Calcium Oxalate Crystals: The Irritant Factor in Kiwifruit CONRAD 0. PERERA, IAN C. HALLETT, TUAN T. NGUYEN, and JUDITH C. CHARLES MATERIALS ABSTRACT The cause of irritation in the mouth, when kiwifruit nectars and dried kiwifruit products are ingested, was investigated. Idioblast cells that contain raphide crystals of calcium oxalate were isolated from inner pericarp tissue of the fruit and studied by light and electron microscopy. An experienced panel of 9 judges detected an irritation from sweetened apple puree when isolated raphide crystals were incorporated at the rate of 30 mg oxalate/lOOg of puree. Sensory and microscopic studies showed evidence that the irritation was causedby sharp calcium oxalate crystals exposed during processing. INTRODUCTION THE IRRITANT PROPERTYof certainplantshasbeenthe topic of manystudiessinceBigelow(1818)reportedhis findings of an irritant factor in Arum triphyllumin the earlynineteenth century. In those early studies, several chemical compoundswere implicatedas causativeagentsof the irritation: They included alkaloids, glucosides, sapotoxins and en- zymes(WalterandKhanna,1972). Calciumoxalatein a variety of plantshasbeenwell established (Black, 1918). Kohl (1889) and Ziegenspeck(1915) working independentlyshowedthe fine needle-likecrystals, knownasraphides,oftenfoundin bundles,consistof calcium oxalate.Plantswhich producecalciumoxalatein this form, wheneatenraw, areknown to producea painful sensationin themouth(Black, 1918).Black (1918)in his studyof Dasheen (Colocassia e.scufunta (L.) Schott),concludedthe solecause of theacridtastein Dasheenwasdueto a mechanicalirritation of the mucousmembranes in the mouthby the actionof calcium oxalatecrystals. A vast amountof informationhas beengatheredover the yearson the occurrenceof differentforms of calciumoxalate crystalsin variouspartsof plants(Fassett,1973; Al-Rais et al., 1971;Oke, 1969;Black, 1918;FranceschiandHornerJr., 1980;SakaiandHanson,1974;Libert andFranceschi,1987). However,only a few fruits areknownto containraphides.The presence of these crystals has been documented in pineapple (Miller, 1928)and in the fruit of Monstera deliciosu (Peters and Lee, 1977).Althoughthe calciumoxalatecrystalsin kiwifruit plant havebeenstudiedextensivelyfrom a plant nutrition point of view (Clarket al., 1987),only brief mentionhas beenmadeof theirpresence in thefruit (Strauss,1970;Schmid, 1978;OkuseandRyugo, 1981;Ferguson,1984). Theconsumption of certainprocessed kiwifruit productssuch as nectarsand dried fruit has beenknown for sometime to sometimescausean irritationin the mouth. Our researchwas undertaken with the view of elucidatingthe causeof this irritation. In our report, the term “catch” was usedto describe the irritation of the mucousmembranes in the mouth, caused by the ingestionof processed kiwifruit products. Authors Perera and Charles are with DSIR Fruit and Trees, and author Hallett is with DSIR Plant Protection, Dept. of Scientific & Industrial Research, Private Bag, Auckland New Zealand. Author Nguyen’s present address is N.S. W. Dairy Corp., Chippendale. Svdnev, Australia. 1066-JOURNAL Of FOOD SCIENCE-Volume 55, No. 4, 1990 & METHODS MATURE KIWIFRUITS (Achnidia deliciosa (A. Chev.) C.F. Liang et A.R. Ferguson var. deliciosa) were harvested from the DSIR researchorchard in Kumeu on May 26, 1987.