Figure 1.
A side–by–side comparison of the three natural products shown here illustrates an example of parallel biosynthetic pathways that operated in disparate organisms including marine sponges and myxobacteria. These compounds, arising from the fusion of a triketide with unusual tripeptide moieties, represent the types of biosynthetic products targeted in this study owing to their parallel biogenetic and potential microbial origins. Each compound was previously discovered from the indicated source organisms, and all were subsequently shown to be F–actin stabilizers.
Table 1.
Deployments.
Figure 2.
The artificial sponge was mounted underwater proximal to sponges envisioned to be engaged in secondary metabolite biosynthesis, such as shown in Figure 1. Its components are: particle filters (disposable funnel containing a polyethylene frit, OP–6602–14, ChemGlass), microbial filters (2 µm pore, 50 mm OD, Millex–AP microbial filters, SLA05010, Millipore); a SeaBattery (DeepSea Power & Light); a power supply (self–built); a microdiaphram pump (NF5RP, KNF Neuberger); a hollow–fiber bioreactor (4300–C5011, FiberCell Systems); and parallel–bundle of sep–pak cartridges (ePlastics) containing Amberlite XAD–18 resin (Dow Chemicals). The hollow–fiber bioreactor can act as a bioreactor allowing microbial material to culture inside the artificial sponge. The sep–pak cartridges serve to collect materials either from the seawater or from the microbial content within the hollow–fiber bioreactor. Green arrows indicate flow direction during charging of the hollow–fiber bioreactor from the water column during the inoculation stage (step 1, Figure 3). Blue arrows depict the passage of seawater through the artificial sponge during the incubation stage (step 2, Figure 3). A generic depiction of the anatomy of a sponge is shown within the inset to illustrate the parallel engineered design, codes are: os = osculum, spc = spongocoel, chc = choanocytes, amc = amebocyte, pc = porocytes, sp = spicule, mes = mesohyl, pic = pinacocytes.
Figure 3.
The capture and isolation of natural products through a three–stage, four–step procedure.
The process began with inoculation of the hollow–fiber bioreactor (Step 1) followed by incubating the artificial sponge (Step 2) for 7–14 days (Table 1) proximal to the inoculation site. The systems were moved from their inoculation site to a second incubation site, devoid of sponges and fragile sea organisms, in an effort to prevent damage to marine specimens. After the completion, the sep–pak cartridges were harvested and eluted to provide a crude extract (Step 3). A total of 10 extracts were obtained (Table 1). Next, a reverse–affinity method (Step 4) was deployed to isolate natural products based on their ability to bind to actin. This provided a semi–pure fraction from each extract as illustrated in Figure 4 (code: AC–X–A) from extract SES–009 (Table 1). A detailed description of these steps has been provided in the Materials and Methods.
Figure 4.
Accurate mass analysis on a <10 µg sample: The actin–bound biosynthetic products were isolated from a sample (code: AC–X–A) via the procedure shown in Figure 3, Step 4.
The top panel shows, by Total Ion Chromatogram (TIC) areas, three compounds of interest observed in a ratio of 37∶54∶9 from this material. The three lower panels show AM–MS data for dereplication of compounds at: 2.34 min = jasplakinolide C (2), 2.47 min = jasplakinolide B (3), and 2.85 min = jasplakinolide (1), respectively.
Figure 5.
Final dereplication by NMR analysis: Sample AC–X–A (∼10 µg) containing the mixture shown in Figure 4 was subjected to NMR determinations at 26°C in CD3OD using a high sensitivity 1.7 mm TCI MicroCryoProbe on a 600 MHz Avance Spectrometer (Bruker Biospin).
The annotations shown for the (A) 1H and (B) 1H–1H gCOSY NMR spectra confirm the dereplication assignments proposed in Figure 4 for jasplakinolide B (3) (protons coded as “b”) and jasplakinolide C (2) (protons coded as “c”), while resonances for jasplakinolde (1) could not be unambiguously observed. Ratios of 3 to 2 were estimated to be 60: 40 by peak areas shown in Figure S7 in File S1.