Montivipera bornmuelleri Venom: Inhibitory Effect on Staphylococcus epidermidis and Escherichia coli F1F0-ATPases and Cytotoxicity on HCT116 Cancer Cell Lines

In this work, we pursued the biological characterization of the venom of Montivipera bornmuelleri, a viper from the Lebanese mountains. In relation to its antibacterial potential, the inhibitory effect of this venom on the F1F0-ATPase enzymes of Gram-positive Staphylocoocus epidermidis and Gram-negative Escherichia coli bacteria was examined. In order to determine the degree of cytotoxicity of the venom on the HCT116 human colon cancer cell lines, the biological MTT proliferation and cell viability test were implemented. After validation of the enzymatic F1F0-ATPase model by the spectrophotometric method, using quercetin as the reference ligand, results revealed that M. bornmuelleri venom is able to inhibit the activity of the enzyme of these two bacteria with a concentration of the order of 100–150 µg/mL. In addition, a venom concentration of 10 µg/mL was sufficient to kill the totality of HCT116 cell lines cultivated in vitro. These data show that M. bornmuelleri venom is a mixture of diverse molecules presenting activities of interest, and is a potential source to explore in order to discover new drug candidates.


Introduction
Montivipera bornmuelleri (M. bornmuelleri) is a rather small, venomous snake (maximum length of 75 cm) belonging to the viperidae family [1,2]. It is endemic to the Mount Lebanon region, found only at high altitudes (above 1800 m) and listed as endangered by the IUCN due to its limited geographical range [2]. The venom of M. bornmuelleri is a mixture of different compounds, with the major being Phospholipase A2 (PLA2), serine-protease, metalloproteinase III [1] and L-amino oxidase (LAAO) [3] and certainly other various molecules unexplored to date. This venom as well as its components display a wide range of biological activities. In fact, crude venom of M. bornmuelleri possesses antibacterial activity against Gram-positive and Gram-negative bacteria [4], plays an anticancer effect against HaCaT cells [5], induces upregulation of pro-inflammatory cytokines [6], has pro-and anti-coagulant activities, indirect hemolytic activity [7] and a relaxant effect Since the discovery of penicillin in 1928 by Alexander Fleming [10], many bacteria developed resistance to one or more classes of antibiotics, rendering the treatment of bacterial infections increasingly difficult [11]. In fact, 25000 patients die annually in the EU due to infection by multidrug-resistant (MDR) bacteria, indicating that mortality rates are very high [10]. Therefore, the need for novel antibacterial agents has emerged, especially those based on new bacterial biological processes [11].
Staphylococcus epidermidis, a Gram-positive staphylococcus found on the human skin and mucosal surfaces [12], capable of forming biofilms and causing nosocomial infections [13] and Escherichia coli, a Gram-negative bacillus, a part of the mammalian gut microbiome, but also possessing pathological strains [14], both developed a resistance to antibiotics [13], [15] and both possess an F1F0-ATPase [16,17]. This membrane associated enzyme found in mitochondria, bacteria and thylakoids [18] consists of a membrane-bound F0 sector, a proton channel, and a soluble F1 sector. The soluble portion is responsible for both the hydrolysis and formation of ATP. Both activities are achieved by a rotary mechanism of the F0 and F1 sectors [19]. Its important role in cell life and death makes it an interesting target for new antimicrobial drugs [20].
Cancer has a high incidence and mortality, and both are increasing rapidly worldwide. Among the numerous human cancers, colon cancer is a major cause of death (9.2% mortality rate, after lung cancer) and has a very high incidence (10.2%) in both males and females, preceded by lung and female breast cancer [21]. Some of the current therapies used to treat this type of cancer include radical surgery, considered to be the most effective method of treatment and adjuvant chemotherapy (Palliative chemotherapy, anti-EGFR therapy and anti-angiogenic therapy) [22]. However, it is known that such treatments as chemotherapy have significant side effects and influence quality of life of patients. In fact, many patients express heavy concerns towards the known side effects of chemotherapy, such as loss of hair (50% of patients), constant fatigue (42% of patients) and disturbance of work duties (39% of patients) [23]. Some chemotherapeutic agents may even cause cardiotoxicity and other cardiac side effects [24]. As these types of treatments continue to prove unsatisfactory and even dangerous, it is important to continue screening for new anticancer medication and improved therapies to treat colon cancer. Since then, the search for a natural extract having a cytotoxic effect on the human colon cancer cell lines will undoubtedly contribute to the discovery of new molecules which could possibly be explored as drug-candidates for this type of cancer.
This work aims to assess M. bornmuelleri venom's ability to inhibit two bacteria F1F0-ATPase, S. epidermidis and E. coli and its cytotoxic effect on the HCT116 colorectal cancer cell lines. obtained from Loba Chemie (Mumbai, India). Magnesium chloride MgCl2 was supplied by Acros organics (New Jersey, USA). Hydrochloric acid HCl and sulfuric acid H2SO4 (96.3%) were provided by VWR chemicals.
S. epidermidis: 50 µL from frozen stock tubes are inoculated on nutrient agar (NA) plates. 8-10 colonies of the bacteria from NA culture plates are inoculated in sterile tubes containing 10 mL TSB in order to prepare pre-cultures. Both plates and tubes are incubated for 48 hours with partial aeration at standard conditions and 37°C, and finally stored at 5°C until use.
E. coli: 2-3 colonies of the bacteria from NA culture plates are inoculated in 5 mL of nutrient medium (TSB) and incubated at 37°C for 18h, with slightly open caps. In order to determine the number of bacteria, a UV-VIS spectrophotometer is used to measure the optical density (OD) of the obtained pre-cultures at 620 nm. This value corresponds to the number of bacteria x 10 9 (CFU/mL).

