Home

From chestnut rose, a promising fruit crop of the Rosa genus, powdery mildew disease-resistant and susceptible genotypes and their F1 progeny were used to isolate nucleotide-binding-site (NBS)-encoding genes using 19 degenerate primer pairs and an additional cloning method called overlapping extension amplification. A total of 126 genes were harvested; of these, 38 were from a resistant parent, 37 from a susceptible parent, and 51 from F1 progeny. A phylogenetic tree was constructed, which revealed that NBS sequences from parents and F1 progeny tend to form a mixture and are well distributed among the branches of the tree. Mapping of these NBS genes suggested that their organization in the genome is a "tandem duplicated cluster" and, to a lesser extent, a "heterogeneous cluster." Intraspecific polymorphisms and interspecific divergence were detected by Southern blotting with NBS-encoding genes as probes. Sequencing on the nucleotide level revealed even more intraspecific variation: for the R4 gene, 9.81% of the nucleotides are polymorphic. Amino acid sites under positive selection were detected in the NBS region. Some NBS-encoding genes were meiotically unstable, which may due to recombination and deletion events. Moreover, a transposon-like element was isolated in the flanking region of NBS genes, implying a possible role for transposon in the evolutionary history of resistance genes.

DISCUSSION

Previous studies of R genes or RGHs in model plants have accumulated knowledge on the generation of resistance specificities and the evolutionary dynamics of R genes (Bergelson et al. 2001; Bai et al. 2002; Richly et al. 2002; Caicedo and Schaal 2004; Kuang et al. 2004; Xiao et al. 2004;Mondragon-Palomino and Gaut 2005; Bakker et al. 2006; Friedman and Baker 2007). This study, however, focused on a fruit crop, the chestnut rose. Two cloning methods were used: the first, direct PCR amplification with degenerate primers, was believed to have a risk of biased sampling due to preferential amplification. To broaden the sequences, a second method, the overlapping extension approach, was used to capture as many NBS genes fromthe genome as possible (Xu et al. 2005). Under this strategy, a total of 126 NBS-encoding genes were isolated, fewer than those reported in Arabidopsis and rice. One possible reason for the lower number of NBS genes could be the preferential amplification caused by degenerate primers, although we tried to avoid this as mentioned above. The second reason could be due to the scarce genome information for this fruit tree; no sequence information could be utilized froma public database. Another reason could be the slower diversification rate of NBS genes in wild chestnut rose than in domestic plants such as rice because of the selective pressure.However, on the basis of the 126 cloned genes, this study still provides some interesting knowledge of the characteristics of NBS-encoding genes in chestnut rose, such as genomic organization in a tandem cluster, the proximity to transposon elements, rapid evolution, meiotic instability, etc.

Evolutionary complexities of NBS-encoding genes in chestnut rose: NBS-encoding genes in chestnut rose exhibited a high level of intraspecific polymorphisms. For example, for the R4 gene, 9.81% of nucleotides were polymorphic, of which 64% were nonconservative amino acid changes, suggesting that this gene is maintained for short time periods.

What are the evolutionary forces shaping the polymorphisms in NBS-encoding genes in the chestnut rose genome? Intraspecific variations in NBS gene copy number and size were observed between resistant and susceptible genotypes (Figure 4). The two chestnut rose cultivars Guinong no. 6 and no. 5 were differentiated and selected as recently as 30 years ago. Figure 4 shows that the signature of unequal crossing-over events can produce gene copy-number variation and size difference (Hammond-Kosack and Jones 1997; McDowell and Simon 2006). Such a process almost certainly leads to rapid gene divergence between different genotypes (gene amplification and reduction, respectively), which is compatible with the opinion that NBS genes evolve rapidly. Moreover, the process of unequal crossing over could be facilitated by the repetitive elements (also found in this study) within the RGH clusters in the chestnut rose genome, as suggested by McDowell and Simon (2006) and Friedman and Baker (2007). Unequal crossing-over events are believed to homogenize genes within a genotype, resulting in paralogs being more similar than orthologs (Michelmore andMeyers 1998). However, in contrast to the above expectations, intergenotype comparison of nucleotide identity between resistant and susceptible genotypes turned out to be higher than that within the resistant genotype, a surprising result with orthologs beingmore similar than paralogs. The contradiction implies that there have been other processes that have shaped the NBS gene polymorphisms as well as unequal crossing-over events and that the unequal crossing-over events are not predominantly generational gene variations.

