Gut Bacteria of Columbia livia Are a Potential Source of Anti-Tumour Molecules

Objectives: The overall aim was to determine whether gut bacteria of Columbia livia are a potential source of antitumour molecules. Methods: Faecal and gut microbiota of Columbia livia were isolated, identified and conditioned media were prepared containing metabolites. Growth inhibition, lactate dehydrogenase cytotoxicity and cell survival assays were accomplished against cervical cancer cells. Next, liquid-chromatography mass spectrometry was conducted to elucidate the molecules present. Results: A plethora of bacteria from faecal matter and gastrointestinal tract were isolated. Selected conditioned media exhibited potent anticancer effects and displayed cytotoxicity to cervical cancer cells at IC50 concentration of 10.65 and 15.19 µg/ml. Moreover, cells treated with conditioned media exhibited morphological changes, including cell shrinking and rounding; indicative of apoptosis, when compared to untreated cells. A total of 111 and 71 molecules were revealed from these gut and faecal metabolites. The identity of 60 molecules were revealed including, dihydroxymelphalan. Nonetheless, 122 molecules remain unidentified and are the subject of future studies. Conclusion: These findings suggest that gut bacteria of Columbia livia possess molecules, which may have anticancer activities. Further in silico testing and/or high throughput screening will determine potential anticancer properties of these molecules.


Bacterial identification
Bacterial identification was carried out based on their texture, size, colour, and shape and inoculating them onto separate fresh nutrient agar plates which were successively incubated overnight at 37oC. Bacteria were subcultured until agars with pure cultures were obtained. Pure bacteria were subjected to Gram staining. Next, bacterial identification was conducted using Analytical profile index (API) identification strip; API 20E was used for gram-negative bacteria while API staph was used for gram-positive bacteria as described previously Akbar et al., 2018).

Conditioned medium (CM) preparation
Pure bacterial cultures were inoculated in RPMI-1640 were incubated for 48h, at 37 o C. Following incubation, bacterial cultures were centrifuged at 4 o C, 10,000 x g, for 1h. Supernatant were collected and filtered using sterile 0.22μm pore size cellulose acetate syringe filter, also referred to CM (conditioned media). The CM were quantified for protein estimation using Bradford assay and then stored at -80 o C Akbar et al., 2018).

Growth inhibition assays
Assays to determine inhibition of growth was performed as previously described. Briefly, cells were grown in 96-well plates until 50% confluency was reached. Control wells were trypsinised to determine the cell count, while experimental wells were treated with CM prepared from bacteria isolated from both faeces and gut of C. livia. CM from E. coli K-12 (a non-pathogenic laboratory strain) was used as negative control. The treated cells were incubated at 37oC in a 5% CO 2 with 95% humidity, until the untreated cells became 100% confluent. The cells were then trypsinised with 2.5% trypsin for 15 min and subjected to Trypan blue exclusion assay using a haemocytometer. The growth inhibition effects were established by comparing the number of viable cells of untreated cells (control) and treated cells. To confirm the anticancer activity of the CM, selected CM were concentrated using vacuum concentrator. CM were then quantified through Bradford assay as described above and used for growth inhibitory assays at a concentration of 10µg/ml.

Cytotoxicity assays, cell staining and survival assays
Cells were grown to confluency in 96-wells plates. Next, cells were treated with CM prepared from bacteria isolated from faeces and gut of C. livia for 24 h at 37 o C in a 5% CO 2 with 95% humidity. CM from E. coli K-12 was used as negative control, while the positive control was prepared by treating control cells with 0.2% Triton X-100 for 30 min at 37°C, to induce 100% cell death. Next, the supernatants were collected and lactate dehydrogenase release was determined using cytotoxicity detection kit. The percentage cytotoxicity was calculated as follows: % c y t o t o x i c i t y = ( ( A b s o r b a n c e s a m p l e -Absorbancenegative control)/ (Absorbancepositive control -Absorbancenegative control)) X 100, whereby the negative control comprised of cells treated with RPMI-1640 media only, and the positive control contained cells treated with the detergent: Triton X-100.
Moreover, cell staining was performed. Briefly, cells were fixed with 100% acetone and 100% methanol in a 1:1 ratio for 15min and stained with 0.4% Trypan blue for 15min. Plate was dried and pictures of individual wells were captured. Cell survival assay was also conducted to determine viability of cells treated with CM. Briefly, cells treated with CM were collected and seeded onto new plates containing growth media for 24 h at 37 o C in a 5% CO 2 with 95% humidity, and their re-growth were examined using a light microscope Akbar et al., 2018).