Preparation of the bacterial sample
In order to study the membrane-bound F1F0-ATPase, it must first be isolated from bacterial cells. The protocol used followed the method by Issa et al. [25] with some modifications. A volume of 500 µL of S. epidermidis bacterial suspension was centrifuged at 10000xg and 4°C for 10 minutes. Supernatant was removed and the pellet is resuspended in 500 µL of Tris-HCL (pH 8.5) then sonicated on ice 6 times for 30 seconds each, and 10 seconds of rest between each cycle. The suspension was again centrifuged at 10000xg and 4°C for 10 min and the supernatant removed. The obtained pellet contains the F1F0-ATPase enzyme.
Similarly, 500 µL of E. coli bacterial suspension was centrifuged at 10000xg and 4°C for 15 minutes. 400 µL of a Tris-HCL buffer (50 mM; pH 8.5) containing 4% (v/v) of EDTA was used to resuspend the pellet after the supernatant removed, followed by incubation for 10 minutes at room temperature. A volume of 100 µL of 4% (v/v) glycerol was added to the bacterial suspension. Next, a freeze-thaw cycle is performed on the suspension, which is then sonicated 3 times on ice for one-minute each. The suspension was lastly subjected to another centrifugation at 10000xg and 4°C for 15 minutes and the supernatant is removed. The obtained pellet contains the F1F0-ATPase enzyme and was washed twice with Tris-HCL buffer (50 mM; pH 8.5) to remove glycerol and EDTA.

Phosphate quantification protocol
The hydrolysis of ATP by F1F0-ATPase leads to the formation of ADP and inorganic phosphate (Pi). The latter is used to quantify the enzymatic reaction. A linear regression showing absorbance values as a function of standard Pi solution concentrations ranging from 10 to 50 µM is used to deduce the concentration of Pi. Standard Pi solutions are prepared in tris-HCL buffer (50 mM, pH 8.5) and their concentrations are quantified using the method of Lowry et al. [26]. In each tube containing 1000 µL of the different Pi concentrations, and one tube containing 1000 µL of Tris-HCL buffer serving as a negative control, is added 100 µL of a 1% ammonium molybdate solution and the contents are thoroughly mixed, allowing the formation of a phosphomolybdic acid complex. After adding 100 µL of a 1% ascorbic acid solution to each tube and incubating for 10 min at room temperature, a blue molybdous compound is formed. Both ammonium molybdate and ascorbic acid solutions are prepared in 0.25N of H2SO4. A spectrophotometer is used to measure the optical density of each tube at 700 nm. The assays for each Pi concentration were done in triplicate.

