Figure 1.
Hypothesis of parallel genetic evolution at the Tb1 locus for the adaptation of vegetative branching during maize and pearl millet domestication.
A. The phylogenetic tree shows that Zea mays and Pennisetum glaucum are two wild grasses from the Panicoid sub-family that separated 30 million years ago (dotted lines, scale not respected), wild Z.mays (teosinte) growing in America and wild P.glaucum in Africa. About 9,000–4,000 years ago, they were independently domesticated into maize and pearl millet, respectively. Pictures below the tree illustrate the parallel morphological evolution of both wild progenitors during their domestication, in particular the reduction of tillering and branching. Z.mays and P.glaucum inherited from their most recent common ancestor the orthologous copies ZmTb1 and PgTb1of the developmental gene Tb1 (represented by a hatched box). It was previously shown that ZmTb1 has been targeted by human selection for the reduction of maize branching during domestication. We ask whether PgTb1 was subjected to parallel evolutionary processes for the similar adaptation of branching during the domestication of pearl millet. B. Distribution of the number of tillers and branches in domesticated pearl millet, wild P.glaucum and weedy plants grown in the same location in south Niger. The cultivated field and the wild population were in parapatric situation. Plants present in the field were classified as domestic or as weedy according to farmer's classification. The ability of weedy pearl millets to shed their seeds spontaneously at the maturity stage is one of the main factors used by farmers to recognize them [13]. Histograms show that domestication was associated with a reduction of vegetative branching. These data were obtained on more than 200 plants for the wild and the domestic pearl millets repectively, and more than 150 plants for weedy phenotypes. C. Tillering in young P.glaucum seedlings. At 4 weeks after germination, tillers are visible in wild P.glaucum (left) but not in the Souna domesticated landrace (right). Close-ups after dissection reveal that axillary meristems have developed into an emergent tiller in the wild plant (arrow) but remain dormant as buds with 1 or 2 leaves (arrow) or undeveloped meristems (box) in the cultivated landrace.
Figure 2.
Comparative mapping of domestication QTLs for vegetative branching in cereals.
The orthologous map segments syntenic to the maize Tb1 region are aligned following consensus markers (linked by dotted grey lines). QTLs associated to branching are indicated by their confidence intervals (colored boxes). The respective percent phenotypic variance they explain (R2) is reported alongside the number and effects of other QTLs in the same cross. These QTLs tend to be consistently conserved at similar positions around the mapped or predicted location of Tb1 orthologues in sorghum, foxtail millet and pearl millet, and in some rice crosses involving parents with contrasted vegetative branching architecture.
Figure 3.
Structure and sequence conservation of grass Tb1 orthologues.
A. Phylogenetic tree built from the alignment of Tb1 orthologous sequences (from start to stop codon) using maximum likelihood. Bootstrap values at the nodes were estimated from 500 replicates. B. The position of orthologous Tb1 EST and cDNA from grass species identified by a nBLAST search of the Genbank database are reported relative to the maize ZmTb1 gene. In the schematic representation of the structure of ZmTb1, white boxes stand for exons, lines for UTR and introns. The putative transcription start site (TSS) is indicated by an arrow and the box in interrupted lines is a putative short exon reported by [33] but not supported by any of the EST and cDNA data. C. Analysis of pairwise sequence conservation between ZmTb1 (BAC clone AF464738) and the orthologous regions of sorghum (AF466204), rice (AC091775) and peal millet using the VISTA software [51]. Evolutionary conserved regions (ECR or CNS) were defined by a sliding window analysis with a threshold size of 70 bp and a minimum 70% nucleotide identity.
Figure 4.
Conservation of PgTb1 function during plant development.
A. In-situ hybridization of the PgTb1 mRNA in serial transverse sections of a 10 day-old seedling of (a–c) the Souna landrace, (d–f) a 10 day-old wild seedling and (g) at the upper node of a mature wild plant (ms: main stem; as: axillary branch). Positions of the sections are given on the bottom-right diagram. B. Association of PgTb1 with domestication QTLs for tillering and branching in two domesticated x wild crosses. Souna and Tiotandé are landraces from Niger and Senegal, respectively. We mapped PgTb1 in the reference cross 81b x icmp451 to position the gene on the consensus pearl millet genetic map. The corresponding ZmTb1 locus and associated maize domestication QTLs are taken from ref. [24]. The percentage of variation explained by a QTL (R2) is reported beside its confidence interval (box). Abbreviations: till, tiller number; nb stems, total number of stems (tillers & branches); nb nodes, node number on the main stem; plht, plant height; lbil, average branch internode length on the main stem.
Figure 5.
Molecular polymorphism at the PgTb1 locus.
A. Sliding-window plot of the polymorphism index π in the PgTb1 region. Values were calculated separately in wild (blue) and domesticated (green) samples in a 600 bp window. B. Genetic tree of PgTb1 alleles (right) and for one of the STS loci (left), constructed using the neighbor-joining method and the Kimura-2P distance (gaps excluded). Significant bootstrap support is indicated at the node and was calculated for 1,000 random permutations.
Table 1.
Genetic diversity and tests of selection in PgTb1 and STS control loci.