Elevated Temperature Triggers the Antioxidant Action of Two Edible

Communication Communication AA Lesson Lesson Learnt Learnt from from Food Food Chemistry— Chemistry – Elevated ElevatedTemperature Temperature Triggers Triggers the Antioxidant the Antioxidant Action Actionof ofTwo Two Edible Edible Isothiocyanates: Isothiocyanates: ErucinErucin and and Sulforaphane Sulforaphane Jakub Cedrowski 1 , Kajetan D ˛abrowa 2 , Agnieszka Krogul-Sobczak 1 and 1 2 1 GrzegorzJakub Cedrowski Litwinienko , Kajetan1,* Dąbrowa , Agnieszka Krogul-Sobczak and Grzegorz Litwinienko 1,* 1 Faculty of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, Poland; 1 [email protected] Faculty of Chemistry, University (J.C.); of [email protected] Warsaw, Pasteura 1, 02-093 (A.K.-S.) Warsaw, Poland; 2 [email protected] of Organic Chemistry, (J.C.); Polish [email protected] Academy of Sciences, Kasprzaka (A.K.-S.) 44 /52, 01-224 Warsaw, Poland; 2 [email protected] Institute of Organic Chemistry, Polish Academy of Sciences, Kasprzaka 44/52, 01-224 Warsaw, Poland; * Correspondence:[email protected] [email protected]; Tel.: +48-22-552-6300 * Correspondence: [email protected]; Tel.: +48-22-5526300 Received:Received: 10 10 October October 2020; 2020; Accepted: Accepted: 3 November 3 November 2020; 2020; Published: Published: 6 November6 November 2020 2020

Abstract:Abstract:In thisIn communicationthis communication we demonstrate we demonstrate that two that natural two isothiocyanates, natural isothiocyanates, sulforaphane (SFN)sulforaphane and erucin (SFN) (ERN), and inhibit erucin autoxidation (ERN), inhibit of lipids autoxidation at 140 ◦C but of notlipids below at 140 100 °◦CC. but This not eﬀ ectbelow is due to thermal100 °C. decompositionThis effect is due of ERNto thermal and SFN decomposition to sulfenic acids of andERN methylsulﬁnyl and SFN to sulfenic radicals, acids species and able to trapmethylsulfinyl lipidperoxyl radicals, radicals. species Our observationsable to trap lipidperoxyl shed new light radicals. on thermal Our observations processing of shed vegetables new containinglight on thermal these two processing isothiocyanates. of vegetables containing these two isothiocyanates.

Keywords:Keywords:isothiocyanates; isothiocyanates; sulforaphane; sulforaphane; erucin; erucin; antioxidants; antioxidants; radicals radicals

Isothiocyanates (ITCs) can be found in vegetables in the form of glucosinolates Isothiocyanates (ITCs) can be found in vegetables in the form of glucosinolates (β-thioglucoside-N-hydroxysulfates), and during cutting, slicing, or grinding, ITCs are enzymatically (β-thioglucoside-N-hydroxysulfates), and during cutting, slicing, or grinding, ITCs are evolved to give a gentle, specific flavor and taste. Recently, great attention has been enzymatically evolved to give a gentle, specific flavor and taste. Recently, great attention has paid to naturally-occurring ITCs with methylsulfoxy- and methylthio-moieties (see Figure1), been paid to naturally-occurring ITCs with methylsulfoxy- and methylthio-moieties (see Figure namely sulforaphane (4-methylsulfinylbutyl isothiocyanate, SFN) and erucin (4-methylthiobutyl 1), namely sulforaphane (4-methylsulfinylbutyl isothiocyanate, SFN) and erucin isothiocyanate, ERN) [1–5]. (4-methylthiobutyl isothiocyanate, ERN) [1–5].

a) O S S N C S N C S sulforaphane (SFN) erucin (ERN) b) OH myrosinase O + H2O HO S R HO R N C S + HSO4 + glucose OH N O glucosinolate isothiocyanate (ITC) SO3 Figure 1. Structures of sulforaphane and erucin (a), and reaction scheme of formation of free Figure 1. Structures of sulforaphane and erucin (a), and reaction scheme of formation of free isothiocyanates (ITCs) during hydrolysis of glucosinolates in the presence of myrosinase (b). isothiocyanates (ITCs) during hydrolysis of glucosinolates in the presence of myrosinase (b). Both ITCs are the components of our diet when consuming common Cruciferous vegetables [1,6]: Both ITCs are the components of our diet when consuming common Cruciferous vegetables SFN is present in broccoli, some cultivars of cabbage, and Brussels sprouts while ERN predominantly [1,6]: SFN is present in broccoli, some cultivars of cabbage, and Brussels sprouts while ERN could be obtained from rocket [1,6–8]. Certain amounts of SFN along with ERN, could be provided from cabbage and red cabbage, rutabaga or turnip [6].

