Current progress in Striga management

The Striga, particularly S. hermonthica, problem has become a major threat to food security, exacerbating hunger and poverty in many African countries. A number of Striga control strategies have been proposed and tested during the past decade, however, further research efforts are still needed to provide sustainable and effective solutions to the Striga problem. In this paper, we provide an update on the recent progress and the approaches used in Striga management, and highlight emerging opportunities for developing new technologies to control this enigmatic parasite. Recent progress and the approaches used in Striga management are summarized.


Background
Striga species of the Orobanchaceae family are obligate root parasites that infest staple crops in sub-Saharan Africa (SSA), Middle East, and parts of Asia (Tank et al., 2006;Parker, 2012). Striga hermonthica, Striga asiatica, and Striga gesnerioides are the most economically important parasitic plants among the 42 known Striga species. The three species differ in their host specificity. Striga hermonthica and S asiatica parasitize on cereals and sugarcane (Saccharum officinarum; Parker, 2009;Pennisi, 2010), while cowpea (Vigna unguiculata) is the main host of S. gesnerioides (Ohlson and Timko, 2020). The Striga, particularly S. hermonthica, problem has become a major threat to food security, exacerbating hunger, and poverty in many African countries (Pennisi, 2010;Khan et al., 2014). Although consequences are difficult to measure, a few estimates have indicated that Striga is affecting the life of more than 300 million people in Africa and causing enormous yield losses with a value ranging from 7 to 10 billion US$ annually (Emechebe et al., 2004;Gressel et al., 2004;Ejeta, 2007;Scholes and Press, 2008;Rodenburg et al., 2010). In heavily infested regions, farmers have been forced to abandon cereal cultivation and to switch to other less important crops (Atera et al., 2012a). The severity of Striga depends upon degree of infestation, seed viability, ecotypes, virulence, host crop susceptibility, climatic/edaphic factors, and cultural practices (Rodenburg et al., 2016).
Striga species are among the hardest parasitic plants to control (Berner et al., 1995;Nickrent and Musselman, 2004). Adaptability of Striga to a wide range of hosts and environmental conditions has made it one of the most widespread and successful parasitic plants (Mohamed et al., 2006). In addition, long-term management of Striga is hampered by the tremendous number and longevity of seeds, vast genetic variability, complex life cycle, and subterranean nature of damage (Joel, 2000;Huang et al., 2012a, b). Indeed, it is estimated that about 900,000 S. hermonthica plants can emerge from one hectare of infested sorghum field, which can add about 4.5 Â 10 10 seeds in one growing cycle (Bebawi et al., 1984). Very tiny (0.3 nm Â 0.15 nm) and light (4-7 lg) Striga seeds are easily dispersible in nearby fields through wind, animals, and agricultural tools, thereby gradually enriching seed reserve in the soil (Ejeta, 2007). The seeds remain dormant for a long period and germinate only after exposure to hot and humid conditions followed by perception of host derived germination stimulants, mainly strigolactones (SLs; Yoneyama et al., 2010;Joel and Bar, 2013;Al-Babili and Bouwmeester, 2015). Following germination, Striga radicle grows toward host roots. The perception of host-derived haustorium-inducing factors (HIFs), such as 2,6dimethoxy-1,4-benzoquinone, prevents further growth of the radicle and induces cell expansion and division, and proliferation of hair cells at its tip, forming a haustorium, a special invasive organ that penetrates host roots to enable siphoning off water, minerals, and nutrients. The vital role of the haustorium provides a rarely exploited option for controlling Striga by breeding varieties with low HIF release or developing compounds that specifically inhibit haustorium formation (Cechin and Press, 1993;Shen et al., 2006;Yoshida and Shirasu, 2012;Yoshida et al., 2016;Goyet et al., 2019).
Due to the persistence and severity of the Striga problem, there has been an extensive effort to develop simple, easy, and effective control strategies that can be employed either alone or integrated with existing approaches (Figure 1). In the past decade, intensive research on the interaction of Striga with its host at molecular level has opened up opportunities to develop new management strategies. For sustainable Striga management, any control method should target at least one of the following goals ( Figure 1): (1) PREVENTION: avoiding seed dispersal, for instance, by using clean crop seeds, tools, and fodder, controlling animal grazing or applying phytosanitary/quarantine measures; (2) CONTAINMENT: limiting new seed production by planting resistant (pre-and post-attachment, see below) varieties, using fertilizers, applying herbicides, employing a number of agronomic practices, such as hand weeding, deep sowing, burning, fallowing and soil solarization, and the application of chemicals that reduce the release of germination stimulants by the host; and (3) REDUCTION: reducing Striga seed bank accumulated in infested soils by employing cultural practices, such as trap cropping, inter cropping, and crop rotations, employing microbial agents that impact Striga and/or Striga/host interaction, applying synthetic germination stimulants in host's absence, and by developing specific inhibitors that block germination of preconditioned seeds and decrease their viability. Reduction of seed bank can be also achieved by planting resistant varieties that stimulate Striga seed germination, but withstand the parasitic attack (post-attachment resistance, see below). In this article, we provide an update on the recent progress and the approaches used in Striga management, and highlight emerging opportunities for developing new technologies to control this enigmatic parasite.

