cDNA Cloning, Heterologous Expression, Cytotoxicity, and Inhibitory Effects of a Disintegrin from Bothrops ammodytoides Venom

An mRNA transcript that codes for a Disintegrin and a Metalloprotease from a venom gland of the viper Bothrops ammodytoides was isolated. Vector pCR®2.1-TOPO was used to clone the Disintegrin transcript and then subcloned in the pET-28a vector to express the protein in the E. coli strain BL21. The recombinant disintegrin, HisrDisintegrin, comprises 38 residues at the N-terminal and 75 residues of the Disintegrin, including 14 cysteines or 7 disulfide bonds. HisrDisintegrin was obtained soluble in the intracellular fraction and separated by affinity chromatography. The experimental molecular mass of HisrDisintegrin, 11,750.8 Da, agreed with its theoretical value, including the cleavage of a Met residue caused by a bacterial post-translational modification. HisrDisintegrin was folded in in vitro conditions and then purified, finding a main fraction with a comparable molecular mass. The HisrDisintegrin contains an alpha/beta structure, as observed by circular dichroism. Its biological activity demonstrated the growth reduction of the human endothelial (HMEC-1) cells and the human mammary gland adenocarcinoma (MDA-MB-231). Furthermore, HisrDisintegrin inhibits the adhesion of HMEC-1 and MDA-MB-231 cells to the proteins of the extracellular matrix (laminin (LN), fibronectin (FN), and vitronectin (VN)).


Introduction
Bothrops ammodytoides is the southernmost situated viper in the world that populates from the cold Patagonia of Argentina to the warm desert regions of "El Gran Chaco" and the dry "Pampeana" region. It is a small viper that could cause the typical complications of Bothrops envenomations, (HMEC-1) and human mammary gland adenocarcinoma (MDA-MB-231) cell lines. Hence, we provide the proof of concept that this recombinant disintegrin could be used as a protein lead for anticancer drug development.

Venom Gland
Adult specimens of B. ammodytoides were maintained in good health in plastic cages (27 ºC and 12 h light-dark cycle). Venom was collected by manual extraction, dried under vacuum, and kept at -20ºC. An anesthetized (ketamine-xylazine) B. ammodytoides was subjected to a surgical extraction of one of the two venom glands. RNAlater→ (Thermofisher, Asheville, NC, USA) was used to treat the extracted venom gland and stored at -20ºC. The treated animal stayed healthy and recovered.

Plasmids, Bacterial Strains, and Enzymes
Plasmid pCR®2.1-TOPO® (Invitrogen, CA, USA) was used for cloning the Disintegrin gene and pET28a (Invitrogen, CA, USA) for production of the 6His-tagged recombinant HisrDisintegrin. DNA cloning and plasmid propagation were conducted in Escherichia coli strain XL1-Blue. The expression of the recombinant Disintegrin was performed in E. coli strain BL21. Taq polymerase, T4 DNA ligase, restriction enzymes, and Factor Xa protease (FXa) were provided by New England Biolabs (NEB, MA, USA).
The vector pCR 2.1-TOPO® (TOPO TA Cloning® Kits, Invitrogen; Carlsbad, CA, U.S.A.) was used to clone the purified DNA fragment. E. coli XL1Blue cells were transformed with the new plasmid, grown in Petri dishes with Luria Broth (LB) agar added with ampicillin (100 µg/ mL), and incubated at 37 ºC, 14 h. Colonies were evaluated by colony PCR with oligonucleotides M13 forward (5'-GTAAAACGACGGCCAGT-3') and M13 reverse (5'-CAGGAAACAGCTATGAC-3'). Plasmids from positive colonies with the expected DNA size were purified and sequenced at the Institute of Biotechnology, UNAM.

