Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of foliar glucosinolates in host plant resistance of oilseed rape and mustard to the Bertha… McCloskey, Catherine A. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_fall_phd_mccloskey_catherine.pdf [ 4.73MB ]
JSON: 831-1.0086405.json
JSON-LD: 831-1.0086405-ld.json
RDF/XML (Pretty): 831-1.0086405-rdf.xml
RDF/JSON: 831-1.0086405-rdf.json
Turtle: 831-1.0086405-turtle.txt
N-Triples: 831-1.0086405-rdf-ntriples.txt
Original Record: 831-1.0086405-source.json
Full Text

Full Text

OF BRITISH COLUMBIATHE ROLE OF FOLIAR GLUCOSINOLATES INHOST PLANT RESISTANCE OF OILSEED RAPE AND MUSTARD TOTHE BERTHA ARMYWORM AND THE DIAMONDBACK MOTHbyCATHERINE ANN MCCLOSKEYB.Sc.(Agr.), The University of Guelph, 1983M.Sc., The University of Ottawa, 1986A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESDepartment of Plant ScienceWe accept this thesis as conformingSeptember 1993© Catherine Ann McCloskey, 1993 THEIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of ^Pict h 1 SelehefThe University of British ColumbiaVancouver, CanadaDate  Ott . 4i, /???.DE-6 (2/88)iiAbstract - The relationships between host plant glucosinolatesand feeding and growth of a polyphagous insect, the Berthaarmyworm, Mamestra configurata Walker and field infestationlevels of an oligophagous insect, the diamondback moth,Plutella xylostella L. were investigated. M. configuratareared on eight rape and mustard varieties had significantlydifferent final weights by species order Brassica juncea <Sinapis alba < B. napus < B. rapa. In choice tests, B. junceawas least preferred while S. alba was significantly moreattractive than rape to neonate larvae. Relative consumptionand growth rates of fourth instar larvae were also reduced onB. juncea foliage. Neonate choice and fourth instar growthrates were negatively correlated to concentrations ofisothiocyanate-releasing glucosinolates. In feeding testswith pure compounds in meridic diets, sinigrin and itsmetabolite, allyl isothiocyanate inhibited growth in a dose-dependent manner. No effects were observed with indole-3-carbinol, the metabolite of 3-indolylmethyl glucosinolate.Foliar isothiocyanate-releasing glucosinolates may providesome protection against polyphagous insects. Plant growthstage differences in the relative resistance of brassicaceoushost plants were investigated by feeding B. juncea (relativelyresistant) or B. rapa cv. Tobin (Canola) (susceptible) foliageof three plant growth stages (2-4) to M. configurata.Relative consumption rates did not differ significantlybetween tne plant species. However, with B. rapa, relativeiiiconsumption rate increased with advancing plant growth stage.Relative growth rates on B. juncea foliage were significantlylower than on B. rapa. With B. juncea, relative growth ratedecreased with plant growth stage. Concentrations ofisothiocyanate- and thiocyanate-releasing glucosinolatescorresponded to the trends observed in the nutritionalindices, while levels of total phenols and catechols appearedunimportant. Analyses of total nitrogen in field-collectedplants showed a serial reduction with advancing growth stage.The influence of glucosinolate profile on infestation by P.xylostella was investigated using field plots of six rape andmustard varieties. S. alba plots supported larger populationsthan the other plant species in the study. Parasitism ratesby Diadegma insularis (Hymenoptera: Ichneumonidae) were notsignificantly different between plant species on foliarsamples. On flower head samples, larvae on B. juncea wereparasitised the most, whereas larvae on S. alba wereparasitised the least. The glucosinolate profile of the hostplant may be more important to D. insularis than to P.xylostella.ivTable of ContentsAbstract^ iiTable of Contents^ ivList of Tables viList of Figures^ viiiAcknowledgementsPreface^ xi1.0 Introduction 11.1 Site and Metabolism of Glucosinolates 21.2 Effects of Glucosinolates on Insects 81.3 Glucosinolates and Pest Management 181.4 Objectives 221.5 Biology of Brassica Insects Studied 231.51 The Bertha Armyworm, Mamestra configurata Walker 231.52 The Diamondback Moth,Plutella xylostella Linneaus242.0 Influence of Foliar Glucosinolates in Oilseed 26Rape and Mustard on Feeding and Growth of theBertha Armyworm, Mamestra configurata Walker2.1 Introduction 262.2 Materials and Methods 282.3 Results 372.4 Discussion 593.0 Plant Growth Stage Effects on the Feeding and 62Growth of the Bertha armyworm, Mamestraconfigurata in Canola and Mustard Foliage3.1 Introduction 623.2 Materials and Methods 633.33.44.0vResults^ 66Discussion 81Diamondback Moth, Plutella xylostella L. Larval 84Populations and Parasitism by Diadegma insularis(Cress.) on Host Plants with DifferentGlucosinolate Profiles^4.1^Introduction^ 844.2^Materials and Methods^ 864.3^Results^ 894.4^Discussion 1005.0^Summary and Conclusions^ 102Bibliography^ 106Appendix 1 Response Factors for Desulfoglucosinolates 118Relative to 0-nitrophenyl-p-D-galactopyranosideAppendix 2Appendix 3Appendix 4HPLC-MS Identification of Desulfo-^119glucosinolatesQuantities (umol/g dry weight) of Total^122Glucosinolates, Determined Spectro-photometrically, and Isothiocyanate- andThiocyanate-Releasing Glucosinolates,Determined by HPLC, in Field-Collected FoliageCultivar Means for Neonate Choice, Relative 124Consumption Rate and Relative Growth Rateof Section 2.viList of TablesTable 1^Biological Activity of Glucosinolates^9and Their Fission Products in InsectsTable 2^Effect of Host Species and Cultivar on^43Final Weight of M. configurata LarvaeReared for Seven Days on Intact Plants asDetermined by ANOVA and Orthogonal ContrastsTable 3^Effect of Host Species and Cultivar on^45Neonate M. configurata Leaf Disc ChoiceTests as Determined by ANOVA and OrthogonalContrastsTable 4^Linear Correlation Coefficients for^47Relationship of Foliar Glucosinolates (umol/gdry weight) to Neonate Choice (weighted means ofsquare root-transformed counts) in Leaf Disc AssayTable 5^Effect of Host Species and Cultivar on^50Relative Consumption Rate (RCRi) of FourthInstar M. configurata as Determined by ANOVAand Orthogonal ContrastsTable 6^Effect of Host Species and Cultivar on^52Relative Growth Rate (RGRi) of Fourth InstarM. configurata as Determined by ANOVA andOrthogonal ContrastsTable 7^Linear Correlation Coefficients for^53Relationship of Foliar Glucosinolates (umol/gdry weight) to Relative Consumption Rate(mean RCRi) of Fourth Instar M. configurataTable 8^Linear Correlation Coefficients for^55Relationship of Foliar Glucosinolates (umol/gdry weight) to Relative Growth Rate (mean RGRi)of Fourth Instar M. configurataTable 9^Analysis of Variance for Relative Consumption 68Rate (RCRi) of M. configurata on Foliage ofB. juncea and B. rapaTable 10^Mean Relative Consumption (RCRi) and Growth^69(RGRi) Rates of Fourth Instar M. configurataon Foliage of B. juncea or B. rapaTable 11^Analysis of Variance for Relative Growth Rate 72(Ruxi) or M. configurata on Foliage ofB. juncea and B. rapaviiTable 12^Analysis of Variance for P. xylostella^91Counts per 100 g Fresh Weight of FoliageTable 13^P. xylostella Population Densities (Larvae^93per 100 g plant tissue) on Foliage andFlower Heads of the Different CultivarsTable 14^Analysis of Variance for P. xylostella^94Counts per 100 g Fresh Weight of Flower HeadsviiiList of FiguresFigure 1 Metabolism of Glucosinolates:^ 4(a) Unsubstituted Alkenyl and AromaticSide Chains.Figure 1(b) 8-hydroxylated Alkenyl Side Chains^6Figure 1(c) Indolyl and p-OH-benzyl Side Chains 7Figure 2 Representative HPLC Chromatograms of^38Mustard Species Used in the StudyFigure 3 Representative HPLC Chromatograms of^39Rape Species Used in the StudyFigure 4Figure 5Figure 6Figure 7Figure 8Mean Final Weights of M. configurata Larvae^42Reared for Seven Days on Intact Rosette-StagePlantsNumbers of Neonate M. configurata on Leaf^44Disks in Choice Tests with Field-GrownFoliageMean Consumption Indices for Fourth Instar^48M. configurata Larvae on Field-Grown FoliageMean Relative Growth Rates of Fourth Instar^49M. configurata Larvae on Field-Grown FoliageLinear Regression of Final Larval Weight^56Versus Dietary Concentration of Pure CompoundsAugmented to Meridic DietFigure 9 Percent Mortality Versus Dietary^57Concentration of Pure Compounds Augmented toMeridic DietFigure 10Figure 11Percent Mortality Attributed to Fumigation^58Effect of Allyl Isothiocyanate in AssayUsing Meridic DietMean Relative Consumption Rate of Fourth^67Instar M. configurata Larvae on Foliage ofRosette (2), Stem Elongation (3) and Flowering(4) StagesMean Relative Growth Rate of Fourth Instar^71M. configurata Larvae on Foliage of Rosette (2), Stem Elongation (3) and Flowering (4)Stages.Figure 12Figure 13Figure 14Figure 15Figure 16Figure 17Figure 18Figure 19Figure 20Figure 21Figure 22Figure 23ixRepresentative HPLC Chromatograms of Desulfo- 73glucosinolates for Greenhouse-Grown B. juncea(Commercial Brown Mustard) and B. rapacv. Tobin Canola.Quantities of Isothiocyanate-Releasing Gluco- 74sinolates (Itc Gsl) in Foliage of Rosette (2),Stem Elongation (3) and Flowering (4) StagesDetermined by HPLC.Quantities of Indolyl Glucosinolates (Gsl)^75in Foliage of Rosette (2), Stem Elongation(3) and Flowering (4) Stages Determined by HPLC.Quantities of Phenolic Compounds in Foliage^77of Rosette (2), Stem Elongation (3) andFlowering (4) Stages Determined Spectrophoto-metrically.Quantities of Catecholic Compounds in Foliage 78of Rosette (2), Stem Elongation (3) andFlowering (4) Stages Determined Spectrophoto-metrically.Quantities of Total Nitrogen in Field-^80Collected Foliage of Brassicaceous Plants.P. xylostella Population Densities on Foliage 90and Flower Heads over the Course of theSampling Study.Scatter Plot of Percent Parasitism by^96D. insularis Versus Number of P. xylostellaLarvae per Sample, with Linear Regression Line.Scatter Plot of Percent Parasitism by^97D. insularis Versus Number of P. xylostellaLarvae per 100 grams fresh weight plant tissue,with Linear Regression Line.Percent Parasitism of P. xylostella Larvae by 98D. insularis on Foliage.Percent Parasitism of P. xylostella Larvae by 99D. insularis on Flower Heads.xAcknowledgementsI would like to thank my supervisor, Dr. Murray Isman,for his input on matters of science, financial support duringpart of this degree, and for his patience and understandingwith regard to the time-consuming properties of youngchildren. My supervisory committee - Dr. J.H. Myers (Depts.of Plant Science and Zoology), Dr. G.G.E. Scudder (Dept. ofZoology) and Dr. R.S. Vernon (Agriculture Canada, Vancouver) -asked useful questions and provided advice and researchfacilities when necessary.I gratefully acknowledge the technical assistance andfriendship of Nancy Brard who maintained the culture of M.configurata, assisted with fieldwork and offered many helpfulsuggestions. Ideas were often discussed with Drs. Tom Lowery,Mike Smirle and Opender Koul. Their comradeship was mostbeneficial to the research. The field assistance and goodhumour of Kathy Craig was also most appreciated.A nucleus for the original M. configurata colony wasprovided by Dr. W.J. Turnock, Agriculture Canada (Winnipeg).The rape and mustard seed were gifts from Dr. R.J. Lamb andDr. L. Burgess, Agriculture Canada (Winnipeg and Saskatoon,respectively), the Alberta Wheat Pool, and Northern Sales(Winnipeg). The granular fertilizers wre a gift from AgricoSales, North Delta, B.C. The TreflanLV E.C. was a gift fromElanco Division, Eli Lilly Canada Inc., Scarborough, Ontario.Advice, some reference brassicaceous desulfoglucosinolateextracts and some pure glucosinolate standards were providedby M. Giblin and Dr. E.W. Underhill of the Plant BiotechnologyInstitute, National Research Council (Saskatoon). Thenitrogen analysis was performed by L. Toerper, Department ofSoil Science, U.B.C. Identification of D. insularis wasconfirmed by Dr. Risa Smith, Vedalia Biological Research,Galiano, B.C. This research was funded by the H.R. MacMillanFamily Fellowship to the author and by an NSERC operatinggrant to Dr. Isman.Finally, and above all, I thank my family: my husbandBill for his moral support and encouragement throughout thedegree, and my mother for encouraging me to leave her andpursue the Ph.D. shortly after the death of her husband.xiPrefacePortions of this thesis have been published or aresubmitted for publication at the time of submission to theFaculty of Graduate Studies. The content of Section 2 hasbeen published as "Influence of foliar glucosinolates inoilseed rape and mustard on feeding and growth of the Berthaarmyworm, Mamestra configurata Walker" by Catherine McCloskeyand Murray B. Isman (J. Chem. Ecol. 19:249-266 [1993]). Thecontent of Section 3 has been submitted as "Plant growth stageeffects on the feeding and growth of the Bertha armyworm,Mamestra configurata to canola and mustard foliage" byCatherine McCloskey and Murray B. Isman, to the journal,Entomologia experimentalis et applicata. Dr. Isman supervisedthe research and edited the papers prior to submission forpublication.Murra B. IsmanAssociate Professsor11.0 IntroductionBrassicaceous crops, including oilseed rape (Brassicanapus L. and B. rapa L. and mustards (B. juncea (L.) andSinapis alba L.) are widely cultivated in western Canada.Much of the acreage on the prairies is taken up by canola,defined as oilseed rape cultivars having less than two percenterucic acid (a C-22 monoethenoid fatty acid) in the oil (Daun,1986a) and less than 30 umols per gram of glucosinolates inthe seed meal (Daun, 1986b). Glucosinolates are plantsecondary metabolites derived from amino acids which, alongwith erucic acid, are undesireable constituents of rapeseed.The existence of low-glucosinolate B. napus and B. rapacultivars is the result of plant breeding programs followingscreening for low-glucosinolate germplasm. A Polish B. napuscultivar, Bronowski, was the first low-glucosinolate varietydiscovered (Josefsson and Appelqvist, 1968). This lineeventually led to a Canadian canola cultivar, Tower, and aEuropean cultivar, Erglu (Fenwick and Curtis, 1980). Manyother lines have since been produced.Although canola is susceptible to disease and insectpests, only a modest effort has been directed toward thedevelopment of resistant varieties. Until very recently, mostof the plant breeding efforts were focussed on reducing oreliminating glucosinolates and erucic acid from the seed.Glucosinolates do not affect the quality of the nil - However,they reduce the nutritional quality of the high protein seed2meal left after oil extraction, which is used in livestockrations. Glucosinolates yield antinutritional and toxicmetabolites following hydrolysis by thioglucosideglucohydrolase (myrosinase; E.C., an enzyme which co-occurs with glucosinolates.1.1 Site and Metabolism of GlucosinolatesGlucosinolates and myrosinase co-occur in plants.Therefore, they are maintained apart from each other. Whenthe plant is mechanically damaged, as would occur duringherbivore feeding, the enzyme and substrate mix to producenoxious compounds which serve a defensive function.The physical description of the glucosinolate-myrosinasesystem has been controversial (Chew, 1988), possibly becausetwo different plant systems have been used to study thecompartmentalization of glucosinolates and myrosinase. Onemodel has been formulated using horseradish (Amoracialapathifolia) roots (Grob and Matile, 1979; Matile, 1980;Luthy and Matile, 1984). The "Glucosinolate Bomb" (Matile,1980) is a description of the horseradish root parenchymacells where glucosinolates and ascorbic acid occur in arelatively large vacuole (Grob and Matile, 1979) whilemyrosinase exists freely in the surrounding cytoplasm.Myrosinase is described as cytosolic enzyme which tends toadhere to membranes, such as the vacuolar tonoplast (Luthy andMatile, 1984).^The second model of the glucosinolate-3myrosinase system is derived from the existence of myrosincells in the seedlings of brassicaceous plants (Bones andIversen, 1985; Thangstad et al., 1990; 1991). Myrosin cellsare specialized protein-dense cells (idioblasts), previouslyof unknown function, which contain myrosin grains. Thesegrains are comprised of proteins (Bones and Iversen, 1985)which are assumed to be associated with glucosinolates(Thangstadt et al., 1990). The myrosinase is located in thetonoplast membrane surrounding the myrosin grains (Thangstadtet al., 1990; 1991). Most likely there is no model for theglucosinolate-myrosinase system which encompasses all planttissues.The enzymatic hydrolysis of glucosinolates has been thesubject of many studies, compiled in numerous reviews (seeKjaer, 1976; Benn, 1977; Fenwick and Heaney, 1983; Fenwick etal., 1983; Chew, 1988). The general structure ofglucosinolates (Figure la, I) includes a side chainoriginating from the precurser amino acid side chain, aglucose molecule, ,B-D-linked to a sulfur atom (donated from asulfur-containing amino acid), and a sulfate group covalentlybound to an oxime (the original amino group). Mostglucosinolates are potassium salts, although some have cholineesters of cinnamic or benzoic acids as counter ions (eg. thesinalbin-sinapine pair).The type of metabolite produced upon hydrolysis dependsupon the side chain and the ambient reaction conditions.4N-0S0,-R- C0^ (I)•S -0-D-GlucoseH2O V myrosinaseN-OSO,"^unstable[R. C0• (II)^agluconeR- CH2 = CHCHI- OM or CHI- (IV) pH > 7.5tfr^ 1 H+ or Fe++R-N=C=S (V) R-CEN (NTisothiocyanate^nitrile+ SO42- + SFigure 1. Metabolism of glucosinolates: (a) Unsubstitutedalkenyl and aromatic side chains.S=^+ D-Glucose5Myrosinase cleaves off the glucose moiety leaving anunstable aglucone (II), which spontaneously loses its sulfategroup and undergoes a Lossen rearrangement, forming anisothiocyanate (V). Unsubstituted alkenyl and arylglucosinolates, such as sinigrin (III) or benzyl glucosinolate(IV) normally yield stable isothiocyanates. Certainglucosinolates yield unstable isothiocyanates depending on thestructure of their side chains, and consequently undergofurther rearrangement. A-Hydroxylated alkenyl glucosinolates(Figure lb), such as progoitrin (VII), yield oxizolidine-2-thiones (VIII) at neutral to basic pH range. Glucosinolatespossessing indolyl or hydroxy-aryl side chains, such asglucobrassicin (XI) or sinalbin (XII) (Figure lc), aremetabolized to alcohols (XIII) with the release ofthiocyanate ion (SCN - ) at neutral to basic pH. These mayfurther condense to form diindolylmethane (XIV). Under acidicconditions or in the presence of ferrous ions, glucosinolatesare metabolized to nitriles (VI and X) (Figure la-c) and freesulfur (Uda et al., 1986a, 1986b). Certain plants possess anepithiospecifier protein (ESP) which acts as a co-factor ofmyrosinase to produce epithionitriles (IX) (Figure lb) fromp-hydroxylated glucosinolates. Ferrous ions are essential tothis process (MacLeod and Rossiter, 1985).The various glucosinolate metabolites have differenttypes of biological activity in mammalian systems (see reviewsby Tookey et al., 1980; VanEtten et al., 1980, and Fenwick andR CH2 = CHVCH 2-OH (VII)pH> 7.5.