Isolation and characterization of maize flavonoids

Efforts to identify the flavonoids in maize started in the early twenties when C. E. Sando and H. H. Bartlett (J. Biol. Chem. 54:629-654, 1922) reported the presence of isoquercitrin in extracts of brown husked (a B Pl) maize. Later, C. E. Sando, R. T. Milner and M. S. Sherman (J. Biol. Chem. 109:203-211, 1953) identified chrysanthemin from purple husked (A B Pl) maize. Since that time, interest has been concerned mainly with the anthocyanins, and in many cases reports have indicated only the aglycone base. Recently, characterizations of flavonoids in silks were made by C. S. Levings and C. W. Stuber (Genetics 69:491-498, 1971), and J. Trautman and B. C. Mikula (MGCNL 49:29-32, 1975) reported spectral data for a number of compounds in different tissues, although some of their tentative identifications did not agree totally with their spectral data.

To understand fully the factors controlling maize flavonoid synthesis, all of the flavonoids in maize should be completely characterized. Ideally this requires isolation and purification of enough of each compound to allow characterization by UV and NMR spectra. This task is made simpler if the genotype and tissue of the plant is chosen to maximize the yield of the desired compounds and minimize the yield of other flavonoids that might interfere with their purification.

We have so far identified two compounds by the following procedure: Dried husks from a B Pl P-WW plants (which accumulate mostly flavonol glycosides because the flavone pathway is blocked by the P-WW allele) were extracted in 80% MeOH. The extract was concentrated under reduced pressure; taken up in boiling water for treatment with Celite Analytical Filter Aid; filtered under vacuum and then extracted with ethyl acetate (F. W. Collins and B. A. Bohm, Can. J. Botany 52:307-312, 1974). An additional extraction of the aqueous extract by BuOH improved the yield of the polar flavonoids (G. A. Guppy and B. A. Bohm, Biochem. Syst. Ecol. 4: 231-234, 1976). Initial separation was made on a Sephadex LH-20 column using a linear elution gradient of increasing concentration of methanol in water. Each fraction taken from the Sephadex column was subjected to a further separation on a Polyamid SC-6 column using a linear elution gradient consisting of ethyl acetate: butanone:90% MeOH operated from 50:50:10 to 50:50:100. Final purifications of the compounds were achieved by repeated TLC on Polyamid DC-6.6, using dichlorethane: butanone:H2O:MeOH:HOAc (50:20:5:20:5), on Avicel microcrystalline cellulose, using 15% acetic acid in water or ethyl acetate:formic acid:H2O (10:2:3). Hydroxylation and glycosylation patterns were determined by inspection of Rf, analysis of UV spectra (T. J. Mabry et al., The Systematic Identification of Flavonoids, Springer, New York, 1970) and identification of acid hydrolysis products. Acid hydrolysis was performed using 0.1 N HCl at 1000 C for 1 hr. Aglycones were identified by Rf on Polyamid DC-6.6 TLC using chloroform:butanone:MeOH (70:15:15) and by color in UV light after spraying with 0.1% diphenylboric acid aminoethyl ester in MeOH:H2O (1:1). Sugars were chromatographed on Avicel microcrystalline cellulose TLC in ethyl acetate:pyridine:water (10:3:2) and were detected by spraying with reagent according to T. J. Mabry et al. (The Systematic Identification of Flavonoids, Springer, New York, 1970). NMR spectra were determined using TMS as internal standard.

The two compounds so far identified by this process are quercetin-3-glucoside (isoquercitrin) and isorhamnetin-3-glucoside. The latter compound is reported for the first time in maize. It is usually overlapped by quercetin-3-glucoside on a chromatogram of a crude extract and is easily overlooked. These two compounds correspond to the spots Qu1 and Qu11 on the TLC figure shown in last year's News Letter. There are several other flavonols present in this tissue but in smaller quantities and we have not yet enough material for the NMR spectra.


Oldriska Ceska, Bruce A. Bohm and E. Derek Styles

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors.

Return to the MNL 54 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page