Method for the production of 3-hydroxy-3-methylbutyric acid from acetone and an activated acetyl compound

Abstract

Described is a method for the production of 3-hydroxy-3-methylbutyric acid by enzyme-catalyzed covalent bond formation between the carbon atom of the oxo group of acetone and the methyl group of a compound which provides an activated acetyl group. Also described are recombinant organisms which produce 3-hydroxy-3-methylbutyric acid, and related compositions and methods.

Claims

1. A method of producing 3 hydroxy-3-methylbutyric acid comprising: (a) providing a recombinant microorganism, wherein the recombinant microorganism is genetically modified to overexpress a polynucleotide encoding a 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) synthase (EC 2.3.3.10) as compared to the same recombinant microorganism without the genetic modification, and wherein the recombinant microorganism is genetically modified to overexpress a polynucleotide encoding an acetoacetate decarboxylase as compared to the same recombinant microorganism without the genetic modification; and (b) incubating, in the recombinant microorganism (i) acetone, (ii) acetyl-coenzyme A, and (iii) the HMG-CoA synthase, to thereby produce 3-hydroxy-3-methylbutyric acid.

2. The method of claim 1, wherein the recombinant microorganism produces acetone.

3. A recombinant microorganism which: (a) produces acetone and acetyl-coenzyme A; (b) is genetically modified to overexpress a polynucleotide encoding a 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) synthase (EC 2.3.3.10) as compared to the same recombinant microorganism without the genetic modification; (c) is genetically modified to overexpress a polynucleotide encoding an acetoacetate decarboxylase as compared to the same recombinant microorganism without the genetic modification; and (d) produces 3-hydroxy-3-methylbutyric acid from acetone and acetyl-coenzyme A.

4. The recombinant microorganism of claim 3 which is derived from a microorganism which naturally produces acetone.

5. The recombinant microorganism of claim 3, wherein the recombinant microorganism is derived from a microorganism of the genus Clostridium, Bacillus or Pseudomonas.

6. The recombinant microorganism of claim 5, wherein the microorganism is selected from the species Clostridium acetobutylicum, Clostridium beijerinckii, Clostridium cellulolyticum, Bacillus polymyxa, and Pseudomonas putida.

7. The recombinant microorganism of claim 3, wherein the recombinant microorganism without the genetic modification to overexpress the polynucleotide encoding an acetoacetate decarboxylase does not produce acetone and the recombinant microorganism with the genetic modification to overexpress the polynucleotide encoding an acetoacetate decarboxylase produces acetone from acetoacetate by the acetoacetate decarboxylase.

8. The recombinant microorganism of claim 3, wherein the recombinant microorganism is derived from a microorganism that is capable of photosynthesis.

9. The recombinant microorganism of claim 3, wherein the recombinant microorganism is derived from a microorganism that does not naturally express HMG CoA synthase (EC 2.3.3.10).

10. The recombinant microorganism of claim 3, wherein the recombinant microorganism comprises a heterologous promoter driving the overexpression of the polynucleotide encoding the HMG CoA synthase (EC 2.3.3.10).

11. The method of claim 1, wherein the recombinant microorganism is derived from a microorganism that is capable of photosynthesis.

12. The method of claim 1, wherein the recombinant microorganism without the genetic modification to overexpress the polynucleotide encoding an acetoacetate decarboxylase does not produce acetone and the recombinant microorganism with the genetic modification to overexpress the polynucleotide encoding an acetoacetate decarboxylase produces acetone from acetoacetate by the acetoacetate decarboxylase.

13. The method of claim 1, wherein the recombinant microorganism is derived from a microorganism that does not naturally express HMG CoA synthase (EC 2.3.3.10).

14. The method of claim 1, wherein the recombinant microorganism comprises a heterologous promoter driving the overexpression of the polynucleotide encoding the HMG CoA synthase (EC 2.3.3.10).

15. The method of claim 2, wherein the recombinant microorganism is derived from a microorganism which naturally produces acetone.

16. The method of claim 2, wherein the recombinant microorganism is derived from a microorganism of the genus Clostridium, Bacillus or Pseudomonas.

17. The method of claim 16, wherein the microorganism is selected from the species Clostridium acetobutylicum, Clostridium beijerinckii, Clostridium cellulolyicum, Bacillus polymyxa, and Pseudomonas putida.

