The different serotypes of BoNT are defined by their ability to be neutralized by specific antitoxins; therefore, it is unusual for one antibody to bind multiple serotypes of botulinum neurotoxin. Monoclonal antibody 4E17.1 binds the translocation domain of BoNT, and residues Y750, Y753, E756, and E757 of BoNT/A play important roles in antibody binding. These residues are fairly well conserved among all of the known botulinum neurotoxins; hence, a single monoclonal antibody could bind most of the known botulinum neurotoxins at the same epitopic location.
All currently known subtypes of BoNT/A contain residues Y750, Y753, E756, and E757. BoNT/A1-/A4 were tested for their ability to bind monoclonal antibody 4E17.1. To test for the presence or absence of the extracted toxin on the antibody-coated beads, we used Endopep-MS. BoNT/A1-/A4 were all efficiently extracted from buffer with antibody 4E17.1, and the beads from those extractions all demonstrated the presence of active BoNT/A, meaning that all four of those BoNT/A subtypes bind the 4E17.1 antibody. Although BoNT/A5 was unavailable for testing, it is expected that this serotype will also effectively interact with 4E17.1 on the basis of its amino acid sequence (Y750, Y753, E756, and E757) 
. In fact, the amino acid sequence YNQYTEEE is identical in 63 BoNT/A strains.
Most subtypes of BoNT/B contain three of the four key amino acids in 4E17.1 binding; the exception to this is BoNT/B4, which contains all four amino acids. Five of the currently known subtypes of BoNT/B were available for testing, and all interacted with the 4E17.1 antibody. This interaction indicates that E756 is perhaps not as critical for binding as the other amino acids. E756 is a lysine in most of the BoNT/B serotypes, and a mutation from glutamic acid to lysine is not a conserved mutation. This non-conserved mutation may be responsible for the higher KD
associated with BoNT/B in general for this antibody () 
. Although BoNT/B6 was unavailable for testing, it is also expected to interact with 4E17.1 based on its amino acid sequence (Y750, Y753, K756, and E757) 
. As with BoNT/A, this epitope is highly conserved among 64 BoNT/B strains as YNIYSEKE.
All currently known subtypes of BoNT/E contain all four key amino acids in 4E17.1 binding. Subtypes BoNT/E1–E4 were available for testing, and all four interacted with 4E17.1. Subtypes BoNT/E5 and /E6 were unavailable for testing, but based on the amino acid homology 
, both are expected to interact with 4E17.1. Additionally, the examination of 55 BoNT/E strains demonstrates that this epitope is identical as YNSYTLEE. All currently known subtypes of BoNT/F contain Y750, Y753, and E757, three of the four key amino acids in 4E17.1 binding. Four of the seven subtypes were available for testing, and all four interacted with 4E17.1. This is further evidence that E756 is perhaps not as critical for binding as the other residues, because E756 is an aspartic acid, a conserved mutation, in all serotypes of BoNT/F. Subtypes BoNT/F3, /F4, and /F5 were unavailable for testing, but based on the amino acid homology 
, all are expected to interact with 4E17.1. Furthermore, 60 strains of BoNT/F have this identical epitope YNNYTSDE.
BoNT/C, /D, /G, and TeNT all contain Y750 and Y753, two of the amino acids known to be important in 4E17.1 binding. However, upon testing, it was determined that these toxins do not efficiently interact with the 4E17.1 antibody. Hence, these data show that a minimum of three of the four critical amino acids, Y750, Y753, and E757, are required for efficient binding. Furthermore, BoNT/G contains three of the four key amino acids in 4E17.1 binding including Y750, Y753 and E756, but it does not contain E757 and BoNT/G does not interact with 4E17.1. These additional data show that E756 is not critical for 4E17.1 interactions but that Y750, Y753, and especially E757 are required for efficient binding.
Endopep-MS was developed to detect the presence of BoNT in a clinical or food sample. An antibody affinity concentration/purification step prior to reaction with the substrate is critical to the success of the method when testing complex matrices such as serum or stool extracts. Prior to the discovery of the 4E17.1 antibody, a sample needed to either be split into 4 aliquots or extracted 4 separate times to test for the presence of four serotypes of BoNT. Either option negatively affected the Endopep-MS assay in terms of either sample volume requirements or time as each extraction step is approximately 1 hr. Therefore, the development of a single antibody which can be used to extract four serotypes of BoNT simultaneously, especially the four serotypes commonly involved in human botulism, is beneficial in the Endopep-MS assay to detect BoNT and would be beneficial in other assays which use antibody-affinity for BoNT detection.
We have tested the 4E17.1 antibody with spiked clinical and food samples such as serum, stool extracts, and milk and found that this antibody is particularly effective with serum and milk. For stool extracts with a severe protease composition, we have had recent success with 2M sodium chloride washes to remove excessive proteases from the antibody-coated beads (manuscript in preparation). In the presence of such extreme conditions, BoNT/B and /F in particular unfortunately are removed from the antibody-coated beads, due to the higher dissociation rates 
listed in . Thus, the use of the 4E17.1 antibody may be limited to stool extract samples which contain fewer proteases, such as infant stools. Further studies employing real-world clinical specimens are planned when such samples become available to us. Nevertheless, as the LOD of BoNT/A, /B, /E, and /F are equivalent or near equivalent to those obtained with other serotype-specific antibodies used for toxin extraction, this new antibody offers the opportunity to test for any of those toxin types in an unknown sample.
The opinions, interpretations and recommendations are those of the authors and are not necessarily those of the Centers for Disease Control and Prevention or the US Army.