Performance of SAP low germination stimulant 1 sorghum under natural Striga infestation
The ultimate test for Striga resistance performance under natural pest infestations. We, therefore, evaluated the resistance response of the SAP lgs1 accessions in a Striga- infested field in western Kenya by measuring Striga emergence using the metric of Area Under Striga Number Progressive Curve (AUSNPC); a modification of the area under the disease progress curve (AUDPC) developed for plant pathology studies (Simko & Piepho 2012). In this approach, a quantitative measurement of Striga infestation is determined over time. For our case, we determined Strigainfestation using data collected at 42, 56, 70, 84 and 98 days after planting.
Resistance measured by AUSNPC showed that most SAP lgs1 had highStriga resistance (Figure4a). We would like to point out SAPlgs1 PI533976 which had remarkably lower emergence (p value = 0.00017) compared to the base mean of all accessions. Furthermore, the resistance of PI533976 was higher than that of known Strigaresistant varieties including SRN39, Framida and IS9830. Also striking was the field resistance displayed by the SAP LGS1 PI655054 accession, described earlier in germination bioassays and pot experiments. Field resistance of this genotype affirmed its importance as a Striga resistant accession. Expectedly, some SAP lgs1accessions for example PI656040, PI585295, and PI656010 showed low resistance, even comparable to the susceptible Shanqui Red andLGS1 PI533839. These same genotypes had displayed low resistance in pot experiments. The low resistance could be attributed to low post-attachment resistance. Two of these accessions (PI585295, PI656010) that were assayed biochemically were also shown to produce relatively large quantities of orobanchol (Figure 2c). Even though orobanchol is a less potent stimulator of germination compared to 5-deoxystrigol,Striga populations from western Kenya are particularly sensitive to high concentrations of orobanchol relative to populations from other locations (Haussmann et al. 2004; Bellis et al. 2020).
In addition to Striga resistance screening, field evaluation allowed us to determine the suitability of SAP lgs1 accessions for growth in western Kenya as well as evaluate other desirable agronomic traits (Table 1). Considering that the SAP accessions were selected for photoperiodism sensitivity, we sought to determine days to flowering by measuring maximum days to 50 % flowering (DFL50 %). We found that DFL50 % of the latest and earliest genotypes averaged between 76.67 in PI5610710 and 55 in PI533976 indicating that the SAP accessions are in general early maturing and do not have variation in flowering time. We also measured the associated trait of plant height under field conditions. Results showed that the average plant height ranged from 65 cm to 235 cm. The plant height of the SAP lgs1accessions are comparable to those of sorghum genotypes grown in Kenya such as SRN39 (125.33.73cm), Framida (155.33 cm) and IS9830 (173.67 cm). Lastly, we compared the average yield of SAP lgs1 lines to that of popular Striga -resistant varieties in Kenya. The top 3 highest yielding varieties were SAP lgs1 PI656094 (2.9667g), PI656096 (3.0333g), and SRN39 while the poorest yielding varieties were PI655979, PI561071and PI533839. The yields are comparable to varieties Framida (2.8667g), IS9830 (2.6333g) grown in Kenya indicating that the lines are suitable for adoption in Kenyan agro ecological regions.
The other aspect of our field experiments was to determine the effect ofStriga infestation on growth of the sorghum plants – given thatStriga infection is associated with severe growth retardation (Fujioka et al. 2019a). An effective way to determine the extent to which Striga infection affects growth, and how best an infected host copes with the infection is to correlate host growth andStriga emergence. Hosts that cope well with infection – described as tolerant genotypes – show less severe effects of parasite infection (Mwangangi et al. 2021). We found that Strigainfection reduced yield by up to 27 %. The most severe reduction in growth was in PI576385 (26 %) while SRN39 and Shanqui Red incurred losses of 6 % and 13 %. Remarkably, PI561071 and PI585295 did not suffer significant yield losses. When we correlated sorghum’s yield withStriga emergence (AUSNPC) we found a weak positive correlation between yield loss and AUSNP (R = 0.33, p = 0.25) indicating that most genotypes were tolerant (Figure 4b). PI561071, PI656096, PI6506054 maintained high yields even with some Striga infestation.
To summarize our field experiments, we ranked Striga tolerance, field resistance and other desirable agronomic traits in SAP lgs1lines to find the best performers. To achieve this, we used the Rank summation index (RSI) method proposed by (Mulamba & Mock) where a genotype is assigned a rank based on performance for each trait (Figure 4c). A final score (RSI) is then assigned based on all traits. Using this analysis, the best 3 performing genotypes were SRN39, SAPlgs1 PI533976 (caudatum, USA), and SAP LGS1 PI656054 (Kafir, South Africa).
Although further field evaluation is required, our results show that some SAP lgs1 lines are well suited to environments in Kenya and can potentially be used in Striga management programs. A representative photograph depicting the performance of the Strigaresistant SAP lgs1 PI561071 against the susceptible LGS1PI533839 and Shanqui Red is shown in Figure 4d.