Isolation of Antibiotic 3 R,5 R -Dihydroxyhexanoate Polymers From Endophytic Fungi

Abstract

The extracts of two isolates in our library of fungal endophytes, Annulohypoxylon multiforme (TC2-046) from the medicinal plant Juniperus communis and a sterile filamentous isolate (KP1-131DD) from the marine alga Scytosiphon lomentaria, displayed similar antimicrobial bioactivity profiles with notable high activity against Staphylococcus aureus. Bioassay-guided fractionation led to the isolation of poly(3R,5R-dihydroxyhexanoic acid) oligomers ranging from the trimer to the 29-mer that exhibited significant and selective inhibition of S. aureus and Mycobacterium tuberculosis H37Ra in vitro.

Endophytic fungi are increasingly being recognized as important sources of bioactive natural products, with interest focusing on endophytes of medicinal plants ^{1 -5} and algae. ^{6 -8} In continuation of our investigation of endophytes isolated from Canadian medicinal plants and marine algae, ^{9 -13} the extracts of an Annulohypoxylon multiforme isolate (TC2-046) from Juniperus communis and a sterile gray filamentous fungus (KP1-131DD) obtained from the marine alga Scytosiphon lomentaria attracted our attention as they exhibited remarkably similar bioactivity profiles in our screening assays characterized by selective, high inhibition of Staphylococcus aureus.

Bioassay-guided fractionation of the ethyl acetate (EtOAc) extract of a 2-week, bench-scale (25 L) fermentation of KP1-131DD through liquid–liquid partition, flash chromatography, and normal-phase high-performance liquid chromatography (HPLC) led to the isolation of poly(3R,5R-dihydroxyhexanoic acid) (1; Figure 1) in high yield (23 mg/L, 20% of the extract by mass). Subsequent targeted fractionation of the EtOAc extract of TC2-046 by reversed phase HPLC gave 1 as a chromatographically homogenous peak but at significantly lower overall yield (2 mg/L, 20% of the extract by mass).

Figure 1

The structures of poly(3R,5R-dihydroxyhexanoic acid) (1) and 4R-hydroxy-6R-methyltetrahydropyran-2-one (2).

The structure of 1 was determined by mass spectrometry and NMR spectroscopy (1D and 2D), with high-resolution positive-mode electrospray ionization mass spectroscopy indicating that the oligomer was present in sizes ranging from the trimer to the 29-mer. Saponification of 1 allowed the determination of its absolute stereochemistry to be 3R, 5R through comparison of the NMR and polarimetric data obtained for the exclusive hydrolysis product, 4R-hydroxy-6R-methyltetrahydropyran-2-one (2), with literature values reported for each of the four diastereomeric lactones. ^14,15 Our data were consistent with those from a previous isolation of 1 when it was obtained at lengths ranging from the pentamer to the 27-mer from a Daldinia concentrica endophyte of a Manikara tree from Costa Rica. ^16,17

Although originally isolated from an extract that inhibited the nuclear factor kappa-light-chain-enhancer of activated B cells pathway regulator IκB kinase (IKK), ¹⁶ no bioactivity data has been reported for 1. In our hands, the polymer was found to possess selective inhibitory activity against both S. aureus (minimal inhibitory concentration [MIC]: 400 µg/mL; half maximal inhibitory concentration [IC₅₀ [95% CIs]: 55 [52–59] µg/mL; see supplemental data, figure S1) and Mycobacterium tuberculosis H37Ra (MIC: 400 µg/mL; IC₅₀ [95% CIs]: 67 [51–89] µg/mL; see supplemental data, figure S1) while showing no significant inhibition against Enterococcus faecium, Pseudomonas aeruginosa, Escherichia coli, Candida albicans, Saccharomyces cerevisiae, or Mycobacterium smegmatis. While the production of polyesters and polyhydroxyalkanoates is well documented in bacteria, ¹⁸ it remains uncommon in fungi; although the scale at which 1 can be produced may facilitate its development and application as a biopolymer in the future. ¹⁷

