Resistance breeding and biocontrol of Striga asiatica (L.) Kuntze in maize: a review

ABSTRACT

Purpose: The aims of this article are to highlight pre-breeding procedures for identifying primary sources of Striga-resistance genes and to summarize complimentary breeding techniques that enhance partial resistance of maize varieties against Striga species.

Materials and methods: The paper presented a comprehensive account of Striga screening and controlling techniques and highlighted the potential of integrating partial resistance with FOS to boost maize production and productivity in SSA.

Results: Striga infestation is a major constraint to maize production and productivity in Sub-Saharan Africa (SSA). A lack of Striga-resistant maize varieties and the limited adoption of other control methods hinder effective and integrated control of the parasitic weed in maize and related cereal crops globally. Genetic resistance of maize should be complemented with the use of Fusarium oxysporum f.sp. strigea (FOS), a biocontrol agent known to suppress Striga.

Conclusions: A combined use of genetic resistance and FOS has remained largely unutilized in controlling Striga in Africa. A combination of conventional and molecular Striga-resistance breeding tools as well as the use of FOS are promising methods to effectively control Striga in SSA.

KEYWORDS:

• Biocontrol
• F. oxysporum f.sp. strigea
• screening techniques
• Striga asiatica (L.) Kuntze
• Zea mays. L

Introduction

Maize (Zea mays L., 2n = 2x = 20) is an important crop globally (Ranum et al. 2014). Maize production and productivity in the developing world is low when compared to global average. This is attributed to unavailability of improved genetic resources, poor crop management practices, as well as various biotic and abiotic stresses. Striga asiatica (L.) Kuntze is among major biotic constraints to maize production. Approximately 30 Striga species within the family Scrophulariaceae that parasitize grass species such as maize (Zea mays L.), sorghum (sorghum bicolor (L.) Moench), rice (Oryza sativa) and pearl millet (Pennisetum glaucum (L.) R.Br) have been described (Musselman 1980). Three of these species, namely S. hermonthica, S. asiatica and S. aspera cause heavy yield losses among cereal crops across sub-Saharan Africa (SSA) (Ejeta and Gressel 2007). Out of the three species, S. hermonthica causes the most damage because of its robustness and widespread occurrence in most parts of the SSA region, except in Southern Africa. On the other hand, S. asiatica is predominant and causing severe productivity losses in southern Africa. Vulnerability of maize in this region to S. asiatica is increased by poor nutrient status of communal lands, coupled with frequent droughts.

The parasitic weed extract most of its nutrients from the host crop through the root tissue leading to stunted growth, wilting, chlorotic appearance and poor yields. At maturity, Striga plants releases over 90000 min seeds per plant that are easily dispersed by several agents including man, wind and erosion (Musselman 1980; Ejeta and Gressel 2007). Seeds remain dormant in soils for more than 14 years until favorable conditions are met to triggering germination (Ejeta et al. 2000). This make it extremely difficult to eradicate the weed from soil seed banks of agricultural lands. Adequate moisture and soil temperature would induce germination of Striga seeds. Also, principal chemical agents released by host plants, Strigolactones, trigger Striga seed germination (Ejeta et al. 2000). Seed germination is a critical stage marking the commencement of the deleterious host-parasite relationship. Striga seed germination is followed by haustorial initiation, leading to subsequent attachment of the parasite to susceptible host roots. Striga species are holo-parasites that rely exclusively on host metabolites, from which they extract photo-assimilates in exchange of phytotoxic compounds that cause leaf chlorosis and stunted growth. The extent of damage depends on the timing and level of infection after seed germination. Yield losses of up to 100% have been reported in susceptible cereal cultivars under high infestation levels, particularly under drought and low fertility soils (Amusan et al. 2008).

Effective Striga control among small-scale communal farmers using either cultural or chemical control is currently limited by inadequate resources, knowledge and skills. Early efforts to breed for Striga resistance in maize was limited by lack of sources of resistance to start with. There is, therefore, a need to devise strategies that integrate various breeding procedures and control options for improving maize resistance to Striga to boost productivity. In parallel, long-term research on the biocontrol of Striga species has been undertaken at several research centers in the northern hemisphere (Zahran et al. 2008). To cope up with the current trends in eco-friendly farming practices, research has been directed at perfecting the effectiveness of inundative biological control. Mycoherbicides form a contemporary weed control option due to host specific allelochemicals produced by fungi that are phytotoxic to weeds species. Such chemicals as fusaric acid and fumonisin are immune to resistance breakdown since they are multiple site inhibitors (Gressel 2010). In the developed world, several mycoherbicides have been patented and are being sold commercially for weed management in common row crops. However, large-scale implementation of biocontrol in sub-Saharan Africa has been of little success. This review explores the potential of various pre-breeding procedures to enhance partial resistance of maize varieties against Striga species. The potential to integrate partial resistance with a well-characterised biocontrol agent of S. asiatica, FOS, is also discussed.

