Box–Behnken factorial design to obtain a phenolic-rich extract from the aerial parts of Chelidonium majus L

Description
Box–Behnken factorial design to obtain a phenolic-rich extract from the aerial parts of Chelidonium majus L

Please download to get full document.

View again

of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Shopping

Publish on:

Views: 6 | Pages: 9

Extension: PDF | Download: 0

Share
Tags
Transcript
  Box – Behnken factorial design to obtain a phenolic-rich extractfrom the aerial parts of   Chelidonium majus  L. Clara Grosso a , Federico Ferreres b, n , Angel Gil-Izquierdo b , Patrícia Valentão a ,Maria Sampaio a , Júlio Lima a , Paula B. Andrade a, n a REQUIMTE/Laboratório de Farmacognosia, Departamento de Química, Faculdade de Farmácia, Universidade do Porto, Rua de Jorge Viterbo Ferreira,no. 228, 4050-313 Porto, Portugal b Research Group on Quality, Safety and Bioactivity of Plant Foods, Department of Food Science and Technology, CEBAS (CSIC), P.O. Box 164, 30100 Campus University Espinardo, Murcia, Spain a r t i c l e i n f o  Article history: Received 28 April 2014Received in revised form13 June 2014Accepted 19 June 2014Available online 27 June 2014 Keywords: Box – Behnken design Chelidonium majus HPLC – DAD – ESI/MS n Flavonoids a b s t r a c t A Box – Behnken design (BBD) was developed to study the in 󿬂 uence of four parameters (  X  1 : % methanol;  X  2 : extraction time;  X  3 : extraction temperature;  X  4 : solid/solvent ratio) on two responses, namelyextraction yield and phenolics content of the aerial parts of   Chelidonium majus  L. The model presented agood  󿬁 t to the experimental results for the extraction yield, being signi 󿬁 cantly in 󿬂 uenced by  X  1  and  X  4 .On the other hand a parameter reductionwas necessary to run the model for phenolics content, showingthat only  X  1  and  X  2  had great in 󿬂 uence on the response. Two best extraction conditions were de 󿬁 ned:  X  1 ¼ 76.8% MeOH,  X  2 ¼ 150.0 min,  X  3 ¼ 60.0  1 C and  X  4 ¼ 1:100 and  X  1 ¼ 69.2%,  X  2 ¼ 150 min,  X  3 ¼ 42.5  1 Cand  X  4 ¼ 1:100.Moreover, the HPLC – DAD – ESI/MS n analysis conducted with the center point sample revealed thepresence of 15 alkaloids and 15 phenolic compounds, from which the 9  󿬂 avonoids and 3 hydroxycin-namic acids are described for the  󿬁 rst time. Only phenolic compounds were quanti 󿬁 ed by a validatedHPLC – DAD method, the pair quercetin-3- O -rutinoside þ quercetin-3- O -glucoside dominating all the 29extracts. This study is of great importance for future works that seek to apply the phenolic pro 󿬁 le to thequality control of   C. majus  samples. &  2014 Elsevier B.V. All rights reserved. 1. Introduction The need to optimize process variables is emerging in different 󿬁 elds of research and industry, due to the increasing demand forfast, simple and ef  󿬁 cient methodologies with reduced costs andwastes. The use of statistical tools, such as response surfacemethodology (RSM), allows  󿬁 nding the best set of independentvariables or factors (input variables) that produce the optimumresponse (output variable). The advantage of this kind of approachis that it enables obtaining more information about the variablesand their interactions with fewer experiments than the traditionalunivariate procedures [1,2]. Box – Behnken design (BBD), a type of RSM, is a second-order multivariate technique based