Authors: Luiza Helena Gremski, Fernando Hitomi Matsubara, Hanna Câmara da Justa, Zelinda Schemczssen-Graeff, Antonielle Beatriz Baldissera, Pedro Henrique de Caires Schluga, Isabel de Oliveira Leite, Marianna Boia-Ferreira, Ana Carolina Martins Wille, Andrea Senff-Ribeiro, Silvio Sanches Veiga
Categories: Review, Allergens, Astacins, Hyaluronidases, Knottins, Serine proteases, Serpins, Spider venom, TCTP
Source: The Journal of Venomous Animals and Toxins Including Tropical Diseases
Accidents caused by the bites of brown spiders (Loxosceles) generate a clinical condition that often includes a threatening necrotic skin lesion near the bite site along with a remarkable inflammatory response. Systemic disorders such as hemolysis, thrombocytopenia, and acute renal failure may occur, but are much less frequent than the local damage. It is already known that phospholipases D, highly expressed toxins in Loxosceles venom, can induce most of these injuries. However, this spider venom has a great range of toxins that probably act synergistically to enhance toxicity. The other protein classes remain poorly explored due to the difficulty in obtaining sufficient amounts of them for a thorough investigation. They include astacins (metalloproteases), serine proteases, knottins, translationally controlled tumor proteins (TCTP), hyaluronidases, allergens and serpins. It has already been shown that some of them, according to their characteristics, may participate to some extent in the development of loxoscelism. In addition, all of these toxins present potential application in several areas. The present review article summarizes information regarding some functional aspects of the protein classes listed above, discusses the directions that could be taken to materialize a comprehensive investigation on each of these toxins as well as highlights the importance of exploring the full venom repertoire.
Keywords: Astacins, Serine proteases, Serpins, Knottins, TCTP, Hyaluronidases, Allergens, Spider venom
Accidents involving brown spider bites are endemic in South and Southeast regions of Brazil, where they have caused more than 80,000 notifications over the past ten years [1]. Brown spiders have this name due to the characteristic brown color displayed by their bodies. They are cosmopolitan spiders that are found in all continents, but are more adapted to hot or temperate regions with temperatures ranging from 8 to 43°C [2,3]. These spiders belong to the Sicariidae family and the Loxosceles genus, whose name alludes to the fact that these animals have their legs curled or bent during the rest period (Loxosceles means folded/slanted legs) [4-6]. There are more than 150 species of Loxosceles spiders described in the literature [7].
They are sedentary animals with nocturnal habits, not aggressive, that organize irregular webs, and prefer to inhabit dark places [2,3,8]. However, few species have clinical significance. In the United States of America, mainly in the South region, and in Central America the spiders Loxosceles reclusa, L. arizonica and L. deserta prevail. In South America Loxosceles intermedia, L. laeta and L. gaucho are predominant, which are of great medical importance, especially as they present endemic dissemination in some regions of Brazil, Chile and Peru [3,6,8-10].
Accidents caused by the bites of brown spiders (Loxosceles) are clinically designated as loxoscelism and are characterized by the appearance, in most victims, of skin lesions at the bite site such as necrosis, edema, ecchymosis and erythema, which spread out to neighboring regions (gravitational spread - the hallmark of loxoscelism). Additionally, a massive inflammatory response at the site of the injury and/or its vicinity is reported, with the participation of neutrophils that seem to be responsible for the damage of tissues seen during envenomation [2,3,11,12]. This clinical condition is histologically characterized as an aseptic coagulative necrosis, which consists of a massive destruction of skin structures without direct involvement of infectious agents [13-15]. This picture has a very intriguing mechanism, since the venom does not directly activate the leukocytes involved in the tissue destruction, but instead causes an initial activation of the endothelium, which in turn indirectly activates the leukocytes [16,17]. Signs like itching and skin rash in the vicinity of the bite site are also reported, which suggest an allergenic component in the venom [3,6,8,10,18,19].
At the systemic level, less frequent but more severe alterations are reported, which include intravascular hemolysis, thrombocytopenia, and acute renal failure. These signs can evolve and even lead patients to death [6,8,10,20,21]. Even in small amounts, this venom causes severe effects on patients. It is speculated that a few microliters of venom are injected during the bite, which contains between 20 and 200 micrograms of proteins [3,6,22,23]. The venom of Loxosceles is produced by two venom-producing glands located in the cephalothorax region of the spiders. These glands display a holocrine secretion mechanism, being the venom produced by epithelial cells organized in a secretory epithelial monolayer, which release a large number of secretory vesicles containing the synthesized toxins towards the apical domain of the cells and then to the gland lumen [3,24].
Through omics analyzes (especially proteomics and transcriptomics), it was shown that Loxosceles venoms have two groups of the highly expressed ones, and those expressed in lower amounts (Figure 1). A study regarding the transcripts encoded in the Loxosceles intermedia venom-producing glands showed that among the toxins produced in large quantities, knottins (inhibitor cystine knot peptides or ICKs), comprise about 56% of the transcripts that encode toxins, whereas astacins (metalloproteases) represent about 23% of these transcripts, and phospholipases D (dermonecrotic toxins) account for about 20%. Together, the other families of toxins identified to date - hyaluronidases, serine proteases, serpins (serine protease inhibitors), allergens and translationally controlled tumor proteins (TCTP) - account for about 1.3% of the toxin encoding transcripts [25].