-^H+ or Fe++CH, = CH ^CH2 = CH-CH- CH,CN^OH^(X)O^NH^nitrile^(VIII) +S+s 042 `v Fe++*ESPvSCH24-1 CH-CH-CH,- C=NOH^(a)epithionitrileFigure 1. (b) ,B-hydroxylated alkenyl side chains. 7R=CH21( 'N Njjor(XI)pH > 7.5^H+ or Fe++[R_N=c=s] unstable R—C= Nisothiocyanate^nitrile (VI)+SR— OH +SCN-(XII)(XIV)diindolylmethaneFigure 1. (c) Indolyl and p-OH-benzyl side chains.8Heaney, 1983). Oxizolidine-2-thiones, isothiocyanates andthiocyanate ion are goitrogenic (Langer, 1966). Nitriles havebeen shown to cause liver and kidney lesions in rats.However, the indolyl glucosinolate metabolite, indole-3-carbinol, has been linked to anti-cancer activity of theBrassicaceae, by induction of cytochrome P450-mediateddetoxification of carcinogens (Baldwin and LeBlanc, 1992; Parkand Bjeldanes, 1992).1.2 Effects of Glucosinolates on InsectsGlucosinolates exist in plants as defensive compoundsagainst invading organisms or herbivores. A complete reviewof the biological interactions of glucosinolate-containingplants with insects, fungi and other plants is outside thescope of this thesis. A recent review of the biologicaleffects of glucosinolates has been compiled by Chew (1988).Vershcaffelt (1911) first reported that larvae of Pierisbrassicae and P. rapae (the cabbage white and importedcabbageworm butterflies, respectively) would feed on non-hostplant leaves smeared with the juice of a cruciferous plant(Bunias orientalis) or a solution of pure sinigrin (ally'glucosinolate). Since then there have been many studiespublished on the biological activity of glucosinolates andtheir fission products towards insects (Table 1).Ta•1. Biological Activity of Glucosinolates and Their Fission Products in InsectsBiological Activity Active Compounds^ ReferencesInsect/ArachnidSpeciesAtt actant(h•st plant) allyl isothiocyanateethyl isothiocyanatebenzyl isothiocyanate,ethyl + methylisothiocyanatobutyratePhyllotretacruciferaeHylemya floralisH. brassicaeDelia brassicaeP. striolataD. brassicaeP. cruciferaeP. striolataP. cruciferaeP. striolataFeeny et al., 1970;Pivnick et al. , 1992.Rygg and Somme, 1972.Nair and McEwen,1976.Wallbank and Wheatley,1979; Finch andSkinner, 1982.Pivnick et al. , 1992.Finch and Skinner,1982.Pivnick et al., 1992.Pivnick et al. , 1992.3-methylthiopropyl,^. cruciferae + P. striolatabutyl isothiocyanate,methyl isothiocyanatobutyratetDAtt actant(c•nt.)volatiles:isothiocyanates, nitrilesD. brassicae Finch, 1978.Att actant(habitat of host)allyl isothiocyanate Diaeretiella rapae Read et al., 1970.Det rrent sinigrin Lipaphis erysimi Malik et al.,^1983.2-phenylethylisothiocyanateH. floralis Rygg and Somme, 1972.Ovi ositionst mulant sinigrin ErioischiabrassicaeTraynier, 1965.H. brassicae Nair and McEwen, 1976.Pieris rapae Renwick and Radke,1983; Traynier,1984;Traynier andTruscott, 1991.Plutella xylostella Reed et al., 1989.sinalbinbenzyl glucosinolateD. radicumH. brassicaeP. xylostellaH. brassicaeD. radicumRoessingh et al., 1992.Nair and McEwen, 1976.Reed et al., 1989.Nair and McEwen, 1976.Roessingh et al., 1992.1-40Ov ositionI*s imulant (cont.)P. xylostellaP. rapaeD. radicumH. brassicaeReed et al., 1989.Traynier and Truscott,1991.Roessingh et al., 1992.Nair and McEwen, 1976.P. xylostella^Reed et al., 1989.3-indolylmethylglucosinolate3-methylsulfinylpropyl,3-methylsulfonylpropylglucosinolate1-methoxy-3-indolylmethylmixed indole,indole glucosinolates+ sinigrin2-OH-3-butenyl, 3-butenyl, D. radicum4-pentenyl, 2-phenylethylglucosinolateally' isothiocyanateB-phenylethylamine4-methylthio-3-butenylisothiocyanate, 1-cyano-4-methylthio-3-buteneRoessingh et al., 1992.E. brassicae^Traynier, 1965.D. brassicae^Ellis et al., 1980.Fee 'rig stimulant^sinigrin^ P. rapae^Verschaffelt, 1911.P. brassicae^Verschaffelt, 1911;Thorsteinson, 1953.P. maculipennis^Thorsteinson, stimulantnt.)sinigrin Hydaphis erysimi Nault and Styer, 1972.Brevicoryne brassicaeMyzus persicaeP. cruciferaeP. nemorumHicks, 1974.Nielsen, 1978;Nielsen, 1989.sinalbinP. amoraciaeP. undulataP. tetrastigmaPhaedon cochleariaeCeutorhynchusconstrictusP. brassicaeP. maculipennisP. amoraciaeP. nemorumP. undulataP. tetrastigmaNielsen et al., 1979.Nielsen, 1978.Nielsen et al. , 1989.Thorsteinson, 1953.Nielsen et al., 1979.Nielsen, 1978.Fee(c ••.:- • ing stimulant sinalbin P. cochleariae Nielsen,^1978.0 • nt.)methyl glucosinolate P. brassicae David and Gardiner,1966.P. cruciferae Hicks,^1974.methyl glucosinolate P. amoraciae Nielsen et al., 1979.benzyl glucosinolateP. nemorumP. undulataP. tetrastigmaP. cochleariaeP. brassicaeP. cruciferaeP. amoraciaeP. nemorumP. undulataP. tetrastigmaP. cochleariaeNielsen, 1978.David and Gardiner,1966.Hicks, 1974.Nielsen et al., 1979.Nielsen, 1978.P. maculipennisP. cruciferaeP. amoraciaeP. nemorumP. undulataP. tetrastigmaP. cochleariaeP. cruciferaeP. amoraciaeP. nemorumP. undulataP. tetrastigmaP. cochleariaeP. maculipennisThorsteinson, 1953.Hicks, 1974.Nielsen et al. , 1979.Nielsen, 1978.Hicks, 1974.Nielsen et al., 1979.Nielsen, 1978.Nayar andThorsteinson, 1963.3-methylsulfonylpropylglucosinolate3-methylsulfinylpropylglucosinolate2-hydroxy-3-butenyl >3-methylsulfonylpropyl =2-OH-isobutyl =4-methylthiobutyl >benzyl = sinigrin =sinalbin > 2-phenylethyl =3-butenyl glucosinolateFe=ding stimulant(•ont.)Papilio polyxenesSpodoptera eridaniaS. frugiperdaS. frugiperdaOstrinia nubilalisErickson and Feeny,1974.Blau et al., 1978.Bartelt andMikolajczak, 1989.Bartelt andMikolajczak, 1989.Fee•ing deterrent^sinigrin3-methoxybenzylglucosinolate,3-methoxybenzylisothiocyanateAphis fabae^Nault and Styer,1972.Acyrthosiphon solaniA. pisiumS. frugiperda^Bartelt andMikolajczak, 1989.0. nubilalisGro th inhibition^sinigrin3-methoxybenzylglucosinolate,3-methoxybenzyl,4-methoxybenzyl,benzyl isothiocyanate2-methoxybenzyl,3,4-dimethoxybenzylisothiocyanateOvi idal^isothiocyanates,1-cyano-2-phenylethane,2-phenylethyl isothiocyanateS. frugiperda^Bartelt andMikolajczak, 1989.Dasineura brassicae Ahman, 1986.UiDi ruption ofn•rmal morphogenesisIn ecticidal2-phenylethylisothiocyanate3-butenyl, 2-phenylethylglucosinolatesinigrinManduca sextaP. maculipennisP. polyxenesUjvary et al., 1989.Nayar andThorsteinson, 1963.Erickson and Feeny,1974; Blau et al.,1978.3-methoxybenzylglucosinolateS. frugiperda Bartlet andMikolajczak, 1989.Blatella germanica Lichtenstein et al.,1962; 1964.Drosophila melanogasterEpilachna varivestisMacrosiphum pisiMusca domesticaTetranychus atlanticusTribolium confusuminopus rubriceps^Lowe et al., 1971.I. rubriceps^Lowe et al., 1971.S. frugiperda^Bartelt andMikolajczak, 1989.0. nubilalis2-phenylethylisothiocyanatephenyl isothiocyanate3-methoxybenzylisothiocyanaternIns cticidal(c•nt.)2-methoxybenzyl,4-methoxybenzyl,benzyl isothiocyanateS. frugiperda^Bartelt andMikolajczak, 1989.ction ofxification enzymess^ sinigrin,indole-3-carbinol,indole-3-acetonitrileS. frugiperda^Yu, 1983, 1984.GS -transferaseal•ha-naphthylac tate esterasesinigrin,indole-3-carbinol,indole-3-acetonitrile,2-phenylethyl isothiocyanateindole-3-carbinolindole-3-acetonitrileYu and Hsu, 1985.InddetMF ••ep►xide hydrolase indole-3-carbinol Yu and Hsu, 1985.InhdetMFbition ofxification enzymes2-phenylethylisothiocyanateS. frugiperda Yu, 1983.18Glucosinolates generally act as feeding stimulants andoviposition stimulants for insects which specialize onbrassicaceous host plants.^However, they are feedingdeterrents and growth inhibitors to many polyphagous insects,and toxic to some insects which specialize on non-glucosinolate-containing plants. The volatile metabolites ofglucosinolates are generally attractive to oligophagouscrucifer-feeding insects. Although isothiocyanates andnitriles may be toxic to both crucifer-specialist and non-specialist insects, the former species may avoid intoxicationby behavioural (Ahman, 1985) or physiological (MacGibbon andAllison, 1968) mechanisms.1.3 Glucosinolates and Pest ManagementWith canola, the reduction or elimination ofglucosinolates would hypothetically result in two majorchanges in the insect fauna of the crop: 1. lower infestationlevels of specialist Brassica insects, and 2. increasedpopulation densities of non-specialist feeders normallydeterred by glucosinolates.The first hypothetical change - a reduction in Brassicaspecialist insects - has not occurred on the Canadianprairies, in spite of the extensive use of low-glucosinolaterape cultivars (Lamb, 1989). There is no doubt thatglucosinolates contribute to host plant recognition. However,the expected relationship with Brassica specialist insects to19their host plants - increasing host resistance with decreasingglucosinolate concentration - has not been demonstrated by thepublished results of many field and out-of-door experiments.Rather, the concentration of glucosinolates appears to havelittle influence on the infestation levels of Brassicaspecialists.Nair et al. (1976) found that the oviposition preferenceof the cabbage maggot, Hylemya brassicae was not correlated tolevels of total glucosinolates in the foliage of six hostplants. Gerber and Obadofin (1981) demonstrated that high-and low-glucosinolate oilseed rape cultivars were notsignificantly different as food plants for the red turnipbeetle, Entomoscelis americana. Similarly, Ahman (1982) foundno significant differences between high- and low-glucosinolaterape plants in the level of Brassica pod midge (Dasineurabrassicae) infestation. Lerin (1983) reported that the numberof cabbage seed weevils (Ceuthorrhynchus assimilis) infestingrape plants were not significantly different between high- andlow-glucosinolate cultivars. However, C. napi, anothercrucifer specialist weevil, apparently grows more rapidly onhigh glucosinolate cultivars. Larsen et al. (1983) found nosignificant correlations between glucosinolate content and thenumbers of blossom beetles (Meligethes aeneus) infesting high-and low-glucosinolate rape cultivars. Lamb (1988) found thatneither high nor low glucosinolate levels influenced fleabeetle (Phyllotreta spp.) attack. Surprisingly, the mustard20aphid, Lipaphis erysimi, appears to be adversely affected byglucosinolates in its host plants (Labana et al., 1983; Maliket al., 1983).Recent laboratory studies have shown that the hostselection responses of some crucifer specialists toglucosinolates are not necessarily side-chain- or dose-dependent. Reed et al. (1989) found that the numbers of eggsoviposited by diamondback moths, P. xylostella, on filterpaper treated with eight different glucosinolates were notsignificantly different. However, a significant positiveresponse with dose was observed using sinigrin. Nielsen(1978) determined that although glucosinolates were necessaryas feeding stimulants for four species of chrysomelid beetles,the side chain was unimportant. The horseradish flea beetle,Phyllotreta amoraciae requires the presence of certainflavonol glycosides in addition to glucosinolates in order tofeed (Nielsen et al., 1979). Bodnaryk and Palaniswamy (1990)found that feeding by the flea beetle P. cruciferae was notinfluenced by the glucosinolate side-chain or concentration inBrassica cotyledons. However, in trapping experiments withglucosinolate autolysis products, Pivnick et al. (1992)observed that allyl isothiocyanate was more attractive toPhyllotreta spp. than other volatiles tested. A positiveresponse with dose was also observed with this compound.At the level of the chemoreceptor, Roessingh et al.1992) found tnat cabbage root fly, D. radicum, tarsal21glucosinolate receptors responded differently toglucosinolates with different side chains, and a dose-responseeffect in oviposition behavior was only found for 3-indolylmethyl glucosinolate. However, they also determinedthat the most active fraction of the surface extract ofcauliflower leaves did not contain glucosinolates. Theaforementioned field and laboratory studies call into questionthe importance of glucosinolates in host plant relations ofBrassica specialist insects and suggest the existence ofadditional, non-glucosinolate kairomones.The second change in the insect fauna hypothesized withthe introduction of low-glucosinolate cultivars is an increasein the pest incidence of polyphagous insects. Infestationlevels of non-specialist insects would be expected to increasewith decreasing levels of glucosinolates. This trend has beenreported for the green peach aphid, Myzus persicae (Lerin,1983). However, other polyphagous insects may not beinfluenced by glucosinolate levels. Loschiavo and Lamb (1985)found no significant correlations between survival ordevelopmental time and glucosinolate content for fourcoleopteran pests of stored products. Similarly, Butts andLamb (1990) found that field infestations of Lygus spp. bugswere not significantly different between cultivars of the sameBrassica spp. Further, the survival and developmental time ofnymphs was not influenced by the glucosinolate level of thecultivars.22However, the possibility of non-specialist insects movinginto the canola crop has been realized to a certain degree.Lygus spp. bugs (Butts and Lamb, 1990) and the Northern falsechinch bug, Nysius niger (Pivnick et al., 1991), have recentlybeen reported as pests on canola. The Bertha armyworm,Mamestra configurata Walker is a polyphagous noctuid which hasbecome a pest of oilseed rape since 1960 (Turnock, 1988).1.4 ObjectivesThis thesis was an attempt to examine the role ofglucosinolates as allomones to a polyphagous insect and askairomones to an oligophagous insect. The Bertha armyworm waschosen as the model polyphage because it is a pest ofbrassicaceous crops and is amenable to laboratory culture.The diamondback moth was chosen as the model oligophagebecause it is also a pest of brassicaceous crops and areliable immigration of this species occurs each summer intothe Vancouver area. The objectives of the research were todetermine if foliar glucosinolates affect the consumption andgrowth rate of M. configurata and whether P. xylostella isaffected by host plant glucosinolate profile in the field.These questions were addressed to the potentialexploitation of glucosinolates as resistance factors againstpolyphagous insect pests of Brassica crops. With M.configurata, host plant glucosinolates were considerpn qualitatively and quantitatively. Food consumption and growth23of the larvae were also quantified, and attempts were made tocorrelate these quantities with the concentration ofglucosinolates in host plant foliage and in artificial media.A different approach was taken for the diamondback moth.Since there were consistent natural populations in the area,the abundances of the larvae on host plants with differentglucosinolate profiles were measured and compared. For thelatter study, host plant glucosinolates were consideredqualitatively. The influence of host plant glucosinolateprofile on a natural enemy, Diadegma insularis of thediamondback moth was also investigated .1.5 Biology of Brassica Insects Studied1.51 The Bertha Armyworm, Mamestra configurata WalkerM. configurata is one of the most destructive pests ofbrassicaceous crops, although serious infestations do notoccur every year in western Canada (Lamb et al., 1985). Thebiology of M. configurata has been concisely described byBurgess et al. (1979). The moths, which are nocturnal, emergein mid-June through July on the Canadian prairies. Eggs arelaid on the underside of canola leaves in clusters of severalhundred, one layer thick. In approximately one week theyhatch, and the neonate and early instar larvae (first to thirdor fourth) feed on foliage. The caterpillars are not highlymobile and complete dPfnliatinn i unoommon MajfIr damag is inflicted by the older instar larvae (fourth to sixth), which24feed directly on the seed pods. Mature caterpillars pupate inthe soil and overwinter. M. configurata is univoltine.M. configurata is a nearctic species (Wylie and Bucher,1977). Naturally-occurring biological control agents includethe tachinid fly Athrycia cinerea (Turnock and Bilodeau,1984), the ichneumonid wasp Banchus