Expression Plasmid Construction
A TOPO recombinant plasmid with the confirmed sequence was used to amplify the Disintegrin peptide by PCR with designed oligonucleotides, containing the BamHI site and FXa cleavage site at the 5' end and the HindIII site and stop codons at the 3' end (Dis-pET28-F: 5'-GAGAGGATCCATCGAGGGAAGGGAGGTG-3'; Dis-pET28-Rev: 5'-CAGATGTCTTCTAGTAACTG-CAGCCAAGCTTTCTC-3'). The amplified gene and the pET28a expression plasmid (Invitrogen, Carlsbad, CA, U.S.A.) were digested with BamHI and HindIII enzymes. After digestion, the Disintegrin gene and vector were run and purified from the agarose gel. The digested DNAs were mixed in a ratio of 1:5, and then T4 DNA Ligase Buffer (1X) and T4 ligase (1 Weiss U) (Thermo-Fisher Co.; San José, CA, U.S.A.) were added. The ligation reaction was kept at 16 ºC 16 h. Chemically competent E. coli XL1-Blue cells were transformed with the ligation product (pET28a-Disintegrin). Transformed cells were grown on LB agar and added with 50 µg/mL of kanamycin (Sigma, St. Louis, MO, U.S.A.). After incubation (37 ºC, 14 h), some colonies were evaluated by colony PCR using the oligonucleotides 1522-Fwd (5'-AGATCTCGATCCCGC-3') and 1522-Rev (5'-GACCCGTTTAGAGGC-3'). Positive colonies were seeded in 3 mL of LB, incubated overnight at 37ºC and 180 rpm, and then subjected to plasmid purification and sequencing at the Institute of Biotechnology, UNAM, Mexico.

Expression and Purification of HisrDisintegrin
E. coli strain BL21 harboring the pET28a-HisrDisintegrin expression vector was grown at 37ºC in LB medium (50 µg/mL kanamycin). When the culture absorbance (600 nm) was close to 0.8 units, 1 mM IPTG (isopropylβ-D-thiogalactopyranoside) was added as the inductor and the culture was incubated for 18 h at 16ºC, 200 rpm. Afterward, the culture was centrifuged at 8,000 g for 20 min (JA-10 rotor, Beckman model J2-21). Cells were resuspended in wash buffer (0.05 M Tris-HCl, pH 8.0) and disrupted at 35 kpsi (One Shot cell disruptor, Constant Systems®). The mixture was centrifuged at 12,000 rpm for 15 min (JA-20 rotor, Beckman Coulter Avanti J-30I®). Soluble and insoluble fractions were analyzed by SDS-PAGE.
The recombinant HisrDisintegrin, found in the soluble fraction, was purified by Ni-nitrilotriacetic acid (Ni-NTA) affinity chromatography (Qiagen, CA, USA) using wash buffer and elution buffer B (0.05 M Tris-HCl buffer, 300 mM imidazole, pH 8.0). The elution buffer was removed by reverse-phase HPLC (RP-HPLC) using a C18 analytical reverse-phase column (Supelco 4.6 × 250 mm, 5 μm). The HPLC ran a separation program from 0 to 60% solvent B (0.1% TFA in acetonitrile) for 60 min at 1 mL/min. The fractions were detected at 230 nm. Collected fractions were dried under vacuum. Molecular masses were determined in a Thermo Scientific LCQ Fleet ion trap mass spectrometer (San Jose, CA) using a Surveyor MS syringe pump delivery system and the data acquisition and deconvolution were performed on the Xcalibur Windows NT PC data system.

Circular Dichroism
The secondary structure of the recombinant HisrDisintegrin was analyzed by Circular dichroism (CD) spectroscopy. The spectrum was obtained at room temperature in 1 mm path quartz cells with a wavelength range of 185 to 260 nm (Jasco J-1500 spectrometer, Japan). Data were gathered at a 50 nm/min rate with 0.1 nm intervals. A 0.25 mg/mL protein solution was prepared in 60% trifluoroethanol (TFE) because it enhanced the secondary structure. The average CD corresponded to three CD measures. The secondary structure percentages were then analyzed using online algorithms provided by BeStSel (Beta Structure Selection, webserver http://bestsel.elte.hu/index.php (accessed on 30 January 2023) (Micsonai et al. 2018).

Cell Lines and Culture Conditions
The American Type Culture Collection (ATCC, Manassas, VA, USA) sold the cell lines HMEC-1 (human endothelial) and MDA-MB-231 (human mammary gland adenocarcinoma). HMEC-1 cells were grown in MCDB-131 medium with 10% Fetal Bovine Serum (FBS), 10 ng/mL epidermal growth factor (Sigma-Aldrich), and 1 µg/mL hydrocortisone, then incubated at 37 °C in a humid chamber containing 5% CO 2 . MDA-MB-231 cells were grown in L-15 Leibovitz medium with 10% fetal bovine serum (FBS) and placed in a CO2-free incubator at 37 °C with humid air.

