**4. Discussion**

Clinical symptoms of envenomation include histamine-related responses; although they are less frequent, there are reports of hypersensitivity or even allergic reactions after spider bites [9,10]. The presence of histamine in the envenomation site of Loxoscelism can cause edema and endothelial changes such as increased vascular permeability and vasodilation, which contribute to the systemic dispersion of venom components and exacerbate the inflammatory response triggered by the bite [7,8]. Inflammatory responses can be related to both mast cells and histamine [6,11]. It has already been shown that *L. intermedia* venom increases vascular permeability and induces vascular relaxation in

rats [29] and that these effects are related to the ability of venom to degranulate mast cells and release mediators such as histamine [11].

Different studies that investigated the protective effects of recombinant *Loxosceles* phospholipases-D (PLDs) or even the neutralizing effects of serum produced with these toxins found that the edematogenic activity of *Loxosceles* venoms is particularly difficult to neutralize and raised the possibility that other venom components may be responsible for edema development [30,31]. LiRecTCTP induces an increase in microvascular permeability of skin vessels and is a component of edema formation with an earlier and faster effect when compared to the inflammatory response triggered by whole *L. intermedia* venom in mouse paws [2]. The effects of *L. intermedia* crude venom on RBL-2H3 degranulation (Figure 2C) are related to the histaminergic effects of Loxoscelism. The report of TCTP-related toxins in the *Loxosceles* venoms were previously done by transcriptomic studies using venom gland transcripts [1], by cloning and recombinant expression of LiRecTCTP [2], and recently, TCTP was identified in the proteomic study of the *L. intermedia* venom [32]. Additionally, TCTP was immunodetected in the whole venom of *Loxosceles* species (*L. intermedia*, *L. gaucho*, and *L. laeta*) [3], pointing to conservation and biological relevance. There are described TCTPs from other spiders [33,34], although they are not related to venomous accidents as *Loxosceles* [35].

The toxins purification was a critical procedure for this study: pure and correctly folded LiRecTCTP was obtained (Figure 1). Two other recombinant proteins were used as negative controls to exclude an effect due to a possible prokaryotic contamination from the heterologous expression system: green fluorescent protein (GFP), an innocuous protein, and LiRecDT1 H12A, a mutated isoform derived from LiRecDT1, with drastically reduced activity to residual levels [2,27,36].

In this study of LiRecTCTP's biological role, it was relevant to rule out its possible cytotoxic effects on RBL-2H3, which could cause release of cellular contents and mimic degranulation in spite of a mechanism dependent of LiRecTCTP action. The morphology of RBL-2H3 cells was not altered by LiRecTCTP, cells submitted to treatment with 100 μg/mL did not differ from control cells, and they remained adhered and spread onto the slide, showing philopodia (Figure 2C). The unreduced capacity to metabolize MTT by cells treated with LiRecTCTP implies a specific effect on degranulation of RBL-2H3 cells (Figure 2A). Beta-hexosaminidase activity after LiRecTCTP treatment of RBL-2H3 cells infers that this toxin can directly trigger degranulation process in these basophil lineage cells (Figure 2B). Cromolyn is known as a "mast cell stabilizer." It interferes with the release of inflammatory and other chemical mediators from mast cells and either blocks or reduces the amount released [37]. The cromolyn effect as a mast cell stabilizer is dose-dependent.

Human TCTP was already well-characterized as a histaminergic molecule [38–40] and suggested as a putative target for therapeutics in asthma and allergy [41,42]. Several reports indicate TCTP involvement in the inflammatory response of parasite-infected individuals [43–46].

