Assessment of Striga gesnerioides (Willd.) Resistance and Genetic Characterization of Forty-Six Cowpea ( Vigna unguiculata (L.) Walp.) Genotypes in Ghana

The parasitic weed, Striga gesnerioides , imposes physiological stress on cowpea ( Vigna unguiculata (L.) Walp.) resulting in signiﬁcant yield loss in the regions of northern Ghana. This warranted identiﬁcation of resistant cowpeas for sustainable production. The current work aim was to identify Striga -resistant cowpea genotypes and assess their genetic relatedness. Forty-six (46) cowpea genotypes were screened in pots for their reaction to Striga samples obtained from the upper east, upper west, and northern regions of Ghana and validated with C42-2B and 61R-M2 markers involving DNA ampliﬁcation by PCR assay. Sixteen polymorphic SSR primer pairs were used to assess genetic relatedness among 46 cowpea genotypes. Data were analyzed with PowerMarker V. 3.25 and a dendrogram was generated with MEGA 4. On the whole, 65.2% of the cowpea genotypes had stable resistance to S . gesnerioides from the regions of northern Ghana and 34.8% were susceptible. The C42-2B marker resolved as a single DNA band of 280bp with segregation eﬃciency of 80% and 61R-M2 marker as double DNA bands of 320bp and 380bp with segregation eﬃciency of 60% associated with Striga resistance. Sixteen (16) polymorphic SSR primers distinguished all 46 cowpea genotypes into three clusters. Gene diversity ranged from 0.04 to 0.49 with an average of 0.29. The average allele frequency is 0.78, with a mean genetic diversity of 0.29. Polymorphism information content (PIC) varied from 0.08 to 1.00 with an average of 0.55. Therefore, cowpeas with Striga resistance and other desirable traits can be evaluated and released as varieties for farmers to cultivate.


Introduction
Cowpea (Vigna unguiculata (L.) Walp.) is the second most important legume in the regions of northern Ghana, after groundnuts in terms of area under cultivation, quantity produced, and annual consumption [1]. Ghana is the fifth highest producer of cowpea in Africa [2] with an average of 143,000 MT produced annually on about 156,000 ha of land.
ere has been a projection that the rate of cowpea production for the period between 2010 and 2020 would increase by 11.1% [3]. However, cowpea production in the major cultivation regions of northern Ghana is constrained by both abiotic and biotic factors [4]. One of the major biotic constraints to cowpea productivity among small-holder and resource-poor farmers in the regions of northern Ghana is the attack by parasitic weed, Striga gesnerioides (Willd.) Vatke [5].
S. gesnerioides is a devastating obligate root hemiparasite that primarily parasitizes dicotyledonous species, including cowpea and other legumes [6,7]. Cowpea yield losses associated with S. gesnerioides infestations range from 83 to 100% [5,8]. e degree of damage to cowpea caused by S. gesnerioides is as a result of the close parasitic association between the host cowpea and the parasitic weed. Seeds of S. gesnerioides in the soil germinate in response to specific stimulants produced by cowpea roots [9]. Haustoria developed by the parasites [10] are used to attach the roots and penetrate the vascular tissues to establish vascular connections with the host cowpea [11]. Water, minerals, and organic compounds (photosynthates) are drawn from the cowpea for the development of the parasite [12]. Several control strategies have been developed including cultural practices, the use of chemical control, and breeding for resistance [13]. Control strategies based on the use of herbicides are too expensive for low-input farming systems, whilst cultural practices are inefficient [14]. However, according to [15], host plant resistance has the potential to efficiently control the parasitic weed, which can be accessible to resource-poor farmers as well as environmentally friendly [5].
Seven races of S. gesnerioides have been identified based on host differential response and genetic diversity analysis within the cowpea growing regions of West Africa [16]. e races of Striga are designated as SG1 (Burkina Faso), SG2 (Mali), SG3 (Nigeria and Niger), SG4 and SG4z (Benin), SG5 (Cameroon), and SG6 (Senegal). However, the race of S. gesnerioides across the three regions of northern Ghana has not been confirmed. It is also unclear whether the S. gesnerioides in the northern Ghana is one of the known races in West Africa or a novel biotype. According to [17], race formation in cowpea-Striga association is largely a result of host-driving selection. Identification of race-specific responses in cowpea is relevant for the development of targetresistant genotypes.
Asare et al. [18] showed that only limited sources of resistance exist against S. gesnerioides in Ghanaian cowpea germplasm, and this underscores the need for more extensive analysis of breeding lines and exploitation of exotic genetic resources for cowpea improvement in Ghana. Some recombinant inbred lines (RILs) of cowpea, local cowpea accessions and exotic cowpea genotypes from the International Institute of Tropical Agriculture (IITA), exist in the Department of Molecular Biology and Biotechnology of the University of Cape Coast. However, these cowpea genotypes lack comprehensive assessment of their reactions to Striga infestation and genetic analysis which are necessary to facilitate selection and evaluation towards the release of adaptable cowpeas as varieties. e objective of the current study was to assess the reactions of 46 cowpea genotypes to S. gesnerioides infestation and to determine resistant genotypes.

