Equal Neutralization Potency of Antibodies Raised against Abrin Subunits

Abrin and ricin are potent AB toxins, which are considered biological threats. To date, there are no approved treatments against abrin or ricin intoxications. Previously, we showed that the administration of polyclonal anti-abrin antibodies to mice that were intranasally exposed to abrin, even very late post-exposure, conferred an exceedingly high-level of protection, while following ricin intoxication, similar treatment with anti-ricin antibodies resulted in negligible survival rates. To probe this unexpected difference in protection ability, we first examined whether the efficient anti-abrin-induced protection was due to neutralization of the A-subunit responsible for the catalytic effect, or of the B-subunit, which enables binding/internalization, by evaluating the protection conferred by antibodies directed against one of the two subunits. To this end, we generated and immunized rabbits with chimeric toxins containing a single abrin subunit, AabrinBricin in which abrin A-subunit was linked to ricin B-subunit, and AricinBabrin in which ricin A-subunit is linked to abrin B-subunit. Here, we show that antibodies raised against either AabrinBricin or AricinBabrin conferred exceptionally high protection levels to mice following intranasal exposure to a a lethal dose of abrin, suggesting that the high level of protection conferred by anti-abrin antibodies is not related to the neutralization of a particular subunit.


Introduction
Abrin and ricin, plant toxins derived from rosary peas (Abrus precatorius) and castor bean plant (Ricinus communis), respectively, belong to the type 2 ribosome inactivating (RIP-2) group of proteins. The heterodimeric glycoprotein of both toxins comprises subunit A, carrying the enzymatic activity of the toxin, and subunit B, which is a lectin that binds to eukaryotic cell surfaces via galactose residues and mediates toxin internalization [1]. Upon entry to the cytosol, the A-subunits of both ricin and abrin inactivate ribosomes irreversibly by site-specific depurination of a specific adenine nucleotide within the 28S rRNA, thereby precipitating cessation of cellular protein synthesis [2]. Due to their high availability, ease of preparation and high toxicity, ricin and abrin are considered potential bioterror agents and are classified as Category B agents by the US Center for Disease Control and Prevention (CDC). To date, there are no approved post-exposure therapeutic countermeasures for either toxin.
The toxicity of ricin and abrin depends on the route of exposure, inhalatory exposure being considered the most fatal [2]. The clinical manifestation following intranasal exposure of mice to these toxins is the onset of a localized yet severe pulmonary edematous inflammation, which is accompanied by massive recruitment of neutrophils to the lungs and onset of a turbulent pro-inflammatory cytokine storm within this organ. Pulmonary (intranasal) ricin and abrin intoxications in mice, are similar with regard to pathological features and kinetics [3,4]. However, despite their resemblance in morbidity and mortality, the ability to protect mice against ricin and abrin intoxications by post-exposure antibody-mediated treatment differs radically. In the case of lethal ricin intoxication, rabbit-derived polyclonal anti-ricin antibody-based treatment up to three hours post exposure conferred protection to ≥90% of the mice, nonetheless following treatment at 24 hours survival rates declined dramatically, to 34% [3]. When antibody treatment was administered at 48 hours post exposure to ricin, protection was no more than marginal [5]. In sharp contrast, the administration of polyclonal anti-abrin antibodies to mice intranasally exposed to a lethal dose of abrin, led to very high survival rates (~70%-80%), even when the antibodies were applied as late as 72 hours after intoxication [4].
The goal of the present work was to elucidate whether the efficient protection by polyclonal anti-abrin antibodies can be attributed to the neutralization of a single subunit, in other words, to define the critical neutralizing event that confers high-level of protection: prevention of toxin binding via the B subunit or abolishment of the A subunit-related catalytic activity. A specific conclusion could assist us, for example, to produce a subunit-directed abrin vaccine, which would be as efficient as holotoxin-based vaccine, but much safer. In addition, determining the subunit whose neutralization is critical for protection could assist in designing a more effective anti-ricin immunization protocol.
Production of effective polyclonal antibodies against a specific subunit of the abrin (or ricin) toxin is complicated by the tendency of isolated subunits to adopt an unstable conformation (in particular, B abrin is insoluble in aqueous solutions when separated from the A abrin , due to its hydrophobicity [6]). To overcome this difficulty, we incorporated each of the abrin subunits into a dimeric AB molecular fold by generating two chimeric heterologous toxins, A abrin B ricin and A ricin B abrin , utilizing the monomeric subunit constituents of abrin and ricin. These chimera toxins served as antigens for rabbit vaccination, which resulted in elicitation of antibodies against each of the anti-abrin-subunits.
In the present study, we found that anti-A abrin and anti-B abrin antibodies were equally effective in recognizing and neutralizing abrin in-vitro. Highly efficient protection of mice against pulmonary abrin intoxication, comparable to that conferred by anti-abrin holotoxin antibodies, was obtained by post-exposure application of either one of the anti-subunit directed antisera. These findings strongly suggest that the neutralization of either A abrin or B abrin is sufficient for achieving enhanced protection.

