Marinobazzanan, a Bazzanane-Type Sesquiterpenoid, Suppresses the Cell Motility and Tumorigenesis in Cancer Cells

Marinobazzanan (1), a new bazzanane-type sesquiterpenoid, was isolated from a marine-derived fungus belonging to the genus Acremonium. The chemical structure of 1 was elucidated using NMR and mass spectroscopic data, while the relative configurations were established through the analysis of NOESY data. The absolute configurations of 1 were determined by the modified Mosher’s method as well as vibrational circular dichroism (VCD) spectra calculation and it was determined as 6R, 7R, 9R, and 10R. It was found that compound 1 was not cytotoxic to human cancer cells, including A549 (lung cancer), AGS (gastric cancer), and Caco-2 (colorectal cancer) below the concentration of 25 μM. However, compound 1 was shown to significantly decrease cancer-cell migration and invasion and soft-agar colony-formation ability at concentrations ranging from 1 to 5 μM by downregulating the expression level of KITENIN and upregulating the expression level of KAI1. Compound 1 suppressed β-catenin-mediated TOPFLASH activity and its downstream targets in AGS, A549, and Caco-2 and slightly suppressed the Notch signal pathway in three cancer cells. Furthermore, 1 also reduced the number of metastatic nodules in an intraperitoneal xenograft mouse model.


Introduction
There were an estimated 19.3 million new cancer cases in 2020 and almost 10 million cancer deaths worldwide [1]. Lung cancer is the leading cause of cancer death, with colorectal cancer being the second. Further, gastric cancer is the fourth most common cause of cancer death globally [2]. The World Health Organization estimates that there will be 28.4 million cancer cases worldwide by 2040 [1].
Metastasis refers to secondary tumors which develop in a different part of the body compared to the original cancer. Epithelial-to-mesenchymal transition (EMT) is one of the reasons for metastasis. The invasiveness and metastatic potential of solid tumors increase with EMT. The EMT transcription factors, Snail, Slug, and Twist, are promoted by EMT processes [3]. The Wnt/β-catenin pathway, which promotes the stemness, deterioration, processes [3]. The Wnt/β-catenin pathway, which promotes the stemness, deteri and metastasis of cancer cells, could be another contributing factor to cancer. βpromotes the transcription of a wide range of oncogenes, including c-Myc and Cy in the nucleus [4]. Several human malignancies exhibit an increased ability to inva metastasize when KAI1 (a metastatic suppressor gene) expression is suppressed a ENIN is promoted. Cancer invasion and metastasis are also mediated through t ENIN/AP-1 axis, another signal transduction component [5]. The Notch signalin way is found in a wide range of solid tumors and can be responsible for both cel eration and metastasis, including EMT. The Notch ligand is a single transmembra tein, and activation of Notch occurs when it binds to the Notch ligand of neighborin The transcriptional targets of HES genes can be also regulated by Notch [6]. Th suppressing EMT, the Wnt/β-catenin pathway, KITENIN/AP-1 axis, and the Notch pathway has become an important goal for developing anticancer therapeutics.
The biological and geochemical roles of marine fungi have attracted the atten researchers in many scientific communities [7,8]. Marine fungi and marine-derive are regarded as prolific sources of natural products with unique chemical structu diverse biological activities [9][10][11]. Therefore, many research groups have focuse attention on culturing marine fungi to discover novel natural products [12][13][14]. In ular, the marine-derived genus Acremonium has been studied intensively as a pr producer of natural products with a wide range of bioactivities [15,16]. Previously, pentadepsipeptide, acremonamide, with wound-healing properties has been i from the Acremonium strain CNQ-049 [17].
Thus, as part of continuing efforts to investigate the chemical components of th monium strain CNQ-049, derived from marine sediments collected off the coast of ern California, we isolated a new bazzanane-type sesquiterpenoid: marinobazza Although the cancer-cell cytotoxicity of sesquiterpene lactones and their applicat developing anticancer agents have been extensively explored [18], little is know the bioactivities of bazzanane-type sesquiterpenoids. Hence, the isolation, structu cidation, and anti-cancer activities of compound 1 (Figure 1) are examined herein