&/ThcaverageBrix reading at harvest was 7.5”. They were stored at 1°C until used. The fruits were ripened by dipping for 5 min in an aqueous solution containing 1000 ppm ‘Ethrel’ (Zchlorocthyl phosphonic acid) and holding at 26°C. They were used when a flesh firmness of 0.6 to 0.7 kg pressure was reached as determined by a universal pcnetrometer fitted with an 8 mm diameter probe. The apple puree used to simulate the “catch” was a standard consumer product having a titratable acidity of 0.45% as malic acid which was sweetened to 10”Brix with sugar (New Zealand Apple and Pear Marketing Board). Nectar preparation Ripe fruits were cut in half and the flesh scoopedout by hand using a spoon. It was gently pressedwith a wooden spoon taking care not to break seeds. The seedswere removed using a laboratory pulp finisher having a screen mesh size of 0.6 mm. The titratable acidity and the Brix value (Abbe refractometer at 20°C) were determined on the pulp. The nectarswere preparedas follows: To a weighed amount of fruit pulp, the amounts of water, sugar and citric acid required to produce a finished nectar with pulp content 50%, soluble solids 13”Brix and titratable acidity (as citric acid) 0.825% were calculated and incorporated into the pulp. The contents were homogenized using a Silverson laboratory homogenizer for 3 min. The nectars thus produced were smooth in texture. To produce an enzyme-treated nectar, the deseededpulp was incubatedwith 200 ppm commercial pectinaseenzyme such as Rohapect DSL (Rohm Enzyme, West Germany) at 35°C for 3 hr. The calculated water, sugar and citric acid required to produce a nectar with the desired specifications were then added. The nectars were homogenized 3 min as before and further homogenized using an Ultra Turrex at the maximum speed 2 min to break up crystal bundles and disperse individual crystals uniformly. Isolation of raphide crystals Ripe kiwifruits were halved transversely. The inner-pericarp area adjacent to the seeds (referred to as the locular region) was collected by vacuum suction, using a Pasteur pipette as shown in Fig. 1. The isoIated locular tissue was washed with 1% sodium dodecyl sulfate (SDS) in water to remove pigments and water soluble components. It was then centrifuged at 400 x g for 5 min. The sediment was resuspended in 1% SDS solution as before. The process was repeated five times until a translucent gel-like sediment was obtained. This sediment was suspendedin four parts of dimethyl sulfoxide (DMSO) and stirred overnight at ambient temperature to remove starch granules. It was again centrifuged at 400 x g and the sediment was washed with water at least five times to remove any traces of DMSO. The washed sediment was suspendedin five parts 0.05M citrate buffer at pH 4.2 and incubated with 200 ppm of Rohapect D5L at 35°C for 3 hr to break down cell wall material. It was centrifuged as before and the dull white amorphous sediment was suspendedin the minimum quantity of water. An aliquot was taken for the determination of insoluble oxalate content by a high pressure liquid chromatography (HPLC) procedure. Analysis of oxalic acid The soluble and total oxalic acid in kiwifruit were measured by a modification of the HPLC method of Libert (1981). Ten to fifteen perature 25°C. Oxalic acid and other substanceswere detected at 224 nm. A working standardof oxalic acid was preparedby dilution of stock oxalic acid (4mM) 1:l with 0.5% KH,PO, (pH 2.00) just before injection. The HPLC system was calibrated by injecting 20 PL of working standard. Calibration was completed when results from 10 injections of the standard showed ~~0.05 level of significant difference among the values. Duplicate injections of 20 +I, sampleswere used for analysis. Quantification was achieved by comparing peak heights. The concentrations of the insoluble oxalate (calcium oxalate crystals) was expressed as anhydrous oxalic acid by subtracting the soluble oxalic acid from the total oxalic acid. Sensory evaluation Fig. 1 -Extraction of the locular tissue from kiwifruit. kiwifruits were each cut longitudinally into six segments, and each sample comprised one segment from