Using Reactive Oxygen and Nitrogen Species in Cancer TreatmentView this Special Issue
Antitumoral and Immunomodulatory Effect of Mahonia aquifolium Extracts
The prodrug potential of Mahonia aquifolium, a plant used for centuries in traditional medicine, recently gained visibility in the literature, and the activity of several active compounds isolated from its extracts was studied on biologic systems in vitro and in vivo. Whereas the antioxidative and antitumor activities of M. aquifolium-derived compounds were studied at some extent, there are very few data about their outcome on the immune system and tumor cells. To elucidate the M. aquifolium potential immunomodulatory and antiproliferative effects, the bark, leaf, flower, green fruit, and ripe fruit extracts from the plant were tested on peripheral blood mononuclear cells and tumor cells. The extracts exert fine-tuned control on the immune response, by modulating the CD25 lymphocyte activation pathway, the interleukin-10 signaling, and the tumor necrosis-alpha secretion in four distinct human peripheral blood mononuclear cell (PBMC) subpopulations. M. aquifolium extracts exhibit a moderate cytotoxicity and changes in the signaling pathways linked to cell adhesion, proliferation, migration, and apoptosis of the tumor cells. These results open perspectives to further investigation of the M. aquifolium extract prodrug potential.
In the tumor microenvironment (TME), tumor intrinsic factors and tumor extrinsic factors work together to induce immunosuppression. The composition of TME depends on the cancer types and disease stages. The tumor cells chronically secrete tumor intrinsic factors. Some of them induce reduction of the immune effector cell activity and promote immune evasion by decreasing the expression of antigen-presenting molecules and by expressing neoantigens. In the same time, the tumor cells use autocrine or paracrine signals in order to stimulate the expression of immune checkpoints (ICs) [1, 2] on immune cells and to upregulate immunosuppressive cell recruitment and activation . The tumor cells also secrete cytokines and growth factors that promote tumor growth, angiogenesis, and metastasis .
The tumor extrinsic factors are provided by immune and nonimmune cells . The immune cells of the TME are T cells, B cells, macrophages, monocytes , dendritic cells, and NK cells , and an efficient antitumor immune response implies both helper CD4+ and effector CD8+ T cells activated in proximity to each other . T regulatory cells (Tregs) are a subset of CD4+ T cells, with immunosuppressive activity by inhibiting cytotoxic CD8+ T cells and effector CD4+ T cell activation, via consumption of IL-2, release of TGF-β and IL-10, and IC expression upregulation [4, 8]. B regulatory cells (Bregs), a subset of B cells, and tumor-associated macrophages (TAMs) reduce the activity of cytotoxic CD8+ and CD4+ T effector cells by releasing anti-inflammatory cytokines like IL-10 and by expressing coinhibitory molecules [9–11]. TAMs also stimulate tumor growth and metastasis by secreting matrix metalloproteinases (MMPs) and proangiogenic factors, like VEGF . Other tumor-associated immune cells are myeloid-derived suppressor cells (MDSCs) , a heterogeneous population of myeloid cell precursors that can suppress cytotoxic CD8+ T cell; tumor-associated mast cells (TAMCs), with a controversial immunosuppressive role; and tumor-associated dendritic cells (TADCs), which can inhibit cytotoxic CD8+ T cells by expressing inhibitory molecules and releasing IL-10 and TGF-β .
From the TME nonimmune cells, the most important are cancer-associated fibroblasts (CAFs) and tumor endothelial cells (TECs) . As a response to TME hypoxia and tumor cell intrinsic factor release, normal resident fibroblasts are converted to CAFs [14, 15]. They stimulate tumor growth, invasion and metastasis, MMPs, angiogenesis, and CD8+ T cell apoptosis . TECs are different from the normal epithelial cells due to the morphological abnormalities and induce angiogenesis with new vessel formation that allows tumor cell metastasis. Moreover, active TECs release their own growth and angiogenic factors, further stimulating neighboring tumor cell growth [16, 17].
The cells of the immune system should stop the tumor growth and the progression by recognition and removal of the malignant cells . Instead, it was found that the result of the interaction between immune and nonimmune cells in the TME is tumor-mediated immunosuppression [4, 19]. Furthermore, the tumor-mediated immunosuppression may also reduce cancer therapy efficiency and may induce resistance to therapy.
The genus Mahonia Nuttall has about 70 species, and it is the second largest genus in the Berberidaceae family. Mahonia plants are native to Eastern Asia, North America, and Central America  and have been widely used in traditional medicine for centuries. It was shown that Mahonia species have antioxidant, anti-inflammatory [21, 22], antifungal, antimicrobial , antiproliferative, hepatoprotective, and analgesic effects . Phytochemical analysis proved that alkaloids represent the major constituents of the genus, and some studies reported that they have anticancer effects. For Mahonia bealei and Mahonia oiwakensis, cytotoxic activity against human cancer cells was demonstrated [24, 25]. Previous studies also found in M. aquifolium extracts important quantities of alkaloids with cytotoxic effects on cancer cells . One study demonstrated cytotoxic and antimetastatic effects of M. aquifolium stem-bark extract .
The phytochemical profile of plant extracts differs depending on the plant and the particular organ of a given plant and of the extraction method ; the aim of the study was to test if M. aquifolium bark, leaf, flower, green fruit, and ripe fruit extracts can influence the TME in order to increase the antitumor responses. First, M. aquifolium extract immunomodulatory effects were tested on four human peripheral blood mononuclear cell (PBMC) subsets which have a key role in the adaptive immunity: CD4+ helper T cells, CD8+ effector T cells, CD19+ B cells, and CD14+ monocytes. Secondly, M. aquifolium extract antitumoral effects were tested on three cancer cell lines, DLD-1 colon carcinoma cells, A2780 ovary adenocarcinoma cells, and A375 malignant melanoma cell, and a nonimmune cell, BJ healthy skin fibroblast.
