Synthetic enzymes for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid and their use

Abstract

The present invention relates to synthetic enzymes for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid, the use thereof for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid, DNA coding for these enzymes and microorganisms transformed with this DNA.

Description

[0001] The present invention relates to synthetic enzymes for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid, the use thereof for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid, DNA coding for the aforementioned enzymes and microorganisms transformed therewith.

[0002] The first article relating to the degradation of eugenol was written by Tadasa in 1977 (Degradation of eugenol by a microorganism. Agric. Biol. Chem. 41, 925-929). It describes the degradation of eugenol with a soil isolate which was presumed to be Corynebacterium sp. In this process ferulic acid and vanillin were identified as intermediate degradation products and the subsequent degradation was assumed to proceed via vanillic acid and protocatechuic acid.

[0003] In 1983 another article by Tadasa and Kyahara appeared (Initial Steps of Eugenol Degradation Pathway of a Microorganism. Agric. Biol. Chem. 47, 2639-2640) on the initial steps of eugenol degradation, this time with a soil isolate which was identified to be Pseudomonas sp. In this article eugenol oxide, coniferyl alcohol and coniferylaldehyde were described as intermediates for the formation of ferulic acid.

[0004] Also in 1983 a report by Sutherland et al. appeared (Metabolism of cinnamic, p-coumaric, and ferulic acids by Streptomyces setonii. Can. J. Microbiol. 29, 1253-1257) on the metabolism of cinnamic, p-coumaric and ferulic acids by Streptomyces setonii. In this process ferulic acid was degraded via vanillin, vanillic acid and protocatechuic acid, the ring-cleaving enzymes catechol 1,2-dioxygenase and protocatechuate 3,4-dioxygenase having been indirectly identified in the cell-free extract.

[0005] In 1985 Ötük (Degradation of Ferulic Acid by Escherichia coli. J. Ferment. Technol. 63, 501-506) reported on the degradation of ferulic acid by a strain of Escherichia coli isolated from decaying bark. Here as well vanillin, vanillic acid and protocatechuic acid were found as degradation products.

[0006] isoeugenol) for the microbial oxidation of eugenol and isoeugenol. However, only when isoeugenol was used the process did produce high conversion rates; the results were very poor using eugenol.

[0007] In 1991 EP-A 453 368 appeared [“Production de vanilline par bioconversion de précurseurs benzeniques” (Production of vanillin by the bioconversion of benzene precursors)], in which the reaction to form vanillin from vanillic acid and ferulic acid using a basidiomycete—Pycnoporus cinnabarinus CNCM I-937 and 1-938—was observed.

[0008] In 1992 the Takasago Perfumery Company was granted a Japanese patent (Preparation of vanillin, coniferyl-alcohol and -aldehyde, ferulic acid and vanillyl alcohol—by culturing mutant belonging to Pseudomonas genus in presence of eugenol which is oxidatively decomposed; JP 05 227 980 21.1.1992) for the preparation of vanillin, coniferyl alcohol, coniferylaldehyde, ferulic acid and vanillyl alcohol from eugenol using a Pseudomonas mutant.

[0009] Also in 1992 U.S. Pat. No. 5,128,253 by Labuda et al. (Kraft General Foods) (Bioconversion process for the production of vanillin) was granted, in which a biotransformation process for the production of vanillin was described. Here as well the starting material was ferulic acid and the organisms used were Aspergillus niger, Rhodotorula glutinis and Corynebacterium glutamicum. The crucial feature was the use of sulphydryl components (e.g. dithiothreitol) in the medium. In 1993 the subject matter of the patent also appeared in the form of a publication (Microbial bioconversion process for the production of vanillin; Prog. Flavour Precursor Stud. Proc. Int. Conf. 1992, 477-482).

[0010] EP-A 542 348 (Process for the preparation of phenylaldehydes) describes a process for the preparation of phenylaldehydes in the presence of the enzyme lipoxygenase. Eugenol and isoeugenol are for example used as substrates. We have attempted to rework the process using eugenol, but have not succeeded in confirming the results of the reactions.

[0011] DE-A 4 227 076 [Verfahren zur Herstellung substitutierter Methoxyphenole und dafür geeigneter Mikroorganismus (Process for the production of substituted methoxyphenols and a microorganism suitable for said process)] describes the production of substituted methoxyphenols with a new Pseudomonas sp. The starting material used is eugenol and the products are ferulic acid, vanillic acid, coniferyl alcohol and coniferylaldehyde.

[0012] Also in 1995 a comprehensive review by Rosazza et al. (Biocatalytic transformation of ferulic acid: an abundant aromatic natural product; J. Ind. Microbiol. 15, 457-471) appeared on possible methods of biotransformation using ferulic acid.

[0013] The present invention relates to synthetic enzymes for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid from eugenol.

[0014] Synthetic enzymes according to the invention are for example:

[0015] a) eugenol hydroxylase,

[0016] b) coniferyl alcohol dehydrogenase,

[0017] c) coniferylaldehyde dehydrogenase,

[0018] d) ferulic acid deacylase and

[0019] e) vanillin dehydrogenase.

[0020] The invention also relates to DNA coding for the abovementioned enzymes and cosmid clones containing this DNA as well as vectors containing this DNA and microorganisms transformed with the DNA or the vectors. It also relates to the use of the DNA for the transformation of microorganisms for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid. The invention also relates to partial sequences of the DNA and functional equivalents. Functional equivalents are understood to be those derivatives in which individual nucleobases have been substituted (wobble substitutions) without resulting in any functional changes. In relation to proteins, amino acids can also be substituted without resulting in any functional changes.

