Contents lists available at ScienceDirect

Industrial Crops & Products

journal homepage: www.elsevier.com/locate/indcrop

Lipoxygenase-mediated peroxidation of model plant extractives

Ali H. Tayeba, Hasan Sadeghifara, Martin A. Hubbea, Orlando J. Rojasa,b,⁎

a Department of Forest Biomaterials, North Carolina State University, Raleigh, NC 27513, USAb Department of Bioproducts and Biosystems, School of Chemical Engineering, Aalto University, Espoo 00076, Finland

A R T I C L E I N F O

Keywords:HydroperoxidesLipoxygenaseUnsaturated fatty acidsLipid oxidationFiber processing

A B S T R A C T

Three unsaturated fatty acids, namely 9-cis,12-cis-linoleic acid, 1,2,3-tri-cis, cis-9,12-octadecadienoyl (glyceroltrilinolein) and 1,2,3-tri-cis-9-octadecenoyl (triolein) were selected as models of components of plant extractivesto monitor the hydroperoxygenation induced by soybean lipoxygenase (LOX), which was applied as an oxidativecatalyst at room temperature. The fatty acids were monitored in colloidal dispersions in relation to theirmolecular changes using 1H/13C nuclear magnetic resonance (NMR), Fourier transform infrared (FTIR) and UVspectroscopies. The detection of the hydroperoxy group was limited due to its unstable nature. However, thereduction of protons associated with the diene groups and the substitution of hydroperoxy groups at the allylicpositon in conjugated lipids were detected by the induced chemical shift of HOO-bearing 13C and 1H resonancesand the oxygen absorption owing to changes in the molecule. Moreover, compared to the two other substrates,no oxygen substitution was observed in triolein, in accordance with its lower level of saturation and the absenceof bis-allylic carbon. Our results are of relevance to plant fiber processing, since fatty acids are majorconstituents of hydrophobic deposits that cause a range of manufacturing challenges.

1. Introduction

Soybean lipoxygenase (LOX) is a type of enzyme that is foundwidely in plants, fungi, and animals (Siedow, 1991; Yamamoto, 1992).It is a non-heme iron-containing enzyme mostly responsible forcatalyzing the stereo-selective dioxidation of methyl-interrupted poly-unsaturated fatty acids and their esters, such as linoleic, linolenic andarachidonic acids, which contain a 1,4-cis, cis-pentadiene system, totheir corresponding hydroperoxy derivatives and giving cis-trans con-jugated hydroperoxide (Kuhn and Thiele, 1999). The roles of LOX inplants include responses to wounding and senescence (Kadamne et al.,2011; Porta and Rocha-Sosa, 2002; Siedow, 1991), and linoleic andarachidonic acid are common substrates for LOX in plant and animaltissue, respectively. It was reported that the enzyme is involved ininflammatory processes (Samuelsson et al., 1987), cell membranematuration (Schewe and Kuhn, 1991), atherogenesis, osteoporosis(Cathcart and Folcik, 2000; Cyrus et al., 1999; Klein et al., 2004) andhas important role in plant wilt resistance (Mhaske et al., 2013).

Lipid peroxidation catalyzed by LOX is a two-step process. In thefirst step, an oxidant Fe+3 from active lipoxygenase attacks unsaturatedfatty acids (containing carbon–carbon double bound(s)) to abstract oneallylic hydrogen from carbon, forming the carbon-centered lipid radical(L·). In the next step, the radicalized lipid reacts with the oxygenmolecule (oxygen insertion) leading to a lipid peroxy radical (LOO·)

that can abstract a hydrogen molecule from a neighboring lipidmolecule and form a new lipid radical and a hydroperoxide (LOOH)(Girotti, 1998; Kanner et al., 1987; Yin et al., 2011).

Many studies indicate the ability of LOX to accept simple chainlipids; however, there is some evidence showing that the enzyme iseven able to catalyze the oxygenation of more complex oils in livingsystem and to produce monohydroperoxy derivatives (Feussner et al.,1998). Besides, LOX has been used to oxygenate not only fatty acids,but also ester lipids such as phospholipids or even bio-membranes(Brash et al., 1987; Maccarrone et al., 1994). Plant oils such as soy beanoil are important sources of unsaturated fatty acids such as oleic andlinoleic acyl groups, which make them more prone to enzymaticoxidation (Chin et al., 1992).

Even though the oxidation of oils is usually an indicator of oil’s poorquality and generates unpleasant taste in food stuff, in other instances itcan be advantageous. For example, the reduction of fouling (tacky)compounds in process waters has been reported (Tayeb et al., 2017).Likewise, the modification of fatty acid fractions derived from woodextractives or other sources, through oxidation and radical formationfrom the unsaturated fatty acids, can be used for coupling ontocellulosic surfaces, and in turn, surface hydrophobization can beachieved (Cusola et al., 2014; Garcia-Ubasart et al., 2013).

Among the many analytical techniques, NMR spectroscopy is veryuseful to characterize plant oils and for functional group identification

http://dx.doi.org/10.1016/j.indcrop.2017.04.041Received 28 January 2017; Received in revised form 20 April 2017; Accepted 23 April 2017

⁎ Corresponding author at: Department of Bioproducts and Biosystems, School of Chemical Engineering, Aalto University, Espoo 00076, Finland.E-mail addresses: orlando.rojas@aalto.fi, ojrojas@ncsu.edu (O.J. Rojas).

Industrial Crops & Products 104 (2017) 253–262

Available online 06 May 20170926-6690/ © 2017 Elsevier B.V. All rights reserved.

MARK

(Belury, 2002; Cao et al., 2007; Evans et al., 2002; Gunstone, 1993;Hämäläinen et al., 2001, 2002; Jie and Mustafa, 1997; Knothe andKenar, 2004; Lie Ken Jie, 2001; Pajunen et al., 2008; Prema et al., 2013;Siciliano et al., 2013).

NMR was also used to determine the proportion of acyl group in thesesame oil throughout oxidation under high temperature, and to followthe formation of primary and secondary oxidation compounds in the oil(Guillén and Ruiz, 2004). Analyses of pure model lipids have adoptedNMR for understanding the chemical composition of lipids such assoybean, fish, and vegetable oils. More importantly, 1H NMR with itsshort testing time and high accuracy, has been useful to acquiredetailed chemical information of lipid mixtures. NMR has been usedto study low-density lipoprotein (LDL) peroxidation compounds as wellas thermal stressing of fish oils (Claxson et al., 1994; Haywood et al.,1995; Lodge et al., 1995; Medina et al., 1998; Silwood and Grootveld,1999).

