Leiodermatolide is a structurally unique macrolide, isolated from the deep-water marine sponge Leiodermatium sp., which exhibits potent antiproliferative activity against a range of human cancer cell lines (IC50 <10 nM) and dramatic effects on spindle formation in mitotic cells. Its unprecedented polyketide skeleton and stereochemistry were established using a combination of experimental and computational (DP4) NMR methods, and molecular modelling.

We examined 215 Enterococcus faecalis isolates and found that neither the two-component regulatory locus fsr (E. faecalis regulator) nor gelatinase production was more common in disease-associated isolates than in isolates colonizing healthy individuals (ca. 60 to 65%). The majority of gelatinase-negative isolates, including 14 endocarditis isolates (of 80 isolates tested), contained the previously described 23.9-kb deletion and lacked fsrA and fsrB. While these findings indicate that neither fsr nor gelatinase is required for E. faecalis to cause infection, this study did not address whether fsr or gelatinase affects the severity of disease, as it does in animal models.Enterococci are normally commensal flora, but they can cause a wide range of diseases, including urinary tract infections, bloodstream infections, wound infections, and endocarditis (12, 13). Although there are over 20 species of enterococci (8), Enterococcus faecalis isolates cause the majority of infections (22). Despite the increasing number of cases of enterococcal infection and the remarkable ability of these organisms to resist antibiotics, relatively little about their pathogenesis is known compared to what is known about some other grampositive cocci.Despite its role as a possible virulence factor, gelatinase is not produced by all clinical isolates, and a number of studies have attempted to correlate gelatinase activity with disease caused by E. faecalis (3,5,7,17,20,24,28). Nakayama et al. recently reported that a 23.9-kb chromosomal deletion involving the fsr locus was commonly found in urine isolates and was also found in most gelatinase-negative strains (17). In contrast, another study reported that fsr was present in 100% of 12 endocarditis isolates (20). In the present study, we compared clinical isolates of E. faecalis from patients with endocarditis, urinary tract infections, or blood-borne infections to fecal isolates from healthy volunteers with respect to the presence of fsr and gelatinase production.A total of 215 E. faecalis isolates from clinical samples and from feces of healthy, community-based volunteers, collected between 1974 and 2003, were used in this study. The identification of all isolates as E. faecalis was initially done by biochemical tests and was confirmed by using an intragenic portion of the ace gene (18) as a probe for colony hybridization (data not shown). E. faecalis OG1RF (15), E. faecalis OG1SSp (6), E. faecalis V583 (23), and Enterococcus faecium TX0016 (2) are well-studied strains and were used in this study as controls. Brain heart infusion agar (Difco Laboratories, Detroit, Mich.) or brain heart infusion broth was used for growth.

The dormant phenotype acquired by Mycobacterium tuberculosis during infection poses a major challenge in disease treatment, since these bacilli show tolerance to front-line drugs. Therefore, it is imperative to find novel compounds that effectively kill dormant bacteria. By screening 4,400 marine natural product samples against dual-fluorescent M. tuberculosis under both replicating and nonreplicating conditions, we have identified compounds that are selectively active against dormant M. tuberculosis. This validates our strategy of screening all compounds in both assays as opposed to using the dormancy model as a secondary screen. Bioassayguided deconvolution enabled the identification of unique pharmacophores active in each screening model. To confirm the activity of samples against dormant M. tuberculosis, we used a luciferase reporter assay and enumerated CFU. The structures of five purified active compounds were defined by nuclear magnetic resonance (NMR) and mass spectrometry. We identified two lipid compounds with potent activity toward dormant and actively growing M. tuberculosis strains. One of these was commercially obtained and showed similar activity against M. tuberculosis in both screening models. Furthermore, puupehenone-like molecules were purified with potent and selective activity against dormant M. tuberculosis. In conclusion, we have identified and characterized antimycobacterial compounds from marine organisms with novel activity profiles which appear to target M. tuberculosis pathways that are conditionally essential for dormancy survival.