Gene expression

How plants avoid incest

Different versions of the same gene can be either dominant or recessive. A small non-coding RNA mediates such differences in dominance as part of a system that prevents inbreeding in plants.

Many flowering plants have elaborate systems to distinguish between pollens that are acceptable for fertilization and those that are unsuitable1,2. For plants of the Brassica family, one such system is self-incompatibility — rejection of genetically similar gametes. On page 983 of this issue, Tarutani et al.3 describe how small non-coding RNAs might introduce another level of mate selection through the epigenetic regulation of genes mediating self-incompatibility.

Three genes regulate self-incompatibility in Brassica: SP11/SCR, SRK and SLG1,2. These genes are inherited together as a unit, and each comes in many versions, or alleles, such as SRK1, SRK2 and SRK3; each allele produces a different version of the same protein. Different allele combinations of these genes are called S-haplotypes (Fig. 1a); the S1-haplotype, for example, represents SP11/SCR1, SRK1 and SLG1. When the S-haplotype of the male parent (the pollen-producing anther) matches the S-haplotype of the female parent (the pistil) the pollen is rejected (Fig. 1b). The key determinants of this system are the pollen-specific SP11/SCR protein and its receptor, the pistil-specific SRK protein. Interaction between these proteins activates the pollen-rejection response in the pistil1,2,4. Only proteins encoded by matching alleles from each S-haplotype can interact (for instance, SP11/SCR1 can bind only to SRK1), thus avoiding unnecessary pollen rejection.

Figure 1: | The intricacies of the S-haplotypes.

a, Tarutani et al.3 report that, in addition to the three well-known genes, SP11/SCR, SRK and SLG, a fourth gene, SMI, also controls self-incompatibility in Brassica plants. In this example, two different S-haplotypes are shown, with S9 being pollen-dominant and S60 pollen-recessive. b, When two S-haplotypes of the pollen are equally dominant, pollen rejection by the pistil occurs when the pollen-producing anther shares the same S-haplotype — in this instance S9 or S12 — with the pistil. c, In cases in which there is a pollen-dominant (S9) and a pollen-recessive (S60) S-haplotype, the same rules apply as in b, except when pollen from the S9 S60 anther lands on the S60 S60 pistil. Because the pollen-dominant S9-haplotype suppresses the pollen-recessive S60-haplotype in the anther (through the action of the SMI9 small non-coding RNA), the S60 S60 pistil accepts the S9 S60 pollen.

Another level of complexity is the inherent diploid nature of plants: they carry two alleles of each gene, and so two different S-haplotypes. Normally, either S-haplotype can cause pollen rejection if it is shared by the anther and the pistil (Fig. 1b). However, there are exceptions to this rule: in the anther, one S-haplotype can be dominant to the other, masking the recessive S-haplotype5. For example, Tarutani et al. show that, if the anther genome carries the pollen-dominant S9-haplotype and the pollen-recessive S60-haplotype, pollen rejection occurs only if the pistil has the S9-haplotype. A pistil with the S60-haplotype will accept pollen grains from the S9 S60 anther, because the pollen-recessive S60-haplotype is suppressed (Fig. 1c).

How does the dominant–recessive relationship between these S-haplotypes emerge? The authors3 reveal that a fourth gene, SP11 methylation inducer (SMI), which is tightly linked to the gene trio mediating self-incompatibility, is responsible for this altered trait.

The same team of researchers has previously discovered6 that, when a plant carries a pollen-dominant and a pollen-recessive S-haplotype, the SP11/SCR gene from the pollen-recessive S-haplotype is turned off. The group also found7 that suppression of the pollen-recessive SP11/SCR gene correlates with methylation of the SP11/SCR promoter sequence — a process known to cause gene silencing. The key outstanding question was how this very selective silencing of SP11/SCR occurs. The authors now show3 that SMI is the missing link.

The SMI gene is part of both the genomic region containing dominant S-haplotypes and that carrying the recessive S-haplotypes (Fig. 1a). It encodes a small non-coding RNA (sRNA) that is specifically produced in the anther. Take Figure 1c, for example: Tarutani et al.3 showed that production of an SMI9 sRNA from the pollen-dominant S9-haplotype results in methylation of the promoter of — and so silencing of — the pollen-recessive SP11/SCR60 gene. Accordingly, the SP11/SCR60 pollen protein is not produced to activate SRK60 in the pistil, and so the pollen grains from the S9 S60 plant are accepted by the S60 pistil.

But why does this occur in only one direction — from the dominant to the recessive S-haplotype — when both S-haplotypes contain the SMI gene? Tarutani and co-workers discovered that the pollen-recessive SMI sequences carry an evolutionarily conserved change. When the authors introduced this change into the dominant S9 SMI gene, the resulting sRNA could no longer silence the pollen-recessive SP11/SCR gene.

Another question is why the SMI gene of the pollen-dominant S-haplotype does not also silence its neighbouring SP11/SCR9 gene within the same haplotype. For gene silencing, the SMI sRNA sequence must be closely related to its target sequence. The SMI sRNA of the pollen-dominant S-haplotype is highly complementary to only pollen-recessive, and not pollen-dominant, SP11/SCR promoters3.

The gene-silencing activity of the pollen-dominant SMI genes is therefore due to the essential sequence identity to the pollen-recessive SP11/SCR promoters. Although at first glance this additional level of complexity — layered on top of the self-incompatibility response — seems unnecessary, it is a biologically relevant mechanism. Recently, two studies8,9 examined the role of dominance in this system using theoretical modelling and found that modifiers leading to pollen-dominant S-haplotypes (such as SMI sRNA) would be favoured, because these plants could mate with more plants in the population8. The pressures of inbreeding depression also seem to favour the evolution of dominance among S-haplotypes9.

A remaining puzzle is the observation10 of gene silencing between select combinations of pollen-recessive S-haplotypes, which indicates that other unknown factors might also be involved. Regardless of this, Tarutani and colleagues' paper is an intriguing example of sRNA-mediated regulation of dominant–recessive patterns of Mendelian inheritance.


  1. 1

    Franklin-Tong, V. E. Self-Incompatibility in Flowering Plants (Springer, 2008).

    Google Scholar 

  2. 2

    Chapman, L. A. & Goring, D. R. J. Exp. Bot. 61, 1987–1999 (2010).

    CAS  Article  Google Scholar 

  3. 3

    Tarutani, Y. et al. Nature 466, 983–986 (2010).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Ivanov, R., Fobis-Loisy, I. & Gaude, T. Trends Plant Sci. 15, 387–394 (2010).

    CAS  Article  Google Scholar 

  5. 5

    Thompson, K. F. & Taylor, J. P. Heredity 21, 345–362 (1966).

    Article  Google Scholar 

  6. 6

    Shiba, H. et al. Plant Cell 14, 491–504 (2002).

    CAS  Article  Google Scholar 

  7. 7

    Shiba, H. et al. Nature Genet. 38, 297–299 (2006).

    CAS  Article  Google Scholar 

  8. 8

    Schoen, D. J. & Busch, J. W. Evolution 63, 2099–2113 (2009).

    CAS  Article  Google Scholar 

  9. 9

    Llaurens, V., Billiard, S., Castric, V. & Vekemans, X. Evolution 63, 2427–2437 (2009).

    Article  Google Scholar 

  10. 10

    Kakizaki, T. et al. Plant Cell Physiol. 44, 70–75 (2003).

    CAS  Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Goring, D., Indriolo, E. How plants avoid incest. Nature 466, 926–927 (2010).

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.