Anal. Chem. 2005, 77, 3916-3924
From the Mouse to the Mass Spectrometer: Detection and Differentiation of the Endoproteinase Activities of Botulinum Neurotoxins A-G by Mass Spectrometry Anne E. Boyer,*,† Hercules Moura,† Adrian R. Woolfitt,† Suzanne R. Kalb,† Lisa G. McWilliams,‡ Antonis Pavlopoulos,† Jurgen G. Schmidt,§ David L. Ashley,† and John R. Barr*,†
National Center for Environmental Health, Centers for Disease Control and Prevention, 4770 Buford Highway, NE, Atlanta, Georgia 30341-3724, Battelle Memorial Institute at the Centers for Disease Control and Prevention, and Los Alamos National Laboratory, Los Alamos, New Mexico 87545
We have developed an assay (Endopep-MS) that detects the specific endoproteinase activities of all seven BoNT types by mass spectrometry (MS). Each BoNT type cleaves a unique site on proteins involved in neuronal transmission. Target peptide substrates based on these proteins identify a BoNT type by its enzymatic action on the substrate and the production of two peptide products, which are then detected by matrix-assisted laser desorption/ionization time-of-flight MS or liquid chromatography electrospray ionization MS/MS. We showed the ability to detect all seven toxin types in a multiplexed assay format. The detection limits achieved range from 0.039 to 0.625 mouse LD50/mL for toxin types A, B, E, and F in a buffer system. The Endopep-MS assay is the first to differentiate all seven BoNT types, is sensitive, specific, and has the potential to quantify toxin activity.
Clostridium botulinum, Clostridium baratti, Clostridium butyricum, and Clostridium argentinense produce potent botulinum neurotoxins (BoNTs) which cause the paralytic illness botulism. Seven distinct BoNT types (A-G) are associated with botulism in a variety of animal species. Human illness is usually caused by toxin types A, B, E, and sometimes F, and occurs in three primary clinical forms: infant and adult intestinal colonization, foodborne, and wound.1,2 Toxin types A and B have also been used for therapeutic and aesthetic purposes. Botulism causes severe paralysis, which requires prolonged hospitalization and mechanical ventilation. Besides supportive care, postexposure therapy includes equine-based antineurotoxin immunoglobulins (IgGs), available from the CDC for emergencies.1 These antitoxins * Corresponding authors. E-mail:
[email protected]. For further information, contact: E-mail:
[email protected]. Phone: 770-488-7848. Fax: 770-488-4609. † Centers for Disease Control and Prevention. ‡ Battelle Memorial Institute under contract at the CDC. § Los Alamos National Laboratory. (1) Centers for Disease Control and Prevention. Handbook, Botulism in the United States, 1899-1996. Centers for Disease Control and Prevention, Atlanta, GA, 1998. (2) Werner, S. B.; Passaro, D.; McGee, J.; Schechter, R.; Vugia, D. J. Clin. Infect. Dis. 2000, 31, 1018-1024.
