- Email: [email protected]
Synthesis and characterization of arabinose-palmitic acid esters by enzymatic esterification
Molecular Catalysis 433 (2017) 383–390
Contents lists available at ScienceDirect
Molecular Catalysis journal homepage: www.elsevier.com/locate/mcat
Editor’s choice paper
Synthesis and characterization of arabinose-palmitic acid esters by enzymatic esterification Valeria M. Pappalardo a , Carmen G. Boeriu b,∗ , Federica Zaccheria a , Nicoletta Ravasio a,∗ a b
CNR, Institute of Molecular Sciences and Technologies (ISTM), via C. Golgi 19, 20133 Milan, Italy Wageningen Food & Biobased Research, Bornse Weilanden 9, 6708 WG Wageningen, Netherlands
a r t i c l e
i n f o
Article history: Received 12 January 2017 Received in revised form 14 February 2017 Accepted 20 February 2017 Keywords: Sugar esters Enzymatic catalysis Amphiphilic compounds Biobased products Bio-surfactants
a b s t r a c t The direct esterification of palmitic acid with l-(+)-arabinose has been carried out. The use of Candida antartica lipase B as the catalyst and the choice of suitable solvent and experimental conditions allowed carrying out the reaction successfully. In particular 10% dimethyl-sulfoxide in tert-butanol was found to be the optimal solvent. The product has been fully characterized by means of FTIR, ESI-MS, DSC, mono and bidimensional 1 H and 13 C NMR. These techniques confirm that only the primary alcoholic group was involved in the esterification reaction. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Sugar fatty acid esters are amphiphilic compounds widely employed as food-grade, odourless and tasteless, non-ionic surfactants in a wide range of industrial formulations (chemicals, pharmaceuticals, cosmetics, detergents, oral-care products, lowcalorific sweeteners, medical supplies) in place of the synthetic ones [1]. The use of synthetic surfactants is, indeed, connected with several human health risks, including allergic reactions, eyes and skin irritations, intestinal disorders, cancer [2,3]. By contrast, sugar fatty acid esters are safe for both humans and the environment thanks to the lower toxicity and the higher biodegradability and biocompatibility. Moreover, they are easily digested as a mixture of sugars and fatty acids in the stomach [4] and some of them showed antimicrobial [5,6], anticancer [7] and insecticidal [8] activity. The surfactant properties of these compounds may be predicted by using the hydrophobic-lipophilic balance (HLB) that was introduced by Griffin [9]. The HLB of non-ionic sugar based surfactants depends on the degree of substitution and the ratio between the hydrophilic head (mono-, oligo- or polysaccharides) and the lipophilic tail (amount and length of the fatty acid chain). Interestingly, in contrast to conventional available food emulsifiers it is possible to produce sugar fatty acid esters that cover a very broad
∗ Corresponding authors. E-mail addresses: [email protected] (C.G. Boeriu), [email protected], [email protected] (N. Ravasio). http://dx.doi.org/10.1016/j.mcat.2017.02.029 2468-8231/© 2017 Elsevier B.V. All rights reserved.
HLB range by modifying the sugar head and/or the length and the unsaturation of the fatty acid tail [10]. Thanks to their typical features, the production of sugar-based surfactants on an industrial scale has been attracting growing attention since 1990. Today, a wide number of them are commercially available as emulsifiers and stabilizers. For example, sucrose esters of lauric, palmitic or stearic acid (marked as E473) have been classified as Generally Recognized As Safe, GRAS [11] and can be used in cosmetic formulations because of their mild dermatological properties or in molecular cuisine to produce airs and foams [12]. Sugar fatty acid esters can be produced from renewable and low-cost raw materials (viz carbohydrates, fatty acids or oils) by using both chemical and enzymatic esterification. The chemical route, that is transesterification of the fatty acid methyl ester with the sugar, is the most widespread at an industrial level. However, not only this reaction is poorly selective, but it also involves the use of harmful solvent and production of organic and inorganic wastes [13] difficult to remove from the product. To overcome these limitations, the enzymatic route is an attractive and eco-friendly alternative that requires mild conditions of temperature and pressure. In addition, an important advantage is the higher regioselectivity which leads to a predominant reaction product [1,13]. Among the enzymes, lipases have been widely used to promote the ester bond formation because of their high substrate specificity. Lipases are efficient catalysts for the hydrolysis of esters; however, in the absence of water in neat organic solvents they can catalyse the reverse reaction. To enhance the stability in organic solvents, immobilized lipases have been employed. At the same time, immo-
384
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
bilization allows an easy separation and recover of the enzyme [14]. Several studies had been reported about the enzymatic transesterification of mono- and disaccharide (viz d-(+)-glucose, d-(−)-fructose, sucrose) and fatty acid esters [7,15–17]. In this paper we wish to report on the preparation of l-(+)arabinose-palmitic acid monoesters by using immobilized Candida antarctica lipase B (CALB, Novozyme 435) as a biocatalyst, following the previous research of some of us about the enzymatic synthesis of oligofructose fatty acids monoesters [1]. The effect of the reaction medium, temperature and molar ratio of reagents that strongly affect the activity and specificity of lipases has been investigated. The product was purified by flash chromatography and characterized by NMR, FTIR, ESI-MS analysis. The thermal properties were evaluated by DSC analysis. Among the sugars, we chose l-(+)-arabinose because it is a constitutive component of arabinoxylans and other naturally occurring non-starch polysaccharides, such as pectin and lignocellulose. These biopolymers are easily extracted with water from several biomass residues of the agri-food industry, particularly flaxseed meal, rice bran, wheat bran, barley, corn fiber, and display interesting physiological and functional properties [18]. In this way, we have a natural source for biobased arabinose and the derivatives that can be esterified to give amphiphilic molecules for food and non-food industrial applications. To the best of our knowledge, the direct esterification of l-(+)-arabinose and free palmitic acid has never been investigated up to now. Only a few studies on the synthesis of d-(−)-arabinose esters [19,20] and the direct esterification of other underivatized sugars and free fatty acids have been reported [14]. 2. Experimentals 2.1. Materials and methods Commercial immobilized lipase from Candida antartica lipase B (CALB, Novozyme 435) was a generous gift from Novozymes (Bagsvard, Denmark). Unless otherwise stated, all chemicals were purchased from Sigma-Aldrich and/or from VWR International and were used without further purification. 2.2. Synthesis of L-(+)-arabinose-palmitic acid esters l-(+)-arabinose and palmitic acid (different sugar/acid molar ratios between 1 and 3) were mixed in a 20 ml glass reaction tube by using a magnetic stirrer (450 rpm) changing the solvent (tert-butyl alcohol, TBU, and methyl tert-butyl ether, MTBE, alone or in binary mixture with dimethyl sulfoxide, DMSO, or 200 mM phosphate buffer pH 7,) and the temperature (40–60 ◦ C). After 30 min, molecular sieve (3 Å, 10% w/wpalmiticacid ) and CALB (8% w/wpalmiticacid ) were added to start the reaction and the mixture was stirred over night at the selected temperature. Control experiments were carried out without the enzyme. Then the mixture was cooled down to room temperature to stop the reaction and centrifuged. The solid was shown by IR to contain only the enzyme, molecular sieves and unreacted l-(+)-arabinose, therefore it was discarded. The solvent was removed under reduced pressure. The crude product was analysed by Thin Layer Chromatography (TLC), attenuated total reflection (ATR) Fourier transform infrared (FTIR) and GC-FID analysis. For the semi-preparative scale, in a 250 ml round bottom flask l-(+)-arabinose (1.0 g, 6.7 mmol) and palmitic acid (1.7 g, 6.7 mmol) were mixed at 60 ◦ C by using a magnetic stirrer (450 rpm) in TBU (35 ml) as the solvent, alone or in binary mixture with DMSO (7% and 10%). After 30 min, molecular sieves (3 Å, 10%) and CALB (140 mg) were added and the mixture was stirred o.n. at 60 ◦ C.