Cytotoxicity Assay
MDA-MB-231 and HMEC-1 cells were subjected to different concentrations of HisrDisintegrin, and its cytotoxicity was assessed by the MTT assay (3-(4,5-dimethylthiazol-2-yl)-2,5-diphe-nyltetrazolium bromide, Sigma-Aldrich) (Mosmann 1983). Briefly, 100 µL of trypsinized cells (2.0 × 10 5 cells/mL) in complete media were seeded in each well of a 96-well plate. The cells were treated with HisrDisintegrin at different concentrations (from 0.06 to 16 µM or PBS pH 7.4 as a negative control). The plate was incubated at 37 °C for 24 h. Afterward, 100 µL of MTT reagent (0.5 mg/mL) prepared in each cell culture medium were added to each well with cells and kept at 37 °C for 3 h. 200 µL of isopropyl alcohol and 0.4 N HCl were added to each well after the MTT remotion. The plate was kept in the dark for 30 min at 25 °C. Absorbance was measured at 570 nm (HTS 7000 Bioassay Reader, Perkin Elmer, Waltham, MA, USA). Viable cells, as a percentage, were determined by relating the number of viable cells of the treated cells to the control group (untreated cells) (Zakraoui et al. 2017). The experiment was repeated three times in triplicate at each HisrDisintegrin concentration.

Cell Adhesion Inhibition Assay
The binding inhibition of HMEC-1 and MDA-MB-231 cells to laminin, vitronectin, and fibronectin, caused by HisrDisintegrin was assessed using a previously established protocol (Hammouda et al. 2016;Lima-Dos-Santos et al. 2015;Wierzbicka-Patynowski et al. 1999). 100 µL (1 µg/mL) of either fibronectin, vitronectin, or laminin were added to each well of a 96-well plate, in triplicate, and left at 4 °C overnight. The wells were then blocked with 200 µL of a culture medium supplemented with 2% bovine serum albumin (BSA) at 37 °C for 1.5 h. The cells were harvested, counted, and resuspended in a medium containing 1% BSA at 5 × 10 5 cells/mL. Various concentrations of HisrDisintegrin (0.03, 0.06, 0.125, 0.25, 0.5, and 1 µM) were added to the cell suspension and incubated at 37 °C for 2 h; PBS was used as the negative control. After removing the blocking solution, 100 µL of the cell/HisrDisintegrin mixture and the negative control were added to the fibronectin-, vitronectin-, and laminin-coated wells and incubated at 37 °C for 2 h. The wells were washed three times, and 200 µL of L15 or MCDB-131 medium containing 1.5 mg/mL MTT and 1% BSA were added to the wells and incubated at 37 °C for 3 h. Cell lysis was achieved by adding 100 µL of dimethyl sulfoxide, and gently mixed. Absorbance was read at 570 nm. [(absorbance of negative control -absorbance of cell/HisrDisintegrin sample)/absorbance of negative control] × 100 was used to determine the percentage of inhibition. The experiment was repeated three times, and each HisrDisintegrin concentration for both cell lines was evaluated in triplicate. methods the software Prism 4.0 (Graph Pad Inc., San Diego, CA) was used.

Isolation and Sequence Determination of HisrDisintegrin
The cDNA for HisrDisintegrin with the expected size (225 bp) was amplified by PCR, cloned in pCR®2.1-TOPO® plasmid, and then in the expression vector pET-28a. The amino acid sequence of HisrDisintegrin is 100% identical to the disintegrin fractions of the ADAM (a disintegrin and metalloproteinase) proteins named bothropasin and catrocollastatin isolated from the venoms of B. jajaraca and Crotalus atrox, respectively ( Table 1).
The disintegrin from B. ammodytoides seems to belong to the group 1 of the a disintegrin and metalloproteinase (ADAM) family of proteins including the snake venom metalloproteinases (ADAM/SVPM). They are also classified according to the number of residues and disulfide bridges, in a large group that consist of more than 70 residues cross-linked by seven disulfide bridges (Vasconcelos et al. 2021). The disintegrin from B. ammodytoides is distinguished from other snake disintegrin structures such as rhodostomin and bitistatin from groups 3 and 2, respectively, because of their amino acid identity and their content in disulfide connections (Vasconcelos et al. 2021).

