PHR of Grains of Asian Cultivated Rice Subspecies.
Rice subspecies shown are as follows (from left to right): indica cv MH63, 93-11, Tx7 and Tx9, and japonica cv CJ06, Nipponbare (Nip), Zh11 and Qfn.
(A) The natural color of hulls.
(B) The color of brown rice.
(C) The PHR of hulls.
(D) The PHR of brown rice.

Phr1 Expression Patterns and Alignment of Phr1 Protein Sequences of MH63 and Nipponbare with Wheat PPO.
(A) Phr1 expression pattern revealed by RT-PCR using total RNA isolated from roots (R), culms (C), leaves (L), and panicles 7 and 21 d after flowering (G1 and G2, respectively). Amplification of actin cDNA was used to ensure that approximately equal amounts of cDNA were loaded.
(B) Alignment of Phr1 protein sequences of MH63 and Nipponbare with wheat (T. aestivum, abbreviated as Tae) PPO (GenBank AAS00454). A dot stands for an identical amino acid residue and a dash for a gap. Putative N-terminal signal peptide and copper binding domains (A and B) are indicated in pink and orange, respectively. Black arrow indicates the predicted cleavage site of N-terminal signal peptide, and vertical lines show the intron positions. The amino acid residues encoded by the 18- and 29-bp deletion in japonica cultivars are highlighted in red and yellow, respectively. The blue triangle indicates the position of the 1-bp insertion in Tx36, a japonica cultivar.

The Genealogy of Phr1 Alleles in Cultivars and Their Wild Relatives.
(A) The main panel (large tree, right) shows the optimal tree obtained using the neighbor-joining method with 1000 bootstraps. The bootstrap values are indicated at nodes with at least 50% support. The H18 (Δ18-bearing) haplotypes are labeled red, and the H29 haplotypes are labeled blue. The only 1-bp insertion line is labeled green. Note the strong genealogical clustering of these colored labels. The boxed cluster consists mainly of H18s, with a few non-Δ18-bearing haplotypes embedded within.
(B) In the inset, we zoom in on this cluster using the parsimonious haplotype cladogram. On each branch, the number of nucleotide changes is given. The suffixes -d and -f appended to accession names are abbreviations for deletion and full-length, respectively.

Population Genetic Tests on the Phr1 Sequences for All Three Populations.
(A) Nucleotide diversity (π) at silent sites by the box-and-whisker diagrams for Phr1 (white) and reference genes (gray). The variance is estimated by the bootstrapping procedure. For comparison, the corresponding estimates for 10 reference genes are also given.
(B) and (C) Tajima's D (B) and Fay and Wu's H (C) tests for the Phr1 sequences. The dashed lines indicated the expected values of 0 for Tajima's D and Fay and Wu's H under the standard neutral model. Both D and H statistics for Phr1 are significantly negative in japonica (P < 0.05) but much higher in indica and O. rufipogon by the bootstrapping procedure described in Methods. Significantly negative D and H indicate an excess of very low and very high frequency variants, respectively.

Map-Based Cloning and Confirmation of Phr1.
(A) The Phr1 locus was mapped on the long arm of chromosome 4 (Chr4) between the markers of S100 and S115.
(B) Phr1 was further localized in a single BAC clone within an interval between the markers P80 and P168. BAC1, OSJNBa0058K23; BAC2, OSJNBa0085C12; BAC3, OSJNBa0053K19; BAC4, OSJNBa0060E08; BAC5, OSJNBa0089N06. Numbers indicate the number of recombinants identified from 1386 F2 phr1 mutant plants.
(C) Predicted ORFs highlighted with arrows.
(D) The Phr1 structure and the complementation construct pC13Phr1. The start codon (ATG) and the stop codon (TGA) are indicated. Purple boxes stand for the coding sequence, blue for 5′ and 3′ untranslated regions, and black lines between boxes for introns. The mutation site in japonica cultivar Nipponbare is also indicated. Structure of the control plasmid pC13p, which contains the promoter region and a truncated Phr1 gene that encodes the first 195 amino acid residues.
(E) The comparison of PPO activities between MH63 and CJ06. One unit of PPO activity was defined as the amount of the enzyme that gives a change in absorbance of 0.001 per min.
(F) Complementation test. PHR-negative Nipponbare (Nip) (1) becomes PHR-positive when transformed with the plasmid pC13Phr1 (3), but transformed with the control vector (2) is shown the same as Nip.
(G) Confirmation of transgenic lines by specific PCR amplification of the Phr1 fragments with (bottom band) or without 18-bp deletion (top band). The DNA template was extracted from MH63, Nip, and transgenic lines (TL).

Discoloration of Transgenic Grains during Storage.
(A) to (C) The comparison of the hull color between untransformed Nip (Nip) and transgenic lines of Nip transformed with the functional Phr1 gene (TL) after storage for 1 month (A), 6 months (B), and 12 months (C).
(D) to (F) The comparison of the coarse grain color between untransformed Nip and transgenic lines of Nip transformed with the functional Phr1 gene after storage for 1 month (D), 6 months (E), and 12 months (F).
(G) The PHRs of coarse grains of untransformed Nip and transgenic Nip.

Distributions of the Fst Statistic in Phr1 vis-á-vis 10 Reference Genes and Randomly Chosen Gene Fragments.
(A) Fst of Phr1 between japonica and O. rufipogon.
(B) Fst of Phr1 between japonica and indica.
(C) Fst of the 10 reference genes between japonica and O. rufipogon.
(D) Fst of the 10 reference genes between japonica and indica.
(E) Fst of randomly chosen gene fragments referred to as STS from Caicedo et al. (2007) between japonica and O. rufipogon.
(F) Fst of the same STS between japonica and indica. Phr1 sequences from each population were resampled following a multinomial distribution with 100 permutations (see Methods).
The Fst profiles of Phr1 ([A] and [B]) show a peak near 1, which is distinctive from the two reference data sets ([C] and [D] for reference genes; [E] and [F] for the STS loci).