An unconventional proanthocyanidin pathway in maize

Abstract

Proanthocyanidins (PAs), flavonoid polymers involved in plant defense, are also beneficial to human health and ruminant nutrition. To date, there is little evidence for accumulation of PAs in maize (Zea mays), although maize makes anthocyanins and possesses the key enzyme of the PA pathway, anthocyanidin reductase (ANR). Here, we explore whether there is a functional PA biosynthesis pathway in maize using a combination of analytical chemistry and genetic approaches. The endogenous PA biosynthetic machinery in maize preferentially produces the unusual PA precursor (+)-epicatechin, as well as 4β-(S-cysteinyl)-catechin, as potential PA starter and extension units. Uncommon procyanidin dimers with (+)-epicatechin as starter unit are also found. Expression of soybean (Glycine max) anthocyanidin reductase 1 (ANR1) in maize seeds increases the levels of 4β-(S-cysteinyl)-epicatechin and procyanidin dimers mainly using (-)-epicatechin as starter units. Introducing a Sorghum bicolor transcription factor (SbTT2) specifically regulating PA biosynthesis into a maize inbred deficient in anthocyanin biosynthesis activates both anthocyanin and PA biosynthesis pathways, suggesting conservation of the PA regulatory machinery across species. Our data support the divergence of PA biosynthesis across plant species and offer perspectives for future agricultrural applications in maize.

Fig. 1. Ectopic expression of ZmANR1 fails to generate PA precursors in high anthocyanin producing tobacco.

a Transcript levels of ZmANR1 and ZmANR2 in developing seeds (14-days after pollination) of maize cultivars ST (Suntava) and BM (Black Mexican). Data are presented as mean ± S.D. (n = 3, independent biological replicates). Differences in transcript levels of ZmANR1 and ZmANR2 between ST and BM maize are significant (P < 0.01) as determined by two-tailed Student’s t-test. ZmEF1α was used as the reference gene. b Confocal microscopy images of GFP-tagged GmANR1 and ZmANR1 showing their subcellular localization in A. thaliana protoplasts. GFP-tagged GmANR1 and ZmANR1 appear green. Autofluorescence of chloroplasts appears red. Images are representative of three independent replicates. c Images of tobacco flowers expressing PAP1, PAP1 with AtANR, PAP1 with GmANR1, and PAP1 with ZmANR1 before (top) and after (bottom) DMACA staining. The distribution of PAs in tobacco flowers is indicated as purple coloration after DMACA staining. AtANR and GmANR1 were cloned from Arabidopsis thaliana (Col-0) and Glycine max (cv Clark), as described (d) Selected ion chromatogram of catechin and epicatechin (left, m/z = 289.0718 ± 10 ppm), 4β-(S-cysteinyl)-catechin and 4β-(S-cysteinyl)-epicatechin (middle, m/z = 408.0759 ± 10 ppm), procyanidin dimers B1 and B2 (right, m/z = 577.1360 ± 10 ppm) in PA extracts from tobacco flowers. e Transcript levels of AtANR, GmANR1 and ZmANR1 in tobacco plants expressing PAP1, PAP1 with AtANR, PAP1 with GmANR1, and PAP1 with ZmANR1. Data are presented as mean ± S.D. (n = 3, independent biological replicates).Asterisks indicate significant difference relative to the untransformed control at P < 0.01 as determined by Student’s t-test. Transcripts not detected are labeled as n.d. NtEF1α was used as the reference gene. Source data for Fig. 1a and Fig. 1e are provided in the Source Data file.

Fig. 2. Analysis of PA starter and extension units and anthocyanins in maize seeds.

Left panels, selected ion chromatograms of catechin and epicatechin (m/z = 289.0718 ± 10 ppm). Middle panels, selected ion chromatograms of 4β-(S-cysteinyl)-catechin and 4β-(S-cysteinyl)-epicatechin (m/z = 408.0759 ± 10 ppm). Right panels, selected ion chromatograms of cyanidin 3-O-glucoside (m/z = 447.0940 ± 10 ppm). SD, chemical standards; c, catechin; epi, epicatechin; c-cys, cysteinyl-catechin; epi-cys, cysteinyl-epicatechin; C3G, cyanidin 3-O-glucoside. Maize varieties are FBLL; AREQ, Arequipa; ST, Suntava; OSA, Osage; BM, Black Mexican.