Results and Discussion
Marinobazzanan (1) was obtained as a pale-yellow amorphous powder. The molecular formula of 1 was determined as C 15 Figures S1 and S2). The molecular formula and the HSQC NMR spectroscopic data of 1 suggested that compound 1 possessed a bicyclic ring system. 3), C-4 (δ C 134.0), C-6 (δ C 36.8), and C-14 (δ C 17.1). In addition, the HMBC NMR correlations from H-4 (δ H 6.66) to the carbonyl carbon C-12 (δ C 172.2) allowed the connection of the amide group to ring B. Meanwhile, the connection of C-6/C-7 was secured by observing the HMBC correlations from H-13 (δ H 1.20) to C-6 (δ C 36.8), C-7 (δ C 48.5) and from H-14 (δ H 0.92) to the same two carbons. Furthermore, the attachment of a hydroxy group at C-10 was assigned through the presence of the carbon chemical shifts at δ C 75.6. In addition, a chlorine atom at C-9 was established by considering the isotope ratio (3:1) of the two protonated adduct [M + H] + and [M + H+2] + in the low-resolution electrospray ionization mass spectrum (LR-ESI-MS) and the chemical shift at δ C 69.2 in the chlorinated methine, which completed the planar structure of 1, as shown in Figure 2. The relative stereochemistry of 1 was determined by analysis of the NOESY NMR spectroscopic data. A NOESY correlation ( Figure S6) between H-9 (δH 4.29) and H-14 (δH 0.92) indicated that these protons should be located on the same face of the molecule Meanwhile, a NOESY crosspeak between H-10 (δH 3.86) and H-13 (δH 1.20) as well as its The relative stereochemistry of 1 was determined by analysis of the NOESY NMR spectroscopic data. A NOESY correlation ( Figure S6) between H-9 (δ H 4.29) and H-14 (δ H 0.92) indicated that these protons should be located on the same face of the molecule. Meanwhile, a NOESY crosspeak between H-10 (δ H 3.86) and H-13 (δ H 1.20) as well as its coupling constant ( 3 J H−H = 10.0 Hz) established their syn relationship. Therefore, the relative configuration of 1 was assigned as 6S*, 7S*, 9S*, and 10S* (Figure 2b).
The absolute configuration of 1 was established using the modified Mosher's method [19,20], combined with a comparison between the measured and calculated vibrational circular dichroism (VCD) spectra [21]. First, esterification of 1 with (R)-and (S)-MTPA-Cl (α-methoxy-α-(trifluoromethyl) phenylacetyl chloride) yielded the (S)-and (R)-MTPA esters of 1 (1a and 1b, respectively). Analysis of the 1 H NMR ( Figures S7 and S8) ∆δ (S−R) values revealed a consistent sign distribution, thus verifying the R configuration at C-10 ( Figure 3). Therefore, the absolute stereochemistry of the three chiral centers of ring A in 1 was determined as 7R, 9R, and 10R. The absolute configurations of C-6 were also confirmed by comparing the experimental and calculated VCD spectra. The VCD spectra were calculated using density functional theory (DFT) at the B3LYP/6-31+G(d) level using the Gaussian 09 software (Gaussian, Inc., Wallingford, CT 06492, United States) and the calculated VCD spectrum of the (6R, 7R) configuration showed good agreement with the experimental spectrum of 1, with a confidence level of 87% ( Figure 4). The relative stereochemistry of 1 was determined by analysis of the NOES spectroscopic data. A NOESY correlation ( Figure S6) between H-9 (δH 4.29) and H 0.92) indicated that these protons should be located on the same face of the m Meanwhile, a NOESY crosspeak between H-10 (δH 3.86) and H-13 (δH 1.20) as we coupling constant ( 3 JH−H = 10.0 Hz) established their syn relationship. Therefore, t tive configuration of 1 was assigned as 6S*, 7S*, 9S*, and 10S* (Figure 2b).
The absolute configuration of 1 was established using the modified Mosher's [19,20], combined with a comparison between the measured and calculated vib circular dichroism (VCD) spectra [21]. First, esterification of 1 with (R)-and (S)-M (α-methoxy-α-(trifluoromethyl) phenylacetyl chloride) yielded the (S)-and (R)-M ters of 1 (1a and 1b, respectively). Analysis of the 1 H NMR ( Figure S7-S8) Δδ(S−R revealed a consistent sign distribution, thus verifying the R configuration at C-10 3). Therefore, the absolute stereochemistry of the three chiral centers of ring A i determined as 7R, 9R, and 10R. The absolute configurations of C-6 were also co by comparing the experimental and calculated VCD spectra. The VCD spectra w culated using density functional theory (DFT) at the B3LYP/6-31+G(d) level us Gaussian 09 software (Gaussian, Inc., Wallingford, CT 06492, United States) and culated VCD spectrum of the (6R, 7R) configuration showed good agreement w experimental spectrum of 1, with a confidence level of 87% ( Figure 4).  Bazzanane-type sesquiterpenes have been mainly found in liverworts, including Bazzania genus. After the first reported isolation of bazzanene from Bazzania pomp (Lac.) Mitt. in 1969 [22], only a few bazzanane-type sesquiterpenes have been repo indicating that they are very rare in nature [23,24]. The most similar compound to 1 isolated from the New Zealand liverwort Frullania falciloba [25]. However, previousl Bazzanane-type sesquiterpenes have been mainly found in liverworts, including the Bazzania genus. After the first reported isolation of bazzanene from Bazzania pompeana (Lac.) Mitt. in 1969 [22], only a few bazzanane-type sesquiterpenes have been reported, indicating that they are very rare in nature [23,24]. The most similar compound to 1 was isolated from the New Zealand liverwort Frullania falciloba [25]. However, previously reported bazzanane-type sesquiterpenes are neither chlorinated nor possess an amide group in the molecule, as in 1. Furthermore, 1 is the first bazzanane-type sesquiterpenoid to be isolated from a strain of the genus Acremonium and the first reported chlorinated bazzanane-type sesquiterpene with an amide group in its structure.
