SummaryMicroRNAs (miRNAs) are endogenous small RNAs that can have large-scale regulatory effects on development and on stress responses in plants. The endemic rust fungus Cronartium quercuum f. sp. fusiforme causes fusiform rust disease in pines, resulting in the development of spindle-shaped galls (cankers) on branches or stems. This disease is the most destructive disease of pines in the southern USA. To test whether miRNAs play roles in fusiform rust gall development, we cloned and identified 26 miRNAs from stem xylem of loblolly pine (Pinus taeda), which belong to four conserved and seven loblolly pine-specific miRNA families. Forty-three targets for nine of these 11 families were experimentally validated in vivo. Sequence analysis suggested that the target cleavage site may be determined not only by the miRNA sequence but also by the target sequence. Members of three loblolly pine-specific miRNA families target a large number of non-protein coding transcripts, and one of these families could also initiate secondary phased production from its target of a putative trans-acting short interfering RNA (ta-siRNA). Expression of 10 of these 11 miRNA families was significantly repressed in the galled stem. PCR-based transcript quantification showed complex expression patterns of these miRNAs and their targets in the galled tissues and in tissues surrounding the gall. We further predict 82 plant disease-related transcripts that may also response to miRNA regulation in pine. These results reveal a new genetic basis for host-pathogen interactions in the development of fusiform rust gall.

BackgroundGenome evolution in the gymnosperm lineage of seed plants has given rise to many of the most complex and largest plant genomes, however the elements involved are poorly understood.Methodology/Principal Findings Gymny is a previously undescribed retrotransposon family in Pinus that is related to Athila elements in Arabidopsis. Gymny elements are dispersed throughout the modern Pinus genome and occupy a physical space at least the size of the Arabidopsis thaliana genome. In contrast to previously described retroelements in Pinus, the Gymny family was amplified or introduced after the divergence of pine and spruce (Picea). If retrotransposon expansions are responsible for genome size differences within the Pinaceae, as they are in angiosperms, then they have yet to be identified. In contrast, molecular divergence of Gymny retrotransposons together with other families of retrotransposons can account for the large genome complexity of pines along with protein-coding genic DNA, as revealed by massively parallel DNA sequence analysis of Cot fractionated genomic DNA.Conclusions/SignificanceMost of the enormous genome complexity of pines can be explained by divergence of retrotransposons, however the elements responsible for genome size variation are yet to be identified. Genomic resources for Pinus including those reported here should assist in further defining whether and how the roles of retrotransposons differ in the evolution of angiosperm and gymnosperm genomes.

Genomic mapping has been used to identify a region of the host genome that determines resistance to fusiform rust disease in loblolly pine where no discrete, simply inherited resistance factors had been previously found by conventional genetic analysis over four decades. A resistance locus, behaving as a single dominant gene, was mapped by association with genetic markers, even though the disease phenotype deviated from the expected Mendelian ratio. The complexity of forest pathosystems and the limitations of genetic analysis, based solely on phenotype, had led to an assumption that effective long-term disease resistance in trees should be polygenic. However, our data show that effective long-term resistance can be obtained from a single qualitative resistance gene, despite the presence of virulence in the pathogen population. Therefore, disease resistance in this endemic coevolved forest pathosystem is not exclusively polygenic. Genomic mapping now provides a powerful tool for characterizing the genetic basis of host pathogen interactions in forest trees and other undomesticated, organisms, where conventional genetic analysis often is limited or not feasible.

Fusiform rust disease remains the most destructive disease in pine plantations in the southern United States. Our ongoing research is designed to identify, map, and clone the interacting genes of the host and pathogen. Several resistance (R) genes have been identified and genetically mapped using informative pine families and single-spore isolate inoculations. In addition, we are mapping the first of many expected corresponding avirulence (Avr) genes in the fungal pathogen. The Avr genes condition avirulence ⁄ virulence and avirulence is required for an incompatible reaction (i.e., no-gall development) to take place within an inoculated tree that carries resistance at the corresponding R gene. We provide an overview of our methodology for identifying and mapping R and Avr genes, an update of our current progress, and a brief discussion of two approaches for predicting R gene genotypes of uncharacterized parental trees and for estimating the efficacy of specific pine genotypes at various planting locations. This paper emphasizes the critical importance of controlled genetic materials of both the host and pathogen for elucidating the genetic nature of resistance and virulence in coevolved forest pathosystems.

We propose a method for defining DNA markers linked to Cronartium quercuum f. sp. fusiforme avirulence (Avr) genes. However, before this method can be successfully employed, a spore competition study was needed to determine the genetic composition of single pycnial drops and multiple drops on single galls when using the standard inoculation procedure, whether virulent (avr1) basidiospores ever predispose some resistant (Fr1/fr1) trees to infection by avirulent (Avr1) basidiospores, and whether avr1 and Avr1 basidiospores equally infect susceptible (fr1/fr1) trees. Results of this study suggest that multiple infections within a single gall are common using the concentrated basidiospore system, resulting on average in >4 infection events per tree. Due to multiple infections within a single gall, an individual pycnial drop cannot be assumed to consist of spores from only a single haploid pycnium. Roughly 57% of the drops harvested were found to consist of more than one haploid genotype, most likely due to the physical mixing of spores from genetically different pycnia. Most importantly, although multiple infections do occur in the formation of a single gall, there is no evidence to suggest that the genetics of the proposed gene-for-gene interaction are compromised. Only avr1 basidiospores were observed to cause infection on Fr1/fr1 trees, whereas both avr1 and Avr1 basidiospores were observed to cause infection on fr1/fr1 trees, albeit not at equal frequencies.