Biochemical and Electrophysiological Characterization of Two Sea Anemone Type 1 Potassium Toxins from a Geographically Distant Population of Bunodosoma caissarum

Sea anemone (Cnidaria, Anthozoa) venom is an important source of bioactive compounds used as tools to study the pharmacology and structure-function of voltage-gated K+ channels (KV). These neurotoxins can be divided into four different types, according to their structure and mode of action. In this work, for the first time, two toxins were purified from the venom of Bunodosoma caissarum population from Saint Peter and Saint Paul Archipelago, Brazil. Sequence alignment and phylogenetic analysis reveals that BcsTx1 and BcsTx2 are the newest members of the sea anemone type 1 potassium channel toxins. Their functional characterization was performed by means of a wide electrophysiological screening on 12 different subtypes of KV channels (KV1.1–KV1.6; KV2.1; KV3.1; KV4.2; KV4.3; hERG and Shaker IR). BcsTx1 shows a high affinity for rKv1.2 over rKv1.6, hKv1.3, Shaker IR and rKv1.1, while Bcstx2 potently blocked rKv1.6 over hKv1.3, rKv1.1, Shaker IR and rKv1.2. Furthermore, we also report for the first time a venom composition and biological activity comparison between two geographically distant populations of sea anemones.

In this study, we report for the first time the characterization of the -neurotoxic fraction‖ from the venom of B. caissarum SPSPA population, and under the same experimental conditions, we compare it to the population found in the state of Sã o Paulo littoral (southeast coast of Brazil; S23°56′, W45°20′) ( Figure 1). Furthermore, we present the purification, biochemical analyses and electrophysiological characterization of two new type 1 sea anemone toxins, as well as their relationship with other known toxins based on sequence, structural and evolutionary analyses.