A methyl thiazolyl tetrazolium (MTT) cytotoxicity bioassay was used to evaluate the effect of 1 treatment in various concentrations (10,25,50, and 100 µM) on the viability of AGS (gastric cancer), A549 (lung cancer), and Caco-2 (colorectal cancer) cells. The cell viability of A549 was unaffected during the treatment with 10-50 µM of 1 for 48 h; however, the viability decreased marginally during the treatment with 100 µM of 1, as shown in Figure 5. Similarly, the viability of AGS and Caco-2 did not decrease with treatment of 10-25 µM of 1 but decreased significantly at concentrations of 50-100 µM. Thus, these observations demonstrate that treatment with 1 is relatively non-toxic toward A549, AGS, and Caco-2 cells at concentrations less than 25 µM. Cell viability was measured using MTT assay. The data represent the mean ± standard devi n = 3. * p < 0.05; ** p < 0.01.
Next, migration and invasion assays were performed using non-toxic concentra (1, 2.5, and 5 μM) to determine whether 1 inhibits the motility of A549, AGS, and C cells. As shown in Figure 6, 1 displayed a dose-dependent inhibitory effect on the m tion of all three cell types at concentrations from 1 to 5 μM (Figure 6a,b) alongside dependent inhibition of invasion by each cell type by up to ~45% at 5 μM concentra (Figure 6c,d).
In addition, the potential anti-tumorigenic activity of 1 was evaluated by exam the soft agar colony formation of A549, AGS, and Caco-2 cells exposed to non-toxic centrations (1, 2.5, and 5 μM). As shown in Figure 6e  Cell viability was measured using MTT assay. The data represent the mean ± standard deviation, n = 3. * p < 0.05; ** p < 0.01.
Next, migration and invasion assays were performed using non-toxic concentrations (1, 2.5, and 5 µM) to determine whether 1 inhibits the motility of A549, AGS, and Caco-2 cells. As shown in Figure 6, 1 displayed a dose-dependent inhibitory effect on the migration of all three cell types at concentrations from 1 to 5 µM (Figure 6a,b) alongside dose-dependent inhibition of invasion by each cell type by up to~45% at 5 µM concentrations (Figure 6c,d).
In addition, the potential anti-tumorigenic activity of 1 was evaluated by examining the soft agar colony formation of A549, AGS, and Caco-2 cells exposed to non-toxic concentrations (1, 2.5, and 5 µM). As shown in Figure 6e To determine whether the suppression of A549, AGS, and Caco-2 cell motility a tumorigenicity in the presence of 1 involves the epithelial-mesenchymal transition (EM the expression of EMT effectors and transcription factors were examined. As shown Figure 7a, 1 decreased mRNA expression of the mesenchymal marker N-cadherin but creased that of the epithelial marker E-cadherin in all three cell types. Further, 1 sign cantly decreased the mRNA expression of the EMT transcription factors, Snail, Slug, a Twist, in all three cell types, as shown in Figure 7b. In summary, these results indicate t 1 modulates the expression of the EMT effector N-cadherin by downregulating the tr scription factors Snail, Slug, and Twist. To examine whether the suppression of A549, AGS, and Caco-2 cell motility and tumorigenicity in the presence of 1 involves KITENIN and AP-1, the protein and mRNA expression levels of KITENIN and AP-1, including their activities, were examined. Epidermal growth factor increases KITENIN-mediated AP-1 activity, and there is an inverse relationship between KAI and KITENIN [26]. As shown in Figure 8a, at a concentration of 5 μM, 1 suppressed AP-1 activity. Figure 8b,c show that 1 suppressed the activity of the KITENIN 3′-UTR, while the KITENIN promoter did not show significant change. Figure  8d,e indicates that the protein level of KITENIN decreased in A549, AGS, and Caco-2 during treatment. The mRNA expression level of KITENIN was also suppressed in A549, AGS, and Caco-2, whereas the mRNA level of KAI1 was increased by 1 (Figure 8f). As a result, 1 decreased cell motility by downregulating the expression level of KITENIN while upregulating that of KAI1. (b) relative mRNA expression of the EMT transcription factors Snail, Slug, and Twist. The mRNA levels were normalized against that of glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Data represent the mean ± standard deviation, n = 3. * p < 0.05; ** p < 0.01; *** p < 0.001.
To examine whether the suppression of A549, AGS, and Caco-2 cell motility and tumorigenicity in the presence of 1 involves KITENIN and AP-1, the protein and mRNA expression levels of KITENIN and AP-1, including their activities, were examined. Epidermal growth factor increases KITENIN-mediated AP-1 activity, and there is an inverse relationship between KAI and KITENIN [26]. As shown in Figure 8a, at a concentration of 5 µM, 1 suppressed AP-1 activity. Figure 8b,c show that 1 suppressed the activity of the KITENIN 3 -UTR, while the KITENIN promoter did not show significant change. Figure 8d,e indicates that the protein level of KITENIN decreased in A549, AGS, and Caco-2 during treatment. The mRNA expression level of KITENIN was also suppressed in A549, AGS, and Caco-2, whereas the mRNA level of KAI1 was increased by 1 (Figure 8f). As a result, 1 decreased cell motility by downregulating the expression level of KITENIN while upregulating that of KAI1. We performed TOPFLASH reporter assays to assess whether treatment with 1 modulates β-catenin-mediated and/or KITENIN-mediated signaling activity. Treatment with 5 μM of 1 significantly decreased β-catenin-mediated TOPFLASH activity on HEK293T by 30% (Figure 9a). In addition, Figure 9b-d indicates that treatment with 1 decreased mRNA expression and the protein level of β-catenin on AGS and Caco-2 cells; however, treatment with 1 did not significantly affect the mRNA expression and protein level of βcatenin in A549 cells.