flavescens (Turnock andBilodeau, 1984; Arthur and Mason, 1985) and the braconid waspMicroplitis mediator (Arthur and Mason, 1986). However,tillage and chemical control are the methods also employed forthe M. configurata on canola (Burgess et al., 1979; Harris,1985).1.52 The Diamondback Moth, Plutella xylostella LinneausP. xylostella is a well-known pest of brassicaceous cropsworldwide. This insect does not overwinter in Canada in largenumbers. Each summer moths immigrate from the United States.The biology and life cycle of this lepidopteran in Canada issummarized in Burgess et al. (1979). Eggs are laidindividually on the upper surfaces of the leaves and theneonate larvae burrow into the leaf. The later instars(second to fifth) feed mainly on the undersides of the leaves.The damage to the foliage is normally not significant.However, the larvae will feed on the flowers and young podsresulting in reduced yields. Pupation occurs in silkencocoons attached to the plants. Although P. xylostella hasIrevm yeneLdtions each year in Canada, only the second25(normally the last week of July) is economically important.Pesticides are normally recommended for control (Burgess etal., 1979; Harris, 1985).262.0 Influence of Foliar Glucosinolates in Oilseed Rape andMustard on Feeding and Growth of the Bertha Armyworm, Mamestraconfigurata Walker2.1 IntroductionRecent advances in the genetic engineering of higherplants, coupled with the renewed enthusiasm for low inputsustainable agriculture have emphasized the advantages of hostplant resistance: environmental neutrality, relativestability, and low cost in the long-term. Allelochemicalshave been proposed or recognized as resistance factors inplants to insects in a large number of cases (Smith, 1989).Allelochemicals in plants may affect both the palatability ofthe substrate as well as the growth of the herbivore, twoimportant parameters of insect resistance. In this mode,glucosinolates have been shown to function as allomonesagainst some non-specialist insects (Nault and Styler, 1972;Blau et al., 1978).Isothiocyanates, produced by some glucosinolates uponhydrolysis by endogenous myrosinase (E.C., have beenshown to be toxic to some insects (Lichtenstein et al., 1962;1964; Nayar and Thorsteinson, 1963; Lowe et al., 1971; Ahman,1986; Bartelt and Mikolajczak, 1989). The presence ofglucosinolates in the foliage of the crop may be advantageousif the consumption and growth of foliage-feeding insects canbe inhibited. In the present study, foliar glucosinolates are27qualitatively and quantitatively examined, and theirrelationship to food consumption and growth of Mamestraconfigurata is examined.Eight cultivars representing four species of cultivatedcanola and mustard were chosen. The criteria for selectionwere their differences in glucosinolate profile and theavailability of sufficient quantities of seed. These includedtwo cultivars of Brassica juncea, commercial Brown mustard andthe yellow seeded Oriental mustard variety, Lethbridge 22A;the white mustard, Sinapis alba cv. Gisilba; three B. napuscultivars, the high glucosinolate oilseed rape Midas and twocanola cultivars, Westar and Regent; and two B. rapa canolacultivars, Candle and Tobin. Two types of glucosinolates,based on the type of fission products yielded upon autolysis,dominate in the foliage of the plants studied: stableisothiocyanate-releasing glucosinolates and thiocyanate (SCN) -releasing glucosinolates. While isothiocyanates are known tobe toxic, the biological activity of thiocyanate-releasingglucosinolates and their corresponding R-alcohols towardsinsects is less understood.At the pH range of the lepidopteran gut, allylisothiocyanate and indole-3-carbinol are the expectedmetabolites of allyl glucosinolate (sinigrin) and 3-indolylmethyl glucosinolate (glucobrassicin), respectively.These were commercially available examples of the two types ofglucosinolate metabolites of interest.^Sinigrin, also28commercially available, is used as an example of an intactglucosinolate. This study examines the relationship betweenbiological activity against M. configurata and concentrationsof total glucosinolates, isothiocyanate-releasingglucosinolates and thiocyanate-releasing glucosinolates.Plant foliage effects were verified with feeding studies usingpure compounds added to meridic diet.2.2 Materials and MethodsInsect culture. A laboratory culture of Mamestra configuratawas maintained at 20 °C with a 16:8 (hr light-dark) photoperiodon an agar-based meridic diet (Velvetbean caterpillar diet,BioSery No. F9795, BioServ, Frenchtown, N.J., U.S.A.)augmented with 1.5% Vanderzant Vitamin mixture and 1% alfalfameal. Lamb's quarters (Chenopodium album L.) were presentedto moths for oviposition (Bucher and Bracken, 1976). Eggswere removed from the leaves prior to hatching so that insectsused for experiments had no previous experience with plantmaterial.Insect growth on greenhouse-grown plants. Seeds of Brassicanapus cv. Westar, Regent and Midas, B. rapa cv. Candle andTobin, B. juncea cv. Lethbridge 22A, and commercial brownmustard and Sinapis alba cv. Gisilba were sown in vermiculite.Cotyledon stage plants were individually transplanted into 10cm pots of a homogenous, sterilized soil mix and grown to theour-leat rosette stage in the Agriculture Canada Research29Station greenhouse under high-intensity sodium vapour lamps(40,000 lux at plant level). The B. juncea and S. albacultivars were sown one week after the B. napus and B. rapacultivars were sown so that the plants would be approximatelythe same size (4-leaf rosette stage) simultaneously. Soluble20-20-20 (N:P:K) fertilizer was applied weekly.Twenty-seven (B. napus and B. rapa) or 20 (B. juncea andS. alba) days post-seeding, five neonate M. configurata larvaewere placed on each of ten rosette-stage plants per variety.A screen-bottomed, clear plastic beverage cup was invertedover each plant and embedded into the soil to confine theinsects. The plants and insects were transferred to acontrolled environment growth chamber with a 16:8 (hr light-dark) photoperiod (16,000 lux) and a constant temperature of20°C for seven days, after which time the insects wereweighed. The data was subjected to ANOVA with orthogonalcontrasts for differences based on species and cultivar withinspecies. Each insect was considered an experimental unit.(Separate ANOVAs for differences between plants of individualcultivars yielded no significant differences.) Plants wereplaced on two benches within the growth chamber. Possiblevariation due to light quality and air flow within the growthchamber was considered in the ANOVA.Biological assays with field-grown foliage. Seeds of thepreviously described varieties were sown in the field at thePlant Science Field Laboratory, University of British Columbia30campus, Vancouver. Four blocks which included plots of eachcultivar were sequentially planted to ensure that foliage ofthe desired plant growth stages were available for testingwhen insect stocks were available. The pre-emergent herbicidetrifluralin (Treflan 0 E.C.) was applied to control weeds.Soluble 20-20-20 (N:P:K) fertilizer was applied weekly toensure that basic nutrients would be available throughout thegrowth stages. Blocks were longitudinally sub-divided intostrips such that individual plants were sampled once only.Three broad plant growth stages (Harper, 1973) wereexamined in the study: rosette (stage 2), stem elongation(stage 3) and flowering (stage 4). Two blocks (= replicatesin time) were sampled for each growth stage except theflowering stage due to limitations in the insect culture. Allof the plant species in the study had characteristic leaftypes for each of these growth stages except for S. alba. Theyoungest fully-expanded leaf characteristic of the growthstage was taken from each plant sampled. One replicate of onegrowth stage was sampled at each sampling interval, and thefollowing biological assays were carried out within a periodof three days:i) Neonate choice test. Leaf discs were punched from foliageusing a 1.6 cm cork borer. Discs of each cultivar (eight intotal) were arranged randomly and evenly spaced around theperimeter of a Petri dish (14 cm) on a moist filter paper disc(12.5 cm, Whatman no. 1). One hundred neonate M. configurata31larvae were introduced into the center of each dish which wascovered and left in darkness at 20 °C for 16 hours, after whichtime the numbers of larvae on each leaf disc were recorded.The darkness was necessary to prevent the larvae from movingtowards the direction of the light source in the growthchambers. Ten replicate petri dishes (1000 insects) were setup for each sampling interval. Because the treatmentvariances were proportional to the means, the data was square-root transformed ([x + 0.51 05 ) prior to analysis.ii) Fourth instar nutritional indices. Individual leaves wereexcised from the plants and placed in a 10 cm Petri dish on amoist filter paper disc. A fourth instar M. configuratalarva, within a weight range of 13 and 22 mg, was placed onthe leaf and allowed to feed for two days at 20 °C with a 16:8(hr light-dark) photoperiod. The following nutritionalindices (Waldbauer, 1968) were calculated on a dry weightbasis using the insect's weight at the start of the feedingperiod as the reference weight (Farrar et al., 1989):Relative Consumption Rate (RCRi)= food ingested / insect initial weightno. of daysRelative Growth Rate (RGRi)= weight gained / initial of daysSixteen insects were tested at each sampling date (replicate).Fresh weight to dry weight conversion factors were based onthe average of 10 untreated insects and leaves of eachcultivar per replicate. For 1paf nnnverRinn factors, leaves were weighed at the start of the feeding period, placed in32petri dishes without larvae, and dried at the end of thefeeding period. This was done to account for leaf weight lossdue to respiration during the course of the experiment, asdetermined in preliminary trials.For the two described assays, ANOVA with orthogonalcontrasts was performed for each plant growth stage. Thecultivar means were regressed against glucosinolateconcentration for each block.Glucosinolate analysis. Bulk samples of foliage (a minimum of20 plants) from each sampling interval were collected forglucosinolate analysis. Foliage was quick-frozen with liquidN2 , lyophilized, and ground to a homogenous powder in ablender (maximum particle size 1 mm). Freeze-dried materialwas stored in a -20°C freezer prior to analysis.Glucosinolates were extracted using the followingprotocol. One hundred mg samples of plant material wereplaced in centrifuge tubes and heated for two minutes in aboiling water bath. Three ml of boiling 70% methanol wasadded to each sample. The slurry was kept at the solventboiling point for two minutes with constant mixing and thencentrifuged for ten minutes. The pellet was re-extractedtwice more in the same manner, and the supernatants pooled.Methanol was removed by heating in a 40 °C water bath under astream of N2 . The aqueous solution was made up to 6 ml withwater, and a 125 ul aliquot of 0.6 M barium-lead acetatesolution (0.3 M Ba(CH3C00) 2 and 0.3 M Pb(CH3C00) 2) was added to33precipitate phenolics and free sulphate (McGregor, 1985). Thesolution was then centrifuged, and the supernatant used forglucosinolate determination. Extraction efficiency averaged83% as determined by HPLC of recovered desulfosinigrin fromsinigrin-spiked Westar Canola foliar slurries. Reportedquantifications are the means of two separate extractions andanalyses.Total glucosinolates. Total glucosinolates were quantified bythe thymol method (Brzezinski and Mendelewski, 1984; Tholen etal., 1989). The supernatant was applied to a 100 mg (dryweight) DEAE-Sephadex A-25 minicolumn regenerated with water.The columns were washed with 2 x 1 ml of 30% w/v formic acidand 4 x 1 ml water, and glucosinolates were eluted with 5 x 1ml 0.3 M potassium sulphate. The volume of the eluate wasstandardized by removing the water by freeze-drying and takingup in precisely 5 ml of water. A 0.5 ml aliquot was mixedwith 100 ul of ethanolic 6% thymol and 2 ml 78% v/v sulphuricacid, and this solution was incubated for 35 minutes in a 93 °Cwater bath. The absorbance of the solution was read at 505 nmon a Pye Unicam PU 8620 UV/VIS/NIR spectrophotometer, and thelinear regression equation of a standard curve of sinigrin wasused for quantification.Individual glucosinolates. These were determined by HPLC oftheir desulfo-derivatives combining the methods of McGregor(1985) and Bjerg and Sorensen (1987). The supernatant wasapplied to a 100 mg (dry weight) minicolumn of DEAE Sephadex34A-25 regenerated in 0.02 M sodium acetate buffer (pH 5.0).The minicolumn was washed with 4 x 1 ml of water and 4 x 1 mlof the sodium acetate buffer. A 0.5 ml aliquot of arylsulfatase (E.C. (Sigma, H-1) solution, prepared as perMcGregor (1985) was applied to the column. Columns wereplaced in darkness for 20 hours, after whichdesulfoglucosinolates were eluted with 4 x 1 ml water. Thesamples were concentrated to dryness using a freeze-drier andthen taken up in 400 ul water in an ultrasonic bath. Twentyul aliquots were analysed by HPLC.The HPLC system consisted of a Rheodyne Model 7125syringe loading sample injector, two Waters Model 510 pumpsand a Waters 490 multiwavelength (UV-Vis) detector.^Thesystem was controlled and data recorded by Waters Expertchromatography software (version 2.0) on a DigitalProfessional 350 personal computer linked to the HPLC by aWaters 840 interface module. The analytical column used waseither a Waters Nova-Pak 0 or a Phenomenex Bondclone (bothC18, 3.9 x 150 mm). This was preceded by a 2 x 20 mm guardcolumn packed with C18 pellicular packing (PerisorbUpchurch Scientific). The solvent program consisted of waterand acetonitrile at the following ratios:^99:1(water:acetonitrile) for 8 minutes, followed by a lineargradient to 76:24 at 34 minutes, and held at this ratio until36 minutes.^The flow rate was 0.6 ml/minute, and theabsorbance of desulfoglucosinolates was monitored at 226 nm.35Quantifications were based on response factors relativeto an internal standard of 0-nitropheny1-1 3-D-galactopyranoside(Sigma) as described by McGregor (1985). This involvedcollecting individual desulfoglucosinolates separated by HPLC.The concentration of a solution of the pure desulfo-glucosinolate was determined spectrophotometrically by thethymol method as described for quantification of totalglucosinolates, above. However, rather than using a standardcurve of sinigrin, the absorption coefficient (Abs. units /umol glucose) of a standard glucose solution (0.0003 M) wasused for quantification. An aliquot (40 ul) of knownconcentration of 0-nitrophenyl-p-D-galactopyranoside (0.006 M)was added to an aliquot (100 ul) of the desulfoglucosinolatesolution and this was injected into the LC. Thedesulfoglucosinolate response factor relative to 0-nitrophenyl-p-galactopyranoside is calculated using the ratioof peak areas of the two compounds. Response factors weredetermined twice and the average taken as the true responsefactor. These are listed in Appendix 1. Published responsefactors from the literature (McGregor, 1985; Muuse and Van derKamp, 1986) show that glucosinolates of the same general side-chain structure have very similar if not identical responsefactors. For this reason, the response factors determined forallylglucosinolate and 3-indolylmethylglucosinolate were usedfor all monoalkenyl and indolyl glucosinolates, respectively.Identification of major peaks was confirmed by HPLC-MS as36per the methods of Hogge et al. (1988a and 1988b). Tables ofexpected ion molecular weights of the desulfoglucosinolatesare located in Appendix 2. The quantities of isothiocyanate-or thiocyanate-releasing glucosinolates from individual peakswere added and these values used in the correlation analyseswith the bioassays.Assay of pure compounds in meridic diet. Sinigrin monohydrate(Sigma), allyl isothiocyanate (Eastman) and indole-3-carbinol(Sigma) were tested against M. configurata in meridic diets atconcentrations of 0.5, 1.0 and 1.5 umols/gram wet weight.Sinigrin and indole-3-carbinol were dissolved in methanol andadded to the dry mix of the standard diet previously describedand the methanol allowed to evaporate for five hours prior tomixing with the agar-water portion. A control diet wasprepared by adding methanol only. Allyl isothiocyanate, whichis extremely volatile, was added directly to the molten dietprior to gelling. The solidified diets were divided into 30ml plastic cups and a single neonate larva was introduced intoeach cup. The tightly capped cups were placed in a coveredplastic box humidified by moist (distilled water) laboratorytissue (Kimwipes, Kimberly-Clark) lining the bottom. Theboxes were placed inside a controlled environment chamber at23 °C with a 16:8 (hr light-dark) photoperiod. After eightdays, the insects were weighed. Forty insects were used foreach treatment and the mean larval weights for each treatmentwere regresses against concentration for each compound.37Allyl isothiocyanate fumigation assay. Uniform cylindricalportions (1.0 cm diameter, 0.5 cm height) of allylisothiocyanate-containing diet (1.5 umol/gram wet diet) andcontrol diet were cut from solidified diet using a cork borer.The average weight for 10 portions was 0.4299 grams, withSEM = 0.0093 grams (2.17 %). One neonate M. configurata larvawas placed into each of twenty rearing cups per treatment.The treatments consisted of (1) a portion of allylisothiocyanate-containing diet, (2) a portion of allylisothiocyanate-containing diet plus a portion of control diet,and (3) a portion of control diet only. The insects were heldat 23 °C with a 16:8 (hr light-dark) photoperiod for eightdays, after which survivors were counted.2.3 ResultsHPLC glucosinolate profiles of the plants studied arequalitatively similar to those published for B. juncea and B.napus cv. Midas (Sang et al., 1984). The dominantglucosinolate in B. juncea is allyl glucosinolate (sinigrin,1); in S. alba, hydroxybenzyl glucosinolate (sinalbin, 2)dominates (Figure 2). Among the rape species, the indolylglucosinolates predominate, particularly 3-indolylmethylglucosinolate (8) and its 4-hydroxy analog (4) (Figure 3).Quantifications of total intact glucosinolates andindividual desulfoglucosinolates were very rpprnciunihlP Within each method. The HPLC method yielded results consistent with3830 3510^15^20^25Retention time (minutes)Brassica juncea1511Sinapis alba 21. allyl (sinigrin)3. 