Method for the regioselective deacetylation of mannosylerythritol lipids

Abstract

The present invention relates to methods for the enzymatic deacetylation of mannosylerythritol lipids produced by fermentation using lipases. More in particular, the present invention relates to a method for the enzymatic deacetylation of mannosylerythritol lipids using a hydrolyzing enzyme in an organic solvent containing only low amounts of water, preferably, no water. It further provides the use of organic solvents, containing only low amounts of water, more preferably, no water, for the enzymatic deacetylation of mannosylerythritol lipids.

Claims

1. A method for the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ia), to obtain a mannosylerythritol lipid represented by formula (Id), ##STR00020## the method comprising: incubating a mannosylerythritol lipid represented by formula (Ia), with a lipase in an organic solvent; wherein the organic solvent is selected from the group consisting of C2-C8 alcohols, and combinations thereof; and wherein the organic solvent contains less than 10% water.

2. The method of claim 1, wherein the C2-C8 alcohol is selected from the group consisting of linear and branched C2-C8 alcohols.

3. The method of claim 1, wherein the organic solvent contains less than 9% water.

4. The method of claim 1, wherein the C2-C8 alcohol is selected from the group consisting of ethanol, 1-propanol, 1-butanol, 1-pentanol, 1-hexanol, 1-heptanol, 1-octanol, isopropanol, isobutanol, t-butanol, 2-pentanol, isoamyl alcohol, 2-ethylhexanol, cyclohexanol, and benzylalcohol.

5. The method of claim 1, wherein the organic solvent contains less than 3% water.

6. The method of claim 1, wherein the lipase is a Candida antarctica lipase B.

7. A method for the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ic), to obtain a mannosylerythritol lipid represented by formula (Id), ##STR00021## the method comprising: incubating a mannosylerythritol lipid represented by formula (Ia), with a lipase in an organic solvent; wherein the organic solvent is selected from the group consisting of C2-C8 alcohols, and combinations thereof; and wherein the organic solvent contains less than 10% water.

8. The method of claim 7, wherein the C2-C8 alcohol is selected from the group consisting of linear and branched C2-C8 alcohols.

9. The method of claim 7, wherein the organic solvent contains less than 9% water.

10. The method of claim 7, wherein the C2-C8 alcohol is selected from the group consisting of ethanol, 1-propanol, 1-butanol, 1-pentanol, 1-hexanol, 1-heptanol, 1-octanol, isopropanol, isobutanol, t-butanol, 2-pentanol, isoamyl alcohol, 2-ethylhexanol, cyclohexanol, and benzylalcohol.

11. The method of claim 7, wherein the organic solvent contains less than 3% water.

12. The method of claim 7, wherein the lipase is a Candida antarctica lipase B.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) With specific reference now to the figures, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of the different embodiments of the present invention only. They are presented in the cause of providing what is believed to be the most useful and readily description of the principles and conceptual aspects of the invention. In this regard no attempt is made to show structural details of the invention in more detail than is necessary for a fundamental understanding of the invention. The description taken with the drawings making apparent to those skilled in the art how the several forms of the invention may be embodied in practice.

(2) FIG. 1: Relative solvent polarities extracted from Table A-1 (Reichardt, 2003))

(3) FIG. 2: Example 1HPLC results of the enzymatic convertion of MEL-B to MEL-D in 90% ethanol and absolute ethanol.

(4) FIG. 3: Example 2HPLC results of the enzymatic conversion of MEL-A to MEL-C in solvents with different polaritiesshort term follow-up.

(5) FIG. 4: Example 2HPLC results of the enzymatic conversion of MEL-B to MEL-D in solvents with different polaritiesshort term follow-up.

(6) FIG. 5: Example 2HPLC results of the enzymatic conversion of MEL-A to MEL-C in alcohols with different carbon chain lengthlong term follow-up.

(7) FIG. 6: Example 2HPLC results of the enzymatic conversion of MEL-B to MEL-D in alcohols with different carbon chain lengthlong term follow-up.

(8) FIG. 7: Example 4HPLC results of the enzymatic deacetylation of MEL-A, -B, -C, -D mix in 2-ethyl hexanol.

(9) FIG. 8: Example 5HPLC results of the enzymatic deacetylation of MEL-A and MEL-B in different commodity alcohols (conversion after 25 h).

