1 Introduction
Epidemiological studies as well as the data from preclinical research have documented that phytochemicals from dietary sources like fruits, vegetables, medicinal plants, and tea exert beneficial effects on organisms and their systematic and long-term intake (4–5 times a week for several years) is linked with reduced cancer risk (Kubatka et al., 2015; Kubatka et al., 2016a; Pourzand et al., 2016; Mao et al., 2017). Notably, the protective effects of phytochemicals are substantially associated with reducing oxidative stress and inflammation (Kubatka et al., 2021). Anticancer characteristics of phytochemicals and their derivatives are exerted by their acting as potent cell signaling modulators. Phytochemicals have exhibited significant oncostatic activities by restraining the initiation and progression of malignancy by affecting mechanisms such as cell cycle, programmed cell death, neovascularization, the activity of stem cells, and metastasis (Abotaleb et al., 2018; Liskova et al., 2019; Liskova et al., 2020a; Liskova et al., 2020b; Tesfaye et al., 2020; Brockmueller et al., 2021). Importantly, phytochemicals demonstrate potent anticancer activities via epigenetic modulation, which include chemical modifications of histones, and expression changes in microRNA, oncogenes, and tumor-suppressor genes (Zwergel et al., 2016; Samec et al., 2019a; Samec et al., 2019b; Jasek et al., 2019).
Medicinal plants have great potential due to their pleiotropic effects on the organism. Salvia officinalis L. [Lamiaceae; Salviae officinalis herba]is the typical medicinal plant in the Mediterranean and Middle East regions, which has been recently naturalized in almost the whole world. S. officinalis L. is rich in phytochemical cocktail containing essential oils based on monoterpene derivatives (camphor, cineol, borneol derivatives, thujone) and sesquiterpenes (caryophyllene); the plant commonly contains labiate tannins (rosmarinic acid, salvianolic acid), diterpenoid bitter substances carnosol derivatives), and flavonoids (cirsiliol, luteolin and apigenin derivatives, and phenolic glycosides). In traditional medicine, S. officinalis has been applied in the management of various types of pathologies, such as inflammation, rheumatism, gout, ulcers, diarrhea, tremor, paralysis, seizure, hyperglycemia, and dizziness (Ghorbani and Esmaeilizadeh, 2017). Anti-oxidant and anti-inflammatory characteristics of S. officinalis (Bonesi et al., 2017) also correlate with its anti-cancer properties that were documented by numerous studies. S. officinalis ethanol and acetone extracts demonstrated antiproliferative effects on hepatocellular carcinoma cells by evaluating cell viability, apoptosis, cell morphology, cellular ATP, and lactate dehydrogenase levels (Jiang et al., 2017). In this regard, the half inhibition concentration (IC50) values (μg/mL) of extracts from S. officinalis leaves on HepG2 and WRL-68 cells after 48 h of treatment were 64.4 and 67.1 in methanol extract and 90.0 and 87.5 in acetone extract. Moreover, S. officinalis suppressed the cell cycle and induced programmed cell death of human colorectal cell lines through the downregulation of the MAPK/ERK pathway (Jiang et al., 2017). Another study found that S. officinalis methanolic extract blocks the proliferation of lymphoma and leukemic cell lines through induced cell death (Zare Shahneh et al., 2013). More specifically, extract at 50–800 μg/mL in a dose and time-dependent manner decreased the proliferation of KG-1A, U937, and Raji cells by more than 80% (p < 0.01). IC50 values showed ascending order after 24 h treatment: KG-1A (214.377 μg/mL), U937 (229.312 μg/mL), and Raji (239.692 μg/mL) when compared to paclitaxel. Moreover, Yanagimichi et al. (Yanagimichi et al., 2021) reported that cirsiliol, luteolin, and carnosol represent significant inhibitory compounds of S. officinalis in STAT3 cell signaling in HepG2 cells. Carnosol, at 30 μM, reduced phosphorylated STAT3 by about 80% but at a dose of 90 μM downregulated also the total STAT3 level. HepG2 cells treated with luteolin at 30 μM did not show changes in phosphorylated/total STAT3 but a higher dose of 90 μM suppressed phosphorylated STAT3 in comparison with total STAT3. Cells treated with cirsiliol at 90 μM did not show a decrease in phosphorylated and total STAT3, however, authors described clear downregulation of STAT3-responsive reporter expression. In addition, S. officinalis ethyl acetate extract and its constituents (carnosic acid and carnosol) inhibited PGE2 accumulation by directly targeting mPGES-1 in a cell-free assay and thus may crucially contribute to its anti-cancer characteristics (Bauer et al., 2012). Component evaluations found that the diterpenes carnosic acid and carnosol as potential bioactive molecules decreased mPGES-1 activity with IC50 values of 5.0 μM.
The hypothesis of this study was based on the knowledge that the application of appropriate plant nutraceuticals rich in specific phytochemicals with the potential additive or synergistic mode of action can show significant antitumor effects, as our group has already demonstrated in the past (Kubatka et al., 2015; Kubatka et al., 2016a; Kubatka et al., 2016b; Kubatka et al., 2017a; Kubatka et al., 2017b; Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). This study aimed to analyze the anticancer activities of S. officinalis haulm in therapeutic (allograft) and chemopreventive breast carcinoma (BC) animal models. In the mechanistic analyses, we evaluated clinically validated cancer biomarkers of cell death, proliferation, vascularization, inflammation/oxidative stress, stem cells, and epigenetic and plasma metabolomic changes. Moreover, histopathological features of tumor specimens such as the high/low-grade carcinomas ratio, tumor necrosis ratio, and mitotic activity index were assessed. Finally, to deepen the mode of anticancer action of S. officinalis, the parameters of cell growth and cell deaths were performed using human hormone-sensitive MCF-7 and metastatic triple-negative MDA-MB-231 BC cell lines.
