Striga Germination Stimulant Analysis

Abstract

Striga is a flowering parasitic plant that can infect a broad range of crops including sorghum, maize and rice. The life cycle of Striga is intricately linked to its host, using host derived signals to initiate germination and attach to its host. Host susceptibility to Striga is thus directly correlated to the production and exudation of these signals. In this chapter we discuss the strigolactones, the most important class of Striga germination stimulants. We review the structural diversity of the strigolactones, their transport and biological relevance in plant and rhizosphere, and their importance for host specificity. In addition, we describe methods to analyse germination stimulant production by a Striga host: how to, reproducibly, collect the root exudate of a host, how to process these exudates, partially purify them and analyse them using analytical chemistry or a high-throughput image analysis-based germination bioassay.

Introduction

Selected Striga resistant cereal mutants derived from gamma irradiated stocks (Chapters “Screening for Resistance to Striga Hermonthica in Mutagenized Sorghum and Upland Rice in Burkina Faso”–“Phenotyping for Resistance to Striga Asiatica in Rice and Maize Mutant Populations in Madagascar”) determined through laboratory assays to have lower Striga germination stimulant activity (Chapter “An Agar-based Method for Determining Mechanisms of Striga Resistance in Sorghum”) than their unmutagenized progenitors may be further characterized by examining the strigolactones in their root exudates. This chapter provides further details of the parasitic association of Striga with its cereal hosts with particular emphasis on strigolactones, the germination stimulants in host root exudates. Protocols for collecting and chemical and biological characterization of root exudates are also described.

Striga life cycle. The life cycle of Striga commences with the germination of its seed, followed by its attachment to the host, marking the start of the parasitic phase. Before the seed can respond to germination stimuli, various metabolic processes occur in what is known as the “conditioning” phase. Full imbibition, which occurs within a day, must be followed by several days (optimally 10–14) of sustained moisture and warmth to become responsive to germination stimulants. If conditioned seed does not encounter a germination stimulant, it eventually returns to an unresponsive state of secondary dormancy (Matusova et al. 2004). With germination, the Striga radicle emerges from the seed coat and grows up to a centimeter. Once the radicle encounters a potential host root, its apex forms an haustorium that aids attachment and penetration of the epidermis, cortex and ultimately stele where xylem connection is established with the host (LeBlanc et al. 2012). A Striga shoot will only emerge with successful vascular connection. Germination without attachment and vascular contact with a suitable host will kill Striga since it cannot survive on its own seed reserves. An established parasite eventually emerges from the ground, flowers and sets up to 200,000 seeds/plant. This leads to the formation of vast seed banks, which pose a significant challenge in the management of parasitic weeds, as these seeds can remain dormant and viable in the soil for many years until they encounter a host.

Scientists had long ago observed that the root exudates of clover, maize, sorghum, and linseed contain compounds that induce the germination of parasitic plant seeds. (Saunders, 1933; Brown et al. 1949, 1951). The first witchweed (Striga lutea Lour.) germination stimulant was isolated from the root exudate of cotton (Gossypium hirsutum L.) and named strigol (Cook et al. 1966, 1972). Over 30 different molecules with similar properties to strigol, collectively referred to as strigolactones (SLs), have been discovered in the plant kingdom since then.