3916 Analytical Chemistry, Vol. 77, No. 13, July 1, 2005
neutralize the free toxin in the bloodstream and are most effective if administered within 24 h of symptom onset.1,3 The IgGs administered are polyvalent for the most commonly occurring types, A, B, and E. Polyvalent IgGs are essential since the toxin type is rarely known at this point. However, the main drawback to antitoxin use is that 9% of treated cases are complicated by serum sickness and anaphylaxis.1 Fortunately, botulism is rare with approximately 120 cases reported annually in the United States.2,4 However, even a single case represents a public health emergency, as it may foretell an outbreak. Because of their extreme toxicity, availability, and ease of preparation, BoNTs have been identified as a major biothreat agent.5 Therefore, prompt reporting, epidemiological investigation, and diagnostic confirmation are essential to determine the source, prevent additional cases, and provide optimal treatment. To date, the only widely accepted test for the identification of BoNTs in both clinical specimens and food is the mouse bioassay.6 One advantage of the mouse bioassay is that it is quite sensitive. It also provides the definition of toxicity, as mouse LD50 for the units (U) of toxin activity used for medical and research applications. It is this biophysical measure of toxin activity that has made the mouse indispensable for BoNT analyses. However, for all BoNT applications, the mice must weigh between 20 and 30 g, no more or less, to be moderately accurate in their measure of toxin activity.1 This stringent requirement for the age and size simply adds to the difficulties of using mice for BoNT analysis. Because mice quickly outgrow their diagnostic size, they cannot be stockpiled for emergency responses. Additionally, although mice often exhibit signs of botulism within a few hours after a BoNT sample injection, it requires 4 days to confirm a negative result, and it can take several days to determine the toxin level and the type. However, the primary drawback is that it requires (3) Caya, J. G.; Agni, R.; Miller, J. E. Arch. Pathol. Lab. Med. 2004, 128, 653662. (4) Shapiro, R. L.; Hatheway, C.; Swerdlow, D. L. Ann. Intern. Med. 1998, 129, 221-228. (5) 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.’ et al. JAMA, J. Am. Med. Assoc. 2001, 285, 1059-1070. (6) Kautter, D. A.; Solomon, H. M. J. Assoc. Off. Anal. Chem. 1977, 60, 541545. 10.1021/ac050485f CCC: $30.25
© 2005 American Chemical Society Published on Web 06/01/2005
many mice per specimen with a lethal end result. Of course, an in vitro alternative to animal use is preferred. One that can improve the efficiency and reduce the time required for testing. However, of the in vitro assays that have been designed, none meet the four assay requirements which include (1) measuring active toxin, with (2) sensitivity, (3) speed, and (4) specificity. Immunoassays for BoNT detection have been developed.7-10 However, only toxin types A, B, E, and F are validated and only for culture supernatants.10 Additionally, the immunoassays are not as sensitive as the mouse bioassay; there is cross-reactivity between toxin types; and false positives occur with nontoxigenic cultures.10 Furthermore, they fail to measure toxin activity, which limits their universal application for BoNT analysis. Activity-based BoNT assays have also been developed that are based on the different cellular activities of each toxin type.11-17 As a group, the BoNTs are zinc-dependent metalloproteinases. Produced as a nontoxic single chain, endogenous proteases cleave the progenitor toxin producing the dichain active toxin. The heavy chain is responsible for binding and internalization, and the light chain contains the zinc proteinase activity18 (Figure 1A). In the neuronal cytosol, the light chain cleaves specific proteins involved in forming the soluble N-ethylmaleimide-sensitive fusion attachment protein receptor (SNARE) complex. This complex is required for fusion of the synaptic vesicle with the presynaptic plasma membrane and communication between neurons18 (Figure 1B). Cleavage of