Method for producing antibiotic from pleosporaceae ulocladium

Abstract

Fermentation processes that use fungi of the genus Ulocladium for producing novel compounds that include an 8- or 9-membered cyclic structure, wherein an (E)-1,4-dioxo-but-2-ene moiety is embedded in the cyclic structure. The compounds may be used as medicaments, in particular as medicaments for treating and preventing microbial infections, such as infections by multidrug resistant and/or Gram-positive bacteria.

Claims

1. A method for producing a compound comprising an (E)-7-hydroxy-4-oxo-hept-2-enoic acid lactone ring, the method comprising: (a) culturing a fungus of the fungal Family of Pleosporaceae of the genus Ulocladium, in a medium conducive to the production of the compound; and, (b) optionally recovery of the compound.

2. The method according to claim 1, wherein the culturing is at a temperature less than 30 C.

3. The method according to claim 1, wherein the compound is recovered by solvent extraction.

4. The method according to claim 3, wherein the solvent is ethyl acetate.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1: MS/MS spectrum of the compound of formula (I), showing fragments of 155.1 (M+H), 309.2 (2M+H) and 617.4 (4M+H).

(2) FIG. 2: Growth kinetics of E. faecium (15A623) in the presence of the compound of formula (I).

(3) FIG. 3: Growth kinetics of E. faecium (16D030) in the presence of the compound of formula (I).

(4) FIG. 4: Growth kinetics of MRSA USA300 in the presence of the compound of formula (I).

(5) FIG. 5: Growth kinetics of MSSA in the presence of the compound of formula (I).

(6) FIG. 6: Growth kinetics of S. epidermidis in the presence of the compound of formula (I).

(7) FIG. 7: Growth kinetics of Listeria monocytogenes in the presence of the compound of formula (I).

(8) FIG. 8: Growth kinetics of Streptococcus group A in the presence of the compound of formula (I).

(9) FIG. 9: Growth kinetics of Streptococcus group B in the presence of the compound of formula (I).

(10) FIG. 10: Growth kinetics of Streptococcus pneumoniae in the presence of the compound of formula (I).

(11) FIG. 11: LC-MS spectrum of the hydrogenated derivative of the compound of formula (I) (2690-f49-hydro). The mass the compound of formula (I)+2 is indicated in the circle.

EXAMPLES

(12) Growth of Fungi

(13) The fungus Ulocladium lanuginosum, strain CBS 102.26 (obtainable from Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands; www.cbs.knaw.nl) was cultured on a malt-extract agar plate for 7 days at 25 C. The fungus was subsequently inoculated in 3 flasks containing 50 ml Czapek-dox broth (CDB)+0.5% yeast extract. Each flask was incubated at a different temperatureroom temperature (RT), 25 C. or 30 C.for 7 days. After 7 days, the fungal media were filter-sterilized (0.45 m pore filter). A 1 ml aliquot of the filtrate was made and the rest of the filtrate was extracted 3 with ethyl acetate (EtOAc). The EtOAc was evaporated with a rotation evaporator and the residue was dissolved in 1 ml 5% DMSO in H.sub.2O. The filtrates and extracts were tested on Methicillin resistant Staphylococcus aureus (MRSA; porcine strain) in broth as follows.

(14) Antibiotic Assay

(15) Bacteria are cultured o/n at 37 C. in Luria Broth (LB). This culture is diluted 100 in fresh LB. This fresh culture is then used to test the filtrate and extract. The filtrate was tested in serial dilutions starting at 2 and further diluted with a factor 2 until 256. The extract was tested starting at 10 dilution and further diluted with a factor 2 until 1280. The bacteria with filtrate or extract were incubated o/n at 37 C. The read-out of the assay is by eye, if the bacteria have visibly grown, the sample is scored as non-inhibitory. If the broth is completely clear, the sample is scored as inhibitory. Using this assay, we established that the extract from fungi grown at RT inhibited growth. The extract from fungi grown at 25 C. was less potent and the extract from fungi at 30 C. had no effect.