Extraction and Purification of Abrin and Ricin
Crude abrin was prepared from Abrus precatorius seeds essentially as described previously [7,8]. Briefly, seed kernels were soaked in 5% acetic acid/phosphate buffer (Na 2 HPO 4 , pH-7.4) overnight and homogenized in a Waring blender. Proteins obtained from 80% ammonium sulfate precipitation were centrifuged and dialyzed extensively against PBS.
Crude ricin was prepared from seeds of endemic Ricinus communis, essentially as described before [9]. Briefly, seeds were homogenized in a Waring blender in 5% acetic acid/phosphate buffer (Na 2 HPO 4 , pH-7.4), the homogenate was centrifuged and the clarified supernatant containing the toxin was subjected to ammonium sulfate precipitation (60% saturation). The precipitate was dissolved in PBS and dialyzed extensively against the same buffer.
For the preparation of purified toxins, crude abrin and ricin were loaded consecutively on 2 columns, the first column containing activated Sepharose which binds and thereby depletes the Abrus precatorius Agglutinin (APA) and ricinus communis agglutinin (RCA), respectively, and the second column, containing α-lactose (lactamyl) agarose (Sigma-Aldrich, Rehovot, Israel), which binds the toxins. Proteins bound to the lactamyl agarose column were eluted with 0.5M Galactose in PBS.

Subunit Purification
Purified abrin was loaded on a lactamyl agarose column (1 mg pure abrin:2 mL lactamyl agarose) and the column was washed with 1M Trizma-HCl pH-8 (Sigma-Aldrich, Rehovot, Israel), which was diluted 1:10 in PBS (abrin washing buffer). For toxin reduction and collection of A abrin , the column was rinsed with 1% v/v β-mercaptoethanol in abrin washing buffer. Each 1 mL A abrin was neutralized with 25 µL of 1M KH 2 PO 4 . The column was washed with PBS (pH 7), and B abrin was eluted with 0.5M galactose in the presence of 1% v/v β-mercaptoethanol. A abrin and B abrin were concentrated in 10 kDa cutoff Amicon-Ultra centrifugal filters.
Purified ricin was loaded on a lactamyl agarose column (1 mg pure ricin:1 mL lactamyl agarose) and the column was washed with 1M Trizma-HCl pH-9 (Sigma-Aldrich, Rehovot, Israel) diluted 1:10 in PBS (ricin washing buffer). For toxin reduction and the collection of A ricin , the column was rinsed with 1% v/v β-mercaptoethanol (Sigma-Aldrich, Rehovot, Israel) in ricin washing buffer. Each 1 mL A ricin was neutralized with 100 µL of 1M KH 2 PO 4 . The column was then washed with PBS (pH-7), and B ricin was eluted with 0.5M galactose (Sigma-Aldrich, Rehovot, Israel), in the presence of 1% v/v β-mercaptoethanol. A ricin and B ricin were concentrated in 10 kDa cutoff Amicon-Ultra centrifugal filters (Mercury, Rosh Haayin, Israel).
All subunit purification processes were conducted at 4 • C, and the subunits were stored under this temperature until further used.

