Screening Maize ( Zea mays ) Genotypes for Tolerance to Witchweed ( Striga asiatica L . Kuntze ) Infection

Striga asiatica (witchweed) is a parasitic weed that is a serious threat to maize (Zea mays L.) production in semi-arid environments in Africa. A pot experiment was conducted under glasshouse conditions to screen nine maize cultivars for their tolerance / tolerance to Striga infection at the University of Zimbabwe in the 2012 / 2013 rainfall season. Striga did not significantly (P > 0.05) reduce the final maize height of the maize cultivars, with the exception of three maize cultivars, namely, PHB3253, PHB30G19 and PHB30B50. Furthermore Striga did not significantly (P > 0.05) reduce the stem biomass, leaf biomass and cob biomass of all the maize cultivars evaluated. Generally the maize cultivars had significantly (P < 0.05) higher root / shoot ratios in the Striga infected plants compared to the uninfected plants. It was concluded that all the maize genotypes used in this study could possibly be tolerant to S. asiatica.


Introduction
Maize (Zea mays L.) is the third largest grain crop in the world after wheat (Triticum aestivum) and rice (Oryza sativa) (Chantereau & Nicou, 1994).It is the staple food of the 12.5 million inhabitants of Zimbabwe and is grown throughout the country including dry marginal areas of Natural Regions IV and V that receive between 450 mm to 650 mm rainfall annually.Apart from rainfall unreliability, maize production especially in the smallholder sector is affected by other biotic factors that include pests, diseases and weeds.Amongst the economically important weeds of maize is Striga asiatica (L.) Kuntze, an obligate root parasitic weed of cereals that belongs to the Orobanchaceae family.S. asiatica is the only economically important species of the Striga genus that is common in Zimbabwe (Mabasa, 2003).Striga is also a major weed of important cereal crops like sorghum (Sorghum bicolour L.) and millets (Pennisetum americanum L. and Eleusine corocana L. (Gaertn)), the major cereals grown in the savannah and Sahel regions of Africa (Lendzemo, 2004).Mabasa (1991) reported that maize is more susceptible to S. asiatica compared to pearl millet and sorghum.Yield reductions of between 10 and 80 % have been reported but can sometimes reach 100 % in susceptible maize cultivars under severe infestation resulting in field abandonments (Haussamann et al., 2000;M'boob, 1994;Odhiambo & Ransom, 1994).Technologies for the control of Striga developed so far have not been widely adopted, because there is a mismatch between technologies and the farmers' socio-economic conditions (Debrah, 1994).These range from cultural methods, chemical and genetic transformation of hosts.There has been little impact of the Striga control methods because farmers continue to grow crops which are very sensitive to Striga.It has to be noted that a small amount of Striga biomass could result in high grain yield losses in Striga sensitive cereals (Gurney et al., 1999).As a result smallholder farmers abandon these methods or simply do not invest time and money in adoption, because effective Striga control cannot be guaranteed in the current season with concomitant pay-offs in yield (Hearne, 2009).
The use of resistant / tolerant varieties appears to be the most promising and economically feasible means of combating Striga in resource poor farming systems.Striga tolerance in maize refers to the reaction of varieties that are parasitized to the same extent as susceptible ones but suffer less damage whereas resistance refers to crop varieties showing less attack, usually in terms of numbers of parasite attached or emerged (Mabasa, 2003).Although genes that confer resistance to Striga in maize have been identified in maize line ZD05, such genes have not yet been incorporated in commercial varieties (Kling et al., 2000).It has been reported that inbred ZD05 was able to demonstrate a great degree of resistance through three mechanisms; avoidance through a less branched root architecture, some ability to resist attachments of nearby germinated Striga and a kind of incompatibility that does not support normal growth of attached parasites (Rich & Ejeta, 2008).The possibility of transferring such strong resistance into cultivated maize through breeding programs to build durable resistance appears likely.In the evaluations which took into account the presence and absence of Striga, the sorghum cultivar Ochuti tolerated Striga in Kenya (Gurney et al., 1995).Similar findings were obtained in the maize cultivar Staha in Tanzania (Gurney et al., 2002), maize genotypes CG4141 and R201 (Musambasi, 1997;Mabasa, 2003) in Zimbabwe, a land race sorghum cultivar, Tiemaringfing in Mali (Ast et al., 2000), the following sorghum varieties KSV-4, NR71150 and NR71182 when 50 to 250 kg / ha Nitrogen (N) was added.
It has been reported that increasing N supply to maize results in improved host performance under Striga infection.Previous studies carried out to screen maize genotypes for tolerance to S. asiatica in Zimbabwe were carried out under high N (Chivinge et al., 1995).Such high N applications rates are out of the reach of the majority of farmers and Striga management practices that involve an increase in the amount of N applied are likely not to receive widespread adoption among resource poor farmers.Mabasa (2003) reported that the majority of farmers in communal areas apply 30 kg or less N. Therefore varieties that perform better under low N conditions are likely to receive wide spread uptake by farmers.Currently the research focus of maize breeding in Zimbabwe has been to develop varieties that do well under low moisture and low nitrogen conditions.It has been reported that there are similarities between the physiological effects, specifically changes in ABA levels in plants caused by drought on cereals and those caused by Striga.Varieties with tolerance to drought and low N conditions are likely to suffer less damage from Striga infection than varieties susceptible to drought.Such varieties have been demonstrated to produce good yields especially under the basin planting.Planting basins are permanent holes (15cm x 15 cm x 15 cm) in which crop seeds are dry planted for the purposes of moisture conservation to ensure that the crop germinates with the early rains.Apart from increased moisture conservation, there are reduced chances of Nitrogen loss because all the fertilisers are spot applied in the planting basin where the risks of fertiliser losses due to sheet erosion are low.The drought and low nitrogen tolerant varieties developed so far have shown potential to produce good yields in marginal areas but their tolerance to Striga has not been studied.Such vital information would be very useful in enabling farmers to select varieties that can withstand both abiotic stress and Striga infection.Therefore there could be merit in screening land races and recently developed maize genotypes with the hope of identifying some that are tolerant to Striga infection.The objective of the study was to screen maize genotypes for tolerance / resistance to S. asiatica infection under N levels that mimic typical smallholder farmer conditions in Zimbabwe.

