Continuous preparation of flavour ‐ active acetate esters by direct biocatalytic esterification

Mycelium ‐ bound lipase(s) from Aspergillus oryzae catalysed direct esterification of isoamyl alcohol and cinnamyl alcohol with acetic acid in organic solvent, showing high stability towards substrates and products. Water produced during the esterification did not significantly affect the equilibrium of the reaction, allowing for high conversions. These features were exploited for preparing flavour ‐ active acetate esters (isoamyl and cinnamyl acetate) in batch and continuous systems. A continuous stirred tank membrane reactor (CST ‐ MR) was developed securing good reactor productivity and high biocatalyst stability.

different flavour esters. [ 5,6,7,8,9,10 ] Direct esterification with free carboxylic acids is a preferred option since many acids are available as natural molecules, but the biotransformation is often hampered by formation of water, which may increase water activity of the medium thus negatively affecting the equilibrium of the reaction. [ 11,12,13,14 ] Different strategies have been developed for controlling water activity of the medium and improving the final conversion in the desired esters. A large-scale enzymatic production of low molecular weight flavour esters in organic solvent was developed, where solutions for the elimination of substrate and product inhibitions are presented; water produced during the process was continuously removed by azeotropic distillation or by using molecular sieves, hence enabling high yields of the produced esters.
[ 15 ] A major drawback of direct acetylation with acetic acid is the deactivation of lipases in the presence of the free acid. [ 16 ] Lipase stabilisation in the presence of acetic acid has often been achieved through immobilisation. [ 17,18,19,20,21,22,23,24 ] Novozym 435 (the commercial preparation of lipase B from Candida antarctica, CALB, immobilised on a macroporous acrylic resin) has been often used as stable and efficient biocatalyst for acetate esters preparation. [ 19,22 ] Towards the development of a continuous methodology, different type of reactors can be set up for the production of esters by direct esterification catalysed by lipases in organic solvents. [ 25,26 ] The final process productivity depends on the activity and stability of the biocatalyst, whereas molar conversion can be modulated by applying the appropriate residence time in the bioreactor. Continuous stirred tank membrane reactor (CST-MR) has been scarcely used for the continuous enzymatic synthesis of esters, albeit they may offer several advantages (enhanced mass transfer, improved liquid mixing and reduced clogging) over packed bed reactors.
[ 27, 28,29,30 ] Moreover, the use of a suited membrane allows for continuous removal of the organic liquid phase containing the product, while retaining the biocatalyst inside the reactor. [ 31 ] Mycelium-bound fungal carboxylesterases have been often used for flavour ester production by direct esterification. [ 28 ] Dry mycelium of moulds (eg Aspergillus oryzae) can be effectively used for different biotransformations, [ 33,34 ] including direct esterification of different alcohols with remarkable advantages, such as easy preparation of the biocatalyst, high stability in organic solvents and high resistance to the inactivation caused by carboxylic acids (including acetic acid); moreover, this biocatalyst allows direct esterification with high molar conversions, enabled by favourable water partition. [ 35,36,37 ] In this work, we have investigated the continuous preparation of flavour-active acetate esters (isoamyl and cinnamyl) using mycelium-bound lipase(s) from Aspergillus oryzae in a continuous stirred tank membrane reactor (CST-MR). F Fi ig g. . 1 1 0.2 g/L, pH 5.8) for 48 h at 28°C on a reciprocal shaker (100 rpm). The mycelium suspension was recovered by vacuum filtration using a Buchner funnel and paper filter and washed with distilled water, and lyophilised.
2.2. B Ba at tc ch h b bi io ot tr ra an ns sf fo or rm ma at ti io on ns s Different amounts of dry (lyophilised) mycelium of A oryzae were suspended in 2.5 mL of n-heptane for 30 minutes and the reactions started by adding different concentrations of isoamyl alcohol and acetic acid; the reaction mixture was magnetically stirred at the desired temperature. Samples (0.25 mL) were taken at different intervals, added to an equal volume of an internal standard solution (noctanol) in n-heptane and analysed by GC. Recovery and purification were carried out as previously reported. [ 7 ] 2.3. C Co on nt ti in nu uo ou us s r re ea ac ct ti io on n p pr ro oc ce ed du ur re e The vessel used in this study was a glass-made membrane reactor constructed by the glass workshop of University of Milan. The working volume of the reactor was 200 mL (a picture and a cross section of the membrane reactor are given in Figure 1). The membrane bioreactor was composed with a thermostated water-jacketed glass vessel and an ultrafiltration module (0.1 µm) mounted on the bottle cap. The final suspension was maintained under agitation using a magnetic stirrer.