[0021] The invention also relates to the individual steps for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid from eugenol, i.e. in concrete terms:

[0022] a) the process for the production of coniferyl alcohol from eugenol carried out in the presence of eugenol hydroxylase;

[0023] b) the process for the production of coniferylaldehyde from coniferyl alcohol carried out in the presence of coniferyl alcohol dehydrogenase;

[0024] c) the process for the production of ferulic acid from coniferylaldehyde carried out in the presence of coniferylaldehyde dehydrogenase;

[0025] d) the process for the production of vanillin from ferulic acid carried out in the presence of ferulic acid deacylase;

[0026] e) the process for the production of vanillic acid from vanillin carried out in the presence of vanillin dehydrogenase.

[0027] After NMG mutagenesis mutants with defects in individual stages of the catabolism of eugenol were obtained from the eugenol-utilising Pseudomonas sp. strain HR 199 (DSM 7063). Using total DNA of wild-type Pseudomonas sp. HR 199 partially digested with EcoRI a gene library was constructed in the pVK100 cosmid, which has a broad host spectrum and can also be replicated in stable form in pseudomonads. After packaging in 1-phage particles the hybrid cosmids were transduced to E. coli S17-1. The gene library comprised 1330 recombinant E. coli S17-1 clones. The hybrid cosmid of each clone was transferred by conjugation into two eugenol-negative mutants (mutants 6164 and 6165) of the Pseudomonas sp. HR 199 strain and tested for a possible capacity for complementation. In this test two hybrid cosmids (pE207 and pE115) were identified, the obtainment of which restored mutant 6165's capacity to utilise eugenol. One hybrid cosmid (pE5-1) resulted in the complementation of mutant 6164.

[0028] The complementing capacity of plasmids pE207 and pE115 was attributed to a 23 kbp EcoRI fragment (E230). A physical map of this fragment was prepared and the fragment completely sequenced. The genes vanA and vanB which code for vanillate demethylase were localised in a 11.2 kbp HindIII subfragment (H110). Another open reading frame (ORF) was found to be homologous to g-glutamyl cysteine synthetase produced by Escherichia coli. An additional ORF, which was homologous to formaldehyde dehydrogenases, was identified between the aforementioned ORF and the vanB gene. Two additional ORF's were found to be homologous to the cytochrome C subunit or the flavoprotein subunit of p-cresol methylhydroxylase, respectively produced by Pseudomonas putida. In the Pseudomonas sp. HR 199 strain, these ORF's code for a new not previously described eugenol hydroxylase which converts eugenol into coniferyl alcohol via a quinone methide derivative by a process analogous to the reaction mechanism of p-cresol methyl hydroxylase. Another ORF of an unknown function was identified between the genes of the two subunits of eugenol hydroxylase. An ORF which was homologous to lignostilbene-a,b-dioxygenase was identified in a 5.0 kbp HindIII subfragment (H50). In addition one ORF was identified which was homologous to alcohol dehydrogenases. The structural gene vdh of vanillin dehydrogenase was identified in a 3.8 kbp HindIII/EcoRI subfragment. Upstream of this gene an ORF was localised which was homologous to enoyl-CoA hydratases produced by various organisms.

[0029] The complementing capacity of plasmid pE5-1 was attributed to the joint obtainment of the 1.2 and 1.8 kbp EcoRI fragments (E12 and E18). Fragment E 12 was completely, and fragment E 18 partially, sequenced. The structural gene cadh of coniferyl alcohol dehydrogenase, which contained an EcoRI cleavage site, was localised in these fragments. Using chromatographic methods the enzyme was isolated from the soluble fraction of the crude extract of cells of Pseudomonas sp. HR 199 grown on eugenol. An oligonucleotide sequence was deduced from the specific N-terminal amino acid sequence. A corresponding DNA probe hybridised with fragment E12, in which the region of the cadh gene encoding the N-terminus was localised.

[0030] A eugenol- and ferulic acid-negative mutant (mutant 6167) was complemented by obtaining a 9.4 kbp EcoRI fragment (E 94) of the hybrid cosmid pE5-1. A physical map of this fragment was prepared. The complementing property was localised in a 1.9 kbp EcoRI/HindIII subfragment. This fragment had incomplete ORF's (they extended beyond the EcoRI and HindIII cleavage sites) which were homologous to acetyl-CoA acetyl transferases of various organisms and to the “medium-chain acyl-CoA synthetase” produced by Pseudomonas oleovorans. Fragment E 94 was completely sequenced. Downstream of the aforementioned ORF's an ORF was located which was homologous to &bgr;-ketothiolases. The structural gene of coniferylaldehyde dehydrogenase (caldh) was localised in a central position of fragment E 94. Using chromatographic methods the enzyme was isolated from the soluble fraction of the crude extract of cells of Pseudomonas sp. HR 199 grown on eugenol.

[0031] The conjugative transfer of hybrid cosmid pE207 into a large number of Pseudomonas strains resulted in the heterologous expression of the van A, van B and vdh genes and the eugenol-hydroxylase genes in the transconjugants obtained. The obtainment of the plasmid of one strain allowed it to grow using eugenol as its carbon and energy source.

[0032] Materials and Methods

[0033] Growth conditions of the bacteria. Strains of Escherichia coli were grown at 37° C. in a Luria-Bertani (LB) or M9 mineral medium (Sambrook, J. E. F. Fritsch and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2nd Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.). Strains of Pseudomonas sp. and Alcaligenes eutrophus were grown at 30° C. in a nutrient broth (NB, 0.8% by weight) or in a mineral medium NM) (Schlegel, H. G. et al. 1961. Arch. Mikrobiol. 38: 209-222). Ferulic acid, vanillin, vanillic acid and protocatechuic acid were dissolved in dimethyl sulphoxide and added to the respective medium in a final concentration of 0.1% by weight. Eugenol was added to the medium directly in a final concentration of 0.1 vol.-%, or applied on filter paper (circular filters 595, Schleicher & Schuell, Dassel, Germany) to the lids of MM agar plates. For the growth of transconjugants of Pseudomonas sp., tetracyline and kanamycin were used in final concentrations of 25 &mgr;g/ml and 300 &mgr;g/ml, respectively.

