6-D-Glucopyranosyl Fatty Acid Esters from Brassica Napus Pollen
Total Page:16
File Type:pdf, Size:1020Kb
Phyrochemisrry. 19i8. Vol. Ii. pp.1187-1189. © Pergamon Press Ltd. Printed in England. 0031-9422/78,0701-1187502.00,0 6-D-GLUCOPYRANOSYL FATTY ACID ESTERS FROM BRASSICA NAPUS POLLEN MICHAEL D. GROVE,* GAYLAND F. SPENCER,* PHILIP E. PFEFFER,t NAGABHUSHANAM MANDAVA,t J. DAVID WARTHEN, JR.t and JOSEPH F. WORLEyt * Northern Regional Research Center. Agricultural Research Service, U.S. Department ofAgriculture. Peoria, IL 61604: t Eastern Regional Research Center. Agricultural Research Service. U.S. Department of Agriculture. Philadelphia. PA 19118; t Beltsville Agricultural Research Center. Agricultural Research Service, U.S. Department of Agriculture. Beltsville. MD 20705. U.S.A. (Revised received 6 January 1978) Key Word Index-Brassica napus; Cruciferae: rape pollen: identification: glucosyl esters: fatty acids. Abstract-6-D-Glucopyranosyl esters of palmitic, oleic, linoleic and linolenic acids were identified in Brassica napus (rape) pollen. These esters are inactive as plant growth promoters in the bean second-internode bioassay. INTRODUCTION acids as oleic. linoleic and linolenic acids. The identitv of o-glucose in hydrolysates of esters subsequently In previous work, fractionation of Brassica nap tiS (rape) isolated by PLC was established by TLC, GLC and pollen gave a preparation called 'brassins' which exhibited colour reaction. novel plant growth-promoting activity, and which Individual glucosyl esters were separated by PLC contained a series of glucosyl esters of various fatty on AgN03-Si gel. Although the layers exhibited gray acids [1]. When bioassayed by the bean second-internode streaks indicating reduction of Ag ion, the reaction was method [2J, brassin preparations cause both elongation only partial. Because reduction of Ag ion was not and thickening of the internode. As the concentration of observed with synthetic 1- and 2-substituted glucosyl the active pr~lciple increases, swelling and curvature esters, an alternate site of fatty acid attachment in the of the stem are enhanced and often give rise to split natural material was indicated. Three bands were internodes. The esters \vere reported earlier [1 J to be removed which afforded the following major glucosyl I-glucosyl esters mainly on the basis of MS evidence esters in order of decreasing R (: palmitate, linoleate and obtained from a mixture. Individual esters were not linolenate. Alternatively. glucosyl esters could be sepa separated. and there was no correlation of biological rated by reversed phase HPLC on ,u-Bondapak C , 18 activity \vith any physical measurement [3]. As a con Esters were eluted in the order: linolenate, linoleate. tinuation of this research, we report here a series of new palmitate and oleate (contaminated with palmitate). 6-glucosyl esters from rape pollen which are distinct HPLC ofa biologically active fraction that also contained from the unidentified active principle, Esters identified as glucosyl esters resulted in elution of material that exhi 6-o-glucopyranosyl linolenate, palmitate, linoleate and bited brassin growth-promoting activity well before the oleate are inactive in the bean second-internode bioassay. elution of glucosyllinolenate. GC-MS of TMSi derivatives of natural glucosyl RESULTS AND DISCUSSION esters (see Experimental) gave ions for M + minus Me and/or one and two TMSiOH groups. In addition An iso-PrOH extract of H,O washed Brassica napus to acylium ions (RCO+) indicative of the fatty acid, (rape) pollen was fractionated by countercurrent distribu- ions corresponding to rearranged RCO,DMSi + were tion (CCD) and column chromatography on Si gel. also observed. The derivatized esters exhibited two The bean second-internode bioassay was used to follow GLC peaks of variable intensity representing the fractionation. Rechromatography of those fractions ex- anomeric mixture; the first peak due to the fl-anomer hibiting significant brassin-type activity afforded an (see below) was usually less intense. The MS of the first inactive major fraction which was eluted just prior G LC peak exhibited strong ions at m/e 217 (100 %) and to the growth-promoting material. The IR (Nujol) 204 (ca 50 %) while the second peak showed these spectrum ofthe inactive fraction showed ester absorption ions with the intensities reversed [m/e 217 (50 %), at 1730 cm - 1. Transesterification with NaGMe gave 204 (100 %)J. These ions represent three- and two Me esters which were analyzed by GLC on the basis carbon fragments from the carbohydrate ring [5]. of equivalent chain length. The following fatty acid In comparison, MS of synthetic TMSi-I-glucopyranosyl Me esters were characterized (percent composition palmitates showed the ion at m/e 217 three times stronger given in parentheses): 14: 0(0.6), 16: 0(39), 18: 0(0.8), than the 204 ion. 18: 1(3), 18:2(11.8), 18:3(44.8). Assignments were con- The 90 MHz PMR spectrum (CsDsN) of natural firmed by GC-MS which showed the appropriate glucosyl palmitate showed it to be an anomeric mixture M + for each ester. Double bond positions in unsaturated of 75 % C( (<5 5.88, J = 3.5 Hz): 25 % fl (<55.32, J = 7.0 Hz). esters were established by GC-MS of TMSi derivatives In contrast, anomeric protons for synthetic 1- and 2 of the alcohols obtained by per-hydroxylation with glucopyranosyl palmitates [6. 