3-butenyl5. ONPGaI (internal std.)7. benzyl9. 2-phenylethyl11. 1-methoxy-3-indolylmethyl2. OH-benzyl (sinalbin)4. 4-0H-3-indolylmethyl6. 4-pentenyl8. 3-indolylmethyl10. 4-methoxy-3-indolylmethyl0^5^10^15^20^25^30^35Retention time (minutes)Figure 2. Representative HPLC chromatograms of mustardspecies used in the study. Peak numbers identifydesulfoglucosinolates.4839Brassica napus Westar^s549 10^1110^15^20^25^30^35Retention time (minutes)Brassica napus Midas5910^114 A^400^5 10^15^20^25^30^35Retention time (minutes)Brassica rapa Candle515^20^25Retention time (minutes)510 30 35Figure 3.^Representative HPLC chromatograms of rapespecies used in the study. Peak numbers as in Figure 2.40other published quantifications (Jurges, 1978; MacfarlaneSmith and Griffiths, 1988; McGregor, 1988; Reed et al., 1989;Bodnaryk, 1991). However, from Appendix 3, it appears thatthe thymol method for total glucosinolate quantificationyields much higher values than would be expected when comparedwith the HPLC quantifications of isothiocyanate- andthiocyanate-releasing glucosinolates. The magnitudes of thedifferences are variable, but generally smallest for S. albaand greatest for B. rapa, suggesting that inherent plantfactors, rather than experimental inconsistency, may accountfor most of the difference. The following possibilities maybe additive: 1. Other glucosinolates (unidentified peaks inthe HPLC chromatograms) were not quantified by the HPLCmethod; 2. Other glucosinolates were not visible using a UVdetector, for example butyl glucosinolate in B. juncea (Roggeet al., 1988a), were not quantified by the HPLC method; 3.The differential affinity of sulfatase for differentglucosinolates (Quinsac and Ribaillier, 1985; Buchner, 1987)may have resulted in incomplete desulfation of certainglucosinolates in the presence of other glucosinolates; and 4.Other interfering glycosylated sulfates may have been includedin the ion exchange purification of intact glucosinolates.Nevertheless, the trends for both methods were consistent witheach other (that is, high glucosinolate plants were high, andlow glucosinolate plants were low).The growth rate of neonate M. configurata as measured by41larval weight after seven days was influenced by the cultivarupon which the larvae fed (Figure 4). Orthogonal contrasts(Table 2) revealed that there were significant differencesbetween mean larval weights by plant species in the order: B.juncea < S. alba < B. napus < B. rapa. There were nosignificant differences between cultivars within B. juncea orB. napus. However, within B. rapa, larvae reared on Tobinwere significantly larger than those reared on Candle.The relative acceptability of the various host plants toneonate larvae is indicated in Figure 5. The means withresults of Tukey's studentized range test are listed inAppendix 4(a). Larval distribution on host plants of allgrowth stages tested was similar to the ranking of larvalweights in the growth experiment (Table 3). The number oflarvae found on the B. juncea leaf discs was significantlylower than on the other species. However, S. albaconsistently attracted significantly more insects than theother species. In agreement with the results of the growthexperiment, B. rapa was more acceptable than B. napus rosettestage foliage. This ranking was reversed for stage 3 foliage,and there was no significant difference between theattractiveness of the two species for flowering stage foliage.Within B. napus, the numbers of larvae on the high-glucosinolate cultivar Midas did not differ significantly fromthe low-glucosinolate cultivars, Regent and Westar. However,Westar attracted significantly more larvae than Regent.42Mean Larval Weight (mg)25 -Brown L.22A^Gisiba^Midas Regent Wester^Candle TobinCultivarFigure 4. Mean final weights of M. configurata larvaereared for seven days on intact rosette-stage plants.Vertical lines above bars represent SEM.Table^2.^Effect^of^Host^Species^andconfigurata Larvae Reared for Seven DaysANOVA and Orthogonal Contrasts.Source^ dfCultivaron IntactSSon^FinalPlants asMSWeight^of^M.Determined byF^Pr>FTotal 282 0.017788Growth Chamber Bench 1 0.000019 0.000019 0.38 0.5367Cultivars 7 0.004156 0.000594 11.95 0.0001B.juncea vs others 1 0.001767 0.001767 35.57 0.0001S. alba vs B. napus + B. rapa 1 0.000384 0.000384 7.72 0.0058B. napus vs B. rapa 1 0.000695 0.000695 13.99 0.0002B. juncea: Brown vs Lethbridge 22A 1 0.000048 0.000048 0.96 0.3288B. napus: Midas vs Regent + Westar 1 0.000005 0.000005 0.10 0.7526Regent vs Westar 1 0.000044 0.000044 0.89 0.3458B. rapa: Candle vs Tobin 1 0.001074 0.001074 21.63 0.0001Error 274 0.013612B. napus1^I^nyI^ny1^nyI Midas Regent WesterCultivar1B. rapsGlallba1Candle Tobin20_B. Juncea10Brown L.22A1660 , NMean No. Larvae25 -S. alba44Growth Stage:I= 2 (Rosette) L 1 3 (Stem Elongation) MI 4 (Flowering)Figure 5. Numbers of neonate M. configurata on leaf disksin choice tests with field-grown foliage. Vertical linesabove bars represent SEM.45Table 3.^Effect of Host Species and Cultivar on Neonate M.configurata Leaf Disc Choice Tests as Determined by ANOVA andOrthogonal ContrastsContrast^ Growth Stage2^3^4B. juncea vs others ** ** *S. alba vs B. napus + B. rapa ** * **B. napus vs B. rapa ** ** nsdB. juncea: Brown vs Lethbridge 22A nsd nsd nsdB. napus: Midas vs Regent + Westar nsd nsd nsdRegent vs Westar nsd ** nsdB. rapa: Candle vs Tobin nsd nsd nsd* = significant difference, p < 0.05** = significant difference, p < 0.01nsd = no significant difference46Acceptability to neonates was not correlated to foliarconcentrations of total glucosinolates (Table 4). In general,the isothiocyanate-releasing glucosinolates had a negativeinfluence on acceptability, while the thiocyanate-releasingglucosinolates had a positive influence. The influence ofglucosinolates on acceptability to neonates was more obviouswith stage 3 and stage 4 foliage than with rosette foliage.For stage 3 plants, the numbers of larvae attracted to theleaf disks was negatively correlated to the isothiocyanate-releasing glucosinolate concentration in the foliage, and thiswas significant over both blocks of the experiment.The consumption and growth rates of fourth instar M.configurata were also influenced by the host species andcultivar (Figures 6 and 7). The cultivar means of RCRi andRGRi with the results of Tukey's studentized range test arelisted in Appendix 4, (b) and (c), respectively. Predictably,lower consumption rates (RCRi) were observed on B. juncea thanon the other host plants (Table 5), even though the differencewas of borderline significance (p = .0552) for rosette stagefoliage. However, the attractiveness of S. alba observed inthe neonate choice tests was not reflected by the fourthinstar consumption rates (Figure 6). In fact, the mean RCRion S. alba was significantly lower than on B. napus and B.rape stage 3 foliage. There were no significant differencesin RCRi between B. napus and B. rapa for any growth stage.47Table 4. Linear Correlation Coefficients for Relationship ofFoliar Glucosinolates (umol/g dry wt.) to Neonate Choice(weighted means of square root-transformed counts) in LeafDisc AssayGrowth Stage^Total^Isothiocyanate- Thiocyanate-releasing^releasing2 - RosetteBlock 1Block 2r = -.300p = .470(n = 8)r = .063p = .893(n = 7)r = -.409p = .314(n = 8)r = -.524p = .183(n = 8)r = .840p = .009(n = 8)r = .476p = .232(n = 8)3 - Stem elongationr = -.557p = .151(n = 8)r = -.692p = .057(n = 8)r = -.265p = .528(n = 8)r = -.787p = .020(n = 8)r = -.766p = .027(n = 8)r = -.533p = .174(n = 8)r = .634p = .091(n = 8)r = .383p = .348(n = 8)r = .801p = .017(n = 8)Block 1Block 24 - FloweringBlock 148RCRi (mg/mg/day) B. napusa rapsB. Juncos,Brown L.22A^GIsIlba^Midas Regent Wester^Candle TobinCultivarGrowth Stage:ESSI 2 (Rosette) 0 3 (Stem Elongation) /11 4 (Flowering)Figure 6. Mean consumption indices for fourth instar M.configurata larvae on field-grown foliage. Vertical linesabove bars represent SEM.5430490. juncosRGRi (mg/mg/day)1-B. napus-0.2^' IB. rapaS. albaBrown L.22A^GIsJibe^Midas RegentWestar^Candle TobinCultivarGrowth Stage:ESSI 2 (Rosette) EJ 3 (Stem Elongation) 11111 4 (Flowering)Figure 7. Mean relative growth rates of fourth instar M.configurata larvae on field-grown foliage. Vertical linesabove bars represent SEM.50Table 5. Effects of Host Species and Cultivar on RelativeConsumption Rate (RCRi) of Fourth Instar M. configurata asDetermined by ANOVA and Orthogonal ContrastsContrast Growth Stage2^3 4B. juncea vs others nsd ** *S. alba^vs B. napus + B. rapa nsd ** nsdB. napus vs B. rapa nsd nsd nsdB. juncea: Brown vs Lethbridge 22A * nsd nsdB. napus: Midas vs Regent + Westar * nsd nsdRegent vs Westar nsd nsd nsdB. rapa: Candle vs Tobin nsd nsd nsd* = significant difference, p < 0.05** = significant difference, p < 0.01nsd = no significant difference51Relative growth rates (RGRi) of fourth instar larvae werereduced on B. juncea foliage compared to the other plantspecies for stage 3 and 4 foliage (Figure 7), and thesedifferences were highly significant (Table 6). However, nosignificant differences were detected for the rosette stagefoliage in any of the contrasts. S. alba did not appear toinhibit the growth of fourth instar larvae as observed withneonates. The comparison of B. napus with B. rapa alsoyielded no significant differences, although with stage 3foliage, the growth rate of larvae on B. napus was somewhatreduced compared to that on B. rapa (p = 0.06). This was mostlikely due to the influence of the high-glucosinolate B. napuscultivar Midas, which yielded a significantly reduced RGRicompared with the low-glucosinolate cultivars Regent andWestar.Effects on fourth instar M. configurata can be linked tothe glucosinolate composition of the plants in some cases.However, a relationship between RCRi and glucosinolate contentis not well-defined. Consumption rate tends to be inverselyrelated to the concentration of total glucosinolates, butthere was no significant linear correlation except in thesecond block of the stem elongation stage foliage (Table 7).Simple linear correlations between RCRi and isothiocyanate-and thiocyanate-releasing glucosinolates were also equivocal.The relationship of RGRi and glucosinolate contentappears to be relatively simple. RGRi is significantly52Table 6.^Effects of Host Species and Cultivar on RelativeGrowth^Rate^(RGRi)^of^Fourth^Instar^M.^configurata^asDetermined by ANOVA and Orthogonal ContrastsContrast Growth Stage2^3^4B. juncea vs others nsd ** **S. alba^vs B. napus + B. rapa nsd nsd nsdB. napus vs B. rapa nsd nsd nsdB. juncea: Brown vs Lethbridge 22A nsd nsd nsdB. napus: Midas vs Regent + Westar nsd * nsdRegent vs Westar nsd nsd *B. rapa: Candle vs Tobin nsd nsd nsd* = significant difference, p < 0.05**nsd= significant difference, p= no significant difference< 0.0153Table 7. Linear Correlation Coefficients for Relationship ofFoliar Glucosinolates (umols/g dry wt.) to RelativeConsumption Rate (mean RCRi) of Fourth Instar M. configurataGrowth Stage^Total^Isothiocyanate- Thiocyanate-releasing^releasing2 - RosetteBlock 1Block 2r = -.611P = .108(n = 8)r = -.669p = .100(n = 7)r = -.642p = .086(n = 8)r = 0p = .966(n = 8)r = .123p = .773(n = 8)r = .176p = .675(n = 8)3 - Stem elongationBlock 1Block 24 - Floweringr = -.580p = .132(n = 8)r = -.840p = .009(n = 8)r = -.266p = .524(n = 8)r = -.75p = .032(n = 8)r = -.804p = .016(n = 8)r = -.173p = .682(n = 8)Block 1^r = -.574^r = -.530^r = .207p = .136 p = .176 p = .620(n = 8) (n = 8) (n = 8)54negatively correlated to total glucosinolate content in themore mature plant growth stages, although the results aredifferent from eachother in the rosette stage (Table 8). Theinconsistency is most likely due to the unusually high, yetreproducible values for total glucosinolates in the Block 2rosette foliage (Appendix 3). The cause of this anomaly isnot known. However, these correlations appear to be mainlydue to the concentration of isothiocyanate-releasingglucosinolates, as RGRi is also significantly negativelycorrelated to the levels of these glucosinolates, and not tothiocyanate-releasing ones.The data generated by feeding studies using intact planttissue is partially supported by the results of the feedingstudy using pure compounds in meridic diet. The negativegrowth responses of neonate M. configurata larvae to equimolarconcentrations of sinigrin and its metabolite, allylisothiocyanate are similar (Figure 8), in that significantlinear relationships between larval weight and dietaryallelochemical concentration were obtained. However, allylisothiocyanate in the diets resulted in abnormally highmortality rates (Figure 9), which was partially due to thefumigation effect of this volatile compound (Figure 10). Inthe fumigation test with allyl isothiocyanate, 27% of themortality in the treatment of allyl isothiocyanate alone wasdue to intoxication by feeding, while 55% of the mortality wasdue to rumigation.55Table 8. Linear Correlation Coefficients for Relationship ofFoliar Glucosinolates (umols/g dry wt.) to Relative GrowthRate (mean RGRi) of Fourth Instar M. configurataGrowth Stage^Total^Isothiocyanate- Thiocyanate-releasing^releasing2 - RosetteBlock 1Block 2r = -.909p = .002(n = 8)r = 0p = .991(n = 7)r = -.60p = .116(n = 8)r = .095p = .827(n = 8)r = -.032p = .953(n = 8)r = -.434p = .284(n = 8)3 - Stem ElongationBlock 1Block 24 - Floweringr = -.817p = .013(n = 8)r = -.857p = .007(n = 8)r = -.887p = .003(n = 8)r = -.778p = .023(n = 8)r = .253p = .545(n = 8)r = -.063p = .886(n = 8)Block 1^r = -.807^r = -.743^r = .145p = .016 p = .035 p = .731(n = 8) (n = 8) (n = 8)Mean Larval Weight (mg)40 -A0.2^0.4^0.6^0.8^1^1.2Dietary Conc. (umolig wet diet)1.61.4560 81nIgrIn^0 Allyl Ito^A Indolo-3-oarblnolr ■ -.997, p ■ .003^r • -.967, p ■ .033^r • .046, p • .956Figure 8. Linear regression of final larval weight versusdietary concentration of pure compounds augmented tomeridic diet.% Mortality578060402017c74-71^IV\N.