Activation of basophils and mast cells can be monitored through different approaches, including morphological changes, phenotypic changes, and quantification of secreted mediators. IL-4 is produced by basophils in the early response to a stimulus, and IL-3, produced by mast cells, is involved in severe hypersensitivity reactions. Functionally, mast cells, and basophils overlap in their ability to produce several mediators, including histamine and granule proteases. IL-3, IL-4 and IL-13 cytokines act as immunomodulators of other immune cells in the inflammatory and allergic signaling pathway and play pivotal roles in exacerbating the inflammatory responses in vivo [47,48]. Expression of IL-3, IL-4 and IL-13 was induced by LiRecTCTP in RBL-2H3 cells, which indicates that this protein could be involved in the cutaneous inflammatory and histaminic conditions of *Loxosceles* envenomation (Figure 3B). As these cytokines are able to recruit inflammatory immune cells to the bite site, LiTCTP could contribute to the exacerbated inflammatory response by stimulating the production and release of these cytokines. The expression of these cytokines by the RBL-2H3 cells is related to their activation by LiRecTCTP, as indicated by increased Ca+<sup>2</sup> influx and beta-hexosaminidase activity (Figure 3A,B).

We used different histamine receptors inhibitors in order to evaluate histaminergic effects of LiRecTCTP: H1R to H4R [49]. Alterations in vascular permeability, measured by the leakage of administered Evans blue from vessels (Figure 4), and the edematogenic e ffect of LiRecTCTP (Figure 5) were performed in mice previously treated or not treated with the anti-histaminic drugs. Cimetidine presented minimum inhibition e ffect of LiRecTCTP; it could not block toxin-induced permeability and modestly reduced paw edema only in the first 10 min after toxin inoculation. Absence of cimetidine inhibition is explained by the fact that its targets (H2 receptors) are mainly localized in the stomach, brain, and cardiac tissue, but typically not in the skin. E ffects of promethazine are related to the blockage of H1R activation by histamine, the previous treatment with this drug resulted in inhibition of LiRecTCTP action on permeability and edema. H1 receptors are expressed by a broad range of cells, including airway and vascular smooth muscle cells, endothelial cells, monocytes, neutrophils, and T and B cells [49]. Thioperamide is a dual H3-H4 receptors antagonist. H3 receptors are irrelevant in the context of our experiments as these histamine receptors are almost exclusively expressed in the nervous system.

On the other hand, histamine H4 receptors are described mainly expressed in cells of the human immune system and influence their cytokine production mediating several e ffects on chemotaxis [50]. Thioperamide treatment significantly decreased mice response to the histaminergic e ffects of LiRecTCTP in vascular permeability and paw edema. When the e ffects of LiRecTCTP on histamine release were blocked by cromolyn, toxin biological e ffects in animals were almost abrogated: results were similar to the negative controls. These results obtained in animal models point to an in vivo histamine-induced effect by LiRecTCTP, related to mast cell degranulation and histamine e ffects on H1 and H4 receptors. It is important to mention, in the regard of the pro-inflammatory response triggered by venom and observed in Loxoscelism, that IL-3, IL-4, and histamine can upregulate H1 receptor gene expression. This positive feedback loop could contribute to the exacerbated inflammatory condition seen in dermonecrotic lesions resulted from *Loxosceles* bites. The involvement of H4 receptor in the histaminergic e ffects of LiRecTCTP should also be highlighted as this receptor is emerging as an essential receptor for the chemoattraction of immunologically relevant cells, contributing to an extended inflammatory cascade.

The investigation of LiRecTCTP participation in dermonecrosis lesion development and clinical features was performed using a well-established in vivo protocol. LiRecDT1 inoculation into rabbit skin, as expected and well-described, caused a characteristic dermonecrosis: a lesion with gravitational spreading, leukocyte infiltration of dermis with the prevalence of neutrophils (PMN), and increased capillary permeability in mice [28,51]. When LiRecTCTP was injected together with the dermonecrotic toxin there was a clear dose-dependent enhancement of all these features at the site of injection: hemorrhage and the gravitational spreading are more intense when compared to LiRecDT1 by itself, and in histopathological analysis, there is an increased number of PMN and more points of fibrin network formation in the connective tissue as a result from an increased microcapillary permeability and disruption of vessel walls (Figure 7). The more prominent cutaneous e ffect could be explained by the histaminergic e ffect of LiRecTCTP, which may have contributed to the systemic dispersion of LiRecDT1 toxin. The synergistic e ffect of both toxins was also revealed by the marked edema and increased dermonecrotic lesions in rabbit skin.