Host resistance to Striga infestation in maize

Development of resistance or tolerance to S. asiatica in locally adapted maize varieties will have a major impact on the productivity of maize. Striga resistant maize varieties form the basis of an integrated weed management approach. However, resistance to the parasitic weed is polygenic and influenced by numerous additive and non-additive gene actions, which complicates resistance breeding (Badu-Apraku 2007; Menkir and Kling 2007). Tolerance refers to the ability of a maize genotype to produce relatively better grain yield and biomass under Striga infestation compared to susceptible genotypes. Nevertheless, tolerance can result in the accumulation of very large reserves of Striga seed in the soil. On the other hand, resistant genotypes employ physiological and biochemical mechanisms such as inhibiting Striga attachment and reduced Striga reproduction. Striga rating and emergence counts have been used as indices for tolerance and resistance in improving maize against S. hermonthica (Badu-Apraku and Akinwale 2011). As suggested by Menkir et al. (2004) and Rodenburg and Bastiaans (2011), a successful breeding program is one that can identify and combine Striga resistance and tolerance. It is therefore essential for any breeding program to have a broad pool of genetic variation at the disposal of the breeder (Menkir and Kling 2007).

Potential sources of Striga resistance

Genetic improvement of any trait is dependent on the availability of genetic diversity. Potential sources of resistance to Striga have so far been found across a range of maize heterotic groups (Table 1). Maize breeders have the liberty of selecting desirable genes stacked in these materials within gene libraries. Other important sources include wild relatives such as Tripsacum dactyloides, Zea diploperennis, landraces and synthetics. Crop wild relatives provides a broad genetic pool for breeding purposes. Resistance to S. hermonthica have successfully been transferred from Zea diploperennis into maize, leading to the development of resistant inbred lines and synthetics (Rich and Ejeta 2008). Most importantly, genes encoding the production of reduced haustorial factors that inhibit parasitic attachments of S. asiatica on Tripsacum dactyloides have been revealed (Gurney et al. 2003). Although there is little expressed resistance to Striga among maize landraces in Africa, some Striga resistant landraces have recently been reported in Kenya (Midega et al. 2016). Farmers had been constantly selecting the most Striga resistant landraces over many generations, hence the likelihood of some levels of horizontal resistance.

Table 1. Summary genetic sources of Striga resistance.

Germplasm	Current sources	References
Wild maize relatives	T. dactyloides- source of Lhf genes for haustorial developmental barriers	Gurney et al. (2003)
	Z. diploperensis- major source of resistance in maize	Amusan et al. (2008)
Landraces	Sources of horizontal resistance	Midega et al. (2016)
Open pollinated varieties	IITA populations – e.g. TZL comp1 synw-1 and Acr94TZE Comp s-w	Menkir and Kling (2007)
Inbred lines	IITA and CIMMYT lines (characterized and non-characterized)	Menkir (2006), Karaya et al. (2012)
Hybrids	Resistant commercial genotypes e.g. Pioneer Hybrids and CGIAR varieties	Ransom et al. (1990), Chitagu et al. (2014), Akinwale et al. (2014)

Abbreviations: CIMMYT, International Maize and Wheat Improvement Centre; IITA, International Institute of Tropical Agriculture; Lhf, Low haustorial factors; OPV, open pollinated variety.

Potential genetic variability for S. asiatica resistance can also be harnessed from open pollinated, synthetic and composite maize populations. Since Striga is largely a small-scale farmers’ problem, resistance in open pollinated maize populations must be a priority for government and NGO scientists with the agenda of enhancing food security for resource poor farmers. Since southern African open pollinated varieties (OPVs) have been characterized for tolerance and resistance to drought, low soil fertility, soil acidity, and diseases, it would also be important to evaluate them for their response to Striga infection. OPVs have been a useful resource in the production of genetically diverse inbred lines in maize pre-breeding programs. They are also widely used by world’s leading research organizations like the International Maize and Wheat Improvement Centre (CIMMYT) and the International Institute of Tropical Agriculture (IITA) (Warburton et al. 2008; Karaya et al. 2012; Semagn et al. 2012). An ideal blend of inbred lines can combine both resistance and tolerance attributes in resultant hybrids.

