|Home | About | Journals | Submit | Contact Us | Français|
A new proline-containing cycloheptapeptide, euryjanicin A (1), has been isolated from the marine sponge Prosuberites laughlini indigenous to Puerto Rico, and its structure established by an X-ray crystal structure determination. The absolute configuration of each amino acid residue was determined by Marfey’s method.
Marine sponges, especially those found in tropical ocean areas, continue to represent the single most prolific source of structurally novel natural products of marine origin.1 Sponge-derived secondary metabolites have also attracted considerable attention because of their relevant and diverse pharmacological actions.2 During the last 25 years a large body of evidence has been accumulated to indicate that certain marine sponges (especially sponges belonging to the class Demospongiae) also contain amino acid derivatives, usually as trace components, with significant cancer cell growth inhibitory properties.3 Examples of such biologically active peptides are the axinellins,4 kapakahines,5 microsclerodermins,6 polytheonamides,7 papuamides,8 stylisins,9 hymenamides,10 wainunuamide,11 and dominicin.12 Additional noteworthy examples include the laxaphycins13 and discodermins14 (they inhibit tumor promotion), calyculin15 (tumor promoting) and the anti-thrombin cyclotheonamides.16
As part of our ongoing campaign to find new natural products with potential anticancer activity from marine organisms found near Puerto Rico, a small library of sponge extracts from this location was recently screened for anti-cancer activity. Considerable activity was detected in the crude extract of Prosuberites laughlini (Díaz, Alvarez & Van Soest, 1987) (Order Hadromerida, Family Suberitidae).17 The sponge was collected in April 2006 at a depth of about 15 m by scuba from El Natural reef, off the municipality of Aguadilla, in the northwestern coast of Puerto Rico (18°27′46.90″N, 67°10′06.11″W). A freeze dried sample (623 g) was repeatedly extracted with MeOH (16 L), and the extracts were combined, concentrated in vacuo (83 g), and partitioned between H2O (1.5 L) and EtOAc (3 × 1 L), followed by n-butanol (3 × 1 L). The EtOAc and n-butanol extracts were concentrated under vacuum to give 6.1 g and 14.8 g of residue, respectively. Significant anti-cancer activity was detected in the EtOAc fraction; upon treatment with 1.0 mg/mL extract, reduction of cell viability was 15, 10, and 19% for DU145 prostate cancer, A2058 melanoma, and MDA-MB-435s breast cancer types, respectively. The EtOAc extract was subjected to normal-phase silica gel (200 g) column chromatography (CC) using a mixture of CHCl3 and MeOH which had been previously saturated with dry NH3 (80/20). Similar fractions were pooled together on the basis of TLC analysis. Subsequent evaluation of the fraction eluting first (2.8 g) by 1H and 13C NMR indicated that this pooled fraction contained a mixture of small peptides. Further fractionation was performed by normal-phase silica gel (100 g) CC with a mixture of hexane and EtOAc (95/5). Close inspection by NMR indicated that sub-fractions 10 (183 g) and 11 (156 mg) appeared to contain most of the peptides observed in the initial NMR analysis. Purification of sub-fraction 10 was achieved by C18 silica gel (5 g) reversed-phase CC using a mixture of MeOH and H2O (85/15) as eluent; this yielded the new compound euryjanicin A (1, 44 mg, 0.05%) as well as the known cyclic octapeptide dominicin (2, 33 mg, 0.04%).12 Also isolated was a new cycloheptapeptide, which is currently still under investigation and will be reported separately.
Euryjanicin A (1),18 was obtained as optically active, [α]D20 −18.0° (c 1.0, CHCl3), colorless crystals that showed a pseudomolecular ion peak at m/z 827 in the ESIMS spectrum. The molecular formula of 1 was determined to be C44H58N8O8 by the HRESIMS [m/z 827.4439 (M+H)+, C44H59N8O8, Δ −1.7 mDa], requiring 20 sites of unsaturation. The intense IR absorptions at 3316 and 1644 cm−1 were attributed to amino and amide carbonyl groups, respectively, and the UV absorptions at λmax 281 (ε 1800) and 290 (ε 1600) nm indicated the presence of aromatic ring(s). The 1H NMR spectrum of 1 (Figure 1A) was indicative of a peptide. Standard amino acid analysis of the acid hydrolysate of 1 revealed the presence of one mole each of valine (Val), phenylalanine (Phe), tryptophan (Trp), serine (Ser), and isoleucine (Ile), and two moles of proline (Pro) residues, which subsequently, were all confirmed to be of L-configuration by Marfey’s derivatization and subsequent HPLC analysis with standards.19,20 Euryjanicin A (1) was negative to a ninhydrin test, which together with its solubility in organic solvents and the 20 degrees of unsaturation, indicated a cyclic structure.
