There are two major groups of antidotes, antibodies and chemical compounds. The history of using antibodies as effective antidotes against toxins can be traced back to 1890 [29
], when antiserum from a tetanus-immune animal protected tetanus toxin-mediated mortality of naïve animals. Since then, antibodies have played a pivotal role in neutralizing toxins [30
]. There are several advantages for antibodies as antidotes as compared to the chemical antidotes [32
]. Firstly, antibodies have a long half-life in the body. Secondly, antibodies are natural and nontoxic products. Lastly, current biotechnology allows the development of antibodies possessing a defined specificity against most toxins.
In order to develop highly potent antiricin neutralizing mAbs, mice have been needed to be immunized by an immunogenic dose of ricin, typically 5μ
g per mouse. However, 5μ
g ricin is equivalent to 25 × LD50 if administered as a primary immunization dose to the mouse. We have observed that mice could survive a large dose of ricin poisoning if the mice were poisoned by a stepwise increase of the ricin dose. In this way, mice were immunized by the i.p. route with ricin from 0.2 × LD50 to 25 × LD50 and a high antiricin antibody titer was obtained (data not shown).
The production of mAbs using hybridoma technology was invented by George Köhler and César Milstein in 1975 [35
]. Since then, this hybridoma technology has been improved greatly by different novel methods for fusing, growing, selecting, cloning, and screening hybridoma clones. Even so, the selection, cloning, and screening are often regarded as the main bottleneck in the development of effective hybridoma clones for antibodies. In our study, we combined a methylcellulose-based semisolid hybridoma selective medium with ricin (2.5
ng/mL) for hybridoma selection, cloning, and screening in one single step to arrive at the desired ricin resistant hybridoma clones. In the unique step, single-hybridoma cells were distributed evenly in the semisolid medium and only ricin resistant single-hybridoma cells could grow to form monoclonal colonies. The ricin resistant hybridoma clones were then transferred into 96-well plates for further selection in liquid medium with a higher concentration of ricin (5
ng/mL). As a result, 25 hybridoma clones were resistant to two cycles of ricin poisoning (semisolid and liquid medium). Twenty clones secreted antiricin antibodies and among these, 12 clones secreted antiricin neutralization mAbs. The result is an improvement over traditional approach to develop antiricin neutralizing antibody hybridomas [24
]. The best 4 antiricin neutralizing hybridoma clones were selected, all of which were found to be RTB specific. Although RTB itself is not toxic, it plays a pivotal role in ricin toxicosis. RTB binding to galactose residues on the cell surface is involved in not only triggering cellular uptake of ricin [2
], but also facilitating transport of RTA from the ER to the cytosol [3
], where RTA exerts enzymatic toxicity. Theoretically, RTB is the logical target for neutralizing antibodies, as these would block the entry of ricin into cells and the transportation of RTA to the cytosol. However, it seems to be more difficult to develop anti-RTB neutralizing mAbs than anti-RTA neutralizing mAbs. One of the reasons is that to date the immunodominant epitopes on RTB have been found not to provide neutralizing protection. In other words, RTB is poor in elicit antiricin neutralizing antibodies although it is highly immunogenic in eliciting nonneutralizing antiricin antibodies [24
]. To date, only a few anti-RTB neutralizing antibodies have been reported [20
RTB is a galactose-specific lectin with 262 residues folded into two globular domains. Each domain is formed by similar folding topologies via two intrachain disulfide bridges and responsible for binding to one terminal galactose residue on the cell surface [28
]. The two galactoside binding pockets are structurally similar and formed by a sharp bend in Asp-Val-Arg tripeptide [26
]. The antigen-binding sites of antibodies are much bigger than RTB galactose binding pockets. Interestedly, all our anti-RTB mAbs did not bind to RTB when the four intrachain disulfide bridges in RTB were broken and the domain structures were disturbed, indicating that all four mAbs most likely bind to conformational epitopes on RTB. It is somewhat perplexing how a RTB-specific neutralizing antibody achieves ricin neutralizing function, given that RTB has two galactose binding sites that work independently and are separated by distance [39
]. Since the two RTB domains are homologous and structurally similar, it is possible that our RTB-specific mAbs bind to conformational epitopes sharing the resemblance between two domains and covering the galactoside binding pockets so as to block both two galactose binding pockets and then interrupt ricin binding to cells. In addition, another possibility also exists that our mAbs bind to somewhere else but not galactose binding pockets of RTB and then interrupt the transport of RTA from the ER to the cytosol. Nevertheless, these hypotheses need further experiments to confirm.