Evaluation of genetic resources for Striga resistance

Various controlled environment and field screening methodologies have been developed and are applied in Striga improvement programs. Evaluation of germplasm for resistance to parasitic weeds can be carried out in controlled and field environments (Table 2). Controlled environments involve laboratory and greenhouse conditions under artificial infestation, while field experiments are carried out either in hotspot areas with supplementary infestation to increase the selection pressure.

Table 2. Summary of controlled and field environment screening protocols for Striga resistance.

Screening environment	Technique	Procedure	Selection criteria	Reference(s)
Laboratory	Agar gel assay	Host roots and Striga grown in Agar	Germination stimulant activity	Hess et al. (1992)
	Paper roll assay	Host and Striga rolled between germination paper	Early post infection resistance mechanisms	Ejeta (2007)
	Extended agar gel assay	Host roots and Striga grown in Agar	Hypersensitive reactions and incompatibility response	Mohamed et al. (2010)
	Rhizometrics	Host roots grown in root chamber	Hypersensitive reactions and incompatibility response	Cissoko et al. (2011)
Glass house	Pot/poly bag	Host and Striga grown in polybag	Resistance and tolerance traits	Rao (1985)
	Eplee bag	Striga seeds geminated within micron-mesh bags	Germination stimulant activity under natural conditions	Yonli et al. (2006)
Field screening	On site and on farm evaluation	Uniform infestation with Striga using appropriate experimental design Involves severity and damage rating	Resistance and tolerance traits	Haussmann et al. (2000)

Laboratory screening methods

Laboratory experiments have been designed to identify resistance components that combine to provide the host’s overall resistance expression during parasitic establishment (Ejeta and Gressel 2007). Several laboratory techniques have been developed to identify discrete mechanisms that confer components of resistance to S. hermonthica in sorghum. In vitro growth systems allow examination of the architecture of host roots and their biochemical mechanisms of resistance. Some laboratory assays allow studying of the release of germination inhibitors and haustorial initiation factors, as well as hypersensitive reactions (Ejeta et al. 2000). Further, the paper roll assay was developed to analyze the early stages of Striga infection (Ejeta et al. 2000; Haussmann et al. 2000). In this case, pre-conditioned Striga seeds are exposed to light, then spread evenly on a germination paper moistened with distilled water. The Striga seeds together with lined host seeds are then rolled between the germination papers. Observations are then made after three weeks when the papers are unrolled to reveal the extent of parasitic attachments on host roots and early resistance mechanisms.

The agar gel assay evaluates induced Striga germination and haustorial formation under controlled, reproducible laboratory conditions (Hess et al. 1992). The assay is useful for screening maize genotypes with a high degree of success in identifying resistant varieties (Reda et al. 1994). Many, especially those emanating from wild relatives Z. diploperennisis and T. dactyloides, have been identified using this technique. The technique was latter modified into the extended agar gel assay, whose purpose is to analyze post infection interactions between the host and Striga (Mohamed et al. 2010). Individual genotypes’ capacity to induce germination of Striga seeds evenly spread in a petri dish containing water ager is assessed using the assay. Selection is based on maximum germination distance of the Striga seed from the host’s root. Host defense mechanisms such as the hypersensitive reactions and incompatibility response in sorghum genotypes were understood through this methodology. Similarly, significant progress in the development of rice resistance to Striga has been achieved through evaluation of post attachment reactions using rhizotrons (Swarbrick et al. 2009; Cissoko et al. 2011; Rodenburg et al. 2017). Rhizotrons vary in their degree of sophistication but are essentially designed to allow microscopic observations on roots grown in different media as the plant grows. Effective use of rhizotrons is augmented with use of light or transmission microscopy or with root image analysis computer softwares such as WinRHIZO (Judd et al. 2015). Simple mini- rhizotrons made of poly glass plastic are cheap to contract and conducive for replicated multi-treatment experiments (Judd et al. 2014). Use of rhizotrons could allow fast and non-destructive selection for root architectural traits associated with Striga resistance and avoidance in maize.