Preparation of Chimeric Toxins
Chimera A abrin B ricin preparation: the subunits were mixed (20% excess A abrin ) and extensively dialyzed (10 kDa cutoff) against 10mM phosphate buffer (1M phosphate buffer solution, pH-7.4, diluted 1:100 in DDW) containing 100 mM galactose, at 4 • C for 7 days, and additional 2 days against the same buffer without galactose. A abrin B ricin was loaded on a lactamyl agarose column and the column was washed with PBS in order to remove residual monomeric A abrin . A abrin B ricin eluted with 0.5M galactose in PBS was dialyzed against PBS and kept frozen until used.
Chimera A ricin B abrin preparation: the subunits were mixed (20% excess Aricin) and extensively dialyzed (10 kDa cutoff) against PBS for 2 days at 4 • C. A ricin B abrin was loaded on a lactamyl agarose column and the column was washed with PBS in order to remove residual monomeric A ricin . A ricin B abrin eluted with 0.5M galactose in PBS was dialyzed against PBS and kept frozen until used.

Gel Electrophoresis
Samples were visualized using Coomassie Blue stained non-reducing 10% polyacrylamide gel, which was subjected to sodium dodecylsulphatepolyacrylamide gel electrophoresis (SDS-PAGE) under reducing or non-reducing conditions.

In-Vitro Abrin Neutralization Assay
Ub-FL cells (HeLa cells stably expressing ubiquitin-luciferase [10]), a kind gift from Professor Piwnica-Worms (University of Texas, MD Anderson Center, Austin, TX, USA), were seeded in a 96-well plate (1.5 × 10 4 cells/well) and incubated over night at 37 • C. Cell-culture medium was removed and serial dilutions of anti-holotoxin abrin , anti-A abrin , or anti-B abrin antisera in the presence or absence of abrin (7 ng/mL) were added to the plate. Twenty four hours later, cell-culture medium was removed, and the cell medium containing the proteasome inhibitor MG132 (Sigma, C2211 1 µM) was added for another hour. After medium aspiration, cells were lysed by the addition of 50 µL lysis buffer (Promega, E1941). Residual luciferase activity was determined by mixing equal volumes of the cell lysate and luciferase assay reagent (Promega, E1483) followed by immediate measurement of the luminescence levels. Residual luciferase activity was expressed as percent of luminescence levels obtained from untreated cells.

Epitope Recognition
Maxisorp 96-well microtiter plates were coated overnight with 5µg/ mL Streptavidin (Sigma, S0677), then washed and blocked as described above. Biotin labeled 15aa length peptides (Table S1) (5 µg/mL in PBST) were then added for 20 minutes. The plates were washed, anti-holotoxin abrin /anti-A abrin B ricin /anti-A ricin B abrin antisera were diluted in PBST and added to the plates for one-hour incubation. The plates were washed and incubated with AP conjugated anti-rabbit antibodies (Jackson, 111-055-003) for 15 minutes followed by detection with SIGMAFAST p-Nitrophenyl phosphate (Sigma, N1891).

Animal Studies
Animal experiments were performed in accordance with the Israeli law and were approved by the Ethics Committee for animal experiments at the Israel Institute for Biological Research. The treatment of animals was in accordance with regulations outlined in the USDA Animal Welfare Act and the conditions specified in the Guide for Care and Use of Laboratory Animals (National Institute of Health, 1996). New Zealand white rabbits (Charles River Laboratories Ltd., UK) weighing 2.5 to 3 kg were immunized in order to produce rabbit-derived polyclonal antibodies. Female CD-1 mice (Charles River Laboratories Ltd., UK) weighing 27-32 g were used for survival studies.
Prior to all studies in mice and rabbits, the animals were habituated to the experimental animal unit for 5 days. All mice were housed in filter-top cages in an environmentally controlled room and maintained at 21 ± 2 • C and 55 ± 10% humidity. Lighting was set to mimic a 12/12 hours dawn to dusk cycle. Animals had access to food and water ad libitum.