% I Control^0.6 1.0Dietary Conc. (umol/g wet diet)Sinlorin 0 Allyl Ito^Indole-3-carblnolFigure 9. Percent mortality versus dietary concentrationof pure compounds augmented to meridic diet.1.6% Mortalityloo -58806040200Allyl ITC^Allyl ITC + Control^ControlDiet treatmentESO Actual mortality MI Corrected mortalityFigure 10. Percent mortality attributed to fumigationeffect of allyl isothiocyanate in assay using meridicdiet. Corrected mortality = Abbott's (1925) formula hasbeen applied.59No growth or mortality responses with dose were observedwith indole-3-carbinol, the degradation product of 3-indolylmethyl glucosinolate, which suggests that indolylalcohols are not biologically active at the concentrationstested.2.4 DiscussionThe feeding studies using meridic diets and variouscanola and mustard cultivars have clearly established thatsinigrin and its metabolite, allyl isothiocyanate, adverselyaffect the growth of neonate and fourth-instar M. configuratalarvae. These effects may be due in part to reduced rates offeeding on substrates that contain sinigrin. B. juncea, whichcontains very high levels of sinigrin, is relatively resistantto this polyphagous insect. This mustard was also found to beless preferred by the flea beetle, Phyllotreta striolata, thanB. oleracea, B. napus and B. rapa (Lamb and Palaniswamy,1990). Currently, B. juncea is being considered fordevelopment as an oilseed because of its superior agronomicperformance compared to the current canola species (Woods etal., 1991). The expression of high levels of sinigrin in thefoliage may be viewed as another desirable trait of thismustard species.The effects of thiocyanate-releasing glucosinolates arenot as well defined as the effects of isothiocyanate-releasingglucosinolates. Bioassay of indole-3-carbinol incorporated60into artificial diet suggests that this type of glucosinolatemetabolite is relatively innocuous to neonate M. configuratalarvae. The leaf disk choice tests suggest that hydroxybenzylglucosinolate (sinalbin) may be a feeding stimulant forneonate larvae. Yet, larval growth on intact S. alba plants,which produce predominately sinalbin, was relativelyinhibited. Sinalbin has previously been shown to be a factorof antixenotic and antibiotic resistance to Bertha armywormand flea beetles (Bodnaryk, 1991). It should be noted thatbenzyl glucosinolate occurs in significant amounts (0.7 - 3.9umol/g dry weight, depending on growth stage) in S. alba cv.Gisilba. Benzyl isothiocyanate has been shown to be toxic toEuropean corn borer and fall armyworm (Bartlet andMikolajczak, 1989). It is possible that benzyl glucosinolatemay be partially responsible for the growth inhibitory effectsobserved with M. configurata on intact plants. The feedingstimulant effect of S. alba to neonate larvae may be due tothe presence of relatively high glucoside concentrations withlower levels of associated isothiocyanates.Thiocyanate-releasing glucosinolates in foliage do notappear to stimulate consumption by fourth-instar M.configurata. Rather, fourth instar consumption rate wasreduced on stage 3 foliage of S. alba. Since there was nocorresponding reduction in larval growth rate attributable tothiocyanate-releasing glucosinolates, it is probable thatother factors also affect the consumption of foliage.61In the Brassicaceae, glucosinolates are the dominantgroup of secondary plant substances and should be given firstconsideration in questions of insect-plant interactions.However, substrate texture, nutrients, and other types ofallelochemicals can also affect the response of insects tothese plants. The data from this study suggests that plantgrowth stage may modify the relative expression ofallelochemical effects. This was further investigated (seeSection 3).Presently, rapeseed plant breeding efforts are directedtowards eliminating glucosinolates due to theirantinutritional effects on livestock which consume the seedmeal. Although the canola cultivars are nutritionallysuperior to the mustards in this regard, these data suggeststhat isothiocyanate-releasing glucosinolates in the foliagemay benefit the crop by providing a degree of protection frompolyphagous insects like the Bertha armyworm. To this end, auseful target for genetic engineering may be the expression ofisothiocyanate-releasing glucosinolates specifically in thefoliage coupled with the elimination of glucosinolates fromthe seed.623.0 Plant Growth Stage Effects on the Feeding and Growth ofthe Bertha Armyworm, Mamestra configurata in Canola andMustard Foliage3.1 IntroductionGlucosinolates have been studied extensively asallelochemicals mediating insect-plant interactions.Isothiocyanates, the most common fission products ofglucosinolates, are known to be insecticidal (see Section 1.2,Table 1). However, little is known of the biological activityof indolyl glucosinolates and their metabolites towardsinsects.Responses of insects to specific allelochemicals can varywidely and may be confounded by other plant factors, such asnutrient content (Campbell and Duffey, 1981), and the presenceof other allelochemicals (Koul et al., 1990). It appears thatthe relative response of the Bertha armyworm, Mamestraconfigurata to glucosinolate levels in the foliage of variousbrassicaceous host plants varies with plant growth stage(Section 2). However, the plants used in those experimentswere field-grown, with the different plant growth stagessampled and bioassayed at different dates during the growingseason. Consequently, the variation due to growth stageeffects could have been confounded by environmental effectssuch as climate and photoperiod. This study examinesdifferences between plant growth stages in relative resistance63to M. configurata and compares the differences toallelochemical content. Greenhouse-grown foliage of threeplant growth stages of two host plant species were examinedand compared: Brassica juncea (Commercial Brown mustard),which was found to be relatively resistant to M. configurata,and B. rapa cv. Tobin, which is relatively susceptible.Although glucosinolates are an obvious group ofallelochemicals to consider in these two plant species,phenolic constituents in the foliage were also quantified.Phenolic compounds have demonstrated biological activitytowards insects (Todd et al., 1971; Dreyer and Jones, 1981;Duffey and Isman, 1981; Isman and Duffey, 1982a; 1982b; Cole,1984; 1985; Lindroth and Peterson, 1988; Classen et al., 1990;Guerra et al., 1990).3.2 Materials and MethodsPlant growth. Brassica juncea and B. rapa cv. Tobin were sowndirectly into 0.141 m2 flats (6.5 cm deep) of a homogenoussoil mixture at a density of 48 plants per flat. The plantswere grown in the Department of Plant Science greenhouse undernatural light augmented with high-intensity sodium vapourlamps to provide a 16:8 (hr light-dark) photoperiod. Thelight intensity at plant level was variable but in the rangeof 16,146 lux. Water was supplied daily and water-soluble 20-20-20 (N-P-K) fertilizer was applied weekly. During thecourse of the plant growth for each replicate, one application64of the aphid-selective pesticide Pirimor ® (pirimicarb) wasnecessary to suppress infestations of green peach aphid, Myzuspersicae. Seeds were sequentially sown so that three growthstages - rosette (stage 2), stem elongation (stage 3) andflowering (stage 4) (Harper, 1973) - were availablesimultaneously for bioassay and phytochemical analysis.Foliage characteristic of each of these growth stages washarvested for insect bioassay and analysis for glucosinolatesand phenolic compounds. The youngest, fully-expanded leaf ofeach plant was sampled. Samples for phytochemical analyseswere quick-frozen in liquid N2 , lyophilized, ground in ablender to a maximum particle size of ca 1 mm and stored withdessicant at -20 °C. The experiment was conducted as tworeplicates separated by two months.Insect feeding assays. Individual fourth instar M.configurata larvae were confined for two days with an excisedleaf of B. juncea or B. rapa representing one of the threegrowth stages of interest. Relative consumption rate (RCRi)and relative growth rate (RGRi) were calculated as describedin Section 2.2. Ten insects were used for each of the sixcultivar-growth stage combinations within each replicate.ANOVA was performed on the main effects (cultivar and growthstage) and interaction terms. Tukey's studentized range testwas performed on RCRi and RGRi for each cultivar separately toexamine growth stage means.Phytochemical analysis. The lyophilized foliage was extracted65and analysed for glucosinolates by HPLC as described inSection 2.2 for individual glucosinolates. For phenoliccompounds, a 100 mg sample of the lyophilized foliage wasextracted three times in 3 ml of boiling 70% methanol for twominutes, with centrifugation between extractions. Thesupernatants were pooled and made up to 25 ml with deionizedwater.i) Total phenols. To a 0.5 ml sample of the diluted extractwas added 2.5 ml deionized water and 0.5 ml of Folin-Ciocaulteau reagent (Fisher Phenol reagent, diluted 3x to 1 NHC1), and after three minutes, 0.5 ml of 1.0 M Na 2CO3. Onehour later, the absorbance of the solution was read at 725 nmon a Pye Unicam PU 8620 UV/VIS/NIR spectrophotometer. Thelinear regression equation of a standard curve of sinapic acidwas used to quantify total phenols.ii) Total catechols. To a 1.0 ml sample of the dilutedextract was added 1.0 ml of deionized water, 1.0 ml of 4.6 NH2SO4 and 3 ml of 10% (w/v) aqueous ammonium molybdate. After10 minutes, the absorbance of the solution was read at 375 nm.The linear regression equation of a standard curve ofquercetin was used to quantify total catechols.Total nitrogen. Composite samples of lyophilized and powderedfoliage of field-grown rape and mustard cultivars of thedifferent growth stages (see Section 2) were analysed fortotal nitrogen content using the methods of Parkinson andAllen (1975). The samples were oven dried for three hours at6670 °C prior to digestion. Five ml of concentrated H 2SO4 wereadded to 1.0 gram samples of foliage, followed by 2 x 1 mlaliquots of a lithium sulfate-peroxide mixture (7.0 g Li 2SO4+ 0.21 g selenium powder + 175 ml 30% H 202). The samples wereheated discontinuously on a 360 °C heat block digester andcooled slightly before additional 2 x 1 ml aliquots of thelithium sulfate-peroxide mixture was added. The samples weredigested on the block for 1.5 hours. After this time, 0.5 mlof 30% H202 was added to the samples and these were thendigested for a further half hour on the block. This last stepwas repeated and the digests cooled to room temperature.Demineralized water was added to make up a final volume of 100ml. Aliquots were analysed on a Technicon Autoanalyser IIusing standards of (NH 4 ) 2SO4 for quantification (wavelength =660 nm).3.3 ResultsThe relative consumption rates (RCRi) for fourth instarM. configurata larvae on foliage were clearly affected by theplant growth stage, although these effects were expresseddifferently for the two plant species (Figure 11). Whilethere were no significant species differences in RCRi (Table9), there was a significant 'species x growth stage'interaction. Within B. juncea treatments, there were nosignificant growth stage differences in RCRi, as determined byTukey's studentized range test (Table 10).(2) (3) (4)^(2)^(3)Growth StageRCM (mg/mg/day)5-4B. Juncea^ B. rape13(4)67Figure 11. Mean relative consumption rate of fourthinstar M. configurata larvae on foliage of rosette (2),stem elongation (3) and flowering stages. Vertical linesabove bars represent SEM.68Table 9.^Analysis of Variance for Relative Consumption Rate(RCRi) of M. configurata on Foliage of B. juncea and B. rapeSource^ df^SS^MS^F^Pr>FTotal 119 273.34Replicates 1 0.89 0.89 0.49 0.4836Species 1 0.26 0.26 0.15 0.7039Stage 2 18.20 9.10 5.08 0.0078Replicate x^Species 1 4.70 4.70 2.62 0.1083Species x Stage 2 49.84 24.92 13.91 0.0000Replicate x Stage 2 2.36 1.18 0.66 0.5241Error 110 197.09 1.7969Table 10. Mean Relative Consumption (RCRi) and Growth (RGRi)Rates of Fourth Instar M. configurata on Foliage of B. junceaor B. rapa 1B. juncea^B. rapaRCRiRosette 2.304a 1.219aStem elongation 2.144a 1.621aFlowering 1.698a 3.585bRGRiRosette 0.209b 0.273aStem elongation 0.165b 0.235aFlowering 0.040a 0.161a1. Means within each species followed by the same letter arenot significantly different by Tukey's studentized range test(p = 0.05).70However, the RCRi for larvae fed B. rapa foliage increasedwith advancing plant growth stage, and was significantlydifferent with stage 4 foliage.With respect to relative growth rate (RGRi) (Figure 12),there were also highly significant differences among speciesand growth stages (Table 11). As expected, the RGRis oflarvae were significantly lower on the B. juncea than on theB. rapa (Figure 12). The RGRi for larvae fed B. junceadecreased with advancing plant growth stage. The differencewas significant between stage 3 and 4 foliage (Table 10).There were no significant differences in RGRi within the B.rapa treatments. There is the appearance of a decreasingtrend in relative growth rate (Figure 12). However, the ANOVA(Table 11) revealed a significant 'Replicate x Stage'interaction term, which is related to inconsistent growthstage trends between replicates in the RGRi for B. rapatreatments. The effects of growth stage in B. rapa onrelative growth rate are therefore inconclusive.The glucosinolate profiles for B. juncea and B. rapa(Figure 13) are compatible with data for field-grown foliage(Section 2, Figures 2 and 3). Quantifications of totalisothiocyanate-releasing and indolyl glucosinolates arepresented in Figures 14 and 15, respectively. Theglucosinolate content of B. juncea increases with advancingplant growth stage, especially among the isothiocyanate-releasing glucosinolates (Figure 14).(2) (3) (3) (4)(4)^(2)Growth StageRGRi (mg/mg/day)0.35 -B. rapa0.3 -0.25 -0.2 -0.15 -0.1 -0.06 -B. JunceaFigure 12. Mean relative growth rate of fourth instar M.configurata larvae on foliage of rosette (2), stemelongation (3) and flowering (4) stages. Vertical linesabove bars represent SEM.7172Table 11.^Analysis of Variance for Relative Growth Rate(RGRi) of M. configurata on Foliage of B. juncea and B. ragaSource df SS MS F Pr>FTotal 119 4.44Replicates 1 0.14 0.14 4.81 0.0304Species 1 0.22 0.22 7.36 0.0077Stage 2 0.42 0.21 7.03 0.0015Replicate x Species 1 0.02 0.02 0.63 0.4300Species x Stage 2 0.02 0.01 0.33 0.7235Replicate x Stage 2 0.38 0.19 6.37 0.0026Error 110 3.25 0.0373Brassica juncea132^ 85^10^15^20^25^30^35Retention Time (minutes)Brassica rapa 235^6 '^8■•••■ki■■•••••••■•••■ ^0^5^10^15^20^25^30^35Retention time (minutes)1. ally!^ 2. 4-0H-3-indolylmethyl3. ONPGaI (internal std.)^4. 4-pentenyl5. 3- indolylmethyl^6. 2-phenylethyl7. 4-methasy-3-indolylmethyl^8. 1-methoxy-3-indolylmethylFigure 13.^Representative HPLC chromatograms ofdesulfoglucosinolates for greenhouse-grown B. juncea(Commercial Brown mustard) and B. rapa cv. Tobin canola.A00cr.0B. juncea007426201510Ito Gal Conc. (umol/g dry wt.)B. raga50(2)^(3)^(4)^(2)^(3)^(4)Growth Stage^0  Block 1^^ Block 2 ED MeanFigure 14.^Quantities of isothiocyanate-releasingglucosinolates in foliage of rosette (2), stem elongation(3) and flowering (4) stages determined by HPLC.75Indolyl Gal Conc. (umol/g dry wt.)3.53 0B. raga2.52 0B. Juncea1.5 01o o0 00.5 0 000(2)^(3)^(4)^(2)^(3)^(4)Growth Stage^0  Block 1^0 Block 2 = MeanFigure 15.^Quantities of indolyl glucosinolates infoliage of rosette (2), stem elongation (3) and flowering(4) stages determined by HPLC.76In contrast, the glucosinolate content of B. rapa decreases,especially among the indolyl glucosinolates (Figure 15).Therefore, the declining growth of M. configurata on B. junceawith advancing plant growth stage may be related to increasingconcentrations of isothiocyanate-releasing glucosinolates. InB. rapa, isothiocyanate-releasing glucosinolates remain atrelatively low levels throughout the growth of the plant. Incontrast, larval consumption rate increases with advancingplant growth stage in B. rapa and decreasing levels of indolylglucosinolates. In the B. juncea treatments, the larvalconsumption rates are statistically similar (Table 10), inagreement with the relatively consistent levels of indolylglucosinolates in this species.The quantifications for total phenols and total catecholsare presented in Figure 16 and 17, respectively. B. junceafoliage of different plant growth stages does not differappreciably in total phenolic content (Figure 16), althoughthe concentration of catechols is slightly higher in stemelongation stage foliage than in stage 2 and 4 foliage (Figure17). These compounds appear relatively unimportant in therelationship of M. configurata to this mustard. In B. rapa,total phenols and total catechols tend to decrease withadvancing plant growth stage. This trend parallels theincreases in RCRi with advancing plant growth stage andsuggests that foliar phenolic compounds may have someinfluence on consumption by M. configurata.Phenol Conc. (umol/g dry wt.)7060 B. Juncea0B. raps0^o50 -9-- 0^o 00403020100(2)^(3)^(4)^(2)^(3)^(4)Growth Stage^o Block 1^^ Block 2 ED MeanFigure 16. Quantities of phenolic compounds in foliage ofrosette (2), stem elongation (3) and flowering (4) stagesdetermined spectrophotometrically.7778Catechol Conc. (umol/g dry IA)7060 B. luncea5040oao^ B. rapsoo0 003000020100(2)^(3)^(4)^(2)^(3)^(4)Growth Stage0 Block 1^^ Block 2^= MeanFigure 17. Quantities of catecholic compounds in foliageof rosette (2), stem elongation (3) and flowering (4)stages determined spectrophotometrically.79The levels of foliar nitrogen for the field-collectedbrassicaceous plants are shown in Figure 18. For mostcultivars, nitrogen levels decrease with advancing plantgrowth stage. Over all the cultivars analysed, there werehighly significant differences between growth stages (F[2,14]= 8.60; p = .00366). Rosette stage foliage was generallyhigher in total nitrogen than stage 3 and 4 plants (F[1,14] =10.29; p = .00632) and stage 3 foliage contained significantlyhigher levels of nitrogen than stage 4 foliage (F[1,14] =6.92; p = .01976). There is agreement between total nitrogencontent and the decreasing trend of RGRi with advancing plantgrowth stage seen in Section 2 with larvae fed field-grownfoliage. The parallel trend shown in the present greenhousestudy suggests a possible role for nitrogen in the response ofM. configurata to its host plants.Gisliba^Midas Regent WesterCultivarCandle TobinBrownLeth.22A80% Nitrogen (dry wt.)M Rosette (2) ED Stem elongation (3) MI Flowering (4)Figure 18.