Glasshouse screening

Screening in pots has also been a vital component of Striga resistance evaluations. Pots have been extensively used for screening for varietal resistance, nutritional inter -relationships between host and parasite, growth stimulant analysis, and herbicide effectiveness. Various pot screening techniques such as the ‘poly bag’ and seed pan have been described in detail by Rao (1985). The methods are commended for their effectiveness in screening for sorghum resistance to S. hermonthica. Of importance is the development of the ‘Eplee bag’ pot screening technique developed by Eplee (1992). Striga seeds are put in small bags made of micron mesh material, tied to strings and buried within the vicinity of the crop roots. At a given time, the strings are pulled to observe Striga germination. The method can also be utilized under field conditions to observe germination of Striga under natural conditions. Sevaral studies demonstrate the validity of the Eplee bag technique as a screening method (Gurney et al. 1995; Ahonsi et al. 2002; Yonli et al. 2006). The most important aspect to glasshouse evaluations is its compatibility with experiments on the efficacy of bio control agents such as Fusarium oxysporum f. sp. strigea. The technique allows a continuous non-disruptive assessment of the plant’s rhizosphere as demonstrated by Ahonsi et al. (2002) and Yonli et al. (2006) in their evaluation of potential bio control agents in controlling S. hermonthica.

Field techniques

Confounding effects of environmental conditions on polygenic inheritance of traits associated with Striga resistance make field screening indispensable despite the advances made through laboratory and pot experiments. The art of increasing accuracy and efficiency in field screening for Striga resistance has been in perfection over years. Achieving uniform infestations in Striga research is possible as demonstrated by yesteryears and modern day researchers (Rao 1985; Haussmann et al. 2000; Badu-Apraku et al. 2006). Berner et al. (1997) suggested mixing of Striga seed with fine sand at a proportion of 1: 99. A mixture of 10 g of seed and 5 kg fine sand when evenly mixed can deliver almost 3000 Striga seeds through a tablespoon load per planting station.

Resistance and tolerance to Striga in the field is quantified by the Striga emergence counts, Striga severity index, Striga damage rating, area under Striga number progress curve, area under Striga severity curves, grain and plant biomass yield (Haussmann et al. 2000). In maize improvement, an efficient rating scale must be used to estimate the breeding value of an individual genotype to Striga. Striga damage rating score, Striga emergence and agronomic traits that contribute to grain yield are widely used in selection for resistance in maize. Regardless of the shift in focus from selection for tolerance to that in favor of resistance, Striga damage rating score still forms a basis of maize improvement for Striga resistance (Menkir 2006). Striga hermonthica tolerance is controlled by additive genes that confer durable polygenic resistance (Kim and Adetimirin 1997). Tolerance as determined by the Striga damage rating is based on a scale of 1–9 where 1 refers to no damage symptoms and the genotype is regarded as highly tolerant, whereas 9 denotes highly susceptible and severely damage (Kim and Adetimirin 1997). A low Striga damage rating has been found to be directly linked to an increase in grain yield under the same level of infestations as the susceptible maize genotypes (Badu-Apraku 2007; Menkir and Kling 2007). However, selection for tolerance leads to the accumulation of a high Striga seed bank.

Genetic variants in crops have been advanced for further improvement on the basis of their ability to suppress parasitic attachment leading to a fewer number of emerged Striga plants than susceptible and tolerant genotypes (Rich and Ejeta 2008). Maize genotypes developed from T. dactyloides and Z. diploperennis seem to inherit both mechanistic and biochemical suppressant traits that inhibit Striga parasitism. However, as observed in sorghum and maize, the mode of inheritance of resistant genes is controlled by few dominant genes and this does not offer durable resistance to Striga across environments. It is against this background that maize bred for resistance against S. hermonthica is not guaranteed to succeed against damage by S. asiatica. Hence, there is a need to identify local genetic material for improving African maize populations for resistance to S. asiatica. Striga research in cereals especially in maize aims to reduce the loss in grain yield under high Striga infested zones. Maize yield is a function of minor genes that are inherited quantitatively in an additive manner and functionally related to other agronomic traits (Adetimirin et al. 2000). However, Striga damage rating and the number of emerged Striga plants are inversely related to maize yield (Badu-Apraku and Akinwale 2011).