on three-level incomplete factorial design. BBD has been widely applied inthe past decade to optimize the extraction procedure of bioactivecompounds from natural sources, such as phenolic compounds [3 – 8].The determination of phenolic compounds involves a generalanalytical strategy that, besides a recovery step, includes theirstructural characterization and quanti 󿬁 cation. Among the analy-tical methodologies available, the most widely employed are basedon reversed-phase high-performance liquid chromatography (RP-HPLC) coupled to diode array detection (DAD) and/or massspectrometry (MS) with atmospheric pressure ionization techni-ques, i.e., electrospray ionization (ESI) or atmospheric pressurechemical ionization (APCI). HPLC – MS, particularly HPLC coupled totandem MS (HPLC – MS n ), has been recognized as the best tool toanalyze biological samples due to its selectivity, sensitivity andspeed of analysis [9,10]. HPLC – MS n allows obtaining more infor-mation concerning molecular structure, such as the type of aglycone moiety and substituents present, the stereochemicalassignment of terminal monosaccharide units, the sequence of the glycan component, the interglycosidic linkages and the pointsof attachment of the substituents to the aglycone [9].The aim of this study was to develop a BBD to optimize theextraction of phenolic compounds from  Chelidonium majus  L. andto perform a systematic characterization of its phenolic pro 󿬁 le byHPLC – DAD – ESI/MS n . Only some free hydroxycinammic (caffeic,  p -coumaric and ferulic acids) and hydroxybenzoic (gentisic andContents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/talanta Talanta http://dx.doi.org/10.1016/j.talanta.2014.06.0430039-9140/ &  2014 Elsevier B.V. All rights reserved. n Corresponding authors. Tel.:  þ 351 220428654; fax:  þ 351 226093390. E-mail addresses:  federico@cebas.csic.es (F. Ferreres),pandrade@ff.up.pt (P.B. Andrade).Talanta 130 (2014) 128 – 136   p -hydroxybenzoic acids) acids, as well as  󿬂 avonoids aglycones(quercetin, kaempferol and apigenin) were previously identi 󿬁 edby HPLC – DAD, after enzymatic hydrolysis [11,12]. Additionally,Hahn and Nahrstedt [11] isolated and characterized four hydro-xycinnamic acid derivatives, namely (  )-2-( E  )-caffeoyl- D -glycericacid, (  )-4-( E  )-caffeoyl- L  -threonic acid, (  )-2-( E  )-caffeoyl- L  -threonic acid lactone and ( þ )-( E  )-caffeoyl- L  -malic acid, by NMR.On the other hand, the most studied bioactive compounds fromthis species are the isoquinoline alkaloids, which are found in highamounts and revealed antimicrobial, anticancer, anti-in 󿬂 amma-tory, adrenolytic, sympatholytic, anticholinesterase and anti-MAO-A activities [13 – 17].Taking into account that the most important factors affectingthe extraction performance are the type of solvent, extractiontime, extraction temperature and solid/solvent ratio [5], our  󿬁 rstpurpose was to develop a four-factor BBD to  󿬁 nd the optimumextraction conditions for obtaining phenolic compounds from  C.majus . Moreover, the second goal was to develop and validate anHPLC – DAD method for phenolic compounds quanti 󿬁 cation. 2. Materials and methods  2.1. Plant materialC. majus  aerial parts (Lot. 1044 05 13) were purchased fromMorais e Costa & CA. Lda (Porto, Portugal). After being powderedto a mean particle size below 910  m m, the plant material wasstored desiccated at room temperature, protected from light. Thevoucher specimen (Cm-A-042013) was deposited at the Labora-tory of Pharmacognosy of the Faculty of Pharmacy of PortoUniversity.  