Analogues of the Potent Antitumor Compound ... - ACS Publications

Note pubs.acs.org/jnp

Analogues of the Potent Antitumor Compound Leiodermatolide from a Deep-Water Sponge of the Genus Leiodermatium Amy E. Wright,* Jill C. Roberts, Esther A. Guzmán, Tara P. Pitts, Shirley A. Pomponi, and John K. Reed Harbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, Florida 34946, United States S Supporting Information *

ABSTRACT: Two new analogues of the potent antitumor compound leiodermatolide, which we call leiodermatolides B and C, have been isolated from specimens of a deep-water sponge of the genus Leiodermatium collected off Florida. The compounds were purified using standard chromatographic methods, and the structures defined through interpretation of the HRMS and 1D and 2D NMR data. Leiodermatolide B (2) lacks the C-21 hydroxy group found in leiodermatolide and has equal potency as the parent compound, providing a simpler analogue for possible clinical development. It inhibits the proliferation of the AsPC-1 human pancreatic adenocarcinoma cell line with an IC50 of 43 nM. Leiodermatolide C (3) has a modified macrolide ring and is over 85-fold less potent with an IC50 of 3.7 μM against the same cell line. These compounds add to the knowledge of the pharmacophore of this class of potent antitumor agents.

P

reviously, we reported the structure and biological activity of leiodermatolide, 1, from a deep water sponge of the genus Leiodermatium.1 Leiodermatolide shows potent cytotoxicity against a panel of human tumor cell lines, induces cell cycle arrest at the G2/M checkpoint in the tumor cell lines tested, and shows tumor reduction in an animal model of pancreatic cancer.2 Leiodermatolide modifies tubulin dynamics2 but does not directly bind to tubulin3 or affect the assembly of purified tubulin in vitro.2 It has been shown to amplify centrosomes, and it has been proposed that it may work through centrosome declustering,3 making it different from other antimitotic compounds, and thus a potential new lead for anticancer drug discovery. The potent biological activity of leiodermatolide has led to its total synthesis by the Fürstner3,4 and Paterson groups5−7 with additional significant contributions from other laboratories.8−10 A series of analogues synthesized by the Fürstner laboratory has provided additional information on leiodermatolide’s cellular effects and the active pharmacophore.3 This paper describes the isolation, structure elucidation, and cytotoxicity of two analogues of leiodermatolide, leiodermatolides B (2) and C (3), which have been isolated from the sponge Leiodermatium sp. and which shed further light on the active pharmacophore. Compounds 2 and 3 were purified from combined specimens of the deep-water sponge Leiodermatium sp. collected using the Johnson-Sea Link manned submersible at depths ranging from © 2017 American Chemical Society and American Society of Pharmacognosy

390 to 425 m on the Miami Terrace, Florida. The specimens were exhaustively extracted with EtOH, the dried extract was partitioned between EtOAc and H2O, and the organic phase was chromatographed under vacuum column chromatography Special Issue: Special Issue in Honor of Phil Crews Received: December 10, 2016 Published: January 30, 2017 735

DOI: 10.1021/acs.jnatprod.6b01140 J. Nat. Prod. 2017, 80, 735−739

Journal of Natural Products

Note

Table 1. NMR Spectroscopic Data for Compounds 2 and 3 2a atom

δC, type

1 2a 2b 3a 3b 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20a 20b 21 22 23 24a 24b 25 26a 26b 27 28 29 30 31 32a 32b 33 34

172.4, C 34.2, CH2d 22.5, CH2 125.9, 137.5, 48.8, 78.4, 39.6, 68.0, 128.7, 126.4, 124.6, 138.0, 35.2, 82.9, 132.8, 130.4, 128.5, 133.3, 34.4,

CH C CH CH CH CH CH CH CH CH CH CH C CH CH CH CH2d

34.2, CHd 34.2, CHd 85.7, CH 27.9 CH2 9.9, CH3 11.5, CH3 16.8, 12.7, 16.7, 12.1, 13.8, 34.0,

CH3 CH3 CH3 CH3 CH3 CH2d

3b δH (J In Hz)

δC, type 174.1, C 31.8, CH2

2.28, m 1.97, m 2.18, m 2H

27.8, CH2

5.07, bdd (8.9, 5.5) 2.44, 3.24, 1.72, 5.86, 5.50, 6.35, 6.50, 5.33, 2.95, 5.04,

mc brd (10.3) q (7.2) d (10.3) t (10.3) t (11.3) t (11.3) t (10.3) tq (10.3, 6.7) d (10.4)

