Apoptotic Induction in Human Cancer Cell Lines by Antimicrobial Compounds from Antarctic Streptomyces fildesensis (INACH3013)

Abstract

The Antarctic Streptomyces fildesensis has been recognized for its production of antimicrobial compounds with interesting biological activities against foodborne bacteria and multi-resistant strains, but not for its potential antiproliferative activity and mechanisms involved. Two bioactive ethyl acetate extract (EAE) fractions were purified via thin-layer chromatography and High-Performance Liquid Chromatography (HPLC), showing that orange-colored compounds displayed antimicrobial activity against pathogenic bacteria even after shock thermal treatment. The UV–VIS features of the active compounds, the TLC assay with actinomycin-D pure standard, Fourier transform infrared (FTIR) spectra and the ANTISMASH analysis support the presence of actinomycin-like compounds. We demonstrated that S. fildesensis displays antiproliferative activity against human tumor cell lines, including human breast cancer (MCF-7), prostate cancer (PC-3), colon cancer (HT-29) and non-tumoral colon epithelial cells (CoN). The half-maximal effective concentrations (EC₅₀) ranged from 3.98 µg/mL to 0.1 µg/mL. Our results reveal that actinomycin-like compounds of S. fildesensis induced apoptosis mediated by caspase activation, decreasing the mitochondrial membrane potential and altering the cell morphology in all tumoral and non-tumoral cell lines analyzed. These findings confirm the potential of the psychrotolerant Antarctic S. fildesensis species as a promising source for obtaining potential novel anticancer compounds.

1. Introduction

Cancer and multi-resistant pathogens are a major global health problem [1]. Prostate, colorectal and breast cancers are the most common diagnosed cancers in the world [1]. Finding new antibiotics and effective anticancer drugs is a major research focus.

The genus Streptomyces has attracted significant interest due to its ability to synthesize different biologically active secondary metabolites, including more than 80% of known antibiotics, antitumoral molecules and growth effectors, among other compounds. Consequently, the genus Streptomyces is considered a primary source of useful drugs to improve human life that cannot be matched by any other organism [2,3]. These bioactive compounds are mainly produced by the activation of cryptic gene clusters that are not active under normal conditions [4]. Microorganisms that inhabit extreme environments have evolved to cope with abiotic stresses in their environment, offering huge potential for the discovery of bioactive compounds. In this sense, the Streptomyces species, derived from extreme sources, is still an extraordinary reservoir of novel biosynthetic gene clusters with the potential for developing anticancer drugs [5].

Despite efforts to explore the biotechnological potential of polar streptomycetes, there is still very little information regarding its chemical characteristics and potential bioactivity. Recently, two psychrotrophic Antarctic Streptomyces fildesensis strains (INACH3013 and So13.3) have been reported to inhibit the growth of several pathogens, including the foodborne pathogens Staphylococcus aureus ATCC 25923 and MRSA [6,7]. The S. fildesensis strain So13.3 contains biosynthetic gene clusters [BGCs] for the synthesis of diverse antibiotics, including streptorubin, surfactin, sarmentoside, prenyletin, dactinomycin, actinomycin and piericidin [7]. A previous report described the presence of an actinomycin-D-related analog in the crude extract of phylogenetically related S. fildesensis. This extract shows pronounced antiproliferative activity in tumor cell lines [8].

Actinomycin is a class of chromopeptide lactone antibiotics produced by several species of Streptomyces. Among the 30 different molecules described, actinomycin D is one of the most studied molecules and is widely used for tumor treatment [9,10]. Its anticancer role has been associated with the inhibition of RNA Polymerase I (RNA Pol I) due to its ability to intercalate double-stranded DNA, blocking ribosome biogenesis and inducing nucleolar stress response [10,11]. Despite current actinomycin studies, there are still some knowledge gaps regarding its chemical structure and its relationship with antitumoral action mechanisms, which are very important aspects to understand in deep biological activities.

The production of bioactive compounds in Antarctic Streptomyces requires exploring not only new strains but also experimental procedures to pose Antarctic streptomycetes as a source of novel compounds. With this in mind, the present work aimed to characterize the activity of two major purified fractions present in the cell-free supernatant of the Streptomyces fildesensis strain INACH3013 against different human cancer cell lines. Based on these results, the putative mechanisms involved in apoptosis, as well as the potential application of these bioactive compounds, are discussed.

2. Materials and Methods

2.1. Streptomyces fildesensis INACH3013 Growth Conditions and Extracts

The psychrotrophic S. fildesensis INACH3013, isolated from soil of the Antarctic Fildes Peninsula (62°12′26.4″ S, 58°58′28.7″ W) [6], was grown in triplicate in 200 mL of Oatmeal medium (30 g/L) at 12 °C with shaking (180 rpm) for 5 days [12]. Cell-free supernatants (CFS) were obtained by centrifuging the cultures at 8000× g for 10 min. Secondary metabolites were extracted from the CFS by using ethyl acetate 1:1 (v/v) as described previously [13]. The CFS and ethyl acetate extract (EAE) were lyophilized and stored at −20 °C until use.

