Screening and Evaluation for Antixenosis Resistance in Wheat Accessions and Varieties to Grain Aphid, Sitobion miscanthi (Takahashi) (Hemiptera: Aphididae)

The grain aphid, Sitobion miscanthi causes serious damage by removing nutritional content from wheat plants and transmitting viral diseases. The use of resistant wheat cultivars is an effective method of aphid management. To identify S. miscanthi resistant cultivars, preliminary antixenosis resistance screening was conducted on 112 Ethiopian and 21 Chinese wheat accessions and varieties along with bioassay to test for further antixenosis resistance, identification of aphid feeding behavior using electrical penetration graph (EPG), and imaging of leaf trichome densities using a 3D microscope. According to antixenosis resistance screening, one highly-resistant, 25 moderately-resistant, and 38 slightly-resistant wheat cultivars to S. miscanthi were identified. Aphid choice tests showed that Luxuan266, 243726, and 213312 were the least preferred after 12, 24, 48, and 72 h of S. miscanthi release. Longer duration of Np, longer time to first probe, and shorter duration of E2 waveforms were recorded in Lunxuan266, 243726, and 213312 than in Beijing 837. The trichome density on adaxial and abaxial leaf surfaces of Lunxuan266, 243726 and 213312 was significantly higher than on those of Beijing 837. We concluded that Lunxuan266, 243726, and 213312 were antixenosis resistant to S. miscanthi based on the choice test, EPG results, and leaf trichome densities.


Introduction
Wheat (Triticum aestivum) is a major and healthy cereal crop that is consumed as a staple human food in many parts of the world [1]. China is among the major producers and consumers of wheat in the world [2]. Wheat is cultivated in almost all provinces of China and its production has rapidly grown within the past thirty years [3]. Ethiopia is one of the largest wheat producers in sub-Saharan countries [4]. Wheat accounts for about 18% of total cereal production and is the main crop for food security in Ethiopia [5]. However, the production and productivity of wheat are highly challenged by biotic and abiotic stresses. Insect pests and diseases are among the biotic stresses that cause wheat damage [6]. Among insect pests, aphids cause an average yield loss of 15% in China [7] Aphids are a major agricultural insect pest throughout the world. Cereal aphid causes serious damage by removing nutritional contents from the host plant [8]. They are also vectors of the barley yellow dwarf virus [9]. Aphids penetrate the plant epidermis with their stylet and push it through the parenchyma extracellular tissue until they reach the vascular bundles. When probing and feeding, aphids secrete saliva containing effector enzymes into their host plants to change the host cell's processes and facilitate infestation [10]. Gel and watery saliva are the key factors in the process of aphid-plant interactions [11]. Watery salivary is used to detoxify secondary metabolites such as plant phenols and alkaloids [11]. The aphid clones of some species differ in probing and feeding behavior on some infested

Preliminary Antixenosis Resistance Screening
The results of preliminary antixenosis resistance screening after 24 and 48 h of winged S. miscanthi release are presented in Table 1. After 48 h from winged aphid release, the 136 wheat accessions and varieties were categorized into one highly-resistant variety (Lunxuan 266), 25 moderately-resistant accessions and varieties (five Chinese and 20 Ethiopian), and 38 slightly-resistant accessions and varieties (eight Chinese and 30 Ethiopian), 29 slightly-susceptible accessions and varieties (four Chinese and 25 Ethiopian), 19 moderately-susceptible accessions and varieties (1 Chinese and 18 Ethiopian), and 21 highly-susceptible accessions and varieties (3 Chinese and 18 Ethiopian). Then, differences in the number of winged S. miscanthi settling on the wheat accessions and varieties were evaluated. The numbers of winged S. miscanthi settling on the wheat accessions were clustered using the heatmap TBtool for additional clarification of the antixenosis resistance screening ( Figure S1). Groups 1 and 2 were comprised of 10 and 23 wheat accessions, respectively, with high and moderate antixenosis resistance, such as Lunxuan 266, 243726, and 213312. Significant differences in the mean number of S. miscanthi populations were observed among the screened wheat accessions and varieties.  Table 2 shows the mean number of adult S. miscanthi settling per plant within 48 h and produced per plant within 28 days in the antixenosis and antibiosis resistance screening of wheat accessions and varieties. The results indicated that the wheat accessions and varieties were divided into high and moderate antixenosis resistance groups, whereas in the case of antibiosis resistance, these wheat accessions and varieties were grouped into slightly resistant, slightly susceptible, and moderately susceptible. Therefore, the antixenosis and antibiosis resistance mechanisms of these wheat accessions and varieties were not found to be on the same level of the resistance scale.

