Hydrogen peroxide from l-amino acid oxidase of king cobra (Ophiophagus hannah) venom attenuates Pseudomonas biofilms

Because of the high incidence of Pseudomonas aeruginosa biofilms-related nosocomial infections, venoms from common Thai snakes were tested. Although venoms from king cobra (Ophiophagus hannah; OH) and green pit viper (Trimeresurus albolabris) showed the broadest antibacterial spectrum, OH venom demonstrated more profound anti-biofilm activities against P. aeruginosa. Additionally, purified l-amino acid oxidase from OH venom (OH-LAAO), using a three-step chromatography and protein identification, reduced biofilm mass as indicated by the downregulation of several genes, including the genes for biofilm synthesis (algD and pslB) and biofilm regulators (algU, gacA, and siaD). Moreover, OH-LAAO disrupted Pseudomonas-preformed biofilms via upregulation of several genes for biofilm dispersion (nbdA, bdlA, and dipA) and biofilm degradation (endA and pslG), resulting in a reduction of the biofilm biomass. Due to the antimicrobial effects and anti-biofilm activities (reduced production plus increased dispersion) neutralized by catalase, a hydrogen peroxide (H2O2)-degrading enzyme, the enhanced H2O2 by OH venom might be one of the anti-biofilm mechanisms. Hence, OH-LAAO was proposed as a novel agent against Pseudomonas biofilms for either treatment or prevention. More studies are interesting.

. Antimicrobial, anti-biofilm, and biofilm eradication activities of crude snake venoms, including Naja kaouthia (NK), Ophiophagus hannah (OH), Dabonia russellii (DR), Trimeresurus albolabris (TA) venoms. The minimum inhibitory concentrations (MICs) and the minimum bactericidal concentrations (MBCs) were determined for antimicrobials. The minimum biofilm inhibitory concentrations (MBICs) was performed for anti-biofilm activity. The minimum biofilm eradication concentrations (MBECs) was determined for biofilm eradication activity. 'ND: Not determined (no biofilm production from S. aureus SA1, SA2, and SA3); all of the strains are clinical isolates, except for the ATCC strains and PAO1 (comercially available strains); PACL is a clinical-isolated P. aeruginosa that can produce more biofilms after stimulated by chlorhexidine (an-antiseptic) referred to as "chlorhexidine-treated P. aeruginosa (C_PACL)". www.nature.com/scientificreports/ TA venoms had bactericidal activities against P. aeruginosa (MBC range 0.06 to 0.25 mg/mL). Additionally, all venoms exhibited bactericidal effects on S. aureus (MBC range 0.0038 to 0.25 mg/mL). Therefore, most snake venoms were effective against S. aureus. The venoms from OH and TA were categorized as broad-spectrum antimicrobial agents against both Gram-positive (S. aureus) and Gram-negative bacteria (P. aeruginosa).