Figure 1. Overview of the toxins present in the venom of Loxosceles spiders. Venom components can be divided into two major (i) highly expressed toxins (phospholipases D, knottins or ICK peptides and metalloproteases); and (ii) low-expressed toxins (translationally controlled tumor proteins - TCTP, allergens, hyaluronidases, serine proteases and serine protease inhibitors). Phospholipases D can induce all the main effects associated with the whole venom, in addition to displaying insecticidal activity. Recombinant phospholipase D can trigger a dermonecrotic lesion, the hallmark of Loxoscelism (light blue panel - Vuitika et al. [123]). In comparison to phospholipases D, all the other venom components have been less studied. Knottins, also known as ICK peptides, are associated with insecticidal activity. Metalloproteases have been linked up to the hydrolysis of extracellular matrix elements facilitating the spread of other toxins, as well as to the induction of deleterious effects on endothelial cells worsening the tissue damage caused by the venom itself (source: da Silveira et al. [40]). TCTP proteins act as a histamine-releasing factor, degranulating mast cells and triggering inflammatory events after the envenoming (source: Justa et al. [19]). Allergens also participate in the inflammatory process, stimulating the degranulation of mast cells and increasing vascular permeability (source: Justa et al. [19]). Hyaluronidases hydrolyze hyaluronic acid and then elicit the gravitational spread of the dermonecrotic lesion (source: Ferrer et al. [68]). Serine proteases hydrolyze gelatin and may be involved with the extracorporeal digestion of prey (source: Veiga et al. [51]). Serpins (serine protease inhibitors) have been poorly characterized to date, but have already shown to be able to inhibit components (serine proteases) of the complement system. Center image from Chaim et al. [114].
Among the toxins with high expression, phospholipases D are undoubtedly the most studied and well characterized molecules from the biochemical and functional standpoints. They are highly conserved toxins among the various species of Loxosceles spiders described in the literature, comprising a family of toxins with intra- and inter-species occurrence coverage. These toxins represent approximately 16% of the transcripts from the venom-producing glands of L. laeta [26], about 15% of the transcripts produced by glands of L. similis [27], in addition to the 9% of total transcripts in the glands of L. intermedia [25]. They are related to the uncontrolled activation of inflammatory response that appears after envenoming, which results in the skin injury, in addition to triggering the systemic deleterious processes previously mentioned. Herein, phospholipases D will not be explored, since updated data on them were extensively reviewed in recent publications [6,17,28,29].
In the present study, we will discuss the literature related to the least studied toxins found in the Loxosceles venoms, which have little data available on their participation in the biology of venoms. Two families of toxins belonging to the group of highly expressed toxins (knottins and astacins) will be addressed as well as the other families of toxins belonging to the group of low-expressed toxins (hyaluronidases, allergens, TCTP, serine proteases, and serpins) (Figure 1). Although these toxins are not individually involved in all venom deleterious activities like phospholipases D, recent studies point out that some of these molecules have a relevant participation in the framework of loxoscelism pathophysiology, in addition to being a potential target for the development of biotechnological resources or for the understanding of molecular and cellular processes.
This systematic review was elaborated based on articles retrieved from the electronic databases PubMed and Google Scholar. Original articles, review studies and case reports published until January 15, 2021 were considered after being identified using as keywords “Loxosceles” and the class of a specific toxin (i.e., metalloprotease, serine protease, hyaluronidase, allergen, TCTP, serpin or ICK peptide) combined. After the identification of the articles, exclusion criteria were sequentially applied in order to select appropriate articles regarding the aim of this review. In the first step, articles repeated between the databases accessed were removed as well as articles with titles showing no relation to the toxins investigated. Only articles in English and Spanish were considered for further analyses whereas theses and dissertations shown by the search engine Google Scholar were also disregarded.
Subsequently, the articles selected in the previous step were screened based on their abstract, resulting in the removal of articles that do not contain information supporting the aim of this review or were published in low-quality journals (with no impact factor assigned). The remaining articles were submitted to full-text appreciation followed by discarding of articles that do not involve the context intended (i.e., characterization of specific toxins present in the venom of Loxosceles spiders) or that do not add any new information to the literature of Loxosceles toxins herein described. Finally, the remaining 55 articles were included in this review. Other 65 articles were biased, selected and included since they were fundamental to set the background and historically, structurally, biochemically or biologically characterize each class of toxin. Figure 2 summarizes the steps adopted in order to produce this review.
Figure 2. Flow chart highlighting the methodology used in order to produce this systematic review.
The existence of proteases in brown spider venoms was originally reported over 40 years ago. The first descriptions showing the presence of proteases in the brown spider venoms were made using the crude venom of Loxosceles reclusa, which showed proteolytic activity on larvae of Heliothis viresceus and Musca domestica [30]. These data on proteolytic activities in brown spider venoms were confirmed by the use of synthetic substrates derived from the L-aminoacyl-beta-naphthylamide, which were hydrolyzed by L. reclusa venom [31]. The discovery of metalloproteases (astacins) in the venoms of brown spiders has the participation of several Brazilian researchers, and was described almost twenty years later, in studies that initially used the crude venom from Loxosceles intermedia. Through kinetics experiments of protein substrate degradation using fibronectin or fibrinogen incubated with crude venom, results showed the presence of proteolytic activity on these proteins (Figure 3).
Figure 3. Experimental strategies used to study some toxins present in the venom of Loxosceles spiders. (A) De Castro et al. [110] reported the obtainment of native toxins (knottins or ICK peptides) purified from L. intermedia crude venom by means of sequential chromatographic approaches. The native toxins were then biologically characterized as insecticidal molecules and their amino acid sequences were identified by sequencing. (B) Hyaluronidases from L. intermedia were first studied by Da Silveira et al. [^66^] using biochemical methods (zymography). These analyses showed molecules displaying hydrolytic activity in the venom upon hyaluronic acid and chondroitin sulfate. Toxin identification using biochemical assays was also described for metalloproteases. Feitosa et al. [^32^] showed that the crude venom of L. intermedia degraded human fibronectin in proteolytic digestion kinetic experiments. Further analyses showed that metalloprotease inhibitors blocked crude venom ability to hydrolyze fibronectin, which together with zymography experiments using gelatin as a substrate, pointed out the presence of metalloproteases in the studied venom. (C) Identification and characterization of toxins can also be performed using transcriptome analysis, followed by recombinant protein-production techniques and studies for the characterization. Sade et al. [87] and Boia-Ferreira et al. [88] reported biological and biochemical characterization of a recombinant L. intermedia TCTP expressed in bacterial model identified in the venom-producing gland transcriptome. Ferrer et al. [^68^] produced a recombinant hyaluronidase in bacterial model followed by in vitro and in vivo analysis; the sequence coding this hyaluronidase was also identified in the transcriptome of L. intermedia venomous gland. In addition, transcriptomic findings revealed a sequence of an allergen encoded in L. intermedia venom-producing gland. Based on this information, Justa et al. [19] recombinantly expressed this toxin using baculovirus-infected insect cells and characterized its functionality. Venom producing glands (C) from Chaim et al. [114].