[0034] Nitrosoguanidine mutagenesis. The nitrosoguanidine mutagenesis of Pseudomonas sp. HR 199 was carried out using a modified method according to Miller (Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.). Instead of the citrate buffer, a potassium phosphate (PP) buffer (100 mM, pH 7.0) was used. The final concentration of N-methyl-N′-nitro-N-nitrosoguanidine was 200 &mgr;g/ml. The mutants obtained were screened with regard to the loss of their capacity to utilise eugenol, ferulic acid, vanillin and vanillic acid as growth substrates.

[0035] Qualitative and quantitative detection of metabolic intermediates in culture supernatants. Culture supernatants were analysed by high-pressure liquid chromatography (Knauer HPLC) either directly or after dilution with twice-distilled water. Chromatography was carried out on Nucleosil-100 C18 (7 &mgr;m, 250×4 mm). The solvent used was 0.1 vol.-% formic acid and acetonitrile.

[0036] Purification of coniferyl alcohol dehydrogenase and coniferylaldehyde dehydrogenase. The purification processes were carried out at 4° C.

[0037] Crude extract. Cells of Pseudomonas sp. HR 199 grown on eugenol were washed in a 10 mM sodium phosphate buffer with a pH of 7.5, resuspended in the same buffer and disrupted by being passed through a French press (Amicon, Silver Spring, Md., USA) twice at a pressure of 1,000 psi. The cell homogenate was subjected to ultracentrifugation (1 h, 100,000×g, 4° C.), the soluble fraction of the crude extract being obtained as the supernatant.

[0038] Anion exchange chromatography on DEAE Sephacel. The soluble fraction of the crude extract was dialysed overnight against a 10 mM sodium phosphate buffer with a pH of 7.5 containing 100 mM NaCl. The dialysate was applied to a DEAE Sephacel column (2.6 cm×35 cm, bed volumn [BV]: 186 ml) equilibrated with a 10 mM sodium phosphate buffer of a pH of 7.5 containing 100 mM NaCl at a flow rate of 0.8 ml/min. The column was washed with two bed volumes of a 10 mM sodium phosphate buffer with a pH of 7.5 containing 100 mM NaCl. The elution of coniferyl alcohol dehydrogenase (CADH) and coniferylaldehyde dehydrogenase (CALDH) was carried out with a linear salt gradient of 100 to 500 mM NaCl in a 10 mM sodium phosphate buffer with a pH of 7.5 (2×150 ml). 5 ml fractions were collected. Fractions with high CADH and CALDH activities were combined in the corresponding DEAE pools respectively.

[0039] Gel filtration chromatography on Sephadex G200. The CADH DEAE pool was concentrated in a 50 ml Amicon ultrafiltration chamber via a Diaflo ultrafiltration membrane PM 30 (both from AMICON CORP., Lexington, USA) at a pressure of 290 kPa to a volume corresponding to approx. 2% of the Sephadex G200-BV. The concentrated protein solution was applied to a Sephadex G200 column (BV: 138 ml) equilibrated with a 10 mM sodium phosphate buffer with a pH of 7.5 containing 100 mM NaCl and eluted with the same buffer at a flow rate of 0.2 ml/min. 2 ml fractions were collected. Fractions with a high CADH activity were combined in the Sephadex G200 pool.

[0040] Hydrophobic interaction chromatography on butyl Sepharose 4B. The CADH Sephadex G200 pool was adjusted to 3 M NaCl and then applied to a butyl Sepharose 4B column (BV: 48 ml) equilibrated with a 10 mM sodium phosphate buffer with a pH of 7.5 containing 3 M NaCl (flow rate: 0.5 ml/min). The column was then washed with 2 BV of a 10 mM sodium phosphate buffer with a pH of 7.5 containing 3 M NaCl (flow rate: 1.0 ml/min). CADH was eluted with a linearly decreasing NaCl gradient of 3 to 0 M NaCl in a 10 mM sodium phosphate buffer with a pH of 7.5 (2×50 ml). 4 ml fractions were collected. Fractions with a high CADH activity were combined in the HIC pool and concentrated as described above.

[0041] Chromatography on hydroxyapatite. The CALDH DEAE pool was concentrated to 10 ml in a 50 ml Amicon ultrafiltration chamber via a Diaflo ultrafiltration membrane PM 30 (both from AMICON CORP., Lexington, USA) at a pressure of 290 kPa. The concentrated protein solution was applied to a hydroxyapatite column (BV: 80 ml) equilibrated with a buffer (10 mM NaCL in a 10 mM sodium phosphate buffer with a pH of 7.0) (flow rate: 2 ml/min). The column was then washed with 2.5 bed volumes of a buffer (flow rate: 2 ml/min). CALDH was eluted with a linearly increasing sodium phosphate gradient of 10 to 400 mM NaP (in each case containing 10 mM NaCL) (2×100 ml). 10 ml fractions were collected. Fractions with high CALDH activity were combined in the CALDH HA pool.