The reaction was controlled by using TLC. After cooling to room temperature, the mixture was centrifuged. The solid phase was discarded and the upper phase was dried under vacuum to give a white and odourless powder. The crude product was purified by flash chromatography and characterized by ATR-FTIR, ESI-MS, NMR and GC-FID analyses. 2.3. Thin layer chromatography Analytical Thin Layer Chromatography TLC was performed on Silica Gel 60 F254 precoated aluminum sheets (0.2 mm layer; Merck, Darmstadt, Germany). Components were separated by using CH2 Cl2 :MeOH (12:1) and detected by spraying the sheet with a p-anisaldehyde/H2 SO4 solution (ethanol:anisaldehyde:sulphuric acid:acetic acid, 90:5:5:1) followed by heating at 105 ◦ C for 5–10 min for the detection of the sugars. At the same time, another plate was developed with a ceric sulfate/ammonium molybdate or with a KMnO4 solution and heated at about 150 ◦ C for few seconds to detect also palmitic acid. The Rf of monoester was 0.2. 2.4. Purification methods Purification of products was accomplished by flash chromatography (silica gel 60, 40–63 mm, Merck). The crude product was eluted on a silica gel column with ethyl ether:n-hexane:formic acid (1:1:0.02), to remove unreacted palmitic acid and l-(+)-arabinose, followed by elution with ether:n-hexane:methanol (4.7:4.7:0.6). 2.5. Gas chromatography analysis The monoester content in the crude of reaction was evaluated according to the GC methods of Cramer et al. [21] with slight modification. The analysis of arabinose, palmitic acid and reaction mixtures was performed on an Agilent 6890 Gas Chro® matography system by using a Alltech Heliflex AT-5 capillary column (30 m × 0.32 mm ID × 0.25 m), with split injection and FID detection. Before injection, unsubstituted hydroxyl groups were silylated according to Degn et al. [22]. To 5–10 mg of crude reaction, 110 L of heptane:pyridine (2:1 v/v) containing octyl-ˇ-D-glucopyranoside (OGP, 5.9 mg/ml, 0.02 mol/ml) and heptadecane (4.66 mg/ml, 0.019 mol/ml) as internal standards was added together with 100 L BSTFA (1%, v/v TMCS). The mixture was stirred at 70 ◦ C for 30 min and, subsequently, 1 L of the resulting solution was injected with a 10 L glass syringe (Hamilton). The injection temperature was 220 ◦ C and the detector temperature was 250 ◦ C. A pressure of 21 psi, a gas flow of 1.7 mL/min (He) and a split flow of 106 mL/min were applied. The temperature profile of the analysis oven was: 1 min at 90 ◦ C, heating to 250 ◦ C at a rate of 30 ◦ C/min and kept constant for 4 min, finally heating to 310 ◦ C at a rate of 15 ◦ C/min and kept constant for 5 min. OGP was used as internal standard to evaluate the monoester concentration and compare the yields of the crude products. 2.6. Structural characterization of 5-O-palmitoyl-l-(+)-arabinosyl ester Mass Spectroscopy – Mass spectrum of monoester was recorded ® on a Q-TOF Micro-Waters Mass Spectrometer interfaced to an Electro Spray Ionization (ESI) by using a positive ion mode. The sample was injected to the mass spectrometer in a MeOH solution. NMR Spectroscopy – High-resolution 1 H and 13 C NMR spectra were acquired at 600.33 and 150.95 MHz, respectively, on a Bruker Advance 600 spectrometer (Bruker, Karlsruhe, Germany) interfaced with a workstation running a Windows operating system and equipped with a TOPSPIN software package. 20 mg of monoester was dissolved in 0.6 ml of DMSO-d6 and the spectra were recorded
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
at 27 ◦ C. Chemical shifts (ı) were given in parts per million (ppm) and referenced to the solvent signals [ıH 2.50 and ıC 39.50 ppm from Tetramethylsilane (TMS)]. 13 C NMR signal multiplicities were based on attached proton test (APT) spectra and assigned on the basis of 1 H-13 C correlation experiments (Heteronuclear Multiple Quantum Correlation spectroscopy, HMQC, and Heteronuclear Multiple Bond Correlation spectroscopy, HMBC). 1 H signals were assigned by using 1 H–1 H correlation experiments (Correlation SpectroscopY, COSY, and TOtal Correlation SpectroscopY, TOCSY). ATR-FTIR spectra – ATR-FTIR spectra were recorded on a Biorad FTS-40 spectrophotometer equipped with a Specac Golden Gate ATR platform with diamond crystal. DSC-analysis – Differential Scanning Calorimetry (DSC) measurements were carried out on 5 mg of sample, by using a DSC 3 Mettler Toledo Analytical Instrument, from 30 to 250 ◦ C at a heating rate of 5 K min−1 under a synthetic air atmosphere.
3. Results and discussion We have focused on the preparation of l-(+)-arabinose/palmitic acid monoesters by using immobilized Candida antarctica lipase B (Novozyme 435) as a biocatalyst. In the enzymatic esterification the substitution should involve the primary hydroxyl group of the sugar which is more accessible than the secondary groups to the active site of the enzyme (Scheme 1). Since the activity and specificity of lipase are strongly dependent on temperature, reaction medium and molar ratio of reagents, we investigate the optimal esterification conditions by using commercial immobilized lipase B from Candida antarctica in accordance with a previous work of some of us [1] about the preparation of oligofructose fatty acids monoesters. The temperature affects the solubility of the substrates and products, the activity of the enzyme and, therefore, the reaction rate and the position of the equilibrium [16]. Lipases are known for their thermal stability especially after immobilization. For example, Novozyme 435 can be used up to 80 ◦ C without notable loss of activity [23]. Although each enzyme has an optimum working temperature, the optimum reaction temperature may be different. The reaction medium must be able to dissolve enough amounts of both the substrates and, at the same time, it should not affect lipase activity and selectivity. As stated above, the esterification is usually promoted by using organic solvents. Among them tertbutanol (TBU) and methyl tert-butyl ether (MTBE) are widely employed. However, the main problem is the low solubility of sugars. To increase the sugar solubility, hydrophilic cosolvent (such as dimethylsulfoxide, dimethylformamide or pyridine) could be used. Nevertheless, these solvents have a strong inactivating effect on the enzyme activity because they interact and strip off water from the protein surface and the active site, imposing a higher structural rigidity [24]. The presence of low water amounts is indeed required to keep the enzyme active [14], although higher water levels would favour the hydrolysis rather than the esterification. To remove the water excess, which is generated during the process as a by-product, evacuation under vacuo, open test tubes, molecular sieves [16], reduce pressure [25] or azeotropic distillation [26] have been used. To optimize temperature and reaction medium, we performed first trial reactions in a 12-Reactor Carousel Reaction System by using equimolar amount of l-arabinose and palmitic acid in presence of CALB and molecular sieves. The reactions were carried out at two different temperatures, 40 and 60 ◦ C, in TBU and MTBE alone or in binary mixture with DMSO or phosphate buffer at pH 7 (that is close to the optimum pH of the enzyme). The conversion of larabinose was evaluated by TLC and FTIR analysis. The IR spectra of the crude products (Fig. 1) showed the typical vibration of the
385
ester group (1734 cm−1 ) in all the mixtures containing TBU alone or in mixture with DMSO (results not reported). On the contrary this vibration was not identified in the reactions performed in MTBE or in the presence of phosphate buffer. In particular, the highest intensity of both TLC spots and IR ester vibration was observed at 60 ◦ C by using TBU and 10% DMSO as cosolvent. The highest temperature, indeed, should increase solubility of both substrates together with lipase activity and reaction rate. At the same time, also DMSO should increase arabinose solubility. Sugar solubility is the main driving force of this reaction, but the amount of DMSO in the solvent mixture has to be carefully tuned as not only this solvent will inactivate the enzyme, but it has also been shown that lowering the percentage of DMSO will promote formation of the diesters with respect to monoesters [27,28]. We therefore carried out a second trial at 60 ◦ C to investigate the influence of the amount of DMSO in the solvent mixture and the influence of the amount of l-arabinose (Table 1). Since the esterification is an equilibrium reaction, an excess of one of the reagents should strongly influence sugar conversion [24]. The reactions were performed by using three sugar/acid molar ratios, 1:1, 2:1 and 3:1 and followed by TLC. The crude products were compared by FTIR (Fig. 2) and the yields were determined by GC-FID analysis. The maximum yield was obtained by using 10% DMSO in TBU (Table 1, entry 3). For higher percentage of DMSO, both conversion and selectivity to monoester collapsed thus leading to a much lower yield (Table 1, entry 4) due to inactivation of the catalyst. Alike, the catalyst reused showed a significant loss in activity (see SI). On the other hand the use of 7% DMSO in TBU as reaction medium and different amount of l-(+)-arabinose (Table 1, entries 5 and 6) to increase CALB activity and the sugar conversion by reducing the DMSO amount, showed that the yield can be increased by using a double amount of arabinose (Table 1, entry 2 vs 5). The progress of the reactions was evaluated by TLC and the dried products were analysed by FTIR and GC-FID. All the reactions were stopped at 24 h, since a light increase of monoester was observed at longer reaction time, associated with the enhancement of byproducts (evaluated by using the spot intensities of TLC). The peaks area were compared to a standard curve, that was obtained by using different concentrations of 5-O-Palmitoyl-l(+)-arabinose, previously purified by flash chromatography and subsequently silylated at the unsubstituted hydroxyl groups before injection. In particular, the elution time of monoester was 12.43 and 13.00 min (as a mixture of ␣ and  anomer), whereas the elution time of unreacted l-arabinose and palmitic acid were 4.86–4.98 min (␣ and  anomer) and 6.46 min, respectively. The chromatograms showed an increasing amount of by-products with the increase of the sugar excess. The formation of monoester is less dependent on the molar ratio of reagents. To isolate and characterize the products and understand whether the monoester substitution involves the primary hydroxyl group, we performed the reaction in a semi-preparative scale at the previously optimized conditions (1:1 molar ratio, 60 ◦ C, TBU/10%DMSO, Novozyme 435, molecular sieves). The crude was purified by flash chromatography, and 5-O-Palmitoyl-l-(+)arabinose was isolated as a white powder in 20% yield. In the ESI-MS spectrum (Fig. 3) the peak with the greatest intensity, the Base Peak, at m/z 411.57 corresponds to the molecular ion [M + Na]+ . Moreover, the peaks at m/z 443.63 and m/z 800.14 represent sodium adducts of the molecular ions species [M + MeOH + Na]+ and [2 M + Na]+ , respectively. This is consistent with the expected molecular mass for 5-O-Palmitoyll-(+)-arabinose, 388.54 g/mol. In the 1 H and 13 C NMR spectra two sets of proton and carbon signals derived from arabinofuranoside moiety were detected, suggesting that the product was a mixture of ␣- and -monoester. The
386
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
Scheme 1. Esterification reaction of palmitic acid with l(+)- arabinofuranose.
