Alkaloids in Seneclo and Crotalaria

When tested for ornithine and arginine decarboxylases, pyrrolizidine alkaloid-bearing Senecio riddellii, S. longilobus (Compositae), and Crotalaria retusa (Leguminosae) plants exhibited only ornithine decarboxylase activity. This contrasts with previous studies of four species of pyrrolizidine alkaloid-bearing Heliotropium (Boraginaceae) in which arginine decarboxylase activity was very high relative to that of ornithine decarboxylase. Unlike Heliotropium angiospermum and Heliotropium indicum, in which endogenous arginine was the only detectable precursor of putrescine channeled into pyrrolizidines, in the species studied here using difluoromethylornithine and difluoromethylarginine as the enzyme inhibitors-endogenous ornithine was the main if not the only precursor of putrescine converted into the alkaloid aminoalcohol moiety. In S. riddeliji and C. retusa at flowering, ornithine decarboxylase activity was present mainly in leaves, especially the young ones. However, other very young organs such as inflorescence and growing roots exhibited much lower or very low activities; the enzyme activity in stems was negligible. There was no correlation between the enzyme activity and polyamine or alkaloid content in either species. In both species only free polyamines were detected except for C. retusa roots and inflorescence-with relatively very high levels of these compounds-in which conjugated putrescine, spermidine, and spermine were also found; agmatine was not identified by HPLC in any plant organ except for C. retusa roots with rhizobial nodules. Organor age-dependent differences in the polyamine levels were small or insignificant. The highest alkaloid contents were found in young leaves and inflorescence. Our previous study on PA'-bearing Heliotropium angiospermum and H. indicum (Boraginaceae) plants exposed to 14Clabeled CO2 proved that endogenous Put formed from an endogenous source is the precursor of the alkaloid aminoalcohol (necine) moiety. This confirms the biosynthetic pathway postulated from studies with exogenous Put or its possible precursors such as Orn, Arg, Spd, or Spm introduced into Senecio plants, as discussed in Birecka et al. (5). The effects ofDFMO and DFMA, specific enzyme-activated irreversible inhibitors of ODC and ADC, respectively, on necine radioactivity in the treated plants showed that endogenous Arg is the only detectable precursor of Put channeled into the necines of the two borages that exhibited not only ADC but also-although low-ODC activity. However, Abbreviations: PA, pyrrolizidine alkaloid; Put, putrescine; Spd, spermidine; Spm, spermine; Agm, agmatine; Om, ornithine; Arg, arginine; ODC, ornithine decarboxylase; ADC, arginine decarboxylase; DFMO, D,L-a-difluoromethylornithine; DFMA, D,L-a-difluoromethylarginine; Rt, retention time; TLE, thin layer electrophoresis. the same borages were able to decarboxylate exogenous Orn to Put that in turn was converted into necines. Thus, in vivo transformation of an exogenous putative precursor does not always indicate the compound formed in situ that normally serves as a precursor. This is especially true in the case of Put, which can derive not only from Orn but also from Arg via Agm, the product of Arg decarboxylation by ADC. Our attempt to identify the in situ precursor(s) of the Put channeled into pyrrolizidines in Senecio vulgaris were unsuccessful due to the extremely low PA content of the shoots. We report here the effects of DFMO and DFMA on necine biosynthesis in S. riddellii, S. longilobus (Compositae), and Crotalaria retusa (Leguminosae) plants. These species occasionally produce extremely high levels of PAs, up to 9 to 17% of leaf dry weight (23). In previously analyzed samples of S. riddellii and S. longilobus plants from New Mexico the PA contents were about 1.5 and 2.8%, respectively (10). Since there is little information about polyamines, including Put, ADC, and ODC activities, or the site(s) of alkaloid biosynthesis in PA-bearing plants, and since such information is available only for Heliotropium (5, 7-9), a more detailed examination of PA-bearing plants seemed important. Flowering S. riddellii and C. retusa were chosen for this work. MATERIALS AND METHODS Plant Material. Seeds of S. riddellii were collected near Tucumcari, NM; young plants of S. longilobus were collected at the same location in late June 1986. Seeds of C. retusa P1-274951 were obtained from the Plant Materials Center U.S.D.A., Brooksville, FL. The total alkaloid contents of the S. riddellii and C. retusa seeds were 1.4 and 2.1 %, respectively, when calculated as monocrotaline. At sowing, the seeds of C. retusa were inoculated with nitrogen-fixing bacteria, 'EL' inoculant from Nitragin, Clearwater, FL. At flowering, dense oblong nodules with leghemoglobin in the center were found on the roots. In addition, three Avena sativa cultivars were used: Garry, a cross of Victory x (Victoria x Hajira Banner), Ogle, and Victory. The seeds of the first two cultivars were supplied by the Carolina Biological Company and the Victory seeds by Svalof AB, International, Svalof, Sweden. The alkaloid-bearing plants were grown in the greenhouse as described previously (9); C. retusa plants did not receive N fertilizers. At the stage of main shoot flowering six plants each of S. riddellii and C. retusa were sampled in two replicates. The organs were separated, weighed, cut, and stored frozen for analyses of basic amino acid, polyamine, and alkaloid contents as well as for ODC and ADC activities. On the same day side shoots of S. riddellii, S. longilobus, and