any one of these proteins prevents the SNARE complex formation, which in turn inhibits acetylcholine release and results in flaccid paralysis. Although the different toxin types have the same neuronal cellular target, each type has a unique target site(s) of endoproteinase activity (Figure 1B). BoNT-A, -C, and -E cleave synaptosomal-associated protein (SNAP-25); whereas BoNT-B, -D, -F, and -G cleave vesicle-associated membrane protein (VAMP-2). BoNT-C is the only type reported to cleave two proteins; it also cleaves syntaxin.18,19 Several innovative activity-based assays reported during the past decade use protein or peptide BoNT type-specific substrates and monitor cleavage in different ways.11-17 One assay was able to detect low levels of BoNT-B in foods by antibody purification (7) Shone, C.; Wilton-Smith, P.; Appleton, N.; Hambleton, P.; Modi, N.; Gatley, S.; Melling, J. Appl. Environ. Microbiol. 1985, 50, 63-67. (8) Poli, M. A.; Rivera, V. R.; Neal, D. Toxicon 2002, 40, 797-802. (9) Ferreira, J. L. J. AOAC. Int. 2001, 84, 85-88. (10) Ferreira, J. L.; Maslanka, S.; Johnson, E.; Goodnough, M. J. AOAC. Int. 2003, 86, 314-331. (11) Hallis, B.; James, B. A. F.; Shone, C. C. J. Clin. Microbiol. 1996, 34, 19341938. (12) Ekong, T. A. N.; Feavers, I. M.; Sesardic, D. Microbiology 1997, 143, 33373347. (13) Wictome, M.; Newton, K. A.; Jameson, K.; Dunnigan, P.; Clarke, S.; Gaze, J.; Tauk, A.; Foster, K. A.; Shone, C. C. FEMS. Immunol. Med. Microbiol. 1999, 24, 319-323. (14) Wictome, M.; Newton, K.; Jameson, K.; Hallis, B.; Dunnigan, P.; Mackay, E.; Clarke, S.; Taylor, R.; Gaze, J.; Foster, K.; Shone, C. Appl. Environ. Microbiol. 1999, 65, 3787-3792. (15) Schmidt, J. J.; Stafford, R. G.; Millard, C. B. Anal. Biochem. 2001, 296, 130-137. (16) Liu, W.; Montana, V.; Chapman, E. R.; Mohideen, U.; Parpura, V. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 13621-13625. (17) Schmidt, J. J.; Stafford, R. G. Appl. Environ. Microbiol. 2003, 69, 297303. (18) Schiavo, G.; Matteoli, M.; Montecucco, C. Physiol. Rev. 2000, 80, 717766. (19) Wictome, M.; Shone, C. C. Symp. Ser. Soc. Appl. Microbiol. 1998, 27, 87S97S.
Figure 1. Botulinum neurotoxin structure and molecular targets. (A) The single-chain progenitor toxin is cleaved by endogenous proteases to the active dichain form consisting of the heavy chain, responsible for binding and internalization, and the light chain, carrying the zinc-dependent endoproteinase activity. (B) The synaptic vesicle carrying acetylcholine (Ach) neurotransmitter fuses with the neuronal plasma membrane to transmit the nerve impulse across the synaptic junction. This fusion requires formation of the SNARE complex, an interaction between VAMP-2 on the synaptic vesicle and Syntaxin and SNAP-25 on the neuronal membrane. The botulinum neurotoxins (BoNTs) cleave the peptide bonds at specific sites at on the three proteins. Cleavage of any one of these proteins prevents vesicle membrane docking and nerve impulse transmission.
of the toxin, followed by a peptide cleavage assay, and antibody detection of the cleaved product in an ELISA format.14 Another unique assay used fluorigenic peptide substrates for toxin types A, B, D, and F, which fluoresce when cleaved by the BoNT.17 These assays possess an excellent, rapid, and simple format, have proven useful for rapid screening of inhibitors, and have been applied to food for one toxin type.14 However, they have not yet been multiplexed for the detection of any toxin type in a single reaction or applied for BoNT detection and typing of unknown samples. The cleavage sites for BoNT-A and -C on SNAP-25 and BoNT-D and -F on VAMP-2 are separated by only one amino acid. It would indeed be a challenge to construct a single fluorigenic substrate that can differentiate two toxin types whose cleavage sites differ by only one amino acid, such as BoNT-A and -C and BoNT-D and -F. Also, antibodies, as used for cleaved peptide product detection of BoNT-B,14 could not be easily produced to differentiate cleaved product peptides differing by one amino acid. Although not difficult in a microtiter plate format, these assays must