gramineaccumulation in leaves barleygrownunderhigh ...plantphysiol. vol.71, 1983 8h.1% 4-. 39 a-d,i...

Plant Physiol. (1983) 71, 896-9040032-0889/83/7 1/0896/09/$00.50/0

Gramine Accumulation in Leaves of Barley Grown under High-Temperature Stress'

Received for publication June 21, 1982 and in revised form December 1, 1982

ANDREW D. HANSON, KIMBERLY M. DITZ, GEORGE W. SINGLETARY, AND TIMOTHY J. LELANDMSU-DOE Plant Research Laboratory/Crop and Soil Sciences Department, Michigan State University, EastLansing, Michigan 48824

ABSTRACT

The indole alkaloid gramine is toxic to animals and may play a defensiverole in plants. Under certain conditions, shoots of barley cultivars such as'Arimar' and CI 12020 accumulate gramine (NN-dimethyl-3-aminometh-ylindole) and lesser amounts of its precursors 3-aminomethylindole (AMI)and N-methyl-3-amithyllndole (MAMI); other cultivars such as'Proctor' do not. When grown at optimal temperatures (21°C/16°C, day/night), Arimar contained a high level of gramine in the first leaf (approxi-mately 6 milligrams per gram dry weight), but progressively less accumu-lated in successive leaves so that the gramine level in the shoot as a wholefell sharply with age. In Arimar and CI 12020 plants transferred at thetwo- to three-leaf stage from 21°C/16°C to supra-optimal temperatures(:300C/250C), there was massive gramine accumulation in leaves whichdeveloped at high temperature, so that gramine level in the whole shootremained high (about 3-8 milligrams per gram dry weight).

Proctor lacked both constitutive gramine accumulation in the first leafand heat-induced gramine accumulation in later leaves. The followingevidence indicates that this results from a lesion in the pathway of synthesis(tryptophan - - AMI -. MAMI -. gramine) between tryptophan andAMI. (a) Proctor and Arimar leaves readily absorbed I14Cjgramine, butneither cultivar degraded it extensively. (b) Arimar leaf tissue incorporatedI14Ciformate label into the N-methyl groups of gramine and MAMI, andconverted Imethylene-4Citryptophan to AMI, MAMI, and gramine; Proc-tor leaf tissue did not, even when a trapping pool of unlabeled gramine wassupplied. (c) Proctor converted I14CIMAMI to gramine as actively asArimar. (d) Proctor incorporated 1'4Clformate label into gramine andMAMI when supplied with AMI; the ratio 114Cjgramine/i14CIMAMI fellwith leaf age, suggesting that the two N-methylations involve differentenzymes. Inasmuch as Proctor leaf tissue did not methylate added trypt-amine or tyramine, the N-methyltransferase(s) of gramine synthesis maybe substrate specific.

In sterile culture at optimal temperatures, 10 millimolar gramine did notaffect autotrophic growth of Arimar or Proctor plantlets or heterotrophicgrowth of callus. At supra-optimal temperature, plantiet growth was re-duced by gramine although callus growth was not. We speculate thatgramine-accumulating cultivars may suffer autotoxic effects at high leaftemperatures.

The simple indole alkaloid gramine (N,N-dimethyl-3-amino-methylindole) is a well-known constituent of young shoots ofcertain barley (Hordeum vulgare L.) cultivars where it can reachconcentrations of about 8 mg g-' dry weight at the one- to three-leaf stage (e.g. 3, 9, 23, 24). High concentrations of gramine and

' Research conducted under Contract DE-ACO2-76ERO1338 from theUnited States Department of Energy. Michigan Agricultural ExperimentStation Journal Article No. 10459.

other indolealkylamines can occur also in foliage of the foragegrass Phalaris arundinacea L. (reed canarygrass); in this crop, theindole alkaloids are important anti-quality components becausethey are toxic to grazing animals and adversely affect their healthand weight gain (17). Gramine has been shown also to havephytotoxic effects on chickweed (22). Although in reed canary-grass it is established that the levels of gramine and other indolealkaloids are subject to both genetic (18) and environmental (15,16) control, little is known of genetic or environmental regulationof gramine in barley (12). Correspondingly little is known forbarley about the physiological, ecological, and agronomic signifi-cance of gramine accumulation. Roles as a feeding deterrent forherbivores (1), as a nematotoxin (3), and as an allelopathic sub-stance (22) have been suggested. Note that all these suggestionsinvoke potent physiological activity-if not outright toxicity-forgramine in plants and animals and thus imply that barley is eitherinsensitive to the alkaloid or able to sequester it in a harmlessstate.

In the gramine biosynthesis pathway established for youngbarley shoots (Scheme 1), the indole ring, the methylene side-chain (and possibly the amino N) are derived from tryptophan (8,10, and references cited therein). The first stable intermediateidentified after tryptophan is AMI2, which is methylated stepwiseto MAMI and then to gramine; shoots of gramine-containingbarley cultivars have small pools of AMI and MAMI (8, 12, 20,24).

Catabolism of supplied and endogenous gramine can certainlyoccur in barley (6, 9, 24), although net degradation rates ofendogenous gramine may be quite low (s5% d-') (9, 24). There islittle information about the catabolic pathways; the methylenecarbon of the side-chain can be oxidized to C02, possibly via 3-indolecarboxylic acid, and there may be some recycling of theindole nucleus to tryptophan (6).

In barley, a brief seedling phase of net gramine synthesis hasusually been considered to give way to a phase ofslow net graminedegradation as the plant develops, so that gramine concentrationsdecline steadily after about the two-leaf stage to reach negligiblevalues by the five- to ten-leaf stage (9, 24). Our recent results withbarley cultivars Arimar and Maraini grown in both laboratoryand field generally conformed to this pattern under cool, near-optimal conditions, but not under warmer conditions (12). Ele-vated temperatures promoted active net gramine synthesis in theshoots, causing gramine concentrations to remain high in quitemature (five- to six-leaf stage) plants. We also found great geneticdiversity for gramine concentration among cultivars of barley andraces of Hordeum spontaneum, the wild progenitor (12). Certaingenotypes (e.g. cv Proctor) never contained detectable gramine

2 Abbreviations: AMI, 3-aminomethylindole; MAMI, N-methyl-3-ami-nomethylindole; PC, paper chromatography; DMAC, p-dimethylamino-cinnamaldehyde; TLE, thin-layer electrophoresis; BES, N,N-bis[2-hy-droxyethyl]-2-aminoethanesulfonic acid.