Zymograms with copolymerized gelatin identified proteases with molecular masses between 30 and 35 kDa, and zymograms with copolymerized fibronectin or fibrinogen showed proteases with 22-28 kDa. In both cases, only divalent metal chelators such as ethylenediaminetetraacetic acid (EDTA) and 1,10-phenanthroline - but not inhibitors of other classes of proteases such as phenylmethylsulfonyl fluoride (PMSF), aprotinin, leupeptin, and pepstatin A - blocked these proteolytic activities, confirming that the proteases involved on such as activities are metalloproteases [32]. The presence of metalloproteases in the venoms of brown spiders were also demonstrated by other authors, who showed metalloprotease-dependent proteolytic activities in L. intermedia venom on the substrates entactin and on heparan-sulfate proteoglycan protein core synthesized by endothelial cells [33,34], in addition to proteolytic activity on fibrinogen [35]. In addition, it was found that a metalloprotease of L. gaucho venom acts upon red cell band 3 transmembrane protein [36]. Then, two important descriptions were the first showing the biological conservation of these metalloproteases in venoms of different species of Loxosceles spiders, such as L. rufescens [37], L. gaucho, L. deserta and L. laeta [38], indicating the importance of metalloproteases in the biology of these animals. Another important description used extract of L. intermedia venom-producing glands, discarding criticisms of a possible contamination of venoms by the digestive secretions of spiders used in the previous articles, which were obtained by electrostimulation, and then proving that metalloproteases are components of venoms and not digestive contaminants [39].
The classification of these metalloproteases as astacins was first described by the cloning of an isoform of these enzymes from a cDNA library obtained from the venom-producing gland of L. intermedia, which identified a metalloprotease with 30 kDa, containing a sequence signature of astacins. This metalloprotease - which was named LALP (from Loxosceles astacin-like protease) - was obtained in its recombinant form by heterologous expression using bacterial model and proved to have gelatinolytic, fibronectinolytic and fibrinogenolytic properties [40]. Transcriptomic studies of the venom-producing glands of L. laeta, which showed the presence of 8.2% of transcripts for metalloproteases [26], and L. intermedia, which showed about 9.8% of the transcripts for the same family of toxins, bring into view further evidence that strongly suggests the presence of metalloproteases in different brown spider venoms [25]. Metalloproteases are also expressed in high amounts in the venom gland of L. gaucho, as found by a transcriptome analysis [41]. A recent study analyzed the Peruvian L. laeta transcripts focusing on LALPs, and found 9 putative sequences coding for astacine-like metalloproteases highly similar to LALP1 from L. intermedia [42]. In addition, the authors compared the activities of Brazilian and Peruvian L. laeta venom upon fibrinogen and gelatin/collagen, and concluded that the Peruvian venom have a higher activity upon these molecules than the Brazilian one [42]. A proteomic analysis of the L. intermedia crude venom using mass spectrometry also described these enzymes as components of the venom [43].
Finally, astacins were described as forming an intra and inter-species family of toxins in the brown spider venoms, which contain at least six different isoforms as showed through an elegant experiment using two-dimensional zymogram copolymerized with gelatin [44,45]. These data were later confirmed by molecular cloning of members of metalloprotease family, strengthening the existence of various isoforms of these proteases in the brown spider venoms [46]. The existence of families of astacins in the venoms of brown spiders was also demonstrated by the use of a monoclonal antibody produced against a L. intermedia LALP isoform, which cross-reacted with crude venoms of L. laeta and L. gaucho and neutralized the proteolytic effects of these enzymes in the venoms [47]. The biological conservation and the presence of astacin isoforms in different brown spider venoms have already pointed out the participation of these proteolytic enzymes in the biological events related to brown spider venom. However, what are the true biological functions of these molecules?
The first aspect that can be discussed is undoubtedly the participation of astacins as regulating molecules of extracorporeal digestion performed on prey captured by these spiders, which are carnivorous and considered very efficient predators that ingest food in liquid form [48]. The presence of these endo-proteases in the venom helps to degrade protein components of the obtained preys, and thus, facilitate the feeding of these spiders. Another possibility is the proteolytic action of these metalloproteases on the components of blood vessel endothelial cells, such as heparan-sulfate proteoglycan, involved in endothelial cell biology and blood vessel stability. This mechanism was previously investigated [34,35] and could induce signs related to blood vessel disorders after envenoming, such as increased capillary permeability, edema, and ecchymosis. Moreover, the action of astacins on extracellular matrix components involved in blood vessel stability, blood clotting, platelet adhesion and aggregation - such as fibronectin, fibrinogen and entactin - could explain other events that occur after envenoming, such as hemorrhage and even the reported difficulty in healing the skin lesions triggered by accidents [32-35].
The proteolytic action of brown spider venoms has also been demonstrated on basement membranes, structures with enormous importance in the organization of various tissues [49]. Together with the proteolytic action on soluble components of the extracellular matrix, these activities could explain some toxic effects caused by the venoms as previously discussed. By making the extracellular matrix loosened and disorganized, astacins could also promote the spread of other toxins to the nearby blood vessels and other parts of the bodies of injured victims, enhancing deleterious systemic effects during envenoming. Finally, metalloproteases present in the venoms could act as molecular scissors that cleave precursor molecules of the venom and/ or confined to the hosts, thus activating the toxic activities seen after accidents. An example that illustrate this mechanism is that of zymogens, produced by leukocytes, which could be activated during the unregulated inflammation that occurs after accidents, as previously hypothesized [40].