Crude snake venoms exhibited anti-biofilm activity and partially destroyed the established biofilms. Due to the effect on both Gram-positive and Gram-negative bacteria, OH and TA venoms were further tested for anti-biofilm activities (inhibition of biofilm formation) as determined by the minimum biofilm inhibitory concentrations (MBICs). As such, OH and TA venoms at the MIC (1 × MIC) or twofold higher (2 × MIC), identical to the MBIC range between 0.06 to 0.25 mg/mL, were adequate to prevent Pseudomonas biofilm (incubated venoms together with P. aeruginosa) (Table 1). Moreover, OH and TA venoms inhibited the biofilms of S. aureus ATCC 29213 (MBIC = 0.0038 mg/mL). Notably, the venoms without the antimicrobial effect did not exhibit anti-biofilm formation, especially in E. coli ATCC 25922 and three E. faecalis isolates (Table 1). Therefore, the biofilm prevention effect of OH and TA venoms against P. aeruginosa was mainly due to antimicrobial activities. Subsequently, to test the ability of snake venoms to eliminate biofilms (administration of venoms onto the preformed biofilms) as indicated by the minimum biofilm eradication concentrations (MBECs) were determined using bacterial-preformed biofilms in 96-well culture plates. Although all venoms had no biofilm eradication effect at the MBECs higher than 0. 25  l-amino acid oxidase from O. hannah venom (OH-LAAO) as a potential active substance for antimicrobial and anti-biofilm activities against P. aeruginosa. The crude OH venom was purified by three-step chromatography (see "Methods" Section). The first step was the analytical gel filtration using the Sephadex G-75 column chromatography (Fig. 3A), exploring the effective fractions. Among 5-peaks of the eluted proteins, peak-1 elution demonstrated the most prominent bactericidal activity against P. aeruginosa PAO1 and PACL (MIC and MBC at 10 µg/mL) ( Table 2) that also showed the most prominent anti-biofilm effect on 24 h biofilms (biofilm prevention experiments) ( Fig. 3B-E) and 24 h-established biofilms (biofilm eradication experiments) ( Fig. 3F-I), implying biofilm prevention and biofilm detachment properties. These results suggested that antimicrobial, anti-biofilms, and biofilm eradication effects of OH venom were in the peak-1 elution from the first step of purification. Additionally, the specific LAAO activity from these 5 peaks was remarkedly observed only in peak-1 (11.17 ± 1.31 U/mg) ( Supplementary Fig. S5). In the second step, the peak-1 elution from the Sephadex G-75 column chromatography was separated and purified for LAAO enzyme using the Resource Q column chromatography, which showed 3 peaks of the elution ( Supplementary Fig. S6). The peak-1 eluted from the Resource Q column demonstrated a remarkable LAAO activity at 15.99 ± 1.18 U/mg ( Supplementary Fig. S7). The last step of purification used the HiTrap™ Heparin column chromatography (see "Methods" Section), which showed two peaks of the elution ( Supplementary Fig. S8). The highest LAAO activity was observed in the peak-1 elution at 21.72 ± 1.39 U/mg ( Supplementary Fig. S9). In summary, using the three-step purification provided a significantly higher specific activity of LAAO (Supplementary Fig. S10 and Table S2) for further experiments.
Then, an approximately 65 kDa protein dominant in the peak-1 elution from the last step purification using the HiTrap™ Heparin column chromatography (Fig. 3J) was identified by N-terminal amino acid sequence using nano-High Performance Liquid Chromatography (nano-HPLC) mass spectrometry and database search (see "Methods" Section). The purified OH-LAAO was matched with LAAO submitted in the UniProt database (Uni-Prot database: P81383) by 52% of LAAO amino acid sequences (Fig. 3K). The purified OH-LAAO (approximately 65 kDa) had a higher molecular weight of LAAO in the database (P81383) (55.977 kDa), indicating the posttranslation modification (glycosylation) 21 www.nature.com/scientificreports/  www.nature.com/scientificreports/ HiTrap™ Heparin