Materials and Methods
A total of forty-six (46) cowpea genotypes were obtained from the Department of Molecular Biology and Biotechnology, University of Cape Coast. Twenty-eight (28) of the cowpea genotypes were recombinant inbred lines developed from a cross between IT97K-499-35 (resistant parent) and Apagbaala and SARC-LO2 (susceptible parents); fourteen (14) were cowpea genotypes from International Institute of Tropical Agriculture (IITA), Nigeria, the three parental genotypes (IT97K-499-35, Apagbaala, and SARC-LO2) as well as the local landrace, GH3684.

Pot Screening.
e pot culture screening method used by [5,17]  e potted soil medium was inoculated with Striga seeds, and four seeds of each cowpea genotype were sown per pot and replicated three times. e seedlings were thinned out at two weeks (14 days) after sowing and three plants were maintained per pot. e soil was kept moist by watering regularly every two days or when necessary. Hand picking of weeds was done when necessary. e days to emergence of Striga and the number of Striga per pot were recorded after 6 weeks. e plant-soil mass was removed from each pot and gently agitated to loosen the soil mass. e roots were washed thoroughly free of soil and examined using hand lens for the presence of necrotic hypersensitive lesions, attachment of S. gesnerioides, and tubercles. Cowpea plants that favoured attachment and emergence of S. gesnerioides were classified as susceptible and those that were intact and free from infestation, thus without any Striga attachment, were categorized as resistant.

DNA Extraction.
e genomic DNA from each plant was fixed on FTA plant card (Bioneer) and processed as described by [19,20], with slight modifications. Each isolated leaf from cowpea seedlings was cleaned with 70% ethanol and then placed over the marked circle with the underside of the leaf facing down on top of the FTA matrix card. e leaf was overlaid with parafilm, and a small porcelain pestle was used to apply moderate pounding over each sample circle area to burst the cell walls of the plant tissue. e back of the FTA card matrix was checked to observe the plant tissue extract drawn through the matrix. e FTA card was air-dried at room temperature for about one hour thirty minutes after the plant tissue extract transfer was completed. e FTA matrix card was placed on FTA sample mat and 2.0 mm Harris micro-punch tool was used to isolate several discs from the center of the dried sample area into 1.5 ml microfuge tube. Alcohol was used to wipe the punch tip after picking samples from a particular cowpea genotype before being reused to pick samples from another cowpea genotype. e leaf discs in each tube were washed with 70% ethanol for 5 minutes and repeated until the disc turned white. About 200 μl of FTA purification reagent was added to each tube, capped, inverted twice, and incubated for 5 minutes at room temperature. e FTA reagent was pipetted up and down twice to ensure that the disc remained in the tube. A pipette was used to remove and discard the FTA purification reagent. is was repeated for two FTA reagent washes. e discs were allowed to completely air dry for a minimum of one hour at room temperature and stored at 4°C until being ready for PCR amplification.