Experimental

General

All antimicrobial standards used as positive controls in susceptibility assays, except for rifampin, were obtained from Sigma Aldrich (Oakville, Ontario). Rifampin was obtained from Fisher Scientific (Ottawa, Ontario). Solvents for extraction and isolation were purchased from Fisher Scientific (Ottawa, ON, Canada), and deuterated solvents for NMR spectroscopy were purchased from Sigma-Aldrich (Oakville, ON, Canada). Solid-phase extraction was performed using Sep-Pak C18 Cartridges (55-105 µm, 125 Å, 2 g; Waters, NA, USA). Flash chromatography was performed using a Biotage Flash+ chromatography system fitted with SiliaSep silica cartridges (40-63 µm, 60 Å, 25 g; SiliCycle, QC, Canada). Semi-preparative normal-phase HPLC was performed on a Phenomenex Luna silica column (250 × 10 mm, 10 µm, 100 Å) using a Waters 510 isocratic pump and a Waters R401 refractive index detector. Semi-preparative reversed-phase HPLC was performed on a Phenomenex Luna C18 column (250 × 10 mm, 10 µm, 100 Å) using an Agilent 1100 HPLC system comprising a G1311A quaternary pump and a G1315C diode array detector. Optical rotations were recorded on an Optical Activity Ltd. AA-10 polarimeter at 589 nm. NMR spectra were recorded on an Agilent 400-MR DD2 instrument and were calibrated to residual protonated solvent. HRMS data were recorded on a Thermo LTQ Exactive instrument with an ESI source.

Endophyte Isolation

TC2-046 was isolated from the leaves of J. communis (New Brunswick Museum voucher specimen NBM VP-37482) in August 2010. ⁹ Leaf surfaces were sterilized by immersing in 5.25% aqueous sodium hypochlorite (NaOCl) for 5 seconds and rinsed with sterile distilled water (H₂O) for 10 seconds and then immersed in 70% ethanol (EtOH) for 10 seconds and rinsed with sterile distilled H₂O and blotted dry on autoclaved paper towel. Sterile tissue was immediately aseptically cut into pieces (5 mm × 5 mm) that were placed onto 2.0% malt extract agar, and incubated at room temperature under ambient light. Endophytic fungi were subcultured onto fresh 2.0% malt extract agar until pure cultures were obtained.

Scytosiphon lomentaria was collected from the shore of Green’s Point, L’Etete, New Brunswick, Canada (45°02.363′N, 066°53.483′W) in July 2013. Portions (5 cm in length) of algal tissue were individually sterilized by immersing in 6.0% NaOCl for 10 seconds and rinsed with sterile distilled H₂O for 10 seconds and then immersed in 70% EtOH for 15 seconds and rinsed with sterile distilled H₂O and blotted dry on autoclaved paper towel. Sterile tissue was immediately aseptically cut into pieces (approximately 5 mm × 5 mm), placed onto 2.0% malt extract agar, and incubated at ambient room temperature. Endophytic fungi were subcultured onto fresh 2.0% malt extract agar until a pure culture was obtained.

Identification of Endophytes

TC2-046 was identified as A. multiforme through examination of spore morphology and colonies grown on malt extract, potato dextrose, cornmeal, and Czapek–Dox agars. The taxonomic classification was confirmed by comparison of the internal transcribed spacer (ITS) and 5.8S rRNA gene DNA regions with corresponding sequences available in the GenBank database (National Center for Biotechnology Information, US National Library of Medicine, Bethesda, MD, USA). The genomic DNA of TC2-046 was isolated using a DNEasy plant mini kit (Quigen, Toronto, Ontario) as directed by the manufacturer; the ITS gene was amplified by PCR using the ITS1 and ITS4 universal fungal primers (Invitrogen, Burlington, Ontario) as previously described, ⁶ and the amplified ITS DNA was sequenced by Genome-Québec (Montreal, Québec). The TC2-046 DNA sequence was checked for ambiguity before being compared with existing GenBank sequence data using BLAST. The ITS gene sequence of TC2-046 was found to have >99% homology with numerous conspecific A. multiforme isolates and has been deposited in GenBank (accession number: KC916700).