(10) FIG. 9: Example 6HPLC results of the enzymatic conversion of MEL-A to MEL-C in the commodity alcohols, 2-ethylhexanol and isoamyl alcohol compared to 1-pentanol.

(11) FIG. 10: Example 6HPLC results of the enzymatic conversion of MEL-A to MEL-C and MEL-D in isoamyl alcohol.

(12) FIG. 11: Example 6HPLC results of the enzymatic conversion of MEL-A to MEL-C and MEL-D in isoamyl alcohol.

(13) FIG. 12: Example 7HPLC results of the water concentration profile of the enzymatic deacetylation of MEL-B in pentanol (conversion after 24 hours).

(14) FIG. 13: Example 8HPLC results of the water activity profile of the enzymatic deacetylation of MEL-A to MEL-C in 2-ethylhexanol (conversion after 24 hours).

(15) FIG. 14: Example 9HPLC results of the effect of the alcohol chain length on the deacetylation of MEL-B in the presence of 10% water or without water.

DETAILED DESCRIPTION OF THE INVENTION

(16) As already indicated herein before, the present invention provides a method for the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (I)

(17) ##STR00008##
wherein R.sub.1 and R.sub.2 are each independently selected from H and Ac; at least one of R.sub.1 and R.sub.2 is Ac; and n=6-10
said method comprising: incubating a mannosylerythritol lipid represented by formula (I), with a hydrolyzing enzyme in an organic solvent containing less than 10% water; in particular said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof.

(18) Hence, the present invention comprises a process for regiospecific deacetylation of MEL-A and -B to form MEL-C and -D, by using an enzyme that hydrolyses ester bonds. The hydrolyzing enzyme is preferably a lipase, more preferably Novozym? 435 and the reaction occurs in an organic solvent such as THF, toluene or cyclohexane, or alcohols such as ethanol, propanol, butanol, pentanol, hexanol, heptanol and octanol, in which all MEL-types dissolve; however in low amounts or water (i.e. less than 10%), preferably in the absence of water.

(19) Typically, following Rau et al.'s publication (Rau et al., 2005), for the fermentative synthesis of MELs, Pseudozyma aphidis is first grown on a seed culture medium containing in one liter, 30 g glucose, 1 g NH.sub.4NO.sub.3, 0.3 g KH.sub.2PO.sub.4, 1 g yeast extract, pH 6.0 not adjusted, incubation time 2 days. The main culture medium contains in one liter, 80 ml vegetable oil (either soybean oil or rapeseed oil), 2 g NaNO.sub.3, 0.2 g KH.sub.2PO.sub.4, 0.2 g MgSO.sub.4.7H.sub.2O, 1 g yeast extract, pH 6.2 not adjusted. The seed culture is inoculated from agar slants and incubated at 30? C. and shaken or stirred at a speed depending on the volume of the flask or reactor. The main culture is inoculated with 10% of its volume with the seed culture and incubated at 27? C. The culture suspension is extracted with either ethyl acetate or MTBE followed by paper filtration of the organic phase. The organic phase is then evaporated by rotary vacuum concentration. The remaining sticky phase is washed with 600 ml n-hexane-methanol-water (1:6:3) to remove the remaining oil and fatty acids (top phase). The aqueous phase (bottom phase) is washed twice with 100 ml n-hexane. After removal of methanol and water by vacuum evaporation, the remaining product contains a mix of MEL-A, -B, -C and -D. This mix is used as reference for TLC analyses.

(20) To further purify MEL-A, -B, -C and -D, this MEL-mix is separated by column chromatography using silica gel as the stationary phase and different proportions of chloroform and acetone for elution, as described previously by Kitamoto et al., 1990.

(21) In this research, 12 enzymes were tested for their ability to selectively deacetylate MELs without affecting the other ester bonds in the glycolipid structure. Using Candida antarctica lipase B (Novozym? 435), the regioselective deacetylation of conventional MELs was successful and resulted in two MEL types: MEL-C and MEL-D

(22) ##STR00009## ##STR00010##

(23) The successful reaction with Candida antarctica lipase B (Novozym? 435) was further optimized by screening solvents with different polarities, resulting in a 100% conversion and a shorter reaction time. In particular, solvents being alcohols with a carbon chain length of C2 to C8 appeared to be suitable; more in particular long-chain alcohols, i.e. n-alcohols with a carbon chain length of C5 to C8, proved to be the most suitable; and the reaction time could be reduced to hours instead of days as for the process of Fukuoka et al., 2011.