Study of the inhibition the membrane-bound S. epidermidis and E. coli ATP synthases by quercetin
After obtaining a pellet containing the isolated F1F0-ATPase enzyme as previously described in section 2.3, inhibitory assays was carried out using quercetin, a natural inhibitor of the enzyme, at different volumes, for both S. epidermidis and E. coli. The pellet was first resuspended in Tris-HCL buffer (50 mM; pH 8.5) and MgCl2 (15 mM) and incubated at 37°C for 10 minutes. Then to each sample was added a different volume of quercetin solution so that different concentrations are obtained (1-500 µM) as well as 100 µM of ATP to start the reaction. The concentration of ATP used was saturating for the enzyme. Next, each tube was incubated at 37°C for 40 min, and 1% (v/v) of SDS was added to each sample in order to stop the reaction. The samples were then centrifuged at 10000xg for 10 minutes and finally the pellet was removed and the supernatant, containing the Pi released by the enzymatic reaction was tested as described in section 2.4 to study the inhibition of the enzyme. Two control tubes were prepared, one lacking the inhibitor quercetin to obtain the maximum activity of the enzyme, and one lacking both the inhibitor and ATP, was used as a blank. All tests were performed in triplicate.
Inhibitory concentration (IC50) value of quercetin was calculated using GraphPad Prism (GraphPad Software, San Diego, USA). This was achieved by using a curve of the enzymatic activity (%) as a function of log of different concentrations of quercetin (µM).

Study of the potential inhibitory effect of M. bornmuelleri venom on the membrane-bound S. epidermidis and E. coli F1F0-ATPases
The pellet containing the isolated F1F0-ATPase as described in section 2.3 is tested with M. bornmuelleri venom as described in section 2.4 to study its effect on the enzyme. The same protocol was applied on both S. epidermidis and E. coli membrane-bound F1F0-ATPase: To each tube containing the membrane-bound enzyme are added different concentrations of M. bornmuelleri venom (0-200 µg/mL) prepared in Tris-HCL buffer (50 mM; pH 8.5), MgCl2 (15 mM), Tris-HCL buffer (50 mM; pH 8.5) and lastly ATP (100µM, a saturating concentration) to start the reaction, and for a final volume of 1 mL for each tube. Samples are then incubated at 37°C for 40 minutes, and 1% (v/v) of SDS is added to each sample in order to stop the reaction. Lastly, the tubes are centrifuged at 10000xg and 4 °C for 10 minutes. Two blanks, one containing Tris-HCL MgCl2 and ATP and the other without ATP were prepared. A positive control tube, containing the enzyme, Tris-HCL, MgCl2, and ATP was prepared to determine the maximum activity of the enzyme. All assays were done in triplicate.
For both bacteria, curves of the enzymatic activity (%) as a function of different concentrations of the venom (µg/mL) are established.

HCT116 Cell culture
To the medium used for HCT116 Cell line culture were added 1% Penicillin Streptomycin (100 U·mL−1) and 10% heat-inactivated fetal bovine serum (FBS) in order to amplify it. DMEM acquired at Sigma Aldrish (Gibco Dulbecco's Modified Eagle Medium) was used to culture the HCT116 Cell line at 37°C in a humidified atmosphere with 5% CO2 and 95% air.

Cellular viability assay
The MTT test for cellular viability was performed as follows: 96 well plates were used to seed HCT116 cells at a density of 10 4 cells/well, when they reached 60-80% confluency the cells were treated for 24 hours and experiments were carried out with different concentrations of the venom extract. The medium was then discarded and to each well was added 100 µL of MTT solution. Finally, the absorbance was measured by ELISA READER at 570 nm. All tests were done in triplicate.

Validation of the phosphate dosage method
The linearity of measurement of the inorganic phosphate (Pi) resulting from the ATP hydrolysis during the enzymatic reaction catalyzed by F1F0-ATPase enzyme is essential to achieve quantitative analysis on the assays. So, a suitable colorimetric method was chosen as previously described by Lowry et al 1945 [26] with some optimizations elaborated previously by our team [27]. Experiments were performed with solutions of low concentrations of Pi of 10, 20, 30, 40 and 50 µM.
Results represented in Figure 1 show a positive linear correlation between the optic density (OD) and the Pi concentration (R 2 > 0.9954) with a good repeatability of the measurements since the relative standard deviation (RSD %) is less than 3% for each standard Pi solution assayed in triplicate. This regression will be used to quantify the Pi released during enzymatic assays catalyzed by the S. epidermidis and E. coli membrane-bound F1F0-ATPases.