However, despite the number of studies available on lipid oxidation,less information exists on the peroxidation of wood extractives cata-lyzed by soybean lipoxygenase. This might be due to the fact that theoxidation and peroxidation reaction products are usually mixtures thatare quite difficult to separate. Besides, most of the derived components,in particular lipid hydroperoxides, are very unstable, having a short lifetime, and they are usually studied as hydroxy derivatives, which areexpected to be more stable (Pajunen et al., 2008).

Pertinent studies highlighted the biological importance and thepossible application of LOX in modifying and degrading unsaturatedfatty acids in wood extractives (Gutiérrez et al., 2009; Marques et al.,2011; Zhang et al., 2007), Here, we further examine the effect of LOXon the molecular structure of three simple unsaturated oils present inwood extractives. This was accomplished in the presence of activeenzyme, and the separation of the modified lipid was achieved bysolvent phase separation. The 1H and 13C NMR of the conjugated dieneallylic hydroperoxides were assigned to determine the influence of theperoxy radical on the chemical structure of the oils.

In order to obtain direct evidence and compare the effect of LOX onlipid radicalization and oxidation, the three substrates considered inthis study were characterized by 13C/1H NMR before and after theenzymatic treatment upon dissolution in deuterated solvent. The studywas complemented by measurements via Fourier transform infrared(FTIR) and UV spectroscopy. The obtained molecular level analysis isessential to understanding the structural changes that lipids undergoafter treatment with the enzyme and to determining the effect ofoxidation on water solubility, hydrophobicity, molecular weight, etc.

2. Materials and methods

Soybean lipoxygenase was obtained from Sigma Aldrich as glycinetype I-B with 150,000 U/mg activity and 108 kDa molecular mass. Purelinoleic acid, glycerol trilinolein and triolein were also purchased fromSigma Aldrich and used as received. All the reagents and buffers wereprepared with Millipore Milli-Q water. Boric acid and potassiumhydroxide were obtained from Fisher Scientific for preparation of thebuffer solutions.

2.1. Lipid oxidation by lipoxygenase

The lipid substrates were subjected to oxidation with lipoxygenase.The enzyme/substrate ratio, temperature, and pH were kept constantthroughout the experiments. Experiments were carried out in anambient atmosphere of air with magnetic stirring in an open reactionvial (inner volume 20 mL). 400 mg of lipid was dispersed in 20 mL of0.2 M boric buffer (pH 9.0) under agitation at 250 rpm (magneticstirring). In order to fully disperse the substrate, 10 min sonication wasapplied. Separately, 20 mg (185 nM) of soybean lipoxygenase wasdissolved in 0.2 M boric buffer (pH 9.0) vial (20 mL) using a magneticstirrer and was then added to the vial containing the dispersed lipid and

stirred at 250 rpm. The reaction time for linoleic acid was 0.5, 2, 24 hand 1 week. However, 2 h reaction time was used for the other twolipids. After the completion of the reaction time, the dispersion wasacidified with 1.0 M HCl solution to pH 4 to stop the enzymaticreaction. The treated lipid was then extracted through phase separationusing three times chloroform extraction in a 100 mL separation funnel.Subsequently, the organic solvent was dried with NaSO4, evaporatedunder reduced pressure and the separation yield was determined. Forcomplete drying, the extracted lipid was left in the vacuum oven for24 h. Finally, the residue was dissolved in deuterated solvent forsubsequent NMR spectroscopy.

2.2. LOX reaction followed by UV spectrophotometry

LOX reaction with linoleic acid substrate was determined by UVspectroscopy using a Single-Beam UV/Visible Spectrophotometer(Holman et al., 1969). UV absorption at 234 nm was measured todetect enzyme activity. The production of hydroperoxide groups andconjugated diene systems in the solution induces UV absorbance at234 nm for the trans–trans isomers and 236 nm for the cis-trans isomers(Chan and Levett, 1977).

2.3. 1H and 13C analyses of the lipids

The lipids were analyzed using proton NMR; carbon NMR was alsoused for the linoleic acid. The measurements were acquired on a Bruker300 MHz spectrometer equipped with a quad probe dedicated to 31P,13C, 19F, and 1H acquisition at 28 °C. NMR spectra were recorded whenthe NMR sample was prepared and placed into a normal 5 mm NMRtube by dissolving 20 mg of extracted lipid in 1 mL deuterated solvent.The tube was sealed with a Teflon cap and secured with paraffin film.The total number of scans for all the experiments was 256 with anacquisition time of 1.6 s. Linoleic acid samples were analyzed using a500 MHz 13C NMR spectrometer operating with topspin 3.2 software.For each test 100 mg of pure oil was dissolved in 0.8 mL DMSO in a5 mm NMR tube. All the spectra were collected with a 1.6 s delay timeand 5000 scans and signals phase/baseline were corrected.

2.4. Fourier transform infrared spectroscopy (FTIR)

The infrared absorption spectra of the oxidized and unmodifiedsamples were obtained by using a Perkin Elmer Frontier FTIR spectro-meter at a wavelength resolution of 4 cm−1 and using 64 scans persample. Extracted lipid samples were dried overnight in a vacuum ovenat 40 °C before subsequent FTIR analysis, which used the same as thatNMR analysis.

3. Results and discussion

The experimental conditions were optimized in preliminary experi-ments. The main variables included the substrate type and the timeused in the enzymatic reaction. Soybean lipoxygenase in plants has asingle polypeptide chain with a molecular mass of 94–104 kDa, and itwas used in our work because, compared to other types of enzymes, it iseasier to purify and has been characterized already (Shibata et al.,1987). The goal of this study was to monitor the effect of LOX ondifferent unsaturated fatty acids through multiple techniques. Inaddition, we address the question whether the enzyme is able toproduce hydroperoxide groups in other substrates.