(24) When looking at the relative polarities of different alcohols (FIG. 1), It was further found that the more apolar the solvent, the better the results were and that this was also related to the polarity of the MEL, more in particular, MEL-A is less polar than MEL-B, so for its deacetylation a less polar alcohol is needed to reach the same reaction rate as with MEL-B. This trend is probably a consequence of the substrate's solubility: the better soluble the higher the reaction rate.

(25) In the context of the present invention, the term solvent is to be understood as being a substance that dissolves a mannosylerythritol lipid, thereby resulting in a solution. A mix of another organic solvent which dissolves MELs in combination with an alcohol can be used as well as long as there is enough alcohol present to participate in the reaction.

(26) In a particular embodiment of the present invention, said organic solvent is selected from the list comprising THF, toluene, cyclohexane, short-chain alcohols (1-3 C atoms) or long-chain alcohols (4-10 C atoms). More in particular, said organic solvent is selected from the list comprising linear and branched C2-C8 alcohols; in particular ethanol, propanol, isopropanol, 1-butanol, 2-pentanol, iso amylalcohol, t-butanol, pentanol, heptanol, octanol.

(27) In the context of the present invention, the term alcohol, is meant to include monohydric alcohols (i.e. having one functional OH group), as well as polyhydric alcohols (i.e. having 2 or more functional OH groups). Examples of monohydric alcohols include ethanol, propanol, isopropanol, 1-butanol, 2-pentanol, isoamyl alcohol, t-butanol, pentanol, heptanol, octanol. Examples of polyhydric alcohols, such as bihydric alcohols include glycols such as ethylene glycol, propylene glycol and butylene glycol. In a particular embodiment, the alcohols of the present invention are monohydric alcohols. In another particular embodiment, the alcohols of the present invention are polyhydric alcohols.

(28) Especially, the used solvents, i.e. alcohol+water (less than 10%) have a polarity less than that of ethanol. Hence, the reaction is performed in as little water as possible (preferably in the absence of water) in particular for those alcohols, which are rather polar, such as ethanol.

(29) In a particular embodiment of the present invention, said organic solvent preferably contains less than 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1% water, most preferably, said organic solvent contains no additional water.

(30) In the context of the present invention, the hydrolysing enzyme is preferably a lipase; even more preferably a Candida antarctica lipase B preferably immobilized for easier recovery and reuse. More in particular, the lipase is selected from the list comprising; Nozozym? 435, CaIB immo Plus, CaIB immo 8285, CaIB immo 8806, CaIB immo 1090, CaIB immo 5587, CaIB immo 5872, . . . ; preferably Novozym? 435. Lipases are generally classified using the following enzyme classification No EC 3.1.1.3.

(31) The process of the present invention was shown to be suitable for the conversion of: MEL-A to MEL-C, MEL-B to MEL-D, MEL-C to MEL-D and MEL-A to MEL-D. The conversion of MEL-A to MEL-D is regioselective and occurs in a 2-step process i.e. the initial conversion of MEL-A to MEL-C, followed by the conversion of MEL-C to MEL-D.

(32) Hence, in a particular embodiment, the present invention provides a method for the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ia), to obtain a mannosylerythritol lipid represented by formula (Ic),

(33) ##STR00011##
said method comprising: incubating a mannosylerythritol lipid represented by formula (Ia), with a hydrolyzing enzyme in an organic solvent containing less than 10% water; in particular said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof.

(34) In yet a further embodiment, the present invention provides a method for the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ib), to obtain a mannosylerythritol lipid represented by formula (Id),

(35) ##STR00012##
said method comprising: incubating a mannosylerythritol lipid represented by formula (Ib), with a hydrolyzing enzyme in an organic solvent containing less than 10% water; in particular said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof.

(36) In another particular embodiment, the present invention provides a method for the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ia), to obtain a mannosylerythritol lipid represented by formula (Id),

(37) ##STR00013##
said method comprising: incubating a mannosylerythritol lipid represented by formula (Ia), with a hydrolyzing enzyme in an organic solvent containing less than 10% water; in particular said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof.

