Maintenance of genetic variation in plants and pathogens involves complex networks of gene-for-gene interactions

Abstract

The RPP13 [recognition of Hyaloperonospora arabidopsidis (previously known as Peronospora parasitica)] resistance (R) gene in Arabidopsis thaliana exhibits the highest reported level of sequence diversity among known R genes. Consistent with a co-evolutionary model, the matching effector protein ATR13 (A. thaliana-recognized) from H. arabidopsidis reveals extreme levels of allelic diversity. We isolated 23 new RPP13 sequences from a UK metapopulation, giving a total of 47 when combined with previous studies. We used these in functional studies of the A. thaliana accessions for their resistance response to 16 isolates of H. arabidopsidis. We characterized the molecular basis of recognition by the expression of the corresponding ATR13 genes from these 16 isolates in these host accessions. This allowed the determination of which alleles of RPP13 were responsible for pathogen recognition and whether recognition was dependent on the RPP13/ATR13 combination. Linking our functional studies with phylogenetic analysis, we determined that: (i) the recognition of ATR13 is mediated by alleles in just a single RPP13 clade; (ii) RPP13 alleles in other clades have evolved the ability to detect other pathogen ATR protein(s); and (iii) at least one gene, unlinked to RPP13 in A. thaliana, detects a different subgroup of ATR13 alleles.

Figure 1

Neighbour‐joining tree of RPP13 nucleotide sequences inferred using paup* 4.0b10. The HKY85 substitution model was assumed. This model allows for unequal base frequencies and a different rate for transitions versus transversions. Bootstrap proportions of 1000 bootstrap replicates > 50% are indicated on the branches. The recognition capabilities of the RPP13 alleles are indicated by colours as follows: RPP13 recognizes ATR13 (red); RPP13 confers resistance by non‐ATR13 recognition (brown); non‐RPP13 recognition of ATR13 (unknown R gene) (green). Black denotes no recognition except for ‘*’ which denotes accessions not tested. For Rld‐2, RPP13 resistance was demonstrated by inoculation of HRI3860::RPP13‐Rld‐2 with Wela3.

Figure 2

Recognition responses of ATR13 alleles by Arabidopsis thaliana lines. A selection of representative examples of leaves bombarded with ATR13 alleles and stained for β‐glucuronidase (GUS). Three distinct phenotypes were observed: no response (N) gives rise to 300–1000 blue‐stained cells per leaf; full response (F) generates less than 10 blue‐stained cells per leaf; intermediate response (I) gives 40–150 blue‐stained cells per leaf.