Host resistance
Deployment of resistant varieties is generally considered as the most economical, practical, and suitable long-term approach for controlling Striga (Hearne, 2009;Mandumbu et al., 2019). There are two main types of host resistance, that is, pre-attachment resistance resulting from SL profiles with low seed germinating activity, and post-attachment resistance that is based on hypersensitive response (HR)/incompatible response (IR). The recent identification of genes mediating the synthesis of different SLs opens up the possibility of modifying the SL profile of host plants by using genetic engineering and gene editing tools and, hence, increasing their resistance (Wakabayashi et al., 2019(Wakabayashi et al., , 2020. Genetic resistance can be employed independently or as a central element of an integrated Striga management (ISM). Here, we present the latest knowledge on Striga resistance in maize (Zea mays), sorghum (Sorghum bicolor), pearl millet (Pennisetum glaucum), rice (Oryza sativa), and cowpea (V. unguiculata).
Maize has surpassed the traditional cereals in SSA, with the highest cultivation area of about 39 million ha in 2018 (FAOSTAT, 2018). Considering the importance of this crop, significant progress has been achieved in identifying a number of Striga resistant maize varieties and hybrids with different types of resistance (Akaogu et al., 2013(Akaogu et al., , 2019Badu-Apraku et al., 2019;Menkir and Meseka, 2019). For instance, Amusan et al. (2008)  susceptible inbred line 5057. This resistance in ZD05 has been attributed to multilevel post-attachment barriers, particularly physiological or biochemical incompatibility to parasite growth and development. It is likely that this resistance is controlled by different genes that can be employed for durable and stable resistance. Similarly, Nyakurwa et al. (2018) compared quality protein maize (QPM) with non-QPM genotypes for S. asiatica resistance/tolerance under field/pot and agar gel assay conditions. They found several QPM genotypes with considerable levels of tolerance (i.e. no impairment of growth and yield despite of infestation), although they showed susceptibility similar to that of other genotypes in agar gel assays. However, the mechanisms underlying the tolerance of QPM genotypes remain elusive. In a similar study, Midega et al. (2016) evaluated six landraces for Striga infection under pot and natural field infestation and showed low Striga emergence, relative to hybrids. In addition to investigating resistance in the field, a number of pre-attachment resistance studies based on SL analysis have been conducted in the past few years. For instance, Karaya et al. (2012) screened a collection of 420 maize landraces, populations, and inbred lines and identified several landraces with low Striga germinating activity and reduced release of the highly active strigol, however, without further investigation of responsible genetic factors. Later, the effect of SL composition on Striga/maize interaction was further demonstrated by comparing the two cultivars Pioneer 3253 (Striga-susceptible) and KSTP94 (resistant), which showed that 5-deoxystrigol was exclusively released by the susceptible variety, while sorgomol was the main SL in root exudates of the resistant one (Yoneyama et al., 2015). Interestingly, the two varieties were indistinguishable with respect to arbuscular mycorrhizal (AM) symbiosis, indicating that the effect of SL composition on the interaction with AM fungi is different from that on Striga attack (Yoneyama et al., 2015). Furthermore, a study of S. hermonthica infestation using rhizotron unraveled the presence of post-attachment resistance in the aforementioned cultivar KSTP94, demonstrated by lower number of Striga attachments and biomass in comparison to the susceptible maize inbred line CML144 (Mutinda et al., 2018). Similarly, Gasura et al. (2019) evaluated 30 maize inbred lines for S. asiatica resistance in pots and using agar gel assays, and identified seven inbred lines with low germination stimulant production and decreased root attachment and emergence under pot conditions, while four inbred lines exhibited low Striga attachment.
Recently, a genome-wide association study for S. hermonthica resistance in maize identified significant loci on chromosomes 3, 9, and 10, which are related to plant defense. A total of 24 single nucleotide polymorphisms (SNPs; under Striga infested conditions) and 11 SNPs (under Striga-free conditions) showed significant association with grain yield and number of ears per plant. The identified loci and candidate genes (GRMZM2G060216, GRMZM2G057243, and GRMZM2G164743) could be excellent breeding source for the development of Striga-resistant maize genotypes through marker-assisted selection (MAS) in SSA (Adewale et al., 2020). Moreover, as mentioned above, the ability of cultivars, such as KSTP94, to resist Striga infestation at pre-and post-attachment stages makes them a very suitable genetic source for resistance breeding. In addition, the identification of genetic factors underlying the resistance will allow stacking them by using genetic engineering/genome editing technologies to generate highly resistant varieties to Striga infestation.
