Chem. Res. Toxicol. 1991,4, 237-240
237
Metabolism of L-Canavanine and L-Canaline in the Tobacco Budworm, Heliothis virescens [Noctuidae] Milan A. Berget and Gerald A. Rosenthal**ti* The Graduate Center for Toxicology and The T . H . Morgan School of Biological Sciences, University of Kentucky, Lexington, Kentucky 40506-0054 Received November 15, 1990 The metabolism of L-canavanine and L-canaline were investigated in larvae of the tobacco budworm, Heliothis virescens [Noctuidae]. H. virescens larvae were treated with L-[ 1,2,3,414C]canavanineor ~-[U-l~C]canaline with sufficient cold carrier to provide 5 mg g-' canavanine or a molar equivalent of canaline (3.81mg g 3 . The preponderant catabolite in both canavanineand canaline-treated larvae was [ 14C]homoserine. Other minor metabolites derived from canavanine included [ 14C]aspartate/asparagine, [14C]glutamate/glutamine, [ 14C]2-aminobutyrate, [ 14C]ornithine, [14C]proline,and [14C]isoleucine. Canaline yielded [14C]glutamate/glutamine, [ 14C]aspartate/asparagine, and [14C]-2-aminobutyrate, Our current studies support the belief that this destructive insect tolerates L-canavanine and L-canaline because of its ability to reductively cleave these potentially insecticidal natural products to L-homoserine and guanidine or ammonia, respectively.
Introduction
(11). Our results demonstrate that no more than 9% of the administered canavanine is catabolized to canaline and [14C]urea(11). On the other hand, this insect converts about 70% of the administered ~-[guanidinooxy-"C]canavanine to [14C]guanidine(11). The presence of the latter reaction pathway in a eukaryote is unusual. Such distinct eukaryotes as higher plants (12,13), canavanine-utilizing insects (14), and the rat (15) commonly utilize arginase for canavanine degradation. Only procaryotes use a pathway that relies upon an enzyme other than arginase for canavanine detoxification (5). Thus, H.uirescens is the first eukaryote identified that employs a detoxification system not based upon arginase. To achieve a basic understanding of canavanine detoxification in H. uirescens and to compare canavanine and canaline degradation, we synthesized L-[ 1,2,3,4-"C]canavanine and ~-[U-'~C]canaline. The use of these radiolabeled compounds permitted us to ascertain the metabolic disposition of the aliphatic carbons of both canavanine and canaline.
L-Canavanine,~-2-amino-4-(guanidinooxy)butyric acid, is one of over 250 nonprotein amino acids found in higher plants (I). This structural analogue of L-arginine serves as a reactant in any enzymatic reaction where arginine is a substrate, including activation and aminoacylation by arginyl-tRNA synthetase (2). These arginyl-tRNA synthetase mediated reactions ultimately produce structurally (3) and functionally ( 4 ) aberrant canavanine-containing proteins. Canavanine can cause antimetabolic effects in viruses, bacteria, and other plants and animals (5), and it can play an important role in higher plants chemical defense against herbivores, especially insects (5, 6). While nearly all insects are sensitive to canavanine's toxic effects, there are two insects that have demonstrated resistance to its toxicity: a bruchid beetle, Caryedes brasiliensis, and a weevil, Sternechus tuberculatus (6). Larvae of these canavanine-utilizing insects feed exclusively on canavanine-producing plants and avoid such deleterious effects of canavanine consumption as aberrant Materials and Methods protein production (7). Indeed, C.brasiliensis has become so intimately dependent on its canavanine-containinghost Insects. H.uirescens larvae, obtained from a continuous colony maintained at the University of Kentucky, were reared as deplant that it utilizes canavanine as a dietary source of scribed previously (9). Drugs were administered by parenteral nitrogen and carbon (8). injection, using a volume equivalent to no more than 5% fresh The tobacco budworm, Heliothis uirescens, is also rebody weight (v/w). The injected larvae, 260-300 mg fresh weight, sistant to high levels of dietary canavanine (9). This rewere no more than 6 h into the second day of the terminal larval sistance is notable in that this destructive pest does not instar. feed on any mature plant that contains detectable canaChemicals. canavanine (free base) was isolated from acevanine (10). H. virescens exhibits an LCm for canavanine tone-defatted jack bean seeds, Canaualia ensiformis, by ion-exof 300 mM or approximately 40% dietary canavanine by change chromatography and purified by repetitive crystallization dry weight (9). The larvae neither excrete nor sequester (16). L-Canaline (free base) was prepared by the method of Rosenthal (17). canavanine; parenterally injected canavanine is cleared ~-[U-"C]Canalinewas produced from ~-[U-~'C]homoserine. from the hemolymph with a tllP of 135 min (9). This preparation involved the successive synthesis of ~ - 2(ben4 The common pathway for L-canavanine degradation is zyloxycarbonyl)amino]-4-hydroxybutyric acid, ~ - 2 -(benzyloxy[ by catalytic hydrolysis conducted by arginase (EC 3.5.3.1) carbonyl)amino]-4-butyrolactone,benzyl L-2-[ (carbobenzyloxy)to yield L-canaline and urea. We evaluated the relative amino]-4-hydroxybutyrate, and benzyl L-2-[ (carbobenzyloxy)importance of arginase to H. virescens in canavanine amino]-4-[ (p-tolylsulfonyl)oxy]butyrate. The aminooxy function metabolism by utilizing ~-[guanidinooxy-~~C]canavanine was then introduced with benzohydroxamic acid to form benzyl *To whom correspondence should be addressed. 'The Graduate Center for Toxicology. *The T. H. Morgan School of Biological Sciences.
~-2-[(carbobenzyloxy)amino]-4-(benzamidooxy)butyrate.The latter compound was deprotected by refluxing with 19% (w/v) ethanolic HCl and 3 N HCl to form ~-[U-"C]canaline(18). L[ 1,2,3,4-14C]Cana~anine was prepared from ~-[U-l~C]canaline and
0893-228x/91/2704-0237$02.50/00 1991 American Chemical Society
238 Chem. Res. Toxicol., Vol. 4, No. 2, 1991 characterized as described by Rosenthal and Berge (13). L[~g~anidinooxy-'~N]Canavanine was prepared by the method of Rosenthal et al. (12). Scintillation medium (Ecolume) was purchased from ICN, Irvine, CA. All other chemicals were obtained from Aldrich Chemical Co., Milwaukee, WI. L-[ 1,2,3,4-'qC]Canavanine a n d ~-[U-'%]Canaline Treatment a n d S a m p l e Preparation. Fifty H. virescens larvae were each injected with 7.2 kBq of L-[ 1,2,3,4-14C]canavaninewith sufficient cold carrier to provide 5 mg/g canavanine fresh body weight. The 50 larvae were divided into 10 groups; each group (n = 5) was processed as a single sample. All determinations were done in duplicate. One pooled group of five treated larvae were sacrificed 0, 135, 270, 405, or 810 min postinjection. These experimental times were equivalent approximately to 0, 1, 2, 3, or 6 half-lives for canavanine in H. uirescens (9). Each treatment group was placed in a 30-mL metabolic cage immediately after injection, and respiratory [14C]C02was collected. Air, drawn from the cage a t a rate of 6 mL/min under vacuum, was conveyed into a C 0 2 trap containing 5 mL of methoxyethanol and 2-aminoethanol[2:1 (v/v)]. Samples of the trapping solution (3 X 100 pL) were quantitated by liquid scintillation spectroscopy. All treated larvae were stored at -60 "C. The injection site of the treated larvae was sealed with bee's wax to prevent hemolymph leakage. However, the treated larvae often responded to the injection trauma by regurgitation that produced an oral exudate. To account for the loss of administered drug as oral exudate, the metabolic cage was rinsed repeatedly with water (final volume 5 mL); three 500-pL samples of the combined