Novel aliphatic epoxide hydrolase activities from dematiaceous fungi

ELSEVIER

FEMS Microbiology Letters 141 (1996) 239-243

Novel aliphatic epoxide hydrolase activities from dematiaceous fungi Gideon Grogan ‘, Stanley M. Roberts b, Andrew J. Willetts at* a Department

of Biological Sciences, b Department

Washington Singer Laboratories. of Chemistry,

University of Exeter,

University of Liverpool, Liverpool M9

Exeter, 3BX,

Devon EX4 4QG, UK

UK

Received 17 May 1996; accepted 30 May 1996

Abstract Epoxide hydrolases were found to be constitutively expressed in dematiaceous fungi coincident with secondary metabolite pigment production in stationary or idiophase. Washed-cell preparations of two fungi, Ulocladium atrum CMC 3280 and Zopfiella karachiensis CMC 3284, exhibited affinity for 2,2-dialkylated oxiranes, for which contrasting enantioselectivities were observed, but not for aromatic styrene oxide or alicyclic cyclohexene oxide type substrates. Lyophilised preparations of soluble epoxide hydrolase activities proved to be effective catalysts for the mild hydrolysis of aliphatic epoxides. Keywords:

Dematiaceous fungi; Biotransformation; Chiral epoxide; Epoxide hydrolase

1. Introduction Chiral epoxides and chiral vicinal diols are important intermediates in the synthesis of pharmaceutical compounds. Recent reports of enantioselective epoxide hydrolases (EH) from microorganisms [1,2] have stimulated interest in the potential utility of these enzymes as tools for the resolution of racemic epoxides of interest, being more abundant and accessible than the much-studied analogous activities isolated from mammalian liver [3]. Reports of these microbial activities are still relatively rare, however, and currently reported sources are largely restricted to actinomycetic bacteria [4,5] and a few fungi including strains of Fusarium solani [6] and Helminthosporium sativum [7].

With the notable exception of the epoxide hydrolase from Aspergillus niger LCP 521 [1,8], the nature and synthetic utility of fungal EH enzymes with respect to a wide range of useful xenobiotic substrates has received little attention. The dematiaceous hyphomycetes, of which the Helminthosporium strain is an example, have been shown to catalyse a number of interesting biotransformations including sulfoxidation [9] Baeyer-Villiger oxygenation [lo] and asymmetric Diels-Alder addition [l 11. A screen of over 50 fungi for EH activity using 1,2-epoxyoctane revealed that two such dematiaceous hyphomycetes, Ulocladium atrum CMC 3280 and Zopjiella karachiensis CMC 3284, were found to express the highest levels of activity. The results of initial studies into the nature of these activities are presented herein.

* Corresponding author. Fax: +44 (1392) 264 668 0378-1097

/ 96/ $12.00

Copyright 0 1996 Federation of European Microbiological Societies. Published by Elsevier Science B.V

PIISO378-1097(96)00228-5

240

G. Grogan et al. IFEMS

Microbiology

Letters 141 (1996)

239-243

2. Materials and methods

2.3. Preparation

2. I. Chemicals

Fungal mycelium from 4 1 of culture medium was harvested as above, washed and homogenised with buffer. The homogenate was divided into 40 ml aliquots of cells which, after freezing, were disrupted using a French press. Cell debris was removed by centrifugation (13,000 r.p.m. for 40 min at 4”C), and the pooled supernatants filtered to remove residual particles. A 30-80% ammonium sulfate cut of the supernatant was obtained and, after the protein pellet was dissolved and dialysed against buffer, the active enzyme solution was lyophilised for 3 days. Typical activities of these preparations were approximately 8 mU EH mg-’ Zopjiella extract and 5 mU EH mg-l Ulocladium extract with respect to 1,2epoxyoctane.

All 2,2-alkylated oxiranes and corresponding racemic diols were synthesised according to the methods of Mischitz et a1.[2]. All other materials were purchased from Aldrich Chemical Company (Poole, Dorset, UK). 2.2.