6.02, 6.27, 5.61, 2.26, 1.97, 2.01, 1.89, 4.03, 1.66, 1.63, 0.96, 1.39,

d (11.0) dd (15.1, 11.0) dt (15.1, 7.6) m m m ddq (4.1, 8.0, 6.8) ddd (8.0, 7.2, 4.1) m m t (7.5) s

1.09, 1.05, 0.84, 1.75, 0.97, 2.40, 2.31,

d (6.2) d (7.5) d (6.2) s d (6.8) dd, (17.5, 5.7) dd (17.5, 7.0)

171.2, C 157.6, C

88.1, CH 152.2 C 40.2, CH 74.8, CH 44.5, CH 71.9, CH 131.0, CH 128.2, CH 126.0, CH 138.2, CH 36.0, CH 84.6, CH 133.6, CH 131.2, CH 131.1, CH 127.7, CH 39.7, CH2 72.9, 44.9, 85.8, 28.1,

C CH CH CH2

9.6, CH3 113.6, CH2 17.0, 13.0, 16.7, 12.2, 11.7, 43.4,

CH3 CH3 CH3 CH3 CH3 CH3

δH (J In Hz) 2.27, 2.22, 1.88, 1.80, 4.18,

m m m m t (6.2)

2.17, 3.74, 2.02, 5.53, 5.50, 6.45, 6.53, 5.27, 3.07, 4.93,

m brm m m t (11.0) t (11.0) t (11.0) t (10.3) tq (10.3, 6.8) d (10.3)

6.09, 6.35, 5.81, 2.44, 2.20,

d (11.0) dd (14.8, 11.0) ddd (15.1, 7.6, 6.2) dd (14.4, 7.6) m

1.85, 4.00, 1.83, 1.60, 1.00, 5.12, 5.10, 1.14, 1.09, 0.84, 1.76, 1.02, 2.74, 2.32,

m ddd (10.1, 7.6, 3.1) m ddq (14.4, 7.5, 6.8) t (6.5) s s d (6.9) d (6.9) d (6.8) s d (6.9) d (16.5) d (16.5)

173.8, C 159.9, C

a1

H and 13C NMR data were recorded in CD2Cl2 at 600 and 125 MHz, respectively. b1H and 13C NMR data were recorded in methanol-d4 due to decomposition of the compound in CD2Cl2. cOverlaps with 32a coupling constants cannot be measured. dAssignments may be interchanged; there is insufficient resolution in the HSQC spectrum to distinguish these overlapping carbons.

as double doublets for 2, suggesting that the change in structure is replacement of the C-21 hydroxy group with a hydrogen atom. Full analysis of the HSQC, DQF COSY, and 1DDPFGSE-TOCSY experiments supports this hypothesis. The 13C NMR spectrum (Table 1) clearly showed the loss of the C-21 nonprotonated carbon observed at δC 72.0 ppm in leiodermatolide. Proton integration, g-DQF-COSY, and gHSQC spectra further supported the substitution by identifying a new methine group (δH 2.01, δC 34.2). The g-DQF-COSY spectrum allowed for assignment of the side chain with the following couplings: H-17→ H-18 →H-19 → H20ab → H-21 → H-32ab. The DQF-COSY spectrum clearly shows H-22 → H-23 → H24ab → H3-25, suggesting that the δ-lactone remains in 2. H-22 and H-21 are very close in chemical shift and also overlap with H-2ab and H-20b, making it difficult to clearly

conditions on a silica gel stationary phase to yield a fraction that by HPLC displayed the characteristic UV spectrum of leiodermatolide. Semipreparative HPLC on a C-18 reversedphase column led to the purification of leiodermatolide (1), 2, and 3 from this fraction. Inspection of the 13C NMR spectrum coupled with HRMS data suggested a molecular formula of C34H51NO7 for 2. The difference in molecular formula from that of leiodermatolide is loss of one oxygen atom. Comparison of the 1H NMR spectrum of 2 to that of leiodermatolide reveals that the major differences between the two spectra are found in the region between 1.5 and 3.0 ppm (Supporting Information Table S1 and Figure S4). The most noticeable change is that the isolated AX spin system observed for H2-32 in leiodermatolide is now observed as an AMX system with H-32a and -b both appearing 736