2 Material and methods
2.1 Rodent models
Female BALB/c mice (Velaz, Prague, Czech Republic) aged 10 weeks and weighing between 17–19 g and female Sprague-Dawley rats (Charles River Laboratories, Sulzfeld, Germany) aged 5 weeks and weighing between 125–140 g were used in the study. Rats were acclimated to a controlled vivarium environment that included a 12-h artificial light cycle, a 23°C ± 2°C temperature range, and a 40%–60% relative humidity range. The animals were fed ad libitum with a Ssniff® diet (R-Z/M-Z low-phytoestrogen; Soest, Germany) and were provided with unlimited access to water. N-nitroso-N-methylurea (NMU, Sigma, Deisenhofen, Germany) was applied to induce mammary gland cancer in rats. Specifically, the carcinogen was injected intraperitoneally (50 mg/kg as a single dosage) at 42. postnatal day. The factor significantly increasing the inducibility of mammary carcinogenesis in the Sprague-Dawley rats represents a period of carcinogen administration in early puberty during postnatal days 40–46 in comparison with the period after postnatal day 50 (Russo et al., 1990; Kubatka et al., 2002). This research method imitates premenopausal women who have a higher risk of BC etiology. A syngeneic mouse model represented the treatment BC model. To induce mammary carcinogenesis, 1 × 105 of 4T1 cells per animal (mouse mammary adenocarcinoma) was subcutaneously injected into the abdomen mammary gland area (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). Salvia officinalis L. [Lamiaceae; Salviae officinalis herba] haulm powder (Calendula, Nová Ľubovňa, Slovakia; site of origin: North-East Slovakia) administration in the allograft mouse model (treatment study) began on the day that the 4T1 cells were inoculated and lasted for 15 days. In rats, S. officinalis administration (chemoprevention study) started 1 week before carcinogenesis induction and lasted for 14 weeks consecutive weeks. In both animal models, S. officinalis was applied in (a) low concentration 1 g/kg–0.1% (w/w) (SAL 0.1) and (b) high concentration 10 g/kg–1% (w/w) (SAL 1). The lower dose of salvia (0.1% in the diet) was derived from human dosing typical for the Mediterranean diet (about 2 g per day). Due to the different pharmacokinetics and pharmacodynamics of S. officinalis in humans and rodents, we used also 10 times higher doses in the diet in both rodent studies (1%). Moreover, the higher dose of salvia was an insurance dose, since the dosing of S. officinalis through diet in rodent BC models has not been published so far, and the potential effect of salvia in rodents could have been masked in the low-dose group. The haulm was processed into the diet by “cold pelleting procedure.” Mice (n = 78) and rats (n = 75) were randomly divided into three experimental groups. (1) control group without S. officinalis in the diet (this group was set as blank control in which the background and the diet was applied with no tested sample); (2) administration of S. officinalis in the diet at a lower dose (SAL 0.1); and (3) administration of S. officinalis in the diet at a higher dose (SAL 1). Tumor incidence represents the percentage of tumor-bearing animals per group, tumor frequency was expressed as the number of tumors evaluated for all animals in the group, and the latency period was determined by the period from the administration of the carcinogen to the appearance of the first tumor in the animal.
From the fourth day following the inoculation of 4T1 cells in mice, the tumor growth (volume) was observed three times each week. The rats were palpated each week beginning the fifth week after carcinogen application to check for each mammary tumor’s presence, size, and location (palpable if tumor diameter is greater than 0.4–0.5 cm). Within 24-hour intervals, dietary intake was monitored twice in mice and four times in rats. As a result, the average daily doses of S. officinalis administered to mice and rats in each group were calculated. At the end of both experiments, rodents were euthanized by quick decapitation and mammary lesions were excised and measured (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
2.2 Histopathology and immunohistochemistry of rodent tumor samples
Each tissue sample of rodent (mouse and rat) carcinoma was routinely formalin-fixed and paraffin-embedded. In the mice model, the average tumor area (the count of all tumor areas divided by the number of tumors) and the average area of necrosis (the count of all necrosis divided by the number of tumors) of each tumor were determined. Subsequently, the necrosis ratio was defined as the ratio of the average area of necrosis to the average area of tumors. The area of the tumor and the extent of necrosis were established in the histological slide. If the tumor had a circular outline, the area was determined as πr2 (r = radius); if the tumor showed the characteristic of an ellipse, the area was determined as π x a x b (a, b = semi-axes); if the tumor showed the characteristic of a rectangle, the area was determined as a x b (a, b sides of the rectangle) and if the tumor showed the shape of a trapezoid, the area was determined as (a + c) x v/2. If the tumor reached the size of one of the high power fields (HPFs) (e.g., with ×4 objective), the area was determined based on this microscopic field. The contours of the necrosis were outlined in the histological slide and the area of the necrosis was usually determined as their extent according to the size of the HPF, or multiple of the area of the HPF (with objective 4x = 23.76 mm2, with objective 10x = 3.8 mm2), with objective 20x = 0.95 mm2). The smallest evaluable necrosis was its extent in one HPF (0.24 mm2; i.e., at magnification ×400 with a diameter of the field of view 0 .5 mm). If the necrosis was smaller than the HPF, it was evaluated as punctiform.
Tumor samples (from rats) were evaluated complexly during the microscopic examination, at first in low magnification (×40), the overall structure of the tumor was recorded, focusing on growth microarchitectural characteristics (e.g., assessment of the proportion of glandular and solid components, and necrosis for histological grading). Subsequently, detailed cellular characteristics (i.e., degree of cellular atypia and number of mitotic figures) were assessed at higher magnifications (×100 and ×200). The mitotic score (both in rats and mice) was determined by the number of mitotic figures found in 10 consecutive HPFs in the most mitotically active part of the tumor. Only identifiable mitotic figures were counted; hyperchromatic, karyorrhetic, or apoptotic nuclei were excluded. In rare cases, where the total tumor area from mice was less than the sum of 10 HPFs, the number of mitoses in five HPFs was evaluated and multiplied by two.
The criteria for standardized classification of mammary tumors were assigned for the classification of each rat mammary tumor. Rat mammary tumors were also sub-divided as low-grade or high-grade (tumor grading as an additional parameter). Also, the standard diagnostic classification method was utilized to select categorization criteria (cell atypia, solidization, index of mitotic activity, and necrosis): firstly, solidization if >30% of tumor sample shows solid growth; secondly, a high index of mitotic activity if ≥10 mitoses is observed in 10 HPFs; thirdly, necrosis if the occurrence of comedo (not infarct) was confirmed (Kubatka et al., 2012). Accordingly, high-grade carcinomas were considered tumors with ≥2 positive criteria, and low-grade carcinomas were considered tumors with ≤1 positive criterion. In mice tumor samples, the mitotic activity index and all tumor area/necrosis ratios were assessed (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