Microorganisms

Ulocladium atrum CMC 3280 and Zopfiella kaCMC 3284 were obtained from Chiroscience Ltd. (Cambridge, UK). Cultures were maintained on malt-extract agar and grown on a medium consisting of 7.5 g L-’ corn-steep powder (Ferntech) and 10 g L-’ glucose, adjusted to a pH of 4.85 at 25°C. A 0.5 cm square of fungal mat was homogenised in 5 ml of sterile buffer and 0.5 ml of this was used to inoculate 60 ml of medium in a 250 ml Erlenmeyer flask. For growth curves, biomass from 60 ml cultures was harvested daily for dry mass and specific activity determinations. For larger fermentations, 60 ml of culture grown for 3 days was transferred to 1 1 of sterile medium in a 2 1 Erlenmeyer flask. After a further 3-4 days growth, at which point the culture medium had turned black, it was harvested by filtration through muslin, washed with 50 mM phosphate buffer, pH 8.0 (containing 1 mM EDTA and 1 mM dithiothreitol), and wrung dry. A litre of medium typically yielded 30 g of mycelium. rachiensis

Table 1 Hydrolysis

of some 2,2_dialkylated

oxiranes

by whole-cell

preparations

of crude extracts

2.4. General epoxide hydrolase assay Whole cells (washed in buffer and resuspended in an equivalent growth volume of buffer) or enzyme solutions (1 ml) were challenged with 10 mM 1,2epoxyoctane from an ethanolic stock solution. After 30 min, a 200 11 aliquot was removed and extracted into 200 ~1 diethyl ether. The composition of the mixture was determined by injection of 1 ~1 of the organic phase onto a 25 m BP 10 capillary GC column using endo-bicyclo[2.2.1 Jheptan-2-01 as internal standard.

of Vlocladium atrum CMC 3280 and ZopJiella karachiensis CMC

3284

2-ethyl-2-pentyl

Zopfielln karachiensis CMC

Vlocladium atrum CMC 3280

Epoxide

oxirane

t (h)

c (%)

e.e. E.

4.5

32.3

3284

e.e. E.

e.e. D.

e.e. D.

t (h)

c (%)

27.4

57.5

3.0

43.4

35.2

45.9

(R)42.5

2.0

49.7

(W17.3

(s)17.5

2-methyl-2-pentyl

oxirane

3.0

49.1

(0 44.0

2-methyl-2-heptyl

oxirane

3.0

33.4

(XI15.5

(R)30.9

2.0

47.9

(RF 8.9

(s)8.2

2-methyl-2-nonyl

oxirane

6.0

32.2

(s)7.7

(RI16.2

4.0

40.3

(s)16.2

(R)24.0

(s)-

(RI-

(s)-

(RI-

t = time; C = conversion; e.e. E = enantiomeric excess residual epoxide; e.e. D = enantiomeric mycelimn and 15 mM substrate in 50 mM phosphate buffer, pH 8.0.

excess product diol. Each reaction contained

10 g

G. Grogan et al. / FEMS Microbiology Letters 141 (1996) 239-243 600

magnesium sulfate, filtered and the solvent removed by vacuum. Diols and epoxides purified by flash silica chromatography using petroleum ether/diethyl ether as solvent. Enantiomeric excesses were determined by chiral GC using a Chirasil-Dex column (Chrompack Inc.). Absolute configurations were assigned by comparison of product retention times with authentic standards. For soluble EH biotransformations of terminal aliphatic epoxides, 5 mg lyophilised epoxide hydrolase was rehydrated per ml of buffer for 30 min, and substrate added at 10 mM from a 1 M stock ethanolic solution. Samples were analysed as described.

500

400

P E %

241

300

ct

3. Results and discussion 200

loo

a 0

I

2

3 Growth

4

5

6

7

8

time (days)

Fig. 1. Growth/EH activity profile for Ulocladium atrum CMC 3280. Biomass from 60 ml culture was harvested at daily intervals for dry mass (m) and specific activity (4) determinations.

2.5.

Biotransformations

For large-scale biotransformations: to 10 g mycelium in a 1 1 Erlenmeyer flask was added 500 ml of 50 mM phosphate buffer, pH 8.0. This mixture was shaken at 150 r.p.m. for 30 min at 25°C after which the epoxide substrate was added to a final concentration of 15 mM, in ethanol to 0.5% the final reaction volume. At intervals, 500 pl aliquots of the aqueous reaction mixture were removed and extracted with 500 ul of diethyl ether. The composition of the organic phase was determined by GC on a 25 m BP10 capillary column. At the appropriate time, the fungus was removed by filtration and the filtrate extracted with 3 X 150 ml of diethyl ether. The combined organic phases were dried using anhydrous