Polyclonal Anti-Chimera Antibody Production
Rabbits were immunized with A abrin B ricin or A ricin B abrin in a stepwise manner, injection 1 containing 1 µg toxin in Freund's adjuvant, and injections 2-5 (with incomplete Freund's adjuvant) in increasing concentrations (4, 25, 50, and 100 µg toxin/rabbit) with 4-week intervals between injections. Blood samples were collected (1 week after injection) to ascertain anti-chimera antibody titer build-up.

Characterization of the Chimeric Toxins
Chimeric abrin/ricin toxins were generated from isolated A and B-subunits of abrin and ricin, based upon a method developed by Olsnes et al. [11], with slight modifications (for details see "Materials and Methods"). Under non-reducing conditions, the chimeric A abrin B ricin and A ricin B abrin toxins, as well as reconstituted ricin and abrin toxin molecules generated by co-incubation of A and B-subunits of the same source, appear as a main band of approximately 60 kDa on SDS-PAGE ( Figure 1A). The formation of A abrin B ricin was incomplete, as indicated by detection of a weak~30 kDa band, in addition to the major band. Interestingly, this observation was also reported previously [11]. Since excess Asubunits are removed prior to the final stage of the experimental protocol for chimeric toxin generation, this~30 kDa band probably represents monomeric B ricin . As seen in Figure 1B, reduction with β-mercaptoethanol of reconstituted ricin and abrin (lanes 1 and 2, respectively) resulted in the generation of two typical major bands, somewhat different in molecular weights (the molecular weight of A abrin is less than that of A ricin , the molecular weight of B ricin is slightly less than that of B abrin ). Reduction of the chimera toxins (lanes 3 and 4) resulted in electrophoretic migrating bands, which confirm the composition of each toxin.
Antibodies 2020, 9, x FOR PEER REVIEW 5 of 10 1A). The formation of AabrinBricin was incomplete, as indicated by detection of a weak ~30 kDa band, in addition to the major band. Interestingly, this observation was also reported previously [11]. Since excess A-subunits are removed prior to the final stage of the experimental protocol for chimeric toxin generation, this ~30 kDa band probably represents monomeric Bricin. As seen in Figure 1B, reduction with β-mercaptoethanol of reconstituted ricin and abrin (lanes 1 and 2, respectively) resulted in the generation of two typical major bands, somewhat different in molecular weights (the molecular weight of Aabrin is less than that of Aricin, the molecular weight of Bricin is slightly less than that of Babrin). Reduction of the chimera toxins (lanes 3 and 4) resulted in electrophoretic migrating bands, which confirm the composition of each toxin.

Abrin Subunit Recognition and Neutralization
To produce polyclonal antibodies directed specifically against Aabrin or Babrin, rabbits were immunized with the chimeric toxins, AabrinBricin and AricinBabrin, respectively. The immunization protocol included repeated injection with increasing antigen doses, at four-week intervals, up to 100 µg antigen/rabbit. Hyperimmune sera directed against each of the chimeric toxins was collected and characterized for its ability to recognize Aabrin or Babrin in an enzyme linked immunosorbent assay (ELISA). Accordingly, increasing concentrations of either the anti-chimera antisera were incubated on abrin-coated plates, after which abrin-bound antibodies were detected using secondary anti-rabbit antibodies. The two antisera displayed ELISA titers of similar magnitude against the corresponding abrin subunits, 4.5 × 10 5 and 8 × 10 5 ELISA units, against Aabrin and Babrin, respectively (Table 1). Next, we asked whether these anti-subunit antibodies possess the capability to neutralize abrin in an invitro cell culture system that may potentially predict their efficacy in-vivo. In this assay, the neutralization potency of an antiserum is measured by monitoring its ability to block abrin-induced luciferase synthesis arrest in HeLa Ub-FL cells [12,13]. Abrin was incubated with increasing concentrations of the relevant antiserum, and added to the Ub-FL cells. Twenty four hours later, residual intracellular luciferase levels were measured and neutralizing antibody titers were determined for each antisera. As can be seen (Table 1), both subunit-directed anti-chimera antibodies possess the ability to effectively neutralize abrin in cell culture (~1 × 10 3 neutralizing antibody titers for both anti-chimera antitoxins). Table 1. Anti-abrin titers of rabbit-derived anti-AabrinBricin and anti-AricinBabrin antisera.