Choice Test for Further Antixenosis Identification
There was a significant difference (p < 0.05) in the percentage of winged S. miscanthi settling on the 14 wheat accessions and varieties tested 12, 24, 48, and 72 h after aphid release ( Figure 1A-D). All of the tested accessions and varieties showed significant differences from Beijing 837. After 12 h., the percentage of aphids settled on Lunxuan266, 243726, 213312, and 7276 was significantly lower than the percentage settled on 243714, 207845, 227068, and Beijing 837 ( Figure 1A). After 24 h, the percentage of aphids settled on Lunxuan266, 243726, and 213312 was significantly lower than the percentage settled on 7248, 7407, 243714, 207845, 227068, and Beijing 837 ( Figure 1B). After 48 h, the percentage of aphids settled on Lunxuan266, 243726, and 213312 was significantly lower than the percentage settled on 203971, 243714, 207845, 227068, and Beijing 837 ( Figure 1C). After 72 h, the percentage of aphids settled on Lunxuan266, 243726, and 213312 was significantly lower than the percentage settled on 207045 and Beijing 837 ( Figure 1D). The results of the choice test indicated that Lunxuan266, 243726, and 213312 were the least preferred plants among the 14 wheat accessions and varieties tested.  12,24,48 and 72 h, respectively. Bars represent the mean percentage of aphids settled (mean ± ME, n = 16). Different letters above the bars indicate significant differences. Table 3, there were significant differences among the wheat accessions and varieties for the EPG parameters associated with the total duration of Np, time to the first probe, duration to first probe, mean duration of E1, time to the first E2, number of E2, the total and mean duration of E2, and time to first sustained E2 (≥10 min) (p < 0.05). The total duration of the Np for Lunxuan266, 243726, and 213312 was longer than that of Beijing 837. The time to the first probe of S. miscanthi feeding on Lunxuan266 and 243726 was longer than that of Beijing 837. The mean duration of E1 waveforms of S. miscanthi feeding on 213312 and Lunxuan266 was significantly longer than that of Beijing 837. The time to reach the first E2 and first sustained E2 (≥10 min) was longer on Lunxuan266 than on Beijing 837. Furthermore, the total and mean duration of E2 waveforms of S. miscanthi feeding on the test accessions was shorter than on Beijing 837. Therefore, the test accessions showed epidermal and phloem-based resistance to S. miscanthi damage.

Trichome Density
As shown in Figure 2, the trichome density on the upper (adaxial) leaf surfaces of wheat accessions and varieties with antixenosis resistance (Lunxuan266, 243726, and 213312) was significantly higher than on Beijing 837 (p < 0.05). Furthermore, significantly higher trichome densities were recorded on the lower (abaxial) leaf surface of Lunxuan266, 243726, and 213312 than on Beijing 837 (p < 0.05). The variations in trichome density could be caused by the genetic differences among the evaluated wheat plants. As a result, the increased trichome densities on the test accessions and varieties act as a physical barrier, preventing aphid-stylet penetration for feeding.

Correlation Analysis between Number of Winged Sitobion micanthi Settling and EPG Parameters on Wheat Seedlings
As shown in Table 4, the number of winged S. miscanthi settling on wheat leaves 12, 24, 48, and 72 h after release was negatively correlated with Np, E1, C, and Pd, and positively correlated with E2 and G waveforms. The number of S. miscanthi settling on the wheat leaves was negatively correlated with Np and E1 waveforms to a significant extent (p < 0.05), and positively correlated with E2 to a significant extent. These results suggest the presence of some resistance factors in the epidermis and/or mesophyll cell and phloem of wheat leaves that have repellent or deterrent effects on winged S. miscanthi.

