Total phenolics and condensed tannins in field pea (Pisum sativum L.) and grass pea (Lathyrus sativus L.)

Seed samples of seventeen field pea cultivars grown at five locations, and nine grass pea lines grown at two locations, in western Canada during 1993 and 1994 were analysed for total phenolics and condensed tannins. Each location in each year was
of 6
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.
Related Documents
   Euphytica  101:  97–102, 1998.  97 c    1998  Kluwer Academic Publishers. Printed in the Netherlands. Total phenolics and condensed tannins in field pea (  Pisum sativum  L.) andgrass pea (  Lathyrus sativus  L.) Xiaofang Wang 1 , Thomas D. Warkentin 2 ;   , Colin J. Briggs 1 , B. Dave Oomah 2 ,Clayton G. Campbell 3 & Sheila Woods 4 1 Faculty of Pharmacy, University of Manitoba, Winnipeg, MB, Canada, R3T 2N2;  2  Agriculture and Agri-Food Canada, Morden Research Centre, Unit 100-101, Route 100, Morden, MB, Canada, R6M 1Y5;  3 Kade Research, Ltd. 135-13 St., Morden, MB, Canada, R6M 1E9;  4  Agriculture and Agri-Food Canada, Cereal Research Centre,195 Dafoe Road, Winnipeg, MB, Canada, R3T 2M9; (    author for correpondence) Received 12 September 1996; accepted 30 September 1997 Key words:  condensed tannins, field pea, grass pea,  Pisum sativum  L.,  Lathyrus sativus  L., total phenolics Summary Seedsamplesofseventeenfieldpeacultivarsgrownatfivelocations,andninegrasspealinesgrownattwolocations,in western Canada during 1993 and 1994 were analysed for total phenolics and condensed tannins. Each locationin each year was consideredas one environment.Total phenolicsin field pea differedsignificantly among cultivars,ranging from 162 mg/kg DM (dry matter) (CE, catechin equivalents) for AC Tamor to 325 mg/kg DM (CE) forRichmond. Field pea had barely detectable levels of condensed tannins. Total phenolics in grass pea ranged from868 mg/kg DM (CE) for L880388 to 2059 mg/kg DM (CE) for LS89110. Condensed tannins in grass pea rangedfrom0.89g/kgDM(CE)forL880388to5.18g/kgDM(CE)forLS89125.Cultivarhadalargerrelativecontributionto total phenolic levelsin field pea and to total phenolicand condensed tannin levels in grass pea than environment.Total phenolic and condensed tannin levels were not correlated with seed yield and seed protein content in fieldpea or grass pea. Levels of total phenolicsand condensedtannins were positively correlatedin grass pea. Grass peaseeds with darker seed coat colour contained higher levels of condensed tannins. Introduction Field pea ( Pisum sativum  L.) is a major pulse crop inwestern Canada which diversifiescroppingoptionsforcereal growers. The area in production has increasedfrom 74,000 ha in 1985 to 800,000 ha in 1995(Saskatchewan Agriculture and Food, 1995). Themajority of the Canadian field pea crop is exported;the largest buyer is the western European compoundfeed industry. In addition, field pea is sold to manycountries for human consumption. Efforts are under-way in Canada to develop new export markets and todevelop the domestic feed market. Grass pea (  Lath- yrus sativus  L.) (chickling vetch or khesari) has highyield potential, drought tolerance and is an essentialfood crop for animals and humans in some countriesin West Asia and North Africa (Spencer et al., 1986).It could become a useful rotation crop in the brownsoil zoneofthe Canadianprairies, anarea thatlacksanadapted annual legume alternative. The first grass peacultivar in Canada, X850002,was recentlyreleased bythe Morden Research Centre.Improving the nutritional quality of field pea andgrass pea will improvethe market value of these cropsforexportanddomesticuse. Althoughthe proteincon-tent of legume seeds is high (field pea: 22–25%; grasspea: 20–30%), their protein quality is not ideal. Thishas