Abrin Subunit Recognition and Neutralization
To produce polyclonal antibodies directed specifically against A abrin or B abrin , rabbits were immunized with the chimeric toxins, A abrin B ricin and A ricin B abrin , respectively. The immunization protocol included repeated injection with increasing antigen doses, at four-week intervals, up to 100 µg antigen/rabbit. Hyperimmune sera directed against each of the chimeric toxins was collected and characterized for its ability to recognize A abrin or B abrin in an enzyme linked immunosorbent assay (ELISA). Accordingly, increasing concentrations of either the anti-chimera antisera were incubated on abrin-coated plates, after which abrin-bound antibodies were detected using secondary anti-rabbit antibodies. The two antisera displayed ELISA titers of similar magnitude against the corresponding abrin subunits, 4.5 × 10 5 and 8 × 10 5 ELISA units, against A abrin and B abrin , respectively (Table 1). Next, we asked whether these anti-subunit antibodies possess the capability to neutralize abrin in an in-vitro cell culture system that may potentially predict their efficacy in-vivo. In this assay, the neutralization potency of an antiserum is measured by monitoring its ability to block abrin-induced luciferase synthesis arrest in HeLa Ub-FL cells [12,13]. Abrin was incubated with increasing concentrations of the relevant antiserum, and added to the Ub-FL cells. Twenty four hours later, residual intracellular luciferase levels were measured and neutralizing antibody titers were determined for each antisera. As can be seen (Table 1), both subunit-directed anti-chimera antibodies possess the ability to effectively neutralize abrin in cell culture (~1 × 10 3 neutralizing antibody titers for both anti-chimera antitoxins). Table 1. Anti-abrin titers of rabbit-derived anti-A abrin B ricin and anti-A ricin B abrin antisera.

Abrin Linear-Epitope Recognition by Anti-A abrin -and Anti-B abrin -Polyclonal Antibodies
We have recently determined the immunodominant epitopes of abrin in sera of rabbits that were immunized with abrin (manuscript in preparation). We now sought to characterize the antibody response elicited against A abrin and B abrin , when these abrin-derived subunits are embedded within the chimeric toxins. By allowing each of the anti-chimera antisera to react with overlapping 15-mer peptides spanning the protein sequence of either A abrin or B abrin , we could determine to which linear epitopes each sera is directed (Figure 2). It was found that the anti-A abrin B ricin antiserum recognizes five epitopes of A abrin ( Table 2). While epitopes 1, 2, and 4 were previously mapped as immunodominant epitopes of anti-abrin antisera, epitopes 3 and 5 were discerned only by anti-A abrin B ricin antisera. It should be noted however, that the signals of the sera reacting with the three known epitopes were high whereas the signal with the hitherto undetected two epitopes was relatively low (close to background level), suggesting that these two epitopes have minor contribution to the overall activity of the sera toward abrin. Mapping the epitopes of the anti-A ricin B abrin reveals six epitopes (Table 2) all of which were previously mapped as immunodominant epitopes of B abrin . It should be noted that epitope 9 is based on signal of the sera binding to a single peptide (No. 37). The sequence DGSI also appears as a part of a larger epitope sequence (epitope 10), suggesting that the binding of the antibodies to peptide 37 is due to possible cross-reactivity with this sequence. Taken together, these results indicate that the overall structure and immunogenicity of the chimeric subunits are similar to that of the native toxin.
Antibodies 2020, 9, x FOR PEER REVIEW 6 of 10 We have recently determined the immunodominant epitopes of abrin in sera of rabbits that were immunized with abrin (manuscript in preparation). We now sought to characterize the antibody response elicited against Aabrin and Babrin, when these abrin-derived subunits are embedded within the chimeric toxins. By allowing each of the anti-chimera antisera to react with overlapping 15-mer peptides spanning the protein sequence of either Aabrin or Babrin, we could determine to which linear epitopes each sera is directed (Figure 2). It was found that the anti-AabrinBricin antiserum recognizes five epitopes of Aabrin (Table 2). While epitopes 1, 2, and 4 were previously mapped as immunodominant epitopes of anti-abrin antisera, epitopes 3 and 5 were discerned only by anti-AabrinBricin antisera. It should be noted however, that the signals of the sera reacting with the three known epitopes were high whereas the signal with the hitherto undetected two epitopes was relatively low (close to background level), suggesting that these two epitopes have minor contribution to the overall activity of the sera toward abrin. Mapping the epitopes of the anti-AricinBabrin reveals six epitopes (Table 2) all of which were previously mapped as immunodominant epitopes of Babrin. It should be noted that epitope 9 is based on signal of the sera binding to a single peptide (No. 37). The sequence DGSI also appears as a part of a larger epitope sequence (epitope 10), suggesting that the binding of the antibodies to peptide 37 is due to possible cross-reactivity with this sequence. Taken together, these results indicate that the overall structure and immunogenicity of the chimeric subunits are similar to that of the native toxin.