16S rDNA sequencing of bacteria with active CM
Selected bacteria that exhibited activity were subjected to 16S rDNA sequencing. Bacteria were grown in nutrient broth at 37 o C with constant shaking. Bacteria were then pelleted by centrifugation and subjected to DNA extraction using QIAGEN DNA extraction kit, as detailed in the manufacturer's instructions (Akbar et al., 2019). The extracted bacterial DNA was amplified using 16S amplification using Taq DNA Polymerase 2X-preMix and a pair of 16S rDNA Universal primers; 27F (5'-AGAGTTTGATCCTGGCTCAG-3') forward primer and 1492R (5'-GGTTACCTTGTTACGACTT-3') reverse primer, with conditions: 1 cycle at 95oC for 5 minutes, amplification step; 30 cycles (i) 95 o C for 30 seconds, (ii) 55 o C for 30 seconds, and (iii) 72 o C for 1 minute and final step; 1 cycle at 72 o C for 5 minutes. Gel electrophoresis was conducted using 1% agarose gel in 1x Tris-acetate-EDTA (TAE) buffer at 100 V for 40min. Amplified DNA was sequenced via Sanger sequencing. The nucleotide sequences obtained were aligned using ChromasPro software and were then blasted into the National Center for Biotechnology Information (NCBI) database for Standard Nucleotide BLAST, for 16S RNA sequences, to obtain the identity of the bacteria with maximum percentage nucleotide match (Akbar et al., 2019).

IC 50 determination of active CM through MTT cell viability assay
The IC 50 value of active CM was determined using MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay as previously described by (Akbar et al., 2019). Briefly, cells were grown to 70% confluency in 96-wells plates and treated with active CM at various concentrations; 2.5, 5.0, 10 and 20 µg/ml in growth medium. Post incubation, positive control cells were Microchannel Plate detector, while the blank expended after each sample was of composition 50% MeOH + 50% MiliQ water. Chromatogram were generated from LC-MS which were used to determine the identity of the molecules from the Metlin_AM_PCDL-N-170502.cdb database. SciFinder database was then used to determine whether the identified molecules had previously reported biological activities.

Various bacteria were isolated from faeces and gastrointestinal tract of C. livia
Several Gram-positive and Gram-negative bacteria were isolated from the faeces and gastrointestinal tract of C. livia (Table 1). Gram-positive bacteria isolated were coagulase negative Staphylococcus spp., Gram positive bacilli, and Lysinibacillus boronitolerans, while the Gramnegative bacteria were Escherichia coli, and Pseudomonas aeruginosa. While from the gastrointestinal tract, Grampositive bacteria; Gram positive bacilli, Lysinibacillus fusiformis, Lysinibacillus boronitolerans, coagulase negative Staphylococcus spp., Lysinibacillus sphaericus, Streptococcus group B, and Streptococcus saprophyticus and Gram-negative bacteria were Escherichia coli, and Sphingobacterium multivorum. As CLF05 and CLG01 showed potent activity, 16S rDNA sequencing was performed on these two bacteria. The result revealed that CLF05 and CLG01 are Bacillus cereus ATCC 14579 and Bacillus velezensis strain FZB42, respectively (Table 1). treated with 0.2% Triton X-100 for 30 min at 37°C, to induce 100% cell death. The percentage cell viability was calculated as follows:

% Cell viability = [(Absorbance sample -Absorbance blank)/ (Absorbance negative control -Absorbance blank)] X 100
Identification of molecules(s) exhibiting anticancer activity through Liquid chromatography-mass spectrometry (LC-MS) Active CM were subjected to LC-MS analysis as described previously . Molecules were extracted from active CM through solvent extraction using chloroform in 1:3 ratio of chloroform to CM. The molecules dissolved in chloroform were subjected to evaporation and resuspended in 1:1 ratio of methanol to type 1 water. The samples were subjected to Agilent 1290 Infinity LC system coupled to Agilent 6520 Accurate-Mass Q-TOF mass spectrometer with dual ESI. The molecules were separated through Agilent Zorbax Eclipse XDB-C18 column with a particle size of 3.5 micron and Narrow-Bore of 2.1x150mmat 25oC using solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in Acetonitrile) for a total run time of 30 minutes. The separated molecules were then ionized by means of ESI + jet stream ion mode with the QQQ analyzer, with parameters: capillary voltage at 4500 V, sheath gas flow at 8 L per min, fragmentor voltage 135 V, gas temperature at 350ºC, gas flow at 8 L per min, and nebulizer gas at 40 psi and detector used was MCP Figure 1. Growth Inhibition Effect of Conditioned Media (CM) Prepared from Bacteria Isolated from the Faecal Sample of Pigeon against HeLa Cells. A) Semi-confluent HeLa cells were incubated with CM and growth inhibition was determined as described in Materials and Methods. B) Using an inverted microscope, images of the cells were taken at x250. The results are representative of several experiments performed in duplicate. P value was determined using two sample T-test, two-tailed distribution, (*) is <0·05. Negative control (0h) is number of cells at the beginning of experiment, while negative control (24h) is untreated cells. The data are compared between negative control (24h) and CLF/CLG. Detailed CLF and CLG nomenclature is described in Table 1.

CLG01 and CLF05 produced damage to HeLa cells
Microscopic images for HeLa cells treated with CLF05 showed change in morphology and the cells appeared round, as compared to untreated cells (negative control), albeit no LDH release was determined. In contrast, CM for faeces depicted morphology similar to the negative control. Additionally, wells containing cells treated with CLF05 remained stained similar to the positive control.

Concentrated CM inhibited growth of cancer and normal cells
At a concentration of 10µg/ml, CLF05 inhibited growth of HeLa, MCF-7, PC3 and HaCaT cells by 100, 100, 90.4 and 82.8% respectively, while CLG01 inhibited growth of aforementioned cells by 100, 91.6, 80.0 and 35.1% respectively (Figure 3). The quantitative data were further supported by microscopic images obtained post-treatment of normal and cancer cells. MTT assay results revealed that CLF05 and CLG01 affected viability of HeLa cells at an IC 50 concentration of 10.65 and 15.19 µg/ml respectively (Figure 4).

CLF05 and CLG01 possess 111 and 71 molecules
CLF05 and CLG01 were subjected to LC-MS. Figure  6 shows spectra (negative and positive ion polarity) of molecules detected from CLF05 and CLG01. A total of 111 molecules were detected from CLF05, out of which 54 molecules were identified (Table 2 and Supplementary  Table 1), while the remaining molecules are unidentified and potentially novel. Moreover, out of the 54 identified    Table 2).

Figure 2. Growth Inhibition Effect of CM Prepared from Bacteria Isolated from the Gastrointestinal Ttract of Pigeon
against HeLa cells. A) Semi-confluent HeLa cells were incubated with CM and growth inhibition was determined as described in Materials and Methods. B) Using an inverted light microscope, images of the cells were taken at x250. The results are representative of several experiments performed in duplicate. P value was determined using two sample T-test, two-tailed distribution, (*) is <0·05. Negative control (0h) is number of cells at the beginning of experiment, while negative control (24h) is untreated cells. The data are compared between negative control (24h) and CLF/CLG. Detailed CLF and CLG nomenclature is described in Table 1.

Discussion
There is continued rise in morbidity and mortality associated with cancer, despite the wide range of available treatment options, hence it is imperative to discover and develop new anticancer agents. Our previous work highlighted that animals that thrive in polluted environments, such as crocodiles, cockroaches, water monitor lizards and snakes, may possess anticancer and antibacterial mechanisms to ward off disease Jeyamogan et al., 2017;Mosaheb et al., 2018;. In addition to the animal lysates, the faecal and gut microbiota is of particular interest to us, as the gut microbiota is known to play an important role in regulating the behaviour and health of its host Heyde and Ruder, 2015;Alarcón et al., 2016). Anticancer effects from avian species remain unexplored and were the subject of present study. A repertoire of bacteria was isolated from both the faecal and gut microbiota of C. livia.
The results revealed that B. cereus and B. velezensis exhibited potent effects against cancer cell lines tested. CM from CLF05 (B. cereus) and CLG01 (B. velezensis) inhibited growth of HeLa cells by more than 40%. Moreover, they inhibited almost 100% cell growth of cancer cells at a concentration of 10µg/ml and affected the viability of HeLa cells at IC50 concentrations of 10.65 and 15.19 µg/ml, respectively. Furthermore, microscopic images revealed that CLF05 and CLG01 produced damage to HeLa cells. Although LDH release was negligible but CM affected the integrity human cells indicating that cell death might have resulted from apoptosis. Thus, further experiments are needed to determine apoptosis in treated cells.
Previous work has reported that Bacillus species produce metabolites with anticancer properties, which support our findings (Ferdous et al., 2018). Bioactives such as ε-Poly-L-lysine, Surfactin, Leodoglucomide B, Bacillistatins-1 and 2 and Mixirins A, B and C synthesised by bacteria have shown to inhibit growth and induce morphological changes in cancer cells (Arnold et al., 1979;Harris and Pierpoint, 2012;Pettit et al., 2009;Tareq et al., 2012;Wu et al., 2017). Previously, anticancer potential of metabolites secreted by B. cereus isolated from soil sample were tested against human liver cell lines and cancer cell lines (Kumar et al., 2014). The results revealed that metabolites from B. cereus was cytotoxic to cells with IC50 of 225.4 µg/mL, 152.2 µg/mL, and 152.2 µg/mL against HepG2, Hep2 and Chang liver cells, respectively. These findings suggest that metabolites released by B. cereus exhibit anticancer activity by inducing apoptosis in cancer cells (Kumar et al., 2014).
The LC-MS revealed 111 molecules including dihydroxymelphalan. Dihydroxymelphalan, also known as melphalan is a widely used anticancer drug. It works by causing inter-strand cross-links in DNA strands; resulting in inhibition of DNA and RNA synthesis (Kuczma et al., 2016;Rahaman et al., 2018). Moreover, melphalan upregulates expression of reactive oxygen species which in turn triggers apoptosis in cancer cells by activating caspase-9 (Kuczma et al., 2016;Rahaman et al., 2018). Since melphalan detected in CM from CLF05 induces apoptosis, it supports our speculation that CLF05 induces damage in cancer cells through apoptosis. The other molecules that were elucidated are tazobactam-m1, citric acid, 3-furoic acid, 4-hydroxyphenylglyoxylate, diethanolamine, methionyl-hydroxyproline, bifenazate, diethyltoluamide, metabutethamine, rivastigmine, tropicamide, xylylcarb, and daimuron. These molecules exhibited activities ranging from antimicrobial, hypolipidemic, cholinesterase inhibitor, to inhibitor of fatty acid oxidation, however, their anticancer effects need to be determined (Halstenson et al., 1994;El Baaboua et al., 2018;In et al., 2013;Hall et al., 1985;Stephens et al., 1985;Leung et al., 2005;Deblander et al., 2015;Asano et al., 1997;Alberts et al., 1979;Allen et al., 1970;Ochiai et al., 2007;Sudakin and Osimitz, 2010;deShazo and Nelson, 1979;Khoury et al., 2018;Manny et al., 2001). Similarly, none of the molecules from CLG01 were reported to possess anticancer effects and this will be determined in future studies. Of note, 57 molecules detected from CLF05 and 65 molecules detected from CLG01 respectively remain unknown and are potentially innovative anticancer agents. Moreover, only aerobic culturable bacteria were isolated in this study and future studies are needed to isolate anaerobic bacteria and unculturable bacteria, that might also be a principal resource of anticancer molecules.
In conclusion, for the first time, we have showed that gut and faecal microbiota of avian species are a potential source of anticancer properties. We have elucidated various molecules that could serve as possible drug leads; but further research is needed to achieve these prospects. It will be imperative to investigate anticancer effects of bacteria against other cancer cells. Our results further augment our speculation that animals residing in unsanitary habitats, that are detrimental to humans, maybe in fact a large unexploited repertoire for innovative pharmaceutical drugs.