Rare genetic variation at Zea mays crtRB1 increases β -carotene in maize grain

Articles

Figure 1 Carotenoid biosynthetic pathway and Zea mays crtRB1 gene structure. (a) Simplified carotenoid biosynthetic pathway in maize and Arabidopsis5,6,12,13 CRTRB, in blue, represents the nonheme di-iron β-carotene hydroxylase (BCH) family in maize, which has at least five members (supplementary table 1); the orthologous family in Arabidopsis has two members (BCH1 and BCH2). Carotenoid intermediates highlighted in red are compounds detected by HPLC in this study. (b) Zea mays crtRB1 is the target gene in the present study. The

sequenced region is framed in gray, translated exons are depicted as black boxes and the putative start of transcription (TSS) and poly(A) sites are indicated. Polymorphisms found in original P1 sequence alignments are marked in the diagram, and those that are significantly associated with changes in βC, βC/βCX, βC/Z and βC/ALL are labeled with asterisks. GGPP, geranylgeranyl pyrophosphate; PSY, phytoene synthase; PDS, phytoene desaturase; Z-ISO, ζ-carotene isomerase; ZDS, ζ-carotene desaturase; CRTISO, carotenoid isomerase; LCYE, lycopene ε-cyclase; LCYB, lycopene β-cyclase; CRTRB, β-carotene hydroxylase family; CYP97A, β-carotene hydroxylase (P450); CYP97C, ε-carotene hydroxylase (P450); ZEP1, zeaxanthin epoxidase; VDE1, violaxanthin de-epoxidase; ABA, abscisic acid.

a

βCRTRB (BCH1/2)CYP97A

βCRTRB (BCH1/2)CYP97AZEP1

αCRTRB (BCH1/2)

CYP97A/C

CYP97C

ZEP1

b

*5′ TE insertion

alleles:2: 206 bp3: 0 bp397 or 206 bp

*3′ TE insertion

alleles:1: 0 bp2: 325 bpD6

325 or 1,250 bp12 bp

as derived traits βC/βCX (60%), βC/Z (42%) and βC/ALL (i.e., βC + βCX + Z + α-carotene + lutein) (42%) in 2003. InDel4 and 3′TE explain 7–27% of phenotypic variation in these four traits (Table 1 and Supplementary Table 1). Similar significant associations were found between these three polymorphisms and nearly all of these traits in each of the years tested in P1 (Table 1), as well as in P2 and P3 (Supplementary Tables 2 and 3). The 5′TE-InDel4 and the InDel4-3′TE polymorphisms were in linkage disequilibrium (LD) (r2 = 0.23 and 0.12, respectively), but the 5′TE-3′TE polymorphisms were not (r2 = 0.02). Variation in significance levels across years for each panel is partly due to differences in allele frequency and environmental effects during each growing season.

All three crtRB1 polymorphisms affect conversion of βC to Z, but the 5′TE has the most pronounced effect (Supplementary Table 4), lead-ing to an average increase of 6.50 µg g 1 βC above the average effect of the unfavorable allelic class (1.50 µg g 1). The significant associations between the three polymorphisms with βC and βC/ALL show that the crtRB1 allelic state influences both the absolute amount of βC and its proportion relative to total carotenoids, indicating that the effect of crtRB1 is likely not attributable to an overall increase in carotenoid

© 2010 Nature America, Inc. All rights reserved.

D1D333 bp6 bpD5390 bp

–1,121 –689–1,325TSS

D2*D418 bp12 bp

1,596G/T

3,440Poly(A)

accumulation. The 206-bp insertion allele of 5′TE (allele 2; Fig. 1b), leading to higher βC concentrations, was detected only in temperate germplasm at low frequency (2.9%, absent and 1.9% in P1, P2 and P3, respectively). The frequencies of the three 3′TE alleles were more evenly distributed within populations, with the most favorable (allele 1, no insertion) present at 20.1%, 4.6% and 18.0% in P1, P2 and P3, respectively (Supplementary Table 5). crtRB1 haplotypes were used to determine the joint effect of 5′TE, InDel4 and 3′TE allelic states on carotenoid phenotypes. Of the seven observed haplotype classes in P1, βC was highest when the favorable alleles of 5′TE (allele 2) and 3′TE (allele 1) were combined (haplotypes 2, 12, 1 and 2, 0, 1), shown in Table 2. The combination of three crtRB1 polymorphisms accounted for 40% of the phenotypic variation for βC, 80% for βC/βCX, 68% for βC/Z and 50% for βC/ALL in P1 in 2003. In P3, the

table 1 crtRB1 polymorphisms associated with carotenoid traits in the P1 diversity panel

Environment:

Average

Polymorphic siteaAlleles in seriesbobservation no.:5′TE

1/2/3

βCβC/βCXβC/ZβC/ALLβCβC/βCXβC/ZβC/ALLβCβC/βCXβC/ZβC/ALL

2002c42n.s.0.0417n.s.0.06530.0498

8.64 × 10–50.02020.01630.0034

4.94 × 10–40.08280.0215

2003c,d168

3.67 × 10 141.74 × 10 334.72 × 10 213.99 × 10 202.40 × 10 44.90 × 10 139.10 × 10 101.46 × 10 086.11 × 10 62.54 × 10 108.16 × 10 77.86 × 10 10

2004c1663.25 × 10 7n.s.0.0163

2.20 × 10 72.00 × 10 50.0248

1.52 × 10 45.0610 81.56 × 10 6

n.s.0.0017

5.15 × 10 8

2005c153n.s.

1.46 × 10 63.53 × 10 40.0064n.s.0.0035n.s.0.01210.0421.

1.01 × 10 123.53 × 10 68.94 × 10 5

R2 (2003)e17032%60%42%42%7%27%21%16%10%26%17%18%

1.26 × 10 146.53 × 10 341.27 × 10 202.73 × 10 205.65 × 10 41.12 × 10 131.33 × 10 103.30 × 10 81.46 × 10 45.88 × 10 121.77 × 10 76.06 × 10 8P (2003)f

InDel412/0

3′TE1/2/3

aOnly significant polymorphic sites are shown. bAlleles in series are listed for each polymorphism, and favorable alleles (higher β-carotene content) are in boldface type. 5′TE allelic series:

1, 397-bp insertion; 2, 206-bp insertion; 3, 0-bp insertion. InDel4 allelic series: 12-bp or 0-bp insertion. 3′TE allelic series: 1, no insertion; 2, 325-bp insertion; 3, 1,250-bp insertion. cP value from association analysis carried out using the mixed model incorporating population structure and kinship, using data from 4 different years. dAn outlier was excluded in all 2003 analyses. eR2 values from analysis of variance (ANOVA) of 2003 data showing percentage phenotypic variation explained. fP value from ANOVA analysis of 2003 data.

n.s., not significant at α = 0.05; βC, β-carotene µg g 1; βC/βCX, β-carotene over β-cryptoxanthin; βC/Z, β-carotene over zeaxanthin; βC/ALL, β-carotene over total carotenoids (β-carotene, β-cryptoxanthin, α-carotene, zeaxanthin and lutein).

Nature GeNetics