Primer Screening.
A total of 100 SSR primers were screened for polymorphism involving two cowpea parental genotypes, IT97K-499-35 and SARC-LO2, to ensure optimal performance. Optimal PCR amplification was achieved within the range of 55 to 60°C annealing temperatures. Sixteen SSR polymorphic primer pairs (Table 1) were selected and used for genetic analysis of the cowpeas.

Polymerase Chain Reaction (PCR) Analysis.
AccuPower Taq PCR Premix containing Taq DNA Polymerase, dNTPs, reaction buffer, tracking dye, and patented stabilizer ordered from Bioneer was used in this work. 10 μl of the complete PCR amplification mixture including 1 μl of primer pair was added directly to the PCR tube containing the dried FTA disc. Each of the 16 primer pairs (Table 1) was used to amplify genomic DNA of the 46 cowpea genotypes.
PCR amplification was carried out in Bio-Rad T100 ™ thermal cycler (Applied Biosystems). PCR conditions involved denaturing at 94°C for 3 minutes, annealing at 55-60°C (Table 1) for each primer pair for 30 seconds, and extension at 72°C for 30 seconds. is cycle was repeated 35 times, with a final extension at 72°C for 10 minutes, and PCR products were stored at 4°C.

Gel Electrophoresis.
e 2% agarose gel was casted in a tray (27.5 cm × 24.5 cm) with 15-well-forming comb inserted to create wells. A 40 ml agarose gel was prepared by dissolving 0.8 g of the agarose in 40 ml of ×1 TBE buffer in a microwave. e mixture was stained with 3.0 μl ethidium bromide. e mixture was then poured into the tank and distributed across the whole surface and allowed to solidify. e whole assembly was transferred into electrophoretic tank after the comb was removed; the assembly was submerged in ×1 TBE buffer. e PCR products were loaded into the wells. e lid of the electrophoresis tank was then fixed. e PCR products were resolved for 1 hr. at 120 mA and 90 V and visualized on a UV transilluminator (M-15; UVP, Upland, CA, USA). e DNA bands in the gel were photodocumented with a digital camera (Sony SELP1650, ailand). e size of DNA bands in base pairs was determined using the 100 bp DNA standard ladder (N0551S, Bioneer).

Data Analysis.
Morphological and physiological data collected from the pot experiment were subjected to the Analysis of Variance (ANOVA) using General Statistics (Genstat) analytical software (version 12.1.0.3338). Varietal means were compared using the Least Significant Difference (LSD) at 5% level of probability. e scoring and analysis of molecular data followed the format used by [21], with slight modifications. All genotypes were scored for the presence and absence of DNA bands. Only clear and repeatable polymorphic DNA bands were scored as "+" for presence of band associated with Striga-resistant and "− " for absence of band associated with Striga-susceptible. e 100 bp standard ladder was used to determine the size of polymorphic DNA bands across the cowpea genome. Data matrix was created and used to calculate the genetic distance and similarity using PowerMarker V. 3.25. e related genetic parameters including the number of polymorphic bands, alleles per locus, genetic diversity, and polymorphism information content (PIC) are generated through the similarity matrixes for cluster analysis using PowerMarker V. 3.25. e Unweighted Pair Group Method with Arithmetic Mean (UPGMA) on the similarity indices was performed to identify genetic variation patterns among cowpea genotypes, and the resulting dendrogram generated was observed in MEGA 7 software.