(38) In yet another particular embodiment, the present invention provides a method for the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ic), to obtain a mannosylerythritol lipid represented by formula (Id),

(39) ##STR00014##
said method comprising: incubating a mannosylerythritol lipid represented by formula (Ia), with a hydrolyzing enzyme in an organic solvent containing less than 10% water; in particular said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof.

(40) As evident from the examples that follow herein after, and as discussed herein before, the inventors have found that the use of the claimed organic solvents, in limited amounts of water, is highly beneficial in an enzymatic deacetylation reaction of mannosylerythritol lipids.

(41) Therefore, in a further aspect, the present invention provides the use of an organic solvent, in the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (I)

(42) ##STR00015##
wherein R.sub.1 and R.sub.2 are each independently selected from H and Ac; at least one of R.sub.1 and R.sub.2 is Ac; and n=6-10
wherein said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof; and wherein said organic solvent contains less than 10% water.

(43) In a particular embodiment, the present invention provides the use of an organic solvent in the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ia), to obtain a mannosylerythritol lipid represented by formula (Ic) wherein said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof; and wherein said organic solvent contains less than 10% water

(44) ##STR00016##

(45) In a particular embodiment, the present invention provides the use of an organic solvent in the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ib), to obtain a mannosylerythritol lipid represented by formula (Id) wherein said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof; and wherein said organic solvent contains less than 10% water

(46) ##STR00017##

(47) In a particular embodiment, the present invention provides the use of an organic solvent in the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ia), to obtain a mannosylerythritol lipid represented by formula (Id) wherein said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof; and wherein said organic solvent contains less than 10% water

(48) ##STR00018##

(49) In a particular embodiment, the present invention provides the use of an organic solvent in the enzymatic deacetylation of a mannosylerythritol lipid represented by formula (Ic), to obtain a mannosylerythritol lipid represented by formula (Id) wherein said organic solvent is selected from the list comprising C2-C8 alcohols, and combinations thereof; and wherein said organic solvent contains less than 10% water

(50) ##STR00019##

EXAMPLES

Example 1: Omitting Water from Fukuoka et al.'s (2011) Reaction Solvent Results in a Vast Improvement of the Deacetylation Reaction of Conventional MEL-B Using Novozym? 435

(51) In a first attempt for the enzymatic deacetylation of the conventional MEL-B, Fukuoka et al.'s method using 90% ethanol as the reaction solvent was replicated. (Fukuoka et al., 2011) However, this resulted in a very low and slow conversion. FIG. 1 shows an 18% conversion after 24 h and no further increase after 72 h. This means that somehow the enzyme was inactivated after 24 h, probably due to the detrimental effect of the solvent, 90% ethanol+10 water, on the enzyme and its carrier. A parallel experiment was carried out where water was omitted from the reaction solvent. Absolute ethanol was used here resulting in a vast improvement of the deacetylation of the conventional MEL-B. After 24 h the conversion reached 91%. (FIG. 2)

(52) Materials and Methods

(53) Two 14 g/l MEL-B solutions were prepared, one in 90% ethanol and one in absolute ethanol (<0.1% water). The deacetylation reactions were initiated by adding 50 g/l Novozym? 435 (Lipase acrylic resin from Candida antarctica, ?5,000 U/g, Sigma-Aldrich, Belgium) and incubating at 60? C. and 200 rpm. Samples were taken periodically and analysed by thin layer chromatography (TLC) and high performance liquid chromatography (HPLC). Results from TLC are not shown. The reactions were terminated after 72 h.

(54) TLC Analysis:

(55) Qualitative results for each reaction were obtained by spotting a 1 ?l of sample on a TLC (thin layer chromatography) plate (TLC plates silica gel 60, Merck) followed by elution in a TLC chamber using dichloromethane-acetone (60:40) as the solvent system. Subsequently, the compounds were located by charring at 110? C. for 5 min after spraying an orcinol (0.1% orcinol in 5% H.sub.2SO.sub.4) solution.