2.2. Thin-Layer Chromatography Direct Bioautography (TLC)

Thin-layer chromatography (TLC) was performed to separate and screen the most active compounds through the bioautography detection method [14]. Samples of 10 µL of EAE (45 µg) were loaded in triplicate onto TLC plates and were developed with a mixture of ethyl acetate and methanol (9:1 v/v) as the mobile phase. After drying, the compound spots were visualized under visible and short UV (366 nm) light. Ten spots were cut, eluted and used for further assay against Staphylococcus aureus ATCC 6538P. The strain was grown in 1.3% Mueller–Hinton (MH) broth at 37 °C for 6 h and was grown in molten MH agar up to an OD₆₀₀ of 0.001. The molten agar was further poured into an empty petri dish with spots and was allowed to solidify. The inhibition zones (mm²) were visualized after incubation at 37 °C for 24 h. The inhibition zone was used as an indicator of the antimicrobial activity associated with each spot. The two major colored spots with remarkable activity (retention factor = R_f = 0.65 and R_f = 0.72) were chosen and processed for purification. Authentic actinomycin D (Thermo Fisher, Grand Island, NY, USA) was used as the standard to perform one-dimensional TLC (1D-TLC; Figure 1H) and two-dimensional TLC (2D-TLC; Figure S1).

2.3. Purification and Partial Characterization by High-Performance Liquid Chromatography (HPLC)

TLC spots with remarkable antimicrobial activity (R_f = 0.65 and R_f = 0.72) were resuspended in 60% acetonitrile. The separations and analysis of each spot’s main compounds were performed via high-performance liquid chromatography (HPLC) using a Jasco PU 980 (Spectroscopic Tokyo, Japan) provided with a UV detector (Jasco UV-970), an integrator (Jasco 807-IT) and a reverse phase column, Kromasil (100-5c18 250 × 4.6 mm). Fraction separation was performed in the gradient mobile phase (acetonitrile, ACN): ultra-pure water, acidified with trifluoroacetic acid (0.1% v/v) from 60% ACN to 85% ACN at a flow rate of 1 mg/mL. The volume of the injected sample was 100 μL, and the running time was 20 min. The absorption spectrum of the resulting fractions was analyzed in the UV–visible range (200–650 nm). The purified HPLC fractions were lyophilized for further analyses (thermal stability, cell viability assay, caspase activity and mitochondrial membrane potential).

2.4. FTIR Characterization

Fourier-transform infrared (FTIR) spectra for the characterization of the purified HPLC compounds obtained in this work were recorded in the spectral range of 4000–400 cm⁻¹ (using KBr as reference) using a Perkin Elmer Frontier spectrophotometer (Perkin Elmer, Waltham, MA, USA).

2.5. Thermal Stability Assay in Bioactive Compounds

The thermal stability of compounds at different temperatures (20, 60 and 100 °C) and exposure times (30, 60, 90 and 120 min) was evaluated using a SmartBlock DWP 1000 Thermoblock (EppendorfLtd., Stevenage, UK). Lyophilized ethyl acetate extracts were resuspended in ethanol (1×) and diluted in water (100×) up to a final concentration of 5.7 mg/mL, and the aqueous solutions (10 μL) were inoculated over a molten agar medium seeded with S. aureus to evaluate the thermal response based on the produced inhibition halo [6]. Furthermore, purified HPLC fractions were incubated at 20 and 100 °C for 1 h, and the thermal stability was further determined in triplicate following the procedure described above. Due to different concentrations recovered from each spot, the results are expressed as the relative percentage of inhibition relative to the corresponding control at 20 °C.

2.6. Identifying Biosynthetic Gene Clusters (BGCs)

The presence of actinomycin-like secondary metabolite biosynthetic gene clusters was determined using AntiSMASH 5.0. (https://antismash.secondarymetabolites.org/, accessed on 1 April 2021) using strict parameters with all features selected [15]. This result was compared with the complete genome of the sister strain, So13.3. The BLAST-based (ANIb) and MUMmer-based (ANIm) whole-genome average nucleotide identity between strains was calculated with JspeciesWS [16]. In silico DNA–DNA hybridization (isDDH, GLM-based) was calculated with the genome-to-genome distance calculator (GGDC) [17]. Synteny analysis and the visualization of the main BGCs between both genomes of S. fildesensis strains, So13.3 [7] and INACH3013 [18], were performed with TREBOL software (www.inf.imo-chile.cl/software/trebol.html accessed on 1 April 2021), using DC-MEGABLAST with an e value cut-off < 0.0001.