Antioxidants 2020, 9, x; doi: FOR PEER REVIEW www.mdpi.com/journal/antioxidants Antioxidants 2020, 9, 1090; doi:10.3390/antiox9111090 www.mdpi.com/journal/antioxidants Antioxidants 2020, 9, 1090 2 of 8

Talalay and co-workers [2,3,7] discovered and documented that among natural ITCs sulforaphane is the most potent inducer of carcinogen-detoxifying enzymes (e.g., quinone reductase, glutathione-S-transferase, heme oxygenase), and SFN is considered as one of the most potent chemopreventive phytochemical [2,7,9]. Such activity of SFN is due to induction of Phase 2 enzymes of the xenobiotic metabolism (this process involves Nrf2 signaling) and inhibition of Phase 1 enzymes. SFN and ERN are also recognized as inducers of some apoptotic pathways, anti-inflammatory effects, inhibition of angiogenesis, and their H2S-donating ability might be important in redox homeostasis agents [2,4,10–14]. ERN, a less studied analogue of SFN, also has a considerable impact on human health, being a potential chemopreventive and anticancer agent [3,8,12,14]. Impressive health benefits of ITCs have been connected with their ability to modulate cellular redox status due to induction of Phase 2 cytoprotective enzymes [15] and such an effect has been named “the indirect antioxidant action” because there is no direct scavenging of radicals by ITCs [16]. The attempts to demonstrate that ITCs are able to effectively react with oxygen-centered radicals like peroxyls were unsuccessful (although glucoerucin, the parent glucosinolate of ERN, efficiently decomposes hydrogen peroxide and tert-butylperoxide proving that glucoerucin is a preventive antioxidant [17]). There is no experimental evidence that ITCs can inhibit or retard the already ongoing peroxidation because their reaction with peroxyl radicals is too slow to compete with the chain-propagation, thus, neither ERN nor SFN are radical-trapping (chain-breaking) antioxidants [17,18]. According to our best knowledge, there is no information about the antioxidant activity of SFN and ERN at temperatures higher than 40 ◦C including thermal behaviour of this class of compounds at elevated temperature, i.e., under conditions similar to those in standard cooking in the presence of lipids as food constituents which are the most sensitive to oxidation. We selected linolenic acid (98%, all-cis-9,12,15-Octadecatrienoic acid, LNA) as an example of highly unsaturated lipid C18:3, and sunflower oil (SUN) as an example of edible oil frequently used in food preparation. The main fatty acids found in SUN were: palmitic acid (6.2% 0.5%), ± oleic acid (33.4 % 2.0%), stearic acid (4.8% 0.4%), linoleic acid (55.5% 3.4%), and linolenic acid ± ± ± (0.11% 0.04%), see details of GC analysis in Supplementary material (Figures S1 and S2 and Table S1). ± We used Differential Scanning Calorimetry in non-isothermal mode, with small 5-10 mg samples heated with heating rate 2 K/min for SUN and 2.5 K/min for LNA under oxygen flow 6 L/h (experimental details are included in the Supporting Material). Under such conditions the recorded DSC plots represent thermal effects of formation of hydroperoxides (primary products of oxidation) [19], see Figure2. The DSC curves recorded for oxidation of pure (neat) LNA and SUN indicate that those two lipids have different oxidative stability: oxidation of LNA starts at ca. 90 ◦C while oxidation of SUN starts at temperature 140 ◦C (see vertical dotted lines in both panels of Figure2). In the Supplementary Material the kinetic parameters calculated by iso-conversional method are presented in details: Ea for oxidation of pure LNA (79 5 kJ/mol) and overall rate constant, k, for its oxidation at 50 C (i.e., during the ± ◦ lag phase) is 7.3 10 3 min 1 and does not change for oxidation of LNA containing ERN or SFN. × − − For sunflower oil the kinetics are quite different: Ea for oxidation of pure SUN is 103 4 kJ/mol and ± increases to 119 8 kJ/mol in the presence of 10 mM ERN or 5 mM SFN. Similarly, k calculated at 100 C ± ◦ (lag phase) for oxidation of pure SUN is 5.8 10 3 min 1 but in the presence of 10 mM ERN is almost × − − two-fold smaller (2.9 10 3 min 1), see Supplementary Material (Tables S2–S9 and Figures S3–S9). × − − Complete data will be presented and discussed (together with kinetics of oxidation of other lipid matrices) in a full length article. Antioxidants 2020, 9, x FOR PEER REVIEW 3 of 8 Antioxidants 2020, 9, 1090 3 of 8