water rinses were quantitated by liquid scintillation spectroscopy. In canaline-treated larvae, 50 H. virescens larvae were each injected with 8.4 kBq of ~-[U-'~C]canaline with sufficient cold carrier (3.81 mg/g) to provide the molar equivalent of the canavanine dosage. The larvae, divided into 10 groups as described above, were manipulated identically with those larvae receiving canavanine with the exception that they were sacrificed a t 0,45, 90, 135, or 270 min postinjection. These exposure times were equivalent to 0, 1, 2, 3, or 6 half-lives of canaline in H. uirescens (9). Each group of frozen, pooled larvae (n = 5) was ground in excess acetone with a Sorvall Omni-mixer a t full speed for 20 s. The homogenate was filtered and the resulting acetonedefatted powder allowed to dry overnight a t 22 "C. The acetone-defatted powder was extracted by mechanical stirring for 24 h with 150 mL of 50% aqueous ethanol containing 0.1 M sulfuric acid. The extract was centrifuged for 20 min at lOOOOg and the supernatant solution evaporated by rotary evaporation in vacuo. The resulting residue was dissolved in a minimum amount of deionized water, treated with sufficient saturated Ba(OH)2 to bring the pH to 4.0, and placed overnight at 4 "C. The resulting BaSO, was removed by centrifugation as described above, and the supernatant solution was filtered. The filtrate was taken to dryness by rotary evaporation in vacuo, the residue dissolved in a minimum amount of water, and the solution adjusted to pH 3.2 with 1 N HCl. The sample was applied to a 25 X 70 mm column of Dowex-50 (H+), washed with 0.8 L of deionized water, and eluted with 0.5 L of 200 mM NH,. The amino acid containing effluent was filtered, concentrated by rotary evaporation in vacuo, and stored at -60 "C. Homoserine Lactonization. The amino acid containing samples were processed chemically to permit separation of homoserine from glutamine prior to automated amino acid assay. Glutamine and homoserine were not separated by the commercial resin used in our automated amino acid analyzer. To overcome this deficiency, the samples were treated chemically to convert homoserine to its lactone, a compound sufficiently more basic to be resolved readily from glutamine. This procedure also converted glutamine and asparagine to their corresponding acid. Thus, our results were expressed as glutamine/glutamic acid and asparagine/aspartic acid. The thawed, amino acid containing fraction was dissolved in 9 mL of 19% (w/w) ethanolic HCl and refluxed for 90 min a t 105 "C. This sample was dried by rotary evaporation in vacuo, deionized water added, and the drying process repeated twice. The residue was refluxed in 9 mL of 3 N HCI for 90 min at 115 "C. This homoserine lactone containing sample was dried by
Berge and Rosenthal rotary evaporation in vacuo; the drying process was repeated after deionized water addition until excess HCl was removed fully. The residue was dissolved in deionized water, adjusted to p H 3.2, applied to a 25 X 70 mm Dowex-50 (NH4+)column, washed with 0.8 L of water, and developed with 0.5 L of 200 mM NH3. The deionized water wash contained all the acidic and neutral amino acids except homoserine. The basic effluent contained the basic amino acids plus homoserine; homoserine lactone was delactonized in situ during column development. Both the deionized water wash and basic eluent were concentrated and assayed by automated amino acid analysis as well as HPLC after amino acid dansylation. Amino Acid Identification. Identification of the 14C-labeled amino acids was established by ion-exchange chromatography with a Dionex D-300 automated amino acid analyzer, employing a lithium citrate buffer system. The column effluent was not reacted with ninhydrin. Rather, the spent column effluent, collected a t 2-min intervals, was assayed