(56) HPLC Analysis:

(57) The quantitative follow-up of each reaction was carried out by HPLC (High-performance liquid chromatography) analysis using a Nova Pak? Silica Column (WATERS?, 60 ?, 4 ?m, 3.9 mm?150 mm) coupled to an ELSD (Evaporating Light Scattering Detector), with a mobile phase flow-rate of 0.6 ml/min and a column temperature of 30? C. The used gradient consisted of chloroform and methanol (99:1 to 0:100) over 15 minutes. Samples were prepared by first evaporating the solvent in a vacuum centrifuge (9 mbar, 60? C.) and diluting them in chloroform to the appropriate concentration for HPLC analysis. Samples were filtered prior to analysis with 0.2 ?m PTFE syringe filters to remove solids which could bring damage to the HPLC column.

Example 2: Screening of Different Organic Solvents with Different Polarities (Log P) for the Regiospecific Deacetylation of Purified MEL-A and MEL-B

(58) Materials and Methods

(59) THF, toluene, cyclohexane, ethanol, butanol, pentanol, hexanol, heptanol and octanol containing 50 g/l immobilized enzyme (Novozym? 435, ?5000 PLU/g, Sigma-Aldrich, Belgium) and 14 g/l of conventional MEL-A (4-O-[(4,6-di-O-acetyl-2,3-di-O-alka-noyl)-?-D-mannopyranosyl]-meso-erythritol) or conventional MEL-B (4-O-[(6-mono-O-acetyl-2,3-di-O-alka-noyl)-?-D-mannopyranosyl]-meso-erythritol), were tested each for their suitability as a solvent for enzymatic deacetylation. The reaction temperature was 60? C. in each case and the shaking speed 200 rpm.

(60) Results

(61) In the pure solvent, i.e. without the addition of water, after 24 h the conversion of MEL-A to MEL-C was 76% with octanol, 77% with heptanol, 78% with hexanol, 72% with pentanol, 60 with butanol, 47% with ethanol and 36% with THF (FIG. 3). After 5 days (120 h) 95-100 conversion of MEL-A to MEL-C was achieved with pentanol, hexanol, heptanol and octanol (FIG. 5). After 24 h the conversion of MEL-B to MEL-D was complete (>99%) for n-alcohols with a carbon chain length of 5 to 8, namely pentanol, heptanol and octanol. In butanol the conversion reached 95%, in ethanol 91% and in THF 85% (FIG. 4 and FIG. 6).

(62) For both reactions, n-alcohols with a carbon chain length of C5 to C8 proved to be the most suitable.

(63) The difference in kinetics between the deacetylation on C6 on MEL-A and MEL-B could be explained by the difference in solubility of both substrates in the reaction medium or by steric hindrance on the MEL-A molecule due to the acetylation on C4.

Example 3: Deacetylation of the Fermentation Product Consisting of a Mixture of MEL-A, -B, -C and -D in Butanol

(64) Materials and Methods

(65) A first TLC sample was prepared by dissolving 0.25 g of the MEL-mix containing MEL-A, -B, -C and D in 10 ml of ethyl acetate. 5 g of the MEL-mix was dissolved in 100 ml of butanol in a 200 ml erlenmeyer flask with screw cap. 5 g of Novozym? 435 were added. The whole was shaken at 180 movements per minute in an incubator at 60? C.

(66) After 2 days and 5 hours (53 hours), the reaction was stopped. The solution was filtered through a paper filter on a funnel.

(67) The filtrate was then filtered through a PTFE filter (0.2 ?m) and the collected volume was placed in a vacuum centrifuge at maximum speed (1500 rpm) and 57? C. for 21 hours. A second TLC sample was then prepared by dissolving 0.25 g of the remaining sticky brown syrup into 10 ml of ethyl acetate.

(68) Results:

(69) After 53 hours, the substrate has been almost completely converted into MEL-C and -D (TLC data not shown). Results from example 1 where MEL-A is completely converted into MEL-C in 5 days, suggest that complete conversion of the MEL-mix into MEL-C and MEL-D could be achieved using a slightly longer reaction time and pentanol instead of butanol as the reaction medium.

Example 4: Deacetylation of the Fermentation Product Consisting of a Mixture of MEL-A, -B, -C and -D in the Commodity Alcohol 2-Ethylhexanol

(70) Materials and Methods

(71) 5 ml of a 14 g/l MEL reference containing MEL-A, -B, -C and -D was dissolved in 2-ethylhexanol. The preparation of this MEL reference was described in the detailed description of the invention. The deacetylation reaction was initiated by adding 50 g/l of Novozym? 435 and incubating at 60? C. and shaken at 270 rpm in closed glass GC headspace vials. Periodically samples were withdrawn for TLC and HPLC analyses. HLPC and TLC analyses were conducted according to Example 1.