Venom Purification and Biochemical Characterization of BcsTx1 and BcsTx2
Sea anemone venom extraction by electric stimulus provides a massive release of proteins, peptides and low molecular weight compounds from the nematocysts [23][24][25]. When this toxic mixture is applied to a Sephadex G-50 gel-filtration column, the peptide content of the venom is separated from enzymes, such as phospholipases and cytolysin [26][27][28]. Gel filtration of B. caissarum venom on Sephadex G-50 yielded five fractions named Fraction I to V (FrI-FrV) (Figure 2A), as previously described for the venom of B. cangicum [26] and B. caissarum population from the southeastern coast of Brazil [29]. Gel-filtration Fraction III (FrIII) from B. caissarum SPSPA population had the highest neurotoxicity when tested on swimming crabs (Callinectes danae) (data not shown), and it was further purified by reverse-phase high performance liquid chromatography (rp-HPLC) ( Figure 2B). Elution peaks, labeled as 1 and 2 ( Figure 2B), were able to fully block the insect channel Shaker IR and, thus, were subjected to a second purification step, leading to the pure toxins BcsTx1 and BcsTx2 ( Figure    Interestingly, the venom of B. caissarum population from the Southeastern coast of Brazil shows hemolytic activity and one actinoporin, named Bcs I, had been purified and biochemically characterized [30,31]. However, neither the whole venom, nor the fraction II (FrII) of SPSPA population (Figure 2A), showed cytolytic activity when tested on erythrocytes (data not shown). Up to date, cytolysin has been found in all classes of cnidarians, and more than 32 species of sea anemones have been reported to produce lethal cytolytic peptides [28,32]. Also, it has been shown that one species of sea anemone (e.g., Actinia equina) can produce more than one isoform, while others are devoid of any cytolytic activity (e.g., Anemonia viridis) [28,33]. Also, the incidence of cytolytic activity in corals (Anthozoa and Hydrozoa) is high, resembling the sea anemones, where cytolysins are widespread [32]. Gunthorpe and colleagues compared the bioactivity of aqueous extracts of scleractinian corals (Cnidaria, Anthozoa, Hexacorallia) from different families and concluded that the occurrence of cytolytic activity do not differ significantly among the genera and the species considered, except for the extracts of colonies of Goniastrea australensis, where intraspecific differences were found [34].
The rp-HPLC profile of fraction III (FrIII) of the SPSPA B. caissarum population yielded a very similar profile to that from the Southeastern coast of Brazil [35], suggesting that both populations releases a similar pattern of neurotoxic peptides ( Figure 3). Until now, only two toxins from B. caissarum venom have been investigated: (i) BcIII that belongs to type 1 neurotoxins and bind at site 3 of the voltage-gated sodium channel (Na V ), delaying the inactivation process [29], and (ii) BcIV, which does not have its exact target determined, yet. However, experiments using crab leg sensory nerve suggest a Na V -activity [35]. A superimposition of both rp-HPLC profiles of -neurotoxic fractions‖ from B. caissarum populations allows us to point out the following: (1) SPSPA sea anemone population has a BcIII-like toxin and (2), at the same retention time of BcIV, no elution peak is observed on the chromatographic profile of the SPSPA population. To our knowledge, such a degree of intraspecific variation in the peptide composition of sea anemone venom is novel.
Moran and colleagues [36,37] analyzed the evolution of a voltage-gated Na + channel neurotoxin genes family from three genetically and geographically distinct populations of the sea anemone Nematostella vectensis [38,39] and from single specimens of Actinia equina and Anemonia viridis.
Genomic data indicated much higher similarity among toxin genes within each species than to toxin genes of other species, suggesting that related neurotoxin genes family in sea anemones are subjected to a concerted evolution [40]. The authors also demonstrated that evolution driven by positive Darwinian selection would have occurred, as observed by the numerous substitutions in the putative neurotoxin genes from A. equina and A. viridis.  [29] and BcIV [35]. Red dotted line is the FrIII chromatographic profile of the SPSPA population. -Neurotoxic fractions‖ were submitted to rp-HPLC chromatography using a semi-preparative CAPCELL PAK C-18 column (1 × 25 cm, Shiseido Corp.), and their components were eluted with a linear gradient from 10% to 60% of acetonitrile containing 0.1% TFA, as described in the Experimental Section.
Intraspecific diversity in the venom composition of various animal species, such as cone snails [41,42], bees [43], ants [44,45], spiders [46,47], scorpions [48,49] and snakes [50,51], have been reported using biochemical, pharmacological, proteomic and/or transcriptomic approaches. Abdel-Rahman and colleagues [52] used a combination of proteomic and biochemical assays to examine variations in the venom composition of the vermivorous Conus vexillum taken from two distinct geographical locations and concluded that the venom is highly diversified. Moreover, intraspecific variation in the peptides present in the venom from two species of fish-hunting cone snails (C. striatus and C. catus) has been reported. However, the venom compositions of individual snails of both species remained quite constant over time in captivity [42]. In contrast, proteomic analyzes of the venom of several specimens of a piscivorous cone snail (C. consors) revealed dramatic variations over time, which could be related to dynamics of peptide production by the secretory epithelium in the venom gland [53].
Similarly to cone snails, venom variability in specimens of Tityus serrulatus scorpion, collected within the same geographical area, has been shown. Specimens showed venom constituent variations, which were related to extraction events and to dynamics in gland production and peptide maturation [54,55]. Furthermore, investigation of intraspecific venom variation of four different populations of Scorpio maurus palmatus from geographically distant locations revealed highly significant differences among all populations and within each population studied. This may be due to geographic differential distribution of prey species, as well as their relative abundance in the environment [49,56]. Also, it has been demonstrated that ontogenetic variation of viperid snakes (Chordata, Reptilia, Viperidae) venoms could be related to differences between the feeding habits of juvenile and adult snakes, suggesting that variation in venom composition may reflect natural selection for greater efficiency in killing and digesting different prey types within the same location or in different locations [57][58][59].
Thus, the relationship between geographic distance and patterns of venom composition implicates spatial scale and localized ecological and genetic factors, such as gender, elapsed time after capture, dynamic expression of the gland and peptide maturation, genetic variation, environmental conditions, seasonality and geographical locations. In the current work, these factors were not standardized (except for venom collection and sea anemone size, presuming a similar age of specimens of each population), and additional studies will be necessary in order to assess more precisely these variations in venom composition and to enhance our understanding of the forces driving sea anemone venom evolution.
Interestingly, type 1 sea anemone toxins could be classified as belonging to the six-cysteine (SXC) protein domain, whose first members were identified in surface coat components of the dog ascaridid Toxocara canis (Nematoda, Secernentea) and, later, also have been identified in many additional nematodes [70,71]. This domain is composed of short (36 to 42 amino acids) peptides, with six conserved cysteines, that can be found in many parasitic nematodes, such as Ascaris suum and Necator americanus [72]. The physiological role of these peptides has not been established, yet; however, it is believed that they might interfere with the local and systemic immune system and with gut muscles of the host [73]. As already mentioned, sea anemone type 1 toxins possess a conserved -functional dyad‖ motif, which is not universally present in nematodes ( Figure 5). However, if we observe the basic and aromatic residues (lysine and phenylalanine) of the putative protein from Ascaris suum, we might suggest a possible K V channel blocker activity. Thus, considering that, throughout evolution, proteins found in venoms are the result of toxin recruitment events in which a protein gene involved in a regulatory process is duplicated and the new gene is selectively expressed in the venom apparatus [74], we may suppose that the existence of the SXC domain in different phyla reflect their common ancestry.

