N-chlorotaurine (NCT) was obtained from two different sources. For dimerization assays, NCT was freshly prepared from the equimolar reaction between taurine (Sigma-Aldrich) and NaOCl (Sigma-Aldrich) in water (Klamt and Shacter, 2005). In the other assays, NCT was used as a crystalline sodium salt (MW 181.57 g/mol) (Gottardi and Nagl, 2002). Solutions of 250 mM NCT were freshly prepared in water and used for necessary dilutions in the experiments. H₂O₂ (30% v/v) was purchased from Sigma-Aldrich and fresh solutions of 500 mM were prepared in water before each experiment. All the other reagents were purchased from Sigma-Aldrich unless otherwise specified.

The recombinant actin-depolymerizing factor of N. caninum (NcADF) was expressed in a bacterial model as described, with minor modifications (Baroni et al., 2018). Briefly, Escherichia coli BL21(DE3) transformed with pET28(a)_NcADF recombinant plasmid was cultured in Luria Bertani (LB) medium and the protein expression was induced with 0.2 mM isopropyl β- d-1-thiogalactopyranoside (IPTG) at 18°C during 18 h. Bacteria were collected and sonicated (Branson Sonifier SLPe, Emerson Electric Co.) in lysis buffer (50 mM Tris, pH 7.0, 300 mM NaCl, 10% glycerol, 20 mM imidazole and Roche cOmplete mini protease inhibitor). The N-terminal 6X-his-tagged protein was purified by immobilized metal affinity chromatography (IMAC) (HisPur Ni-NTA Resin, Thermo-Fisher Scientific) and stored at -70°C in a storage buffer (20 mM Tris, pH 7.0, 30 mM NaCl, 5% glycerol; with or without 0.2 mM dithiothreitol - DTT) after quantification at 280 nm (ϵ_280nm = 13,980 M^-1 cm^-1) (Nanodrop 2000c, Thermo-Fisher Scientific).

The purified recombinant NcADF (25 µM) was oxidized with NCT or H₂O₂ as previously described (Eitzinger et al., 2012; Luo et al., 2014). NcADF was incubated with several concentrations of NCT or H₂O₂ for 30 min at room temperature. Subsequently, the solution was dried for 10 min at CentriVap Vacum Concentrator (Labconco) and the remaining protein was suspended in sample buffer (10% glycerol, 80 mM Tris, pH 6.8, 2% SDS, 0.01% bromophenol blue) with or without reductants. The solutions were incubated at 96°C for 3 min and resolved by SDS-PAGE. For NCT inactivation with sodium thiosulfate, a step was added to the protocol. After incubation with 10 mM NCT, the NcADF solutions were incubated for 10 min with 30 mM sodium thiosulfate.

NcADF was oxidized with 50 mM H₂O₂ (as described in 2.3) and 25 µM of the protein was resolved by SDS-PAGE without reductants. The gel content was transferred to the PVDF membrane (Immobilon 0.45 μM, Millipore) and NcADF was detected with anti-NcADF serum (1:15,000) (Baroni et al., 2018) followed by the incubation with anti-mouse IgG secondary antibody (anti-mouse IgG – whole molecule – peroxidase, an antibody produced in rabbit) (1:10,000). The transferred membranes were incubated with a chemiluminescent horseradish peroxidase (HRP) substrate (SuperSignal West Pico Chemiluminescent Substrate, Thermo-Fisher Scientific) and exposed to radiographic films.

The oxidation of NcADF cysteines was detected by the 5,5′-dithiobis-(2-nitrobenzoic acid) (DTNB or Ellman`s reagent) after incubation with NCT or H₂O₂. The recombinant NcADF (25 µM) was previously reduced with 2 mM tris(2-carboxyethyl)phosphine (TCEP) for 1 h at room temperature followed by desalinization in Sephadex G-25 (packed in PD-10, Cytiva Lifesciences) to remove TCEP from the solution. NcADF was eluted in 20 mM Tris pH 7.0 in 500 µl aliquots, which were tested with Bradford reagent (Bio-Rad) for the presence of the protein. To confirm when TCEP emerges from the desalinization column, a control solution without protein but containing TCEP was desalted in parallel and the presence of the reductant in the aliquots was tested with 2 mM DTNB. The aliquots with NcADF and without TCEP were stored at -70°C until its use. The reduced NcADF was incubated with NCT or H₂O₂ for 30 min and incubated, in duplicates, with 2 mM DTNB for 15 min in a 96-well plate (Non-treated surface; Thermo-Fisher Scientific). The reaction product of DTNB with the sulfhydryl groups of the cysteines was detected by absorbance at 512 nm (Biotek Epoch, Agilent) after 15 min incubation under light protection.