2. Materials and Methods
2.1. Plant Material
Fresh Mahonia aquifolium (Pursh) Nutt. bark, leaves, flowers, and fruits were purchased from the A. Borza Botanical Garden “Babes-Bolyai” University of Cluj-Napoca, Romania, between April and June 2018 and extracted in the Mycology Laboratory of “Babes-Bolyai” University, Cluj-Napoca, Romania, as previously described by a modified Squibb repercolation method with 70% ethanol (Merck, Bucuresti, Romania), producing the following extracts of M. aquifolium: bark extract 1 : 1.5 (g : mL) (1), leaf extract 1 : 1.2 (g : mL) (2), flower extract 1 : 1 (g : mL) (3), green fruit extract 1 : 1 (g : mL) (4), and ripe fruit extract 1 : 1 (g : mL) (5). The phytochemical analysis of the extracts has been performed and previously published [26, 30]. The plants were taxonomically identified and authenticated, and voucher specimens (number 665978) were deposited in “Alexandru Borza” Botanical Garden Herbarium, “Babes-Bolyai” University of Cluj-Napoca, Romania. The stock solutions were diluted with Phosphate-Buffered Saline Solution (PBS, from Sigma-Aldrich Company, St. Louis, USA), to obtain for each extract a series of stock concentrations from 500 to 10 μg plant/mL.
2.2. Immunomodulatory Effects
2.2.1. Isolation of PBMC Subsets
The biologic system used for testing was the suspensions of human PBMC, obtained by venipuncture from a 25-year-old healthy male volunteer, who gave his informed written consent before the blood collection, according to the approvals from the Ethical Committee of the Institute of Oncology “Prof. Dr. Ion Chiricuta” from Cluj-Napoca (IOCN), Romania, member of OECI. CD4+ helper T cells, CD8+ effector T cells, CD19+ B cells, and CD14+ monocytes were separated as previously described  (see supplemental data (available here)).
2.2.2. PBMC Cytotoxicity Test
To assess cytotoxicity serial dilutions from M. aquifolium extracts, stock solutions were prepared, in order to obtain 5 successive concentrations, from 20 μg/mL to 1 μg/mL. The cytotoxicity of the extracts was assessed using the MTS viability dye (CellTiter 96 Proliferation Assay, manufactured by Promega Corporation, Madison, WI, USA), as previously described  (see supplemental data).
2.2.3. Detection of IL-10-Positive and CD25-Positive Cells
After a prolonged exposure to the extracts 1-5 at a subcytotoxic concentration, CD4+, CD8+, CD19+, and CD14+ cell activation pathways through CD25 and interleukin-10 (IL-10), an immune-activating cytokine implicated in cancer immunotherapy , were evaluated by flow cytometry (see supplemental data).
2.2.4. Tumor Necrosis Factor Alpha (TNF-α) Production
The soluble form of tumor necrosis factor alpha (TNF-α) production was measured, knowing that this inflammatory cytokine has an important role in tumor proliferation, metastasis, and neoangiogenesis . The evaluation of the secreted TNF-α level through ELISA testing (kit acquired from Hycult Biotech, Uden, The Netherlands) was performed according to the manufacturer indication (see supplemental data).
2.3. Antitumoral Effect Evaluation
2.3.1. Cell Cultures
The human cell lines used in the present study were DLD-1 colon carcinoma and the A2780 ovary adenocarcinoma cell lines acquired from the European Collection of Authenticated Cell Cultures (ECACC) through Sigma-Aldrich, St. Louis, USA; the A375 malignant melanoma and the BJ healthy skin fibroblast cell lines were from the American Type Culture Collection (ATCC) acquired through LGC Standards GmbH, Wesel, Germany (see supplemental data).
2.3.2. Cytotoxicity Test
For the cytotoxicity testing, the colorimetric assay based on the tetrazolium dye 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction to its purple colored formazan product was used. Because this process occurs only in the mitochondria of the living cells, the amount of living cells can be tracked by the color intensity of the samples. All experiments were performed in triplicates. The samples were analyzed as previously described  (see supplemental data).
2.3.3. Protein Content of the Samples
For proteomic methods, the tumor and normal cells were seeded on 6-well plates at a concentration of 105 cells/mL, and after 24 hours, they were treated with a 10 μg/mL solution of each extract. The supernates and cell lysates were kept at -20°C until analysis. To evaluate the supernatants and cell homogenate total protein content, the Bradford assay was used. The calibration curve was prepared using seven serial dilutions starting from 100 μg/mL proteins to 1.56 μg/mL proteins (see supplemental data).
2.3.4. Intracellular Caspase-3 and Caspase-8
The intracellular caspase-3 and caspase-8 were measured with an ELISA method, and the caspase concentration provided by the quantitative measurement (ng/mL) was normalized according to the total protein content of each lysate sample, and in this way, the caspase level was expressed as ng/mg protein, to ensure an accuracy of the assessment (see supplemental data).
2.3.5. Soluble Intracellular Adhesion Molecule-1 (ICAM-1) and Vascular Cell Adhesion Molecule-1 (VCAM-1/CD106)
ICAM-1 was determined by using a human ELISA kit (E-EL-H2585 from Elabscience Biotechnology Co. Ltd., Houston, TX, USA), and VCAM-1/CD106 was measured with a human ELISA kit (E-EL-H5587, from Elabscience Biotechnology Co. Ltd., Houston, Texas) according to the manufacturer indications. ICAM-1 and VCAM-1/CD106 were expressed as ng/mg protein (see supplemental data).
2.3.6. Matrix Metalloproteinase-9 (MMP-9)
MMP-9 was assessed with a human MMP-9 Platinum ELISA kit (MBS2016/2, from Affymetrix, through eBioscience, Vienna, Austria), according to the manufacturer’s indications. MMP-9 was expressed as ng/mg protein (see supplemental data).
2.4. Statistical Analysis
GraphPad Prism 5 software (from GraphPad Software Inc., La Jolla, CA, USA) was used to compute IC50 values (nonlinear regression of the concentration versus normalized response) and Spearman nonparametric correlations, to compare values with the one-way analysis of variances, followed by Bonferroni or Dunnett multiple comparison posttest, and to compute the mean values and the standard error of the mean (SEM).
To examine the strengths of associations between the results, specifically Pearson correlations, we have used Statistica 12.0 for Windows (Stat-soft, Inc., USA). Multivariate data analysis was performed on the entire antioxidant and hematological parameters determined in this study using PCA (principal component analysis) incorporated in Statistica 12.0 software.
3. Results and Discussion
3.1. Immunomodulatory Effect
3.1.1. PBMC Cytotoxicity Test
From each Mahonia aquifolium extract, five concentrations were tested, between 1 and 20 μg/mL in the cell culture media. All concentrations exhibited low cytotoxicity against PBMC, including the 20 μg/mL concentration. The survival rates of the cells treated with the highest concentration were shown in Figure 1(a) (see supplemental data Table 5).