896

www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from Copyright © 1983 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

GRAMINE IN BARLEY

ICIII37H2IIJHFCOH-2 +CH3 +CH3 ,CH3

HJ CH OOCH-NH -*---R-CH2-NH2 P R-CH2-NH-CH3 - R-CH2-NH CH3

Tryptophan AMI MAMI Gramine

Scheme 1

(-30 Ag g-' dry weight) whereas others (e.g. Arimar, Maraini, CI12020, and four H. spontaneum races) always did so, with concen-trations at the four-leaf stage ranging from 2 to 10 mg g-' dryweight. Gramine accumulation thus appeared to be both a consti-tutive feature of young seedlings and a heat-inducible character-istic of older plants, with both types of accumulation absent fromcertain genotypes.

In this study, we characterized gramine accumulation in relationto shoot age and to heat-stress in cv Arimar and CI 12020, andshowed cv Proctor to lack gramine because of a lesion in biosyn-thesis. To probe the physiological significance of gramine, wetested its effects on barley plantlets and callus in sterile culture.We also investigated the pH-dependence of the octanol/waterpartition coefficient of gramine; this coefficient describes thedistribution of a compound between a model organic phase andan aqueous phase and can indicate the probable distribution of acompound in vivo between membrane and aqueous tissue com-ponents.

MATERIALS AND METHODS

Plant Material and Growth Conditions. The seed sources andprovenances for the spring barley cv Proctor (CI 11806), Arimar(CI 13626), and Mahouma 3445 (CI 12020) have been givenearlier (12). Plants were routinely grown four per pot in a soil mix(12) and irrigated each 2nd d with half-strength Hoagland solu-tion; standard growth chamber conditions were: 16-h d, 21 °C, RH60%o, 200 ,uE m-2 s-1 PAR/8-h night, 16°C. For the time courseexperiment of Figure 1, plants for harvest at 4 and 7 d were grownin flats (six rows, 20 plants/row). For the growth-temperatureexperiments of Figures 2 and 3, plants were first grown (in mostcases, for 10 d ) in the standard growth chamber conditions. Theywere then thinned to two per pot and placed in growth chambersunder various temperature regimes for a further 14 d (10°C/5°C,150C/100C, 21°C/160C, 30°C/250C, 35°C/300C, 38°C/330C).For all regimes, the irradiance and daylength were as given above;daytime RH was adjusted to give a vapor pressure deficit approx-imating that of the standard growth chamber conditions (about 10mbars). Plants were irrigated with half-strength Hoagland solutionliberally each day throughout the temperature treatments, allow-ing about 1 pot volume of solution to drain through each pot.

Alkaloid Extraction and Determination. Whole shoots or leaveswere freeze-dried, weighed, and ground in a Wiley mill to pass a40-mesh screen. A weighed sub-sample (usually 100 mg) ofgroundmaterial was then extracted as detailed previously (12). Repre-sentative samples in each batch were spiked with a small quantityof [14Clgramine (e.g. 1.5 ,ug, 3 nCi) for estimation of gramine

recovery, which averaged about 70o. The gramine contents ofalkaloid fractions were estimated from the A at 270 nm; valuesare expressed as gramine equivalents, corrected for recovery (12).Representative samples were separated by PC in the 'Isobuffsystem (23) followed by spraying with DMAC reagent (12) toconfirm that gramine and its precursors AMI and MAMI wereresponsible for theA at 270 nm. Because Proctor alkaloid fractionsshowed very little A at 270 nm, and gramine was not detectableby PC in this genotype (12; detection limit = 30 ,ug g-' dry weight),fluorimetry (4) was applied in a sensitive isotope dilution assay

for gramme in selected Proctor samples (detection limit about 1,ug g- dry weight). In this assay, 250-mg samples of freeze-driedmaterial were spiked with 1 ,ug (1.3 nCi) of ['4C]gramine andextracted using the standard procedure (12), except that the con-centrations of H2SO4 and NaOH were increased 2.5-fold. Thealkaloid fractions were subjected to PC in the Isobuff system, andthe zone corresponding to the R, for gramine was eluted. Analiquot of the eluate was taken for scintillation counting, and theremainder was used for fluorescence measurements, made inMcIlvaine's citrate-phosphate buffer, pH 5.6 (4) with a System4000 scanning spectro-fluorimeter (SLM Instruments, Urbana, IL)by the dual channel ratiometric acquisition method.

Radiochemicals and Alkaloid Precursors. [14C]Formate (54.8 or59.1 ,uCi ,imol1') and L-[methylene-'4CJtryptophan (58.2 ,uCiumnol-) were from Amersham Corp.; radiochemical purity of theI14Cjtryptophan was verified by TLE on cellulose in 1.5 N formicacid at 2 kv for -12 min. [14C]Gramine (0.23 or 0.25 ,uCi /Lmol-')and [14C]MAMI (0.45 ,uCi ,imol-') were isolated from first leaftissue of Arimar supplied with ['4C]formate. Thirty 1-cm leafsegments (2 or 3/leaf, from 6-d-old shoots) were infiltrated with1 p1 [14C]formate solution (0.4 ,uCi, 59.1 ,uCi /Lmol-1) per segment,and incubated for 24 h on moist filter paper in darkness at 21 °C/16°C. The segments were frozen in liquid N2, and extracteddirectly by standard procedure; the alkaloid fraction was separatedby PC in the Isobuff system. Gramine and MAMI zones werelocated by UV absorption and by autoradiography, eluted andchecked for radiochemical purity by TLC on silica gel G inmethanol:acetone:concn. HCI (90:10:4, v/v) and by TLE as above.For one preparation of [14C]gramine, TLE as above was used ona preparative scale to upgrade radiochemical purity. The label inthe ['4C]gramine preparations was shown to be predominantly(-900%b) in the N-methyl groups, as described below.

Authentic gramine (Sigma) was recrystallized from acetone.Tryptamine HCI (NBC) and tyramine (Sigma) were recrystallizedfrom 96% ethanol and checked for chromatographic homogeneityin the PC Isobuff system. AMI was synthesized using a modifi-cation ofthe Mannich reaction, as follows. Glacial acetic acid (117mmol) was added to NH40H (57% solution, 53 mmol) at 4°C. Tothis solution was added formaldehyde solution (36%, 53 mmol) at4°C. This mixture was then added to 26 mmol of indole. Thereaction was run for 12 h at 4°C, and then warmed to 25°C for anadditional 4 h. AMI was separated from the reaction mix by twocolumn chromatography steps (silica gel 60, followed by SephadexLH-20), and checked for chromatographic homogeneity in the PCIsobuff system. Overall yield of purified AMI was 2.4%."C Tracer Experiments. Plants for labeling experiments were

grown at 21°C/16°C. All experiments using [14CJformate werewith 1-cm segments cut from the center of the first leaf blade,usually of 6-d-old plants (1-3 segments/blade), infiltrated with 1p1/segment of a solution containing 0.4 to 0.5 ,uCi I14Clformate inH20. In some cases, segments of Proctor were preinfiltrated withunlabeled gramine (5 nmol/segment) or unlabeled alkaloid pre-cursors (10 nmol/segment) dissolved in 1 id of K-phosphate (20mK pH 7), supplied 1.5 to 3 h before the I14Clformate. Segmentswere incubated on moist filter paper in darkness, usually for 24 h,at 21°C/16°C.

