Bioorganic & Medicinal Chemistry 17 (2009) 63486353

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry

Microbial transformation of isosteviol oxime and the inhibitory effects on NF-jB and AP-1 activation in LPS-stimulated macrophages q
Shwu-Fen Chang b, Bo-Hon Chou a, Li-Ming Yang a,d,*, Feng-Lin Hsu c, Wen-Kuang Lin a, Yi Ho e, Shwu-Jiuan Lin a,*
a

Department of Medicinal Chemistry, College of Pharmacy, Taipei Medical University, Taipei 110, Taiwan Division of Cell and Molecular Biology, Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, Taipei 110, Taiwan c Graduate Institute of Pharmacognosy, College of Pharmacy, Taipei Medical University, Taipei 110, Taiwan d Division of Medicinal Chemistry, National Research Institute of Chinese Medicine, Taipei 112, Taiwan e Department of Pharmaceutics, College of Pharmacy, Taipei Medical University, Taipei 110, Taiwan
b

a r t i c l e

i n f o

a b s t r a c t
Microbial transformation of isosteviol oxime (ent-16-E-hydroxyiminobeyeran-19-oic acid) (2) with Aspergillus niger BCRC 32720 and Absidia pseudocylindrospora ATCC 24169 yielded several compounds. In addition to bioconverting the D-ring to lactone and lactam moieties, 4a-carboxy-13a-hydroxy13,16-seco-ent-19-norbeyeran-16-oic acid 13,16-lactone (7) and 4a-carboxy-13a-amino-13,16-secoent-19-norbeyeran-16-oic acid 13,16-lactam (10), one known compound, ent-1b,7a-dihydroxy-16-oxobeyeran-19-oic acid (6), and ve new compounds, ent-7a-hydroxy-16-E-hydroxyiminobeyeran-19-oic acid (3), ent-1b,7a-dihydroxy-16-E-hydroxyiminobeyeran-19-oic acid (4), ent-1b-hydroxy-16-E-hydroxyiminobeyeran-19-oic acid (5), ent-8b-cyanomethyl-13-methyl-12-podocarpen-19-oic acid (8), and ent-8b-cyanomethyl-13-methyl-13-podocarpen-19-oic acid (9), were isolated from the microbial transformation of 2. Elucidation of the structures of these isolated compounds was primarily based on 1D and 2D NMR, and HRESIMS data, and 35 were further conrmed by X-ray crystallographic analyses. Additionally, the inhibitory effects of all of these compounds were evaluated on NF-jB and AP-1 activation in LPS-stimulated RAW 264.7 macrophages. Among the compounds tested, 5 and 10 signicantly inhibited NF-jB activation, with 5 showing equal potency to dexamethasone; 3 and 69 signicantly inhibited AP-1 activation, particularly 8, which showed more inhibitory activity than dexamethasone. 2009 Elsevier Ltd. All rights reserved.

Article history: Received 3 June 2009 Revised 14 July 2009 Accepted 15 July 2009 Available online 18 July 2009 Keywords: Microbial transformation Isosteviol oxime Diterpenoid AP-1 NF-jB

1. Introduction Glucocorticoids (GCs) such as dexamethasone have long been used as effective immunosuppressive and anti-inammatory agents.1 The immunosuppressive and anti-inammatory actions of glucocorticoid hormones are mediated by their trans-repression of nuclear factor-kappa B (NF-jB) and activator protein (AP)-1 transcription factors.24 NF-jB is a pivotal mediator of the human immune response regulating the transcriptions of various proinammatory and inammatory mediators.5 Activation of NF-jB was reported to induce the transcriptions of multiple proinammatory mediators, such as inducible nitric oxide synthase (iNOS), tumor necrosis factor (TNF)-a, and interleukin (IL)-1a, which are involved in the pathogenesis of inammatory diseases.6 AP-1 is a
q A portion of this work was presented in the form of an abstract at the 50th ASP Annual Meeting, Honolulu, HI, USA, June 27July 1, 2009. * Corresponding authors. Tel.: +886 2 27361661x6133; fax: +886 2 28264276 (S.-J.L.); tel.: +886 2 28201999x8551 (L.-M.Y.). E-mail addresses: [email protected] (L.-M. Yang), [email protected] (S.-J. Lin).

key mediator of cytokine signaling and is required for the activation of numerous proinammatory genes,7 a group of dimeric factors constituted by members of the Jun and Fos families of DNA-binding proteins.8 Because GC therapy is still accompanied by a wide range of adverse effects,9 the inhibition of NF-jB and/ or AP-1 transcriptional activation may become an alternative attractive approach for developing novel immunoinammatory agents.10 A microbe is a biocatalyst, which possesses a multi-enzyme system. Thus, microbial transformation is an important technique for structurally modifying natural and synthetic compounds due to its signicant regio- and stereoselectivities.11 Fungi are widely used in microbial transformation studies, since their versatile enzymatic reservoir allows them to modify a diverse array of molecules.12 Isosteviol (1), with a rigid skeleton comprising four fused rings similar to the steroid skeleton, possesses many biological activities.13 In general, natural products are very useful as template molecules for producing new drugs.14 Small modications in the chemical structure of a compound can modify its biological activities.15 According to the literature, a number of biologically important

0968-0896/$ - see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.bmc.2009.07.029

S.-F. Chang et al. / Bioorg. Med. Chem. 17 (2009) 63486353

6349

properties of steroids are dependent upon structural features of the steroid D-ring.16 Transformation of cyclic ketones into oximes by aqueous alkaline media, frequently used for preparing oximes of natural compounds, provides a way of changing functional groups, sizes, and the stereochemistry of the D-ring.16 Thus, isosteviol oxime (ent-16-E-hydroxyiminobeyeran-19-oic acid) (2) was prepared by reacting 1 with hydroxylamine hydrochloride. The activity of 2 in preventing cholera toxin-induced intestinal uid secretions was reported.17 However, the microbial transformation of 2 has never been described in the literature. In order to obtain new functionalized analogues with new biological activities, microbial transformation of 2 was undertaken. A number of microorganisms were screened for their ability to biotransform 2. Aspergillus niger BCRC 32720 and Absidia pseudocylindrospora ATCC 24169 were selected for preparative-scale fermentation because they reproducibly converted 2 into many metabolites. Because the skeleton of tetracyclic diterpenoids possesses similarities to steroids18 and in a continuing search for potential anti-inammatory agents,13 the inhibitory effects of 110 on the expressions of NF-jB and AP-1 target genes in lipopolysaccharide (LPS)-stimulated RAW 264.7 macrophages were also evaluated. This article deals with the production, isolation, structure elucidation, and biological evaluation of these compounds. 2. Results and discussion 2.1. Bioconversion of isosteviol oxime (2) by Aspergillus niger BCRC 32720 and Absidia pseudocylindrospora ATCC 24169 Isosteviol oxime (2) was prepared as previously reported19 and identied by 1D and 2D NMR spectroscopy, and HRESIMS. The bioconversion of 2 by Asp. niger and Abs. pseudocylindrospora reproducibly produced diverse products. Thus, they were selected for the preparative-scale biotransformation of 2. Incubation of 2 with Asp. niger for 6 days led to the formation of ent-7a-hydroxy-16-Ehydroxyiminobeyeran-19-oic acid (3), ent-1b,7a-dihydroxy-16-Ehydroxyiminobeyeran-19-oic acid (4), ent-1b-hydroxy-16-E-

hydroxyiminobeyeran-19-oic acid (5), ent-1b,7a-dihydroxy-16oxobeyeran-19-oic acid (6), 4a-carboxy-13a-hydroxy-13,16-secoent-19-norbeyeran-16-oic acid 13,16-lactone (7), ent-8b-cyanomethyl-13-methyl-12-podocarpen-19-oic acid (8), ent-8b-cyanomethyl-13-methyl-13-podocarpen-19-oic acid (9), and 4a-carboxy13a-amino-13,16-seco-ent-19-norbeyeran-16-oic acid 13,16-lactam (10) (Fig. 1). Incubation of 2 with Abs. pseudocylindrospora for 6 days yielded 3, 4, and 6. Among these, 35, 8, and 9 are new compounds, whereas 6, 7, and 10 have previously been reported.13,20,21 The structures of these isolated compounds were elucidated through analysis of 1D and 2D NMR spectroscopic data, with 35 being further conrmed by X-ray crystallographic studies. Compound 3 displayed [M+H]+ at m/z 350.2309 on HRESIMS, corresponding to the molecular formula C20H31NO4, consistent with its 13C NMR and DEPT spectra. The 1H and 13C NMR spectra indicated 20 carbons attributable to six quaternary carbons (including one carbonyl carbon), three CH3, eight CH2, and three CH (including one oxymethine). These data suggested that 3 contains one more oxygen atom than 2. Analyses of the HSQC and HMBC spectra of 3 revealed that the new proton resonance at d 4.01 showed connectivities with C-5 (d 48.1), C-9 (d 50.1), and C15 (d 37.4). Thus, the additional OH group in 3 occurred at C-7. The orientation of the OH group at C-7 followed from the multiplicity of the H-7 signal in the 1H NMR spectrum, a broad singlet signal, indicating that H-7 was a-oriented.22 Furthermore, the structure of 3 was conrmed by an X-ray crystallographic experiment (Fig. 2). Thus, 3 was characterized as ent-7a-hydroxy-16-Ehydroxyiminobeyeran-19-oic acid. Compound 4 had the molecular formula of C20H31NO5, as evidenced by the HRESIMS [m/z 366.2313 (M+H)+] and NMR spectra. The 13C NMR and DEPT spectra of 4 possessed similar features as 2 except for the disappearance of two CH2 signals and the presence of two new CH resonances at d 82.1 and 76.6. This suggests that 4 is a dihydroxylated metabolite of 2. In the HMBC spectrum, the proton resonance at d 3.92 showed connectivities with CH3-20 (d 10.4), C-2 (d 31.4), C-10 (d 45.3), and C-9 (d 51.1). Thus, one of

O O COOH COOH

NH O

10

C D A B
COOH

O NH2OH HCl
1 2 3 4 18

11 20 10 5 19 9

12 13 14 15

17 16

R1 N OH
COOH

N OH R2

8 6 7

COOH

3: R1 = H, R2 = OH 4: R1 = R2 = OH 5: R1 = OH, R2 = H

11 20 2 34 18 1 10 5 19 9

12 13 14

17

OH CN
COOH COOH

8 6 7

15 16

CN

COOH

OH

9
Figure 1. Structures of compounds 110.

6350

S.-F. Chang et al. / Bioorg. Med. Chem. 17 (2009) 63486353

Figure 2. Perspective drawing of the X-ray structure of 3.

Figure 4. Perspective drawing of the X-ray structure of 5.

the OH groups is at C-1. An a-orientation of the OH group was established by NOE correlations of dH 3.92 with H-2b and H-5b. The location of the second OH group at C-7 was deduced by 2D NMR data. In the HMBC spectrum, the proton resonance at dH 4.02 showed connectivities with C-15 (d 37.9), C-8 (d 47.2), C-5 (d 46.8), and C-9 (d 51.1). This indicated that the OH group is at C-7. H-7 resonated as a broad singlet at d 4.02 indicating that the proton was a-oriented.22 An X-ray crystallographic analysis was further performed to conrm the structure of 4 (Fig. 3). Thus, 4 was characterized as ent-1b,7a-dihydroxy-16-E-hydroxyiminobeyeran-19-oic acid. The HRESIMS of 5 displayed a quasi-molecular ion [M+H]+ at m/z 350.2366, which indicated a molecular formula of C20H31NO4 in combination with the 1H and 13C NMR data. The 13C NMR and DEPT spectra revealed 20 signals due to six quaternary, three CH3, eight CH2, and three CH carbons. The 1H and 13C NMR spectra, when compared to 2, showed new peaks at dH 3.72 and dC 81.9, respectively, indicating hydroxylation at a CH2 carbon. The HMBC spectrum showed correlations of dC 81.9 with CH3-20 (d 1.47), H-2 (d 2.60), H-3 (d 2.50), and H-5 (d 1.19). This indicated that an OH group had been introduced at C-1. In the COSY spectrum, the methine proton at C-1 (d 3.72) resonated as a doublet of doublets (J = 11.0, 4.0 Hz) due to coupling with the protons of the neighboring C-2. This indicated that the proton is in a b-orientation.22 Additionally, a single-crystal X-ray crystallographic experiment was used to conrm the structure of 5 (Fig. 4). Thus, 5 was established as ent-1b-hydroxy-16-E-hydroxyiminobeyeran-19-oic acid. Compounds 8 and 9 displayed [MH] at m/z 314.2112 and 314.2118 (calcd for C20H28NO2 314.2120), respectively, in the negative-mode HRESIMS, corresponding to the molecular formula C20H29NO2, consistent with the 13C NMR and DEPT spectra. The 13 C NMR and DEPT data of 8 and 9 displayed 20 carbons attributable to six quaternary carbons (including one cyano and one carbonyl carbon), three CH3, eight CH2, and three CH carbons. Comparison of the HSQC and HMBC data of 8 and 9 with that of 2 revealed that no additional OH group had been introduced. In the HSQC and HMBC spectra of 8 and 9, the proton resonance at

d 5.35 (dC 121.3) showed connectivities with d 24.0 (C-17), 52.4 (C-9), and 46.6 (C-14) in 8; the proton resonance at d 5.51 (dC 131.7) showed connectivities with d 23.7 (C-17), 32.3 (C-12), 37.5 (C-8), and 39.0 (C-7) in 9. These suggested the presence of double bonds at C-12 in 8, and C-13 in 9. No connectivity was found between dC 120.1 and d 5.35 (H-12), 1.81 (H-14), or 1.65 (H-17) in 8 or between dC 120.2 and d 1.92 (H-12), 5.51 (H-14), or 1.62 (H-17) in 9, suggesting that the bond between C-16 and C-13 had been cleaved. Accordingly, an abnormal Beckmann rearrangement of 2 to yield ring-cleaved carbonitrile products of 8 and 9 had occurred.16 Thus, 8 and 9 were characterized as ent-8bcyanomethyl-13-methyl-12-podocarpen-19-oic acid and ent-8bcyanomethyl-13-methyl-13-podocarpen-19-oic acid, respectively. 2.2. Biological evaluation The inhibitory effects of compounds 110 on both NF-jB and AP-1 activations were evaluated in LPS-stimulated RAW 264.7 macrophages using an NF-jB- or AP-1-mediated luciferase reporter gene assay. Among the compounds tested, 5 and 10 signicantly inhibited NF-jB activation, with 5 showing equal potency to dexamethasone. On the other hand, 3 and 69 showed signicant inhibition of AP-1 activation, particularly 8, which showed greater inhibitory activity than dexamethasone (Tables 3 and 4). 3. Conclusion In summary, the two selected lamentous fungi possess the ability to perform hydroxylation (36), Beckmann rearrangement (10), and an abnormal Beckmann rearrangement (8 and 9). According to the literature,21 the formation of lactone 7 might be from the introduction an OH group into nitrile carbocation and subsequently cyclization to yield imidate, which was hydrolyzed to 7. On the other hand, the results also suggested that substrate 2 might be hydrolyzed to isosteviol (1) rst, and then converted to the 6 and 7 by hydroxylation and BaeyerVilliger reaction, respectively. Although substrate 2 can be transformed into lactone and lactam derivatives under concentrated hydrochloric acid in an ampoule at 180 C,21 this is the rst report to yield these two products by microbes. Asp. niger also has the abilities to yield ring-cleaved carbonitrile products and also to cause stereoselective hydroxylation at the 1a- and/or 7b-positions. Results obtained from the inhibitory effects on NF-jB or AP-1 activation in LPS-stimulated RAW 264.7 macrophages showed that 5 possessing 1a-OH on substrate 2 exhibited an equal potency of inhibition toward NF-jB activation as dexamethasone; 8, with a ring-cleaved carbonitrile, exhibited more-signicant inhibition toward AP-1 activation than dexamethasone. Thus, the suppression of NF-jB or AP-1 activation in LPS-stimulated RAW 264.7 macrophages by compounds 5 and 8 may have value for protection against inammation. This study also demonstrates the use of microbial transformation techniques

Figure 3. Perspective drawing of the X-ray structure of 4.

S.-F. Chang et al. / Bioorg. Med. Chem. 17 (2009) 63486353 Table 1 1 H NMR chemical shifts of compounds 25, 8, and 9 (C5D5N, d values in ppm)a,b Position 1 2 3 5b 6 7 9b 11 12 14 15 17 18-CH3 20-CH3 N-OH
a b c

6351

2 1.531.73 m 0.86 td (12.0, 4.0) 2.112.23 mc 1.371.50 mc 2.43 br d (12.8) 1.011.07 mc 1.12 m 2.112.23 mc 2.02 m 1.531.73 mc 1.371.50 mc 1.011.07 mc 1.531.73 mc 1.371.50 mc 1.531.73 mc 1.371.50 mc 1.371.50 mc 1.22 dd (10.8, 2.8) 3.32 dd (18.4, 2.8) 2.19 br d (18.4) 1.28 s 1.35 s 1.09 s 12.22 s
c

3 1.741.80 m 1.07 td (13.2, 3.9) 2.26 m 1.481.56 mc 2.502.61 mc 1.18 dd (13.3, 4.0) 2.38 d (1.9) 2.502.61 mc
c

4 b3.92 dd (11.3, 4.5) 2.652.75 m 1.952.02 mc 2.542.57 mc 1.42 dd (13.6, 4.0) 2.542.57 mc 2.652.75 mc 2.542.57 mc a4.02 br s 2.35 dd (12.3, 4.3) 3.28 m 1.952.02 mc 1.801.86 mc 1.66 td (12.5, 5.0) 2.09 dd (11.3, 2.6) 1.801.86 mc 3.55 dd (18.3, 2.6) 2.45 d (18.3) 1.37 s 1.48 s 1.56 s 12.27 s
c

5 b3.72 dd (11.0, 4.0) 2.60 m 2.00 m 2.50 d (13.2) 1.28 dd (13.2, 3.5) 1.19 br d (11.6) 2.32 m 2.10 d (12.6) 1.64 d (13.0) 1.481.62 mc 1.481.62 mc 3.23 m 1.801.88 mc 1.801.88 mc 1.481.62 mc 1.481.62 mc 1.36 dd (11.0, 2.5) 3.52 dd (18.0, 2.5) 2.31 d (18.0) 1.32 s 1.39 s 1.47 s 12.36 s

8 1.61 m 0.85 m 2.142.25 mc 1.48 br d (14.0) 2.412.45 mc 1.041.21 mc 1.041.21 mc 2.142.25 mc 1.952.04 mc 2.142.25 mc 1.041.21 mc 1.26 dd (12.5, 5.1) 1.952.04 mc 1.67 m 5.35 s 2.142.25 mc 1.81 br d (17.2) 2.73 dd (16.7, 1.25) 2.412.45 mc 1.65 s 1.35 s 0.96 s

9 1.70 d (13.0) 0.82 td (13.0, 4.0) 2.222.33 mc 1.53 m 2.44 d (13.0) 1.051.13 mc 1.051.13 mc 2.222.33 mc 2.02 m 2.13 dt (16.2, 2.85) 1.231.32 mc 1.051.13 mc 1.64 m 1.231.32 mc 1.92 m 5.51 s 2.90 2.53 1.62 1.34 0.94 d (16.2) d (16.2) s s s

a4.01 br s
1.90 m 1.68 m 1.481.56 mc 1.741.80 mc 1.481.56 mc 1.99 dd (11.3, 2.8) 1.741.80 mc 3.40 dd (18.5, 2.8) 2.36 d (18.5) 1.37 s 1.46 s 1.23 s 12.32 s

Assignments based on DEPT, HSQC, and HMBC. Signal multiplicity and coupling constants (Hz) are in parentheses. Overlapping signals.

Table 2 13 C NMR chemical shifts of compounds 25, 8, and 9 (C5D5N, d values in ppm)a Carbon no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
a

Table 3 Data of compounds 110 on an NF-jB-mediated luciferase reporter gene assaya Compound 1 2 3 4 5 6 7 8 9 10 Control Dex Luciferase activity 0.32 0.12 0.15 0.07 0.16 0.05 0.26 0.24 0.13 0.03* 0.16 0.06 0.20 0.04 0.18 0.05 0.22 0.06 0.15 0.03* 0.21 0.04 0.13 0.04*

2 40.2 19.6 38.6 43.8 57.1 22.4 41.4 40.8 55.2 38.5 20.8 40.0 43.5 56.8 37.6 167.6 23.0 29.4 180.1 13.7

3 40.8 20.3 39.2 44.0 48.1 31.2 76.2 46.3 50.1 39.0 21.1 40.5 44.1 54.0 37.4 168.1 23.8 29.9 181.1 14.2

4 82.1 31.4 37.5 43.8 46.8 31.1 76.6 47.2 51.1 45.3 24.7 41.3 44.3 54.4 37.9 168.4 23.6 29.8 181.0 10.4

5 81.9 31.3 37.3 44.1 56.8 23.1 42.7 42.2 56.6 45.3 24.9 41.2 44.2 57.7 38.8 168.6 23.4 29.8 180.5 10.4

8 40.6 20.2 38.9 44.3 57.6 21.0 39.9 35.9 52.4 38.2 23.0 121.3 131.4 46.6 20.8 120.1 24.0 29.8 180.4 14.2

9 40.4 20.1 39.1 44.4 57.6 21.0 39.0 37.5 55.2 38.2 18.3 32.3 134.8 131.7 25.1 120.2 23.7 29.6 180.3 14.7

a The concentration of each test compound was 10 lM. All rey luciferase activities were normalized to Renilla luciferase activity. The data were expressed as multiples of luciferase activity compared to the no-treatment (control) group. Dexamethasone (Dex) is the reference compound. Each value is the average of the rey/Renilla luciferase ratio and presented as the mean SEM (n = 3). * Signicantly different equals p <0.05, using Students t-test for paired samples.

Assignments are based on DEPT, HSQC, and HMBC.

as a useful means for chemists to prepare diverse new derivatives of diterpenoids. 4. Experimental 4.1. General Melting points were determined using a Yanagimoto micromelting point apparatus and are uncorrected. Optical rotations were determined on a JASCO DIP-1020 digital polarimeter. 1H, 13 C NMR, DEPT, and 2D NMR spectra were recorded on a Bruker AVANCE DRX 500 spectrometer. Chemical shifts are reported in parts per million (ppm) with respect to the corresponding solvent as the internal standard, and coupling constants (J) are in hertz

(Hz). Low- and high-resolution ESI mass spectra were recorded using a VG Platform Electrospray ESI/MS spectrometer. Column chromatography (CC) was performed with Kieselgel silica (70 230 and 230400 mesh, Merck, Darmstadt, Germany). HPLC was performed on a Hitachi L-2130 (Tokyo, Japan) using a Betasil Silica-100 column (250 10 mm, 5 lm, at a ow rate of 2 mL/min) (Thermo Scientic, Waltham, MA, USA) equipped with a refractometer detector (Hitachi L-2490). X-ray single-crystal diffraction was measured on a Siemens SMART CCD XRD. Fractions were monitored by TLC (Merck 1.05554), and spots were visualized by heating Si gel plates sprayed with 10% H2SO4. 4.2. Substrate Isosteviol oxime (2) was prepared as previously reported.19 Signals for the proton and carbon of 2 were initially assigned by 1D and 2D NMR, and HRESIMS data.

6352

S.-F. Chang et al. / Bioorg. Med. Chem. 17 (2009) 63486353

Table 4 Data of compounds 110 on an AP-1-mediated luciferase reporter gene assaya Compound 1 2 3 4 5 6 7 8 9 10 Control Dex Luciferase activity 3.41 0.60 4.87 2.15 2.69 0.45* 3.37 0.70 3.52 0.85 2.98 0.35* 2.38 0.18* 2.22 0.35* 2.85 0.48* 3.19 0.72 3.94 0.53 2.33 0.36*

increasing polarity) to give 452 mg of a white solid. The white solid was further puried by HPLC on a semi-preparative column to give 6 (210 mg) and 3 (180 mg). Fraction 4 (35 mg) was subjected to repeated semi-preparative HPLC separation (CH2Cl2isopropanol, 9:1) to give 4 (17 mg). Fraction 2 (810 mg) was applied to a silica gel column (230400 mesh, 2 55 cm) eluted with n-hexane EtOAc (2:1) to give two fractions (2-1 and 2-2). Fraction 2-2 (760 mg) was recrystallized with EtOAc and MeOH, and 710 mg of 2 was obtained as white crystals. 4.4.1. Compound 3 1 White crystals, mp 214216 C; a25 D 48.7 (c 0.5, MeOH); for H and 13C NMR spectra, see Tables 1 and 2; HRESIMS m/z calcd for C20H32NO4 [M+H]+ 350.2331, found 350.2309. 4.4.1.1. X-ray crystallographic data of 3. C20H31NO4, M 349.46, monoclinic, P21, a 8.64800 (10) , b 8.89210 (10) , c 11.69010 (10) , V 898.030 (16) 3; Z 2, Dcalcd 1.292 g cm3, F(0 0 0) 380, k (Mo Ka) 0.71073 , T 295(2) K, 7082 reection collected. Final GooF 1.093, nal R indices R1 0.0312, wR2 0.0864, 238 parameters, l 0.089 mm1, R indices based on 3957 reections with I > 2r(I) absorption corrections applied. Complete crystallographic data of 3 have been deposited in the Cambridge Crystallographic Data Centre as supplementary publication number CCDC 722590.24 4.4.2. Compound 4 1 White crystals, mp 217219 C; a25 D +30.8 (c 0.5, MeOH); for H and 13C NMR spectra, see Tables 1 and 2; HRESIMS m/z C20H32NO5 [M+H]+ calcd for 366.2280, found 366.2313. 4.4.2.1. X-ray crystallographic data of 4. C20H31NO5, M 365.46, monoclinic, P21, a 7.053 , b 18.44180 (10) , c 7.649 , V 956.783 (5) 3; Z 2, Dcalcd 1.269 g cm3, F(0 0 0) 396, k (Mo Ka) 1.54178 , T 295(2) K, 31,108 reection collected. Final GooF 1.061, nal R indices R1 0.0299, wR2 0.0832, 252 parameters, l 0.734 mm1, R indices based on 3336 reections with I > 2r(I) absorption corrections applied. Complete crystallographic data of 4 have been deposited in the Cambridge Crystallographic Data Centre as supplementary publication number CCDC 722592.24 4.4.3. Compound 5 1 White crystals, mp 161163 C; a25 D 48.2 (c 0.5, MeOH); for H and 13C NMR spectra, see Tables 1 and 2; HRESIMS m/z C20H32NO4 [M+H]+ calcd for 350.2331, found 350.2366. 4.4.3.1. X-ray crystallographic data of 5. C20H32NO4.50, M 358.47, orthorhombic, P212121, a 11.86090 (10) , b 14.08840 (10) , c 23.35940 (10) , V 3903.38 (5) 3; Z 8, Dcalcd 1.220 g cm3, F(0 0 0) 1560, k (Mo Ka) 1.54178 , T 295(2) K, 68,493 reection collected. Final GooF 1.029, nal R indices R1 0.0346, wR2 0.0975, 493 parameters, l 0.689 mm1, R indices based on 7109 reections with I > 2r(I) absorption corrections applied. Complete crystallographic data of 5 have been deposited in the Cambridge Crystallographic Data Centre as supplementary publication number CCDC 722591.24 4.4.4. Compound 8 1 13 C NMR White powder; a25 D 76.5 (c 0.8, MeOH); for H and spectra, see Tables 1 and 2; HRESIMS m/z C20H28NO2 [MH] calcd for 314.2120, found 314.2112. 4.4.5. Compound 9 1 13 C NMR White powder; a25 D +40.0 (c 0.8, MeOH); for H and spectra, see Tables 1 and 2; HRESIMS m/z C20H28NO2 [MH] calcd for 314.2120, found 314.2118.

a The concentration of each test compound was 10 lM. All rey luciferase activities were normalized to Renilla luciferase activity. The data were expressed as multiples of luciferase activity compared to the no-treatment (control) group. Dexamethasone (Dex) is the reference compound. Each value is the average of the rey/Renilla luciferase ratio and presented as the mean SEM (n = 3). * Signicantly different equals p <0.05, using Students t-test for paired samples.