The in vivo
protective efficacy for the four mAbs was first evaluated in a coincubation assay. All of them showed the protection of mice against ricin poisoning. The therapeutic efficacy of the four mAbs for postexposure therapy was then examined in vivo
. The therapeutic efficacy of antiricin antibody-based treatment is largely dependent on timing of administration of rescuing antibody relative to exposure. A relatively wide therapeutic window will provide necessary time for exposed individual to obtain antiricin antibody treatment in the event of a ricin attack. Therefore, the therapeutic window was determined for each antiricin neutralizing mAb. Although all four mAbs showed postexposure therapeutic functions against ricin poisoning, their therapeutic windows were different. When antibody dose was 5μ
g, the best was D9, then B10, followed by A9 and D3 in order. The four mAbs were further characterized. They were isotyped using a mouse IsoStrip kit and all the mAbs showed the same subtype of heavy chain, gamma 1, and the same type of light chain, kappa. Their ricin binding affinities were measured by SPR and they had different KD
s ranging from 2.55 to 36.27
nM. The highest was D9, then D3, followed by B10 and A9 in order. In addition, synergistic ricin neutralization effects among different mAbs were evaluated by pairs in an in vitro
cell-based assay. All of these showed synergistic effect when paired with others. Taken together, the four antiricin RTB neutralizing mAbs appeared different. Their different ricin neutralization activities were more related with their epitope specificities than ricin-binding affinities and not related with their antibody isotype.
Ricin acts very fast and leaves a very short therapeutic window (effective timing of administration of therapeutic antibodies) for postexposure medical countermeasures. The only way to improve the chance of success to rescue subjects from ricin-intoxication is to develop highly potent neutralizing antibody. To date, dozens of antiricin neutralizing mAbs evaluated in vivo
have been reported and it is hard to compare their therapeutic efficacies [20
]. There are many factors, which are attributed to the efficacy outcome, such as mouse strains (inbreed, outbreed) [20
], sex (male, female) [21
] and age (6–8 weeks, 8–12 weeks) [41
], rich challenge doses (5 × LD50, 10 × LD50) [25
] and routes (i.p., intranasal) [20
], mAb compositions (single, cocktail) [20
], and routes (i.p., intravenous) [21
]. These factors should be taken into account when the therapeutic efficacy among different mAbs from different laboratories is compared. There are several publications regarding antiricin neutralizing mAbs evaluation in vivo
in a similar setting to ours, i.p. route used for both ricin challenge and antibody administration [17
]. The best reported result has been that the administration of 10μ
g of antiricin mAb GD12 per mouse as much as 6 hours after ricin challenge (5 × LD50) rescued 100% mice from toxin-induced death over a 3-day period of observation [41
] and 5μ
g GD12 protected mice 24 hours before the ricin challenge [17
]. D9 is twice as potent as GD12 in a postexposure therapeutic setting and much more potent in a preexposure prophylactic setting.
It is necessary to humanize murine mAbs for clinical applications since these antibodies have a serious problem in humans, which is serum sickness due to foreignness to humans [43
]. Currently, therapeutic settings, using antibody-based drugs, require a large dosage (hundreds mg) and multiple doses. As a result, animal antibody's immunogenicity in humans is a critical concern. Repeating administration of these mAbs may result in rapid clearance of the animal antibodies in humans and anaphylaxis, which can sometimes be fatal [44
]. Our data demonstrated that only a very little dose of our antibody D9, such as 5μ
g per mouse (equivalent to 1.4
mg per person), could rescue 100% mice 6 hours after ricin challenge (5 × LD50) and protected mice 6 weeks before the ricin challenge. Although one administration of mAb D9 is impractical to prophylactically protect the public against ricin for a long period of time, it is practical to prophylactically protect first responders and military personnel to entering ricin-contaminated zones to perform their duties within six weeks. Therefore, D9 is an excellent candidate to be humanized as a potent antidote against ricin poisoning for both prophylactic and therapeutic purposes.