Evolutionary analysis and genetic mapping revealed the existence of a "tandem duplicated cluster," where point mutations were observed among the tandem direct repeats. The accumulated mutations are another evolutionary way to increase the complexity of RGHs and an opportunity to produce a new homolog. Moreover, some RGHs share high homology with Pinus RGHs and produced uniform hybridization band among Rosaceae plants (Figure 4), indicating that these genes are ancient and may be evolutionarily maintained by some form of balancing selection (Tian et al. 2002).

Detecting adaptive evolution by comparing amino acid substitution rates (dN) to synonymous substitution rates (dS) has indicated that positive selection has contributed to the evolution of NBS genes in chestnut rose. For many reported R genes, positive selection has been detected primarily in LRR regions. However, positively selected sites were also detected in the NBSencoding region; e.g., Mondragon-Palomino et al. (2002) found that in Arabidopsis five positive selected sites were positioned in the NBS region, implying that the NBS domain may function in determining resistance specificity. This could be determined from domainswapping experiments by replacing the NBS-encoding region of the L10 gene with the equivalent region of L2- or L9-generated new recombinant alleles with novel specificity (Luck et al. 2000). In chestnut rose, the NBS domain may also function in determining the resistance specificity. Further research is required to investigate this hypothesis.

Transposable-like elements appear to have been involved in the evolutionary dynamics of NBS-encoding genes in chestnut rose. A transposable element was isolated proximal to the NBS gene. For most Rosaceae genomes, a large number of gene copies were detected by RFLP analysis. However, on cultivar Guinong no. 1 (intraspecific genotype), peach, and pear, one to three copies were detected. The marked difference in copy number implied that transposable elements could play active roles in the evolutionary history of resistance genes in chestnut rose. In rice, 11 different families of transposable elements were identified at the Xa21 cluster, and the elements appeared to be a major source of variation in this cluster (Song et al. 1998). In addition, in plants, transposable elements are commonly activated by environmental stresses such as pathogen infection (Grandbastien 1998), and it is believed that such activation can increase genomic flexibility with a possible selective advantage.

Together, positive selection, balancing selection, recombination, point mutation, and even transposable elements may constitute the driving forces that shape the complexity, rapid evolution, and even the generation of new resistance specificity of R-gene sequences in chestnut rose.

Meiotic instability of RGH genes: Sequence pairwise comparison revealed that nucleotide identity averaged 52% within resistant parents, 57% within susceptible parents, and 45% within the F1 generation. Statistical analysis showed that the latter was significantly lower than the former two at the 0.05 level, indicating a higher sequence variation in F1 progeny than in parents. Moreover, some sequences from F1 progeny carried repetitive elements.

Evidence from RFLP markers demonstrated that RGH genes are meiotically unstable. New alleles were observed in F1 plants. To confirm this, we used the RGH sequences from F1 plants to design specific PCR primers and to determine the gene status in two parents. Three types were detected; what merits attention is type III where the RGH gene was detected only in some F1 plants but not in either parent (supplemental Figure 4), suggesting that this allele was newly produced during meiosis. However, these are preliminary data for the estimation of meiotic instability. One classical example was Rp1 complex loci in maize, where the homozygous line for the Rp1 locus was used to generate a large number of testcross progeny.

Surprisingly, a high frequency of susceptibles was found in the progeny, indicating the occurrence of meiotically unstable genes (Sudupak et al. 1993). Further research on the Rp1 locus demonstrated that recombination is the primary mechanism of meiotic instability and that such recombination can result in new race specificity (Smith and Hulbert 2005). But for this study, the chestnut rose fruit tree is believed to be highly heterozygous for most gene loci; it is difficult to create a homozygous line to obtain deep insight into the mechanism for meiotic instability. However, further research on sequencing the flanked region around the NBS domain may help us understand the types of recombination and the genetic mechanism for meiotic instability.