Post-Exposure Treatment of Abrin Intoxicated Mice with Anti-Abrin Subunit-Directed Antisera
As the goal of this study was to elucidate which abrin subunit needs to be neutralized in-vivo to promote the observed high level of protection following pulmonary abrin-intoxication [4], mice were intranasally exposed to 2LD 50 of abrin (8 µg/kg), and anti-abrin or anti-abrin-subunit (anti-chimera) antibodies (100 µL intravenously) were administered at 48 hours post exposure. Control intoxicated mice were not subjected to antibody treatment. The data in Table 3 establish that, as in the case of anti-abrin holotoxin antibody treatment, late administrations of either anti-A abrin or anti-B abrin antisera conferred very high protection levels (≥90%) against a lethal abrin intoxication.

Discussion
Previous studies carried out in our laboratory, demonstrated that application of anti-abrin antibodies conferred extremely high survival rates to intranasally abrin-intoxicated mice, even when treatment was performed at very late time after exposure to a lethal dose of the toxin [4]. In contrast, late anti-ricin antibody treatment following lethal ricin intoxication resulted in insignificant protection [3]. This unexpected sharp dissimilarity in antibody-induced protection could stem from a difference in the quality of the anti-abrin and anti-ricin antibody preparations, or from differences in disease progression following exposure to the two toxins.
We have previously shown that anti-ricin and anti-abrin ELISA-and neutralizing-antibody titers, as well as the apparent affinities of the specific fraction of either anti-ricin or anti-abrin antibodies, were very similar [4]. However, this data could not rule out the possibility that neutralization of an essential epitope within a specific subunit, or perhaps neutralization of the functional activity of one of the two subunits, toxin catalytic activity or lectin binding of subunit A and B, respectively, is obligatory for obtaining high-level protection. To evaluate such subunit-related neutralization effects, we sought to produce anti-abrin polyclonal antibodies directed specifically against A abrin or B abrin . An obvious option for generating such antibodies, would be to immunize rabbits with isolated monomeric A abrin or B abrin , however isolated individual subunits, originating from an AB dimeric context, are unstable and their conformational structure is drastically impaired [14]. Indeed, export of the abrin (or ricin) monomeric A-subunit from the endoplasmic reticulum (ER) into the cytosol relies on the fact that following toxin reduction in the ER, the isolated A-chain is identified as a misfolded protein, thereby allowing its retrieval to the cytosol via the ER degradation pathway [15,16]. In this regard, it has been suggested that ribosome depurination within the cytosol by monomeric A-subunit, requires its refolding, with the aid of the Hsc70 chaperone, to a catalytic conformation [17]. With respect to the B-subunit, it has been documented that B abrin is insoluble in aqueous solutions when separated from the A abrin , due to its hydrophobicity [6]. Accordingly, isolated subunit vaccination may lead to the production of antibodies that could not recognize and neutralize the holotoxin, resulting in inability to protect against in-vivo intoxications. Indeed, when recombinant A abrin was used as the immunizing antigen for production of anti-A abrin mAbs, in-vivo neutralization of abrin could be demonstrated only when these mAbs were applied prophylactically, while no post-exposure effects were reported [18,19]. MAbs directed against A abrin -abrus precatorius agglutinin (APA) chimeric toxin (1-123 amino acids of A abrin linked to 124-175 amino acids of APA), in which A abrin is not in an AB holotoxin context, were also found to be protective prophylactically, without any reports regarding post exposure effects [20]. Consequently, to produce an abrin-subunit specific antigen possessing proper abrin-subunit conformational structure for rabbit immunization, we generated chimeric abrin-ricin toxins composed of A abrin or B abrin linked via disulfide bond to their non-homologous counterparts, subunit B ricin or A ricin , respectively. Characterization of the two antisera raised against the two chimeric abrin/ricin toxins demonstrated that these recognized their