Another interesting work that deserves to be cited and is based on the participation of astacins in the toxicity of brown spider venoms used linear sequences of an astacin from the venom of L. intermedia LALP-1 (SLGRGCTDFGTILHE, ENNTRTIGPFDYDSIMLYGAY, and KLYKCPPVNPYPGGIRPYVNV) in the construction of a chimeric recombinant antigen containing, in addition to LALP sequences, the sequences of a hyaluronidase and a phospholipase D present in this venom. This chimera was antigenic and triggered the production of antibodies that neutralized the dermonecrotic activity in rabbits’ skin and inhibited mice lethality induced by the crude venom [50]. Another hybrid immunogen consisting of hydrophilic regions from the metalloprotease LgALP1 from L. gaucho and from a phospholipase D of the same species was constructed, expressed and used to produce antibodies in mice. This antiserum neutralized dermonecrotic (in vivo) as well as fibrinogenolytic and platelet aggregation (in vitro) activities elicited by Loxosceles venoms [41]. Astacins, as well as other toxins present in the venom of Loxosceles spiders, are not feasible to be obtained in their native form through purification from crude venom because of the low yield of the venom extractions. Moreover, these enzymes are difficult to be recombinantly obtained in their soluble and active form [40,46]. Due to this difficulty in isolating these molecules, the knowledge regarding their participation in the envenoming process is based on theoretical possibilities and/or hypotheses related to their biochemical properties, pointing out the lack of robust evidence for final conclusions.
Other proteases described in brown spider venoms belong to the family of the serine proteases. In this case, literature data are even scarcer and are restricted to a few published articles. The first description of serine proteases as part of brown spider venoms was made using crude venom of L. intermedia, which after treatment with thrombin, activated two molecules with gelatinolytic activity at regions of 85 kDa and 95 kDa in a zymogram analysis. These proteases were inhibited by serine proteases inhibitors as PMSF, aprotinin, benzamidine, soybean-trypsin inhibitor and leupeptin, but not by other protease inhibitors as EDTA, 1-10’-phenanthroline, iodoacetamide and pepstatine-A. In addition, such data indicated that these toxins are present in the forms of inactive zymogens, and that after treatment with thrombin, they acquire proteolytic activity [51]. They are enzymes with an optimal pH between 7 and 8 and, interestingly, they do not have a wide spectrum of proteolysis. These enzymes showed proteolytic activity restricted to gelatin, with no catalytic activity on proteins such as BSA, hemoglobin, IgG, casein or laminin [51]. The presence of serine proteases in brown spider venoms was reinforced by transcriptomic analysis of L. laeta venom-producing gland, which indicated the presence of serine proteases transcripts comprising about 0.5% [26], and also of L. intermedia venomous gland, which revealed about 0.3% of transcripts encoding serine proteases [25]. These data indicate inter-species conservation, strengthening the biological importance of these proteases, although in both cases they are low expressed in comparison to other toxins such as the phospholipases D and ICKs.
The results based on transcriptomic analysis were confirmed by proteomic studies, which showed the presence of serine proteases in the crude venom of L. intermedia, indicating that these molecules are indeed components of brown spider venoms [43]. As reported for astacins, the purification of native serine proteases from Loxosceles spiders’ crude venoms is virtually unviable and the production of these enzymes in heterologous models has proved to be quite difficult. For these reasons, the biological functions of serine proteases in the venoms of brown spiders have not been determined yet, leading to only speculative hypothesis regarding their functionality. Based on these theoretical assumptions, as described for the astacins, the serine proteases could act as molecular scissors that work by activating precursors molecules in the crude venoms or in victim’s body, and/or digestive enzymes that participate in the extracorporeal digestion occurring shortly after the envenoming of the prey. However, the precise functions of serine proteases in the envenoming are still to be determined.
Hyaluronidases are found in several animal venoms including those of spiders [52-56] snakes [57], caterpillars [58], and bees [59,60]. Hyaluronidase activity in brown spider venom was first reported in the venom of L. reclusa [61]. Later, other description regarding hyaluronidase in a brown spider venom was made by Wright et al. [62]. In this study, a hyaluronidase was purified from the venom glands of L. reclusa and exhibited activity on hyaluronic acid (optimum pH 5.0 - 6.6) and chondroitin sulfate, being the former the preferred substrate. The authors also tested the in vivo activity of this enzyme and observed the development of a mild erythema in guinea pigs after 6 hours that subsided over the next 24 hours. Although a complete neutralization of purified L. reclusa hyaluronidase by the gamma globulin fraction of a specific antivenom raised against L. reclusa crude venom was observed in vitro, the whole antiserum exhibited only a slight inhibitory effect on the spreading action of the venom [62]. The presence of antibodies in the antivenom that recognize L. reclusa purified hyaluronidase was confirmed later by immunodiffusion studies [63].
At that time, toxinologists already understood that hyaluronidases in spider venoms were not a toxic element per se, but probably act as a spreading factor [64,65]. Later, hyaluronidase activity was described in the venom of various Loxosceles species, e.g. L. deserta, L. gaucho, L. intermedia and L. laeta, and this activity appeared in a hyaluronic acid zymogram as a 44 kDa enzyme [38]. L. intermedia hyaluronidase was further characterized by da Silveira et al. (Figure 3B) [66]. These authors showed that this hyaluronidase is a hydrolase characterized as an endo-b-N-acetyl-D-hexosaminidase, has an optimal activity at 6.0-8.0 pH and hydrolyze both hyaluronic acid and chondroitin sulfate in vitro and hyaluronic acid in vivo [66]. Proteomic and transcriptomic analyses identified few sequences that correspond to hyaluronidase, evidencing that they are low-abundance toxins in Loxosceles venom, comprising 0.13% of L. laeta and 0.05% of L. intermedia venom gland transcripts [25,26,67].