incorporate separate reactions for separate substrates and toxin types because the presence of fluorescence or ELISA-based colorimetric change does not intrinsically specify the BoNT type. The ELISA must also incorporate separate microtiter plate wells and the mouse bioassay separate mice, with separate type-specific antitoxin antibodies to determine the toxin type. Of all the in vitro assays developed to date, detection is possible for no more than four of the seven toxin types. We have turned to an analytical approach, mass spectrometry, which overcomes both challenges for other assays posed above, Analytical Chemistry, Vol. 77, No. 13, July 1, 2005
3917
differentiating between toxin types whose cleavage sites are separated by only one amino acid, and multiplexed detection of the activity and presence of any toxin type in a single reaction. Mass spectrometry is unique in its ability to distinguish between small molecular weight differences and overcomes the difficulty of distinguishing between cleaved product peptides differing by only one amino acid. The inherent mass specificity of MS, combined with well-designed substrate and predicted product peptides, also allows type detection multiplexed in a single reaction. The use of mass spectrometry also has the potential for quantification of toxin activity. Although BoNT-A, -B, -E, and -F typically affect humans and types C and D affect animals and birds, both C and D have been reported in humans.3 BoNT-G has only been implicated once in five human deaths, but its causative role is uncertain.20 When considering a possible intentional release, any toxin type could theoretically be involved. Therefore, we focused on incorporating all seven known toxin types in the Endopep-MS assay. Only four peptide substrates are needed to detect the endopeptidase activity of all seven toxin types. The substrates and type-specific cleavage products are detected by matrix-assisted laser desorption/ionization time-of-flight (MALDITOF) mass spectrometry (MS) and liquid chromatography (LC) electrospray ionization (ESI) tandem MS (MS/MS). While MALDITOF MS has been used previously for sequence characterization of actual BoNT complex proteins,21,22 this is the first published use of MS analytical methods for detecting BoNT endopeptidase activity. MATERIALS AND METHODS Materials and Safety Procedures. All BoNT complexes were provided at 1 mg/mL in 50 mM sodium citrate buffer, pH 5.5 (Metabiologics, Madison, WI). Toxin activities in mouse LD50 or units (U) of specific toxicity were from the provider as follows: BoNT-A at 3.6 × 107, BoNT-B at 1.6 × 107, BoNT-C at 6.0 × 107, BoNT-D at 2.8 × 107, BoNT-E at 2.8 × 107, BoNT-F at 5.5 × 107, and BoNT-G, at 1.75 × 104 U/mg, all at 1 mg/mL. All chemicals were from Sigma-Aldrich (St. Louis, MO) except where indicated. Peptides were synthesized by Los Alamos National Laboratory (Los Alamos, NM) and are listed in Table 1. We followed standard safety handling and decontamination procedures as described for botulinum neurotoxins,1 and because only very low toxin amounts were needed for this work, we maintained neurotoxin complex stocks below the 500-µg level for select agent registration. Endopep-MS Assay. Multiplexed reactions were performed by adding all four peptides 1-4 (Table 1) at 1 nmol each to reaction buffer (0.05 M Hepes, pH 7.3, 25 mM dithiothreitol, 20 mM ZnCl2, and 1 mg/mL bovine serum albumin), spiking BoNTs A-F at 0.20 µg, for a 20-µL final volume, and incubating at 37 °C for 2 h. BoNT-G complex was added at 1 µg and incubated overnight with peptides 1-4 and 7 h for a reaction with only peptide 3. BoNT-A and -B were assayed in one reaction as above. Pairs of toxins, types A and F or B and F were also assayed together using 1 µg of A or B and 0.01 µg of BoNT-F. A 1-µL (20) Sonnabend, O.; Sonnabend, W.; Heinzle, R.; Sigrist, T.; Dirnhofer, R.; Krech, U. J. Infect. Dis. 1981, 143, 22-27. (21) van Baar, B. L. M.; Hulst, A. G.; de Jong, A. L.; Wils, E. R. J. J. Chromatogr., A 2002, 970, 95-115. (22) van Baar, B. L. M.; Hulst, A. G.; de Jong, A. L.; Wils, E. R. J. J. Chromatogr., A 2004, 1035, 97-114.