by liquid scintillation spectroscopy. Amino acid identification was substantiated by HPLC of the dansylated derivatives. Appropriate samples were dansylated (19) and the dansylated amino acids separated with a Beckman Model 324 gradient liquid chromatograph and a Altex Ultrasphere-ODS C,, column (4.6 X 250 mm) using the method of Weiner and Tishbee (20). The effluent, collected a t I-min intervals, was assayed by liquid scintillation spectrometry. Canaline degrades on contact with Dowex-50 cationic exchange resin as well as the commercial resin used in the Dionex D-300 column. Our analytical methods, therefore, could not measure unmetabolized canaline directly. However, interaction of the resin with canaline produced two stable, canaline-resin interaction products. Thus, while we could not assess canaline directly, analysis of these degradation products provided an accurate measure of unreacted canaline. No information is available presently on the chemical nature of these canaline-containing degradation products. ~-[N~,gu~lljdinooxy-~~N]Canavanine Treatment and S a m p l e Preparation. Thirty larvae were injected with 5 pmol After 12 and 24 h, half of ~-[N~guanidinooxy-'~N]canavanine. of the treated larvae were stored a t -60 "C. Each pooled group (n = 15) of ~-[Nguanidinooxy-~~N]canavanine-treated larvae was processed in the same manner as described for [14C]canavanineand ["Clcanaline-treated larvae until the amino acids were eluted from the Dowex-50 (H+) column. The amino acid containing effluent was concentrated by rotary evaporation in vacuo, treated with decolorizing charcoal, filtered, lyophilized, and stored at -60 "C. Mass Spectroscopy. The lyophilized amino acids were derivatized to their N-(trifluoroacetyl) n-butyl ester forms and analyzed by GC-MS as described previously (8).
Results Metabolism of Radiolabeled Canavanine. H.virescens effectively metabolized L-[ 1,2,3,4-14C]canavanine; only 4.5% of the radiolabeled canavanine r e m a i n e d after approximately 6 half-lives (Table I). In fact, after 6 half-lives about 1.5% of the original canavanine would remain. The majority of the 14C-labeleddegradation products occurred in the charged, water-soluble fraction. This fraction consisted of unmetabolized [ 14C]canavanine and six other [ 14C]amino acids that i n c l u d e d the following:
[14C]-
homoserine, [ 14C]aspartate/asparagine, [14C]glutamate/ glutamine, [14C]-2-aminobutyrate, [14C]ornithine, [14C]proline, and [ 14C]isoleucine. [ 14C]Homoserinewas the most a b u n d a n t metabolite of t h i s group. The acetone-soluble fraction, containing metabolites such as triglycerides and fatty acids, contained much less of t h e 14C-labeled degradation products. The 14Ccontent of t h i s fraction r a n f r o m 5% to 8% over the e n t i r e exp e r i m e n t a n d r e m a i n e d relatively steady. The neutral, water-soluble fraction representing the 14C converted to metabolic products such as sugars ranged from 5% to 7%. Both of these fractions m a i n t a i n e d a relatively c o n s t a n t level throughout the experiment. Finally, over 6 half-lives,
Metabolism of L-Canavanine and L-Canaline Table I. Metabolism of L-[ 1,2,3,4-14C]Canavanine in H. virescens % "C recovered' fraction or substance 135 min 270 min 405 min 810 min 5.1 14.5 I. respiratory I4CO2 0.7 3.1 7.8 5.4 11. acetone-soluble 6.5 7.2 fraction 111. water-soluble fraction 5.8 6.5 5.6 4.9 a. neutral b. charged 2.4 0.4 2.2 aspartate/asparagine 0.6 7.0 6.3 4.3 3.6 glutamate/glutamine 0.8 1.1 0.2 0.4 proline 5.2 3.1 3.7 4.6 2-aminobutyrate 2.9 2.1 1.6 1.3 isoleucin 10.4 28.6 39.3 38.2 homoserine 9.5 4.3 4.3 4.2 ornithine 60.3 30.8 14.8 4.5 canavanine 76.5 67.2 85.3 74.0 total 95.0 92.0 IV. total 14Crecovered 98.3 90.8 ~
a Each value is the mean of 2 independent determinations with a pooled group of 5 insects.