column was further used as OH-derived LAAO (OH-LAAO). After purification, OH-LAAO showed increased antimicrobial activities against P. aeruginosa PAO1 and PACL with 5 and 10 µg/mL of MICs and MBCs, respectively (Table 2), and 10 µg/mL of purified OH-LAAO exhibited a prominent anti-biofilm effect ( Table 2). Subsequently, the impacts of the purified OH-LAAO against P. aeruginosa biofilms were investigated. As such, 2.5 µg/mL of purified OH-LAAO (0.5 × MIC) downregulated algD in PAO1 and PACL (planktonic and sessile forms), while pslB was downregulated only in sessile but not planktonic cells, and algU was downregulated in sessile forms ( , and hydrogen peroxide (H 2 O 2 ), from the substrate (l-amino acids) (Fig. 6A). With the incubation of OH-LAAO with either l-lysine or l-argi- Table 2. Antimicrobial, anti-biofilm, and biofilm eradication activities of Ophiophagus hannah (OH) elution peak-1 to peak-5 from the first step purification using the Sephadex G-75 column chromatography and purified OH-l-amino acid oxidase (LAAO) from the three-step purification. The minimum inhibitory concentrations (MICs) and the minimum bactericidal concentrations (MBCs) were determined for antimicrobials. The minimum biofilm inhibitory concentrations (MBICs) was performed for anti-biofilm activity. The minimum biofilm eradication concentrations (MBECs) was determined for biofilm eradication activity. www.nature.com/scientificreports/  www.nature.com/scientificreports/ nine, there was an increased antimicrobial impact against PAO1 planktonic compared to OH-LAAO alone, as indicated by the growth inhibition plate and time-kill assays ( Fig. 6B-C). The antimicrobial property of OH-LAAO was neutralized by catalase (an H 2 O 2 decomposing enzyme) (Fig. 6D-E). Likewise, catalase also neutralized the anti-biofilm effect of OH-LAAO as determined by crystal violet staining and the rescue of several biofilm-associated genes (including algD, pslB, algU, gacA, and siaD) in PAO1 or PACL with some differences between these strains (Fig. 7A-G). In P. aeruginosa, environment sensing proteins (NbdA) recognize several activators, including nitric oxide (NO) and O 2 , that promote several phosphodiesterase enzymes (PDEs) (BdlA and DipA) for c-di-GMP degradation and enhanced biofilm matrix-degrading enzymes, including EndA and PslG, the hydrolyzer against extracellular DNA (eDNA), and PSL hydrolyzer, respectively, resulting in biofilm www.nature.com/scientificreports/ dispersion (a natural process for the biofilm biomass reduction) 23 (Fig. 8A). As such, OH-LAAO promoted biofilm dispersion through the enhanced expression of nbdA, bdlA, dipA, endA, and pslG but was neutralized by catalase ( Fig. 8B-F 29,30 , as these bacteria can be isolated from venoms 31 and some snakebite wounds 32 , indicating a possible natural resistance from the microbial evolution 33 . However, P. aeruginosa, the minor microbiota in snake oral cavity 34 with biofilm production property 10 and pathogenicity 35 , was susceptible to several snake venoms, especially king cobra venoms (OH). Although both OH and TA venoms exhibited broad spectrum antimicrobial activities as indicated by the MICs and MBCs (Table 1), OH venom had a more  (Tables 1 and 2), the venoms and OH-LAAO demonstrated some impacts on biofilms as indicated by the acceptable MBIC values. Indeed, l-amino acid oxidase (LAAOs), an enzyme produced by many living organisms 36 , was identified as a potentially effective compound of OH venom with antimicrobial and anti-biofilm properties, partly through hydrogen peroxide (H 2 O 2 ) production. Although LAAOs have various l-amino acid substrates, especially l-leucine 37,38 , LAAO from the king cobra (OH-LAAO) seems to be very specific to l-lysine and l-arginine 39,40 . Indeed, snake LAAOs rapidly and strongly produce more H 2 O 2 than human LAAOs (encoded by interleukin-4-induced gene 1, IL4I1), but the human LAAOs induced higher indole-3-pyruvate (I3P) that promotes cancer survival 41 . Due to the