Schematic representation and picture of the membrane reactor
The temperature in the reactor was kept constant by circulating water in the jacket. A Gilson Miniplus 2 peristaltic pump controlled incoming liquid flow.
The molar balance for a continuous stirred tank reactor (CSTR) has the following expression, where F stands for the molar flow rate of substrate: A e.Proofing https://wileyproofs.sps.co.in/eproofing_wiley_v3/printpage.php?... (1) (2) (3) One can rewrite as: Finally, rate of the CSTR was calculated using the algebraic Equation 3: The reactor was run for 24-30 hours depending on the residence time of each operation to reach a steady state before collecting samples for analysis.

A An na al ly ys si is s
Isoamyl alcohol and isoamyl acetate concentrations were determined by gaschromatographic (GC) analysis on a Carlo Erba Fractovap GC equipped with a fusedsilica capillary column MEGA-DEX DMP-Beta (dimethyl pentyl-β-cyclodextrin, 25 m × 0.25 mm i.d.), with the injector temperature at 200°C. Oven temperature was a 5°C/min gradient from 40 to 180°C. The retention times were isoamyl alcohol, 6.7 minutes and isoamyl acetate, 7.1 minutes. Cinnamyl alcohol and cinnamyl acetate concentrations were determined with a fused-silica capillary column MEGA-SE30 (100% methyl polysiloxane; 25 m × 0.25 mm i.d.), with the injector temperature at 200°C. Oven temperature was 80°C (2 minutes) and then from 80 to 180°C with a 5°C/min gradient. The retention times were cinnamyl alcohol, 9.9 min and cinnamyl acetate, 13.2 min.

Multisimplex
2.0 software, previously employed for the optimisation of biotransformations. [ 37 ] Response variables were the productivity of the biotransformation (defined as amount of product per amount of biocatalyst per unit of time) and the molar conversion of the alcohol, both determined after 24 hours; the control variables were alcohol concentration, biocatalyst concentration, molar ratio (acid/alcohol) and temperature. The initial levels considered for the optimisation are listed in Table 1. Table 1 Control  (isoamyl alcohol = 54 mmol/L, dry mycelium = 25 mg mL , molar ratio acid/alcohol 1.2/1 at 50°C) were chosen for further experiments, since the lowest amount of biocatalyst still gave the highest amount of isoamyl acetate. Critical parameters were acetic acid concentration and temperature, since significant decreases of the productivity were observed in any trial performed at concentrations above 70 mmol/L and at temperature above 60°C. Figure 3 reports the time-course of the direct esterification of isoamyl alcohol with acetic acid under the optimised conditions.  Direct esterification of isoamyl alcohol with acetic acid in CST-MR was then studied; substrates were continuously added in a n-heptane solution and the composition of the outflow solution monitored by GC. Experiments were initially focused to reach the best compromise between reaction rate and degree of conversion (Table 2).  Experiments carried out with a τ = 500 min allowed for 78%-83% molar conversion, depending on the F in the inlet flow; the best compromise between conversion and rate was found with a F of 5.5 µmol/min (corresponding to a C = 55 mmol/L and C = 66 mmol/L) corresponding to a rate of 0.092 µmol/min. The rate was slightly better than the one observed in batch reactor, showing the good performance in terms of mixing of the CST-MR. Experiments were also carried out at lower residence time (t = 350 min), leading to a significant decrease in the conversion.