Sorghum is the second most important cereal crop after maize in SSA. It was grown on about 30 million ha in 2018 (FAOSTAT, 2018). In the past two decades, significant progress has been made with the identification and characterization of quantitative trait loci (QTLs) associated with Striga resistance in this cereal. Initially, Haussmann et al. (2004) detected QTLs in recombinant inbred lines derived from the cross between IS9830 and N13. By deploying MAS techniques, some QTLs have been transferred into elite sorghum varieties, leading to the development of sorghum cultivars with improved resistance to Striga (Ejeta, 2007). Moreover, by using 328 recombinant inbred lines, derived from a cross between SRN39 (low germination stimulant) and Shanqui Red (high germination stimulant) sorghum, Satish et al. (2012) fine mapped LOW GERMINATION STIMULANT 1 (LGS1) locus to a 400-kb region on chromosome 5. Interestingly, it was later demonstrated that mutations in LGS1 (lgs1-1 to lgs1-5 mutants) lead to a change in the stereochemistry of released SLs, replacing the dominant SL 5-deoxystrigol by the less active orobanchol (Gobena et al., 2017). These findings were confirmed by Mohemed et al. (2018) who showed that sorghum genotypes with high release of 5-deoxystrigol are more susceptible to Striga, compared to orobanchol releasing genotypes. More importantly, LGS1-mediated (loss-of-function) resistance was further characterized in various sorghum landraces with respect to S. hermonthica diversity and geographic distribution (Bellis et al., 2020). These authors reported that LGS1 loss-of-function mutations are adaptive and widely distributed among African landraces across a large region of highly Striga infestation in Africa. Using gene-edited sorghum lines, it has been further shown that the degree of LGS1-mediated resistance depends on parasite genotype and abiotic environment (Bellis et al., 2020). In addition to the well-characterized pre-attachment mechanism, few reports documented the presence of strong HR or incompatibility-based post-attachment resistance (Mohamed et al., 2010a(Mohamed et al., , 2010bMbuvi et al., 2017).
Pearl millet was cultivated on around 22 million ha during 2018, making it the third most important cereal after maize and sorghum in SSA (FAOSTAT, 2018). However, with paucity of reliable and adapted resistance donor sources, research and breeding for Striga resistance in pearl millet is still challenging, compared with other cereals (Wilson et al., 2000(Wilson et al., , 2004Kountche et al., 2013;Sattler et al., 2018).
Although several studies have been conducted to characterize mechanisms and inheritance of resistance to Striga in other cereals (Amusan et al., 2008;Cissoko et al., 2011;Jamil et al, 2011a;Satish et al., 2012), knowledge on individual resistance mechanisms, their genetic and physiological basis are still lacking in pearl millet. To fill this gap and by deploying a genotyping-by-sequencing approach, a genetic map of SNP markers together with single sequence repeats was constructed using a segregating population derived from a cross between a wild relative, resistant to Striga (Wilson et al., 2004) and a cultivated-susceptible pearl millet parent (Moumouni et al., 2015). The availability of such genomic resources will help in identifying and mapping QTLs associated with Striga resistance, which can be deployed in a MAS. Also, using conventional field based breeding, significant progress has been made with the identification of six varieties (M141, M239, M029, M197, M017, and KBH), showing higher yield and less Striga susceptibility (Kountche et al., 2013). The response to five cycles of phenotypic recurrent selection for Striga resistance was evaluated in a diversified pearl millet gene pool developed from these varieties (Kountche et al., 2013). The authors also reported the development of the first Striga-resistant experimental varieties (Kountche et al., 2013).
Rice was grown in 2018 on about 14 million ha in SSA (FAOSTAT, 2018). Both, pre-and post-attachment resistances were detected in this cereal. The Nipponbare rice cultivar shows strong post-attachment resistance to S. hermonthica, likely because of its incapability to build xylem-xylem connections with the parasite (Gurney et al., 2006). To shed light on the genes underlying this resistance, the response of Nipponbare and the susceptible cultivar IAC 165 was investigated using gene expression profiling. This study unraveled an association of the induction of defense genes with resistance, and of that of genes involved in nutrient transport, amino acid metabolism and abiotic stress response with susceptibility (Swarbrick et al., 2008). Genes induced in the resistant cultivar include three with unknown function, which are localized within a major resistance QTL on chromosome 12 and might have a significant contribution to the resistance (Swarbrick et al., 2008). In another study, three Striga resistance QTLs were detected in Koshihikari-Kasalath backcross inbred lines. A QTL of major effect was verified and narrowed down and could be, therefore, a good target for MAS (Swarbrick et al., 2009). Striga attack on the resistant rice cultivar Nipponbare is accompanied by an accumulation of lignin, guaiacyl, and syringyl at the site of Striga infection and by induction of phenylpropanoid pathway genes. The role of lignification in Nipponbare post-attachment resistance was demonstrated through manipulation of genes regulating lignin composition, which led to Striga susceptibility in this cultivar (Mutuku et al., 2019). Besides this structural barrier, post-attachment resistance also depends on a change in the level of defense hormones. Transcriptome analysis of infested roots indicated an involvement of the plant hormones jasmonic acid (JA) and salicylic acid (SA) in Striga postattachment resistance. Mutant analysis confirmed the role of JA, but not that of salicylic acid. However, WRKY45 knockdown-a regulator of the SA/benzothiadiazole-mediated defense response-can lead to Striga susceptibility. This phenotype (susceptibility) could be rescued by exogenous JA application, indicating that WRKY45 contributes to Striga defense by modulating the interaction between JA and SA and positively regulating SA/benzothiadiazole and JA pathways (Mutuku et al., 2015).