The protein level of total, cytoplasmic, and nuclear β-catenin was examined to test whether 1 affected the nuclear/cytoplasmic distribution of β-catenin in A549 cells. As shown in Figure 10a,b, treatment with 5 μM of 1 did not affect the level of total β-catenin, whereas it decreased the β-catenin nuclear to cytoplasmic ratio remarkably in A549 cells compared to DMSO. To further test the effect of 1 on the levels of downstream target genes of β-catenin, qRT-PCR analysis was performed. As shown in Figure 10c, treatment with 5 μM of 1 suppressed the mRNA expression of cyclin-D1 in A549 and CD44 in Caco-2, while treatment with 5 μM of 1 suppressed the mRNA expression of c-Myc and CD44 in AGS cells for 24 h. Moreover, the β-catenin downstream target genes, including c-Myc, CD44, We performed TOPFLASH reporter assays to assess whether treatment with 1 modulates β-catenin-mediated and/or KITENIN-mediated signaling activity. Treatment with 5 µM of 1 significantly decreased β-catenin-mediated TOPFLASH activity on HEK293T by 30% (Figure 9a). In addition, Figure 9b-d indicates that treatment with 1 decreased mRNA expression and the protein level of β-catenin on AGS and Caco-2 cells; however, treatment with 1 did not significantly affect the mRNA expression and protein level of β-catenin in A549 cells. (Figure 10d). Treatment with 5 μM of 1 suppressed the mRNA expression of cyclinand CD44 in A549, while 5 μM of 1 suppressed the mRNA expression of CD44 in Cac cells for 48 h (Figure 10d). These results show that 1 decreased β-catenin-mediat TOPFLASH activity and its downstream targets in three cancer cells. Figure 9. The effects of 1 on the β-catenin-mediated TOPFLASH activity: (a) the β-catenin-media TOPFLASH activity in HEK293T cells; (b) relative β-catenin in A549, AGS, and Caco-2 cells; Western blot analysis of β-catenin in A549, AGS, and Caco-2 cells; (d) relative protein levels of catenin. Data represent the mean ± standard deviation, n = 3. * p < 0.05; ** p < 0.01; *** p < 0.001. Figure 9. The effects of 1 on the β-catenin-mediated TOPFLASH activity: (a) the β-catenin-mediated TOPFLASH activity in HEK293T cells; (b) relative β-catenin in A549, AGS, and Caco-2 cells; (c) Western blot analysis of β-catenin in A549, AGS, and Caco-2 cells; (d) relative protein levels of β-catenin. Data represent the mean ± standard deviation, n = 3. * p < 0.05; ** p < 0.01; *** p < 0.001.
The protein level of total, cytoplasmic, and nuclear β-catenin was examined to test whether 1 affected the nuclear/cytoplasmic distribution of β-catenin in A549 cells. As shown in Figure 10a,b, treatment with 5 µM of 1 did not affect the level of total β-catenin, whereas it decreased the β-catenin nuclear to cytoplasmic ratio remarkably in A549 cells compared to DMSO. To further test the effect of 1 on the levels of downstream target genes of β-catenin, qRT-PCR analysis was performed. As shown in Figure 10c  We examined the effect of 1 on the Notch signal pathway in A549, AGS, and Caco-2 cells and found that treatment with 5 μM of 1 significantly decreased the relative CSL activity in HEK293T cells by approximately 20% (Figure 11a). In addition, the relative Hes-1 activity in HEK293T and mRNA expression of Hes-1 in AGS, A549, and Caco-2 cells decreased during treatment with 5 μM, as shown in Figure 11b,c. The level of expression of HES correlated with NICD and CSL. Treatment with 1 decreased the level of Cleaved Notch1 in A549, AGS, and Caco-2 cells (Figure 11d,e). This result indicates that 1 slightly suppressed the Notch signal pathway in A549, AGS, and Caco-2 cells. We examined the effect of 1 on the Notch signal pathway in A549, AGS, and Caco-2 cells and found that treatment with 5 µM of 1 significantly decreased the relative CSL activity in HEK293T cells by approximately 20% (Figure 11a). In addition, the relative Hes-1 activity in HEK293T and mRNA expression of Hes-1 in AGS, A549, and Caco-2 cells decreased during treatment with 5 µM, as shown in Figure 11b Peritoneal carcinomatosis occurs when gastric cancer metastasizes to the cavity [27]. Peritoneal carcinomatosis is a hallmark of advanced peritoneal tumo sion, and peritoneal recurrence from gastric cancer occurs due to resistance to c apy. On day 28 after inoculation of AGS-iRFP, treatment of 10 mg/kg of 1 sig reduced the three different categories of the number of tumor nodules in the compared to the control ( Figure 12). As shown in Figure 12b, treatment with 10 1 reduced the number of nodules with diameters >1 compared to other group group, 8.4 ± 1.67 and 5 mg/kg of 1 group, 8.2 ± 1.92 versus 10 mg/kg of 1, 5 ± 2.35 The number of nodules with diameters ranging from 1 to < 5 mm in the grou mg/kg of 1 was also lower than those in the other groups (control group, 8.4 ± mg/kg of 1 group, 8.2 ± 1.92 versus 10 mg/kg of 1, 5 ± 2.35, p < 0.05). As a result, titative data showed that the total number of metastatic nodules was significantl in the mice treated with 10 mg/kg of 1 compared to those in the control group (F Peritoneal carcinomatosis occurs when gastric cancer metastasizes to the peritoneal cavity [27]. Peritoneal carcinomatosis is a hallmark of advanced peritoneal tumor progression, and peritoneal recurrence from gastric cancer occurs due to resistance to chemotherapy. On day 28 after inoculation of AGS-iRFP, treatment of 10 mg/kg of 1 significantly reduced the three different categories of the number of tumor nodules in the mesentery compared to the control ( Figure 12). As shown in Figure 12b, treatment with 10 mg/kg of 1 reduced the number of nodules with diameters >1 compared to other groups (control group, 8.4 ± 1.67 and 5 mg/kg of 1 group, 8.2 ± 1.92 versus 10 mg/kg of 1, 5 ± 2.35, p < 0.05). The number of nodules with diameters ranging from 1 to < 5 mm in the group with 10 mg/kg of 1 was also lower than those in the other groups (control group, 8.4 ± 1.67 and 5 mg/kg of 1 group, 8.2 ± 1.92 versus 10 mg/kg of 1, 5 ± 2.35, p < 0.05). As a result, the quantitative data showed that the total number of metastatic nodules was significantly reduced in the mice treated with 10 mg/kg of 1 compared to those in the control group (Figure 12c).

