity: evaluation of the lethality assay. Toxi-col Appl Pharmacol 1994;128:69–77.

23. Pearce LB, Borodic GE, Johnson EA, First ER, MacCallum R. Themedian paralysis unit: a more pharmacologically relevant unit ofbiologic activity for botulinum toxin. Toxicon 1995;33:217–227.

24. Sesardic D, Mclellan K, Ekong TA, Das RG. Refinement andvalidation of an alternative bioassay for potency testing of thera-peutic botulinum type A toxin. Pharmacol Toxicol 1996;78:283–288.

25. Aoki KR. Preclinical update on BOTOX(R) (botulinum toxin typeA) purified neurotoxin complex relative to other botulinum neuro-toxin preparations. Eur J Neurol 1999;6(Suppl. 4):S3–S10.

26. Aoki KR. A comparison of the safety margins of botulinum neu-rotoxin serotypes A, B, and F in mice. Toxicon 2001;39:1815–1820.

27. Pearce LB, First ER, MacCallum RD, Gupta A. Pharmacologiccharacterization of botulinum toxin for basic science and medicine.Toxicon 1997;35:1373–1412.

28. Shone C, Appleton N, Wilton-Smith P, Hambleton P, Modi N,Gatley S, et al. In vitro assays for botulinum toxin and antitoxins.Dev Biol Stand 1986;64:141–145.

29. Ekong TA, Mclellan K, Sesardic D. Immunological detection ofClostridium botulinum toxin type A in therapeutic preparations.J Immunol Methods 1995;180:181–191.

30. Sesardic D. Requirements for valid alternative assays for testing ofbiological therapeutic agents. Dev Biol Stand 1996;86:311–318.

31. Sheridan RE, Deshpande SS, Smith T. Comparison of in vivo andin vitro mouse bioassays for botulinum toxin antagonists. J ApplToxicol 1999;19:S29–S33.

32. Amersdorfer P, Wong C, Smith T, Chen S, Deshpande S, SheridanR, et al. Genetic and immunological comparison of anti-botulinumtype A antibodies from immune and non-immune human phagelibraries. Vaccine 2002;20:1640–1648.

33. Dressler D, Bigalke H. Botulinum toxin antibody type A titres aftercessation of botulinum toxin therapy. Mov Disord 2002;17:170–173.

34. Shone CC, Melling J. Inhibition of calcium-dependent release ofnoradrenaline from PC12 cells by botulinum type-A neurotoxin.Long-term effects of the neurotoxin on intact cells. Eur J Biochem1992;207:1009–1016.

35. Boyd RS, Duggan MJ, Shone CC, Foster KA. The effect ofbotulinum neurotoxins on the release of insulin from the insuli-noma cell lines HIT-15 and RINm5F. J Biol Chem 1995;270:18216–18218.

36. Chaddock JA, Purkiss JR, Friis LM, Broadbridge JD, Duggan MJ,Fooks SJ, et al. Inhibition of vesicular secretion in both neuronaland nonneuronal cells by a retargeted endopeptidase derivative ofClostridium botulinum neurotoxin type A. Infect Immun 2000;68:2587–2593.

37. Pahlman S, Mamaeva S, Meyerson G, Mattsson ME, Bjelfman C,Ortoft E, et al. Human neuroblastoma cells in culture: a model forneuronal cell differentiation and function. Acta Physiol ScandSuppl 1990;592:25–37.

38. Hall YHA, Chaddock JA, Kirby ER, Moulsdale HJ, AlexanderFCG, Marks J, et al. Use of a cellular assay to detect neutralisingantibodies to botulinum neurotoxin. Naunyn Schmiedebergs ArchPharmacol 2002;365:R23.

39. Schiavo G, Poulain B, Benfenati F, DasGupta BR, Montecucco C.Novel targets and catalytic activities of bacterial protein toxins.Trends Microbiol 1993;1:170–174.

40. Sutton RB, Fasshauer D, Jahn R, Brunger AT. Crystal structure ofa SNARE complex involved in synaptic exocytosis at 2.4 Aresolution. Nature 1998;395:347–353.

41. Shone CC, Quinn CP, Wait R, Hallis B, Fooks SG, Hambleton P.Proteolytic cleavage of synthetic fragments of vesicle-associatedmembrane protein, isoform-2 by botulinum type B neurotoxin. EurJ Biochem 1993;217:965–971.

42. Schmidt JJ, Bostian KA. Endoproteinase activity of type A botu-linum neurotoxin: substrate requirements and activation by serumalbumin. J Protein Chem 1997;16:19–26.

43. Soleilhac JM, Cornille F, Martin L, Lenoir C, Fournie-Zaluski MC,Roques BP. A sensitive and rapid fluorescence-based assay fordetermination of tetanus toxin peptidase activity. Anal Biochem1996;241:120–127.

44. Corran PH, Sesardic D. A capillary electrophoresis-based assay ofbotulinum type A toxin proteolytic activity. Tranter, H. S. 167.1996. Proceedings of Biomedical Aspects of Clostridial Neurotox-ins International Conference, Oxford UK.

45. Schmidt JJ, Stafford RG, Millard CB. High-throughput assays forbotulinum neurotoxin proteolytic activity: serotypes A, B, D, andF. Anal Biochem 2001;296:130–137.

