Fucoxanthinol Promotes Apoptosis in MCF-7 and MDA-MB-231 Cells by Attenuating Laminins–Integrins Axis

Simple Summary: Carotenoids are potential candidates for preventing breast cancer (BC), a major malignancy affecting women worldwide with high incidence and mortality rates. Some studies have demonstrated that fucoxanthin, a marine carotenoid, and its major metabolite fucoxanthinol (FxOH), promoted apoptosis in representative human BC cells (MCF-7 and MDA-MB-231 cells). However, the effects of Fx and FxOH in those cells still remain fragmentary. Herein, we investigated the comprehensive mechanisms underlying FxOH-induced apoptosis in MCF-7 and MDA-MB-231 cells. Consequently, it was suggested that FxOH promoted apoptosis in MCF-7 and MDA-MB-231 cells by modulating the extracellular matrix–integrin axis, and the downstream signals: cell cycle, STAT, TGF- β , RAS/Rho, MAPK, and/or DNA repair. Thus, FxOH may exert preventive effects on BC by modulating some core signals involved in apoptosis induction. Abstract: Fucoxanthinol (FxOH), the main metabolite of the marine carotenoid fucoxanthin, exerts anti-cancer effects. However, fragmentary information is available on the growth-inhibiting effects of FxOH on breast cancer (BC). We investigated the growth-inhibiting effects of FxOH on human BC cells (MCF-7 and MDA-MB-231 cells), and the underlying mechanisms, differently from previous studies, by using comprehensive transcriptome analysis. The molecular mechanisms of FxOH were evaluated using ﬂow cytometry, microarray, Western blotting, and gene knockdown analyses. FxOH (20 µ M) signiﬁcantly induced apoptosis in MCF-7 and MDA-MB-231 cells. Transcriptome analysis revealed that FxOH modulated the following 12 signaling pathways: extracellular matrix (ECM), adhesion, cell cycle, chemokine and cytokine, PI3K/AKT, STAT, TGF- β , MAPK, NF- κ B, RAS/Rho, DNA repair, and apoptosis signals. FxOH downregulated the levels of laminin β 1, integrin α 5, integrin β 1, integrin β 4, cyclin D1, Rho A, phosphorylated (p)paxillin (Tyr 31 ), pSTAT3(Ser 727 ), and pSmad2(Ser 465/467 ), which play critical roles in the 12 signaling pathways mentioned above. Additionally, FxOH upregulated the levels of pERK1/2(Thr 202 /Tyr 204 ) and active form of caspase-3. Integrin β 1 or β 4 knockdown signiﬁcantly inhibited the growth of MCF7 and MDA-MB-231 cells. These results suggest that FxOH induces apoptosis in human BC cells through some core signals, especially the ECM–integrins axis, and the downstream of cell cycle, STAT, TGF- β , RAS/Rho, MAPK, and/or DNA repair signals.


Introduction
According to the GLOBOCAN database for 2020, female breast cancer (BC) is estimated to be the most common cancer among newly diagnosed cancer cases worldwide (approximately 2.3 million new cases). In addition, female BC is the leading cause of cancer in women, with high incidence and mortality rates [1]. In the United States, the incidence of BC is increasing, and is estimated to remain higher than that of other cancers by 2040 [2]. The 5 year relative survival rates of BC patients with localized, regional, and distant stages are 99%, 86%, and 29%, respectively (combined survival rates of all stages = 90%) [3].
Carotenoids, which are a category of tetraterpenoids and fat-soluble pigments, are abundant in fruits, vegetables, and algae, and confer health benefits to humans. The pigments are important phytochemicals for antioxidants with radical scavenging [4] and singlet oxygen-quenching potentials [5]. Increasing evidence from human studies suggests that some carotenoids, such as α-carotene, β-carotene, and lycopene, which are found in various fruits and vegetables, exert preventive effects. Therefore, several epidemiological approaches aimed to investigate the preventive effects of carotenoid intake on BC. However, the evaluation of carotenoids as anti-cancer agents is still classified as "Limited-suggestive decreases risk" [6]. On the other hands, carotenoids, including α-carotene, β-carotene, lycopene, lutein, β-cryptoxanthin, violaxanthin, neoxanthin, and fucoxanthin (Fx), ingested by humans exhibit a wide range of polarity. To date, the preventive effects of highly polar carotenoids, such as neoxanthin and Fx, on BC have not been previously investigated.
The highly polar carotenoid Fx is a representative marine carotenoid and exhibits photosynthetic and photoprotective activities in brown algae and microalgae [7]. Fx is abundant in dietary marine algae such as Undaria pinnatifida (Japanese name, wakame) and Himanthalia elongata (sea spaghetti) [8], with a wide range of 0.3-18.6 mg Fx/g dry weight.
This study investigated the effects of FxOH on the transcriptome profiles of MCF-7 and MDA-MB-231 cells, and elucidated the novel molecular mechanisms differently than the previous reports.