Breeding techniques for S. asiatica resistance in maize

Conventional breeding

When a potential source of resistance is identified, the next critical step in the breeding program depends on the breeder’s ability to reach out to resistance genes and incorporate them into better adapted varieties. Conventional breeding techniques have been predominantly used in conferring superior combinations of Striga resistance alleles among susceptible cultivated crops (Menkir et al. 2004). Recurrent selection, half-sib selection, full-sib, S1 family selection schemes and hybrid breeding have been successfully utilized in developing resistance to most virulent Striga species in legumes and cereal. It is relevant to explore the applicability of most conventional breeding techniques as they have been utilized in various Striga resistance breeding programs.

Striga resistance traits have been accumulated successfully through recurrent selection in cereal crops. The IITA maize program has managed to enhance resistance to S. hermonthica through cyclical selection of genotypes with combined traits for low damage rating and reduced Striga attachments (Badu-Apraku and Akinwale 2011). Through recurrent selection, genetic gains in grain yield in segregating populations has been reported by Menkir et al. (2004) and Badu-Apraku et al. (2006). The IITA suggests pooling and mating maize populations of different genetic makeup with contrasting maturity groups and grain colors to produce Striga resistance breeding populations. The half sibling selection scheme as described by (John and Sleeper 1995) can be an easier route in developing composite populations with at least moderate resistance to S. asiatica. However, the full sib and selection from S1 progeny test allows for an increased scope of variability in progeny from source populations and greater control over pollen (Hallauer 1992). This might translate into an increased frequency of favorable alleles for Striga resistance in populations under selection (Menkir et al. 2004).

Quantitative trait loci (QTLs) for resistance to S. hermonthica have been identified from local populations including wild relatives and successfully transferred through backcross breeding into adaptable maize populations (Rich and Ejeta 2008). Germplasm derived through the backcross method form the basis for cultivar advancements towards achieving polygenic resistance to S. hermonthica. Such inbreds from Zea diploperennis and tropical maize have been essential in the development of S. hermonthica resistant open pollinated populations like Zea diplo SYNW-1, TZL Comp SYNW-1. This has been a key resource for communal maize production systems. Partial resistance to S. hermonthica was also observed in backcross hybrids from a resistant donor T. dactyloides (Gurney et al. 2003). The backcross breeding procedure is straight forward if a source population or donor with a high frequency of desirable alleles for Striga resistance is available. Rapid progress can be achieved in building resistance to S. asiatica if a donor exhibiting high dominance for S. asiatica resistance genes is identified. Under such a scenario, ideal recurrent parents would be genotypes combining early maturity and high yield (Badu-Apraku et al. 2006).

Despite the seed cost cutting effect and yield stability benefits associated with recurrent use of synthetic maize populations, the superiority in performance of hybrid cultivars is being realized with an increasing trend among southern African farmers. The desire to increase maize yields under marginal growing conditions and a rise in literacy can be the major reasons behind the increase towards complete adoption of hybrid technology in countries like Zimbabwe. Heterosis of hybrid varieties can be useful in mitigating the effect of S. asiatica on crop productivity. In line with the increased use of hybrid maize seed in west and central Africa, IITA has managed to accumulate S. hermonthica resistance in maize hybrids by crossing diverse inbreds lines (Menkir et al. 2004). This produce S. hermonthica resistant hybrids capable of suppressing parasite emergence, with some producing high grain yield under high infestation levels (Karaya et al. 2012). The rapid progress in development of resistant S. hermonthica hybrids in IITA programs can be attributed to the availability of stable resistant genotypes that have been used as testers for evaluating the broad pool of inbred lines for their general combining abilities (Menkir et al. 2004). Such resources are still limited in southern Africa. Extensive screening is among the few available means of selecting potential inbred lines for hybridization. However, assuming the presence of pleiotropic gene effects, inbred lines developed by CIMMYT Southern Africa Early Maize Breeding Program should be evaluated for their response to S. asiatica parasitism paving way for resistant hybrid cultivar development.

Marker assisted breeding for Striga resistance

Marker assisted selections have been an indispensable element of most breeding programs because it reduces selection errors associated with phenotypic evaluations. Progress in Striga resistance breeding has been rapid in sorghum than in any other crop. Owing to the relentless efforts of scientists in international institutes such as the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) and Purdue University, molecular markers linked to Striga resistance in sorghum have been identified. The partitioning of Striga resistance into various components based on the notion that resistance to the parasite is a combination of mechanism that influence the development of the parasite. Haussmann et al. (2001), Mutengwa et al. (2005) and Satish et al. (2012) reported the control of the low germination trait in sorghum by a recessive gene (lgs). Six resistance QTL were also identified by Grenier et al. (2007) in sorghum.