Figure 2. Thermal effects of non-isothermal oxidation (heating rate β = 2.5 K/min) of pure linolenic acid (LNA) and LNA containing 1, 5 and 10 mM sulforaphane (SFN)/erucin (ERN) (panel a). Thermal effects Figure 2. Thermal effects of non-isothermal oxidation (heating rate β = 2.5 K/min) of pure of non-isothermal oxidation (β = 2.0 K/min) of pure SUN and SUN containing 1, 5 and 10 mM SFN/ERN linolenic acid (LNA) and LNA containing 1, 5 and 10 mM sulforaphane (SFN) / erucin (ERN) (panel b). The DSC curves were vertically shifted for clarity. The number above each DSC curve (panel a). Thermal effects of non-isothermal oxidation (β = 2.0 K/min) of pure SUN and SUN denotes SFN/ERN concentration (in mM). Vertical dotted lines at 90 ◦C and 140 ◦C are introduced as a containing 1, 5 and 10 mM SFN / ERN (panel b). The DSC curves were vertically shifted for guide to the eye to observe a shift of temperatures of start of oxidation. clarity. The number above each DSC curve denotes SFN/ERN concentration (in mM). Vertical ° ° In thisdotted communication lines at 90 C and we limit140 C our are observations introduced as to a theguide slowest to theβ eye= 2 to and observe 2.5 K a/min, shift whenof the conditionstemperatures are relatively of start close of oxidation. to thermal equilibrium. Non-isothermal experiments allow for a direct comparison of thermal behaviour of two lipid matrices (LNA and SUN) regardless of their different In this communication we limit our observations to the slowest β = 2 and 2.5 K/min, when oxidative stability, parametrized as different temperatures of the start of oxidation. The addition of the conditions are relatively close to thermal equilibrium. Non-isothermal experiments allow antioxidant or retardant of oxidation should enhance the oxidative stability (increasing lag phase, for a direct comparison of thermal behaviour of two lipid matrices (LNA and SUN) regardless inhibition phase) and during the non-isothermal oxidation such lag phase should be manifested as of their different oxidative stability, parametrized as different temperatures of the start of the increasing temperature when the oxidation starts, as we observed many times in our previous oxidation. The addition of antioxidant or retardant of oxidation should enhance the oxidative works with lipids containing phenolic antioxidants like 2,6-di-tert-butyl-4-methylphenol (BHT) or stability (increasing lag phase, inhibition phase) and during the non-isothermal oxidation such tocopherol [20–23]. However, the thermal behaviour of LNA containing SFN or ERN (at concentrations lag phase should be manifested as the increasing temperature when the oxidation starts, as we 1–10 mM) presented in Figure2a indicates that these two isothiocyanates do not improve the oxidative observed many times in our previous works with lipids containing phenolic antioxidants like stability of polyunsaturated lipid because extrapolated start of oxidation is usually at ca. 90 ◦C 2,6-di-tert-butyl-4-methylphenol (BHT) or tocopherol [20–23]. However, the thermal behaviour (for 10 mM SFN the oxidation starts at even lower temperature). In contrast, sunflower oil that is of LNA containing SFN or ERN (at concentrations 1–10 mM) presented in Figure 2a indicates resistant towards oxidation up to temperature 140 ◦C is additionally stabilized when SFN or ERN are that these two isothiocyanates do not improve the oxidative stability of polyunsaturated lipid added, see Figure2b. For 10 mM concentration of ERN, the prolongation of oxidative stability of SUN because extrapolated start of oxidation is usually at ca. 90 °C (for 10 mM SFN the oxidation is ca. 5 ◦C and taking into account the non-isothermal character of the experiment, such difference is a starts at even lower temperature). In contrast, sunflower oil that is resistant towards oxidation clear demonstration of antioxidant contribution of SFN and ERN to the overall kinetics of thermal up to temperature 140 °C is additionally stabilized when SFN or ERN are added, see Figure 2b. oxidation of SUN. For 10 mM concentration of ERN, the prolongation of oxidative stability of SUN is ca. 5 °C and The first hypothesis, that ITCs could be activated by the products of SUN decomposition, taking into account the non-isothermal character of the experiment, such difference is a clear was excluded because both LNA and SUN produce very similar lipidperoxyl radicals as intermediates demonstration of antioxidant contribution of SFN and ERN to the overall kinetics of thermal and similar hydroperoxides as primary products of autoxidation. We also excluded the explanation that oxidation of SUN. the antioxidant effect is due to the decomposition of the isothiocyanate functional group, because alkyl The first hypothesis, that ITCs could be activated by the products of SUN decomposition, isothiocyanates are stable and -N=C=S is not eliminated during rather harsh conditions, as evidenced was excluded because both LNA and SUN produce very similar lipidperoxyl radicals as by GC analysis [24], isothiocyanates survive high temperature of GC injector (280 ◦C) and temperature intermediates and similar hydroperoxides as primary products of autoxidation. We also of GC column (increasing up to 220 ◦C). Thus, the functionality other than -N=C=S must be responsible excluded the explanation that the antioxidant effect is due to the decomposition of the for the observed “antioxidant activation” of SFN and ERN at higher temperatures, and we put forward isothiocyanate functional group, because alkyl isothiocyanates are stable and -N=C=S is not the hypothesis that CH (S=O)- and CH S- groups are responsible for the observed antioxidant action eliminated during rather3 harsh conditions,3 as evidenced by GC analysis [24], isothiocyanates of both ITCs at temperatures close to 140 ◦C. In our investigations we were inspired by the works on survive high temperature of GC injector (280 °C) and temperature of GC column (increasing up the antioxidant action of diallyl thiosulfinate (allicin—a component of garlic and onion) published to 220 °C). Thus, the functionality other than -N=C=S must be responsible for the observed recently by Ingold and Pratt [25,26]. Following the excellent review by Block [27], Ingold and Pratt “antioxidant activation” of SFN and ERN at higher temperatures, and we put forward the re-vitalized the idea based on older observation of the antioxidant action of sulfoxides in tetralin hypothesis that CH3(S=O)- and CH3S- groups are responsible for the observed antioxidant action of both ITCs at temperatures close to 140 °C. In our investigations we were inspired by Antioxidants 2020, 9, x FOR PEER REVIEW 4 of 8 Antioxidants 2020, 9, x FOR PEER REVIEW 4 of 8 the works on the antioxidant action of diallyl thiosulfinate (allicin — a component of garlic and onion)the published works on therecently antioxidant by Ingold action and of Pratt diallyl [25,26]. thiosulfinate Following (allicin the excellent — a component review byof garlicBlock and Antioxidants[27],onion) Ingold2020 published, and9, 1090 Pratt recentlyre-vitalized by Ingoldthe idea and based Pratt on [25,26]. older observation Following the of theexcellent antioxidant review action by Block4 of 8 of sulfoxides[27], Ingold in andtetralin Pratt made re-vitalized by Koelewijn the idea andbased Berger on older [28] observation who explained of the their antioxidant inhibitory action effectof duesulfoxides to formation in tetralin of transient made bysulfenic Koelewijn acids andduring Berger thermal [28] decompositionwho explained of their sulfoxides inhibitory madevia Cope-type byeffect Koelewijn due toelimination formation and Berger (Equation of [28 transient] who (1)). explained sulfenic theiracids inhibitory during thermal effect due decomposition to formation of of sulfoxides transient sulfenicvia acids Cope-type during elimination thermal decomposition (Equation (1)). of sulfoxides via Cope-type elimination (Equation (1)).