2.7. Cell Viability and Cytotoxicity Assay against Cancer Cell Lines

The cell lines used for the viability and cytotoxicity assay were human breast cancer (MCF-7), prostate cancer (PC-3), colon cancer (HT-29) and non-tumoral colon epithelial (CCD841) CoN. The cell lines were maintained at 37 °C in Ham’s-F12: DMEM High Glucose medium (Gibco, San Diego, CA, USA) and were supplemented with 10% (v/v) fetal bovine serum (FBS), 100 U/mL penicillin, 100 µg/mL streptomycin and 1 mM glutamine in a humidified 5% CO₂ incubator. Serial dilutions of the extracts (CFS and EAE), as well as the purified HPLC fractions (R_f = 0.65 and R_f = 0.72), were prepared to achieve final concentrations ranging from 500 µg/mL to 0.0125 µg/mL. Cells were seeded in 96-well plates at 3 × 10³ cells per well, containing the serial dilutions of each extract or fraction, and were then incubated in replicates (n = 5 per treatment) for 24 h. Cell viability was determined via the Sulforhodamine B (SRB) (Sigma, St. Louis, MO, USA) dye assay [19]. For this, cell lines were seeded at a density of 3 × 10³ cells per well and were incubated for 24 h in 96-well plates. Then, cells were treated with CFS, EAE, R_f = 0.65 and R_f = 0.72 extracts for 24 h. After the treatment, cells were fixed by the addition of 25 µL of cold 50% (wt/vol) trichloroacetic acid and were incubated for 60 min at 4 °C. Then, plates were washed with deionized water and dried for 24 h. An amount of 50 µL of SRB solution (0.1% wt/vol in 1% acetic acid) was added to each well and incubated for 30 min at room temperature. Unbound SRB was removed by washing with 1% acetic acid. Plates were air-dried, and bounded staining was solubilized with 100 µL of Tris base (10 mM). Optical densities were read in an automated spectrophotometer plate reader at 540 nm. Finally, the percentage of cell viability and half-maximal effective concentration (EC₅₀) was determined. For the EC₅₀ calculation, the percentages were plotted against the logarithm of extract concentrations.

2.8. Morphological Characterization of Cell Apoptosis

Changes in cell morphology were analyzed using an inverted microscope (Olympus IX81, Olympus America, Center Valley, PA, USA). Morphological changes in the nuclear chromatin of cells undergoing apoptosis were evaluated using Hoechst 33342 nuclear fluorescent dye [20,21]. An amount of 1 × 10⁴ cells/mL of CoN, HT-29, PC-3 and MCF-7 cell lines were seeded independently in 24-well plates and exposed to the selected compounds for 24 h; then, 1 µM of Hoechst 33342 was added to each well and incubated for 30 min in the dark at room temperature. The cells were further washed with cold phosphate-buffered saline (PBS), and cell viability was determined with the number of blue fluorescent cells as a proxy of the dye incorporation using a fluorescence microscope (Olympus IX81).

2.9. Analysis of Mitochondrial Membrane Potential (MMP)

Rhodamine 123 (Thermo Fisher Scientific), a cationic voltage-sensitive probe that reversibly accumulates in mitochondria, was used to detect changes in transmembrane MMP. CoN, PC-3 and HT-29 cell lines were seeded in 24-well plates (8 × 10⁴ cells per well). Exponentially growing cells were incubated with EAE, R_f = 0.65 and R_f = 0.72 for 12, 24 and 48 h. Cells were stained with 1 mΜ of rhodamine 123 at 37 °C for 60 min. Cells were detached from the plate and washed with ice-cold PBS. Samples were analyzed via flow cytometry (BD FACSJazz™ Cell Sorter) using a 488 nm laser and by collecting the fluorescence at 585 nm. Data are expressed as the percentage of cells that were positive for rhodamine 123 dye.

2.10. Analysis of Caspase Activity

Caspase activity was determined using the fluorescent inhibitor of caspases tagged with fluorescein isothiocyanate, FITC-VAD-FMK (CaspACE™ FITC VAD-FMK In Situ Marker, Promega, Madison, WI, USA) according to the recommended manufacturer instructions. Flow cytometry was performed to measure the activity of total caspases [22,23]. Briefly, CoN, PC-3 and HT-29 cell lines were exposed to EAE, R_f = 0.65 and R_f = 0.72 fractions for 48 h. Cells were incubated with CaspACE™ in the dark for 20 min at room temperature. Subsequently, the medium was removed, and cells were washed twice with PBS. Exposed cells were collected via trypsinization and centrifugation (7 min at 1500× g). The supernatant was discarded, and the pellet was resuspended in PBS and analyzed via flow cytometry using the 488 nm excitation laser and by collecting the fluorescence at 692 nm. The results of caspase activity are expressed as the times of induction in comparison with the control treatment.