been attributed to two factors: the deficiency of sulfur-containing amino acids (Bressani et al., 1973;Elias & Bressani, 1974) and the presence of antinu-tritional factors such as condensed tannins which canreduce protein intake by precipitating proteins, resist-ing digestive enzymes and increasing fecal nitrogen(Gatel & Grosjean, 1990).Low-molecular-weight phenolic compoundscausea bitter taste in foods and feeds. Bitterness in ani-  98mal feeds often leads to reduction in feed consump-tion (Price & Butler, 1980). Tannins are water-solublepolymeric phenolics with relatively high molecularweight which can precipitate proteins (Haslam, 1989).Hydrolysable and condensed tannins are two groupsofpolyphenols.Hydrolysabletanninsarepolyestersof sugars with gallic acid or ellegic acid. Condensed tan-nins are formed by the polymerization of flavan-3-ols(catechins) or flavan-3, 4-diols (anthocyanidins). Tan-nins can bind to amide groups of enzymes and otherproteinsbyhydrogenbondingtoforminsolubletannin-proteincomplexesmakinglarge blocksof amino acidsresistant to the digestive enzymes of monogastric ani-mals (Sosulski, 1979). Growth retardation has beenobserved in chicks and rats fed diets containing high-tannin sorghum (Chang & Fuller, 1964; Jambunathan& Mertz, 1973). The poor palatability of high-tannindiets can be ascribed to tannin’s astringent property,which is a consequence of its ability to bind with theprotein of saliva and mucosal membranes (Mehanshoet al., 1987).Deshpande & Campbell (1992) reported that grasspea lines grown at a single location in Manitoba var-ied quite widely in levels of condensed tannins andphenolics. To facilitate the breeding of field pea andgrass pea cultivars with improved nutritional quality,it was necessary to survey existing germplasm for lev-els of relevant antinutritional compounds. Thus, theobjectives of this study were to determine the relativeimportanceof genotypeand environmenton total phe-nolics and condensed tannins in field pea and grasspea, and to determine whether these compounds arecorrelated with seed protein content and yield. Materials and methods Seed samples of seventeen field pea cultivars com-monly grown in western Canada were obtained fromregional adaptation trials conducted at five locationsin Manitoba (Arborg, Dauphin, Minto, Rosebank andThornhill) during 1993 and 1994. For grass pea, seedsamples were obtained from nine breeding lines fromtheAgricultureandAgri-FoodCanada,Mordenbreed-ingprogramwhichhadbeengrowninpreliminarytestsin Morden and Portage la Prairie, Manitoba during1993 and 1994. Each location in each year was con-sidered as one environment. Each environment wasidentified by the location and the year (1993 or 1994).For both field pea and grass pea, a single compositesample of each genotype was obtained from each tri-al. Field pea cultivars used included yellow and greencotyledon types; all cultivars had white seed coats.Samples of grass pea had a range of seed coat coloursfrom white through dark brown.Whole grain samples of all field pea and grass peaseeds were ground to pass through a 20-mesh screenin a Wiley Intermediate Mill equipped with stainlesssteel blades. Moisture contents of sample flours weredetermined by the AOAC method (1990). Condensedtannin contents of ground samples were determinedaccording to the method of Burns (1971) as modifiedby Price et al. (1978) using 1% HCl in methanol asan extracting solvent and vanillin (0.5%) hydrochlo-ride (4%) in methanol as a chromogen. Appropriatesample blanks were used to eliminate interference byseed pigments. Sample size was 2.0 g for field peaand 0.1 g for grass pea; smaller samples were usedfor grass pea since concentrations of condensed tan-nins were greater than in field pea. Sample