Cell Viability Assay
MCF-7 and MDA-MB-231 cells were seeded into a 24-well plate containing culture medium at a density of 5 × 10 4 cells/mL (2.5 × 10 4 cells/well). The cells were allowed to adhere for 1 day. The medium was then replaced with fresh culture medium containing FxOH (final concentration: 5.0 or 20.0 µM) or vehicle alone (dimethyl sulfoxide (DMSO)), and the cells were incubated for 1 or 2 days. Cell viability was measured using a WST-1 assay. The absorbance at 450 nm of the mixture was measured using an enzyme-linked immunosorbent assay plate reader TECAN (TECAN Japan, Tokyo, Japan).

Analysis of Apoptosis-Associated Nuclear Alteration
MCF-7 and MDA-MB-231 cells were seeded into a 24-well plate containing culture medium at a density of 5 × 10 4 cells/mL (2.5 × 10 4 cells/well). The cells were allowed to adhere for 1 day. The medium was then replaced with fresh culture medium containing FxOH (final concentration: 20.0 µM) or vehicle alone (DMSO), and the cells were incubated for 2 days. The cells were incubated with Hoechst33342 (Ho342, Dojindo Laboratories, Kumamoto, Japan) at 37 • C for 10 min. Apoptosis-associated chromatin condensation and nuclear fragmentation were assessed using the fluorescence microscope Nikon TE2000 (Nikon, Melville, NY, USA).

Analyses of Apoptotic-Like Cell Body and Cell Cycle Phases
MCF-7 and MDA-MB-231 cells were seeded into 10 cm plates containing culture medium at a density of 5 × 10 4 cells/mL (50 × 10 4 cells/plate). The cells were allowed to adhere for 1 day. The medium was then replaced with fresh culture medium containing FxOH (final concentration: 20.0 µM) or vehicle alone (DMSO), and the cells were incubated for 2 days. The cells were dissociated into a single-cell suspension, fixed with cold 70% ethanol for 30 min, incubated with ribonuclease A (Nacalai Tesque, Kyoto, Japan) at 37 • C for 20 min, and stained with propidium iodide (Sigma-Aldrich, St Louis, MO, USA) at 4 • C for 30 min. The number of cells with apoptosis-like bodies (sub-G1) and at different cell cycle phases (G1, S, and G2/M) were counted using a FACSaria-III flow cytometer (BD Biosciences, San Jose, CA, USA).

Extraction and Purification of Total RNA
MCF-7 and MDA-MB-231 cells were seeded into 10 cm plates containing cell culture medium at a density of 5 × 10 4 cells/mL (50 × 10 4 cells/plate). The cells were allowed to adhere for 1 day. The medium was then replaced with fresh culture medium containing FxOH (final concentration: 20.0 µM) or vehicle alone (DMSO), and the MCF-7 and MDA-MB-231 cells were incubated for 2 and 1 days, respectively. The cells were trypsinized, washed twice with phosphate-buffered saline (PBS), incubated with RNA later overnight at 4 • C, and stored at −80 • C until total RNA extraction. Total RNA was extracted and purified using RNeasy Mini Kit, RNase-Free DNase Set, and QIA shredder (QIAGEN, Valencia, CA, USA), following the manufacturer's instructions. The concentration of RNA was determined using a NanoDrop ND-1000 (NanoDrop, Wilmington, DE, USA). Additionally, total RNA was quantified using agarose gel electrophoresis with DynaMarker RNA High for Easy Electrophoresis.