(72) Results

(73) In less than 24 h MEL-A and MEL-B could be completely converted into MEL-C and -D. After 24 h the product contained 58% MEL-C and 42% MEL-D. As the reaction proceeded, MEL-C was gradually converted into MEL-D to end with a product containing 37% MEL-C and 63 MEL-D after 120 h (5 days). The HPLC results are shown in FIG. 7.

Example 5: Deacetylation of MEL-A and -B in Different Commodity Alcohols, Using Novozym? 435

(74) Materials and Methods

(75) Different branched commodity alcohols were used for the deacetylation of MEL-A and MEL-B: isopropanol (C3), isobutanol (C4), tertiary butanol (C4) and 2-pentanol (C5). Pentanol and propanol were also used in this experiment to compare the non-branched with the branched alcohols. 14 g/l of MEL-A and MEL-B were respectively dissolved in 1 ml of each solvent. 0.014 g of MEL was weighed in a 2 ml centrifuge tube with screw cap and 1 ml of solvent was added and the whole was vortexed. A first sample was taken (t=0 h) after which 50 g/l of Novozym? 435 was added to initiate the reactions. The reactions were started by putting the centrifuge tubes in a heating block at 60? C. and 1000 rpm. For TLC and HPLC analyses 51 ?L of sample was taken just before the addition of enzyme and after 25 h. TLC and HPLC methods are described in Example 1.

(76) Results

(77) HPLC results showed that the conversion of MEL-B was much faster than that of MEL-A, as noticed already in Example 2. After 25 h, all the solvents showed a complete conversion of MEL-B. For MEL-A, the highest conversions were achieved in 2-pentanol (84%), isobutanol (81%) and pentanol (80%) (FIG. 8).

Example 6: Deacetylation of MEL-A in the Commodity Alcohols, 2-Ethylhexanol and Isoamyl Alcohol, Using Novozym? 435

(78) Materials and Methods

(79) The deacetylation of MEL-A was carried out in 2 commodity alcohols, 2-ethylhexanol (C8) and isoamyl alcohol (C5). The reaction in pentanol was run in parallel as a reference.

(80) A 14 g/l MEL-A solution was made with 5 ml of each alcohol in glass reaction vials. To each vial 0.25 g of Novozym? 435 was added to initiate the reactions. The vials were incubated at 60? C. and shaken at 240 rpm. During the reactions samples of 100 ?l were taken from which 1 ?L was used for TLC analysis. From the remaining amount, the solvent was evaporated in a vacuum centrifuge at 60? C. and 9 mbar. The sample was then redissolved in 1 ml of chloroform (the same solvent as used for HPLC analysis) and 50 times diluted.

(81) Results

(82) TLC results showed that the conversion of MEL-A to MEL-C was slower in isoamyl alcohol as compared to 2-ethylhexanol. In 2-ethylhexanol, MEL-A was almost completely converted into MEL-C after 24 h. After 96 h both reaction mixtures presented significant amounts of MEL-D. In isoamyl alcohol almost all MEL-A was converted into MEL-D (TLC data not shown).

(83) The TLC results were confirmed by HPLC analysis. FIG. 9 shows a comparison of the conversion of MEL-A in the 3 solvents. The conversion is much faster in 2-ethylhexanol (92 after 24 h) compared to 1-pentanol (55%) and isoamyl alcohol (9%). However, Novozym? 435 expresses a different selectivity in 2-ethylhexanol and isoamyl alcohol. The deacetylation of MEL-A in 2-ethylhexanol occurred in 2 steps; first the R.sub.1 acetate on the mannose moiety was cleaved to produce MEL-C with a 87% MEL-C yield after 24 h. Subsequently the R.sub.2 carbon was deacetylated and yielded 37% MEL-D after 96 h (FIG. 10). In isoamyl alcohol the initial MEL-C formation rate was much lower, but the formation of MEL-D already took place when only 29% of MEL-C was formed resulting in a much higher MEL-D yield of 93% after 96 h. (FIG. 10 and FIG. 11)

Example 7: Effect of WaterDeacetylation of MEL-B in Pentanol with Different Water Concentrations, Using Novozym? 435

(84) Materials and Methods

(85) 5 reactions in pentanol were carried out with different water concentrations (0%, 1%, 3%, 5% and 10% (v/v). 14 g/l of MEL-B and 50 g/l of Novozym? 435 were used to carry out the reactions. 0 to 10% of water was added to the MEL-B solution in pentanol. For the 0% water in pentanol, molecular sieve (12% w/v) was added to dry the solvent. The reaction vials were incubated at 60? C. during 24 h. The first 5 hours, every hour a sample of 100 ?l was taken. These samples were used for TLC and HPLC analysis.