flours wereextractedwith 10 mL extractingsolventand shakenona reciprocatingshaker for20 min at roomtemperature.A water bath at 30    C was used for the 20 min intervalbetween mixing reagents and reading the absorbance.Absorbance at 500 nm was read in 1 cm cuvettes on aBeckman Model 50 spectrophotometer zeroed againsta reagent blank.Total phenolic contents were determined asdescribed by Price & Butler (1977) using the Prussianblue assay. Sample size was 1 g for field pea and 0.1 gfor grass pea; smaller samples were used for grass peasince concentrations of total phenolics were greaterthan in field pea. Sample flours were extracted with5 mL of absolute methanol and shaken on a recipro-cating shaker for 30 min. An aliquot was centrifugedat 13,000 rpm for 10 min (Biofuge, Baxter DiagnosticCorp., Ontario, Canada). The supernatant was dilut-ed 100 times with distilled water mixed with 3 mLof 0.1 M FeCl 3  in 0.1 N HCl for 3 min, followed bythe timed addition of 3 mL of 0.008 M K 3  Fe(CN) 6 .The absorbance was read after 10 min at 720 nm ona Spectronic 501 spectrophotometer which had beenzeroed with water (Oomah et al., 1995). A blank of identical composition,but omitting the sample extract,was analysed and subtracted from all other readings.Allsampleswereextractedwithinthreedaysofmillinginto flour. The standard curves for total phenolics andcondensedtanninwere preparedwith catechin (+) (3.5moles of H 2 O per mole of catechin, Sigma Chemi-cal Co.), and the results were expressed in terms of catechin equivalents on a dry weight basis, i.e., con-densed tannin was expressed as g catechin per kg dry  99 Table 1 . Total phenolics levels of field pea cultivars grown at ten environment conditionsCultivar Breeder Total phenolics in catechinequivalents (mg kg    1 DM)Mean Estimate(square root)AC Tamor Agriculture and Agri-Food Canada 162 12.9Spring D Danisco, Denmark 167 13.0Fluo Blondeau, France 167 13.1Celeste Nickerson, France 182 13.6Montana Cebeco, Netherlands 193 13.8Bohatyr Selgen, Czech Republic 191 13.9Carneval Sval¨of-Weibull, Sweden 192 13.9Titan Agriculture and Agri-Food Canada 193 13.9Baroness Sharpes, UK 201 14.2Highlight Sval¨of-Weibull, Sweden 211 14.4Miko IHAR, Poland 217 14.6Trump Agriculture and Agri-Food Canada 228 15.0Patriot Sval¨of-Weibull, Sweden 247 15.4Orb Sharpes, UK 248 15.5Danto Prodana, Denmark 257 15.7Express Sval¨of-Weibull, Sweden 280 16.3Richmond Sval¨of-Weibull, Sweden 325 17.3Mean 215 14.5Standard error of difference 0.7Estimate based on random effects model analysis of variance. Least significant differencebetween estimates at 5% level is 1.4 for pre-planned comparisons. Degrees of freedom for ttest between cultivars are 144. flour, total phenolics were expressed as mg catechinper kg dry flour. Protein content (N    6.25) was deter-minedaccordingtotheKjeldahlmethodwithaTecatordigester and a Kjeltec (System 1002) distillation unit(Tecator AB, H¨ogan¨as, Sweden).Seed coat colour of grass pea was determinedusing a Hunterlab colour difference meter Model D25(Hunter Associated Laboratory, Inc., McLean, Vir-ginia) calibratedby a standardwhite plate (L = 90.67).Three readings were measured on each 25 g sample.Seed coat colour was expressed as mean Hunter L val-ues. L value represents lightness and varies from 100for perfect white to 0 for black.All chemicals used were of reagent grade. Allchemicalanalyseswereperformedatleastinduplicate.Data were analysed using a randomeffects model, andvariancecomponentsandcultivarmeanswereestimat-edusingtheSASprocedureMixed(SASInstitute,Inc.,1992). A square root transformation was applied to allvariables except L value, to stabilize the variances.Correlation analysis was conducted on the estimatedcultivar means using SAS (SAS Institute, Inc., 1990). Results and discussion Field pea.  