A screen for pre-attachment resistance unraveled NEw RICe for Africa (NERICA) cultivars, such as NERICA1, and their parent CG14, which release low amounts of SLs (Jamil et al., 2011a). In another study, rice high tillering cultivars, for example, Super Basmati or TN1, showed low SL production and Striga infection, in contrast to low tillering rice varieties, for example, IAC-165 or IAC-1246 (Jamil et al., 2012a). Interestingly, Cissoko et al. (2011) reported that the abovementioned NERICA cultivars also exhibited post-attachment resistance to S. hermonthica and S. asiatica, caused by incompatibility response or lack of xylem-xylem connections to rice endodermis. Both resistance studies were further validated and confirmed under field conditions (Atera et al., 2012b;Rodenburg et al., 2015Rodenburg et al., , 2017. Similarly, the upland rice variety Umgar was characterized by pre-and postattachment resistance to S. hermonthica under lab, pot, and field conditions (Samejima et al., 2016a).
Cowpea is an important vegetable and food legume in many African countries but its yield is severely affected by Striga gesnerioides (Parker, 2009). About seven distinct races of S. gesnerioides (SG1-SG6 and SG4z) were classified based on their genetics and parasitism on cowpea in West Africa (Botanga and Timko, 2005;Botanga and Timko 2007). Cowpea resistance to S. gesnerioides is conferred by single dominant genes in a race-specific manner (Timko and Singh, 2008;Timko et al., 2012). In addition, many cowpea landraces and local accessions possess post-attachment resistance that is based on HR at the site of attachment (Timko and Singh, 2008). Indeed, Li and Timko (2009) identified and characterized RSG3-301 from the cowpea cultivar B301, which is involved in S. gesnerioides resistance. Silencing of RSG3-301 in B301 plants caused susceptibility to S. gesnerioides race SG3 due to reduced HR. These findings led to the conclusion that the race-specific Striga resistance in cowpea is likely an effector-triggered immunity that activates intracellular NLR proteins (RSG3-301) upon the recognition of pathogen/parasite effectors. Supporting the race-specific interactions in S. gesneriodes-cowpea associations, the cultivar B301 was, however, susceptible to the S. gesnerioides race SG4z producing a small soluble effector protein at high amounts in haustoria, which is transferred to the host root (Huang et al., 2012a, b). This protein can suppress the host innate immunity by binding to a host BTB-BACK domain-containing ubiquitin E3 ligase homolog POB1 (POZ/BTB containing protein 1). Overexpression of VuPOB1 led to a reduction of SG4z parasitism due to increased HR, while its silencing caused susceptibility, suggesting that VuPOB1 might be a positive regulator of the HR response (Su et al., 2020). As mentioned above, S. gesnerioides races and their distribution are dynamic systems, influenced by genetic drift and gene flow. Ohlson and Timko (2020) recently investigated S. gesnerioides diversity and known sources of resistance in cowpea. They collected 58 unique S. gesnerioides populations from 9 West African countries and screened 7 cowpea lines for resistance. Results obtained showed that none of the cowpea lines was resistant to all S. gesnerioides populations and that there is no S. gesnerioides population that can overcome the resistance of all seven cowpea lines. Analysis of single sequence repeats of the Striga populations unraveled high differentiation and suggested that genetic relatedness is generally a result of geographic proximity rather than of host compatibility. This study indicates that generating a broad-spectrum and durable S. gesnerioides cowpea-resistant lines requires stacking of multiple resistance genes (Ohlson and Timko, 2020).

Cultural and agronomic practices
Trap cropping, sowing of false host such as cowpea, groundnut, sesame, and cotton to stimulate suicidal germination and to improve soil fertility were reported to be an effective way of seed bank depletion (Atera et al., 2013;Goldwasser and Rodenburg, 2013). Combining trap crops and nitrogen fertilizers was also reported to significantly decrease Striga seed bank (Tadesse, 2018). Cover cropping, sowing crops for the protection and enrichment of the soil showed Striga suppression directly through mulching, induction of suicidal germination, or its shading effect (Pickett et al., 2010;Goldwasser and Rodenburg, 2013;Randrianjafizanaka et al., 2018). Intercropping of cereals with legumes or a trap crop such as Desmodium spp. (Push-Pull) reduced Striga emergence by improving soil fertility, organic matter, and soil moisture content and releasing allelochemicals, such as C-glycosylflavonoids, isoflavanones, isoschaftoside, phenolics, 3,4-dihydroxybenzoic acid, which might impact Striga germination, growth, or development (Makoi and Ndakidemi, 2012;Khan et al., 2014;Pickett et al., 2014;Hooper et al., 2015;Midega et al., 2017;Hailu et al., 2018). A combination of herbicide-resistant maize varieties intercropped with legumes appeared more effective against Striga (Kanampiu et al., 2018).