The sample selection (the most relevant section of mammary tumor in a paraffin block, including typing characteristics) for immunohistochemical analysis was based on specific criteria: the representation (substantial) of vital tumor epithelial component without any regressive changes (e.g., necrosis). The detection of markers (precisely selected for mechanistic study) was performed via an indirect immunohistochemical method on whole paraffin sections with commercially available rat-specific antibodies (Thermo Fisher Scientific, Rockford, IL, USA; Santa Cruz Biotechnology, Paso Robles, CA, USA; GeneTex, Irvine, CA, USA; Dako, Glostrup, Denmark; Boster Biological Technology, Pleasanton, CA, USA; Bioss, Woburn, MA, USA; Abcam, Cambridge, MA, USA). Immunohistochemical staining was performed according to the manufacturer’s recommendations (Autostainer Link 48/Hermes/). Each primary antibody concentration was as follows: Bax 1:200 (sc-526); Bcl-2 1:200 (sc-492); cleaved caspase-3 1:500 (catalogue no. ab2302); Ki-67 1:50 (M7248 01); VEGFA 1:150 (sc-57496); VEGFR-2 1:80 (sc-6251); MDA 1:1000 (ab6463); EpCam 1:160 (ab71916); ALDH1A1 1:500 (pa532127); CD133 1:150 (ab19898); CD44 1:200 (pa1021-2); CD24 1:200 (gtx37755); H3K4m3 1:500 (ab8580); H3K9m3 1:400 (ab8898); H4K16ac 1:200 (ab109463), H4K20m3 1:300 (ab9053). The visualization of primary antibodies (with diaminobenzidine tetrahydrochloride as a substrate) was performed using the secondary staining system EnVision, Dual Link System-HRP, (Dako North America, Carpinteria, CA, USA, cat. No. K060911). The samples with the omission of primary antibodies were assigned as negative controls. The evaluation of immunohistochemically detected expression of selected antigens was performed using a precise morphometric method. Specifically, the sections were first screened and microscopically analyzed (Olympus BX41N, magnifications of digital images ×400). The quantification of protein expression was obtained as the average percentage of antigen-positive area in standard fields (0.5655 mm2) of tumor cells’ hot spot areas. Using the morphometric method, three hot spots were analyzed per tumor sample. The morphometric analysis of digital images was performed using QuickPHOTO MICRO software, version 3.0 (Promicra, Prague, Czech Republic). The obtained values were further compared between tumor tissue specimens from experimental groups with salvia in diet (SAL 0.1 and SAL 1) and non-treated (control) tumor tissue specimens. At least 60 tumor samples were analyzed for each marker (960 tumor slides for 16 markers) (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
2.3 Analysis of miRNA expression
Total RNA was isolated from tumor tissues using a commercial miRVana microRNA isolation kit (Thermo Fisher Scientific, Waltham, MA, USA), which is in detail described in the supplementary protocol. Subsequently, the NanoDrop ND-2000 spectrophotometer (Thermo Scientific, Wilmington, Delaware, USA) was utilized for the RNA quantification, followed by reverse transcription via TaqMan Advanced miRNA cDNA Synthesis Kit (Applied Biosystems, Life Technologies, Carlsbad, CA, USA). The cDNA samples were kept at −20°C until further analyses. Quantitative real-time PCR was carried out using miRNA-specific TaqManTM advanced miRNA assays kit (Applied Biosystems, Life Technologies, Carlsbad, CA, USA) for the tumor-suppressive miRNA (miR-22, miR-34a, miR-145) and the oncogenic miRNA (miR-21, miR-155, and miR-210). MiR-191-5p was chosen as an internal control to normalize the quantities of cDNA in the samples by relevant publications and protocol recommendations. After that, quantitative real-time PCR was performed using the AB7500 real-time system from Applied Biosystems, Life Technologies, Carlsbad, California, USA. Each qPCR reaction was utilized in duplicates and the resultant Cq values were then averaged (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). Fold changes were calculated in Excel using the standard ∆∆Cq method.
A list of catalogue numbers and specific ID’s of used miRNA assays.
| Assay’s name |
Catalog numbers |
Specific ID |
| rno-miR-155-5p |
A25576 |
002571 |
| rno-miR-145-5p |
A25576 |
002278 |
| rno-miR-21-5p |
A25576 |
000397 |
| rno-miR-210-3p |
A25576 |
000512 |
| rno-miR-22-5p |
A25576 |
002301 |
| rno-miR-34a-5p |
A25576 |
000426 |
| hsa-miR-191-5p |
A25576 |
002299 |
2.4 DNA extraction and bisulfite conversion
Homogenization and disruption of fresh frozen tissue were provided using the mechanical disruptor TissueLyser LT (Qiagen, Germany) designed for low-to-medium disruption of different types of tissue. In the first step, we placed a tissue sample (∼100 mg) and a 5 mm stainless steel bead into a precooled 2 mL round bottom tube. Subsequently, animal tissue was homogenized in 200 mL lysis buffer at 50 Hz for 1 min. In the next step, we added 20 proteinase K for the digestion of the sample and incubated it at 56°C overnight. Purification of total DNA from animal tissue was mediated using a DNeasy blood and tissue kit (Qiagen, Germany) according to the manufacturer’s protocol. The concentration of the samples was determined by Qubit™ 3.0 fluorometer (Thermo Fisher Scientific) with Qubit dsDNA BR assay kit (Thermo Fisher Scientific). Subsequently, isolated DNA (at a concentration of at least 50 ng/μL) was bisulfite-treated using an EpiTect bisulfite kit (Qiagen, Germany) according to the manufacturer’s protocol (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
2.5 CpG assays and determination of methylation status by pyrosequencing
For the pyrosequencing analysis of selected regions of target genes: PTEN (6 CpG sites), TIMP3 (3 CpG sites), RASSF1A (3 CpG sites), ATM (4 CpG sites), and PITX2 (5 CpG sites), we used commercially available CpG assays (PyroMark CpG Assay, Qiagen, Germany). Primer sequences are available in the supplementary material of the manuscript. Briefly, bisulfite-converted DNA was used as a template and amplified using a PyroMark PCR kit (Qiagen, Germany). PCR conditions were as follows: initial denaturation of DNA for 15 min at 95°C, 45 cycles (96°C for 30 s, 56°C for 30 s, and 72°C for 30 s) followed by final extension at 72 °C for 10 min. PCR products were visualized using gel electrophoresis (1.75% agarose gel). The PCR products were further processed according to the manufacturer’s protocols and analyzed by PyroMark Q96 ID System (Qiagen, Germany) and PyroMark Gold Reagents (Qiagen, Germany). PyroMark Q96 software version 2.5.8 (Qiagen, Germany) automatically calculated the methylation level of selected CpG dinucleotides (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
A list of catalogue numbers and specific ID’s of used assays.
| Assay’s name |
Catalog numbers |
Specific ID |
| Rn_Pten_03 |
978746 |
PM00450450 |
| Rn_Npat_02 |
978746 |
PM00592487 |
| Rn_Pitx2_08 |
978746 |
PM00519141 |
| Rn_Timp3_06 |
978746 |
PM00574896 |
| Mm_Rassf1_G2_PM |
978746 |
PM00416297 |
2.6 Serum cytokine levels in rats
Blood samples (after clot formation) were centrifuged at 2,000 x g for 10 min. Consequently, serum samples were collected and evaluated for the following cytokines: IL-6, IL-10, TNF-α, and TGF-β. The levels of cytokines were analyzed using ELISA in vitro kits (Abcam, Cambridge, MA, USA) designed for quantitative measurement of cytokines in rat serum: Rat IL-6 in vitro ELISA Kit (ab234570), Rat IL-10 in vitro ELISA Kit (ab214566), Rat TNF-α in vitro ELISA Kit (ab236712), and Rat TGF-β in vitro ELISA Kit (ab119558) (Molitorisova et al., 2021).
2.7 Plasma metabolome profiling in tumor-bearing rats
EDTA-coated tubes were used to collect blood samples, which were centrifuged at 4°C and 2,000 rpm (380 g-forces) for 20 min. Consequently, plasma was stored at −80°C until further analysis. Plasma denaturation (Nagana Gowda et al., 2015) and further processing were carried out as follows: data acquisition, handling, and evaluation were performed according to the study by Baranovicova et al. (Baranovicova et al., 2018). We used plasma from tumor-bearing rats, only a few samplings were omitted due to insufficient amount, hemolysis, or blood clotting. The usual labeling of BCAAs for branched-chain amino acids -leucine, isoleucine, and valine, further BCKAs for their 2-oxoderivates, also branched-chain keto acids - ketoleucine (2-oxoisocaproate), ketosoleucine (3-methyl-2-oxopentanonate), and finally ketovaline (2-oxoisovalerate) was applied (Baranovicova et al., 2018).