Table 11. Metabolism of ~-[U-"C]Canaline in Heliotbis virescens % "C recovered' fraction or substance 45 min 90 min 135 min 270 min I. respiratory "Cop 0.0 0.1 0.1 0.5 11. acetone-soluble fraction 6.7 6.0 5.5 4.7 111. water-soluble fraction a. neutral 4.7 13.1 23.4 22.4 b. charged aspartate/asparagine 0.1 0.3 0.3 0.4 glutamate/glutamine 2.9 13.3 16.7 18.8 2-aminobutyrate 0.4 0.7 0.9 1.2 homoserine 30.9 31.4 36.3 43.9 26.6 13.4 3.8 canalineb 45.6 total 79.9 72.3 67.6 68.1 unknown 1 25.8 12.9 5.5 1.7 unknown 2 19.8 13.7 7.9 2.1 IV. total 14C recovered 91.3 91.5 96.6 95.7 ~~
Each value is the mean of 2 independent determinations conducted with a pooled group of 5 insects. Canaline was monitored as the canaline-Dowex-50 resin degradation product as described under Materials and Methods. (I
H . virescens converted 14.5% of the injected [14C]canavanine to 14C02. Metabolism of Radiolabeled Canaline. H.uirescens metabolized 96% of the injected ~-[U-'~C]canaline in approximately 6 half-lives (270 min). As with canavanine, a majority of the 14C-labeled,degradation products from canaline-treated larvae were charged and water-soluble (Table 11). The charged, water-soluble fraction contained four 14C compounds: [ 14C]homoserine, [ 14C]glutamate/ glutamine, [ 14C]aspartate/asparagine,and [ 14C]-2-aminobutyrate. Unlike canavanine, canaline did not support the production of [14C]ornithine, [14C]proline, or [14C]isoleucine. Canaline-treated insects converted only 0.5% of the injected 14Cto 14C02in about 6 half-lives (270 min). The 14Ccontent of the acetone-soluble fraction obtained from canaline-treated H.virescens larvae ranged from 5 % to 7%. The neutral water-soluble fraction was higher in canaline- as compared to canavanine-treated larvae since the neutral water-soluble 14Ccontent reached 22% of the injected dose of 6 half-lives (270 min). The above in vivo studies of L-canavanine degradation revealed that L-homoserine is the principle product of canavanine catabolism. Homoserine could be produced either by reductive cleavage or by hydrolysis of canavanine. These catabolic reactions would also generate guanidine or hydroxyguanidine, respectively. Analysis of the me-
Chem. Res. Toxicol., Vol. 4, No. 2, 1991 239 tabolism of L-[guanidinoo~y-'~C]canavanine disclosed that guanidine was the principal radiolabeled reaction product. No [ 14C]hydroxyguanidine was found in L- [guanidino~xy-'~C]canavanine-treatedlarvae (11). To determine if the guanidinooxy nitrogens of canavanine, released BS guanidine, were diverted into biosynthetic pathways yielding a useful product such as an amino acid, and aswe prepared L-[N,guanidin~oxy-~~N]canavanine sessed the capacity of the heavy nitrogen to support amino acid synthesis by H. uirescens. Our mass spectrometric analysis of the isolated amino acids of L-[Nguanidinoo~y-'~N]canavanine-treated larvae failed to disclose any significant incorporation of 15Ninto de novo synthesized amino acids. Analysis of the frass of canavanine-treated larvae failed to disclose significant guanidine. Thus, this nitrogen-rich canavanine degradation product appeared to be neither metabolized nor excreted but rather was sequestered apparently within the larval hemolymph.