Study Site
A greenhouse pot experiment was conducted at the University of Zimbabwe (17.78 °S, 31.05°E, and 1523 meters above sea level), Crop Science department during the 2012 / 2013 rainy season.The experiment was laid out as a Randomised Complete Block Design (RCBD) with three blocks.Blocking was done according to the position of the pots from the window.The treatment structure was a 9 x 2 factorial.The first factor was maize genotypes (nine maize genotypes) and the second factor was the Striga level (infested and uninfested maize genotypes).Maize genotypes that were used are shown in Table 1.

Experimental Procedure
Fifty four black polythene bags, each measuring 180 x 140 x 320 mm were filled with sandy soil up to three quarters.The top five centimetres of soil in 27 of the polythene bags was thoroughly mixed with 0.04 g (approximately 9800 seeds) of S. asiatica seeds and the other 27 polythene bags remained uninfested and were used as the control.The Striga seed used had been collected from Striga plants associated with maize in Rushinga smallholder farming area in May 2012 in Zimbabwe.The Striga seed was not allowed to precondition prior to the introduction of the maize seed (Parkinson, 1985).Three seeds of maize were planted in each polythene bag at the depth of 5 centimetres on the 14 th of November 2012 and the plants emerged on the 20 th of November 2012.At planting, 8 g of compound D (7 % N: 14 % P 2 O 5 : 7 % K 2 O) was banded below the maize seed in order to promote early root development.Thinning of maize seedlings was done one week after crop emergence (WACE) to leave one plant per polythene bag.Weeds other than S. asiatica that had emerged were hand pulled after every three days and the maize plants were irrigated using a watering can fitted with a fine rose every other day.Top dressing of maize plants was done using Ammonium Nitrate (AN) (34.5 % N) at 6 WACE at the rate of 30 kg per hectare (one gram of AN was added applied per polythene bag).Harvesting was done on the 20 th of February 2013 by cutting maize stems at ground level using a pair of secateurs and the roots were carefully washed with water to remove the soil.The leaves, stems, cobs and roots from each polythene were placed in separate envelopes.
The number of Striga plants that had emerged was recorded at weekly intervals starting from 7 WACE up to harvesting of maize.The roots of maize were carefully washed and the number of Striga attachments was determined by physically counting the Striga attachments on the maize roots at harvesting.Maize leaf, stem, cob and root biomass was determined at physiological maturity by oven drying all the leaves, stems, cobs and roots separately at 80°C for 72 hours.The maize height measurements were taken from the ground level to the growing tip using a 30 centimeter ruler weekly starting at 40 days after planting (DAP).