2.8 Cell lines, cell cultures, and experimental design
BC cell lines MCF-7 (ER+, PR+, HER2−) and MDA-MB-231 (ER−, PR−, HER2−) (both lines: American Type Culture Collection, ATCC; Manassas, VA, USA) were used in all in vitro experiments. In addition, a non-cancerous MCF-10A (human mammary gland epithelial cells) were used as a normal breast epithelial model. MCF-7 cells were maintained in DMEM medium (GE Healthcare, Piscataway, NJ, USA), MDA-MB-231 in RPM1 1640 medium (Biosera, Kansas City, MO, USA), and MCF-10A in DMEM F12 medium (Biosera, Kansas City, MO, USA) + suppl. insulin, EGF- epithelial growth factor and HC-hydrocortisone (all Sigma, Steinheim, Germany). The 10% FBS (Fetal bovine serum, Gibco) and antibiotic/antimycotic solution (1× HyClone™; GE Healthcare, Chicago, IL, USA) were used to supplement culture media. Basic cultivation conditions were a 5% CO2-containing atmosphere, humidified air, and 37°C temperature. A trypan blue exclusion test was used to estimate the viability of used cells (≥95%). For the flow cytometry experiments, the MCF-7 (3 × 105) and MDA-MB-231 (1 × 105) cells were seeded in Petri dishes. After 24 h seeding and initial colony growth in a complete cultivation medium, the cells were treated with the SPGE (Calendula, Nová Ľubovňa, Slovakia) for 24, 48, and 72 h before analysis (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
2.9 Cytotoxicity assay
The cytotoxic effects of SPGE and propylene glycol (concentration range 10–1,000 μg/mL) on MCF-7, MDA-MB-231, and MCF-10A cells were analyzed by resazurin metabolic assay. The 10 µL of the final concentration of 40 µM resazurin solution (Sigma) was added to each well after 72 h of incubation with SPGE and propylene glycol. After 1 h metabolization, the fluorescent product resorufin was analyzed and fluorescence was measured by the automated CytationTM 3 cell imaging multi-mode reader (Biotek, Winooski, VT, USA) at 560 nm excitation/590 nm emission filter set. The results were expressed as a fold of the control in which the fluorescence of the control wells was taken as 100%. The numbers of experimental repeats were set (n = 3) (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
2.10 5-Bromo-20-deoxyuridine (BrdU) cell proliferation assay
Cell proliferation was analyzed by the BrdU incorporation ELISA assay following the manufacturing instructions (Roche Diagnostics GmbH, Mannheim, Germany). The incorporation of BrdU into genomic DNA during cell growth was quantified and analyzed after SPGE extract treatment (10–1,000 μg/mL) in our experiments. After the substrate degradation, the color intensity changes were read at 450 nm (reference wavelength: 690 nm) by CytationTM 3 Cell Imaging Multi-Mode Reader (Biotec). The numbers of experimental repeats were set (n = 3). Based on Resazurine and BrdU assays, the final IC50 was calculated and determined: 160 μg/mL (MCF-7 cells) or 130 μg/mL (MDA-MB-231 cells) (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
2.11 Flow cytometry analyses protocol
The floating and adherent cells (MCF-7 or MDA-MB-231) were harvested and mixed for every sample after 24, 48, and 72 h following the treatment (SPGE extract final dilutions 160 or 130 μg/mL). Then, the samples were centrifuged, washed in PBS, and again resuspended in PBS according to the following staining protocol. For intracellular staining, cells were fixated by 4% paraformaldehyde (15 min), followed by washing and permeabilization with 0.5% TWEEN for 15 min at room temperature. The samples were divided for particular staining as is listed in the table below. Cells were stained for 15–30 min in the dark at room temperature. After a washing step (PBS), samples were resuspended in PBS, and the FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA, USA) was used for fluorescence analyses (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
A list of flow cytometry analyses and staining.
| Analyses |
Staining solution |
Company |
| Cell cycle * |
Triton X-100 10% solution ribonuclease A 0.5 mg/mL propidium iodide–PI 0.025 mg/mL In 500 µL PBS |
Sigma-Aldrich, Steinheim, Germany |
| Apoptosis |
Annexin V-Alexa Fluor 647 1:100 (cat. no. A23204) |
Thermo Scientific, Rockford, IL, USA |
| PI (5 mg/mL) 1:500 |
Sigma-Aldrich |
| Mitochondrial membrane potential |
TMRE (tetramethylrhodamine ethyl ester per chlorate) final conc. 0.1 µM |
Molecular Probes, Eugene, OR, USA |
| Caspase activation |
Cleaved caspase-3 rabbit mAb PE conjugate 1:100 (#9978) |
Cell Signaling, Danvers, MA, USA |
| Cleaved caspase-7 rabbit mAb PE conjugate 1:100 (#42542) |
| Proteins analysis |
Cleaved PARP rabbit mAb PE conjugate 1:100 (#8978) |
| Bcl-2 mouse mAb PE conjugated 1:100 (#26295) |
| Phospho-Bcl-2 (Ser 70) rabbit mAb Alexa Fluor 488 conjugate 1:200 (#2834) |
*After harvesting, the cold 70% ethanol was used to fix cell suspension and samples were kept at −20°C overnight.
2.12 The examinations of plant secondary metabolites in salvia extract
The propylene glycol extract of S. officinalis L. (Lamiaceae) herb solvent (Batch No. S-02-01-11-10-21; Calendula a.s., Nová Ľubovňa, Slovakia) was used for the following studies. The extract was produced in an extraction reactor by extracting haulm of S.officinalis in the extraction agent propylene glycol at a temperature of 80°C. After cooling, the mixture was filtered and the liquid extract was packed in shipping containers. Analytical HPLC measurements were obtained on an Agilent 1260 chromatographic system (1260 Vial sampler G7129A, 1260 Quat Pump G7111B, 1260 MCT G7116A, 1260 DAD HSG7117C, Agilent Technologies, Waldbronn, Germany) with MS AB SCIEX Triple Quad 3500 system (Framingham, USA). InfinityLab Poroshell 120EC-C18 (4.6 × 100 mm, 2.7 μm) column with InfinityLab Poroshell 120EC-C18 (4.6 × 5 mm, 2.7 μm) guard column were used, with a gradient elution of A: MeOH with 0.1% HCOOH (v/v) and 1 mM HCOONH4, and B: water with 0.1% HCOOH (v/v) and 1 mM HCOONH4; A:B 0 min 10:90 (v/v), in 36th min 100% A, in 50th min 100% A. The flow rate was 0.3 mL/min, column block temperature was 30°C. MS conditions: curtain gas N2 25 L/min, temperature 450°C, gas 1 50 L/min, gas 2 40 L/min, ion spray voltage −4500 V, scan mode m/z 50-1,000, scan rate 1,000 Da/s, solvent delay time 4 min compounds were identified by comparison of retention times, UV, and MS profile with standards. The quantification was carried out by using calibration curves constructed based on measurements of corresponding standards. Analytical GC/MS analysis was carried out the same as described previously (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
2.13 Statistical analyses
Data from rodent studies are shown as mean ± SEM. The one-way analysis of variance (ANOVA), Student’s t-test, and Mann–Whitney test were used as statistical methods in data analysis. The tumor volumes were counted by the formula: V = π × (S1)2 × S2/12 (S1, S2 are tumor diameters; S1 < S2). Data from in vitro studies are shown as mean ± SD and were analyzed by ANOVA test followed by the Bonferroni multiple comparisons test. In addition, we used ANOVA and Student–Newman–Keuls multiple comparison tests in the fluorescence assay analysis (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). Differences between groups were considered significant when p ≤ 0.05. In the mentioned data analyses, GraphPad Prism Comparison Software (version 5.01, La Jolla, CA, USA) was used.
