2.2.1. Analytical Assessment

As the monomerized ricin antigen proved to elicit high anti-ricin antibody titers in rabbits, which conferred high level protection to ricin-intoxicated mice, we prepared deactivated monomerized ricin antigen at large amount for the immunization of a horse while carefully monitoring the various steps comprising the ricin purification and monomerization process (Figure 1A). The source material, crude ricin, displays a two-main band appearance of approximately 120 and 65 kDa on SDS-PAGE, representing *Ricinus communis* agglutinin (RCA) and ricin, respectively (lane 2), while the gel-filtration purified ricin appears as a single ~65 kDa band (lane 3) and the monomerized ricin subunits, generated by reduction and alkylation of purified ricin, appear as a ~30 kDa band doublet, representing RTA and RTB (lane 4). Next, we examined whether the purified subunits retain the RTA-dependent catalytic and RTB-dependent lectin activities. To this end, the monomerized subunit mixture was applied to an α-Lactose-Agarose column. The monomerized ricin Ultra Performance Liquid Chromatography (UPLC) chromatogram (Figure 1(B-1)), contains three main peaks, at 150, 170 and 210 s. The flow through contained only the last major peak (Figure 1(B-2)), which could be related to alkylated-RTA. The two other peaks that were attached to the column were collected only upon elution with 0.5 M galactose (Figure 1(B-3)), indicating that these two peaks represent the RTB subunit, which retains its lectin activity. To appreciate the functionality of alkylated-RTA, we assessed the activity of the alkylated-toxin in a cell free system using the reticulocyte lysate-based transcription and translation (TnT) assay. In the TnT assay (Figure 1C), wherein the toxin's activity is quantified by its ability to inhibit ribosomal translation of mRNA encoding for luciferase enzyme [14,15], not only did monomerized ricin prevent luciferase from being synthesized (ED50 = 0.6 ng/mL), it was also found to be a more potent inhibitor (~7 fold) than native ricin (ED50 = 3.9 ng/mL). Preservation of the biological activity of the monomerized ricin subunits would imply that their structural conformations did not alter in any significant manner during the monomerization process. The functional conservation of the biological activities of the two alkylated subunits may lead one to expect that antibodies raised against the toxin monomers, will interact with the holotoxin, a mandatory prerequisite for passive immunization against ricin intoxications.

**Figure 1.** Characterization of the monomerized ricin vaccine. (**A**) SDS-PAGE analysis of the ricin purification and monomerization process (0.25 μg/lane) samples. Lane 1, Marker; Lane 2, Crude ricin; Lane 3, Purified ricin; Lane 4, Monomerized ricin. (**B**) UPLC chromatograms of alkylated-ricin and its purified subunits. 5 μL of the tested samples were injected into the UPLC columns (2.1 × 50 mm) and eluted at a flow rate of 0.4 mL/min. The chromatograms were monitored at 215 nm. 1. Alkylated-ricin preparation. 2. Alkylated-RTA. 3. Alkylated-RTB. (**C**) Catalytic activity assessment of ricin and monomerized ricin in a cell free system. The catalytic activities of purified ricin holotoxin (black line) and monomerized ricin (blue line) were determined using the transcription and translation (TnT) assay. Luminescence of untreated reticulocytes was considered as 100% protein synthesis. (**D**) In vitro activity of pure and alkylated ricin. Cultured HEK-293-AChE cells were incubated with increasing concentrations of pure (black line) and alkylated (blue line) ricin. The residual AChE activity in the culture medium was determined and expressed as the percent activity determined for untreated cells. (**E**) Toxicity following monomerized ricin administration to mice. Monomerized ricin (40 μg/kg body weight) was intraperitoneally administered to mice (*n* = 10), and body weights were determined at the indicated time points (blue line). Phosphate buffered saline (PBS) injected mice served as control (black line). Animals were observed for a 14 days period after alkylated ricin or PBS were injected.