Microarray Analysis
Gene expression was comprehensively analyzed using the microarray Clariom S human assay with an optimal enzyme and reagent kit (Thermo Fisher Scientific, Carlsbad, CA, USA). Total RNA (500 ng) was mixed with poly(A) control RNAs. First-strand complementary DNA (cDNA) was enzymatically synthesized from the total and control RNA mixtures, followed by the synthesis of double-stranded cDNA from the first-strand cDNA. Single-stranded complementary RNA (cRNA) was generated from double-stranded cDNA using an in vitro transcription method. Second-cycle single-strand-cDNA was generated from a single-strand cRNA template. The resulting single-stranded cDNA was enzymatically fragmented, labeled with biotin, and hybridized to a Clariom S Human Array. The microarray was washed, stained with the accessory reagents using Affymetrix Fluidics Station 450 (Affymetrix, Santa Clara, CA, USA), and scanned using the Affymetrix GeneChip Scanner 3000 system (Affymetrix). Gene expression profiles were measured using Transcriptome Analysis Console (TAC) software (version 4.0.2; Applied Biosystems, Foster City, CA, USA). The significant differentially expressed genes between FxOH-treated and control cells were extracted based on the following criteria: fold-change, ≥2.0 or ≤−2.0-fold; p < 0.05 (one-way analysis of variance (ANOVA); exact p values (obtained using an exact test with edge R in the TAC software)). Principal coordinate analysis (PCoA) plots, volcano plots, hierarchical clustering, and the distribution of the top 30 gene sets were displayed using TAC software, based on Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. Gene set enrichment analysis (GSEA) was performed using GSEA software (ver. 4.0.3; Broad Institute of Harvard University and Massachusetts Institute of Technology, Cambridge, MA, USA) [27,28].

Analysis of Protein Expression and Activation
MCF-7 and MDA-MB-231 cells were seeded into 10 cm plates containing culture medium at a density of 5 × 10 4 cells/mL (50 × 10 4 cells/plate). The cells were allowed to adhere for 1 day. The medium was then replaced with fresh culture medium containing FxOH (final concentration: 20.0 µM) or vehicle alone (DMSO), and the MCF-7 and MDA-MB-231 cells were incubated for 2 and 1 days, respectively. The cells were trypsinized, washed twice with PBS, and stored at −80 • C until total protein extraction. Whole proteins were lysed using a lysis buffer. The protein concentration in the lysate was determined using Bradford assay (Bio-Rad, Hercules, CA, USA). Whole cellular proteins (10 µg) were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis using a 10% gel. The resolved proteins were transferred onto a Hybond polyvinylidene difluoride membrane (Amersham Bioscience, Little Chalfont, UK). The membrane was washed with Tris-buffered saline containing 0.1% Tween 20 (TBS-T), and incubated with 1% (w/v) bovine serum albumin (BSA) in TBS-T (1% BSA/TBS-T) at room temperature for 1 h. Next, the membrane was incubated with primary antibodies in 1% BSA/TBS-T at 4 • C overnight, followed by incubation with horseradish peroxidase-conjugated anti-mouse or anti-rabbit secondary antibodies in TBS-T at room temperature for 1 h. Immunoreactive bands were visualized using a luminol-enhanced chemiluminescence assay (Millipore, Billerica, MA, USA). After the observation, the membrane was re-probed and washed with TBS-T, and then applied to the next protein measurement. β-actin expression was used as a loading control per proteins set at different membrane.

Gene Knockdown Experiments
Twenty-seven mer of Dicer-substrate short interfering RNAs (dsiRNAs) targeting the coding sequences of integrin β1 and β4 mRNAs in Homo sapiens were designed by Integrated DNA Technologies (Coralville, IA, USA). The designed integrin β1 and β4 dsiRNAs were as follows: integrin β1 dsRNA, 5 -ACU CUU GUC AGC UAA GGU CAC AUT G-3 ; integrin β4 dsiRNA, 5 -CGA GAA GCU UCA CAC CUA UUU CCC T-3 ; negative control (NC), 5 -GUG UUC UAC ACC AUU ACU CAA UUC UUA-3 . The dsiRNA/Lipofectamine RNAiMAX complex was prepared in Opti-MEM I, following the manufacturer's instructions. MCF-7 and MDA-MB-231 cells were seeded into 10 cm plates containing cell culture medium at a density of 6 × 10 4 cells/mL (60 × 10 4 cells/plate). The cells were allowed to adhere for 1 day. The dsiRNA or NC/Lipofectamine RNAiMAX complex was added to 10 mL of culture medium (final concentration of dsiRNA or NC: 10 nM) for 1 day. Integrin β1 or β4 knockdown MCF-7 and MDA-MB-231 cells were seeded into a 24-well plate at a density of 5 × 10 4 cells/mL (2.5 × 10 4 cells/well), and allowed to adhere for 1 day. The growth of the treated cells was measured using a WST-1 assay.