Fallow and crop rotation were found not only to improve soil fertility and crop yield, but also to lower Striga infestation in Cameroon (Ayongwa et al., 2010). In a similar study, a reduction in Striga seed bank through fallowing was reported in Mali and Niger (Van Mourik et al., 2011). Rotation of a nonhost legume crop can considerably reduce the Striga seedbank in infested fields, leading to decreased infestation and significantly enhanced yield, compared with continuous cereal cultivation (Franke et al., 2018). Recently, Kountche et al. (2019) suggested field partitioning into two sections where suicidal germination agents (see below) and existing integrated Striga and soil management practices can be rotated to sustain Striga seed bank reduction. Hand pulling, uprooting Striga by hand or hand tools, is still considered as the cheapest traditional Striga control method (Goldwasser and Rodenburg, 2013). It is recommended to apply this method before Striga flowering, to prevent further seed setting and seed bank accumulation (Ayongwa et al., 2010;Sibhatu, 2016). However, hand weeding is laborious, time-consuming and less effective in reducing damage to standing crop (Mahuku et al., 2017). Fertilizer application and organic amendments showed a negative impact on Striga emergence (Ayongwa et al., 2011). The reduction of Striga infestation of rice upon fertilizer application is likely caused by a decrease in SL exudation (Jamil et al., 2011b). Similarly, nitrogen-phosphate-potassium fertilizer, microdosing of di-ammonium phosphate, and phosphate-based seed priming have been shown to reduce SL release and Striga parasitism in sorghum, pearl millet, and rice (Jamil et al., 2012b(Jamil et al., , 2014a(Jamil et al., , 2014bIsah et al., 2013).
Recently, combining conservation agriculture practices, such as cover cropping and fertilizer applications, with Striga-resistant varieties was found to alleviate Striga impact on rice and maize (Abdallah et al., 2015;Rodenburg et al., 2020). Although most of the cultural practices are less expensive and helpful in reducing parasitic seed bank and improving soil fertility and soil texture, they are constrained by low farmer acceptance, the need for introducing additional crops, labor forces, and financial resources (Murage et al., 2011).

Biocontrol by microbiome
Myco-herbicides developed from the fungus Fusarium oxysporum showed Striga inhibition by reducing its attachment to cereals and decreasing seed bank in infested soils (Rebeka et al., 2013;Zimmermann et al., 2016;Bàrberi, 2019). Some F. oxysporum strains produce high amounts of the amino acids L-leucine and L-tyrosin which are toxic to Striga-but not to maize-as they disrupt the tightly regulated free amino acid homeostasis. In addition, methionine released by F. oxysporum strains can be converted by soil microbes into the germination stimulant ethylene, causing suicidal germination of Striga seeds (Nzioki et al., 2016;Rubiales et al., 2018). Seed coating with F. oxysporum (FOXY2) was proposed as an effective way to deal with Striga under field conditions (Elzein et al., 2010;Ndambi et al., 2011Ndambi et al., , 2012Rebeka et al., 2013;Watson, 2013). Later PSM197 and FOXY2 were encapsulated in a granular formulation (PESTA) for easy application/longevity, and a reduction of 75% in Striga emergence was observed in maize and sorghum crops (Schaub et al., 2006). FOXY2 was further classified as F. oxysporum f. sp. strigae (FOS) based on its highly selective inhibition of S. asiatica and S. hermonthica emergence (Mrema et al., 2018;2020;Shayanowako et al., 2020). The characteristics that make mycoherbicides striking bioagents against Striga include host specificity, high aggressiveness, easy mass production, genetic diversity, and long storage life (Rebeka et al., 2013;Nzioki et al., 2016). Fungal delivery of primary inoculum on toothpick, multiplication of secondary field inoculum and farmers training are key components of this approach (Nzioki et al., 2016). The application of FOXY T14 led to an increase of 42%-56% in crop yield and reduction of 80% in Striga infestation under experimental conditions. However, F. oxysporum has not been extensively used in real fields, which might be due to a low effectiveness (Nzioki et al., 2016). Moreover, F. oxysporum might cause diseases, such as Fusarium wilt and dieback, in solanaceous crops (Zarafi et al., 2015).