Phenotypic Screening.
e 46 cowpea genotypes expressed resistance or susceptible responses across the spectrum of S. gesnerioides from Walewale (northern region), Manga (upper east region), and Lawra (upper west region). In all, 65.2% (30) of the cowpea genotypes were resistant to S. gesnerioides and 34.8% (16) were susceptible (Table 2). e cowpea genotypes that expressed complete resistance to the "witchweed" in Ghana were not associated with the emergence of S. gesnerioides (Figure 1(b)), no attachment of S. gesnerioides to the roots, or necrotic hypersensitive lesions on the roots. Susceptible cowpea genotypes were characterized by an average of 15 emerged S. gesnerioides seedlings per pot (Figure 1(a)) and attachment of tubercles to the roots. e seedlings of germinated Striga emerged on the surface of the soil after 31 days (Table 3) of sowing. e Striga-infested cowpea plants expressed varied symptoms including stunted growth, leaf necrosis, chlorosis, senescence, defoliation and reduced size of young leaves, low flowering, and pod formation.

Molecular Screening.
e amplicons of C42-2B marker across the genome of 46 cowpea genotypes appeared as single DNA bands of 280 bp, associated with resistant cowpea genotypes but that of 61R-M2 marker was expressed as double DNA bands of 320 bp and 380 bp associated with resistant cowpea genotypes. Susceptible cowpea genotypes showed only single DNA band of 380 bp for 61R-M2 or no band at all (Table 3).

SSR Polymorphism.
e 16 informative SSR primers distinguished all the 46 (100%) genotypes of the cowpea including those from same parents and those with similar seed size and seed coat colour. e sizes of polymorphic amplicons (DNA bands) ranged from 120 bp for SSR-6315 to 750 bp for SSR-6375. e number of alleles detected per primer pair varied from a minimum of 2 to a maximum of 22 with an average of 5.25. e allele frequencies yielded by the 16 SSR primers ranged from 0.57 to 0.98 with an average of 0.78. Gene diversity also ranged from 0.04 to 0.49 with an average of 0.29. e PIC varied from 0.08 to 1.00 with an average of 0.55, from a total of 8.76 (Table 4). Fifty percent of the primers had PIC of 0.5 or above. ere was significant (P ≤ 0.05) correlation between the allele frequency and the gene diversity. ere was, however, no significant (P ≥ 0.05) correlation between the allele frequency and the PIC (P-value � 0.322; r � 0.265) and between the gene diversity and PIC (P-value � 0.246; r � − 0.308).

Phenotypic Analysis.
e overall resistance or susceptibility expressed by the 46 cowpea genotypes to the S. gesnerioides from Walewale (northern region), Manga (upper east region), and Lawra (upper west region) gave implication that probably the S. gesnerioides samples may be of the same race or the cowpea genotypes may have multiple resistance to the parasitic weed in Ghana. Indeed, [18] observed similar responses to S. gesnerioides from Bawku among recombinant inbred lines (RILs) of cowpea. is suggests that the Ghanaian biotypes of S. gesnerioides may have similar virulence [5]. Besides, the Striga-resistant cowpeas may possess the Striga-race-specific resistant gene  TGA AAA TTG AAC  57  GCA TGT TGC TTT GAC AAT  International Journal of Agronomy to combat the Ghanaian isolate of the parasite. In addition, the cowpeas genotypes from IITA and the local landrace GH3684 may have similar Striga-resistant trait as IT97K-499-35. e Striga-resistant cowpeas in the current study showed robust and healthy growth forms compared to the Striga-susceptible cowpeas (Figure 1). e enhanced vigorous growth of Striga-resistant cowpea genotypes may be due to adequate biomass accumulation [22] that could be translated into grain yield contrary to the Striga-susceptible cowpeas.