2.2. Chemicals and standards Methanol (MeOH) Lichrosolv and acetic acid 100% were purchasedfrom Merck (Darmstadt, Germany) and VWR (Fontenay-sous-Bois,France), respectively. Caffeic acid and quercetin-3- O -rutinoside wereobtained from Sigma-Aldrich (St. Louis, MO, USA) and kaempferol-3- O -rutinoside, isorhamnetin-3- O -glucoside and isorhamnetin-3- O -rutinoside were from Extrasynthèse (Genay, France).  2.3. Extraction procedure All extractions were performed with ca. 1 g of plant material.Different combinations of four parameters were tested to extractphenolic compounds from  C. majus , namely % of MeOH (0 – 100%),extraction time (30 min of sonication, followed by 30 – 90 min of stirring maceration plus 30 min of sonication), temperature (25 – 60  1 C) and solid/solvent ratio (1:50 – 1:150). After, the solvent wasevaporated under reduced pressure and the extracts were storedat   20  1 C until further use.  2.4. Factorial design The software Design Expert (version 6.0.8, Stat-Ease Inc.,Minneapolis, MN, USA) was used for experimental design, dataanalysis and model building.BBD was employed to  󿬁 nd the optimum extraction conditionsfor obtaining the highest extraction yield and amount of phenoliccompounds.BBD requires an experiment number according to N  ¼ 2 k ð k  1 Þþ cp where  k  is the number of factors (or independent variables) and  cp is the number of the center points.The independent variables chosen for this study, namelypercentage of MeOH (  X  1 ), extraction time (  X  2 ), extraction tem-perature (  X  3 ) and solid – solvent ratio (  X  4 ), were evaluated at threedifferent levels (  1, 0, 1) and coded according to the followingequation:  x i ¼  X  i   X  0 Δ  X  i ¼ 1 ; 2 ; 3 ; 4where  x i  is the coded value of an independent variable,  X  i  is theactual value of an independent variable,  X  0  is the actual value of anindependent variable at the center point and  Δ  X   is the stepchange value of an independent variable. The coded and uncodedlevels of the four independent variables are given in Table 1. Intotal, 29 experiments were performed in triplicate, with  󿬁 verepetitions of the center point (Table 2).In order to predict the optimal responses (extraction yield andcontent of phenolic compounds), the following second-orderpolynomial equation was used to  󿬁 t the experimental data: Y  ¼  α  0 þ  ∑ 4 i ¼ 1 α  i  X  i þ  ∑ 4 i ¼ 1 α  ii  X  2 i  þ  ∑ 3 i ¼ 1 ∑ 4  j ¼ i þ 1 α  ij  X  i  X   j  Table 1 Coded and uncoded levels of the four independent variables. Independent variables Low Center High  X  1  (% MeOH)   1 (0%) 0 (50%)  þ 1 (100%)  X  2  (extraction time)   1 (90 min) 0 (120 min)  þ 1 (150 min)  X  3  (extraction temperature)   1 (25  ○ C) 0 (42.5  ○ C)  þ 1 (60  ○ C)  X  4  (solid/solvent ratio)   1 (1:50) 0 (1:100)  þ 1 (1:150)  Table 2 Extraction yields (%) and phenolic compounds content (mg/kg of extract) as afunction of the four independent variables. Run  X  1 a  X  2  X  3  X  4  Extraction yield Phenolics content1  100% MeOH 90 42.5 1:100 10.4 2544.8 2  100% Water 90 42.5 1:100 21.3 372.1 3  100% MeOH 150 42.5 1:100 10.3 3746.8 4  100% Water 150 42.5 1:100 21.7 828.2 5  50% MeOH 120 25.0 1:50 14.6 4537.0 6  50% MeOH 120 60.0 1:50 15.6 2496.5 7  