The role of Loxosceles venom hyaluronidases as spreading factors was first demonstrated when a recombinant L. intermedia hyaluronidase was produced [68]. Dietrich’s hyaluronidase, as it was named, was expressed in E. coli cells and subjected to in vitro refolding in order to obtain a soluble and active 6xHis tagged enzyme with ~45 kDa (Figure 3C). This recombinant enzyme retained linear antigenic determinants from native hyaluronidases of Loxosceles crude venom as demonstrated by immunoassays. Finally, Dietrich’s hyaluronidase increased the area of dermonecrosis and enhanced edema induced by a recombinant phospholipase D, as well as triggered the gravitational spreading of the lesion. These data proved the role of Loxosceles hyaluronidases as a spreading factor of other toxins near the bite site [68]. Two other spider hyaluronidases were produced as recombinant CsHyal (from Cuppienius salei), which was produced in E. coli and further refolded, and BvHyal (from Brachypelma vagans), which was expressed using baculovirus system in insect cells [69,70]. CsHyal potentiated the insecticidal activity of neurotoxins in invertebrate preys, and authors speculated that this glycosidase may act as a spreading factor that enhance the activity of neurotoxic venom compounds [70].
Recently, a novel isoform of Loxosceles intermedia venom hyaluronidase was produced in a baculovirus-infected insect cells system and named LiHyal2 [71]. This recombinant glycosidase was produced as an active glycosylated enzyme and the biological characterization of LiHyal2 confirmed its ability in acting as a spreading factor.
By using two linear epitopes of Dietrich’s hyaluronidase (NGGIPQLGDLKAHLEKSAVDI and ILDKSATGLRIIDWEAWR) combined with epitopes of other Loxosceles toxins (e.g. astacin-like protease and phospholipase D), Lima et al. [50] produced a recombinant multiepitopic protein named rMEPLox (recombinant MultiEpitopic Protein derived from Loxoscelic toxins). Antibodies against this protein efficiently neutralized hyaluronidase activity of L. intermedia venom [50]. These results represent one of the various possibilities of using Loxosceles hyaluronidases as biotools for therapeutical applications. These enzymes are related to several physiological and pathological processes. In fact, venom hyaluronidases are of interest also because they belong to the same class of mammal hyaluronidases, as mentioned earlier. The use of a recombinant human PH20 hyaluronidase (rHuPH20; Halozyme Therapeutics, Inc.) to overcome the resistance to bulk fluid flow in the subcutaneous space and favor drug delivery, dispersion, and absorption, is currently FDA-approved. It acts by degrading hyaluronic acid, facilitating the route of administration and optimizing the dosage of subcutaneous therapies [72].
A recombinant human hyaluronidase was also proved to be effective and secure as a facilitating agent for subcutaneous immunoglobulin in a retrospective, multicenter study (fSCIG; HyQviaR ®) in elderly patients with primary or secondary immunodeficiencies [73]. They are also widely applied in the field of dermatology, to degrade hyaluronic acid filler to reverse cutaneous augmentation with this glycan [74]. In short, the production of recombinant spider venom hyaluronidases is a promising alternative since they can contribute to understand the role of these glycosidases in the venom and to the development of specific therapies to treat loxoscelism, besides having various potential applications for the pharmaceutical industry [75,76].
Accidents involving bites caused by bees and ants, in addition to exposures to animals such as cockroaches and mites can generate serious allergic reactions such as fever, edema, vertigo and anaphylactic shock [77-81]. In spiders, little is known about allergenic molecules and their biological activities. However, in accidents involving spiders from the Loxosceles genus, symptoms at the cutaneous tissues as itch, erythema, edema, cutaneous rash, and in some cases acute generalized exanthematous pustulosis are common, suggesting allergic reactions in some instance [6,28,82,83]. The presence of toxins with hyaluronidase activity in the venoms of Loxosceles spiders, which are strong allergenic factors in bee venoms, was initially described by studying the crude venom of L. reclusa [62]. Later, toxins characterized as hyaluronidases were identified in venoms of other Loxosceles species, as mentioned earlier [37,38,66], however there was never a direct correlation of these toxins with allergenic activities in these venoms.
The first scientific report demonstrating the presence of allergens in venoms of brown spiders was obtained through proteomic analysis of the crude venom of L. intermedia using MS/MS mass spectrometry [43]. The authors pointed out the presence of a toxin similar to the mite allergen of group 7 and suggested this toxin could be directly related to allergenic activities seen in injured patients. In this same way, by analyzing the transcriptome of the venom-producing glands of L. laeta, it was shown that 0.6% of toxin coding transcripts were related to a toxin characterized as a venom allergen III, which is conserved in animals with an allergenic potential such as the ant Solenopsis invicta, the wasps Vespula vulgaris and Dolichovespula maculata and the mosquito Aedes aegypti [26]. This toxin could hypothetically be related, according to the authors, to allergic reactions described during the envenoming caused L. laeta venom.
The presence of allergenic factors in the venoms of Loxosceles spiders was also demonstrated in the study regarding the transcriptome of L. intermedia venom glands, in which 0.2 % of transcripts encoding toxins were described as being allergens [25]. The similarity of these allergenic toxins from L. intermedia with other allergens present in the venoms of the spider Lycosa sigoriensis and the scorpoion Opisthacanthus cayaporum as well as mite allergens of Ixodes scapularis and Argas monolakensis suggests the involvement of this toxin in the possible allergic responses seen after accidents. Finally, the existence of an allergen in the venoms of Loxosceles spiders was greatly suggested by a study using molecular biology techniques, which reported cloning of an allergen named as LALLT (Loxosceles allergen-like toxin) [19]. This toxin was cloned, expressed in the baculovirus/insect cells system, purified and some of its biological activities were reported (Figure 3C). The allergen from L. intermedia venom has a molecular mass of 42 kDa and presents epitopes that cross reacted with anti-venom sera developed using crude venoms from L. laeta and L. gaucho, suggesting that these molecules are conserved in different species of the Loxosceles genus. This biological conservation was also revealed by analysis involving multiple sequences alignment of cDNA-deduced amino acid sequences for LALT orthologues from L. laeta and from L. gaucho venom, reinforcing the existence of a family of allergens in the venoms of brown spiders [19].