Discussion L-[ 1,2,3,4J4C]Canavanine and ~-[U-'~C]canaline were injected into H . uirescens larvae and their metabolic products determined. This investigation represents the first study of the fate of the aliphatic carbons of canavanine as well as canaline in a living organism. In both [14C]canavanine-(Table I) and [14C]canaline(Table 11) treated insects, [14C]homoserineis the preponderant product of canavanine catabolism. [14C]Canavanine additionally produces small but detectable levels of [14C]glutamate/glutamine, [14C]aspartate/asparagine, [14C]-2-aminobutyrate,[14C]ornithine, [14C]proline,and [14C]isoleucine. It is not known presently how these catabolic products are formed, but some of these amino acids can be converted to 2-oxo derivatives which can then enter the Krebs cycle. Such metabolic events would account for the observed radiolabeled C02. [ 14C]Canalir>.e also produces small but detectable amounts of [ 14C]glutamate/glutamine,[ 14C]aspartate/asparagine, and [14C]-2-aminobutyrate. Very little C 0 2 is formed. The latter finding may be explained by canaline's potent ability to inhibit a wide range of aminotransaminases and decarboxylases due to its ability to bind with the pyridoxal phosphate moiety of these enzymes (21). The 14C-containing,neutral water-soluble fraction is higher significantly than the same fraction is in their canavanine-treated larval counterparts. While we do not have an explanation, this observation makes it evident that this insect utilizes distinct degradation pathways in processing these two nonprotein amino acids. In both [14C]canavanine- (Table I) and [14C]canaline(Table 11) treated insects, [14C]homoserineis the primary product of canavanine catabolism. Since homoserine is the preponderant product of both canavanine and canaline catabolism, the present evidence suggests that H. virescens is drawing upon a reductase that can act upon the oxygen-nitrogen linkage of the guanidinooxy or aminooxy group to yield homoserine and guanidine or ammonia, respectively.
H2NC(NH2)=NOCHZCH,CH(NH2)COOH
+
L-canavanine HOCH2CH2CH(NH2)COOH + HPNC(NHZ)=NH L-homoserine guanidine
-
H2NOCH2CH2CH(NH2)COOH L-canaline HOCH2CH2CH(NH,)COOH + NH3 L- homoserine
240 Chem. Res. Toxicol., Vol. 4, No. 2, 1991
L-Canavanine and L-canaline can be highly deleterious to insects (22,23);yet, H. uirescens tolerates remarkably high levels of these normally protective natural products (9). Possession of a reductase able to convert canavanine and canaline to homoserine provides this insect with a biochemical mechanism for the effective detoxification of these pernicious substances. We are currently isolating and characterizing the canavanine-metabolizing enzyme of the larvae. The failure of H. uirescens to incorporate any of L[~gu~nidinooxy-'~N]canavanine's I5N into an amino acid is not unexpected. This finding is compatible with previous metabolic studies, employing L-[guanidinooxy14C]canavanine(1I), which established that the guanidinooxy moiety is converted to guanidine and/or urea. Since the heavy nitrogen atom is part of the guanidinooxy moiety, the labeled nitrogen is transferred to guanidine or urea. There was no evidence of guanidine metabolism by H.uirescens, and these larvae contain only trace urease activity (6). Thus, the 15N atom was inaccessible and unusable to support the nitrogen metabolism of the larvae.
Acknowledgment. This work was supported by National Science Foundation Grant DCB-8901749. Registry No. L-Canavanine, 543-38-4;L-canaline, 496-93-5; arginase, 9OOO-96-8; L-homcwerine, 672-15-1; guanidine, 113-00-8; ammonia, 7664-41-7.