Note

assign a scalar coupling between H-21 and H-22. Similarly the chemical shifts for C-20, C-21, C-22, and C-32 are all observed between 33.5 and 34.0 ppm, and H-20ab, H-21, H-22, H-23, H24ab, H3-31, and H-32ab all have correlations in the g-HMBC spectrum to carbons in this range, making assignments to particular carbons unreliable. In the 1H NMR spectrum observed in methanol-d4 H-21 appeared as a multiplet with no apparent large couplings, suggesting that it adopts an equatorial position. The coupling constant between H-21 and H-32a is 6 Hz, while to H-32b it is 7 Hz, which is larger but consistent with an equatorial assignment. In a 1D-DPFSGE NOE spectrum, irradiation of H-23 resulted in enhancement of H-20a and -b as observed in leiodermatolide, further supporting that the C-20 side chain is in the axial position. The chemical shifts, NOE data, and scalar couplings observed for H-2 to H19 match those of leiodermatolide, suggesting that the remaining stereocenters in 2 have the same relative configuration as leiodermatolide (Supporting Information Table S1). The final proposed structure is 2, and we propose the common name leiodermatolide B for this compound. Inspection of the 13C NMR spectrum coupled with HRMS data for 3 suggested a molecular formula of C34H51NO10 for 3. Comparison of the 1H and 13C NMR data (Table S1) of 3 with those observed for leiodermatolide suggested that the difference in the structures is associated with the C-4−C-5−C-26 olefin (Figures S17, S24, and S26). Interpretation of the gHSQC spectrum clearly showed the loss of the C-26 methyl and C-4 olefinic methine resonances with the gain of an oxygenated methine (δH 4.18 and δC 88.1) and resonances attributable to an exocyclic olefin (δH 5.12 and 5.10 s; δC 113.6). The C-5 olefinic carbon also shifted from δC 137.5 to δC 152.2, which is consistent with the presence of an exocyclic olefin. Full interpretation of the DQF-COSY, HSQC, and HMBC spectra all suggested that the only change in the carbon skeleton of 2 was the replacement of the C-4−C-5−C-26 triad with a C-4 oxygenated methine adjacent to the C-5−C-26 exocyclic olefin. The molecular formula determined by HRMS had one more oxygen atom than would be expected for the C-4 allylic alcohol. Multiple measurements of the HRMS consistently returned a formula with the additional oxygen atom. Inspection of the spectra for other parts of the molecule showed very similar chemical shifts and scalar coupling patterns to the parent leiodermatolide, leaving us to conclude that 3 must contain a peroxide functionality at C-4. The observed chemical shifts were consistent with those reported for the products of 1O2 reaction with model homoallylic esters similar to that found in leiodermatolide (Figure 1).11 It is possible that 3 is an artifact formed during the chemical isolation via autooxygenation of leiodermatolide, although it was detected by HPLC in extracts made on shipboard. During the structure elucidation of leiodermatolide, the Paterson group attempted to prepare modified Mosher esters without success. One reaction product they isolated appeared to be a leiodermatolide derivative bearing a chlorine atom at C-4 (Paterson, personal communication), suggesting that this center may be prone to ene reactions in which an enophile adds to the double bond and creates the exocyclic olefin. For molecule 3, the changes in C-4−C-5−C-26 result in a conformational change of the macrolide versus that of leiodermatolide. In leiodermatolide, H-8 did not show scalar coupling to either H-7 or H-9 (it was at 90 deg to both of its vicinal protons). In 3, the 2D-DQF-COSY correlations between

Figure 1. (A) 13C chemical shifts for carbons bearing a hydroperoxide functionality resulting from 1O2 reaction with model homoallylic esters; (B) 13C chemical shifts observed in 3 for the putative reaction of leiodermatolide with 1O2 to form 3.