Despite the benefit of accelerated breeding for resistance to Striga using molecular markers, maize breeders working on Striga resistance have been reluctant to utilize this technology. This is possibly because the maize genome is complexed by about 78% repetitive DNA and retrotransposons. The polygenic nature of Striga resistance further complicates marker assisted Striga resistance breeding (Badu-Apraku 2007; Menkir and Kling 2007). Also, phenotyping large pools of germplasm for Striga resistance is expensive, making it a challenge to generate sufficient data for high resolution maker-trait-association and QTL discovery. Currently, there are limited reports on QTL conditioning Striga resistance in maize. However, a recent study involving marker-assisted recurrent selection for grain yield under drought stress and Striga infestation elucidated the importance of this technique in accumulating favorable genes for quantitative traits (Abdulmalik et al. 2017). Generally, maize breeding for Striga resistance has over relied on field screening whose accuracy can be confounded by a plethora of uncontrolled variations. As reported by Ejeta and Gressel (2007) some Striga resistance genes are recessive hence without molecular mapping some sources of resistance may be discarded. Hence, the current S. hermonthica resistant genetic resource developed through lengthy and costly breeding cycles might represent half of its potential. The search for Striga resistance QTL in maize may also adopt the same approach of assaying for resistance QTL for low induction of Striga germination and for genomic regions associated with field resistance. This can be achieved by genotyping the maize gene pool at different phases of parasitic development. Crosses can be made among recombinant inbred lines (RIL) derived from susceptible and resistant backgrounds as those reported in Menkir et al. (2004). This effort also requires the use of molecular markers for diversity analysis and heterotic grouping helpful in selecting parents for bi-parental population development (Mengesha et al. 2017). The resultant hybrids can then be evaluated for histological response mechanism to parasitic infection in agar gel assay tests before fingerprinting for the lsgs gene as described by Haussmann et al. (2004) and Mutengwa et al. (2005).

Role of genetic engineering in improving maize resistance to S. asiatica

Over the years’ scientists have been increasing research against parasitic weed species including Striga through the use of technological advancements that enable insertions, deletions of targeted genes, meddling of specific protein sequences and regulation of plant metabolites involved in synthesis of cellular structural constituencies (Aly 2007; Li and Timko 2009; Bandaranayake and Yoder 2013; Kirigia et al. 2014). Mediated gene silencing through ribonucleic acid interference (RNAi) has emerged as one of the most important features of the recent technological advances in the control of both Orobancea and Schrophularacea species in cultivated legumes and cereals (Aly et al. 2009; Runo et al. 2011). Since its discovery and subsequent use in plant virology, gene silencing by RNAi has been utilized at least more than once in developing transgenic maize resistant to both S. hermonthica and S. asiatica. Regardless of the varying degrees of success recorded by various workers in attempts to generate Striga resistance, all reports reflect a great potential of RNAi in developing transgenic maize capable of suppressing the prolificacy of the parasite. Interfering double stranded ribonucleic acid (dsRNA) constructs in maize plants could silence expression of genes responsible for vulnerability to S. hermonthica parasitism (Kirigia et al. 2014). The novelty of the authors’ work is their ability to locate and silence carotenoid sites controlled by 9-cis-e poxycarotenoid cleavage dioxygenases (NCEDs) genes responsible for synthesis of Strigaloctones. Runo et al. (2011) provide further proof of the feasibility of RNAi in engineering resistance to Striga in maize. Genetic modification through mutagenesis has also fostered development of a potential solution to Striga endemic in maize production system through discovery of maize variants with resistance and compatibility to actolactate synthesis (ALS) inhibitors (Kanampiu et al. 2001). A series of research work spanning more than a decade since the discovery of the imidazolinone-resistant (IR) gene led to the release of a commercial maize hybrid by BASF dubbed the Striga-way, which first proved its efficacy against S. asiatica in USA. Striga-way hybrids traded as Ua kayongo ‘Striga killer’ in Kenya have reported extensive increase in grain yields which successively translated into high incomes among its adopters (Ransom et al. 2012). Conventional incorporation of the IR gene in adapted tropical elite maize lines has also produced OPVs and Hybrids with outstanding levels of resistance to Striga.