Germination stimulants. After its discovery in cotton, strigol was subsequently identified in the root exudates of Striga hosts such as maize (Zea mays L.), sorghum (Sorghum bicolor (L.) Moench), and proso millet (Panicum miliaceum L.) (Siame et al. 1993). Two additional germination stimulants were subsequently identified in the root exudates of sorghum, sorgolactone (Hauck et al. 1992), and from cowpea [Vigna unguiculata (Walp)], alectrol (which was later identified as orobanchyl acetate; Ueno et al. 2015; Xie et al. 2008). Both share basic chemical structure to strigol. In 1995, Butler coined the name strigolactones for these strigol-related compounds (Butler 1995). Some years later, orobanchol, a germination stimulant for the broomrape, Orobanche minor (Smith), was isolated from the root exudate of red clover (Trifolium pratense L.) (Yokota et al. 1998). This showed that SLs are used by both Striga and Orobanche spp. as germination cues to ensure germination in the presence of a host root. Subsequently, sorgomol (Fig. 1; Table 1) was identified in the root exudate of Sorghum bicolor and shown to induce germination in Striga hermonthica and O. minor (Xie et al. 2008). Fabacol, fabacyl acetate (germination stimulants of O. minor) and strigone (a highly active germination stimulant of S. hermonthica) were isolated from the root exudates of pea (Pisum sativum L.) (Xie et al. 2009) and from the non-host, Houttuynia cordata (Kisugi et al. 2013), respectively.

Fig. 1

Chemical structures of sorghum strigolactones and the synthetic strigolactone, GR24

Table 1 Optimized UPLC-MS/MS conditions for analysis of selected strigolactones

The discovery of new, non-canonical SLs that do not possess the consistent ABCD-ring structure (as shown in Fig. 1) was a fascinating and unexpected development. One such example is the compound methyl carlactonoate (MeCLA), which has been found in Arabidopsis (van Ha et al. 2014; Seto and Yamaguchi 2014). Other examples are heliolactone, identified in the root exudate of sunflower (Ueno et al. 2014) and zealactone, identified in maize root exudate (Xie et al. 2017) that induces S. hermonthica germination (Charnikhova et al. 2017). Avenaol was identified in the root exudate of black oats (Avena strigosa Schreb.) and is a strong germination stimulant for Phelipanche ramosa, while it is virtually inactive in S. hermonthica and O. minor (Kim et al. 2014). To date, all examined plant species have been found to release a combination of multiple SLs. The SL profiles can vary among different species, and sometimes, even within different cultivars of the same species. Moreover, the quantities and proportions of SLs can fluctuate under varying growth conditions and developmental stages (Yoneyama et al. 2008).

Other roles of strigolactones. For many years, plant scientists were puzzled by the question of why plants exude SLs into the rhizosphere, thereby inviting parasitic weed attacks. In 2005, Akiyama et al. finally provided an answer to this quandary by demonstrating that SLs play a crucial role in establishing beneficial arbuscular mycorrhizal (AM) fungi symbiosis (Akiyama et al. 2005). Strigolactones stimulate hyphal branching in AM fungi, enabling them to extend towards and colonize the host root. This symbiotic relationship is a vital factor in the growth and survival of terrestrial plants, as demonstrated by its prevalence in approximately 80% of all plant species (Bonfante and Genre 2015; Gutjahr and Parniske 2013).

This mutualistic relationship plays a crucial role in providing the host plant with minerals that are absorbed by the fungal hyphae, effectively expanding the plant's root surface area and allowing it to exploit a much larger volume of soil. In exchange, the heterotrophic fungal partner receives reduced carbon compounds from the plant's photosynthesis to support its own growth (Gutjahr and Parniske 2013; Bonfante and Genre 2015). The benefits of AM symbiosis explain why plants, particularly under phosphorus deficiency, release SLs into the soil (Gutjahr 2014; Khosla and Nelson 2016). It is plausible to assume that root parasitic weeds, which evolved much later than AM symbiosis, have co-opted the SL signal as a reliable indicator of host presence. This was achieved through the evolution of a highly sensitive SL detection system coupled with the induction of seed germination. It is worth noting that the function of SLs as a plant hormone was reported three years after their role in AM symbiosis was revealed (Gomez-Roldan et al. 2008; Umehara et al. 2008). The identification of SLs as a plant hormone regulating shoot branching and tillering was made possible by the availability of mutants with increased shoot branching and tillering from various plant species. Through genetic analysis and grafting studies, these mutants were classified into two groups: those deficient in the synthesis of a presumed mobile shoot branching-inhibitory signal and those affected in the perception of this signal. The biosynthetic mutants lack SLs, and application of the SL analog GR24 restores their branching and tillering phenotypes. This finding showed that SLs are the elusive shoot branching inhibitory signal. The supposed perception mutants, on the other hand, did not respond to GR24, which further supports the conclusion that SLs are the signal in question. (Gomez-Roldan et al. 2008; Umehara et al. 2008). Thus, SLs are a new plant hormone that plays an important role in the adaptation of plants to nutrient deficiency, regulating senescence, root development, and stem secondary growth (Koltai 2011; Ruyter-Spira et al. 2013; Brewer et al. 2013; Al-Babili and Bouwmeester 2015; Waters et al. 2017; Jia et al. 2018). Moreover, SLs seem to play a role in the plant response to biotic and (other) abiotic stresses (van Ha et al. 2014; Torres-Vera et al. 2014; Wang et al. 2015; Decker et al. 2017).