3918
Analytical Chemistry, Vol. 77, No. 13, July 1, 2005
aliquot of each reaction was added to 19 µL of R-cyano-4hydroxycinnamic acid (CHCA) at 5 mg/mL in 50% acetonitrile, 0.1% trifluoroacetic acid, and 1 mM ammonium citrate (CHCA matrix), for analysis by MALDI-TOF MS as described below. A timed reaction was performed with 100 U of BoNT-A complex spiked into 20 µL of reaction buffer with 1 nmol of peptide 1 and 0.05 nmol of peptide 5 (Table 1) as an internal standard (ISTD). MS analysis of the reaction over time was achieved by taking 1-µL aliquots at timed intervals and spiking immediately into 19 µL of CHCA matrix, which stops the reaction at that moment. To differentiate BoNT concentrations, BoNT-A complex was spiked at 50 U up to 1000 U in reaction buffer with 1 nmol of peptide 1 and 0.05 nmol of peptide 5. Reactions were incubated for 2 h and stopped simultaneously by adding 180 µL of CHCA matrix. All samples in matrix were spotted and analyzed by MALDI-TOF MS. Assay absolute limits of detection (LOD) were determined by serial 10-fold dilutions of BoNT complexes A-F in reaction buffer from 100 down to 0.01 U/µL, and BoNT-G complex from 17.5 down to 0.5 U/µL. Then 1 µL of toxin at each dilution was added to 19 µL of reaction buffer containing only the specific peptide 1, 2, 3, and 4 for the diluted toxin type. These reactions were incubated at 37 °C for 17 h to ensure the reaction had gone to completion, then 2 µL was spiked into 18 µL of CHCA matrix, spotted, and analyzed by MALDI-TOF MS. Functional LOD were determined for BoNT-A, -B, -E, and -F by serially diluting each 1:1 in dH2O. Toxin types A, B, and E were diluted from 20 U/mL down to 0.039 U/mL, and type F was diluted from 25 U/mL down to 0.781 U/mL, all with a dilution at 1 U/mL included for a point at the mouse detection limit. A 168µL aliquot at each dilution was added to 10 times concentrated reaction buffer with 10 nmol of the type-specific peptide described, in a total 200-µL reaction incubated at 37 °C. Half of each reaction was stopped by the addition of 5 µL of 10% acetic acid at 4 and 10 h for BoNT-A and at 4 and 17 h for BoNT-B, -E, and -F, for analysis by MALDI-TOF MS and LC-ESI-MS/MS. For MALDI analysis, 2 µL of the reaction was diluted in 18 µL of CHCA matrix and then spotted for MALDI-TOF MS analysis. For all MALDI-TOF MS analyses, the reaction mix was diluted in CHCA matrix as indicated, applied at 0.5 µL/spot to a 192-spot MALDI plate (Applied Biosystems, Framingham, MA), and then mass spectra were collected from 650 to 4500 mass/charge (m/ z) or as described, in MS positive ion reflectron mode on the Applied Biosystems 4700 proteomics analyzer. The instrument uses a nitrogen laser at 337 nm, and each final mass spectrum was an average of 2400 laser shots. The LC-ESI-MS/MS system consisted of an API 4000 triple quadrupole instrument with a standard ion spray interface (MDS Sciex, Toronto, ON, Canada) and a Shimadzu (Kyoto, Japan) modular LC. All other LC and chromatography solvents and parameters used were described previously, using 1-mm C18 columns, a 50 µL/min flow rate, and a 33-min cycle time per sample.23 The sample injection size was 50 µL. Synthetic product and substrate peptides were used to determine and optimize mass transitions for each substrate and product peptides that were expected from cleavage by BoNT-A, -B, -E, and -F. At least two (23) Noort, D.; Fidder, A.; Hulst, A. G.; Woolfitt, A. R.; Ash, D.; Barr, J. R. J. Anal. Toxicol. 2004, 28, 333-338.
Table 1. Peptide Sequences for the BoNT Endopep-MS Assay, Including Substrates and Predicted Amino Terminal (NTP) and Carboxy Terminal (CTP) Peptide Cleavage Products and Accurate Monoisotopic Uncharged Masses for Each Toxin Typea
a Superscript letters denote the cleavage site for the toxin type. Substrate sequences for BoNT-A, -C, and -E were derived from the human SNAP-25 protein (P60880); peptide 1 (187-203) with modifications of biotin()-KG (K189fR and K201fR)GGK-()Biotin and peptide 2 (156-186). The substrate sequences for BoNT-B, -D, and -F are from human VAMP-2 (NP_055047); peptide 3 (59-93) and peptide 4 (35-74). Peptide 5 is used as an internal standard. Peptide 6 is the same as peptide 1 with two substitutions A195fG, M202fX (X ) norleucine).