7J were observed as OsO4 [4]. These procedures identified the unsaturated follows: I-C(, <5 6.92 (J = 3.5 Hz): I-fl, <56.41 (J = 6.7 Hz): 1187 1188 Short Reports 2-y.:2-f3 (1:1 mixture), (j 6.10 (J=3.5Hz), (j 5.38 TLC, 95 ~io EtOH-toluene, 1: 3) just prior to the active material. (J = 8 Hz). The 220 MHz PMR spectrum of natural Combined inactive fractions from 6 kg pollen (230 mg) were glucosyl palmitate was compared with that of synthetic rechromatographed on Si gel (23 g) to give the glucosyl esters 6-y.-D-glucopyranosyl palmitate [8J. Except for dif (190mg). ferences due to anomeric composition, the spectra were Fatty acid analysis. A soln of glucosyl esters (40 mg) in identical. Substitution at C-6 was verified by the observa CHoClo-MeOH was treated with 0.5 M NOlO Me (2 ml) at 60' for 20 min. The soln was cooled, acidfied with 4 N HCl, diluted tion of the lovv field resonances of the C-6 methylene with HoO and extracted with hexane to give the Me esters: protons at (j 4.3 (Me CO-d ) in agreement with pre 2 6 IR (Cd.) 1740 cm- 1. no trans unsaturation. GLC analvsis viously reported values [9J. However, a detailed analysis was at 180' on a 3 m ~ 4 mrn column packed with 5% LAC-2 of the coupling constants could not be readily performed R446. GLC separation of TMSi derivatives of the alcohols due to the complexities given by the anomeric mixture obtained from OsO" per-hydroxylation of unsaturated Me and the degeneracy of the C-6 methylene proton reso esters [4J was on a 1.2 m x 4 mm column packed with 3 % nances. Heating the syntheticy.-anomer in C D N-D 0 OV-I programmed from 150 to 250' at 4'/min. s s 2 at 50' for 15 min gave a 60:40y.:f3-anomeric distribution Sugar identification. Samples of hexose esters from PLC that more closely resembled the natural material. (0.6 mg) in 2 N HCI (l ml) were heated under reflux for 1 hr. cooled, washed with hexane and evap. to dryness under red: Similarly, a sample consisting of at least 90 %y.-anomer pres. The residues were dissolved in H,O (30 pI) and applied was converted to a 50:50 mixture after 3 days in CsDsN to Lilly Tes-tape R urine sugar analysis paper to give an immedi at room temperature. GC-MS of the TMSi derivative ate positive reaction for D-glucose. The sugar migrated with of this mixture now showed a corresponding increase D-glucose on TLC using 0.1 NH 3B03-Si gel G (n-BuOH in the first GLC peak representing the f3-anomer and MeoCO-H,O: 4:5:1). GLC analvsis of the TMSi derivative exhibiting the intense mle 217 (100 ~~) ion. The mutarota on Ii 1.2 m'x 2 mm column pack~d with 3 ~!~ Dexsil 300 and tion of glucose in pyridine explains why 6-glucosyl programmed from 100 to 200' at 3'/min confirmed the presence palmitate was obtained as an anomeric mixture [1 OJ. of D-glucose. The PMR spectra of natural 6-glucosyl linoleate and Separation ofglucosyl esters. PLC: Solns of esters in MeOH CHCI3 (l :9) were applied to 0.5 mm thick layers of AgN03-Si Iinolenate likewise showed the predominant y.-anomeric gel G (5: 95) which were developed with EtOH-toluene (l: 3). protons at (j 5.88 (J = 3.8 Hz) and (j 5.87 (J = 3.5 Hz) The plates were dried in a stream of No and I em of each side respectively. The amount off3-anomer was notdetermined was sprayed with 0.2 % 2',7'-dichlorofluorescein in EtOH. because of proximity to the olefinic proton signals Three bands observed under long UV light were removed and centered at (j 5.49. extracted with MeOH-CHCI3 (l :9). Each extract was re· These esters are new natural products, but are closely chromatographed on a Si gel column (I g) with EtOH-toluene related to the steryl-f3-D-( 6-0-fattyacyl)-glucopyranosides (l :9). Glucosyl linolenate and linoleate were obtained as oils. which are well-known plant constituents [11 J. Although Combined samples of glucosyl palmitate were subjected to we were able to separate the brassin growth-promoting PLC on Si gel G. Material migrating 5.5-9 em from the origin was extracted and crystallized from EtOH :mp 129-132': mmp activity from the reported 6-glucosyl esters, we were \vith synthetic 6-y.-D-glucopyranosyl palmitate was not de unable to confirm by PMR the presence of 1-glucosyl pressed.