In all PBMC subpopulations, M. aquifolium extracts 1-5 treatment induced a similar growth inhibition pattern. Extract 1 exerted the highest inhibitory effect, followed by 2 (Figure 1(b)), while 3, 4, and 5 caused less than 10% cell loss after 24-hour exposure. Consequently, for further testing, we used all extracts at the concentration of 20 μg/mL. The principal component analysis (PCA) clearly indicated (Figure 1(b)) the similarities between 1 and 2 effects, these extracts being in the same quadrant of the loading plot, at distance to the group of extracts 3, 4, and 5, placed in another quadrant; further, 1 and 2 are distanced to more than 90 degrees to the least active 3 and 5 M. aquifolium extracts.
3.1.2. IL-10-Positive and CD25-Positive Cells
The immunomodulatory effect of M. aquifolium extracts was evaluated by monitoring the number of CD25+ and IL10+ cells within a population of 10000 PBMC. CD25 and IL10 membrane marker expression was assessed by flow cytometry, and each data represents the median value of 10000 measurements, provided by the BD FACSDiva version 6.1 software. For each sample, two independent evaluations were made; their mean value was calculated with the column statistics, and the comparison between values was provided by the one-way analysis of variance in the 95% interval (Table 1) (see supplemental data Figures 10–13).
Following the 24-hour exposure of the PBMC subsets to 1-5 M. aquifolium extracts, both the CD25 membrane marker and the IL-10 intracytoplasmic marker were modulated distinctly in each cell subpopulation. In untreated T cells, the basal level of the two molecules was modest, while in B lymphocytes and monocytes the basal values were higher (Table 1).
The effect of 1, the bark extract, was the prominent regulator of the lymphocyte activation through the CD25 pathway. Extract 1 distinguishes itself from other extracts through its lower chlorogenic acid content (see Supplemental data Table 7) and through the presence of berbamine, jatrorrhizine, palmatine, and berberine, four compounds that are not present in the extracts 2 to 5 [26, 30].
Extract 1 upregulated the CD4+CD25+, CD8+CD25+, and CD14+CD25+ phenotype expression, and in this manner, 1 triggered the activation of both tumor suppressor and effector lymphocyte subsets and also the CD14+CD25+ monocytes, the main population of blood monocytes implicated in antitumor response . CD14+ cells were proven to act against the tumor cells and the metastatic processes, directly by an antibody-mediated mechanism  and indirectly by activating the natural killer cells .
T cells which expressed CD4+ and CD25+ are essential in self-recognition and are known as T regulatory cells  and maintain the self-tolerance control in the immune response against infections, transplantation antigens, and tumor-associated antigens . CD4+ T cells play antitumor roles through various mechanisms, and some studies claim that they could be efficient even in antitumoral immunotherapy like CD8+, previously known as the gold standard . Although tumor-infiltrating CD4+CD25+ lymphocytes can contribute to the progression of the disease , they may act against tumors indirectly, helped by NK cells, or by inactivating the IL-10 pathway .
The fact that 1 exerts a significant enhancement of CD8+CD25+ activated phenotype converges to the antitumor potential of the bark extract, since CD8+ are important effectors of the antitumor cellular immunity in several cancers , counteracting the metastatic potential of tumor cells , and it was proven that their presence is related to the patient survival . The increase of CD25+IL-10+ phenotype occurred following the treatments with 1, the only extract able to increase CD25, or concomitant with the increase of IL-10 induced by 2-5 in CD4+ and CD19+ (Table 1). IL-10 sustain the toxicity in CD8+ cells,  and in high concentrations, IL-10 is able to inhibit tumor growth by inhibiting the angiogenesis and the production of reactive oxygen species .
Extracts 2, 3, 4, and 5 could not influence significantly CD4+ or CD8+ cells. Instead, they influenced the double-positive populations and the IL-10 expression in these two subsets. Extract 2 increased the CD25+IL-10+ expression only on cytotoxic CD8+ T lymphocytes, without augmenting the helper CD4+ activation. Extracts 3 and 5 enhanced the CD25+IL-10+ activation in cytotoxic CD8+ cells but parallel in the regulatory CD4+ T cells. Overall, the PCA indicated that the outcome of the extracts on CD4+ subpopulation was opposed to CD8+, and the effect of 1-5 on CD14+ is opposed to CD19+ cells (Supplemental data Figure 14).
In CD19+ B cells, all extracts 1-5 performed likewise: they exerted a strong downregulatory effect on CD25+ cells, but the CD25+IL-10+ phenotype was upregulated (Table 1). The CD19+ expression is essential in the B cell-mediated immune response . The increase of CD19+ cells and the regulatory CD19+CD25+IL-10+ B cell overexpression enhance the immunoglobulin production, but in tumors, it is responsible for metastatic growth support  as well, by costimulating the Treg cells. The CD19+CD25+IL-10+ phenotype was confirmed to be protector against inflammation  and promotes allograft survival in transplantation .
In CD14+, monocytes 2-5 had a strong downregulatory effect through CD25 and IL-10 decrease (Table 1), acting contrary to extract 1.
The IL-10 expression decreased following the treatment with extract 1 in CD4+, CD8+ T cells, and monocytes and strongly upregulated in CD19+B cells (Figure 2). In IL-10 modulation as well, extracts 2, 3, 4, and 5 act convergently: they caused a significant increase of IL-10 phenotype in helper CD4+ T cells and CB19+ B cells and decline in effector CD8+ T cells and CD14+ monocytes.
It was a strong correlation between the CD25+ and CD25+IL-10+ phenotype in all subsets (Spearman nonparametric correlation, ) which means that the two subpopulations tend to increase or to decrease in the same time. In all PBMC subsets, the treatment with 1-5 induced no significant correlation between CD25+ and IL-10+ expression (nonparametric two-tailed correlation, Spearman ).
