ENZYMATIC LIPOPHILIZATION OF VITAMIN C WITH LINOLEIC ACID: DETERMINATION OF ANTIOXIDANT AND DIFFUSION PROPERTIES OF L-ASCORBYL LINOLEATE

Lipophilic derivatives of vitamin C are additives with antioxidant properties, attractive for application in food, cosmetics and pharmaceutics. They could be synthesized in lipase-catalyzed processes by using various acyl donors. Hereby, we present application of linoleic acid, which is polyunsaturated fatty acid essential in human nutrition, for esterification of vitamin C catalyzed by immobilized enzyme preparation Novozym 435 in acetone. Highest specific ester yield, 9.7 mmol/g of immobilized lipase, was accomplished with 0.15 M of vitamin C, 0.6 M of linoleic acid, 3 g/l of enzyme and 0.07% (v/v) of water, at 60 °C. NMR analyses of purified product proved that synthesized molecule was identical to 6-O-ascorbyl linoleate. Capacity of ester for scavenging 2,2-diphenyl-1picrylhydrazyl radicals was two times higher comparing to parent molecule, vitamin C. Its diffusion coefficient, determined using Franz cell and cellulose acetate membrane, was 40% higher than palmitate and 62% higher than oleate. Obtained results showed that L-ascorbyl linoleate could be successfully synthesized in biocatalyzed processes. Furthermore, it was demonstrated that it possess high potential for application in different lipophilic products due to its liposolubility, high antioxidant efficiency and good diffusion properties.


INTRODUCTION
Vitamin C is natural molecule known for its biological functions and antioxidant properties.However, it is unstable and poorly liposoluble, therefore unsuitable for application in lipophilic products.Products with high lipid content are susceptible to deterioration caused by oxidation of unsaturated fatty acids which they contain.To alter polarity of L-ascorbic acid and improve its stability while saving its beneficial properties, variety of L-ascorbyl fatty acid esters were so far synthesized via esterification and transesterification using sulfuric acid (Cousins et al., 1977;Nickels and Hackenberger, 1986) or lipases as catalysts (Karmee, 2009;Stojanović et al., 2013a).At this moment, L-ascorbyl palmitate and stearate are being commercially produced in chemically catalyzed processes.In recent years, enzymatic pro-cesses for the synthesis of valuable products, including fatty acid ascorbyl esters, have attracted attention of researchers, since they are efficient, selective, eco-friendly and, if properly optimized, highly cost-effective.
According to literature data, the most efficient biocatalyst of the process is lipase isoenzyme B from Candida antarctica (Karmee, 2009;Stojanović et al., 2013a).Liquid enzyme preparations are not adequate for application, due to negative influence of water present in reaction medium on equilibrium position.Immobilized enzymes are almost exclusively applied since they could be easily separated from reaction mixture and reused in repetitive reaction cycles leading to decrease of process costs and increase of commercial competitiveness (Karmee, 2009  .Different oils, fatty acids and their -methyl, -ethyl, and -vinyl esters were so far used as substrates (Karmee, 2009;Stojanović et al., 2013a;Bezbradica et al., 2017).Saturated fatty acids were predominantly applied, although it is well known fact that unsaturated ones, particularly PUFA-s (polyunsaturated fatty acids) have more beneficial effects on human health.Linoleic acid is unsaturated omega-6 fatty acid with two cis double bonds which is essential for human nutrition.In recent years, ascorbyl linoleate was synthesized enzymatically by several authors (Song et al., 2004;Song et al., 2006;Karmee, 2009).It is proven that linoleic acid reduces inflammation and prevents trans-epidermal moisture loss when applied topically and therefore, it is common ingredient of beauty products (Bradoo et al., 1999;Song and Wei, 2002).
However, just like the other PUFAs, it is prone to autooxidation, which could be prevented by transforming it into ascorbyl ester (Adamczak and Bornscheuer, 2009).Its ascorbyl ester is valuable substance for food, cosmetic, and pharmaceutical industry owing to its nutritional, antioxidant, bleaching, skin conditioning, and skin protecting function (Ando et al., 1992;Song et al. 2004).
The primary purpose of this study was to synthesize L-ascorbyl linoleate using immobilized Candida antarctica lipase B as a catalyst (Scheme 1.), to purify it, and to characterize it in terms of antioxidant and diffusion properties.
First part of the research was focused on the optimization of essential esterification parameterslimiting substrate concentration, substrates molar ratio, water content, and enzyme amount in order to maximize conversion degrees and ester yields.Afterwards, product was isolated and its structure was confirmed by NMR.Purified ascorbyl linoleate was further used for determination of its free radical scavenging capacity.Diffusion coefficient of synthesized molecule was determined, as well.