(1) (1) (1) The rate constants for unimolecular decomposition of di-tert-butylsulfoxide is 1.3 × 10−5 s−1 at 60 °C Theand rate for constantsless reactive for unimoldi-n-dodecylsulfoxideecular decomposition is 10− 5of s− 1di- attert 130-butylsulfoxide °C [28]. Koelewijn is 1.3 ×and 10− 5 s−1 The rate constants for unimolecular decomposition of di-tert-butylsulfoxide is 1.3 10 5 s 1 at Bergerat 60also °C measured and for theless rate reactive constants di-n -dodecylsulfoxidefor reaction of sulfenic is 10 acids−5 s−1 at with 130 alkylperoxyl °C [28]. Koelewijn× radicals− − and 60 C and for less reactive di-n-dodecylsulfoxide is 10 5 s 1 at 130 C[28]. Koelewijn and Berger also (Equation◦ Berger 2): also measured the rate constants for reacti− on− of sulfenic◦ acids with alkylperoxyl radicals measured the rate constants for reaction of sulfenic acids with alkylperoxyl radicals (Equation (2)): (Equation 2): R-SOH + R’OO• → non-radical products (2) R-SOHR-SOH+ R’OO + R’OO• •non-radical → non-radical products products (2) (2) to be higher than 107 M−1⋅s−1 at 60 °C. Sulfenic→ acids are transient species and if peroxyl radicals are tonot be higheraccessible, than7 101they7 M1 −1⋅undergos−1 at 60 °C.self-condens Sulfenic acidsation are totransient thiosulfinates species and(here, if peroxyl S-methyl radicals to be higher than 10 M− s− at 60 ◦C. Sulfenic acids are transient species and if peroxyl radicals are methanethiosulfinate, Equation· (3)): not accessible,are not theyaccessible, undergo self-condensationthey undergo self-condens to thiosulﬁnatesation (here,to thiosulfinates S-methyl methanethiosulﬁnate, (here, S-methyl methanethiosulfinate, Equation (3)): Equation (3)): 2CH3SOH → CH3(S=O)SCH3 + H2O (3)

2CH32CHSOH3SOHCH →3 (SCH=O)SCH3(S=O)SCH3 + H3 +2 OH2O (3)(3) S-methyl methanethiosulfinate easily decomposes→ via Cope-type elimination (the process is S-methylmuchS-methyl faster methanethiosulfinate formethanethiosulfinate thiosulfinates easily than easily decomposes for dialkylsulfoxides)decomposes via Cope-type via Cope-type into elimination another elimination sulfenic (the process (the acid process isand much is fasterthioformaldehyde formuch thiosulfinates faster for (Equation thanthiosulfinates for (4)) dialkylsulfoxides) [29–31]: than for dialkylsulfoxides) into another sulfenic into acidanother and thioformaldehydesulfenic acid and (Equationthioformaldehyde (4)) [29–31]: (Equation (4)) [29–31]: CH3(S=O)SCH3 → CH3SOH + H2C=S (4) CH3(S=O)SCH3 CH→3SOH + H2C=S (4) therefore, this is an additional CHchance3(S=O)SCH to →trap3 peroxyl CH3SOH radicals + H2C=S because sulfenic acid is (4) therefore, this is an additional chance to trap peroxyl radicals because sulfenic acid is “recovered” as a “recovered”therefore, as thisa second is an generation additional of chanceintermed toiates trap of thermalperoxyl decompositionradicals because of sulfoxides.sulfenic acid is second generation of intermediates of thermal decomposition of sulfoxides. “recovered”Another possible as a second explanation generation of antioxidant of intermed behaviouriates of thermal of SFN decomposition at elevated temperature of sulfoxides. is Another possible explanation of antioxidant behaviour of SFN at elevated temperature is direct direct thermolysis,Another possible with formation explanation of methylsulfinyl of antioxidant radicals behaviour CH of3SO SFN• that at areelevated able to temperature scavenge is thermolysis, with formation of methylsulfinyl radicals CH3SO• that are able to scavenge peroxyl peroxyldirect radicals. thermolysis, Methylsulfinyl with formation radicals of can methylsulfinyl be also produced radicals from CH sulfenic3SO• that acid are reacting able to scavengewith radicals. Methylsulfinyl radicals can be also produced from sulfenic acid• → reacting• with peroxyl peroxylperoxyl radicals radicals. (Equation Methylsulfinyl (2) modified radicals to the can form: be also CH produced3SOH + R’OO from sulfenic CH3SO acid + reactingR’OOH). with radicals (Equation (2) modified to the form: CH3SOH + R’OO• CH3SO• + R’OOH). The analogous The peroxylanalogous radicals process (Equation was described (2) modified by Prattto the et form: al. [26], CH→ 3SOHwho +studied R’OO• the→ CHmechanism3SO• + R’OOH). of process was described by Pratt et al. [26], who studied the mechanism of antioxidant action of antioxidantThe analogous action of process S-prop ylwas propanethiosulfinate, described by Pratt C et3H 7al.(S=O)SC [26], 3whoH7, as studied saturated the analogue mechanism of of S-propyl propanethiosulfinate, C3H7(S=O)SC3H7, as saturated analogue of allicin and other secondary allicinantioxidant and other action secondary of S-prop metabolitesyl propanethiosulfinate, of plants belonging C3H7 (S=O)SCto Allium3H genus7, as saturated. They computed analogue of metabolites of plants belonging to Allium genus. They computed thermodynamic parameters for thermodynamicallicin and other parameters secondary for model metabolites methanesul of plantsfenic belonging acids undergoing to Allium subsequent genus. They reactions computed model methanesulfenic acids undergoing subsequent reactions with methylsulfinyl radicals as reactive withthermodynamic methylsulfinyl parametersradicals as for reactive model methanesulintermediatesfenic able acids to undergoing trap alkylperoxyl subsequent radicals reactions intermediates able to trap alkylperoxyl radicals (Equation (5)). (Equationwith methylsulfinyl(5)). radicals as reactive intermediates able to trap alkylperoxyl radicals (Equation (5)).