3.3.
I Is so oa am my yl l a ac ce et ta at te e p pr re ep pa ar ra at ti io on n: : s st ta ab bi il li it ty y o of f t th he e b bi io oc ca at ta al ly ys st t i in n t th he e c co on nt ti in nu uo ou us s r re ea ac ct to or r The continuous operation aimed at preparing isoamyl acetate was carried out for 10 days under optimised conditions (inflow solution containing 55 mmol/L isoamyl alcohol and 65 mmol/L acetic acid, τ = 500 min at 50°C); no decrease of the molar conversion was observed at the end of the process, thus indicating a notable operational stability of the catalyst. Table 3 summarises the results of the continuous biotransformation, which allowed for a remarkable overall reactor productivity (defined here as the amount of isoamyl acetate produced for volume of reactor) of 169 mg mL .  A closed-loop reactor was also set up by recirculating the outflowing solution; after 16 hours of operation, the mycelium-bound lipase(s) was able to catalyse a conversion of 98% of isoamyl alcohol into the corresponding acetate ester.
3.4. C Ci in nn na am my yl l a ac ce et ta at te e p pr re ep pa ar ra at ti io on n: : b ba at tc ch h a an nd d c co on nt ti in nu uo ou us s r re ea ac ct to or r Biocatalytic preparation of cinnamyl acetate was also studied; cinnamyl acetate is characterised by a typical pineapple flavour.
[ 1 ] Direct acetylation of cinnamyl alcohol was accomplished applying the optimised conditions employed for the direct esterification of isoamyl alcohol. Figure 4 shows the time-course of the batch biotransformation.
Esterification between cinnamyl alcohol and acetic acid catalysed by mycelium-bound lipase(s) from A oryzae. Reaction conditions: initial alcohol concentration 54 mmol/L; initial acetic acid concentration 65 mmol/L in n-heptane containing 25 g L of biocatalyst at 50°C (C = 55 mmol/L, F = 5.5 µmol/min, C = 66 mmol/L, Q = 0.1 mL min , τ = 500 min at 50°C) and in the same CST-MR. Also, in this case, the biocatalyst was stable over 10 days of reaction; Table 4 summarises the data of the continuous bioprocess.

C Co on nc cl lu us si io on n
The mycelium-bound lipase(s) activity of Aspergillus oryzae was exploited for the direct esterification of isoamyl and cinnamyl alcohol with acetic acid for the preparation of the corresponding flavour-active esters. Remarkably, similar acetylation rates and molar conversions were observed, although two alcohols with quite different structure were employed. It is worthy to underline that the biotransformation occurred in n-heptane without any particular system for the removal of the water produced; it has previously suggested that the mycelia supply a hydrophobic micro-environment, thus disfavouring water access to the enzymes which catalyses the reaction. [ 38 ] A stirred tank membrane reactor (CST-MR) was set up for carrying out continuous biotransformations; reusability of the biocatalyst reduces the cost of fermentation, necessary for its production. Indeed, the CST-MR guaranteed 10 days of operation without any significant loss of the biocatalyst activity; reactor productivity (defined as the amount of ester formed per volume of reactor) was 169 mg/mL for isoamyl acetate and 223 mg/mL for cinnamyl acetate. In conclusion, a highly effective continuous production of isoamyl and cinnamyl acetate was obtained with dry mycelium, without any necessity of costly or laborious enzymatic purification, and without the need of water sorption during the prolonged bioprocess.

C CO ON NF FL LI IC CT T O OF F I IN NT TE ER RE ES ST T
The authors have no conflicts of interest to declare. e.Proofing https://wileyproofs.sps.co.in/eproofing_wiley_v3/printpage.php?...