Fig. 1. Solvent effect in the esterification of palmitic acid with arabinose.
Table 1 Yield in arabinose ester under different conditions (palmitic acid 1.0 mmol, 256 mg; Candida antartica Lipase B (CALB) Novozyme 435, 8% w/wpalmiticacid , 20 mg; molecular sieves (3 Å, 10%), 60 ◦ C, 24 h). Entry 1 2 3 4 5 6
Molar ratio Ara/PA
l-(+)-arabinose (mg)
Solvent (l) TBU
DMSO
Yield (%)
Palmitic acid Conv %
Sel %
1:1 1:1 1:1 1:1 2:1 3:1
150 150 150 150 300 450
5000 4650 4500 4250 4650 “
– 350 500 750 350 “
6.13 7.83 20.72 5.50 9.32 7.82
7.70 13.91 21.69 10.31 12.26 18.07
79.61 56.29 95.53 53.35 76.02 43.28
spin system for each anomer was assigned with the aid of 1D and 2D NMR spectra and literature data [29]. As previously reported, although the chemical shifts of proton at H-1 position of ␣- and -arabinose rings are clearly separated, 1 H NMR spectrum is not suitable to assign the anomeric configuration as the peaks showed very close J1,2 coupling constants (Table 2 and Figs. S1–S7). Moreover, the other 1 H NMR signals are partially overlapped each other. On the other hand, 13 C NMR spectroscopy gives a clear distinction of the signals of the two anomers. Indeed, the 1,2-trans configuration of ␣-arabinose shifts the C-1 resonance signal to lower field (of about 7 ppm) from that of 1,2-cis configuration of -arabinose [29]. For these reasons, in the 13 C spectrum, signals at ␦ 102.53 and 96.55 were assigned at ␣-monoester and -monoester, respectively. The
others 13 C and 1 H chemical shifts were attributed on the basis of correlations in 2D NMR spectra, HSQC, COSY and TOCSY. The relative proportion of ␣- to - anomers was estimated of roughly 3 to 1 by integrating the intensity of the 2 typical proton signals at ␦ 5.01 (33%) and 4.92 (67%) in the 1 H NMR spectrum that correspond to - and ␣-arabinose rings, respectively. Compared to unsubstituted l-(+)-arabinose, the chemical shift of protons at 5 position of both anomers is shifted to lower fields because of the deshielding effect of carbonyl group. Similarly, all 1 H-13 C correlations in the 2D HSQC NMR spectrum are shifted to lower fields. These results confirm that the substitution occurred only at 5 position.
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
Fig. 2. Effect of TBU/DMSO ratio and alcohol/acid in the esterification of palmitic acid with arabinose.
Fig. 3. ESI-MS spectrum of 5-O-Palmitoyl-l-(+)-arabinose. The peaks at m/z 315.50 and 259.39 derived from the ionization source of the mass spectrometer.
387
388
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
Table 2 13 C and 1 H chemical shifts values of 5-O-Palmitoyl-l-(+)-arabinose.a C NMR (150.95 MHz) ␦13 C (ppm) ␣-anomer
-anomer
1 2 3 4 5a 5b C1 OH C2 OH C3 OH
102.51 79.96 77.62 83.03 64.65
96.53 77.23 75.93 79.58 66.34
Fatty acid moiety
␣-anomer
-anomer
C O C2 C3 C4 C5 and C13 C6 C7-12 C14 C15 C16
173.28 33.89 24.89 28.91 29.14 29.33 29.48, 29.45 31.74 22.53 14.36
173.31
13
Sugar moiety
a
H NMR (600.33 MHz,) ␦1 H (ppm) ␣ anomer 1
4.93 (d, 1H, J = 2.8 Hz) 3.93 (dd, 1H, J = 6.9, 2.3 Hz) 3.60 (dd, 1H, J = 6.9, 5.1 Hz) 3.72 (dd, 1H, J = 5.0, 2.8 Hz) 4.20 (dd, 1H, J = 11.1, 2.5 Hz) 3.98–3.94 (m, 1H) 6.26 (br.s, 1H) 5.34 (br.s, 1H) 3.43 (br.s, 1H)
 anomer 5.02 (d, 1H, J = 4.4 Hz) 3.70 (dd, 1H, J = 6.5, 4.4 Hz) 3.79 (t, 1H, J = 6.3 Hz) 3.66 (m, 1H, J = 7.6, 6.1, 3.8 Hz) 4.20– 4.17 (m, 1H) 3.97– 3.91 (m, 1H)
2.29 (t, 2H, J = 7.4 Hz) 1.51 (quint, 2H, J = 7.2 Hz) 1.30–1.20 (m, 24H)
0.85 (t, 3H, J = 7.0 Hz)
Abbreviations: s = singlet; d = doublet; t = triplet; quint = quintet; dd = doublet of doublets; m = multiplet; br.s = broad signal.
Fig. 4. Overlapped ATR-FTIR spectra of palmitic acid, l-(+)-arabinose and 5-O-Palmitoyl-l-(+)-arabinose.