The novelty of IR resistant maize in S. asiatica infested areas in southern African communities is based on its synergy with very low volumes of Imidazolinones applied at about 45 g ha⁻¹ as a seed coating agent (Gressel 2009; Ransom et al. 2012). This seed treatment concept has been widely tested under diverse environments across and outside Africa (Kanampiu et al. 2001, 2003; De Groote 2007; Menkir et al. 2010). Almost all evaluations maintain the same conclusion of prolonged suppression from parasitic infection that last close to about two thirds of the maize growing season. When farmers purchase coated IR maize seed they also reduce herbicide costs, environmental risks such as contamination of water bodies and harmful effects of improper herbicide handling, which is common among small-scale farmers. Another important attribute derived from localized application of Imidazolinones on maize seed is its compatibility with intercropping systems. This allows the inclusion of legumes and cucurbits commonly grown together with maize in countries like Zimbabwe. Kanampiu et al. (2002) showed that sensitive crops such as common bean and cowpeas are unaffected by the high concentration of Imazapyr when planted at least 15 cm away from the coated seed.

In as much as the genetic engineering is remarkable and relevant in addressing the Striga weed challenge, it has also met both unjustified and fair criticism from social, political and scientific spheres. Countries like Zimbabwe, Malawi and Zambia maintain an anti-genetically modified organism stance, which hinder the local use of unique bio-technologies such genetic engineering. Such policies have completely increased the vulnerability of communal farmers to biotic and abiotic agents of crop failures that are incessant in Southern Africa. However, with regard to herbicide resistance, regional policies concur with the insight raised in Ejeta and Gressel (2007) that ‘one must consider the possibility of needing other types of resistance, as herbicide resistance may eventually breakdown’.

An integrated approach to Striga resistance

The existence of variants among S. asiatica with differences in size and flower color as observed by Cochrane and Press (1997) may reflects on the differences in pathogenicity amongst the parasite ecotypes. Thus, to avoid loss of cultivar resistance, breeding efforts must be complemented with other control methods (Rubiales et al. 2009). Cultural practices such as hand weeding, crop rotation, trap cropping, intercropping, soil fertility improvement, soil tillage and various planting methods are among traditional Striga control strategies. While crop rotations involving non-host crops is mostly recommended, it don’t always fit within the cropping patterns of Southern African communal farmers since preference is given to staple crops such as maize, which ae always at deficit. An ideal method to include as a synergist to host resistance against S. asiatica in maize would therefore be one that does not compete for space with the main crop and must be self-perpetuating. Since low available soil N is a characteristic feature in communal lands and a major proponent of the efficacy of S. asiatica on maize, use of N fixing leguminous trees (Faidherbia albida), soil amendments involving kraal manure and biochar are equally viable options. These options are self-perpetuating and bring almost permanent solutions. However, they both take a lot of time and effort to establish. Therefore, use of a biocontrol agent, Fusarium oxysporum f. sp. strigae (FOS) can achieve control in the shortest possible time.

Fusarium oxysporum f. sp. strigea as Biocontrol agent

Fusarium oxysporum f.sp. strigea is a soil borne fungal pathogen that has shown immense potential in suppressing emergence and fecundity of both S. hermonthica and S. asiatica as evidenced by the vast array of studies confirming its efficacy on Striga species (Elzein and Kroschel 2004; Marley and Shebayan 2005; Schaub et al. 2006; Elzein et al. 2010; Zarafi et al. 2015). Fusarium sp. effects control on Striga species through germination and photosynthetic inhibition by producing phytotoxic compounds such as fumonisin B₁ (Elzein and Kroschel 2004)_. FOS is excluded from the strains producing fumonisins that are pathogenic to animals and humans but phytotoxic to Striga species especially S. asiatica (Elzein et al. 2006, 2010). Most importantly, effects of FOS on Striga are almost equals to those of IR technologies since it can also achieve high rates of control at low doses when delivered as seed treatment through film coating and in Pesta formulations (Elzein and Kroschel 2004). This has the major advantage of making the technology commercially viable and adopted by commercial seed producers. In the event of a lapse in below ground resistance component of FOS, liquid formulations on emerged Striga have also been reported to be effective in curtailing productive development of the parasite (Marley and Shebayan 2005). Therefore, the biocontrol agent has all the attributes that can fit perfectly with outcomes of breeding efforts against Striga such as resistance (partial or complete) and tolerance. Currently, the fungal pathogen has been widely tested in controlling S hermonthica in sorghum and in maize (Avedi et al. 2014).