General Experimental Procedures
Optical rotation was acquired using a Kruss Optronic P-8000 polarimeter with a 5-cm cell. The UV spectrum was recorded in methanol (MeOH) on a Scinco UVS2100, and the VCD spectra were measured using a BioTools dualPEM ChiralIR spectrophotometer. The IR spectrum was collected on a Varian Scimitar Series. The 1 H and 2D NMR spectra were recorded at 400 and 500 MHz in CD 3 OD, containing Me 4 Si as the internal standard on Varian Inova spectrometers. The 13 C NMR spectra were acquired at 75 MHz on a Varian Inova spectrometer. The high-resolution mass spectrum was obtained on a JMS-700 (JEOL) mass spectrometer, and the low-resolution LC-MS data were measured using the Agilent Technologies 1260 quadrupole and Waters Micromass ZQ LC/MS system with a reversedphase column (Phenomenex Luna C 18 (2), 100 Å, 50 mm × 4.6 mm, 5 µm), at a flow rate of 1.0 mL/min, at the National Research Facilities and Equipment Center (NanoBioEnergy Materials Center) at Ewha Womans University. Medium-pressure liquid chromatography (MPLC) was performed on a Biotage Isolera One system (SE-751 03 Uppsala, Sweden), using Biotage SNAP KP-Sil, with a step gradient solvent of dichloromethane (DCM) and methanol (MeOH). The fractions were purified by reversed-phase high-performance liquid chromatography (HPLC) (Phenomenex Luna C 18 (2), 100 Å, 250 nm × 10 mm, 5 µm).