Molecular Analysis.
e resistance status of the cowpea genotypes to S. gesnerioides was validated with C42-2B and 61R-M2 markers. e results are consistent with those of Asare et al. [18], who used C42-2B to screen some breeding lines of cowpea. e marker segregation efficiency of 83.3% for C42-2B was better than that of 60% for 61R-M2 in identifying Striga-resistant cowpea genotypes. However, 92.6% discriminatory efficiency of SSR-1 marker was reported among cowpea breeding population [18]. e cowpea genotypes UCC-11, UCC-24ʹ, UCC-56, UCC-122, UCC-221, UCC-226, UCC-241, and UCC-328 had stable resistance to S. gesnerioides. e stability of resistance to S. gesnerioides demonstrated by the cowpeas suggests that the Striga-resistant gene might be fixed in the F 9 recombinant inbred lines. e current study observed double DNA bands of 380 bp and 320 bp for the 61R-M2 marker associated with S. gesnerioides-resistant cowpea genotypes.
is was consistent with the similar result by Omoigui et al. [23], who also observed the double DNA band characteristic of 61R-M2 marker in Striga-resistant cowpea genome of F 1 and F 2 populations from a cross between Borno Brown × IT97K-499-35 and Borno Brown × B301. Four of the cowpea genotypes (UCC-56, UCC-489, IT07K-297-13, and IT11K-321-2) had the C42-2B and 61R-M2 markers but were susceptible against S. gesnerioides in the pot analysis. However, three cowpea genotypes (UCC-473,  were resistant to the sampled S. gesnerioides but did not amplify for any of the markers (Table 2). Unlike SSR-1 primer where the marker is within the Striga race 3 (SG3) gene of B301 (1 cM), the C42-2B and 61R-M2 primers have their markers close to the resistant gene, with genetic distances of 8.5 cM and 3.5 cM, respectively [23]. e Striga-resistant gene can be separated from the marker during crossover. e marker could, therefore, be present, but the gene controlling the resistance to S. gesnerioides may be absent. is could account for the reason why some of the cowpea genotypes (UCC-56, UCC-489, IT07K-297-13, and IT11K-321-2) had the marker of both primers present but susceptible phenotypically.

Genetic Analysis.
e genetic diversity and phylogenetic relationships of cowpea genotypes from Ghana have been evaluated using SSR markers [24,25]. In this study, all the 16 SSR primer combinations used gave amplification products with 100% polymorphism. e 16 selected microsatellites (SSR) markers differentiated the cowpea genotypes and clustered them differently. ere was increasing similarity among the cowpea genotypes from IITA and the inbred lines from UCC. SARC-LO2 and Apagbaala, as well as UCC-497 and UCC-514, were the most genetically similar with a genetic distance of 0.0476. e longest genetic distance (0.500) existed between UCC-11 and UCC-24; UCC-11 and UCC-532; and UCC-32 and UCC-473. UCC-11, UCC-24, and UCC-32 are progenies from a cross between IT97K-499-35 × SARC-LO2, while UCC-473 and UCC-532 are progenies from a cross between IT97K-499-35 × Apagbaala. Most of the cowpea genotypes are also populations from the cross between IT97K- Table 2: e responses of cowpea genotypes to S. gesnerioides infestation.