Screening of the reference inhibitor quercetin on S. epidermidis and E. coli F1F0-ATPase enzymes
The natural quercetin was chosen as reference inhibitor of the two bacteria F1F0-ATPases in order to validate the experimental model. The dose-response curves representing the enzymatic activity (%) of S. epidermidis and E. coli F1F0-ATPases as a function of log of reference inhibitor concentrations are represented in Figure 2. a and b, respectively. The corresponding IC50 values obtained are 24.2 ± 2.2 M for S. epidermidis and 30.0 ± 1.5 M for E. coli. These values are in good agreement with those reported in literature [28] which validate the developed method. Good repeatabilities were obtained on assays with RSD < 5% for both bacteria.

Effect of the M. bornmuelleri venom on the membrane S. epidermidis and E. coli. F1F0-ATPase enzymes
The M. bornmuelleri venom test on the membrane S. epidermidis and E. coli F1F0-ATPase enzymes shows that this extract is a potentially active inhibitor on those enzymes.  are expressed as mean ± SD (n = 3). Quercetin was used as a positive control. One-way ANOVA-test: ns (no significant), * p < 0.05; *** p < 0.001 when compared with the control. The M. bornmuelleri venom exerts a dose dependent inhibition on ATPase activity.

Effect of the M. bornmulleri venom on HCT116 cancer cell lines
The cytotoxic activity of M. bornmuelleri on human colon cancer HCT116 cells was exanimed using MTT (bromure de 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium) assay. Cancer cells were exposed to increasing concentrations of M. bornmuelleri venom (0.5, 1, 2.5, 5, 10, 50 and 50 µg/mL) for 24h. Results were expressed as the percentage of cell viability in comparison with the untreated control cells with 100% of viability. The data showed that the M. bornmuelleri venom inhibits the total cell viability of HCT116 cells with a low concentration of 10 µg/mL of the M. bornmuelleri venom. In fact, this concentration was able to induce a significant decrease in cell viability (95 % of viability) in comparison to the control (Figure 3, a). The results reveal the strong cytotoxic effect of the Lebanese snake M. bornmuelleri venom on HCT116 cells with an IC50 of 6.9 µg/mL calculated by Prism software (Figure 3, b).

