The FABMS of 2 displayed three ion peaks at m/z 380.9/382.9/384.9 in the ratio of 1:2:1, indicating the dibrominated nature of the molecule. The HRFABMS data of 2 suggested the molecular formula C₁₂H₁₉⁷⁹Br₂N₂O₂ as established by the molecular ion peak at m/z 380.9822 [M + H]⁺, being larger than that of 1 [18] by 30 mass units, suggesting the presence of an additional methoxyl moiety in the molecule. In comparison with compound 1 [18], the absence of the signals of H₂-7/C-7 in compound 2 and the appearance of new oxygenated methine at δ 4.78 (H-7)/70.4 (CH, C-7) and signal for a methoxyl moiety at δ 3.19 (H₃-12)/57.6 (C-12) a three-proton singlet at δ 3.19 supported the presence of the methoxyl moiety in 2. The ¹H and ¹³C NMR data of 2 (Table 1) together with the HSQC revealed the presence of four methylenes (C-8, C-9, C-10 and C-11), two aromatic methines (C-2 and C-6), one oxygenated aliphatic methine (C-7) and four quaternary carbons (C-1, C-3, C-4 and C-5). The ¹H and ¹³C NMR data of the basic skeleton of 2 are in good agreement with those reported for moloka’iamine (1) [18] suggesting that 2 possesses the moloka’iamine skeleton [18]. Moreover, the ¹H NMR signal at δ 4.78 (dd, J = 9.5 and 3.0 Hz, H-7) together with carbon signal at δ 70.4 (CH, C-7) suggested the oxygenation of C-7. Beside the vicinal COSY correlations with H-7, a geminal correlation between the protons at C-8 was observed with a coupling constant value of 13.0 Hz [4]. The HMBC data confirmed the location of a methoxyl functionality on the ethylamine moiety at C-7 through the cross peaks of H₃-12/C-7, H-7/C-12, H-7/C-2, C-6, H-7/C-1, H-7/C-8, and H₂-8/C-1 (Table 1). Thus, compound 2 was assigned as 3-(4-(2-amino-1-methoxyethyl)-2,6-dibromophenoxy)propan-1-amine.

The HRFABMS analysis of 3 (Figure 1) confirmed the molecular formula C₁₃H₁₇⁷⁹Br₂N₂O₂ as established from the HRFABMS ion peak at m/z 390.9683 corresponding to [M + H]⁺, requiring two degrees of unsaturation more than 2. Again, the appearance of three ion peaks cluster at m/z 390.9/392.9/394.9 in the ratio 1:2:1 suggested the presence of two bromines in 3. The ¹H NMR spectra (in CD₃OD) together with the HSQC revealed the presence of two methyl groups (δ_H/C 1.43/20.0 and 1.43/21.3), two methylenes (δ_H/C 3.40/37.2 and 3.16/49.8), three oxygenated methines (δ_H/C 4.0/73.7, 3.60/74.6 and 3.75/72.4), and six quaternary aromatic carbons between 103.5 and 152.5 ppm, corresponding to the fully substituted aromatic moiety in 3 (Table 1). Interpretation of the COSY experiment revealed the presence of two spin-coupling systems within 3 (Figure 2). The first spin-coupling system could be traced from H-11 (δ 4.00, d) to H-10 (δ 3.60, dd), which further couples to H-9 (δ 3.75, m). In addition, H-9 showed vicinal coupling to the methyl protons at C-13 (δ 1.43, d), while the second system consists of two triplets of the methylene protons at C-7 (δ 3.40, t, J = 6.5 Hz) and C-8 (δ 3.16, t, J = 6.5 Hz) (Table 1) of the dihydroindole moiety. Interpretation of HSQC and HMBC (Table 1 and Figure 2) data allowed the assignment of the dihydroindole fragment with a fully substituted benzene ring. The presence of three oxygenated methines (C-9, C-10 and C-11) in the side chain of the molecules and the need for six degrees of unsaturation in 3 support the presence of an epoxy ring at C-10/C-11 to complete the number of unsaturation in the molecule and the structure of 3. Moreover, the mass fragment ion peaks at m/z 374.9 [M − NH₂ + H]⁺ and 332.9 [M − C₂H₄NO + H]⁺ supported the presence of the oxirane functionality at C-10 and 11. HMBC experiment allowed the assignment of the quaternary carbons of the benzene moiety. For example, HMBC cross peaks of H₃-12/C-3, H₃-12/C-2, H₃-12/C-1, H₂-7/C-2, H₂-7/C-1, H₂-7/C-6, H₂-7/C-8, H₂-8/C-1 and H₂-8/C-7 supported the unambiguous assignment of these carbons (Table 1 and Figure 2). The relative stereochemistry at side chain of 3 was assigned from the coupling constant values of H-9, H-10 and H-11 [19,20]. The coupling constant value of 3.2 Hz between H-10 and H-11 reflects their trans configuration [19]. Furthermore, the coupling constant of 10.0 Hz between H-9 and H-10 supported their trans (E)-configuration as well [20]. The strong NOESY correlation between H-9 (d 3.75) and H-10 (d 4.00) supported the assignment. An MM2 ChemBio3D Ultra 11.0 (ChemBioOffice) energy-minimized drawing for compound 3 was created which show the significant NOESY correlation between H-9 and H-11 (Figure 3). Surprisingly, no NOESY correlations were observed between the H₃-13 and any of the protons on the side chain, which could due to the ring constrain which arise from the substitution on the dihydroindole moiety.