18. The method of claim 2, wherein the recombinant microorganism without the genetic modification to overexpress the polynucleotide encoding an acetoacetate decarboxylase does not produce acetone and the recombinant microorganism with the genetic modification to overexpress the polynucleotide encoding an acetoacetate decarboxylase produces acetone from acetoacetate by the acetoacetate decarboxylase.

19. The method of claim 2, wherein the recombinant microorganism is derived from a microorganism that is capable of photosynthesis.

20. The method of claim 2, wherein the recombinant microorganism is derived from a microorganism that does not naturally express HMG CoA synthase (EC 2.3.3.10).

21. The method of claim 2, wherein the recombinant microorganism comprises a heterologous promoter driving the overexpression of the polynucleotide encoding the HMG CoA synthase (EC 2.3.3.10).

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1: Chemical structure of 3-hydroxy-3-methylbutyric acid (also referred to as beta-hydroxyisovalerate)

(2) FIG. 2: Reaction scheme of the reaction catalysed by HMG-CoA synthase

(3) FIG. 3: Reaction scheme of the reaction catalysed by HMG-CoA lyase

(4) FIG. 4: Reaction schemes of the reactions of the pksX pathway including the reaction catalysed by the PksG protein

(5) FIG. 5: Reaction scheme of the reaction of the conversion of acetone and a compound containing an activated acetyl group into 3-hydroxy-3-methylbutyric acid X stands for SCH.sub.2CH.sub.2NHCOCH.sub.2CH.sub.2NHCOCH(OH)C(CH.sub.3).sub.2CH.sub.2OPO.sub.2HOPO.sub.2HC.sub.10H.sub.13N.sub.5O.sub.7P (coenzyme A), SCH.sub.2CH.sub.2NHCOCH.sub.2CH.sub.2NHCOCH(OH)C(CH.sub.3).sub.2CH.sub.2OPO.sub.2H-polypeptide (acyl-carrier protein), SCH.sub.2CH.sub.2NHCOCH.sub.2CH.sub.2NHCOCH(OH)C(CH.sub.3).sub.2CH.sub.2OH (pantetheine), SCH.sub.2CH.sub.2NHCOCH.sub.3 (N-acetyl-cysteamine), SCH.sub.3 (methane thiol), SCH.sub.2CH(NH.sub.2)CO.sub.2H (cysteine), SCH.sub.2CH.sub.2CH(NH.sub.2)CO.sub.2H (homocysteine), SCH.sub.2CH(NHC.sub.5H.sub.8NO.sub.3)CONHCH.sub.2CO.sub.2H (glutathione), SCH.sub.2CH.sub.2SO.sub.3H (coenzyme M) and OH (acetic acid).

(6) FIG. 6: Mass spectra of commercial available 3-hydroxy-3-methylbutyrate

(7) FIG. 7: Mass spectra of formation of 3-hydroxy-3-methylbutyrate from acetyl-CoA and acetone in the presence of Hmg-CoA synthase from Gallus gallus (P23228).

(8) FIG. 8: Mass spectra of the control assay without enzyme.

(9) FIG. 9: Michaelis-Menten plot for the reaction with the HMG CoA synthase of S. epidermidis described in Example 7

EXAMPLES

(10) The following examples serve to illustrate the invention.

Example 1

Bioinformatic Method Used to Create HMG-CoA Synthases and HMG-CoA Lyases Database

(11) A panel of 12 HMG-CoA synthases and 8 HMG-CoA lyases were selected to create a non-redundant set of proteins aiming to represent the diversity of these enzyme classes as found across eukaryotic organisms. These proteins were identified by performing multiple sequence-based and text-based searches on the Universal Protein Resource Database Uniprot (www.uniprot.org). They all contain unique features such as conserved protein domains and motifs characteristic to the enzyme class of interest. In order to effectively cover the sequence diversity without having to screen a large set of proteins, the initial pool of enzymes was narrowed down by grouping them into clusters of sequences with more than 85% homology and then selecting one single candidate sequence representative of each cluster. Protein sequence identity ranged from 30% to 80% and from 50% to 80% between any two proteins from the HMG-CoA synthases panel and the lyases panel respectively.

(12) The same approach was applied to select the HMG-CoA synthases and HMG-CoA lyases from prokaryotic organisms. The created set contained 50 proteins homologues to HMG-CoA synthases, including pksG proteins, and 59 proteins homologues to HMG-CoA lyases.

Example 2

Cloning, Expression and Purification of a Collection of HMG-CoA Lyases and HMG-CoA Synthases

(13) Gene Cloning:

(14) The nucleic acid sequences coding for HMG-CoA synthase and lyase from eukaryotic organism were optimized for E. coli codon preference and the genes were obtained by chemical synthesis (GeneArt, reagents).