these protons and the splitting patterns for the protons clearly indicate that these protons are coupled (Figures S19, S20, S21, and S28). Although the conformation of the macrolide ring has changed, NOE data for 3 (Table S3) are consistent with the same relative configuration as found in leiodermatolide for the remaining stereocenters in the molecule. Interpretation of the 2D-NOESY spectrum suggests that H-4 is on the top face of the molecule. H-4 has correlations to H-26a, H-2ab, H-3ab, and H-6. During the isolation of leiodermatolide it was noted that NMR spectra collected in methanol-d4 resulted in broadening of the resonances for C-7 to C-9, which were thought to participate through hydrogen bonding of the carbonyl and the hydroxy proton on C-7 in a ring-like structure. Similar broadening of the NMR resonances was observed for 3, providing additional support for the assignment of the relative configuration of these centers. The final proposed structure is 3, and we propose the common name leiodermatolide C for this compound. As with leiodermatolide, it is not possible to correlate the configuration of stereocenters in the macrolide with those of the δ-lactone for either analogue. We show the same absolute configuration as has been determined through total synthesis for leiodermatolide, but confirmation requires total synthesis. The specific rotation for 2 ([α]24D −83.4 (c 0.26, MeOH)) is quite similar to that of leiodermatolide ([α]24D −84 (c 0.34, MeOH)), while that of 3 ([α]24D −54 (c 0.08, MeOH)) is substantially different but still negative. Leiodermatolide and compounds 2 and 3 were tested for cytotoxicity against the AsPC-1 human pancreatic adenocarcinoma cell line using an MTT assay. The dose required to inhibit 50% of the cell proliferation for each compound (IC50) was determined. Leiodermatolide and 2 had low nM level activity against the cell line, with IC50 values of 46 ± 2 and 43 ± 4 nM, respectively. Compound 3 was over 85-fold less potent, with an IC50 of 3.7 ± 0.1 μM. Taken together these data suggest that modification of the C-4 to C-6 portion of the macrolide reduces activity and is critical to the biological activity of leiodermatolide. Modification of the homoallylic ester leads to significant changes in the conformation of the macrolide versus that observed in the parent leiodermatolide, and it is possible that this ring conformation may contribute to the biological activity. The Fürstner group prepared a series of analogues of leiodermatolide3 demonstrating that modification of the C-7 alcohol through ether formation results in loss of activity. The carbamate can be replaced with acetate, but loss of both the carbonyl functionality at C-9 and the C-7 hydroxy led to loss of activity. Significant modification of the δ-lactone also 737



Note

graphic conditions using a step gradient of EtOAc in heptane as follows: fraction 1, heptane/EtOAc (4:1 v/v) 200 mL; fraction 2, heptane/EtOAc (3:2 v/v) 200 mL; fraction 3, heptane/EtOAc (2:3 v/ v) 200 mL; fraction 4, heptane/EtOAc (1:4 v/v) 200 mL; fraction 5, EtOAc 200 mL. Leiodermatolide and the two analogues (2 and 3) were detected in fraction 4 and were further purified by semipreparative HPLC (Vydac C-18 Protein and Peptide column (10 mm × 250 mm), isocratic elution with H2O/CH3CN (1:3 v/v), flow = 3 mL/min, detected by UV at λ = 230 nm, tR = 13.0 min for leiodermatolide (1), tR = 15.0 min for compound 2, tR = 11.0 min for compound 3). A total of 3.4 mg of 2 (2.85 × 10−4% of wet weight of sponge) and 1.4 mg of 3 (1.2 × 10−4% of wet weight) were obtained. This separation also yielded 11.1 mg of leiodermatolide (9.3 × 10−4% of wet weight of sponge). Leiodermatolide B (21-deoxyleiodermatolide) (2): amorphous, white solid; [α]24D −83.4 (c 0.26, MeOH); UV (MeOH) λmax (log ε) 244 (4.6) nm; IR (KBr) νmax 3364, 2928, 1726, 1369, 1205, 1147, 1033, 989, 745 cm−1; 1H and 13C NMR data, Table 1; HRMS m/z 608.3561 [M + Na]+ (calcd for C34H51NO7Na, 608.356324 Δ = −0.1 mmu). Leiodermatolide C (3): amorphous, white solid; [α]24D −54 (c 0.08, MeOH); UV (MeOH) λmax (log ε) 244 (4.6) nm; IR (KBr) 3363, 2930, 1703, 1596, 1376, 1249, 1153, 1033, 1003, 916, 780, 749, 678 cm−1; 1H and 13C NMR data, Table 1; HRMS m/z 656.3419 [M + Na]+ (calcd for C34H51NO10Na, 656.3411, Δ = 0.8 mmu). Cytotoxicity Assays. Compounds 1−3 were evaluated for their effects on proliferation of ASPC-1 human pancreatic adenocarcinoma cells (ATCC No. CRL-1682). The ASPC-1 cell line was obtained from the American Type Culture Collection. Details of the cytotoxicity assays were described in an earlier publication.15 Assays were run in 96-well plates. After 72 h of incubation with the compound cytotoxicity was measured spectrophotometrically using MTT as the indicator. All samples were assayed a minimum of three times to derive the final IC50 value. Results are presented as the average value ± standard deviation.