50% MeOH 120 25.0 1:150 21.5 6703.4 8  50% MeOH 120 60.0 1:150 20.9 3731.9 9  50% MeOH 120 42.5 1:100 21.5 3612.5 10  100% MeOH 120 25.0 1:100 9.4 3004.0 11  100% Water 120 25.0 1:100 21.8 324.0 12  100% MeOH 120 60.0 1:100 10.3 4397.3 13  100% Water 120 60.0 1:100 20.5 304.9 14  50% MeOH 90 42.5 1:50 15.3 2034.9 15  50% MeOH 150 42.5 1:50 16.3 3567.3 16  50% MeOH 90 42.5 1:150 22.6 2644.4 17  50% MeOH 150 42.5 1:150 21.4 4639.1 18  50% MeOH 120 42.5 1:100 19.3 4200.3 19  100% MeOH 120 42.5 1:50 9.8 2781.1 20  100% Water 120 42.5 1:50 17.1 432.3 21  100% MeOH 120 42.5 1:150 11.4 1934.7 22  100% Water 120 42.5 1:150 22.4 306.3 23  50% MeOH 90 25.0 1:100 17.6 3468.9 24  50% MeOH 150 25.0 1:100 20.7 3664.4 25  50% MeOH 90 60.0 1:100 17.6 4239.7 26  50% MeOH 150 60.0 1:100 21.7 3985.3 27  50% MeOH 120 42.5 1:100 20.4 3331.0 28  50% MeOH 120 42.5 1:100 18.5 2973.0 29  50% MeOH 120 42.5 1:100 19.6 2954.2 a  X  1  –  solvent (% MeOH);  X  2  –  extraction time (min);  X  3  –  extractiontemperature ( 1 C);  X  4  –  solid/solvent ratio. C. Grosso et al. / Talanta 130 (2014) 128 – 136   129  where  Y   represents the response variables,  α  0  is a constant,  α  i ,  α  ii and  α  ij  are the linear, quadratic and interactive coef  󿬁 cients,respectively.  X  i  and  X   j  are the independent variables [5,18].Only for the amount of phenolic compounds a transformationsquared root was required, since the ratio of maximum andminimum response was higher than 10.  2.5. HPLC  – DAD – ESI/MS  n qualitative analysis The center point extract was dissolved in MeOH/water mixture(1:1) and submitted to sonication, centrifugation at 12,000 rpmand  󿬁 ltration through 0.2  m m membrane.Chromatographic analyses were carried out on a Luna C 18 (2) 100A column (150  1.0 mm, 3  m m particle size; Phenomenex,Maccles 󿬁 eld, UK). The mobile phase consisted of two solvents:water (1% acetic acid) (A) and methanol (B), starting with 20% Band using a gradient to obtain 50% B at 30 min and 80% B at40 min. The  󿬂 ow rate was 20  m L/min and the injection volume3  m L. Spectral data from all peaks were accumulated in the range240 – 400 nm and chromatograms were recorded at 280, 320 and350 nm. The HPLC – DAD – ESI/MS n analyses were carried out in anAgilent HPLC 1200 series equipped with a diode array detector andmass detector in series (Agilent Technologies, Waldbronn, Ger-many). The HPLC consisted of a binary pump (model G1376A), anautosampler (model G1377A) refrigerated at 4  1 C (G1330B), adegasser (model G1379B) and a photodiode array detector (modelG1315D). The HPLC system was controlled by ChemStation soft-ware (Agilent, v. B.01.03-SR2). The mass detector was a Bruker iontrap spectrometer (model HCT Ultra) equipped with an electro-spray ionization interface and was controlled by LCMSD software(Agilent, v. 6.1). The ionization conditions were adjusted at 300  1 Cand 4.0 kV for capillary temperature and voltage, respectively.The nebulizer pressure and  󿬂 ow rate of nitrogen were 5.0 psiand 3 L/min, respectively. The full scan mass covered the rangefrom  m/z   100 up to  m/z   1200. Collision-induced fragmentationexperiments were performed in the ion trap using helium as thecollision gas, with voltage ramping cycles from 0.3 up to 2 V. Massspectrometry data were acquired in the negative ionization modefor the study of phenolic compounds and in the positive mode foralkaloids. MS 2 