(15) The genes encoding for HMG-CoA synthases and lyases from prokaryotic organisms were cloned from genomic DNA of different origins by routine recombinant techniques. These genes were then inserted in a His-tag containing pET 25b and pET 22b vectors (Novagen, Inc), respectively, for eukaryotic and prokaryotic organisms.

(16) Overexpression in E. coli:

(17) Plasmids are electroporated into E. coli BL21 bacteria (Novagen) that are then spread on an ampicillin containing LB-Agar Petri dish. The cultures are grown at 30 C. on TB medium, containing 0.5 M sorbitol, 5 mM betaine, 100 g/ml ampicillin under moderate shaking. When OD (600 nm) reached 0.8, IPTG is added to a final concentration of 1 mM, and expression is run for 16 hours at 20 C. under moderated shaking. The bacteria cells are then harvested by centrifugation at 4 C., 10.000 rpm, 20 minutes and frozen at 80 C.

(18) Cell Extract Preparation:

(19) Cell extracts are prepared by resuspending 1.6 g of cell pellet in 5 ml 50 mM Na.sub.2HPO.sub.4 buffer, containing 300 mM NaCl, 5 mM MgCl.sub.2, 1 mM DTT pH 8. 20 l lysonase (Novagen) is then added to the preparations, which are incubated for 10 min at room temperature and 20 min on ice. The cell lysis is achieved by triple sonication treatment of 5 minutes in ultrasonic water-bath on ice and homogenization of extract between each pulse. The crude extracts are then clarified by centrifugation at 4 C., 10.000 rpm, 20 minutes.

(20) Protein Purification:

(21) The clear supernatants are loaded onto the PROTINO-1000 Ni-IDA column (columns for the purification of proteins, Macherey-Nagel) which enables the specific immobilization of proteins carrying 6-histidine tails. The columns are washed and the enzymes are eluted with 4 ml 50 mM Na.sub.2HPO.sub.4 buffer, containing 300 mM NaCl, 5 mM MgCl.sub.2, 1 mM DTT, 250 mM imidazole pH 8. The enzyme containing fractions are then concentrated and desalted on Amicon Ultra-4 10 kDa filter unit (membranes for filtration, dialysis; Millipore) and resuspended in 250 l 40 mM Tris-HCl pH8, containing 0.5 mM DTT. The protein concentration is determined by the Bradford method.

(22) The homogeneity of purified enzymes varied from 20% to 75%.

Example 3

Measure of the HMG-CoA Synthase Activity Using Natural Substrates Acetoacetyl-CoA and Acetyl-CoA

(23) The HMG-CoA synthase activity is measured according to Clinkenbeard et al. (J. Biol. Chem. 250 (1975), 3108-3116). The standard assay medium mixture for HMG-CoA synthases contains 40 mM Tris-HCl pH 8, 1 mM MgCl.sub.2, 100 M acetoacetyl-CoA, 200 M acetyl-CoA, 0.5 mM DTT in a total volume of 1 ml. Mitochondria HMG-CoA synthases are assayed in the absence of MgCl.sub.2 to avoid the inhibition observed for this enzyme (Reed et al., J. Biol. Chem. 250 (1975), 3117-3123). Reaction is initiated by addition of 0.02 mg/mL enzyme.

(24) A Control assay was carried out in the absence of enzyme. HMG-CoA synthase activity was measured by monitoring the decrease in absorbance at 303 nm that accompanies the acetyl-CoA-dependent disappearance of the enolate form of acetoacetyl-CoA. To account for non-specific disappearance of acetoacetyl-CoA, results obtained in a control assay lacking enzyme were subtracted from results obtained in test samples. The apparent absorption coefficient for acetoacetyl-CoA under the assay conditions was 5600 M.sup.1.Math.cm.sup.1. One enzyme unit represented the disappearance in 1 min of 1 mol of acetoacetyl-CoA.