897



Plant Physiol. Vol. 71, 1983

8h

.1%

4-.

39

A-

D,i

aw

w

z

61-

4

2

0

0.3 '

2c0.2 A /

0.1

04 12 20

ARIMARIM

n PROCTOR

O_O_ O_o- _.+_ _ _0

4 8 12 16 20

DAYS AFTER PLANTINGFIG. 1. Time courses of indole alkaloid concentration in shoots of

barley cv Arimar and cv Proctor, grown at 21°C/16°C. Indole alkaloidswere estimated from A27o of alkaloid fractions; for each time, two or fouralkaloid determinations were made on material pooled from 4 to 60 shoots.Paper chromatograms sprayed with DMAC showed that gramine was themajor (>80%) indole alkaloid in Arimar shoots of all ages; MAMI andAMI were the other indole alkaloids present. Chromatograms of Proctoralakloid fractions gave no DMAC-positive reaction. By isotope-dilution/fluorescence assay, 7-d-old Proctor shoots (arrow) were found to containonly I fg g-' dry weight of gramine. Inset gives dry weights of the seedlingshoots.

For experiments with ["4Cgramine,["C]MAMI, and ["Citrp-

tophan, attached second leaves of 12- or 13-d-old plants wereused. Labeled substances were supplied in a 3-pl droplet of 20 mmK-phosphate (pH 7), applied either to the cut tip of a bladetrimmed to 5 or 10 cm in length or to an abraded spot sited 2 to3 cm above the point of blade emergence from the sheath of thefirst leaf. In experiments with l4qCtryptophan, leaves of someProctor plants were supplied with 50 nmol of unlabeled gramine

2 or 3 h before [14C]tryptophan addition. Plants were incubated ingrowth chambers at 21°C/16°C for up to 2 d, and the secondleaves were harvested. In experiments with [14C]gramine, theunabsorbed radioactivity remaining at the end of the experimentin free-space of the fed blade was determined as described previ-ously (1 1).As appropriate, authentic alkaloids or leaf tissues of Arimar

were added to labeled Proctor leaf tissues to provide carrier for"4C-alaloids during extraction. Alkaloid fractions were routinelyseparated by PC in the Isobuff system and by TLC on silica gel Gin methanol:acetone:concn. HCI (90:10:4, v/v). Radioactive zoneswere located by autoradiography 'and eluted for 14C assay byscintillation counting. Identification of "4C-metabolites was basedon co-chromatography or co-electrophoresis with authentic stan-dards in at least two systems. In the case of [14C]MAMI, for whichno standard was available, identification was based on comigration

06

04

8

z

0

r.-02

0

8

6

4

2

(-I

w

0

25

20

1-5

1[0

0-5

o10 20 30 40

DAYTIME TEMP (°C)FIG. 2. Effects of growth temperature on dry weight gain (A), indole

alkaloid concentration (B), and net indole alkaloid synthesis (C) in shootsof three barley cultivars. All plants were grown for 10 d at 21°C/16°C,and then transferred to the various temperature regimes for 14 d beforeharvest. Night temperatures were 5°C below day temperatures. Dryweights (mg) for 10-d-old shoots: Proctor = 26; Arimar = 32; CI 12020 =38. Alkaloid concentrations (mg gramine eq g-' dry weight) of 10-d-oldshoots: Proctor < 0.2; Arimar = 5.2; CI 12020 = 5.2. Proctor shoots grownat 30°C/25°C (arrow) were checked for traces of gramine by isotope-dilution/fluorescence assay, and found to contain <1 pg g-' dry weight.Data in A and C are means of three replicates; each replicate comprisedthe two plants from a pot; for A, the dry weights of 10-d-old shoots were

subtracted from the weights at harvest. Data in C were derived bysubtracting the alkaloid contents of 10.d-old shoots from those of shootsharvested after the 14-d temperature treatments.

with the MAMI zone of Arimar leaf extracts. For [14Cigramine,the identity was further checked by alkylation with iodomethane,which yielded the labeled quatemary ammonium derivative ofgramine.To confirm that the label in ["CJgramine isolated from Arimar

leaves supplied with [14Ciformate was in the N-methyl groups, thedimethylamino function was liberated by heating in base. About10 nCi of [i4C]gramine was mixed with 1 ,umol ofauthentic carriergramine-HCl in a total volume of 10 ml H20, and samples weretaken for measurement of specific activity. The remaining solutionwas made alkaline with 0.2 ml of 2 N NaOH, and subjected to

A. SHOOT GROWTHARIMAR

0112020

c(- POTOR

El SH ALKALOID C0C

C 12020

^ ~~~~~PROCTOR-C. NET ALKALOID -

SYNTHESIS \

a \\/

C I12020

ARMAR

II I~~~~~~~~~~~~~~~~~~~~~~~~~898 HANSON ET AL.



GRAMINE IN BARLEY

6

5

4

^ 3L-

.0 2

w

-j

J

aw

w

z

7

6

5

4

3

2

1 2 3 4 5 l6LEAF NUMBER

FIG. 3. Indole alkaloid concentrations in individual leaves from themain culms of Arimar barley plants grown at optimal and supraoptimaltemperatures. Plants were grown at 21°C/16°C for 10 d (A) or 18 d (B),and then either maintained at 21°C/16°C or transferred to 30°C/25°Cfor a further 14 d. Data are means of duplicate determinations on leaves(blades + sheaths) pooled from eight plants (A) or 12 plants (B). Leaves3 through 5 or 6 were smaller in the 30°C/25°C plants, and completedexpansion slightly earlier. (In A, for the 22°C/16°C plants the expandedpart of leaf 5 was pooled with leaf 4 for analysis, and in the 30°C/25°Cplants the expanded part ofleaf6 was pooled with leaf 5). The approximatedevelopmental stage of the main culms at harvest is shown schematically;shaded areas indicate tissues that were emerged and essentially expandedat the time of transfer to 30°C/250C. In B, horizontal arrows denote thealkaloid levels in the first and second leaves at 18 d.

steam distillation. The dimethylamine in the steam distillate (yieldabout 70%) was trapped in 2 ml ice-cold 1 N HCI. After reducingthe trap to dryness and redissolving in 10 mm HCI, samples weretaken for scintillation counting, for dimethylamine assay with theFolin reagent (2), and for TLC confirmation that dimethylaminewas the only labeled compound and the only amine present.