Effects of FxOH on the Transcriptome Profiles of MCF-7 Cells
Next, the effects of 20 µM FxOH on the transcriptome profiles of MCF-7 cells were evaluated. The PCoA plot revealed that the transcriptome profile of FxOH-treated cells (group 1) was distinct from that of control cells (group 2) ( Figure 3A). Hierarchical cluster analysis of 3545 genes revealed different clusters between groups 1 and 2 ( Figure 3B). Volcano plots of differentially expressed genes between group 1 and group 2 revealed that the frequency of significantly upregulated genes with both high fold-change and p-value was higher than that of downregulated genes ( Figure 3C). However, the number of downregulated genes (1966 genes) was higher than that of upregulated genes (1579 genes) ( Figure 3D). Pathway analysis demonstrated that, of the top 30 pathways, 18 were associated with cancer development, including vascular endothelial growth factor receptor (VEGFR), micro RNA regulation (miR), nuclear receptors, phosphatidylinositol-3 kinase/protein kinase B (PI3K/AKT), endothelin, adhesion, cell cycle, mitogen-activated protein kinase (MAPK), interleukin (IL)-18, integrated BC, NF-E2-related factor 2 (NRF2), glia-cell-derived neurotrophic factor (GDNF), and epidermal growth factor receptor (EGFR) ( Figure 3E, black circle).

Effects of FxOH on the Transcriptome Profiles of MDA-MB-231 Cells
The effects of 20 µM FxOH on the transcriptome profiles of MDA-MB-231 cells were also evaluated. The PCoA plot revealed that the transcriptome profile of FxOH-treated cells (group 1) was distinct from that of control cells (group 2) ( Figure 4A). Hierarchical clustering analysis of 2995 differentially expressed genes revealed different clusters between groups 1 and 2 ( Figure 4B). Volcano plots of differentially expressed genes between groups 1 and 2 revealed that the frequency of significantly upregulated genes with both high foldchange and p-value was higher than that of downregulated genes ( Figure 4C). However, the number of downregulated genes (1702 genes) was higher than that of upregulated genes (1293 genes) ( Figure 4D). Pathway analysis demonstrated that, of the top 30 pathways, 21 were associated with cancer development, including VEGFR, microRNA regulation, nuclear receptors, PI3K/AKT, adhesion, cell cycle, endothelin, MAPK, IL-18, transforming growth factor (TGF)-β, integrated breast cancer, DNA repair, DNA damage, small cell lung cancer, GDNF, insulin, and vitamin D receptor ( Figure 4E, black circle).

Bioinformatics Analyses of Transcriptome Profiles of FxOH-Treated MCF-7 and MDA-MB-231 Cells
GSEA of 3545 differentially expressed genes in FxOH-treated MCF-7 cells revealed that among cancer-related signaling pathways, cell cycle, DNA replication, cytokine-cytokine receptor interaction, and protein export were strongly enriched, whereas extra cellular matrix (ECM)-receptor interaction, NOD-like receptor signaling pathway, and MAPK signaling pathway were weakly enriched (Table 1 and Figure 5). Meanwhile, GSEA analysis of 2995 differentially expressed genes in FxOH-treated MDA-MB-231 cells revealed that among cancer-related signaling pathways, DNA replication and MAPK signaling pathways were strongly enriched, whereas cell adhesion molecular cell adhesion molecules (CAMS), ECM-receptor interaction, cytokine-cytokine receptor interaction, insulin signaling pathway, and toll-like receptor signaling pathway were weakly enriched (Table 2 and Figure 6).