4 Discussion
Meta-analyses of clinical trials assessing regular consumption of plant foods and/or plant-derived molecules on the BC risk point to beneficial effects in the population (Farvid et al., 2021; Poorolajal et al., 2021). Other meta-analyses revealed that dietary management in BC patients based on vegetable and fibre intake provides useful lifestyle intervention against BC development (Wang et al., 2022). Despite constant progress in the advancement of anti-cancer strategies, the development of cancer resistance remains the main reason for BC-related mortality. In this regard, comprehensive preclinical research documented that targeting specific molecular pathways by phytochemicals to enhance therapeutic response by reversing the resistance and/or increasing sensitivity of cancer cells towards currently used therapeutic procedures could represent the perspective way in the management of BC (Liskova et al., 2021). To find novel medicinal plants with clinically beneficial anti-cancer activities, we evaluated the oncostatic potential of S. officinalis in therapeutic and preventive rodent models of BC and human cell BC lines.
The doses of salvia administered in the diet in both types of selected rodent models were based on our previous rich experience with the application of plant nutraceuticals in food (Kubatka et al., 2017b; Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). The dose of salvia in the SAL 0.1 group (0.1% in the diet) is similar to the human dosing per day typical for the Mediterranean diet. Moreover, due to the supposed different pharmacokinetics and pharmacodynamics of phytochemicals present in salvia between humans and rodents, we used also a ten times higher dose, Data from this study showed significant anti-cancer activities of S. officinalis in rodent BC models. In the mouse 4T1 syngeneic study, a non-significant (but apparent) decline in tumor volume in mice after salvia treatment was associated with a significant reduction in necrosis ratio within the tumor sample and mitotic activity index. Our recent experiments using the same murine model and doses of dietary administered Rhus coriaria L. fruit peels, Cinnamomum zeylanicum L. bark, and Thymus vulgaris L. haulm showed significant therapeutic effects after comparison with control animals (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). The mentioned anti-cancer effects of plants used as medicinal plants or spices found in our laboratory were comparable with the data gained with conventional synthetic drugs analyzed using the same 4T1 BC murine model (Demečková et al., 2017; Solár et al., 2017; Grasselly et al., 2018). Even though the most sensitive parameter of experimental carcinogenesis, tumor frequency, was not changed, the rat BC model of primary chemoprevention in this study revealed a significant lengthening of tumor latency after higher salvia dose and significant shifted high-grade mammary carcinomas to low-grade lesions after application of both salvia doses when compared to controls. Simultaneously, no parameter was negatively affected by S. officinalis, which is a result demonstrating the chemopreventive effect of salvia in the BC model. Using the same rat model and similar doses of natural substances, significant chemopreventive effects and improvement in histopathological characteristics of tumors after an application of dark fruits, various medicinal plants, or spices were found in our laboratory (Kubatka et al., 2015; Kubatka et al., 2016a; Kubatka et al., 2016b; Kubatka et al., 2017a; Kubatka et al., 2017b; Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b) and by other authors (Singletary et al., 1996; Ravoori et al., 2012; Jeyabalan et al., 2014; Bishayee et al., 2016).
Based on our LC-DAD-MS analysis, the dominant metabolites present in SPGE are phenolic derivatives represented by rosmarinic acid and triterpenes represented by ursolic and oleanolic acids. Organic Anion Transporters are the major transporters that may modulate the pharmacokinetics of rosmarinic acid (Kang et al., 2021). In addition, intestinal transport of rosmarinic acid was mainly realized by conjugated forms with glucuronic acid or sulfate across Caco-2 cells through passive diffusion (Woottisin et al., 2022). PXR-ABC drug transporters are responsible for the membrane transport of CYP-mediated ursolic acid in vivo and in vitro (Jinhua et al., 2020). The research indicates that oleanolic acid monoglucoside is transported through the cell membrane by an active, carrier-mediated mechanism; on the other hand, the transport of monoglucuronide is a passive, carrier-dependent process (Szakiel and Janiszowska, 2002). These compounds could be assigned as responsible for the observed effects of salvia extract in both animal and cell models. Anticancer activities of rosmarinic acid were described and reviewed repeatedly, including effects in models based on BC cell lines and animals (Sharma et al., 2022; Zhao et al., 2022). As described previously, the bioavailability of triterpenes was analyzed by numerous studies (Furtado et al., 2017) and several studies also analyzed the bioavailability of phenolic acids, such as rosmarinic acid (Guan et al., 2022). However, the direct bioavailability of rosmarinic acid is relatively low and metabolic degradation producing smaller compounds like ferulic, caffeic, and 3,4-dihydroxyphenyl lactic acids (Hitl et al., 2021) was described. Numerous research data concluded higher tumor-suppressive effects of natural phytochemical cocktails occurring in whole foods intake when compared to single phytochemicals (Kapinova et al., 2017). It seems logical that synchronous targeting of multiple signaling molecules within the cell associated with carcinogenesis via numerous bioactive substances present in medicinal plants may represent a more effective clinical approach in cancer management when compared to monotherapy. As far as we know, there is no data on the adverse effects associated with long-term S. officinalis administration during its usage in humans for many centuries (Tildesley et al., 2005). The usage of S.officinalis in common doses within the Meditteranean diet is very safe; however, there might be undesirable side effects in excessive amounts, which can be attributed to the high content of thujone (Walch et al., 2011). However, the application of data gained from preclinical research to cancer patients is a very difficult task for oncologists. In this regard, translational research must solve several important issues crucial for the use of medicinal plants in clinical practice: (1) pharmacokinetic parameters of individual phytochemicals/metabolites; (2) sufficient levels of phytochemicals/metabolites in human plasma (promising method seems to be nanotechnology and advanced drug formulations); (3) effective and safe dosing; (4) combined application with the conventional drugs to increase the effects of anticancer therapy; (5) re-sensitizing of the chemotherapy/radiotherapy-resistant cancers; (6) impact of phytochemicals on cell signaling pathways to affect specific stages of carcinogenesis such as cell invasion/metastasis or disease relapse; (7) understanding the individual characteristics of patients within the multi-omics data integration for a more preventive, predictive, and personalized medical approach using plant medicinal foods.