The infrared spectra of palmitic acid, l-(+)-arabinose and 5-OPalmitoyl-l-(+)-arabinose are shown in Fig. 4. In particular, the vibration absorbance at 1736 cm−1 of the purified product suggests that the esterification has occurred. Moreover, the typical vibrations of both starting materials are evident in this spectrum: a broad OH stretching at 3400 cm−1 , characteristic of l-(+)-arabinose,
and intense peaks at 3000–2700 cm−1 , typical of the methyl and methylene groups of palmitic acid. Fig. 5 shows the DSC curve for 5-O-Palmitoyl-l-(+)-arabinose. After the glass transition of the amorphous component at about 55 ◦ C, the curve presents an endothermic peak at 79.6 ◦ C that should correspond to the melting of the crystalline fraction. The melting is followed by an endothermic oxidative decomposition that
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
389
Fig. 5. DSC curve of 5-O-Palmitoyl-l-(+)-arabinose.
is perhaps caused by the sugar moiety. Melting temperature of monoester is lower than that of pure l-arabinose (164–165 ◦ C, literature data) probably because of the presence of the lipophilic tail of palmitic acid, which has the melting point at 62.9 ◦ C. 4. Conclusions The direct esterification of l-(+)-arabinose and free palmitic acid has been carried out in the presence of Novozyme 435 by carefully tuning experimental conditions. The synthesized 5-O-Palmitoyl-l-(+)-arabinose monoester could be used as safe non-ionic surfactant in a wide range of industrial formulations. Moreover, the ability of Novozyme 435 to recognize l-(+)-arabinose may be useful for the preparation of food-grade arabinoxylan-palmitic acid conjugates that may find manifold food and non-food industrial application in place of the synthetic esters. Acknowledgements The authors thank COST Action TD1203, EUBIS for a short term mobility grant to V.M.P. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mcat.2017.02. 029. References [1] S.E.H.J. Van Kempen, C.G. Boeriu, H.A. Schols, P. De Waard, E. Van Der Linden, L.M.C. Sagis, Novel surface-active oligofructose fatty acid mono-esters by enzymatic esterification, Food Chem. 138 (2–3) (2013) 1884–1891.
[2] P. Floyd, P. Zarogiannis, K. Fox, Non-surfactant organic ingredients and zeolite-based detergents Final report prepared for the European Commission, Risk & Policy Analysts Limited (RPA), 2006. [3] P.B. Singh, H.S. Saini, Exploitation of agro-industrial wastes to produce low-cost microbial surfactants, in: S.K. Brar, G.S. Dhillon, C.R. Soccol (Eds.), Biotransformation of Waste Biomass into High Value Biochemicals, Springer, New York, New York, NY, 2014, pp. 445–471. [4] T. Vijai Kumar Reddy, G. Sandhya Rani, R.B.N. Prasad, B.L.A. Prabhavathi Devi, Green recyclable SO3 H-carbon catalyst for the selective synthesis of isomannide-based fatty acid monoesters as non-ionic bio-surfactants, RSC Adv. 5 (2015) 40997–41005. [5] T. Watanabe, S. Katayama, M. Matsubara, Y. Honda, M. Kuwahara, Antibacterial carbohydrate monoesters suppressing cell growth of Streptococcus mutans in the presence of sucrose, Curr. Microbiol. 41 (3) (2000) 210–213. [6] L. Xin, Antimicrobial structure-efficacy relationship of sugar fatty acid esters, J. Chem. Pharm. Res. 6 (5) (2014) 944–946. [7] S.W. Chang, J.F. Shaw, Biocatalysis for the production of carbohydrate esters, N. Biotechnol. 26 (3–4) (2009) 109–116. [8] M. Ferrer, J. Soliveri, F.J. Plou, N. López-Cortés, D. Reyes-Duarte, M. ˜ A. Ballesteros, Synthesis of sugar esters in Christensen, J.L. Copa-Patino, solvent mixtures by lipases from Thermomyceslanuginosus and Candida antarctica B, and their antimicrobial properties, Enzyme Microb. Technol. 36 (4) (2005) 391–398. [9] W.C. Griffin, Classification of surface-active agents by HLB, J. Soc. Cosmet. Chem. 1 (1949) 311–326. [10] N.R. Pedersen, R. Wimmer, J. Emmersen, P. Degn, L.H. Pedersen, Effect of fatty acid chain length on initial reaction rates and regioselectivity of lipase-catalysed esterification of disaccharides, Carbohydr. Res. 337 (13) (2002) 1179–1184. [11] GRAS Notice (GRN) No. 514. 2014. http://www.fda.gov/Food/ IngredientsPackagingLabeling/GRAS/NoticeInventory/default.htm. [12] http://www.molecularrecipes.com/hydrocolloid-guide/sucrose-ester-sucro2/. [13] I.S. Yoo, S.J. Park, H.H. Yoon, Enzymatic synthesis of sugar fatty acid esters, J. Ind. Eng. Chem. 13 (1) (2007) 1–6. [14] L.A.M. Van Den Broek, C.G. Boeriu, Enzymatic synthesis of oligo- and polysaccharide fatty acid esters, Carbohydr. Polym. 93 (1) (2013) 65–72. [15] J.F. Kennedy, H. Kumar, P.S. Panesar, S.S. Marwaha, R. Goyal, A. Parmar, S. Kaur, Enzyme-catalyzed regioselective synthesis of sugar esters and related compounds, J. Chem. Technol. Biotechnol. 81 (6) (2006) 866–876. [16] A.M. Gumel, M.S.M. Annuar, T. Heidelberg, Y. Chisti, Lipase mediated synthesis of sugar fatty acid esters, Process Biochem. 46 (11) (2011) 2079–2090.
390
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
[17] T. Polat, R.J. Linhardt, Syntheses and applications of sucrose-based esters, J. Surfactants Deterg. 4 (4) (2001) 415–421. [18] M.S. Izydorczyk, Arabinoxylans, in: G.O. Phillips, P.A. Williams (Eds.), Handbook of Hydrocolloids, 2nd edition, Woodhead Publishing Cambridge, England, 2009, pp. 653–692. [19] A. Sharma, S. Chattopadhyay, Lipase catalysed acetylation of carbohydrates, Biotechnol. Lett. 15 (11) (1993) 1145–1146. [20] W. Tsuzuki, Y. Kitamura, T. Suzuki, S. Kobayashi, Synthesis of sugar fatty acid esters by modified lipase, Biotechnol. Bioeng. 64 (3) (1999) 267–271. [21] J.F. Cramer, M.S. Dueholm, S.B. Nielsen, D.S. Pedersen, R. Wimmer, L.H. Pedersen, Controlling the degree of esterification in lipase catalysed synthesis of xylitol fatty acid esters, Enzyme Microb. Technol. 41 (3) (2007) 346–352. [22] P. Degn, L.H. Pedersen, J. Duus, W. Zimmermann, Lipase-catalysed synthesis of glucose fatty acid esters in tert-butanol, Biotechnol. Lett. 21 (4) (1999) 275–280. [23] Á. Cruz-Izquierdo, L.A.M. van den Broek, J.L. Serra, M.J. Llama, C.G. Boeriu, Lipase-catalyzed synthesis of oligoesters of 2,5-furandicarboxylic acid with aliphatic diols, Pure Appl. Chem. 87 (1) (2015) 59–69. [24] R. Croitoru, F. Fit¸igˇau, L.A.M. Van Den Broek, A.E. Frissen, C.M. Davidescu, C.G. Boeriu, F. Peter, Biocatalytic acylation of sugar alcohols by 3-(4-hydroxyphenyl)propionic acid, Process Biochem. 47 (12) (2012) 1894–1902.
ˇ [25] S. Sabeder, M. Habulin, Zˇ . Knez, Lipase-catalyzed synthesis of fatty acid fructose esters, J. Food Eng. 77 (4) (2006) 880–886. [26] Y. Yan, U.T. Bornscheuer, L. Cao, R.D. Schmid, Lipase-catalyzed solid-phase synthesis of sugar fatty acid esters – removal of byproducts by azeotropic distillation, Enzyme Microb. Technol. 25 (8–9) (1999) 725–728. [27] R. ter Haar, H.A. Schols, L.A.M. van den Broek, D. Sa˘glam, A.E. Frissen, C.G. Boeriu, Harry Gruppen, Molecular sieves provoke multiple substitutions in the enzymatic synthesis of fructose oligosaccharide-lauryl esters, J. Mol. Catal. B: Enzym. 62 (2010) 183–189. [28] F.J. Plou, M.A. Cruces, M. Ferrer, G. Fuentes, E. Pastor, M. Bernabé, M. Christensen, F. Comelles, J.L. Parra, A. Ballesteros, Enzymatic acylation of diand tri-saccharides with fatty acids: choosing the appropriate enzyme, support and solvent, J. Biotechnol. 96 (2002) 55–66. [29] T. Usui, S. Tsushima, N. Yamaoka, K. Matsuda, K. Tuzimura, H. Sugiyama, S. Seto, K. Fujieda, G. Miyajima, Determination of anomeric configuration of furanoside derivatives by carbon-13 NMR, Agric. Biol. Chem. 38 (7) (1974) 1409–1410.