^Quantities of total nitrogen in field-collected foliage of brassicaceous plants.B. rapaB. napus_B. juncea S. alba10860813.4 DiscussionAlthough the Bertha armyworm is a polyphagous insect,brassicaceous plants are among its preferred hosts. Thisstudy demonstrates that plant growth stage affects the feedingand growth of Bertha armyworms. Foliar allelochemicalcontent, which changes during plant growth, influences theinsect's responses (Section 2). Isothiocyanate-releasingglucosinolates, which are known to be insect growthinhibitors, may partially account for plant growth stageeffects in B. juncea. These compounds are certainly importantwhen comparing interspecific host plant effects (Section 2).However, total nitrogen content is also likely to affectlarval growth rate. The importance of nitrogen to insectgrowth is well-documented (McNeill and Southwood, 1978;Mattson, 1980; Myers, 1985) and the reduction in totalnitrogen levels in more mature foliage compared to young,actively growing foliage is also well-known (Mattson, 1980;Bowers et al., 1991). With larvae fed B. rapa foliage, RGRidecreases in spite of increasing RCRi, which suggests that thefoliage becomes less nutritious with growth stage.The data also suggests that indolyl glucosinolates arebiologically active towards M. configurata. A significant,inverse relationship was found between relative consumptionrate and levels of indolyl glucosinolates in B. rapa.Although levels of total phpnnlc and 'atecholc ;gem alsoinversely related to consumption rate in the B. rapa82treatments, the magnitude of the differences in phenolsbetween stages is much less than that for the indolylglucosinolates. In Section 2, it is shown that indole-3-carbinol, a major metabolite of 3-indolylmethyl glucosinolate,does not inhibit larval growth of M. configurata. However, itis possible that larval consumption is influenced by intactindolyl glucosinolates.In the earlier experiments (Section 2) it was observedthat plant growth stage affected the magnitudes of bothnutritional indices and the relative importance ofglucosinolates in the relative resistance of the various hostplant cultivars. The data in this section demonstrates thatlevels of glucosinolates, phenolic compounds and nitrogenchange with plant growth stage. These differences may wellaccount for the previous observations. Althoughglucosinolates are the obvious chemical group to focus on whenstudying the Brassicaceae, it is clear that other intrinsicfactors influence insect responses and must be considered whenstudying phytochemical mechanisms of insect resistance.The significance of M. configurata's differentialresponses to plant growth stage may be appreciated in light ofthe insect's life cycle and mode of damage. In Canada, thisspecies is univoltine and overwinters as a pupa. As aconsequence, the damaging larval stages (instars 4-6) are notpresent until the crops have progressed to the stem elongationand flowering stages. These larvae tend to move up into the83canopy and graze on the developing siliques, causingshattering later as the crop dries (Burgess et al., 1979;Lamb, 1989). This study suggests that as the larvae grow, andtheir demand for nourishment increases, the host plant foliagebecomes less nutritious. The larvae are then compelled tomigrate to the protein-dense siliques.Investigations of insect-plant chemical interactions haveoften focussed on the role of allelochemicals as pivotalmediators. However, since basic nutrition is a primary needfor insects, it should never be ignored when consideringinsect-plant interactions. This experiment has underlined theroles of both glucosinolates and nitrogen in mediating larvalgrowth and feeding of the Bertha armyworm.844.0 Diamondback Moth, Plutella xylostella Larval Populationsand Parasitism Rates by Diadegma insularis Cress. on HostPlants with Different Glucosinolate Profiles4.1 IntroductionGlucosinolates and their breakdown products are importantkairomones for crucifer specialist insects. The action ofmyrosinase (E. C. 3 . 2 . 3 . 1) on unsubstituted alkenyl and aromaticglucosinolates normally results in the release ofisothiocyanates, which attract oligophagous crucifer insects(Feeny et al., 1970; Rygg and Somme, 1972; Nair and McEwen,1976; Finch, 1978; Pivnick et al., 1992). Glucosinolates arefeeding stimulants (Thorsteinson, 1953; Nayar andThorsteinson, 1963; David and Gardiner, 1966; Nault and Styer,1972; Hicks, 1974; Nielsen, 1978; 1989) and ovipositionstimulants (Traynier, 1965; 1984; Nair and McEwen, 1976;Renwick and Radke, 1983; Reed et al., 1989; Traynier andTruscott, 1991) for many pests of the Brassicaceae.The diamondback moth, Plutella xylostella, does notoverwinter in large numbers in Canada, but rather migrateseach summer from the south (Burgess et al., 1979). Hostselection must therefore depend on long-range attractioninitially. Although the moths are attracted by host plantvolatiles (Palaniswamy et al., 1986; Reed et al., 1989;Pivnick et al., 1990), recent work by Reed et al. (1989) hasdetermined that the oviposition response of diamondback moths85to glucosinolates is not significantly affected by the side-chain. While their study established a linear response todose with sinigrin, the major glucosinolate occurring in B.juncea, diamondback moth also infests canola species (B. napusand B. rapa), which do not produce sinigrin.The mutualistic relationships that exist between someplants and the natural enemies of herbivores which attack themare often neglected in plant resistance studies. Allylisothiocyanate, the metabolite of sinigrin (allylglucosinolate) has been shown to be a kairomone forDiaeretiella rapae, a parasitoid of the green peach aphid,Myzus persicae (Read et al., 1970). P. xylostella larvae arecommonly parasitised by the solitary endoparasitoid, Diadegmainsularis (Cress.) (Hymenoptera: Ichneumonidae). While hostplant nutritional quality has been shown to influence thelevel of parasitism by this wasp (Fox et al., 1990), noinformation exists on the importance of glucosinolates in hostselection.Large monocultures of canola and cultivated mustardsoffer essentially no choice to diamondback moths or theirnatural enemies for oviposition once the habitat has beenfound. This study was intended to determine if brassicaceaoushost plants having different foliar glucosinolate profiles(see Section 2) are equally infested by diamondback moth inthe field, and if parasitism by Diadegma insularis isinfluenced by the host plant. By using relatively small86plots, the experiment was designed to provide choice, whichcould increase the sensitivity to treatment effects.4.2 Materials and MethodsExperimental plots were grown at the U.B.C. Department ofPlant Science field station, Vancouver, B.C. Following soiltesting, granular fertilizers were incorporated at thefollowing rates: 58.88 kg/ha N as urea, 59.52 kg/ha K asmuriate of potash and 1.14 kg/ha Bo as borate. The selectivepre-emergent herbicide Treflan(p(trifluralin) was applied at2.0 1/ha prior to seeding.The experiment was conducted as a split-block design(Little and Hills, 1978) consisting of two blocks in locationand time. A third block was planted in late August but wasabandoned due to low diamondback moth populations andinfection by Beet Western Yellows virus. Each block consistedof six main plots of 'cultivar' and five sub-plots of`sampling date'. The following species/cultivars were chosenfor the study because of previously reported differences inglucosinolate profile and susceptibility to the Berthaarmyworm (Section 2): Sinapis alba cv. Gisilba, Brassicajuncea cv. Lethbridge 22A, B. napus cv. Midas (highglucosinolate) and Westar (low glucosinolate) and B. rapa cv.Candle and cv. Tobin (both low gluocosinolate, although Candleis somewhat more resistant than Tobin to M. configurata).Prior to the commencement of the experiment, assorted trap87plants consisting of B. oleracea (broccoli, Brussels sprouts,and cabbage), B. napobrassica (rutabaga) and B. chinesis (bokchoy) were used to attract a base population of P. xylostellato naturally infest the experimental plots. These weretransplanted as rosette stage plants along three sides of thefirst block at a distance of two meters.Main plots (six meters west to east, three meters northto south) were comprised of 31 three-meter rows sown 20 cmapart. Within each block, the main plots were oriented northto south such that each block was 6 meters west to east by 23meters north to south. Although a precision belt seeder (StanHay) was used for seeding, rows had to be thinned by hand toa uniform density of ca 30 plants per row due to sub-optimalgermination rates with the older cultivars. A one-meter stripof bare soil was left between each main plot and a two-meterstrip between the blocks. The first block was planted on June18, 1990. The second block was planted on July 12, 1990, twometers east of the first block. Locations for cultivars andsampling dates were selected using a random numbers table.The main plots were divided into subplots (strip plots) offive three-meter rows, bounded by a guard row on each side,for weekly destructive sampling. Plants at the ends of rowswere not sampled.Sampling began at three weeks post-seeding (rosettestage).^Each sample consisted of 100 individual leavesL.ullecLed randomly within the subplots. At six, seven and88eight weeks, 100 flower heads (top 12 cm) were also removedfrom randomly selected plants. The diamondback moth larvaewere counted and reared in the laboratory on foliage of thesame cultivar to either adulthood or the emergence ofparasitoids, which were also counted. In order to compensatefor the morphological differences between the plant species inthe study, numbers of larvae per 100 leaves or flower headswas converted to numbers of larvae per 100 grams of freshtissue, as the weights of plant samples were in the range of100 grams.Larval numbers per 100 grams were square-root transformed([x + 0.05]•) prior to ANOVA because the variances were foundto be proportional to the means. Leaf counts were analysedsepearately from flower head counts. In earlier treatment ofthe data, block effects on flower head counts were found to benot significant. Therefore, they are analysed here as sixcontinuous equally-spaced samples, even though the intervalbetween the two blocks is ten days rather than one week.Tukey's studentized range test was performed on cultivarmeans. In addition, individual degree of freedom tests(orthogonal contrasts) were performed on cultivar means oflarval data to address the following questions:B. juncea versus others: Do high levels of isothiocyanate-releasing glucosinolates, particularly sinigrin, affectinfestation levels of P. xylostella?S. alba versus B. napus + B. rapa: Are there differences89between white mustard (with high levels of sinalbin) and therape species (lacking sinalbin)?Midas versus Westar, Candle + Tobin: Are there differencesbetween high and low glucosinolate rape cultivars?Westar versus Candle + Tobin: Of the low glucosinolatecultivars, is there a difference between B. napus and B. rape?Candle versus Tobin: Is there a difference between these twoB. rapa cultivars?Percent parasitism in each sample was linearly regressedagainst both larval counts and counts per 100 g fresh weightof plant tissue in order to determine if host densityinfluences parasitism rate. One-way ANOVA was performed onarcsine-transformed values of percent parasitism (sine -1 [%parasitism /100) 0.5) for cultivar effects. Samples with larvalcounts of zero were omitted from the analysis of % parasitism.Leaf and flower head data were analysed separately. Tukey'sstudentized range test was performed on cultivar means.4.3 ResultsThe second generation of P. xylostella larvae was sampledover both blocks of the experiment (Figure 19). However, thepopulation maximum occurred during the sampling of the secondblock, as indicated by greater numbers of larvae found in bothleaf and flower head samples. Block effects for numbers of P.xylostella on foliage were highly significant (Table 12).eultivar means for leaf and flower head counts are35 Mean No. Larvae/100 g fr. wt.3025201510Block 2Block 10Jul 9^Jul 16 Jul 23 Jul 30 Aug 6 Aug 13 Aug 20 Aug 27Sample Date—0— Leaf Samples --e-- Flower HeadsFigure 19. P. xylostella population densities on foliageand flower heads over the course of the sampling study.9091Table 12. Analysis of Variance for P. xylostella Counts per100 g Fresh Weight of FoliageSource of Variation df SS MS F Pr>FSubplots 59 137.77 2.34Main plots 11 30.10 2.74Blocks 1 21.18 21.18 55.51 0.0007Cultivars 5 7.01 1.40 3.67 0.0898L vs G+M+W+C+T 1 1 0.52 0.52 1.37 0.2950G vs M+W+C+T 1 4.17 4.17 10.94 0.0213M vs W+C+T 1 0.00 0.00 0.00 1.0000W vs C+T 1 1.04 1.04 2.74 0.1590C vs T 1 1.27 1.27 3.32 0.0128Main plot error(A) 5 1.91 0.38Sample date 4 48.90 12.23 1.49 0.3551Linear trend 1 27.42 27.42 3.33 0.1419Quadratic 1 19.43 19.43 2.36 0.1990Cubic 1 1.05 1.05 0.13 0.7393Deviations 1 1.00 1.00 0.12 0.7445Cultivar x Sampledate Interaction20 15.52 0.78 1.50 0.1871Strip plot error(B) 4 32.89 8.22Subplot error(C) 20 10.36 0.521. L = B. juncea cv. Lethbridge 22A, G = S. alba cv. Gisilba,M = B. napus cv. Midas, W = B. napus cv. Westar,C = B. rape cv. Candle and T = B. rapa cv. Tobin.92presented in Table 13. There were no significant cultivardifferences for leaf counts (Table 12). However, there wasstatistical significance in one of the contrasts: S. albaversus B. napus + B. rapa, which suggests that S. alba may bea superior host plant for P. xylostella compared to the othersin the study. The contrast of B. juncea versus S. alba + B.napus + B. rapa was not statistically significant. Therefore,high foliar levels of isothiocyanate-releasing glucosinolates,notably sinigrin, apparently had little influence on P.xylostella populations in the field. The lack of statisticalsignificance in the contrast of the high glucosinolate rapecultivar Midas versus the canola cultivars Westar, Candle andTobin also supports the argument that foliar glucosinolatelevels are not important.For flower head counts, there were no significantcultivar differences (Table 14). The orthogonal contrastsrevealed similar trends to the leaf counts, in that only thecontrast of S. alba versus B. napus + B. rapa approachedsignificance (p = .0946).The effects of sampling date on P. xylostella counts werefound to be not statistically significant in leaf samples, butapproached significance in the flower head samples. A lineartrend accounted for most of the variation that was found inthe foliar samples (Table 12), while a quadratic trendaccounted for most of the variation in the flower head samples(Table 14).93Table 13. P. xylostella Population Densities (Larvae per 100g plant tissue) on Foliage and Flower Heads of the DifferentCultivars. 2Cultivar Foliage Flower HeadsB. juncea cv. Lethbridge 22A 9.92 6.83S. alba cv. Gisilba 12.14 11.02B. napus cv. Midas 7.58 4.69B. napus cv. Westar 6.08 8.04B. rapa cv. Candle 6.84 6.84B. rapa cv. Tobin 9.83 7.531. Data are presented as weighted means of square root-transformed counts.2. No significant cultivar effects were demonstrated byTukey's studentized range test.Table 14.^Analysis of Variance for100 g Fresh Weight of Flower HeadsSource of Variation^df^SSP. xylostellaMS^F94Counts perPr>FCultivars 5 3.94 0.79 0.99 0.4431L vs G+M+W+C+T 1 1 0.07 0.07 0.09 0.7642G vs M+W+C+T 1 2.40 2.40 3.02 0.0946M vs W+C+T 1 1.33 1.33 1.67 0.2081W vs C+T 1 0.09 0.09 0.11 0.7382C vs T 1 0.05 0.05 0.06 0.8102Sample date 5 10.22 2.04 2.57 0.0523Linear trend 1 0.72 0.72 0.90 0.3520Quadratic 1 7.65 7.65 9.62 0.0047Cubic 1 1.42 1.42 1.79 0.1930Deviations 1 0.002 0.002 0.00 0.9650Cultivar x Sample 25 19.88 0.80date (Error)1. Cultivar abbreviations as in Table 12.95Parasitism by D. insularis was highly variable and foundto be independent of host density by simple linear regression(Figures 20 and 21), in agreement with the findings of Fox etal. (1990). There were no significant cultivar differences infoliar samples (F[5,48] = 0.35; p = 0.8791) (Figure 22).However, there were significant cultivar effects in the flowerhead samples (F[5,29] = 3.76; p = .0096) (Figure 23): larvaereared from the B. juncea cv. Lethbridge 22A samples wereparasitised at the highest rate, while those from S. alba cv.Gisilba were parasitised the least.120 % Parasitism0^ 0r ■ -.097, p ■ .367O ^ 0^ 0^ ^^ Booo^0 0 ^ ^ooo 0 0O 08B0 ^00O D ^O ®oo010^20^30^40No. of P. xylostella larvae/sampleFigure 20.^Scatter plot of percent parasitism by D.insularis versus number of P. xylostella larvae persample, with linear regression line.96100 ^00080604020El]0% Parasitism0 0^ ^^ ^ r • -.106, p • .324^ 0 ^®®O ^ 00 0 0^^1:10 43^s^ CO1:1^00010^20^30^40^50No. of P. xylostella larvae/100 g fr.wt.Figure 21.^Scatter plot of percent parasitism by D.insularis versus number of P. xylostella larvae per 100grams fresh weight plant tissue, with linear regressionline.9712010080604020% Parasitism (weighted mean)70 -aLeth.22A^Gisilba^Midas^Westar^Candle^TobinCultivarFigure 22. Percent parasitism of P. xylostella larvae byD. insularis on foliage. There was no significantdifference among cultivars.9880504030201070 % Parasitism (weighted mean)-99b605040302010Leth.22A Gisilba^Midas^Wester^Candle^TobinCultivarFigure 23. Percent parasitism of P. xylostella larvae byD. insularis on flower heads. Bars labelled with the sameletter are not significantly different from each other byTukey's studentized range test.1004.4 DiscussionThis field study has shown that in a choice situation, P.xylostella populations are not significantly influenced by thehost plant's glucosinolate profile. The high foliar levels ofsinigrin in B. juncea do not appear to make this plant moreattractive than the other species. This agrees with thefindings of Reed et al. (1989), who determined by laboratorychoice tests that the diamondback moth oviposition response isnot affected by the glucosinolate side chain, or concentrationat realistic levels in the plant.The comparatively high larval density we observed on S.alba is similar to that reported by Palaniswamy et al. (1986).In their laboratory study, S. alba (formerly B. hirta) cv.Ochre plant extracts (volatiles only) were significantly moreattractive to diamondback moths and elicited higheroviposition responses than extracts of B. napus cv. Regent andfaba bean (Vicia faba) in choice tests. Our results, however,appear to contradict those reported by Reed et al. (1989) withrespect to S. alba. While their study presents convincingevidence of a non-anionic oviposition inhibitor to diamondbackmoth in aqueous S. alba cv. Ochre extracts, it appears fromour data that the biological activity of the compound may notbe expressed in intact S. alba cv. Gisilba plant tissue.Aqueous extracts of cabbage and other brassicaceous hostplants have previously been shown to possess oviposition-deterrent properties not evident in the intact plants toward101the cabbage looper, Trichoplusia ni (Renwick and Radke, 1981).Host plant glucosinolate profile may play a moreimportant role for the parasitoid D. insularis than for P.xylostella. Although there were no significant differences inpercent parasitism due to cultivar in the foliar samples, theflower head samples yielded higher rates of parasitism on B.juncea than on the other plants in the study. Since thismustard possesses the highest level of isothiocyanate-releasing glucosinolates (predominantly sinigrin) of all theplants in the study, it appears that sinigrin and/or itsvolatile metabolite, allyl isothiocyanate, may be important inthe host selection sequence of D. insularis. The fact thatthese host plant effects on D. insularis are only evident inthe flower head samples, and not the foliar samples, maysimply be an indication of the parasitoid's attraction toflowers as a source of nectar once inside the host habitat.This study of P. xylostella and D. insularis providesadditional evidence that the quality and quantity ofglucosinolates is less important to oligophagous cruciferinsects than previously regarded. However, parasitoidsspecializing on these herbivores may be more sensitive toallelochemicals as host selection kairomones. Considering thelevels of natural parasitism already occurring, theconservation and even promotion of indigenous biologicalcontrol agents should be considered in plant breeding efforts.1025.0 Summary and ConclusionsThe results of the experiments contained in this thesiscan be summarized by the following interpretive points:1) B. juncea foliage is more resistant to neonate and fourthinstar Bertha armyworm, M. configurata larvae than S. alba andthe canola species, B. napus and B. rapa. The antixenoticeffect of B. juncea towards neonate larvae was demonstrated bycomparatively low numbers of larvae on this species in amultiple choice leaf disk assay. A growth inhibitory effecttowards neonate larvae was demonstrated by reduced weight gainin larvae reared on intact plants.