cognate subunits to a similar extent and that both anti-chimeric toxin antibodies possessed potent in-vitro anti-cytotoxicity abilities. Most importantly, we found that following intranasal exposure of mice to abrin, both subunit-directed antisera, anti-A abrin B ricin (anti-A abrin ) and anti-A ricin B abrin (anti-B abrin ), were highly protective (≥90% survival) even when treatments were applied as late as 48 hours following intoxication. This high level of protection is due to the antibodies elicited against the abrin subunit of the chimeric toxins and not to those formed against the ricin subunit, since anti-ricin antibodies were completely ineffective when given to mice even at early hours following intranasal exposure to a lethal dose (2LD 50 ) of abrin (data not shown).
Taken together, it seems unlikely that the difference in protection conferred by anti-abrin and anti-ricin antibodies against abrin and ricin intoxications, respectively, is related to variances in the quality of the two antibody preparations. Rather, it seems that the striking difference in the ability to protect against ricin-and abrin-challenges stems from dissimilarities in disease progression. Supporting this argument, we have previously shown that following intranasal intoxication, the in-vivo catalytic performance of abrin, i.e., ribosomal depurination of pulmonary tissue, is significantly lower than that observed following ricin intoxication. In particular, the depurination level of pulmonary epithelial cells was markedly reduced following abrin intoxication, approximately five times lower than after ricin intoxication (16% and 81% depurination following exposure to abrin and ricin, respectively [21]). We further determined by FACS analysis that the epithelial cell damage following intoxications with abrin (unpublished data) and ricin [22] was mostly confined to the type II epithelial cells, responsible for lung tissue regeneration [23]. The injury to the epithelial cell sub-population was correlative to the trend described above for the whole-epithelial population, namely, the specific insult to the type II cells following abrin intoxication was less prominent than the damage observed following ricin exposure (unpublished data). Thus, anti-abrin and anti-ricin antibodies may halt further toxin-induced damage in an equally efficient manner, yet the anti-abrin antibodies will provide greater protection to abrin-intoxicated subjects due to the reduced levels of irreversible cellular damage inflicted by this particular toxin at a given post-exposure time-point. Furthermore, the notion that antibody-related protection ability is mostly due to toxin pathology progression and not to antibody quality, is in line with experiments recently carried out in our laboratory, in which mice were intoxicated with lethal doses of the chimeric toxins, and then treated with their corresponding anti-chimeric antibodies. Preliminary results suggest that significantly higher survival rates are achieved following late antibody treatment of mice that were intranasally intoxicated with the chimera toxin A ricin B abrin , even though both anti-A abrin B ricin and anti-A ricin B abrin conferred equally high levels of protection to abrin-intoxicated mice. Future studies will be required for characterization of the different damage patterns in the pulmonary cell subpopulations following intranasal intoxications with the two chimeric toxins.
In parallel to the results described above, recent studies carried out in our laboratory established that the low-density lipoprotein Receptor Related-protein 1 (LRP-1) is the major receptor for ricin in murine lung cells (Falach et al, manuscript in preparation). A different receptor for abrin (and A ricin B abrin ), if indeed detected, may further explain the dissimilarity between the disease progressions following pulmonary abrin and ricin intoxications.
In summary, both anti-A abrin -and anti-B abrin directed rabbit-derived polyclonal antibodies confer extremely effective protection against abrin lethality when administered lately following intranasal abrin intoxication in mice, allowing us to conclude that the significant difference in protection levels following antibody treatment after intranasal abrin and ricin intoxications stems from a dissimilarity in disease progression inflicted by these two toxins.