Phytotoxicity Tests. Sterile embryos were prepared from seedsof Proctor and Arimar as described by Olien and Smith (21). Forcallus initiation experiments, embryos were cut into four piecesand placed in 6-cm Petri plates (1 embryo/plate) containing 10 mlof l-B5 medium (7) supplemented with 0, 1, or 10 mm gramine-HCI and solidified with 8 g 1` agar. The gramine-HCl was filter-

sterilized as a 500 mM stock (pH 5) and was added to the othermedium constituents during cooling after autoclaving. Plates werewrapped in aluminum foil and incubated in darkness at 25°C or32°C/27°C. The explants were harvested after 2 weeks, rinsedwith distilled H20 until the A270 of the rinsing water wasnegligible, fresh-weighed, and freeze-dried. For autotrophic plan-tlet growth experiments, intact sterile embryos were first germi-nated in darkness for 2 d at 25°C on B5 minerals, vitamins, andsucrose plus 8 g 1` agar (7), and were then transferred to 2.5- x20-cm tubes containing 10 ml of B5 minerals solidified with 8 g1-1 agar supplemented with 0, 1, or 10mm filter-sterilized gramine-HC1. Sterile plantlets were cultured for 2 weeks in the normalgrowth chamber conditions (16 h day, PAR 200 ,E m-2 s-1) at21°C/16°C or 32°C/27°C, fresh-weighed, and freeze-dried. En-dogenous gramine levels in freeze-dried callus or plantlet shoottissue were determined as described above. Fresh weight and dryweight data were subjected to a two-way analysis of variance.

K., Measurements. 1 -Octanol/water partition coefficients wereestimated according to (14), using A270 to quantify gramine inaqueous and octanol phases. The pH range was obtained with thefollowing buffer solutions: pH 2.6 to 7.6, 0.1 M citric acid-0.2 MNa2HPO4; pH 8.6 to 10.6, 0.05 M glycine-NaOH; pH 6.2 to 8.2, 0.1M BES-NaOH; pH 11.6, 0.1 M Na2HPO4-NaOH. Measurementswere made at room temperature (about 22°C), starting withgramine concentrations ranging from 0.8 to 6.3 mg ml-' in theoctanol phase; K., was shown to be independent of gramineconcentration in this range.

RESULTS AND DISCUSSION

Gramine Concentration during Seedlng Growth. When grownat 1 °C/16°C, shoots ofArimar seedlings showed trends in gramineconcentration similar to those reported for several other barleycultivars (e.g. cv Certina [9], cv Champlain [24]). Gramine, whichwas not chromatographically detectable in dry seeds, was synthe-sized actively by emerging shoots; gramine concentration peakedat 7 d and thereafter declined steadily (Fig. 1). From day 4 to day7, the rate of net gramine synthesis was about 3.5 mg g-1 dryweight d-', but had dropped to about 0.2 mg g-1 d-l between days14 and 21. The decline with time of gramine concentration in theshoot as a whole reflected a progressive reduction in gramineconcentrations in leaves 2 through 6 (Fig. 3); there was no net lossof gramine from the first or second leaves as the plants aged (Fig.3B). Figure 1 shows that, in contrast to Arimar, the alkaloidfractions of Proctor shoots contained very little material absorbingat 270 nm (indole alkaloids). Analysis of 7-d-old Proctor with theisotope-dilution/fluorescence assay showed only a trace of gram-ine (1 ,ug g-1), a concentration some I04-fold lower than that inArimar shoots of the same age.

Effects of Temperature on Growth and Gramine Concentration.Six growth-temperature regimes were tested using Proctor andArimar, and also CI 12020, a high-gramine cultivar unrelated toArimar (Fig. 2). All cultivars grew best at 21°C/16°C; 30°C/25°Cwas somewhat supraoptimal and 35°C/30°C highly so, but clearheat injury symptoms (chlorosis ofthe expanding leaves) appearedonly at 38°C/33°C. Chlorosis increased in severity in the orderProctor < Arimar < CI 12020. Proctor shoots contained virtuallyno indole alkaloids at any growth temperature. Gramine concen-tration in Arimar and CI 12020 was affected by temperature inthe same way: gramine concentration remained low at suboptimaland optimal growth temperatures, increased sharply as tempera-ture became moderately supraoptimal, and began to fall when theheat stress was severe. Figure 2C demonstrates that these effectson gramine concentration were due to enhanced net graminesynthesis (at average rates of up to 0.9 mg g-1 d-l) by shoots at30°C/25°C and 35°C/30°C, followed by a fall in net synthesisrate coinciding with the appearance of chlorosis. The increasedgramine concentration in shoots grown at 30°C/25°C reflected

899



Table I. Fate of[Methyl-_4CJGramine Supplied to Attached Second Leaf Blades ofArimar and Proctor Barley PlantsThe [14Cjgramine doses fed corresponded to about 10 to 20% of the endogenous gramine content of an Arimar second leaf blade. Results are from

several similar tests during a 6-month period. In each test, [methyl-_4Cqgramine (0.23 or 0.25 nCi nmol-') was supplied to the expanding second leafblade of one 12- or 13-d-old Arimar or Proctor plant. At harvest, the second leaf blades were taken for estimation of 14C in free space, and forsubsequent extraction. The remaining plant parts were freeze-dried and their 14C content was determined after combustion. Data are included forunlabeled Proctor blades spiked with [I4Clgramine at the start of extraction, and processed alongside fed blades. These spiked samples indicate theI14Clgramine recovery to be expected from the extraction and fractionation procedures.

14C Distribution

NumbeRane of Rangeo14LabeledCultivar of 1C]GramineRangeof o Total '4C Blade 2

Shoot Compoundsof['4CJGramie Incubation Recovery mn AlkaloidTestsa Doses Times Alkaloid (-Blade 2) Fractionb

Free space fraction + Roots

nmol/blade h % of 14C suppliedArimar 3 32-48 25-56 91 + 3 21 ± 3 55 ± 3 <1 GramineProctor 5 32-67 24-56 93 ± 5 18 ± 3 58 ± 3 <1 GramineProctor spiked 3 9-24 73 ± 4 Gramine

a Because of all combinations of ['4C]gramine dose/feeding method/incubation time tested gave similar results, data are summarized as the meanssSE for all tests.b Analyzed by TLC or PC. Up to four minor labeled zones (one comigrating with dimethylamine) were sometimes detected, but were considered not

to be metabolites of gramine because comparable minor labeled zones were present also in accompanying spiked extracts.