Fig. 3. Formation of PA precursors in different maize varieties expressing GmANR1.

Selected ion chromatograms of catechin and epicatechin (m/z = 289.0718 ± 10 ppm), as well as 4β-(S-cysteinyl)-catechin and 4β-(S-cysteinyl)-epicatechin (m/z = 408.0759 ± 10 ppm) in seeds of untransformed and transgenic ST (a) and OSA (c) maize. Peak areas of 4β-(S-cysteinyl)-epicatechin and ratio of 4β-(S-cysteinyl)-epicatechin/4β-(S-cysteinyl)-catechin in ST, ST-GmANR1, OSA and OSA-GmANR1 are shown in histograms. Data are presented as mean ± S.D. (n = 3, independent biological replicates). Asterisks denote significant difference relative to wild-type ST or OSA at P < 0.01 determined by two-tailed Student’s t-test. b Phloroglucinolysis of PAs extracted from seeds of ST, ST-GmANR1 and HiII-GmANR1 using procyanidin dimer B2 as standard. Left, selected ion chromatograms of epicatechin monomers (m/z = 289.0718 ± 10 ppm); right, selected ion chromatograms of epicatechin-phloroglucinol (m/z = 413.0876 ± 10 ppm). d MS/MS spectra of 4β-(S-cysteinyl)-epicatechin in the standard and PAs extracted from seeds expressing GmANR1. SD, chemical standards; ST, Suntava maize; OSA, Osage maize; HiII-GmANR1, HiII maize expressing GmANR1 (line GmANR1-24); ST-GmANR1, seeds from genetic cross between ST and GmANR1-24; OSA-GmANR1, seeds from genetic cross between OSA and GmANR1-24. Red arrows indicate peak representing cysteinyl-epicatechin. Source data for Fig. 3a and Fig. 3c are provided in the Source Data file.

Fig. 4. Expression of SbTT2 or SbMYB5 reduces growth and enhances anthocyanin accumulation in yellow-seeded maize.

a Top, images of seeds from untransformed B104, transgenic lines expressing SbTT2 (SbTT2-21 and SbTT2-31), and transgenic lines expressing SbMYB5 (SbMYB5-12 and SbMYB5-51); middle, zoom-in images of seeds in the top row showing differences in seed size and color; bottom, images of maize coleoptile at 5 days after seed germination showing differences in color. b Seed weight of untransformed B104 and transgenic lines expressing SbTT2 or SbMYB5. Data are presented as mean ± S.D. (n = 6, independent biological replicates). Asterisks denote significant difference relative to untransformed B104 at P < 0.01 determined by two-tailed Student’s t-test. c Images of maize silk showing differences in color. Source data for Fig. 4b are provided in the Source Data file.

Fig. 5. PA and PA precursor content and composition in seeds of B104 maize expressing SbTT2 or SbMYB5.

a Transcript levels of SbTT2, SbMYB5, ZmANR1 and ZmANR2 in developing seeds (14-days after pollination) of untransformed B104, SbTT2-OX and SbMYB5-OX analyzed by qRT-PCR. Data are presented as mean ± S.D. (n = 3, independent biological replicates). ZmEF1α was used as reference gene. Asterisks indicate significant difference relative to the untransformed control at P < 0.01 as determined by two-tailed Student’s t-test. b Contents of soluble and insoluble PAs in seeds of B104, SbTT2-OX and SbMYB5-OX. Data are presented as mean ± S.D. (n = 3, independent biological replicates). Asterisks indicate significant difference relative to the untransformed control at P < 0.01 as determined by two-tailed Student’s t-test. c Selected ion chromatograms of catechin and epicatechin (m/z = 289.0718 ± 10 ppm), as well as 4β-(S-cysteinyl)-catechin and 4β-(S-cysteinyl)-epicatechin (m/z = 408.0759 ± 10 ppm) in seeds of B104, SbTT2-OX (line SbTT2-31) and SbMYB5-OX (line SbMYB5-51). Peak areas of epicatechin (epi), 4β-(S-cysteinyl)-catechin (c-cys), and 4β-(S-cysteinyl)-epicatechin (epi-cys) in B104 and SbTT2-OX are shown in the histograms. Data are presented as mean ± S.D (n = 3, independent biological replicates). Asterisks denote significant difference relative to untransformed B104 at P < 0.01 determined by two-tailed Student’s t-test. Compounds not detected are labeled as n.d. d Phloroglucinolysis of PAs extracted from seeds of B104, SbTT2-OX and SbMYB5-OX using procyanidin dimers B2 and B3 as references. Left, selected ion chromatograms of catechin and epicatechin monomers (m/z = 289.0718 ± 10 ppm); right, selected ion chromatograms for detection of catechin-phloroglucinol and epicatechin-phloroglucinol (m/z = 413.0876 ± 10 ppm). Source data for Figs. 5a–c are provided in the Source Data file.