BcsTx1 and BcsTx2 Pharmacological Profiles
Sequence alignment and phylogenetic analysis ( Figure 4A,B) indicated that BcsTx-1 and -2 are new members of the type 1 (subtype 1b) toxins from sea anemones that are known to be potent inhibitors of K V channels. The pharmacological profile of BcsTx-1 and -2 were determined on a wide range of twelve K V channels (rK V 1.1, rK V 1.2, hK V 1.3, rK V 1.4, rK V 1.5, rK V 1.6, rK V 2.1, rK V 3.1, rK V 4.2, rK V 4.3, hERG and the insect channel Shaker IR; r: rat and h: human). Channels were expressed in X. laevis oocytes, and their currents were recorded by using two-electrode voltage-clamp technique. BcsTx1 (0.5 µM) inhibited rKv1.1, rKv1.2, rKv1.3, rKv1.6 and Shaker IR channels with 44% ± 2%, 100%, 100%, 88% ± 3% and 64% ± 4%, respectively ( Figure 6). BcsTx2 (3 µM) showed an effect on potassium currents inhibiting 96% ± 2.1%, 100%, 100%, 98% ± 1.75% and 94% ± 2% of rK V 1.1, rK V 1.2, hK V 1.3, rK V 1.6 and Shaker IR, respectively (Figure 7). Type 1 toxins, such as BgK and ShK, have been extensively characterized. BgK was found to block K V 1.1-3 and K V 1.6 channels with potencies in the nanomolar range [60]. ShK was originally found to block K V 1.3 channels [69,75], but also blocks K V 1.1-4 and K V 1.6 [61,76]; and more recently, it has been found that ShK shows activity against K V 3.2 channels [77]. Both BgK and ShK block intermediate conductance K (Ca) channels [78]. Some of the other type 1 toxins were indirectly assayed by competitive inhibition of the binding of 125 I-dendrotoxins, allowing the conclusion that they will show activity on K V 1.1, K V 1.2 and/or K V 1.6, since dendrotoxins only block the current of these K V channels. The AsKs toxin has been characterized as a blocker of K V 1.2 current expressed in Xenopus oocytes, and no biological activity has been reported to FC850067, FK724096, FK755121 and FK747792 [13,[62][63][64][65]. Thus, it is worth mentioning that our work represents the first electrophysiological characterization of type 1 sea anemone toxin activity on cloned Shaker IR insect channel.  Representative whole-cell current traces in the absence and in the presence of 3 µM BcsTx2 are shown for each channel. The dotted line indicates the zero-current level. The * indicates steady state current traces after application of 3 µM BcsTx2. This screening carried out on a large number of K V channel isoforms belonging to different subfamilies shows that BcsTx2 selectively blocks Shaker channels subfamily.
In order to characterize the potency and selectivity profile, concentration-response curves were constructed for BcsTx1. IC 50 Figure 8A and Table 1). A concentration-response curve was also constructed to determine the concentration at which BcsTx2 blocked half of the channels. The IC 50 values calculated are 14.42 ± 2.61 nM for rK V 1.1, 80.40 ± 1.44 nM for rK V 1.2, 13.12 ± 3.29 nM for hK V 1.3, 7.76 ± 1.90 nM for rK V 1.6 and 49.14 ± 3.44 nM for Shaker IR ( Figure 8B and Table 1). Similar to BgK, the BcsTx-1 and -2 potencies are within the nanomolar range and are more potent when compared to type 2 sea anemone toxins, such as kalicludines (AsKC1-3), which blocks K V 1.2 channels with IC 50 values around 1 µM [63]. In general, previous work has shown that type 1 sea anemone toxins are more potent than type 2, and it has been proposed in the literature that toxins with a -functional dyad‖ are more potent, because it provides a secondary anchoring point, contributing to a higher toxin affinity [68,79]. However, APEKTx1, a type 2 toxin from A. elegantissima, is a selective blocker of K V 1.1, with an IC 50 value of 1 nM, and the existence of a -functional dyad‖ has not been shown [80]. Moreover, the electrophysiological characterization of the scorpion toxins Pi1 and Tc32 (from Pandinus imperator and Tityus cambridgei, respectively), which are known to potently inhibit K V 1 channels, suggested that other amino acids, rather than those of the -functional dyad‖, are also involved in both potency and selectivity of the K V channel isoforms [81,82]. Although, it is worth noting that the -functional dyad‖ of α-KTx family of scorpion toxins is very important for high affinity block and selectivity [83]. For instance, toxin Pi2 (α-KTx7.1), from the venom of P. imperator, has a -functional dyad‖ formed by Lys27 and Trp8 and is able to block K V 1.2 current with an IC 50 value (0.032 nM) comparable to BcsTx1 [84]. Also, MgTX (α-KTx2.2) toxin, from Centruroides margaritatus, binds with very high affinity to K V 1.6 (IC 50 value of 5 nM), and the role of the side chain of the dyad lysine (Lys27) as a critical residue to the binding of the toxin to the ion conduction pathway of the channel was proposed [85].
In order to elucidate whether BcsTx-1 and -2 block the current through a physical obstruction of the Shaker IR channel pore or act as gating modifiers, current-voltage (I-V) experiments were performed. The currents were inhibited at the test potentials from −90 to 100 mV, and the inhibition was not associated with a change of the shape of the I-V relationship. The control curve and the curve in the presence of BcsTx1 (500 nM) were characterized by a V 1/2 values of 20.85 ± 0.69 mV and 22.62 ± 0.73 mV, respectively. Moreover, the control curve and the curve in the presence of BcsTx2 (50 nM) were characterized by a V 1/2 values of 18.49 ± 1.49 mV and 23.88 ± 1.57 mV, respectively. The V 1/2 of activation was not significantly shifted (p < 0.05), and thus, channel gating was not altered by BcsTx1 and BcsTx2 binding ( Figure 8C,D). Additionally, BcsTx-1 and -2 shows a non-dependence of voltage for the blockage on a wide range from −10 mV to 50 mV ( Figure 8E,F); the blockage effect was reversible, and a complete recovery was observed after washout, suggesting an extracellular site of action ( Figure 8G,H). To date, type 1 sea anemone toxins have been described to act solely through a K V channel pore-blocking mechanism [1].