Soil microbes like plant growth-promoting bacteria (PGPR), employment of AM fungi and some bacterial strains caused considerable reduction in Striga germination, attachment, and emergence (Lendzemo et al., 2009;Babalola, 2010;Hassan and Babiker, 2011;Mazaheri-Naeini et al., 2015). The potential of AM fungi in alleviating Striga infection has been indicated in a number of studies (Akiyama et al., 2005;Lendzemo et al., 2006;Bouwmeester et al., 2007;Xie et al., 2010). AM fungi not only enhance cereal growth and performance to withstand Striga damage but also facilitate host plant's uptake of water, phosphorus (P), and micronutrients from the soil through the wide net of extraradical fungal hyphae (Bonfante and Genre, 2010). Increased uptake of P through symbiotic interaction by AM fungi could ultimately reduce SLs exudation by the host in the soil, thereby lowering Striga infection (Lendzemo et al., 2007;López-Ráez et al., 2011).
In a screening study, four PGPR suppressed Striga infestation in sorghum. Moreover, application of the strain B. subtilis GBO3 led to the death of 35%-59% of emerging Striga tubercles and to a 23% reduction in Striga attachment (Mounde et al., 2015). Striga seed germinating activity of sorghum root exudates decreased significantly upon treatment with Pseudomonas bacterial suspensions, which might be due to degradation of SLs (Ali et al., 2013). As shown for isolates of Bacillus, Streptomyces, and Rhizobium genera, the production of compounds with antibiotic activity and of extracellular enzymes, such as xylanases, pectinases, and amylases, can directly cause Striga seed decay (Neondo et al., 2017). In addition, soil microbes could impact Striga by releasing amino acids, such as tyrosine, leucine, and/or methionine (Vurro et al., 2009) or by producing secondary metabolites, such as anthranilic acid, b-lactone derivatives sesquiterpenoids, tricothecenes, which might interfere with SL perception (Tyc et al., 2017). Similarly, there is some evidence for SL degradation by fungi, which might lead to reduction in Striga germinating activity of released exudates (Boari et al., 2016). The germinating activity of root exudates and, hence, infestation by Striga can be also affected by soil microbes that modulate root architecture and growth (Huang et al., 2014).
The usage of soil microbes (AM fungi, PGPR, and other bacterial strains) is now considered as a promising, costeffective, and environmentally safe approach for combating Striga (Watson, 2013;Samejima and Sugimoto, 2018). However, a number of biotic and abiotic factors can affect the efficacy of this approach, especially under field conditions. Finding a suitable inoculum medium, its mass production, suitable formulation, storage, shelf-life, consistency, and compatibility of applied microbes with the host, and the maintenance of their activity in infested soils must be taken into consideration. The validation and further development of this microbes-based biocontrol approach still requires intensive research and field testing under varying climatic and edaphic conditions (Müller-Stöver et al., 2016;Mohammadi, 2019).

Use of herbicides and suicidal agents
Seed coating of imazapyr-resistant maize with imidazolinone herbicides, such as imazapyr, imazapic, pyrithiobac, and imazaquin, caused a reduction in Striga emergence throughout the planting season and led to a three-to four-fold increase in maize yield (Menkir et al., 2010;Chikoye et al., 2011;Habimana et al., 2014;Makumbi et al., 2015). Similarly, treatment of cowpea seeds with imazaquin at 0.24 kg a.i. ha -1 significantly decreased Striga infection in different cowpea genotypes (Lado et al., 2018). Albeit promising results, availability of IR resistance seeds to the farmers, application technology, the risk of generating resistance in Striga itself, and the impact on environment are important issues that need to be considered (Ransom et al., 2012).
Application of synthetic germination stimulants to deplete Striga seed bank in infested soils by inducing suicidal germination in the absence of host has recently gained a lot of attention (Samejima et al., 2016b;Zwanenburg et al., 2016;Kountche et al., 2019). Several SL analogs (Fig. 2) have been developed and evaluated Mwakaboko and Zwanenburg, 2011;Kgosi et al., 2012;Boyer et al., 2014;Screpanti et al., 2016;Vurro et al., 2016;Jamil et al., 2018Jamil et al., , 2020Uraguchi et al., 2018;Prandi and McErlean, 2019). Functional tests revealed large variation in the efficiency of these analogs in exerting different SL functions, including the induction of Striga seed germination under lab conditions. Generally, SL analogs/mimics are quite variable in their structure but have in common the D-ring and ether bridge characteristic for natural SLs Fukui et al., 2013Fukui et al., , 2017Cala et al., 2016;Takahashi et al., 2016;Oancea et al., 2017;Jamil et al., 2020). Structure-activity relationship studies have shown that minor structure modification can lead to significant changes in the biological activity (Cohen et al., 2013;Zwanenburg et al., 2013;Jamil et al., 2018Jamil et al., , 2019Jamil et al., , 2020. Following the discovery of the central SL biosynthesis intermediate carlactone (Alder et al., 2012), a carlactone-based SL analog, nitro-phenlactone was developed. Nitro-phenlactone showed high parasitic seed germinating activity in Phelipanche ramosa, which was at least comparable to GR24, but not in Striga (Jia et al., 2016). Subsequently, a series of carlactonoic acid-based analogs, called methyl-phenlactonoates (MPs), were developed and tested . The three analogs MP1, MP3 and MP16 showed high suicidal germination activity on S. hermonthica in lab, greenhouse and field trials performed in Burkina Faso Kountche et al., 2019). Similarly, a carbamate-derived SL analog T-010 was applied in an infested field in Sudan, which led to a significant reduction in Striga emergence and better growth of sorghum (Samejima et al., 2016b). Recently, a chemical screen for SL agonists, followed by structural modifications, led to a highly active compound that triggers Striga seed germination under lab conditions at femtomolar range concentrations (Uraguchi et al., 2018). However, the question about the activity of this analog in field is still open.
