Plant biology

Genetics of high-rise rice

Article metrics

When subject to flooding, deepwater rice survives by shooting up in height. Knowledge of the genetic context of this and other responses to inundation will be a boon in enhancing rice productivity.


Deepwater rice lives up to its name: this variety can outgrow slowly rising floodwaters of up to 4 metres in depth. On page 1026 of this issue, Hattori and colleagues1 describe how they have identified two genes, SNORKEL1 and SNORKEL2, that contribute to this spectacular elongation response.

Rice — the seed of Oryza sativa — feeds billions. Although productivity per hectare has more than doubled since the 1960s, a further doubling will be necessary to meet projected requirements by 2050 (refs 2, 3). More than 30% of Asian and 40% of African rice acreage is cultivated in either lowland paddies (15–50 centimetres deep) or deepwater paddies (depth of more than 50 cm). But lack of control of water depth in rain-fed paddies can be a serious problem: in some areas, water levels rise progressively during the growing season and can reach several metres; in others, flash flooding can fully submerge plants for days or weeks. High-yielding rice varieties cannot survive either extreme of inundation. As a result, some flood-prone areas are planted with traditional local varieties that display a remarkable capacity for flooding-induced elongation — of up to 25 cm per day — or that can tolerate submergence for up to 15 days. But the high-yielding varieties are typically five times more productive than these flood-tolerant plants.

Hattori et al.1 pinpoint three chromosomal segments that regulate flooding-induced stem elongation in deepwater rice. When these regions were bred into a high-yielding, non-elongating rice (O. sativa subsp. japonica), the offspring gained the capacity to elongate in response to partial submergence. Hattori et al. focused on one segment of chromosome 12 that contributes 30% of the deepwater response, and identified a pair of genes, SNORKEL1 and SNORKEL2, that stimulate underwater shoot elongation. Both genes belong to the large family of ethylene-response-factor (ERF) transcription factors, and are highly expressed in submerged stems. The SNORKEL genes are absent or non-functional in rice that lacks a deepwater elongation response, including japonica subspecies and Oryza nivara, one of two wild progenitors of domesticated rice. The SNORKEL locus seems to have been introduced from the other wild relative, Oryza rufipogon — there are populations of this species that have functional SNORKEL genes and display deepwater elongation.

Flooding triggers stem and leaf elongation by prompting a cellular increase in the volatile hormone ethylene, which promotes a decline in the hormone abscisic acid, followed by increased synthesis of, and response to, gibberellic acid4,5 (Fig. 1). The interaction of gibberellic acid with its protein receptor induces the degradation of DELLA transcription factors, which normally limit cell elongation. Hattori et al.1 found that the activation of SNORKEL1 and SNORKEL2 is induced by ethylene, which accumulates in submerged organs, mainly as a consequence of the slow diffusion of gases in water as opposed to in air4. The action of the two SNORKEL genes apparently occurs downstream of ethylene but upstream of gibberellic acid (Fig. 1a). Further study is needed to see if SNORKEL directs the marked rise in bioactive gibberellic acid that Hattori et al.1 see in submerged tissues of deepwater rice.

Figure 1: Ethylene and flooding-tolerance strategies in rice.

The elongation of stem and leaf cells is positively regulated by the hormone gibberellic acid (GA). In normal circumstances, a second hormone, abscisic acid (ABA), inhibits GA activity. When plants are submerged, ethylene, a gaseous plant hormone, accumulates owing to its slow outward diffusion in water. This promotes the breakdown of ABA, increasing the production of, or responsiveness to, GA, ultimately stimulating cell elongation. a, The escape strategy of deepwater varieties involves fast stem elongation to rise above the water level. Elongation growth and possibly GA accumulation or action are stimulated by transcription factors encoded by two ethylene-regulated genes, SNORKEL1 and SNORKEL2 (SK1 and SK2) (dashed arrow)1. b, In the quiescence strategy of submergence-tolerant varieties, shoot elongation is suppressed so as to conserve carbohydrates and increase survival under flash-flood conditions. GA signalling and thus elongation are inhibited by the ethylene-induced action of a SUBMERGENCE gene (SUB1A-1) on the growth-inhibiting genes SLENDER RICE-1 (SLR1) and SLR LIKE-1 (SLRL1).

