Sheep Monoclonal Antibodies Prevent Systemic Effects of Botulinum Neurotoxin A1

Botulinum neurotoxin (BoNT) is responsible for causing botulism, a potentially fatal disease characterized by paralysis of skeletal muscle. Existing specific treatments include polyclonal antisera derived from immunized humans or horses. Both preparations have similar drawbacks, including limited supply, risk of adverse effects and batch to batch variation. Here, we describe a panel of six highly protective sheep monoclonal antibodies (SMAbs) derived from sheep immunized with BoNT/A1 toxoid (SMAbs 2G11, 4F7) or BoNT/A1 heavy chain C-terminus (HcC) (SMAbs 1G4, 5E2, 5F7, 16F9) with or without subsequent challenge immunization with BoNT/A1 toxin. Although each SMAb bound BoNT/A1 toxin, differences in specificity for native and recombinant constituents of BoNT/A1 were observed. Structural differences were suggested by pI (5E2 = 8.2; 2G11 = 7.1; 4F7 = 8.8; 1G4 = 7.4; 5F7 = 8.0; 16F9 = 5.1). SMAb protective efficacy vs. 10,000 LD50 BoNT/A1 was evaluated using the mouse lethality assay. Although not protective alone, divalent and trivalent combinations of SMabs, IG4, 5F7 and/or 16F9 were highly protective. Divalent combinations containing 0.5–4 μg/SMAb (1–8 μg total SMAb) were 100% protective against death with only mild signs of botulism observed; relative efficacy of each combination was 1G4 + 5F7 > 1G4 + 16F9 >> 5F7 + 16F9. The trivalent combination of 1G4 + 5F7 + 16F9 at 0.25 μg/SMAb (0.75 μg total SMAb) was 100% protective against clinical signs and death. These results reflect levels of protective potency not reported previously.

well-characterized product, devoid of contaminating proteins. The cost-effectiveness of this approach, however, relies on the generation of potent high affinity mAbs. The difficulty, however, has been identifying mAbs that protect against >1000 LD50 when administered at relatively low doses, i.e., <0.5 mg/kg.
Here, we describe the derivation, characterization and in vivo efficacy of six sheep monoclonal antibodies (SMAbs) derived from immunization with BoNT/A1 toxoid, HcC or LHn with or without subsequent challenge immunization with BoNT/A1 toxin. Alone, these SMAbs were found to be poorly protective; however, when administered in bi-or tri-valent combinations, selected SMAbs provided 100% survival against 10,000 LD50 BoNT/A1 when administered at doses as low as 0.75 μg/mouse or 0.0375 mg/kg.

Specificity for BoNT/A1 Hc and Lc
Western blot analysis was performed to evaluate the specificity of SMAbs 1G4, 2G11, 4F7, 5E2, 5F7 and 16F9 for BoNT/A1 Hc and Lc ( Figure 3). Based on the binding patterns revealed, SMAbs 2G11 and 5E2 bind BoNT/A1 Hc and SMAb 4F7 binds BoNT/A1 Lc. No binding was observed for SMAbs 1G4, 5F7 and 16F9, thus specificity for BoNT/A1 Hc and Lc could not be determined. Previous attempts to identify linear protective epitopes of BoNT/A1 suggest these epitopes are easily denatured and may be discontinuous or conformational [10]. Alternatively, SMAbs 1G4, 5F7 and 16F9 may be unable to bind insoluble BoNT/A1. Western blot analysis to evaluate SMAb BoNT/A1 Hc and Lc specificity. Bands corresponding to the Hc and Lc BoNT/A1 were identified based on published molecular weights [15]. Shown is a representative of three replicates.
1G4-bt or 5F7-bt Figure 5. Determination of SMAb pI by IEF. Shown is a representative of two replicates.