(25) TABLE-US-00003 TABLE 1 Physiological activity of some purified HMG-CoA synthases or enzymes homologous to HMG CoA synthases Physiological Uniprot activity Accession mol/min .Math. mg number Organism protein P54961 Blattella germanica (German 0.02 cockroach) P23228 Gallus gallus (Chicken) 0.02 Q01581 Homo sapiens (Human) 0.03 P54873 Arabidopsis thaliana 1.19 P54871 Caenorhabditis elegans 0.23 P54874 Schizosaccharomyces pombe 0.61 (Fission yeast) P54839 Saccharomyces cerevisiae (Baker's 0.28 yeast) P54872 Dictyostelium discoideum (Slime 0.09 mold) Q86HL5 Dictyostelium discoideum (Slime 0.02 mold) Q9M6U3 Brassica juncea 0.02 A5FM54 Flavobacterium johnsoniae 0.02 Q03WZ0 Leuconostoc mesenteroides 0.28 Q2NHU7 Methanosphaera stadtmanae 0.02 Q8CN06 Staphylococcus epidermidis 0.07 Q03QR0 Lactobacillus brevis 0.18 A6UPL1 Methanosarcina mazei 0.01 B2HGT6 Mycobacterium marinum 0.01 Q4L958 Staphylococcus haemolyticus 0.18 Q4A0D6 Staphylococcus saprophyticus 0.08 Q1GAH5 Lactobacillus delbrueckii 0.32

Example 4

Measuring of the HMG-CoA Lyase Activity Using Natural Substrate HMG-CoA

(26) HMG-CoA lyase activity is measured according to Mellanby J et al. (Methods of Enzymatic Analysis; Bergmeyer Ed. (1963), 454-458). The complete reaction mixture (1 ml) containing 40 mM Tris-HCl pH 8, 1 mM MgCl.sub.2, 0.5 mM DTT, 0.4 mM HMG-CoA, 0.2 mM NADH, 5 units of 3-hydroxybutyrate dehydrogenase is incubated for 5 min before adding 0.005 mg/ml of HMG-CoA lyase and then the progress of the reaction is monitored by the decrease in absorbance at 340 nm. A control assay was carried out in the absence of enzyme.

(27) To account for non-specific disappearance of NADH, results obtained in a control assay lacking enzyme were subtracted from results obtained in test samples. Specific activities were calculated as mol NADH/min.Math.mg protein.

(28) TABLE-US-00004 TABLE 2 Physiological activity of some purified HMG-CoA lyases Physiological Uniprot activity Accession mol/min .Math. mg number Organism protein A8WG57 Danio rerio (Zebrafish) 4.05 (Brachydanio rerio) Q29448 Bos taurus (Bovine) 5.79 B6U7B9 Zea mays 13.31 A5FHS2 Flavobacterium johnsoniae 2.89 A1VJH1 Polaromonas naphthalenivorans 34.92 A9IFQ7 Bordetella petrii 9.84 A9IR28 Bordetella petrii 1.74 A1VT25 Polaromonas naphthalenivorans 0.39

Example 5

3-hydroxy-3-methylbutyrate Production

(29) The complete reaction for 3-hydroxy-3-methylbutyrate synthesis contained 40 mM Tris-HCl pH 8, 5 to 50 mM acetyl-CoA, 100 to 500 mM acetone, 1 MgCl.sub.2 (except for mitochondria HMG-CoA synthase), 0.5 mM DTT and enzyme varying in the range from 0.2 to 8 mg/ml. Control reactions were carried in the absence of enzyme and one of the substrates.

(30) The progress of synthesis was followed by analyzing aliquots taken after increasing period of incubation at 30 or 37 C. Typically, an aliquot of 50 l was removed after 48 h of incubation, heated for 1 min at 100 C. to eliminate the proteins, centrifuged and the supernatant was transferred to a clean vial for HIV detection by mass spectrometry. A solution of 3-hydroxy-3-methylbutyrate was prepared in 40 mM Tris-HCl pH 8, 1 mM MgCl.sub.2, 0.5 mM DTT, heated as described early and used as reference.

(31) The samples were analyzed on a PE SCIEX API 2000 triple quadrupole mass spectrometer (mass spectrometer, Perkin-Elmer) in negative ion mode with H.sub.2O/acetonitrile=60/40 containing 0.1% triethylamine as mobile phase, flow rate was 40 l/min. 10 l of each supernatant were mixed with an equal quantity of mobile phase and directly injected into the mass spectrometer. The presence of [3-hydroxy-3-methylbutyrate-H].sup. ion was monitored.