LALLT has 18 cysteine residues and belong to the CAP superfamily, showing significant identity to other allergens from spiders, scorpions, mites and ticks. Experiments of biological characterization of LALLT showed that this recombinant molecule caused edema in the skin of rabbits, increased vascular permeability and triggered paw edema in mice, besides increasing calcium influx and inducing release of beta-hexosaminidase from mast cells (RBL-2H3) in vitro. Finally, LALLT caused degranulation of rat mesentery mast cells [19]. In this same study, histological analysis of the skin of rabbits exposed to recombinant LALLT revealed edema and an infiltrate of inflammatory cells on the dermis, findings that are common in the histopathological analysis of samples from patients with allergies caused by arthropods bites [84].
All the data exposed here support the existence of allergens in the venom of Loxosceles spiders, providing valuable information that can assist in the treatment of accidents in the hospital or outpatient settings. The fact that these toxins belong to the group of low expressed toxins in Loxosceles venoms has to be taken into account, since their low concentrations can make these allergenic responses uncommon, restricted to more susceptible patients.
Loxosceles intermedia TCTP protein, LiRecTCTP, was identified in the cDNA library of the brown spider venom gland of L. intermedia [25]. This protein exhibits a high degree of similarity with tick TCTPs (~70%), which are described as histamine release factors [85-87]. The complete sequence identified in the cDNA library contains 536 bp, which encodes a 172-amino acid protein with a predicted molecular mass of 22.3 kDa and a pI of 4.7 (mature TCTP) [87]. Additionally, the transcriptome study of the L. intermedia venom gland identified 0.2% of the total transcripts corresponding to TCTP protein transcripts [25]. In 2012, Sade et al. [87] performed cloning, heterologous expression, purification and functional characterization of the L. intermedia TCTP (Figure 3C). The recombinant protein, expressed in E. coli with a 6 His-tag at the N-terminus, was called LiRecTCTP. Purification involved two chromatography steps - an affinity chromatography using Ni-NTA agarose resin with subsequent ion exchange chromatography (DEAE-sepharose) [87]. Recently Boia-Ferreira et al. [88] have standardized a new purification protocol with higher yield and purity using the Akta purified system and affinity chromatography (Ni-NTA agarose). Secondary structures and solubility analyses of LiRecTCTP were performed by using circular dichroism spectroscopy and showed the proper folding features of the recombinant protein [88].
Envenomation by Loxosceles spiders can cause hypersensitivity and allergic reactions. The cutaneous symptoms generated by the venom include erythema, edema, itching and pain. Rattmann et al. [89] demonstrated that the L. intermedia venom triggers mast cell activation and histamine-dependent effects. Initial inflammation events, such as increased vascular permeability, were related to the participation of histaminergic and serotonergic receptors. As mentioned earlier, many symptoms observed during loxoscelism can be mimicked by phospholipase D toxins (PLDs), the most characterized and studied family of toxins in Loxosceles venoms. However, recombinant PLDs are not able to induce paw edema with the same intensity as the crude venom, pointing to histaminergic events in the increased formation of edema during envenomation [6,87].
Brown spider venom was shown to be capable of causing regulated release of mast cell mediators, mainly histamine, responsible for inducing vasodilation in experiments with rat aorta using a chamber for an isolated organ [89]. Paludo et al. [90] identified the presence of histamine in the venom in sufficient quantities to exert inflammatory effects. Despite this, the dialyzed venom, without the presence of histamine, was still capable of exerting a certain histamine-dependent inflammatory effect, due to some other component present in the venom, acting directly on mast cells [89,90]. First Sade et al. [87] and then Boia-Ferreira et al. [88] demonstrated that the TCTP of L. intermedia participates in the exacerbated inflammatory process resulting from LiRecTCTP causes in vivo increased vascular permeability and edema in mice, in a time and concentration dependent manner [87,88]. Therefore, these results suggest the LiTCTP may be the first and fastest component to induce edema formation in loxoscelism pathophysiology [87]. Boia-Ferreira et al. [88] also demonstrated that LiRecTCTP is capable of activation mast cells (de-sensitized RBL-2H3 cells) leading to degranulation in vitro.
Expression profile analysis via quantitative real-time PCR showed that LiRecTCTP induced the cellular expression of cytokines involved in allergic and parasitic processes such as IL-3, IL-4, and IL-13 in cultured RBL-2H3 cells. In vivo assays showed that when LiRecTCTP was injected in mice together with inhibitors of histamine receptors (H1, H2, H3 and H4) a reduction in vascular permeability and edema was observed when compared to isolated toxin, confirming that this toxin is responsible for inducing these deleterious histaminergic effects [88]. H1 and H2 receptors inhibitors (prometazine and thioperamide, respectively) have been shown to significantly reduce the effects of LiRecTCTP on increasing vascular permeability. The degranulation inhibitor cromolyn, in turn, was able to abrogate the edematogenic effect prompted by LiRecTCTP. Furthermore, experiments of dermonecrosis using rabbits demonstrated a synergism between LiRecTCTP and a recombinant phospholipase D toxin. These data emphasizes LiRecTCTP relevant participation in the inflammatory and histaminergic cutaneous effects of loxoscelism for acting as a histamine release factor and thus contributing to the systemic dispersion of other venom components [88].