In Vivo Neutralization of BoNT/A1
Using the mouse lethality assay, a series of in vivo studies were performed with various combinations of SMAbs 1G4, 2G11, 4F7, 5F7, 5E2 and 16G9. Animals were scored for both survival and development of clinical signs consistent with botulism. An initial study indicated that in combination, at a dose of 10 µg each, SMAbs 1G4 + 2G11 + 4F7 + 5F7 + 5E2 (total SMAb dose = 50 µg/mouse), prevented both clinical signs and death vs. 10,000 LD50 BoNT/A1 (data not shown). Thus, a dose-response study was conducted in which the combination of SMAbs 1G4 + 2G11 + 4F7 + 5F7 + 5E2 was administered at doses of 1, 2 and 5 μg each (total SMAb dose = 5, 10 and 25 μg/mouse, respectively). Although mice that received 25 µg total SMAb were fully protected against development of clinical signs and death, those that received 10 μg total SMAb were protected against death, but developed moderate to severe clinical signs of botulism; only 20% of mice that received 5 µg total SMAb survived ( Figure 6, Panel A).
To determine whether all five SMAbs within this combination were necessary for protective efficacy, two studies were conducted in which one or two SMAbs were eliminated from the combination. For each of these studies, a dose of 4 μg of each SMAb was used vs. 10,000 LD50 BoNT/A1. The first study involved evaluating the effect of eliminating a single SMAb. Mice that received either all five SMAbs (1G4 + 2G11 + 4F7 + 5F7 + 5E2; total SMAb dose = 20 µg) or a combination of four SMAbs that included both 1G4 and 5F7 along with any two additional SMAbs were fully protected against development of both clinical signs and death. In contrast, mice which received a combination lacking both 1G4 and 5F7 did not survive, indicating that both 1G4 and 5F7 were necessary for protective efficacy ( Figure 6, Panel B). The second study involved evaluating the effect of eliminating two SMAbs. Mice that received 1G4 + 5F7 +/-any additional SMAb were fully protected against development of both clinical signs and death. In contrast, mice that received the combination of 2G11 + 4F7 + 5E2 did not survive. This study, therefore, demonstrated that SMAbs 1G4 and 5F7 in combination were both necessary and sufficient for protective efficacy ( Figure 6, Panel C).

Sheep Monoclonal Antibodies (SMAbs)
SMAbs were generated via standard methods [17] using lymphocytes isolated from the axillary (Sheep #1 and #2) or superficial cervical (Sheep #3 and 4) lymph nodes. Briefly, lymph nodes were removed and crushed to release lymphocytes into Dulbecco's Modified Eagle's Medium (DMEM) + 1% Penicillin/Streptomycin. The cell suspension was pelleted via centrifugation at 1200 rpm for 10 minutes to isolate the lymphocytes. Isolated lymphocytes were fused to 1C10, a proprietary, non-productive, sheep heteromyeloma, produced by inter-species back-fusion to the mouse myeloma NS1 (Bioventix, Ltd., Farnham Surrey, UK) by standard methods [17]. Supernatants from hypoxanthine-aminopterin-thymidine (HAT)-selected hybridomas were successively screened by ELISA on microtiter plates coated with 1 µg/mL BoNT/A1 and developed with AP-labeled rabbit anti-sheep IgG (Southern Biotechnology Associates #6150-04, Birmingham, AL) or donkey anti-sheep IgG (Sigma-Aldrich #A5187, St. Louis, MO). Stable, positive clones were selected by subconing at least twice by soft-agar cloning [17,18]. A total of six SMAbs, derived from lymphocytes isolated from the superficial cervical lymph nodes of Sheep #1-4, were selected for further study. Purified SMAb was isolated from the corresponding hybridomas culture supernatants via Protein A chromatography using a Prosep A column (Millipore, Billerica, MA). Purified SMAbs were quantitated via A 280 .

Western Blot Analysis of SMAbs
The subunit specificity of each SMAb was evaluated by Western blot. BoNT/A1 was electrophoresed by SDS-PAGE on a 12%, 10 x 14 cm acrylamide slab gel under reducing conditions and then electrophoretically transferred to a 0.2 µm nitrocellulose membrane (BioRad, Hercules, CA). Following electrophoretic transfer, the membrane was soaked 3 h in BLOTTO (2% non-fat milk + 5 mM NaN 3 + TBS-T). A 25 lane miniblotter apparatus (Immunetics, Boston, MA) was used to divide the membrane into discrete lanes. The membrane was washed five times (three times with TBS + 0.1% Tween-20, then twice with BLOTTO) between each of the following steps. SMAbs were incubated at 10 µg/mL in BLOTTO in individual lanes in the presence of membrane-bound BoNT/A1 for 2-3 h at room temperature. A 1:1500 dilution of AP-labeled rabbit anti-sheep IgG (Southern Biotechnology Associates #6150-04, Birmingham, AL) was added to each lane. The blot was developed using Sigma-Aldrich Fast™ BCIP/NBT substrate (Sigma-Aldrich, St. Louis, MO).

Sandwich ELISA
Sandwich ELISAs were performed to determine whether SMAbs 1G4, 2G11, 4F7, 5E2, 5F7 and 16F9 recognized similar or different epitopes on BoNT/A1. This ELISA was performed by coating microtiter plates with an SMAb, which was used to capture BoNT/A1. Biotinylated detection SMAbs (1G4-bt or 5F7-bt) were then serially diluted across the microtiter plate.