In an envenomation event, LiTCTP is probably acting in edema formation and permeability alterations, corroborating with the huge inflammatory condition of Loxoscelism. LiTCTP could also facilitate the di ffusion of other toxins, ultimately promoting venom components spreading from the bite site. Other brown spider toxins that promote extracellular matrix degradation and remodeling, such as metalloproteinases and hyaluronidases, are also implicated in this process [8,52,53]. This combined e ffect of LiRecDT1 and LiRecTCTP was reproduced in the Evans blue assay. Vascular permeability was highly augmented by the use of both toxins in the inoculation site (Figure 6C). The evaluation of myeloperoxidase activity was crucial to investigate the inflammatory response in mice, as these animals do not develop dermonecrotic lesions (reason still not fully understood) [28,54]. The level of myeloperoxidase (MPO) activity in a sample is directly proportional to the number of neutrophils present, representing the inflammation status [27,55,56]. Results of MPO activity show LiRecTCTP increases the inflammation triggered by LiRecDT1, data that corroborate with its relevant

participation in the pro-inflammatory response seen in dermonecrotic lesions (Figure 6B). Concerning the ratio LiRecDT1/LiRecTCTP used in our experiments, transcriptome analysis of *Loxosceles* venoms glands showed that phospholipases-D transcripts such as LiRecDT1 are much more abundant that TCTP transcripts [1]. However, higher concentrations of LiRecDT1 would impair the detection of the synergistic activity between these toxins, inflammatory e ffects generated only by PLDs would mask the allergenic and inflammatory activities generated by LiTCTP.

Data presented herein confirm previous data that suggested LiRecTCTP histaminergic action by its e ffect on vascular permeability and edema formation [2]. We sugges<sup>t</sup> that the participation of LiTCTP in the exacerbated inflammatory response is based on LiRecTCTP's direct e ffect on mast cells and histamine release. Moreover, LiRecTCTP could be considered for potential applications as biotool, as for instance in basophil activation tests or allergy diagnosis and in vitro testing [8,57].

Altogether, the e ffects observed for LiRecTCTP, resulting in increased inflammatory response, capillary permeability and edema, as well as acting synergistically with dermonecrotic toxin LiRecDT1, unveil LiTCTP's role as an additional spreading factor present in the venom of *Loxosceles* spiders. Along with the classic spreading agents described in the venom [13,53], LiTCTP through its histaminergic mechanisms can contribute to the spread of other toxins from the bite site, accentuating the local and even systemic post-envenoming condition.

**Author Contributions:** Conceptualization, A.S.-R., S.S.V., O.M.C., and M.B.-F.; Methodology, M.B.-F., K.G.M., A.B.C.B., A.C.M.W., L.P.d.S., L.D., L.V. and B.S.; Analysis, M.B.-F., K.G.M., B.S., K.C.B., A.B.C.B., A.C.M.W., L.P.d.S., L.D., O.M.C., and L.V.; Writing (Original Draft Preparation), A.S.-R. and M.B.-F.; Writing (Review and Editing), A.S.-R., M.B.-F., L.H.G., S.S.V., and O.M.C.; Visualization, M.B.-F.; Supervision, A.S.-R., L.H.G., and S.S.V.; Project Administration, A.S.-R., O.M.C., and L.H.G.; Funding Acquisition, A.S.-R., O.M.C., L.H.G., and S.S.V.

**Funding:** This work was supported by grants from CAPES, CNPq, UFPR (Federal University of Paraná), FUNDAÇ ÃO ARAUC ÁRIA-PR/SETI-PR/SESA-PR/MS-Decit/PPSUS, Brazil.

**Conflicts of Interest:** The authors declare no conflict of interest.