3.1. Characterization of pure lipids

After LOX treatment, modified lipids were successfully separated(yield 90%) using a nonpolar solvent (chloroform). Fig. 1 shows the 1HNMR spectra of the three model lipids before enzymatic oxidativetreatment. These experiments were performed to provide reference

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

254

spectra prior to modification through enzymatic reaction. Modifiedlipids should have similar spectra with the main difference being thepresence of new peaks associated with the protons added to the lipidchain. The region between 0.5 and 5.5 ppm contains all the typicalproton NMR features of oils (Guillén and Ruiz, 2004). Proton intensityin the 1H NMR spectra is directly related to protons concentration in thesample, and the correlation coefficient is the same for all protons;therefore, it was possible to determine the proportion of differentfunctional groups, such as OH and OOH. During the oxidation processcatalyzed by soybean lipoxygenase, different reactions take place thatcan lead to the formation of oxidative products. Therefore, in order toidentify related groups, first, all the protons associated with eachcarbon atom were identified in the unmodified oil, according to thereported values, and compared to data from the treated oils.

The oxidation reaction catalyzed by LOX leads to the formation ofCeOOH groups on the fatty acid as the primary product and thecongugated diene system (Fig. 2a). It should be expected that the 1HNMR spectrum of the substrate throughout the oxidation processprovides information about the changes in the concentration of protons,especially those that are associated with the double bonds. Further-more, it is also expected that any new CeOOH group can be assigned itsown unique proton signal. Therefore, such data can be used todetermine the reaction yield, since as the oxidation process advances,the intensity of the signal corresponding to the CeOOH group increasesand the intensity of the signal corresponding to the double bonddeclines. Since different substrates were used, different intensity andpeak position were expected. There are six double bonds in glycerol

trilinoleate and 3 double bonds in glycerol trioleate molecules, sotheoretically this means that the first lipid should be a better substratesince LOX radicalizes the lipid through the allylic carbon associatedwith the double bond and glycerol trilinoleate has lower degree ofsaturation.

The formation of the conjugated linoleic acid results from lipox-ygenase treatment, which produces, as a function of reaction time, anincreased UV absorbance at 234 nm (Fig. 2b). During the process ofoxidation catalyzed by the lipoxygenase, the enzyme reacts with thesubstrate by abstracting one proton from the bis-allylic methyleneposition (De Groot et al., 1975) and forms an enzyme-carbon radicalcomplex. The mechanism of formation of this complex is not exactlyknown, but the role of iron for the formation of such complex (LOX-radicalized lipid) has been confirmed. Moreover, the produced complexis extremely unstable and can easily react with the dioxygen to form anew proxy radical complex (LOX-LOO·). The group that is produced bythe reaction can be stabilized during the LOX catalytic cycle via anintra-complex electron transfer process, which reduces the radical to itsanion (LOO−) (Zoia et al., 2011). As shown in Fig. 2a, this is theintermediate product for the formation of the hydroperoxy group;however it is not clear what portion of the LOO− groups forms the HOOgroups and what portion remains unchanged.

3.2. Identification of hydroperoxy groups

The proton NMR approach for the determination of enzymatically-added HOO group and its derivatives in the oils was based on the

Fig. 1. 1H NMR spectra of pure a) linoleic acid, b) glycerol trilinoleate and c) glycerol trioleate dissolved in deuterated solvent (*: peaks are assigned to the residual solvent).

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

255

integration of the signal of the ending methyl protons (CH3) of the neatsubstrate. The methyl group in the region of 0.85-0.95 (Table 1)involves three hydrogens, and since pure oils were used, one canemploy the intensity of the peak to determine the number of protons forthe other carbons in the oil chain. For all the oils studied in this work,the peaks were normalized based on the integrated intensity of methylprotons (with an assumed value of 1) and all the other protons wereevaluated based on this assumption. For example, the number ofprotons associated with the olefinic group (CH]CH) in the linoleicacid was determined to be 1.19 × 3= 3.6, which compared to the 3protons in CH3, i.e., 1 × 3 = 3, is reasonable. There should be 3protons per methyl group and 4 protons per vinyl group (HeC]CeH)in each lipid molecule, which leads to a ratio of 3:4. We used the ratioof methyl protons to the rest of the functional groups protons in the oilmolecule to monitor any possible addition or reduction of hydrogensafter treatment of the oil with soybean lipoxygenase. As can be seen(Fig. 1), the most important functional groups in the linoleic acid arethose involving CeOeOH and C]C, which are key for understandingand monitoring the performance of LOX catalytic reaction to increasethe rate of peroxidation. The 1H NMR results are shown as two spectra(Figs. 3 and 4) for better observation of a region of particular interest(Fig. 4 is the enlargement of the region 3–7 ppm for the samespectrum).

Even though the hydroperoxy groups are unstable and they usuallycan react and form other secondary and tertiary components, Figs. 3and 5, show relatively well-defined spectra both in carbon and protonNMR. The primary oxidation products that resulted from the enzymetreatment at room temperature, indicated conjugated diene and HOOgroups (Chan and Levett, 1977; Claxson et al., 1994).

As indicated in Table 1, the reaction between enzyme and lipid

caused a chemical shift at 8.3 ppm, which is induced by the HOO groupon the oil chain. Fig. 4 displays the region between 3 and 7 ppm of the1H NMR spectra of untreated and treated linoleic acid. Other than thevinyl group peak, this region is free of signal for the unoxidizedsubstrate; however, after oil oxidation with LOX, proton associatedfor conjugated diene signal is observed. This indicates that even at anearly stage of the oxidation process, 1H NMR spectroscopy was able todetect the reaction product. This is in agreement with (Knothe andKenar, 2004; Martínez-Yusta et al., 2014) who reported the productionof HOO groups in plant oil under thermal stress. Based on theintegration technique, it was realized that the intensity of the HOOpeak was about 10% of that of CH3 (methyl ending group). Therefore,since each methyl group contains three protons, it can be concludedthat the intensity attributed to HOO (at 8.3 ppm) is equivalent to 30%of one proton. The intensity of the other protons after the enzymaticreaction is summarized in Table 1.