KP1-131DD did not produce fruiting bodies with characteristic morphological features on malt extract, potato dextrose, cornmeal, or Czapek–Dox agars. Attempts to identify the isolate through DNA sequencing were unsuccessful as DNA from the ITS region could not be isolated and amplified after repeated attempts using published procedures. ⁶ On malt extract agar, the isolate grew as a flat, filamentous, gray colony with a rough surface and undulating margin that lacked spores and produced an orange pigment that diffused into the growth medium. On potato dextrose agar, the isolate grew as a flat, filamentous, gray colony with a rough surface, and filiform margin that lacked spores. On cornmeal agar, the isolate grew as a flat, filamentous, beige colony with a smooth surface and filiform margin that lacked spores. On Czapek–Dox agar, the isolate grew as a flat, filamentous, gray colony with a rough surface and undulating margin that lacked spores. Microscopy revealed the isolate to be sterile with septate hyphae and brown pigmentation.

Fermentation and Extraction

KP1-131DD was fermented in 2.0% malt extract broth (25 L; 250 × 100 mL batches in 250 mL Erlenmeyer flasks stoppered with foam baffles) at room temperature with shaking (150 rpm) for 2 weeks. The cultures were then sonicated for 30 seconds; the mycelia and cell debris separated from the spent broth by filtration and the broth was extracted with EtOAc (3 × 300 mL EtOAc per liter of broth). The organic fractions were combined and concentrated in vacuo (2.70 g). TC2-046 was fermented (1 L; 10 × 100 mL) and extracted (10 mg) in the same fashion as KP1-131DD.

Bioassay-Guided Fractionation of KP1-131DD

The extract (2.70 g) was dissolved in 9:1 methanol (MeOH)/H₂O (200 mL) and extracted with hexanes (3 × 100 mL) before being diluted with H₂O (100 mL) and extracted with dichloromethane (CH₂Cl₂) (3 × 100 mL). The aqueous fraction was concentrated in vacuo, dissolved in H₂O (200 mL) and extracted with EtOAc (3 × 100 mL) and n-BuOH (3 × 100 mL). The 5 fractions were concentrated in vacuo with the CH₂Cl₂ fraction (1.39 g) showing antimicrobial activity of S. aureus. The CH₂Cl₂ fraction (1.13 g) was separated by normal-phase flash column chromatography (column eluted with 100% hexanes to 100% EtOAc in 10% increments, 1:1 MeOH/EtOAc, and 100% MeOH) with the 12th fraction of the column (1:1 MeOH/EtOAc) giving 1 (579 mg).

Bioassay-Guided Fractionation of TC2-046

The extract (10 mg) was loaded onto a 2 g C-18 Sep-Pak Cartridge and eluted first with H₂O (14 mL), then MeOH (14 mL) and then EtOAc (14 ml). The MeOH fraction (8 mg) was subjected to reversed-phase HPLC (gradient elution from 95:5 H₂O:acetonitrile to 100% acetonitrile over 10 minutes and then held at 100% acetonitrile for 5 minutes) to give 1 (2 mg) as a single homogenous peak.

Poly(3R,5R-Dihydroxyhexanoic Acid) (1)

IR (NaCl, thin film): 3426, 2987, 2932, 1723, 1387, 1273, 1176, 753 cm^{-1 ; 1}H NMR (400 MHz, CD₃OD, supplemental data, figure S2): 5.10 (1H, m, H-5), 4.09 (1H, m, H-3), 2.51 (1H, dd, J = 15.2, 4.7 Hz, H-2a), 2.41 (1H, dd, J = 15.2, 8.1 Hz, H-2b), 1.86 (1H, ddd, J = 14.0, 7.9, 7.0 Hz, H-4a), 1.68 (1H, ddd, J = 13.8, 6.5, 5.1 Hz, H-4b), 1.28 (3H, d, J = 6.3 Hz, H-6); ¹³C NMR (100 MHz, CD₃OD, supplemental data, figure S3): 172.8 (C-1), 70.1 (C-5), 66.6 (C-3), 43.9 (C-2), 43.7 (C-4), 20.2 (C-6); HRESIMS (see supplemental data, table S1 for complete list of data): m/z 431.1888 [M+Na⁺] (calcd for trimer, C₁₈H₃₂O₁₀Na⁺, 431.1909), 561.2521 [M+Na⁺] (calcd for tetramer, C₂₄H₄₂O₁₃Na⁺, 561.2545), 691.3151 [M+Na⁺] (calcd for pentamer, C₃₀H₅₂O₁₆Na⁺, 691.3180), 1895.9455 [M + 2H⁺] (calcd for 29-mer, C₁₇₄H₂₉₄O₈₈ ²⁺, 1895.9369).