(86) Before the addition of Novozym? 435, a sample was taken and analyzed using TLC (data not shown). The results showed that the start product contained MEL-B only.

(87) Results

(88) After a reaction time of 24 h a clear trend was visible: the lower the water concentration, the higher the conversion of MEL-B. Based on the quantitative HPLC results, the water concentration profile could be determined for the deacetylation of MEL-B in pentanol (FIG. 12). Here as well it was clear that a lower water concentration promoted a higher the conversion. At a water concentration of 10%, the conversion was even reduced to 0%.

Example 8: Effect of WaterDeacetylation of MEL-A Under Different aw Values, Using Novozym? 435

(89) Introduction

(90) The activity of an enzyme is correlated to the water that is bound to it. It is difficult to measure and control the amount of water bound to the enzyme. Control of water activity (a.sub.w) comes in handy as a way to control bound water and achieve reproducible results, especially in organic solvents that are water immiscible. An a.sub.w value of 0 corresponds to a dry system and a value of 1 to a water-saturated system. At equilibrium the a.sub.w is equal in all phases.

(91) Materials and Methods

(92) A 14 g/l stock solution of MEL-A in 2-ethylhexanol was prepared by dissolving 0.7 g of MEL-A in a 50 ml flask. From this stock solution 5 ml was brought into six 10 ml GC headspace vials. Six times 0.25 g of Novozym? 435 were weighed as well.

(93) Next, the stock solutions and the enzymes were put in jars with saturated salt solutions to pre-equilibrate at the corresponding water activity levels during 18 h at room temperature (Table 1).

(94) TABLE-US-00001 TABLE 1 Vial Solvent Salt a.sub.w 1 2-ethylhexanol LiBr 0.064 2 2-ethylhexanol LiCl 0.113 3 2-ethylhexanol MgCl.sub.2 0.328 4 2-ethylhexanol Mg(NO.sub.3).sub.2 0.529 5 2-ethylhexanol NaCl 0.753 6 2-ethylhexanol K.sub.2SO.sub.4 0.973 7 2-ethylhexanol 0

(95) Before the start of the reaction, a seventh reaction vial was filled with 5 ml of MEL-A stock solution and molecular sieve to reach an a.sub.w of 0. The reactions were initiated by adding the enzymes to the vials with the MEL-A stock solutions and by incubating them at 60? C. and shaking them at 240 rpm. Samples were analysed by HPLC.

(96) Results

(97) FIG. 13 shows the conversion of MEL-A as a function of the water activity. A clear trend is noticeable revealing that at higher a.sub.w values the conversion decreases. 2 points at very low a.sub.w values deviate due to sample taking. At such low moisture levels the a.sub.w can be greatly influenced. At the lowest a.sub.w this did not pose a problem because the molecular sieve controlling the a.sub.w was present in the reaction mix during the entire time. From these results it could be concluded that the optimal a.sub.w for the deacetylation of MEL-A in 2-ethylhexanol using Novozym? 435 is 0. Results from Example 6 already revealed a similar trend based on water percentage added to the reaction mix.

Example 9: Effect of WaterDeacetylation of MEL-B in MeOH, EtOH, PrOH and PeOHwith 10% Water and without Water, Using Novozym? 435

(98) This experiment was set up to verify if the negative effect of water on the deacetylation of MELs in lower alcohols (C1-C3) is as pronounced as in higher alcohols (C5 for example).

(99) Materials and Methods

(100) A series of solutions of 14 g/l MEL-B were prepared in 5 ml of each solvent in glass reaction vials. A second series was prepared in the same way whereafter 10% (v/v) of water was added to the reaction vials. Afterwards from each vial a sample was taken and Novozym? 435 (50 g/l) was added. The reaction vials were incubated at 65? C. and shaken continuously at 240 rpm. Another sample was taken after 22 h when the reaction was ended.

