Variation Between Three Eragrostis tef Accessions in Defense Responses to Rhopalosiphum padi Aphid Infestation

Abstract

Tef (Eragrostis tef), a staple crop that originated in the Horn of Africa, has been introduced to multiple countries over the last several decades. Crop cultivation in new geographic regions raises questions regarding the molecular basis for biotic stress responses. In this study, we aimed to classify the insect abundance on tef crop in Israel, and to elucidate its chemical and physical defense mechanisms in response to insect feeding. To discover the main pests of tef in the Mediterranean climate, we conducted an insect field survey on three selected accessions named RTC-144, RTC-405, and RTC-406, and discovered that the most abundant insect order is Hemiptera. We compared the differences in Rhopalosiphum padi (Hemiptera; Aphididae) aphid performance, preference, and feeding behavior between the three accessions. While the number of aphid progeny was lower on RTC-406 than on the other two, the aphid olfactory assay indicated that the aphids tended to be repelled from the RTC-144 accession. To highlight the variation in defense responses, we investigated the physical and chemical mechanisms. As a physical barrier, the density of non-granular trichomes was evaluated, in which a higher number of trichomes on the RTC-406 than on the other accessions was observed. This was negatively correlated with aphid performance. To determine chemical responses, the volatile and central metabolite profiles were measured upon aphid attack for 4 days. The volatile analysis exposed a rich and dynamic metabolic profile, and the central metabolism profile indicated that tef plants adjust their sugars and organic and amino acid levels. Overall, we found that the tef plants possess similar defense responses as other Poaceae family species, while the non-volatile deterrent compounds are yet to be characterized. A transcriptomic time-series analysis of a selected accession RTC-144 infested with aphids revealed a massive alteration of genes related to specialized metabolism that potentially synthesize non-volatile toxic compounds. This is the first report to reveal the variation in the defense mechanisms of tef plants. These findings can facilitate the discovery of insect-resistance genes leading to enhanced yield in tef and other cereal crops.

Keywords: aphid; cereal crop; electrical penetration graph; green leaf volatile; insect behavior; trichome; volatile organic compounds.

FIGURE 1

Insect abundance on the three tef accessions from the field survey. Pie chart of the insects that were collected in three locations in the field (total of 3 m² and normalized to 1 m²) sorted by orders. The insect families monitored in each order are included in the table. Sampling was performed prior to flowering (late May 2019).

FIGURE 2

Aphid performance and preference for 1-month-old tef plants. (A) A Y-tube olfactometer choice bioassay was used to determine the aphid preference. Bars represent the average number of aphids (mean ± SE, n = 5). In each replicate, 30 aphids were tested. The asterisk indicates significantly different choices as determined by the chi-square goodness of fit test at P < 0.05. (B) A non-choice bioassay was used to determine the differences in aphid performance between the three tef accessions. The whole-plants were infested with 20 adult R. padi aphids for 4 and 7 days, then the total number of adult and nymphs was counted (mean ± SE, n = 14–15). On the top, a summary of the two-way ANOVA, comparing the aphid reproduction among the three accessions at two infestation time periods 4 and 7 days (p < 0.05). Different letters above the bars indicate significant differences, using one-way ANOVA followed by TukeyHSD test separately for each time point, corrected with the false discovery method.

FIGURE 3

Trichome density of three tef leaves. Bars represent the average number of trichome density per mm² (mean ± SE, n = 10). On the top, a summary of the two-way ANOVA, comparing the number of trichome among the three accessions at three leaf sections (p < 0.05). Different letters above the bars indicate significant differences, using one-way ANOVA followed by TukeyHSD test separately for each time point, corrected with the false discovery method.

FIGURE 4

Heatmap of the VOC profile of aphid-infested and untreated control tef plants. The VOCs were selected using two-way ANOVA comparing the three accessions and the aphid treatment. The Euclidean distance with Ward’s minimum variance method was calculated using the default parameters of the MetaboAnalyst software, and the graph was created in R and presented in average values. Colors correspond with concentration values (autoscaled parameters), where red indicates high levels, and blue indicates low levels (n = 4–5 biological replicates).

FIGURE 5

Heatmap of the central metabolites profile of aphid-infested and untreated control tef plants. The Euclidean distance with Ward’s minimum variance method was calculated using the default parameters of the MetaboAnalyst software, and the graph created in R. Colors correspond with concentration values (autoscale parameters), where red indicates high levels and blue indicates low levels (n = 4–5 biological replicates).

FIGURE 6

Transcriptomic overview of RTC-144 tef leaves infested with R. padi aphid for different periods. (A) PCA plot was generated using 35,284 genes. (B) Venn diagram illustrating the number of genes up- or down-regulated by aphid infestation in the time course. p < 0.05 FDR, and absolute fold change > 2 (n = 3 biological replicates for each time point).