Correlation Analysis between Number of Aphids Settling and Trichome Density on Wheat Accessions and Varieties
As shown in Table 5, the number of winged S. miscanthi settling on wheat leaves after 12, 24, 48, and 72 h was significantly negatively correlated (p < 0.05) with trichome density on the adaxial and abaxial leaf surfaces of the four wheat accessions and varieties evaluated. These results suggest that the higher trichome density on leaves of Luxyaun266, 243726, and 213312 hinders the settlement and feeding of winged S. miscanthi.

Preliminary Antixenosis Resistance Screening
Our study suggested that some of the wheat accessions and varieties included in the antixenosis resistance screening were less preferred by the winged S. miscanthi, which is similar to previous findings [32]. Thus, the screened 136 wheat accessions and varieties were categorized as one highly resistant, 25 moderately resistant, 38 slightly resistant, 29 slightly susceptible, 19 moderately susceptible, and 21 highly susceptible. The smaller number of winged S. miscanthi settling on highly and moderately resistant accessions and varieties indicates antixenosis resistance. Many studies have demonstrated that plant olfactory cues and some chemicals such as hydroxamic acid affect the preference of aphids for plants [33]. The physical structures of plants, chemicals such as repellents and deterrents, and antifeedant effects on herbivores contribute to antixenosis resistance [34]. Most insects have innate behaviors to orient towards or away from plants using odor clues emitted by the plants; for example, European corn borer (ECB) larvae and adults accepted maize, but rejected spinach because it produces phytoecdydtrroids, which is toxic and deterrent to ECB [35]. Thus, these wheat accessions and varieties that were less preferred by S. miscanthi could have inherited repellent behavior.
One highly-resistant and 12 moderately-resistant wheat accessions and varieties in the antixenosis resistance screening were also evaluated for antibiosis resistance to S. miscanthi, and were slightly resistant, slightly susceptible, and moderately susceptible ( Table 2). Our findings indicate that antixenosis and antibiosis resistance factors are not necessarily found in the same accession and that they do not depend on each other, which is consistent with previous reports [36,37].

Aphid Bioassays for Further Antixenosis Resistance Tests
Variations in antixenosis resistance were found among different plants of some species [38]. Plants emitted a variety of VOCs to protect themselves from herbivores, which also stimulate the neighboring healthy plants to produce defense VOCs (18). However, the amount released of each VOC varies qualitatively and quantitatively depending on the duration after herbivore damage [18,39]. The insect response to individual VOCs depends on the concentration with which the insects can be attracted or repelled [35,40]. For example, Sitophilus granaries and Tribolium confusum were repelled by blending plant VOCs (aliphatic alcohols, aliphatic aldehydes, aliphatic ketones, and aromatics) at >100 ng min −1 concentrations, while they were attracted at 10 ng min −1 concentrations [39,40]. Green leaf volatiles (GLVs) are the herbivore-induced plant volatiles (HIPVs) that attract insect herbivores directly and/or indirectly at higher concentrations [39,41]. For example, (Z)-3-hexenyl acetate, (Z)-3-hexenal, (E)-2-hexenal, (E)-2-hexen-1-ol, and (Z)-3-hexenol were attractive to ECB at higher concentrations, but repellent at low concentrations [42]. Previous studies have shown variations in antixenosis resistance among wheat cultivars for winged S. miscanthi settling 24 h after release [41], and 24, 48, and 72 h after release [43]. Similarly, variations were observed in some winged Rhopalosiphum padi and Schizaphis graminum settling among different wheat accessions [44,45]. Our results indicated that Lunxuan266, 243726, and 213312 were less preferred, whereas Beijing 837 was more preferred 12, 24, 48, and 72 h after winged S. miscanthi release. Plants with antixenosis resistance release increased VOCs in response to aphid infestation, which disturbs aphid settlement [46]. For example, VOCs such as methyl-salicylate (MeSA), linalool, and E-β-farnesene emitted from Prunus apices have repellent or dispersive effects on aphids [47]. The increased release of (E)-β-farnesene from wheat plants repels aphid pests and attracts the natural enemies of aphids [43]. Plants release EBF, which as an alarm pheromone for aphids and defends plants from aphid infestation by deterring aphids from settling and inducing more winged offspring that help to leave the plant [48]. Therefore, the variations in preference levels of winged S. miscanthi to the wheat accessions and varieties could be due to the variations in the types and concentrations of VOCs produced among the wheat plants evaluated.