The available spectroscopic data was not enough for the determination of the absolute configuration of the chiral centers C-9-C-11 of 3 and therefore, they were left ambiguous. Ceratinine B possesses an unprecedented dibrominated dihydroindole skeleton with an epoxy ring on the side chain and a fully substituted aromatic ring of the indole skeleton. This is the first report of this brominated indole alkaloids in members of the Verongid sponges. Therefore, compound 3 was found to be 3-(1-(5,7-dibromo-4-methylindolin-6-yloxy)ethyl)oxiran-2-amine.

HRFABMS analysis of 4 (Figure 1) confirmed the molecular formula C₁₂H₁₈Br₂N₃O₂ as established from the ion peak at m/z 393.9772 corresponding to [M + H]⁺, suggesting five degrees of unsaturation, being one more than 2. Again, the FABMS cluster at m/z 393.9, 395.9 and 397.9 [M + H]⁺ suggested a dibrominated species. A comparison of its NMR data (¹H, ¹³C, HSQC and HMBC) to that of 2 (Table 2) established the absence of the methoxyl moiety at C-7 of the side in 2. Furthermore, the appearance of an additional ¹³C NMR signal at δ 160.3 (qC, C-12), was assigned as a carbonyl group of a carbamide moiety attached C-11, completing the degrees of unsaturation and the molecular formula of 4. The mass fragment ion at m/z 350 [M − CONH₂ + H]⁺ supports the presence of the of the carbamide moiety at one side of the molecule This attachment was secured from HMBC cross-peak of H₂-11 to C-12 (δ 160.3, Table 2). The assignment of all protonated carbons was securely established from the HSQC data and the HMBC experiment secured the assignment of quaternary carbons (Table 2). Thus, compound 4 was proved to be 1-(3-(4-(2-aminoethyl)-2,6-dibromophenoxy)-propyl)urea.

The molecular formula of 5 (Figure 1) was assigned as C₁₄H₁₈⁷⁹Br₂N₃O₄ as established from the molecular ion peak at m/z 449.9655 [M + H]⁺, suggesting six degrees of unsaturation, two more than 4. Again, the dibrominated nature of 5 was confirmed from the FABMS cluster at m/z 449.9, 451.9 and 453.9. A comparison of the NMR data (Table 2) of 5 (¹H, ¹³C, HSQC and HMBC) against those of 4 established identical similarity of all signals. Furthermore, the appearance of NMR signals at δ 160.4 (qC, C-113), 162.5 (qC, C-14) and 8.03/162.4 (H-12/C-12) were assigned as an oxalamide and a formamide moiety, respectively, completing the degrees of unsaturation and the structure of 5. In addition, the mass fragment ion at m/z 378 [M − COCONH₂ + H]⁺ supports the presence of the oxalamide moiety at one side of the molecule. The attachment of these moieties to the terminal amines at C-11 and C-8, respectively, were secured from HMBC cross-peaks of H₂-11/C-13, H₂-8/C-12 and H-12/C-8 (Table 2). HSQC and HMBC experiments secured the assignments of protonated and quaternary carbons (Table 2). The chemical shifts of the oxalamide moiety are in good agreement with the reported data in the literature [4]. Thus, compound 5 was assigned as N-(3-(2,6-dibromo-4-(2-formamidoethyl)-phenoxy)propyl)oxalamide.

