ASSESSMENT OF STRIGA RESISTANCE IN WILD RELATIVES OF SORGHUM UNDER FIELD CONDITIONS

The witch-weeds (Striga spp.) are destructive root parasites of many crops. They result in considerable crop damage, especially in the semiarid tropic parts of the world. The parasite control is difficult due to the complexity of the parasite life cycle, and the large number of seeds produced by the parasite with prolonged viability. The most promising way of controlling the parasite is through the development of the resistant crop varieties. Identification of different sources of resistance will enhance breeding for resistant varieties. Wild relatives of sorghum are rich in genetic diversity and have a broad genetic base including novel and valuable traits like Striga and disease resistance. In this context, 55 wild sorghum lines were collected from three regions of Sudan including eastern Sudan (Gadaref), central Sudan (Gezira), and western Sudan (North Kordofan). The collected germplasm was assessed for Striga resistance using artificial infestation. The results showed a significant difference in the number of Striga emerged plants compared to the checks. Hence, the significant difference was observed in the number of days to 50% plants to reach flowering, plant height, and grain yield per hectare. The wild relatives were also morphologically characterized and the result showed 55 lines structured in six groups independently from their geographical regions.

Identification and utilization of conserved genetic material of wild species help to sustain crop improvement, particularly in breeding for biotic and abiotic stresses.Information from genetic diversity studies permits the classification of genotypes into heterotic groups which are important for hybrid development and estimating the relative strengths of the factors affecting the genetic makeup as mutation, natural selection, migration, and genetic drift.Understanding of genetic variability is useful to create segregating populations with maximum genetic variability for further selection (Barrett and Kidwell, 1998).This information is also useful for better understanding of evolutionary trends and will help in gene bank management and strategies for collection and conservation of the germplasm.
Sorghum is originated in the area bordering between Ethiopia and Sudan (Dogget, 1988).In Sudan, wild relatives of sorghum have been little investigated, and few results have been published.In this context, we collected wild relatives of sorghum from Striga (Striga hermonthica) infested fields of eastern (Gadaref), central (Gezira) and western (North Kordofan) parts of Sudan.
The objectives of this study were to assess Striga resistance in wild relatives of sorghum as well as to characterize and determine the genetic diversity of Sudanese wild sorghum accessions.

Plant material
Fifty-five wild relatives of sorghum entities were collected from three areas in Sudan: central (Gezira), eastern (Gadaref area), and western (North Kordofan), which can be described as irrigated, high and low rainfall areas, respectively (Figure 1).Four cultivated sorghum cultivars were used as out-groups.

Field testing
A selected wild relative of sorghum (55 lines) collected germplasm was tested in the Striga infested plot at the Elobied Agricultural Research Station farm.Data was collected for the number of emerged Striga plants, 45 and 60 days after crop emergence.The progenies were also evaluated for days to 50% flowering, plant height, and yield.Four lines SRN39-40, Tex623B, Tex623A and Carper-R were used as control.A randomized complete block design (RCBD) with three replicates was used.
The selected genotypes were also morphologically characterized using ten parameters including: awn, midrib color, senescence, waxy layer, plant height, exertion, plant color, 50% flowering, seed shattering, and head shape.
The statistical analysis was carried out using Statistix8.1 and GenSTAT for mean separation and clusters respectively.

50% flowering
The results showed significant differences in all tested lines, comparing with the resistant check SRN-39.Differences appeared with the use of LSD at the 0.05 level of probability (Tables 1 and 4).

Plant height
Field screening results showed a significant difference (Table 1) for all wild relatives and the resistant cultivar SRN-39.The other three lines as Tex623A, Tex623B and Carber-R were wiped out.The examined wild sorghum accessions and the resistant control SRN-39 showed a normal plant height.Differences appeared using the least significant differences (LSD) at the 0.05 level of probability (Table 5).

