PerspectivePlant Science

Unfallen Grains: How Ancient Farmers Turned Weeds into Crops

See allHide authors and affiliations

Science  02 Jun 2006:
Vol. 312, Issue 5778, pp. 1318-1319
DOI: 10.1126/science.1128836

Some 10,000 years ago during the agricultural revolution, ancient farmers bred hundreds of wild species into the domesticated crops on which humans are dependent today. During this process, these ancient peoples saved seeds from plants with favored traits to form each subsequent generation, and over time they converted slender and unpromising wild species into reliable, bountiful crops. Variants or mutants of genes that conferred favorable phenotypes rose in frequency over time, while variants that best adapted plants to life in the wild were removed by selection from the domesticated population.

Foremost among the creations of ancient plant breeders are the cereals—rice, wheat, and maize, a triumvirate that provides more than 50% of the calories consumed by humans. As compared to their progenitors, these cereals have more and larger grains, thicker stalks, seed that thresh freely from the chaff, and improved flavor. The cereals, and most other crops, share one additional feature that is central to domestication: Their grains remain attached to the plant for harvest by humans rather than falling (shattering) from the plant, as required for wild species to produce their next generation. Although quantitative trait locus (QTL) mapping (1) has convincingly shown that the evolution of domestication traits such as the loss of shattering arose through a relatively small number of gene changes, the nature of these genes and the molecular changes within them is not well understood.

Gathering grain.

Two agricultural workers harvesting rice in Yangshuo, Guangxi Province, China.


In research published in Science earlier this year (2) and on page 1392 of this issue (3), groups in the United States and Japan take two big steps toward bridging the gap between domestication QTLs and domestication genes. Li et al. (2) cloned shattering4 (sh4), a gene first identified by this group as a major QTL controlling 69% of the variance for shattering in crosses of wild and cultivated indica rice. This team was able to localize the causative difference to a 1700-base pair region, and then to demonstrate that a single amino acid change is principally responsible for the loss of shattering. An extraordinary feature of the cultivated allele of sh4 is that it severely weakens but does not eliminate shattering. Thus, the grains are retained on the plants long enough for harvest, but then they can be removed easily by threshing.

Konishi et al. (3) identified a second major shattering QTL (qSH1) in a cross of the two independently domesticated forms of rice, indica and japonica. This QTL controls 68% of the variation for shattering in this hybrid population, and the authors cloned the gene and mapped the causative difference to a single nucleotide change. An absolutely exquisite result is that this single nucleotide change in cultivated rice obliterates a cis-regulatory element required for the expression of qSH1 in the abscission layer, which is needed for the grain to break away from the plant. Other cis-regulatory elements in qSH1 are conserved between wild and cultivated rice, and thus qSH1 still fulfills its other functions in the inflorescence of cultivated rice.

These two rice genes join the growing number of plant domestication QTLs cloned to date. In 1997, the maize gene tb1 was reported as the first domestication QTL to be cloned (4). tb1 controls the complex differences in plant architecture between maize and its progenitor, teosinte. In 2000, a major QTL (fw2.2) contributing to the massive increase in fruit size that was a central feature of tomato domestication was cloned (5). In 2005, the maize domestication QTL tga1, which provides naked grains to maize (as opposed to the covered grains of teosinte), was cloned (6). And thus far in 2006, in addition to the two rice shattering genes, cloning of the wheat Q gene was reported (7). Q controls the compaction and fragility of the ear of wheat and also the ease with which the grain can be separated from the chaff.

A notable feature of this list of six domestication genes is that five of the six encode transcription factors that regulate other (target) genes by directly binding to their DNA. Transcription factors represent only about 5% of genes in plant genomes (8, 9) but 83% of the domestication genes listed above. Interestingly, the five domestication transcription factors belong to five separate families: TCP (tb1), SBP (tga1), AP2 (Q), MYB3 (sh4), and HOX (qSH1). This suggests that the exaggerated proportion of transcription factors among domestication genes is the product of some general feature of transcription factors and not of one particular class of transcription factors. The predominant role of transcription factors in domestication mirrors their equally large role in controlling plant development (10), which supports the view that they have properties that predispose them to become the major genes contributing to morphological evolution in plants (11).

Another remarkable feature of this list is that the domesticated alleles of all six genes are functional. If domestication involved the crippling of precisely tuned wild species, one might have expected domestication genes to have null or loss-of-function alleles. Rather, domestication has involved a mix of changes in protein function and gene expression. As a consequence of domestication, sh4 shows changes in protein function and expression level (2), qSH1 shows a change in the spatial pattern of its expression (3), tb1 shows increased expression (4), tga1 shows a change in protein stability or protein function (6), fw2.2 shows a heterochronic shift in its expression (5), and Q shows changes in protein function and gene expression (7). Given that the cultivated allele of not one of these six domestication genes is a null, a more appropriate model than “crippling” seems to be adaptation to a novel ecological niche—the cultivated field. Tinkering and not disassembling is the order of the day in domestication as in natural evolution, and Darwin's use of domestication as a proxy for evolution under natural selection was, not surprisingly, right on the mark.

A consequential question now is whether modern plant breeders might borrow from the playbook of their Neolithic predecessors. Might one tinker with the expression patterns or protein functions of known domestication genes to create superior alleles? Can every transcription factor in the genome be manipulated in a systematic manner to generate a pool of new trait variation? Knowledge of past successes should help to intelligently guide future crop improvement.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
View Abstract

Stay Connected to Science

Navigate This Article