Compound 6 (Figure 1) was purified as an amorphous powder. Its FABMS cluster at m/z 450.9/452.9/454.9 required two bromines, and the entire formula C₁₅H₂₁⁷⁹Br₂N₂O₄ [M + H]⁺ as established from the HRFABMS ion peak at m/z 450.9873. The molecular formula of 6 suggests six degrees of unsaturation, being two more than 2. Comparison of its NMR data (¹H, ¹³C, HSQC and HMBC) with those of 5 revealed the similarity to the structure of 5 (Table 2). However, instead of the terminal oxalamide moiety at C-11 in 5, new NMR signals for an ethylcarbamate moiety at δ 159.2 (qC, C-13), 4.05/61.7 (q, CH₂) (H₂-14/C-14) and 1.22/15.0 (t, CH₃) (H₃-15/C-15) were observed in the spectra of 6, completing the structure of 6. The chemical shift values of the signals of the ethylcarbamate moiety on the terminal amine are in good agreement with reported data [2]. Furthermore, the mass fragment ion at m/z 378 [M − COOCH₂CH₃ + H]⁺ supports the presence of the ethylcarbamate moiety at one side of the molecule. The COSY, HSQC and/or HMBC correlations secured the placement of formamide and ethylcarbamate moieties at C-9 and C-11, respectively. HMBC correlations of H₂-8/C-12, H-12/C-8, H₂-11/C-13, H₂-14/C-13, H₂-14/C-15, H₃-15/C-14 (Table 2) supported the assignment of these moieties and their attachment to the terminal amines. Accordingly, compound 6 was assigned as ethyl 3-(2,6-dibromo-4-(2-formamidoethyl)phenoxy)-propylcarbamate.

Compounds 1 [18], 7 [4], 8 [4], 9 [2], 10 [12], and 11 [12] were identified by comparison of their spectroscopic data with those previously reported in the literature.

The antimigratory activity of compounds 1–3, 5, and 7–11 against the highly metastatic MDA-MB-231 human breast cancer cell line was evaluated using the wound-healing assay (WHA, Figure 4) [21,22]. The wound-healing assay is a simple tool to investigate the in vitro directional cell migration [21,22,23]. The scratched tumor cell monolayer heals the wound in a specific fashion. The ability of the compounds to inhibit the migration of the highly metastatic MDA-MB-231 human breast cancer cells into the wound is measured by counting the number of the cells in the wound after 24 h incubation. The higher the antimigratory activity of the compound, the lower the number of the cells migrated into the wound. Therefore, this assay is widely used to study cell migration. However, little number of the cells may be found in the wound due to the effect of the compounds on their viability. The antimigratory activity of all compounds was assessed in the wound-healing assay at a 10 and 30 µM doses (Figure 4A), except compound 9, which was tested at several concentrations below 10 µM (Figure 4B and Figure 5) against MDA-MB-231 cells, to avoid false positive activity due to their possible cytotoxicity. The activity was compared to a 30 µM dose of the antimetastatic lead 4-S-ethylphenylmethylene hydantoin (S-ethyl), which was discovered based on marine natural products [22,23]. The effect of the compounds 1–3, 5, and 7–11 on the viability of the highly metastatic MDA-MB-231 human breast cancer cells was evaluated by MTT assay (Figure 6) [21,22]. The compounds showed no effect on the viability of MDA-MB-231 cells up to a 50 µM dose, indicating the lack of cytotoxicity toward this cell line. Only compounds 9 showed cytotoxic activity against MDA-MB-231 cells at 30 µM but showed no toxicity at 10 µM.