Oncostatic characteristics of plant medicinal foods are well evidenced to be based on usual mechanisms of action such as programmed cell death, effects on cell cycle/proliferation and angiogenesis, or antioxidant activities that are induced by specific mixtures of secondary metabolites present in specific plants. Apoptosis represents a strictly organized process that is controlled by several cellular proteins and signaling pathways (Abotaleb et al., 2018; Abadi et al., 2022). Pro-apoptotic proteins are classified into caspases and Bcl2 family molecules. Caspases as cysteine proteases play a crucial role in apoptosis initiation. Caspases are usually present in the cell as inactive heterodimers - procaspases. Regarding the role of Bcl2 proteins in apoptosis, under stress cellular conditions these molecules are overexpressed to interact and sequester the pro-apoptotic Bax and Bak proteins and thus preserve the integrity of the mitochondrial membrane. Apoptosis can be mediated via two signaling pathways, intrinsic and/or extrinsic. The intrinsic pathway is modulated by the mentioned molecules of the Bcl2 family, whereas increased levels of the Bax/Bcl2 heterodimer activate the caspase-3 and subsequently initiate apoptotic changes in cancer cells (Abotaleb et al., 2018). Numerous research data described significant regulatory effects of phytochemicals in Bax/Bcl-2/caspase-3 signaling on cancer (Khan et al., 2021; Liu et al., 2021; Salamatullah et al., 2021). This study documented a significant rise in the Bax/Bcl-2 ratio in rat BC specimens in vivo after the treatment with both salvia doses. This result correlated with the upregulated expression of cleaved caspase-3 in the group with a higher salvia dose. We found similar data also in our previous experiments using the same rat model in which we demonstrated a significant correlation between caspase-3 expression and an increase in Bax/Bcl-2 ratio after the chemoprevention with dark fruit, sumac peels, oregano haulm, clove buds, and cinnamon bark (Kubatka et al., 2016b; Kubatka et al., 2017a; Kubatka et al., 2017b; Kubatka et al., 2020a; Kubatka et al., 2020b). The pro-apoptotic activities of salvia extract in this study were also assessed in our parallel evaluation using MCF-7 and MDA-MB-231 BC cells in vitro. In this regard, we analyzed early and late apoptosis mechanisms. The data indicated that SPGE treatment damaged the mitochondria membrane’s normal state in early events accompanied by release and a decrease in Bcl-2 expression. Alongside the increased Bcl-2 phosphorylation, the loss of MMP occurred very early. It is known that mitochondrial dysfunctions induce apoptosis and loss of MMP or release of anti- and pro-apoptotic mitochondrial proteins (Bcl-2, Bad, Bax, cytochrome c) may be the first feature of intrinsic cell death (Wang and Youle, 2009). Furthermore, we observed late apoptosis markers, i.e., activated caspase-3 and -7, and cleavaged PARP protein after salvia extract treatment in both BC cell lines. The similar apoptotic manifestation was found also after treatment with different natural extracts performed in our laboratory (clove buds, thyme, cinnamon, or sumac) (Kubatka et al., 2017b; Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b).
Isolated phytochemicals or medicinal plants demonstrate a significant impact on the division of BC cells (Adams et al., 2010; Han et al., 2021; Shehata et al., 2021). In this regard, phytochemicals target numerous signaling pathways and mechanisms such as PI3 kinase, Nrf2, COX-2, NF-κB, poly-ADP-ribosylation, Plk1, STAT3, Hedgehog, Wnt, or epigenetic changes that are included in the modulations of cell cycle and proliferation (Wang et al., 2012). The immunohistochemical analysis of the expression of nuclear protein Ki67, a highly valid proliferation marker, revealed a boundary-significant dose-dependent decline in rat tumor samples of both treated groups by 20.5% and 21.5% vs. control tumors. Moreover, salvia in both doses markedly decreased the mitotic activity index and necrosis ratio (only in higher doses) in 4T1 tumors. In addition, using resazurin viability assay and BrdU analysis, we found similar data as in rodent studies in vivo. SPGE declined the cell viability, metabolic activity, and proliferation in both human BC cell lines. The blocking of the cell cycle by salvia extract was confirmed by the G2/M phase arrest and subG0/G1 sub-population increase in MCF-7 and MDA-MB-231 BC cells. The different effect of S. officinalis on ER+ (MCF-7) and TNBC (MDA-MB-231) cells in vitro was partially manifested. Triple-negative BC cells were more resistant than ER+ BC cells, and we found a slightly stronger G2/M block, which is related to a time shift in other monitored parameters. It is well known that persistent or transient G2/M arrest is a prerequisite step for apoptosis and proliferation inhibition elicited by varying doses of substances with anticancer potential (Shu et al., 1997). Targeting the S and G2 checkpoints is also attractive for cancer therapy because the loss of G1 checkpoint control is a common feature of cancer cells, making them more reliant on the S and G2 checkpoints to prevent DNA damage–triggered cell death.
However, our previous studies analyzing the anticancer activities of Chlorella pyrenoidosa L. and Thymus vulgaris L. haulm showed clear discrepancies in markers of cell proliferation between in vitro and animal cancer models (Kubatka et al., 2015; Kubatka et al., 2019). The reason for such conflicting results and different therapeutic outcomes using mentioned medicinal plants is the heterogeneous genotype/phenotype of cancer cells in human lines and NMU-induced rat adenocarcinomas.
Several hypotheses have been described regarding the key role of tumor angiogenesis in tumor development and metastasis, while preventive/therapeutic inhibition of angiogenesis may represent an outstanding clinical approach in the management of carcinogenesis. Oncology research described numerous candidate molecules including phytochemicals that effectively affect the neovascularization linked with cell signaling molecules such as VEGF, FGF, EGF, or HGF (Niland and Eble, 2019). In addition, multiple factors such as tyrosine kinase receptors, integrins, and neuropilins interact with VEGF/VEGFR signaling and induce cancer initiation. Numerous bioactive plant-derived molecules can bind with VEGF-stimulating factors and suppress them, thereby inhibiting cancer development (Parveen et al., 2019). Salvia in our study did not modulate either VEGF nor VEGFR-2 expression levels in rat tumor samples. Using the same model, sumac in our previous study showed the same result. Conversely, in our two other recent animal studies, we observed a downregulation in VEGFR-2 expression after thyme (Kubatka et al., 2019) and VEGF expression after cinnamon treatment in rat tumor tissues (Kubatka et al., 2020a). Significant anti-angiogenic effects of natural cocktails of phytochemicals present in medicinal plants were demonstrated in our older rodent cancer studies (Kubatka et al., 2016b; Kubatka et al., 2017a; Kubatka et al., 2017b). However, comprehensive analyses on the effect of single phytochemicals/plant foods on the mechanisms of anti-cancer action and complex cell signaling within the process of angiogenesis and vasculogenesis are strongly warranted.