Protons corresponding to bis-allylic (]CHeCH2eCH]) were ob-served in the range of 2.7-2.9, the intensity of which declined with theadvance of the reaction. There was also a significant decline in thesignal of proton associated to olefinic (HeC]CeH) groups as theoxidation proceeded. The proton signal intensity for both groupscontinued to decrease with additional reaction time and the signalassociated with vinyl group almost disappeared after one week. Thisindicates that, although the majority of oxidation process was catalyzedwithin the first 2 h, the reduction of carbon double bond signalcontinued until complete degradation. It is expected that the intensityof HOO grows as the oxidation process advances; however, it actuallydecreased after 24 h under oxidation by LOX. The reason for thisreduction in HOO groups can be explained by their instability, sincethey can react with other components. This is in agreement with(Guillén and Ruiz, 2004), who identified the presence of HOO.However, as expected, the signals corresponding to carbon doublebonds at 5.3 ppm and the bis-allylic group at 2.75 ppm (-CH2-unitbetween two isolated double bonds) continued to decline, whichindicate the continuation of lipid breakdown. It is known that theformation of HOO is accompanied by the generation of conjugateddienes, which can be either trans-trans or cis-trans isomerization(Vasankari et al., 2001). Close to the olefinic carbon, at around5.3 ppm, small but distinguishable signals were observed after LOXtreatment. The intensity of these signals increased significantly after24 h (Fig. 4) which shows production of the conjugated diene.

3.3. Linoleic acid analysis by 13C NMR

Most of the 13C NMR signals were between 13.5 and 35 ppm, mainlyascribed to the, CH3, CH2 and allylic carbon atoms of the fatty acid. As

Fig. 2. Substrate oxidation mechanism through carbon-based radicalization (a). The UV spectrum of conjugated linoleic acid with a maximum at around 234 nm. Upon oxidation, theabsorption peak is shifted by 5–10 nm (b).

Table 1Assignment of signals of the 1H NMR spectra for the linoleic acid (integrated values wereused).

Chemical shift(ppm)

Functional groups Proton intensity for linoleic acid

LA LA-LOX(2 h)

LA-LOX(24 h)

LA-LOX(1 week)

0.85–0.95 eCH3 (ending methyl) 1 1 1 11.9 eCH2eCH]CH (allylic) 1.29 1.01 0.65 0.192.7–2.9 ]CHeCH2eCH= (bis

allylic)0.62 0.42 0.26 0.02

5.3–5.4 CH]CHe (olefinic) 1.19 0.90 0.63 0.258.3 OOH (hydroperoxy

group)– 0.10 0.03 –

6.0–6.5 H (Conjugated diene) – 0.10 0.53 0.35

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

256

shown in Fig. 5, all of the carbon signals and their positions wereidentified for linoleic acid (Gomez et al., 2011; Sacchi et al., 1998,1997). The signals at 128.0 and 132.0 ppm were assigned to the vinyliccarbon atoms, and the signal at 174 ppm was identified as carbonyl

(C]O) carbon atom. It was also confirmed that the first signal at14 ppm is associated with the terminal CH3 of the linoleic acid chain.Other signals in the region below 34 ppm were assigned to the CH2

carbon atoms.The effects on the chemical shifts of the neighboring carbon upon

the substitution of a hydroperoxy group in linoleic acid can be observedin Table 2. The signal at 71 ppm is attributed to carbon atomsassociated with these groups after the enzymatic oxidation of oil. Theshift probably depends on the position of HOO group and on thegeometry of the diene structure.

All the above carbon NMR signals were in agreement with thereported unmodified structure of linoleic acid (Davis et al., 1999;Gomez et al., 2011). However, the signal in the regions of 71 wasassigned to carbon atoms bonded to the oxygen of the hydroperoxide orhydroxide groups, which were not observed in the unmodified sample.This is in agreement with the observed peak for linoleic acid in theproton NMR region between 8.1-8.3 ppm. Besides, another signal wasalso seen in the 60 ppm area that can be attributed to the produced OHgroup on the lipid chain. Here it was again observed that the intensityof the vinyl group carbons decreased as the oxidation reactionadvanced, to the point that they completely disappeared after 1 week.Furthermore, the intensity of signals for carbons associated with C11

(bis allylic) and C8,14 (di-allyic) shows the same reduction (Fig. 3) as theoxidation proceeds, in agreement to the proton NMR results.

3.4. Glycerol oxidation with soybean lipoxygenase

3.4.1. Glycerol trilinoleateThe catalytic peroxygenation of two types of glycerol with different

degrees of saturation was also carried out by LOX in order to bettermonitor the effect of enzyme on more complex substrates as well as toestablish the effect of the number of C]C bonds over the peroxidationof the oil. There is limited NMR data on the LOX oxidation of glyceroltrilinoleate and glycerol trioleate. Fig. 6, shows a 1H NMR spectrum ofpure trilinoleate as well as that of the modified oil. No signal was

Fig. 3. Quantitative 1H NMR spectrum covering 0.0–9.0 ppm of a) pure unmodified linoleic dissolved in DMSO, b) extracted linoleic acid after being treated with LOX for 2 h, c) afterbeing treated for 24 h, and d) after 1 week.

Fig. 4. Enlargement of region between 3 and 7 ppm of the 1H NMR spectra of Fig. 3, a)pure unmodified linoleic acid, b) extracted lipid after being treated with LOX for 2 h, c)24 h and d) 1 week. The peak assignment for the conjugated diene was seen at6.0–6.5 ppm, the peak in 5.3 is representing the CH=CH which is weaker in the modifiedsample.

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

257

observed in the area above 6 ppm for this oil.Here the same experimental conditions (room temperature and pH

9.0) was applied for trilinoleate. Under oxidative condition, somefunctional groups evolved to produce new components, which can bereflected in the intensity of the glycerol proton signal in the region ofthe spectra in which these signals appear. Table 3, shows the spectraregion of the glycerol before and after LOX treatment.

As can be seen in Fig. 6, the intensity of the signals associated withmethyl and glycerol groups, corresponding to 0.85–0.95 and 4.1–4.3,remained unchanged. However, the intensities of signals for othergroups were changed. The bis-allylic (CHeCH2eCH ]) signal declinedfrom 0.45 to 0.22 ppm and that of the olefinic group (-CH]CHe) was

reduced from 1.16 to 0.58 ppm, all indicating that as the oxidationprocess advanced, the substrate became more saturated and, as a result,the peaks associated to these groups decreased. This is in agreementwith the occurrence of an oxidation process and its effect on the protonsthat belong to the carbon double bond. Besides, the degradation of theglycerol under the oxidation condition induced the formation of newcompounds, namely, primary oxidation compounds such as hydroper-oxy or hydroxyl groups. There is the possibility that HOO wassupported on saturated and unsaturated acyl chains, without theformation of a conjugated diene system. It was reported in the literaturethat 2 hydroxy groups can be supported in the same acyl chain (Coxonet al., 1984; Neff et al., 1988; Schneider et al., 2005, 2004; Sun andSalomon, 2004). However, the formation of each of the mentionedfunctional groups depends on the nature of the lipid and the oxidationcondition. It is reported that the degradation rate of these componentsis very fast, and it is difficult to detect all secondary oxidation productson the lipid chain (Guillén and Ruiz, 2004). Nevertheless, some of theprotons of these components might give signals in the proton NMRspectra that will not overlap with the acyl groups, and so theiridentification is possible.