Initially, subereamolline A (9) showed significant antimigratory activity at 10 µM (Figure 4B). Therefore, it was reevaluated at six different concentrations (10, 5, 2.5, 1.25, 0.625, and 0.3125 µM) in order to calculate its IC₅₀ value (Figure 5). At the nanomolar dose level (0.3125 µM), the percent migration of the highly metastatic MDA-MB-231 human breast cancer cell line was about 52%. Therefore, it is expected its IC₅₀ value will be around 400 nM. Aerothionin (10) with terminal 1-oxa-2-azaspiro[4.5]deca-2,6,8-trienecarboxamido groups at each side showed no activity. This clearly indicates the importance of the terminal ethyl carbamate moiety in subereamolline A because it is identical in structure to 10 except this moiety. This may also imply the need for low molecular size group as bulky groups like 1-oxa-2-azaspiro[4.5]deca-2,6,8-trienecarboxamido may hinder the binding at the target receptor. Homoaerothionin (11) was also inactive; suggesting the 4-carbons n-butyl group connecting the two NH groups was preferred over the 5-carbon n-pentyl group.

Moloka’iamine (3-(4-(2-aminoethyl)-2,6-dibromophenoxy)propan-1-amine, 1) was the most active, inhibiting >50% of the migration at 30 and 10 μM, while ceratinine C ((2R,3S)-3-(1-(5,7-dibromo-4-methylindolin-6-yloxy)ethyl)oxiran-2-amine, 3) was the least active (Figure 4). This suggests the preference of the dibromophenoxy group over the dibromo-4-methylindolin-6-yloxy group for the antimigratory activity. The low activity of hydroxymoloka’iamine (7) compared to 1 suggested that C-8 hydroxy group adversely affect the activity.

The anti-invasive activity of compounds 1–3, 5, and 8–11 was assessed using the Cultrex^® BME cell invasion assay against the highly metastatic MDA-MB-231 human breast cancer cell line (Figure 7) [24]. The activity was compared to a 50 µM dose of the antimetastatic lead 4-S-ethylphenylmethylene hydantoin (S-ethyl) [22,23]. The compounds have been screened at 10 µM dose, homoaerothionin (11) has shown 50% invasion whereas compounds 3 and 5 showed medium activity. Compound 9 was cytotoxic at the tested concentration and, thus, it was re-evaluated at 2 µM dose which has shown 38% invasion. Subereamolline A (9) was screened at four different concentrations (2, 1.5, 1, and 0.5 µM) and its IC₅₀ value was 1.7 µM. This clearly shows the potential of subereamolline A as possible scaffold for the future design of breast cancer migration and invasion inhibitors.

Optical rotations were recorded on a JASCO DIP-730 digital polarimeter. Ultraviolet spectra were recorded on a Hitachi 300 spectrometer. NMR spectra were obtained in CD₃OD on Bruker Avance DRX 400 Spectrometers at 400 MHz for ¹H NMR and 100 MHz for ¹³C NMR. NMR chemical shifts are expressed in parts per million (ppm) referenced to CD₃OD solvent signals (δ 3.29 for ¹H and δ 49.0 for ¹³C) or DMSO-d₆ signals (2.49 ppm for ¹H and 39.9 ppm for ¹³C). Positive ion FAB mass spectral data were obtained with a Micromass Q-tof equipped with lockspray mass spectrometer using Leucine Enkaphalin at m/z 556.2771 [M + H]⁺ as a reference mass. Pre-coated silica gel G-25 UV₂₅₄ plates were used for thin layer chromatography and silica gel 60 Å, 230–400 μm mesh (E. Merck) and Sephadex LH-20 (Pharmacia) employed for column chromatography.