The crystals of 1 gave a sharp, well-resolved peak when analyzed by reversed-phase HPLC using a variety of solvent systems, in addition to a sharp melting point range (mp 174–176 °C) and a single, clean molecular ion in the HRESI. Despite these indications of purity, the initial structural approaches to euryjanicin A (1) employing high field 2D-NMR techniques were complicated by the existence of a mixture of slowly interconverting conformers in varying proportions depending on the solvent used (e.g., CDCl3, DMSO-d6, acetone-d6, MeOH-d4, pyridine-d5, MeCN-d3, benzene-d6). Attempts to simplify the 1H NMR spectra by cooling (to −15 °C) and heating (to 60 °C), or adjusting the pH by addition of TFA or Et3N, did not alleviate the problem, and typically resulted in further loss of resolution which precluded structural studies. Likewise, many of the resonance lines in the 13C NMR spectra were doubled, tripled, or broadened when the spectra were recorded in the same solvents (Figure 1B). Because the 1D and 2D NMR spectra of euryjanicin A in these organic solvents were complex and difficult to interpret, only a handful of NMR spectral assignments could be made with complete confidence.21 Therefore, efforts to determine the amino acid sequence by 2D NMR experiments were abandoned.
Crystals suitable for X-ray crystallographic analysis were obtained by slow recrystallization from MeOH(NH3). The results obtained from this analysis (Figure 2) defined the relative stereochemistry at all centers, and the absolute configuration shown was selected to agree with the known amino acid centers.22 The geometry of the peptidic linkages about the proline residues was also assigned on the basis of the X-ray diffraction method, which showed that the two proline peptide bonds were cis. The cis peptide bonds and bifurcated transannular hydrogen bonds between the C44 carbonyl oxygen (O8) and N4–H (2.50 Å) and N5–H (2.43 Å) set the sharp turn conformation of the macrocyclic ring. An additional hydrogen bond (2.18 Å) was pinpointed between primary hydroxyl oxygen O4 of the Ser4 residue and N6–H. This turn results in the side chains of the only two aromatic amino acids, L-Trp1 and L-Phe5, being projected in the same direction. The absolute structure 1 of euryjanicin A was thereby completed and thus confirmed as cyclo(Trp1–Pro2–Ile3–Ser4–Phe5–Val6–Pro7). The fact that other investigators have isolated similar cyclic peptides from extracts of cultured marine bacteria and fungi, suggests a possible microbial or symbiotic origin for metabolites euryjanicin A (1) and dominicin (2).23
Unlike the EtOAc extract, strong anti-tubercular activity was detected in the n-butanol partition fraction; thus a small portion of the crude extract (2 g) was subjected to C18 silica gel (50 g) reversed-phase CC using a mixture of MeOH and H2O (85/15). Similar fractions were combined on the basis of TLC and 1H NMR analyses. The first (most polar) fraction to elute out (0.90 g) was further chromatographed over normal-phase silica gel (100 g, dry packing), initially with a mixture of CHCl3 and MeOH(NH3) (90/10), and later using gradients of increasing polarity. This procedure yielded the known compounds hymenidin (3, 66 mg, 0.08%) and monobromoisophakellin (4, 57 mg, 0.07%).24,25
Pure euryjanicin A (1) and dominicin (2) were evaluated for their inhibitory activity against cell viability in melanoma (A2058), breast cancer (MDA-MB-435s), and prostate cancer (DU145). Surprisingly, none of them displayed significant inhibitory activity.9 On the other hand, in the National Cancer Institute’s one-dose (10−5 M) 60-cell-line assay, euryjanicin A (1) inhibited a non-small cell lung and renal cancer cell line: NCI-H522 displayed growth at 72% of the mean and UO-31 at 73% of the mean. Dominicin (2), in turn, inhibited a leukemia (SR) and a renal (UO-31) cancer cell line, displaying growths at 80% and 32% of the mean, respectively. Preliminary tests for anti-tuberculosis activity of alkaloids 3 and 4 demonstrated that these compounds significantly inhibited the growth of Mycobacterium tuberculosis H37Rv (MIC values = 6.1 and 64.0 μg/mL, respectively). On the other hand, alkaloids 3 and 4 lacked significant anti-plasmodial activity as their IC50s against the malaria parasite Plasmodium falciparum(chloroquine-resistant W2 clone) were ≥ 10 μg/mL.
We thank Dr. Ileana I. Rodríguez for helping out during the collection of P. laughlini, Dr. Hong Zhao for assistance with the X-ray data, and the UPR-MARC and UPR-RISE Fellowship Programs for financial support to JV and BV, respectively. Sponge extracts were screened for anti-cancer activity by Dr. Sangkil Nam at the City of Hope Beckman Research Institute and for anti-TB activity by Dr. Yuehong Wang at the Institute for Tuberculosis Research of the University of Illinois. Anti-cancer, anti-mycobacterial, and anti-malarial bioassays of the pure natural products were conducted at the National Cancer Institute (NCI), the Institute for Tuberculosis Research of the UIC, and the Instituto de Investigaciones Avanzadas y Servicios de Alta Tecnología (Panama), respectively. Major financial support was provided by the NIH-MBRS SCORE program (Grant S06GM08102) of the University of Puerto Rico.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.