**3. Discussion**

Shiga toxin (Stx) is central to the development of hemolytic uremic syndrome (HUS). The supportive treatment is the current default procedure for STEC-infected patients, also, the administration of some antibiotic classes could increase the Stx production or release, which could lead to a risk of catastrophic consequences with HUS development, making this treatment option highly controversial [6,10]. Therefore, it is mandatory to develop either an effective treatment or a prevention method for the deleterious effects of Stx intoxication [19]. The STEC prevention is focused on individual and industry levels, such as hygiene procedures, meat processing protocols, and slaughterhouse maintenance, for example. Regarding therapy, it is focused mainly on inpatient supportive care, even though some strategies are in development aiming at different stages of infection, such as bacterial growth control without increasing Stx secretion, toxin trafficking interference and cellular response to the toxin. Moreover, it is worth mentioning the challenges regarding a therapeutic approach against Stx-induced symptoms, especially for clinical trials, such as the low incidence of STEC infections and HUS, the lack of highly specific diagnostic screening, and the narrow therapeutic window (onset of disease 3 days after infection, HUS development one week after the first symptom), which is also hypothetical [29]. Thus, despite great achievements towards a therapeutic tool against Stx, a specific treatment remains elusive.

Specific antibodies against Stx as a tool either to prevent or treat the HUS disease process is a promising approach [20]. Indeed, some other recombinant antibodies have also shown neutralizing ability against Stx2 in vitro or in vivo. Such as, the family Stx2Bbinding VHHs which were constructed with one anti-Stx2B VHH, and two copies were fused to one anti-human albumin VHH, neutralizing Stx2 in vitro [24]. Another VHH also protected mice against Stx2 intoxication, but it was not humanized [30]. Concerning scFv, the one described by Maa et al. [31] and Luz et al. [25], neutralizes the cytotoxic ability of Stx2 in vivo and in vitro, respectively, however, none of them is a human antibody, produced in a bacterial system, which impairs their use as therapeutic agents and costs of production.

The recombinant human Fab and F(ab')2 fragments characterized by Akiyoshi et al. [32] showed neutralizing capacity, however, the production was dependable on mammalian cells (CHO), having a high cost for obtaining as with hybridoma technology. Therefore, the library F [27] was employed to select specific Fab high binders against Stx toxins. Two phage clones showed high affinity and binding ability against Stx2. The FabC11:Stx2 was the first to be characterized and showed cross-reactivity with Stx1 besides being able to prevent Stx2 toxicity to human kidney cells and in mice [21,22].

Herein, the other Fab selected against Stx2 (FabF8:Stx2) was characterized and employed in different cell assays. The variety of toxin subtypes that could be expressed by a diverse set of STEC serotypes able to express one or more toxin types at the same time is a major challenge for antibody-based Shiga toxin neutralizers, once to be universally effective, should neutralize multiple Stx1 and Stx2 subtypes [14]. In the present study, using the gold standard Vero cell neutralization assay, we observed that FabF8:Stx2 neutralized the cytotoxicity of 23 of 27 supernatants from Stx2 or Stx1/Stx2-producing STEC strains. These strains belong to different serotypes and present diverse *stx* subtypes, it is worth mentioning that no differences were found with neutralization percentage and its *stx* subtype, even though the FabF8:Stx2 was generated against a Stx2a toxin, some strains harboring *stx2a* gene were not neutralized whereas two non-Stx2a producing were neutralized. This kind of investigation is not commonly employed, usually, most neutralization assays are tested against one type and/or one subtype of the toxin, therefore in this work, for the first time, we showed how one recombinant monoclonal antibody neutralizes different Stx combinations obtained from the STEC isolates culture.

The human microvascular endothelial cells are an excellent cell model for in vitro therapeutic studies once it can express 50-fold higher Gb3 levels compared to the endothelial cells of large vessels [33]. In this sense, previously, we developed human glomerular endothelial cells (HGEC) primary cultures and demonstrated the decrease of cell viability by apoptosis and endothelial injury like that documented in kidney biopsies of HUS patients after incubation with Stx2 [28]. In this work, additionally, we assayed FabF8:Stx2 antibodies on HGEC exposed to Stx2 and we were able to corroborate their great effectiveness on the protection of Stx2 cytotoxicity on HGEC, in about 80–90% at the pre-incubation condition. These results were coincident with the high capacity of these antibodies to prevent HGEC apoptosis in about 75–90% under pre-incubation and co-incubation conditions. Furthermore, at pre-incubation, cell detachment was avoided in approximately 60–65% and swelling, in about 90–95%.

Previously, we demonstrated that by using 10 μg/mL FabC11:Stx1/Stx2 we observed lower protection of the HGEC viability (54.0% and 52.0%, for pre-incubation and coincubation, respectively) compared to the same concentration of FabF8:Stx2 when cells were exposed to a 1 CD50 of Stx2, therefore, preventing Stx2 toxicity on human kidney cells [26]. Additionally, FabC11:Stx1/Stx2 cell detachment protection was also lower than FabF8:Stx2, which showed protection levels of 43.5% under pre-incubation and 36% under co-incubation conditions. With respect to swelling, although we demonstrated good prevention, pre-incubation: 97.0% and co-incubation: 81.0%, it is noteworthy that this protection had been obtained with a higher concentration of FabC11:Stx1/Stx2 (10 μg/mL) compared with FabF8:Stx2 (1 μg/mL). Our results conclusively demonstrate the efficacy of

FabF8:Stx2 antibodies to avoid the cytotoxic effects of Stx2 on human renal microvascular endothelial cells, one of the principal target cells for this toxin, raising the possibility of the development of a new therapeutic molecule against Stx2 toxicity.

#### **4. Conclusions**

The present work showed encouraging results about the effectiveness of FabF8:Stx2 antibodies to neutralize the cytotoxic effects of both purified Stx2 and Stx2 or Stx1/Stx2 produced by STEC strains. Thus, they could be a promising therapeutic strategy to prevent kidney damage and the subsequent development of HUS. Future studies will be focused on analyzing the efficacy of FabF8:Stx2 in in vivo models.