Contents lists available at ScienceDirect
Molecular Catalysis journal homepage: www.elsevier.com/locate/mcat
Editor’s choice paper
Synthesis and characterization of arabinose-palmitic acid esters by enzymatic esterification Valeria M. Pappalardo a , Carmen G. Boeriu b,∗ , Federica Zaccheria a , Nicoletta Ravasio a,∗ a b
CNR, Institute of Molecular Sciences and Technologies (ISTM), via C. Golgi 19, 20133 Milan, Italy Wageningen Food & Biobased Research, Bornse Weilanden 9, 6708 WG Wageningen, Netherlands
a r t i c l e
i n f o
Article history: Received 12 January 2017 Received in revised form 14 February 2017 Accepted 20 February 2017 Keywords: Sugar esters Enzymatic catalysis Amphiphilic compounds Biobased products Bio-surfactants
a b s t r a c t The direct esterification of palmitic acid with l-(+)-arabinose has been carried out. The use of Candida antartica lipase B as the catalyst and the choice of suitable solvent and experimental conditions allowed carrying out the reaction successfully. In particular 10% dimethyl-sulfoxide in tert-butanol was found to be the optimal solvent. The product has been fully characterized by means of FTIR, ESI-MS, DSC, mono and bidimensional 1 H and 13 C NMR. These techniques confirm that only the primary alcoholic group was involved in the esterification reaction. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Sugar fatty acid esters are amphiphilic compounds widely employed as food-grade, odourless and tasteless, non-ionic surfactants in a wide range of industrial formulations (chemicals, pharmaceuticals, cosmetics, detergents, oral-care products, lowcalorific sweeteners, medical supplies) in place of the synthetic ones [1]. The use of synthetic surfactants is, indeed, connected with several human health risks, including allergic reactions, eyes and skin irritations, intestinal disorders, cancer [2,3]. By contrast, sugar fatty acid esters are safe for both humans and the environment thanks to the lower toxicity and the higher biodegradability and biocompatibility. Moreover, they are easily digested as a mixture of sugars and fatty acids in the stomach [4] and some of them showed antimicrobial [5,6], anticancer [7] and insecticidal [8] activity. The surfactant properties of these compounds may be predicted by using the hydrophobic-lipophilic balance (HLB) that was introduced by Griffin [9]. The HLB of non-ionic sugar based surfactants depends on the degree of substitution and the ratio between the hydrophilic head (mono-, oligo- or polysaccharides) and the lipophilic tail (amount and length of the fatty acid chain). Interestingly, in contrast to conventional available food emulsifiers it is possible to produce sugar fatty acid esters that cover a very broad
∗ Corresponding authors. E-mail addresses: [email protected] (C.G. Boeriu), [email protected], [email protected] (N. Ravasio). http://dx.doi.org/10.1016/j.mcat.2017.02.029 2468-8231/© 2017 Elsevier B.V. All rights reserved.
HLB range by modifying the sugar head and/or the length and the unsaturation of the fatty acid tail [10]. Thanks to their typical features, the production of sugar-based surfactants on an industrial scale has been attracting growing attention since 1990. Today, a wide number of them are commercially available as emulsifiers and stabilizers. For example, sucrose esters of lauric, palmitic or stearic acid (marked as E473) have been classified as Generally Recognized As Safe, GRAS [11] and can be used in cosmetic formulations because of their mild dermatological properties or in molecular cuisine to produce airs and foams [12]. Sugar fatty acid esters can be produced from renewable and low-cost raw materials (viz carbohydrates, fatty acids or oils) by using both chemical and enzymatic esterification. The chemical route, that is transesterification of the fatty acid methyl ester with the sugar, is the most widespread at an industrial level. However, not only this reaction is poorly selective, but it also involves the use of harmful solvent and production of organic and inorganic wastes [13] difficult to remove from the product. To overcome these limitations, the enzymatic route is an attractive and eco-friendly alternative that requires mild conditions of temperature and pressure. In addition, an important advantage is the higher regioselectivity which leads to a predominant reaction product [1,13]. Among the enzymes, lipases have been widely used to promote the ester bond formation because of their high substrate specificity. Lipases are efficient catalysts for the hydrolysis of esters; however, in the absence of water in neat organic solvents they can catalyse the reverse reaction. To enhance the stability in organic solvents, immobilized lipases have been employed. At the same time, immo-
384
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
bilization allows an easy separation and recover of the enzyme [14]. Several studies had been reported about the enzymatic transesterification of mono- and disaccharide (viz d-(+)-glucose, d-(−)-fructose, sucrose) and fatty acid esters [7,15–17]. In this paper we wish to report on the preparation of l-(+)arabinose-palmitic acid monoesters by using immobilized Candida antarctica lipase B (CALB, Novozyme 435) as a biocatalyst, following the previous research of some of us about the enzymatic synthesis of oligofructose fatty acids monoesters [1]. The effect of the reaction medium, temperature and molar ratio of reagents that strongly affect the activity and specificity of lipases has been investigated. The product was purified by flash chromatography and characterized by NMR, FTIR, ESI-MS analysis. The thermal properties were evaluated by DSC analysis. Among the sugars, we chose l-(+)-arabinose because it is a constitutive component of arabinoxylans and other naturally occurring non-starch polysaccharides, such as pectin and lignocellulose. These biopolymers are easily extracted with water from several biomass residues of the agri-food industry, particularly flaxseed meal, rice bran, wheat bran, barley, corn fiber, and display interesting physiological and functional properties [18]. In this way, we have a natural source for biobased arabinose and the derivatives that can be esterified to give amphiphilic molecules for food and non-food industrial applications. To the best of our knowledge, the direct esterification of l-(+)-arabinose and free palmitic acid has never been investigated up to now. Only a few studies on the synthesis of d-(−)-arabinose esters [19,20] and the direct esterification of other underivatized sugars and free fatty acids have been reported [14]. 2. Experimentals 2.1. Materials and methods Commercial immobilized lipase from Candida antartica lipase B (CALB, Novozyme 435) was a generous gift from Novozymes (Bagsvard, Denmark). Unless otherwise stated, all chemicals were purchased from Sigma-Aldrich and/or from VWR International and were used without further purification. 2.2. Synthesis of L-(+)-arabinose-palmitic acid esters l-(+)-arabinose and palmitic acid (different sugar/acid molar ratios between 1 and 3) were mixed in a 20 ml glass reaction tube by using a magnetic stirrer (450 rpm) changing the solvent (tert-butyl alcohol, TBU, and methyl tert-butyl ether, MTBE, alone or in binary mixture with dimethyl sulfoxide, DMSO, or 200 mM phosphate buffer pH 7,) and the temperature (40–60 ◦ C). After 30 min, molecular sieve (3 Å, 10% w/wpalmiticacid ) and CALB (8% w/wpalmiticacid ) were added to start the reaction and the mixture was stirred over night at the selected temperature. Control experiments were carried out without the enzyme. Then the mixture was cooled down to room temperature to stop the reaction and centrifuged. The solid was shown by IR to contain only the enzyme, molecular sieves and unreacted l-(+)-arabinose, therefore it was discarded. The solvent was removed under reduced pressure. The crude product was analysed by Thin Layer Chromatography (TLC), attenuated total reflection (ATR) Fourier transform infrared (FTIR) and GC-FID analysis. For the semi-preparative scale, in a 250 ml round bottom flask l-(+)-arabinose (1.0 g, 6.7 mmol) and palmitic acid (1.7 g, 6.7 mmol) were mixed at 60 ◦ C by using a magnetic stirrer (450 rpm) in TBU (35 ml) as the solvent, alone or in binary mixture with DMSO (7% and 10%). After 30 min, molecular sieves (3 Å, 10%) and CALB (140 mg) were added and the mixture was stirred o.n. at 60 ◦ C.