Growth/activity profiles for Ulocladium atrum CMC 3280 (Fig. 1) and Zopjiella karachiensis CMC 3284 (Fig. 2) revealed that after 2-3 days growth and without specific induction, large amounts of EH activity were expressed. Dematiaceous hyphomycetes are characterised by dark conidia or conidiophores [12] and the production of enzyme by the fungi under consideration was coincident with the production of those dark pigments produced as secondary metabolites in stationary or idiophase, under conditions of excess available carbohydrate, where the absence of one or more other nutritional factors has become limiting. It is possible to speculate that their expression in idiophase may be related to the processing of secondary metabolites, such as some polyketides and sesquiterpenes (e.g. trichodiol), which have been shown to possess epoxide functionalities [13]. Fungal epoxide hydrolases in the plant pathogen Fusarium solani has already been implicated in the detoxification of host epoxyacids during pathogenesis [6], but, in contrast to the current study, this activity was induced by growth on the polymer cutin, containing 1%hydroxy-9, lo-epoxyoctadecanoic acid, and only a small fraction (6%) of the EH activity was reported to be expressed constitutively. Washed whole-cell preparations of Ulocladium and Zopjiella were used to study the synthetic potential of these EH activities, using substrates previously demonstrated to engender enantioselectivity in microbial EH enzymes [1,2]. When challenged with an

242

G. Grogan et al. IFEMS

Microbiology

homologous series of 2,2-alkylated oxiranes (Table l), it was observed that the enantioselectivity of the Ulocludium enzyme (UEH) is compromised with the increase in alkyl chain length, in direct contrast to results obtained with Rhodococcus NCIMB 11216 (REH), where enantioselectivity improves with this progression [2]. This enzyme also afforded enantiocomplementary products to REH, although with much reduced optical purity. ZopJiella karachiensis (ZEH) enzyme, afforded the same series of epoxides as REH when challenged with 2-ethyl-2-pentyloxirane and 2-methyl-2-pentyloxirane, but then switched selectivity from (R)- to (S)- as the chain increases from C5 to CT, at which point, an extension appeared to engender improved enantioselectivity, as with REH. The possibility of the contribution of multiple EH activities with opposing selectivities in these organisms, as is observed in mammalian liver [14-l@ is currently under investigation. In contrast to epoxide hydrolases expressed by Aspergillus niger LCP 521 and Beauveriu szdjiirescens ATCC 7159 [l], whole-cell UEH and ZEH demonstrated very poor activity towards the aromatic epoxide, styrene oxide, and the alicyclic epoxides cyclohexene oxide and lmethylcyclohexene oxide, even at low concentrations (5 mM). The value of hydrolytic enzymes, such as lipases, as commodity chemicals available as easy to use powders is beyond dispute [3]. To date, the utility of epoxide hydrolases in this vein has not been investigated to any depth. Lyophilised preparations of cell extracts from Zopjiella and Ulocladium were Table 2 Activity of lyophihsed preparations terminal aliphatic and Spiro-oxiranes

of EH activities

from

Letters 141 (1996)

500

450

400

350

300 a d 2

250

c! 200

150

100 -2 50

-0

0 0

challenged

1,2-Epoxybutane 1,2-Epoxyhexane 1,2-Epoxyoctane 1,2-Epoxydecane 1,2-Epoxydodecane 1,2_Epoxytetradecane 1,2-Epoxyhexadecane Cyclohexene oxide Styrene oxide 1-Oxaspiro[2,5]octane

0.0 2.1 5.1 4.1 1.0 0.3 0.0 0.0 0.0 4.9 enzyme and 10 mM substrate

3

4

5

6

CMC 3280 EH

with a series of terminal

CMC 3280 and Zopjella

Enzyme activity (mU mggr lyophilised

5 mg ml-’

2

7

8

Fig. 2. Growth/EH activity profile for Zopfiella karachiensis CMC 3284. Biomass from 60 ml culture was harvested at daily intervals for dry mass (m) and specific activity (+) determinations.

Substrate

Assays contained

I

Growth time (days)

Ulocladium atrm

Ulocladium atrm

239-243

Zopfiella karachiemis

buffer, pH 8.0.

epox-

CMC 3284 toward

preparation)

0.0 2.0 8.2 10.5 4.0 0.7 0.0 0.0 0.0 5.5 in 50 mM phosphate

karachiensis

aliphatic

CMC 3284 EH

G. Grogan et al. I FEMS Microbiology

ides (Table 2). Each enzyme was shown to hydrolyse Cs and Cl0 aliphatic epoxides with most preference, with little or no activity detected toward epoxybutane or epoxyhexadecane. The enzymes were also able to hydrolyse a simple achiral Spiro-oxirane, a function often observed in fungal secondary metabolites [13]. In conclusion, it is possible that a ‘generic’ aliphatic EH activity is expressed constitutively to a high degree in the dematiaceous hyphomycetes. The exhibition of a wide variety of enantioselectivities expressed by these fungi therefore identifies them as a potentially valuable reservoir of EH activities for future target-directed screening.