3.1.3. Tumor Necrosis Factor Alpha (TNF-α) Production
TNF is a multifunctional cytokine that has important roles in cell survival, proliferation, differentiation, and death. As a proinflammatory cytokine, TNF may be implicated in inflammation-induced carcinogenesis. TNF exerts its functions by activating distinct nuclear factor-κB (NF-κB), an antiapoptotic signal, and c-Jun N-terminal kinase (JNK), a cell death signal. So, TNF is a double-edged sword with pro- or antitumorigenic effects , which in the early stages can contribute to antitumor response enhancement, while in late stages, it could maintain the tumor growth . All PBMC subsets have had a certain basal TNF-α level (Figure 3), since CD4+ and CD8+ T lymphocyte subsets, B cells, and CD14+ monocytes are TNF-α secretory cells . Following the treatment, TNF-α was modulated distinctively in each PBMC population. Extracts 1 and 2 had no effect on T or B cells, but they upregulate the TNF-α in CD14+ cells. In CD14+ monocytes, the main population of blood monocytes implicated in antitumor response, the direct cytotoxicity against the target cells is mediated by TNF-α, IL-12, reactive oxygen species, and reactive nitrogen species . Since 1 and 2 modulated significantly the cytokine production only in CD14+ monocytes, it is expected that these two extracts will enhance the CD14+ cell antitumor capacity. Extract 3 was completely inert in the PBMC biologic system. Surprisingly, extract 4 obtained from green fruits significantly decreases the TNF-α production of monocytes, while in all the other studied subpopulations causes a very significant increase of TNF-α. Novel studies suggested that the increase of TNF-α secreted by the tumor-infiltrating T cells and monocytes may enhance the tumor cell death without associated systemic toxicities . The TNF-α secreted by T lymphocytes can synergize with chemotherapy to strengthen tumor cell death mechanisms and the induced oxidative stress , and by enhancing the TNF-α production in CD4+ and CD8+ lymphocytes, extract 4 could balance the regulatory mechanisms induced through CD25. Extract 5 inhibited the TNF-α secretion in CD4+ T cells, and in B cells, a stimulation was observed.
Previous studies reported that in CD4+ and CD8+ T cells the IL-10+ phenotype is suppressed by exogenous TNF-α . This was confirmed by the decrease of IL-10 positivity in CD8+ cells treated with 4, concomitant with the highest TNF-α secretion. It is known that TNF-α downregulates the function of tumor suppressor CD4+CD25+ lymphocytes , but overall, no correlation was found between TNF-α secretion and the IL-10+ or CD25+IL10+ expression.
As a conclusion, M. aquifolium extracts 1-5 exhibit moderate toxicity against PBMC subpopulations without selectivity towards any of the four studied subsets, and their modulator effect through CD25, IL-10, and TNF-α was well-balanced. The bark extract (1) effect was distinct, as regards the cytotoxicity, and its capacity to prime the helper CD4+ cells via CD25 activation, but this was compensated by the enhancement of CD8+ cytotoxic T lymphocyte activity. Also, 1 stimulated the immunoglobulin (the B cell secretor) through CD25 and IL-10 and triggered and amplified the TNF-α production in CD14+ monocytes only. The leaf extract (2), the richest in chlorogenic acid, rutin and isoquercitrin, has the capacity to activate the effector CD8+ lymphocytes and the B cells, deactivate the CD14+ monocytes, and in the same time increase a fold higher their TNF-α production.
TNF-α secretion was correlated with the CD25+ and CD25+IL-10+ phenotype but not with the CD25-IL-10+ cells (Supplemental data Figure 15). The PCA statistic confirms once again that CD25+ and CD25+IL-10+ populations were strongly correlated. In the tridimensional PCA (Figure 4), TNF-α variations were considered as a principal component; the concentrations of TNF-α were plotted with different colors, against the CD25 and IL-10 parameters. We can conclude that above the 100% increase of TNF-α, both IL-10 and CD25/IL-10 phenotype are overexpressed. Where the TNF-α is stationary towards the control values, then IL-10 is negatively correlated with CD25+ expression.
The flower extract (3) has the most moderate effect, it does not influence the T cells or monocyte activation or the TNF-α production; the only enhancement of activation was in B cells through the IL-10; therefore, the application of these extracts as part of a potential cancer treatment should not exhibit any unwanted effect on antitumor immunity.
There was a significant difference between green fruit extract 4 and ripe fruit extract 5, even if their composition is quite similar , the only notable difference being the higher chlorogenic acid level in 5. Extract 4 activates via IL-10 and not via CD25 the helper lymphocytes, deactivating the effector lymphocytes and monocytes. The only cell population which could be activated through 4 was the B subset, where the TNF-α production was also very high, costimulating the B cell responsiveness. Following the treatment with extract 5, the same outcome occurred at a smaller scale.
3.2. Antitumoral Effect
3.2.1. Cytotoxicity Test
The toxicity of the M. aquifolium bark (1), leaf (2), flower (3), green fruit (4), and ripe fruit (5) extracts was assessed and quantified using the mathematic parameter half inhibitory concentration IC50, which represents the amount of compound necessary to eliminate 50% from the cells subjected to the treatment. Therefore, a small IC50 value is a characteristic for the toxic compounds (Table 2).
In vitro, extract 1 exhibited the best cytotoxicity, corresponding to the smallest IC50 value in the whole series (Table 2). This tendency was convergent in all cell lines, the IC50 values being around 10 μg/mL, which is a moderate cytotoxicity according to the pharmacologic standards. Only extract 1 displays a dose-response relationship which can be described by the sigmoidal curve; in all the rest of extracts, the sigmoidal curves are distorted, because none of the tested concentrations was able to inhibit 100% of the cells (see supplemental data Figure 16). This effect can be correlated to the higher alkaloid content in the M. aquifolium bark extract . Even the highest concentration of the extract was not able to inhibit 50% of the cells when the treatment was made with extracts 3 (in all cell lines), 4, and 5 (each in two cell lines); therefore, the IC50 values above 50 μg/mL provided by the biostatistics software are hypothetical values, obtained by extrapolation.
The PCA statistics indicated few convergences in the extract IC50 values, when all the cell lines were analyzed (Figure 2(a)); A2780 and A375 outcome was well correlated, since the two values are in the same quadrant. The best selectivity of the extracts was observed in the DLD-1 and BJ contrast (Figure 2(b)).
Extract 2 has had the second-best cytotoxicity, but the IC50 values were twice as big as in 1. Extract 3 presented the lowest toxicity, while the IC50 values of 4 and 5 were in-between. The extract 2, but especially 4, showed a good selectivity towards the tumor cell lines, since in the normal BJ skin cell line, its IC50 value was higher than in tumor cells, which anticipate a lower toxicity towards normal skin fibroblasts following the administration versus the tumor cells with lower IC50. The cytotoxicity was dependent on the extracts and not on the cell type.
Based on the IC50 values (Table 2), for further testing, the 10 μg/mL concentrations were used for each extract, ensuring in this way a good proportion of living cells for the mechanistic studies, even in the extract 1-treated cells.