(5)(5) (5) The above mechanism explains the antioxidant action of allicin analogues at low The above mechanism explains the antioxidant action of allicin analogues at low temperatures, temperatures, but the same chemistry should be also valid for the antioxidant action of but the sameThe chemistry above shouldmechanism be also explains valid for the the antioxidantantioxidant action action of sulforaphaneof allicin analogues at temperatures at low sulforaphane at temperatures above 140 °C. abovetemperatures, 140 C. but the same chemistry should be also valid for the antioxidant action of Similar◦ antioxidant behaviour of both ITCs,° regardless of different oxidation state of sulfur Similarsulforaphane antioxidant at temperatures behaviour of above both ITCs,140 C. regardless of different oxidation state of sulfur atom atom in ERN and SFN, is not surprising because at high temperature and in the presence of O2, in ERN andSimilar SFN, is antioxidant not surprising behaviour because of at both high ITCs, temperature regardless and of in different the presence oxidation of O ,state sulfides of sulfur can sulfides can be easily oxidised to sulfoxides. Moreover, formation of unstable sulfenic2 acid was be easilyatom oxidised in ERN to and sulfoxides. SFN, is not Moreover, surprising formation because of at unstable high temperature sulfenic acid and was in the also presence observed of for O2, also observed for cysteine in the systems with induced stress (elevated temperature is not cysteinesulfides in the can systems be easily with oxidised induced to stress sulfoxides. (elevated Mo temperaturereover, formation is not necessary)of unstable [32 sulfenic]. The electronacid was necessary) [32]. The electron withdrawing group at β position (β-EWG) greatly facilitates the withdrawingalso observed group atforβ positioncysteine (inβ-EWG) the systems greatly wi facilitatesth induced the process,stress (elevated but Cope-type temperature elimination is not process, but Cope-type elimination not assisted by β-EWGβ was alsoβ observed by Kubec et al. not assistednecessary) by β-EWG [32]. The was electron also observed withdrawing by Kubec group et al. [at33 ] forposition S-methylcysteine ( -EWG) greatly sulfoxide facilitates heated the at process, but Cope-type elimination not assisted by β-EWG was also observed by Kubec et al. 120 ◦C for 1 h (complete conversion in the presence of 10% of water, for the dry compound the process was “somewhat slower”). On the other side, generation of methylsulfinyl radical was proposed by Jin et al. [24] on the basis of careful analysis of the products of thermal decomposition of sulforaphane Antioxidants 2020, 9, x FOR PEER REVIEW 5 of 8

[33] for S-methylcysteine sulfoxide heated at 120 °C for 1 h (complete conversion in the presence of 10% of water, for the dry compound the process was “somewhat slower”). On the other side, generation of methylsulfinyl radical was proposed by Jin et al. [24] on the basis of careful analysis of the products of thermal decomposition of sulforaphane in water at 50–100 °C. Both mentioned works considered decomposition in a polar environment, thus, we decided to check the products of thermal decomposition of neat ERN and SFN at temperatures of 100 ° andAntioxidants 160 C.2020 , 9, 1090 5 of 8 The results of the GC/MS analysis are presented in Table 1 and Figure 3. Indeed, for both, ITCs we found three compounds that are the same as described by Jin et al. [24] for thermal in water at 50–100 C. Both mentioned works considered decomposition in a polar environment, thus, decomposition of SFN,◦ that is a proof that ERN readily undergoes oxidation to SFN. we decided to check the products of thermal decomposition of neat ERN and SFN at temperatures of 100 and 160 ◦C. TableThe results1. Names, of the GCretention/MS analysis times areand presented mass spectral in Table 1data and for Figure three3. Indeed, products for both,of thermal ITCs we ° ° founddecomposition three compounds of neat that SFN are / ERN the same at temperatures as described 100 by JinC and et al. 160 [24 C.] for thermal decomposition of SFN, thatCompound, is a proof Symbol, that ERN readily undergoes oxidation to SFN.MS Spectral Data m/z (Relative Experiment 5 RT 1 Intensity) Table 1. Names, retention times and mass spectral data for three products of thermal decomposition of 4-methylthio-3-butenyl neat SFN/ERN at temperatures 100 C and 160 C. 161 [(M+2)+, 9], 87 (22), 85 (24), isothiocyanate 2 3A, SFN(100◦ °C, 160 °C),◦ ERN(160 °C) 72 (42), 61 (100), 45 (34) RT = 34.2 min MS Spectral Data m/z Compound, Symbol, R 1 Experiment 5 T (Relative Intensity) but-3-enyl 113 (M+, 50), 85 (10), 72 (100), 4-methylthio-3-butenylisothiocyanate 3 3B, SFN(100 °C, 160 °C), ERN(160 °C) + 2 161 [(M+2) , 9], 87 (22), 85 (24), isothiocyanate 3A, SFN(100 ◦C, 160 ◦C), ERN(160 ◦C) 55 (30), 39 (71) RT = 11.6 min 72 (42), 61 (100), 45 (34) RT = 34.2 min but-3-enylS-methyl 128 [(M+2)+, 8], 126 (M+, 73), 113 (M+, 50), 85 (10), 72 (100), methylthiosulfonateisothiocyanate 3 3B ,4 SFN(160SFN(100 °C),C, ERN(160 160 C), ERN(160°C) C) 111 (11), 81 (5), 79 (50), 64 (40), ◦ ◦ ◦ 55 (30), 39 (71) 3CR,T R=T =11.6 10.5 min min 47 (10), 45 (20) 128 [(M+2)+, 8], 126 (M+, 73), 111 S-methyl1 RT = retention methylthiosulfonate time. 2 Identified4 basing on NIST Mass Spectrometry Data Center, SRD1A, NIST SFN(160 ◦C), ERN(160 ◦C) (11), 81 (5), 79 (50), 64 (40), 47 (10), 3C,R = 10.5 min 3 MS numbersT 5649 and 414629 (accessed on 14.08.2020). Identified basing45 on (20) NIST Mass 1 2 4 SpectrometryRT = retention Data time. IdentifiedCenter, basingSRD on69, NIST NIST Mass MS Spectrometry number Data 414530 Center, (accessed SRD1A, NIST on MS 14.08.2020). numbers 5649 Identifiedand 414629 basing (accessed on on NIST 14.08.2020). Mass3 Identified Spectrometry basing on Data NIST Center, Mass Spectrometry SRD 69, DataNIST Center, MS SRDnumber 69, NIST 236079 MS number 414530 (accessed on 14.08.2020). 4 Identified basing on NIST Mass Spectrometry Data Center, SRD 69, 5 (accessedNIST MS numberon 14.08.2020). 236079 (accessed This on column 14.08.2020). indicates5 This column in which indicates experiment in which experiment the compounds the compounds were detectedwere detected for thermal for thermal decomposition decomposition of of SFN SFN/ERN. /ERN.

Figure 3. Proposed paths of thermal oxidative conversion of erucin (ERN) to sulforaphane (SFN) and Figure 3. Proposed paths of thermal oxidative conversion of erucin (ERN) to sulforaphane subsequent decomposition to the products detected by GC/MS and showed in Table1. (SFN) and subsequent decomposition to the products detected by GC/MS and showed in Table 1.All three compounds, 3A, 3B and 3C, were found in the samples containing SFN and ERN heated at 160 ◦C, but S-methyl methylthiosulfonate (3C) was not detected at lower temperatures, perhapsAll three because compounds, of a slower 3A reaction, 3B and at 100 3C◦,C, were where found the amount in the samples of 3C was containing below the detection SFN and limit. ERN heatedMoreover, at SFN160 undergoes°C, but furtherS-methyl processes: methylthiosulfonate dehydratation and (3C formation) was not of 3A detected(we detected at bothlower temperatures,E/Z isomers), perhaps product ofbecauseβ-elimination of a slower3B (second reaction product at 100 of this°C, reaction,where the sulfenic amount acid, of was 3C not was belowdetected the because detection of its limit. instability) Moreover, and cleavage SFN ofundergoes S-C bond withfurther generation processes: of methylsulfinyl dehydratation radical and formationwhich, after of recombination3A (we detected with both methylsulfanyl E/Z isomers), radical product (CH3S• ),of gives β-elimination S-methyl methanethiosulfinate, 3B (second product further oxidized to 3C. Compound 3A was reported by Papi et al. [18] as exhibiting selective cytotoxic/apoptotic activity, however, the same authors noticed that regardless of the reaction with Antioxidants 2020, 9, 1090 6 of 8 diphenylpicrylhydrazyl radicals, 3A was not a good chain-breaking inhibitor of low temperature autoxidation of styrene and methyl linoleate. In conclusion, our experiments with thermal oxidation of two model lipids indicate that erucin and sulforaphane are kinetically neutral below 100 ◦C but at elevated temperatures these isotiocyanates decompose (with kinetically significant rates) to sulfenic acids and/or methylsulfinyl radicals that are good radical trapping agents, resulting in better oxidative stability of the lipid system. Thus, ITCs are examples of biocompounds which (in contrast to phenolic antioxidants) become chain-breaking antioxidants above 120 ◦C, that is, at temperatures corresponding to the frying process. A similar effect of inhibition of autoxidation at temperatures above 140 ◦C by SFN and ERN was observed by us during thermal oxidation of soy lecithin. Detailed description with the kinetic parameters for spontaneous and inhibited oxidation and comparison for LNA, SUN and soy lecithin will be described in a full length article.