TCTP-related proteins were also identified in the venom of other Loxosceles spiders (L. laeta and L. gaucho) by immunoblot cross-reactivity assays [91]. Literature on TCTP from spiders is scarce but some sequences were identified in the venom of spiders from different species [92-95], and in gland secretions of ixodid tick parasites, which are also arachnids [96]. Concerning the biological and evolutionary purpose of this toxin to be present in Loxosceles venom we must highlight that L. intermedia TCTP does not present a signal peptide for endoplasmic reticulum translocation, this toxin can be secreted by exosomes and also by the holocrine secretion pathway [24], as other constituents of whole venom [87].
Studies on TCTP as a venom toxin are very few and its biological and evolutionary role as a venom component in prey capture remains still unknown [97]. In contrast, as a multifunctional protein involved in several biological processes, its biotechnological potential is enormous and yet to be further explored.
Another group of toxins found in the venoms of Loxosceles spiders includes protease inhibitors of the serine protease family characterized as serpins [28,98]. The first evidence in the literature pointing out the existence of protease inhibitors in Loxosceles spiders was described using crude venom of L. reclusa, which showed the presence of a potent inhibitory activity on the complement-dependent hemolysis (an event highly dependent of serine proteases). This component of the venom showed properties such as not being dialyzable, but can be excluded from the venom by means of a gel filtration chromatography using Sephadex G-75, in addition to being stable under a broad range of pH [99]. Serine protease inhibitors were initially described in the venoms of brown spiders by means of transcriptome analyzes of L. laeta venom-producing glands. In this study, transcripts encoding serine protease inhibitors corresponded to 0.6% of the total transcriptome [26]. Serine proteases inhibitors were additionally identified through proteomic analyzes using crude venom of L. intermedia that was submitted to sequential chromatography steps by using cation exchange and reverse phase to purify proteins and peptides that were identified by mass spectrometry MS/MS. These studies reported the presence of molecules in the venom characterized as trypsin inhibitor-like protein and serine protease inhibitor protein [43].
The presence of serine proteases inhibitor transcripts in the venom glands of Loxosceles spiders was later identified in the transcriptome analysis of the venom-producing gland of L. intermedia, which showed the presence of 0.1% of toxin encoding transcripts identified as serine protease inhibitors [25]. Finally, a further biochemical characterization supporting the presence of serine protease inhibitors in the venom of Loxosceles spiders is currently being carried out by using a recombinant serine protease inhibitor from L. intermedia [100]. The reasons for the presence of serine protease inhibitors in the venom of Loxosceles spiders, as well as the physiological targets of these molecules are still unknown. It was proposed that these toxins, through their inhibitory activities on proteases, could protect the integrity of other venom components, and thus increase the useful life of venom toxins exposed to an external proteolytic environment, for instance when the venom is released to protect the spider against predators or to kill their prey [43].
An interesting fact that suggests the biological importance of serine protease inhibitors for Loxosceles spider venom is that toxins found in the venom of L. laeta are similar to serine protease inhibitors from different animals as Mus musculus, Aedes aegypti, Branchiostoma lanceolatum, Gallus gallus and Boophilus microplus [26]. This also is valid for serine protease inhibitors described in the L. intermedia venom, which are quite similar to inhibitors found in mammals such as Mus musculus and Pan troglodytes, or in the tick Ambliomma americanum [25]. These toxins fit into the families of molecules with low expression in the venoms of Loxosceles spiders [25] and perhaps that is why they have been little studied so far. However, the identification of transcripts coding for serpins present in the venom-producing glands, and the production of recombinant molecules as tools, will contribute and help elucidate the functions for these inhibitors, as well as contribute to structural analysis to understand the relation between structure and function of these toxins. Finally, but not less important is the possible uses of these molecules, since there are numerous examples of biotechnological applications of analogs of serpins in the control of blood clotting, anti-tumor activity and viral infection treatments [98,101-105].
Inhibitory Cystine Knot (ICK) peptides are single-chain molecules enriched in cysteine residues, which establish intramolecular disulfide bonds. The disulfide bonds are organized in a specific pattern in which two of them together with the peptide backbone form a ring that is crossed by the third disulfide bond. This disulfide bonds’ arrangement creates a pseudo-knot framework, which is why these peptides are also known as knottins [106-108]. A large number of studies have already shown that ICK peptides display insecticidal activity [109-111]. For this reason, the foremost function regarding ICK peptides in spider venoms concern the predation for feeding purposes, especially insects [112-114]. It is also due to this insecticidal activity that ICK peptides have been biotechnologically explored in order to develop alternative bioinsecticides to the harmful chemical compounds still used [113,115].
The ICK peptides from Loxosceles spiders were first studied by De Castro et al. [110] who fractioned the crude venom of L. intermedia and identified a fraction with insecticidal activity (Figure 3A). Further chromatographic steps sequentially performed allowed the purification of three peptides named LiTx1, LiTx2 and LiTx3 with insecticidal activity (induced flaccid paralysis) on the larvae of economic interest Spodoptera frugiperda. By using amino acid sequencing and molecular biology methodologies, De Castro et al. [110] obtained the coding sequence for these peptides, which revealed that these peptides are produced as prepropeptide precursors (signal peptide, propeptide and mature peptide). Later, the authors included a sequence related to a fourth isoform (LiTx4) on the GenBank, which has not had its insecticidal activity tested yet.
The transcriptome of the venom-producing glands of L. intermedia published by Gremski et al. [25] revealed that the majority of the sequences expressed regarding toxins was related to ICK peptides (55.9%). Sequences with high identity with the LiTx1-4 peptides represented 53.5%, and 2.4% of that transcripts showed significant similarity with a neurotoxic ICK peptide from the spider Macrothele gigas [116]. In addition to the annotation of the sequences mentioned, Gremski et al. [25] analyzed the venom protein content by SDS-PAGE, which suggested that ICK peptides are massively predominant in the venom of L. intermedia.