Electrical Penetration Graph
Wheat plants emit repellent volatile compounds that impede aphids' settling and feeding behaviors [49]. EPG helps to determine the effects of plant repellents and antifeedants on the feeding behavior of sap-sucking insects [50]. In our study, the total duration of Np waveforms on the test accessions was significantly longer than om Beijing 837. Similar results have been reported on Solstice, W064, W068, and W591 wheat landraces against R. padi [51]. The choice test for S. miscanthi also supported these results and indicated a clear lack of preference for the test accessions compared to Beijing 837. These results indicate the presence of repellent volatile chemicals originating from the host plant [52]. Additionally, the aphids probed more slowly on the aphid-resistance wheat accessions and varieties compared to the susceptible accession Beijing 837, which suggests that resistance factors may also be present on the surfaces of leaf tissue. This result is also supported by the higher trichome density on the leaf surfaces of aphid-resistance wheat accessions and varieties compared to Beijing 837. The aphid S. miscanthi feeding on 213312 and Lunxuan 266 showed a longer total duration of E1 than on Beijing 837. The longer duration in E1 suggests the presence of some toxic and/or deterrent chemicals preventing the phloem sap from being ingested [53]. The time taken to reach the first E2 and first sustained E2 (≥10 min) on Lunxuan 266 was longer than on Beijing 837. Moreover, all of the test accessions and varieties showed a shorter total and mean duration of E2 than Beijing 837, which is coherent with the studies on host-plant resistance occurring in the phloem [53]. The shorter duration of E2 is an indication of antixenosis because the phloem activates a deterrent to aphid feeding [54]. Therefore, the longer duration of Np, time to the first probe, E1, time for the first sustained E2 (≥10 min), and shorter duration of E2 in the tested accessions indicate the antixenosis resistance to S. miscanthi feeding, which is most likely attributable to the presence of repellent factors in the wheat seedlings. Aphids feeding on host plants with epidermal cell and phloem-based resistance demonstrate the repellent and deterrent effects of plant repellence volatiles, which is consistent with our findings.
Some wheat volatiles are repellent to winged aphids alighting and feeding [55]. The host plants containing higher concentrations of deterrent chemicals negatively affected aphid feeding behaviors [49]. The number of winged S. miscanthi settling on wheat leaves 12, 24, 48, and 72 h after release showed a significant negative correlation with Np and E1, and a significant positive correlation with E2 waveforms. The results indicated that the number of winged aphids settling on the wheat accessions may be determined by repellent volatiles such as (E)-β-farnesene and the MeSA content of the plant, which is consistent with the previous report [56]. We have tried to collect volatile chemical components from wheat seedlings, but failed, as few volatiles were identified using GC/MS. It was speculated that the number of wheat seedlings used for emission collection was not large enough. Therefore, it is necessary to identify the volatile components of wheat that have repellent effects on S. miscanthi through further study.

Trichome Density
Trichomes are among the physical defensive mechanisms of plants, impeding insect movement, feeding, development, and reproduction [57]. The effects of such defensive structures against aphids has been reported in wheat and E. tef [57,58]. A higher trichome density on the upper leaf surface of the PI 137739 wheat genotype is antixenotically resistant to RWA [59]. The trichome densities of the same plant species can differ due to growth stage, genetic makeup, and leaf location, and density is also induced by insect damage [60]. In our study, we found higher trichome densities on the upper and lower leaf surfaces of Lunxuan 266, 243726, and 213312 compared to Beijing 837, which indicates that genetic variation affects trichome density. Wheat plants with dense trichomes can hamper cabbage whitefly, Aleyrodes proletella, from reaching its exact probing site [36]. Trichome density affects the feeding behaviors of R. padi and limits its performance in wheat plants [57]. Similarly, wheat cultivars with a higher number of trichomes extended the time for R. padi penetration [61]. In our EPG result, the time to the first probe on test accessions was longer than on Beijing 837. Thus, the cause for the lower number of winged S. miscanthi settling on test accessions and varieties could be caused by the higher number of trichomes on wheat seedlings. The number of winged S. miscanthi settled on the wheat accessions and varieties was negatively correlated with trichome density. The leaf trichome density of nine cotton varieties showed a negative correlation with aphid abundance in the plants [62]. Therefore, our results indicate that a higher trichome density on leaves may play an important role in the reduction in winged aphids settling on the tested wheat accessions and varieties.