Fig. 6, shows that the evolution of oxidation that is the origin of theproton signal at 3.8 ppm, attributed to the proton added on the lipidafter the oxidation. However the proton associated to the group thatwas added to the linoleic chain is seen in the region between 8.0-8.5. Itis not fully understood whether HOO would remain on the chain or thepeak in region 3.8 ppm belongs to the OH group that was formed as asecondary oxidation product. In either case, it can be concluded, fromthe reduction of protons in the bis-allylic, di-allylic or vinylic groups,that the degradation of the lipid was started by changes in the oil

Fig. 5. 13C NMR spectrum covering of a) pure unmodified linoleic dissolved in DMSO, b) extracted linoleic acid after treatment with LOX for 2 h and c) after treatment for 1 week (*:Residual solvent).

Table 2Chemical shift (δ) and assignment of the main resonances in the 13C NMR spectrum ofpure modified linoleic acid after 2 h enzymatic treatment. Chemical shifts identified fromliterature values (Sacchi et al., 1997).

Functional groups Chemical shift δ (ppm)

C18 (eCH3, methyl group) 13.5C17 (CH2eCH3) 22.45C3 25.5C11 (]CHeCH2eCH], bis allylic) 28.45C4-7 29C8-14 (e(CH2)n diallylic) 30.1C16 31C2 35CeOOH* 71C9,10,12,13 (HC]CHe olefinic) 129–131C1 (COOH, carbocylic group) 174

* Proposed, CeO shows a signal around same chemical shift.

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

258

structure. It was also seen some small peaks next to C]C as the mainsignal declines that can be due to the formation of a conjugated dienesystem as the oxidation proceeds. The quantification of these groupscan be carried out relative to the main functional group, e.g., endingmethyl group, which is present in the lipid.

3.4.2. Glycerol trioleateTrioleate has a very similar structure to trilinoleate, with the only

difference that it has one double bond in each chain. This oil was alsosubmitted to oxidation by LOX to evaluate the effect of the number ofdouble bonds on the oxidation yield. As shown in Fig. 7, all the protonsignals were identical as those observed for trilinoleate, except for the(HeC]CeH) group, which was weaker (0.72 ppm) compared to thecorresponding signal in trilinoleate (1.19 ppm) (Table 4).

As observed, upon oil treatment with lipoxygenase, no changesoccurred in the lipid structure. All the signals were similar, and there

was no indication of the presence of conjugated diene systems, whichexemplifies the resistance of this type of oil to the LOX-inducedoxidation.

3.5. FTIR analysis

FTIR was used to confirm the NMR analyses; it has been used widelyto evaluate plant and fruit oil fractions, such as linoleic acid (Kadamneet al., 2011; Kadamne and Proctor, 2009; Setyaningrum et al., 2013;Siddiqui and Ahmad, 2013). Fig. 8a–c, shows the FTIR spectra of thethree unsaturated oils studied herein. The analytical evaluation of theoil spectra, in term of functional groups corresponding for absorption ofcertain frequency/wavenumbers, is given in Table 5. The structuralcharacteristics of the oils were confirmed by FTIR spectroscopy in therange 600–4000 cm−1, as shown. The IR absorption spectra of each oilbefore and after treatment were combined to compare the effects ofoxidation on the oil structure. All the signals were identified andcompared to the literature (Gomez et al., 2011). The peak at about1710 cm −1 is due to the C]O stretching of carbonyl functionality ofthe carboxylic group, whereas the peak at 2920 and 2855 cm−1

corresponded to the CH2 and CH3 groups of the aliphatic chain. Noneof the intensities for these groups changed after the LOX treatment,which is in agreement with both carbon and proton NMR results.

The peak at 3010 cm−1 was identified as the C]CeH (stretchingCeH attached to the vinyl group), which almost disappeared in linoleicacid and trilinoleate but did not change in the case of the trioleate. Thisindicates partial removal of the ]CeH bond by addition of oxidationproducts on the oil chain. Two peaks in 945 and 970 were observed inlinoleic acid that can be attributed to the formation of the cis and trans

Fig. 6. 1H NMR spectrum covering 0.0–6.0 ppm of a) pure unmodified glycerol trilinoleate dissolved in CDCl3, b) treated with LOX for 2 h.

Table 3Assignment of signals of the 1H NMR spectra for the trilinoleate (data associated withliterature values (Vigli et al., 2003)).

Chemical shift(ppm)

Functional groups Proton intensity for trilinoleate

unmodified oxidized, (2 h)

0.85–0.95 eCH3 (ending methyl) 1 12.7–2.9 ]CHeCH2eCH= (bis

allylic)0.45 0.22

3.6–3.7 *OH/OOH(?) – 0.544.1–4.3 CH2COOR (glycerol group) 0.53 0.535.3–5.4 CH]CHe (olefinic) 1.16 0.58

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

259

conjugated linoleic acid respectively (Kadamne and Proctor, 2009;Poiana et al., 2015).

There was a strong band at about 1024 cm−1 on the modifiedtrilinoleate, indicating the presence of new groups on the lipidstructure. This peak, however, was not observed in the two other oils.As reported by (Guillén and Cabo, 1998), this can be associated withthe CeO bond, as was also observed in 1H NMR spectrum. There was a1H NMR signal at 3.6 ppm that arose in an area different than that ofthe region expected for the HOO group. The FTIR spectrum (Fig. 8b)supports the hypothesis that an unstable, short-lived HOO groupunderwent a secondary reaction, leading to formation of a CeOH groupon the lipid chain, rather than a HOO group. In the same figure, it isshown that the peak at 3009 cm−1 for trilinoleate, is associated with]CeH, disappeared after the oxidation process. Finally, Fig. 8c showsthe FTIR analysis of glycerol trioleate in which the two spectra (of theoriginal versus treated oil) was very similar. All the identified groups inthe spectra were the same for both the unmodified and modified oil.