The reaction was controlled by using TLC. After cooling to room temperature, the mixture was centrifuged. The solid phase was discarded and the upper phase was dried under vacuum to give a white and odourless powder. The crude product was purified by flash chromatography and characterized by ATR-FTIR, ESI-MS, NMR and GC-FID analyses. 2.3. Thin layer chromatography Analytical Thin Layer Chromatography TLC was performed on Silica Gel 60 F254 precoated aluminum sheets (0.2 mm layer; Merck, Darmstadt, Germany). Components were separated by using CH2 Cl2 :MeOH (12:1) and detected by spraying the sheet with a p-anisaldehyde/H2 SO4 solution (ethanol:anisaldehyde:sulphuric acid:acetic acid, 90:5:5:1) followed by heating at 105 ◦ C for 5–10 min for the detection of the sugars. At the same time, another plate was developed with a ceric sulfate/ammonium molybdate or with a KMnO4 solution and heated at about 150 ◦ C for few seconds to detect also palmitic acid. The Rf of monoester was 0.2. 2.4. Purification methods Purification of products was accomplished by flash chromatography (silica gel 60, 40–63 mm, Merck). The crude product was eluted on a silica gel column with ethyl ether:n-hexane:formic acid (1:1:0.02), to remove unreacted palmitic acid and l-(+)-arabinose, followed by elution with ether:n-hexane:methanol (4.7:4.7:0.6). 2.5. Gas chromatography analysis The monoester content in the crude of reaction was evaluated according to the GC methods of Cramer et al. [21] with slight modification. The analysis of arabinose, palmitic acid and reaction mixtures was performed on an Agilent 6890 Gas Chro® matography system by using a Alltech Heliflex AT-5 capillary column (30 m × 0.32 mm ID × 0.25 m), with split injection and FID detection. Before injection, unsubstituted hydroxyl groups were silylated according to Degn et al. [22]. To 5–10 mg of crude reaction, 110 L of heptane:pyridine (2:1 v/v) containing octyl-ˇ-D-glucopyranoside (OGP, 5.9 mg/ml, 0.02 mol/ml) and heptadecane (4.66 mg/ml, 0.019 mol/ml) as internal standards was added together with 100 L BSTFA (1%, v/v TMCS). The mixture was stirred at 70 ◦ C for 30 min and, subsequently, 1 L of the resulting solution was injected with a 10 L glass syringe (Hamilton). The injection temperature was 220 ◦ C and the detector temperature was 250 ◦ C. A pressure of 21 psi, a gas flow of 1.7 mL/min (He) and a split flow of 106 mL/min were applied. The temperature profile of the analysis oven was: 1 min at 90 ◦ C, heating to 250 ◦ C at a rate of 30 ◦ C/min and kept constant for 4 min, finally heating to 310 ◦ C at a rate of 15 ◦ C/min and kept constant for 5 min. OGP was used as internal standard to evaluate the monoester concentration and compare the yields of the crude products. 2.6. Structural characterization of 5-O-palmitoyl-l-(+)-arabinosyl ester Mass Spectroscopy – Mass spectrum of monoester was recorded ® on a Q-TOF Micro-Waters Mass Spectrometer interfaced to an Electro Spray Ionization (ESI) by using a positive ion mode. The sample was injected to the mass spectrometer in a MeOH solution. NMR Spectroscopy – High-resolution 1 H and 13 C NMR spectra were acquired at 600.33 and 150.95 MHz, respectively, on a Bruker Advance 600 spectrometer (Bruker, Karlsruhe, Germany) interfaced with a workstation running a Windows operating system and equipped with a TOPSPIN software package. 20 mg of monoester was dissolved in 0.6 ml of DMSO-d6 and the spectra were recorded
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
at 27 ◦ C. Chemical shifts (ı) were given in parts per million (ppm) and referenced to the solvent signals [ıH 2.50 and ıC 39.50 ppm from Tetramethylsilane (TMS)]. 13 C NMR signal multiplicities were based on attached proton test (APT) spectra and assigned on the basis of 1 H-13 C correlation experiments (Heteronuclear Multiple Quantum Correlation spectroscopy, HMQC, and Heteronuclear Multiple Bond Correlation spectroscopy, HMBC). 1 H signals were assigned by using 1 H–1 H correlation experiments (Correlation SpectroscopY, COSY, and TOtal Correlation SpectroscopY, TOCSY). ATR-FTIR spectra – ATR-FTIR spectra were recorded on a Biorad FTS-40 spectrophotometer equipped with a Specac Golden Gate ATR platform with diamond crystal. DSC-analysis – Differential Scanning Calorimetry (DSC) measurements were carried out on 5 mg of sample, by using a DSC 3 Mettler Toledo Analytical Instrument, from 30 to 250 ◦ C at a heating rate of 5 K min−1 under a synthetic air atmosphere.
3. Results and discussion We have focused on the preparation of l-(+)-arabinose/palmitic acid monoesters by using immobilized Candida antarctica lipase B (Novozyme 435) as a biocatalyst. In the enzymatic esterification the substitution should involve the primary hydroxyl group of the sugar which is more accessible than the secondary groups to the active site of the enzyme (Scheme 1). Since the activity and specificity of lipase are strongly dependent on temperature, reaction medium and molar ratio of reagents, we investigate the optimal esterification conditions by using commercial immobilized lipase B from Candida antarctica in accordance with a previous work of some of us [1] about the preparation of oligofructose fatty acids monoesters. The temperature affects the solubility of the substrates and products, the activity of the enzyme and, therefore, the reaction rate and the position of the equilibrium [16]. Lipases are known for their thermal stability especially after immobilization. For example, Novozyme 435 can be used up to 80 ◦ C without notable loss of activity [23]. Although each enzyme has an optimum working temperature, the optimum reaction temperature may be different. The reaction medium must be able to dissolve enough amounts of both the substrates and, at the same time, it should not affect lipase activity and selectivity. As stated above, the esterification is usually promoted by using organic solvents. Among them tertbutanol (TBU) and methyl tert-butyl ether (MTBE) are widely employed. However, the main problem is the low solubility of sugars. To increase the sugar solubility, hydrophilic cosolvent (such as dimethylsulfoxide, dimethylformamide or pyridine) could be used. Nevertheless, these solvents have a strong inactivating effect on the enzyme activity because they interact and strip off water from the protein surface and the active site, imposing a higher structural rigidity [24]. The presence of low water amounts is indeed required to keep the enzyme active [14], although higher water levels would favour the hydrolysis rather than the esterification. To remove the water excess, which is generated during the process as a by-product, evacuation under vacuo, open test tubes, molecular sieves [16], reduce pressure [25] or azeotropic distillation [26] have been used. To optimize temperature and reaction medium, we performed first trial reactions in a 12-Reactor Carousel Reaction System by using equimolar amount of l-arabinose and palmitic acid in presence of CALB and molecular sieves. The reactions were carried out at two different temperatures, 40 and 60 ◦ C, in TBU and MTBE alone or in binary mixture with DMSO or phosphate buffer at pH 7 (that is close to the optimum pH of the enzyme). The conversion of larabinose was evaluated by TLC and FTIR analysis. The IR spectra of the crude products (Fig. 1) showed the typical vibration of the
385
ester group (1734 cm−1 ) in all the mixtures containing TBU alone or in mixture with DMSO (results not reported). On the contrary this vibration was not identified in the reactions performed in MTBE or in the presence of phosphate buffer. In particular, the highest intensity of both TLC spots and IR ester vibration was observed at 60 ◦ C by using TBU and 10% DMSO as cosolvent. The highest temperature, indeed, should increase solubility of both substrates together with lipase activity and reaction rate. At the same time, also DMSO should increase arabinose solubility. Sugar solubility is the main driving force of this reaction, but the amount of DMSO in the solvent mixture has to be carefully tuned as not only this solvent will inactivate the enzyme, but it has also been shown that lowering the percentage of DMSO will promote formation of the diesters with respect to monoesters [27,28]. We therefore carried out a second trial at 60 ◦ C to investigate the influence of the amount of DMSO in the solvent mixture and the influence of the amount of l-arabinose (Table 1). Since the esterification is an equilibrium reaction, an excess of one of the reagents should strongly influence sugar conversion [24]. The reactions were performed by using three sugar/acid molar ratios, 1:1, 2:1 and 3:1 and followed by TLC. The crude products were compared by FTIR (Fig. 2) and the yields were determined by GC-FID analysis. The maximum yield was obtained by using 10% DMSO in TBU (Table 1, entry 3). For higher percentage of DMSO, both conversion and selectivity to monoester collapsed thus leading to a much lower yield (Table 1, entry 4) due to inactivation of the catalyst. Alike, the catalyst reused showed a significant loss in activity (see SI). On the other hand the use of 7% DMSO in TBU as reaction medium and different amount of l-(+)-arabinose (Table 1, entries 5 and 6) to increase CALB activity and the sugar conversion by reducing the DMSO amount, showed that the yield can be increased by using a double amount of arabinose (Table 1, entry 2 vs 5). The progress of the reactions was evaluated by TLC and the dried products were analysed by FTIR and GC-FID. All the reactions were stopped at 24 h, since a light increase of monoester was observed at longer reaction time, associated with the enhancement of byproducts (evaluated by using the spot intensities of TLC). The peaks area were compared to a standard curve, that was obtained by using different concentrations of 5-O-Palmitoyl-l(+)-arabinose, previously purified by flash chromatography and subsequently silylated at the unsubstituted hydroxyl groups before injection. In particular, the elution time of monoester was 12.43 and 13.00 min (as a mixture of ␣ and  anomer), whereas the elution time of unreacted l-arabinose and palmitic acid were 4.86–4.98 min (␣ and  anomer) and 6.46 min, respectively. The chromatograms showed an increasing amount of by-products with the increase of the sugar excess. The formation of monoester is less dependent on the molar ratio of reagents. To isolate and characterize the products and understand whether the monoester substitution involves the primary hydroxyl group, we performed the reaction in a semi-preparative scale at the previously optimized conditions (1:1 molar ratio, 60 ◦ C, TBU/10%DMSO, Novozyme 435, molecular sieves). The crude was purified by flash chromatography, and 5-O-Palmitoyl-l-(+)arabinose was isolated as a white powder in 20% yield. In the ESI-MS spectrum (Fig. 3) the peak with the greatest intensity, the Base Peak, at m/z 411.57 corresponds to the molecular ion [M + Na]+ . Moreover, the peaks at m/z 443.63 and m/z 800.14 represent sodium adducts of the molecular ions species [M + MeOH + Na]+ and [2 M + Na]+ , respectively. This is consistent with the expected molecular mass for 5-O-Palmitoyll-(+)-arabinose, 388.54 g/mol. In the 1 H and 13 C NMR spectra two sets of proton and carbon signals derived from arabinofuranoside moiety were detected, suggesting that the product was a mixture of ␣- and -monoester. The
386
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
Scheme 1. Esterification reaction of palmitic acid with l(+)- arabinofuranose.