S. officinalis, through a high flavonoid and phenolic acid content, demonstrates effective anti-inflammatory and antioxidant characteristics (Kolac et al., 2017). Chronic, non-physiologic inflammation is associated with the pathophysiology of numerous diseases via inducing oxidative stress and consequent cellular damage. We have found a significant decline in serum TGF-β cytokine levels and borderline significant decreases in serum IL-6 and TNF-α cytokine levels. TGF-β is a multifunctional cytokine that behaves as a tumor promoter or tumor suppressor in dependence on a cellular stage and content. Scientific data showed that TGF-β is actively produced by tumor cells (mainly in the later stages of carcinogenesis) and contributes to cell growth, migration, and metastatic spread, and in addition, represses host-tumor immune responses (Kim et al., 2021). Similarly, TNF-α stimulates BC growth through the positive feedback loop of TNFR1/NF-κB (and/or p38)/p-STAT3/HBXIP/TNFR1 (Cai et al., 2017). Despite the knowledge that TNF-α supports apoptosis, the knock-out of this cytokine can upregulate different pro-apoptotic signaling in certain BC subtypes that regulate specific molecular mechanisms that direct the cancer cells toward death (Pileczki et al., 2012). Our results support the ambivalent effects of TNF-α on cancer apoptosis (Kokolakis et al., 2021). Besides, circulating inflammatory cytokines, e.g., TNF-α, IL-6, and IL-10 are discussed as potential prognostic oncomarkers in BC patients due to their contribution to cancer proliferation and invasion (Wang and Yang, 2017; Ahmad et al., 2018). The anticancer activities of S. officinalis in vivo and in vitro associated with cytokine downregulation in this study, point to a rationale for cytokine blockade in cancer research and clinical practice. In this regard, further evaluation of anti-inflammatory drugs, including phyto-substances, in cancer chemoprevention and treatment is needed.
Alterations in cardinal cellular biomolecules, i.e., DNA, proteins, and lipids may be induced by imbalanced redox status in the cell. Such changes are principally associated with the processes of carcinogenesis (Günes et al., 2020). Antioxidant potential of plant-derived substances may play a prominent role in the prevention of cancer via the stable preservation of the mentioned cellular biomolecules. Comprehensive data from our laboratory demonstrated that whole plant foods/substances (applied dietary in relatively low doses), such as young barley leaves, dark fruit peels, clove buds, thyme haulm, cinnamon bark, and sumac fruit, significantly suppressed oxidative damage of lipids or proteins in cancer cells in vivo (Kubatka et al., 2016a; Kubatka et al., 2016b; Kubatka et al., 2017b; Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). According to the ORAC ranking, salvia belongs among plants with the highest antioxidant capacity (Tundis et al., 2020). With this presumption, we have assessed cellular MDA levels (a well-established marker of oxidative damage of lipids) after salvia treatment. Our data confirmed that salvia at a higher dose decreased MDA levels in tumor cells by 30.5% vs. control. Based on our outcome, we believe that the antioxidant mode of action (geno-protection) typical for salvia represents one of several mechanisms of its anticancer effect found in this study.
Recent scientific data precisely described the role of phytochemicals in targeting specific signaling pathways, e.g., Notch, Wnt, or Hedgehog and via this mechanism suppress the vitality of cancer stem cells (CSC) (Liskova et al., 2019; Deldar Abad Paskeh et al., 2021). CSCs play an important role in cancer development by impacting several processes such as apoptosis and the development of drug resistance (Yang et al., 2020). Well-established parameters of BC CSCs include overexpression of CD44+/CD24−/low phenotype, EpCAM, CD133, ALDH1, and nestin; all these markers correlate with poorer BC prognosis (Shima et al., 2017). Importantly, CD24, CD44, CD133, EpCAM, and ALDH1 have been shown as valid BC CSCs markers in chemocarcinogen-induced cancer in rodents (Rennó et al., 2015). Salvia in this study showed apparent anti-CSCs activity by reducing expressions of EpCam and ALHDH1 proteins in tumor cells in vivo. Previously, we have confirmed the anti-CSC effects of oregano, clove buds, thyme, cinnamon, and sumac in the same animal cancer model (Kubatka et al., 2017a; Kubatka et al., 2017b; Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). We assume that the anti-CSCs activities of mentioned medicinal plants are affected via the regulation of multiple cell signaling pathways and mechanisms induced by specific mixtures of phytochemicals present in medicinal plants. It seems that the natural cocktails of bioactive molecules demonstrate better oncostatic potential (including the anti-CSC activities) in comparison with the single phytochemicals (Kapinova et al., 2017). In addition, scientific research has revealed anti-CSC effects of plant metabolites not only in BC but also in numerous other cancer types (Liskova et al., 2019). These data demand the need for in-depth analyses and the consequent translation to oncology practice (Naujokat and McKee, 2020).
Abnormal epigenetic modifications including histone chemical alterations, miRNA-modulated post-transcriptional changes, and promoter DNA methylation can alleviate crucial tumor suppressor genes for signal transducers, transcription factors, nuclear receptors, apoptosis-inducing and DNA repair gene products, and cell cycle regulators and thus markedly contribute to initiation and development of cancer (Thakur et al., 2014; Varghese et al., 2020). Aberrant posttranslational chemical modifications of histones were described as factors included in the pathogenesis of numerous types of chronic diseases. In this regard, they could be used as valid prognostic and predictive markers in clinical practice, including oncology (Romanowska and Joshi, 2019). Targeting the histone-modulating enzymes by single phytochemicals or their specific mixtures may provide a novel and progressive medicinal approach to the management of chronic diseases. In this study, salvia in both doses significantly decreased H3K4m3 levels and elevated H4K16ac in rat tumor specimens. Using the same BC model, we revealed beneficial alterations in histone chemical modulations in cancer cells after chronic application of clove buds (Kubatka et al., 2017b), thyme (Kubatka et al., 2019), cinnamon (Kubatka et al., 2020b), and sumac (Kubatka et al., 2020a). In the study of other authors, resveratrol induced cytotoxicity in MDA-MB-231 and MCF-7 cells via declining the H4R3me2s and H3K27me3 levels and increasing H3K9ac and H3K27ac levels (Chatterjee et al., 2019). These data correlated with an upregulated BRCA1, p53, and p21 expression. Another in vitro study revealed reduced levels of histone deacetylase in MCF-7 and MDA-MD-231 lines after the combined application of sulforaphane and withaferin A. Moreover, upregulated histone methylations were associated with cytotoxic effects, apoptosis, and blocking of the cell cycle in BC cells (Royston et al., 2017). Other authors found that epigallocatechin gallate (EGCG) reactivated silenced tumor suppressor genes via decreasing histone deacetylase activity and increasing the acetylation in lysine 9 and 14 on H3 and lysine 5, 12, and 16 on H4 and via decreased methylation of lysine 9 on H3 (Nandakumar et al., 2011). Even though our study did not reveal the precise mechanisms by which salvia modulates posttranslational histone modifications in used BC model in vivo, the comprehensive preclinical and clinical data point to dietary phytochemicals as a progressive tool to control cancer (Uramova et al., 2018; Samec et al., 2019a). A deeper knowledge of the mechanisms linked with the posttranslational histone modifications may disclose new molecular targets for plant-derived substances, therefore, further research is warranted.