Collection and Phylogenetic Analysis of the Strain CNQ-049
The marine-derived Acremonium sp. CNQ-049 was isolated from marine sediment collected off the coast of Southern California. The strain CNQ-049 was identified as Acremonium sp. with 99.4% similarity to that of Acremonium fusidiodes, based on 18S rRNA gene sequence analysis (GenBank accession number KP131520.1).

VCD Analysis and Calculations
The conformational assignments for the C-6 and C-7 positions of 1 were performed using the Macromodel software (Version 9.9, Schrodinger LLC.) with "Mixed torsional/Lowmode sampling" in the GAFF force field. The experiments were conducted in the gas phase with the 50 kJ/mol energy window limit and a maximum of 10,000 steps to thoroughly examine all low-energy conformers. The Polak-Ribière conjugate gradient method was utilized for the minimization processes with 10,000 maximum iterations and a 0.001 kJ (mol Å) −1 convergence threshold on the root mean square gradient. Conformers within 10 kJ/mol of each global minimum for the 6R,7R, and 6S,7S forms of 1 were used for calculating the gauge-independent atomic orbital shielding constant, without geometric optimization, by employing the TmoleX Version 4.2.1 software (COSMOlogic GmbH & Co. KG) at the B3LYP/6-31 + G(d) level in the gas phase. A sample of 1 (5.0 mg) was dissolved in CDCl 3 (150 µL) and placed in a BaF2 cell with a path length of 100 µm, and data were acquired on a BioTools dualPEM ChiralIR spectrophotometer. The spectra were collected in 12 blocks, and each block was acquired for 3120 scans.