Total phenolic levels in field pea differedsignificantly among the 17 cultivars tested (P < 0.05(Tables1,2).Meanvaluesrangedfrom162mg/kgDM(CE, catechinequivalents)forACTamorto 325mg/kgDM (CE) for Richmond. The level of total pheno-lics in a cultivar did not appear to be associated withthe breeder that developedit. For example, the Sval¨of-WeibullcultivarRichmondhadthehighestlevel, whileanother Sval¨of-Weibull cultivar, Carneval, was amongthe lowest. Condensed tannin levels in field pea werebarelydetectable.Analysisofvariancefortotalpheno-licsoffieldpea(Table2)showedthatcultivar(30.7%of variation) had a larger relative contribution than envi-ronment (18.1%).The correlation coefficients for total phenolics,seed yield, and seed protein content in field pea aresummarized in Table 3. Levels of total phenolics andseed yield were not correlated, nor were total pheno-lics andproteincontent.Protein contentand seedyieldwerenegativelycorrelatedinfieldpea(P<0.01)which  100 Table 2 . Variance components and indicator of heritability for total phenolics of field pea cultivars grownat ten environment conditionsCovariance parameter Estimate Standard error Probability Percent of totalCultivar 1.747 0.721 0.016 30.7Environment 1.029 0.566 0.069 18.1Interaction 2.917 0.344 0.000 51.2Heritability indicator (%) 37.4 10.2Square root transformed. Interaction includes error. Heritability indicator is cultivar divided by cultivar plusinteraction. Asymptotic covariance between cultivar and interaction variance components is – 0.0118. Table 3 . Correlation coefficients for total phenolics, seed yield,andseedprotein content infield peabetween cultivars (mean of10environments, n = 17). All variables are square root transformedSeed yield Seed protein contentTotal phenolics    0.150 0.244Seed yield    0.778 a a  P < 0.01. was likely induced by environmental variation acrosslocations. Grass pea.  The means of total phenolics and con-densed tannins in grass pea did not differ significantlyamong nine lines grown at two locations in 1993 and1994 (P > 0.05) (Tables 4, 5), possibly because of thesmall sample size. The mean values of total phenolicsvaried widely, ranging from 868 mg/kg DM (CE) forL880388 to 2059 mg/kg DM (CE) for LS89110. Themean values of condensed tannins also varied wide-ly, ranging from 0.89 g/kg DM (CE) for L880388 to5.18 g/kg DM (CE) for LS89125 (Table 4). The linesLS89110,LS89125,LS90040,LS90043andLS90045had high levels of condensed tannins. It should be not-ed that sampling a larger germplasm pool may haverevealed even greater variability for these traits.The results of analysis of variance for total phe-nolics and condensed tannins of grass pea (Table 5)showed that cultivar (26.6% of variation) had a larg-er relative contribution to total phenolics than envi-ronment (9.0% of variation). Similarly, for condensedtannins, cultivar (65.1%of variation) had a much larg-er relative contribution than environment (10.9% of variation).The correlation coefficients for total phenolics,condensed tannins, seed yield, seed protein content,and seed coat colour in grass pea are summarized inTable 6. Seed yield was not correlated with either totalphenolics or condensed tannins in grass pea. Totalphenolics and condensed tannins were positively cor-related (P < 0.01). A similar observation was madeby Deshpande & Campbell (1992). Protein contentsin grass pea were not correlated with total phenolics,condensed tannins or seed yield. A highly significantnegative correlation existed between condensed tan-nins and lightness of seed coat colour (P < 0.01), i.e.,grasspea lineswith darkerseed coatscontainedhigherlevels of condensedtannins(Table 6). A similar obser-vation was made by Deshpande & Campbell (1992).Total phenolics and L value were also negatively cor-related. There was no correlation between L value andseed yield or seed protein content.The initial emphasis of the grass pea breeding pro-gramhascentredonthereductionoreliminationoftheneurotoxin, 3-N-oxalyl-L-2,3-diaminopropanoic acid(ODAP) from seeds. Improvement of other traits,including purification of flower and seed coat colour,received lower priority. For this reason, all grass pealines analysed in this study were non-uniformfor seedcoat colour. Those with the greatest proportionof dark coloured seed coats (> 95%) had high levels of con-densed tannins and total phenolics. Similar observa-tions were made in common bean (Bressani & Elias,1980) and grass pea (Deshpande & Campbell, 1992).In addition to the nine grass pea lines described in theinitialanalysis,seedsamplesofsixgrasspealinesfromthe Morden breeding program grown in co-operativetests in Morden during 1993–1994 were also tested.Line LS87099 had lower levels of total phenolics andcondensed tannins (219 mg/kg DM (CE), 0.20 g/kgDM (CE), respectively) than lines in the initial analy-sis. LS87099 had about 98% of seeds with white seedcoats. Allcultivarsoffieldpeatestedhadwhiteflowersand white seed coats and contained condensed tanninlevelsnearzero,similartoresultssummarizedbyGatel& Grosjean (1990) for European cultivars. The dark seeded cultivar Sirius, which was used as an internalstandardintheseanalyses,hadhigherlevels(3.32g/kg  101 Table 4 . Total phenolics, condensed tannins, and seed coat colour (L value) in grass pea linesgrown at four environment conditionsLine Total phenolics Condensed tannins L valuein catechin equivalents in catechin equivalents(mg kg    1 DM) (g kg    1 DM)Mean Estimate Mean Estimate Mean(square root) (square root)L880388 868 31.3 0.89 1.01 49.2L880294 878 31.8 2.00 1.42 43.3LS89026 1200 35.1 2.15 1.49 49.0L900431 1385 36.8 3.09 1.74 46.8LS90045 1464 37.5 3.93 1.93 35.8LS90040 1547 38.2 5.06 2.19 35.4LS90043 1603 38.5 4.37 2.06 36.1LS89125 1623 38.5 5.18 2.19 39.0LS89110 2059 41.7 4.27 2.04 37.7Mean 1403 36.6 3.44 1.79 41.4Standard error of difference 3.7 0.17Estimate based on random effects model analysis of variance. Least significant difference betweenestimates at 5% level in pre-planned comparisons is 7.6 for total phenolics, 0.36 for condensedtannins. Degrees of freedom for t-test between lines are 24. Table 5 . Variance components and indicator of heritability for total phenolics and condensed tannins of grass pea lines grown at four environment conditionsCovariance parameter Estimate Standard error Probability Percent of totalTotal phenolicsLine 18.0 14.8 0.223 26.6Environment 6.1 9.1 0.498 9.0Interaction 43.4 12.5 0.001 64.3Heritability indicator (%) 29.3 19.2Condensed tanninsLine 0.180 0.098 0.067 65.1Environment 0.030 0.031 0.327 10.9Interaction 0.066 0.019 0.001 24.0Heritability indicator (%) 73.1 12.4Square root transformed. Interaction includes error. Heritability indicator is line divided by line plusinteraction. Asymptotic covariance between line and interaction variance components is    39.3 for totalphenolics,    0.914    10    4 for condensed tannins. Table 6  . Correlation coefficients for total phenolics, condensed tannins, seed yield, seed protein content, andseed coat colour (L value) between grass pea lines (means over 4 environments, n = 9). All variables exceptcolour are square root transformedCondensed tannins Seed yield Seed protein content L valueTotal phenolics 0.889 a   0.423 0.282    0.707 b  Condensed tannins 0.256 0.242    0.846 a  Seed yield 0.653    0.405Seed protein content    0.070 a  P < 0.01, b   P < 0.05.
Similar documents
View more...
Related Search
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