was carried out in the automatic mode.  2.6. HPLC  – DAD quantitative analysis For phenolic compounds quanti 󿬁 cation, 20  m L of each extract(29 in total) were analyzed in triplicate on an analytical HPLC unit(Gilson), using a Luna C 18  column (250  4.60 mm, 5  m m particlesize; Phenomenex, Torrance, CA, USA). The mobile phase consistedof water (1% acetic acid) (A) and methanol (B), starting with 20% Band using a gradient to obtain 50% B at 30 min and 80% B at40 min. The  󿬂 ow rate was 0.8 mL/min. Detection was achievedwith a Gilson diode array detector. Spectral data from all peakswere collected in the range of 200 – 400 nm. The data wereprocessed on Unipoint System software (Gilson Medical Electro-nics, Villiers le Bel, France). Peak purity was checked by thesoftware contrast facilities. Phenolic compounds quanti 󿬁 cationwas achieved by the absorbance recorded in the chromatogramsat 320 nm (for hydroxycinnamic acids) and 350 nm (for  󿬂 avo-noids) relative to calibration curves carried out with  󿬁 ve concen-trations (in triplicate) of each standard.Compounds  9 ,  12 ,  13  and  22  were not quanti 󿬁 ed because theywere present in trace amounts. Compounds  19  and  25  werequanti 󿬁 ed as kaempferol-3- O -rutinoside, the pair  20 þ 21  as quer-cetin-3- O -rutinoside, compounds  23 ,  24  (or  23 þ 24 ) and 26 þ  Ac þ 27  as caffeic acid, compound  28  as isorhamnetin-3- O -glucoside and compound  29  as isorhamnetin-3- O -rutinoside.With the purpose of validating the HPLC – DAD method, linear-ity, limit of detection (LOD), limit of quanti 󿬁 cation (LOQ), precisionand accuracy were determined. Linearity was evaluated from thecorrelation coef  󿬁 cients ( R 2 ) of the regression curves obtained foreach standard.LOD and LOQ were calculated from the residual standarddeviation ( σ  ) of the regression curves and the slopes ( S  ), accordingto the following equations: LOD ¼ 3.3 σ  / S   and LOQ  ¼ 10 σ  / S  .Precision (reproducibility) was determined by calculating therelative standard deviation (R.S.D.) from repeated injections of thesample corresponding to the center point (extract 18). Intradayprecision was calculated from  󿬁 ve replicate injections performedin the same day, while interday precision was determined with 󿬁 ve injections done in 5 consecutive days.For accuracy (recovery) evaluation, extracts at the center pointwere spiked with isorhamnetin-3- O -glucoside at three differentlevels: low (1.6  10  3 mg/mL), medium (1.6  10  2 mg/mL) andhigh (8.2  10  2 mg/mL). This compound was chosen because itwas the only one with available standard that did not co-elutedwith others. 3. Results and discussion  3.1. Phenolic compounds and alkaloids characterization The screening of the hydromethanolic extract of the aerial partsof   C. majus  by RP-HPLC – DAD – ESI/MS n revealed a chromatographicpro 󿬁 le (350 and 280 nm) in (Fig. 1) which  󿬁 rst part correspondedto peaks with UV spectra of alkaloids and that ionized in thepositive mode. The second part of the chromatogram showed Fig. 1.  