Table II. Conversion ofL -[Methylene-'4CJTryptophan to Indok Alkaloids by Attached Second LeafBlades ofArimar and Proctor Barley Plants

The ['4C]tryptophan dose (15 or 18 nmoL specific activity 58 nCi nmol') was about equal to the endogenoustryptophan content ofbarley second leafblades (estimated from Table III ofRef. 23). In experiment A, expandingblades of 13-d-old plants were fed via the cut blade tip; a 50-nmol trapping pool of unlabeled gramine was fedto one-half of the Proctor leaves 3 h before [14CJtryptophan addition. In all treatments of experiment A, <5% ofthe ['Cltryptophan remained unmetabolized at the end of the experiment. In experiment B, expanding blades of12-d-old plants were fed via an abrasion at the blade base; the gramine trap was supplied 2 h before the[4C]tryptophan. All data are means for duplicate plants, and are not corrected for incomplete alkaloid recovery.

14C Incorporation intoExperi- Cultivar and ["4C]Tryptophan Incubation Indole alkaloidsament Treatment Dose Time

AMI MAMI GraminenCi/blade h nCi/blade

A Arimar 5.0 8.7 3.9Proctor 860 22 <0.6 <0.9 <0.3Proctor + 50 nmol <0.6 <0.9 <0.3gramine

B Arimar 7.6 11.2 13.8Proctor 1,020 48 <0.8 <0.5 <0.7Proctor + 50 nmol <0.8 <0.5 <0.7gramine

a Detection limits for "4C labeling of all three alkaloids were taken as 2 x the background "4C activity presentin PC or TLC zones of alkaloid fractions from unlabeled blades spiked with the [14CJtryptophan dose beforeextraction.

increased gramine accumulation by all of the leaves which ex-panded during the high-temperature treatment (Fig. 3). Leavesthat were fully expanded or nearly so (leaves 1, 2, and 3 in Fig.3B) at the time of transfer to high temperature accumulated littleor no additional gramine, although both leaves 2 and 3 were ableto do so if they were still expanding at transfer (Fig. 3A).Metabolsm of [Methyl-'4CjGramine. The very low gramine

content of Proctor might result either from a lack of syntheticcapacity or from a degradative capacity sufficiently high to preventbuild-up of a pool of gramine. To test whether Proctor was ableto degade gramine more rapidly than Arimar, physiological dosesof [1 C]gramine were supplied to attached second leaf blades(Table I). Both cultivars took up most (about 80%) of the4C]gramine from free-space, but neither cultivar metabolized the

[I4C]gramine appreciably during incubations lasting up to 56 h.

There was almost no export of 14C from the fed blades. BecauseTable I indicates that no more than 10% of the [14CJgraminesupplied to Proctor underwent metabolic degradation, an upperbound for the gramine-degrading capacity of Proctor can be set atabout 2 nmol leaf' d-', equivalent to about 10 jg g-1 dry weightd-'. Inasmuch as this upper bound for the degradation rate is atleast 20x lower than the rate of net gramine synthesis in youngshoots of Arimar plants (Fig. 1), rapid degradation of gramine isunlikely to account for the gramine-deficiency of Proctor.Metabolsm of [14ClTryptophan and I[4CIMAMI. Table II

shows that when [I4C]tryptophan was supplied to attached, ex-panding second leaf blades, Arimar incorporated 14C into AMI,MAMI, and gramine (total [14C]tryptophan conversion to indolealkaloids = 2.3%), as do other gramine rich cultivars (8, 9). Proctordid not incorporate measurable 14C from ["4CItryptophan into

900 HANSON ET AL Plant Physiol. Vol. 71, 1983



GRAMINE IN BARLEY

Table III. Conversion ofJ'4CJMAMI to Gramine by Attached Second LeafBlades ofArimar and Proctor BarleyPlants

The ['4CJMAMI dose (9 nmoL specific activity 0.45 nCi nmol') was estimated to be about one-half theendogenous MAMI content of an Arimar second leaf blade (see Fig. 5 in Ref. 24). In both experiments A and B,the [14CJMAMI was supplied to the expanding second leaves of one Arimar and one Proctor plant, via anabrasion at the blade base. Unlabeled blades spiked with ['4CJMAMI or [14Clgramine before extraction wereprocessed alongside the fed blades.

Exeri- Cultivar Plant [14CJMAMI Incubation 14C Incorporationment Age Dose Time into Graminea

d nCi/blade h nCi/bladeA Arimar 13 3.94 57 0.73

Proctor 13.40.74

B Arimar 12 3.99 48 0.73Proctor 1239 81.25

a Corrected for traces of 14C (<0.22 nCi/blade) from unmetabolized [14C]MAMI that were recovered in PC orTLC gramine zones, and for gramine recovery from PC or TLC zones, which averaged 48% for the accompanyingsamples spiked with [14C]gramine.

Table IV. Incorporation of'4Cfrom [j4CIFormate into Indole Alkaloids by Segments of First Leaves of6-Day-Old Arimar and Proctor Barley Plants

The [14C]formate (55 or 59 nCi nmol-) in aqueous solution was infiltrated into batches of 10 or 30 1-cm leafsegments (1 pl/segment) cut from the center of expanding first leaves. Gramine and AMI were dissolved in K-phosphate and were infitrated (1 pI/segment) 1.5 to 3 h before the [14CJformate. The control Proctor segmentswere infitrated with K-phosphate; control Arimar segments were not. Incubation was for I d in darkness. Dataare not corrected for incomplete alkaloid recovery. Experiment B was repeated three times, with very similarresults.