3.2.2. The Protein Content of the Samples
In order to identity the expression or production of certain molecules implicated in apoptosis, intercellular communication, and tumor dissemination, two types of samples were collected: the supernatants to identify soluble proteins secreted by the treated cells and cell homogenates, respectively, to identify the molecules implicated in the intracellular signaling. The protein content of the samples derived from supernatants and cell lysates was measured (Table 3) as described in Materials and Methods. The total protein concentration (μg/mL) itself is not an indicator of cytotoxicity; instead, it was used to normalize the quantity of the target proteins after they were measured by immune-enzymatic methods (see supplemental data Table 7 and Figure 17).
3.2.3. The Extrinsic Apoptosis Induction
The caspase cascade is the basic apoptotic signaling pathway, involved in both intrinsic and extrinsic apoptosis. Caspases are essential components of the apoptotic process. Caspase-8 is an initiator caspase of the extrinsic apoptotic pathway, while caspase-3 has an effector role, being activated by caspase-8 or other caspases or chemokines.
Some of the studied M. aquifolium extracts were capable to enhance the TNF-α expression in PBMC. On the extrinsic apoptotic pathway, the TNF-α increase implicates the proteolysis or cleavage of caspase-8, which further can activate caspase-3. Therefore, the intracellular level of these two caspases was evaluated following the 24-hour in vitro treatment of tumor (A2780, DLD-1, and A375) and normal (BJ) cell lines with extracts 1-5. The active or cleaved form of the two proteins is of interest in the apoptotic process, and the increase of the cleaved caspase expression is known to be an indicator of the apoptotic process (data provided by Metacore of Clarivate Analytics: Apoptosis Pathway Map, http://pathwaymaps.com/maps/373).
The apoptotic process initiation through caspase-8 occurred in all tumor cell lines for all extracts 1-5 (Figure 5), with statistical significance in the 95% confidence interval (one-way analysis of variance, Dunnett posttest). In normal BJ cells, only extract 1 effect was significant, but in the opposite direction, meaning caspase-8 decrease. This indicates a selectivity of the extracts regarding the cell death mechanism induction.
The caspase cascade was evolving through caspase-8 towards the programmed cell death, but not in all treatments, or in all cell lines, the proapoptotic signal was propagated. The active caspase-3 increased following the exposure to 1 in all tumor cell lines, and 2 induced and increased in DLD-1 and A375 cells, and this denotes that the caspase cascade was influenced by 1 and 2 to evolve towards the extrinsic apoptotic pathway. In DLD-1 colon carcinoma, even 3, 4, and 5 caused caspase-3 cleavage; therefore, it is very likely the extrinsic apoptosis triggered by any of extracts 1-5 in K-ras mutant aggressive DLD-1 cells. In BJ cells, only extract 2 augmented the caspase-3 expression.
It was found that berberine, the main alkaloid from M. aquifolium extracts, mostly the bark extract, enhanced the expression and activation of caspase-3 and caspase-8 in ovarian cancer cells . This finding may be an explanation for the highest caspase activation by extract 1 with the most important berberine content . Once again, extract 1 was constantly active against all cell lines and selective towards normal BJ cells.
3.2.4. Modulation of Adhesion Molecules
The intracellular adhesion molecule-1 (ICAM-1) and the vascular adhesion molecule-1 (VCAM-1) are implicated in tumor growth, extravasation, and dissemination. ICAM-1 is expressed on fibroblasts, keratinocytes, endothelial cells, and leukocytes, but also in many types of tumors . ICAM-1 enhances tumor cell adherence to endothelial cells favoring tumorigenesis . Moreover, it was reported an association between tumor relapse or drug resistance and ICAM-1 overexpression . In inflammation, VCAM-1 is predominantly expressed on endothelial cells, but in cancer, it plays a dual role: increases the ability to metastasize and rises tumor-associated monocytes and macrophage recruitment. In tumors like breast, renal, and gastric carcinoma, aberrant VCAM-1 expression was found on tumor-associated vasculature and on tumor cells. In other tumor-associated vessels of some human malignancies has been reported downregulation of ICAM-1 and VCAM-1 . The soluble form of the two proteins can act as biomarker in clinical diagnosis and treatment follow-up, and generally, the elevated values indicate a poor prognosis.
The extracts did not act identically against the ICAM-1 secreted by tumor cells (Figure 6). Extract 1 acts only on DLD-1 cells, generating a significant drop of ICAM-1; 3 influenced only the A375 cell line, while 4 and 5 were efficient in DLD-1 cells.
The extracts 1-5 were more active against VCAM-1, and in this case, extract 1 was capable to reduce the VCAM-1 secretion in all tumor cell lines (A2780, DLD-1, and A375). Extract 2 and even the less active extract 3 reduced VCAM-1 in DLD-1 and A375 cells.
Interestingly, all extracts have had an inhibitory effect against the ICAM-1 and VCAM-1 secreted by BJ normal fibroblasts in vitro. This property could be used to initiate further studies, since the elevated ICAM-1 or VCAM-1 levels are a characteristic of benign skin disorders; therefore, the M. aquifolium extracts gain a new perspective for future applications.
The PCA model is a useful tool to examine the multifaceted biologic effect of many natural extracts . This method highlighted the relationship between the 1-5 extract capacity to trigger apoptosis and to influence the adhesion molecules which give the tumor cells the capacity to migrate (Figure 7): for strong VCAM-1 increase, up to 380 units, the expression of caspase-3 and ICAM-1 was directly proportional, while to a moderate increase of VCAM-1, the two parameters were divergent. VCAM-1 could increase up to 400 units, or it could be diminished up to 800-unit weight against the untreated controls in different tumor cell types, at the same time as the correlation remains indirect between ICAM-1 and caspase-3 (Supplemental data Figure 18).
3.2.5. The Effect on Matrix Metalloprotease-9 (MMP-9)
MMPs, synthesized by neoplastic and stromal cells, are zinc-dependent proteases, which play a role in extracellular matrix remodeling and cancer progression, invasion, and metastasis. Because MMP-9 has been found overexpressed in tumor tissues compared with the adjacent nontumor tissues, it can be a prognostic molecular biomarker . Against MMP-9 overexpression, the best cytotoxic extracts 1 and 2 do not act in a significant manner, and 3 as well has no effect (Figure 8). The only exception is 1, which was able to inhibit MMP-9 in melanoma cell A375. Instead, extracts 4 and 5 were more active against the MMP-9 overgrowth. Extract 5 inhibited MMP-9 in all tumor cell lines, while 4 was active only in A2780. In normal BJ cells, none of the treatments caused significant drop in MMP-9 levels.