each fruit, pooled and homogenized in a Waring blendor. The puree was further homogenized for 2 min using an Ultra Turrex at the maximum speed to produce a uniform sample. Soluble oxalic acid. A known weight (about 50g) of the sample was mixed with about 40 mL of water and incubated in a 100°C water bath for 15 min. After cooling to room temperature, the volume was made up to 100 mL and the suspension centrifuged at 10000 rpm for 15 min at 10°C in a J-13 rotor on a Beckman Model J2-21 centrifuge. Thirty mL of the supernatant was made up to 50 mL with 0.5% KH2P04 buffer at pH 2.00. This was filtered through a 0.45 p,rn Millipore filter and through a Sep-Pak Cl8 cartridge (Waters Associates, Milford, MA). An aliquot of 20 ~.LLwas injected into the column for HPLC. Total oxalic acid. A known weight (about 50g) of the sample was mixed with 10 mL of SN HCI and 30 mL water. The mixture was incubated in a 100°C water bath 30 min. After cooling to room temperature, the volume was made up to 100 mL. It was centrifuged as before, and 15 mL of the supernatantwas adjusted to pH 2.00 with 5N KOH and made up to 50 mL with 0.5% KH,P04 buffer at pH 2.00. The solution was filtered through a 0.45 )*rn millipore filter and a Sep-Pak Cl8 cartridge as before. An aliquot of 20 p,L was injected into the HPLC. The total oxalic acid of the isolated raphide crystal suspensionwas determined in the same manner. A known weight (about 2g) of the suspensionwas incubated with 1 mL 5N HCl and about 10 mL water in a 100°C water bath. After cooling to room temperature, the volume was made up to 2.5 mL. The suspension was centrifuged as before and 20 mL of the supernatantwas adjusted to pH 2.00 and made up to 50 mL with 0.5% KHZPOl buffer at pH 2.00. The solution was filtered as before and an aliquot of 20 ~.LLwas injected into the HPLC. HPLC equipment and conditions A Waters HPLC system equipped with a Wisp 710B automatic sample injector, a model 6000A solvent delivery system, a column heating device, a model 730 data module, a model 720 system controller and a Lambda-Max model 480 UV-Visible variable wavelength detector was used. A 250 x 4.6 mM i.d., 5 PM, end-capped Spherisorb ODS 11 column (Alltech Associates Inc., Deerfield, IL) was used. A Guard Pak precolumn module with a Cl8 cartridge (Waters Associates, Milford, MA) was connected in front of the analytical column. The mobile phase was 0.5% (w/v) KH,PO, buffered to pH 2.00 with ortho-phosphoric acid. It was millipore-filtered and degassed. The flow rate was 0.6 mL/min at 750 psi and column tem- Nine panelists, screened for their ability to perceive “catch” in kiwifruit products, were chosen for this study. A training sessionwas held to describe the sensation of “catch” in kiwifruit nectars. This was followed by a panelist discussion to reach a consensus opinion on the definition of “catch.” A paired comparison test was used to evaluate samples of apple puree, with and without addedoxalate crystals isolated from kiwifruit. Oxalate crystals were added at a level of 30 mg oxalic acid per 100 mL apple puree in the test sample. Thirty five mL each of the control (apple puree) and the test sample (apple puree + added oxalate ctystals) were presentedin random order, coded with three digit numbers. The panelists were asked to taste the sample on the left first, to stir the puree before tasting and to allow at least 30 min between evaluation of the first and second sample. They were asked to circle the code number of the sample which produced the sensation of “catch” as defined in the training session. Scoring of enzyme treated nectars Forty-nine untrained volunteers from the staff at Mount Albert Research Centre participated in this panel. Each panelist was given