Grain yield
Analysis of variance of the grain yield revealed significant differences (P= 0.05) between the wild relatives and the resistant check SRN-39 in Striga infested plots (Tables 1 and 6).Mean grain yield for each genotype in the experiment is presented in Table 1.The highest grain yield was produced by the resistant cultivar SRN-39, while the wild relatives showed lower grain yield (Table 1).Cluster analysis and the effect of cultivated region The cluster analysis developed based on ten morphological characters included: awn, midrib color, senescence, waxiness, plant height, exertion, plant color, 50% flowering, seed shattering, and head shape.The results showed the accessions structured in 4 major groups (Figure 2).The wild relatives were structured in different groups and subgroups independently from their geographical regions.This work is part of the project that aimed to introgress Striga resistance into Hageen Dur1 (F1 hybrid generation), which is one of the hybrids released for commercial use, showing good performance in irrigated and high rainfall areas of Sudan.To achieve this goal, 55 wild relatives of sorghum were collected, assessed for Striga resistance and morphologically characterized.Striga resistance in wild and related species has not been fully exploited, and a few surveys of wild sorghums for Striga resistance have been reported (Deodikar, 1951;Lane et al., 1994;Mohamed et al., 2003).
Through this work, 55 wild relatives of sorghum were tested in Striga infested plots, with the resistant cultivar, SRN-39, as a resistant control.However, the parental lines of Hageen Dura-1, Tex623A, Tex623B and Carber-R, were used as the susceptible control.
Counting Striga plants is considered as one of the field measurements that can determine host plant resistance.Our data indicated that the susceptible checks, Tex623A, Tex623B and Carber-R, had the highest number of the emerged Striga plants among the tested genotypes.This may be due to production or exudation of the stimulant from the three checks.However, wild sorghum showed a lower number of Striga emerged plants, some of which showed immunity in which no Striga plants were observed surrounding the host plant.Wild sorghums may be sources of unique resistance traits lacking in cultivars since they have evolved under selective pressures imposed by Striga spp.(Rich et al., 2004).It is noteworthy that Striga resistant sorghum varieties may produce the low number of Striga plants and the number of viable seeds in the soil, but they are often not locally adapted and morphologically inferior (Haussmann et al., 2004).The wild relatives were surveyed and tested in the field, but they were not examined for the resistance mechanisms.In the previous studies, several mechanisms of Striga resistance were mentioned in sorghum and its wild relatives (Ejeta et al., 2000;Gurney et al., 2002;Heller and Wegmann, 2000).
The susceptible control showed no flowering even one hundred and twenty days after germination; it was wiped out by the end of the season.The observed results correspond with the previously published reports stating that sorghum yield losses may reach 100% on heavily infested soils (Parker and Riches, 1993;Khidir, 1983;Obilana, 1983).The results confirmed the high susceptibility of the cultivars, Tex623A, Tex623B and Carber-R, and a high level of infestation in the naturally infested plot.
A significant difference was observed in plant height among these genotypes.The susceptible ones were poorly developed and stunted and genotypes with resistance had normal height.This result is coherent and corresponding with our previous field evaluation of 19 sorghum lines developed by using marker assisted selection in our local breeding program (Gamar and Abdalla, 2013).
The study also measured the grain yield.The three susceptible parents showed no seeds (highly affected by Striga damage), while the resistant control SRN-39 achieved the highest grain yield.The wild relatives of sorghum showed a normal feature of growing, which means the tested lines were not affected by Striga infection.Empirical breeding for Striga resistance in field crops has relied on selection of host plants that allow the emergence of few parasitic plants and show little or no loss in productivity of the crop (Mohamed et al., 2003).
The cluster analysis of 55 wild relatives revealed that Tex623A and Tex623B were structured in one group, while Carper-R, the restorer line, was structured in a different group.This showed the ability of morphological markers to discriminate between the tested lines.The wild relatives of sorghum were structured in different groups independently from their geographical regions and Striga resistance.

Conclusion
The study revealed the Striga resistance in wild relatives of sorghum under field conditions.The wild relatives of sorghum showed immunity against Striga invasion.The lines showed that immunity was expected to have an impact on breeding for Striga resistance in sorghum.

Table 1 .
Results of the wild relatives of sorghum in the Striga infested field.
Legend: C: Central, G: Gezira, W: Western, O: Out-group.**Means with different superscript letters in the same column are significantly different at P = 0.05.

Table 2 .
ANOVA table of the Striga 1 st count.

Table 3 .
ANOVA table of the Striga 2 nd count.

Table 4 .
ANOVA table of days to 50% flowering.

Table 5 .
ANOVA table of plant height.

Table 6 .
ANOVA table of plant height.