A phenol amine molecule from Salinivenus iranica acts as the inhibitor of cancer stem cells in breast cancer cell lines

In recent years, the anticancer properties of metabolites from halophilic microorganisms have received a lot of attention. Twenty-nine halophilic bacterial strains were selected from a culture collection to test the effects of their supernatant metabolites on stem cell-like properties of six human cancer cell lines. Human fibroblasts were used as normal control. Sphere and colony formation assay were done to assess the stem cell-like properties. invasion and migration assay, and tumor development in mice model were done to assess the anti-tumorigenesis effect in vitro and in vivo. The metabolites from Salinivenus iranica demonstrated the most potent cytotoxic effect on breast cancer cell lines (IC50 = 100 µg/mL) among all strains, with no effect on normal cells. In MDA-MB-231 cells, the supernatant metabolites enhanced both early and late apoptosis (approximately 9.5% and 48.8%, respectively) and decreased the sphere and colony formation ability of breast cancer cells. Furthermore, after intratumor injection of metabolites, tumors developed in the mice models reduced dramatically, associated with increased pro-apoptotic caspase-3 expression. The purified cytotoxic molecule, a phenol amine with a molecular weight of 1961.73 Dalton (IC50 = 1 µg/mL), downregulated pluripotency gene SRY-Box Transcription Factor 2 (SOX-2) expression in breast cancer cells which is associated with resistance to conventional anticancer treatment. In conclusion, we suggested that the phenol amine molecule from Salinivenus iranica could be a potential anti-breast cancer component.

Salinivenus iranica, Salinibacter ruber, and the cancer stem cell-like properties of breast cancer cell lines.Sphere and colony formation determine the self-renewal potential of cancer cells 12,13 .Therefore, we evaluated the effect of Salinibacter ruber and Salinivenus iranica SMs on sphere and colony formation ability of MDA-MB-231, MDA-MB-468, and MCF-7 cells.Prior to treatment, all three human breast cancer cell lines exhibited stem-like properties, as evidenced by their ability to form colonies and large spheres.These characteristics are indicative of the presence of stem-like cells in the cell lines.Our results indicated that cells in all treated groups with both SMs lost their repopulation and could not form large spheres.Salinivenus iranica and Salinibacter ruber reduced the spheroid formation ability to about 3.5, 8.4, and 2.7 folds in MDA-MB-231, MDA-MB-468, and MCF-7 cells, respectively (p < 0.001).In addition, Salinivenus iranica and Salinibacter ruber SMs significantly decreased the size of spheres about 1.4-fold in MDA-MB-231 cells, 2.2 folds in MDA-MB-468, and 1.5-fold in MCF-7 cells (Fig. 2a,1-6).The results of the colony formation test were a bit different as Salinivenus iranica reduced the percentage and size of colonies in all treated cells significantly in MDA-MB-231, MDA-MB-468, and MCF-7 cells, (p < 0.001).While Salinibacter ruber only declined colonies in MCF-7 cells (1.8-fold, p < 0.001) with no reduction in the size of colonies (Fig. 2a, 7-12).The type of colonies also is a determinative factor for the presence of stem-like cells in a cell population as the higher number of holoclones (compact colonies with distinct margins) means that cancer stem cells were higher in the cell pools.The results exhibited that the number of holoclones reduced in Salinivenus iranica treated cells from 24.50 and 31.12 to 5.50 and 12.62 in MDA-MB-231 and MCF-7, respectively (p < 0.001) along with the reduction in meroclones (loose colonies) and paraclones (colonies with both compact and loose cells) in MDA-MB-468 cells.Salinibacter ruber only reduced the number of holoclones in MCF-7 cells from 31.12 to 16.75 (p < 0.001).However, it significantly enhanced the number of paraclones in MDA-MB-231 cells with no significant effect on the number of meroclones (Fig. 2b).
Our study found that treatment with Salinibacter ruber resulted in a significant decrease in the migration of breast cancer cells (p < 0.001).However, the invasion of MDA-MB-231 and MCF-7 cells was reduced by 60% (p < 0.001) and 77.5% (p < 0.001), respectively, while the invasion of MDA-MB-468 cells was increased by 89.1% (p < 0.001).Notably, the invasion index dramatically increased in cells treated with Salinibacter ruber, from 0.15 to 0.56 (p < 0.001) in MDA-MB-231 cells, from 0.04 to 0.60 (p < 0.001) in MDA-MB-468 cells, and from 0.74 to 6.00 (p < 0.001) in MCF-7 cells.Conversely, treatment with Salinivenus iranica had different effects on the invasion and migration of breast cancer cells.The migration of MDA-MB-231 and MDA-MB-468 cells increased by approximately two-fold (p < 0.001), while the migration of MCF-7 cells decreased by approximately 75% (p < 0.001).Specifically, Salinivenus iranica reduced the invasion of MDA-MB-231 and MCF-7 cells by 85.0% (p < 0.001) and 53.1% (p < 0.01), respectively, but did not significantly affect the invasion potential of MDA-MB-468 cells.www.nature.com/scientificreports/Furthermore, Salinivenus iranica reduced the invasion index of MDA-MB-231 treated cells from 0.16 to 0.01 (p < 0.001).However, it significantly (p < 0.01) enhanced the invasion potential of MCF-7 by approximately twofold, with no significant change in the invasion index of MDA-MB-468 cells (Fig. 2c).As the decrease in invasion index is an indicator of stem-like cells in cancer cell populations, and its reduction showed the reduction of cancer stem cells 14 , the results from Salinivenus iranica were better in reducing the cancer stem cells.Overall, because of the more desirable results of Salinivenus iranica SM on inhibiting cancer stem cells (sphere and colony formation and invasion index), this strain was selected for future experiments.

Salinivenus iranica's supernatant metabolites induce late apoptosis and S phase inhibition in breast cancer cells.
The apoptosis assay was done on treated and untreated breast cancer cell lines to find the effect of Salinivenus iranica SM on the induction of apoptosis.As shown in Fig. 3a, the percentage of both early  and late apoptosis was extremely enhanced in all treated cells, which were associated with overexpression of proapoptotic caspase-3 (CASP3) (a marker of late apoptosis) (p < 0.001).Resistance to traditional anticancer therapy is related to SOX2 expression.Therefore, inhibiting SOX2 expression may decrease the malignant characteristics associated with breast cancer, such as invasion, migration, proliferation, stemness, and chemoresistance 15 .As a result, SOX2 expression was monitored.Expression of the SOX2 gene was significantly down-regulated in MDA-MB-231 and MDA-MB-468 cells (p < 0.001, Fig. 3a).

Salinivenus iranica inhibits breast tumor growth in vivo.
We injected 1.25 × 10 6 of 4T1 cells (mice breast cancer cell line) subcutaneously to develop breast cancer tumors in BALB/c mice.When the tumor size reached 50-60 mm 3 (about three days post cellular injection), Salinivenus iranica SM at a dose of 17.5 mg/ kg (Equivalent to 100 µg/mL in vitro) was injected intratumorally, just for one time.Interestingly, the size of injected tumors was smaller than controls post seven days of SM injection (181.60 mm 3 vs.661.74mm 3 ).The expression of ki67 and sox2 (markers of cell proliferation and pluripotency) was downregulated.However, cdh1 (migration-related gene) was significantly upregulated in injected tumors (p < 0.01, Fig. 3c).
A phenol Amine molecule acts as the effective part of Salinivenus iranica metabolites.To find the effective part of Salinivenus iranica's SM, the crude metabolite was fractionated by four solvents: hexane, ethyl acetate, butanol, and water.It was separated into thirty-one fractions.A solution containing 100 µg/mL of separately fraction was prepared, and the viability of three human breast cancer cell lines was evaluated in their presence.Fraction number 6 of hexane had decreased the viability of all cell lines to 1% after 48 h, while the other 30 fractions did not exhibit any significant effect on cell lines viability (Fig. 4a).As the fraction had only one visible peak, the effective anticancer metabolite was purified in this phase.The mass spectrophotometry defined that this molecule's molecular weight (MW) was about 1961.78Da (Fig. 4b).In Fourier Transform Infrared Spectroscopy (FT-IR) analysis, three types of functional groups were observed in this molecule (Fig. 4c).1: In 2950 -2850 cm −1 , Alkyl C-H Stretch was distinguished, which are generally less useful in defining structure as a consequence of fairly ubiquitination.2: In 3500-3300 cm −1 , Amine N-H Stretch was detected, which is a pri-    To determine the anticancer activity of purified phenol amine molecule, MDA-MB-231, MDA-MB-468, and MCF-7 breast cancer cell lines and one normal human fibroblast cell line (HFF-5) were subjected to treatment by different concentrations of this molecule.After 48 h, none of the concentrations had a cytotoxic effect on fibroblasts, while the 1 µg/mL of the phenol amine molecule significantly reduced the viability of all breast cancer cells to 50% (Fig. 5a).Also, this molecule significantly downregulated the expression of the pluripotency

Discussion
Using natural compounds to target cancer stem cells could help decrease the severe side effects of chemotherapy and provide a potential anti-cancer response 8,10,16 .Therefore, in the present study, we screened supernatant metabolites (SM) of at least twenty-nine strains of halophilic bacteria isolated from different hypersaline lakes of Iran to find the effective anti-cancer strain that could potentially target most cancer cell lines, including breast (MDA-MB-468 and MCF-7), lung (A549) and prostate (DU145, PC3, and LNCaP) cancer cells.Our initial screening indicated that the SM of Salinibacter ruber and Salinivenus iranica reduced the viability of breast cancer cell lines at lower concentrations while it was ineffective on normal cells (HFF-5).At higher concentrations, these bacteria were able to decrease the viability of all selected cancer cell lines.This suggested that their effects might be dose-dependent.SMs enhanced the percentage of early and late apoptosis and the expression level of CASP3 in breast cancer cells.Sagar et al. observed that halophilic bacterial extracts were toxic at the dose of 200 and/or 500 µg/mL on HeLa, MCF-7, and DU145 cells 17 and induced apoptosis in cancer cells 18 .Apoptotic cell death is induced by caspase-3, which is an executioner caspase and a crucial mediator of intrinsic and extrinsic pathway-activated apoptosis.Upon activation of pro-apoptotic proteins, caspase-3 cleaves proteins such as cell cycle proteins, and DNase proteins and leads to apoptotic cell death 19 .More so, caspase-3, recognized as an important early biomarker for evaluating chemotherapy-induced cell death, is mainly responsible for tumor cells' response to chemotherapy via the cell apoptosis pathway and activation 20 .Furthermore, the molecule, PPDHMP, from Staphylococcus sp.strain MB30 (a marine bacterium) induces apoptosis in the lung (A549) and cervical (HeLa) cancer cells through increases in the expression of CASP3 21 .Moreover, the number of cells in the cell cycle's S phase (synthesis phase) significantly reduced in MDA-MB-231 and MCF-7 cells post-treatment with Salinivenus iranica.Therefore, Salinivenus iranica could inhibit the synthesis phase in breast cancer cells.
Interestingly, Salinivenus iranica SMs significantly reduced the number and size of spheres and colonies in all breast cancer cell lines, which were associated with the downregulation of SOX2 as a pluripotency gene.SOX2 is an essential gene in breast cancer cells that plays an early role in breast carcinogenesis and induces the metastatic potential in these cells 22 , and its inhibition could prevent tumor initiation of human breast cancer 23 .In contrast, Salinibacter ruber SMs potentially reduced the number and size of spheres and had no reduction effect on colony formation ability of all breast cancer cell lines.Both sphere and colony formation can indicate the presence of cancer stem cells in breast cancer cell lines 12,24 .More so, holoclones, which are compact colonies with specific borders, are an important marker of cancer stem cells 24 .Salinivenus iranica reduces the number of holoclones in MDA-MB-231 and MCF-7 cells.In contrast, the Salinibacter ruber reduced these numbers only in MCF-7 and enhanced the number of paraclones in MDA-MB-231, which is a pattern of induction of differentiation.Moreover, cancer stem cells are assumed as metastasis inducers in breast cancer 16 , and the invasion index is an exhibitor of invasion in cancer cells.Our results indicated that the Salinibacter ruber significantly increased this index in all breast cancer cells, but Salinivenus iranica reduced this index in MDA-MB-231 cells.Therefore, we suggested that Salinivenus iranica is a more reliable strain for inhibiting cancer stem cells.In vivo examinations showed that the tumor size decreased in the presence of Salinivenus iranica SM.Halophilic bacterial extracts exhibited tumor growth inhibition ability in previous studies.It was reported that Streptomyces sp.ANAM-5 and AIAH-10 in their co-culture produced some metabolites which had decreased tumor growth in Ehrlich Ascites Carcinoma cell-bearing albino mice 25 .The expression of ki67 was significantly down-regulated in treated tumors compared to the untreated ones.As ki67 is the marker of proliferation 26 , this revealed the role of the SM in the proliferation inhibition of breast cancer cells.Also, the expression of sox2 (a pluripotency gene) was reduced in treated tumors, which approved the in vitro studies.The upregulation of gene cdh1, which encodes E-cadherin, is important in inhibiting metastasis in the breast cancer tumor cells because the inhibition of cdh1 causes the initiation of epithelial-mesenchymal transition and metastasis 27 .The expression of cdh1 was enhanced in mouse breast cancer tumors after Salinivenus iranica treatment, and it was similar to the effect of Salinomycin on mouse breast cancer stem cells 28 .This showed the similarities between Salinomycin and Salinivenus iranica SM in inhibiting breast cancer stem cells.Moreover, the purified phenol amine molecule reduced the viability of breast cancer cells at a lower concentration with a significant decrease in SOX-2 expression.This revealed that the purified phenol amine was the effective molecule that performed the principal role in inhibiting breast cancer stem cells in Salinivenus iranica SM.

Conclusions
This is the first study to investigate the anti-cancer stem cell effects of halophilic bacterial metabolites.The findings revealed that metabolites produced from Salinivenus iranica have significant anti-cancer and anti-cancer stem cell activities.Furthermore, the anti-cancer stem cell ability of the purified phenol amine molecule was discovered in this study.The apoptosis was induced by overexpression of CASP3, and the pluripotency was reduced by the downregulation of SOX2.In addition, the SM decreased the invasion index of breast cancer cell lines and demonstrated these results in tumors of the mouse model.

Methods
Bacterial culture, supernatant isolation, and metabolite extraction.Twenty-nine halophilic strains were selected from the Iranian Biological Resources Center (IBRC) microorganism's bank.All strains cultured in Sea Water Nutrient (SWN) medium with the total salt of 3% or 7% or in Modified Growth Medium (MGM) with 23% total salt or Yeast Malt Agar (YM Agar or ISPII medium, Table 1).All the materials and reagents were purchased from Merck (E.Merck, Darmstadt, Germany).One loopful of agar slant culture of each www.nature.com/scientificreports/strain for metabolite production was inoculated into 25 mL of proper medium.The inoculated medium was incubated at 34 or 40 °C for 12 h on a rotary shaker at 150 rpm and then transferred into a 225 mL medium to cultivate at the same temperature for 3 or 7 days.The culture was taken and centrifuged at 4000 g for 40 min.The cell-free supernatant was filtered with Whatman no.1 filter paper.An equal volume of ethyl acetate was added to the upper layer with the ratio of 1:1 and was shaken well for 2 h at room temperature, and was allowed to settle overnight at 4 °C.The upper organic layer containing supernatant metabolites was collected in a clean, dry bottle and then evaporated by BUCHI Rotavapor R_114 (Switzerland) until 1 mL of total volume remained in the rotary balloon.The remaining volume was transferred to a sterile vial with identified weight and evaporated again to dry completely.The vial which contained the dried metabolite was weighted, and the pure weight of the metabolite was calculated.The SM was dissolved in DMSO (Merck, Germany), serving as a stock solution, which was later diluted to a final solvent concentration.The SM was tested as the weight of total SM/volume 9 .The SM of each inoculated medium was isolated and used as a control negative.MTT assay.The effect of SMs on cell viability was evaluated using the MTT-assay by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT, Biosera, Austria), and the results were expressed as viable cell percentages compared to the control.A 96-well plate was seeded with approximately 5000 cells per well and allowed to adhere for 24 h.Three replicates of each plate were incubated with different concentrations of metabolite extracts (0, 10, 100, 200, 400, and 800 µg/mL) or purified molecules (100, 50, 25, 10, 5, 2.5, 1, 0.5, 0.25, and 0.1 µg/mL), and viability was measured after 48 h 30 .The final concentration of DMSO was 1mM which was not cytotoxic for our cell lines (data have not been shown).After treatment, 10 μL of 5 mg/mL solution of MTT in PBS was added to each well.The plate was then incubated for 3 h at 37 °C.Finally, the medium was removed, and 100 μL of the DMSO was added to each well to solubilize the blue formazan.Dye absorbance was measured at 560 nm.One chamber of each 96 well plate contained only DMSO, and its OD 560 was assumed blank.The resulting ODs (control and treated groups) from each plate were first subtracted from the blank of the same plate, and then the mean of OD 560 for control and treated groups was calculated.The percentage of viable cells was calculated via mean OD 560 (Treated group)/ mean OD 560 (Control group) × 100.The effect of SM from inoculated mediums was also measured in the same way, and no significant changes were observed (data have not been shown).
Sphere formation assay.Sphere formation capacities were assessed after 48 h of SM pretreatment.To perform this assessment, cells were seeded at a density of 10 6 cells/well in 6-well plates and allowed to adhere for 24 h.Then the cells were treated with bacterial metabolite at 0 (as control) and 100 µg/mL (IC50 dose).After 48 h, cells were harvested for sphere formation assay.To create a single-cell suspension for sphere formation, the single-cell suspension was prepared by enzymatic (1 × Trypsin-EDTA, Sigma Aldrich) and manual disaggregation (25-gauge needle).Cells were seeded in culture dishes coated with (2-hydroxyethyl methacrylate) (poly-HEMA, Sigma) and cultivated in serum-free DMEM supplemented with B27 (GIBCO, Karlsruhe, Germany), 20 ng/ml EGF and bFGF (Royan Institute, Tehran, Iran), and PenStrep.Cells were grown for 7 days and maintained in a humidified incubator at 37 °C and an atmospheric pressure of 5% (v/v) carbon dioxide/air.After 7 days, the spheres with a diameter of > 50 µm were counted using an eyepiece graticule.The percentage of spheres pre-and post-treatment was calculated by dividing the number of spheroids by the number of seeding number × 100 and is referred to as the percentage of sphere formation.Values were expressed as means ± SD of at least three independent experiments.

Colony formation assay.
Colony formation capacities were assessed after 48 h of SM pretreatment.The numbers of 200 cells from pretreated and un-pretreated (Control) groups were plated in each well of six-well plates in DMEM medium supplemented with 10% (v/v) FBS and allowed to grow for 12 days.After this, the upper media was removed, and the chamber was washed twice with PBS.Then the colonies were fixed with paraformaldehyde 4% and stained with crystal violet.The colonies were counted under light microscopy, and the percentage of colonies pre-and post-treatment was calculated by dividing the number of colonies by the number of seeding number × 100, which is referred to as the percentage of colony formation.Values were expressed as means ± SD of at least three independent experiments.
Invasion and migration assay.Invasion and migration assays were done after 48 h of SM pretreatment.
To assess migration, 125 × 10 3 were seeded in the serum-free medium into the upper chamber of the transwell filter (pore size 8 µm, Corning; Germany).The bottom of the wells (24-well plate) was filled with DMEM medium supplemented with 20% (v/v) FBS.To assess invasion, the same procedure was done with only one difference: Transwell filters were coated with 0.5 mg/mL of Matrigel later, before seeding.For both migration and halophilic and halotolerant bacterial strains on the viability of normal and tumor cells.The numbers are referred as ≤ IC50.-: The SM did not reduce the viability of cells to ~ 50% in any of examined concentrations.Fibroblast (HFF-5), prostate cancer (PC3, LNCaP, and DU145), lung cancer (A549), and breast cancer (MDA-MB-468, MDA-MB-231 and MCF-7) cell lines.SWN Sea Water Nutrient Medium, MGM Modified Growth Medium, ISPII Yeast Malt Medium, NT Not Tested.Significant values are in bold.

Figure 2 .
Figure 2. Effect of S. ruber and S. iranica SM (100 µg/mL) on cancer stem cell properties of breast cancer cell lines after 48h treatment.(a) Sphere formation.1.Sphere formation in S. ruber.2. Sphere formation in S. iranica.3. Percentage of sphere formation in S. ruber.4. Percentage of sphere formation in S. iranica. 5. Size of spheres in S. ruber.6. Size of spheres in S. iranica.(b) Colony formation.1. Colony formation in S. ruber.2. Colony formation in S. iranica.3. Percentage of colony formation in S. ruber.4. Size of colonies in S. ruber. 5. Percentage of colony formation in S. iranica.6. Size of colonies in S. iranica.7. The shape of colonies in S. ruber.8.The shape of colonies in S. iranica.9. Percentage of holoclones in colonies in S. ruber.10.Percentage of meroclones

Figure 3 . 2 .Figure 4 . 2 .
Figure 3.Effect of S. iranica SM on apoptosis and cell cycle of breast cancer cell lines and Tumorigenicity of the SM in breast cancer mice model.(a) Effects on apoptosis.1. Apoptosis test of Salinivenus iranica SM on breast cancer cells.The number of cells in early (downright quadrate) and late (upright quadrate) apoptosis increased in treated cells in comparison with controls.2. The percentage of early apoptosis.3. The percentage of late apoptosis.4. Quantitative real-time PCR analysis of CASP3 in breast cancer cell lines.5. Quantitative real-time PCR for SOX2 (a pluripotency gene) in breast cancer cell lines.(b) Effects on cell cycle.1.The cell cycle of Salinivenus iranica SM on breast cancer cells.2. The number of cells in the S phase of the cell cycle after 48 h treatment with 100 µg/mL of Salinivenus iranica.3. The number of cells in the G1 phase was up-regulated in MDA-MB-468 (p < 0.5) and MCF-7 (p < 0.001) significantly but was not changed on MDA-MB-231.4. The number of cells in the G2 phase was up-regulated in MDA-MB-468 significantly (p < 0.05) but was not changed in MDA-MB-231 and MCF-7 cells.(c) Tumorigenicity in breast cancer mice model.1. Viability of mouse breast cancer cell line (4T1) in presence of 10, 100, and 200 µg/mL of Salinivenus iranica SM. 2. Breast cancer mice models and their breast cell line tumors in control and treated (SM injected) groups.3. Mean tumor volume of 4T1 cell line.4. Quantitative real-time PCR for ki67 (a marker of proliferation), sox2 (a pluripotency gene) and cdh1 (a migration-related gene) in 4T1 tumors.n = 4, *p < 0.05, **p < 0.01, ***p < 0.001.

Table 1 .
Effect of SM of 1. Sphere formation in S. ruber.2. Percentage of meroclones in colonies in S. iranica.14.Percentage of paraclones in colonies in S. iranica.(c) Migration and invasion.1. Migration test in S. ruber.2. Invasion test in S. ruber.3. Migration test in S. iranica.4. Invasion test in S. iranica. 5. Migration number of Breast cancer cell lines in S. ruber.6. Invasion number of Breast cancer cell lines in S. ruber.7. Invasion index of Breast cancer cell lines in S. ruber.8. Migration number of Breast cancer cell lines in S. iranica.9. Invasion number of Breast cancer cell lines in S. iranica.10.Invasion index of Breast cancer cell lines in S. iranica.n = 4, **p < 0.01, ***p < 0.001.
Sphere formation in S. iranica.3. Percentage of sphere formation in S. ruber.4. Percentage of sphere formation in S. iranica. 5. Size of spheres in S. ruber.6. Size of spheres in S. iranica.(b) Colony formation.1. Colony formation in S. ruber.2. Colony formation in S. iranica.3. Percentage of colony formation in S. ruber.4. Size of colonies in S. ruber. 5. Percentage of colony formation in S. iranica.6. Size of colonies in S. iranica.7. The shape of colonies in S. ruber.8.The shape of colonies in S. iranica.9. Percentage of holoclones in colonies in S. ruber.10.Percentage of meroclones in colonies in S. ruber.11.Percentage of paraclones in colonies in S. ruber.12. Percentage of holoclones in colonies in S. iranica.13.