Wound Healing Assay
MDA-MB-231 or HMEC-1 cells were plated (2 × 10 5 cells/ mL) in a 96-well microtiter plate and incubated at 37 °C overnight. The confluent monolayer was scratched with a sterile 200 µL pipet tip at the midline of each well (Galan et al. 2008;Ren et al. 2006). A specific culture medium was used to wash the detached cells. The remaining cells were treated with 200 µL of fresh culture medium containing 10% FBS. 100 µL of PBS 0.01 M, pH 7.4, or HisrDisintegrin 0.04 µM prepared in the same PBS were added to the control and treatment groups to stimulate cell migration. The cells were then incubated at 37 °C in a humidified chamber, and images were captured at 0, 12, and 24 h using a microscope. The percentage of cell migration was calculated using the equation: [(C -E)/C] X 100, where C is the distance (µm) between the cell edges at t = 0 h of the control, and E is the distance between the cell edges of the monolayer at the final time (t = 24 h) of the incubation. The images were analyzed using Image J software, and the data were plotted as the percentage of cell migration of the control group vs. the treatment with HisrDisintegrin. The experiment was repeated three times, and the assays were performed in triplicate.

Statistics
Results were expressed as mean and standard deviation, or as mean with 95% confidence intervals. For all the statistical Table 1 Comparation of Disintegrins from various snake 1 Disintegrin from Bothrops ammodytoides; Bothropasin from Bothrops jararaca (PBD: 3DSL); Catrocollastatin from Crotalus atrox (PBD: 2DW0); Rhodostomin (PBD: 2LJV) from Calloselasma rhodostoma; Bitistatin (PBD: 2MP5) from Bitis arietans. 2 The percent identity (ID) is shown based on the disintegrin from this work. The sequence alignment was obtained from https://www.ebi.ac.uk/Tools/msa/clustalo/. Gaps (hyphens) have been introduced to enhance similarities. Conserved and related amino acids in each group of similar peptides are marked by asterisk, colons, and dots, respectively. (Clement et al. 2012b(Clement et al. , 2015(Clement et al. , 2016Estrada et al. 2007). HisrDisintegrin contains 14 cysteines and could form up to 135,135 isoforms considering scrambling disulfide pairing possibilities. Clearly, in the snake venom gland, the molecular machinery in charge of the disintegrin expression directs the formation of a correct folding.

Secondary Structure of HisrDisintegrin
Here the CD spectrum of HisrDisintegrin was obtained in solution using 60% TFE, which promotes hydrogen bond formation mimicking in-cell solvation for protein folding (Fig. 3). According to the CD deconvolution program BeStSel (Micsonai et al. 2018), the secondary structure content (%) of this recombinant disintegrin was α-helix 18.6, β-antiparallel 16.3, β-parallel 0, β-turns 14.2 and random structures 51. So far, this recombinant disintegrin contain a significative proportion of α-helix and antiparallel structures, here characterized for having negative ellipticities at 203-229 nm giving a characteristics CD spectrum of an α/β protein. The root-mean-square deviation (RMSD) of the HisrDisintegrin CD experimental data compared to the fitted data predicted by the model was 0.22.
Although a reliable estimation of secondary structure content from the CD spectra cannot be calculated for proteins with the mixed α and β elements such secondary structure estimation, it could be used as a relative measure of structural changes of an unfolded protein (Khrapunov 2009). So that, the CD of a protein could be a reliable mean for determination of secondary structure of proteins that has been obtained using recombinant techniques (Greenfield