one 50 mL sample each of the kiwifruit nectars preparedwith and without enzyme treatment, as described. The samples were presented in random order. The panelists were asked to evaluate the strength of any irritation perceived using a nine-point scale (1 = no irritation; 9 = extreme irritation). As these panelists were not trained, the term “irritation” was used to avoid any misunderstandingof the term “catch.” In order to minimize the carry-over effect of the irritation, the panelists were asked to allow at least 30 min between tasting the two samples. The results were subjected to analysis of variance. Light microscopy Samples of fresh kiwifruit tissue were examined from fruit at harvest maturity (firmness about 5 to 6 kG using a Universal Penetrometer and Brix value of over 6.50), or eating ripeness (firmness about 0.5 to 0.9 kG). Hand sections of unfixed whole fruit were observed without staining. Mechanically disrupted locules and enzyme treated idioblast cells were observed unstained using bright field and differential interference contrast microscopy. Physically extracted crystals were placed in solutions of 80% acetic acid or 1N HCI and the resulting changeswere observed. Scanning electron microscopy Examination was carried out on fractures of freeze-dried fresh material and critical-point dried glutaraldehyde-fixed material both from the inner pericarp of Hayward variety and of freeze-dried pulp. Fresh material was rapidly frozen with liquid nitrogen, fractured with a cooled blade and freeze-dried. Glutaraldehyde-fixed material was dehydrated in an ethanol series and critical-point-dried using carbon dioxide as the transition fluid in a Samdri-780 critical-point drier. Samples were fractured after drying. All material was sputter-coated with gold before observation in a Philips PSEM505 scanning electron microscope. RESULTS & DISCUSSION PRELIMINARYmicroscopicexamination of thenectarsshowed the presenceof idioblast cells (cucumber-shaped cells having transparent cell walls) containingraphidecrystals(Fig. 2). These Volume 55, No. 4, 19904OURNAL OF FOOD SCIENCE-7067 CALCIUM OXALATE IRRITANT IN K/W/FRUIT... Fig. 2-Light of systemls. micrograph of an idioblast Bar = 0.7 mm. cell containing a bundle Fig. 5-High pressure liquid chromatogram crystals after solubilizing in HCI. of isolated raphide Table I- Oxalate in six cultivars of kiwifruit’ Name of cultivar Fig. 3-Light rioe kiwifruit. micrograph of a thin transverse Bar = 1.0 mm cross section of a Hayward Abbott Allison DCWney Bruno Jones Soluble oxalateb 31.9 18.5 19.0 15.8 18.8 41.4 Total Insoluble 57.6 37.0 39.5 44.3 65.2 55.1 25.7 18.5 20.5 28.5 36.4 13.7 oxalateb oxalateC a Expressed as mG oxalic acid/lOOf of fruit. b Average of two replicates. c Obtained by subtracting soluble oxalate from total oxalate. Fig. 4-Light micrograph tals. Bar = 0.1 mm. of the residue of isolated raphide crys- cellsweresimilar in appearance to thosereportedin the family Araceaewith slight structuraldifferences(Sakaiet al., 1972; Sakai and Hanson, 1974; Sakai et al., 1984). Each idioblast cell containedlong needle-shaped raphidecrystals, arranged longitudinally and packed into a tight bundle. Each crystal bundlewas embeddedin mucilagesurroundedby an elongated 1068-JOURNAL OF FOOD SCIENCE-Volume 55, No. 4, 1990 ellipsoidal transparentcell wall. The cell walls were removed by incubatingwith pectinase.However, the raphidecrystals were not releasedas they remainedembeddedin the mucilage which was not degradedby the enzyme.When a shearforce was applied to a suspensionof idioblastsin which the cell walls had been removedor weakenedby pectinaseactivity, viscosity increased.The mucilagewithin the idioblastwas insolublein water but was partially solublein 