References (1) Rosenthal, G. A. (1982)Plant Nonprotein Amino and Imino Acids. Biological, Biochemical, and Toxicological Properties,
Academic Press, New York. (2) Mitra, S. K.,and Mehler, A. H. (1967)The arginyl transfer ribonucleic acid synthetase of Escherichia coli. J . Biol. Chem. 242,5490-5494. (3) Rosenthal, G. A., Reichhart, J.-M., and Hoffmann, J. A. (1989) L-Canavanine incorporation into vitellogenin and macromolecular conformation. J . Biol. Chem. 264, 13693-13696. (4) Rosenthal, G. A., Lambert, J., and Hoffmann, D. (1989)L-Canavanine incorporation into protein can impair macromolecular function. J . Biol. Chem. 264,9768-9771. (5) Rosenthal, G. A. (1977)The biological effects and mode of action of L-canavanine, a structural analogue of L-arginine. Q.Reu. Biol. 52, 155-178. (6) Bleiler, J., Rosenthal, G. A,, and Janzen, D. H. (1988)Biochemical ecology of canavanine-eating seed predators. Ecology 69, 427-433. (7) b e n t h a l , G. A., Berge, M. A., Bleiler, J. A., and Rudd, T. (1987) Avoidance of aberrant protein production and an organism's
Berge and Rosenthal ability to utilize or tolerate L-canavanine. Experientia 43, 558-561. (8) Rosenthal, G.A., Hughes, C. G., and Janzen, D. H. (1982)LCanavanine, a dietary nitrogen source for the seed predator, Caryedes brasiliensis (Bruchidae). Science 217,353-355. (9) Berge, M.A., Rosenthal, G. A., and Dahlman, D. L. (1986) Tobacco budworm, Heliothis uirescens [Noctuidae], resistance to L-canavanine, a protective allelochemical. Pestic. Biochem. Physiol. 25,319-326. (10) Lincoln, C. (1972)Distribution, abundance and control of Heliothis spp. in cotton and other host plants. In Southern Cooperative Series Bulletin 169, pp 2-7, Oklahoma Agricultural Experiment Station, Oklahoma State University, Stillwater, OK. (11) Berge, M.A., and Rosenthal, G. A. (1990)Detoxification of L-canavanine by the tobacco budworm, Heliothis uirescens [Noctuidae]. J . Agric. Food Chem. 38,2061-2065. (12) Rosenthal, G.A., Berge, M., Ozinskas, A., and Hughes, C. H. (1988)Ability of L-canavanine to support nitrogen metabolism in the jack bean, Canaualia ensiformis (L.) DC. J . Agric. Food Chem. 36, 1159-1163. (13) Rosenthal, G.A., and Berge, M. (1989)Catabolism of L-canavanine and L-canaline in the jack bean, Canaualia ensiformis (L.) DC. J . Agric. Food Chem. 37,591-595. (14) Rosenthal, G.A., and Janzen, D. H. (1983)Arginase and L-canavanine metabolism by the bruchid beetle, Caryedes brasiliensis. Entomol. Exp. Appl. 34,336-337. (15) Thomas, D. A., and Rosenthal, G. A. (1987)Metabolism of L-[guanidinooxy-"C]-canavanine in the rat. Toxicol. Appl. Pharmacol. 91,406-414. (16) Rosenthal, G. A. (1977)Preparation and colorimetric analysis of L-canavanine. Anal. Biochem. 77,147-151. (17) Rosenthal, G. A. (1973)The preparation and colorimetric analysis of L-canaline. Anal. Biochem. 51, 354-361. (18) Ozinskas, A. S.,and Rosenthal, G. A. (1986)Synthesis of Lcanaline and y-functional2-aminobutyricacid derivatives. J. Org. Chem. 51, 5047-5050. (19) Zanetta, J. P.,Vincendon, G., Mandel, P., and Gombos, G. (1970)The utilization of 1-dimethylaminonaphthalene-5-sulphonyl chloride for quantitative determination of free amino acids and partial analysis of primary structure of proteins. J. Chromatogr. 51, 441. (20) Weiner, S., and Tishbee, A. (1981)Separation of Dns-amino acids using reversed-phase high-performance liquid chromatography: a sensitive method for determining N-termini of peptides and proteins. J. Chromatogr. 213,501-506. (21) Katunuma, N., Okada, M., Matsuzawa, T., and Otauka, Y. (1965)Studies on ornithine keto acid transaminase. 11. Role in metabolic pathway. J . Biochem. (Tokyo) 57,445. (22)Rosenthal, G.A. (1988)Biochemical insight into the protective efficacy of L-canavanine, a toxic higher plant metabolite. Bioscience 38,104-109. (23) Rosenthal, G.A,, and Dahlman, D. L. (1990)Interaction of L-canaline with ornithine aminotransferase of the tobacco hornworm, Manduca sexta [Sphingidae]. J.Biol. Chem. 265,868-873.