specific transitions for each peptide, substrate, N-terminal, and C-terminal product peptides were monitored by ESI-MS/MS, although data are only included for one transition and product. (Mass ion transitions for LC-ESI-MS/MS analysis of BoNT-A, -B, -E, and -F are available as Supporting Information (S-1). RESULTS AND DISCUSSION Endopep-MS Assay Design. BoNTs are highly specific proteases, and three primary BoNT protein targets, SNAP-25, VAMP-2, and Syntaxin, are known. Specific BoNT cleavage sites and the minimal sequences necessary for BoNT binding and cleavage have been deduced.15,17,18,24 We took advantage of the BoNT specificity to design an assay that can detect and differentiate any one or multiple BoNT types simultaneously. The presence of all seven BoNT types, A-G, can be determined by virtue of their proteinase action on one of the four peptide substrates, shown in Table 1, along with expected BoNT type-specific cleavage products and predicted monoisotopic uncharged masses. The type(24) Vaidyanathan, V. V.; Yoshino, K.; Jahnz, M.; Dorries, C.; Bade, S.; Nauenburg, S.; Niemann, H.; Binz, T. J. Neurochem. 1999, 72, 327-337.
specific production of cleavage products with unique masses coupled with detection by high-resolution MALDI-TOF MS makes it possible to multiplex toxin type detection. Two SNAP-25-based peptides were used, one for BoNT-A and -C (peptide 1) and one for BoNT-E (peptide 2) (Table 1). Peptide 1 is a biotinylated version of the minimum substrate for BoNT-A with linker amino acids included. Since the cleavage sites for BoNT-A and -C are separated by only one amino acid, peptide 1 was a good substrate for both toxin types. The biotins were incorporated in peptide 1 for potential postreaction purification of the substrate and cleaved products. So far, this purification step has not been necessary. Interestingly, it was discovered that for BoNT-C the biotinylated peptide was a much better substrate than the identical sequence without biotins. Two VAMP-2-based peptides were used as substrates, one for BoNT-B and BoNT-G (peptide 3) and one for BoNT-D and BoNT-F (peptide 4). The cleavage sites for BoNT-D and -F are also separated by only one amino acid and for BoNT-B and G by five amino acids, which enables one substrate to be used for two toxin types in each case. Analytical Chemistry, Vol. 77, No. 13, July 1, 2005
3919
Figure 2. Multiplexed detection of BoNT-A-G by MALDI-TOF MS. Multiplex reactions containing all four peptides, 1-4, in BoNT reaction buffer, were spiked with 0.2 µg of toxin type(s); BoNT-A (A), BoNT-B (B), BoNT-C (C), BoNT-D (D), BoNT-E (E), BoNT-F (F), 1 µg BoNT-G (G), 0.2 µg each BoNT-A and BoNT-B (I), and H2O (No Toxin) (J). Peptide 3 alone was spiked with 1 µg of BoNT-G (H). The reactions were incubated 2 h, except for BoNT-G incubated 17 (G) and 7 h (H), spotted, and then analyzed by MALDI-TOF MS from m/z 650 to 4500, with resulting spectra shown. The substrate peaks (Sub), N-terminal (NTP) and C-terminal (CTP) peptide cleavage products are labeled and color coded by type. The inset mass spectrum for BoNT-E distinguishes the BoNT-E N-terminal cleaved peptide from the BoNT-A substrate. The Y-axis from m/z 3000 to 4500 was amplified for detail.