differences in LAAOs between snakes and other hosts, l-lysine and l-arginine which were possibly more specific to OH-LAAO were used in this study. Despite H 2 O 2 neutralization by several antioxidants in most of the living cells 42 , LAAO from venomous animals produces high H 2 O 2 to induce damage in most cells 36,43 . Because H 2 O 2 -mediated cell surface injury can overcome gene-mediated antibiotic resistance, engineering LAAOs specific to bacterial cells are interesting for combating pan-drug-resistant bacteria, especially against P. aeruginosa, with a less adverse effect on the host's cells. Notably, sub-lethal OH-LAAO was used in our anti-biofilm experiments to avoid interference from microbicidal activity, then the demonstrable anti-biofilm property should be a direct effect on anti-biofilms, but not a simple antimicrobial activity. Although several sub-lethal stress (oxidative stress, H 2 O 2 , and ultraviolet A) can also enhance alginate and PSL production in P. aeruginosa biofilms 44,45 , H 2 O 2 levels from OH-LAAO might be high enough to induce bacterial adverse effects.  Mean ± SEM is presented with the one-way ANOVA followed by Tukey's analysis (*p ˂ 0.05 considered statistically significant). www.nature.com/scientificreports/ l-amino acid oxidase from O. hannah venom (OH-LAAO) versus several gene-associated P. aeruginosa biofilms. In harsh environments, Pseudomonas biofilm synthesis is initiated by i) alginate synthesis through alg operon expression from the release of sigma factor (algU) from mucA (an algU negative regulator) and ii) PSL production from psl operon activation (a c-di-GMP dependent process) through diguanylate cyclase (siaD) and gacA (transcriptional regulator)/rsmZ (non-coding small RNA) pathway 47,48 . Here, OH-LAAO inhibited biofilm production in H 2 O 2 dependent manner, and the effect was neutralized by catalase (an H 2 O 2 neutralizing enzyme) as concluded in Fig. 9 (left side). Moreover, OH-LAAO attenuated the 24 h preformed biofilm, partly through induced biofilm dispersion (the reduced biofilm biomass through the conversion from sessile into planktonic bacteria that are separated from the biofilms 23 ). The biofilm dispersion is initiated by nbdA (activated regulatory sensor) that triggers phosphodiesterases (bdlA and dipA) to digest c-di-GMP for inducing degradation of extracellular DNA (eDNA) and PSL (the important biofilm components) through endonuclease A and matrix-degrading enzyme (pslG), respectively 23 . Indeed, OH-LAAO promoted biofilm dispersion (Fig. 8), which was neutralized by catalase, suggesting a H 2 O 2 -dependent mechanism, as concluded in Fig. 9 (right side). Nevertheless, the limitations of biofilm-degrading enzymatic treatment are a necessity of multiple enzymes due to numerous types EPS in biofilms from different organisms and different EPS compositions in each step of biofilm 16 . Interestingly, H 2 O 2 alone from OH-LAAO suppressed several EPS genes (algD and pslB) and enhanced numerous EPS-degrading enzymes (endA and pslG) without a necessity for enzyme combination suggesting the potentially effective biofilm eradication. Although spreading of the infections through the dispersed cells from biofilm 23 , the OH-LAAO antimicrobial activity might reduce this concern. Despite the necessity for further in vivo experiments, OH-LAAO might be useful for the prevention and treatment of biofilms in catheters (catheter-lock solution).

Antimicrobial and anti-biofilm parameter (µg/mL)
In conclusion, antimicrobial and anti-biofilm activities of crude OH venoms against P. aeruginosa were possibly mainly due to LAAO that promoted H 2 O 2 -associated down-regulation of several genes for biofilm production. Moreover, H 2 O 2 generated by LAAO of OH venom had biofilm eradication activity via increased expressions of several genes for biofilm dispersion. The extraction or synthesis of OH-LAAO is a clinically fascinating candidate for anti-biofilms against Pseudomonas spp. More studies are interesting.