[0042] Gel filtration chromatography on Superdex HR 200 10/30. The CALDH HA pool was concentrated to 200 &mgr;l (Amicon ultrafiltration chamber, ultrafiltration membrane PM 30) and applied to a Superdex HR 200 10/30 column (13V: 23.6 ml) equilibrated with a 10 mM sodium phosphate buffer with a pH of 7.0. CALDH was eluted with the same buffer at a flow rate of 0.5 ml/min. 250 &mgr;l fractions were collected. Fractions with high CALDH activity were combined in the CALDH Superdex pool.

[0043] Determination of coniferyl alcohol dehydrogenase activity. The CADH activity was determined at 30° C. by means of an optical enzymatic test according to Jaeger et al. (Jaeger, E., L. Eggeling and H. Sahm. 1982. Current Microbiology. 6: 333-336) with the aid of a ZEISS PM 4 spectrophotometer fitted with a TE converter (both from ZEISS, Oberkochen, Germany) and a recorder. The reaction mixture with a volume of 1 ml contained 0.2 mmol of Tris/HCl (pH 9.0), 0.4 &mgr;mol of coniferyl alcohol, 2 &mgr;mol of NAD, 0.1 mmol of semicarbazide and a solution of the enzyme (“Tris”=tris(hydroxymethyl)-aminomethane). The reduction of NAD was monitored at 1=340 nm (e=6,3 cm2/&mgr;mol). The enzyme activity was recorded in units (U), 1 U corresponding to that quantity of enzyme which metabolises 1 &mgr;mol of substrate per minute. The protein concentrations in the samples were determined according to the method described -by Lowry et al. (Lowry, O. H., N. J. Rosebrough, A. L. Farr and R. J. Randall. 1951. J. Biol. Chem. 193: 265-275).

[0044] Determination of the coniferylaldehyde dehydrogenase activity. The CALDH activity was determined at 30° C. by an optical enzymatic test with the aid of a ZEISS PM 4 spectrophotometer fitted with a TE converter (both from ZEISS, Oberkochen, Germany) and a recorder. The reaction mixture of a volume of 1 ml contained a 10 mM Tris/HCl buffer (pH 8.8), 5.6 mM coniferylaldehyde, 3 mM NAD and a solution of the enzyme. The oxidation of coniferylaldehyde to form ferulic acid was monitored at 1=400 nm (e=34 cm2/&mgr;mol). The enzyme activity was recorded in units (U), 1 U corresponding to that quantity of enzyme which metabolises 1 &mgr;mol of substrate per minute. The protein concentration in the samples was determined according to the method described by Lowry et al. (Lowry, O. H., N. J. Rosebrough, A. L. Farr and R. J. Randall. 1951. J. Biol. Chem. 193: 265-275).

[0045] Electrophoretic methods. The separation of protein-containing extracts was carried out in 7.4% by weight polyacrylamide gels under native conditions according to the method described by Stegemann et al. (Stegemann et al. 1973. Z. Naturforsch. 28c: 722-732) and under denaturing conditions in 11.5% by weight polyacrylamide gels according to the method described by Laemmli (Laemmli, U.K. 1970. Nature (London) 227: 680-685). Serva Blue R was used for non-specific protein staining. For specifically staining coniferyl alcohol, coniferylaldehyde and vanillin dehydrogenase the gels were placed for 20 mins in a new 100 mM PP buffer (pH 7.0) and then incubated at 30° C. in the same buffer, to which 0.08% by weight of NAD, 0.04% by weight of p-nitroblue-tetrazolium chloride, 0.003% by weight of phenazine methosulphate and 1 mM of the respective substrate had been added, until the corresponding coloured bands appeared.

[0046] The transfer of proteins from polyacrylamide gels to PVDF membranes. Proteins were transferred from SDS polyacrylamide gels to PVDF membranes (Waters-Milipore, Bedford, Mass., USA) with the aid of a semidry fast blot device (B32/33 from Biometra, Göttingen, Germany) according to the manufacturer's instructions.

[0047] Determination of N-terminal amino acid sequences. The determination of N-terminal amino acid sequences was carried out with the aid of a protein peptide sequencer (of type 477 A, Applied Biosystems, Foster City, USA) and a PTH analyser, according to the manufacturer's instructions.

[0048] Isolation and manipulation of DNA. The isolation of genomic DNA was carried out by the method described by Marmur (Marmur, J. 1961. Mol. Biol. 3: 208-218). Megaplasmid DNA was isolated according to the method described by Nies et al. (Nies, D., et al. 1987. J. Bacteriol. 169: 4865-4848). The isolation and analysis of other plasmid DNA or DNA restriction fragments, the packaging of hybrid cosmids in 1-phage particles and the transduction of E. coli. was carried out by standard methods (Sambrook, J. E. F. Fritsch and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2nd Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.).

[0049] Transfer of DNA. The preparation and transformation of competent Escherichia coli cells was carried out by the method described by Hanahan (Hanahan, D. 1983. J. Mol. Biol. 166: 557-580). Conjugative plasmid transfer between plasmid-containing Escherichia coli S17-1 strains (donor) and Pseudomonas sp. strains (recipient) and Alcaligenes eutrophus (recipient) was carried out on NB agar plates according to the method described by Friedrich et al. (Friedrich, B. et al. 1981. J. Bacteriol. 147: 198-205) or by a “minicomplementation method” on MM agar plates using 0.5% by weight of gluconate as the carbon source and 25 &mgr;g/ml of tetracylin or 300 &mgr;g/ml of kanamycin. In this process cells of the recipient were applied in one direction in the form of an inoculation line. After 5 minutes cells of the donor strains were then applied in the form of inoculation lines crossing the recipient inoculation line. After incubation for 48 h at 30° C. the transconjugants grew directly downstream of the crossing point, whereas neither the donor nor the recipient strain was capable of growth.

