inhibitory effect of marinades with hibiscus extract on formation of heterocyclic aromatic amines and sensory quality of fried b

1、Inhibitory effect of marinades with hibiscus extract on formation of heterocyclic aromatic amines and sensory quality of fried bscavenging effect of antioxidants on pyrazine cation radicals thatparticipate in the formation of HAA has been demonstrated by adecrease in electron spin resonance signals

2、in the present of anti-oxidants (Kikugawa, 1999). The application of red wine, beer,plant extracts, or combined spice mixtures containing naturalantioxidants inhibited formation of HAA when used as marinadesprior to frying (Busquets, Puignou, Galceran, & Skog, 2006; Gibis,2007; Melo, Viegas,

3、 Petisca, Pinho, & Ferreira, 2021b; Nerurkar,Le Marchand, & Cooney, 1999; Smith, Ameri, & Gadgil, 2021). Thisbrief pretreatment has the advantage that the products are notoverly spiced and do not develope negative sensory characteristicsas only the surface is treated.The obje

4、ctive of this study was to examine the feasibility ofinhibiting HAA formation in fried beef patties using marinadescontaining hibiscus extract. For the determination of the antioxi-dant capacity, the water phase of marinades with hibiscus extractwas analyzed by the TEAC-Assay and the amounts of phen

5、oliccompounds determined with the FolinCiocalteau-reagent. Thesensory quality of the patties after frying was determined using aranking test, by a trained sensory panel.2. Materials and methods2.1. MaterialsTheHAA-standardsIQ(2-amino-3-methylimidazo4,5-fquinoline), IQx (2-amino-3-methylimidazo4,5-f

6、quinoxaline),MeIQ (2-amino-3,4-dimethylimidazo4,5-f quinoline), MeIQx (2-amino-3,8-dimethylimidazo4,5-f(2-amino-3,4,8-trimethylimidazo4,5-f quinoxaline), 7,8-DiMeIQx(2-amino-3,7,8-trimethylimidazo4,5-f quinoxaline), PhIP (2-ami-no-1-methyl-6-phenylimidazo4,5-b pyridine), Trp-P-1 (3-amino-1,4-dimethy

7、l-5H-pyrido4,3-bmethyl-5H-pyrido4,3-b indole), Glu-P-1 (2-amino-6-methyldi-pyrido1,2-a:30,20-d imidazole), Glu-P-2 (2-Aminodipyrido1,2-a:30,20-dimidazole),AaC(2-amino-9H-pyrido2,3-bMeAaC (2-amino-3-methyl-9H-pyrido2,3-b indole), harmane(1-methyl-9H-pyrido3,4-b indole), and norharmane (9H-pyr-ido3,4-

8、b indole) with the following end concentrations in thestandard mixture 19.7, 22.8, 21.3, 13.6, 12.9, 12.5, 6.5, 5.9, 7.4,9.4, 5.3 and 5.0 ng in 100 lL methanol, were obtained from TorontoResearch Chemicals (Ontario, Canada). All stock solutions of eachsubstance were corrected by means of the extinct

9、ion coefficient(Hatch, Felton, Stuermer, & Bjeldanes, 1984). Caffeine (internalstandard 2.5 lg/mL in ultrapure water and methanol, 1 + 1, v/v),harmane and norharmane (standard mixture: 4.9 and 4.1 ng in100 lL methanol) were purchased from SigmaAldrich (Taufkir-quinoxaline),4,8-DiMeIQxindole)

10、,Trp-P-2(3-amino-1-indole),chen, Germany). Diatomaceous earth isolute was obtained fromSepartis, Germany, and propylsulfonic acid (PRS) (100 mg) and C18 Bond Elut?cartridges (500 and 100 mg) were purchased fromVarian Inc., USA. All other chemicals were analytical or gradientgrade for HLPC (Merck, Da

11、rmstadt, Germany).ABTS2,20-azino-bis(3-ethylbenzothiazoline-6-sulfonate),Trolox6-hydroxy-2,5,7,8-tetramethyl-chroman-2-carboxylicacid and ABAP 2,20-azobis(2-amidinopropane) dihydrochloridewere used for the determination of the antioxidant capacity(TEAC-Assay) and the FolinCiocalteu-reagent, respecti

12、vely. Thesechemicals were obtained from Sigma (St. Louis, MO, USA).For the analysis of free amino acids Ala (alanine), Arg (arginine),Asp (aspartic acid), Cys (cysteine), Glu (glutamic acid), Gly (gly-cine), Ile (isoleucine), Leu (leucine), Lys (lysine), Met (methionine),norleucine, Phe (phenylalani

13、ne), Pro (proline), Ser (serine), Thr(threonine), Trp (tryptophan), Tyr (tyrosine), and Val (valine), acet-yl chloride, trifluoric acetic anhydride and 2,6-di-tert-butyl-p-cre-sol were obtained from Sigma (St. Louis, MO, USA), while cationexchanger Dowex 50 WX8 (50100 mesh) was from Serva (Heidel-be

14、rg, Germany). 1-Butanol, 2-propanol, diethyl ether, methanol,2,4,6-trinitrophenol, petroleum ether were supplied by Merck(Darmstadt, Germany).2.2. Sample materials and marinade formulationDeep-frozenHitburger (Gro?ostheim, Germany) (70 g, 8 mm thick ? 113 mm ?105 mm) with the same charge number.The

15、hibiscus extract (Hibiscus sabdariffa) was purchased fromPlantextrakt (Vestenbergsgreuth, Germany). Hibiscus extract wasused in marinades at concentrations of 0.2, 0.4, 0.6, and 0.8 g per100 g marinade. For manufacture of the marinades, the water sol-uble extracts were first dissolved in pure water.

16、 Marinade emul-sions were then produced by homogenizing sun flower oil,emulsifier (citric acid esters of mono- and diglycerides of fattyacids, E472c, Cognis, Illertissen, Germany) and the respective di-luted extract (67.5 g/0.5 g/32 g) using a high-shear disperser (Ul-tra-Turrax, Janke and Kunkel, S

17、taufen, Germany). In addition tothe control marinade that did not contain any extract, a furthercontrol was only broiled using sun flower oil.beefpattieswereobtainedfromSalomon2.3. Determination of optimal heating timeTwo grill plates of a double contact grill (Nevada, Neum?rker,Hemer, Germany) were

18、 heated to 230 ?C. The deep-frozen pattieswere coated with refined sunflower oil and covered on both sidesby tin foil. For the determination of the optimal heating time, theFig. 1. Formula of commonly found HAA in fried meat products prepared under household conditions.736M. Gibis, J. Weiss/Meat Sci

19、ence 85 (2021) 735742patties were broiled for 120, 130, 140, 150, 160, 170 and 180 s. Foreach heating time, eight patties were fried. These were then mixedfor the chemical analysis. For the sensory tests, eight patties ateach heating time were fried and each patty was cut into four.2.4. Frying of ma

20、rinated pattiesFrying conditions similar to those used in the determination ofthe optimal heating time were applied. Prior to frying, the deep-frozen patties were coated with either refined sunflower oil orthe respective marinades at 5 g coating per side. The beef pattieswere then covered on both si

21、des by tin foil and broiled for 160 s.Patties reached a core temperature of 72 ?C and a surface temper-ature of <190 ?C at the end of the frying process. For each formula-tion, eight patties were fried, mixed together and analyzed. For thesensory tests, eight patties of each formulation were

22、fried and eachpatty cut into four pieces.2.5. Determination of HAAThe 15 polar and non polar HAA were determined using a mod-ified HPLC method (Gibis, 2007) based on a previously publishedmethod (Gross & Grueter, 1992). HAA were extracted from pattiesusing solid phase extraction. The samples

23、 were homogenized withsodium hydroxide (1 mol/L). The homogenate was divided intofour equal 20 g portions. Two portions were spiked with 100 lLof the standard mixture. Each of the four portions was mixed withdiatomaceous earth. HAA were extracted with dichloromethanewith 5% toluene (100 mL) and abso

24、rbed onto a coupled precondi-tioned cation exchanger PRS cartridge. After drying of the PRS car-tridge, and washing with hydrochloric acid (0.1 M HCl), the nonpolar HAA were eluted with a mixture of 15 mL 0.1 M HCl/metha-nol (2:3, v/v). For clean up, the non polar HAA were adsorbed ontoa C18 cartrid

25、ge (500 mg). After washing with ultrapure water anddrying, the non polar HAA were eluted with 1.2 mL a methanol/ammonia (25%) mixture (90:10, v/v) into a vial. After drying, theeluate was re-dissolved in 100 lL caffeine solutions (internal stan-dard). The polar HAA were eluted from the PRS cartridge

26、 withammonia acetate (20 mL 0.5 M, pH 8) and adsorbed onto a coupledpreconditioned C18 cartridge for clean up (100 mg, 1 mL methanoland 1 mL ultrapure water). After washing and drying, the polarHAA were eluted (with 0.8 mL methanol/ammonia (25%) mixture(90:10, v/v) and re-dissolved in 100 lL interna

27、l standard afterdrying.The HPLC equipment consisted of a Gynkotek HPLC-system(Gynkotek, Germering, Germany) Pump M480, autosampler Gina50, degasser (DG 1310 S), joined with a fluorescence detector (RF1002), diode array detector (UVD 320) and the Gynkosoft chroma-tography data system (version 5.50).

28、The HAA were analyzed usinga column, TSK-gel?ODS-80TM 2504.6 mm, 5 lm (Tosoh Biosci-ence, Stuttgart, Germany) and a guard column Supelguard? LC-18-DB (Supelco, USA). The mobile phase consisted of eluent A:0.01 M triethylamine phosphate buffer (pH 3.2), eluent B: 0.01 Mtriethylamine phosphate buffer

29、(pH 3.6), and eluent C: acetonitrile.HAA were separated with a gradient program at 1 mL/min 8275%A, 10% B and 815% C from 0 to 10 min; 8575% B and 1525% Cfrom 10 to 20 min, 05% A, 7555% B and 2545% C from 20 to29 min; 582% A, 5510% B and 458% C from 29 to 33 min. Forthe regeneration of the HPLC-colu

30、mn, the mobile phase was 75%C for 4 min and then was equilibrated for 10 min with the startingconditions to 52 min.UV-detection was at 258 nm and 3D-field for spectral plots(200360 nm). The adjustment of fluorescence detector (ex/em)was from 0 min (360/450 nm), 14 min (300/440 nm), 22 min(265/410 nm

31、), 24 min (305/390 nm), 25.5 min (265/410 nm) and28 min (335/410 nm). The peaks of HAA, as well as norharmaneand harmane in samples were identified by comparing the reten-tion times and UV-spectra with standards. Quantification was per-formed by the standard addition method (two samples and twowith

32、standard mixture spiked samples).2.6. Determination of antioxidant capacity by TEAC and FolinCiocalteu assaysFor the TEAC (Trolox-Equivalent-Antioxidant-Capacity)-Assay,0.20.8 g of hibiscus extract was made to 100 g with ultrapurewater, and then ultrasonicated for 10 min. Solutions were filteredand

33、subjected to a modified TEAC test method (Schilling et al.,2007), i.e. the antioxidant capacity was determined with theABTS*+radical cation assay (2,20-azino-bis(3-ethylbenzthiazo-line)-6-sulfonic acid diammonium salt). The TEAC-Assay repre-sents a second electron transfer based method in which ther

34、adical cation 2,20-azino-bis(3-ethylbenzothiazoline-6-sulfonate)ABTS*+ is discolored after reduction to ABTS. For each extract con-centration, different dilutions were prepared. A phosphate bufferwas prepared by mixing 818 mL of a sodium hydrogen phosphatesolution (66 mmol/L Na2HPO4) with 182 mL of

35、a potassium dihy-drogen phosphate (KH2PO4) solution (66 mmol/L), 8.766 g sodiumchloride (150 mmol) and adjusted to pH 7.4 with phosphoric acid.ABTS solution (0.5 mL) in phosphate buffer (20 mmol/L) was mixedwith 100 mL of an ABAP solution in phosphate buffer (2.5 mmol/L)and heated at 60 ?C for 15 mi

36、n in a water bath. The reaction wasinitiated by adding 1.96 mL of the ABTS*+solution to 40 lL of thesample, Trolox (standard) solutions or ultrapure water was usedas a control. After standing for 6 min at room temperature, theabsorbance was measured at 734 nm (photometer HP 8453, Agi-lent Technologi

37、es, Waldbronn, Germany). Aqueous solutions ofTrolox(6-hydroxy-2,5,7,8-tetramethyl-chroman-2-carboxylicacid) in a concentration range 501000 lmol/L were used for cali-bration of the TEAC-Assay. The results were calculated as lmol ofTrolox (Trolox antioxidant equivalents) per g of marinade.The FolinCi

38、ocalteu-Assay is an electron transfer based assayand measures reducing capacity. The content of total phenoliccompounds was determined by application of FolinCiocalteu-re-agent. The absorbance at 720 nm was determined after 60 minwith a UVvisible spectrophotometer (HP 8453, Agilent Technolo-gies, Wa

39、ldbronn, Germany) and calculated as gallic acid (gallic acidmg/g marinade) (Singleton, Orthofer, & Lamuela-Raventos, 1999).Solutions of gallic acid in the range 540 mg/L gallic acid were usedas calibration standards. For each extract, different dilutions wereproduced. Only dilutions with abs

40、orbances in the calibration rangewere used. All measurements were in duplicate.2.7. Determination of weight loss and colorFor the determination of weight loss, the patties (n = 8) wereweighed before frying and 1 h after heat treatment. For color mea-surement, the patties were analyzed by a Chroma Me

41、ter CR 200(Minolta, Osaka, Japan) 1 h after heating to determine their L*-val-ues (brightness). Eight patties at each heating time were measuredthree times.2.8. Determination of precursors and main compositionsThe contents of glucose and creatine/creatinine were deter-mined as described in Roche Dia

42、gnostics GmbH (Roche, Mann-heim, Germany) (Wahlefeld, Holz, & Bergmeyer, 1974). Thecomposition of the raw patties in terms of moisture, fat, protein,minerals (ash) and 4-hydroxyproline was determined by OfficialGerman Methods (Anonym, 2005).The determination of free amino acids was carried o

43、ut by GC(Brückner & Hausch, 1989). The internal standard 1-norleucineM. Gibis, J. Weiss/Meat Science 85 (2021) 735742737was added to each sample and the mixture was extracted with70 mL ultrapure water and 30 mL methanol for 60 min at 50 ?C.The mixture was concentrated in vacuum to appro

44、ximately 5 mL;then 5 mL of a saturated solution of picric acid added and the mix-ture mixed for 5 min and centrifuged. The supernatant was filteredinto a separating funnel and extracted three times with 20 mL por-tions of petroleum ether/diethyl ether (1:1, v/v) to remove lipo-philic compounds. The

45、aqueous phase was passed through acolumn packed with a Dowex 50 WX 8 cation exchanger. The ad-sorbed amino acids were desorbed with 30 mL of 4 M NH4OHand the eluate evaporated to dryness; the residue was derivatizedto n-trifluoric acid 2-propyl esters. In a reaction vial, 0.5 mL 2-pro-panol/acetylch

46、loride (8:2, v/v) was added to the residue. The vialwas closed with a Teflon-lined screw-cap and heated at 100 ?Cfor 1 h. Solvents were removed in a stream of nitrogen, 150 lLdichloromethane and 50 lL trifluoracetic anhydride were addedand the mixture heated at 150 ?C for 5 min. Solvents were remove

47、din a stream of nitrogen, the residue was dissolved in 100200 lL ofdichloromethane and 1 lL samples were injected into the gas chro-matograph, this was repeated three times. For each sample themeans and standard deviations were calculated. Gas chromatogra-phy was performed on a GC 14-A (Shimadzu Inc

48、., Kyoto, Japan)equipped with a flame ionization detector (FID) and an Apex Chro-matography Data System (Autochrom Inc., Milford, USA) using aOV 1701 column (25 m ? 0.25 mm ID) (Machery and Nagel, Dür-ren, Germany). Nitrogen was the carrier gas at 1 kg cm?2. Detectorand injector temperatures we

49、re 260 ?C, the split mode was 1:34.The temperature program was 5 min at 80 ?C then 8 ?C/min to105 ?C, 1.5 ?C/min to 122 ?C, 8 ?C/min to 210 ?C, then 20 min at210 ?C. All analyses were in duplicate.2.9. Sensory testsThe panelists first evaluated beef patties that had not been mar-inated to determine

50、the preferred (optimal) frying time. Sensoryscores were rated using a 10 cm scale for color (010, very lightto very dark), for flavor and odor (010, too little frying to over-done). The panelists (n = 15) ranked the samples by a mark onthe scale.Sensory panelists (students or experts) were trained u

51、singhibiscus extract marinades for fried beef patties. After frying thebeef patties from the most preferred frying time were served hotto the panelists. The panelists (n = 10) had to evaluate the pattiesin a hedonic rank sum test (a = 0.05). In addition, they had to de-scribe any impressions of nega

52、tive characteristic odors and flavors.2.10. Statistical analysisAll data for the concentrations of HAA, color and sensory anal-ysis were analyzed for normal distribution and subjected to ananalysis of variance using the GLM and Tukey-test (a = 0.05) ofSAS program version 9.0 (SAS Institute INC., Car

53、y, USA).Pearson correlation coefficients (r-values) between color, glu-cose, creatine, creatinine, free amino acids and HAA formation dur-ing frying of the beef patties were also calculated (SAS program).3. Results and discussion3.1. Sensory determination of preferred frying timeInitially, the panel

54、ists decided on the optimal frying time basedon the color of the fried patty. Then they evaluated odor and flavorto their decision. The color of all patties looked brown and appet-isingly. Patties got the highest scores for color and flavor at fryingtimes of 150 and 160 s (Fig. 2). The panelists wer

55、e not able to sig-nificantly differentiate between products heated for these twotimes. The other frying times were significantly differentiated fromthese two times (p < 0.05). Some panelists preferred the darker andoverdone products, because they liked a well done crust andFig. 2. Hedonic sen

56、sory test of beef patties broiled at different heating times (means with the same letters are not significantly different, p < 0.05, n = 15).Fig. 3. Weight loss and L*-value (lightness) of beef patties fried and coated withsunflower oil as a function of frying time (n = 8).738M. Gibis, J. Wei

57、ss/Meat Science 85 (2021) 735742material and the first frying time was approximately 50% with theexception of alanine and glycine. With further increased fryingtime, only a slight decrease in the concentrations of all amino acidswas observed and eventually concentrations remained constant.The concen

58、tration of glucose in the raw material was reduced byapproximately 85% for the longest frying time.The reduction of creatine, glucose and free amino acidscorrelated with the formation of HAA including norharmane andharmane. With the exception to 4,8-DiMeIQx, correlation coeffi-cients between glucose

59、, creatine, total free amino acids and differ-ent HAA and broiling times were ?0.76 to ?0.86, ?0.74 to ?0.8,and ?0.6 to ?0.76, respectively (p < 0.001). The correlation coeffi-cients between creatinine and different HAA were positive at0.690.88 (p < 0.001) due to the formation of this compound bycyclization of creatine during heating. However, the formation ofHAA is only one reaction within the very complex Maillard reactionscheme. Most compounds generated from free amino acids andglucose are Maillard browning substances. In other studies, creat-inine was described as a pre