Methods
Bacterial strains and crude snake venoms. The clinically isolated strains of P. aeruginosa (PACL), S. aureus (SA1, SA2, and SA3), and E. faecalis (EF1, EF2, and EF3) were isolated from the patients in the King Chulalongkorn Memorial Hospital, Bangkok, Thailand. All methods and experimental protocols were approved by l-amino acid oxidase (LAAO) purification from OH venom. The OH crude venom was purified by a three-step procedure of column chromatography using the ÄKTA purification machine (GE Healthcare Life Sciences, Solingen, Germany) to identify the potent fraction against P. aeruginosa. Firstly, 200 mg of crude OH venom, 20 mg/mL in 10 mM phosphate buffer saline (PBS) pH 7.3, was loaded to the Sephadex G-75 column (3.2 cm × 150 cm) (GE Healthcare Life Sciences), eluted with 10 mM PBS, and measured for protein concentration by the UV absorbance (A 280 nm ; mAU). The fractions belonging to the same A 280 nm peak were pooled, including peak-1 (fraction 5-6), peak-2 (fraction 8), peak-3 (fraction 9-10), peak-4 (fraction [11][12][13][14], and peak-5 (fraction 20-22). All peaks were determined for LAAO activity (mentioned later). Notably, peak-1 from the first step demonstrated the most positive specific LAAO activity among 5 peaks at 11.17 ± 1.31 U/mg. Secondly, the peak-1 elution from the Sephadex G-75 column chromatography was concentrated and dialyzed against 50 mM Tris-HCl, pH 8.0. The concentrated peak-1 elution (6 mL) was subsequently purified using the Resource Q column (1 mL column) (GE Healthcare Life Science), which was equilibrated with 50 mM Tris-HCl, pH 8.0, and eluted with a linear gradient using 0-1 M sodium chloride (NaCl) and 50 mM Tris-HCl, pH 8.0 before measuring protein concentration (A 280 nm ). The Resource Q column fractions that were belong to the same A 280 nm peak, including peak-1 (fraction 12-13), peak-2 (fraction 14), and peak-3 (fraction 15-17), were pooled for LAAO activity determination (mentioned later). The peak-1 eluted from the Resource Q column chromatography (the second step) showed the highest specific LAAO activity among these three peaks at 15.99 ± 1.22 U/mg. Thirdly, the peak-1 from Resource Q column chromatography was concentrated and dialyzed with 50 mM Tris-HCl, pH 8.0 for further purification (the third step purification). Briefly, 5 mL of the concentrated peak-1 was loaded to the equilibrated HiTrap™ Heparin affinity chromatography column (1.6 cm × 2.5 cm, 5 mL) (GE Healthcare Life Science) with 50 mM Tris-HCl (pH 8.0), and eluted with a linear gradient using 0-1 M NaCl with 50 mM Tris-HCl, pH 8.0. Proteins in eluted fractions were measured by A 280 nm . Two peaks of A 280 nm , including peak-1 (fraction 3-7) and peak-2 (fraction 20-22), were observed and collected for LAAO activity assay (mentioned below). Between the peak-1 and peak-2 from the third step purification, the peak-1 showed the greatest specific LAAO activity (21.72 ± 1.39 U/mg) and was used for further experiments as "purified OH-LAAO". l-amino acid oxidase (LAAO) activity test. The crude OH venom and the purified fractions (mentioned above) were determined for LAAO activity using l-leucine as a substrate in the standard enzymatic assay as previously described 50 The specific LAAO activity was calculated using the following equation: Antimicrobial activity of snake venoms. Crude freeze-dried venoms dissolved in sterilized distilled water (DW) were evaluated for antimicrobial activity using the broth microdilution assay. Briefly, 25 µL of crude venoms (20 mg/mL) was added to the final volume of 200 µL of LB broth (Difco, MD, USA) in the 96-well plates, for a maximum concentration of crude venoms at 0.25 mg/mL to avoid the too less of bacterial nourishments in the broth. Then, bacterial isolates (final concentration 1 × 10 5 CFU/well) were co-cultured in LB broth supplemented with two-fold serial dilutions of crude venoms, followed by incubation at 37 °C for 18 h. The minimal inhibitory concentrations (MICs), defined as the lowest concentration that inhibits bacterial growth, were visually evaluated. Next, the bacterial growth from the broth microdilution method was inoculated on LB agar (Difco) to determine the minimum bactericidal concentrations (MBCs). The MBCs were the lowest venom concentrations that kill bacteria, resulting in no visible growth on LB agar.

Identification of the isolated l-amino acid oxidase from OH venom (OH-LAAO
Anti-biofilm activity testing (biofilm prevention). For biofilm prevention, bacteria (10 8 CFU/mL) were cultured in 96-well plates containing LB broth supplemented with a serial dilution of either crude venoms or purified OH-LAAO at 37 °C for 24 h. After incubation, the bacterial supernatants were removed by gently pipetting, and the biofilms were washed twice with sterilized distilled water (DW). The biofilms were stained with 0.1% crystal violet (CV) for 15 min and washed twice with DW. The minimum biofilm inhibitory concentrations (MBICs), defined as the lowest concentration that inhibits bacterial biofilm formation, were visually evaluated. The CV-stained biofilms were measured at the OD 590 nm (CV detection) to quantify the biofilm masses.