Migration Assay
Migration assays were performed in non-coated Transwell chambers [30]. The cells were seeded at a density of 2.5-3 × 10 5 cells/well in RPMI 1640/DMEM containing 0.2% BSA in the upper compartment of the chamber. The lower chamber was filled with 600 µL RPMI 1640/DMEM containing 0.2% BSA and fibronectin as a chemoattractant. The cells were cultured either in the absence or presence of compound 1 (1, 2.5, and 5 µM) for 24 h and were fixed using a Diff-Quick kit. The cells in the upper chamber were counted using a Nikon Eclipse 400 fluorescence microscope (Nikon Instech, Co., Ltd.) and i-Solution FL Auto Software (IMT i-Solution Inc.; five fields/chamber)

Soft Agar Colony Formation Assay
The cancer cells were suspended at a density of 2.5-3 × 10 3 cells/well in 1.0 mL of soft agar (0.35% soft-agar solution diluted 2-fold with 2 × DMEM/RPMI) and planted onto 1 mL of soft agar (0.5% agarose solution diluted 2-fold with 2 × DMEM/RPMI) in a 12-well plate and cultured for three weeks [31]. The cells were fed twice per week with cell-culture media, compound 1 (1, 2.5, and 5 µM), or DMSO. The surface areas of the colonies in the five fields per well were estimated using a Nikon Eclipse 400 fluorescence microscope (Nikon Instech, Co., Ltd.) and i-Solution FL Auto Software (IMT i-Solution Inc.; five fields/chamber). Three replications were performed.

Quantitative Real-Time PCR
The total RNA was isolated from A549, AGS, and Caco-2 cells using RNAiso Plus (Takara, Otsu, Japan) according to the manufacturer's instructions. Moloney murine leukemia virus reverse transcriptase (Invitrogen, Carlsbad, CA, USA) was used to convert 1 µg of RNA into cDNA. The dye SYBR Green (Enzynomics, Seoul, Republic of Korea) was used to analyze relative gene expression. Further, the qRT-PCR reaction and analysis were performed using CFX (Bio-Rad, Hercules, CA, USA).

Western Blotting
The A549, AGS, and Caco-2 cells were treated with 1 for 24 h, and washed twice with ice-cold phosphate-buffered saline (PBS). Lysis buffer was used for extraction. In some experiments, cytoplasmic and nuclear extracts were separated with the NE-PER nuclear and cytoplasmic extraction kit (Pierce Biotechnology, USA), and the extracted protein was separated using SDS-PAGE. The density of the bands was measured using the Multi Gauge 3.0 (Fujifilm, Tokyo, Japan) software, and the bands' relative density was calculated based on the density of the control bands during loading in each sample.

Lentiviral Transduction
The human gastric-cancer AGS cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 Medium (Gen Depot, Barker, TX, USA), supplemented with puromycin solution. After 12 h, the lentiviral vector was transfected into the human gastric-cancer cell line by polybrene, and the returned cells were incubated. After 72 h, the single cells were then plated in individual wells of a 96-well plate and incubated for 7-10 days.

Animal Studies
The AGS-iRFP cell suspension (1 × 10 7 cells in 0.1 mL PBS per mouse) was implanted into the abdominal cavity of male BALB/c nude mice (five weeks old), obtained from Central Lab. Animal Inc. (Seoul, Korea). All in-vivo experiments were performed according to the Guiding Principles for the Care and Use of Animals (DHEW publication, NIH 80-23) and were approved by the Sunchon National University Research Institutional Animal Care and Use Committee. The mice were randomly assigned to one of three groups: control, 5 mg/kg of 1, and 10 mg/kg of 1 by IP treatment. All animals were examined by measuring weight change. The treatment was initiated a week after the AGS-iRFP cells were injected. On day 28 after tumor inoculation, all the mice were sacrificed. The fluorescence area of the images alongside the representative images was then obtained through a fluorescence-labeled organism bioimaging instrument system.

Conclusions
In conclusion, this study described a new bazzanane-type sesquiterpenoid, marinobazzanan (1), from the genus Acremonium. This is the first Acremonium-derived bazzanane-type sesquiterpenoid isolated with chemical-structure modifications, such as chlorination and amination. Marinobazzanan (1) was shown to inhibit cancer-cell migration and invasion at non-toxic concentrations of 1, 2.5, and 5 µM and downregulate the transcription factors Snail, Slug, and Twist. In addition, marinobazzanan (1) decreased cell motility by downregulating the expression level of KITENIN while upregulating that of KAI1. Furthermore, the new compound modulated the expression of β-catenin by downregulating downstream target genes. Marinobazzanan (1) was also shown to have a reduced number of metastatic nodules in an intraperitoneal xenograft mouse model. Together, these findings suggest that 1 exhibits potent anticancer activity against cancer cells in vitro and anti-cancer activity for the peritoneal carcinomatosis model in vivo.

Conflicts of Interest:
The authors declare no conflict of interest.