(16) Purification of Compounds with Antibiotic Activity

(17) Based on the first results, 20 jars of 50 ml CDB+0.5% yeast extract were inoculated with fungus Ulocladium lanuginosum CBS 102.26 and incubated at RT for 7 days. The medium was filter sterilized as described, a 1 ml aliquot was made of the filtrate, to be able to test the starting material, and the remainder was extracted 3 with EtOAc. The EtOAc was evaporated and the residue was dissolved in 1 ml DMSO. The extract was then fractionated using a C18 Reprosil column on a Prep. HPLC (see Table 4 for apparatus information) with UV-detection between 214 nm and 254 nm. The buffers used are described in Table 1, the buffer B gradient runs as described in Table 2.

(18) TABLE-US-00001 TABLE 1 HPLC buffers Buffer A Buffer B 95% MiliQ 5% MiliQ 5% Acetonitrile 95% Acetonitrile 0.1% Trifluoroacetic Acid 0.1% Trifluoroacetic Acid

(19) TABLE-US-00002 TABLE 2 HPLC gradient Time (min.) Percentage buffer B 0-5 0 5-85 0-100 85-90 100 90-95 100-0 95-100 0

(20) The 95 fractions collected were pooled per 6 and the pooled fractions were dried in a speed-vac o/n. The fraction residues were dissolved in 1 ml 5% DMSO and tested on E. faecium (15A623) in serial dilutions starting at 2. Pooled fraction 9 was the only one that inhibited bacterial growth, so the individual fractions in pooled fraction 9 (fractions 49-54) were tested further. 1 ml of each fraction was dried in a speed-vac o/n and each fraction residue was dissolved in 1 ml 5% DMSO to be tested on E. faecium. Fraction 49 was the only fraction that inhibited bacterial growth, which concurs with the UV detection of the HPLC fractionation (data not shown). The fraction was then tested on 19 additional strains of bacteria (Table 3). The filtrate and extract were found to inhibit the Gram-positive bacteria, but not the Gram-negative bacteria.

(21) Identification of the Compound with Antibiotic Activity

(22) The sample was run on analytical HPLC (see Table 4 for apparatus information), which indicated the compound was >95% pure (data not shown). An aliquot of fraction 49 was analyzed using LC-MS (see Table 4 for apparatus information) and high resolution mass spectrometry. The calculated mono-isotopic mass was 308.1067 with a molecular formula prediction of C.sub.16H.sub.20O.sub.6. However, the MS/MS spectrum measured with the high resolution MS suggested that the calculated mass was actually 2M. The spectrum shows peaks at 155.1 (representing M+H), 309.2 (representing 2M+H) and 617.4 (representing 4M+H) (FIG. 1), suggesting that the exact mass is actually 154.05 (M) with molecular formula C.sub.8H.sub.10O.sub.3. UV-VIS was also measured, revealing a UV-max at 221 nm (data not shown).

(23) TABLE-US-00003 TABLE 3 Antibiotic activity of purified fraction on different strains of bacteria Minimal concentration for Gram + antibacterial Bacterium ESBL* or activity Staphylococcus epidermidis Gram+ 18.8 mg/l (08A1071) Methicillin-sensitive Staphylococcus Gram+ 4.69 mg/l aureus 476 (S0101), ATCC BAA- 1721 Methicillin-resistant Staphylococcus Gram+ 18.8 mg/l aureus (MRSA) ST8: USA300 (S1474) FPR3757, NARSA collection NRS482 Enterococcus faecium (15A623) Gram+ 37.5 mg/l Enterococcus faecium (16D030) Gram+ 75.0 mg/l Streptococcus group A (23M092) Gram+ 75.0 mg/l Streptococcus group B (05A396) Gram+ 75.0 mg/l Streptococcus pneumoniae (14B186) Gram+ 75.0 mg/l Listeria monocytogenes (41-4a) Gram+ 75.0 mg/l Klebsiella pneumoniae (MR44) Gram No effect at 1.5 g/l Acinetobacter calcoaceticus Gram No effect at 1.5 g/l (15A600), NCCB 100514 Pseudomonas aeruginosa (04A191), Gram No effect at 1.5 g/l NCCB 100515 Stenotrophomonas maltophilia Gram No effect at 1.5 g/l (20A226), NCCB 100516 Klebsiella pneumonia (CP14) TEM-18 Gram No effect at 1.5 g/l Klebsiella pneumonia (JS006) SHV-18 Gram No effect at 1.5 g/l Klebsiella pneumonia (JS264) CTX-M- Gram No effect at 1.5 g/l 15 Escherichia coli (JS004) TEM-3 Gram No effect at 1.5 g/l Escherichia coli (CP131) SHV-3 Gram No effect at 1.5 g/l Escherichia coli (JS203) CTX-M-1 Gram No effect at 1.5 g/l *Extended-spectrum beta-lactamase
Elucidation of the Structure of the Bioactive Compound

