The Invasive Anemone Condylactis sp. of the Coral Reef as a Source of Sulfur- and Nitrogen-Containing Metabolites and Cytotoxic 5,8-Epidioxy Steroids

The Condylactis-genus anemones were examined for their proteinaceous poisons over 50 years ago. On the other hand, the current research focuses on isolating and describing the non-proteinaceous secondary metabolites from the invasive Condylactis anemones, which help take advantage of their population outbreak as a new source of chemical candidates and potential drug leads. From an organic extract of Condylactis sp., a 1,2,4-thiadiazole-based alkaloid, identified as 3,5-bis(3-pyridinyl)-1,2,4-thiadiazole (1), was found to be a new natural alkaloid despite being previously synthesized. The full assignment of NMR data of compound 1, based on the analysis of 2D NMR correlations, is reported herein for the first time. The proposed biosynthetic precursor thionicotinamide (2) was also isolated for the first time from nature along with nicotinamide (3), uridine (5), hypoxanthine (6), and four 5,8-epidioxysteroids (7–10). A major secondary metabolite (−)-betonicine (4) was isolated from Condylactis sp. and found for the first time in marine invertebrates. The four 5,8-epidioxysteroids, among other metabolites, exhibited cytotoxicity (IC50 3.5–9.0 μg/mL) toward five cancer cell lines.


Cytotoxicity Testing
The cancer cell lines: NCI-H661 (human lung carcinoma), KB (human oral epidermoid carcinoma), Hepa59T/VGH (human liver carcinoma), HeLa cells (human cervical cancer), and Med (human medulloblastoma) were purchased from the American Type Culture Collection (ATCC). Cytotoxicity assays of the substances were performed using MTT colorimetric method [34]. The compound is considered inactive when IC 50 > 20 µg/mL. The positive control used is mitomycin.

Results and Discussion
The frozen bodies of Condylactis sp. were sliced and extracted with ethanol (EtOH). The solvent-free crude extract was triturated in methanol (MeOH). The MeOH-soluble portion was then fractionated on Sephadex LH-20 to yield three fractions. A fraction that exhibited an in vitro cytotoxicity against Hepa59T/VGH and HeLa cells (IC 50 18.9 and 18.3 µg/mL, respectively), was subjected to chromatographic isolation using a combination of a normal phase (NP)-HPLC and MCI gel ® CHP20P columns to yields compounds 1−3, 5−10. Moreover, columns of Diaion ® HP-20P and MCI gel ® CHP20P were also utilized to obtain compound 4 from a major non-cytotoxic fraction MeOH-soluble fraction. The molecular structures of 1-10 ( Figure 1) were established mainly by MS and NMR spectroscopic analyses (Table 1 and Supplementary Figures S1-S27), and by comparison of their NMR data with those in the literature and databases. The four epidioxysteroids, among other metabolites, showed significant cytotoxic activities against five cancer cell lines ( Table 2).
Compound 1 was obtained as an off-white powder. Its molecular ion peak at m/z 240.0467 in the high-resolution electron impact mass spectrometry (HREIMS) indicated a molecular formula of C 12 H 8 N 4 S and eleven degrees of unsaturation. The 13 C NMR data revealed the presence of two pyridinyl groups, equivalent to eight unsaturation degrees, from the two sets of five carbon signals (C-2 to C-6 and C-2" to C-6") appearing from δ C 153.0 to 123.9 as those of the co-isolated compound 2 (Table 1). This was further supported by the ion peaks that appeared at m/z 162 in EIMS after the elimination of a pyridinyl (C 5 H 4 N) + group. Therefore, the remaining two sp 2 quaternary carbon signals at δ C 171.4 and 185.8 should be incorporated in a diunsaturated heterocyclic ring having one sulfur and two nitrogen atoms to form a thiadiazole core. The appearance of twelve 13 C signals in the molecule of 1 instead of six omits the possibility of the presence of the symmetrical 1,2,5or 1,3,4-thiadiazol as a linking unit for 1. The analysis of the COSY) and HMBC spectra ( Figure 2) established the structure of compound 1 as a 1,2,4-thiadiazole-based alk1aloid. The 3 J CH correlations of the two sets of pyridinyl protons (H-2 /H-4 and H-2 /H-4 ) to carbons C-3 (δ C 171.4, C) and C-5 (δ C 185.8, C) designated linking the position of pyridinyl groups to C-3 and C-5 of thiadiazole ring, respectively. Moreover, a fission of the molecule into [C 6 H 4 N 2 ] + at m/z 104 and [C 