Discussion
Montivipera bornmuelleri is among three viper species found in Lebanon and inhabiting mountainous regions at altitudes above 1800 m [2]. Our previous studies aimed to characterize this snake and its venom in order to prove its important therapeutic applications. Indeed, an ecological study found that the snake prefers rocky regions with few thorny cushion plants and its maximum size is around 60 cm [2]. Its venom was analyzed and it was shown that it contains mainly three enzymes: PLA2, Metalloprotease III and serine protease, which possess many biological activities. In fact, the venom displayed both pro-and anti-coagulant activities depending on its concentration, as well as indirect hemolytic activity, possibly due to the presence of PLA2 [7]. It was shown by in vivo experiments on mice that it induced an up-regulation of pro-inflammatory cytokines (INFγ, TNF-α, IL-1β, IL-4 and IL-17) while also down-regulating the anti-inflammatory cytokine IL-10 [6]. In addition, the M. bornmulleri venom exhibits antibacterial effect against both Gram-positive and Gram-negative bacteria, as well as antifungal activity [4]. Moreover, given the resistance to antibiotics of a multitude of bacteria and now more than ever, the need to discover new antibacterial agents [37] such as those capable of acting on the F1F0-ATPase enzymes which are essential for the pathogenic microorganisms survival, we were interested in testing the venom effect of M. bornmuelleri on the F1F0-ATPases of bacteria from both Gram-positive (S. epidermidis) and Gram-negative (E. coli) genera. In agreement with our previous results [4], we found here that the venom inhibited the activity of the F1F0-ATPase enzyme presented in the membranes of these two bacteria. This membrane-bound enzyme plays an important role in cell life and death by both synthesizing and hydrolyzing ATP. Since ATP deprivation is linked to cell death [20], M. bornmuelleri venom having the ability to inhibit the activity of this enzyme translates its ability as an effective antimicrobial agent. Thus, we showed that this crude venom managed to reduce the enzymatic activity to a minimum of 20% at concentrations of 100 to 150 µg/mL. However, this inhibitory effect was only assayed on isolated F1F0-ATPase of both bacteria, and not on total cells. The ability of the venom to inhibit the ATPase of Gram-negative bacteria, which are more resistant to antibiotics than Gram-positive ones due to the presence of an outer membrane that must be passed by drugs to attain their target [29], with the same level of inhibition (maximum of 80%) than Gram-positive bacteria confers it a broad spectrum of activity and an important property. Indeed, it can be used against both types of bacteria instead of acting against only one, and this expands its potential applications. Another potential benefit for this activity, is the fact that the transfer of antibiotic resistance genes is possible between Gram-positive and Gram-negative bacteria [30] or between the same types of bacteria, such as S. epidermidis which is able to transfer these genes to S. aureus, and thus spread antibiotic resistance [13], so an antibacterial agent that is not susceptible to resistance by the various bacteria proves to be even more useful. PLA2, an enzyme purified from Protobothrops mucrosquamatus snake venom is able to control both Gram-positive (Bacillus subtilis) and Gram-negative (Pseudomonas aeruginosa) bacterial strains [31]. Metalloproteinase isolated from Agkistrodon halys, inhibited also Gram-positive (Staphylococcus aureus) and Gram-negative (Proteus vulgaris) bacteria [32]. LAAO, purified from Bothriechis schlegelii venom exerts too an inhibitory effect against both Grampositive (S. aureus) and Gram-negative (Acinetobacter baumannii) bacteria [33]. All of these enzymes are found in M. bronmuelleri venom and could be responsible for its antibacterial activity, which is in accordance with these findings in the literature [2].
In terms of cytotoxicity, our previous studies showed that the M. bornmuelleri venom has a specific cytotoxicity against benign A5 and low-grade malignant II4 cancer cell lines, which was higher than that observed with HaCaT keratinocytes [5] as well as B16 skin melanoma cells and 3-MCA-induced murine fibrosarcoma cell lines, which were more sensitive to the venom when overexpressing ovalbumin. However, it did not manage to reduce tumor size in vivo [34]. It has been shown that the venom possesses a vasorelaxant effect on K + depolarized smooth muscle and inhibits the contraction of smooth muscle tissue induced by CaCl2, phenylephrine and AngI [8] but did not have a strong cytotoxicity towards human erythrocytes [3]. Here, different concentrations of the crude venom of M. bornmuelleri were tested against human colon cancer cell line HCT116 and it was found that a low concentration of 10 µg/mL was able to totally reduce cancer cell viability (by 95%) compared to untreated control cells and with an IC50 value of 6.9 µg/mL. This is in agreement with the literature, where the snake venom toxin from Vipera lebetina turanica was also tested on HCT116 colon cancer cells and reduced their viability with an IC50 value of 1.14 µg/mL. This cell growth inhibition was achieved by inducing apoptosis in the treated cells through increase in ROS generation, upregulation of death receptors DR4 and DR5, cleavage of caspases-3, 8 and 9 and activation of the JNK pathway [35]. M. bornmuelleri venom could act through similar pathways and it would be beneficial to elucidate its mechanism of action in future studies. In addition, crude venom from the Malaysian Cryptelytrops purpureomaculatus demonstrated cytotoxic activity in SW480 and SW620 colon cancer cell lines with EC50 values of 29.43 µg/mL and 23.19 µg/mL respectively, which is in good agreement with our results. In this study, LAAO was also isolated from the venom and decreased the viability of the SW480 and SW620 colon cancer cell lines with lower EC50 values of 13.56 µg/mL and 13.17 µg/mL respectively through apoptosis [36]. The purification of a single component from M. bornmuelleri venom in order to test it on HCT116 cells could similarly prove to possess an enhanced cytotoxic effect compared to the crude venom. However, another study in which Malaysian mangrove pit viper Trimeresurus purpureomaculatus venom was assayed on HT-29 colon cancer cell lines, showed that viability was reduced with a low IC50 value of 0.42 µg/mL, suggesting that high cytotoxicity could result from synergistic interactions between various compounds present in the venom instead of only one component. Indeed, the different fractions obtained by HPLC had demonstrated each variable cytotoxic activity [37]. Since colon cancer is a major cause of cancer-related death and its incidence and mortality are increasing quickly [21], and its treatment faces the problem of the resistance of this disease to chemotherapy [38], which is also unsatisfactory since it has many adverse side effects and lowers quality of life of patients [23], finding new treatments for this life-threatening disease is an important step that must be taken to improve public health, and natural sources such as snake venoms, and specifically M. bornmuelleri venom as we showed in this study, prove to be of special interest to discover such therapeutic agents.