(24) The entire fraction 49 was subsequently evaporated in the speed-vac and the residue was dissolved in 400 l DMSO-d6. The yield of the fraction was 15 mg dry weight (from 1 L fungal culture). A .sup.1H-NMR spectrum and a .sup.13C-NMR spectrum were measured. The .sup.13C-NMR spectrum revealed 8 carbon atoms, which together with the mono-isotopic mass measurements confirmed the calculated mono-isotopic mass of 154.0534 with molecular formula C.sub.8H.sub.10O.sub.3. We then went on to perform several 2D-NMR experiments, namely Correlation Spectroscopy (COSY), Total Correlation Spectroscopy (TOCSY), Heteronuclear Single-Quantum Correlation Spectroscopy (HSQC), Heteronuclear Multiple-Bond Correlation spectroscopy (HMBC) and a .sup.13C-Distortionless Enhancement by Polarization Transfer (.sup.13C-DEPT). The NMR data (not shown) led us to the structure (E)-8-methyl-7,8-dihydro-2H-oxocine-2,5(6H)-dione, i.e. the structure formula (I):

(25) ##STR00001##

(26) Formula (II) depicts the structure of (Z)-8-methyl-7,8-dihydro-2H-oxocine-2,5(6H)-dione, a compound identical to the compound of formula (I) except that the double bond between carbons 3 and 4 is in a cis configuration, rather than in a trans configuration as in formula (I).

(27) The compound of formula (I), which we name Ulocladine (currently renamed to Delphiton), has not been reported before. However, a similar structure has been published previously (J. F. Grove (1964), Metabolic Products of Stemphylium radicinum. Part I. Radicinin., Journal of the Chemical Society: 3234-3239; D. C. Aldridge and J. F. Grove, (1964), Metabolic Products of Stemphylium radicinum. Part II. ()-7-Hydroxy-4-oxo-oct-2-enoic Acid Lactone, Journal of the Chemical Society: 3239-3241; B. Lygo and N. O'Conner (1990), Epoxide Ring Opening by Dianions Derived from -Ketosulphones. Synthetic Studies on the 7-Hydroxy-4-oxopropenoate System and Medium-Ring Lactones, Synlett: 282-284). In these cases, however, it concerns a variant of the molecule in which the double bond is in a cis configuration, i.e. a molecule having the structure of formula (II). A preparation obtained from the fungus Stemphylium radicinum, comprising the compound of formula (II), is disclosed to have antagonistic effects toward the fungal plant pathogen Phytophtora erythroseptica.

(28) In the compound of formula (I), the double bond between carbons 3 and 4 is in a trans configuration. The J-coupling between the two vinyl protons of 16 Hz as observed with NMR confirmed the trans configuration. In the cis-configuration, the J-coupling would be significantly smaller.

(29) Growth Kinetics of Gram-Positive Bacteria in the Presence of the Compound of Formula (I)

(30) After elucidating the structure, we studied the growth kinetics of several Gram+ bacteria in the presence of 50, 100 and 200 g/ml of the compound of formula (I), in order to determine whether the compound was bacteriostatic or bactericidal. The bacteria were cultured at 37 C. for 24 hrs (with the exception of Streptococcus pneumoniae which was cultured until 6 hrs) in the presence of the compound and samples were taken for colony counting at 0, 1, 3, 6 and 24 hrs.

(31) More specifically, bacterial isolates as indicated below were grown on blood-agar plates. Liquid cultures were inoculated in Mueller-Hinton Broth and incubated at 37 C. o/n. The o/n cultures were used to prepare a 0.05-0.1 McFarland suspension in Mueller Hinton Broth, which was incubated at 37 C. until 0.5 McFarland (110.sup.6 CFU/ml). These suspensions were used to inoculate 5 ml pre-warmed Mueller-Hinton Broth with approximately 1.510.sup.5 CFU/ml with the appropriate antibiotic and antibiotic concentration. Directly after addition of the bacterial suspension a 0.25 ml sample was taken. Further 0.25 ml samples were taken after 1, 3, 6 and 24 hrs. Samples were added to 2.25 ml physiological salt solution. This suspension was used to prepare plates for CFU determination. The lower limit of detection was 40 CFU/ml. Bacterial suspension without antibiotics was checked for normal growth after 24 hrs. Efficient bactericidal activity is defined as at least 3 log.sub.10 reduction of CFU.