Murine Lethality Assay
The murine lethality assay was used to examine the ability of SMAbs either alone or in combination to neutralize the effects of BoNT/A1 in vivo. These studies were conducted at Tufts Cummings School of Veterinary Medicine under an approved animal care and use protocol. Groups of 5-10 female CD-1 mice ~20 g each (Charles River, Wilmington, MA) were used. One day prior to each study, mice were weighed and sorted to reduce inter-group weight variation. BoNT/A1 was dosed on a pg/g basis at 10,000 LD50 (calculated based on the LD50 dose determined via conventional mouse lethality assay; for the BoNT/A1 utilized in these studies, 1 LD50 = 0.9 pg/g mouse weight) using the average weight of all mice included within each experimental study. Combinations of SMAbs 1G4, 2G11, 4F7, 5E2, 5F7 and 16F9 at combined doses of 0-50 μg/mouse were evaluated. SMAb combinations were pre-incubated with BoNT/A1 in PBS containing 0.
Using a murine model of botulism, combinations of SMAbs 1G4, 2G11, 4F7, 5E2, 5F7 and 16F9 were evaluated for the ability to neutralize BoNT/A1 in vivo. All in vivo studies were performed at a dose of 10,000 LD50 BoNT/A1. Mice were scored for both survival and development of clinical signs. Absence of clinical signs and death were indicative of 100% protection; presence of clinical signs in the absence of death was indicative of partial protection.
Of the six SMAbs evaluated, only 1G4, 5F7 and 16F9 were found to contribute toward protective efficacy. Interestingly, SMAbs 1G4, 5F7 and 16F9 exhibited little or no binding to insoluble forms of BoNT/A1. In contrast, SMAbs 2G11, 4F7 and 5E2 appeared to bind plate-bound BoNT/A1 quite well, but possibly were unable to bind soluble BoNT/A1. Thus, the relative ability of 1G4, 5F7 and 16F9 vs. 2G11, 4F7 and 5E2 to contribute or not contribute to the protective efficacy in vivo may be related to the ability or inability, respectively, to bind soluble BoNT/A1, which would be the form present in vivo.
Similar to the findings of Nowakowski et al. [12] and Cheng et al. [13], these SMAbs were only highly protective when administered as divalent or trivalent combinations. The trivalent combination of SMAbs 1G4 + 5F7 + 16F9 provided 100% protection against both clinical signs and death when 1G4 + 5F7 + 4F7 + 2G11 + 5E2 at a dose of 10 μg/SMAb appeared to be less protective (mild clinical signs observed) than the combination of 1G4 and 5F7 at 2 μg/SMAb (no clinical signs observed).
Similar to studies utilizing mAbs C25, S25 and 3D12, recognition of a single epitope defined by SMAbs 1G4, 5F7 or 16F9 was not sufficient; instead, protective efficacy relied on recognition of two or more protective epitopes [12]. We were unable to determine Hc vs. Lc binding specifity of SMabs 1G4, 5F7 and 16F9. Neutralizing antibodies such as SMAbs 1G4, 5F7 and 16F9 that lack the ability to bind BoNT within the format of a Western blot have previously been described [20,21]. It has been hypothesized that lack of Hc or Lc recognition may be the result of binding conformational epitopes that are destroyed via SDS denaturation and/or are associated with regions at or near the Hc-Lc junction [21]. Given SMAbs 1G4, 5F7 and 16F9 belong to the same antigen specificity group, the epitopes recognized by these SMAbs are likely within a similar region. Recognition of adjacent or overlapping epitopes on the Hc has been hypothesized to provide increased blockade of the Hc receptor binding region [12,22].
The availability of protective SMAbs provides the opportunity for immunotherapy of individuals exposed to BoNT and prevention of subsequent development of botulism. A single dose of a protective divalent or trivalent SMAb preparation administered soon after exposure would be sufficient to prevent morbidity and/or mortality due to BoNT. SMAb administration is expected to be associated with adverse effects similar to that of equine antisera, including serum sickness and anaphylaxis. In addition, it is anticipated that induction of anti-sheep antibodies would limit use to once per lifetime. Unlike equine antisera, however, the SMAbs can be chimerized to produce a product that is unlikely to induce such adverse effects and that can be administered more than once. Thus, these SMAbs provide an unlimited, potentially safer alternative to both the currently available human and equine polyclonal BoNT therapeutics.