The oxidation of NcADF cysteines was evaluated by alkylation using biotin-iodoacetamide, as described (Hill et al., 2009). NcADF was reduced by TCEP and then desalinized (see DTNB assay). Subsequently, the recombinant protein was oxidized with NCT or H₂O₂ for 30 min at room temperature. The excess of NCT and H₂O₂ were inactivated with 12 mM histidine/methionine and 2 U catalase, respectively, for 20 min. After inactivation, the protein solutions were incubated with 0.5 µM biotin-iodoacetamide (bio-IAM) for 20 min at room temperature. Bio-IAM was quenched by the addition of 30 mM β-mercaptoethanol (Gibco Thermo-Fisher Scientific). Afterward, the reactions were dried in the vacuum concentrator and the contents were suspended in a non-reductant sample buffer at a final concentration of 25 µM NcADF. The samples were resolved by SDS-PAGE and transferred to PVDF membranes, which were stained with DB-71 (0,008% w/v Direct Blue 71, 10% acetic acid, 40% ethanol). The alkylation of cysteines was detected by streptavidin-peroxidase (1:20,000; Thermo-Fisher Scientific) and observed after luminol incubation in radiographic films.

The rabbit actin from skeletal muscle was isolated from acetone powder as previously described by (Pardee and Spudich, 1982). The “acetone powder” was produced using the skeletal muscles from a rabbit acquired in a local butchery. Briefly, the powder was suspended in G buffer (2 mM Tris, pH 8.0, 0.5 mM ATP, 0.2 mM CaCl₂, 0.5 mM DTT) and actin was isolated after induction of polymerization with 50 mM KCl. Then, polymerized actin was separated from tropomyosin in 800 mM KCl. The filaments of actin were sedimented by ultracentrifugation at 100,000 g (TL-100 Ultracentrifuge; Beckman Coulter Inc.; rotor 70.1 TI) at 4°C for 1 h. The sediment containing actin filaments was suspended in G buffer and dialyzed against the same buffer at 4°C for 48 h for actin depolymerization. Afterward, the solution was again ultracentrifuged at 100,000 g for 1 h and the actin in the supernatant was concentrated, quantified (ϵ_290nm = 26.600 M^-l cm^-1), and stored at -70°C in small aliquots.

The previously reduced recombinant NcADF was oxidized with NCT or H₂O₂ (description in 2.3) in G buffer (without DTT) and the excess of the oxidants was inactivated with histidine/methionine solution or catalase, respectively. The reduced NcADF control was obtained after incubation of the protein with G buffer containing DTT. Afterward, the oxidized and reduced protein was serially diluted (20-0.312 µg/ml). NcADF was used in the microplate protein-binding assay, which was performed as described before, with modifications (Biesiadecki and Jin, 2011). Briefly, 100 µl of 2.5 µg/ml monomeric actin (G-actin in G buffer) was immobilized in a 96-well plate (NUNC MaxiSorp flat-bottom, Thermo-Fisher Scientific) at 4°C for 18 h. The wells were washed once with 200 µl G buffer-T (G buffer with 0.1% Tween-20) and blocked overnight at 4°C with 150 µl 3% BSA in G buffer-T. After the blocking, the wells were washed three times with G buffer-T followed by the addition of NcADF dilutions, in duplicate, and incubated at 4°C for 18 h. After three washes, the wells were incubated with anti-NcADF serum (1:1,000, in G buffer-T with 0.1% BSA) for 1 h at room temperature, followed by incubation with anti-mouse secondary antibody (1:10,000) for 45 min. The colorimetric signal was produced after the addition of TMB (OptEIA TMB Substrate Reagent Set, BD Biosciences) and quantified at 650 nm after 30 min incubation. The results were analyzed using the nonlinear fit log vs. response procedure in Prism Software (GraphPad). The statistical analysis was performed using One Way ANOVA.