Biofilm eradication activity testing (experiments on 24 h preformed biofilms). The biofilm
eradication activity of either crude venoms or purified OH-LAAO was tested using 24-h preformed bacterial biofilms. As such, bacteria (10 8 CFU/mL) were grown in 96-well plates with LB broth at 37 °C for 24 h. Bacterial supernatants (containing bacterial planktonic forms) were removed by aseptically pipetting, and the 24-h biofilms were washed with sterilized DW. The biofilms were treated with LB broth containing either snake venoms or purified OH-LAAO for 24 h at 37 °C. After treatment, the remaining biofilms were stained with 0.1% crystal violet (CV) for 15 min and washed twice with DW. The minimum biofilm eradication concentrations (MBECs), defined as the lowest concentration that completely eradicates bacterial biofilm, were visually evaluated. Moreover, the remaining biofilms were quantified by CV measurement at the OD 590nm .
Bacterial gene expressions. The expression of genes encoding for biofilm-related proteins of P. aeruginosa was determined by quantitative real-time reverse transcription-polymerase chain reaction (qRT-PCR) using primers listed in Supplementary Table S2. Briefly, the total RNA of planktonic (bacteria without biofilms) or sessile (bacteria with biofilms) forms were extracted by TRIzol ® Reagent (Invitrogen, Carlsbad, CA, USA) and was converted to cDNA using RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific, Vilnius, Lithuania). The numbers of gene transcripts were quantified by the QuantStudio6 (Applied Biosystems, Foster City, CA USA) using the SYBR ® Green PCR Master Mix (Applied Biosystems). The relative number of transcripts was normalized with 16S rRNA and calculated using the 2 −ΔΔct method.

OH-LAAO and l-amino acids combination in bacterial inhibition study.
Because the substrates of LAAO were l-amino acids, the combination of OH-LAAO and l-amino acids were used against P. aeruginosa by the growth inhibition plate assay. Briefly, the lawn of P. aeruginosa (10 8 CFU/mL) on LB agar was spotted by OH-LAAO (0.025-0.1 µg/spot) with or without l-amino acids (l-lysine or l-arginine; Sigma Chemical, MO, USA) (2.5 mM/ spot). After incubation at 37 °C for 24 h, the inhibition zone was visually observed. The increased inhibition of OH-LAAO and l-amino acids was confirmed by the time-kill assay. Briefly, P. aeruginosa (10 6 CFU/ mL) was grown in LB broth supplemented with OH-LAAO (at 0.5 × MIC) with or without 250 mM of l-amino acids (l-lysine or l-arginine) at 37 °C with shaking at 120 rpm. The viable bacterial cells at 0, 15, 30, 45, 60, 120, 240, and 360 min were quantified with the drop plate method. Accordingly, bacterial suspensions were diluted serially with sterilized 0.85% NaCl, followed by dropping on LB agar to quantify bacterial colony. Furthermore, catalase, a hydrogen peroxide decomposer, was used in the time-kill study to evaluate if hydrogen peroxide was the potent antibacterial product of OH-LAAO. In brief, P. aeruginosa (10 6 CFU/mL) was grown in LB broth supplemented with snake venoms (either crude OH venom or OH-LAAO) (at 1 × MIC) and catalase (0.1 mg/mL) (SKU, Yiwu, China). The viable bacteria were determined by the drop plate method as described above.
Statistical analysis. Statistical analysis was performed using the Statistical Package for Social Sciences software (SPSS 22.0, SPSS Inc., IL, USA) and Graph Pad Prism version 7.0 software (La Jolla, CA, USA). All data were presented as mean ± standard error (SE). The differences between two groups and multiple groups were examined for statistical significance by a paired t-test or one-way analysis of variance (ANOVA) with Tukey's analysis, respectively. A p-value < 0.05 was considered statistically significant.

Data availability
All data generated or analyzed during this study are included in this published article and its supplementary information files.