#### 2.2.2. In-Vitro and In-Vivo Toxicity of the Monomerized Ricin

Unlike the catalytic activity of ricin in acellular expression systems such as the TnT-based analysis described above, which is dependent on the RTA unit itself, the catalytic activity of ricin in a cell-based system, requires the presence of intact dimeric toxin molecules, since it is the lectin function of the attached RTB which enables RTA internalization into cells. To evaluate the residual cellular toxicity of the monomerized ricin, cultured HEK-293-acetylcholinesterase (AChE) cells, which produce and secrete recombinant AChE to the medium in a constitutive manner [16], were incubated with different concentrations of native or monomerized ricin, and secreted AChE levels were determined. As seen clearly (Figure 1D), protein (AChE) synthesis inhibition by monomerized ricin was dramatically reduced, the IC50 of monomerized ricin being 260 fold higher than that of native ricin (4729 ± 573 versus 18 ± 3 pg/mL, respectively), indicating a loss-of-activity of >99.5%. Following this in vitro assay and prior to horse immunization, we conducted an in vivo safety test in mice. To this end, 40 μg/kg monomerized-ricin (a 10-fold higher dose than the intended initial equine-vaccination dose) was injected intraperitoneally to mice, and body weights were monitored for 14 days. No body weight loss was recorded within this period of time (Figure 1E), nor were any noticeable side effects observed (data not shown).

#### *2.3. Anti-Ricin Titer Buildup Following Horse Immunization*

Although the antigen was shown to be safe, as an extra precaution, we devised a vaccination protocol for immunizing a horse, comprising a low initial dose, followed by increasing doses of the monomerized ricin antigen. Serum samples were collected three weeks after each injection, to determine ELISA and neutralizing Ab titer buildup (Figure 2). Repeated vaccinations at intervals of 3 weeks resulted in increasing antibody titers up to 11 weeks, after which antibody titers began to level off. To induce a robust immunological response, we discontinued boosting while monitoring antibody titers on a monthly basis. During a period of 3 months in which the horse was not immunized, titer levels significantly decreased, from ~10<sup>5</sup> to ~104, after which immunization was recommenced with monthly doses of 10 mg monomerized ricin. This immunization regime led to the generation of anti-ricin antibodies at titer levels that were almost one order of magnitude higher than the titer before resuming immunization. Thus, following a vaccination period of 25 weeks, high and stable ELISAand neutralizing-antibody titers of 0.64 × 10<sup>6</sup> and 1.28 × 10<sup>6</sup> units/mL, respectively, were reached.

**Figure 2.** Titer buildup in the horse serum after vaccination with monomerized ricin immunization. The reactivity profile of the antibodies elicited by immunization were determined by enzyme linked immunosorbent assay (ELISA, black curve) and by in vitro ricin neutralization assay (blue curve).

#### *2.4. In Vitro and In Vivo Efficacy of the Anti-Ricin Antitoxin*

Pooled horse hyperimmune plasma served as the source material for the production of concentrated anti-ricin F(ab')2-based antitoxin. The motivation to use a F(ab')2 fragment as an antitoxin, was the potential of equine IgG antibodies to cause serum sickness [17], as the crystallizable fragment (Fc) present in IgG, may induce inflammation and unwarranted immune responses [18,19]. To characterize the F(ab')2-based antitoxin, we evaluated its neutralizing potency in the cultured HEK-293-AChE cell system. To this end, ricin was mixed with increasing concentrations of anti-ricin F(ab')2 and then added to the cells. As seen (Figure 3A), the antitoxin inhibited the activity of ricin and restored protein synthesis in a dose dependent manner. The antitoxin dose needed to neutralize 50% (ED50) of the toxin was approximately ~0.7 nM.

To determine the efficacy of this equine-derived antitoxin, mice were intranasally challenged with a lethal dose of ricin, and 24 h later were subjected to F(ab')2 antitoxin treatment via the intranasal route. Overall (Figure 3B), the treatment led to significantly high surviving ratios (62%). Since the clinical treatment of ricin-intoxicated subjects is expected to be via the intravenous route, an additional group was treated intravenously. As can be seen, the surviving ratios (65%) are practically the same as those obtained following intranasal treatment. To determine the efficacy of a late time intervention, mice were intranasally intoxicated with ricin and treated intravenously 48 h post exposure with the same horse antitoxin. Even at this late time point, considerable surviving ratios (34%) were obtained.