6 H 4 N 2 S] + at m/z 136 was observed in the EIMS of 1 (Figure 3). The latter fragment ion further yielded [C 6 H 4 NS] + at m/z 122 after the loss of one nitrogen atom. This fragmentation pattern and others could further support the identity of metabolite 1 as 3,5-bis(3-pyridinyl)-1,2,4-thiadiazole, but not 3,5-bis(3-pyridinyl)-1,2,3-thiadiazole. This 1,2,4-thiadiazole-based alkaloidal structure of compound 1 could be formed via oxidative dimerization of thioniacinamide (2). This dimerization process could be initialized in nature by the formation of thioamide S-oxide [35] after acidic oxidation of 2 or in the presence of an electrophilic environment [36]. Consequently, a plausible biosyn-thetic pathway of 1 from 2 was suggested as shown in Scheme 1 and further established the formation of the 1,2,4-thiadiazole moiety, but neither 1,2,3-, 1,3,4-, nor 1,2,5-thiadiazole, as a linking unit in the molecule of 2. Finally, a comparison of the NMR data of compound 2 with those of a synthetic product prepared by copper(II)-mediated homocoupling of thioamides revealed the same chemical identity [37]. However, the full assignment of 1 H and 13 C NMR signals was reported herein by the analysis of 2D NMR correlations for the first time. Despite the fact that the non-pyridinyl substituted-1,2,4-thiadiazoles are very rare, being only reported in nature from a fungus [38], a plant [39], and an ascidian [40], this is the first report to isolate a 3,5-bispyridinyl-substituted 1,2,4-thiadiazole (1) as a new type of alkaloid from nature.  Compound 1 was obtained as an off-white powder. Its molecular ion peak at m/z 240.0467 in the high-resolution electron impact mass spectrometry (HREIMS) indicated a molecular formula of C12H8N4S and eleven degrees of unsaturation. The 13 C NMR data  comparison of the NMR data of compound 2 with those of a synthetic product prepared by copper(II)-mediated homocoupling of thioamides revealed the same chemical identity [37]. However, the full assignment of 1 H and 13 C NMR signals was reported herein by the analysis of 2D NMR correlations for the first time. Despite the fact that the non-pyridinyl substituted-1,2,4-thiadiazoles are very rare, being only reported in nature from a fungus [38], a plant [39], and an ascidian [40], this is the first report to isolate a 3,5-bispyridinylsubstituted 1,2,4-thiadiazole as a new type of alkaloid (1) from nature.    (Table 1) in the 13 C NMR spectrum and the analysis of 2D NMR spectral correlations ( Figure 2) were thus in a good agreement for a 3-pyridinecarbothioamide or thionicotinamide (2). To the best of our knowledge, compound 2 is a natural product isolated for the first time from a living organism. This compound might   (Table 1) in the 13 C NMR spectrum and the analysis of 2D NMR spectral correlations ( Figure 2) were thus in a good agreement for a 3-pyridinecarbothioamide or thionicotinamide (2). To the best of our knowledge, compound 2 is a natural product isolated for the first time from a living organism. This compound might Compound 2 was obtained as pale yellow prisms and possessed a molecular formula of C 6 H 6 N 2 S from its HREIMS (m/z 138.0243). Its IR absorption bands at ν max 3229 and 3026 (broad) cm −1 along with characteristic absorptions at 1678, 1452, and 1311 cm −1 , suggesting the presence of aromatic or heteroaromatic thioamides [41,42]. The ion peaks displayed in the EIMS at m/z 122 [M-NH 2 ] + and 78 [C 5 H 4 N] + indicated that the remaining C and S atoms should be consistent with a thiocarbonyl functionality to fulfill the fifth unsaturation degree in the molecule. The four downfield shifted protons at δ H 7. 42-9.11 together with the four sp 2 methine (δ C 123.7-152.5, 4C) and the two sp 2 quaternary (δ C 136.3 and 200.4, 2C) carbons (Table 1) in the 13 C NMR spectrum and the analysis of 2D NMR spectral correlations ( Figure 2) were thus in a good agreement for a 3-pyridinecarbothioamide or thionicotinamide (2). To the best of our knowledge, compound 2 is a natural product isolated for the first time from a living organism. This compound might be derived from thionation of the co-existing nicotinamide (3) [43] through a reaction with H 2 S from the conversion of a sulfur source of the ocean by sulfate-reducing bacteria [44] or similar symbiotic bacteria associated with sea anemones [45,46] (Scheme 1).