MiRNAs play a crucial role in the modulation of gene expression; therefore, miRNAs are substantially involved in numerous physiological and pathophysiological processes including carcinogenesis. MiRNAs act as either potential oncogenes or tumor-suppressor genes; thus, in many cases, their role in the pathogenesis of BC and clinical utility remains disputable (Zhang et al., 2022). Prognostic miRNA data are crucial for answering very complex questions concerning specific BC patient outcomes. Specific miRNAs have been introduced as valid biomarkers in BC management including miR21, miR155, miR210, miR22, miR34a, and miR145. Specifically, miR21 and miR155 function as oncogenic miRNAs, on the other hand, miR22, miR34a, and miR145 are documented as tumor suppressors in breast carcinogenesis. MiR210 is characterized as oncogenic under hypoxic conditions in the cell (Qin et al., 2014). This study demonstrated significant decreases in oncogenic miR21 and tumor-suppressive miR145 and an apparent (but non-significant) decrease in oncogenic miR155 expression when compared to controls. Significant effects of T. vulgaris, C. zeylanicum, and R. coriaria on the expression of specific miRNAs in the same BC model were found also in our recent studies (Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). Based on not completely coherent data within the comprehensive BC research (including data from our laboratory), robust clinical and preclinical validation focused on a deeper understanding of molecular mechanisms is needed to precisely define the role of specific miRNA signatures and particular miRNAs in BC and its specific subtypes in the context of disease diagnosis, prognosis, and preventive strategies (Zografos et al., 2019).
Evaluating the methylation of the patterns of TSG promoter regions using standardized CpG islands in rat tumor cells in vivo treated with salvia showed a significant decline in ATM and PTEN promoters. Considerable, but non-significant decrease in six CpG islands of the TIMP3 promoter region after salvia treatment was found in this study. We have documented similar data using the same chemopreventive model of BC. Natural bioactive metabolites present in Syzygium aromaticum, T. vulgaris, C. zeylanicum, and R. coriaria initiated significant downregulation in the methylation status of clinically validated TSG promoters (clove: RASSF1; thyme: ATM, RASSF1, PTEN, and TIMP3; cinnamon: ATM and TIMP3; sumac: ATM, PTEN, and TIMP3 (Kubatka et al., 2017b; Kubatka et al., 2019; Kubatka et al., 2020a; Kubatka et al., 2020b). As we documented in our laboratory, demethylating plant-derived substances (as a cocktail of bioactive secondary metabolites in whole plants) by affecting the activity of DNA methyltransferases can modulate specific gene expression by reversing the abnormal epigenetic changes that are common cancer inducers. Future oncology research must be unambiguously aimed toward the clinical sphere, applying a more personalized approach to appoint the beneficial role of plant-derived substances as significant epigenetic modulators in cancer or high-risk individuals (Jasek et al., 2019).
Cancer, comprising consecutive genetic and epigenetic modifications, leads to multiple aberrant metabolic routes in the cells and tissues. Comprehensive scientific data supports the potential of phytochemicals to meddle with several key metabolic routes, e.g., tumour glycolysis and the modulation of hypoxia (Pralea et al., 2022; Samec et al., 2023). Zhang et al. highlighted the relevance of BCAA metabolic reprogramming in BC. The authors showed that the proliferation of BC cells through mTOR-modulated mitochondrial biogenesis is stimulated by BCAT transaminase 1, which catalyzes the conversion of BCAAs to BCKAs (Zhang and Han, 2017). In our study, after treatment of animals with induced BC by S. officinalis, we observed reduced amounts of all three BCKAs in blood plasma which would be in link with the decrease in BCAT activity in conditions of the suppressed BC growth in rats.
Cancer cells rely heavily on glucose and convert it rapidly to pyruvate via glycolysis. Instead of pyruvate entering the TCA cycle, cancer cells turn most of the pyruvate into lactate which is released into the environment, then reuptaken from circulation and converted back to pyruvate to fuel the TCA cycle. It has been shown that the turnover of circulating lactate is the highest of all metabolites and lactate is the main source for the TCA cycle in normal tissue as well as in tumors (Hui et al., 2017). In the SAL 0.1 group, blood plasma levels of lactate and pyruvate were observed to be decreased in comparison to the non-treated group, suggesting less metabolically active tumorous tissue. The lowered levels of pyruvate influence the functioning of TCA which is manifested in decreased citrate plasma levels. The importance of metabolites involved in energy energy-gaining TCA cycle in comprehensive cancer development and progression was demonstrated also for succinate, accumulated in blood plasma in the SAL 0.1 group. Succinate is interconnected with inflammatory responses in macrophages and other immune cells. In addition to its role in TCA metabolism, succinate acts as a regulatory signal enhancing Il-1β expression and inhibits the hydroxylation of HIF-1α (Cordes et al., 2016). Various mechanisms could be proposed as the cause of succinate accumulation, including increased glutamine anaplerosis and oxidation in the TCA cycle. Just this seems to be likely as we observed in parallel a significant decrease in glutamine levels in the SAL 0.1 group. Although reduced circulating glutamine could be linked to the increased glutamine demand by BC cells, this seems to be improbable regarding applied S. officinalis treatment showing cancer-suppressing effects. Instead, in line with increased succinate, reduced glutamine levels are rather associated with enhanced immune response, since glutamine is known to serve as an essential fuel for all immunocompetent cells. On the other hand, in the SAL 1 group, the accumulation of circulating pyruvate was observed, suggesting its decreased utilisation, probably due to impaired transport into the mitochondria. The decreased citrate level in this group supports the thesis of the alterations in pyruvate conversion, and together with decreased succinate levels also slowdown of TCA cycle, which is, in contrast to SAL 0.1, not supplemented by circulating glutamine. In both groups, observed changes are likely linked with the phytopharmacological effect of S. officinalis affecting the AcetylCoA biosynthesis which is an important molecule that connects lipid metabolism with histone acetylation, forms a more complex regulatory mechanism that influences cancer growth, proliferation, and metastasis (He et al., 2023).
Finally, SAL 0.1 rats showed decreased plasma levels of essential amino acid histidine, suggesting accelerated histamine production. Histamine is a molecule with dual effects in carcinogenesis: although some studies support its pro‐tumorigenic effects, many experimental and clinical findings show that it is a crucial mediator of immune cell responses, participating in anti-tumour immunity (Nguyen and Cho, 2021). Considering the other results from our work, we believe that histidine depletion in SAL 0.1 animals may participate in the antitumorous effect of phytopharmacological treatment, although the exact mechanism is not known. SAL 0.1 animals also showed accelerated utilization of circulating tyrosine, which is a precursor of dopamine and other catecholamines, which were reported to exhibit an inhibitory effect on tumour growth in the breast (Jayachandran et al., 2023). In summary, observed plasma metabolic changes in BC rats treated group with S. officinalis did not follow the same pattern in both doses, where differences, particularly in levels of metabolites involved in energy metabolism, may be linked with additional, but so far unknown effects of phytochemicals on plasma metabolome.