^These effects wereconsistent with the effects on fourth instar larvae asmeasured by nutrition indices: relative consumption andgrowth rates were reduced in larvae fed B. juncea foliagecompared to S. alba, B. napus and B. rapa.2) The biological parameters measured in M. configurata werenot correlated to levels of total glucosinolates in mostcases. However, significant negative linear correlations didoccur between neonate attraction and fourth instar relativegrowth rate and the quantity of isothiocyanate-releasingglucosinolates.^The adverse effects of isothiocyanatestowards M. configurata were corroborated by feeding studiesusing pure compounds added to meridic diets.^Indole-3-carbinol, the expected metabolite of 3-indolylmethylglucosinolate, was not biologically antivP, cbquimolar concentrations of sinigrin and allyl isothiocyanate inhibited103larval growth.3) Plant growth stage affects the response of fourth instarM. configurata to host plant foliage. Relative consumptionrates of larvae fed B. rapa cv. Tobin foliage increased withadvancing plant growth stage (rosette, to stem elongation, toflowering stages), while the concentration of indolylglucosinolates correspondingly decreased with plant growthstage.^Relative growth rates of larvae fed B. juncea(commercial brown mustard) foliage decreased with advancingplant growth stage, while the concentration of isothiocyanate-releasing glucosinolates correspondingly increased.^Theconcentration of foliar phenolic compounds do not appear toplay an important role in plant growth stage effects. Totalfoliar nitrogen, which sequentially decreases from rosette tostem elongation to flowering stage may be a more significantfactor.4) While glucosinolates may be necessary kairomones for thediamondback moth, P. xylostella, the glucosinolate profile ofthe host foliage appears to be of little importance to naturalinfestation of this insect in the field. However, S. albaplots supported larger populations of larvae than plots of B.juncea, B. napus and B. rapa cultivars.^In flower headsamples, higher levels of larvae were parasitised by theichneumonid D. insularis on B. juncea than on the otherplant species.104The results of the feeding studies with M. configuratasuggest that glucosinolates, particularly the type that yieldstable isothiocyanates, may be useful plant defense compoundsagainst polyphagous insects that invade brassicaceous crops,such as the Bertha armyworm. On the other hand, it has beenargued that plants containing high levels of glucosinolatesmay be more attractive to Brassica specialist insects thanplants with low levels of glucosinolates. The field study ofP. xylostella has shown that this is not the case for thisinsect. The glucosinolate profile of the host plant is lessimportant in host plant suitability for this herbivore thanwould be expected. There is also evidence that sinigrin asthe dominant glucosinolate in B. juncea may play a beneficialrole as a host selection kairomone for the parasitoid D.insularis.Hypothetically, the "ideal" Canola would be a modifiedversion of B. juncea: the storage of glucosinolates in theseed would need to be blocked while maintaining high levels ofisothiocyanate-releasing glucosinolates in the foliage. Thiswould result in a "canola" plant which is fairly resistant topolyphagous insects and attractive to the natural enemies ofBrassica specialists.Apart from glucosinolates, many brassicaceous plantspossess other herbivore and fungal defensive characteristics,such as phytoalexins and protease inhibitors. The inductionof insect resistance by mechanical damage (Bodnaryk, 1992) is105one example of host plant resistance potential yet to beexploited. Uncultivated Brassica plants also possess usefulresistance factors yet to be identified. Since Brassica cropswill likely be grown as monocultures for many years, therewill doubtless be no shortage of insects to try to manage,mandating on-going research on the insect-plant chemicalinteractions in the Brassicaceae.106BibliographyAbbott, W.S. 1925. A method of computing the effectivenessof an insecticide. J. Econ. Entomol. 18:265-267.Ahman, I.^1982.^A comparison between high and lowglucosinolate cultivars of summer oilseed rape (Brassica napusL.) with regard to their levels of infestation by the Brassicapod midge (Dasineura brassicae Winn.). Z. ang. Ent. 94:103-109.Ahman, I. 1985. Oviposition behavior of Dasineura brassicaeon a high- versus low-quality Brassica host. Entomol. exp.appl. 39:247-253.Ahman, I. 1986. Toxicities of host secondary compounds toeggs of the Brassica specialist Dasineura brassicae. J. Chem.Ecol. 12:1481-1488.Arthur, A.P. and P.G. Mason. 1985. The life history andimmature stages of Banchus flavescens (Hymenoptera:Ichneumonidae) a parasitoid of the Bertha armyworm, Mamestraconfigurata (Lepidoptera: Noctuidae) in western Canada. Can.Ent. 117:1249-1255.Arthur, A.P. and P.G. Mason. 1986. Life history and immaturestages of the parasitoid Microplitis mediator (Hymenoptera:Braconidae), reared on the Bertha armyworm Mamestraconfigurata (Lepidoptera: Noctuidae). Can. Ent. 118:487-491.Baldwin, W.S. and G.A. LeBlanc. 1992. The anti-carcinogenicplant compound indole-3-carbinol differentially modulatesP450-mediated steroid hydroxylase activities in mice. Chem.-Biol. Interact. 83:155-169.Bartelt, R.J. and K.L. Mikolajczak.^1989.^Toxicity ofcompounds derived from Limnanthes alba seed to fall armyworm(Lepidoptera: Noctuidae) and European corn borer (Lepidoptera:Pyralidae) larvae. J. Econ. Entomol. 82:1054-1060.Benn, M.H. 1977. Glucosinolates. Pure Appl. Chem. 49:197-210.Bjerg, B. and H. Sorensen. 1987. Quantitative analysis ofglucosinolates in oilseed rape based on HPLC ofdesulfoglucosinolates and HPLC of intact glucosinolates, pp.125-150, in J.-P. Wathelet (ed.). Glucosinolates in Rapeseeds:Analytical Aspects. Martinus Nijhoff, Dordrecht.107Blau, P.A., P. Feeny, L. Contardo and D.S. Robinson. 1978.Allylglucosinolate and herbivorous caterpillars: A contrast intoxicity and tolerance. Science 200:1296-1298.Bodnaryk, R.P. 1991. Developmental profile of sinalbin (p-hydroxybenzyl glucosinolate) in mustard seedlings, Sinapisalba L., and its relationship to insect resistance. J. Chem.Ecol. 17:1543-1556.Bodnaryk, R.P. 1992. Effects of wounding on glucosinolatesin the cotyledons of oilseed rape and mustard. Phytochemistry31:2671-2677.Bodnaryk, R.P. and P. Palaniswamy. 1990. Glucosinolatelevels in cotyledons of mustard Brassica juncea L. and rape,Brassica napus L. do not determine feeding rates of fleabeetle, Phyllotreta cruciferae (Goeze). J. Chem. Ecol.16:2735-2746.Bones, A. and T.H. Iversen. 1985. Myrosin cells andmyrosinase. Israel J. Bot. 34:351-375.Bowers, M.D., N.E. Stamp and E.D. Fajer. 1991. Factorsaffecting calculation of nutritional indices for foliage-fedinsects: An experimental approach. Entomol. exp. appl.61:101-116.Brzezinski, W. and P. Mendelewski. 1984. Determination oftotal glucosinolate content in rapeseed meal with thymolreagent. Z. Pflanzenzuchtg. 93:177-183.Bucher, G.E. and G.K. Bracken. 1976. The Bertha armyworm,Mamestraconfigurata (Lepidoptera: Noctuidae). Artificial dietand rearing technique. Can. Ent. 108:1327-1338.Buchner, R. 1987. Comparison of procedures for optimumextraction, purification and analysis of desulfo indolylglucosinolates. pp. 76-89, in J.-P. Wathelet (ed.).Glucosinolates in Rapeseeds: Analytical Aspects. MartinusNijhoff, Dordrecht.Burgess, L., J. Dueck, G.A. Petrie and L.G. Putnam. 1979.Insect pests and diseases of rape and mustard. Publ. No. 48,The Rapeseed Association of Canada. 32 pp.Butts, R.A. and R.J. Lamb. 1990. Comparison of oilseedBrassica crops with high or low levels of glucosinolates andalfalfa as hosts for three species of Lygus (Hemiptera:Heteroptera: Miridae). J. Econ. Entomol. 83:2258-2262.Campbell, B.C. and S.S. Duffey.tomatine-induced toxicity toexiguae, by phytosterols in thezea. J. Chem. Ecol. 7:927-945.1081981. Alleviation of oc-the parasitoid, Hyposoterdiet of the host, HeliothisChew, F.S. 1988. Biological effects of glucosinolates. pp.155-181, in H.G. Cutler (ed.). Biologically Active NaturalProducts: Potential Use in Agriculture. American ChemicalSociety, Washington.Classen, D., J.T. Arnason, J.A. Serratos, J.D.H. Lambert, C.Nozzolillo and B.J.R. Philogene. Correlation of phenolic acidcontent of maize to resistance to Sitophilus zeamais, theMaize Weevil in CIMMYT's collections. J. Chem. Ecol. 16:301-315.Cole, R.A.^1984.^Phenolic acids associated with theresistance of lettuce cultivars to the lettuce root aphid.Ann. Appl. Biol. 105:129-145.Cole, R.A. 1985. Relationship between the concentration ofchlorogenic acid in carrot roots and the incidence of carrotfly larval damage. Ann. Appl. Biol. 106:211-217.Daun, J.K. 1986a. Erucic acid levels in Western CanadianCanola and rapeseed . J. Amer. Oil Chem. Soc. 63:321-324.Daun, J.K. 1986b. Glucosinolate levels in Western Canadianrapeseed and Canola . J. Amer. Oil Chem. Soc. 63:639-643.David, W.A.L. and B.O.C. Gardiner.^1966.^Mustard oilglucosides as feeding stimulants for Pieris brassicae larvaein a semi-synthetic diet. Entomol. exp. appl. 9:247-255.Dreyer, D.L. and K.C. Jones. 1981. Feeding deterrency offlavonoids and related phenolics towards Schizaphis graminumand Myzus persicae: Aphid feeding deterrents in wheat.Phytochemistry 20:2489-2493.Duffey, S.S. and M.B. Isman. 1981. Inhibition of insectlarval growth by phenolics in glandular trichomes of tomatoleaves. Experientia 37:574-576.Ellis, P.R., R.A. Cole, P. Crisp and J.A. Hardman. 1980. Therelationship between cabbage root fly egg laying and volatilehydrolysis products of radish. Ann. Appl. Biol. 95:283-289.Erickson, J.M. and P. Feeny. 1974. Sinigrin: A chemicalbarrier to the black swallowtail butterfly, Papilio polyxenes(Lepidoptera: Papilionidae). Ecology 55:103-111. 109Farrar, R., J. Barbour and G.G. Kennedy. 1989. Quantifyingfood consumption in insects. Ann. Entomol. Soc. Am. 82:593-598.Feeny, P., K.L. Paauwe and N.J. Demong. 1970. Flea beetlesand mustard oils: Host plant specificity of Phyllotretacruciferae and P. striolata adults (Coleoptera:Chrysomelidae). Ann. Entomol. Soc. Amer. 63:832-841.Fenwick, G.R. and R.F. Curtis. 1980. Rapeseed meal and itsuse in poultry diets. A review. Anim. Feed Sci. Technol.5:255-298.Fenwick, G.R. and R.K. Heaney. 1983. Glucosinolates andtheir breakdown products in cruciferous crops, foods andfeedingstuffs. Food Chemistry 11:249-271.Fenwick, G.R., R.K. Heaney and W.J. Mullin.^1983.Glucosinolates and their breakdown products in foods and foodplants. CRC Critical Reviews in Food Science and Nutrition18:123-201.Finch, S. 1978. Volatile plant chemicals and their effect onhost plant finding by the cabbage root fly (Delia brassicae).Entomol. exp. appl. 24:350-359.Finch, S. and G. Skinner. 1982. Trapping cabbage root fly intraps baited with plant extracts and with natural andsynthetic isothiocyanates. Entomol. exp. appl. 31:133-139.Fox, L.R., D.K. Letourneau, J. Eisenbach and S. Van Nouhuys.1990. Parasitism rates and sex ratios of a parasitoid wasp:effects of herbivore and plant quality. Oecologia 83:414-419.Gerber, G.H. and A.A. Obadofin. 1981. Growth, development,and survival of the larvae of the red turnip beetle,Entomoscelis americana (Coleoptera: Chrysomelidae), onBrassica campestris and B. napus (Cruciferae). Can. Ent.113:395-406.Grob, K. and P. Matile.^1979.^Vacuolar location ofglucosinolates in horseradish root cells. Plant Sci. Lett.14:327-335.Guerra, D.J., J.T. Cothren and J.R. Phillips.^1990.Influence of selected phenolic compounds on development ofbollworm (Lepidoptera: Noctuidae) larvae. J. Econ. Entomol.83:2115-2118.Harper, F.R. 1973. A key to standa dir- Ze the aPSnriptinn ofgrowth stages in turnip rape, Brassica campestris. Can. PlantDis. Surv. 53:93-95.110Harris, J.L. 1985. Field crop insect control. Co-operativeagricultural extension service publication. SaskatchewanAgriculture, Regina.Hicks, K.L. 1974. Mustard oil glucosides: Feedingstimulants for adult cabbage flea beetle, Phyllotretacruciferae (Coleoptera: Chrysomelidae). Ann. Entomol. Soc.Amer. 67:261-264.Hogge, L.R., D.W. Reed and E.W. Underhill. 1988a. Theidentification of desulfoglucosinolates using thermosprayliquid chromatography/mass spectrometry. J. Chromatogr. Sci.26:348-351.Hogge, L.R., D.W. Reed, E.W. Underhill and G.W. Haughn.1988b. HPLC separation of glucosinolates from leaves andseeds of Arabidopsis thaliana and their identification usingthermospray liquid chromatography/mass spectrometry. J.Chromatogr. Sci. 26:551-556.Isman, M.B. and S.S. Duffey. 1982a. Phenolic compounds infoliage of commercial tomato cultivars as growth inhibitors tothe fruitworm, Heliothis zea. J. Amer. Soc. Hortic. Sci.107:167-170.Isman, M.B. and S.S. Duffey. 1982b. Toxicity of tomatophenolic compounds to the fruitworm, Heliothis zea. Entomol.exp. appl. 31:370-376.Josefsson, E. and L.-A. Appelqvist. 1968. Glucosinolates inseed of rape and turnip rape as affected by variety andenvironment. J. Sci. Food Agric. 19:564-570.Jurges, V.K. 1978. Glucosinolates in the growing andflowering plants of Brassica napus and Brassica campestris.Proceedings 5th International Rapeseed Conference. Malmo ,Sweden. Vol. 2, pp. 57-60.Kjaer, A. 1976. Glucosinolates in the Cruciferae. pp. 207-220, in J.G. Vaughan, A.J. MacLeod and B.M.G. Jones (eds.).The Biology and Chemistry of the Cruciferae. Academic Press,London.Koul, 0., M.J. Smirle and M.B. Isman. 1990. Synergism of anatural insect growth inhibitor is mediated by bioactivation.Experientia 46:1082-1084.Labana, K.S., K.L. Ahuja, M.L. Gupta and K.S. Brar. 1983.Preliminary studies on the chemical basis of resistance inBrassica species to mustard aphid (Lipaphis erysimi). Proceedings 6th Congres International sur le Colza(International Rapeseed Conference). Vol. 2, pp. 1132-1134.111Lamb, R.J.^1988.^Susceptibiltiy of low- and high-glucosinolate oilseed rapes to damage by flea beetles,Phyllotreta spp. (Coleoptera: Chrysomelidae). Can. Ent.120:195-196.Lamb, R.J.^1989. Entomology of oilseed Brassica crops.Annu. Rev. Entomol. 34:211-229.Lamb, R.J. and P. Palaniswamy. 1990. Host descrimination bya crucifer-feeding flea beetle, Phyllotreta striolata (F.)(Coleoptera: Chrysomelidae). Can. Ent. 122:817-824.Lamb, R.J., W.J. Turnock and H.N. Hayhoe.^1985.^Wintersurvival and outbreaks of Bertha armyworm, Mamestraconfigurata (Lepidoptera: Noctuidae), on canola. Can. Ent.117:727-736.Langer, P. 1966. Antithyroid action in rats of small dosesof some naturally occurring compounds. Endocrinology 79:1117-1122.Larsen, L.M., J.K. Nielsen, 0. Olsen, L. Pedersen and H.Sorensen.^1983.^Studies of insects associated withcrucifers: Host plant selection as a function ofallelochemicals present in the plants. Problems related torape containing high and low concentrations of glucosinolates.Proceedings 6th Congres International sur le Colza(International Rapeseed Conference). Vol. 1, pp. 270-275.Lerin, J. 1983. Selection varietale du colza et problemesentomologiques. Proceedings 6th Congres International sur leColza (International Rapeseed Conference). Vol. 2, pp. 1111-1113.Lichtenstein, E.P., F.M. Strong and D.G. Morgan.^1962.Identification of 2-phenylethylisothiocyanate^as aninsecticide occurring naturally in the edible part of turnips.J. Agric. Food Chem. 10:30-33.Lichtenstein, E.P., D.G. Morgan and C.H. Mueller.^1964.Naturally occurring insecticides in cruciferous crops. J.Agric. Food Chem. 12:158-161.Lindroth, R.L. and S.S. Peterson. 1988. Effects of plantphenols on performance of southern armyworm larvae. Oecologia75:185-189.Little, T.M. and F.J. Hills.