46. Hallis B, James BA, Shone CC. Development of novel assays forbotulinum type A and B neurotoxins based on their endopeptidaseactivities. J Clin Microbiol 1996;34:1934–1938.

47. Ekong TA, Feavers IM, Sesardic D. Recombinant SNAP-25 is aneffective substrate for Clostridium botulinum type A toxin endo-peptidase activity in vitro. Microbiology 1997;143:3337–3347.



48. Sesardic D, Corran PH, Mclellan K, Feavers IM, Ekong TAN. Invitro assays for estimating the activity of therapeutic preparationsof botulinum toxin. In: van Zutphen LFM, Balls M, editors. Ani-mal alternatives, welfare and ethics. Amsterdam: Elsevier; 1997. p1033–1038.

49. Sesardic D. Alternatives in testing of bacterial toxins and antitox-ins. In: Brown F, Hendriksen C, Sesardic D, Cussler K, editors.Advancing science and elimination of the use of laboratory animalsfor development and control of vaccines and hormones. Basel:Karger; 2002. p 103–110.

50. Jones RGA, Landon J. Enhanced pepsin digestion: a novel processfor purifying antibody F(ab�)(2) fragments in high yield fromserum. J Immunol Methods 2002;263:57–74.

51. Clayton MA, Clayton JM, Brown DR, Middlebrook JL. Protectivevaccination with a recombinant fragment of Clostridium botulinumneurotoxin serotype A expressed from a synthetic gene in Esche-richia coli. Infect Immun 1995;63:2738–2742.

52. Hallis B, Fooks S, Shone CC, Hambleton P. Characterization ofmonoclonal antibodies to botulinum type A neurotoxin. In: Das-Gupta BR, editor. Botulinum and tetanus neurotoxins. New York:Plenum Press; 1993. p 433–436.

53. Cenci Di Bello I, Poulain B, Shone CC, Tauc L, Dolly JO.Antagonism of the intracellular action of botulinum neurotoxintype A with monoclonal antibodies that map to light-chainepitopes. Eur J Biochem 1994;219:161–169.

54. Atassi MZ, Dolimbek BZ, Hayakari M, Middlebrook JL, WhitneyB, Oshima M. Mapping of the antibody-binding regions on botu-linum neurotoxin H-chain domain 855-1296 with antitoxin anti-bodies from three host species. J Protein Chem 1996;15:691–700.

55. Wu HC, Yeh CT, Huang YL, Tarn LJ, Lung CC. Characterizationof neutralizing antibodies and identification of neutralizing epitopemimics on the Clostridium botulinum neurotoxin type A. ApplEnviron Microbiol 2001;67:3201–3207.

56. Nowakowski A, Wang C, Powers DB, Amersdorfer P, Smith TJ,Montgomery VA, et al. Potent neutralization of botulinum neuro-toxin by recombinant oligoclonal antibody. Proc Natl Acad SciU S A 2002;99:11346 –11350.

57. Li L, Singh BR. Isolation of synaptotagmin as a receptor for typesA and E botulinum neurotoxin and analysis of their comparativebinding using a new microtiter plate assay. J Nat Toxins 1998;7:215–226.

58. Singh AK, Harrison SH, Schoeniger JS. Gangliosides as receptorsfor biological toxins: development of sensitive fluoroimmunoas-says using ganglioside-bearing liposomes. Anal Chem 2000;72:6019–6024.

59. Doellgast GJ, Brown JE, Koufman JA, Hatheway CL. Sensitiveassay for measurement of antibodies to Clostridium botulinumneurotoxins A, B, and E: use of hapten-labeled-antibody elution toisolate specific complexes. J Clin Microbiol 1997;35:578–583.

60. Palace J, Nairne A, Hyman N, Doherty TV, Vincent A. A radio-immuno-precipitation assay for antibodies to botulinum A. Neu-rology 1998;50:1463–1466.

61. Szilagyi M, Rivera VR, Neal D, Merrill GA, Poli MA. Develop-ment of sensitive colorimetric capture ELISAs for Clostridiumbotulinum neurotoxin serotypes A and B. Toxicon 2000;38:381–389.

62. Dressler D, Dirnberger G. Botulinum toxin antibody testing: com-parison between the immunoprecipitation assay and the mousediaphragm assay. Eur Neurol 2001;45:257–260.

63. Hanna PA, Jankovic J, Vincent A. Comparison of mouse bioassayand immunoprecipitation assay for botulinum toxin antibodies.J Neurol Neurosurg Psychiatry 1999;66:612–616.

64. Lindsey CY, Smith LA, West MW, Boles JW, Brown JE. Evalu-ation of a botulinum fragment C-based ELISA for measuring thehumoral Immune response in primates. Biologicals 2003;31:17–24.

65. Poli MA, Rivera VR, Neal D. Development of sensitive colori-metric capture ELISAs for Clostridium botulinum neurotoxin se-rotypes E and F. Toxicon 2002;40:797–802.

66. Winsnes R, Hendriksen C, Sesardic D, Akkermans A, Daas A.Serological assays as alternatives to the Ph Eur challenge test forbatch release of tetanus vaccines for human use. Dev Biol Stand1999;101:277–288.



detection of antibodies against botulinum toxins

Documents