Although a large number of synthetic stimulants have been developed and proposed for usage as suicidal germination agents, there are only very few studies that assess the practicability and success of the suicidal germination approach and the activity of these stimulants in infested fields (Samejima et al., 2016b;Zwanenburg et al., 2016;Kountche et al., 2019). Indeed, the evaluation of the syntheticstimulant-based suicidal germination strategy requires an application protocol suitable for infested regions in Africa, in addition to efficient and easy-to-synthesize SL analogs. Such a protocol needs to include an appropriate formulation and to account for local conditions, including less developed infrastructure. Given the water scarcity in the Sahel zone, the realization of the suicidal germination concept in the arid and semi-arid Striga-infested regions in SSA requires also a method that does not consume large amounts of water. Recently, Kountche et al. (2019) established a pipeline for assessing the germinating activity of SL analogs, which includes tests under lab conditions and in greenhouse, and mini and large field trials in Africa. For application in field, they developed a protocol for rain-fed agriculture, which requires a minimum of water and exploits rainfall for diluting and distributing of sprayed SL analogs. Following this protocol, the application of three SL analogs, that is, MP1, MP3, Nijmegen-1, at a final concentration below one micromole led to an up to 55% and 65% reduction in Striga emergence in an infested sorghum and pearl millet field, respectively . These results and the study of Samejima et al. (2016b) demonstrate that the suicidal germination is a practicable and promising strategy to deplete Striga seed banks in infested African fields.

Development of novel chemicals for Striga control
Recent knowledge on SL biosynthesis and perception, and the availability of sequenced genomes have paved the way for developing new alternatives to combat root parasitic plants by chemicals, which decrease host SL release by inhibiting biosynthetic enzymes, down-regulating related transcripts, or blocking SL perception in the parasite.
Inhibitors of SL biosynthesis reported so far target either the formation of carotenoids, the precursor of SLs, or SL biosynthesis itself. Application of fluridone or norflurazon, which inhibit the desaturation of phytoene in carotenoid biosynthesis (Zheng et al., 2020), to rice plants decreased SL production and, hence, Striga germination (Jamil et al., 2010). However, treatment with carotenoid biosynthesis inhibitors may cause unwanted effects, given the diverse functions of carotenoids (Felemban et al., 2019;Zheng et al., 2020). Ito et al. (2011) developed the triazole-derivative TIS13 as inhibitor of SL biosynthesis and demonstrated its effect in decreasing rice SL release. On the basis of these results, a new set of structurally related SL biosynthesis inhibitor candidates were synthesized, including TIS108 that showed considerable reduction in the release of Striga germination stimulants without affecting the host . Further development of TIS108 led to KK05 that caused higher reduction of 4-deoxyorobanchol release in rice exudates (Kawada et al., 2019).