Procedure for ester synthesis
Esterification reactions were conducted in 100 ml sealed Erlenmeyer flasks, in thermostated orbital shaker at 60 °C and 200 rpm for 72 h.Reaction mixtures consisted of vitamin C, linoleic acid and water (amounts defined for each experiment individually), while organic solvent -acetone was added to reach the final volume of 10 ml.Reactions were initiated by adding predefined amounts of biocatalyst, while control samples did not contain immobilized lipase and were exposed to same treatment as all other samples.All experiments were conducted in duplicates and average values are presented in graphs.Deviations were less than 5%.

HPLC analyses
Quantitative analyses of reactants and products were performed by HPLC on Dionex Ultimate 3000 Thermo Scientific (Waltham, USA) system and a reverse phase column (Hypersil GOLD C18, 150 mm × 4.6 mm, 5 μm).During reaction, 50 μl of reaction mixtures were withdrawn for quantitative analyses.Reaction mixture samples were diluted 50-100 times with methanol before analyses.Injection volumes were 15 μl.As a mobile phase, methanol/formic acid, 100/0.1% (v/v) with a flow rate of 0.5 ml/min, was used.UV-VIS detector was used and ester was detected at 235 nm.

Purification of product
Reaction mixture was filtered, evaporated under vacuum, diluted 10 times comparing to initial mixture and subjected to HPLC system with fraction collector.Semipreparative reverse-phase column (Hypersil Gold ODS, 250 mm × 10 mm, 5 µm, Thermo Fisher Scientific, Waltham, USA) was used.Mobile phase used for preconditioning of column and elution was acetonitrile/water 95/5% (v/v) with 0.1% of formic acid.Elution flow rate was 6 ml/min and injection volume of samples was 1 ml.Fractions of pure product were collected in portions of 0.5 ml, retested on analytical column and evaporated to constant weight.Detection was done at 235 nm.

Antioxidant activity determination
Antioxidant activity of L-ascorbyl linoleate was measured spectrophotometrically by standard DPPH (2,2-diphenyl-1-picrylhydrazyl) method based on ability of tested compounds to act as electron acceptors and to reduce stable DPPH radical (Sharma and Bhat, 2009).Reaction mixture consisted of 200 µl of analyzed sample with concentration of 0-5 mg/ml of ester, 200 µl of 0.15 mM DPPH solution in methanol and 600 µl of pure methanol.Control samples were composed from 800 µl of methanol and 200 µl of DPPH solution.Mixtures were vortexed for 2 minutes and left for 30 minutes at room temperature in the dark.Absorbance was measured afterwards at 517 nm, and free radical scavenging capacity (FRSC) was calculated as follows: Where A c is absorbance of control sample and A s is absorbance of analyzed sample.Final result was expressed IC 50 (half minimal inhibitory concentration) value which represents concentration of antioxidant which lowers initial concentration of DPPH radicals by 50%.

Diffusion studies
Diffusion studies were conducted in jacketed Franz diffusion cell (PermeGear, Inc., Hellertown, PA, USA) made of two compartments separated with acetatecellulose membrane (pore size of 0.2 μm).
As samples, placed in the donor compartment, approximately 2 g of reaction mixtures free of catalyst and solvent (removed by vacuum evaporation) were used.
The receptor compartment was filled with medium (50% ethyl alcohol) and continuously stirred at 400 rpm using magnetic bead (Klimundová et al., 2006).The samples were taken in appropriate time intervals during 24 h from receptor chamber and subjected to HPLC analyses for determination of ascorbyl ester concentrations.
The diffusion coefficients were calculated using Fick's second law, as described by Pjanović et al. (Pjanović et al., 2010): and its linearized form Where β (24900 min/m 2 ) is geometric constant related to a cell type, t (min) is time, C (g/l) is concentration, subscripts R and D refer to receptor and donor compartment, respectively, and super-script 0 refers to initial conditions (at zero time).