Insects and Plants
The experiment was conducted inside a greenhouse at the Langfang Research Station (39 •  Wheat accessions (112 from the Ethiopian Institute of Biodiversity /EIB/, Addis Ababa, Ethiopia) and wheat varieties (21 from the Institute of Crop Sciences, CAAS, Beijing, China) were used for screening of antixenosis resistance to winged S. miscanthi. One highlyresistant variety (Lunxuan266) and eleven moderately-resistant accessions (243726, 213312, 7276, 243710, 8324, 7407, 203971, 243714, 7298, 207845, and 227068) from the preliminary screening were used for aphid bioassay to conduct further tests on antixenosis resistance against winged S. miscanthi. Both EPG recording and trichome density scanning were performed on the slightly-preferred Lunxuan266, 243726, and 213312 wheat accessions, and varieties were taken from the aphid bioassay test. Beijing 837 was included in all activities as a negative control.

Preliminary Antixenosis Resistance Screening of Wheat Accession and Varieties against Sitobion miscanthi
Six seeds of each of the 133 wheat accessions and varieties used for the antixenosis resistance screening were grown in plastic pots as described above, and replicated two times. After germination, the seedlings were thinned to maintain four seedlings per pot. For simplification, the 23 pots containing wheat seedlings at two-leaf visible stages (12 days after germination) [64] were grouped as one and arranged at random in a circle inside a 2 × 2 × 1.5 m rectangular gauze cage to protect against the winged aphids escaping [41]. Two hundred thirty (230) winged S. miscanthi having emerged within 24 h were released into the center of the experimental plants in the cage. The greenhouse was dark to avoid the artificial orientation of aphids to the screened wheat accessions and varieties in response to light. The number of adult aphids that settled on each accession and variety was counted and recorded after 24 and 48 h. The number of aphids were counted between 10:30 a.m. and 1:30 p.m., when plants are active in odor releasing.
Aphid indexes were calculated as the average number of winged aphids settled per plant of each wheat accession and variety over the average number of winged aphids settled per plant for all the wheat accessions and varieties screened. The calculated aphid index was expressed as 0 (immunity) to 6 (highly susceptible), to indicate the aphid infestation severity according to the Painter's method [65], with the aphid index being 0 = immunity; 1 = high resistance (HR), with aphid indexes ranging from 0.01 to 0.30; 2 = moderate resistance (MR), with aphid indexes ranging from 0.31 to 0.60; 3 = slight resistance (LR), with aphid indexes ranging from 0.61 to 0.90; 4 = slight susceptibility (LS), with aphid indexes ranging from 0.91 to 1.20; 5 = moderate susceptibility (MS), with aphid indexes ranging from 1.21 to 1.50; 6 = high susceptibility (HS), with aphid indexes exceeding 1.50.

Aphid Bioassay for Further Antixenosis Resistance Test (Choice Test)
One highly and 12 moderately-resistant wheat accessions and varieties selected from preliminary screening, plus one susceptible variety (Beijing 837) were used for the choice test. Six seeds of each treatment were planted in each plastic pot. After germination, the seedlings were thinned to maintain four seedlings per pot. Fourteen pots containing four wheat seedlings at the two-leaf stage were randomly arranged in a circle within a rectangular gauze cage 2 m × 2 m × 1.5 m. Then, 140-winged S. miscanthi have emerged within 24 h were released into the center of the test plants in the cage. The greenhouse was dark to avoid the artificial orientation of aphids to the test plants in response to light. The number of aphids that settled on each treatment was counted and recorded after 12, 24, 48, and 72 h [66]. The counting was performed between 10:30 a.m. and 1:30 p.m. Each treatment was replicated 16 times. The number of winged aphids settled per treatment was divided by the total number of aphids settled per plants of each replication to calculate the proportion of S. miscanthi settling on each treatment.