1% SDS. Earlier attemptsto isolatetheraphidecrystalsby the method of Tang andSakai(1983)were unsuccessful.Two immiscible organic solventsof different densitieswere used to separate raphidecrystalsfrom cell wall debris in taro. However, no separationwaspossiblein kiwifruit asthecrystalswereembedded in the mucilagewhich tendedto emulsify the solvents. This mucilageresisteddegradationby enzymessuch as pectinase,cellulaseand hemicellulase. Microscopicexaminationof thin transversesectionsof fresh fruit indicatedthe majority of the raphide-containingidioblast cells were locatedin the locular region of the fruit, adjacent to the seeds(Fig. 3). Microscopicexaminationof the residue after enzymatictreatmentof the idioblast cells showedlarge numbersof raphidecrystals(Fig. 4). HPLC analysisof this residuefor organicacidsshowedonly onepeakcorresponding Fig. 6-SEM of raphide bundle embedded in the mucilage within the idioblast cells, obtained from freeze-dried kiwifruit. Bar = 0.1 mm. Table 2-Sensory scores of enzyme-treated -t 1.2 1.9 t 0.9 0.2 Enzyme-treated Control SEDa a Standard nectars Mean score % SD Samnla error of difference 5.1 between the two mean scores. to oxalic acid (Fig. 5). The isolatedraphidecrystalswere insoluble in 80% acetic acid but soluble in 1N HCl without evolutionof carbondioxide indicatingthe crystalswere not calciumcarbonate. Analysesof severalcultivars of kiwifruit for soluble,total and insolubleoxalatesare reportedin Table 1. Ail oxalate analysesareexpressed in termsof anhydrousoxalic acid. The differentkiwifruit cultivars differed.considerably in their oxalatecontents,but the valuesobservedfor Haywardcultivar werein closeagreement with thosepublishedearlier(Turner, 1980). Simulation of “catch” The isolatedraphidecrystals,when disperseduniformly in the applepureeat aboutthe samelevel of insolubleoxalate foundin freshkiwifruit (30mg oxalic acidper lOOg),imparted a “catch” to it. The tastingpanel,all perceivedthe “catch” at that level of additionof the raphidecrystals(p<O.O02). The“catch” wasnot eliminatedby boilingtheisolatedcrystals in water or in 80% alcohol. This is in sharpcontrastto theresultsof Moy et al. (1979)who foundwith tarothemouth irritationwas removedby boiling in wateror aqueousalcohol. Plantsbelongingto thegenusDieffenbrachiaarereportedto be poisonousandcauseitching, swellingandsalivationwhen juice of the plant comesin contactwith the skin or mucous membrane(Walter andKhanna,1972).In a studyof this genus, Walter andKhanna(1972)reporteda proteolyticenzyme which may act on the sensitivetissueinjuredby the raphides to causetheirritation. Kiwifruit alsohasa powerfulproteolytic enzymeknown as actinidin(Ferguson,1984).However,it is unlikely that the rigorousextractionprocedurewe usedwould leaveany of the enzymein the final crystal isolate.Analysis of the crystal isolatefor proteinwas negative.This indicated theirritationsimulatedin theapplepureewasdueto theadded raphidecrystals. Scoring of enzyme treated nectars Our studywas conductedto providefurtherevidenceof the natureof the factor responsiblefor the irritation. It was also usedto estimatethe proportionof the populationwho were ableto perceivethe “catch.” Isolationof idioblastsin large enoughquantitiesfor testsusingnearly50 panelistswould be laboriousand time consuming.Therefore,we usedenzymetreatedandmechanicallyshearednectarfor the largerstudies dueto its easeof preparationin largequantities. Kiwifruit nectarspreparedby enzymedigestionof the idioblast cell wall membranesfollowed by high shearforce were comparedwith a control nectarpreparedwithout this treatment. Seventy-fourpercentof the panelistsperceivedan increasedirritationin theenzyme-treated nectar(p<O.OOl).Mean scorefor panelistswas5.1 for