(101) Results

(102) HPLC results showed that in the pure alcohols the conversion increased in function of the alcohol chain length, with a 100% conversion in pentanol, whereas in combination with 10% water the conversion decreased down to 0% in function of the alcohol chain length (FIG. 14). Generally, the conversions were significantly higher without the addition of water.

Example 10: Screening of Different Enzymes for the Deacetylation of MELs

(103) Materials and Methods

(104) As an example for the deacetylation of MELs, MEL-B was used to screen a variety of enzymes. A 14 g/l solution of MEL-B in 1-pentanol was prepared and divided over 22 glass vials. To each vial 50 g/l (or more for enzymes with a low specific activity) of 22 different enzymes were added to initiate the reaction (Table 2). The reaction was carried out at 60? C. and 270 rpm. From each vial a sample was taken before the start of the reaction and one after 23 hours. 1 ?l of each sample was spotted on a TLC plate for analysis. The MEL reference containing MEL-A, -B, -C and -D was also spotted for identification. TLC was carried out according to EXAMPLE 1.

(105) Results

(106) Under the used reaction conditions, 7 from the 22 screened enzymes were able to reach (almost) complete conversion of MEL-B to MEL-D in only 23 hours (+++). Esterzyme B2 gave a significant conversion as well(++). Lipozyme? TL IM and RM IM, Transzyme A2 and Esterzyme B1 generated a slight conversion (+). For the other enzymes, no conversion could be detected (?).

(107) TABLE-US-00002 TABLE 2 Enzyme Company Deacetylation Lipase from Aspergillus oryzae Sigma-Aldrich ? Co. Protease from Streptomyces griseus Sigma-Aldrich ? Co. Protease from Aspergillus saitoi Sigma-Aldrich ? Co. Candida antarctica lipase B Sigma-Aldrich +++ (Novozym? 435) Co. Lipase from Candida rugosa Biocatalysts ? (Lipomod? 34P) Ltd. Lipase from Rhizomucor miehei, Novozymes + immobilized (Lipozyme? RM IM) A/S Lipase from Thermomyces lanugenosus, Novozymes + immobilized (Lipozyme? TL IM) A/S Lipase from Rhizopus oryzae, Amano ? powder (Lipase DF Amano 15-K) Enzyme Inc. Lipase from Penicillium camembertii, Amano ? powder(Lipase G Amano 50) Enzyme Inc. Candida antarctica lipase A, c-LEcta GmbH ? powder (CalA lyo, FG) Esterase from Penicillium roqueforti, Biocatalysts ? powder (Lipomod? 338 MDP) Ltd. Candida antarctica lipase A, aqueous Novozymes ? suspension (Novocor? ADL) A/S Transzyme A1 Transbiodiesel ? Ltd. Transzyme A2 Transbiodiesel + Ltd. Esterzyme B1 Transbiodiesel + Ltd. Esterzyme B2 Transbiodiesel ++ Ltd. CALB immo 8806 Purolite Ltd. +++ CALB immo 5587 Purolite Ltd. +++ CALB immo PLUS Purolite Ltd. +++ CALB immo 5872 Purolite Ltd. +++ CALB immo 1090 Purolite Ltd. +++ CALB immo 8285 Purolite Ltd. +++

REFERENCES

(108) Fukuoka, T., Yanagihara, T., Imura, T., Morita, T., Sakai, H., Abe, M., & Kitamoto, D. (2011). Enzymatic synthesis of a novel glycolipid biosurfactant, mannosylerythritol lipid-D and its aqueous phase behavior. Carbohydrate Research, 346(2), 266-271. Retrieved from http://www.sciencedirect.com/science/article/pii/S0008621510005094 Jos?, C., Bonetto, R. D., Gambaro, L. a., Torres, M. D. P. G., Foresti, M. L., Ferreira, M. L., & Briand, L. E. (2011). Investigation of the causes of deactivation-degradation of the commercial biocatalyst Novozym? 435 in ethanol and ethanol-aqueous media. Journal of Molecular Catalysis B: Enzymatic, 71, 95-107. doi:10.1016/j.molcatb.2011.04.004 Reichardt, C. (2003). Solvents and Solvent Effects in Organic Chemistry (Third edit.). doi:10.1002/9783527632220