2.11. Statistical Analyses

Statistics were performed using R v3.6.3 [24]. Statistical differences among treatments were evaluated using t-tests or one-way and two-way ANOVAs, followed by Tukey’s HSD (honestly significant difference) test for multiple comparisons using the dplyr package [25]. A regression analysis was used to determine EC₅₀ values using the tidyverse, ggplot2 and drc packages [26,27,28].

3. Results

3.1. Purification and Characterization of R_f = 0.65 and R_f = 0.72 Fractions by HPLC

The R_f = 0.65 and R_f = 0.72 fractions with antimicrobial activity were purified and characterized (Figure 1). The lyophilized extracts appeared as an orange-yellow powder. Both fractions turned from yellow to red when reacting with 20% sulfuric acid, and no reaction with 0.1% ninhydrin in ethanol was observed. Absorption spectra in the UV–visible range (220–500 nm) of both fractions revealed two peaks at 240 nm and 445 nm for both extracts (Figure 1B). The ratios of the absorbance at 240 nm to that at 445 nm were 1.3 and 1.35 for R_f = 0.65 and R_f = 0.72, respectively. When TLC was performed using actinomycin D as a reference, it exhibited the same R_f found for the R_f = 0.65 fraction (Figure 1H).

3.2. Thermal Stability of Bioactive Compound from S. fildesensis INACH3013

Streptomycete EAE was highly stable because exposure to 60 °C for 120 min did not induce a significant decrease in antimicrobial activity (Figure 1E; one-way ANOVA, F(4,10) = 1.20, p = 0.254). In the same way, no antimicrobial activity decrease was observed when the EAE was exposed to 100 °C for 30 min (Figure 1E).

However, the antimicrobial activity of the EAE decreased to 48.5% after 120 min of incubation at 100 °C (Figure 1E; one-way ANOVA, F(4,10) = 29.2, p < 0.001). Regarding the purified fractions, the reduction in antibacterial activity with the high-temperature treatment (100 °C for one hour) was significant for the analyzed fractions (one-way ANOVA, F(3,8.37) = 35.7, p < 0.001). The observed loss of activity after the 1 h treatment ranged between 46.8% and 39.8%, which correlates with the degree of polarity of each fraction (R_f = 0.72 > R_f = 0.65), with the non-polar fraction being more sensitive to temperature (Figure 1F). No significant differences among the fractions in terms of antimicrobial activity was observed (Student’s df = 8, p = 0.302; Figure 1F).

3.3. Identification of Secondary Metabolite Biosynthetic Gene Clusters

The strain INACH3013 harbors 33 gene clusters for the biosynthesis of secondary metabolites (Table S1), including genes for the potential production of antibiotics such as actinomycin. Other BGCs with high similarity to well-known BGCs corresponded to ectoine, melanin, geosmin, alkylresorcinol, spore pigment, pristinol and hopene, with gene similarities to known BGCs ranging from 73% to 100%. For the BGCs in S. fildesensis So13.3, 29 secondary metabolite regions were found in this sister strain (Table S1).

3.4. Cytotoxic Effect of Bioactive Compounds from S. fildesensis INACH3013

The CFS, EAE extract and TLC purified fractions (R_f = 0.65 and R_f = 0.72) exerted differential cytotoxicity against a series of challenged cancer cell lines (

Table 1. Cytotoxicity of Streptomyces extracts and fraction. The half-maximal effective concentration (EC₅₀) values of the cell-free supernatant (CFS), ethyl acetate extracts (EAE) and the R_f = 0.65 and R_f = 0.72 fractions recovered from TLC spots against MCF-7, PC-3 and HT-29 cells and from non-tumoral colon epithelial cells CCD841 (CoN) used as the control were calculated as described in Materials and Methods as the mean ± standard error of the mean (SEM), n = 5.

Sample	EC50 (µg/mL)
	CoN F = 58.13; p < 0.0001	MFC-7 F = 37.95; p < 0.0001	PC-3 F = 14.25; p < 0.0001	HT-29 F = 43.42; p < 0.0001
CFS F = 0.3557; p = 0.8738	17.52 ± 2.18 g,h,i	12.97 ± 1.99 j,l,m	11.95 ± 2.84 n,o	13.42 ± 1.96 r,s,t
EAE F = 6.953; p < 0.0001	0.49 ± 0.14 d,f g	0.06 ± 0.01 d j	0.35 ± 0.04 n,p	0.10 ± 0.02 f r
Rf = 0.65 F = 25.97; p < 0.0001	1.50 ± 0.28 a h	2.04 ± 0.19 b l	7.71 ± 1.13 a,b,c p,q	1.16 ± 0.25 c s
Rf = 0.72 F = 5.335; p = 0.0097	0.16 ± 0.02 i	0.14 ± 0.01 m	0.19 ± 0.01 e o,q	0.09 ± 0.01 e t

(One-way ANOVA with post hoc Tukey’s HSD test by row: ^{a, b, c} p < 0.001; ^{d, e} p < 0.01; ^f p < 0.05; by column: _{g, h, i, j, l, m, n, o, r, s, t} p < 0.001; _{p, q} p < 0.05).