For this assay, 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride (EDC) was used as a zero-length crosslinker. NcADF was oxidized with NCT or H₂O₂ in G buffer and the excess oxidants were inactivated as previously described in this study. The oxidized or reduced NcADF was incubated with equimolar G-actin for 18 h at 10°C and 350 rpm (Eppendorf ThermoMixer). After the reaction, 1 mM EDC was added to the protein mixture and incubated for 15 min at 37°C. The addition and incubation with EDC were repeated one more time. Subsequently, 9 mM glycine was added and the reactions were resolved by SDS-PAGE. The gels were stained with Coomassie G-250.

The polymerization assay was performed as described in Baroni et al. (2018), with modifications. In a 96-well black plate, 15% PI-actin (Pyrene-labeled actin from rabbit muscle, BioVision) was added (final concentration of 10 µM) to an equimolar concentration of NcADF previously oxidized with NCT. The mixture of proteins was incubated for 10 min at room temperature. Afterward, 10X ME (500 mM MgCl₂, 2 mM EGTA) was added to the proteins. After a 2-min incubation, the actin polymerization was induced with 10X KMEI (500 mM KCl, 10 mM MgCl₂, 10 mM EGTA, 100 mM imidazole, pH 7.0). The fluorescence (ex/em 365/407 nm) was detected in a microplate reader (Spectramax i3, Molecular Devices LLC).

The global alignment of NcADF (ToxoDB ID NCLIV_012510/GenBank ID XP_003881486) and human cofilin 1 (GeneBank ID 1072) amino acid sequences was performed using MegAlign 7.1.0 (Lasergene, DNASTAR) and visualized in ESPript (Robert and Gouet, 2014). The software Pymol 1.1r.1 (Molecular Graphics System, Schrödinger, LLC) was employed to align and visualize the tertiary structures of human cofilin 1 (PDB: 1q8x) and NcADF (Baroni et al., 2018).

To evaluate the susceptibility of the recombinant NcADF to oxidation, the protein was treated with N-chlorotaurine (NCT) or H₂O₂ and observed by SDS-PAGE. NcADF was expressed as an 18-kDa protein (Baroni et al., 2018) and the treatment with a large range of H₂O₂ concentrations induced the formation of a band with 36 kDa ( Figure 1A , lanes 5-8), indicating that, under oxidation, NcADF forms dimers. The treatment with 150 mM H₂O₂ produced a less intense dimer band compared to 50 mM of the oxidant ( Figure 1A , lanes 5 and 6), indicating that higher concentrations of H₂O₂ may be altering the protein activity and producing an excess of oxidation. The detection of this protein by the NcADF antiserum in both bands confirmed the dimerization ( Figure 1B ). To assess if the dimer formation was reversible, the oxidized NcADF was, subsequently, reduced with dithiothreitol (DTT) ( Figure 1A , lanes 1-4). In this condition, the intensity of the 36-kDa bands was reduced when compared to the counterparts without DTT, confirming the reversibility of the dimerization.

NCT was previously shown to oxidize human cofilin1 (Klamt et al., 2009). In order to evaluate the NcADF dimerization by NCT, several concentrations of this substance were tested in recombinant NcADF produced without DTT in the storage buffer. Under non-reduced conditions, a discrete dimer band was formed ( Figure 1C , lane 1). The use of lower concentrations of NCT did not alter the formation of dimers, but concentrations above 2 mM induced more intense bands of 36 kDa ( Figure 1C , lanes 9-11). To assess whether the formation of dimers by NCT was also reversible, as observed with H₂O₂, the oxidized NcADF was reduced by tris(2-carboxyethyl)phosphine (TCEP), which disrupts the disulfide bonds, after the oxidation with NCT. In parallel, the excess of NCT in the solution was inactivated by sodium thiosulfate (ST) (Böttcher et al., 2017). With NCT inactivation, the presence of a higher concentration of TCEP significantly reduced the intensity of the dimer and oligomers bands ( Figure 1D , lane 7) compared to the oxidation control ( Figure 1D , lane 3). Moreover, the decrease in band intensity was less intense in ST-free samples ( Figure 1D , lanes 11-14) compared to the ST supplemented ones ( Figure 1D , lanes 4-7).

