Identification of QTLs for Reduced Susceptibility to Rose Rosette Disease in Diploid Roses

Abstract

Resistance to rose rosette disease (RRD), a fatal disease of roses (Rosa spp.), is a high priority for rose breeding. As RRD resistance is time-consuming to phenotype, the identification of genetic markers for resistance could expedite breeding efforts. However, little is known about the genetics of RRD resistance. Therefore, we performed a quantitative trait locus (QTL) analysis on a set of inter-related diploid rose populations phenotyped for RRD resistance and identified four QTLs. Two QTLs were found in multiple years. The most consistent QTL is qRRV_TX2WSE_ch5, which explains approximately 20% and 40% of the phenotypic variation in virus quantity and severity of RRD symptoms, respectively. The second, a QTL on chromosome 1, qRRD_TX2WSE_ch1, accounts for approximately 16% of the phenotypic variation for severity. Finally, a third QTL on chromosome 3 was identified only in the multiyear analysis, and a fourth on chromosome 6 was identified in data from one year only. In addition, haplotypes associated with significant changes in virus quantity and severity were identified for qRRV_TX2WSE_ch5 and qRRD_TX2WSE_ch1. This research represents the first report of genetic determinants of resistance to RRD. In addition, marker trait associations discovered here will enable better parental selection when breeding for RRD resistance and pave the way for marker-assisted selection for RRD resistance.

Keywords: QTL; Rosa; emaravirus; plant breeding; virus resistance.

Figure 1

Distribution of rose rosette disease. (A) C_t value; (B) rosette ratings; (C) severity ratings in 2019, 2020, and 2021 across diploid rose families grown in Crossville, Tennessee; and (D) examples of the RRD severity scale with symptoms indicated by black circles. In violin plots, width of each shaded portion reflects the proportion of samples in that area. Black triangles indicate the median. Within each trait for a given year, the median, first interquartile, and third interquartile were equivalent; thus, only the median was plotted. Samples that were negative for RRV as determined by qRT-PCR were assigned a C_t value of 40. Rosettes were scored on a scale of 0–3, where 0 = no rosettes, 1 = one rosette, 2 = two rosettes, and 3 = three or more rosettes. Severity was rated on a scale of 0–3, where 0 = no symptoms, 1 = one shoot with symptoms, 2 = two shoots with symptoms, and 3 = three or more shoots with symptoms. Photographs of infected plants courtesy of Jennifer Olson, Oklahoma State University.

Figure 2

Alignment of the consensus map developed for three diploid rose populations with the Rosa chinensis genome assembly of Hibrand Saint-Oyant et al. [17] per linkage group (LG). Centimorgans (cM) are plotted on the x-axis and physical position (Mbp) on the y-axis.

Figure 3

Positions and posterior intensities of identified QTL for (A) RRD C_t value and severity on chromosome 5, (B) RRD severity on chromosome 1, (C) RRD severity on chromosome 3, and (D) severity in 2021 on chromosome 6 in diploid rose populations. Multiple lines for the same color reflect the results of individual runs in FlexQTL™. Figure constructed with MapChart v2.32 (Wageningen University and Research, Wageningen, The Netherlands).

Figure 4

Effects of qRRV_TX2WSE_ch5 diplotypes on C_t value in diploid rose populations. Levels not connected by the same letter are significantly different at p < 0.05 (Steel–Dwass nonparametric multiple comparison test).

Figure 5

Effects of qRRV_TX2WSE_ch5 diplotypes on overall severity BLUEs in diploid rose populations. Levels not connected by the same letter are significantly different at p < 0.05 (Steel–Dwass nonparametric multiple comparison test).

Figure 6

Effects of qRRD_TX2WSE_ch1 diplotypes on overall severity BLUEs in diploid rose populations. Levels not connected by the same letter are significantly different at p < 0.05 (Steel–Dwass nonparametric multiple comparison test).

Figure 7

Sources of haplotypes for qRRV_TX2WSE_ch5 and qRRD_TX2WSE_ch1 identified in diploid rose populations. Haplotypes are color-coded as follows: black, original score; blue, original score changed by PediHaplotyper; yellow, original score removed by PediHaplotyper; pink, score manually imputed. Red and blue lines indicate female and male parents, respectively. SEB-ARE and SOB-ARE indicate Swamp Rose EB ARE and Swamp Rose OB ARE, respectively.

Figure 8

Pedigree of the TX2WSE diploid rose families (indicated in black) used in this study. Progeny numbers indicate the number phenotyped in Crossville, Tennessee for RRD resistance. Direct parents are indicated with gray, and red and blue lines indicate female and male parents, respectively. Founders and progenitors are indicated in white. SEB-ARE and SOB-ARE indicate Swamp Rose EB ARE and Swamp Rose OB ARE, respectively. Figure created with Pedimap v1.2 (Wageningen University and Research, Wageningen, The Netherlands) [33].