HPLC-UV (350 and 280 nm) pro 󿬁 le of   C. majus  extract obtained with extraction conditions at the center point. Identity of compounds as in Tables 1 and 2. Ac:uncharacterized cinnamoyl acid derivative. C. Grosso et al. / Talanta 130 (2014) 128 – 136  130  other kind of compounds, less abundant or in trace amounts,possessing UV spectra of   󿬂 avonoids and that suffered ionization inthe negative mode.  3.1.1. Alkaloids The study of alkaloids was carried out by comparison of theirUV and MS spectra with literature data [19 – 26]. As so, dihydro-berberine ( 1 ), protopine ( 2 ), allocryptopine ( 3 ), chelidonine ( 4 ),coptisine ( 6 ), tetrahydrocoptisine ( 10 ), tetrahydroberberine ( 11 ),berberine ( 14 ), norchelidonine ( 15 ), chelerythrine ( 18 ), which havebeen already described in this species [27 – 32, among others],were tentatively characterized. Besides these, other alkaloids werealso detected, though their complete characterization was notpossible: alkaloid A ( 5 ), alkaloid B ( 7 ), alkaloid C ( 8 ), alkaloid D( 16 ) and alkaloid E ( 17 ) (Fig. 1, Table 3).  3.1.2. Phenolic compounds As indicated above,  󿬂 avonoids were found in low or even traceamounts. Therefore the majority of their UV spectra could not beproperly observed, although band I was at ca. 350 nm (Table 4),indicating that the hydroxyl at position 3 of the  󿬂 avonol was notfree. The structural characterization of the  󿬂 avonoids was mainlybased on their MS spectra. Four diglycosides ( 19 ,  20 ,  25  and  29 )and  󿬁 ve monoglycosides ( 21 ,  22  and  26 – 28 ) (Fig. 1, Table 4) were detected, corresponding to kaempferol ( 19 ,  25  and  27 ) ([Aglyc-H]-at  m/z   285), quercetin ( 20 – 22  and  26 ) ([Aglyc-H]  at  m/z   301) andisorhamnetin ( 28  and  29 ) ([Aglyc-2H/H]  at  m/z   314/315) deriva-tives (Table 4). In the MS 2 of the diglycoside  19 , besides the ion of the deprotonated aglycone ( m/z   285), ions at  m/z   447 (base peak)and 431 were also observed, which correspond to the loss of rhamnosyl (  146) and hexosyl (  162) radicals, respectively. Theloss of those radicals was not accompanied by the loss of water(  164/  180), indicating the absence of interglycosidic linkage[33,34]. This fragmentation is typical of di- O- glycosides with sugarresidues linked to different phenolic hydroxyls. Moreover, thepreferential fragmentation of   󿬂 avonol-3,7-di- O- glycosydes occursat the C7-OH, giving rise to the base peak [35] and, therefore,compound  19  can be tentatively characterized as kaempferol-3- O -hexoside-7- O -rhmanoside. Kite and Veitch [36] differentiatedkaempferol-3- O -glucoside-7- O -rhamnoside from kaempferol-3- O -galactoside-7- O -rhamnoside by the relative abundance of theion at  m/z   431, which corresponds to ca. 50% for the glucosylderivative, while for the galactosyl one it is  o 10%. Thus, com-pound  19  can be kaempferol-3- O -glucoside-7- O -rhamnoside. TheMS fragmentation pattern of compound  25  was similar to that of compound  19 , but the two sugars were equal (rhamnose) and theion at  m/z   431 ([(M-H)-146]  ) was observed as base peak(Table 3). So, it can be labeled as kaempferol-3,7-di- O -rhmanoside( 25 ). The diglycosides  20  and  29  are two rhamnosyl-hexosides,and in their MS fragmentations practically just the ion of theirdeprotonated aglycones (base peak) was observed, at  m/z   301(quercetin) and 315 (isorhamnetin), respectively. This type of diglycosides fragmentation, in which ions resulting from the break  Table 3 R t, UV and MS: [M] þ /[M þ H] þ and MS 2 [M] þ /[M þ H] þ data of alkaloids from  C. majus .  