). The CFS exerted the least cytotoxic effect among all the tested cell lines.

Higher cytotoxic activity was achieved by both the EAE and the R_f = 0.72 fraction. Independently from the cellular line assayed, the cytotoxicity of the R_f = 0.72 fraction was significantly higher than the cytotoxicity of the R_f = 0.65 fraction, ranging between 9 and 40 times higher. As

Table 2. Effect of metabolites on the mitochondrial membrane potential (MMP). Percentage of cells labeled with rhodamine 123 after treatment with total ethyl acetate extract (EAE), fraction R_f = 0.65 and fraction R_f = 0.72 for 12 h, 24 h and 48 h (mean ± standard deviation; n = 5).

Treatments		Mitochondrial Membrane Potential (%)
		Cell Lines
		CoN	PC-3	HT-29
	12 h	69.6 ± 0.61	34.3 ± 0.52	26.0 ± 0.47
EAE	24 h	48.1 ± 0.31	34.4 ± 0.67	30.6 ± 0.75
	48 h	32.1 ± 0.34	46.9 ± 0.97	62.1 ± 3.21
	12 h	63.2 ± 0.54	35.9 ± 1.30	26.4 ±0.70
Rf = 0.65	24 h	36.4 ± 0.32	81.0 ± 1.36	82.6 ± 0.12
	48 h	15.6 ± 1.15	91.3 ± 4.17	98.8 ± 0.15
	12 h	21.2 ± 0.90	44.7 ± 1.16	37.5 ± 0.24
Rf = 0.72	24 h	20.2 ± 1.37	43.6 ± 0.48	28.0 ± 0.28
	48 h	22.8 ± 0.69	36.1 ± 0.15	24.0 ± 0.13
Two-way ANOVA (treatment vs Cell line)		F(4,36) = 278.3, p-value < 0.0001	F(4,36) = 116.4, p-value < 0.0001	F(4,36) = 429.9, p-value < 0.0001
One-way ANOVA EAE		F(2,12) = 1820, p-value < 0.0001	F(2,12) = 94.87 p-value < 0.0001	F(2,12) = 104.5 p-value < 0.0001
One-way ANOVA Rf = 0.65		F(2,12) = 995.2 p-value < 0.0001	F(2,12) = 124.5 p-value < 0.0001	F(2,12) = 8220 p-value < 0.0001
One-way ANOVA Rf = 0.72		F(2,12) = 1.631 p-value = 0.2362	F(2,12) = 41.11 p-value < 0.0001	F(2,12) = 943.4 p-value < 0.0001

shows, the cytotoxicity of the R_f = 0.72 fraction (measured as EC₅₀) was significantly higher than the cytotoxicity found in the total EAE for all cellular lines except MFC-7.

The comparison of cytotoxicity between EAE and the R_f = 0.72 fraction was 3 times significantly higher in the CoN cell line (t = 2.333, df = 8; p = 0.024) and 1.8 times higher in the PC-3 cell line (t = 3.881, df = 8; p = 0.0023) but was not significant in the HT-29 cell line (t = 0.4472, df = 8; p = 0.3333). However, for MFC-7, the cytotoxic activity was 2.3 times significantly lower (t = 5.657, df = 8; p = 0.0002) in comparison with the total EAE. Because the main activities are associated with EAE, R_f = 0.65 and R_f = 0.72, we used them for the subsequent analysis.

3.5. Induction of Cell Apoptosis by Bioactive Compounds

Morphological changes were observed in all analyzed cell lines (CoN, HT-29, PC-3 and MCF-7) after the treatment with the EAE and with the R_f = 0.65 and the R_f = 0.72 fractions (Figure 2). Untreated cells (controls) appeared as normal, angular and spindle-shaped, whereas prominent changes were observed in cells after 24 h of exposure to the extracts, adopting a spherical shape that tended to increase the aggregation distribution, indicating that cells appeared to have improved intercellular adhesion (mainly for R_f = 0.72), nuclear margination, nuclear fragmentation, chromatin condensation, secondary necrotic cells, apoptotic bodies and late apoptotic cells. These results were confirmed when cells were labeled with Hoechst 33342 (Figure 3E–H). Cell shrinkage with lesser cytoplasm mass was also observed (indicated by arrows; Figure 3), with their number varying accordingly to the cancer cell line (Figure 3). The most prominent effect was observed in the HT-29 cell line, whereas the PC-3 cell line was the least sensitive to the extracts (Figure 2 and Figure 3; Table 1).

