**2. Tumor Growth Inhibiting Cephalostatins from the South African Marine Tube Worm** *Cephalodiscus gilchristi*

The first large-scale collections of African marine invertebrates solely for the purpose of new drug discovery were coordinated by Professor G. R. Pettit of Arizona State University, USA, over three decades ago off South Africa's temperate southern coast [5]. Cephalostatin 1 (**1**, Figure 2) was isolated in low yield (*ca.* 2.3 × <sup>10</sup>−7%) from two separate and substantial SCUBA collections (166 and 450 kg (wet weight) collected in 1981 and 1990, respectively) of the hemichordate marine tube worm *Cephalodiscus gilchristi* (Figure 1a) [6]. Cephalostatin 1 has emerged as one of the most potent cell growth-inhibiting secondary metabolites ever screened by the U.S. National Cancer Institute (NCI) (ED50 0.1–0.0001 pM in a P338 leukemia cell line) [6,7].

Of immediate interest to those exploring this compound's tumor growth inhibitory activities was, first, the comparative GI50 values (quantification of the concentration required to inhibit cellular growth by 50%) of **1** (GI50 1.2 nM) with commercially available anticancer drugs, e.g., taxol (**2**, GI50 29 nM), cisplatin (**3**, GI50 2000 nM) and 5-fluorouracil (**4**, 24,000 nM), and, second, the 275-times higher concentration of **1** required to kill 50% of cancer cells (LC50 330 nM) relative to the amount required for 50% cell growth inhibition [6]. In addition, the application of the NCI's COMPARE algorithm [8] to the GI50 data acquired for **1** indicated that this novel bis-steroidal pyrazine alkaloid possesses a unique mechanism of action against the proliferation of cancer cells in the NCI's *in vitro* 60 cancer cell line screen, and therefore, not surprisingly, **1** is increasingly proving to be a valuable tool for the discovery of new apoptosis signaling pathways [9]. Vollmar and co-workers' early studies into cephalostatin's apoptotic mechanism of action established that **1** promotes the release of Smac (second mitochondria-derived activator of caspase) through the dissipation of mitochondrial membrane potential [6,9,10] as part of a novel apoptosome-independent, caspase-9-mediated apoptotic pathway [6]. Furthermore, Shair and co-workers have shown that **1** also selectively binds to oxysterol binding protein (OSBP) and OSBP-related protein 4L (ORP4L) [11] and drew attention to these proteins, whose role in cancer cell survival was little known at the time. A further eighteen naturally-occurring and semi-synthetic analogues of **1** have subsequently been reported (1988–2012) in the chemical and patent literature (e.g., U.S. Patents 4873245, 5047532, 5583224 and WO 8908655). The isolation, structure elucidation, synthesis and bioactivity of this cohort of cephalostatins has been comprehensively reviewed along with the closely-related bis-steroidal pyrazine alkaloids, the ritterazines, e.g., ritterazine G, (**5**) from the Japanese ascidian (tunicate), *Ritterella tokioka* [6]. Since the publication of Iglesias-Arteaga and Morzycki's extensive review [6], the chemical structure of the twentieth member of the cephalostatin series, cephalostatin 20 (**6**), has recently been reported by Pettit *et al.* [12]. Compound **6**, the 9 -α-hydroxy analog of cephalostatin 9 (**7**), was isolated in low yield (1 × <sup>10</sup>7%) from the combined bioactive (cytotoxic to P338 murine lymphocyte cells) fractions from

the original extract of *C. gilchristi* [5] nearly a quarter of a century ago. Interestingly, the cell growth inhibitory activities of **6** and **7** against six human tumor cell lines was 100–1000-times less active than **1** in the same tumor cell panel, thus underlining the importance of an intact spirostanol structure in the southern unit of cephalostatins to the growth inhibition activities of these compounds [12].

**Figure 2.** Chemical structures of compounds **1**–**8** and **10**.

Significant effort [6,13,14] has been directed towards the total enantioselective syntheses of **1** over the last two decades. Following on from their first 65-step convergent total synthesis of **1** and potently active cephalostatin/ritterazine hybrids [15], Fuchs and co-workers have recently reported the first convergent total synthesis of 25-*epi* ritterostatin GN1N **8** [16] from commercially available dihydroxyhecogenin acetate (**9**, Figure 3). Fuchs and co-workers identified the key step in their synthesis as a chiral ligand ((DHQ)2PHAL)-mediated dihydroxylation reaction, which introduced the 25-*epi* functionality into the north segment (analogous to the north unit of cephalostatin) [16]. Compound **8**, structurally incorporating the north units of both **1** and **5**, exhibited a mean GI50 (0.48 nM) in a panel of eight cancer cell lines and was 30-fold more active than ritterostatin (**10**), also screened in the same cell line panel [16].

**Figure 3.** Chemical structures of synthetic intermediates **9** and **11**.

The daunting synthetic challenges of cephalostatin molecular architecture continue to inspire the synthesis of simpler analogs with similar bioactivities to **1** [6]. The latest target in this series, [5.5]-spiroketal (**11**, Figure 3), which shares the steroidal scaffold of the northern hemisphere of **1** with an intact 1,6-dioxaspiro[5.5]nonane side chain, but with a diminished oxygenation pattern, was synthesized by Pettit *et al.* in seven steps from **9** in an overall 4.6% yield [17]. Although **11** and several synthetic precursors of this compound were not cytotoxic to P388 leukemia cells, Pettit *et al.* suggested that **9** was potentially useful as a synthetically-accessible starting point for further synthetic modification into both symmetrical and asymmetrical trisdecacyclic bis-steroidal pyrazine congeners of **1** [17] through well-established pyrazine ring construction protocols [16].