a Compound  R t (min) UV (nm) [M] þ /[M þ H] þ m/z   MS 2 [M] þ /[M þ H] þ m/z   (%) 1  Dihydroberberine 16.1 290 338 190(100) 2  Protopine 17.1 288 354 190(100), 149(80) 3  Allocryptopine 17.5  —  370 352(50), 188(100) 4  Chelidonine 17.9  —  354 323(100), 305(40), 295(10), 275(50) 5  Alkaloid A 18.4  —  338 190(100) 6  Coptisine 18.7 268, 348, 360, 460 320 293(100) 7  Alkaloid B 19.4  —  370 339(100), 321(25), 290(50) 8  Alkaloid C 19.6  —  354 190(100) 10  Tetrahydrocoptisine 20.4 288, 348 324 176(100), 149(60) 11  Tetrahydroberberine 20.9  —  340 176(100), 149(20) 14  Berberine 25.5 264, 276sh, 338, 346, 426 336 321(100), 292(10) 15  Norchelidonine 25.7 290 340 322(100) 16  Alkaloid D 26.3 290, 320 356 338(100), 308(40) 17  Alkaloid E 26.7 252, 300, 380 368 350(100), 306(20), 276(10) 18  Chelerythrine 27.2 272, 285sh, 320, 338, 426 348 333(100), 318(10), 304(20) a Main observed fragments. Other ions were found but they have not been included.  Table 4 R t, UV and MS: [M  H]  and MS 2 [M  H]  data of   󿬂 avonoids from  C. majus. a Compound b R t (min) UV (nm) [M  H]  ,  m/z   MS 2 [M  H]  ,  m/z   (%)-146 -162 [Aglc-H/2H]  19  kaempf-3-glc-7-rhmn 28.4  —  593 447(100) 431(47) 285(33) 20  querct-3-rhmn(1 - 6)glc 29.4 256, 264sh, 298sh, 354 609 301(100) 21  querct-3-glc 29.8  —  463 301(100) 22  querct-3-gluc 30.5  —  477 301(100) 25  kaempf-3,7-di-rhmn 32.5 264, 316sh, 348 577 431 (100) 285(50) 26  querct-3-rhmn 33.7  —  447 301(100) 27  kaempf-3-glc 34.0  — c 447 285(100) 28  isorhamnt-3-glc 34.4  —  477 314(100) 29  isorhamnt-3-rhmn(1 - 6)glc 34.7  —  623 315(100) a Main observed fragments. Other ions were found but they have not been included. b kaempf: kaempferol; querct: quercetin; isorhamnt: isorhamnetin; glc: glucoside; gluc: glucuronide; rhmn: rhamnoside. c Compound  27  co-eluted with a cinnamoyl acid derivative not fully characterized, which did not allowed to observe its UV spectrum. C. Grosso et al. / Talanta 130 (2014) 128 – 136   131  of the interglycosidic linkage are not observed, is typical of the 1 - 6 linkage, which is hard to be broken [33,34]. Compound  20 chromatographically matched with the rutin (quercetin-3- O -rhamnosyl(1 - 6)glucoside) standard. Thus, and taking intoaccount its chromatographic mobility, compound  29  can belabeled as isorhamnetin-3- O -rhamnosyl(1 - 6)glucoside.As above indicated, the remaining  󿬂 avonoids are monoglyco-sides. Compounds  21 ,  27  and  28  presented similar MS spectrum,characterized by the loss of a 162 amu fragment (hexosyl radical)to give rise to the ions of their deprotonated aglycones as basepeak ( 21 ,  m/z   301 [quercetin-H]  ;  27 ,  m/z   285 [kaempferol-H]  ; 28 ,  m/z   314 [isorhamnetin-2H]  ). The chromatographic mobilityin reverse phase of compound  21  relative to that of rutin ( 20 )indicates that it should be quercetin-3- O -glucoside. Moreover, therespective chromatographic mobility of compounds  21 ,  27  and  28 showed that they are substituted by the same hexose. Thus,compounds  27  and  28  can be labeled as kaempferol-3- O -glucosideand isorhamnetin-3- O -glucoside, respectively. The MS fragmenta-tion of compound  22  exhibited the loss of 176 amu (glucuronoylradical), indicating that it is quercetin-3- O -glucuronide. The loss of  Fig. 2.  Three - dimensional (3D) response surface and contour plots for extraction yield. (A)  X  1   X  2 ; (B)  X  1   X  3 ; (C)  X  1   X  4 ; (D)  X  2   X  3 ; (E)  X  2   X  4 ; and (F)  X  3   X  4 . C. Grosso et al. / Talanta 130 (2014) 128 – 136  132
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks