Cowpea Vigna unguiculata (L.) Walp.)  is an important crop  in  the  semi-arid  tropics  including  parts   of   Asia, Africa, Southern Europe, Southern United States, Central and South America (Singh, 2005; Timko  et al.,  2007a). It is an affordable source of quality protein for rural and urban dwellers in Africa (Ajeigbe et al., 2012; Dube and Fanadzo, 2013). The dry grain protein concentration oscillates from 21 to 33% (Abudulai et al., 2016).  It is well adapted to hectic environments where several crops fail to grow well (Bisikwa et al., 2014; Ddamulira et al., 2015). According to FAO, cowpea was cultivated on about 12.08 million hectare in Africa in 2016 with a total production of 5.83 million hectare in West Africa, predominantly in Nigeria, Niger, Burkina Faso, Mali and Senegal (FAOSTAT, 2018). Currently, cowpea yields are estimated around 300 to 500 kg ha^-1 on farmer’s field in Sub Saharan Africa (SSA) while its yield potential isup to 3000 kg ha^-1 in optimum growing conditions (Tanzubil et al., 2008).

Cowpea production is mostly affected by major constraints. Parasitic plants are a major constraint to today’s agriculture with most crop species being potential hosts (Westwood et al., 2010).  Striga gesnerioides, is a key threat to cowpea production throughout West and Central Africa (Omoigui et al., 2017).  It is one of the greatest devastating parasitic weed in most parts of the world. It is an obligate root parasitic flowering weed that belongs to the Orobanchaceae family (Parker, 2012). Out of about 30 Striga species which have been identified, Striga gesnerioides is the only Striga species that is virulent to dicots (Mohamed and Musselman, 2008). Striga gesnerioides is a major limitation to cowpea production in Africa (Timko et al., 2007b), causing considerable yield losses (Aggarwal and Ouédraogo, 1989).  The extent of the damage in cowpea due to S. gesnerioides could be up to 70% depending on the extent of damage and level of infestation (Alonge et al., 2004). On susceptible cultivars, yield losses can reach up to 100% when S. gesnerioides population is over 10 emerged shoots per plant (Kamara et al., 2008). Omoigui et al. (2009) reported that yield losses caused by Striga in dry savannah of SSA are estimated in millions of tons annually and the prevalence of Striga infested soils is steadily increasing. Methods including improved cultural practices and the use of chemicals to control S. gesnerioides are available but most of them are ineffective whilst others are not affordable for small-scale farmers of Sub Saharan Africa (Singh et al., 1997; Timko et al., 2007b). The long viability of the seeds and the subterranean nature of the initial stages of parasitism make the control of the parasite by conventional approaches challenging (Berhane, 2016). In general, S. gesnerioides control is difficult to achieve due to the close association with its host (Lane et al., 1997). The identification of sources of S. gesnerioides resistance and their incorporation into breeding schemes would be a useful approach to combat the damage caused by the parasite in cowpea fields (Omoigui et al., 2017).

The objective of this study was to evaluate the field performance of 251 recombinant inbred lines (RILs) under S. gesnerioides infestation in Northern Ghana  and to assess yield loss due to Striga infestation.

 

The experiments were conducted from July 2015 to April 2016 at the Manga Station of Council for Scientific and Industrial Research-Savannah Agricultural Research Institute (CSIR-SARI). Manga near Bawku in the Upper East Region is geographically located within latitude 11.02° and longitude 0.27°, with an altitude of 224 m above sea level. The area is situated in the Sudan Savanna agro-ecological zone of Ghana. The mean rainfall of the area during the period of the experiment was approximately 44.33 mm. The average annual temperature was about 29.44°C, the highest being observed from March to April 2016. The relative humidity (RH) of the location fluctuated significantly, dropping in the dry season and rising during the rainy season with an average humidity of 55.4%. The study was conducted in two stages. The first stage was carried out in the field and the second stage in pots experiment.

Planting materials

Two hundred and fifty one recombinants inbred lines (RILs) at F₈ generation (F₈) (Table 1) derived from a cross between two cowpea lines, ‘Sanzi’ (susceptible to Striga) and ‘IT97K-499-35’ (resistant to Striga) (Omoigui et al., 2009), were used in the study.

 

 

Field experiment

The field study was carried out under rain fed conditions (between July and September) and under irrigation during the dry season. In a preliminary screening, each of the RILs and the parents (Sanzi and IT97K-499-35) were planted in a 2-meter single row plot without any replication on a field known to be a Striga hot spot.  During this preliminary screening, data collected included days to 50% flowering, presence or absence of Striga plants, number of Striga per plants, total number of Striga per plot and Striga height. The presence or absence of Striga was recorded by visual observation on the different plots from thirty five (35) days after planting (DAP) up to maturity.

Pot experiment

Pot experiment was carried out to confirm the resistance or otherwise of the sixty-nine (69) the RILs that were identified as Striga resistant in the field experiment (Table 2). The pots were filled with top soil and then artificially infested with 5 g of Striga seeds. The top most, (1/3) portion of the soil per pot was mixed with the 5 g of Striga seeds. The infested pots were watered and allowed to condition before planting of the cowpea seeds. The pots were arranged in a randomized complete block designs with three replications. Thirty five days after planting (DAP), the pots were monitored on daily basis to check for Striga emergence. At maturity, the early pods were harvested on single plant basis to get some seeds from each plant. This was followed by gently washing the soil off the roots of the plants to confirm or otherwise that there were no Striga attachment to the roots of those that did not record Striga emergence.

 

 

Yield loss assessment due to Striga infestation

Twelve RIL’s were selected based on their good agronomic traits on the field  (white  seed  coat,  big  size and early maturity). The 12 consisted of five Striga resistant lines, five Striga susceptible lines and the two parents (IT97K-499-35 and Sanzi) as checks (Table 3).

 

 

The experiment was designed as a split plot with four replications. The Striga treatments (infested and no infestation) were the main plots while the 12 lines were the sub plots. The soil used to fill the pots were steam sterilised at 100°C to get rid of all Striga seeds. A metallic barrel was used for the sterilization of the soil. A wire mesh was fitted at 1/3 of the length of the barrel from the bottom. This served as a separator between the soil and the water. The setup was placed on fire. Water was poured in the barrel to fill up to the level where the wire mesh is fitted, jute sack was then laid over the wire mesh before filling the remaining two thirds with soil. The soil was covered with jute sack. The steam generated from the boiling water was allowed to pass through the soil for an hour and half to heat up the soil up to 100°C. The fire was put off upon attaining the 100°C to  allow  the  soil  to  cool  down. The  soil was then scooped and spread on a plastic sheet to allow it to further cool down under shade before filling the plastic pots.

Forty-eight pots were infested with 5 g with S. gesnerioides. The other forty-eight pots were not infested with Striga seeds. All the pots were watered to field capacity and allowed to drain for 24 h before planting. The pots were irrigated as when it is needed and kept weed-free through hand pulling. Monitored spray was done against insects. From thirty-five days after planting, Striga emergence was recorded on daily based on visual observation. The other agronomic data collected included first day of flowering, days to 50% flowering, plant height, number of peduncles per plant and days to maturity.

The post-harvest data collected included dry pod weight, grain weight, number of seeds per pod, hundred seed weight as well as fresh and dried biomass weight. The dried biomass was obtained after drying in an oven for 24 h to a constant weight.  

Yield loss assessment due to Striga infestation was estimated using the formula:

YL: Yield losses.

Statistical analysis

All field data collected were subjected to analysis of variance (ANOVA) using the GenStat analytical software (version 12.1.0.3338). Varietal means were compared using least significant difference at 5% level of probability (LSD 5%).

 

 

 

 

Cowpea RILs reaction to natural Striga gesnerioides in the field screening

The result showed that sixty-six (66) RILs out of the 251 (26.29%) used for this trial were resistant. (Table 4). Striga plants emerged from the soil of the plots containing susceptible RILS.

 

 

Reaction of cowpea RILs to artificial Striga gesnerioides in pot experiments

The results of artificial inoculation showed that 27 RILs were found to be resistant (no Striga emergence or Striga attachment) whiles 39 were susceptible (having Striga emergence or Striga attachment at the roots level) (Table 5). The number of days to flowering and maturity varied from 35 to 55 and 60 to 86, respectively.

 

 

Evaluation of Striga promising lines in yield loss

Genotypes varied significantly in terms of days to 50% flowering and maturity under both infested and not infested (Table 6). Days to flowering and maturity varied from 41 to 55 and 63 to 73 days after planting. Under Striga infestation, the genotype 191 was the earliest to flower at 44 days after planting (Table 6). Under no Striga infestation, the genotypes 25 and 112A flowered earlier than the rest of the genotypes (41 and 43 days). 155A2 flowered 50 days after planting. The remaining genotypes were considered as medium maturity cultivars based on the days to flowering (43 to 49 days).

 

 

Under Striga infestation, all the resistant lines significantly (P<0.001) flowered and matured  almost at the same time as in no Striga infested pots whiles the susceptible lines delayed in flowering and maturity (Table 6).

The resistant genotype 19B for instance flowered at 48 DAPS and matured at 65 and 66 days DAP in the non-infested and Striga infested pots, respectively.

The susceptible genotype 12B flowered at 48days after planting under no Striga infestation and 55 DAP under Striga infestation.  The days to maturity were 65 and 73 days non-infested and Striga infested pots respectively (PË‚ 0.001).

Seed yield and dry biomass per hectare

The analysis of variance revealed significant differences between the progenies under Striga infestation and no Striga infestation (Table 9)

Among the RILs, 16A, under no infestation produced the highest grain yield (754.2 kg ha^-1) followed by 25 (473.3 kg ha^-1) and 19B (470.1 kg ha^-1) (Table 7). The genotype, 155A2, a resistant cultivar recorded the lowest grain yield (320.1 kg ha^-1). The cultivar 16A which recorded the highest yield under no infestation (754.2 kg ha^-1) also recorded the significant yield under Striga infestation. (750 kg ha^-1). Ironically, the susceptible cultivar 25, one of the highest grain producers under no Striga infestation (436.1 kg ha^-1) also had one of the lowest grain yield under the infestation (338.6 kg ha^-1). In general, the reduction in grain yield was higher in the susceptible progenies than the resistant ones.

 

 

Dry biomass yield showed significant differences among the Striga infested (P Ë‚ 0.001) and non-infested conditions. The mean values of dry fodder yield were 1507 kg ha^-1 under no Striga condition and 1126 kg ha^-1 in the infested conditions. The progenies with the highest dry biomass under no infestation conditions were 155A2 and 12 B with 2234 and 1901 kg ha^-1, respectively. The lowest fodder yield was recorded for the cultivar 25 with yield of 876 kg ha^-1. The dry biomass yields for the other genotypes ranged from 1076 to 1812 kg/ha.

Under the Striga infestation condition, the genotype 155A2 still recorded the highest fodder likewise in the no infestation. The dry fodder yield of genotype 12B drastically dropped from 1901 kg ha^-1 in the non-infested condition to 1002 kg ha^-1 under the infested condition. The genotype 16A also recorded good production of fodder in both infested (1813 kg ha^-1) and no infested condition (1898 kg ha^-1).

Plant height and number of pods per plant

The number of pod per plant was significantly different in both infested (P Ë‚ 0.001) and non-infested environment. The analysis of variance indicated a significant difference for the plant height in both infested and non-infested environments.

The resistant plants were taller than the susceptible RILs under Striga infested condition (Table 8). The resistant parent IT97k-499-35 was the tallest plant (35.61 cm) followed by 16A and then 191 with plant heights of 31.63 and 31.53 cm, respectively. The susceptible cultivars 12B recorded the shortest plants height with 15.74 cm, in the no Striga infested pots. Striga susceptible genotypes were shorter compared to the resistant RILs.

 

 

Among the progenies, the highest mean number of pods per plant (10 pods) was recorded in the susceptible genotype, 22A, under no Striga infestation, but produced 7 pods under Striga infested condition. However, for the resistant RIL 19B, the mean number of pods was not affected when grown on Striga infested soils.

Grain yield and dry biomass loss due to Striga gesnerioides

In general the grain and biomass yield loss were higher in the susceptible lines compared to the resistant RILs. For the resistant RILs  the dry grain yield losses ranged from 4.5 kg ha^-1 (0.55%) to 14.5 kg ha ^-1 (3.08%) (Table 9). In the susceptible genotypes, grain yield losses oscillated from 134.7 kg ha^-1 (28.45%) to 262.5 kg ha^-1 (58.88%).

 

 

The highest grain yield loss (58.88%) was recorded for the susceptible RIL 12B followed by the susceptible line 22A which registered 37.9% grain yield loss. Grain yield losses for the resistant progenies were found to be between 0.55% for the cultivar 16 A to 3.08% for the genotype 19B. The resistant line 16A also showed a lower yield loss (0.55%) than the resistant parent (1.51%).

Dry biomass losses for the susceptible progenies ranged from 889 kg ha^-1 (47.29%) to 664 kg ha^-1 (61.71%) for the cultivars 12B and 112A, respectively.

Similarly  for  the  dry  grain yield, the resistant RILs did not show any significant biomass losses. With regard to the biomass losses, the cultivar 155A2 performed better in both Striga infested (2209 kg ha^-1) and non-infested (2234 kg ha^-1) then to the resistant parent IT97K-499-35, and also recorded the least biomass loss (1.1%) (Table 9).

 

 

 

 

 

 

 

 

Field screening for cowpea genotypes resistant to Striga gesnerioides

The field study recorded high emergence of Striga gesnerioides per plot (243 shoots) of the susceptible lines and this is similar to observations in other studies (Carsky et al., 2003; Kamara et al., 2008). The high Striga emergence observed on the Striga Susceptible lines was an indication that the site was really a hot spot for S. gesnerioides and the field had been infested with high concentration of Striga seeds over the years. However, a rigorous screening of the 66 genotypes in artificially infested soils in pot experiments revealed that only 10.75% were truly resistant to Striga. A susceptible genotype could be heavily infested underground without any Striga emergence as a results of several factors. According to Kim et al. (2002), one of the major limitations of  screening  under  natural  infestation  is  the variability in Striga seeds dissemination and cultivars escaping infestation. Striga sp. seeds need warm stratification for a certain time at a right temperature (approximately 30°C) before the seeds start responding to germination stimulants (Matusova et al., 2004). The high interference such as soil and climatic factors observed in the field makes the field screening less accurate (Baptiste et al., 2013).

Pot screenings

Field screening under artificial infestation is not always practical due to the fact that it can cause Striga seeds spreading to novel regions and it is moreover not consistent because breeders do not have any control of the parasite density and distribution (Haussmann et al., 2000). Pot screening has been operative as an alternative technique to confirm uniform infestation of Striga seeds.

After the pot experiment, the number of resistant lines was reduced from sixty-six (66) RILs to twenty seven (27) RILs after the pots experiment. This is essentially due to the high level of infestation (five grams of Striga seed per pot), the uniformity and a better control of the environment. A previous study done by Baptiste et al. (2013), confirmed the reliability of the pot screening compared to field screening.

The increased number of susceptible recombinant inbred lines found among the 66 could also be implied that these genotypes though showed no emerged seedlings of Striga had Striga attached to their roots. According to Ba (1983), some cowpea genotypes stimulate the Striga to germinate and penetrate their root tissues, but the Striga fails to grow more.

After both field and pot screening for Striga resistance, and taking into consideration farmers preferred traits, the genotypes 16A, 19B, 35, 155A2 and 191 were identified as promising Striga resistant lines.

Striga infestation delayed the flowering and maturity of susceptible cowpea genotypes. The susceptible genotypes also experienced huge reduction in grain yield and dry biomass in the Striga infested environment compared to when they were grown under no Striga environment. The study also confirmed that Striga infestation induces stunted growth hence the significant reduction of plant height at 50% flowering recorded for the susceptible genotypes. It also had an effect on the production of number of pods per plant. These data corroborated with previous studies (Press, 1995; Alonge, 1999; Gworgwor et al., 1991), which produced similar results. The stunted growth of genotypes, 12B, 22A, 25, 112A and 211A, could be attributed to the competition between the host and Striga for resources. The reduced vegetative growth of the susceptible varieties resulted in reduced leaf area, photosynthetic capacity and therefore affected flowering, podding and seed production (Alonge, 1999). According to Press (1995), the lower biomass accumulation by the susceptible genotypes could be the result of competition among the host and the weed for solutes, as well as carbon, water and minor rate of photosynthesis in the leaves of Striga infested plant. The reduced photosynthesis might have resulted in lower number of pods per plant and translocation of photosynthates to the sink.

Graves et al. (1992), showed that the low chlorophyll content which characterizes susceptible genotypes may account for the reduced development of the susceptible cowpea genotypes causing a decrease in both grain and biomass yield. The low biomass yield could also be attributed to the reduced shoot growth of the susceptible genotypes. The same phenomenon has also been reported for cereals infected with Striga hermonthica and for some cowpea genotypes infected with S. gesnerioides (Graves et al., 1992).

The resistant cultivars showed a relatively good growth compared to the susceptible lines in the infested pots. The relative good growth and the reduced export of assimilate to the weed would have ensured sufficient biomass accumulation and seed development as suggested by Gworgwor et al. (1991) on S. gesnerioides. The superior growth of some genotypes like 16A, 19B, 35, 155A2 and 191 indicated the positive relationship between crop vigour and crop performance even in Striga infested pots.

Grain and biomass loss due to S. gesnerioides

This current study has shown that all the resistant cowpea cultivars (16A, 19B, 35, 155A2 and 191) exhibited lower grain and dry biomass loss compared to the susceptible ones (12B, 22A, 25, 112A and 211A) indicating that these cultivars could play an essential role in controlling Striga in the endemic areas.

The susceptible genotypes recorded an average yield loss of 37.66% for dry grain yield which is quite consistent with the yield loss of 31±4% with a range of 26 to 65% observed by Aggarwal and Ouedraogo (1989). According to these authors, the loss could be attributed exclusively to the genotype effect as a consequence of Striga direct parasitism of susceptible cowpea lines (Muleba et al., 1996). S. gesnerioides diverts the host nutrient into themselves via the haustorium which establishes contact with the host tissues (xylem and phloem) (Okonkwo and Nwoke 1978; Okwonkwo, 1966). Consequently, this competition among host and parasite for water, and essential metabolites could be the explanation for the yield loss (Stewart and Press, 1990).  Setty and Nanjapp (1985) and Kuijt (1969), reported that the osmotic pressure of the parasite is higher in both leaf and root than its host making the Striga more competitive. The use of high yielding Striga resistant varieties coupled with good agronomic practices can therefore help to reduce the yield losses in soil infested with Striga in the traditional farming systems.

 

The study revealed different reactions of cowpea RILs to S. gesnerioides during the field and the pot experiments. Out of the 251 RILs used, 27 RILS were found resistant similar to the resistant parent (IT97K-499-35), whiles 224 RILs were susceptible.

Yield loss assessment showed that the Striga resistant genotypes suffered less yield loss compared to the susceptible ones and therefore resistant genotypes can be one of the best means to minimize yield loss. These genotypes that expressed complete resistance are potential lines that will serve as resistant genotypes. The latest discovery of new sources of resistance to Striga provides an excellent way to supply farmers with new genotypes to replace their susceptible varieties.

 

The authors have not declared any conflict of interests.