Genotype
Phenotype International Journal of Agronomy 499-35 × SARC-LO2 and IT97K-499-35 × Apagbaala and therefore may share similar genetic traits. e common donor, IT97K-499-35, could influence the genetic relationship among the cowpea inbred lines in particular. e genetic diversity across the cowpea genome in the current study is low. is was in line with a similar report by Asare et al. [24], who also observed a lower level of genetic variability among Ghanaian cowpea accessions. Kuruma et al. [26] also found low level of genetic diversity among cowpea accessions in Kenya using molecular markers. e low genetic diversity among the cowpea genotypes could be as a result of common ancestral origin and self-pollination mechanism in cowpea. e number of alleles per locus of 2 to 22 with an average of 5.25 detected in the current study is in line with other recent reports. Li et al. [27] assessed the genetic similarities and relationships among 48 wild cowpea lines using SSR primers and detected between 4 and 13 alleles with an average of 7.5 alleles. Indeed, the authors in [28] observed that the number of alleles ranged from 1 to 9 per SSR primer combination in cowpea germplasm from Senegal. Sawadogo et al. [29] indicated that the 16 SSR primers used to assess the genetic diversity of cowpea cultivar in Burkina Faso generated a range of alleles between 5 and 12 fragments with an average of 8.2 bands per primer combination among cowpea genotypes. A combination of 25 S.gesnerioides    International Journal of Agronomy informative SSR primers were used to analyse Ghanaian cowpea germplasm and yielded 1 to 6 alleles per primer pair with a mean of 3.8 [24]. e assessment of 48 accessions of cultivated cowpea sampled from West Africa, Northeast Africa, Central Africa, and Southern Africa using 12 SSR markers revealed that the number of alleles per locus ranged from 2 to 5 with a total of 37 alleles generated for the primers [30]. e polymorphism information content (PIC) is a means of detecting alleles and distribution of their frequencies [31]. Data reported by [24] showed PIC ranging between 0.07 and 0.66 with a mean of 0.38. Sixteen SSR primers used to assess the genetic diversity and phylogenetic relationship among 252 cowpea genotypes in Sudan, yielding PIC between 0.33 and 0.83 with an average of 0.56 [31]. Genetic diversity studies among 32 cultivating cowpea, using 22 SSR primers, showed a PIC range from 0.25 to 0.63 with an average of 0.45 [21]. e mean PIC value (0.55) recorded in the current study is, therefore, in line with the results obtained from previous reports. e study in [30] produced a PIC valve ranging from 0.08 to 0.60 with a mean valve of 0.34 from a total of 4.47. Also, Larweh et al. [32] noted that polymorphism information content (PIC) values ranged from 0.32 to 0.36 with a mean of 0.34 in cowpea breeding lines. Ali et al. [33] reported that a mean PIC value ≥ 0.5 is highly informative, 0.25∼0.50 reasonably informative, and <0.25 slightly informative, and loci (marker) with many alleles and a PIC value near 1 are most desirable.  is means that the primers used for the current study were highly informative. Gene diversity observed in this study was 0.29 on average ranging from 0.04 to 0.49. In Senegal, cowpea gene diversity varied from 0.08 to 0.42 with a mean of 0.28 [34], whereas in Ghana cowpea germplasm gene diversity ranged from 0.12 to 0.68 with an average of 0.44 [24]. e results of gene diversity reflect the proportion of polymorphic loci across the genome. erefore, according to the result of the current study, the markers used were almost as polymorphic as those used by [24,34].

Conclusion
e responses of 46 cowpea genotypes to S. gesnerioides samples from upper east (Manga), upper west (Lawra), and northern regions (Walewale) of Ghana were similar which implies that the S. gesnerioides in Ghana may be of the same biotype or the cowpea genotypes may have the same response to different races of S. gesnerioides. On the whole, 65.2% of the cowpea genotypes were resistant to S. gesnerioides associated with C42-2B and 61R-M2 markers and 34.8% were susceptible. e segregation efficiency of 80% of C42-2B marker (single DNA band of 280 bp) was better than that of 60% of 61R-M2 (double DNA bands of 320 bp and 380 bp) for identification of Striga-resistant cowpea genotypes. e 16 SSR primers were informative and could distinguish all the 46 cowpea genotypes into three major clusters in a dendrogram. e allele frequencies yielded by the SSR primers ranged from 0.57 to 0.98 with an average of 0.78. Gene diversity also ranged from 0.04 to 0.49 with an average of 0.29. e PIC varied from 0.08 to 1.00 with an average of 0.55 with a mean genetic diversity of 0.29.

Data Availability
e data used to support the findings of this study can be accessed from the corresponding author upon request. Disclosure e authors take full responsibility for any error.

Conflicts of Interest
e authors declare that they have no conflicts of interest regarding the publication of this paper.