reduces biological activity, but the C-21-deoxy compound (2) was not reported in their study. To summarize, isolation and structure elucidation of metabolites from the sponge Leiodermatium revealed two new analogues of leiodermatolide, which we call leiodermatolides B and C. Removal of the C-21 alcohol functionality has no effect on potency against the AsPC-1 tumor cell line, while modifications of the C-4−C-5−C-26 triad results in near total loss of activity. It is possible that the latter functionality is involved in binding to the molecular target through an ene-type reaction.

■

EXPERIMENTAL SECTION

General Experimental Procedures. Optical rotations were measured on a PerkinElmer 343 digital polarimeter. UV spectra were collected on a Hitachi U-3010 spectrophotometer. IR spectra were collected on a Thermo Nicolet IR100 spectrometer with potassium bromide discs. NMR data was collected on a JEOL ECA600 spectrometer operating at 600 MHz for 1H and 150.9 MHz for 13 C (g-HSQC, g-HMBC, 2D-g-DQF-COSY, 2D-NOESY, 1DDPFGSE-NOE). Chemical shifts were recorded using an internal deuterium lock for 13C and residual 1H in CD2Cl2 (δH 5.32, δC 54.0) or methanol-d4 (δH 3.31, δC 48.0). The edited-g-HSQC spectrum was optimized for 140 Hz, and the g-HMBC spectrum was optimized for 8 Hz. High-resolution mass spectrometry was performed using a JEOL AccuTOF DART mass spectrometer. Biological Material. A number of sponge specimens were evaluated in this study. All of them have been classified as an unidentified species of Leiodermatium Schmidt, 1870 [phylum Porifera, class Demospongiae, order Tetractinellida, family Azoricidae].12 All specimens studied were collected using the Johnson-Sea Link manned submersible at depths between 390 and 425 m on the Miami Terrace, off the coast of Ft. Lauderdale (26°01.277′ N latitude and 79°49.266′ W longitude). Leiodermatium is part of a polyphyletic group of sponges, the lithistids, or rock sponges, characterized by choanosomal articulated spicules called desmas that form a rigid skeleton.13 The genus has a worldwide distribution; two species (L. lynceus and L. pfeif ferae) are reported from the tropical western Atlantic;14 however, a thorough taxonomic revision of the genus is needed.13 The specimens in this study are foliated, rippling plates, ∼25 cm diameter. The plate is ∼4 mm thick, very hard in consistency, and tan in color. The convex side is relatively smooth, and the inner side has small raised pores. The choanosomal skeleton consists of rhizoclone desmas that are strongly branched, and the tips are divided into fine spiny processes. No microscleres are apparent, although some oxeas, 200− 350 μm in length, are present. The plate morphology, raised pores, and the presence of oxeas make this species most closely related to Leidermatium lynceus, identified from Barbados.14 The other species reported to occur in the tropical western Atlantic, Leiodermatium pfeifferae, is not well described, so it is impossible to determine if the specimens in this study are similar to L. pfeif ferae. A museum voucher specimen of the primary specimen used in the study is deposited at the Harbor Branch Oceanographic Museum, catalog number 003:01035, and is available for inspection upon request. Isolation of 2 and 3. These two compounds were originally detected with the extraction of a 1037 g sample of Leiodermatium sponge from the Miami Terrace (1-VI-04-2-005, stored frozen at −20 °C until workup). Due to low yields, additional samples collected from the same site on the Miami Terrace off the coast of Ft. Lauderdale were used in the purification of 2 and 3 used in the NMR studies. A typical fractionation procedure follows. Three specimens of Leiodermatium sp. (HBOI samples: 25-V-04-2-002 (389 g), 4-VIII05-2-001 (285 g), and 1-VI-04-2-005 (517 g)) were extracted exhaustively with a mixture of EtOAc/EtOH (9:1) using a Waring blender. Extracts were concentrated using distillation under reduced pressure and then partitioned between EtOAc and H2O. The EtOAc partition was chromatographed on a silica gel stationary phase (Kieselgel 60H, 150 mL column) under vacuum column chromato-