Fig. 1. Solvent effect in the esterification of palmitic acid with arabinose.
Table 1 Yield in arabinose ester under different conditions (palmitic acid 1.0 mmol, 256 mg; Candida antartica Lipase B (CALB) Novozyme 435, 8% w/wpalmiticacid , 20 mg; molecular sieves (3 Å, 10%), 60 ◦ C, 24 h). Entry 1 2 3 4 5 6
Molar ratio Ara/PA
l-(+)-arabinose (mg)
Solvent (l) TBU
DMSO
Yield (%)
Palmitic acid Conv %
Sel %
1:1 1:1 1:1 1:1 2:1 3:1
150 150 150 150 300 450
5000 4650 4500 4250 4650 “
– 350 500 750 350 “
6.13 7.83 20.72 5.50 9.32 7.82
7.70 13.91 21.69 10.31 12.26 18.07
79.61 56.29 95.53 53.35 76.02 43.28
spin system for each anomer was assigned with the aid of 1D and 2D NMR spectra and literature data [29]. As previously reported, although the chemical shifts of proton at H-1 position of ␣- and -arabinose rings are clearly separated, 1 H NMR spectrum is not suitable to assign the anomeric configuration as the peaks showed very close J1,2 coupling constants (Table 2 and Figs. S1–S7). Moreover, the other 1 H NMR signals are partially overlapped each other. On the other hand, 13 C NMR spectroscopy gives a clear distinction of the signals of the two anomers. Indeed, the 1,2-trans configuration of ␣-arabinose shifts the C-1 resonance signal to lower field (of about 7 ppm) from that of 1,2-cis configuration of -arabinose [29]. For these reasons, in the 13 C spectrum, signals at ␦ 102.53 and 96.55 were assigned at ␣-monoester and -monoester, respectively. The
others 13 C and 1 H chemical shifts were attributed on the basis of correlations in 2D NMR spectra, HSQC, COSY and TOCSY. The relative proportion of ␣- to - anomers was estimated of roughly 3 to 1 by integrating the intensity of the 2 typical proton signals at ␦ 5.01 (33%) and 4.92 (67%) in the 1 H NMR spectrum that correspond to - and ␣-arabinose rings, respectively. Compared to unsubstituted l-(+)-arabinose, the chemical shift of protons at 5 position of both anomers is shifted to lower fields because of the deshielding effect of carbonyl group. Similarly, all 1 H-13 C correlations in the 2D HSQC NMR spectrum are shifted to lower fields. These results confirm that the substitution occurred only at 5 position.
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
Fig. 2. Effect of TBU/DMSO ratio and alcohol/acid in the esterification of palmitic acid with arabinose.
Fig. 3. ESI-MS spectrum of 5-O-Palmitoyl-l-(+)-arabinose. The peaks at m/z 315.50 and 259.39 derived from the ionization source of the mass spectrometer.
387
388
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
Table 2 13 C and 1 H chemical shifts values of 5-O-Palmitoyl-l-(+)-arabinose.a C NMR (150.95 MHz) ␦13 C (ppm) ␣-anomer
-anomer
1 2 3 4 5a 5b C1 OH C2 OH C3 OH
102.51 79.96 77.62 83.03 64.65
96.53 77.23 75.93 79.58 66.34
Fatty acid moiety
␣-anomer
-anomer
C O C2 C3 C4 C5 and C13 C6 C7-12 C14 C15 C16
173.28 33.89 24.89 28.91 29.14 29.33 29.48, 29.45 31.74 22.53 14.36
173.31
13
Sugar moiety
a
H NMR (600.33 MHz,) ␦1 H (ppm) ␣ anomer 1
4.93 (d, 1H, J = 2.8 Hz) 3.93 (dd, 1H, J = 6.9, 2.3 Hz) 3.60 (dd, 1H, J = 6.9, 5.1 Hz) 3.72 (dd, 1H, J = 5.0, 2.8 Hz) 4.20 (dd, 1H, J = 11.1, 2.5 Hz) 3.98–3.94 (m, 1H) 6.26 (br.s, 1H) 5.34 (br.s, 1H) 3.43 (br.s, 1H)
 anomer 5.02 (d, 1H, J = 4.4 Hz) 3.70 (dd, 1H, J = 6.5, 4.4 Hz) 3.79 (t, 1H, J = 6.3 Hz) 3.66 (m, 1H, J = 7.6, 6.1, 3.8 Hz) 4.20– 4.17 (m, 1H) 3.97– 3.91 (m, 1H)
2.29 (t, 2H, J = 7.4 Hz) 1.51 (quint, 2H, J = 7.2 Hz) 1.30–1.20 (m, 24H)
0.85 (t, 3H, J = 7.0 Hz)
Abbreviations: s = singlet; d = doublet; t = triplet; quint = quintet; dd = doublet of doublets; m = multiplet; br.s = broad signal.
Fig. 4. Overlapped ATR-FTIR spectra of palmitic acid, l-(+)-arabinose and 5-O-Palmitoyl-l-(+)-arabinose.
