Dissecting Resistance to Phytophthora cinnamomi in Interspecific Hybrid Chestnut Crosses Using Sequence-Based Genotyping and QTL Mapping

Abstract

The soilborne oomycete Phytophthora cinnamomi—which causes root rot, trunk cankers, and stem lesions on an estimated 5,000 plant species worldwide—is a lethal pathogen of American chestnut (Castanea dentata) as well as many other woody plant species. P. cinnamomi is particularly damaging to chestnut and chinquapin trees (Castanea spp.) in the southern portion of its native range in the United States due to relatively mild climatic conditions that are conductive to disease development. Introduction of resistant genotypes is the most practical solution for disease management in forests because treatment with fungicides and eradication of the pathogen are neither practical nor economically feasible in natural ecosystems. Using backcross families derived from crosses of American chestnuts with two resistant Chinese chestnut cultivars Mahogany and Nanking, we constructed linkage maps and identified quantitative trait loci (QTLs) for resistance to P. cinnamomi that had been introgressed from these Chinese chestnut cultivars. In total, 957 plants representing five cohorts of three hybrid crosses were genotyped by sequencing and phenotyped by standardized inoculation and visual examination over a 6-year period from 2011 to 2016. Eight parental linkage maps comprising 7,715 markers were constructed, and 17 QTLs were identified on four linkage groups (LGs): LG_A, LG_C, LG_E, and LG_K. The most consistent QTLs were detected on LG_E in seedlings from crosses with both ‘Mahogany’ and ‘Nanking’ and LG_K in seedlings from ‘Mahogany’ crosses. Two consistent large and medium effect QTLs located ∼10 cM apart were present in the middle and at the lower end of LG_E; other QTLs were considered to have small effects. These results imply that the genetic architecture of resistance to P. cinnamomi in Chinese chestnut × American chestnut hybrid progeny may resemble the P. sojae–soybean pathosystem, with a few dominant QTLs along with quantitatively inherited partial resistance conferred by multiple small-effect QTLs.

“A reference genome assembly and adaptive trait analysis of Castanea mollissima ‘Vanuxem,’ a source of resistance to chestnut blight in restoration breeding”

“…Contigs from the version 3.2 de novo assembly were initially anchored to chromosomal locations based on the order of DNA marker sequences on the chestnut research community's reference genetic linkage map for the Vanuxem genotype (Kubisiak et al 2013). To increase the number of anchored contigs, DNA markers were integrated from 10 additional genetic maps: an expanded version of the original Vanuxem genetic map (Supplementary File 1 and 2), eight maps from three American chestnut backcross families segregating for P. cinnamomi resistance (Zhebentyayeva et al 2019), and a dense Quercus robur genetic map (Bodénès et al 2016). Previous analysis of C. mollissima by C. dentata crosses and comparative genomics with Q. robur reveal highly conserved colinearity across chromosomes (Bodénès et al 2012), lending confidence these maps will anchor contigs to the correct location.…”

“…Pairwise nucleotide diversity in C. dentata was lower than that for C. mollissima by 1 nucleotide per 5-kb window (p < 0.001, Wilcoxon signed-rank test). Leveraging data from the previous QTL mapping studies of P. cinnamomi resistance in Chinese/American hybrid families (Zhebentyayeva et al 2012(Zhebentyayeva et al , 2019Olukolu et al 2012;Santos et al 2017), we ran statistical tests for natural selection across LG_E that had three strong QTL intervals associated with P. cinnamomi resistance. Using sequence-based markers from our mapping analyses and local blast alignment tools, we delineated the QTL intervals on the contigs anchored to Chinese chestnut LG_E chromosome and searched for genome-wide outliers from the selection scan in these QTL regions.…”

“…The most striking difference in nucleotide diversity between C. dentata and C. mollissima was observed in the middle of chromosome E which colocalizes with qPcE.2, the most stable QTL detected over multiple years in progeny derived from two Chinese chestnut sources of resistance to P. cinnamomi (Zhebentyayeva et al 2019). Two genes were annotated in this region-a putative ALA-interacting subunit 2, homolog of the ligand-effect modulator 3 (AT5G46150) in Arabidopsis that encodes a protein of unknown function with transmembrane activity; and an ortholog of ornithine deltaaminotransferase (KEGG:AT5G46180) involved in arginine and proline metabolisms and transcriptionally upregulated in response to osmotic stress and non-host disease resistance (Anwar et al 2018).…”

A plan to diversify a transgenic blight‐tolerant American chestnut population using citizen science

“…An additional generation of genome‐wide selection against founder parent genome was performed in the T 5 generation in the scenario ES + LD T 2 to T 4 + FP T 5 . Markers were simulated at 1 cM intervals over 12 chromosomes and a total genome length of 869 cM, whichis the length of the 'Cranberry' genetic map aligned to v.0.5 of the C. dentata genome (Zhebentyayeva et al, ; Westbrook et al, ).…”

“…After ES, three progeny per cross with the largest scores were selected.An additional generation of genome-wide selection against founder parent genome was performed in the T 5 generation in the scenario ES + LD T 2 to T 4 + FP T 5 . Markers were simulated at 1 cM intervals over 12 chromosomes and a total genome length of 869 cM, whichis the length of the 'Cranberry' genetic map aligned to v.0.5 of the C. dentata genome(Zhebentyayeva et al, 2019;Westbrook et al, 2019).Each marker-assisted simulation scenario was repeated 50 times to simulate random crossover events in the progeny. The average, maximum, and minimum proportions of 2N genome length inherited from the founder parent were estimated for the carrier chromosome and for the entire genome over the 50 replicates.…”