Bioinformatics Analysis
Molecular Models of BcsTx-1 and -2 Venomous animals produce a wide variety of neurotoxins with different types of amino acid sequences, secondary structures and disulfide bridge frameworks, and none of them is definitively associated with a particular animal species or ion channel selectivity [79]. Type 1 sea anemone toxins are associated with the αα type of family fold. BgK toxin has a -helical cross-like‖ motif in which one α-helix is disposed perpendicular to the others [67] ( Figure 9A) and ShK has a -helical capping‖ motif (3 10 αα), since one α-helix (formed by three amino acid residues) caps the other two helical structures [86]. The molecular models of BcsTx-1 and -2 ( Figure 9B,C) were constructed using BgK as template, and the quality of the models were analyzed using PROCHECK [87]. BcsTx-1 and -2 share 55.3% and 62% of sequence identity with BgK, respectively. BcsTx1 and BcsTx2 analyses revealed that 87.1% and 90.0% of residues are located in the most favored regions, 12.9% and 6.7% are located in additionally allowed regions and 0% and 3.3% are located in generously allowed regions of the Ramachandran diagram, respectively [88]. The secondary structure of both toxins consists of three α-helical segments; the first α-helix comprises the amino acids 8-17, the second comprises the residues 24-29 and the third α-helix consists of the amino acids 31-34. Despite the overall moderate identity between these three toxins, the residues of the second and third α-helices are highly identical. BgK second α-helix shares 83.3% and 100% of identity to BcsTx-1 and -2, respectively, and the third is 100% identical within the three toxins.

Sea Anemone Collection, Venom Isolation and Neurotoxins Purification
Specimens of the sea anemone Bunodosoma caissarum (3.5-4.0 cm of diameter) were collected at the Saint Peter and Saint Paul Archipelago (N0°55′, W29°20′), Brazil. The sea anemones were maintained in aquarium for 24 h, and then the venom was obtained by electrical stimulation of the animals, according to the method of Malpezzi et al. [23]. The venom was fractionated first by gel-filtration chromatography using a Sephadex G-50 column (1.9 × 131 cm, GE Healthcare, Uppsala, Sweden), and afterwards, the fraction containing the neurotoxic peptides was submitted to reverse-phase HPLC chromatography in an ÄKTA Purifier system (GE Healthcare, Uppsala, Sweden) using a semi-preparative CAPCELL PAK C-18 column (1 × 25 cm, Shiseido Corp., Kyoto, Japan). Elution was done in a linear gradient from 10% to 60% of acetonitrile containing 0.1% TFA at a flow rate of 2.5 mL/min during 40 min, and the peptides were monitored at UV 214 nm. Pure BcsTx1 and BcsTx2 were obtained using an analytical CAPCELL PAK C-18 column (0.46 × 15 cm, Shiseido Corp., Kyoto, Japan) and different gradients of the solvent described above, at a flow rate of 1 mL/min. The protein content of the pure peptides was estimated by the bicinchoninic acid assay (BCA) method (Pierce, Rockford, IL, USA).