3.6. Reduction in Mitochondrial Membrane Potential (MMP) by Bioactive Compounds

Cancer cell lines treated with the EAE and the R_f = 0.65 and R_f = 0.72 fractions from S. fildesensis INACH3013 showed a reduction in the MMP. This reduction was statistically significant (p < 0.0001) for the CoN, PC-3 and HT-29 cells lines (Table 2). Moreover, the MMP reduction varied significantly according to the exposure time and the utilized fraction (two-way ANOVA using all cell lines—EAE: F(4,36) = 1226.11, p < 0.0001; R_f = 0.65: F(4,36) = 1296.27, p < 0.0001; R_f = 0.72: F(4,36) = 322.53, p < 0.0001).

In the case of CoN cells, the MMP decreased significantly when increasing the time of exposure to EAE and R_f = 0.65 (one-way ANOVA—EAE: p < 0.0001; R_f = 0.65: p < 0.0001). The percentage of labeled cells labeled reached a maximum value of 32% at 48 h after treatment with EAE. The treatment of R_f = 0.65 decreased the MMP percentage significantly over time, reaching 15% at 48 h. The R_f = 0.72 fraction exerted a significant reduction in MMP in CoN cells (two-way ANOVA, p <0.0001), reaching 21% at 12 h, with no significant changes over time (24 h, 48 h; one-way ANOVA, p = 0.2362). For the PC3 cell line, the maximal reduction was observed at 48 h when exposed to both the EAE and R_f = 0.65 (46.9% and 91.3%, respectively).

Conversely, there was a decrease in the percentage of reduction when PC3 cells were treated with the R_f = 0.72 extract over time. A similar trend was observed in the HT-29 cells, with a maximal potential reduction (98.8%) when cells were treated with the R_f = 0.65 extract for 48 h.

For the PC3 cells, treatment with EAE induced a decrease in the MMP, reaching a level of 34% and maintaining this level for 12–24 h. Then, this effect tended to be reversed, increasing significantly to 47% at 48 h (one-way ANOVA, p < 0.0001).

In a similar way, the R_f = 0.65 fraction reached 36% at 12 h; then, this result increased significantly from 81% to 91% at 24 and 48 h, respectively (one-way ANOVA, p < 0.0001). The fraction R_f = 0.72 reached MMP values of 44% and 43% at 12 and 24 h, respectively. Moreover, the R_f = 0.72 fraction decreased significantly to 36% at 48 h (one-way ANOVA, p < 0.0001).

After treatment with EAE extract, the percentage of HT-29 cells marked with rhodamine 123 reached 26% at 12 h. However, this result tended to be reversed, significantly increasing the percentage of cells marked to 82% and 99% at 24 h and 48 h, respectively. Contrary to treatment EAE and R_f = 0.65, in cells treated with R_f = 0.72, it was possible to first observe a decrease in MMP that corresponded to 37.5% after 12 h and then a new significant decrease of 28% and 24% at 24 h and 48 h, respectively (one-way ANOVA, p < 0.0001).

3.7. Caspase Induction in CoN, PC-3 and HT-29 Cancer Cell Lines

As shown in

Table 3. Caspase induction in CoN, PC-3 and HT-29 cell lines. Time induction values and fold increases relative to the control after cell treatment for 48 h with total ethyl acetate extracts (EAE), R_f = 0.65 fraction and R_f = 0.72 fraction, as described in Materials and Methods.

	Fold Induction of Caspases
Treatments	Cell Lines
	CoN	PC-3	HT-29
EAE	0.0	0.1	1.6
Rf = 0.65	0.3	1.3	1.3
Rf = 0.72	0.5	0.1	1.3

, treatment with the extracts exerted a significant induction in the activity of caspases in all analyzed cell lines but mainly in the HT-29 cell line. EAE was the most effective treatment in inducing caspase activity in HT-29 cells, with a fold caspase induction of 1.6. R_f = 0.65 was the only effective extract against the PC-3 and HT-29 cell lines, with a fold caspase induction of 1.3. The three fractions did not induce the caspase activity in CoN cells.

4. Discussion

Species of the genus Streptomyces have been responsible for producing two-thirds of all therapeutic bioactive metabolites [4], which, in many cases, have shown high toxicity, being effective on Gram (+), Gram (−) and human pathogenic fungi [2,6,7,14]. The same toxic effect has been observed on different human cancer cell lines [5,9,11,13]. This broad spectrum of action is an important characteristic that must be met by metabolites that are synthesized as chemical defenses [4].

