A Survey of Marine Natural Compounds and Their Derivatives with Anti-Cancer Activity Reported in 2010

Although considerable progress in oncology therapeutics has been achieved in the last century, cancer remains one of major death causes in the World and for this reason, the development of novel cancer drugs remains a pressing need. Natural marine compounds represent an interesting source of novel leads with potent chemotherapeutic or chemo-preventive activities. In the last decades, structure-activity-relationship studies have led to the development of naturally-derived or semi-synthetic analogues with improved bioactivity, a simplified synthetic target or less toxicity. We aim here to review a selection of natural compounds with reported anticancer activity isolated of marine sources and their associated analogues published in 2010.


Introduction
Despite the continuous and important advances in biomedical research, the World Health Organization predicts that there will be more then 11 million cancer-related deaths per annum by 2030 [1]. Recent research highlights the isolation of promising compounds with effective anticancer activities from natural sources. An example of these compounds is trabectedin (PharmaMar's Yondelis ® ) [2], which represents the first anticancer drug isolated from a marine source. Almost 50 percent of the antitumor agents approved over the last 50 years have consisted of compounds either derived from OPEN ACCESS natural sources or (hemi-) synthetic analogues of these products [3]. Natural compounds remain a rich source of promising chemotherapeutic or chemo-preventive agents [4][5][6][7][8].
The sea covers over seventy percent of the Earth's surface, and ecosystems such as coral reefs contain high levels of biodiversity compared to rainforests. The sea contains many untapped sources of drugs with promising activities due to the extensive variety of marine habitats (influenced by factors such as UV-exposure, the presence of sunlight, and salinity levels) [9]. Over 2700 scientists from over 80 nations, who assessed the diversity, distribution and abundance of marine life, conducted a marine census. The census resulted in the discovery of over 6000 potentially new species [10][11][12][13]. As a consequence of this research effort, it is clear that the marine environment represents a largely unexploited reservoir of unknown natural compounds, which need to be evaluated for potential medicinal applications.
Natural derivatives of potent bioactive compounds from marine organisms can be bio-synthesized. It is well known that in some cases, like that of the plant-derived polyphenol curcumin, the synthetic analog exerts a higher activity compared to the parent compound [14,15]. In addition to natural analogs, chemical modification is an extensive and exceptionally powerful tool for the development of novel drug candidates [16]. Structure-activity-relationship studies of marine compounds can lead to the design of analogs that have greater activity together with a simplified synthetic approach, as reported with bryostatin 1, a compound produced by the marine bryozoan Bugula neritina, which has been studied for several years by the Wender research group [17,18].
Many reviews on natural compounds from marine environments have been published [6,[19][20][21][22][23][24][25]. Here, however, we will focus only on selected marine anti-cancer agents and analogues either discovered or synthesized in 2010 or whose biological activity was discussed in that year.

Marine Natural Compounds and Their Derivatives Published in 2010
2.1. Cryptosphaerolide (1) Fenical et al. isolated cryptosphaerolide (1, Figure 1), an ester-substituted sesquiterpenoid, in 2010 from the ascomycete strain CNL-523 (Cryptosphaeria sp.) [26]. This marine product exerts cytotoxicity (IC 50 of 4.5 μM) on the HCT-116 colon carcinoma cell line. A biochemical study revealed that this compound inhibited myeloid leukemia cell differentiation protein Mcl-1, a critical player involved in life/death decisions of individual cells [27], with an IC 50 of 11.4 μM. Studies on a hydrolyzed analog of this compound demonstrated that the presence of a hydroxylated ester side chain, linked to the core sesquiterpenoid group, is responsible for the observed anti-cancer activity [26] ( Figure 2).

Manzamine A (2)
The alkaloid manzamine A (2, Figure 1), which has been isolated from various marine sponges, exerted a cytotoxic effect against AsPC-1 pancreatic cancer cells, with an IC 50 in a range of 4.2 μM, after 3 days of treatment [28]. Additionally, it was reported that manzamine A inhibited AsPC-1 pancreatic cancer cell migration in vitro, and it decreased their overall metastatic potential. Fluorescent microscopy after staining with annexin V indicated an onset of apoptosis.

HESA-A, a Drug from Herbal-Marine Origin
HESA-A (patented by researchers in Iran) is composed of both plant and marine materials, including material from Penaeus latisculatus (king prawn), Carum carvi (Persian cumin) and Apium graveolens (celery). HESA-A consists of both mineral and organic constituents and a small amount of water (45%, 50% and 5%, respectively) [35]. The exact biological targets of HESA-A have not been determined to date, but it is thought that this multi-component drug acts using a variety of pharmacological mechanisms [36]. Its efficiency as a non-toxic, chemotherapeutic agent has been confirmed recently in various in vivo and pre-clinical studies [35,[37][38][39][40].