**Figure 3.** In vitro and in vivo efficacy of the anti-ricin F(ab')2-based antitoxin. (**A**) In vitro ricin neutralization. Ricin (2 ng/mL) was mixed with increasing concentrations of the antitoxin. The mixtures were added to cultured HEK-293-AChE cells, and the residual AChE activity in the culture medium was determined 18 h later. (**B**) Kaplan–Meier survival curves of mice intoxicated with ricin and subjected to anti-ricin antibody treatment. Mice intranasally intoxicated with ricin (7 μg/kg body weight) were not treated (black line; *n* = 14), treated intranasally at 24 h post exposure (blue line; *n* = 29), treated intravenously at 24 h post exposure (red line; *n* = 26), or treated intravenously at 48 h post exposure (green line; *n* = 38) with horse derived F(ab')2 anti-ricin antitoxin. Animals were observed for a 14-day period after ricin challenge.

#### *2.5. Horse Antitoxin Attenuates Ricin-Induced Pulmonary Damage Markers*

To determine the effect of horse derived F(ab')2 anti-ricin antitoxin administration on pathological markers, intranasally-intoxicated mice (7 μg ricin/kg) were intravenously treated 24 h later with the antitoxin and bronchoalveolar lavage fluids (BALFs) collected at 72 h post-intoxication were analyzed for damage markers. A prominent hallmark of pulmonary ricinosis in mice is the presence of exceptionally high levels of the pro-inflammatory cytokine interleukin-6 (IL-6) in BALF [20–23]. Remarkably (Table 2), treatment with the horse derived F(ab')2 anti-ricin antitoxin induced a sharp attenuation in IL-6 levels (~90% reduction); ~450 and ~3550 pg/mL of IL-6 were measured in the BALFs of antitoxin-treated and non-treated ricin-intoxicated mice, respectively.


**Table 2.** Pro-inflammatory and damage markers in the bronchoalveolar lavage fluids (BALFs) of mice following ricin intoxication.

\* *p* < 0.05 between tested group and naïve; \*\* *p* < 0.01 between tested group and naïve; && *p* < 0.01 in comparison to ricin intoxicated mice. a *n* = 4; b *n* = 6; c *n* = 5. BALF samples were collected from a naive mice, or b,c ricin intoxicated mice 72 h following exposure.

Altered lung fluid balance, leading to increased permeability pulmonary edema, is a major pathophysiological characteristic of intranasal ricin intoxication and high levels of protein were reported to be present in the BALF sampled from the inflamed lungs [6,21,22], indicating that the lung–blood barrier has been disrupted. Accordingly, edema markers in mice lungs were assessed. In addition to overall protein level, we have previously demonstrated that increased pulmonary edema can be well monitored by determining cholinesterase (ChE) levels in BALF. Normally, ChE is confined to the bloodstream, ye<sup>t</sup> appears at elevated levels in the BALF following disruption of the pulmonary

epithelial–endothelial barrier [20]. In the antitoxin-treated mice, total protein and ChE measurements in BALF collected at 72 h post-exposure, were significantly lower in comparison to untreated mice (63% reduction in total protein, from ~5.5 to ~2 mg/mL, and 78% reduction in ChE, from ~300 to ~70 mU/mL).

Previous studies carried out in our laboratory established that xanthine oxidase (XO), an oxidative stress marker, which may also contribute to edema formation, is dramatically elevated in BALFs of mice intranasally intoxicated with ricin, at 72 h post exposure [20–22]. We therefore measured XO levels in BALFs of ricin intoxicated mice that were treated with the horse antitoxin. Indeed, XO levels were significantly reduced, by ~60% following antitoxin administration (~1.7 and ~4.0 mU/mL were measured in BALFs of antitoxin-treated and non-treated ricin-intoxicated mice, respectively) (Table 2).