The infrared spectra of palmitic acid, l-(+)-arabinose and 5-OPalmitoyl-l-(+)-arabinose are shown in Fig. 4. In particular, the vibration absorbance at 1736 cm−1 of the purified product suggests that the esterification has occurred. Moreover, the typical vibrations of both starting materials are evident in this spectrum: a broad OH stretching at 3400 cm−1 , characteristic of l-(+)-arabinose,
and intense peaks at 3000–2700 cm−1 , typical of the methyl and methylene groups of palmitic acid. Fig. 5 shows the DSC curve for 5-O-Palmitoyl-l-(+)-arabinose. After the glass transition of the amorphous component at about 55 ◦ C, the curve presents an endothermic peak at 79.6 ◦ C that should correspond to the melting of the crystalline fraction. The melting is followed by an endothermic oxidative decomposition that
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
389
Fig. 5. DSC curve of 5-O-Palmitoyl-l-(+)-arabinose.
is perhaps caused by the sugar moiety. Melting temperature of monoester is lower than that of pure l-arabinose (164–165 ◦ C, literature data) probably because of the presence of the lipophilic tail of palmitic acid, which has the melting point at 62.9 ◦ C. 4. Conclusions The direct esterification of l-(+)-arabinose and free palmitic acid has been carried out in the presence of Novozyme 435 by carefully tuning experimental conditions. The synthesized 5-O-Palmitoyl-l-(+)-arabinose monoester could be used as safe non-ionic surfactant in a wide range of industrial formulations. Moreover, the ability of Novozyme 435 to recognize l-(+)-arabinose may be useful for the preparation of food-grade arabinoxylan-palmitic acid conjugates that may find manifold food and non-food industrial application in place of the synthetic esters. Acknowledgements The authors thank COST Action TD1203, EUBIS for a short term mobility grant to V.M.P. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mcat.2017.02. 029. References [1] S.E.H.J. Van Kempen, C.G. Boeriu, H.A. Schols, P. De Waard, E. Van Der Linden, L.M.C. Sagis, Novel surface-active oligofructose fatty acid mono-esters by enzymatic esterification, Food Chem. 138 (2–3) (2013) 1884–1891.
[2] P. Floyd, P. Zarogiannis, K. Fox, Non-surfactant organic ingredients and zeolite-based detergents Final report prepared for the European Commission, Risk & Policy Analysts Limited (RPA), 2006. [3] P.B. Singh, H.S. Saini, Exploitation of agro-industrial wastes to produce low-cost microbial surfactants, in: S.K. Brar, G.S. Dhillon, C.R. Soccol (Eds.), Biotransformation of Waste Biomass into High Value Biochemicals, Springer, New York, New York, NY, 2014, pp. 445–471. [4] T. Vijai Kumar Reddy, G. Sandhya Rani, R.B.N. Prasad, B.L.A. Prabhavathi Devi, Green recyclable SO3 H-carbon catalyst for the selective synthesis of isomannide-based fatty acid monoesters as non-ionic bio-surfactants, RSC Adv. 5 (2015) 40997–41005. [5] T. Watanabe, S. Katayama, M. Matsubara, Y. Honda, M. Kuwahara, Antibacterial carbohydrate monoesters suppressing cell growth of Streptococcus mutans in the presence of sucrose, Curr. Microbiol. 41 (3) (2000) 210–213. [6] L. Xin, Antimicrobial structure-efficacy relationship of sugar fatty acid esters, J. Chem. Pharm. Res. 6 (5) (2014) 944–946. [7] S.W. Chang, J.F. Shaw, Biocatalysis for the production of carbohydrate esters, N. Biotechnol. 26 (3–4) (2009) 109–116. [8] M. Ferrer, J. Soliveri, F.J. Plou, N. López-Cortés, D. Reyes-Duarte, M. ˜ A. Ballesteros, Synthesis of sugar esters in Christensen, J.L. Copa-Patino, solvent mixtures by lipases from Thermomyceslanuginosus and Candida antarctica B, and their antimicrobial properties, Enzyme Microb. Technol. 36 (4) (2005) 391–398. [9] W.C. Griffin, Classification of surface-active agents by HLB, J. Soc. Cosmet. Chem. 1 (1949) 311–326. [10] N.R. Pedersen, R. Wimmer, J. Emmersen, P. Degn, L.H. Pedersen, Effect of fatty acid chain length on initial reaction rates and regioselectivity of lipase-catalysed esterification of disaccharides, Carbohydr. Res. 337 (13) (2002) 1179–1184. [11] GRAS Notice (GRN) No. 514. 2014. http://www.fda.gov/Food/ IngredientsPackagingLabeling/GRAS/NoticeInventory/default.htm. [12] http://www.molecularrecipes.com/hydrocolloid-guide/sucrose-ester-sucro2/. [13] I.S. Yoo, S.J. Park, H.H. Yoon, Enzymatic synthesis of sugar fatty acid esters, J. Ind. Eng. Chem. 13 (1) (2007) 1–6. [14] L.A.M. Van Den Broek, C.G. Boeriu, Enzymatic synthesis of oligo- and polysaccharide fatty acid esters, Carbohydr. Polym. 93 (1) (2013) 65–72. [15] J.F. Kennedy, H. Kumar, P.S. Panesar, S.S. Marwaha, R. Goyal, A. Parmar, S. Kaur, Enzyme-catalyzed regioselective synthesis of sugar esters and related compounds, J. Chem. Technol. Biotechnol. 81 (6) (2006) 866–876. [16] A.M. Gumel, M.S.M. Annuar, T. Heidelberg, Y. Chisti, Lipase mediated synthesis of sugar fatty acid esters, Process Biochem. 46 (11) (2011) 2079–2090.
390
V.M. Pappalardo et al. / Molecular Catalysis 433 (2017) 383–390
[17] T. Polat, R.J. Linhardt, Syntheses and applications of sucrose-based esters, J. Surfactants Deterg. 4 (4) (2001) 415–421. [18] M.S. Izydorczyk, Arabinoxylans, in: G.O. Phillips, P.A. Williams (Eds.), Handbook of Hydrocolloids, 2nd edition, Woodhead Publishing Cambridge, England, 2009, pp. 653–692. [19] A. Sharma, S. Chattopadhyay, Lipase catalysed acetylation of carbohydrates, Biotechnol. Lett. 15 (11) (1993) 1145–1146. [20] W. Tsuzuki, Y. Kitamura, T. Suzuki, S. Kobayashi, Synthesis of sugar fatty acid esters by modified lipase, Biotechnol. Bioeng. 64 (3) (1999) 267–271. [21] J.F. Cramer, M.S. Dueholm, S.B. Nielsen, D.S. Pedersen, R. Wimmer, L.H. Pedersen, Controlling the degree of esterification in lipase catalysed synthesis of xylitol fatty acid esters, Enzyme Microb. Technol. 41 (3) (2007) 346–352. [22] P. Degn, L.H. Pedersen, J. Duus, W. Zimmermann, Lipase-catalysed synthesis of glucose fatty acid esters in tert-butanol, Biotechnol. Lett. 21 (4) (1999) 275–280. [23] Á. Cruz-Izquierdo, L.A.M. van den Broek, J.L. Serra, M.J. Llama, C.G. Boeriu, Lipase-catalyzed synthesis of oligoesters of 2,5-furandicarboxylic acid with aliphatic diols, Pure Appl. Chem. 87 (1) (2015) 59–69. [24] R. Croitoru, F. Fit¸igˇau, L.A.M. Van Den Broek, A.E. Frissen, C.M. Davidescu, C.G. Boeriu, F. Peter, Biocatalytic acylation of sugar alcohols by 3-(4-hydroxyphenyl)propionic acid, Process Biochem. 47 (12) (2012) 1894–1902.
ˇ [25] S. Sabeder, M. Habulin, Zˇ . Knez, Lipase-catalyzed synthesis of fatty acid fructose esters, J. Food Eng. 77 (4) (2006) 880–886. [26] Y. Yan, U.T. Bornscheuer, L. Cao, R.D. Schmid, Lipase-catalyzed solid-phase synthesis of sugar fatty acid esters – removal of byproducts by azeotropic distillation, Enzyme Microb. Technol. 25 (8–9) (1999) 725–728. [27] R. ter Haar, H.A. Schols, L.A.M. van den Broek, D. Sa˘glam, A.E. Frissen, C.G. Boeriu, Harry Gruppen, Molecular sieves provoke multiple substitutions in the enzymatic synthesis of fructose oligosaccharide-lauryl esters, J. Mol. Catal. B: Enzym. 62 (2010) 183–189. [28] F.J. Plou, M.A. Cruces, M. Ferrer, G. Fuentes, E. Pastor, M. Bernabé, M. Christensen, F. Comelles, J.L. Parra, A. Ballesteros, Enzymatic acylation of diand tri-saccharides with fatty acids: choosing the appropriate enzyme, support and solvent, J. Biotechnol. 96 (2002) 55–66. [29] T. Usui, S. Tsushima, N. Yamaoka, K. Matsuda, K. Tuzimura, H. Sugiyama, S. Seto, K. Fujieda, G. Miyajima, Determination of anomeric configuration of furanoside derivatives by carbon-13 NMR, Agric. Biol. Chem. 38 (7) (1974) 1409–1410.