The SNORKEL locus is the second example of a multigene region in rice that encodes ERFs and regulates underwater growth. The SUBMERGENCE-1 (SUB1) gene region, situated on chromosome 9, encodes two or three ERFs (SUB1A, SUB1B and SUB1C) that determine the response to complete submergence6. SUB1 was identified in a traditional variety prized for its ability to endure more than two weeks of inundation. In contrast to the rapid stem elongation manifested in submerged deepwater rice, lines that have the SUB1A-1 gene can survive the stress by limiting ethylene-activated elongation growth (Fig. 1b), thereby conserving precious carbohydrates for regrowth when the flood recedes6,7. This is accomplished by minimizing the decline in a DELLA protein (SLENDER RICE-1, SLR1) and a related non-DELLA protein (SLR LIKE-1, SLRL1) in submerged shoots8. Strikingly, therefore, ethylene can trigger antithetical outcomes: the promotion (SNORKEL) or suppression (SUB1) of underwater elongation.

Fast submergence-induced shoot elongation is also a characteristic of some wild plants that grow in flood-prone ecosystems. Together with other traits, such as the possession of aerenchyma (the snorkel-like conduits for fast gas diffusion)9, underwater elongation determines the distribution and abundance of species in river floodplains and similar environments. Fast underwater growth in wild plants is also regulated by the interaction between ethylene and gibberellic acid, with downstream target proteins such as the cell-wall-loosening expansins also having a role4. Submergence-induced elongation is observed in plants inhabiting locations characterized by prolonged but shallow floods. By contrast, species found in places where deep transient floods occur limit shoot elongation during submergence4 — like rice that adopts the quiescence strategy.

Wetland species that are quiescent during submergence also accumulate ethylene in submerged tissues and can respond to gibberellic acid under non-submerged conditions4. This suggests that ethylene-driven shoot elongation, and thus the ecological distribution of many wetland plants, is regulated by molecular components that control the production and response to gibberellic acid in an ethylene-dependent manner. We speculate that transcription factors, related to the SNORKEL and SUB1 ERFs, regulate submergence responses — and thus survival in flood-prone environments — across plant species. The presence, function, timing and tissue-specificity of ERF expression could be targets of natural selection, and might thereby determine the remarkably varied growth responses mediated by ethylene at different doses and in distinct species10.

Hattori and colleagues' delineation1 of genes that control deepwater elongation has a practical edge: it provides a way to increase grain yields in areas prone to deep flooding by introducing the fast-elongation trait into high-yielding cultivars. This will complement the successful development of SUB1 rice varieties that provide robust submergence tolerance in areas susceptible to flash floods6. Yields of rice cultivated in deep water may not reach those achieved in shallow paddies because carbon must be allocated to rapidly elongating underwater organs, and tall, grain-laden plants can topple if floodwaters recede. Nevertheless, the introduction of SNORKEL and SUB1 into high-yielding varieties, using advanced breeding strategies, promises to improve the quality and quantity of rice produced in marginal farmlands.


  1. 1

    Hattori, Y. et al. Nature 460, 1026–1030 (2009).

  2. 2

    Khush, G. S. Nature Rev. Genet. 2, 815–822 (2001).

  3. 3

    Food and Agriculture Organization of the United Nations. The State of Food and Agriculture 2008. Biofuels: Prospects, Risks and Opportunities (FAO, 2008).

  4. 4

    Bailey-Serres, J. & Voesenek, L. A. C. J. Annu. Rev. Plant Biol. 59, 313–339 (2008).

  5. 5

    Kende, H., van der Knaap, E. & Cho, H.-T. Plant Physiol. 118, 1105–1110 (1998).

  6. 6

    Xu, K. et al. Nature 442, 705–708 (2006).

  7. 7

    Fukao, T. et al. Plant Cell 18, 2021–2034 (2006).

  8. 8

    Fukao, T. & Bailey-Serres, J. Proc. Natl Acad. Sci. USA 105, 16814–16819 (2008).

  9. 9

    Colmer, T. D. Plant Cell Environ. 26, 17–36 (2003).

  10. 10

    Pierik, R. et al. Trends Plant Sci. 11, 176–183 (2006).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Voesenek, L., Bailey-Serres, J. Genetics of high-rise rice. Nature 460, 959–960 (2009) doi:10.1038/460959a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.