14C Incorporation intoExperi- Cultivar Treatment 114C]Formate Indole Alkaloidsament Cultivar Treatment Dose

MAMI Gramine

nCi/segment nCi/segmentA Arimar Control 1.55 12.4

Proctor Control 0.01 0.01Proctor +Gramine, 5 nmol 370 0.01 0.01

B Arimar Control 1 1.17 12.1Proctor Control 0.01 0.02Proctor +AMI, 10 nmol 490 4.98 8.57

a Analyzed by PC (exp. A), and PC + TLC (exp. B). No [1'4CAMI was detected in Arimar.

gramine or the other indole alkaloids, even when a large trappingpool of gramine was present. These results establish that Proctoris deficient in at least the early part of the gramine biosynthesispathway, i.e. that part leading from tryptophan to AMI. A blockin the synthesis path upstream from AMI is also strongly impliedby the observation that Proctor lacks AMI and MAMI as well as

gramine (Figs. 1 and 2; Ref. 12).To find whether the last step in the gramine synthesis pathway4

the methylation of MAMI-was also deficient in Proctor, [ 'C]MAMI was supplied (Table III). Proctor synthesized at least asmuch ["C]gramine as did Arimar. Even if it is assumed that theadded [i4CJMAMI equilibrated completely with the internal poolof MAMI in Arimar (see Table III), this result still suggests thatProctor has a capacity for methylating MAMI quite comparableto that of Arimar. Note, however, that in both cultivars the rate ofconversion of the added ["CIMAMI to gramine over the 2-dincubation averaged only about 1 to 2 nmol leaf-' d-i (equivalentto about 5 to 10 ,ug gramine leaf' d-') and so was far lower thanthe rates of net gramine synthesis in young Arimar shoots (Fig. 1).Metabolism of 114CIFormate. Because ["Ciformate enters 1-

carbon metabolism in barley (11, 12), it should be readily incor-porated into the N-methyl groups ofMAMI and gramine in tissuessynthesizing these alkaloids; I 4Ciformate labeling of the indolenucleus and side chain of AMI would likely be far slower. Inaccord with these expectations, first leaf tissues from Arimarincorporated label from [14C]formate into both MAMI and gram-ine, but not into AMI (Table IV); the 14C in gramine was largelyconfined (>90%1o) to the N-methyl groups (Table V). In Proctorfirst leaf segments, no label was detected in any indole alkaloidwhen [14C]formate was supplied either in the presence or absenceof a trapping pool of granine (Table IV), which strengthens theconclusion that gramine biosynthesis in this cultivar is blocked atan early step. When Proctor segments were supplied with unla-beled AMI together with [14C]formate, 14C incorporation into bothMAMI and gramine comparable to that in Arimar was obtained(Table IV). This supports and extends the results from feedingI14C]MAMI (Table III); Proctor has significant capacity not onlyfor the second N-methylation in the gramine synthesis pathway,but also for the first.Since Proctor possessed N-methylation capacity, but lacked

901




Table V. Degradation of f" CJGramine Isolatedfrom First Leaf Segments ofArimar Barley Supplied/"CJFormate

[14CjGramine batches I and 2 were from leaf segments incubated for 6 and 24 h after [14Cqformate addition,respectively.

["4CJGramine Batch "CI]Gramine Reaction Mix Reaction Product

No. Specific Total Specific Total SpecificNo. activity gramine activity recovered activitya

nCi nmol' ,umol nCi ,.mol" ,umol nCi umol'1 0.25 0.94 10.7 0.77 9.9 (92%)2 0.23 0.91 10.1 0.71 9.4 (93%)

a Values in parentheses are specific activity of [14Cldimethylamine expressed as a percentage of that of the [14C]gramine degraded; they indicate that up to 8% of the 14C in the [14Cigramine batches was located elsewhere thanthe dimethylamine function. Consistent with this, small amounts of 14C were recovered in a DMAC-positivedegradation product that had lost the N-methyl groups (not shown).

Table VI. Incorporation of "4Cfrom /4"CJFormate into Alkaloids by Segmentsfrom Expanding First Leaves ofCereals, in the Presence and Absence ofAromatic Amines

1[4CIFormate (55 nCi nmol', 490 nCi/segment) was infitrated into batches of four 1-cm segments cut fromthe center of expanding first leaf blades. In experiment A, all plants were 6-d-old; in experiment B, barley andrye (cv Rosen) were 7-d-old, wheat (cv Mexipak) and oats (cv Froker) were 9-d-old. The amines (10 nmol/segment) were in K-phosphate and were infiltrated 2 h before the ["Cjformate. Incubation was for 1 d indarkness. Both experiments were repeated, with very similar results.

14C Incorporation Products Present inGenotype e into Alkaloid Alkaloid Fraction

ment Supplied FraintionAkaoi _________4______Experi-GenoydFractiona "C-Labeled Unlabeled

nCi/segmentA Arimar 25.3 Gramine, MAMI AMI

Proctor 0.10Proctor AMI 15.8 Gramine, MAMI AMIProctor Tryptamine 0.15 Tryptamine

B Arimar 13.3 Gramine, MAMI AMIProctor 0.16Proctor AMI 18.8 Gramine, MAMI AMIProctor Tyramine 0.06 Tyramine,

HordeninebWheat 0.02Wheat AMI 0.00 AMIRye 0.07Rye AMI 0.03 AMIOats 0.05Oats AMI 0.02 AMI

a Corrected for background "C present in alkaloid fractions of unlabeled barley segments spiked with [14Clformate before extraction (= 0.03 and 0.07 nCi/segment in experiments A and B, respectively). Not corrected forincomplete alkaloid recovery.

b 100 jig of hordenine-SO4 was added to the batch of four segments before extraction as carrier for a possiblereaction product.

endogenous pools of AMI and MAMI, in vivo experiments with["Ciformate were devised first to test the substrate specificity ofthe methyltransferase activity, and second to indicate whetherseparate methyltransferases catalyze the two methylation steps.When leaf segments were supplied with tryptamine or tyramine,both of which are normal minor metabolites of barley seedlings,neither amine was detectably methylated (Table VI). Also, firstleaf tissues from wheat, rye, and oats-cereal species that lackgramine-failed to methylate supplied AMI (Table VI). Takentogether, these results imply that the enzyme(s) catalyzing the N-methylations of gramine biosynthesis are unique to this pathwayand are of narrow substrate specificity.To investigate whether the N-methylations are functions of the

same or different enzymes, barley first leaf tissue of various ages

was supplied AMI and [14C]formate. Because net synthesis ofgramine appears to take place in such leaves only during expansion(Fig. 3), it was reasoned that: (a) the activities of enzymes of thesynthesis pathway would likely decay at various rates as the leafaged; (b) if distinct enzymes catalyze each methylation, thendifferences in their rates of decay might generate diagnosticchanges in the amount of 14C accumulated in the intermediate(MAMI) compared to that in the end-product (gramine). Figure4 shows that, on day 6, leaf segments synthesized approximatelytwice as much [14C]gramine as [14CJMAMI, and that whereas[1Cigramine production fell continuously with leaf age, ['4C]MAMI production increased until 20 d, by which time "4CJMAMIhad become the major reaction product. These changes are con-sistent with, but do not establish, the presence of separate enzymes