Fig. 6. Analysis of procyanidin dimers and trimers in maize seeds expressing GmANR1 or SbTT2.

Selected ion chromatograms of procyanidin dimers in standards and PAs extracted from seeds of untransformed and transgenic maize ST (a), OSA (b), and B104 (c). d Selected ion chromatograms of procyanidin trimers in PAs extracted from seeds of Medicago (Mt), soybean (Gm), untransformed ST maize, and ST maize expressing GmANR1. SD, chemical standards; ST, Suntava maize; ST-GmANR1, transgenic maize seeds obtained from crosses between HiII-GmANR1 and ST; OSA, Osage maize; OSA-GmANR1, transgenic maize seeds obtained from crosses between HiII-GmANR1 and OSA; B1-B4, procyanidin dimers B1-B4 (m/z = 577.1360 ± 10 ppm); iso-B2, procyanidin B2 isomer with (+)-epicatechin as the starter unit; iso-B4, procyanidin B4 isomer with (+)-epicatechin as the starter unit; C1, procyanidin C1 trimer (m/z = 865.1981 ± 10 ppm).

Fig. 7. Analysis of anthocyanins and PA precursors in seeds of BZ1 and bz1 mutant maize.

a Selected ion chromatograms of catechin and epicatechin (left, m/z = 289.0718 ± 10 ppm), 4β-(S-cysteinyl)-catechin and 4β-(S-cysteinyl)-epicatechin (middle, m/z = 408.0759 ± 10 ppm), as well as cyanidin 3-O-glucoside (right, m/z = 447.0940 ± 10 ppm) in seeds of BZ1 and bz1 maize. b Anthocyanin contents in BZ1 and bz1 seeds. Data are presented as mean ± S.D. (n = 3, independent biological replicates). Asterisks indicate significant difference between BZ1 and bz1 samples at P < 0.01 as determined by two-tailed Student’s t-test. c Chiral-HPLC analysis of the stereochemistry of epicatechin monomers in bz1 seeds. Red arrow indicates the (+)-epicatechin in PAs extracted from bz1 seeds. d Selected ion chromatograms of procyanidin dimers (m/z = 577.1360 ± 10 ppm) in BZ1 and bz1 maize seeds. SD, chemical standard; (-)-epi and (+)-epi, (-)-epicatechin and (+)-epicatechin; c-cys and epi-cys, 4β-(S-cysteinyl)-catechin and 4β-(S-cysteinyl)-epicatechin; C3G, cyanidin 3-O-glucoside; B3, procyanidin B3 dimer; iso-B2, procyanidin B2 isomer with (+)-epicatechin as the starter unit; iso-B4, procyanidin B4 isomer with (+)-epicatechin as the starter unit. Source data for Fig. 7b are provided in the Source Data file.

Fig. 8. Schematic diagram of PA, anthocyanin and phlobaphene biosynthesis pathways in maize seeds.

CHS, chalcone synthase, c2; CHI, chalcone isomerase, chi1; F3H, flavanone 3-hydroxylase, f3h; F3’H, flavanone 3´-hydroxylase, pr1; DFR, dihydroflavanol 4-reductase, a1; ANS, anthocyanidin synthase, a2; ANR, anthocyanidin reductase; UFGT, UDP-glucose: flavonoid glucosyltransferase, BZ1; GST, glutathione S-transferase, BZ2. Anthocyanins accumulate in aleurone cells, and phlobaphenes in the pericarp.