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.6b01140. Experimental procedures, pictures of the specimens used in the study, and copies of NMR spectra for compounds 2 and 3 (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 772-242-2459. ORCID

Amy E. Wright: 0000-0003-4872-9776 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This publication was made possible by grant 1R01CA093455 from the National Cancer Institute at the National Institutes of Health. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NCI. Funding for this project was also provided by the State of Florida Center of Excellence in Biomedical & Marine Biotechnology (COE-HRE07), NOAA Cooperative Institute for Ocean Exploration, Research and Technology (NA09OAR4320073), and the Health Resources & Services Administration Center for Sustainable Use of Marine Resources (4C76HF00231-01-04). We thank R. Cody of JEOL USA for 738



Note

high-resolution mass spectrometry data. This paper is HBOI Contribution Number 2067.

■

DEDICATION Dedicated to Professor Phil Crews, of the University of California, Santa Cruz, for his pioneering work on bioactive natural products.

■

REFERENCES

(1) Paterson, I.; Dalby, S. M.; Roberts, J. C.; Naylor, G. J.; Guzman, E. A.; Isbrucker, R.; Pitts, T. P.; Linley, P.; Divlianska, D.; Reed, J. K.; Wright, A. E. Angew. Chem., Int. Ed. 2011, 50, 3219−3223. (2) Guzmán, E. A.; Xu, Q.; Pitts, T. P.; Mitsuhashi, K. O.; Baker, C.; Linley, P. A.; Oestreicher, J.; Tendyke, K.; Winder, P. L.; Suh, E. M.; Wright, A. E. Int. J. Cancer 2016, 139, 2116−2126. (3) Mailhol, D.; Willwacher, J.; Kausch-Busies, N.; Rubitski, E. E.; Sobol, Z.; Schuler, M.; Lam, M.-H.; Musto, S.; Loganzo, F.; Maderna, A.; Fürstner, A. J. Am. Chem. Soc. 2014, 136, 15719−15729. (4) Willwacher, J.; Kausch-Busies, N.; Fürstner, A. Angew. Chem., Int. Ed. 2012, 51, 12041−12046. (5) Paterson, I.; Paquet, T.; Dalby, S. M. Org. Lett. 2011, 13, 4398− 4401. (6) Paterson, I.; Ng, K. K. H.; Williams, S.; Millican, D. C.; Dalby, S. M. Angew. Chem., Int. Ed. 2014, 53, 2692−2695. (7) Paterson, I.; Williams, S. Isr. J. Chem. 2016, DOI: 10.1002/ ijch.201600084. (8) Rink, C.; Navickas, V.; Maier, M. E. Org. Lett. 2011, 13, 2334− 2337. (9) Reiss, A.; Maier, M. E. Org. Lett. 2016, 18, 3146−3149. (10) Navickas, V.; Rink, C.; Maier, M. E. Synlett 2011, 2011, 191− 194. (11) Griesbeck, A. G.; Cho, M. Tetrahedron Lett. 2009, 50, 121−123. (12) van Soest, R. (2006). “Leiodermatium Schmidt 1870”. In Van Soest, R. W. M.; Boury-Esnault, N.; Hooper, J. N. A.; Rützler, K.; de Voogd, N. J.; Alvarez de Glasby, B.; Hajdu, E.; Pisera, A. B.; Manconi, R.; Schoenberg, C.; Klautau, M.; Picton, B.; Kelly, M.; Vacelet, J.; Dohrmann, M.; Díaz, M.-C.; Cárdenas, P.; Carballo, J. L. (2016). World Porifera database. http://www.marinespecies.org/porifera/ porifera.php?p=taxdetails&id=132086, Accessed Dec 9, 2016. (13) Pisera, A.; Lévi, C. 2002, In Systema Porifera; Hooper, J. N.A.; Van Soest, R. W. M., Eds.; Kluwer Academics/Plenum Press: New York, pp 299−301 and pp 353−356. (14) Van Soest, R. W. M.; Stentoft, N. Stud. Fauna Curacao Caribb. Isl. 1988, 70, 1−175. (15) Gunasekera, S. P.; Zuleta, I. A.; Longley, R. E.; Wright, A. E.; Pomponi, S. A. J. Nat. Prod. 2003, 66, 1615−1617.

739


Analogues of the Potent Antitumor Compound ... - ACS Publications

Recommend Documents