Data Analysis
Striga and maize data were subjected to Analysis of Variance (ANOVA) using Genstat version 14, Minitab version 16.A repeated measure ANOVA was performed to test the effects on maize cultivar, Striga infection, maize height measuring time (40, 54, 61, 68, 78 and 82 days after planting) and their interactions on maize heights.Mean separation was done using Fischer's protected Least Significance Difference (LSD) at 5 % probability level.

Striga Emergence
Striga emergence was observed at 7 WACE in one pot of LANDRACE 1, one pot of R201 and in all the pots for AG541 only but did not emerge in the other polythene bags.

Maize Leaf Biomass
The interaction of maize genotypes and Striga effects were not significant on leaf biomass (P > 0.05).Striga infestation had no impact on leaf biomass across the maize genotypes (Figure 3).

Maize Stem Biomass
The interaction of maize genotype x Striga effects was not significant (P > 0.05) on stem biomass (Figure 4).However, there were significant (P < 0.05) differences in stem biomass amongst the nine maize genotypes.P2859W and AG 541 had significantly lower stem biomass than PHB 3253, R201 and Landrace 1, but their stem biomass was not significantly different from the other genotypes.
Figure 5.Effect of maize genotypes on maize cob biomass

Maize Root Biomass
The maize genotype x Striga infestation interaction was not significant (P > 0.05) on root biomass (Figure 6).Maize root biomass was not significantly different (P > 0.05) amongst the maize genotypes.Striga infestation significantly influenced root biomass (P = 0.006).The Striga infested maize had significantly higher root biomass as compared to the uninfested maize.

Root / Shoot Ratio
The maize genotype x Striga infestation interaction was significant (P = 0.018) on root to shoot ratio (Figure 7).Root to shoot ratio was significantly higher in infested than uninfested maize in PHB 30G19, PHB 30B50, SIRDA MAIZE, R201, AG 541 and Landrace 1.However, Striga infestation did not cause a significant effect on root to shoot ratio in PHB30D79, PHB 2859W and PHB 3253.

Discussion
Only few Striga plants emerged at 35 days after planting (seven weeks after crop emergence (WACE)) but they did not produce flowers.The parasite did not emerge in most of the Striga infested pots although it managed to attach to the roots of all the maize genotypes used in this study.The delay in emergence can be explained by a delay in the onset of attachment (Ast, 2006) due to the fact that the seed of Striga was not preconditioned when maize seeds were planted in infested polythene bags.These findings imply that host avoidance of the parasite is likely to be an important Striga management strategy, especially in the arid parts of Zimbabwe where some farmers practice early planting in planting basins.Maize planted in planting basins germinates early and is likely to grow for a few weeks before the Striga seed is preconditioned resulting in delayed onset of parasite attachment to the host.
This study also showed that all the genotypes used equally supported few attachments of the parasite to the host.This can be partly explained by the delay in the onset of attachments but the low numbers of attachments observed could be an indication of similarly low stimulant production capacity of the genotypes.It is now known that the production of strigolactones, a group of Striga germination stimulants produced by the host plants are responsible for causing germination of Striga seeds.It is also well documented that the production of strigolactones among host genotypes could account for differences in numbers of Striga attachments among genotypes (Jamil et al., 2011).The capacity of these genotypes to produce stimulants has not yet been studied.There is therefore need for further research to find out whether the lack of differences in attachments on the genotypes was due to similarities in their onset and capacity of stimulant production.
The maize height data over time showed that the maize genotypes differed in their sensitivity to Striga infection.The most sensitive genotypes to Striga infection were PHB3253 and PHB30G19.In these genotypes Striga significantly reduced maize heights compared to uninfested maize.This suggests that these genotypes are susceptible to Striga infection.Striga infection did not affect the height and stem biomass of the genotypes AG541, R201, SIRDA MAIZE, LANDRACE 1, PHB30D79 and P2859W implying that these genotypes are tolerant to S. asiatica infection.Maize stem biomass and height are the most sensitive parameters to Striga infection (Gurney et al., 2003;Mabasa, 2003).Therefore, genotypes whose stem biomass and height were not affected by striga infection could be tolerant to Striga infection (Gurney et al., 1999;Cechin & Press, 1993).However, tolerance of the genotypes to Striga infestation can not only be determined by the low sensitivity of these genotypes to Striga infection in terms of stem biomass and heights.As mentioned earlier on, this sensitivity could be due to late attachment of the parasite to the roots of the host.Striga seeds could have germinated and attached at a later stage on roots developing superficially, when the host had already developed capacity to tolerate the effect of Striga infection (Gurney et al., 1999).These findings are supported by the work of Cechin and Press (1993) who observed that severe damage to host under controlled conditions can only be observed when S. asiatica attachment to host occurs early in the life cycle of the host.However, the reduced effect of Striga on the host cannot be wholly attributed to avoidance due to late attachment but could also indicate the presence of a certain level of tolerance especially given that the height of some of the genotype were reduced by the late Striga attachments.The low sensitivity of the maize genotypes evaluated in this study to Striga infection in terms of stem biomass and stem heights could also be explained by the fact that most of the genotypes used in this study are drought tolerant.The maize genotypes did not express the characteristic wilting symptoms which are associated with the presence of Striga infection.Striga infection is known to cause accumulation of abscisic acid (ABA) in maize genotypes (Taylor et al., 1996) leading to reduced rates of photosynthesis in some maize genotypes.High ABA levels in the plants are responsible for causing stomatal closure and this will limit the entry of carbon dioxide into the leaf resulting in a reduction in the photosynthetic rate of the plant.Maize genotypes which are drought tolerant are known to accumulate less ABA in their leaves after receiving the signal from the roots (Taylor et al., 1996) and their photosynthetic rate is therefore not reduced.
The root biomass of Striga infested maize genotypes was significantly higher than the uninfested maize genotypes and hence resulted in higher root / shoot ratio.The response of root / shoot ratio in two cultivars, R201 and LANDRACE 1 was greater than in the other genotypes, which means that Striga elicited a greater shift in favour of root growth in these two cultivars.It is well documented that Striga infection affects the host in many ways.Firstly there is an increase in host root biomass at the expense of shoot biomass due to competition for photoassimilates, amino acids and water because the parasite will be acting as an additional sink.Secondly, it is reported that Striga infection also causes hormonal imbalances in Striga infected hosts characterised by increased levels of abscisic acid and decreased levels of cytokinins and gibberellins (Ast, 2006).Parker and Riches (1993) reported that these changes in hormonal balance result in the stimulation of root growth and reduced shoot growth in Striga infected hosts.Thirdly the increase in root to shoot ratio in infected hosts could be explained by the fact that most of the Striga remained below the ground and a few managed to emerge above ground and this could have imposed a heavy burden on the maize plants in terms of carbohydrate requirements since Striga plants have high respiration rates (Cechin & Press, 1993).On the contrary emerged Striga plants contain chlorophyll and are able to process part of their carbon through photosynthesis.The impact of Striga on the maize genotypes was not significant in terms of dry matter production.This probably confirms the fact that these maize genotypes are indeed Striga tolerant.

Conclusion and Recommendations
It was observed in this study that the maize genotypes LANDRACE 1, SIRDA MAIZE and R201 are tolerant to S. asiatica infection and have a potential to be grown in Striga endemic areas where most of the resource poor farmers are located.The data on stem heights showed that all the maize genotypes with the exception of PHB3253, PHB30B50 and PHB30G19, were not sensitive to Striga infection.This confirms the hypothesis that the maize genotypes have different sensitivities to Striga infection.The maize genotypes which are not sensitive to Striga infection could be tolerant to Striga and those that are sensitive are Striga susceptible.The genotypes, PHB30D79, P2859W and AG541can also be grown in Striga endemic areas since they indicated some level of tolerance.However, there is a need to evaluate these maize genotypes under field conditions in order to confirm their tolerance to S. asiatica infection.There could also be merit in studying the effects of different strains of Striga asiatica on these maize genotypes.

Figure 1 .
Figure 1.Number of Striga attachments in a pot experiment carried out at the University of Zimbabwe Maize genotypes Number of Striga attachments 0

Figure 3 .
Figure 3.Effect of Striga infestation on leaf biomass of maize genotypes

Figure 4 .
Figure 4. Effect of maize genotypes on maize stem biomass

Figure 6 .
Figure 6.Effect of Striga infestation on maize root biomass

Figure 7 .
Figure 7. Effect of Striga infestation x maize genotype on root/shoot ratio of Striga infested and uninfested maize

Table 1 .
Characteristics of maize genotypes under study Sirda Maize Medium maturity, drought tolerant, resistant to MSV and GSV.AG541 Short seasonLandrace 1 Broad baseR201Early maturity, heat stress tolerance moderate resistance to cob rot and Late blight.