This confirms that no significant structural changes were induced bysoy lipoxygenase on this glycerol under oxidation conditions andconfirms the NMR results that showed no changes in the oil protonsignal.

4. Conclusions

Comparison of the spectroscopic data before and after lipoxygenasetreatment indicated the oxidation of trilinoleate and linoleic acid. Nosignificant changes occurred in the case of trioleate. A reduction in thenumber of vinyl groups and the formation of the conjugated dienesystems throughout the oxidation process was confirmed for trilinoleateand linoleic acid. The fact that trilinoleate was more extensivelyaffected by the enzyme, compared to trioleate, is in agreement withproposed enzyme oxidation pathways (Ayala and Argauelles, 2014).Our results point toward the possibility of changing the chemicalactivity of lipids relevant to extractives from wood in order to reducethe formation of hydrophobic deposits that otherwise occur during fiberprocessing. Lipoxygenase enzymes degrade the oils, making themsuitable for controlling the lipophilic components in papermaking.

Acknowledgements

The authors acknowledge support from United Soybean Board(USB) under Project Number 1640-512-5276 and HYBER’s Academyof Finland’s Centers of Excellence Programme (2014–2019). We aregrateful to Dr. Keith D. Wing for fruitful discussions and advise.

Fig. 7. 1H NMR spectrum covering 0.0–6.0 ppm of a) pure unmodified glycerol trioleate dissolved in CDCl3, b) treated with LOX for 2 h.

Table 4Assignment of signals of the 1H NMR spectra for the trioleate (integrated values were usedand data associated with literature values (Vigli et al., 2003)).

Chemical shift (ppm) Functional groups Proton intensity

unmodified oxidized, 2 h

0.85–0.95 eCH3 (terminal methyl) 1 12.0–2.1 CH2eCH]CH 1.21 1.183.6–3.7 OOH – –4.1–4.3 CH2COOR (glycerol group) 0.53 0.535.3–5.4 CH]CHe(olefinic) 0.72 0.68

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

260

References

Ayala, A., Argauelles, S., 2014. Lipid peroxidation: production metabolism, and signalingmechanisms of malondialdehyde and 4-hydroxy-2-nonenal. Oxid. Med. Cell.Longevity 2014, 31.

Belury, A., 2002. Inhibition of carcinogenesis by conjugated linoleic acid: potential

mechanisms of action. J. Nutr. 132 (10), 2995–2998.Brash, R., Ingram, C., Harris, T., 1987. Analysis of a specific oxygenation reaction of

soybean lipoxygenase-1 with fatty acids esterified in phospholipids. Biochemistry 26(17), 5465–5471.

Cao, Y., Yang, L., Gao, H., Chen, J., Chen, Y., Ren, S., 2007. Re-characterization of threeconjugated linolenic acid isomers by GC–MS and NMR. Chem. Phys. Lipids 145 (2),128–133.

Cathcart, M., Folcik, A., 2000. Lipoxygenases and atherosclerosis: protection versuspathogenesis. Free Radic. Biol. Med. 28 (12), 1726–1734.

Chan, H., Levett, G., 1977. Autoxidation of methyl linoleate. Separation and analysis ofisomeric mixtures of methyl linoleate hydroperoxides and methyl hydroxylinoleates.Lipids 12 (1), 99–104.

Chin, F., Liu, W., Storkson, J., Ha, L., Pariza, W., 1992. Dietary sources of conjugateddienoic isomers of linoleic acid, a newly recognized class of anticarcinogens. J. FoodCompos. Anal. 5 (3), 185–197.

Claxson, A., Hawkes, E., Richardson, P., Naughton, P., Haywood, M., Chander, L.,Grootveld, C., 1994. Generation of lipid peroxidation products in culinary oils andfats during episodes of thermal stressing: a high field 1H NMR study. FEBS Lett. 355(1), 81–90.

Coxon, T., Peers, E., Rigby, M., 1984. Selective formation of dihydroperoxides in the[small alpha]-tocopherol inhibited autoxidation of methyl linolenate. J. Chem. Soc.Chem. Commun. 1, 67–68.

Cusola, O., Roncero, B., Vidal, T., Rojas, O.J., 2014. A facile and green method tohydrophobize films of cellulose nanofibrils and silica by laccase-mediated coupling ofnonpolar colloidal particles. ChemSusChem 7 (10), 2868–2878.

Cyrus, T., Witztum, L., Rader, J., Tangirala, R., Fazio, S., Linton, F., Funk, D., 1999.Disruption of the 12/15-lipoxygenase gene diminishes atherosclerosis in apo E-deficient mice. J. Clin. Invest. 103 (11), 1597–1604.

Davis, L., Mc Neill, P., Caswell, C., 1999. Analysis of conjugated linoleic acid isomers by13C NMR spectroscopy. Chem. Phys. Lipids 97 (2), 155–165.

Fig. 8. FTIR spectra (region between 600 and 4000 cm−1) of a) linoleic acid, b) Trilinoleate and c) Trioleate. The corresponding 2 h enzymatic treatment for each lipid is shown with redline. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 5Functional groups from FTIR spectra of evaluated oils. Data associated withliterature values (Esterbauer et al., 1991; Guillén et al., 1998; Gutiérrez et al.,1998; Lerma-García et al., 2010).

Frequency (cm−1) Functional groups

3460–80 eOH3005–10 ]CeH2922–26 eCeH(CH2)2855 eCeH(CH3)1736–42 eC]O1235–58 eCeO, CH2

1160 eCeO, CH2

1094 eCeO1024 eCeO970 Conjugated diene (cis)810 Conjugated diene (trans)723 e(CH2)e

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

261

De Groot, J., Veldink, A., Vliegenthart, G., Boldingh, J., Wever, R., van Gelder, F., 1975.Demonstration by EPR spectroscopy of the functional role of iron in soybeanlipoxygenase-1. Biochim. Biophys. Acta (BBA) − Enzymol. 377 (1), 71–79.

Esterbauer, H., Schaur, J., Zollner, H., 1991. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic. Biol. Med. 11(1), 81–128.

Evans, M., Brown, J., McIntosh, M., 2002. Isomer-specific effects of conjugated linoleicacid (CLA) on adiposity and lipid metabolism. J. Nutr. Biochem. 13 (9), 508.