Following to the treatment with 1-5 if MMP-9 values were amplified (above 12 units), then ICAM-1 and VCAM-1 were both elevated (Figure 9). Instead, if MMP-9 secretion was moderate, below 10 units, one of variables ICAM-1 and VCAM-1 has increased, simultaneously with the decline of the other parameters.
In conclusion, the extracts 1-5 have good potential to become TME modulatory agents, similar with other Mahonia extracts  described before, due to the immunomodulatory and antitumoral effects. Extract 1 should be employed carefully, and further tested, to estimate whether the increase of the regulatory helper CD4+cell level is counterbalanced with the tumor-suppressing mechanisms in vivo. The extracts 2, 4, and 5 are more suitable to activate the effector CD8+ cells, the monocytes, and B cells, and the utilization of 3 more likely will substantially benefit together with other active compounds, since it will give no secondary effects in the case of systemic administration, together with the standard antitumor drugs or alone. The M. aquifolium extracts also exhibited a moderate cytotoxicity on tumor cells. Among them, extract 1 was prominent as regards the antiproliferative capacity. We demonstrated for the first time that at subcytotoxic concentrations, the tumor cell lines subjected to 1-5 undergo changes in the signaling pathways linked to cell adhesion, proliferation, migration, and apoptosis (Table 4, Figure 9). These results open perspectives to further investigation of the M. aquifolium extract prodrug potential.
Data are available in the manuscript and supplemental data file.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
The authors express their gratitude to Anca-Daniela Farcaş of National Institute for Research and Development of Isotopic and Molecular Technologies, Cluj-Napoca (INCDTIM), for her generous contribution to the PCA statistical analysis of data.
More information about the methods, materials, and results can be found in the Supplemental data folder, as indicated in the main text. (Supplementary Materials)
R. Ruiu, L. Tarone, V. Rolih et al., “Cancer stem cell immunology and immunotherapy: Harnessing the immune system against cancer's source,” in Progress in Molecular Biology and Translational Science, Elsevier Inc., 1st ed edition, 2019.View at: Publisher Site | Google Scholar
N. Curdy, O. Lanvin, C. Laurent, J.-J. Fournié, and D.-M. Franchini, “Regulatory mechanisms of inhibitory immune checkpoint receptors expression,” Trends in Cell Biology, vol. 29, no. 10, pp. 777–790, 2019.View at: Publisher Site | Google Scholar
D. S. Vinay, E. P. Ryan, G. Pawelec et al., “Immune evasion in cancer: mechanistic basis and therapeutic strategies,” Seminars in Cancer Biology, vol. 35, pp. S185–S198, 2015.View at: Publisher Site | Google Scholar
R. Saleh and E. Elkord, “Acquired resistance to cancer immunotherapy: role of tumor-mediated immunosuppression,” Seminars in Cancer Biology, 2019.View at: Publisher Site | Google Scholar
C. Liu, C. J. Workman, and D. A. A. Vignali, “Targeting regulatory T cells in tumors,” The FEBS Journal, vol. 283, no. 14, pp. 2731–2748, 2016.View at: Publisher Site | Google Scholar
S. Paul, S. Chhatar, A. Mishra, and G. Lal, “Natural killer T cell activation increases iNOS+CD206- M1 macrophage and controls the growth of solid tumor,” Journal for Immunotherapy of Cancer, vol. 7, no. 1, pp. 208–213, 2019.View at: Publisher Site | Google Scholar
A. Y. Huang, P. Golumbek, M. Ahmadzadeh, E. Jaffee, D. Pardoll, and H. Levitsky, “Role of bone marrow-derived cells in presenting MHC class I-restricted tumor antigens,” Science, vol. 264, no. 5161, pp. 961–965, 1994.View at: Publisher Site | Google Scholar
K. Shitara and H. Nishikawa, “Regulatory T cells: a potential target in cancer immunotherapy,” Annals of the New York Academy of Sciences, vol. 1417, no. 1, pp. 104–115, 2018.View at: Publisher Site | Google Scholar
A. Sarvaria, J. A. Madrigal, and A. Saudemont, “B cell regulation in cancer and anti-tumor immunity,” Cellular & Molecular Immunology, vol. 14, no. 8, pp. 662–674, 2017.View at: Publisher Site | Google Scholar
M. R. Räihä and P. A. Puolakkainen, “Tumor-associated macrophages (TAMs) as biomarkers for gastric cancer: a review,” Chronic Diseases and Translational Medicine, vol. 4, no. 3, pp. 156–163, 2018.View at: Publisher Site | Google Scholar
A. A. Wolf, A. Yáñez, P. K. Barman, and H. S. Goodridge, “The ontogeny of monocyte subsets,” Frontiers in Immunology, vol. 10, 2019.View at: Publisher Site | Google Scholar
W.-C. Wu, H. W. Sun, J. Chen et al., “Immunosuppressive immature myeloid cell generation is controlled by glutamine metabolism in human cancer,” Cancer Immunology Research, vol. 7, no. 10, pp. 1605–1618, 2019.View at: Publisher Site | Google Scholar
J. Patel, E. N. Bozeman, and P. Selvaraj, “Taming dendritic cells with TIM-3: another immunosuppressive strategy used by tumors,” Immunotherapy, vol. 4, no. 12, pp. 1795–1798, 2012.View at: Publisher Site | Google Scholar
B. Farhood, M. Najafi, and K. Mortezaee, “Cancer-associated fibroblasts: secretions, interactions, and therapy,” Journal of Cellular Biochemistry, vol. 120, no. 3, pp. 2791–2800, 2019.View at: Publisher Site | Google Scholar
V. Petrova, M. Annicchiarico-Petruzzelli, G. Melino, and I. Amelio, “The hypoxic tumour microenvironment,” Oncogenesis, vol. 7, no. 1, p. 10, 2018.View at: Publisher Site | Google Scholar
N. Maishi and K. Hida, “Tumor endothelial cells accelerate tumor metastasis,” Cancer Science, vol. 108, no. 10, pp. 1921–1926, 2017.View at: Publisher Site | Google Scholar