Spongistatin 1 (5)
The Pettit group isolated the macrocyclic lactone spongistatin 1 (5) from a marine sponge of the genus Spongia in 1993; this marine lactone exerted strong cytotoxicity on a panel of 60 types of human cancer cells [41]. Spongistatin 1 was reported to inhibit glutamate-induced tubulin polymerization (IC 50 of 3.6 μM in PtK1 kangaroo rat kidney cells) through its interaction with the Vinca alkaloid domain of tubulin, which leads to the inhibition of mitosis [42].
After treatment of A549 lung cancer cells with spongistatin 1 (1 nM), cell cycle arrest at the G2-M phase, the simultaneous up-regulation of GADD45α-γ and down-regulation of c-Myc were observed [43]. Various studies have reported that this marine product triggers caspase-dependent apoptosis in leukemia cells, even in primary leukemia cell lines, at low concentrations (1 nM) [43][44][45].
This compound did not induce significant apoptosis in healthy peripheral blood cells, highlighting its potential use as a therapeutic drug [44]. Several structure-activity-relationship studies have been conducted to date ( Figure 6); Kishi et al. noted that a C-23 epimer and spongistatin 1 had similar cytotoxic effects [46]. Paterson et al. reported that dehydration of the E-ring (C35-C36) led to an increase in cytotoxic potency, but that altering the side chain resulted in an important loss of activity [47]. A hydrogen-chlorine substitution in spongistatin 1 resulted in a 10-fold reduction in cytotoxicity [41]. In 2008, Heathcock et al. evaluated the toxicity of acyclic spongistatin 2 analogs, which contained only the E-and F-rings, as well as cyclic EF, ABEF and ABCD ring derivatives. In all derivatives tested, cytotoxicity was lost [48]. More recently, Smith et al. noted that an ABEF analog had cytotoxic effects when used in the nanomolar range, although its potency was 1,000 times weaker than that of spongistatin 1 [49]. In summary, these results demonstrate that the ABEF ring system, as well as the triene side chain, are crucial for spongistatin 1 cytotoxicity.

Makaluvamine Analogs
Makaluvamines (20) (a type of pyrroloiminoquinone) were first isolated from marine sponges belonging to the genera Zyzzya and Histodermella [63,64]. These products exert cytotoxic activity against various cell lines by inhibiting DNA topoisomerase II [65]. Anti-cancer activity was observed with bioactive synthetic analogs of these compounds. Imidazoquinoxalinone derivatives 21 have been reported to be less active due to the presence of an electro-deficient benzimidazole ring; the naturally occurring compound contains an indole ring. Under physiological conditions, the cationic makaluvamines were reported to be active, whereas imidazoquinoxalinones analogs are not charged, which explains the decreased activity of the latter [66,67]. Velu et al. evaluated the biological activity of makaluvamines containing various substitutions at the 7-position of the pyrroloiminoquinone ring. Observations from many natural derivatives allowed them to conclude that the presence of functional groups at this position greatly increased cytotoxic potential (Figure 7). In vitro testing with a NCI panel of 60 human cancer cell types indicated that the 7-benzyl-and 7-(4-fluorobenzyl) analogs (BA-TPQ and FBA-TPQ, respectively) exhibited the greatest cytotoxic effects [68,69]. These promising results lead various groups to perform pre-clinical studies. In breast cancer cell lines, both products significantly decreased cancer cell growth, induced apoptosis and caused cell cycle arrest at submicromolar concentrations (0.5 μM). Additionally, it has been reported that the anticancer activity was independent from the activity of p53 in cancer cells [70,71]. FBA-TPQ also strongly inhibited cancer cell proliferation, activated apoptosis and caused cell cycle arrest in prostate cancer cells, in the low micromolar range (2 μM). Furthermore, androgen receptor (AR) and prostate-specific antigen (PSA) levels, as well as the expression of apoptosis-related proteins were reduced [72]. A pharmacological study conducted in mice revealed that intravenously-injected BA-TPQ accumulated in the lungs, kidneys and spleen; it even reached low concentrations in the brain. However, this product was systemically toxic (indicated by animal weight loss) when administered at a concentration of 10 mg/kg [73]. The data presented here clearly indicated that makaluvamine analogs represent a promising choice for future clinical trails, and they may promote the development of novel anticancer drugs.

Cyclopentenones
In 1977, cyclopentenones were shown to possess antitumor properties and to down-regulate cellular metabolism [74,75]. It is has been noted that prostaglandins A 1 , A 2 and J 2 (compounds 22-24), which are known cancer-proliferation inhibitors, carry an α,β-unsaturated cyclopentenone ring linked to alkyl chains [76]. It is not surprising that the bioactivity of prostaglandins is mainly due to the presence of the cyclopentenone ring. An α,β-unsaturated ketone group is thought to act as an important alkylating center through a Michael-type reaction with cysteine [74][75][76]. Detailed mechanistic studies indicated that cyclopentenone (25) caused cell cycle arrest by the repression of cyclin D1, inhibited constitutive NF-κB activity and lead to the induction of apoptosis [77,78]. Novel cytotoxic cyclopentenones, namely didemnenone (26) and trichoderone (27), have been isolated from the didemnid ascidian Lissoclinum sp., and from the marine-derived fungus Trichoderma sp. [79,80]. In both cases, the IC 50 was in the micromolar range. These findings highlight the anticancer potential of cyclopentenone groups, making them interesting compounds on which to focus future studies.