Feussner, I., Bachmann, A., Hohne, M., Kindl, H., 1998. All three acyl moieties oftrilinolein are efficiently oxygenated by recombinant His-tagged lipid bodylipoxygenase in vitro. FEBS Lett. 431 (3), 433–436.

Garcia-Ubasart, J., Vidal, T., Torres, L., Rojas, O.J., 2013. Laccase-mediated coupling ofnonpolar chains for the hydrophobization of lignocellulose. Biomacromolecules 14(5), 1637–1644.

Girotti, W., 1998. Lipid hydroperoxide generation, turnover, and effector action inbiological systems. J. Lipid Res. 39 (8), 1529–1542.

Gomez, A., Abonia, R., Cadavid, H., Vargas, H., 2011. Chemical and spectroscopiccharacterization of a vegetable oil used as dielectric coolant in distributiontransformers. J. Braz. Chem. Soc. 22 (12), 2292–2303.

Guillén, D., Cabo, N., 1998. Relationships between the composition of edible oils and lardand the ratio of the absorbance of specific bands of their fourier transform infraredspectra. role of some bands of the fingerprint region. J. Agric. Food Chem. 46 (5),1788–1793.

Guillén, D., Ruiz, A., 2004. Formation of hydroperoxy- and hydroxyalkenals duringthermal oxidative degradation of sesame oil monitored by proton NMR. Eur. J. LipidSci. Technol. 106 (10), 680–687.

Gunstone, F., 1993. High resolution 13C NMR spectroscopy of lipids. Adv. Lipid Method.Two 4 (2), 1–68.

Gutiérrez, A., del Rıo, C., González-Vila, J., Martın, F., 1998. Analysis of lipophilicextractives from wood and pitch deposits by solid-phase extraction and gaschromatography. J. Chromatogr. A 823 (1), 449–455.

Gutiérrez, A., José, C., Martínez, T., 2009. Microbial and enzymatic control of pitch in thepulp and paper industry. Appl. Microbiol. Biotechnol. 82 (6), 1005–1018.

Hämäläinen, T., Sundberg, S., Mäkinen, M., Kaltia, S., Hase, T., Hopia, A., 2001.Hydroperoxide formation during autoxidation of conjugated linoleic acid methylester. Eur. J. Lipid Sci. Technol. 103 (9), 588–593.

Hämäläinen, T., Sundberg, S., Hase, T., Hopia, A., 2002. Stereochemistry of thehydroperoxides formed during autoxidation of CLA methyl ester in the presence of α-tocopherol. Lipids 37 (6), 533–540.

Haywood, M., Claxson, W., Hawkes, E., Richardson, P., Naughton, P., Coumbarides, G.,Grootveld, C., 1995. Detection of aldehydes and their conjugated hydroperoxydieneprecursors in thermally-stressed culinary oils and fats: investigations using highresolution proton NMR spectroscopy. Free Radic. Res. 22 (5), 441–482.

Holman, T., Egwim, O., Christie, W., 1969. Substrate specificity of soybean lipoxidase. J.Biol. Chem. 244 (5), 1149–1151.

Jie, S., Mustafa, J., 1997. High-resolution nuclear magnetic resonancespectroscopy–applications to fatty acids and triacylglycerols. Lipids 32 (10),1019–1034.

Kadamne, M., Proctor, A., 2009. Measurement of conjugated linoleic acid (CLA) in CLA-rich soy oil by attenuated total reflectance-Fourier transform infrared spectroscopy(ATR-FTIR). J. Agric. Food Chem. 57 (22), 10483–10488.

Kadamne, J., Castrodale, C., Proctor, A., 2011. Measurement of conjugated linoleic acid(CLA) in CLA-rich potato chips by ATR-FTIR spectroscopy. J. Agric. Food Chem. 59(6), 2190–2196.

Kanner, J., German, B., Kinsella, E., Hultin, O., 1987. Initiation of lipid peroxidation inbiological systems. CRC Crit. Rev. Food Sci. Nutr. 25 (4), 317–364.

Klein, F., Allard, J., Avnur, Z., Nikolcheva, T., Rotstein, D., Carlos, S., Orwoll, S., 2004.Regulation of bone mass in mice by the lipoxygenase gene Alox15. Science 303(5655), 229–232.

Knothe, G., Kenar, A., 2004. Determination of the fatty acid profile by 1H NMRspectroscopy. Eur. J. Lipid Sci. Technol. 106 (2), 88–96.

Kuhn, H., Thiele, J., 1999. The diversity of the lipoxygenase family: many sequence databut little information on biological significance1. FEBS Lett. 449 (1), 7–11.

Lerma-García, J., Ramis-Ramos, G., Herrero-Martínez, M., Simó-Alfonso, F., 2010.Authentication of extra virgin olive oils by Fourier-transform infrared spectroscopy.Food Chem. 118 (1), 78–83.

Lie Ken Jie, M., 2001. Analysis of conjugated linoleic acid esters by nuclear magneticresonance spectroscopy. Eur. J. Lipid Sci. Technol. 103 (9), 628–632.

Lodge, K., Sadler, J., Kus, L., Winyard, G., 1995. Copper-induced LDL peroxidationinvestigated by 1H NMR spectroscopy. Biochim. Biophys. Acta 1256 (2), 130–140.

Maccarrone, M., van Aarle, M., Veldink, A., Vliegenthart, G., 1994. In vitro oxygenationof soybean biomembranes by lipoxygenase-2. Biochim. Biophys. Acta (BBA) 1190 (1),164–169.

Marques, G., Molina, S., Babot, D., Lund, H., Río, C., Gutiérrez, A., 2011. Exploring thepotential of fungal manganese-containing lipoxygenase for pitch control and pulpdelignification. Bioresour. Technol. 102 (2), 1338–1343.

Martínez-Yusta, A., Goicoechea, E., Guillén, D., 2014. A review of thermo-oxidative

degradation of food lipids studied by 1H NMR spectroscopy: influence of degradativeconditions and food lipid nature. Compr. Rev. Food Sci. Food Saf. 13 (5), 838–859.

Medina, I., Sacchi, R., Giudicianni, I., Aubourg, S., 1998. Oxidation in fish lipids duringthermal stress as studied by 13C nuclear magnetic resonance spectroscopy. J. Am. OilChem. Soc. 75 (2), 147–154.