C. retusa10, 14, and 8 shoots per treatment, respectively-were exposed to H20, 2 mM DFMO, 2mM DFMA, or both inhibitors together as described previously (5). After 24 h, 2 to 4 shoots per treatment were sampled and 224 www.plantphysiol.org on October 1, 2017 Published by Downloaded from Copyright © 1988 American Society of Plant Biologists. All rights reserved. ORNITHINE DECARBOXYLASE AND PYRROLIZIDINE ALKALOIDS the remaining ones, with their stems in H20, 1 mM DFMO, 1 mM DFMA, or both inhibitors together, were exposed in light to 2.5 mCi of '4C-labeled CO2 in a Plexiglas chamber as described in Birecka and Catalfamo (8). After a 68 h exposure the shoots were sampled, cut, and frozen in two replicates. In addition, each of several shoots of S. riddellii and C. retusa was given 1 ml of 2 mm DFMO or DFMA labeled with about 1 ,uCi of D,L-a-[3,43H]DFMO or D,L-a-[3,4-3H]DFMA, respectively. After the solutions were absorbed (10-12 h) additional portions of 0.5 to 1.0 ml H20 were supplied to the test tubes; 24 h after the beginning of exposure the immersed part of the stem was cut 1.5 cm above the immersion level and sampled separately from the upper part of the stem and the leaves. Oat seedlings were grown in a mixture of vermiculite and peat humus. The first leaves were sampled on the 5th d after sprouting and assayed for ODC and ADC activities. Analysis. Extraction, quantitation, and purification of alkaloids, hydrolysis of labeled necine esters, recovery of labeled free necines and their separation by TLC as well as recrystallization were carried out as described in Birecka and Catalfamo (8). For basic amino acid and polyamine analyses plant tissues were homogenized with 6% HC104. A portion of the homogenate was centrifugated at 1 2,000g and the supernatant was tested for free amino acids and polyamines. Another portion ofthe homogenate was hydrolyzed with 6 N HCI (final concentration) in sealed ampules at 100°C for about 20 h. After centrifugation the hydrolyzate was dried in vacuo and the residue dissolved in 5% HC104. The amino acids were quantified by HPLC as described in Birecka et al. (6) using an ion exchange Whatman SCX column. Diaminopropane, Put, and cadaverine can also be separated on this column. The Rt values for Orn, Lys, Arg, diaminopropane, Put, and cadaverine were 15.6, 18.0, 22.2, 32.6, 35.4, and 40.8 min, respectively. Diaminopropane, Put, Spd, Spm, and cadaverine were analyzed by dansylation. The derivatives were prepared as follows: 0.2 ml of the extract was mixed with 0.15 ml of 1.5 M Na2CO3/0.05 M Na2B407 (pH 10.7) buffer, 0.01 ml of 0.1 mm diaminohexane (as an internal standard), and 0.9 ml of 1% dansyl chloride in acetone and the mixture was allowed to stand at room temperature in the dark overnight. Excess proline (0.075 ml of 15% solution) was added to the mixture and the acetone was removed under a stream of N2. After addition of 0.6 ml H20, 0.1 ml of 2 N NaOH, and 1 ml toluene the sample was vigorously mixed for 30 s and then centrifuged. An aliquot (0.7 ml) of the toluene layer was transferred to a clean tube and evaporated to dryness under N2. After addition of 0.7 ml of acetonitrile the dansyl-polyamines were analyzed by HPLC on a Beckman ODS-I column (4 x 250 mm, 5 juM) using a modification of conditions described by Brown et al. ( 12). The derivatives were detected with Kratos FS 950 fluorometer equipped with 365 nm excitation and 428 nm emission filters. The mobile phases consisted of buffer A, 0.02 M heptane sulfonic acid, and buffer B, acetonitrile. The column was equilibrated with a mixture containing 50% A and 50% B. After injection of a sample buffer B was increased to 80% in 22 min and then to 100% in 10 min where it was maintained for additional 20 min; the original conditions were then restored over 0.5 min. Another sample was injected after equilibration at starting conditions for 10 min. The Rt values for diaminopropane, Put, cadaverine, Spd, and Spm were 12.0, 12.6, 14.0, 22.8, and 29.4 min, respectively. Agm, which does not dansylate quantitatively, was assayed after postcolumn derivatization with o-phthaldialdehyde using an Ultrasphere ODS column as previously (6) and a buffer system described by Seiler et al. (27); the Rt ofAgm was 29.4 min. Soluble protein was assayed using Coomassie blue and BSA as the standard protein. The procedures used for assaying ODC and ADC activities were similar to those previously described for Heliotropium plants (6, 7). Spectrapor membrane tubings with mol wt cut off of about 8,000 were used for dialysis. The radioactivities of the substrates were 0.5 to 2 ,uCi per reaction mixture; the volume of the latter varied between 80 and 150 ,l. In the case of ADC assays the labeled Arg concentration in the reaction mixtures ranged from 0.1 to 0.5 mm. The use of polyvinylpolypyrrolidone (20, 21) during tissue homogenization did not affect the activities of the decarboxylases. TLE of HC104 extracts from shoots exposed to labeled DFMO and DFMA as well as of the reaction mixtures to determine [U-'4]Put and/or [U-'4C]Agm was performed as previously described (6). Liquid scintillation system 725, Nuclear Chicago, was used to measure the radioactivity (5). The differences between replicates in alkaloid, polyamine, and ODC levels did not exceed 13 to 18%. Chemicals. D,L-a-[3,4-3H]DFMO (32 Ci/mmol) and D,L-a[3,4-3H]DFMA (33.3 Ci/mmol) were obtained from New England Nuclear. The sources of the remaining labeled compounds were the same as previously (5). Except for DFMO and DFMA, which were synthesized at Merrell Dow Research Institute, all chemicals were obtained from Sigma.