Matsubara et al. [117] investigated an ICK peptide sharing 86% sequence identity with LiTx3, which was named U2-sicaritoxin-Li1b (U2-SCRTX-Li1b) in agreement with the rational nomenclature developed by King et al. [118]. The authors cloned the sequence and expressed the peptide in bacterial cells, resulting in the production and purification of the first recombinant ICK peptide from Loxosceles’ venoms. Using the recombinant U2-SCRTX-Li1b and hyperimmune sera raised against different Loxosceles spider venoms, the authors performed immunoassays that showed antigenic cross-reactivity, pointing out that ICK peptides constitute a family of toxins widespread throughout the genus [117]. An additional ELISA cross-reactivity analysis using polyclonal antibodies and the recombinant peptide U2-SCRTX-Li1b or whole venom performed by Buch et al. [91] suggested that ICK peptides are present in the venoms of L. intermedia, L. laeta and L. gaucho, reinforcing once more that these peptides belong to a conserved family of toxins in Loxosceles spiders. Interestingly, Buch et al. [91] also carried out western blotting analysis that did not show cross-reactivity between the recombinant peptide and polyclonal antibodies that recognize L. laeta and L. gaucho venoms or between the polyclonal antibodies that recognize the peptide U2-SCRTX-Li1b and L. laeta and L. gaucho whole venoms.
Meissner et al. [119] studied another ICK peptide from L. intermedia, whose sequence was identified in the venom gland transcriptome [25]. This peptide - U2-SCRTX-Lit2 - shares 52% identity with the toxin μ-hexatoxin-Mg2a (µ-HXTX-Mg2a) from M. gigas, in addition to the fact that both peptides contain 10 cysteine residues that establish the same disulfide bond connectivity pattern. A great deal of ICK peptides have been described as highly specific to insects, interacting with ion channels or membrane receptors in their nervous system and then resulting in paralysis and death. In order to investigate the target of U2-SCRTX-Lit2, Meissner et al. [119] performed molecular docking and dynamics analyses using a voltage-gated sodium channel from Spodoptera litura (tobacco cutworm), whose structure has already been determined. The choice of this target was also due to the fact that µ-HXTX-Mg2a was able to cause paralysis in S. litura and was associated with the inhibition of voltage-gated sodium channels (SlNaVSC) in synaptosome preparations obtained from cockroaches by binding to the site 3 of these channels [116]. Bioinformatics data showed that U2-SCRTX-Lit2 presents amino acid residues arranged in a pattern that suggests affinity to the site 3 of the SlNaVSC and revealed that the peptide may act as a steric blocker, hiding the gate access of these channels [119]. Sequence analyses comparisons have pointed out that the peptides LiTx3 and U2-SCRTX-Li1b may act on voltage-gated sodium channels as well [110,117].
Another study regarding ICK peptides from L. intermedia was carried out by Matsubara et al. [111]. In this study, the peptide U2-SCRTX-Li1b was recombinantly expressed in the periplasm of bacterial cells. This strategy was selected because the periplasm of E. coli provides the molecular machinery that assists in the correct formation of disulfide bridges, in contrast to the reducing environment of the cytoplasm that disadvantages the establishment of these structures [120]. After purification, recombinant U2-SCRTX-Li1b was able to cause long-lasting paralysis in sheep blowflies (Lucilia cuprina), which was irreversible even after 72 hours. Therefore, U2-SCRTX-Li1b constitutes the first recombinant ICK peptide from Loxosceles spiders to have its activity determined [111]. Furthermore, the authors carried out a screening of sequences encoding ICK peptides in other two Loxosceles species (L. gaucho and L. laeta) from the total RNA produced in the venom glands of the spiders. This screening of venom-gland transcripts resulted in the obtainment of sequences encoding orthologues of LiTx1-4 peptides, with identities ranging from 83% to 100% compared to the sequences encoding ICK peptides of L. intermedia. All sequences encoding ICK peptides found contain 10 cysteine residues in their mature sequence and exhibit the same predicted disulfide bond connectivity pattern [111].
In 2017, Trevisan-Silva et al [67] published a revealing proteomic analysis of the whole venom of L. intermedia by using a multi-protease, multi-dissociation, bottom-up-to-top-down approach. This study identified ICK peptides from Loxosceles venoms at a proteomic level for the first time, resulting in the identification of peptides with correspondence to LiTx in high abundance as depicted by the L. intermedia transcriptome [25].
ICK peptides have also been described as toxins that help spiders in defending against their predators. In addition, due to the anthropic action, many species of spiders have had their natural habitats destroyed and, consequently, they have been recurrently found in peridomiciliar environments, which facilitates accidents with humans [3,6]. Due to natural interactions with predators and episodic interactions with humans, many species of spiders have evolved ICK peptides with harmful properties to these organisms. Loxosceles bites result in a mild stinging that usually is painless in humans [3,6], which can suggest the existence of molecules with anesthetic or analgesic effects in the venom. Some ICK peptides in other spiders have already proved to display analgesic effects on animal models such as the peptide PcTx1 (µ-TRTX-Pc1a) from the tarantula Psalmopoeus cambridgei, which interacts with acid sensing ion channels and results in analgesic properties in rat models for acute pain when administered intrathecally or intracerebroventricularly [107,121]. Hence, given the great diversity of ICK peptide-coding sequences and the painless aspect of Loxosceles spider bites, the search for peptides with possible analgesic activities remains a promising idea.
Much learning has been gathered about astacins, serine proteases, knottins, TCTP, hyaluronidases, allergens and serpins (Table 1). However, we are still at an initial phase in understanding the full role of these proteins in the brown spider venom and how they can work together to affect the tissue of victims. Novel strategies must be undertaken to overcome the barrier of obtaining enough amount of these toxins to enable further investigation and comprehension of the pathophysiology of loxoscelism. Then, it will be possible to use this information to improve therapeutic strategies for treating affected patients. In addition, a deeper knowledge on functional and structural aspects of these poorly explored toxins will certainly reveal new possible applications in diverse areas.