K. Hida, N. Maishi, D. Annan, and Y. Hida, “Contribution of tumor endothelial cells in cancer progression,” International Journal of Molecular Sciences, vol. 19, no. 5, p. 1272, 2018.View at: Publisher Site | Google Scholar
B. F. Zamarron and W. Chen, “Dual roles of immune cells and their factors in cancer development and progression,” International Journal of Biological Sciences, vol. 7, no. 5, pp. 651–658, 2011.View at: Publisher Site | Google Scholar
M. W. L. Teng, J. Galon, W. H. Fridman, and M. J. Smyth, “From mice to humans: developments in cancer immunoediting,” The Journal of Clinical Investigation, vol. 125, no. 9, pp. 3338–3346, 2015.View at: Publisher Site | Google Scholar
M. W. Dulin and B. K. Kirchoff, “Paedomorphosis, secondary woodiness, and insular woodiness in plants,” The Botanical Review, vol. 76, no. 4, pp. 405–490, 2010.View at: Publisher Site | Google Scholar
L. Zhang, A. S. Ravipati, S. R. Koyyalamudi et al., “Antioxidant and anti-inflammatory activities of selected medicinal plants containing phenolic and flavonoid compounds,” Journal of Agricultural and Food Chemistry, vol. 59, no. 23, pp. 12361–12367, 2011.View at: Publisher Site | Google Scholar
A.-D. Cecan, A. E. Pârvu, M. Pârvu et al., “Mahonia aquifolium flowers extract effects in acute experimental inflammation,” Bulletin of University of Agricultural Sciences and Veterinary Medicine Cluj-Napoca. Food Science and Technology, vol. 75, no. 2, pp. 189–199, 2018.View at: Publisher Site | Google Scholar
A. R. Li, Y. Zhu, X. N. Li, and X. J. Tian, “Antimicrobial activity of four species of Berberidaceae,” Fitoterapia, vol. 78, no. 5, pp. 379–381, 2007.View at: Publisher Site | Google Scholar
W. Hu, L. Yu, and M. H. Wang, “Antioxidant and antiproliferative properties of water extract from Mahonia bealei (Fort.) Carr. leaves,” Food and Chemical Toxicology, vol. 49, no. 4, pp. 799–806, 2011.View at: Publisher Site | Google Scholar
B. S. Wong, Y. C. Hsiao, T. W. Lin et al., “The in vitro and in vivo apoptotic effects of Mahonia oiwakensis on human lung cancer cells,” Chemico-Biological Interactions, vol. 180, no. 2, pp. 165–174, 2009.View at: Publisher Site | Google Scholar
A.-D. Andreicut, A. E. Pârvu, A. C. Mot et al., “Phytochemical Analysis of Anti-Inflammatory and Antioxidant Effects of Mahonia aquifolium Flower and Fruit Extracts,” Oxidative Medicine and Cellular Longevity, vol. 2018, 12 pages, 2018.View at: Publisher Site | Google Scholar
D. Gođevac, A. Damjanović, T. P. Stanojković, B. Anđelković, and G. Zdunić, “Identification of cytotoxic metabolites from Mahonia aquifolium using 1H NMR-based metabolomics approach,” Journal of Pharmaceutical and Biomedical Analysis, vol. 150, pp. 9–14, 2018.View at: Publisher Site | Google Scholar
W. Zhu, J. Hu, X. Wang, J. Tian, and S. Komatsu, “Organ-specific analysis of mahonia using gel-free/label-free proteomic technique,” Journal of Proteome Research, vol. 14, no. 6, pp. 2669–2685, 2015.View at: Publisher Site | Google Scholar
M. Pârvu, A. E. Pârvu, L. Barbu-Tudoran et al., “In vitro effects of Allium obliquum extract on the growth and ultrastructure of Botrytis paeoniae,” Journal of Medicinal Plants Research, vol. 7, no. 17, pp. 1138–1145, 2013.View at: Google Scholar
A. D. Andreicuţ, A. E. Parvu, A. C. Moț et al., “Anti-inflammatory and antioxidant effects of Mahonia aquifolium leaves and bark extracts,” Farmácia, vol. 66, no. 1, pp. 49–58, 2018.View at: Google Scholar
E. Fischer-Fodor, A. Mot, F. Deac, M. Arkosi, and R. Silaghi-Dumitrescu, “Towards hemerythrin-based blood substitutes: comparative performance to hemoglobin on human leukocytes and umbilical vein endothelial cells,” Journal of Biosciences, vol. 36, no. 2, pp. 215–221, 2011.View at: Publisher Site | Google Scholar
H. Zhang, Y. Wang, E. S. Hwang, and Y. W. He, “Interleukin-10: an immune-activating cytokine in cancer immunotherapy,” Journal of Clinical Oncology, vol. 34, no. 29, pp. 3576–3578, 2016.View at: Publisher Site | Google Scholar
Y. Wu and B. P. Zhou, “TNF-α/NF-κB/Snail pathway in cancer cell migration and invasion,” British Journal of Cancer, vol. 102, no. 4, pp. 639–644, 2010.View at: Publisher Site | Google Scholar
M. Perde-Schrepler, A. Florea, I. Brie et al., “Size-Dependent Cytotoxicity and Genotoxicity of Silver Nanoparticles in Cochlear Cells In Vitro,” Journal of Nanomaterials, vol. 2019, 12 pages, 2019.View at: Publisher Site | Google Scholar
A. Szaflarska, M. Baj-Krzyworzeka, M. Siedlar et al., “Antitumor response of CD14+/CD16+ monocyte subpopulation,” Experimental Hematology, vol. 32, no. 8, pp. 748–755, 2004.View at: Publisher Site | Google Scholar
I. O. Gordon and R. S. Freedman, “Defective antitumor function of monocyte-derived macrophages from epithelial ovarian cancer patients,” Clinical Cancer Research, vol. 12, no. 5, pp. 1515–1524, 2006.View at: Publisher Site | Google Scholar
R. N. Hanna, C. Cekic, D. Sag et al., “Patrolling monocytes control tumor metastasis to the lung,” Science, vol. 350, no. 6263, pp. 985–990, 2015.View at: Publisher Site | Google Scholar
A. M. Wolf, D. Wolf, M. Steurer, G. Gastl, E. Gunsilius, and B. Grubeck-loebenstein, “Increase of regulatory T cells in the peripheral blood of cancer patients,” Clinical Cancer Research, vol. 9, no. 2, pp. 606–612, 2003.View at: Google Scholar
S. Chattopadhyay, S. Mehrotra, A. Chhabra, U. Hegde, B. Mukherji, and N. G. Chakraborty, “Effect of CD4+ CD25+ and CD4+ CD25 − T regulatory cells on the generation of cytolytic T cell response to a self but human tumor-associated epitope in vitro,” Journal of Immunology, vol. 176, no. 2, pp. 984–990, 2006.View at: Publisher Site | Google Scholar