DNTB was used to investigate whether NCT and H₂O₂ treatments would oxidize cysteine residues of NcADF. This substance reacts with cysteine’s free sulfhydryl groups to yield the yellow compound 2-nitro-5-thiobenzoic acid (TNB), detected by colorimetry. The oxidized cysteine does not react with DTNB, impairing or reducing the absorbance. Nanomolar concentrations of NCT or H₂O₂ had no significant effect on NcADF thiol oxidation, whereas micromolar concentration of NCT, but not of H₂O₂, was sufficient to oxidize the protein (p ≤ 0.05) ( Figure 2A ). When the protein was treated with 10 mM NCT or H₂O₂, the absorbance was significatively (p ≤ 0.0001) reduced in both treatments compared to the reduced protein ( Figure 2A ). Nevertheless, at the same concentration, the treatment with NCT had a more pronounced effect (p ≤ 0.0001) compared to the treatment with H₂O₂ ( Figure 2A ). To confirm these observations, the levels of cysteine oxidation were also monitored by alkylation with bio-IAM. The decrease of the NcADF band intensity indicates thiol groups´ oxidation of the cysteine residues. The treatment with 2 and 10 mM H₂O₂ reduced the intensity of the bands ( Figure 2B ). Likewise, the NCT treatment completely abrogated the signal with concentrations above 0.08 mM ( Figure 2C ). The results using alkylation confirmed the ones obtained with DTNB. NCT and H₂O₂ oxidized NcADF cysteine residues, but NCT demonstrated a higher oxidation potential.

The ability of NCT- and H₂O₂-treated NcADF to bind to rabbit G-actin was determined using the crosslinker EDC. The binding of NcADF (18 kDa) and actin (42 kDa) produced a band containing the sum of the molecular weight of both proteins (60 kDa). Moreover, binding was proportional to the intensity of this band. The treatment of NcADF with up to 2 mM H₂O₂ did not impair the formation of the 60-kDa band ( Figure 3A ). However, the same concentration of NCT reduced the band intensity in a concentration-dependent manner ( Figure 3B ). In this experiment, the dimer formation was not as clear as in Figure 1 , since the mass of protein applied was lower. The binding of these proteins was also investigated in non-equilibrium conditions. NCT presented higher EC50 (half-maximum effective concentration) compared to H₂O₂ (192.4 nM and 29.53 nM, respectively). In comparison, the EC50 for reduced NcADF was 27.65 nM ( Figure 3C ). These data suggest that the ability of NcADF to bind to G-actin decreases under the oxidation of that protein. To evaluate the role of oxidized NcADF in actin dynamics, the recombinant protein was oxidized or not and applied in the polymerization assay. The reduced NcADF decreased the actin fluorescence, while the oxidized NcADF accelerated the initial rate of polymerization ( Figure 3D ). This alteration in actin polymerization by reduced or oxidized NcADF suggests that the oxidized protein did not bind to actin. Furthermore, the effect of the protein in actin polymerization was produced by the remaining reduced active NcADF.

The oxidation of human cofilin 1 is a known mechanism of protein activity regulation (Samstag et al., 2013; Bamburg et al., 2021). However, considering that NcADF and mammalian cofilin share less than 30% identity (Baroni et al., 2018) we investigated whether the position of cysteine residues in the structures of the two proteins is conserved. The alignment of the primary sequences revealed that only the Cys 58 (Cys 80 in cofilin 1) is conserved ( Figure 4A ). To assess whether the position of cysteine residues is maintained between the tertiary structures, the structure of both proteins was aligned and confirmed that Cys 58 (NcADF) and Cys 80 (cofilin 1) occupy correspondent positions in the molecules ( Figure 4B ). The additional cysteines occupy distinct portions of the molecules, with emphasis on cofilin Cys 139 and 147, absent in NcADF, which were identified as responsible, when oxidized, to reduce the binding of the protein with actin (Cameron et al., 2015).