Expression, Purification, and Protein Folding of HisrDisintegrin
The gene encoding HisrDisintegrin was assembled and subcloned into the expression vector pET-28a (Fig. S1), which produces N-terminal 6His-tagged proteins, for a straightforward purification of the recombinant products by affinity column chromatography Ni-NTA (Ni-nitrilotriacetic acid). A cleavage site for FXa was added between the 6His-tag and the HisrDisintegrin, to separate the 6His-tag from the HisrDisintegrin if necessary (Fig. S1). HisrDisintegrin was found in soluble fraction after expression in E. coli BL21 strain, and it was recovered by Ni-NTA (Fig. 1).
The heterologous expression of HisrDisintegrin in the soluble fraction and its purification process was corroborated by a western-blot assay (anti-6His-tag antibody coupled to alkaline phosphatase). HisrDisintegrin was purified by RP-HPLC, and a fraction with retention time around 35 min or 35% acetonitrile was collected (Fig. 2). The fraction also showed a similar apparent molecular mass SDS-PAGE (15/4%). The protein yield of the HisrDisintegrin fraction was 0.5 mg/L. Furthermore, the experimental molecular mass of the fraction obtained by mass spectrometry was 11,750.8 Da (oxidized form) (Fig. S2) and agreed to the theoretical molecular mass of HisrDisintegrin after the cleavage of a Met residue caused by a bacterial post-translational modification. Therefore, the fraction obtained from RP-HPLC corresponded to the oxidized cystine isoform of the HisrDisintegrin. Multiple cysteine-rich protein isoforms generated during heterologous expression have been reported

Cytotoxicity
The MTT assay was used to assess the effect of HisrDisintegrin on HMEC-1 and MDA-MB-231 cells. Cells treated with different concentrations for 24 h of HisrDisintegrin (from 0.0625 to 16 µM) decreased their cell viability in a dose-dependent manner. HisrDisintegrin at 16, 8, and 4 µM significantly decreased cell viability by 70.8, 67.3, and 54.9% of MDA-MB-231 and 52.5, 38.6, and 36.4% of HMEC-1 cells, respectively (p < 0.001). HisrDisintegrin resulted in 80% cell viability at 1 µM and lower concentrations (Fig. 4). Consequently, the highest selected concentration to evaluate the inhibitory activities of HisrDisintegrin was 1 µM.
Commonly, isolated Disintegrins are known as components without toxicity (Walsh and Marcinkiewicz 2011). However, at cellular level, HisrDisintegrin showed higher cytotoxicity compared to other isolated snake venom disintegrins such as totonacin (Rivas Mercado et al. 2020), tzabcanin (Saviola et al. 2016), vicrostatin (Minea et al. 2012). Non cytotoxic HisrDisintegrin concentrations selected for biological activities were similar to those used for other Bothrops disintegrins, such as Neuwiedin from B. neuwiedi (Lima-Dos-Santos et al. 2015), Colombistatin from 2006). The CD analysis of HisrDisintegrin showed elements of α/β protein, and we can consider that it is folded, utmost if it contains disulfide connections that tightly restrict internal changes. Furthermore, to compare the secondary structure of HisrDisintegrin, the CD spectrum of a classical α/β protein, the scorpion neurotoxin AaH2 (Fig. 3). Similarly, according to the BeStSel deconvolution program (Micsonai et al. 2018), the secondary structure content (%) of AaH2 was α-helix 86.8, β-antiparallel 8.5, β-parallel 0, β-turns 4.7, and random structures 0. AaH2 has 66 amino acids and four disulfide bridges, compared to the 75 amino acids and the seven disulfide connections of Disintegrin. Although AaH2 and Disintegrins are different in their mode of action, they share common structural patterns, such as a large content of α-helix 86.8 and β-antiparallel 8.5; however, in the case of the recombinant Disintegrin, it is less structured for the large content of random structures, which in most analyzed Disintegrins have shown disordered structures; some examples are the 3D structures of rhodostomin (2PJF) and bothropasin (3DSL), obtained by NMR and X-ray diffraction methods, respectively. The CD data of proteins is challenging to compare with X-ray crystallography or NMR (Greenfield 2006).  Results are shown as the mean ± SD. **p < 0.01 and *** p < 0.001. Each concentration was evaluated in triplicate and the experiment was repeated three times Fig. 3 Circular dichroism of HisrDisintegrin and a scorpion neurotoxin. The CD analysis showed higher α-helix content than β-strands demonstrated a significant inhibitory effect of HisrDisintegrin on the adhesion of MDA-MB-231 and HMEC-1 cells to VN and FN. A higher HMEC-1 cells inhibitory effect compared to other reported snake venom disintegrins such as tzabcanin, which inhibits about 70% A-375 cell adhesion to VN (Saviola et al. 2016), totonacin inhibits 60 and 44% MDA-MB-231 and HMEC-1 cell adhesion to VN (Rivas Mercado et al. 2020), more differences are noted when it is compared to other Bothrops disintegrins as DisBa-01 known by its high affinity on FN binding receptor α V β 3 integrin (Kauskot et al. 2008), or Colombistatin that has a higher inhibitory effect on FN and collagen I (Montenegro et al. 2017;Sanchez et al. 2009). Similarly, Bothrasperin inhibits endothelial t-End and melanoma B16 cells to VN, FN, and collagen 70 to 80% (Pinto et al. 2003), with lower inhibitory activity if it is compared to HisrDisintegrin inhibitory effect on VN and FN on HMEC-1 cells (Fig. 5). As VN and FN were more inhibited ligands in both HMEC-1 and MDA-MB-231 cells, and considering VN targets (integrins α V β 3 , α 8 β 1 , α V β 5 , α V β 1 , and α IIb β 3 ) (Barczyk et al. 2010), and FN targets (integrins: α IIb β 3 , α 4 β 1 , α 4 β 7 , α 5 β 1 , α 8 β 1 , α v β 1 , α v β 3 , α v β 6 , and α v β 8 ) (Saviola et al. 2016), the results suggest HisrDisintegrin antagonized one or more of the five mentioned integrin VN receptors, and one or more of the nine integrins FN receptors. Finally, LN stands as the ECM less affected protein as HisrDisintegrin inhibitory effect on MDA-MB-231 and HMEC-1 cells was just about 30% at the highest disintegrin concentration proved, 1 µM. LN targets seven integrin receptors: α1β1, α2β1, α3β1, α6β1, α7β1, α10β1, and α6β4 (Saviola et al. 2016).