The highly metastatic human MDA-MB-231 breast cancer cells were cultured in RPMI 1640 medium containing 10 mM HEPES, 4 mM L-glutamine, 10% fetal bovine serum, penicillin (100 IU/mL), and streptomycin (50 μg/mL), and grown in a 5% CO₂ atmosphere at 37 °C [21]. Cells were plated onto sterile 24-well and allowed to recover for a confluent cell monolayer formed in each well (>95% confluence). Wounds were then inflicted to each cell monolayer using a sterile 200 μL pipette tip. Media were removed, cells were washed twice with PBS, and then fresh media containing test compounds were added to each well. Test compounds were prepared in DMSO stock solution (50 mM). The required test compound concentrations were prepared in serum free media containing 0.5% fetal bovine serum. Initially, the ability of the compounds to inhibit the migration of the cells into the wound was performed at 10 and 30 µM concentrations. The compounds that show promising migration inhibitory activity were tested at 6 non-toxic concentrations, prepared by serial dilution, each in triplicate using DMSO as negative control. The incubation was carried out for 24 h, after which media was removed and cells were fixed and stained using Diff Quick staining (Dade Behring Diagnostics, Aguada, Puerto Rico) [22,23]. The number of cells migrated on the scratched wound were counted under the microscope in three or more randomly selected fields (magnification: 40×). Final results are expressed as mean per 40× field.

The highly metastatic MDA-MB-231 human breast cancer cells were cultured in RPMI 1640 medium containing 10 mM HEPES, 4 mM L-glutamine, 10% fetal bovine serum, penicillin (100 IU/mL), and streptomycin (50 μg/mL), and incubated in a 5% CO₂ atmosphere at 37 °C. For subculturing, cells were rinsed twice with sterile Ca²⁺ and Mg²⁺-free phosphate buffered saline (PBS) and incubated in 0.25% trypsin containing 0.025% EDTA in PBS for 5 min at 37 °C. The released cells were centrifuged, resuspended in fresh media and counted using hemocytometer. Cells were plated onto sterile 96-well plates, at an initial cell count of 20 × 10⁴ cells/mL and allowed for complete attachment overnight. A stock solution of the compounds was prepared in DMSO. About 2 µL of each stock solution was transferred to 998 µL of serum-free medium containing 0.5% fetal bovine serum to obtain 100 µM concentrations (0.2% DMSO). Serial dilutions were then conducted to obtain the desired concentrations for the assay. Cell viable number was determined by the 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazolium bromide (MTT) colorimetric assay after incubation period of 24 h. On the assay day, treatment medium was replaced with fresh control medium before the addition of 1 mg/mL MTT (50 µL/well), and the cells in 96-well plates were incubated at 37 °C for 4 h. The medium was removed, and the MTT crystals were dissolved in DMSO (100 µL/well). The optical density of each sample was read at 570 nm on a microplate reader (Synergy™ 2, Biotek Instruments, Inc.), against a blank prepared from cell-free cultures. The number of cells/well was calculated against a standard curve prepared by plating various concentrations of cells, as determined by hemocytometer, at the start of each experiment.

Anti-invasive activities were measured using Cultrex^® BME cell invasion assay [24]. About 50 μL of basement membrane extract (BME) coat was added per well of the top chamber. After an overnight incubation at 37 °C in a 5% CO₂ atmosphere, 50,000/50 μL of MDA-MB-231 cells in 0.5% FBS RPMI medium were added per well of the top chamber. 150 μL of RPMI medium was then added to the lower chamber. Media contained 10% FBS and penicillin/streptomycin as well as fibronectin (1 μL/mL) and N-formyl-Met-Leu-Phe (10 nM) as chemoattractants. Test compounds were prepared at 6× the desired concentration and 10 μL of each of the compounds was added per well of the top chamber. Cells were incubated at 37 °C under 5% CO₂ which allowed for cell migration from the top to the lower chamber. After 24 h, the top and bottom chambers were aspirated and washed with washing buffer supplied within the kit. About 100 μL of cell dissociation solution/calcein-AM solution was added to the bottom chamber and incubated at 37 °C under 5% CO₂ for 1 h. The cells internalize calcein-AM, and the intracellular esterases cleave the acetomethylester (AM) moiety to generate free calcein. Fluorescence of the samples was determined at λ_excitation 485 nm and λ_emission 528 nm using an ELISA plate reader (BioTek, VT, USA). The numbers of cells that invaded through the BME coat were calculated using a standard curve.