Heteronemin and Semi-Synthetic Derivatives
The pentacyclic scalarane heteronemin (28) was first isolated in large quantities from the sponges Heteronema erecta and Hytios sp. in 1976 [81]. Crews et al. reported heteronemin had cytotoxic effects when applied to brine shrimp and giant kelp (Macrocystis pyrifera) gametes [82]. In cytotoxicity assays, this sesterterpene induced cell death in human thyroid carcinoma cells and an analog, 12-deacetoxy-21-hydroxyheteronemin, exhibited significant cytotoxicity against K562 cells [83,84]. The biological pathways affected by this marine product have since been identified. It has also been shown that heteronemin exerts antitubercular activity by inhibiting farnesyl transferase [85,86]. Furthermore, results from our laboratory have clearly shown that heteronemin attenuates NF-κB pathway activation through the down-regulation of proteasome activity [87]. Heteronemin triggered caspase-dependent apoptosis in K562 cells, and it sensitized K562 cells to TNFα-induced apoptosis [87].
Two structure-activity-relationship studies were published in 2009, with focus on the anti-carcinogenic effects induced by this compound [88,89] (Figure 8). Despite the fact that half of the tested analogs showed little or no cytotoxicity towards normal human oral fibroblasts or monkey kidney epithelial cells, these studies demonstrated that the oxygen atoms at positions C-25 and C-16 were crucial for the cytotoxic activity of heteronemin, whereas the double bond at position C-17-C-24 was of marginal importance [88,89]. To summarize, these promising results indicate that heteronemin and some of its derivatives represent interesting candidates for future chemotherapeutic drug research.

Latrunculin A and B
Latrunculins A and B (compounds 29,30) were first isolated from the Red Sea sponge Negombata magnifica [90,91]. The core structure of both compounds consists of a macrolide fused to a tetrahydropyran moiety, where the latter is linked to a 2-thiazolidinone side chain. These were the first marine natural products reported to bind reversibly to actin, leading to its disorganization [91]. These compounds also exerted potent activity on the angiogenesis, migration and proliferation of cells [92,93]. Latrunculin A has a therapeutic index (T/C) of 146% in mice [94], a remarkable result, considering the actin-active agents jaspamide and cucurbitacin did not have comparable therapeutic indices in case study conducted by scientists in the Developmental Therapeutics Program (DTP) at the NCI [95,96]. X-ray crystallography demonstrated that 2-thiazolidinone fits perfectly into actin pockets, and each polar oxygen, except for the O 2 -ester, forms a hydrogen bond with actin [97]. Several structure-activity-relationship studies have been completed [92,[98][99][100][101]. The deletion of methyl groups from the macrocycle group of latrunculin B resulted in an increased activity and a simplified synthetic target [99]. The composition of latrunculin A's 16-membered macrolide, as well as the conformation of its thiazolidinone ring play critical roles in its anticancer activities, as observed with human solid cancer cell lines HCT-116 and MDA-MB-435 [98]. Carbamate derivatives exerted 2.5-to 5-fold greater anti-invasive activity against the extremely metastatic human prostate PC-3M cancer cells, with lower actin binding properties [101] (Figure 9). El Sayed et al. demonstrated that both 17-O-phenylethyl-and N-hydroxymethyl-analogs of latrunculin A had higher activity than the parent product [100].

Dermacozines
Dermacozines (31), phenazine-type pigments, have been isolated from marine actinomycetes isolated from Mariana Trench sediment from a depth of 10.898 meters by Jaspar's research group [102]. This novel class of phenazines has been characterized and confirmed through in-depth analysis of 1D-,2D-NMR data combined to high-resolution MS, UV-data and CD spectroscopy. Hence these marine products exerted a cytotoxicity activity versus leukemic K562 cancer cells with an IC 50 -range from 7 to 220 μM. A structure-activity-relationship study showed that a carboxamide moiety nor a lactone ring or a benzyl function majorly affected the observed cytotoxicity of the products ( Figure 10). However, an additional carboxylic anhydride linked to the phenazine core structure led to a 20-fold increase in observed cytotoxicity in contrast to an imide ring. The latter did not alter the observed activity [102]. Further mechanistic studies on the pathways affected by dermacozines need to be conducted.  Figure 9. Summary of structure-activity relationship (SAR) results of the marine compound latrunculin A (29) concerning the observed anti-cancer cytotoxicity. The pharmacophores identified during the studies are highlighted in color (X: inactive; , =, : higher, similar, lower cytotoxicity).

Conclusions
This review provides insight into the current literature regarding marine natural products and their derivatives which was published in 2010. The data presented here indicate the great value of natural marine products, as well as their synthetic analogs. The data suggest that these synthetic analogs, in particular, could be important candidates for further studies involving structural modifications to improve the pharmacological profile of native marine metabolites. Furthermore, a simplified analog with equipotent activity can lead to the development of a simplistic synthesis process, which would guarantee a sufficient supply if bioactive products for further investigation. In conclusion, the isolation or modification of novel marine products, as well as their analogs, and the subsequent evaluation of their bioactivity will push the discovery of novel promising chemotherapeutic drugs forward.