Multiplexed Reactions. MALDI-TOF mass spectra are shown for multiplexed reactions with each toxin type added separately to the four-peptide mix (Figure 2A-G) and a control reaction (Figure 2J). Mass spectra of the peptide substrates alone show that the relative intensity of the BoNT-A/C substrate, peptide 1 at 2911.5 m/z is much greater than the intensities of the larger substrates although they are present in equimolar concentrations (Figure 2J). By our experience, the larger peptides produce less intense MS signals by MALDI-TOF MS compared to smaller peptides, but besides the larger masses, specific sequence differences might contribute to the reduced intensities of these peptides. Because peptides are protonated during MALDI ionization, the singly charged peptide peaks were observed as (M + H)+ ions and thus are ∼1 m/z higher than the accurate mass listed in Table 1. Peaks corresponding to doubly charged (M + 2H)2+ ions are also present, as seen for the BoNT-A substrate at 1456.7 m/z (Figure 2J). The mass spectra of toxin type-specific reactions show that for each toxin types A-G, only mass peaks matching the expected products listed in Table 1 are present (Figure 2A-G). For example, only predicted product peaks at m/z 1714.88 and 1215.67 are evident when BoNT-A was present. Additionally, there is no cross reactivity observed between toxin types. After incubation with toxins A-F, the corresponding substrate peak is diminished or absent, which coincides with very high product peptide signals (Figure 2A-F, I). BoNT-G is the 3920 Analytical Chemistry, Vol. 77, No. 13, July 1, 2005
only exception. Because of the low specific toxicity of BoNT-G, a 17-h incubation was required for the reaction with four peptides, which yielded relatively low peptide product signals (Figure 2G). Besides the low toxicity level, the low response for BoNT-G may be due to some affinity of the toxin for the other VAMP-2-based substrate, peptide 4. In fact, when BoNT-G is added to peptide 3 alone, the intensity of the product peaks is higher in 7 h versus 17 (Figure 2H). The inability to distinguish the BoNT-B N-terminal product at m/z 1758.84 and the BoNT-F C-terminal product at m/z 1758.87 is important to note (Figure 2B, F). Conveniently, they can be distinguished by their respective C-terminal and N-terminal products which differ by 273 mass units (Figure 2B, F). This assay offers the distinct advantage that two cleavage products per toxin type rule out any such ambiguities. Future peptide design strategies will allow complete resolution of these two products. We tested the assay after combining two toxin types, BoNT-A and -B, which showed high peak intensity for all four product peaks (Figure 2I). Although the occurrence of two or more toxin types is not frequent, several documented examples of two types affecting humans exist and include any combination of BoNT-A, -B, and -F.25-28 In one fatal case, the strain was originally typed as (25) Hatheway, C. L.; McCroskey, L. M. J. Clin. Microbiol. 1987, 25, 23342338. (26) Fernandez, R. A.; Ciccarelli, A. S.; Arenas, G. N.; Gimenez, D. F. Rev. Argent. Microbiol. 1986, 18, 29-31.
A and later typed by mouse neutralization tests as 99% type A and a minor type F, designated Af.26 As is routine for botulism, this patient did not receive type F antitoxin. We also tested the ability to detect the two toxin types Af and Bf, present at a ratio of 99% A or B and 1% F, in one reaction. These reactions showed the presence of both toxin types within 2-h incubation time. (MALDI-TOF MS data can be viewed as Supporting Information (S-2).) Since F antitoxin is not standard therapy, mixed or single types that involve BoNT-F can represent a clinical emergency. Therefore, the early identification of less common toxin types may help guide appropriate adjunct antitoxin treatment. BoNT-A Toxin Activity Demonstrated by Relative Levels of Cleavage Products. The products of BoNT proteinase activity accumulate over time as the toxin continually cleaves available substrate. Reactions with 100 U of BoNT-A toxin complex and peptide 1 were used to illustrate this relationship, which shows an increase in both product peptides over time (Figure 3A). The product peptide peaks, emphasized in red, are visible within 15min incubation time with 100 U BoNT-A complex and are substantial at 60 min. The product peak areas alone cannot be used for accurate relative toxin levels because spot-to-spot variability in peptide signals from the same sample exists. Therefore, a peptide that is chemically similar to the BoNT-A product sequences, but is neither an inhibitor nor a substrate of the toxin, was used as an ISTD to normalize these differences (peptide 5, Table 1). The ISTD peptide was included at the reaction onset, and the ISTD peak, indicated in green, served as a constant throughout the course of the reaction (Figure 3A). The ratios of product peak areas to ISTD peak areas for each spot analyzed were plotted over time, and both gave R2 values close to 0.99 demonstrating a good fit to a second-order polynomial (Figure 3B), compared to an R2