strengthof irritationfrom treated nectars.This was significantlyhigherthanthe meanscoreof 1.9 for the control nectar (p<O.OOl).The mean score and standarderror of the differencebetweenthe two meanscores of the enzyme-treated nectarand control are shownin Table 2. This differencein intensityof theirritationbetweenthetwo preparations canbe explainedon thebasisof theultrastructure of the idioblast cells and the raphidecrystals found within them. In the control nectar,relatively few crystalswere observedoutsidethe idioblastcells. Thosethat werewithin the idioblastcell structureswereembedded in an envelopeof mucilage.However,enzymedigestionbrokeup the idioblastcell wall. Whena highshearforcewas applied,theraphidecrystals were dispersedthroughoutthe entiremediumin which they were contained.They were, therefore,more proneto cause mechanicalirritation when ingestedand would have greater “catch” comparedto the control. The raphidecrystalsare embeddedin a mucilagewhich is resistantto normal pectolytic enzymes.This explainswhy “catch” is hardlynoticeablein the freshfruit. However,during drying, themucilageshrinksandsomeof thesharpneedleshapedcrystalsprotrudefrom the dried mucilagematrix as shownin Fig. 6. Suchdriedmucilagematrices,holdingsharp protrudingcrystalbundlestogether,will probablycausegreater mechanicalaction on the mucousmembranesthan the free crystals. This may explainwhy more peopleexperiencethe “catch” in productssuchas-driedslicesandfruit leathersthan in fresh fruit or pulp (Perera,1985). Fruit leathersmadeof 100%kiwifruit have considerable“catch.” This can, however,bereducedto a nondetectable level by blendingtwo parts of kiwifruit pulp with threepartsof anothersuitablefruit pulp beforedrying (Perera,1985). The moisturecontentof dried kiwifruit was also notedas an importantfactor affectingthe perceptionof “catch,” whichwasperceivedto a greaterextent at low (e.g. 4%) thanat high(e.g. 15%)moisturelevels.These observations weresupportedby theelectronmicroscopicstudy of the ultra structureof the idioblast cells after drying (Fig. 6). Thesestudiessuggestthe “catch” or irritation observedin certainprocessedkiwifruit productswhen ingested,was due to a simple mechanicalactionof the raphidecrystalson the mucousmembranes in the mouth. REFERENCES Al-R& A.H., Myers, A., and Watson, L. 1971. The isolation and properties of oxalate crystals from plants. Ann. Bot. 35: 1213. Big-slow, J. 1818. Acridity in Arum. J. Am. Mod. Bot. 1: 55. Black, O.F. 1918. CaIcium oxalate in the dasheen. Am. J. Bot. 5: 447. Clark, C.J., Smith, G.S., Fd WaIkcr, G,D. 1987. The form, distribution ;;II sqyal accumulation of calcmm m klwfmt leaves. New Phytol. Fasse& D:W. 1973. OxaIates. Ch.16. In Tozicants Naturally Occurring in Foods, 2nd ed., p. 346, Nat. Acad. Sci., Washington, DC. Ferguson, A.R. 1984. Kiwifruitz A botanical review. Hart. Rev. 6: 1. Franceschi, V.R. and Homer Jr., H.T. 1980. Calcium oxelate crystals in plants. The Bot. Rev. 46: 361. Kohl, F.G. 1889. Anatomisch-physiolo ‘sch untersuchungen der kalksalse und der Kieselsaure in der pflanse. rf arburgp. 91. Quoted in Black, O.F. 1918, J. Bat. 5: 447. Libert, B. 1981. Rapid determination of oxalic acid by reverse phase high erformance liquid chromate aphy. J. Chromatogr. 