PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Org Chem. Author manuscript; available in PMC 2013 April 6.
Published in final edited form as:
PMCID: PMC3321083
NIHMSID: NIHMS362110

Erythrolic acids A-E, Meroterpenoids from a Marine-Derived Erythrobacter sp

Abstract

Erythrolic acids A-E (1–5) are five unusual meroterpenoids isolated from the bacterium Erythrobacter sp. derived from a marine sediment sample collected in Galveston, TX. The structures were elucidated by means of detailed spectroscopic analysis and chemical derivatization. The erythrolic acids contain a 4-hydroxybenzoic acid appended with a modified terpene side chain. The side chain modifications include oxidation of a terminal methyl substituent and in the case of 1–4 addition of a 2-carbon unit to give terpene side chains of unusual length; C22 for 1 and 2, C17 for 3 and C12 for 4. The relative and absolute configurations of the meroterpenoids were determined by coupling constant, NOE and Mosher’s analysis. In vitro cytotoxicity towards a number of non-small cell lung cancer (NSCLC) cell lines revealed only modest activity for erythrolic acid D (4) (2.5 μM against HCC44). The discovery of these unusual diterpenes, along with the previously reported erythrazoles, demonstrate the natural product potential of a previously unstudied group of bacteria for drug discovery. The unusual nature of the terpene side chain, we believe, involves an oxidation of a terminal methyl group to a carboxylic acid and subsequent Claisen condensation with acetyl-CoA.

INTRODUCTION

Meroterpenoids are natural products of mixed biosynthetic origin, containing a terpene element in combination with a carbon skeleton derived from other biosynthetic pathways, such as the shikimate or polyketide pathways.1 There are relatively few meroterpenoids from bacterial sources, with the napyradiomycins, neomarinone, furaquinocin C and azamerone being among the few examples.2 We have found that bacteria of the genus Erythrobacter are prolific producers of meroterpenoids. Erythrobacter are strict aerobic Gram-negative bacteria that are ubiquitous in the marine environment. Many species of Erythrobacter are known to be producers of carotenoids, which are responsible for the smooth red-orange color of the colonies growing on agar plates.3 However, to the best of our knowledge there are no other reports of natural products from this class of bacteria.

As part of our efforts to search for novel bioactive natural products from marine-derived bacteria, we investigated the secondary metabolism from strain SNB-035 that by 16S rRNA analysis was identified as closely resembling Erythrobacter sp. Our studies on this strain reveal that Erythrobacter species are prolific producers of unique bioactive natural products. We have previously reported the isolation of two natural products from this bacterial strain, erythrazoles A and B (6–7), which contain a benzothiazole moiety coupled to a terpene side chain.4 The molecules we report here are related to 6 and 7, displaying similar two-carbon homologated terpenes. Bioassay guided fractionation using the Locus Derepression assay (LDR), which identifies molecules that modulate epigenetic regulation, led to the isolation of five meroterpenoids erythrolic acids A-E (1–5), containing a hydroxybenzoic acid moiety and in the case of 14, a two-carbon homologated terpene side-chain. Of the analogs, only erythrolic acid D (4) was found to have modest cytotoxicity. Although the biological activity of these compounds is modest at best in the assays for which they have been evaluated, the unusual two-carbon homologation provides a terpenoid skeleton that could be further explored.

RESULTS AND DISCUSSION

Marine bacterium SNB-035 was isolated from a sediment sample collected from Trinity Bay, Galveston, TX (29° 42.419′ N, 94° 49165′ W) and isolated on seawater based acidic Gauze media. Analysis of the16S rRNA revealed 98% identity to Erythrobacter citreus. Large-scale shake fermentation (30 L) was carried out to obtain sufficient material for full chemical and biological analysis of the new analogs. The excreted metabolites were collected using XAD-7 resin and the resulting crude extract purified by a combination of solvent/solvent extraction and reversed phase chromatography to give fractions enriched in terpenoid metabolites. Final purification by gradient reversed phase HPLC gave erythrolic acids A (1, 2.5 mg), B (2, 1.5 mg), C (3, 0.7 mg), D (4, 1.5 mg), and E (5, 0.6 mg) (Figure 1).