Breeding maize compatible with Fusarium oxysporum f.sp. strigea

For effective control of Striga, the host must be able to facilitate the fungal antagonist to grow and colonize its roots. To establish a potentially mutualistic association, there must be deliberate selection of maize genotypes capable of maintaining a mycorrhizal symbiotic association with FOS in suppressing S. asiatica. Breeding efforts should, therefore, also focus on developing resistant genotypes promoting proliferation of FOS colonies on the host’s site of infection (Handelsman and Stabb 1996). Thus, the site of host infection within the rhizosphere should facilitate the active growth of FOS, while the fungus should maintain its virulence to the parasite and wholesome to the host plant (Elzein et al. 2010; Rebeka et al. 2013). Combining host resistance and F. oxysporum against S. hermonthica in sorghum by Rebeka et al. (2013), gave an insight on the possibilities of maintaining sufficient fungal colonies capable of sufficient control of Striga in the presence of a cooperative partner. The study reported a significant interaction between sorghum varieties and colony counts intervals suggesting that sporulation of F. oxysporum on sorghum rhizosphere might be genotype dependent. It is important to screen the fungal strains based on their ability to form colonies on different Striga species. Success in achieving this association through breeding lies in the heritability of the trait and mapping of gene associated with hospitality (Handelsman and Stabb 1996). There is a need to identify traits associated with genotype by FOS compatibility when selecting for Striga resistance among maize populations. Variation in suppression of Striga infestation among genotypes treated with FOS suggests the possibility of selecting genotypes that have high genetic potential for compatibility to FOS (Mrema et al. 2017). If genes conditioning FOS compatibility can be introgressed into candidate maize genotypes possessing partial Striga resistance through breeding, considerable integrated Striga control can be achieved using host resistance breeding and the biological agent.

This is possible with near isogenic lines as already proven in sorghum. However, since maize is highly allogamous, maintenance of association with F. oxysporum may be complicated by continuous segregation, especially when targeting synthetic populations. Avedi et al. (2014) reported inefficiency of FOS in controlling S. hermonthica in maize and they attributed this to the failure of the host’s rhizosphere to maintain sufficient pathogen levels that guarantee control of the weed. There is need to ascertain whether synergistic association between maize and F. oxysporum is dependent on host genetic factors or on the Fusarium strain (Foxy 2, PSM197, M12-4A, 4-3-B, F. nyagamai). However, cytological evaluations suggest that the delivery system is a major determinant of the ability of the fungus to colonize the hosts’ root system (Elzein et al. 2010). The intensity of root colonization increases with time following film coating of seeds with FOS. In addition, a mixture of 40% Arabic gum and dried chlamydospores achieves the highest intensity of root colonies (Elzein et al. 2006, 2010).

Conclusions and outlook

To accomplish complete control over Striga in Southern African maize production systems, there is need to complement agronomic practices with building genetic resistance in maize against this noxious root parasite. Achieving improvement of a polygenic trait such as resistance to Striga can only be necessitated by a rigorous search of resistance QTLs and stacking them into preferred maize varieties. This can be achieved through combining laboratory assays with field screening techniques, together with the use of molecular marker assisted selections. Such an approach will enable the discovery of Striga resistant genetic resources that have remained largely unexplored in Southern Africa. Accumulation of the resistance QTLs at present in most local programs may be facilitated by conventional breeding techniques and use of cost effective molecular markers. Most importantly, once partially or completely resistant sources are developed, combing them with other forms of resistance will ensure a durable outcome (Rubiales et al. 2006). Bio-control with FOS remains highly underutilized yet it is valuable and self-sustaining source of resistance in maize. Communal maize production systems can be improved if genetic resistance is included in the blend of ingredients making up an integrated approach against S. asiatica in Southern Africa and beyond.

Acknowledgements

The authors thank the National Research Foundation of South Africa for financial support of the first author. The World Academy of Science (TWAS) is also greatly appreciated for its financial and technical support.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributor

Admire Isaac Tichafa Shayanowako is a Ph.D fellow in Plant Breeding at the University of KwaZulu-Natal in South Africa. He is the main author for the current study. Prof. Shimelis Hussien is a Professor of Plant Breeding at the University of KwaZulu-Natal in South Africa. He is a co-author for the current study. Prof. Mark Laing is a Professor of Plant Pathology at the University of KwaZulu-Natal in South Africa. He is a co-author for the current study. Dr. Learnmore Mwadzingeni is a Post-Doctoral Research Fellow in Plant Breeding at the African Centre for Crop Improvement (ACCI), University of KwaZulu-Natal in South Africa. He is a co-author for the current study.