Acknowledgments

We thank the B.B.S.R.C. and Chiroscience Cambridge, UK, for funding.

Ltd.,

References [1] Pedragosa-Moreau, S., Archelas, A. and Furstoss, R. (1993) Microbiological Transformations 28. Enantiocomplementary epoxide hydrolyses as a preparative access to both enantiomers of styrene oxide. _I. Org. Chem. 58, 5533-5536. [2] Mischitz, M., Kroutil, W. Wandel, U. and Faber, K. (1995) Asymmetric microbial hydrolysis of epoxides. Tet. Asymm. 6, 1261-1272. [3] For a review of the use of hepatic epoxide hydrolases and lipases in organic synthesis, see Faber, K. (1992) in Biotransformations in Organic Synthesis, Springer, Berlin. [4] Hechtberger, P., Wimsberger, G., Mischitz, M., Klempier, N. and Faber, K. (1993) Asymmetric hydrolysis of epoxides using an immobilised enzyme preparation from Rhodococcus sp. Tet. Asymm. 4, 1161-l 164.

Letters 141 (1996)

239-243

243

[5] Nakamura, T., Nagasawa, T., Yu, F., Watanabe, I. and Yamada, H. (1994) Purification and characterisation of two epoxide hydrolases from Corynebacterium sp. strain N-1074. Appl. Environ. Microbial. 60, 46304633. [6] Kollatukudy, P.E. and Brown, L. (1975) Fate of naturally occurring epoxy acids: a soluble epoxide hydrase which catalyses cis- hydration, from Fusarium sol& pisi. Arch. Biochem. Biophys. 166, 599-607. [7] Imai, K., Marumo, S. and Mori, K. (1974) Derivation of (+)- and (-)-C,, juvenile hormone from its racemic alcohol derivative via fungal metabolism. J. Am. Chem. Sot. 96. 5925-5927. [8] Chen, X.J., Archelas, A. and Furstoss, R. (1993) Microbiological transformations 27. The first examples for preparativescale enantioselective or diastereoselective epoxide hydrolyses using microorganisms. An unequivocal access to all four bisabolol enantiomers. .I. Org. Chem. 58, 552885532. [9] Holland. H.L., Brown, F.M. and Larsen. B.G. (1994) Biotransformation of organic sulfides. Part 5. Formation of chiral para-alkyl benzyl methyl sulfoxides by Helminthorporium species NRRL 4671. Tet. Asymm. 5, 1241-1248. [lo] Carnell, A.J. and Willetts, A.J. (1992). Biotransformations by fungi: Regio- plus stereoselective Baeyer-Villiger oxidations by dematiaceous fungi. Biotechnol. Lett. 14, 17-20. [ll] Oikawa, H., Katayama, K., Suzuki, Y. and Ichihara, A. (1995) Enzymatic activity catalysing exo-selective Diels-Alder reaction in solanapyrone biosynthesis. J. Chem. Sot, Chem. Commun. 1321-1322 [12] Ellis, M.B. (1971) Dematiaceous Hyphomycetes, Commonwealth Agricultural Bureau, UK. [13] Turner, W.B. and Aldridge, D.C. (1983) Fungal Metabolites II, Academic Press, London. [14] Ota, K. and Hammock, D. (1980) Cytosolic and microsomal epoxide hydrolases: differential properties in mammalian liver. Science 207, 147991481. [15] Bellucci, G., Chiappe, C., Marioni, F. and Benetti, M. (1991) Regioselectivity and enantioselectivity of the cytosolic epoxide hydrolase-catalysed hydrolysis of racemic monosubstituted alkyloxiranes. J. Chem. Sot. Perkin Trans. 1, 361-363. [16] Bellucci, G., Chiappe, C., Cordoni, A. and Marioni, F. (1994) Different enantioselectivity and regioselectivity of the cytosolic and microsomal epoxide hydrolase-catalysed hydrolysis of simple phenyl substituted epoxides. Tet. Lett. 35, 42194222.