Figure 1
Structures of natural products from Erythrobacter sp. strain SNB-035

The molecular formula of erythrolic acid A (1) was determined to be C29H42O7 (calcd for C29H41O7, 501.2852), based on a high-resolution ESI-MS [M − H] of 501.2849, indicating nine degrees of unsaturation. The 1H NMR spectrum of 1 (Table 1) showed signals for a 1,3,4-trisubstitued benzene ring with protons at δH 7.76 ppm (d, J = 2.0 Hz, H-2), 7.69 ppm (dd, J = 8.4, 2.0 Hz, H-6), and 6.77 ppm (d, J = 8.4 Hz, H-5); three vinyl methyl singlets at δH 1.73, 1.71, 1.61; three olefinic 1H (δH 5.35, 5.34, and 5.21); and twelve allylic methylene protons. Analysis of 13C NMR and HSQC spectra revealed 29 carbons corresponding to two carbonyls (δC 170.6 and 176.7), one oxygenated quaternary sp2 carbon (δC 161.2), eleven sp2 carbons (δC 115.4–140.3), two oxygenated sp3 carbons (δC 75.6 and 78.3) and nine sp3 carbons (δC 16.4–41.1). In the HMBC spectrum of 1, both of the proton signals at δH 7.76 and 7.69 showed correlations to the carbonyl carbon at δC 170.6 and the oxygenated quaternary carbon at δC 161.2, indicating that compound 1 is a 3-substituted 4-hydroxybenzoic acid derivative, with a 22 carbon side chain. A combination of 1H NMR, 13C NMR, 1H-1H COSY and gHMBC data (Table S1) allowed us to assemble an unusual 22 carbon terpene fragment terminating in a carboxylic acid. A few key HMBC correlations (Figure 2) for the terpene portion of 1 were from the H3-26 singlet (δH 1.71, s) to C-9 (δC 123.5), C-10 (δC 137.6), and C-11 (δC 41.0) and from the H3-27 singlet (δH 1.61) to C-13 (δC 125.6), C-14 (δC 136.3), and C-15 (δC 38.1). The saturated portion of the terpene was assigned based on HMBC correlations from the H3-28 singlet (δH 1.07) to C-17 (δC 78.3), C-18 (δC 75.6), and C-19 (δC 39.2) and from the H3-29 singlet (δH 1.73) to C-21 (δC 33.5), C-22 (δC 140.3), and C-23 (δC 118.2). COSY correlations between H-15/H-16 and H-16/H-17 fully establish the saturated spin system. A COSY correlation between H-23 (δH 5.35) and H-24 (δH 3.01) combined with HMBC correlations from both H-23 and H-24 to the carbonyl carbon at δC 176.7 established the side chain, with the unusual two carbon homologation to a C22 terpene.

Figure 2
Key 1H-1H COSY and HMBC correlations for compound 1
Table 1
NMR data for erythrolic acids A (1) and B (2) in CD3OD

The downfield chemical shift of the H2-8 doublet (δH 3.31) and HMBC correlations from H-8 to C-2, C-3, and C-4 indicate the terpene side chain is located at C-3 of the benzoic acid ring. The two carboxylic acids in 1 were further confirmed by conversion of 1 to the bis-methyl ester (1a) with TMS-CHN2, giving rise to two new methyl singlets at δH 3.65 and 3.84 ppm in the 1H NMR. The configuration of the double bonds in 1 were deduced as 9E, 13E, and 23Z based on the chemical shifts5 of C-26 (δC 16.4), C-27 (δC 16.4) and C-29 (δC 23.7) and confirmed by ROESY correlations between H-26/H-8, H-9/H-11, H-27/H-12, H-13/H-15, H-29/H-23, and H-24/H-21.

The absolute configuration of C-17 in 1 was determined by application of the modified Mosher method on its tri-methylated product 1b, which was generated via exhaustive treatment of 1 with TMS-CHN2 (Figure 3).6 Treatment of 1b, in separate experiments, with R-(−)- and S-(+)-α-methoxy-α-(trifluoromethyl)phenylacetyl chloride (MTPA-Cl) produced the corresponding S- and R- Mosher esters 1d and 1e, respectively. Interpretation of the 1H NMR chemical shift differences (Δδ = δSδR) between 1d and 1e, established the absolute configuration of C-17 as S by applying Mosher ester analysis (Figure 3). To assign the relative configuration of the 17,18-diol, the dimethyl ester 1a was transformed into acetonide 1c by treatment with dimethoxypropane in acetone with catalytic pyridinium p-toluenesulfonic acid. 1D NOE experiments (Figure 4 and Table S2) on 1c showed NOE enhancements on the C-28 methyl singlet and CH3-30α by irradiating H-17, enhancement of H-17 and CH3-30β upon irradiating CH3-30α, and enhancement of H-17 and CH3-30α on irradiation the methyl singlet H-28. Full analysis of the NOE data and the confirmation of the five-membered ring acetonide7 indicated the erythro configuration for the 17, 18-diol in 1c, thus establishing the absolute configuration of C-18 as R. This configuration was also supported by the 13C chemical shifts of C-17 and C-18 (δC 78.3 and 75.6).8

Figure 3
Stereochemical determination of 1.
Figure 4
a) NOE correlations for 1c. b) Potential erythro and threo configurations for C17/C18 of 3. Observed NOE correlations establish erythro configuration.

The molecular formula of erythrolic acid B (2) was determined to be C29H40O7 on the basis of a HRESIMS [M − H] of 499.2692 (calcd for C29H39O7, 499.2696), two hydrogens less than 1. The 1H and 13C NMR spectra for 1 and 2 (Table 1) showed very similar signals, with the exception that the H-17 doublet at δH 3.26 for 1 was missing for 2, and the carbon signal at δC 78.3 (C-17) for 1 was replaced by a new carbonyl signal at δC 216.9 for 2. The data indicated that the hydroxyl group at C-17 in 1 has been oxidized to a ketone in 2. This assignment was supported by HMBC correlations from both H3-28 (δH 1.23) and H2-16 (δH 2.67) to the C-17 carbonyl signal at δC 216.9. The full structure of 2 was further confirmed by COSY and HMBC experiments (Table S1). Similarly, the stereochemistry of the double bonds in 2 were determined as 9E, 13E, and 23Z based on the chemical shifts of C-26 (δC 16.4), C-27 (δC 16.4), and C-29 (δC 23.6)5 and ROESY correlations. Biosynthetically, compound 2 is likely to be derived from 1 by oxidation at C-17. Hence, we assigned the stereochemistry of C-18 in 2 as R, the same as in 1.

Erythrolic acid C (3) was determined to have a molecular formula of C24H32O6 based on HRESIMS [M + H]+ of m/z 417.2285 (calcd for C24H33O6, 417.2277), indicating nine degrees of unsaturation. The 1H NMR spectrum of 3 (Table 2) showed typical signals for a prenylated benzoic acid, including a 1,3,4-trisubstitued benzene ring based on 1H NMR signals at δH 7.76 (d, J = 2.0 Hz, H-2), 7.71 (dd, J = 8.3, 2.0 Hz, H-6), and 6.78 (d, J = 8.3 Hz, H-5), as well as two triplet signals for olefinic protons at δH 5.35 and 5.21. By analysis of 1H-1H COSY, HSQC and gHMBC spectra for 3 (Table S3), a seventeen-carbon terpene fragment was established. The key COSY correlations included those for H-8 (δH 3.31)/H-9 (δH 5.35), H-11 (δH 2.08)/H-12 (δH 2.15)/H-13 (δH 5.21), H-15 (δH 2.22, 2.02)/H-16 (δH 1.59, 1.37)/H-17 (δH 3.51), and H-19 (δH 2.34, 1.87)/H-20 (δH 2.60).

Table 2
NMR data for erythrolic acids C (3), D (4) and E (5) in CD3OD

HMBC correlations from H-22 to C-9, C-10 and C-11, from H-23 to C-13, C-14 and C-15, from H-24 to C-17, C-18 and C-19 clearly indicated the presence of a terpene chain. Moreover, the COSY correlations between H-19 and H-20 as well as HMBC correlations from both H-19 and H-20 to the C-21 carbonyl carbon at δC 179.9 revealed that the side chain terminated in a carboxylic acid or ester. The benzoic acid, two double bonds and terminal carbonyl only account for eight of the nine degrees of unsaturation, indicating the presence of an additional ring. The downfield shift of the terminal carbonyl carbon at δC 179.9 and the C-18 oxygenated quaternary carbon at δC 90.5, allowed for formation of the five-membered lactone ring. As a result, the planar structure of compound 3 was established.

The stereochemistry of the double bonds in 3 were determined as 9E, 13E based on the chemical shifts of C-22 (δC 15.9) and C-23 (δC 15.9).5 The relative configuration of C-17/C-18 in 3 was established by a 2D ROESY experiment, with the H3-24 singlet having NOE correlations with H-19β and H-17, and H-19α showing NOE correlations with H-16 and H-17. However, there was no apparent NOE correlation between H-16 and H-24 (Figure 4, Table S3). If the relative configuration with threo, we should not observe an NOE between H-16 and H-19α. Thus the NOE data are consistent with C-17/C-18 having the erythro configuration. Based on the similarity of 1 and 3, we assume the absolute configuration is 17S, 18R.

The molecular formula of erythrolic acid D (4) was determined to be C19H24O5 based on the HRESIMS which gave an [M + Na]+ of m/z 355.1510 (calcd for C19H24O5Na, 355.1521), indicating eight degrees of unsaturation. The 1H NMR spectrum of 4 (Table 2) showed signals for a prenylated benzoic acid, including 1H NMR signals for a 1,3,4-trisubstitued benzene ring at δH 7.76 (d, J = 2.2 Hz, H-2), 7.71 (dd, J = 2.2, 8.4 Hz, H-6), and 6.78 (d, J = 8.4 Hz, H-5), as well as two triplet signals for olefinic protons at δH 5.35 and 5.30. Comparison of the NMR spectra for 4 with 1 and 3 clearly indicated the presence of a monoterpenoid unit with a terminal carboxylic acid. The COSY correlations for H-8 (δH 3.32)/H-9 (δH 5.35), H-11 (δH 2.06)/H-12 (δH 1.56)/H-13 (δH 2.04), and H-15 (δH 5.30)/H-16 (δH 2.98) (Table S4) and HMBC correlations from H-18 (δH 1.72) to C-9, C-10 and C-11, from H-19 (δH 1.73) to C-13, C-14 and C-15, from both H-15 (δH 5.30) and H-16 (δH 2.98) to C-17 (δC 176.7) and from H-8 (δH 3.31) to C-2, C-3 and C-4 establish an unusual 12 carbons terpene fragment located at C-3 of the benzoic acid ring. Similarly, the stereochemistry of the double bonds in 4 was determined as 9E, 14Z based on the chemical shifts of C-18 (δC 16.2) and C-19 (δC 23.7).5

The molecular formula of erythrolic acid E (5) was determined to be C22H30O6 based on HRESIMS with an [M + Na]+ of m/z 413.1936, (calcd for C22H30O6Na, 413.1940). Analysis of 1H, COSY, HSQC and HMBC data (supporting information, Table S5) allowed for the identification of a 4-hydroxybenzoic acid unit and a sesquiterpene side chain. The key COSY correlations for H-15 (δH 2.15, 1.98)/H-16 (δH 1.56, 1.49)/H-17 (δH 3.75)/H-18 (δH 2.40)/H-22 (δH 1.14) as well as key HMBC correlations from H-22 to C-17 (δC 73.1), C-18 (δC 47.0), and C-19 (δC 179.4) and from H-18 to C-17, C-19 and C-22 (δC 12.4) are highly indicative of a sesquiterpene terminating in a carboxylic acid and hydroxylated at C-17. The double bonds in 5 were assigned as 9E, 13E based on the chemical shifts of C-20 (δC 16.1) and C-21 (δC 15.9).5 Due to lack of material we were unable to establish the relative and absolute configuration of C-17/C-18. Compound 5 is the only one of the erythrolic acids isolated to have a traditional terpene backbone, albeit with a terminal methyl group oxidized to a carboxylic acid. This is similar to the terpene framework of erythrazole A (6). As described below, the presence of the terminal carboxylic acid at C-19 in 5 could provide insight into the biosynthesis of the other erythrolic acids.

Compounds 1–4 contain a C22, C17 or C12 terpene side chain, two more carbons than a regular diterpene, sesquiterpene and monoterpene, respectively, which is uncommon in natural products. Mycophenolic acid (MPA) is one of the limited examples of two-carbon homologated terpenes.8 The biosynthesis of MPA has been shown to be derived from 6-farnesyl-5,7-dihydroxy-4-methyl-phthalide by oxidative cleavage of an olefin, thereby reducing the terpene by three-carbons in a “cleavage pathway” (Figure 5).9,10 However, we believe the unusual terpene side chains in the erythrolic acids are generated by an alternative “homologation pathway”. The isolation of 5, which contains a standard length terpene chain, but with oxidation of a terminal methyl group to a carboxylic acid led us to speculate that 5 could be a precursor for a two-carbon homologation to give 3. Specifically in our proposed pathway to 1, C22 terpene 9 may arise from Claisen condensation of acetate with diterpene 8 (Figure 5). Terpene 8 would arise from oxidation of a terminal methyl group and reduction of the olefin. Following the Claisen condensation, subsequent elimination of H2O would result in α,β-unsaturated acid 10, which would isomerize to give the requisite C22 terpene.1 Although we have attempted feeding studies with sodium [1-13C] acetate (100 mg/L), sodium [1,2-13C] acetate (100 mg/L), [U-13C6] glucose (100 mg/L) and sodium [3-13C] pyruvate, our incorporation rates were extremely low and we have been unsuccessful in determining which of the two pathways is responsible for the two additional carbons.

Figure 5
Two biosynthetic pathways to C22 terpene.

The biological activity of the purified compounds was evaluated in the Locus Depression (LDR) assay, in cytotoxicity assays against three non-small cell lung cancer (NSCLC) cell lines and for direct inhibition of HDAC 2. Erythrolic acid D (4) was the only molecule that showed activity, with an IC50 value of 2.4 μM against the NSCLC cell line HCC44 and 3.4 μM against HCC366. 4 is the only erythrolic acid derivative to lack oxygenation along the terpene side-chain, much the same as erythrazole B (7), which also exhibits comparable cytotoxicity. It is possible that the C19 vinyl methyl plays a role in the biological activity.

EXPERIMENTAL SECTION

General Procedures

The optical rotations were recorded with a polarimeter equipped with a halogen lamp (589 nm). 1H and 2D NMR spectral data were recorded at 600 MHz in CD3OD or CDCl3 and chemical shifts were referenced to the corresponding solvent residual signal. 13C NMR spectra were acquired at 100 MHz. and chemical shifts were referenced to the corresponding solvent residual signal. High resolution ESI-TOF mass spectra were provided by The Scripps Research Institute, La Jolla, CA. Low-resolution LC/ESI-MS data were measured using via ESI-MS with a reversed- phase C18 column (Phenomenex Luna, 150 mm × 4.6 mm, 5 mm) at a flow rate of 0.7 mL/min. Preparative HPLC was performed on an Agilent 1200 series instrument with a DAD detector, using a Phenyl-Hexyl column (Phenomenex Luna, 250×10.0 mm, 5μ). ODS (50μ, Merck) were used for column chromatography.

Collection and phylogenetic analysis of strain SNB-035

The marine-derived bacterium, strain SNB-035, was isolated from a sediment sample collected from Trinity Bay, Galveston, TX (29° 42.419′ N, 94° 49165′ W). Bacterial spores were collected via stepwise centrifugation as follows: 2 g of sediment was dried over 24 h in an incubator at 35 °C and the resulting sediment added to 10 mL sH2O containing 0.05% tween 20. After a vigorous vortex for 10 min, the sediment was centrifuged at 2500 rpm for 5 min (4° C). The supernatant was removed and transferred into a new tube and centrifuged at 18,000 rpm for 25 min (4° C) and the resulting spore pellet collected. The re-suspended spore pellet (4 mL sH2O) was plated on a acidified Gauze media, giving rise to individual colonies of SNB-035 after two weeks. Analysis of the 16S rRNA sequence of SNB-035 revealed 98% identity to Erythrobacter citreus.

Cultivation and extraction

Bacterium SNB-035 was cultured in 30 2.8 L Fernbach flasks each containing 1 L of a seawater based medium (10 g starch, 4 g yeast extract, 2 g peptone, 1 g CaCO3, 40mg Fe2(SO4)3·4H2O, 100 mg KBr) and shaken at 200 rpm at 27 °C. After seven days of cultivation, sterilized XAD-7-HP resin (20 g/L) was added to adsorb the organic products, and the culture and resin were shaken at 200 rpm for 2 h. The resin was filtered through cheesecloth, washed with deionized water, and eluted with acetone. The acetone soluble fraction was dried in vacuo to yield 4.1 g of extract.

Isolation

The extract (4.1 g) was partitioned with n-hexane, CH2Cl2, EtOAc, and MeOH/H2O. The EtOAc soluble layer (342 mg) was fractionated by flash column chromatography on ODS (50 μm, 50 g), eluting with a step gradient of MeOH and H2O (30:70–100:0), and 25 fractions (Fr.1–Fr.25) were collected. Fractions 16–18 were combined and purified by reversed phase HPLC (Phenomenex Luna, Phenyl-Hexyl, 250 × 10.0 mm, 2.5 mL/min, 5 mm, UV = 210 nm) using a gradient solvent system from 20% to 88% CH3CN (0.1% TFA) over 30 min to afford erythrolic acid A (1, 2.5 mg, tR = 20.7 min), erythrolic acid B (2, 1.5 mg, tR = 22.3 min). Fractions 7–9 were combined and purified with the same HPLC column using a gradient from 25% to 68% CH3CN (0.1% TFA) over 29 min to afford erythrolic acid C (3, 0.7 mg, tR = 24.3 min), erythrolic acid D (4, 1.5 mg, tR = 19.5 min), and erythrolic acid E (5, 0.6 mg, tR = 20.7 min).

Erythrolic acid A

(1, 2.5 mg) White solid; [α]D + 2.0 (c 0.17, MeOH); UV (MeOH) λmax (log ε) 257 (3.6), 211 (3.9); 1H NMR (600 MHz, CD3OD) and 13C NMR (100 MHz, CD3OD) see Table 1. ESI-MS m/z 525.2 [M + Na]+, 501.2[M − H]. HRESIMS m/z 501.2849 [M − H] (calcd for C29H41O7, 501.2852).

Erythrolic acid B

(2, 1.5 mg) White solid; [α]D +4.1 (c 0.04, MeOH); UV (MeOH) λmax (log ε) 259 (3.8), 217 (4.1), 211 (4.1); 1H NMR (600 MHz, CD3OD) and 13C NMR (100 MHz, CD3OD) see Table 1. ESI-MS m/z 523.2 [M + Na]+, 499.2 [M − H]. HRESIMS m/z 499.2692 [M − H] (calcd for C29H39O7, 499.2696).

Erythrolic acid C

(3, 0.7 mg) White solid; [α]D +13.3 (c 0.04, MeOH); UV (MeOH) λmax (log ε) 257 (3.3), 210 (3.9); 1H NMR and 2D NMR data (600 MHz, CD3OD) see Table 2. ESI-MS m/z 439.2 [M + Na]+ 415.2 [M − H]. HRESIMS m/z 439.2110 [M + Na]+ (C24H32NaO6, calcd 439.20966), 417.2285 [M + H]+( calcd for C24H33O6, 417.2277).

Erythrolic acid D

(4, 1.5 mg) White solid; UV (MeOH) λmax (log ε) 256 (3.9), 214 (4.0); 1H NMR and 2D NMR data (600 MHz, CD3OD) see Table 2. ESI-MS m/z 355.1 [M + Na]+, 331.1 [M − H]. HRESIMS m/z 355.1510 [M + Na]+ (C19H24NaO5, calcd 355.1521), 333.1692 [M + H]+(calcd for C19H25O5, 333.1702).

Erythrolic acid E

(5, 0.6 mg) White solid; [α]20 D +12.5 (c 0.04, MeOH); UV (MeOH) λmax (log ε) 259 (3.6), 210 (4.1); 1H NMR and 2D NMR data (600 MHz, CD3OD) see Table 2. ESI-MS m/z 413.1.2 [M + Na]+ 389.2 [M − H]. HRESIMS m/z 413.1939 [M + Na]+ (C22H30NaO6, calcd. 413.1940), 391.2116 [M + H]+ (calcd for C22H31O6, 391.2120).

Methyl esters of compound 1

Solutions of 1 (1.5 mg) in dry methanol (0.5 mL) were treated with 30 μL trimethylsilyldiazomethane (2 M in diethyl ether) at room temperature for 35 min. The reaction mixture was analyzed by LC–MS and purified by RP-HPLC (Phenomenex Luna, Phenyl-Hexyl, 250 × 10.0 mm, 2.5 mL/min, 5 mm, UV = 210 nm) using a gradient solvent system from 50% to 99% CH3CN (0.1% FA) over 30 min to afford dimethyl-erythrolic acid A (1a, 0.8 mg, tR = 17.6 min) and trimethyl-erythrolic acid A (1b, 0.6 mg, tR = 20.1 min).

Dimethyl-erythrolic acid A (1a)

Yellow solid; 1H NMR (600 MHz, CD3OD) δ 7.74 (1H, d, J = 1.8, H-2), 7.70 (1H, dd, J = 1.8, 8.4 Hz, H-6), 6.79 (1H, d, J = 8.4 Hz, H-5), 5.34 (1H, t, J = 7.0 Hz, H-9), 5.29 (1H, t, J = 7.1 Hz, H-23), 5.21 (1H, t, J = 6.9 Hz, H-13), 3.84 (3H, s, 7-COOCH3), 3.65 (3H, s, 25-COOCH3), 3.31 (2H, m, overlap, H-8), 3.26 (1H, d, J = 10.4 Hz, H-17), 3.05 (2H, d, J = 7.1 Hz, H-24), 2.24/1.98 (2H, m, H-15), 2.15 (2H, q, J = 7.1 Hz, H-12), 2.07 (2H, d, t = 7.1 Hz, H-11), 2.04 (2H, d, t = 7.0 Hz, H-21), 1.71/1.34 (2H, m, H-16), 1.73 (3H, s, H-29), 1.70 (3H, s, H-26), 1.62 (3H, s, H-27), 1.51/1.45 (2H, m, H-19), 1.53/1.44 (2H, m, H-20), 1.07 (3H, s, H-28). ESI-MS m/z 553.2 [M + Na]+, 529.2 [M − H].

Trimethyl-erythrolic acid A (1b)

Yellow solid; 1H NMR (600 MHz, CD3OD) δ 7.87 (1H, br d, J = 8.5 Hz, H-6), 7.78 (1H, br s, H-2), 6.99 (1H, d, J = 8.5 Hz, H-5), 5.34 (1H, t, J = 7.0 Hz, H-9), 5.29 (1H, t, J = 7.1 Hz, H-23), 5.21 (1H, t, J = 6.9 Hz, H-13), 3.90 (3H, s, 4-OCH3), 3.85 (3H, s, 7-COOCH3), 3.65 (3H, s, 25-COOCH3), 3.31 (2H, m, overlap, H-8), 3.25 (1H, d, J = 10.4 Hz, H-17), 3.05 (2H, d, J = 7.1 Hz, H-24), 2.24/1.98 (2H, m, H-15), 2.15 (2H, q, J = 7.1 Hz, H-12), 2.06 (2H, d, t = 7.1 Hz, H-11), 2.03 (2H, d, t = 7.0 Hz, H-21), 1.71/1.34 (2H, m, H-16), 1.73 (3H, s, H-29), 1.70 (3H, s, H-26), 1.62 (3H, s, H-27), 1.51/1.45 (2H, m, H-19), 1.53/1.44 (2H, m, H-20), 1.07 (3H, s, H-28). ESI-MS m/z 567.2 [M + Na]+.

Preparation of Acetonide 1c

Compound 1a (0.6 mg) was treated with 2,2-dimethyoxypropane (0.6 mL) and pyridinium p-toluene sulfonate (3.0 mg), then stirred at room temperature for 12 h under a N2 atmosphere. The reaction solution was evaporated in vacuo and purified by RP-HPLC (Phenomenex Luna, Phenyl-Hexyl, 250 × 10.0 mm, 2.5 mL/min, 5 mm, UV = 210 nm) using a gradient solvent system from 60% to 99% CH3CN (0.1% FA) over 25 min to yield acetonide 1c (0.4 mg, tR = 20.0 min): 1H NMR (600 MHz, CDCl3) δ 7.80 (1H, dd, J = 2.1, 8.3 Hz, H-6), 7.79 (1H, d, J = 2.1, H-2), 6.80 (1H, d, J = 8.3 Hz, H-5), 5.32 (1H, t, J = 7.3 Hz, H-9), 5.32 (1H, t, J = 7.3 Hz, H-23), 5.11 (1H, t, J = 6.9 Hz, H-13), 3.87 (3H, s, 7-COOCH3), 3.67 (3H, s, 25-COOCH3), 3.66 (1H, dd, J = 9.1, 3.5 Hz, H-17), 3.36 (2H, d, J = 7.1 Hz, H-8), 3.02 (2H, d, J = 7.1 Hz, H-24), 2.17/1.98 (2H, m, H-15), 2.13 (2H, m, H-12), 2.09 (2H, m, H-11), 2.02 (2H, m, H-21), 1.59/1.47 (2H, m, H-16), 1.76 (3H, s, H-26), 1.73 (3H, s, H-29), 1.60/1.43 (2H, m, H-20), 1.58 (3H, s, H-27), 1.49/1.14 (2H, m, H-19), 1.40 (3H, s, 30β–CH3), 1.34 (3H, s, 30α–CH3), 1.17 (3H, s, H-28). 1H NMR (600 MHz, CD3OD) and 2D NMR data (in both CDCl3 and CD3OD ) see Table S2. ESI-MS m/z 593.3 [M + Na]+.

Preparation of MTPA Esters 1d and 1e

Compound 1b was divided into two portions, and each was dissolved in 400 μL of dry pyridine in a 2 mL vial. The samples were then treated with 3 μL of (R)-R-methoxy-a-(trifluoromethyl)-phenylacetyl chloride (MTPA-Cl) and 3 μL of (S)-MTPA-Cl at room temperature, respectively. After 12 h, the crude ester mixtures were purified by RP-HPLC with a Luna C-18 column (250 × 10 mm) using 30% MeCN-H2O to 100% MeCN as a linear gradient for 30 min (flow rate 2.5 mL/min) to give (S)-Mosher ester (1d) and (R)- Mosher ester (1e), respectively. ESI-MS (m/z) 761 [M + H]+ (see Supporting Information for NMR data).

Biological Assays

Cell lines were cultured in 10 cm dishes (Corning, Inc.) in NSCLC cell-culture medium: RPMI/L-glutamine medium (Invitrogen, Inc.), 1000 U/ml penicillin (Invitrogen, Inc.), 1 mg/ml streptomycin (Invitrogen, Inc.), and 5% fetal bovine serum (Atlanta Biologicals, Inc.). Cell lines were grown in a humidified environment in the presence of 5% CO2 at 37 °C. For cell viability assays, HCC44, H1395, H2122 and HCC1395 cells (60 mL) were plated individually at a density of 750 and 500 cells/well, respectively, in 384 well microtiter assay plates (Bio-one; Greiner, Inc.). After incubating the assay plates overnight under the growth conditions described above, purified compounds were dissolved and diluted in DMSO and subsequently added to each plate with final compound concentrations ranging from 40 mM to 1 nM and a final DMSO concentration of 0.5%. After an incubation of 96 h under growth conditions, Cell Titer GloTM reagent (Promega, Inc.) was added to each well (10 mL of a 1:2 dilution in NSCLC culture medium) and mixed. Plates were incubated for 10 min at room temperature and luminescence was determined for each well using a multi-modal plate reader. Relative luminescence units were normalized to the untreated control wells (cells plus DMSO only).

Supplementary Material

1_si_001

Acknowledgments

The authors thank Elisabeth Martinez (University of Texas Southwestern Medical Center, Department of Pharmacology) for the LDR assay, Bruce Posner and Shuguang Wei (University of Texas Southwestern Medical Center, Biochemistry) for cytotoxicity assays. We acknowledge the following grants for funding this project: NIH R01 CA149833, P01 CA095471 and the Welch Foundation I-1689. JBM is a Chilton/Bell Foundation Endowed Scholar.

Footnotes

Supporting Information

General procedures, chemical derivatization, data tables and NMR spectra. This material is available free of charge via the Internet at http://pubs.acs.org.

References

1. Dewick PM. Medicinal natural products: A biosynthetic approach. 3. John Wiley and Sons; West Sussex, England: 2009.
2. (a) Shiomi K, Nakamura H, Iinuma H, Naganawa H, Isshiki K, Takeuchi T, Umezawa H, Iitaka Y. J Antibiot. 1986;39:494–501. [PubMed](b) Ishibashi M, Funayama S, Anraku Y, Komiyama K, Omura S. J Antibiot. 1991;44:390–395. [PubMed](c) Funayama S, Ishibashi M, Komiyama K, Omura S. J Org Chem. 1990;55:1132–1133.(d) Hardt IH, Jensen PR, Fenical W. Tetrahedron Lett. 2000;41:2073–2076.(e) Cho JY, Kwon HC, Williams P, Jensen PR, Fenical W. Org Lett. 2006;8:2471–2474. [PubMed]
3. Denner EB, Vybiral D, Koblizek M, Kampfer P, Busse HJ, Velimirov B. Int J Syst Evol Microbiol. 2002;52:1655–1661. [PubMed]
4. Hu Y, MacMillan JB. Org Lett. 2011;13:6580–6583. [PMC free article] [PubMed]
5. Maxwell A, Rampersad D. J Nat Prod. 1989;52:614–618.
6. Ohtani I, Kusumi T, Kashman Y, Kakisawa H. J Am Chem Soc. 1991;113:4092–4096.
7. Nakanishi K, Schooley DA, Kpreeda M, Miura I. J Am Chem Soc. 1972;94:2865–2867.
8. (a) Deouet D, Cauret L, Brosse JC. Eur Polym J. 2003;39:671–686.(b) Kwon HC, Espindola APDM, Park JS, Prieto-Davo A, Rose M, Jensen PR, Fenical W. J Nat Prod. 2010;73:2047–2052. [PubMed]
9. (a) Clutterbuck PW, Percival W, Oxford AE, Raistrick H, Smith G. Biochem J. 1932;26:1441–1458. [PubMed](b) Canonica L, Kroszczynski W, Ranzi BM, Rindone B, Santaniello E, Scolastico C. J Chem Soc, Perkin Trans. 1972;1:2639–2643. [PubMed](c) Muth WL, Nash CH., III Antimicrob Agents Chemother. 1975;8:321–327. [PubMed]
10. (a) Geris R, Simpson TJ. Nat Prod Rep. 2009;26:1063–1094. [PubMed](b) Colombo L, Gennari C, Potenza D, Scolastico C, Aragozzini F, Gualandris R. J Chem Soc, Perkin Trans. 1982;1:365–373.