Mass Spectrometry Analysis
Mass spectrometry analyses were performed on an Ultraflex II TOF/TOF MALDI (Bruker Daltonics, Bremen, Germany) equipped with Nd-YAG Smartbeam laser (MLN 202, LTB) under reflectron mode. The laser frequency was adjusted to 50 Hz. The matrix, α-cyano-4-hydroxycinnamic acid (Sigma-Aldrich Co., St. Louis, MO, USA), was prepared at a concentration of 20 mg/mL in 1:1 acetonitrile containing 0.1% TFA solution. External calibration was performed using peptide calibration standard II (Bruker Daltonics, Bremen, Germany). Sample solution (1 μL) dropped onto the MALDI sample plate was added to the matrix solution (1 μL) and dried at room temperature. Data were analyzed using the FlexAnalysis 3.0 program (Bruker Daltonics, Bremen, Germany).

Electrophysiological Recordings
Two-electrode voltage-clamp recordings were performed at room temperature (18-22 °C) using a Geneclamp 500 amplifier (Molecular Devices, Sunnyvale, CA, USA) controlled by a pClamp data acquisition system (Axon Instruments, Union City, CA, USA). Whole-cell currents from oocytes were recorded from 1 to 3 days after injection. Bath solution was the same ND96 solution described above. Voltage and current electrodes were filled with 3 M KCl. Resistances of both electrodes were kept between 0.8 and 1.0 ΩM. The elicited currents were filtered at 500 Hz using a four-pole lowpass Bessel filter. Leak subtraction was performed using a −P/4 protocol. K V 1.1-K V 1.6 and Shaker IR currents were evoked by 500 ms depolarizations to 0 mV, followed by a 500 ms pulse to −50 mV, from a holding potential of −90 mV. Current traces of hERG channels were elicited by applying a +40 mV prepulse for 2 s, followed by a step to −120 mV for 2 s. K V 3.1, K V 4.2 and K V 4.3 currents were elicited by 500 ms pulses to +20 mV from a holding potential of −90 mV. To assess the concentration-response relationships, data were fitted with the Hill equation: where y is the amplitude of the toxin-induced effect, EC 50 is the toxin concentration at half maximal efficacy, [toxin] is the toxin concentration and h is the Hill coefficient. In order to investigate the current-voltage relationship, current traces were evoked by 10 mV depolarization steps from a holding potential of −90 mV. The activation curves were fitted with a Boltzmann relationship of the form: where V 1/2 is the voltage for half-maximal activation and s is the slope factor. The activation kinetics were obtained by mono-exponential fits to the raw current traces.

Structure Computational Modeling
3D-structures of B. caissarum toxins were modeled using the publicly available program MODELLER9v10 [93]. BcsTx-1 and -2 were modeled using as template BgK, a voltage-gated potassium channel toxin from the venom of the sea anemones Bunodosoma granulifera (PDB code: 1BGK). Models were refined based on predicted secondary structure using SCRATCH Protein Predictor [94] and PROCHECK [87].

Statistical Assessment
Comparison of two sample means was made using a paired Student's t test (p < 0.05). All data represent at least three independent experiments (n ≥ 3) and are presented as the mean ± standard error. All data were analyzed using Clampfit 10.3 (Molecular Devices, Sunnyvale, CA, USA) and Origin 7.5 software (Origin Lab., Northampton, MA, USA).

Conclusions
In summary, we demonstrate, for the first time, a venom composition and biological activity comparison between two geographically distant populations of sea anemones. Moreover, this is the first electrophysiological characterization of a sea anemone type 1 toxin on cloned Shaker IR insect channels, allowing us to suggest that the role of these toxins in the physiology of the sea anemone would be related with predation and defense against predators and highlights the possible application of these peptides as tools for research in neuroscience, as well as in the development of novel insecticides.