Genetic variation and host-parasite specificity of Striga resistance and tolerance in rice: the need for predictive breeding

Abstract

The parasitic weeds Striga asiatica and Striga hermonthica cause devastating yield losses to upland rice in Africa. Little is known about genetic variation in host resistance and tolerance across rice genotypes, in relation to virulence differences across Striga species and ecotypes. Diverse rice genotypes were phenotyped for the above traits in S. asiatica- (Tanzania) and S. hermonthica-infested fields (Kenya and Uganda) and under controlled conditions. New rice genotypes with either ecotype-specific or broad-spectrum resistance were identified. Resistance identified in the field was confirmed under controlled conditions, providing evidence that resistance was largely genetically determined. Striga-resistant genotypes contributed to yield security under Striga-infested conditions, although grain yield was also determined by the genotype-specific yield potential and tolerance. Tolerance, the physiological mechanism mitigating Striga effects on host growth and physiology, was unrelated to resistance, implying that any combination of high, medium or low levels of these traits can be found across rice genotypes. Striga virulence varies across species and ecotypes. The extent of Striga-induced host damage results from the interaction between parasite virulence and genetically determined levels of host-plant resistance and tolerance. These novel findings support the need for predictive breeding strategies based on knowledge of host resistance and parasite virulence.

Keywords: Oryza glaberrima; Oryza sativa; grain yield; photosynthesis; post-attachment resistance; predictive breeding; witchweed.

Figure 1

Rainfall data for field sites in (a) Kyela (Tanzania), (b) Namutumba (Uganda) and (c) Mbita (Kenya) in 2014 and 2015.

Figure 2

Maximum number of emerged Striga plants m⁻² per rice cultivar for field trials at Kyela, Tanzania under Striga asiatica infestation in Kyela in (a) 2014 and (b) 2015, Striga hermonthica infestation in Mbita, Kenya in (c) 2014 and (d) 2015 and S. hermonthica infestation in Namutumba, Uganda in (e) 2014 and (f) 2015. Bars represent ± SE of the least squares (LS) means. Red boxes indicate the position of resistant check genotype NERICA‐2.

Figure 3

Striga numbers and biomass on a subset of rice genotypes observed in the rhizotron system infected with either (a, b) Striga asiatica from Kyela (Sa‐Ky), (c, d) Striga hermonthica from Mbita (Sh‐Mb) or (e, f) Striga hermonthica from Namutumba (Sh‐Na). Bars represent ± SE of the least squares (LS) means. Bars with different lowercase letters are significantly different (P < 0.05).

Figure 4

Resistance phenotypes of WAB935, WAB928, IR38547 and IAC165, screened in the rhizotron systems infected with either Striga asiatica from Kyela (Sa‐Ky), Striga hermonthica from Mbita (Sh‐Mb) or Striga hermonthica from Namutumba (Sh‐Na). Bars: main images, 0.5 cm; inset image, 500 μm.

Figure 5

Rice grain weights under Striga asiatica infestation in Kyela in (a) 2014 and (b) 2015, Striga hermonthica infestation in Mbita, Kenya in (c) 2014 and (d) 2015 and S. hermonthica infestation in Namutumba, Uganda in (e) 2014 and (f) 2015. Bars represent ± SE of the least squares (LS) means. Red boxes indicate the position of resistant check genotype NERICA‐2.

Figure 6

Rice grain yields plotted against maximum number of emerged Striga plants under Striga asiatica infestation in Kyela in (a) 2014 and (b) 2015, Striga hermonthica infestation in Mbita, Kenya in (c) 2014 and (d) 2015 and S. hermonthica infestation in Namutumba, Uganda in (e) 2014 and (f) 2015. Red boxes indicate the position of resistant check genotype NERICA‐2. Roman numerals in (a) refer to different quadrants with groups of genotypes relative to NERICA‐2.

Figure 7

(a) Maximum rice plant height and (b, c) light‐saturated photosynthesis at (b) 30 and (c) 45 d after sowing (DAS) for a subset of genotypes grown in Striga asiatica‐infested (grey bars) and Striga‐free control (white bars) pots. Significant within‐genotype differences between infected and uninfected plants are indicated: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant. Error bars indicate ± SE.

Figure 8

Striga asiatica‐inflicted losses (%) in (a) maximum rice plant height and (b, c) light‐saturated photosynthesis at (b) 30 and (c) 45 d after sowing (DAS), relative to the Striga‐free control plants, plotted against the maximum number of emerged Striga plants, for a subset of genotypes grown in the pot experiment. Genotypes are indicated by different symbols.