Feeding Trial with EPG Recording
The feeding behavior of S. miscanthi on wheat leaves was recorded using the EPG (Giga-8d) method [67]. The EPG method differentiates between the feeding behaviors that indicate the acceptance or rejection of the sucking insects of the plant species, depending on variations in the duration of EPG waveforms [68]. Five seeds of each wheat accession and variety (Lunxuan 266, 243726, and 213312, and Beijing 837) were planted in a plastic pot 10 cm in diameter. Winged S. miscanthi were collected from the rearing cage and starved for about 1.5 h, then connected to 15 µm pieces of 2-3 cm long gold wire using water-soluble conductive silver glue. The silver glue was also used to connect the gold wire to about 3 cm copper wire, which was then connected to a brass pin. The device was attached to an eight-channel 'Giga-8 DC amplifier with 1 GΩ input resistance (EPG-systems, Wageningen, Netherlands) and covered within a grounded Faraday cage to protect against external noise at 21 ± 1 • C. Each aphid was placed on the first leaf of an individual eleven-day-old wheat seedling. A ground electrode was inserted into the soil of each potted plant and connected to the amplifier. Recording was performed continuously for 6 h from 9:00 to 15:00 on a daily basis. Each aphid and plant were used only once. One replicate of each of the four treatments was run per day and the positions of the plants and probe wires were randomized daily. At least 14 replicates were tested per treatment. The recorded EPG waveforms were manually analyzed using the Stylet+ software [69]. EPG parameters were selected from the Sarria Excel Notebook [70]. The waveform patterns were categorized into non-penetration (Np), stylets pathway phase (C), stylets cell punctures (Pd), xylem ingestion (G), phloem salivation (E1), and phloem ingestion (E2).

Trichomes Density
The trichome density of Lunxuan 266, 243726, and 213312, and Beijing 837 wheat seedlings was counted as described in [57]. Samples were collected from the second leaves of 11-day old seedlings. For trichome visualization, the middle part of the leaf sample was placed on glass microscope slides and covered with glass coverslips. The super depth-offield 3D microscope system (vhx-20000) was used for scanning leaf trichomes. The number of trichomes was counted manually from adaxial and abaxial leaf surfaces and the density of trichomes was calculated per mm 2 . For each wheat accession and variety, five biological replicates with two images per leaf were taken.

Data Analysis
The number of aphids counted in the antixenosis resistance screening of wheat accessions and varieties was analyzed using heatmap clustering by TBtools [71]. All statistical analyses were conducted using SPSS 22.0. software (IBM SPSS statistics 22). Before analysis, the normality and homogeneity of variance of the data were checked. The data that did not follow normal distribution were square root transformed. Data on duration and counts were analyzed through a one-way analysis of variance (ANOVA). P values less than 0.05 were treated as statistically significant. The means were compared using Fisher's least significance difference method for the EPG, trichome density, and aphid bioassay response variables. GraphPad Prism 8.0 was used to draw the curves and graphs for all demographic parameters.

Conclusions
We concluded that the preliminary antixenosis resistance screening showed variations in some winged S. miscanthi settling among the 133 wheat accessions and varieties. Out of 14 wheat accessions and varieties tested in the study, Lunxuan266, 243726, and 213312 showed the best antixenosis resistance to S. miscanthi. The EPG result showed a longer duration of Np and time to first probing, and a shorter duration of E2, indicating antixenosis resistance to S. miscanthi feeding, which is probably due to the presence of repellent factors produced from the tested wheat seedlings. The greater number of trichomes on the upper and lower surfaces of the tested wheat accessions and varieties could be among the causes for the lower number of winged S. miscanthi settling on the tested accessions and varieties. The overall results indicate the presence of some S. miscanthi repellent VOCs, as well as trichomes as physical barriers in the tested wheat accessions and varieties. It is suggested to evaluate these wheat accessions and varieties under field conditions for further confirmation of their antixenosis resistance to S. miscanthi infestation.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/plants11081094/s1, Table S1: Number of winged aphids settled per plant counted after 24 and 48 h after release to wheat seedlings, aphid resistance index, and aphid resistance scale of S. miscanthi on the wheat accessions and varieties screened for antixenosis resistance. Figure S1: Dendrogram clustering of 133 wheat accessions based on the number of winged Sitobion miscanthi settled 48 h after aphid release. Figure S2: A scheme of the rectangular gauge cage (2 × 2 × 1.5 m), wheat seedling, and winged S. miscanthi arrangement for the preliminary antixenosis resistance screening and aphid bioassay for further antixenosis resistance tests (choice test).