SL antagonists, which specifically inhibit Striga SL perception, are a promising tool that could be applied in the presence of the host, enabling Striga control throughout the cropping season and complementing the suicidal germination strategy employed in host's absence. The development of specific germination inhibitors has become possible through the identification of Striga SL receptors, particularly the most sensitive one ShHTL7, which are involved in Striga seed germination and differ from the receptors mediating SL response in host plants (Toh et al., 2015). Striga SL receptors that regulate seed germination are supposed to have evolved from the homolog receptor KAI2 that perceives karrikins, smoke derived compounds mimicking a yet unidentified plant growth regulator inducing seed germination in nonparasitic plants (Toh et al., 2014;Waters et al., 2014;Conn and Nelson, 2016). Thus, the Striga ShHTL7 can replace KAI2 and mediate SL-dependent seed germination in Arabidopsis kai2 mutant (Holbrook-Smith et al., 2016). This capability allowed the establishment of a high-throughput screening for chemicals that block the Striga ShHTL7 receptor and act as SL antagonists, which led to the identification of soporidine (SOP; Holbrook-Smith et al., 2016). Evaluation of SOP activity on Striga seed germination indicated its capability to block this process (Tsuchiya et al., 2018). A serendipity discovery of a potent SL antagonist that specifically blocks Striga ShHTL7 was recently reported . The authors aimed at elucidating the structure of ShHTL7 by crystallography and found the receptor tightly bound to the detergent Triton X-100 usually used in protein purification. Functional studies confirmed that Triton X-100 is a SL antagonist that specifically blocks the Striga ShHTL7, but not host SL receptors, and reduce Striga infestation in greenhouse . SOP and Triton X-100 can be considered as lead compounds for the development of efficient Striga herbicides that act by inhibiting seed germination. An improvement of Triton X-100 efficiency might be achieved by modifying it to become a covalently binding SL antagonist. Similarly, 2-methoxy-1-naphthaldehyde as an SL antagonist (Mashita et al., 2016) or simple b-lactones (TFQ0011) that act as covalent and high-efficient inhibitors of SL receptors have been developed based on the general SL perception mechanism (Xiang et al., 2017). Triazole urea compounds, such as KK094, that covalently bind to nonparasitic plant SL receptors have been recently produced and shown to efficiently inhibit rice SL perception (Nakamura et al., 2019). A combination of structural elements of Triton X-100 and KK094 or b-lactone inhibitors might lead to highly efficient Striga-specific herbicides. The recently developed Triton X-100/KK094 hybrid structure KK023-N1 ( Figure 2) is an example for such compounds (Randa et al., manuscript under review).
Zaxinone is a natural growth-regulating apocarotenoid metabolite that has been recently shown to be required for normal rice growth and development. In addition, application of this compound promotes rice root growth and downregulates SL biosynthesis at the transcript level . A greenhouse study demonstrated that zaxinone application can alleviate Striga infestation . These activities point to zaxinone as a suitable candidate for reducing Striga infestation by accelerating host growth improving its performance and decreasing parasitic seed germination. Very recently, a series of easy-to-synthesize Mimics of Zaxinone (MiZax) have been developed and tested in lab and greenhouse. This study unraveled MiZax3 and MiZax5 as potent zaxinone mimics that promote rice growth, reduce SL release and decrease Striga infestation . The simple synthesis protocol and high efficacy make MiZax very promising candidates for field application.

Integrated Striga management
Effective Striga management cannot be achieved by a single control method and requires the integration of different approaches (Figure 1; Ejeta, 2007;Sibhatu, 2016). For instance, complementing host resistance with the use of F. oxysporum caused effective Striga reduction (Mrema et al., 2020;Shayanowako et al., 2020). Similarly, the reduction in Striga infestation achieved through seed coating of imazapyr-resistant hybrid maize can be significantly further increased by exploiting maize Striga-resistance (Kamara et al., 2020). As a further example, a cereal-legume crop rotation can be combined with the application of synthetic germination stimulants to deplete accumulated seed bank in infested soils (Hailu et al., 2018;Kountche et al., 2019). It can be also anticipated that the integration of suicidal germination technology with Striga-specific herbicides and/or zaxinone analogs can be very effective in dealing with the Striga problem in African agriculture.

Concluding remarks and perspectives
A number of Striga control strategies have been proposed and tested during the past decade; however, further research efforts are still needed to provide sustainable and effective solutions to the Striga problem (see Outstanding Questions).
First, further understanding of the molecular and genetic basis of host resistance and host-parasite interaction is needed to breed crops with durable resistance. Use of genomic resources and modern tools, such as targeted gene editing or mutation breeding, can translate this knowledge into resistant crops. Second, rotation and/or intercropping with false host are important and cost-effective components of Striga management. This approach can be a vital element in existing ISM in SSA. Third, farmers willingness, commitment and planning, labor, capital, and input availability in particular cropping system are important factors for the effectiveness of push-pull, seed coating, or toothpick/FOS technologies. Compatibility to climatic and soil factors, famers awareness, dissemination of information, and transfer of technology to smallholders' farmers are essential factors that need further attention. Fourth, the effectiveness of novel chemicals (SL analogs, antagonists, bio-stimulants) might depend upon formulation and method and time of application. A suitable formulation of efficient compounds, mass scale, low-cost synthesis, and practical field application for rain-fed African agriculture are crucial, particularly for seed bank depletion by suicidal agents. Moreover, impact of these chemicals on soil fauna and flora, soil structure, persistency, and residual effects on environment must be investigated prior to their release for on-farm application. Last, but not least, a smart package of technology integrating Striga-resistant cultivars with either fertilizers, mycoherbicides, herbicide-based seed coating, or new chemicals still needs to be worked out to achieve complete and robust control of Striga.

OUTSTANDING QUESTIONS
• What are the molecular factors underlying Striga resistance? • How can we efficiently mobilize the existing omics tools to breed durable Striga-resistant varieties? • What determines the stereochemistry of released strigolactones? • Can we design a microbiome community to suppress Striga impact and enhance host crop productivity in sub-Saharan Africa? • What is the impact of synthetic germination stimulants and strigolactone-related novel chemicals on soil fauna and flora?