A. Perez-Diez, N. T. Joncker, K. Choi et al., “CD4 cells can be more efficient at tumor rejection than CD8 cells,” Blood, vol. 109, no. 12, pp. 5346–5354, 2007.View at: Publisher Site | Google Scholar
T. J. Curiel, G. Coukos, L. Zou et al., “Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival,” Nature Medicine, vol. 10, no. 9, pp. 942–949, 2004.View at: Publisher Site | Google Scholar
M. Oft, “IL-10 : master switch from tumor-promoting inflammation to antitumor immunity,” Cancer Immunology Research, vol. 2, no. 3, pp. 194–199, 2014.View at: Publisher Site | Google Scholar
S. M. A. Mahmoud, E. C. Paish, D. G. Powe et al., “Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer,” Journal of Clinical Oncology, vol. 29, no. 15, pp. 1949–1955, 2011.View at: Publisher Site | Google Scholar
J. Galon, A. Costes, F. Sanchez-Cabo et al., “Type, density, and location of immune cells within human colorectal tumors predict clinical outcome,” Science, vol. 313, no. 5795, pp. 1960–1964, 2006.View at: Publisher Site | Google Scholar
M. E. Dudley, J. R. Wunderlich, P. F. Robbins et al., “Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes,” Science, vol. 298, no. 5594, pp. 850–854, 2002.View at: Publisher Site | Google Scholar
S. Mocellin, F. M. Marincola, and H. A. Young, “Interleukin-10 and the immune response against cancer: a counterpoint,” Journal of Leukocyte Biology, vol. 78, no. 5, pp. 1043–1051, 2005.View at: Publisher Site | Google Scholar
D. Depoil, S. Fleire, B. L. Treanor et al., “CD19 is essential for B cell activation by promoting B cell receptor-antigen microcluster formation in response to membrane-bound ligand,” Nature Immunology, vol. 9, no. 1, pp. 63–72, 2008.View at: Publisher Site | Google Scholar
P. B. Olkhanud, B. Damdinsuren, M. Bodogai et al., “Tumor-evoked regulatory B cells promote breast cancer metastasis by converting resting CD4+ T cells to T regulatory cells,” Cancer Research, vol. 71, no. 10, pp. 3505–3515, 2011.View at: Publisher Site | Google Scholar
S. Sattler, G. S. Ling, D. Xu et al., “IL-10-producing regulatory B cells induced by IL-33 (BregIL-33) effectively attenuate mucosal inflammatory responses in the gut,” Journal of Autoimmunity, vol. 50, pp. 107–122, 2014.View at: Publisher Site | Google Scholar
Y. Wei, D. Zheng, X. Li et al., “Infusion of Dendritic Cells Carrying Donor Lymphocytes Treated With 8-Methoxypsoralen and Ultraviolet A Light Induces CD19+IL-10+ Regulatory B cells and Promotes Skin Allograft Survival,” Transplantation Proceedings, vol. 46, no. 10, pp. 3641–3646, 2014.View at: Publisher Site | Google Scholar
X. Wang and Y. Lin, “Tumor necrosis factor and cancer, buddies or foes?” Acta Pharmacologica Sinica, vol. 29, no. 11, pp. 1275–1288, 2008.View at: Publisher Site | Google Scholar
S. F. Josephs, T. E. Ichim, S. M. Prince et al., “Unleashing endogenous TNF-alpha as a cancer immunotherapeutic,” Journal of Translational Medicine, vol. 16, no. 1, pp. 242–248, 2018.View at: Publisher Site | Google Scholar
A. Bojarska-Junak, J. Rolinski, E. Wasik-Szczepaneko, Z. Kaluzny, and A. Dmoszynska, “Intracellular tumor necrosis factor production by T- and B-cells in B-cell chronic lymphocytic leukemia,” Haematologica, vol. 87, pp. 490–499, 2002.View at: Google Scholar
P. E. Bunney, A. N. Zink, A. A. Holm, C. J. Billington, and C. M. Kotz, “Orexin activation counteracts decreases in nonexercise activity thermogenesis (NEAT) caused by high-fat diet,” Physiology & Behavior, vol. 176, no. 2, pp. 139–148, 2017.View at: Publisher Site | Google Scholar
C. A. Roberts, L. E. Durham, V. Fleskens, H. G. Evans, and L. S. Taams, “TnF blockade maintains an il-10 + phenotype in human effector cD4 + and cD8 + T cells,” Frontiers in Immunology, vol. 8, pp. 1–14, 2017.View at: Publisher Site | Google Scholar
X. Valencia, G. Stephens, R. Goldbach-mansky, M. Wilson, E. M. Shevach, and P. E. Lipsky, “TNF downmodulates the function of human CD4+CD25hi T-regulatory cells,” Blood, vol. 108, no. 1, pp. 253–261, 2006.View at: Publisher Site | Google Scholar
L. Liu, J. Fan, G. Ai et al., “Berberine in combination with cisplatin induces necroptosis and apoptosis in ovarian cancer cells,” Biological Research, vol. 52, no. 1, p. 37, 2019.View at: Publisher Site | Google Scholar
H. Harjunpää, M. Llort Asens, C. Guenther, and S. C. Fagerholm, “Cell adhesion molecules and their roles and regulation in the immune and tumor microenvironment,” Frontiers in Immunology, vol. 10, 2019.View at: Publisher Site | Google Scholar
S. Wang, C. Yin, Y. Zhang et al., “Overexpression of ICAM-1 predicts poor survival in high-grade serous ovarian carcinoma: a study based on TCGA and GEO databases and tissue microarray,” BioMed Research International, vol. 2019, 9 pages, 2019.View at: Publisher Site | Google Scholar
Y. Vedvyas, J. E. McCloskey, Y. Yang et al., “Manufacturing and preclinical validation of CAR T cells targeting ICAM-1 for advanced thyroid cancer therapy,” Scientific Reports, vol. 9, no. 1, pp. 1–15, 2019.View at: Publisher Site | Google Scholar
A. Saplonţai-Pop, A. Moţ, M. Moldovan et al., “Testing antiplatelet and antioxidant activity of the extract of seven varieties of Allium cepa L,” Open Life Sciences, vol. 10, no. 1, pp. 89–98, 2015.View at: Publisher Site | Google Scholar
W. Wang, D. Li, L. Xiang et al., “TIMP-2 inhibits metastasis and predicts prognosis of colorectal cancer via regulating MMP-9,” Cell Adhesion & Migration, vol. 13, no. 1, pp. 273–284, 2019.View at: Publisher Site | Google Scholar