[0050] Hybridisation experiments. DNA restriction fragments were electrophoretically separated in an 0.8% by weight agarose gel in a 50 mM Tris, 50 mM boric acid and 1.25 mM EDTA buffer (pH 8.5) (Sambrook, J. E. F. Fritsch and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2nd Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.). The transfer of the denatured DNA from the gel to a positively charged nylon membrane (pore size: 0.45 &mgr;m, Pall Filtrationstechnik, Dreieich, Germany), the subsequent hybridisation with biotinylated or 32P-labelled DNA probes and the production of these DNA probes was carried out according to standard methods (Sambrook, J. E. F. Fritsch and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2nd Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.).

[0051] The synthesis of oligonucleotides. Using desoxynucleoside phosphoramidites as the starting material, oligonucleotides were synthesised on a 0.2 &mgr;mol scale (Beaucage, S. L., and M. H. Caruthers. 1981. Tetrahedron Lett. 22: 1859-1862). The synthesis was carried out in a Gene Assembler Plus according to the manufacturer's instructions (Pharmacia-LKB, Uppsala, Sweden). The elimination of the protecting groups was carried out by incubation for 15 h at 55° C. in a 25 vol.-% aqueous ammonia solution. The oligonucleotides were finally purified in an NAP-5 column (Pharmacia-LKB, Uppsala, Sweden).

[0052] DNA sequencing. The determination of nucleotide sequences was carried out by the didesoxy chain termination method described by Sanger et al. (Sanger et al. 1977. Proc. Natl. Acad. Sci. USA 74: 5463-5467) using [&agr;-35S]dATP and a T7 polymerase sequencing kit (Pharmacia-LKB). 7-Deazaguanosine-5′-triphosphate was used instead of dGTP (Mizusawa, S. et al. 1986. Nucleic Acids Res.14: 1319-1324). The products of the sequencing reactions were separated in a 6% by weight polyacrylamide gel in a 100 mM Tris/HCl, 83 mM boric acid and 1 mM EDTA buffer (pH 8.3) containing 42% by weight urea, an S2 sequencing apparatus (GIBCO/BRL, Bethesda Research Laboratories GmbH, Eggenstein, Germany) being used according to the manufacturer's instructions. After electrophoresis the gels were incubated for 30 mins in 10 vol.-% acetic acid and, after washing briefly in water, dried for 2 hours at 80° C. Kodak X-OMAT AR X-ray films (Eastman Kodak Company, Rochester, N.Y., USA) were used for the autoradiography of the dried gels. In addition DNA sequences were also determined “non-radioactively” with the aid of an “LI-COR DNA Sequencer Model 4000L” (LI-COR Inc., Biotechnology Division, Lincoln, Nebr., USA) using a “Thermo Sequenase fluorescent labelled primer cycle sequencing kit with 7-deaza-dGTP” (Amersham Life Science, Amersham International plc, Little Chalfont, Buckinghamshire, England), in each case according to the manufacturer's instructions.

[0053] Various sequencing strategies were used: With the aid of synthetic oligonucleotides sequencing was carried out by the “Primer-hopping Strategy” described by Strauss et al. (Strauss, E. C. et al. 1986. Anal. Biochem. 154: 353-360). If only “universal” and “reverse primers” were used hybrid plasmids were used as “template DNA”, the inserted DNA fragments of which had been unidirectionally shortened with the aid of an “Exo III/Mung Bean Nuclease Deletion” kit (Stratagene Cloning Systems, La Jolla, Calif., USA) according to the manufacturer's instructions.

[0054] Chemicals, biochemicals and enzymes: Restriction enzymes, T4 DNA ligase, lambda DNA and enzymes and substrates for the optical enzymatic tests were obtained from C. F. Boehringer & Söhne (Mannheim, Germany) or from GIBCO/BRL (Eggenstein, Germany). [a-35S]dATP and [g-32P]ATP were obtained from Amersham/Buchler (Braunschweig, Germany). NA-type agarose was obtained from Pharmacia-LKB (Uppsala, Sweden). All the other chemicals were from Haarmann & Reimer (Holzminden, Germany), E. Merck AG (Darmstadt, Germany), Fluka Chemic (Buchs, Switzerland), Serva Feinbiochemica (Heidelberg, Germany) or Sigma Chemie (Deisenhofen, Germany).

EXAMPLES

Example 1

[0055] The Isolation of Mutants of the Pseudomonas sp. HR 199 Strain with Defects in the Catabolism of Eugenol

[0056] The Pseudomonas sp. HR 199 strain was subjected to nitrosoguanidine mutagenesis in order to isolate mutants with defects in the catabolism of eugenol. The mutants obtained were classified according to their capacity to utilise eugenol, ferulic acid and vanillin as their carbon and energy source. Mutants 6164 and 6165 were no longer capable of utilising eugenol as a carbon and energy source, although, as in the case of the wild type, they were capable of utilising ferulic acid and vanillin. Mutants 6167 and 6202 were no longer capable of utilising eugenol and ferulic acid as their carbon and energy source, although, as in the case of the wild type, they were capable of utilising vanillin. The abovementioned mutants were used in the subsequent molecular-biological analyses.

Example 2

[0057] Construction of a Pseudomonas sp. HR 199 Gene Library in the Cosmid Vector

[0058] The genomic DNA of the Pseudomonas sp. HR 199 strain was isolated and subjected to partial restriction digestion with EcoRI. The DNA preparation thus obtained was ligated with vector pVK100 cut by EcoRI. The DNA concentrations were relatively high in order to accelerate the formation of concatemeric ligation products. The ligation materials were packaged in 1-phage particles which were subsequently used for transduction of E. coli S17-1. The selection of the transductants was carried out on tetracycline-containing LB agar plates. In this manner 1330 transductants were obtained which contained various hybrid cosmids.