4.3. Microorganisms, incubation, and screening procedures Twenty-one microorganisms including 10 genera (number of species): Aspergillus (four), Absidia (one), Bacillus (one), Beauveria (two), Cunninghamella (four), Mortierella (one), Mucor (two), Nocardia (two), Pseudomonas (two), and Streptomyces (two), obtained from the Division of Medicinal and Natural Products Chemistry, College of Pharmacy, University of Iowa, Iowa City, IA, USA and Bioresources Collection and Research Center, Hsinchu, Taiwan, were used for the preliminary screening of 2. The fermentation protocol for screening was identical to that described previously.23 Culture controls and a substrate control were both run. Eight compounds were reproducibly produced by Asp. niger and Abs. pseudocylindrospora. 4.4. Microbial transformation of 2 by Asp. niger and Abs. pseudocylindrospora Using 24-h-old stage II cultures, a solution of 2 (1.00 g dissolved in 10 mL DMF) was evenly distributed among 100 asks containing stage II cultures. Substrate-containing cultures were incubated for 144 h. Extraction as previously described23 produced 4.4 g of black oil and 3.6 g of brown oil after respective bioconversions by Asp. niger and Abs. pseudocylindrospora. The crude residue from Asp. niger (4.4 g) was subjected to silica gel CC (70230 mesh, 5 90 cm). In total, six fractions were eluted with mixtures of CH2Cl2MeOH (800 mL each of 20:1, 15:1, and 10:1). The elutes were monitored using TLC. Three fractions (13) were obtained on the basis of similar TLC proles. Further chromatography of the fraction 3 (1.02 g) over silica gel (230400 mesh, 3 50 cm) eluted with CH2Cl2isopropanol (160 mL each of 12:1, 10:1, and 8:1) yielded four fractions (3-13-4). Compound 2 of 760 mg was recovered from fraction 3-2 (780 mg). After recrystallization of fractions 3-3 (106 mg) and 3-4 (102 mg) with CH3OH, 98 mg of 3 and 96 mg of 6 were obtained as white crystals and a white powder, respectively. Fraction 4 (85 mg) was chromatographed over silica gel (n-hexaneEtOAc with increasing polarity) to give 71 mg of a white solid. The white solid was further puried by HPLC on a semi-preparative column to obtain 7 (20 mg), 8 (32 mg), 9 (10 mg), and 10 (7 mg). Fraction 5 (108 mg) was subjected to repeated semi-preparative HPLC separation (CH2Cl2 isopropanol, 9:1) to give 4 (80 mg) and 5 (16 mg). The crude residue from Abs. pseudocylindrospora (3.6 g) was puried by CC over silica gel using mixtures of n-hexane, EtOAc, CH2Cl2, and MeOH with increasing polarity to obtain four fractions (14). Fraction 3 (610 mg) was chromatographed over silica gel (CH2Cl2MeOH with

S.-F. Chang et al. / Bioorg. Med. Chem. 17 (2009) 63486353

6353

4.5. Transfection procedures and reporter gene assays Twenty-four hours before transfection, about 1 105 mouse RAW 264.7 macrophage cells per well were seeded in 96-well white plates. The reporter plasmid, pAP-1-Luc or pNF-jB-Luc plasmid together with an internal control plasmid, pGL-hRluc, were transfected into RAW 264.7 cells using lipofectamine plus (Invitrogen, San Diego, CA, USA) according to the manufacturers instructions. At 48 h post-transfection, lipopolysaccharide (LPS) from Escherichia coli (serotype 0111:B4) (Sigma, St. Louis, MO, USA) at a nal concentration of 100 ng/mL was added to the transfected cells for 6 h. After LPS stimulation, a nal concentration of 10 lM of each test compound including the reference compound, dexamethasone (Sigma), in DMSO was added to the cells. Cells were harvested 24 h after treatment, and the reporter activities of rey luciferase expressed from pAP-1-Luc or pNF-jB-Luc, and Renilla luciferase from pGL-hRluc were assayed in a Veritas microplate luminometer (Turner Biosystems, Sunnyvale, CA, USA) using a dual-luciferase reporter assay system (Promega, Madison, WI, USA). 4.5.1. Statistical analysis Data are from at least three individual experiments. Average rey/Renilla luciferase ratios were analyzed by two-tailed Students t-test for paired samples. Signicance was accepted at p <0.05.

Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bmc.2009.07.029. References and notes
1. Barnes, P. J. Eur. Respir. J. 2006, 27, 413. 2. Adcock, I. M. Pulm. Pharmacol. Ther. 2001, 14, 211. 3. Jeon, Y. J.; Han, S. H.; Lee, Y. W.; Lee, M.; Yang, K. H.; Kim, H. M. Immunopharmacology 2000, 48, 173. 4. Ma, W.; Gee, K.; Lim, W.; Chambers, K.; Angel, J. B.; Kozlowski, M.; Kumar, A. J. Immunol. 2004, 172, 318. 5. Makarov, S. S. Arthritis Res. 2001, 3, 200. 6. Christman, J. W.; Sadikot, R. T.; Blackwell, T. S. Chest 2000, 117, 1482. 7. Karin, M.; Liu, Z. G.; Zandi, E. Curr. Opin. Cell Biol. 1997, 9, 240. 8. Jochum, W.; Passegu, E.; Wagner, E. F. Oncogene 2001, 20, 2401. 9. Wang, J. C.; Shah, N.; Pantoja, C.; Meijsing, S. H.; Ho, J. D.; Scanlan, T. S.; Yamamoto, K. R. Genes Dev. 2006, 20, 689. 10. Palanki, M. S. S.; Erdman, P. E.; Gayo-Fung, L. M.; Shevlin, G. I.; Sullivan, R. W.; Suto, M. J.; Goldman, M. E.; Ransone, L. J.; Bennett, B. L.; Manning, A. M. J. Med. Chem. 2000, 43, 3995. 11. Loughlin, W. A. Bioresour. Technol. 2000, 74, 49. 12. Lehmann, L. R.; Stewart, J. D. Curr. Org. Chem. 2001, 5, 439. 13. Chang, S. F.; Yang, L. M.; Lo, J. H.; Liaw, J. H.; Wang, L. H.; Lin, S. J. J. Nat. Prod. 2008, 71, 87. and references cited therein. 14. Venisetty, R. K.; Ciddi, V. Curr. Pharm. Biotechnol. 2003, 4, 153. 15. Braguini, W. L.; Biazon, M. A.; de Oliveira, B. H.; Carnieri, E. G. S.; Rocha, M. E. M. R.; de Oliveira, M. B. M. Toxicol. Lett. 2003, 143, 83. 16. Wang, C.; Jiang, X.; Shi, H.; Lu, J.; Hu, Y.; Hu, H. J. Org. Chem. 2003, 68, 4579. 17. Pariwat, P.; Homvisasevongsa, S.; Muanprasat, C.; Chatsudthipong, V. J. Pharmacol. Exp. Ther. 2008, 324, 798. 18. Hanson, J. R. Nat. Prod. Rep. 1992, 9, 139. 19. Alfonsov, V. A.; Andreeva, O. V.; Bakaleinik, G. A.; Beskrovnyi, D. V.; Gubaidullin, A. T.; Kataev, V. E.; Kovylyaeva, G. I.; Konovalov, A. I.; Korochkina, M. G.; Litvinov, I. A.; Militsina, O. I.; Strobykina, I. Y. Russ. J. Gen. Chem. 2003, 73, 1255. 20. Chou, B. H.; Yang, L. M.; Chang, S. F.; Hsu, F. L.; Lo, J. H.; Liaw, J. H.; Liu, P. C.; Lin, S. J. J. Nat. Prod. 2008, 71, 602. 21. Militsina, O. I.; Kovyljaeva, G. I.; Bakaleynik, G. A.; Strobykina, I. Y.; Kataev, V. E.; Alfonsov, V. A.; Musin, R. Z.; Beskrovny, D. V.; Litvinov, I. A. Mendeleev Commun. 2005, 1, 27. 22. de Oliveira, B. H.; dos Santos, M. C.; Leal, P. C. Phytochemistry 1999, 51, 737. 23. Hsu, F. L.; Hou, C. J.; Yang, L. M.; Cheng, J. T.; Liu, P. C.; Lin, S. J. J. Nat. Prod. 2002, 65, 273. 24. CCDC 722590, 722591, and 722592 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44 1223 336 033; email: [email protected]).

Acknowledgments We thank Ms. Shou-Ling Huang for the NMR data acquisition and Mr. Yi-Hung Liu for conducting X-ray crystallography at the Instrumentation Center of National Taiwan University, Taipei, Taiwan. We also thank Dr. John P. N. Rosazza, Division of Medicinal and Natural Products Chemistry, College of Pharmacy, University of Iowa, Iowa City, IA, for kindly providing the bacterial strains. Financial support from the National Science Council of Taiwan (NSC97-2320-B-038-009 and NSC98-2320-B-038-011-MY3) is gratefully acknowledged.