902 HANSON ET AL.



903GRAMINE IN BARLEY

Table VII. Heterotrophic Growth of Callus and Autotrphic Growth of Plantletsfrom Embryos ofProctor andArimar Barley Cultured with 0, 1, and 10 mM Gramine: Summary ofMean Squaresfrom Analyses of VarianceThe ranges in embryo callus fresh weight at harvest for all temperature and gramine treatment means were 82

to 113 mg for Proctor, 95 to 134 mg for Arimar. Initial embryo fresh weight was 2 to 3 mg. In both cultivars inboth temperature regimes; internal gramine levels of callus (plant water basis) were: <0.2 mm at 0 mM gramine;1 to 2 mm at I mM gramine; 7 to 13 mm at 10 mM gramine. Ranges in plantlet dry weight at harvest for alltemperature and gramine treatment means were 6.0 to 9.5 mg for Proctor, 5.0 to 12.8 mg for Arimar. Initialplantlet dry weight was 4 to 6 mg. In both temperature regimes, internal gramine levels of shoots (plant waterbasis) were: <0.2 mm for Proctor and 3 mm for Arimar at 0 mm gramine; 3 to 5 mm for both cultivars at I mmgramine; 14 to 25 mm for both cultivars at 10 mm gramine.

ReplicatesSource of Variancea

Culture System Parameter per GramineTreatment Cultivars (C) concn. (G) C x G

Embryo callus, 25°C Fresh wt 15 6,822b 2,614 197Embryo callus, 32°C/27°C Fresh wt 5 2,806 1,192 1,733

Plantlets, 210C/160C Dry wt 5 70.7b 1.73 4.25Plantlets, 320C/270C Dry wt 7 4.23b 9.37b 0.76b

' Degrees of freedom associated with F tests were 1 for cultivars, 2 for gramine and C x G.b Significant at the 1% level of probability.

40H0

cn

E

CE

a)

nI)N

G

(n(I)wI

z(I)0

-J

0

30F

20k

10

o-- _5

DA

FIG. 4. Effect of leaf ageinto indole alkaloids by segmAMI. Batches of three 1-cm sleaf of Proctor plants of varinmol/segment), and 2 h laternmol'). Incubation was forduplicate samples; in all casethe mean. The experiment w;

for the sequential methyl;than the first.

Phytotoxicity and Partiwas tested for effects ontrophic plantlet growth at

gramine levels in leaves of Arimar and CI 12020 can reach about10 mm on a plant-water basis (Figs. 1 and 2). In the absence ofgramine, calluses and plantlets of Arimar slightly but consistentlyoutgrew those of Proctor; this inherent genotype difference ingrowth rate accounts for the cultivar variance in Table VII.Gramine did not significantly affect growth ofcallus from embryosof Proctor and Arimar at 25°C or 32°C/27°C, although the

?AMINE internal gramine level in thoroughly rinsed calus was close to orabove the external level (Table VII). Nor did gramine inhibitautotrophic growth of Proctor and Arimar plantlets at 21 °C/ 16°Cdespite presence of gramine in the shoots at levels at least as high

MAMI as those in the media. In contrast, gramine depressed plantletgrowth significantly at 32°C/27°C and, at 10 mK, provokedchlorosis and necrosis; both the growth depression and the visiblesymptoms were more severe in Arimar than Proctor, which isreflected in the significant cultivar x gramine interaction term inTable VII.The sustained, normal proliferation of achlorophyllous callus

tissue exposed to both extemal and internal gramine indicateseither that (a) nonphotosynthetic tissues of barley are insensitiveto physiological concentrations of the native alkaloid and itscatabolites, or (b) if gramine and/or its catabolites are metabolicpoisons, then they are excluded from metabolic compartments.Whatever the mechanisms of tolerance to gramine, the inhibitionof autotrophic plantlet growth at 32°C/27°C-but not 21°C/

0-__ 16°C-shows that they lose effectiveness in green shoot tissue atIII _O high temperatures. Factors, alone or in combination, that could

15 25 35 45 contribute to this include high-temperature potentiation of gram-me toxicity, loss of compartmentation, and enhanced chemical or

(S AFTER PLANTING metabolic breakdown of gramine to toxic products. In any case,

on incorporation of label from I14Clformate autotoxic effects of internally produced gramine might be antici-entsofProctor firstleafbladessuppliedwith pated at high temperature; it is interesting that heat-inducedsgments cut from the central pat ofthe first chlorosis in the experiment of Figure 2 increased in the orderious ages were infiltrated first with AMI (10 roctor < rimar <with ['4CJformate (460 nCi/segment, 55 nCi Some inferences about compartmentation and toxicity ofgram-I d in darkness. Data points are means for ine may be drawn from its partitioning between a model organics, individual samples were within 2.3 nCi of phase (1-octanol) and dilute aqueous buffer solutions (Fig. 5);as repeated twice, with similar results. partitioning behavior was typical for a hydrophobic base. At high

pH when the tertiary amine function (pK = 10) was unprotonated,ations, with the second decaying faster partitioning was about 100:1 in favor of the organic phase. As pH

fell, gramine increasingly entered the aqueous phase, with a 1:1itioning Behavior of Gramine. Gramine distribution occurring at approximately pH 7.7. (It can be simplyheterotrophic callus growth and auto- shown that, at this 1:1 point, pH = pK - log K.w.) It follows that,1 and 10 mm (Table VII). Endogenous within a cell comprising aqueous compartments bounded by mem-

\

t, GI#4II

I GR

II



HANSON ET AL.

pHFIG. 5. Effect of pH upon the partitioning of gramine between organic

(1-octanol) and aqueous phases. K.. = [gramine mii/[grmineau.(, V), pH measured after partitioning was equilibrated; (0, A, E), pHmeasured on buffers before partitioning. Buffers: (0, 0), citrate-phosphate;(V), BES-NaOH; (A), glycine-NaOH; (0), Na2HPO4-NaOH.

branes (organic phases), gramine will tend to migrate to-andaccumulate in-the compartment at the lowest pH, possibly thevacuole. It also follows that, at pH values and tissue gramine

levels in the physiological range, the gramine concentration in cellmembranes themselves could be both significant and highly re-

sponsive to pH.