Optimization of reaction conditions
Lipase catalyzed esterifications are reversible reactions, strongly affected by several experimental factorssubstrates concentration, water content and concentration of biocatalyst.Hence, it is of great importance to optimize all parameters influencing equilibrium yield of ester, as well as duration and costs of biosynthesis process.
At the beginning of the study, concentrations of both reactants were varied within wide ranges, with the aim of their optimization.At first, concentration of acyl acceptor -L-ascorbic acid was kept constant, while influence of linoleic acid concentration on esterification progress was examined.In previous studies, different substrate molar ratios were established as optimal (Karmee, 2009 ).In this research, substrates molar ratio was varied from 1:1 to 1:10 (Fig. 1).
It can be noticed that increase of molar ratio up to 1:4 led to higher conversions (after 48 h maximum conversion of 27.3% was achieved).
Further increase caused steep decrease in accomplished conversions.This is in accordance with some previous reports, and it was probably due to diffusion limitation induced by linoleic acid high viscosity and/or reduced solubility of vitamin C in reaction medium with higher log P value (Burham et al., 2009).
Furthermore, influence of limiting substrate concentration on both conversion degree and product yield was investigated.Lascorbic acid concentration was varied from 0.05 to 0.2 M and obtained results are depicted in Figs.2a and 2b.In Fig. 2a it is presented that highest conversion (62.0%) was obtained at lowest concentration of vitamin C (0.05 M), while the same reaction parameter sho-wed opposite effect on yield of ester (maximum of 50.0 mM of ascorbyl linoleate was synthesized at 0. On the other hand, lipases are known to keep their active conformation only if they are surrounded by monomolecular layer of water (Hsieh et al., 2005).In accordance with that, initial water content was optimized (varied in range 0.02-0.12%(v/v)) and influence of this parameter on reaction progress was monitored.Highest conversion was achieved at 0.07% (v/v) of water (Fig. 3).Although some previous researches implied that Novozym ® 435 contains sufficient amount of water for preservation of its catalytic activity (Salis et al., 2005;Tamalampudi et al., 2008), our study showed that some water in the reaction mixture had positive effect.
Finally, loading of the immobilized lipase was optimized.Concentrations of the catalyst from 3-7 g/l were examined.It has been shown (Fig. 4a) that an increase in lipase loading has positive effect on limiting substrate molar conversion.
However, price of the enzyme is the greatest of all expenses of the process and therefore, its consumption should be minimized.
Hence, yield of the product per mass of lipase (specific yield) was also taken into consideration.As it can be seen in Fig. 4b, maximum specific yield of 9.7 mmol of ester per gram of lipase was accomplished at lowest enzyme concen-tration -3 g/l.

Spectral analyses
L-ascorbyl linoleate synthesized under optimized conditions (0.15 M vitamin C, 1:4 molar ratio, 0.07% (v/v) of water and 3 g/l of enzyme) was collected as a pure substance using semipreparative RP-HPLC and subjected to spectral analyses.
Following NMR spectrum was determined:

DPPH analyses
Primary function of L-ascorbyl linoleate is to prevent deterioration of lipids in oils and emulsions.
To prove that antioxidant properties of vitamin C are preserved after its esterification with linoleic acid, standard method which determines capacity of compound for scavenging DPPH radical was employed and obtained result is expressed as its IC 50 value.
As it can be observed from

CONCLUSIONS
Main goals of this research were examination of the influence of key experimental factors on lipase-catalyzed synthesis of ascorbyl linoleate in acetone, product purification, structural characterization and determination of its antioxidant and diffusion properties.Obtained results are comparable to those achieved by other authors with more expensive solvents and/or activated acyl donors and hence, could be used in further optimization studies or in investigations on larger scale.Moreover, determined antioxidant capacity and diffusion coefficient recommend L-ascorbyl linoleate for application in food, cosmetics and pharmaceutics.

Scheme 1 .
Scheme 1. Lipase-catalyzed esterification of vitamin C with linoleic acid

Figure 1 .Figure 2 .Figure 3 .Figure 4 .
Figure 1.Influence of substrate molar ratio on molar conversion.Experiments were conducted with 0.15 M of vitamin C, 5 g/l of lipase, and 0.07 % (v/v) of water

Table 1 .
DPPH radical scavenging capacity of fatty acid ascorbyl esters

Table 1
(Austria et al., 1997)s in Franz cell were conducted.Based on results presented in Table2., it could be noticed that diffusion coefficient of linoleate was 40% higher than palmitate and 62% higher than oleate, while stearate was not detected in samples taken from receptor compartment within all 24 hours.Such behavior is most likely caused by multiple factor effect, including solubility, molecular weight, log P, etc. Obtained results indicate that L-ascorbyl linoleate is a promising molecule for cosmetic formulations with the function of delivering active substance into deeper skin layers.For its final characterization, experiments with human skin model membranes are still to be performed in the future.Also, different dermocosmetic preparations should be tested since stability and diffusion properties of fatty acid ascorbyl esters are dependent on formulation structural properties(Austria et al., 1997).