210: 540. Ll%ert, B. and Frances&i, V. lY 1987. Oxalate in crop plants. J. Agric. Food Chem. 35: 926. Miller, CD. 1928. Note on the &act of ingesting large amounts of pin& apple juice upon the pH of the urine. J. Home Econ. 20: 498. -Continued on page 1080 Volume 55, No. 4, 1990-JOURNAL OF FOOD SCIENCE-1069 THERMAL DESTRUCTION CYSTEINEICYSTINE RESIDUES... Table 1 -Effect of heating on the amino acid composition (Values are in g/l OOg protein) of soy protein” Amino acid Unheated sample Heated sample 12.15 3.20 5.03 21.03 5.66 3.83 3.75 0.55 4.13 0.90 4.35 7.67 3.66 5.32 2.56 6.18 7.99 0.03 12.02 ASP THR SER GLU PRO GLY ALA CYS VAL MET ILE LEU TYR PHE HIS LYS ARG LAL a Conditions: 6% soy protein isolate solution 3.13 5.11 21.28 5.65 3.81 3.57 0.45 4.12 0.64 4.33 7.61 3.58 5.31 2.50 6.06 7.97 0.04 (pH 8.0) was heated at 100°C for 60 min. ageto form sulfhydrylgroupswhich theqaredestroyed,probably via p-eliminationreaction. To determinewhetherthe thermaldestructionof cysteine and cystineresultedin formationof LAL an 8% soy protein solutionwas heatedat 100°Cfor 60 min andthe aminoacid compositionanalyzed.The aminoacidprofile of unheatedand heatedsoy protein are shownin Fig. 6 and the amino acid compositionsare shownin Table 1. The half-cystinecontent of unheatedandheatedsoy proteinwas 0.55 and0.45 g/lOOg protein, respectively(Table l), indicatingan 18.2% loss of half-cystine.This is in agreement with the 15%lossshownby the NTSB analysis(Fig. 4, 60 min datumpoint). Standard LAL studiesshowedit elutedjust beforethe histidinepeakin theaminoacid elutionprofile. A slight peakcorresponding to the LAL positionwas foundboth in the unheatedand heated samples(Fig. 6, Table 1). The differencein the LAL.content betweentheunheatedcontrolandtheheatedsample,however, was insignificantand did not accountfor the lossesof halfcystine and lysine (Table 1). The lossesof half-cystineand lysin were about0.1 g/lOOg(i.e., 8.26 x lo-“ moles)and 0.12 g/lOOg(i.e., 8.21 X 1O-4 moles) of protein, respectively. However,the increaseof LAL contentwas about0.01 g/lOOg(i.e., 4.29 x 1O-5moles)protein.Onemole of halfcystineandonemoleof lysinewould theoreticallybe expected to yield onemole of lysinoalanine.Hence,the discrepancyin amountof lysinalanineindicatesthethermaldestructionof cystine and cysteinedoesnot involve LAL formation,but may involve other cross-linkingreactionssuchas lanthionineformation(WhitakerandFeeney,1983).However,no newpeak was observedin the amino acid profile (Fig. 6). Either lan- thionineor someotherderivatives,if presentmight haveeluted alongwith the other aminoacids.Friedmanet al, (1984)reportedwhen 1% soy proteinsolution(pH 8.0) was heatedat 75°Cfor 3 hr, therewas a 52% loss of half-cystineand formationof LAL. The only differencebetweentheir studyand our studyis theproteinconcentration andthetemperature. The majorconclusionthatcanbe drawnfromcomparingour results to thoseof Friedmanet al. (1984)is that higherproteinconcentration(i.e., 8%) not only decreases the % destructionof SH + S-S groups,but also influencesthe pathwayof eliminationof cysteineandcystineresidues.Thus, it appearsthat while heatinga 1% soy protein solution (pH 8.0) at 75°C resultsin the formationof LAL, heatingan 8% soy protein solution(pH 8.0) at 100°Cdoesnot resultin LAL formation, but may involveformationof othercrosslinkedderivatives. In summary,our study clearly indicatesthat cysteineand cystineresiduesin soy proteinsundergothermaldestruction undertheconditionsof gelation(8%proteinconcentration, pH 8.0, 1OPC).However,destructionof theseresiduesdoesnot involveformationof LAL. Factorssuchasshortheatingtime, low pH, highersalt concentrationand higherviscosity affect the rateandextentof thermaldestructionof cysteineandcystine. Useof theseresultscan helppreservenutritionalquality in thermalprocessingof soy proteins. REFERENCES Babajimopoulos, M., Damodaran, S., Rizvi, S&H., and Kinsella, J.E. 1983. Effects of various anions on the rheological and gelling behavior of soy proteins: Thermodynamic observations. 3. Agric. Food Chem. 31: 1270. CRC Handbook of Chemistry and Physics. 1985. p. D-232, The Chemical Rubber Co., Cleveland. Damodaran, S. 1985. Estimation of disulfide bonds using 2-nitro-5-thiosulfobenzoic acid: Limitations. Anal. Biochem. 