The thermal degradation of bioactive compounds has been described as an important feature related to their antimicrobial activity [29]. Here, we show that EAE and bioactive TLC bands (R_f = 0.65 and R_f = 0.72) from the INACH3013 strain are thermostable at 100 °C for 30 min. Despite thermal treatment, EAE and bioactive TLC bands showed a significant reduction in antimicrobial activity after prolonged exposure (Figure 1). Additionally, the chemical characterization of those compounds in the two TLC spots with sulfuric acid, ninhydrin and the UV–visible spectra of both HPLC purified fractions showed that the absorbance ratio of 240 nm to that at 445 nm was between 1.30 and 1.50 when observed between 220 and 500 nm (Figure 1B). According to this and considering that R_f = 0.65 used the same migration of actinomycin D as the control, it is possible to suggest that both compounds belong to the actinomycin class due to the color and UV–visible spectra, as described previously by JH Lee and coworkers [30,31]. On the other hand, the FTIR spectrum obtained for the compound was highly consistent with the spectrum obtained in the case of actinomycin D (Figure S1 and Figure 1C). When comparing the spectra, it was possible to observe a marked profile with only slight differences in the area of the bands of the most relevant functional groups, showing that this compound would belong to the actinomycin family. In a broad sense, the signals corresponding to the carbonyls associated with the structures (ester, ketone and amide) could be clearly distinguished in the area of 1600 to 1700 cm⁻¹. In the region of 2900 to 3500 cm⁻¹, we expected to observe bands associated with the vibrations of the aliphatic and aromatic C-H bonds; however, the sample contained traces of water that masked the bands present in that region due to the great intensity of the O-H vibration of water. Actinomycin corresponds to a chromopeptide composed of a heterocyclic chromophore and two cyclic pentapeptide lactone rings with potent antimicrobial and antitumor activity [32,33]. In agreement with previous references, cyclic peptides are well recognized due to their high resistance to thermal degradation [34], as demonstrated by actinomycin D when exposed to 80 °C for a prolonged period (180 min) [35,36].

Genome sequencing has allowed for the generation of large databases with sequences of over 300 Streptomyces spp. An analysis of these data showed that Streptomyces species express only a fraction of their biosynthetic genes under laboratory growth conditions [37].

The antiSMASH analysis has been used as an indirect approach to confirm the presence of the actinomycin cluster. This analysis reveals differences in terms of the number of BGCs but also in the functional assignment of the genes between both S. fildesensis INACH3013 (32 BGCs) and its sister strain, S. fildesensis So13.3 (29 BGCs). However, these differences may obey the incomplete genome of INACH3013, which also generates differences in the putative BGC assignment by antiSMASH, because most of the BGCs detected in INACH3013 are represented in the edge of the contigs of the partial genome (Table S1).

Different strains of the same Streptomyces species usually harbor between 19 and 28 putative secondary metabolite genes clusters, with at least one unique strain-specific biosynthetic gene cluster [37]. INACH3013 and SO13.3 correspond to sister strains, with a difference in % G+C of 0.02, an ANIb of 99%, an ANIm of 99.65 and an iSSDH of 96.50%. BLAST and synteny analysis between both strains revealed that actinomycin D is present in both Streptomyces fildesensis strains. Actinomycin V (structural analog of actinomycin D) was recently reported as the main compound produced by the Streptomyces strain (CMAA-1653, GenBank MH241012) and is also present in the So13.3 strain [7,8], both related to S. fildesensis. It is well known that different strains of the same species of streptomycetes can share common putative BGCs, with at least one unique strain-specific biosynthetic gene cluster as a result of the presence or absence of genes or mutations that result in distinct final structures of metabolites [37,38]. The partial chemical characterization and thermal stability features, in combination with the BCG analysis on the draft genome, reinforce the production of actinomycin-like compounds by the INACH3013 strain.

In this study, the cytotoxicity of CFS, EAE and two purified fractions, R_f = 0.65 and R_f = 0.72, were evaluated via in vitro experiments by using frequently diagnosed human colorectal adenocarcinoma (HT-29), human breast carcinoma (MCF-7), human prostatic adenocarcinoma (PC-3) and the non-tumor CoN cell line. The results indicate different degrees of cytotoxicity on tumor cell lines (Table 1), being similar and non-selective for all cell lines. Considering previous studies on Streptomyces [8,39], it is possible to confirm that the range of the cytotoxic effects of the extracts from the INACH3013 strain were similar for all cell lines. Both the EAE and the R_f = 0.72 fractions showed high toxicity against the HT-29 cell line, with EC₅₀ values of 0.1 and 0.09 µg/mL, respectively, and the CFS extract had an EC₅₀ value of 13.42 µg/mL. These results correspond to the obtained range of values of EC₅₀ when colorectal carcinoma cell lines (HCT-116) were exposed to cell extracts from the genus Streptomyces, ranging from 71.0 to 0.076 µg/mL [40]. Similar values of EC₅₀ were found in the HT-29 and MCF-7 cell lines when exposed to actinomycin V, produced by closely related strains (99.26% of 16S rRNA gene similarity to S. fildesensis) [8,39]. In fact, it has been shown that actinomycin V presents higher toxicity than actinomycin D [39,41], suggesting that R_f = 0.72 could correspond to actinomycin V, as previously described for the related S. fildesensis strain [8]. However, more detailed analyses are needed to elucidate functional groups and chemical structures because it is well known that several Streptomyces species synthesize actinomycins D, V and X2 or X2 and D simultaneously [42,43,44].

