W.C.Dulap et al.I Methods 4(2007)358-376 OH other V-ATP itors (Tabl o。 hutic action V-ATP of membranes of vacuoles,lysosomes,and other cellular OH organelles,as well as on certain special H8 tial for many cellular processes [13)Irregular V-ATPase HO 0 aking V-ATPas 6 the development of novel agents.Conse rug d for phar with the US Nationa lites across three different orders of metazoans suggest Cancer Institute (NIH)(http://ttb.nci.nih.gov/opportuni es/opportunity.php?opp_id- 24/&cat 1c 6 r1410510711nd larme and extracting a sample of Ircinia ramosa with methanol gives (N)d tional polyketide synthase (PKS). omal peptide a dark-green solution containing chlorophyll a and b S hybrid path nated modules, of which each module is responsible for of one complete cycl polyketid modifications [119-121 In this way,polyketide metabolites are assembled through rom the sponge metagenome commor of small ca of the e he cyanobacteria popul (40/40 random clones)within four specimens sampled at of each spe cific amino acid substrate.The vast structura unpublished) stly from variations in t the condensations type of starte cvanobacterium (AY692243.99% cifie secondary structural proc Y2H的c extension.A large number ks( ide (or per ptide)synthesis on a protein template,whereby cyanobacteria form a distinct ster within the Synecho group (Fig. The e apparen ng the appropnate carboxylate acid mig correlates well with the consisteney of chon- identify kev PKS and NRPS dropsin metabolites analysed in corresponding specimens 106.1071.which can be used to screen metagenomes for source oactive meta yet to b hion [122.123Epro Regardless of their true biosynthetic origin,the chon putative peptide synthetase and polyketide synthase genes humsin-class of onn potent inhibitor of ere detected using degener te primer sets Irom strains o an excepuon a distinct and unprecedented biolog ssociated with stromatolites of Western Australia [124). OH O O H N OH OH OH HN O OH O OH O HO O OH H N O O O OH OH OH poecillastrin A (80) Unlike metabolites from the NZ sponge Mycale hentsc￾heli, the distribution of the chondropsin-class of metabo￾lites across three different orders of metazoans suggests the biosynthetic involvement of a common sponge-specific symbiont, as has been implicated in the biosynthesis of sev￾eral other marine PKS metabolites [14,105–107]. Indeed, extracting a sample of Ircinia ramosa with methanol gives a dark-green solution containing chlorophyll a and b (kmax = 666, 616 nm) and carotenoid pigments (kmax = 475–423 nm) that, by unmasking non-soluble cellu￾lar pigments, reveals the pink phycobillin proteins of cya￾nobacteria residing within the light-exposed surface of the sponge. Photomicroscopy confirms the presence of algal symbionts at high density within the ectosome of Ircinia ramosa (Fig. 1). Phylogenetic analysis using cyanobacte￾ria-specific PCR primers to amplify 16S rRNA subunit genes directly from the sponge metagenome [6,85,106,108–111] has revealed the cyanobacteria popula￾tion of the Ircinia ramosa pinacoderm to be monophyletic (40/40 random clones) within four specimens sampled at two reefs and three depths (Dunlap et al., unpublished). 16S rRNA sequence analysis demonstrates that the Ircinia ramosa symbiont is closely affiliated with an uncultured cyanobacterium (AY692243, 99% sequence homology) and related clones (AY692244 and AY692245) from the Palauan sponge Xestospongia exigua [112] and an uncul￾tured cyanobacterium (AJ347056, 98% sequence homol￾ogy) from the Mediterranean sponge Aplysina aerophoba [45]. Viewed by phylogenetic comparison, these symbiotic cyanobacteria form a distinct cluster within the Synecho￾coccus/Prochlorococcus group (Fig. 2). The apparent monophyletic specificity of resident cyanobacteria in Irci￾nia ramosa correlates well with the consistency of chon￾dropsin metabolites analysed in corresponding specimens (Dunlap et al., unpublished), although a strict cyanobacte￾rial source of these bioactive metabolites is yet to be proven. Regardless of their true biosynthetic origin, the chon￾dropsin-class of metabolites are a potent inhibitor of human osteosarcoma cell lines and offer an exceptional opportunity for lead pharmaceutical development. Chon￾dropsin A exhibited a distinct and unprecedented biologi￾cal activity profile in the NCI’s 60-cell anti-tumor screen as compared to other V-ATPase inhibitors (Table 1), sug￾gesting that its mechanism of tumor growth inhibition has a novel therapeutic action. V-ATPases are located on membranes of vacuoles, lysosomes, and other cellular organelles, as well as on certain specialized plasma mem￾branes where these proton pumps regulate pH within the intracellular compartments of eukaryotes, which is essen￾tial for many cellular processes [113]. Irregular V-ATPase activity is thought to contribute to the development of a variety of diseases, such as cancer and osteoporosis [114– 116], making V-ATPase a potential target receptor for the development of novel pharmacological agents. Conse￾quently, the chondropsins offer a new structural lead in drug development [100,117,118] and are available for phar￾maceutical exploitation by a Collaborative Research and Development Agreement (CRADA) with the US National Cancer Institute (NIH) (http://ttb.nci.nih.gov/opportuni￾ties/opportunity.php?opp_id=247&cat_id=46). A vast array of natural products of marine and terres￾trial microbial origin are constructed by large multifunc￾tional polyketide synthase (PKS), nonribosomal peptide synthetase (NRPS), and mixed PKS–NRPS hybrid path￾ways that are distinctive biosynthetic motifs of eukaryotic and prokaryotic microorganisms. PKS and NRPS biosyn￾thetic gene clusters are assembled into repeated, coordi￾nated modules, of which each module is responsible for the sequential catalysis of one complete cycle of polyketide or polypeptide chain elongation together with necessary tailoring enzymes for structural modifications [119–121]. In this way, polyketide metabolites are assembled through a common mechanism by the condensation of small car￾boxylic acids, while nonribosomal peptides are con￾structed in a similar manner by sequential condensation of each specific amino acid substrate. The vast structural variety of polyketides stems mostly from variations in the number of condensations, the type of starter and extender units incorporated, and the extent of stereo-spe￾cific secondary structural processing (ketoreductase, dehy￾drase, cyclase or enoylreductase) occurring during each cycle of ketide extension. A large number of clinically sig￾nificant polyketides (or nonribosomal peptides) are made by modular PKS (or NRPS) assemblies that direct polyke￾tide (or peptide) synthesis on a protein template, whereby each module is responsible for selecting, incorporating and processing the appropriate carboxylate (or amino acid) unit. Accordingly, there are universal degenerate primers available to identify key PKS and NRPS genes [106,107], which can be used to screen metagenomes for biosynthetic pathways leading to the production of poten￾tially useful bioactive metabolites in diverse prokaryotic symbionts, including cyanophytes [122,123]. For example, putative peptide synthetase and polyketide synthase genes were detected using degenerate primer sets from strains of Symploca, Leptolyngbya, Microcoleus, Pleurocapsa and Plectonema species in the cyanobacterial communities associated with stromatolites of Western Australia [124], W.C. Dunlap et al. / Methods 42 (2007) 358–376 369