Mhaske, D., Mahatma, K., Jha, S., Singh, P., Mahatma, L., Parekh, B., Ahmad, T., 2013.Castor (Ricinus communis L.) Rc-LOX5 plays important role in wilt resistance. Ind.Crops Prod. 45, 20–24.

Neff, E., Frankel, N., Fujimoto, K., 1988. Autoxidative dimerization of methyl linolenateand its monohydroperoxides, hydroperoxy epidioxides and dihydroperoxides. J. Am.Oil Chem. Soc. 65 (4), 616–623.

Pajunen, I., Koskela, H., Hase, T., Hopia, A., 2008. NMR properties of conjugated linoleicacid (CLA) methyl ester hydroperoxides. Chem. Phys. Lipids 154 (2), 105–114.

Poiana, A., Alexa, E., Munteanu, M., Gligor, R., Moigradean, D., Mateescu, C., 2015. Useof ATR-FTIR spectroscopy to detect the changes in extra virgin olive oil byadulteration with soybean oil and high temperature heat treatment. Open Chem. 13,689–698.

Porta, H., Rocha-Sosa, M., 2002. Plant lipoxygenases. physiological and molecularfeatures. Plant Physiol. 130 (1), 15–21.

Prema, D., Pilfold, J.L., Krauchi, J., Church, J.S., Donkor, K.K., Cinel, B., 2013. Rapiddetermination of total conjugated linoleic acid content in select canadian cheeses by1H NMR spectroscopy. J. Agric. Food Chem. 61 (41), 9915–9921.

Sacchi, R., Addeo, F., Paolillo, L., 1997. 1H and 13C NMR of virgin olive oil. An overview.Magn. Reson. Chem. 35 (13), S133–S145.

Sacchi, R., Mannina, L., Fiordiponti, P., Barone, P., Paolillo, L., Patumi, M., Segre, A.,1998. Characterization of italian extra virgin olive oils using 1H-NMR spectroscopy.J. Agric. Food Chem. 46 (10), 3947–3951.

Samuelsson, B., Dahlen, S.E., Lindgren, J.A., Rouzer, C.A., Serhan, C.N., 1987.Leukotrienes and lipoxins: structures, biosynthesis, and biological effects. Science237 (4819), 1171–1176.

Schewe, T., Kuhn, H., 1991. Do 15-lipoxygenases have a common biological role? TrendsBiochem. Sci. 16, 369–373.

Schneider, Porter, A., Brash, A.R., 2004. Autoxidative transformation of chiral omega6hydroxy linoleic and arachidonic acids to chiral 4-hydroxy-2E-nonenal. Chem. Res.Toxicol. 17 (7), 937–941.

Schneider, B.W.E., Yin, H., Stec, D.F., Hachey, D.L., Porter, N.A., Brash, R., 2005.Synthesis of dihydroperoxides of linoleic and linolenic acids and studies on theirtransformation to 4-hydroperoxynonenal. Lipids 40 (11), 1155–1162.

Setyaningrum, L., Riyanto, S., Rohman, A., 2013. Analysis of corn and soybean oils in redfruit oil using FTIR spectroscopy in combination with partial least square. Int. FoodRes. J. 20 (4), 1977–1981.

Shibata, D., Steczko, J., Dixon, J.E., Hermodson, M., Yazdanparast, R., Axelrod, B., 1987.Primary structure of soybean lipoxygenase-1. J. Biol. Chem. 262 (21), 10080–10085.

Siciliano, C., Belsito, E., De Marco, R., Di Gioia, L., Leggio, A., Liguori, A., 2013.Quantitative determination of fatty acid chain composition in pork meat products byhigh resolution 1H NMR spectroscopy. Food Chem. 136 (2), 546–554.

Siddiqui, Ahmad, 2013. Infrared spectroscopic studies on edible and medicinal oils. Int. J.Sci. Technol. 2 (6), 1297–1306.

Siedow, J.N., 1991. Plant lipoxygenase: structure and function. Annu. Rev. Plant Physiol.Plant Mol. Biol. 42, 145–188.

Silwood, J., Grootveld, M., 1999. Application of high-resolution, two-dimensional 1H and13C nuclear magnetic resonance techniques to the characterization of lipid oxidationproducts in autoxidized linoleoyl/linolenoylglycerols. Lipids 34 (7), 741–756.

Sun, M., Salomon, G., 2004. Oxidative fragmentation of hydroxy octadecadienoatesgenerates biologically active γ-hydroxyalkenals. J. Am. Chem. Soc. 126 (18),5699–5708.

Tayeb, A.H., Hubbe, M.A., Zhang, Y., Rojas, O.J., 2017. Effect of lipoxygenase on surfacedeposition of unsaturated fatty acids. Langmuir. http://dx.doi.org/10.1021/acs.langmuir.7b00908.

Vasankari, T., Ahotupa, M., Toikka, J., Mikkola, J., Irjala, K., Pasanen, P., Viikari, J.,2001. Oxidized LDL and thickness of carotid intima-media are associated withcoronary atherosclerosis in middle-aged men: lower levels of oxidized LDL with statintherapy. Atherosclerosis 155 (2), 403–412.

Vigli, G., Philippidis, A., Spyros, A., Dais, P., 2003. Classification of edible oils byemploying 31P and 1H NMR spectroscopy in combination with multivariate statisticalanalysis. A proposal for the detection of seed oil adulteration in virgin olive oils. J.Agric. Food Chem. 51 (19), 5715–5722.

Yamamoto, S., 1992. Mammalian lipoxygenases: molecular structures and functions.Biochim. Biophys. Acta (BBA) 1128 (2), 117–131.

Yin, H., Xu, L., Porter, N.A., 2011. Free radical lipid peroxidation: mechanisms andanalysis. Chem. Rev. 111 (10), 5944–5972.

Zhang, X., Nguyen, D., Paice, M.G., Tsang, A., Renaud, S., 2007. Degradation of woodextractives in thermo-mechanical pulp by soybean lipoxygenase. Enzyme Microb.Technol. 40 (4), 866–873.

Zoia, L., Perazzini, R., Crestini, C., Argyropoulos, D.S., 2011. Understanding the radicalmechanism of lipoxygenases using 31P NMR spin trapping. Bioorg. Med. Chem. 19(9), 3022–3028.

A.H. Tayeb et al. Industrial Crops & Products 104 (2017) 253–262

262