(32) A peak corresponding to 3-hydroxy-3-methylbutyrate was observed for the following enzymes:

(33) Blattella germanica (German cockroach) P54961 (SEQ ID NO: 6)

(34) Gallus gallus (Chicken) P23228 (SEQ ID NO: 7)

(35) Homo sapiens (Human) Q01581 (SEQ ID NO: 8)

(36) Arabidopsis thaliana P54873 (CAA58763) (SEQ ID NO: 4)

(37) Caenorhabditis elegans P54871 (SEQ ID NO: 1)

(38) Schizosaccharomyces pombe (Fission yeast) P54874 (SEQ ID NO: 2)

(39) Saccharomyces cerevisiae (Baker's yeast) P54839 (SEQ ID NO: 3)

(40) Dictyostelium discoideum (Slime mold) Q86HL5 (SEQ ID NO: 10)

(41) Leuconostoc mesenteroides Q03WZ0 (SEQ ID NO:)

(42) Staphylococcus epidermidis Q8CN06 (SEQ ID NO: 11)

(43) Lactobacillus delbrueckii Q1GAH5 (SEQ ID NO: 24)

(44) Staphylococcus haemolyticus Q4L958 (198>V difference compared to wild type protein) (SEQ ID NO: 25)

(45) FIGS. 6 to 8 show representative results for commercially available 3-hydroxy-3-methylbutyrate, for the reaction using the HMG CoA synthase from Gallus gallus (P23228) and for the control assay without enzyme.

Example 6

3-hydroxy-3-methylbutyryl-CoA Production Using Lyases

(46) 3-hydroxy-3-methylbutyryl-CoA synthesis is carried out in the presence of radiolabeled [2-.sup.14C] acetone. The complete reaction for 3-hydroxy-3-methylbutyryl-CoA synthesis contains 40 mM Tris-HCl pH 8, 5 to 50 mM acetyl-CoA, 100 to 500 mM acetone, 1 to 10 mM MgCl.sub.2, 0.5 mM DTT and enzyme varying in the range from 0.5 to 7 mg/ml. The formation of product is analyzed after separation of reaction mixture by TLC or HPLC.

(47) 3-hydroxy-3-methylbutyryl-CoA is also analyzed by TLC method (Stadtman E. R., J. Biol. Chem. 196 (1952), 535-546). An aliquot of reaction is deposited on a cellulose plate and chromatographied in the following solvent system: ethanol/0.1 M sodium acetate pH 4.5 (1/1). Co-A and acetyl-CoA are used as internal standards. R.sub.f reported for 3-hydroxy-3-methylbutyryl-CoA is 0.88.

Example 7

Kinetic Parameters for the Enzymatic Reaction Between Acetyl-CoA and Acetone in the Case of HMG Synthases

(48) The kinetic parameters were measured using a variable concentration of acetone and a constant concentration of acetyl-CoA (10 mM) in following conditions:

(49) 40 mM Tris-HCl pH 8

(50) 2 mM MgCl.sub.2

(51) 0-1 M acetone

(52) The final pH was adjusted to 8.

(53) The reaction was initiated by the addition of 3 mg of purified enzyme to the 1 ml reaction mixture. The mixture was then incubated without shaking at 37 C. for 40 h.

Analysis of 3-hydroxy-3-methylbutyrate Production

(54) Thermochemical conditions leading to the decomposition of 3-hydroxy-3-methylbutyrate into isobutene were applied (Pressman et al., JACS, 1940, 2069-2080): the pH of the reaction mixtures was first adjusted to pH 4 using 6N HCl and the samples were then transferred into gas chromatography vials (Interchim). The vials were sealed and incubated at 110 C. for 4 hours, thus leading to the decomposition of 3-hydroxy-3-methylbutyrate into isobutene.

(55) The calibration curve was prepared in the same conditions using commercial 3-hydroxy-3-methylbutyrate.

(56) One milliliter of headspace gas was collected and injected into a HP5890 gas chromatograph (HP) equipped with a FID detector and a CP SilicaPlot column (chromatography column; Varian). Commercial isobutene was used as reference. From the isobutene signal the amount of 3-hydroxy-3-methylbutyrate initially present in the sample was calculated.

(57) The kinetics parameters for several of the studied HMG-CoA synthases are presented in the following Table.

(58) TABLE-US-00005 K.sub.M for acetone, k.sub.cat/K.sub.M 10.sup.6, Organism mM k.sub.cat 10.sup.4, sec.sup.1 mM.sup.1 sec.sup.1 Gallus gallus 250 5 2 Staphylococcus 200 0.6 0.3 epidermidis Schizosaccharomyces 200 0.2 0.1 pombe

(59) For the enzyme from S. epidermidis FIG. 9 shows a corresponding Michaelis-Menten plot.