CONCLUSIONS

If gramine accumulation is a defensive measure against herbi-vory, then the presence of very high gramine levels at the first-leafstage-when the plant is most vulnerable to attack-conforms toa pattern common for chemical defense agents (25). The loss ofthe alkaloid accumulation trait among some cultivated types,without apparent deleterious effect on their performance as crops,

shows gramine to be a secondary compound whose toxic or

deterrent properties impart little benefit to plants grown underpresent agronomic conditions. This, too, conforms to a pattern inphytochemical ecology of domesticated species (13, 19). Themarked stimulation by high temperature ofgra ie accumulationin growing leaves is more unusuaL although various adverseenvironments oftentimes raise levels ofsecondary compounds (e.g.15, 19).Inasmuch as gramine is nontoxic at low or moderate tempera-

ture to the barley plant, it would seem an ideal defensive agent.However, the damaging effect of applied gramine on plantletgrowth at high temperature, and the more severe heat injurysymptoms of the gramine-rich cultivars indicate that autotoxicitycould offset any antifeedant value ofgramine in hot environments.Perhaps the variation in gramine level reported among genotypesof H. spontaneum, the wild progenitor of barley (12), reflects anevolutionary compromise between the probabilities of encounter-ing high temperatures and herbivore attacks; an analogous trade-


off between cyanogenesis and low-temperature injury has beenproposed by Daday (5).

If the interpretations above are correct, it follows that in agri-cultural environments gramine accumulation has only potentiallydeleterious consequences-both for the productivity of livestockgrazing barley and for the heat tolerance of the crop itself. Thatthe almost gramine-free cv Proctor is apparently blocked at onlyone step in the synthesis pathway is consistent with control overthe presence or absence of gramine by a single major gene.Selection against gramine in a plant-breeding program might beworthwhile and straightforward.

Acknowledgments-We thank Dr. Jerry D. Cohen for synthesizing AMI, and Dr.Kenneth L. Poff and Ms. Jan Watson for help with fluorescence assays.

LITERATURE CITED

1. ARNOLD GW, JL HnIL 1972 Chemical factors affecting selection of food plantsby ru ts. In JB Harborne, ed, Phytochemical Ecology. Academic Press,New York, pp 71-101

2. BiAu K, W ROBSON 1957 A modified Folin method for estimating primary andsecondary amines. Chem Ind 1957: 424

3. BRANDT K, H VON EuLER, H HETROM, N LOFGREN 1935 Gramin und zweiBegleiter desselben in Laubblattern von Gerstensorten. Z Physiol Chem 235:37-42

4. Busutrr D, LW AUDUS 1964The use of fluorimetry in the estimation of naturally-occurring indoles in plants. Phytochemistry 3: 395-415

5. DADAY H 1965 Gene frequencies in wild populations of Tri'olium repens L. IV.Mechanism of natural selection. Heredity 20: 355-365

6. DIGENIs GA 1969 Metabolic fates of gramine in barley. I: Mechanism ofincorporation of gramine into tryptophan in barley shoots. J Pharm Sci 58: 39-42

7. GAMBORG OL, LR WETTER 1975 Plant Tissue Culture Methods. NationalResearch Council of Canada, Prairie Regional Laboratory, Saskatoon, Sas-katchewan

8. GowER BG, E LEETE 1963 Biosynthesis of gramine: the immediate precursors ofthe alkaloid. JAm Chem Soc 85: 3683-3685

9. GROss D, H LEHmANN, H-R ScHUTTE 1970 Zur Physiologie der Graminbildung.Z Pflanzenphysiol 63: 1-9

10. GRoss D, H LEHMANN, H-R ScHUTTE 1974 Zur Biosynthese des Gramins.Biochem Physiol Pflanzen 166: 281-287

11. HANSON AD, NA ScoTT 1980 Betaine synthesis from radioactive precursors inattached, water-stressed barley leaves. Plant Physiol 66: 342-348

12. HANSON AD, PL TRAYNOR, KM Dmrz, DA REiCOSKY 1981 Gramine in barleyforage-effects of genotype and environment. Crop Sci 21: 726-730

13. HARoRNE JB 1972 Phytochemical Ecology. Academic Press, New York14. KARICK.HOFF SW, DS BROWN 1979 Determination of octanol/water distribution

coefficients, water solubilities, and sediment/water partition coefficients forhydrophobic organic pollutants. EPA Report 600/4-79-032, EnvironmentalResearch Laboratory, Office of Research and Development, U.S. Environmen-tal Protection Agency, Athens, GA

15. MAJAK W, RE McDIARMID, TW POWELL, AL VAN RyswyiK, DG STOUT, RGWILLiAMs, RE TucKER 1979 Relationships between alkaloids in reed canary-grass (Phalaris arundinacea), soil moisture and nitrogen fertility. Plant CellEnviron 2: 335-340

16. MARTEN GC 1973 Alkaloids in reed canarygrass. In AG Matches, ed, Anti-Quality Components of Forages. Crop Science Society of America, Madison,pp 15-31

17. MARTEN GC, RM JORDAN, AW HOVIN 1981 Improved lamb performanceassociated with breeding for alkaloid reduction in reed canarygrass. Crop Sci21: 295-298

18. MARUM P, AW HOVIN, GC MARTEN 1979 Inheritance of three groups of indolealEaloids in reed canarygrass. Crop Sci 19: 539-544

19. MATCHES AG (ed) 1973 Anti-Quality Components of Forages. Crop ScienceSociety of America, Madison

20. MUDD SH 1961 3-Aminomethylindole and 3-methylaminomethylindole: newconstituents of barley. Nature 189: 489

21. OLEN CR, MN SsssH 1977 Ice adhesions in relation to freeze stress. PlantPhysiol 60: 499-503

22. OVERLAND L 1966 The role of allelopathic substances in the "smother crop"barley. Am J Bot 53: 423-432

23. ScHNEIDER EA, RA GIBSON, F WIGHTMAN 1972 Biosynthesis and metabolism ofindol-3yl-acetic acid. I. The native indoles of barley and tomato shoots. J ExpBot 23: 152-170

24. SCHNEIDER EA, F WIGHTMAN 1974 Amino acid metabolism in plants. V. Changesin basic indole compounds and the development of tryptophan decarboxylasein barley (Hordeum vulgare) dunng germination and seedling growth. Can JBiochem 52: 698-705

25. SWAIN T 1977 Secondary compounds as protective agents. Annu Rev PlantPhysiol 28: 479-501

904

2

1

0:0he 0

I I I I I I

p- ~~~~~~~/

/iA

AlI -

I

4

- 1--I

/.

1 a I I I

-1

-2

2 4 6 8 10 12



gramineaccumulation in leaves barleygrownunderhigh ...plantphysiol. vol.71, 1983 8h.1% 4-. 39 a-d,i...

Documents