^1978.^AgriculturalExperimentation Design and Analysis. John Wiley and Sons, NewYork. xiii + 350 pages. 112Loschiavo, S.R. and R.J. Lamb. 1985. Food preferences,survival, and development of four stored-product pests(Coleoptera) on rapeseed and canola (Brassica spp.). Can.Ent. 117:575-580.Lowe, M.D., R.F. Henzell and H.J. Taylor. 1971. Insecticidalactivity to soldier fly larvae, Inopus rubriceps (Macq.) ofisothiocyanates occurring in "choumoellier" (Brassicaoleraceae cv.). N.Z. J. Sci. 14:322-326.Luthy, B. and P.H. Matile. 1984. The mustard oil bomb:Rectified analysis of the subcellular organization of themyrosinase system. Biochem. Physiol. Pflanzen 179:5-12.Macfarlane Smith, W.H. and D.W. Griffiths. 1988. A time-course study of glucosinolates in the ontogeny of forage rape(Brassica napus L.). J. Sci. Food Agric. 43:121-134.MacGibbon, D.B. and R.M. Allison. 1968. A glucosinolasesystem in the aphid Brevicoryne brassicae. N.Z. J. Sci.11:440-446.MacLeod, A.J. and J.T. Rossiter. 1985. The occurrence andactivity of epithiospecifier protein in some Cruciferae seeds.Phytochemistry 24:1895-1898.Malik, R.S., I.J. Anand and S. Srinivasachar. 1983. Effectof glucosinolates in relation to aphid (Lipephis erysimiKalt.) fecundity in crucifers. Int. J. Tropical Agriculture1:273-278.Matile, P. 1980. "Die Senfolbombe": Zur Kompartimentierungdes Myrosinasesystems. Biochem. Physiol. Pflanzen 175:722-731.Mattson, W.J. 1980. Herbivory in relation to plant nitrogencontent. Ann. Rev. Ecol. Syst. 11:119-161.McGregor, D.I. 1985. Determination of glucosinolates inBrassica seed. Eucarpia Cruciferae Newsl. 10:132-136.McGregor, D.I. 1988. Glucosinolate content of developingrapeseed (Brassica napus L. 'Midas') seedlings. Can. J. PlantSci. 68:367-380.McNeill, S. and T.R.E. Southwood. 1978. The role of nitrogenin the development of insect/plant relationships. pp. 77-98,in J.B. Harbourne (ed.). Biochemical Aspects of Plant andAnimal Coevolution. Academic Press, London.113Muuse, B.G. and H.J. Van der Kamp. 1987. Determination ofglucosinolates in crucifer vegetables for human consumptionand rapeseed. pp. 38-49, in J.-P. Wathelet (ed.).Glucosinolates in Rapeseeds: Analytical Aspects. MartinusNijhoff, Dordrecht.Myers, J.H. 1985. Effect of physiological condition of thehost plant on the ovipositional choice of the cabbage whitebutterfly, Pieris rapae. J. Animal Ecol. 54:193-204.Nair, K.S.S. and F.L. McEwen. 1976. Host selection by theadult cabbage maggot Hylemya brassicae (Diptera:Anthomyiidae): Effect of glucosinolates and common nutrientson oviposition. Can. Ent. 108:1021-1030.Nair, K.S.S., F.L. McEwen and V. Snieckus.^1976.^Therelationship between glucosinolate content of cruciferousplants and oviposition preferences of Hylemya brassicae(Diptera: Anthomyiidae). Can. Ent. 108:1031-1036.Nault, L.R. and W.E. Styer. 1972. Effects of sinigrin onhost selection by aphids. Entomol. exp. appl. 15:423-431.Nayar, J.K. and A.J. Thorsteinson.^1963.^Furtherinvestigations into the chemical basis of insect-host plantrelationships in an oligophagous insect, Plutella maculipennis(Curtis) (Lepidoptera: Plutellidae). Can. J. Zool. 41:923-929.Nielsen, J.K.^1978.^Host plant discrimination withinCruciferae: Feeding responses of four leaf beetles(Coleoptera: Chrysomelidae) to glucosinolates, curcurbitacinsand cardenolides. Entomol. exp. appl. 24:41-54.Nielsen, J.K.^1989.^The effect of glucosinolates onresponses of young Phyllotreta nemorum larvae to non-hostplants. Entomol. exp. appl. 51:249-259.Nielsen, J.K., L.M. Larsen and H. Sorensen. 1979. Host plantselection of the horse radish flea beetle Phyllotretaamoraciae (Coleoptera: Chrysomelidae): Identification of twoflavonol glycosides stimulating feeding in combination withglucosinolates. Entomol. exp. appl. 26:40-48.Nielsen, J.K., A.H. Kirkeby-Thomsen and M.K. Petersen. 1989.Host plant recognition in monophagous weevils: specificity infeeding responses of Ceutorhynchus constrictus and thevariable effect of sinigrin. Entomol. exp. appl. 53:157-166.114Palaniswamy, P., C. Gillott and G.P. Slater. 1986.Attraction of diamondback moths, Plutella xylostella (L.)(Lepidoptera: Plutellidae), by volatile compounds of canola,white mustard, and faba bean. Can. Ent. 118:1279-1285.Park, J-Y. and L.F. Bjeldanes. 1992. Organ-selectiveinduction of cytochrome P-450-dependent activities by indole-3-carbinol-derived products: Influence on covalent binding ofbenzo(a)pyrene to hepatic and pulmonary DNA in the rat.Chem.-Biol. Interactions 83:235-247.Parkinson, J.A. and S.E. Allen. 1975. A wet oxidationprocedure for the determination of nitrogen and mineralnutrients in biological material. Comm. Soil Sci. Plant Anal.6:1-11.Pivnick, K.A., B.J. Jarvis, G.P. Slater, C. Gillott and E.W.Underhill. 1990. Attraction of the diamondback moth(Lepidoptera: Plutellidae) to volatiles of Oriental mustard:The influence of age, sex, and prior exposure to mates andhost plants. Environ. Entomol. 19:704-709.Pivnick, K.A., D.W. Reed, J.G. Millar and E.W. Underhill.1991. Attraction of Northern false chinch bug Nysius niger(Heteroptera: Lygaeidae) to mustard oils. J. Chem. Ecol.17:931-941.Pivnick, K.A., R.J. Lamb and D. Reed. 1992. Response of fleabeetles, Phyllotreta spp., to mustard oils and nitriles infield trapping experiments. J. Chem. Ecol. 18:863-873.Quinsac, A. and D. Ribaillier. 1985. Quantitative analysisof glucosinolates in rapeseed seeds. Optimization ofdesulfation. pp. 85-96, in H. Sorensen (ed.). Advances inthe Production and Utilization of Cruciferous Crops. MartinusNijhoff/DR W. Junk Publishers, Dordrecht.Read, D.P., P.P. Feeny and R.B. Root. 1970. Habitatselection by the aphid parasite Diaeretiella rapae(Hymenoptera: Braconidae) and hyperparasite Charips brassicae(Hymenoptera: Cynipidae). Can. Ent. 102: 1567-1578.Reed, D.W., K.A. Pivnick and E.W. Underhill. 1989.Identification of chemical oviposition stimulants for thediamondback moth, Plutella xylostella present in three speciesof Brassicaceae. Entomol. exp. appl. 53:277-286.Renwick, J.A.A. and C.D. Radke. 1981. Host plantconstituents as oviposition deterrents for the cabbage looper,Trichoplusia ni. Entomol. exp. appl. 30:201-204. 115Renwick, J.A.A. and C.D. Radke. 1983. Chemical recognitionof host plants for oviposition by the cabbage butterfly,Pieris rapae (Lepidoptera: Pieridae). Environ. Entomol.12:446-450.Roessingh, P., E. Stadler, G.R. Fenwick, J.A. Lewis, J.K.Nielsen, J. Hurter and T. Ramp. 1992. Oviposition and tarsalchemoreceptors of the cabbage root fly are stimulated byglucosinolates and host plant extracts. Entomol. exp. appl.65:267-282.Rygg, T. and L. Somme. 1972. Oviposition and larvaldevelopment of Hylemya floralis (Fallen) (Diptera:Anthomyidae) on various varieties of swedes (B. napus) andturnips (B. rapa). Norsk. Entomol. Tidsskr. 19:81-90.Sang, J.P., I.R. Minchinton, P.K. Johnstone and R.J.W.Truscott. 1984. Glucosinolate profiles in seed, root andleaf tissue of cabbage, mustard, rapeseed, radish and swede.Can. J. Plant Sci. 64:77-93.Smith, C.M. 1989. Plant Resistance to Insects. AFundamental Approach. John Wiley and Sons, New York. 286 pp.Thangstad, O.P., T.-H. Iversen, G. Slupphaug and A. Bones.1990. Immunocytochemical localization of myrosinase inBrassica napus L. Planta 180:245-248.Thangstad, O.P., K. Evjen and A. Bones. 1991. Immunogold-EMlocalization of myrosinase in Brassicaceae. Protoplasma161:85-93.Tholen, J.T., S. Shifeng and R.J.W. Truscott. 1989. Thethymol method for glucosinolate determination. J. Sci. FoodAgric. 49:157-165.Thorsteinson, A.J. 1953. The chemotactic responses thatdetermine host specificity in an oligophagous insect (Plutellamaculipennis (Curt.) Lepidoptera). Can. J. Zool. 31:52-72.Todd, G.W., A. Getahun and D.C. Cress. 1971. Resistance inbarley to the greenbug, Schizaphis graminum. 1. Toxicity ofphenolic and flavonoid compounds and related substances. Ann.Entomol. Soc. Amer. 64:718-722.Tookey, H.L., C.H. VanEtten and M.E. Daxenbichler. 1980.Goitrogens. pp. 103-142. In I.E. Liener (ed.). ToxicConstituents of Plant Foodstuffs, 2nd. ed. Academic Press, NewYork.116Traynier, R.M.M. 1965. Chemostimulation of oviposition bythe cabbage root fly, Erioischia brassicae (Bouche). Nature207:218-219.Traynier, R.M.M.^1984.^Associative learning in theovipositional behavior of the cabbage butterfly, Pieris rapae.Physiol. Ent. 9:465-472.Traynier, R.M.M. and R.J.W. Truscott. 1991. Potent naturalegg-laying stimulant for cabbage butterfly Pieris rapae. J.Chem. Ecol. 17:1371-1380.Turnock, W.J.^1988.^Density, parasitism, and diseaseincidence of larvae of the Bertha armyworm, Mamestraconfigurata Walker (Lepidoptera: Noctuidae), in Manitoba,1973-1986. Can. Ent. 120:401-413.Turnock, W.J. and R.J. Bilodeau. 1984. Survival of pupae ofMamestra configurata (Lepidoptera: Noctuidae) and two of itsparasites in untilled and tilled soil. Can. Ent. 116:257-267.Uda, Y., T. Kurata and N. Arakawa. 1986a. Effects of pH andferrous ion on the degradation of glucosinolates bymyrosinase. Agric. Biol. Chem. 50:2735-2740.Uda, Y., T. Kurata and N. Arakawa. 1986b. Effects of thiolcompounds on the formation of nitriles from glucosinolates.Agric. Biol. Chem. 50:2741-2746.Ujvary, I., G. Matolcsy, L. Riddiford, K. Hiruma and K.L.Horwath. 1989. Inhibition of spiracle and crochet formationand juvenile hormone activity of isothiocyanate derivatives inthe tobacco hornworm, Manduca sexta. Pestic. Biochem.Physiol. 35:259-274.VanEtten, C.H., M.E. Daxenbichler, H.L. Tookey, W.F. Kwolek,P.H. Williams and O.C. Yoder.^1980.^Glucosinolates:Potential toxicants in cabbage cultivars.^J. Amer. Soc.Hortic. Sci. 105:710-714.Verschaffelt, E. 1911. The cause determining the selectionof food in some herbivorous insects. K. Acad. VanWetenschappen Proc. Sect. Sci. 13:536-542.Waldbauer, G.P. 1968. The consumption and utilization offood by insects. Adv. Insect Physiol. 5:229-288.Wallbank, B.E. and G.A. Wheatley. 1979. Some responses ofcabbage root fly (Delia brassicae) to allylisothiocyanate andother volatile constituents of crucifers. Ann_ Appl_ Riol-91:1-12.117Woods, D.L., J.J. Capcara and R.K. Downey.^1991.^Thepotential of mustard (Brassica juncea (L.) Coss) as an edibleoil crop on the Canadian prairies. Can. J. Plant Sci. 71:195-198.Wylie, H.G. and G.E. Bucher. 1977. The Bertha armyworm,Mamestra configurata (Lepidoptera: Noctuidae). Mortality ofimmature stages on the rape crop, 1972-1975. Can. Ent.109:823-837.Yu, S.J.^1983.^Induction of detoxifying enzymes byallelochemicals and host plants in the fall armyworm. Pestic.Biochem. Physiol. 19:330-336.Yu, S.J.^1984.^Interactions of allelochemicals withdetoxification enzymes of insecticide-susceptible andresistant fall armyworms. Pestic. Biochem. Physiol. 22:60-68.Yu, S.J. and E.L. Hsu. 1985. Induction of hydrolases byallelochemicals and host plants in fall armyworm (Lepidoptera:Noctuidae) larvae. Environ. Entomol. 14:512-515.118Appendix 1.^Response factors for desulfoglucosinolatesrelative to 0-nitrophenyl-p-D-galactopyranoside at flow rate0.6 ml/minute and wavelength 226 nm.2-propenyl (sinigrin)^ 1.249hydroxybenzyl (sinalbin) 0.595benzyl^ 1.0653-indolylmethyl^ 0.3282-phenylethyl 1.249Appendix 2.(a) HPLC-MS identification of desulfoglucosinolates in aqueous foliar extracts of Brassicajun^ea (commercial brown mustard), Sinapis alba cv. Gisilba, B. napus cv. Midas and Westar,and B. rapa cv. Candle. Relative intensities of diagnostic ions differ from those published(Hogge et al, 1988a) due to a difference in the thermospray source temperature (240 °C ratherthan 265 ° C). * = Out of the range of the scan. ** = Ion not detected.Positive Protonated Molecular Ions and Diagnostic Ions [m/z(%)]Desalfo-glucosinolate(M+H)+^(R-C=^(R-C=^(R-N=C=O + (R-C(=NOH)(R) + (R-C=NOH) + NOH + NOH + 2H) 4"^NH4 )^-S + 2H) +allyl2-073-4-pentenyl3-butenylOH-benzyl4-OH-3-indolyl-methyl4-pentenylbenzyl3-indolylmethyl2-phenylethyl280(63)^41*^84*324(21)^85*^128(17)294 (100) 55 *^98*346(88) 107 *^150(1)385(3)^146**^189(100)308(100) 69 *^112*330(100) 91 *^134 **369(18) 130**^173 (6)344(100) 105 *^148 **85*129 (6)99*151(8)190(22)113 *135(3)174 (24)14 9 **86*130"100*152(2)191(14)114 *136(1)175(100)150(6)101*145 (78)115*167(35)206(7)129(1)151(9)190(15)165(40)118*162(18)132(2)184(51)223"146(1)168(16)207(51)182(45)120(b) Glucosinolates found in aqueous foliar extracts of B.juncea (commercial brown mustard), S. alba cv. Gisilba, B.napus cv. Midas and Westar and B. rapa cv. Candle.Unidentified desulfoglucosinolates are listed as their ScanNumber (associated with retention time). M/z and relativeintensities (%) of the ten most frequently occurring fragmentsare listed for unidentified desulfoglucosinolates.B. junceaScan 54: 198(100), 148(30), 134(25), 154(16), 184(6), 180(6),147(6), 166(3), 130(3), 199(3)allyl3-butenyl3-indolylmethyl2-phenylethylS. albaScan 54:Scan 93:198(100),^154(57),^148(47),^134(37),^147(18),199(12),^184(10),^166(10),^149(8),^180(6)124(100),^226(83),^132(37),^125(7),^180(6),^227(6),133(4),^134(4),^137(4),^146(3)Scan 129: 131(100), 214(31), 180(29),^133(15),^198(15),149(14), 132(13), 310(10),^130(9),^148(6)OH-benzylScan 334: 376(100), 180(91), 300(47),^198(46),^214(40),240(38), 342(38), 360(36),^168(35),^167(34)Scan 353: 320(100), 272(71), 198(48),^180(26),^321(18),199(11), 151(10), 273(6),^288(6),^318(5)benzylB. napus cv. Midas2-0H-4-pentenyl3-butenyl 121B. napus cv. Midas (cont.)Scan 289: 193(100), 214(23), 177(19), 194(15), 176(7), 255(7),178(6), 195(6), 211(5), 210(4)4 -pentenylScan 370: 300(100), 125(42), 301(22), 206(13), 134(9), 247(7),162(7), 305(6), 127(6), 192(5)3-indolylmethyl2-phenylethylB. napus cv. WestarScan 54: 148(100), 198(88), 147(12), 134(11), 210(7), 184(7),154(6), 149(5), 166(3), 199(3)4-0H-3-indolylmethyl3-indolylmethyl2-phenylethylScan 395: 341(100), 319(9), 223(6), 342(5), 191(5), 169(4),193(4), 328(4), 145(4), 248(3)Scan 438: 388(100), 389(19), 150(13), 165(10), 182(9), 167(8),298(7), 195(6), 390(5), 166(4)B. rapaScan 229: 128(100), 145(96) , 214(37), 146(30), 129(21),180(19), 162(17), 198(16), 163(16), 324(10)Scan 328: 164(100), 212(25) , 253(19), 141(13), 124(11),237(11), 134(10), 165(10), 177(9), 198(8)4-0H-3-indolylmethyl4-pentenylScan 453: 432(100), 134(20), 342(17), 317(16), 243(16),143(16), 433(15), 159(15), 344(14), 180(11)122Appendix 3. Quantities (umol/g dry weight) of totalglucosinolates, determined spectrophotometrically, andisothiocyanate- and thiocyanate-releasing glucosinolates,determined by HPLC of desulfo-derivatives, in field-collectedfoliage.Stage 2 (Rosette)Block 1B. juncea commercial brownB. juncea cv. Lethbridge 22AS. alba cv. GisilbaB. napus cv. MidasB. napus cv. RegentB. napus cv. WestarB. rapa cv. CandleB. rapa cv. TobinBlock 2B. juncea commercial brownB. juncea cv. Lethbridge 22AS. alba cv. GisilbaB. napus cv. MidasB. napus cv. RegentB. napus cv. WestarB. rapa cv. CandleB. rapa cv. TobinStage 3 (Stem Elongation)Block 1B. juncea commercial brownB. juncea cv. Lethbridge 22AS. alba cv. GisilbaB. napus cv. MidasB. napus cv. RegentB. napus cv. WestarB. rape cv. CandleB. rapa cv. TobinTotal ITC-type SCW-type67.25 21.44 0.7552.62 15.70 0.8540.03 3.25 22.1716.12 1.81 6.4535.39 0.17 6.8729.43 0.12 8.2828.43 0.49 2.645.40 0.75 2.1786.68 15.02 1.2153.70 12.87 1.4348.99 1.92 20.8371.60 0.38 10.0056.68 0.03 9.3268.91 0.08 5.75o.s. 0.73 4.2487.11 0.86 3.32Total ITC-type SCW-type61.94 24.73 0.6069.53 40.51 0.6646.62 3.70 31.2219.95 4.86 4.097.53 0.36 3.196.25 0.25 3.9812.55 0.50 0.606.74 0.44 0.55123Block 2B. juncea commercial brown 71.51 21.45 0.56B. juncea cv. Lethbridge 22A 74.09 26.90 0.44S. alba cv. Gisilba 44.19 3.96 23.61B. napus cv. Midas 16.58 3.90 2.89B. napus cv. Regent 5.88 0.32 2.77B. napus cv. Westar 7.57 0.28 2.43B. rapa cv. Candle 14.13 1.15 0.67B. rapa cv. Tobin 12.99 0.49 0.41Stage 4 (Flowering)Block 1Total ITC-type SCW-typeB. juncea commercial brown 82.59 14.42 0.64B. juncea cv. Lethbridge 22A 77.58 14.00 0.28S. alba cv. Gisilba 37.56 0.78 6.26B. napus cv. Midas 24.27 4.23 1.68B. napus cv. Regent 5.09 0.33 2.25B. napus cv. Westar 5.09 0.23 3.33B. rapa cv. Candle 5.34 0.27 0.28B. rapa cv. Tobin 4.76 0.26 0.37124Appendix 4.^Cultivar Means for Neonate Choice, RelativeConsumption Rates and Relative Growth Rates of Section 2.a) Neonate M. configurata Leaf Disc Choice Tests (weightedmeans of square root-transformed counts) 1Cultivar^ Growth Stage2^3^4Commercial Brown2 6.0ab 3.3a 8.8abLethbridge 22A2 4.6a 1.8a 4.8aGisilba3 19.9e 15.1cd 16.8bMidas4 9.9bcd 11.9bcd 7.4aRegent4 7.7abc 10.7bcd 8.2abWestar4 7.8abc 16.7d 10.8abCandle5 11.2cd 9.6bc 10.8abTobin5 13.6de 7.7b 9.0abb) Relative Consumption Rate (mg/mg/day) of M. configurata lCultivar^ Growth Stage2^3^4Commercial Brown2 1.803a 1.572ab 1.182aLethbridge 22A2 2.576ab 1.122a 0.790aGisilba3 2.451ab 0.975a 1.636aMidas4 2.318ab 3.462c 2.861aRegent4 2.631ab 2.422abc 2.087aWestar4 3.291b 3.172bc 1.987aCandle5 2.237ab 2.287abc 1.016aTobin5 2.637ab 2.649abc 2.395a125c) Relative Growth RateCultivar(mg/mg/day)2of M. configurata 1Growth Stage3^4Commercial Brown e 0.5377a -0.0374a 0.0150abLethbridge 22A 2 0.5736a -0.0325a -0.0157aGisilba3 0.5637a 0.1312ab 0.0926abcMidas4 0.5985a 0.1442ab 0.2062abcRegent4 0.5814a 0.2947b 0.0888abcWestar4 0.6617a 0.2842b 0.2570cCandle s 0.5841a 0.1122ab 0.1372abcTobin5 0.7073a 0.1870ab 0.2254bc1. Means within each growth stage followed by the same letterare not significantly different from eachother by Tukey'sstudentized range test (p = 0.05).2. Brassica juncea3. Sinapis alba4. B. napus5. B. rapa


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items