145: 200. Friedman, M., L&n, C.E., and Noma, A.T. 1984. Factors governing lysinoalanine formation in soy proteins. J. Food Sci. 49: 1282. Gould, D.H. and MacGregor, J.T. 1977. Biological effects of alkali-treated protein lysinoalanine: an overview. In Protein Crosslinking: Nutritional and Medical Consequences,M. Friedman (Ed.), p. 29. Plenum Press, New York. Koshiyama, I. 1972. Purification and some properties of P-conglycinin in soybean seeds. Int. J. Peptide Protein Res. 4: 167. Than?, v. H. Shibasaki, K. 1976. Major proteins of soybean seeds. A strmghtforward fractionation and their characterization. J. Agric. Food Chem. 24: 1117. Thanh, v.H. and Shibasakj, K. 1978. Major proteins of soybean seeds. Reconstitution of P-conglyanin from its subunits. J. Agric. Food Chem. 26: Gee ---. Thannhauser, T.W., Konishi, Y., and Scheraga, H.A. 1984. Sensitive quantitative analysis of disulfide bonds in polypeptides and proteins. Anal. Biochem. l.?A. - -. -1-Al-. Volkin, D.B. and Klibanov, M. 1987. Thermal destruction processes in nroteins involving cvstine. J. Biol. Chem. 262: 2945. Whitaker J.R. andFeeney, R.E. 1983. Chemical and ph sical modification of moteins bv the hvdroxide ion. CRC Crit. Rev. Food 8.~1.Nutr. 19: 1773. Woddard, C.J.: Short; D.D, Alvarez, M.R., and Reyniers, J. 1975. Biolog ical effects of lysinoalanme. In Protein Nutrition Quality of Foods and Feeds,,Part 2, M. Friedman (Ed.), p. 595. Marcel Dekker, New York. MS received 9129189;accepted 12/18/89. CALCIUM OXALATE IRRITANT IN KIWIFRUIT. . .From page-1069 May, J.H,, Shadbolt, B., Stoewsand, G.S., and Nakayama, T.O.M. 1979. The acridity factor in tam processmg. 3. Food Process. Preserv. 3: 139. Oke, O.L. 1969. Oxalic acid in plants and in nutrition. World Rev. Nutr. Diet. 10: 263. Okuse, I. and Ryugo,, K. 1981. Compositional changes in the developing “hayward” kiwifnut in California. J. Am. Sot. Hart. Sci. 106: ‘73. Peters, R.E. and Lee, T.H. 1977. Composition and physiology of Monstera deliciosa fruit and juice. 3. Food Sci. 42: 1132. Perera, C.O. 1985. Unpublished data. Department of Scientific and Industrial Research, Mount Albert Research Centre, Auckland. New Zealand. Sakai, W.S. and Hanson, M. 1974. Mature raphide and raphide idioblast structure in plants of the edible aroid genera Colocasia, Alocasia and Xanthosoma. Ann. Bot. 38: 739. Sakai, W.S., Hanson, M., and Jones, R.C. 1972. Ra hides with barbs and ooves in Xanthosoma sagittifolium (Araceae). 8.clence 178: 314. S&i, W.S. Sh’Iroma, S.S., and Nago, M.A. 1984. A study of raphide microstruct&e in relation to irritation. Scanning Electron Microscopy 11:979. Schmid, R. 1978. Reproductive anatomy of Actinidia chinensis (actinidiaceae). Botanische Jahrbucher fur systematik. pflanzengeschichte und Pflanzengeographie 100: 149. 1080-JOURNAL OF FOOD SCIENCE-Volume 55, No. 4, 1990 Strauss, D. 1970. Uber die Mikroscopie fremder fruchte 111. Mitteilung. Deutsche Lebensmittel-Rundschau. 66: 260. Tang, C. and Sakai, W.S. 1983. Acridity of tare and related plants. Ch. 6. In Two. A Review of Colocasia esculanta and Its Potentials. Jaw-Kai Wang (Ed.), p. 148. Univ. Hawaii Pres., Honolulu. HI. Turner, N.A. 1980. Micro-determination of oxalate in crude extracts of plant tissues. J. Sci. Food Agric. 31: 171. Walter,, W.G. and Khanna, P.N. 1972. Chemistry of the aroid 1. Dieffenbachla sequine, amoena and picta. Econ. Bot. 26: 364. Ziegenspeck, H. 1915. Ber. Deutsch. Bot. Ges. 32:630. Quoted Black by O.F. 1918. Am J. Bot. 5: 447. MS received 7/23/89; revised l/19/90; accepted l/27/90. The authors acknowledge Kay McMath for her useful suggestions an sensory evaluation techniques.This researchwaspartially fundedby the New ZealandKiwifruit Marketing Board.