Hydrolysis of Poly(3,5-Dihydroxyhexanoic Acid)

Compound 1 (10 mg) was dissolved in MeOH (2.0 mL); 10% aqueous potassium hydroxide (0.5 mL) was added and the mixture stirred at room temperature for 75 hours. The solution was acidified by adding 1 M hydrochloric acid (2.5 mL) and concentrated in vacuo. The reaction product was subjected to normal-phase column chromatography (5 g silica; column eluted with 100% EtOAc and washed with 100% MeOH) with fractions collected in test tubes. The contents of the test tubes were combined according to their thin layer chromatography profiles to give 2 (4 mg) as the only product.

4R-Hydroxy-6R-Methyltetrahydropyran-2-One (2)

$[\alpha {]}_{\mathrm{D}}^{23}$ : +19 (c 0.27, CHCl₃); IR (NaCl, thin film): 3419, 2919, 2856, 1641, 1463, 1379, 1252, 1184 cm^{-1 ; 1}H NMR (400 MHz, CDCl₃, supplemental data figure S4): 4.84 (1H, ddq, J = 11.2, 6.5, 3.1 Hz, H-6), 4.39, (1H, m, H-4), 2.74 (1H, dd, J = 17.7, 5.0 Hz, H-3a), 2.62 (1H, ddd, J = 17.6, 3.7, 1.8 Hz, H-3b), 1.98 (1H, dddd, J = 14.5, 3.8, 3.4, 1.7 Hz, H-5a), 1.74 (1H, ddd, J = 14.5, 11.2, 3.3 Hz, H-5b), 1.41 (3H, d, J = 6.4 Hz, H-7); ¹³C NMR (100 MHz, CDCl₃, supplemental data figure S5): 170.4 (C-2), 72.3 (C-6), 63.0 (C-4), 38.5 (C-5), 37.8 (C-3), 21.5 (C-7); HRESIMS: m/z 131.0704 [M+H⁺] (calcd for C₆H₁₁O₃ ⁺, 131.0703).

Antifungal Bioassay

Antifungal activity against C. albicans (ATCC 14053) and S. cerevisiae (ATCC 9763) was evaluated using a microbroth dilution antibiotic susceptibility assay modified from McCulloch et al. ¹⁹ Immediately prior to use, stock solutions of fungal extracts, fractionation products, or purified natural products were prepared at the desired concentration in sterile-filtered dimethyl sulfoxide (DMSO,40 µL) and diluted with either Difco Sabouraud dextrose broth (C. albicans, 960 µL) or yeast mold broth (S. cerevisiae, 960 µL) (Becton Dickinson, Mississauga, Ontario). The resulting test solutions (100 µL; 4% DMSO) were transferred to the non-peripheral wells of a clear, non-tissue cultured 96-well microtiter plate in triplicate (BD Falcon, Becton Dickinson, Mississauga, Ontario). Each plate contained 3 positive control wells (C. albicans: nystatin, 5.0 µg/mL; S. cerevisiae: amphotericin B, 5.0 µg/mL, 100 µL per well). Wells were then inoculated with suspensions of either C. albicans or S. cerevisiae (100 µL; 1 × 10⁶ colony forming unit [CFU]/mL), to obtain a cell density of 5 × 10⁵ CFU/mL. Sterile water (200 µL) was added to all perimeter wells to reduce evaporation from experimental wells. Each plate contained 3 negative control wells (4% DMSO in appropriate broth [100 µL] inoculated with appropriate fungi [100 µL; 1 × 10⁶ CFU/mL]) and 3 untreated blank wells (2% DMSO in appropriate broth [200 µL]). Initial and final optical densities (OD) were determined for each well by recording absorbance at 600 nm immediately before and after incubation for 24 hours at 37°C using a Molecular Devises Emax microplate reader (Molecular Devices; Sunnyvale, CA, USA).