Supplementary Materials: The following are available online at http://www.mdpi.com/2076-3921/9/11/1090/s1, Scheme S1: pathway of the synthesis of ERN and SFN; Figure S1: GC chromatogram of FAME Mix RM-3 reference mixture; Figure S2: GC chromatogram of FAME obtained from the sunflower oil; Table S1: Relative FAME composition obtained from the sunflower oil; Tables S2–S9 and Figures S3–S9: Experimental procedures (conversion of SUN into fatty acid methyl esters, analysis of fatty acids profile, synthesis of sulforaphane and erucin, NMR and MS analysis of ERN and SFN, experimental procedure of thermal degradation of ERN/SFN and GC-MS analysis of the products, description of DSC measurements, methodology of calculation of kinetic parameters of oxidation) the values of activation energy (Ea), pre-exponential factor Z, and overall rate constant (k) calculated for oxidations of pure LNA/SUN and for oxidations of LNA/SUN containing ERN/SFN; references. Author Contributions: Individual contribution of the authors: Conceptualization, J.C. and G.L.; synthesis of ERN and SFN, K.D.; NMR and MS analysis of ERN and SFN, K.D., J.C.; DSC measurements, J.C., G.L. GC-MS measurements, A.K.-S., J.C.; writing—original draft preparation, G.L., J.C.; writing—review and editing, all authors; supervision, G.L. All authors have read and agreed to the published version of the manuscript. Funding: J.C. gratefully acknowledges financial support from the National Science Center of Poland, NCN grant PRELUDIUM 8 No. 2014/15/N/ST5/02939. The analysis of the products were financed within the NCN grant OPUS 18 No. 2018/31/B/ST4/02354 received by G.L. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

References

1. Fahey, J.W.; Zalcmann, A.T.; Talalay, P. The chemical diversity and distribution of glucosinolates and isothiocyanates among plants. Phytochemistry 2001, 56, 5–51. [CrossRef] 2. Zhang, Y.; Talalay, P.; Cho, C.G.; Posner, G.H. A major inducer of anticarcinogenic protective enzymes from broccoli: Isolation and elucidation of structure. Proc. Natl. Acad. Sci. USA 1992, 89, 2399–2403. [CrossRef] 3. Posner, G.H.; Cho, C.G.; Green, J.V.; Zhang, Y.; Talalay, P. Design and synthesis of bifunctional isothiocyanate analogs of sulforaphane: Correlation between structure and potency as inducers of anticarcinogenic detoxication enzymes. J. Med. Chem. 1994, 37, 170–176. [CrossRef] 4. Dinkova-Kostova, A.T.; Kostov, R.V. Glucosinolates and isothiocyanates in health and disease. Trends Mol. Med. 2012, 18, 337–347. [CrossRef] 5. Palliyaguru, D.L.; Yuan, J.M.; Kensler, T.W.; Fahey, J.W. Isothiocyanates: Translating the Power of Plants to People. Mol. Nutr. Food Res. 2018, 62, e1700965. [CrossRef] 6. Fenwick, G.R.; Heaney, R.K.; Mullin, W.J.; VanEtten, C.H. Glucosinolates and their breakdown products in food and food plants. CRC Crit. Rev. Food Sci. Nutr. 1983, 18, 123–201. [CrossRef] 7. Fahey, J.W.; Zhang, Y.; Talalay, P. Broccoli sprouts: An exceptionally rich source of inducers of enzymes that protect against chemical carcinogens. Proc. Natl. Acad. Sci. USA 1997, 94, 10367–10372. [CrossRef] 8. Azarenko, O.; Jordan, M.A.; Wilson, L. Erucin, the Major Isothiocyanate in Arugula (Eruca sativa), Inhibits Proliferation of MCF7 Tumor Cells by Suppressing Microtubule Dynamics. PLoS ONE 2014, 9, e100599. [CrossRef][PubMed] Antioxidants 2020, 9, 1090 7 of 8