The present study also led to the isolation of nicotinamide (3), which showed similar 1 H and 13 C signals to those of 2, except for those of carbonyl (δ C 167.6, C) and thiocarbonyl (δ C 200.4, C) carbons, respectively. This is further confirmed by direct mass and NMR data comparison with those previously reported [27].
The water-soluble compound (−)-betonicine (4) (Figure 4) proved the 2S and 4R configuration in the structure of 4 which was found to be the same as (−)-betonicine [47] that has been found in higher plants, e.g., Melaleuca spp. [47]. However, the isolation of (−)-betonicine, a pyrrolidine alkaloid, from a marine invertebrate is reported herein for the first time ( [48]. The full assignment of NMR data of 4 is presented herein for the first time at neutral pH (c.f. previous NMR data measured at different acidic pH [47]).
The heterocyclic compounds 1-5 were shown to be non-cytotoxic (IC 50 > 20 µg/mL) against the growth of cancer cell lines such as NCI-H661 (human lung carcinoma), KB (human oral epidermoid carcinoma), Hepa59T/VGH (human liver carcinoma), HeLa (human cervical cancer), and Med (human medulloblastoma) cells. However, it has been reported that compound 1 has much stronger inhibitory activity against aromatase than resveratrol and thus could be served as an effective non-steroidal chemopreventive candidate in the treatment of estrogen-sensitive breast cancer [49]. Thionicotinamide (2) is considered an inhibitor for nicotinamide adenine dinucleotide kinase (NADK) [50], metabolic targets controlling redox co-enzymes. Therefore, although compound 2 displayed no cytotoxicity against the tested cancel lines in this study, a mixture of thionicotinamide (2) with NADK inhibitors might represent an efficacious antitumor combination as it reduces nicotinamide adenine dinucleotide phosphate (NADPH) pools and accelerates the degradation of dihydrofolate reductase (DHFR), synergistically inhibiting the proliferation of cancer cells [51,52]. In addition, a combination treatment of thionicotinamide adenine dinucleotide phosphate (NADPS) with methotrexate (MTX) demonstrated significant synergy in a metastatic colon cancer cell line and was effective in an MTX-resistant leukemic cells [52]. Although the thionicotinamide therapy in tuberculosis is less-effective relative to rifampicin or isoniazid, the use of thioamides such as 2 can overcome multidrug-resistant tuberculosis [53,54]. A pyridine-4-thiocarboxymide analog, ethionamide, a second-line anti-tubercular drug, can be activated by enzymatic oxidation to exert the pharmaceutical effect [55]. A mimicked redox process involving electron transfer reaction of hexacyanoferrate (III) and thionicotinamide (2) also showed the potential for the design of new thioamidecontaining medicines with the aim of fighting the drug-resistant strains of Mycobacterium tuberculosis [53].
The pyrrolidine alkaloid (4) has been shown an unprecedented stimulatory effect similar to N-octanoyl-DL-homoserine lactone (OHL) on bacterial quorum sensing (QS), a clue that could be useful for the development of potential QS inhibitors which are chemically unrelated to acylated homoserine lactones (AHLs) [48]. Therefore, the anemone under investigation is regarded as a rich source for this medicinally important natural product (4), particularly when it could be cultivated.
The acetylated product of 5, uridine triacetate, has been shown to be an effective antidote against life-threatening toxicity and mortality resulting from overdoses or exaggerated susceptibility to the widely used anticancer drugs 5-fluorouracil (5-FU) and capecitabine [56]. It has been approved by FDA since 2018.

Conclusions
The organic soluble extract of the invasive sea anemone Condylactis sp. was chemically investigated for the first time, isolating several S-, N-, and O-heterocyclic metabolites. The 3,5-bis(3-pyridinyl)-1,2,4-thiadiazole alkaloid and its biosynthetic precursor thionicotinamide were recognized herein as natural products discovered for the first time from nature. Moreover, four 5,8-epidioxysteroid metabolites were isolated and significantly inhibited the growth of a limited panel of cancer cell lines and were considered responsible for the cytotoxicity of their organic soluble fraction. (−)-Betonicine was also separated from marine organisms as a major polar N-heterocyclic metabolite for the first time. Aside from the proteinaceous venom of anemones of the genus Condylactis, other organic soluble metabolites are worthy of being studied with the aim of discovering molecules of chemical and pharmaceutical interest. A massive harvest of this Condylactis anemone, invasive to coral reefs, is not only beneficial to the marine reef ecosystem but also makes this environmentally harmful organism a valuable resource of biologically active compounds.