Acknowledgements
The mass spectrometry determination conducted by Dr. Fernando Zamudio is greatly acknowledged. Also, we acknowledge Dr. Adolfo de Roodt, BSc., DVM Vanessa Costa de Oliveira, and DVM Pablo Regner for their help during the extraction of the venom gland and B. ammodytoides venom supply. We additionally acknowledge Dr. Paul Gaytán, M.C. Eugenio López-Bustos and Q.I. Santiago Becerra from Unidad de Síntesis y Secuenciación de ADN at Instituto de Biotecnología. This research was funded by DGAPA-UNAM (grant number IT200321) and by CONACyT (grant number PRONAII 303045).
Data Availability Data will be made available on request.

Declarations
Compliance with Ethical Standards Animal experimental procedures were in accordance with international recommendations and the guidelines of the Good Experimental Practices, under the supervision of the Ethical and Animal Welfare Committee of the Institute of Biotechnology (Ethical approval IT/IBt/Project #411). In brief, proper animal handling, to minimize distress and discomfort, was always conducted towards maximizing the animal welfare during experimentation ac-

Evaluation of the Effect of HisrDisintegrin from B. ammodytoides on MDA-MB-231 and HMEC-1 Cell Migration
An in vitro wound healing assay assessed cell migration, where the cells were scraped from the wells' bottom. Post a 24-hour incubation period, HisrDisintegrin at a concentration of 0.04 µM effectively restrained MDA-MB-231 and HMEC-1 cell migration by 87.8 ± 4% and 76.5 ± 2.5%, respectively (Fig. 6).
The observation that HisrDisintegrin inhibits cell migration proposes that αvβ3 integrin is blocked, although it doesn't exclude the possibility that other integrins are targeted. For example, αvβ5 is prominently expressed in MDAMB-231 cells and is also known to facilitate cancer cell migration (Dang et al. 2006;Pecheur et al. 2002;Rolli et al. 2003;Sloan et al. 2006;Taherian et al. 2011). Similar results on cell migration, especially regarding to high 80-90% inhibition level have been reported for few other viperid snake venom disintegrins, such as crotatroxin 2, vicrostatin r-viridistatin 2, and r-mojastin 1 (Galan et al. 2008;Lucena et al. 2011;Minea et al. 2012).

Concluding Remarks
This work is the first approach to demonstrate the use of the recombinant HisrDisintegrin from B. ammodytoides with the capacity to inhibit cell migration and cell adhesion to ECM proteins. The use of recombinant proteins is becoming more common nowadays in the search for novel leads with