With respect to the induction of apoptosis, it is known that programmed cell death plays a key role in the survival of normal and cancerous cells [45]. In cancer cells, all pro-apoptotic stimulations are inhibited to ensure the survival and proliferation of the tumor [46]. Programmed cell death is an important target when evaluating the potential of new metabolites or anticancer extracts. Apoptosis implies morphological and biochemical changes [45] at the morphological level, including cell shrinkage, improved intercellular adhesion, nuclear condensation, nuclear fragmentation and the loss of adhesion to the extracellular matrix (Figure 2 and Figure 3). Cellular contraction and the loss of adhesion were evident under photonic microscopy, whereas under fluorescence microscopy, after applying nuclear fluorescent dye Hoechst 33342, we observed the condensation of the nucleus and the formation of apoptotic bodies (Figure 3). These changes have been reported in other studies when cells were exposed to actinomycin compounds [39,41]. Actinomycin has been described to be responsible for the down-regulation of the mRNA expression levels of N-cadherin and vimentin, inhibiting the epithelial–mesenchymal transition and consecutively suppressing the migration and invasion of cancer cells by morphological changes [39].

Actinomycin has been reported as a potent inducer of apoptosis in different cells in vitro and in vivo assays. Moreover, different pathways have been described involving a p53-dependent or independent fashion and a tumor-necrosis-factor-related apoptosis-inducing ligand (TRAIL) [10,11,41,47]. At the molecular level, apoptosis can be induced through a signaling pathway associated with the permeability of the mitochondria membrane (intrinsic pathway) that releases pro-apoptotic proteases to the cytosol that activate the caspases, which are ultimately responsible for carrying out apoptosis [48]. In our study, the EAE and the R_f = 0.65 and R_f = 0.72 extracts clearly affected the MMP and also induced the activation of caspases at different degrees (Table 2). However, only the R_f = 0.65 fraction triggered the irreversible induction of apoptosis in the PC-3 and HT-29 cell lines (Table 2), with both having higher values of caspase induction (Table 3). The results suggest that the mechanism of cell apoptosis is similar to what has been reported for the action mode of actinomycin D in human neuroblastoma SH-SY5Y cells, decreasing mitochondrial transmembrane potential and enhancing apoptosis through caspase-dependent and -independent manners [47].

Our findings suggest that the possible mechanism of cell death in MCF-7, PC-3 and HT-29, and in the non-tumor cell line CoN, is associated with this intrinsic pathway linked to the mitochondrion and the activation of caspases triggered by actinomycin.

The current information about the actinomycin biosynthetic gene cluster similarity analysis (89%) of S. fildesensis suggests the potential presence of new structurally related molecules, and further analysis must be conducted in order to elucidate its chemical structures and relationships with the antitumoral action mechanism to investigate its medicinal properties. Moreover, it is important to continue research to explore if this intrinsic pathway linked to the mitochondrion and the activation of caspases triggered by those actinomycins is operating in other cancer cell lines.

5. Conclusions

As expected, for several Streptomyces strains, the Antarctic S. fildesensis INACH3013 is able to produce two compounds (R_f = 0.72 and R_f = 0.65) that belong to the actinomycin family, as confirmed by functional groups via FTIR and chromatography analysis.

Despite both compounds displaying remarkable antimicrobial and cytotoxic activity against human cell lines, the activity of compound R_f = 0.72 was higher in comparison to the compound R_f = 0.65 (related to actinomycin D). Further research using modern techniques such as nuclear magnetic resonance (NMR) could bring new and unambiguous information about its chemical structure.

Actinomycin compounds are well known for their use as anticancer drugs and their antimicrobial activity. The present work confirms both the antimicrobial and cytotoxic activity (apoptosis mediated by caspase activation) of actinomycin compounds produced by S. fildesensis INACH3013, which is relevant not only for the discovery of novel bioactive compounds but also as a new source of actinomycin compounds to address the needs of industrial pharmaceuticals.