Initial OD readings were subtracted from the final readings for each well to obtain the change in OD (∆OD). ∆OD values were corrected for background absorbance of the media by subtracting the mean ∆OD readings of the blanks from the mean ∆OD readings of the control and test wells. The percentage inhibition of fungal growth is defined as:

$$[1-\text{(mean test or positive control ∆OD/mean negative control ∆OD)}]\times 100$$

Antibacterial Bioassay

Antibacterial activity against S. aureus (ATCC 29213), P. aeruginosa (ATCC 10145), E. coli (ATCC 25922), and E. faecium (ATCC 35667) was evaluated in the same manner as described for antifungal assay. The growth medium used for P. aeruginosa, E. coli, and S. aureus was BBL Mueller Hinton II cation adjusted broth (Becton Dickinson, Mississauga, Ontario), whereas BBL Brain Heart Infusion broth (Becton Dickinson, Mississauga, Ontario) was used for E. faecium. Positive controls consisted of a triplicate concentration of antibiotic (1.25 µg/mL, erythromycin for S. aureus; 20 µg/mL and 2.5 µg/mL, gentamicin for P. aeruginosa and E. coli, respectively; 1.25 µg/mL, tetracycline for E. faecium; 100 µL per well) inoculated with a suspension of the appropriate pathogen (100 µL).

Antimycobacterial Bioassay

Antimycobacterial activity against M. tuberculosis strain H37Ra (ATCC 25177) and M. smegmatis (ATCC 70084) was evaluated using a microplate resazurin assay according to O’Neill et al. ²⁰ Immediately prior to use, stock solutions of fungal extracts, fractionation products, or purified natural products were prepared at the desired concentration in sterile-filtered DMSO (40 µL) and were diluted with modified Middlebrook 7H9 broth (960 µL, BBL MGIT, Becton Dickinson, Mississauga, Ontario). The resulting test solutions (100 µL) were transferred to the non-peripheral wells of a black non-tissue culture treated, low-binding, 960-well microtiter plate (VWR, Mississauga, Ontario) in triplicate and inoculated with suspensions of the appropriate organism (100 µL) of cell density 2.0 × 10⁶ cells/mL. Sterile water (200 µL) was added to all perimeter wells to reduce evaporation from experimental wells. The positive control consisted of rifampin (M. tuberculosis, 0.02 µg/mL) or ciprofloxacin (M. smegmatis, 2.5 µg/mL). In addition to the positive controls, negative controls (4% DMSO in modified Middlebrook 7H9 broth [100 µL] inoculated with suspensions of the appropriate organism [100 µL]), blanks (2% DMSO in modified Middlebrook 7H9 broth [200 µL], and test solutions [100 µL] with modified Middlebrook 7H9 broth [100 µL]) were included in triplicate in each plate. Mycobacterium tuberculosis plates were incubated (37°C; 5% CO₂) for 3 days, in a humid environment and M. smegmatis plates were incubated for 1 day (37°C). Following incubation, a solution of resazurin (0.0625 mg/mL) in aqueous Tween 80 (50 µL) was added to all non-peripheral wells. Plates were then incubated for a further 24 hours, sealed with an adhesive polyester film (50 µm; VWR, Mississauga, Ontario), and mycobacterial growth was assessed fluorometrically at 37°C (Molecular Devices Gemini EM dual-scanning microplates spectrofluorometer with a 530 nm excitation filter and a 590 nm emission filter operating in top-scan mode). Fluorescence values were corrected for any background fluorescence of the media and test samples by subtracting mean fluorescence readings of the appropriate blanks from the mean fluorescence readings of the control and test wells. The percentage inhibition of mycobacterial growth was then defined as:

$$[1-(\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{o}\mathrm{s}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{f}\mathrm{l}\mathrm{u}\mathrm{o}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e}∕\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{n}\mathrm{n}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{f}\mathrm{l}\mathrm{u}\mathrm{o}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e})]\times 100$$

Supplemental Material

Supplementary material - Supplemental material for Isolation of Antibiotic 3R,5R-Dihydroxyhexanoate Polymers From Endophytic Fungi

Supplemental material, Supplementary material, for Isolation of Antibiotic 3R,5R-Dihydroxyhexanoate Polymers From Endophytic Fungi by Nicholas J. Morehouse, Andrew J. Flewelling, John A. Johnson and Christopher A. Gray in Natural Product Communications