9. Yagishita, Y.; Fahey, J.W.; Dinkova-Kostova, A.T.; Kensler, T.W. Broccoli or sulforaphane: Is it the source or dose that matters? Molecules 2019, 24, 3593. [CrossRef][PubMed] 10. Hayes, J.D.; Kelleher, M.O.; Eggleston, I.M. The cancer chemopreventive actions of phytochemicals derived from glucosinolates. Eur. J. Nutr. 2008, 47, 73–88. [CrossRef] 11. Houghton, C.A. Sulforaphane: Its “Coming of Age” as a Clinically Relevant Nutraceutical in the Prevention and Treatment of Chronic Disease. Oxidative Med. Cell. Longev. 2019, 2019, 1–27. [CrossRef] 12. Melchini, A.; Traka, M.H. Biological Profile of Erucin: A New Promising Anticancer Agent from Cruciferous Vegetables. Toxins 2010, 2, 593–612. [CrossRef] 13. Cho, H.J.; Lee, K.W.; Park, J.H.Y. Erucin Exerts Anti-Inflammatory Properties in Murine Macrophages and Mouse Skin: Possible Mediation through the Inhibition of NFκB Signaling. Int. J. Mol. Sci. 2013, 14, 20564–20577. [CrossRef][PubMed] 14. Citi, V.; Piragine, E.; Pagnotta, E.; Ugolini, L.; Mannelli, L.D.C.; Testai, L.; Ghelardini, C.; Lazzeri, L.; Calderone, V.; Martelli, A. Anticancer properties of erucin, an H2S-releasing isothiocyanate, on human pancreatic adenocarcinoma cells (AsPC-1). Phytother. Res. 2019, 33, 845–855. [CrossRef] 15. Dinkova-Kostova, A.T.; Talalay, P. Direct and indirect antioxidant properties of inducers of cytoprotective proteins. Mol. Nutr. Food Res. 2008, 52, S128–S138. [CrossRef] 16. Valgimigli, L.; Iori, R. Antioxidant and pro-oxidant capacities of ITCs. Environ. Mol. Mutagen. 2009, 50, 222–237. [CrossRef] 17. Barillari, J.; Canistro, D.; Paolini, M.; Ferroni, F.; Pedulli, G.F.; Iori, R.; Valgimigli, L. Direct Antioxidant Activity of Purified Glucoerucin, the Dietary Secondary Metabolite Contained in Rocket (Eruca sativa Mill.) Seeds and Sprouts. J. Agric. Food Chem. 2005, 53, 2475–2482. [CrossRef] 18. Papi, A.; Orlandi, M.; Bartolini, G.; Barillari, J.; Iori, R.; Paolini, M.; Ferroni, F.; Fumo, M.G.; Pedulli, G.F.; Valgimigli, L. Cytotoxic and Antioxidant Activity of 4-Methylthio-3-butenyl Isothiocyanate from Raphanus sativus L. (Kaiware Daikon) Sprouts. J. Agric. Food Chem. 2008, 56, 875–883. [CrossRef] 19. Litwinienko, G.; Kasprzycka-Guttman, T. A DSC study on thermoxidation kinetics of mustard oil. Thermochim. Acta 1998, 319, 185–191. [CrossRef] 20. Litwinienko, G.; Kasprzycka-Guttman, T.; Studzinski, M. Effects of selected phenol derivatives on the autoxidation of linolenic acid investigated by DSC non-isothermal methods. Thermochim. Acta 1997, 307, 97–106. [CrossRef] 21. Litwinienko, G.; Kasprzycka-Guttman, T. The Influence of some Chain-Breaking Antioxidants on Thermal-Oxidative Decomposition of Linolenic Acid. J. Therm. Anal. Calorim. 1998, 54, 203–210. [CrossRef] 22. Litwinienko, G.; Kasprzycka-Guttman, T.; Jamanek, D. DSC study of antioxidant properties of dihydroxyphenols. Thermochim. Acta 1999, 331, 79–86. [CrossRef] 23. Musialik, M.; Litwinienko, G. DSC study of linolenic acid autoxidation inhibited by BHT, dehydrozingerone and olivetol. J. Therm. Anal. Calorim. 2007, 88, 781–785. [CrossRef] 24. Jin, Y.; Ou, J.; Rosen, R.T.; Ho, C.T. Thermal degradation of sulforaphane in aqueous solution. J. Agric. Food Chem. 1999, 47, 3121–3123. [CrossRef] 25. Vaidya, V.; Ingold, K.U.; Pratt, D.A. Garlic: Source of the Ultimate Antioxidants-Sulfenic Acids. Angew. Chem. Int. Ed. 2009, 48, 157–160. [CrossRef][PubMed] 26. Lynett, P.T.; Butts, K.; Vaidya, V.; Garrett, G.E.; Pratt, D.A. The mechanism of radical-trapping antioxidant activity of plant-derived thiosulfinates. Org. Biomol. Chem. 2011, 9, 3320–3330. [CrossRef] 27. Block, E. The Organosulfur Chemistry of the Genus Allium—Implications for the Organic Chemistry of Sulfur. Angew. Chem. Int. Ed. Engl. 1992, 31, 1135–1178. [CrossRef] 28. Koelewijn, P.; Berger, H. Mechanism of the antioxidant action of dialkyl sulfoxides. Recl. Trav. Chim. Pays-Bas 1972, 91, 1275–1286. [CrossRef] 29. Bateman, L.; Cain, M.; Colclough, T.; Cunneen, J.I. Oxidation of organic sulphides. Part XIII. The antioxidant action of sulphoxides and thiolsulphinates in autoxidizing squalene. J. Chem. Soc. 1962, 3570–3578. [CrossRef] 30. Block, M.R. Chemistry of alkyl thiosulfinate esters. II. Sulfenic acids from dialkyl thiolsulfinate esters. J. Am. Chem. Soc. 1972, 94, 642–644. [CrossRef] 31. Block, M.R. Chemistry of alkyl thiolsulfinate esters. III. tert-Butanethiosulfoxylic acid. J. Am. Chem. Soc. 1972, 94, 644–645. [CrossRef] Antioxidants 2020, 9, 1090 8 of 8

32. Gupta, V.; Carroll, K.S. Sulfenic acid chemistry, detection and cellular lifetime. Biochim. Biophys. Acta Gen. Subj. 2014, 1840, 847–875. [CrossRef] 33. Kubec, R.; Drhová, V.; Velíšek, J. Thermal Degradation of S-Methylcysteine and its Sulfoxide Important Flavor Precursors of Brassica and Allium Vegetables. J. Agric. Food Chem. 1998, 46, 4334–4340. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).