Example 3

[0059] The Identification of Hybrid Cosmids Containing Essential Genes of Eugenol Catabolism

[0060] The hybrid cosmids of the 1330 transductants were transferred conjugatively to mutants 6164 and 6165 by a minicomplementation process. The resulting transconjugants were examined on MM plates containing eugenol for their capacity to grow again on eugenol (complementation of the respective mutant). Mutant 6164 was complemented by the obtainment of hybrid cosmid pE5-1, which contained a 1.2 kbp, a 1.8 kbp, a 3 kbp, a 5.8 kbp and a 9.4 kbp EcoRI fragment in cloned form. The E. coli S17-1 strain containing this hybrid cosmid was deposited at the “Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH” (DSM) under the number DSM 10440. Mutant 6165 was complemented by the obtainment of the hybrid cosmids pE207 or pE115 respectively. The complementing capacity was attributed to a 23 kbp EcoRI fragment which was contained in cloned form in the hybrid cosmid pE207 as the only EcoRI fragment, whereas hybrid cosmid pE115 additionally contained a 3 kbp and a 6 kbp EcoRI fragment. The E. coli S17-1 strain containing hybrid cosmid pE207 was deposited at the DSM under the number DSM 10439.

Example 4

[0061] The Analysis of the 23 kbp EcoRI Fragment (E230) of the Hybrid Cosmid pE207

[0062] Fragment E230 was isolated preparatively from EcoRI-digested hybrid cosmid pE207 and ligated to pBluescript SK−-DNA digested with EcoRI. Using the ligation material E. coli XL1-Blue was transformed. Following “blue-white” selection on LB-Tc-Amp agar plates containing X-Gal and IPTG, “white” transformants were obtained whose hybrid plasmids pSKE230 contained the fragment E230 in cloned form. With the aid of this plasmid and by using various restriction enzymes a physical map of the fragment E230 was prepared (FIG. 1).

[0063] By cloning subfragments of E230 in vectors pVK101 and pMP92, both of which have a broad host specturm and are also stable in pseudomonads, followed by conjugative transfer into mutant 6165, the region complementing mutant 6165 was localised in a 1.8 kbp KpnI fragment (K18). After cloning this fragment in pBluescript SK− the nucleotide sequence was determined, the gene of the cytochrome C subunit of eugenol hydroxylase being identified. The gene product of 117 amino acids had an N-terminal leader peptide (MMNVNYKAVGAS-LLLAFISQGAWA) and 32.9% identity (via a region of 82 amino acids) with the cytochrome C subunit of p-cresol methylhydroxylase produced by Pseudomonas putida (McIntire et al. 1986. Biochemistry 25:5975-5981).

[0064] By cloning the KpnI subfragments of E230 adjacent to K18 in pBluescript SK− and sequencing, additional open reading frames (ORF) were identified, one of which codes for the flavoprotein subunit of eugenol hydroxylase and was highly homologous to the flavoprotein subunit of p-cresol methylhydroxylase produced by Pseudomonas putida. An additional ORF was found to be highly homologous to g-glutamyl cysteine synthetase (the first enzyme in the biosynthesis of glutathione) produced by Escherichia coli (Watanabe et al. 1986. Nucleic Acids Res. 14: 4393-4400).

[0065] In the soluble fraction of the crude extract of E. coli (pSKE230) vanillin dehydrogenase was detected by specific activity staining in a polyacrylamide gel. By subcloning in pBluescript SK− and analysis of soluble fractions of the crude extracts of the transformants obtained, the vanillin dehydrogenase gene (vdh) was localised in a 3.8 kbp HindIII/EcoRI subfragment of E230. The complete nucleotide sequence of this fragment was determined. The molecular weight of the vanillin dehydrogenase was 50,779, as confirmed by SDS polyacrylamide gel electrophoresis. The amino acid sequence was highly homologous to other aldehyde dehydrogenases of various origins.

[0066] Upstream of the vdh gene an additional ORF was identified which was homologous to enoyl-CoA hydratases. The calculated molecular weight of 27,297 was confirmed by SDS polyacrylamide gel electrophoresis.

[0067] By sequencing the 5.0 kbp HindIII subfragment of E230, which had also been cloned in pBluescript SK−, an ORF was identified which was highly homologous to the lignostilbene-a,b-dioxygenase produced by Pseudomonas paucimobilis. By complete sequencing of the fragment E230 two additional ORF's were identified which were homologous to formaldehyde-dehydrogenases (fdh) and alcohol dehydrogenases (adh) (cf. FIG. 1).

Example 5

[0068] The Analysis of the Region of Hybrid Cosmid pE5-1 Complementing Mutant 6164

[0069] Mutant 6164 was complemented by the obtainment of hybrid cosmid pE5-1 which contained a 1.2 kbp (E12), a 1.8 kbp (E18), a 3 kbp (E30), a 5.8 kbp (E58) and a 9.4 kbp (E94) EcoRI fragment in cloned form (FIG. 1). By digesting pE5-1 with EcoRI and subsequent religation a derivative (pE106) of this hybrid cosmid was obtained which only contained fragments E12, E18 and E30. Following conjugative transfer into mutant 6164 this plasmid was however capable of complementing the latter, as a result of which corresponding transconjugants were once again capable of growing on eugenol as a carbon and energy source.

[0070] After digesting plasmid pE106 with EcoRI, gel-electrophoretic separation of the digestion material in a 0.8% by weight agarose gel and transfer of the DNA to a nylon membrane, hybridisation was carried out with a 32P-labelled oligonucleotide probe of the following sequence: 1 ATG CAA CTC ACC AAC AAA AAA ATC GT-3′      G   G   C   T   G   G   T     G   G   C       G   G     G   T   G       G   G             G       G   G             T       G   G

[0071] The sequence of this gene probe had been deduced from the N-terminal amino acid sequence of coniferyl alcohol dehydrogenase (CADH) (see below) purified from Pseudomonas sp. HR 199. With the aid of this probe the region of the cadh gene encoding the N-terminus of the CADH was localised in fragment E12. This fragment and parts of the adjacent fragment E 18 were also sequenced and the complete sequence of the cadh gene thus determined. The amino acid sequence deduced from cadh was homologous to other alcohol dehydrogenases of class I, group II (according to Matthew and Fewson. 1994. Critical Rev. Microbiol. 20(1): 13-56).

Example 6

[0072] Purification and Characterisation of Coniferyl Alcohol Dehydrogenase

[0073] Pseudomonas sp. HR 199 was grown on eugenol. The cells were harvested, washed and disrupted with the aid of a French press. The soluble fraction of the crude extract obtained after ultracentrifugation had a specific activity of 0.24 U/mg of protein. By means of chromatography on DEAE Sephacel an 11.7-fold enrichment of CADH was obtained in a yield of 83.7%. By means of chromatography on Sephadex G200 a 6.8-fold enrichment of CADH was obtained in a yield of 11.2%. By means of chromatography on butyl Sepharose 4B a 70.6-fold enrichment of CADH was obtained in a yield of 7.8%.

[0074] With the aid of this method a preparation was obtained which displayed a band at 27 kDa according to SDS polyacrylamide gel electrophoresis. The purification factor was 64 and the yield 0.8%.

[0075] Optimum Temperature and Thermal Stability

[0076] The optimum temperature for the reaction catalysed with CADH was 42° C. The enzyme was however sensitive to heat. The half-lives were as follows: T1/2 (34° C.)=5 mins, T1/2 (39° C.)=1 min, T1/2 (42° C.)<1 min.

[0077] Optimum pH

[0078] The optimum pH for the reaction catalysed by CADH was 10.9 in a 25 mM MOPS buffer. At higher pH values a decrease in activity due to denaturation was observed.

[0079] Apparent Molecular Weight

[0080] The molecular weight of native CADH was determined with the aid of FPLC by gel filtration on Superdex 200HR 10/30 at 54.9 kDa, which suggests a a2 subunit structure.

[0081] N-Terminal Amino Acid Sequence

[0082] The determination of the N-terminal amino acid sequence of the purified protein revealed the following result: 2 1       5         10            15               20 M Q L T N K K I V V V (G) V (S) ? (R) (I) ? (A) (E)                       (V)   (V)

[0083] (Sequence in the single letter code; ?: definition not possible; ( ): not certain; in the second row an amino acid is mentioned which may also apply)

Example 7

[0084] Purification and Characterisation of Coniferylaldehyde Dehydrogenase

[0085] Pseudomonas sp. HR 199 was grown on eugenol. The cells were harvested, washed and disrupted with the aid of a French press. The soluble fraction of the crude extract obtained after ultracentrifugation displayed a specific activity of 0.43 U/mg protein. By chromatography on DEAE Sephacel a 6.6-fold enrichment of CALDH was obtained in a yield of 65.3%. By chromatography on hydroxy-apatite a 63-fold enrichment of CALDH was obtained in a yield of 33%. By chromatography on Superdex HR 200 an 81-fold enrichment of CALDH was obtained in a yield of 13%. With the aid of this method a preparation was obtained which, according to SDS polyacryamide gel electrophoresis, displayed a band at approx. 49 kDa.

[0086] Optimum Temperature and Thermal Stability

[0087] The optimum temperature of the reaction catalysed by CALDH was 26° C. The enzyme was sensitive to heat. The half-lives were as follows: T1/2 (31° C.)=5 mins, T1/2 (34° C.)=2.5 mins, T1/2 (38° C.)=1 min.

[0088] Optimum pH

[0089] The optimum pH for the reaction catalysed by CALDH was 8.8 in a 100 mM Tris/HCl buffer. At this pH value the enzyme is however already unstable (87 % decrease in activity within 5 mins). At lower pH values the enzyme is more stable (e.g. pH 6.0: 50% decrease in activity within 4 hours).

[0090] Substrate Specificity

[0091] The enzyme not only accepts coniferylaldehyde (100%) but also trans-cinnamaldehyde (96.7%), sinapyl aldehyde (76.7%), p-anisaldehyde (23.1%), benzaldehyde (17.8%), 3,5-dimethoxy-benzaldehyde (7.6%) and 3-hydroxy-benzaldehyde (1.7%) as substrates.

[0092] The KM value of CALDH for coniferylaldehyde is in the range between 0.007 and 0.012 mM at a Vmax of approx. 9 to 15 U/ml. The KM value of CALDH for NAD is 0.334 mM at a Vmax of 14.2 U/ml. Compared with NAD, NADP is accepted at a rate of 4.3%.

[0093] N-Terminal Amino Acid Sequence

[0094] The determination of the N-terminal amino acid sequence of the purified protein revealed the following result: 3 1 S I L G L N G A P V G A E Q L G S A L (D) 20

[0095] (seqence in the one-letter code; ( ): not certain).

Example 8

[0096] Localisation and Sequencing of the Coniferylaldehyde Dehydrogenase Gene (caldh)

[0097] The N-terminal amino acid sequence was definitively assigned to an amino acid sequence deduced from the DNA sequence of fragment E94 of plasmid pE5-1. Thus the CALDH structural gene caldh is localised in E94. The amino acid sequence deduced from caldh was homologous to other aldehyde dehydrogenases.

Example 9

[0098] The Complementation of other Mutants Displaying Defects in the Catabolism of Eugenol using Hybrid Cosmids pE207 and pE5-1

[0099] Following NMG mutagenesis, mutants 6167 and 6202 had been obtained which were no longer capable of utilising eugenol and ferulic acid as their carbon and energy source (see above). The obtainment of plasmid pE207 meant that, after conjugative transfer, mutant 6202 was once again capable of utilising the aforementioned substrates. This mutant is complemented by the gene homologous to enoyl-CoA hydratase.

[0100] The obtainment of plasmid pE5-1 meant that, after conjugative transfer, mutant 6167 was once again capable of utilising the abovementioned substrates. By individually cloning the EcoRI fragments of pE5-1 in pHP 1014 and the conjugative transfer of these plasmids into mutant 6167 the complementing property was localised in fragment E94. A physical map of fragment E94 was prepared after cloning in pBluescript SK− and digestion with various restriction enzymes. By cloning subfragments of E94 in the vectors pVK101 and pMP92, followed by conjugative transfer into mutant 6167, the region complementing mutant 6167 was localised in a 1.9 kbp EcoRI/HindIII fragment (EH19). After cloning this fragment in pBluescript SK− and sequencing, 2 ORF's were identified which were homologous to acetyl-CoA acetyltransferases and to “medium-chain acyl-CoA synthetase” produced by Pseudomonas oleovorans. By completely sequencing fragment E94, additional ORF's were identified which were homo-logous to regulator proteins and a chemotaxis protein (cf. FIG. 1).

Example 10

[0101] Determination of the Chromosomal Coding of the Genes for the Catabolism of Eugenol in Pseudomonas sp. HR 199

[0102] Since Pseudomonas sp. HR 199 has a megaplasmid of a size of approx. 350 kbp, a hybridisation experiment was carried out to examine whether the genes for the catabolism of eugenol were localised in this megaplasmid or in the chromosome. For this purpose megaplasmid preparations of the wild type and of the mutants were separated in an 0.8% by weight agarose gel. The chromosomal and megaplasmid DNA was blotted onto a nylon membrane and then hybridised against a biotinylated HE38 DNA probe. A hybridisation signal was only obtained with the chromosomal DNA and not with the megaplasmid DNA. Thus the genes for the catabolism of eugenol in Pseudomonas sp. HR 199 are coded in the chromosome.

Example 11

[0103] The Heterologous Expression of Genes for the Catabolism of Eugenol from Pseudomonas sp. HR 199 in other Pseudomonas Strains and in Alcaligenes eutrophus.

[0104] The plasmid pE207 and a pVK101 hybrid plasmid containing fragment H110 (pVKH110) were conjugatively transferred to A. eutrophus and into Pseudomonas strains which were not capable of metabolising eugenol, vanillin or vanillic acid. The transconjugants obtained were not only examined for their capacity to grow on MM agar plates containing eugenol, vanillin or vanillic acid but also some transconjugants were incubated with eugenol in an MM liquid medium. By means of HPLC analysis of the culture supernatants some of the transconjugants were found to metabolise eugenol.

[0105] In this analysis the functional expression of the vdh gene in transconjugants of P. stutzeri, P. asplenii, Pseudomonas sp. DSM13, Pseudomonas sp. DSM15a and Pseudomonas sp. D1 was determined.

[0106] Transconjugants of the strain Pseudomonas sp. D1, which contained the plasmid pE207, were capable of growing using eugenol as their carbon and energy source. In corresponding transconjugants of P. testosteroni LMD3324, P. fluorescens TypeB, P. stutzeri DSM 50027, Pseudomonas sp. DSM 1455 and P. fragi DSM3456 functional expression of the eugenol hydroxylase genes was also observed which resulted in the secretion of intermediates of the catabolism of eugenol (coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin, vanillic acid) into the culture medium. Growth of these transconjugants on eugenol was however not observed.

Claims

1. Synthetic enzymes for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid from eugenol.

2. Synthetic enzymes according to claim 1 selected from the group comprising:

a) eugenol hydroxylase,

b) coniferyl alcohol dehydrogenase,

c) coniferylaldehyde dehydrogenase,

d) ferulic acid deacylase and

e) vanillin dehydrogenase.

3. DNA coding for the enzymes according to claims 1 and 2 as well as partial sequences and functional equivalents thereof.

4. Cosmid clones containing the DNA according to claim 3.

5. Vectors containing the DNA according to claim 3.

6. Microorganisms transformed with the DNA according to claim 3.

7. The use of the DNA according to claim 3 for the transformation of microorganisms.

8. The use of microorganisms according to claim 6 for the production of coniferyl alcohol, coniferylaldehyde, ferulic acid, vanillin and vanillic acid.

9. A process for the production of coniferyl alcohol from eugenol, characterised in that the reaction is carried out in the presence of eugenol hydroxylase.

10. A process for the production of coniferylaldehyde from coniferyl alcohol, characterised in that the reaction is carried out in the presence of coniferyl alcohol dehydrogenase.

11. A process for the production of ferulic acid from coniferylaldehyde, characterised in that the reaction is carried out in the presence of coniferylaldehyde dehydrogenase.

12. A process for the production of vanillin from ferulic acid, characterised in that the reaction is carried out in the presence of ferulic acid deacylase.

13. A process for the production of vanillic acid from vanillin, characterised in that the reaction is carried out in the presence of vanillin dehydrogenase.