(32) The bacteria tested were two strains of Enterococcus faecium (FIGS. 2A and 2B), MRSA USA300 (FIG. 2C), MSSA (FIG. 2D), Staphylococcus epidermidis (FIG. 2E), Listeria monocytogenes (FIG. 2F), Streptococcus group A (FIG. 2G), Streptococcus group B (FIG. 2H) and Streptococcus pneumoniae (FIG. 2I).

(33) Based on these time-kill curves, we can conclude that the compound of formula (I) is bacteriostatic on Streptococcus Group B and on both Enterococcus faecium strains but that the compound is bactericidal on MRSA USA300, Staphylococcus epidermidis, MS SA, Streptococcus Group A, Streptococcus pneumoniae and Listeria monocytogenes.

(34) Hydrogenation Inactivates the Antibacterial Activity of the Compound of Formula (I)

(35) The trans configuration of the double bond in the compound of formula (I) is unique and generates a highly reactive moiety, a Michael acceptor for nucleophilic attack. We hypothesized that this was the active core that was responsible for the antibacterial activity. In an approach to confirm the importance of the trans-configuration of the double bond, we hydrogenated the double bond. The hydrogenation reaction was done as previously described by Aldrige and Grove (1964, supra). Briefly, the compound was dissolved in ethyl acetate and was hydrogenated in the presence of 5% palladium-carbon for 96 h. The product was filtered and the ethyl acetate was evaporated off. The residue was then dissolved in acetonitrile and analyzed by analytical HPLC to check the purity. The hydrogenated sample, named 2690-f49-hydro, was also analyzed by LCMS, which revealed the exact mass of the compound of formula (I)+2 (156.84) (FIG. 3), as expected. A .sup.1H-NMR spectrum was measured, wherein the absence of peaks for the vinyl protons between 7 and 8 ppm, confirmed that the double bond in the compound of formula (I) was saturated (data not shown). To estimate the relative concentration of the hydrogenated sample, a standard addition experiment was performed with the compound of formula (I) and 2690-f49-hydro on analytical HPLC. The area under the peaks of both samples were roughly the same (data not shown), suggesting equal concentrations of the compound of formula (I) and its hydrogenated derivative 2690-f49-hydro. The compound of formula (I) was active against E. faecium (15A623), whereas its hydrogenated derivative 2690-f49 hydro, at similar concentrations, was not active. Even at 16-fold higher concentrations, 2690-f49 hydro was not active against E. faecium (15A623). We conclude that the double bond is required for antibiotic activity of the compound of formula (I).

(36) TABLE-US-00004 TABLE 4 Apparatus Apparatus Column Prep. HPLC LC-MS QP8000 prep: Reprosil-PurC18 AQ; Shimadzu 250 22 mm; Screening Device Piston pump: Shimadzu LC- R10.AQ.S2522 8A Particle size: 10 m UV-VIS detector: Shimadzu Pore size: 120 SPD-10A (standard 214 nm and Carbon load: 15% 254 nm) pH range: 1-10 Fraction collector: Gilson Liquid Handler 215 Analytical VP SHIMADZU Reprosil-PUR C18 AQ; HPLC System controller: SCL-10A 250 4.6 mm; Screening Device Pump: LC-10AT R15.AQ.S2546 UV-VIS detector: SPD-10A Particle size: 5 m (standard 214 nm and 254 nm) Pore size: 120 Auto injector: SIL-10AD Carbon load: 15% (sample loop 50 l) pH range: 1-10 LC-MS VP SIMADZU Reprosil-PurC18 AQ; Column oven: CTO-10AS 250 22 mm; Screening Device System controller: SCL-10A R10.AQ.S2522 Pump: LC-10AD Particle size: 10 m UV-VIS detector: SPD-10A Pore size: 120 (standard 214 nm and 254 nm) Carbon load: 15% Auto injector: SIL-10AD pH range: 1-10 (sample loop 50 l) MS: Finnigan LCQ Deca XP Max