Discussion
This study demonstrated that FxOH significantly induced apoptosis in MCF-7 and MDA-MB-231 cells by suppressing core genes, proteins, and signaling pathways, especially the laminins/integrins axis. This is a novel report suggesting the comprehensive mechanisms underlying pro-apoptotic effects of FxOH in human BC cells.
Laminin β1, a major component of ECM proteins, forms a heterotrimeric structure with the other laminin chains, such as α and γ chains. In addition to serving as a component of basement membrane, laminin β1 is involved in tumor development, metastasis, and invasion [32][33][34]. The upregulated expression of laminin β1 in BC is positively correlated with malignancy [34]. The laminins-integrins (e.g., α3, α6, and β4) axis plays an essential role in the adhesion of epithelial cells. In particular, integrin β1 functions as a central adhesion molecule in the interactions involving laminins, collagens, and Arg-Gly-Asp (RGD) peptides [35]. The laminins-integrins axis is a critical trigger that positively or negatively regulates downstream molecules and signals, such as microtubule assembly, paxillin, FAK, steroid receptor coactivator, Ras/Rho, PI3K/AKT, MAPK, NF-κB, NRF2, TP53, mitochondrial apoptosis pathway, reactive oxygen species production, and microRNA regulation [36][37][38]. Aberrant regulation of integrin signaling in the tumor microenvironment is associated with the enhancement of stemness, epithelial-mesenchymal transition, and anoikis resistance in cancer epithelial cells, which enables the accumulation of inflammatory cells and cancer-associated fibroblasts in tumor tissues [36,39]. In the present study, the several other genes encoding laminins, collagens, and integrins are downregulated in FxOH-treated MCF-7 and MDA-MB-231 cells, in addition to genes encoding laminin β1, integrin α5, integrin β1, and integrin β4 (Figure 7 and Figure S1, Tables S2 and S4). Therefore, it was suggested that the suppression of the ECM-adhesion signal was a key biological process for apoptosis or anoikis induction in the FxOH-treated MCF-7 and MDA-MB-231 cells. Interestingly, the activation of ERK1/2 was observed in the both FxOH-treated MCF-7 and MDA-MB-231 cells (Figure 7 and Figure S1). Recently, increasing evidence demonstrated that the activation of ERK regulates dual functions of cell growth and apoptosis in cancer cells by various regulatory factors. Sugiura et al. reported that many natural compounds, such as γ-tocotrienol, curcumin, and shikonin, induced apoptosis with ERK activation [40]. The downregulations of some ERK-suppressing phosphatases, such as serine/threonine phosphatases, tyrosine phosphatases, and dual-specificity phosphatases, leads to ERK1/2 activation in the cytoplasm and nucleus, followed by the induction of apoptosis in cancer cells [40,41]. These ERK-suppressing phosphatases may contribute to the ERK1/2 activation in the FxOH-treated MCF-7 and MDA-MB-231 cells. Furthermore, FxOH downregulated BRCA1 expression in MDA-MB-231 cells, but not in MCF-7 cells (Figure 7 and Figure S1). The inheritable mutation in BRCA1, a DNA break repair protein-encoding gene, is a risk factor for BC [42][43][44]. However, the reasons for differential regulation of BRCA1 by FxOH in MDA-MB-231 and MCF-7 cells are unclear. In addition, FxOH enhances the chromatin condensations, nuclear fragmentations, sub-G1 ratios, and caspase-3 activations in both cells, with the significant downregulation of many genes relat-ing to DNA repair and damage ( Figure 2B,C, Figure 3E, Figure 4E, Figures 7 and S1). These results were consistent with the previous data on MCF-7 and MDA-MB-231 cells [21,22]. The alteration of DNA damage and repair signals may be the frequently altering signals during apoptosis induction in MDA-MB-231 and MCF-7 cells with FxOH treatment.
BC is classified into different subtypes: luminal A, luminal B, HER2-enriched, normallike, and several types of triple negative breast cancers (TNBCs), based on the expression status of estrogen receptors (ERα), human epidermal growth factor receptor 2 (HER2), progesterone receptors (PR), and others. The clinical outcome of TNBCs (ERα − , HER2 − , and PR − ), which constitute 10%-20% of all BCs, is poorer than those of other subtypes [45][46][47][48][49][50][51]. The subtype and molecular characteristics of the cell lines used in this study are as follows: MCF-7 cells, luminal A subtype, ER + , PR + , HER2 − , BRCA1 wildtype , and TP53 wildtype ; MDA-MB-231 cells, TNBC subtype, BRCA1 wildtype , and TP53 mutant [52,53]. Additionally, the invasive capacity of MDA-MB-231 cells is higher than that of MCF-7 cells [54]. These findings indicate that the molecular profiles and malignant properties of MCF-7 cells are distinct from those of MDA-MB-231 cells, which may explain the differential susceptibility of the two cell lines to FxOH. The anti-proliferative effect of FxOH on MDA-MB-231 cells was higher than that on MCF-7 cells (Figure 2A). However, we speculated that FxOH induced apoptosis in both cell types through similar mechanisms, irrespective of cellular characteristics, such as ER, PR, HER2, BRCA1, TP53, and invasion. ECM-integrins are key targets for FxOH. FxOH induces apoptosis or anoikis in human colorectal, mouse pancreatic, and hamster pancreatic cancer cells by attenuating integrin signals [29][30][31]55]. Additionally, FxOH downregulates a chloride intracellular channel 4 signal involved in integrin trafficking and cell adhesion [56]. FxOH may also attenuate the ECM-integrin axis by regulating the trafficking of integrins to the cellular membrane. Furthermore, Fx exerts an anti-fibrotic effect in nasal polyp-derived fibroblasts by suppressing ECM, TGF-β, and PI3K/AKT signaling [57]. Fx and FxOH may target the ECM, irrespective of the cell type. Further investigations are needed to elucidate the molecular mechanisms underlying the effects of FxOH on human BC cells.