Strigolactone biosynthesis. The discovery in 2008 that strigolactones act as the long-sought branching-inhibiting hormone (Gomez-Roldan et al. 2008; Umehara et al. 2008) facilitated the rapid elucidation of the core strigolactone biosynthesis pathway. This pathway involves a ß-carotene isomerase (DWARF27) and two carotenoid cleavage dioxygenases (CCD7 and CCD8), which produce carlactone, the precursor for all known strigolactones. However, biosynthesis of strigolactones diverges strongly after carlactone, leading to the production of different types of strigolactones, including orobanchol-type and non-canonical strigolactones that lack the classic ABC-ring structure of canonical strigolactones. Our understanding of the details of strigolactone biosynthesis after carlactone is limited, with only a few enzymatic steps identified in Arabidopsis towards non-canonical strigolactones and in rice towards orobanchol-type strigolactones (Bouwmeester et al. 2021). Notably, maize strigolactones with unique structures have been identified, but their biosynthesis remains unknown (Charnikhova et al. 2017).

Strigolactone transport. Grafting is a common technique used to study hormone and metabolite transport between plant shoots and roots. Mutants lacking strigolactones (SLs) exhibit a distinct high-branching phenotype. However, when these mutants were grafted onto wildtype rootstocks, the bushy phenotype was fully rescued, indicating that SLs are transported from roots to shoots (Kohlen et al. 2011). Conversely, grafting wildtype scions onto SL-deficient rootstocks did not result in increased branching, suggesting that SLs can also be synthesized in aerial tissues (Kohlen et al. 2011). Grafting studies have also shown that SL biosynthesis intermediates can be transported from roots to shoots. For example, grafting of MORE AXILLARY BRANCHING (MAX) mutants max3 or max4 onto max1 rootstocks restored wildtype branching to the mutant shoots, suggesting that carlactone or its derivatives, which accumulate in max1 rootstocks, are transported into the mutant shoots where they are further metabolized into SLs that inhibit branching (Booker et al. 2005).

Transport of SLs from roots to shoots has been suggested to occur through the xylem (Kohlen et al. 2011). However, labeled orobanchol and 4-deoxyorobanchol, that were applied to the roots, were detected in the shoot, but not in the xylem (Xie et al. 2015). Moreover, transport of the labeled SLs was slower than expected for xylem-mediated movement (Xie et al. 2015). In addition to transport from root to shoot, SLs are exuded into the rhizosphere by specific transporters, such as the Petunia PLEIOTROPIC DRUG RESISTANCE 1 (PhPDR1), an ATP-binding cassette (ABC) transporter (Kretzschmar et al. 2012). PhPDR1 is located in the plasma membrane of sub-epidermal cells in lateral roots, the hypodermal passage cells where AM fungi penetrate the roots. The activity of PhPDR1 may indicate the presence of a local maximum of SL, which could guide AM fungal hyphae towards the hypodermal passage cells. PaPDR1, an ortholog of PhPDR1 found in P. axillaris, exhibits asymmetric localization in different root tissues, with its transporter localized at the apical membrane of root hypodermal cells in root tips, potentially facilitating the transport of SLs towards the shoot (Sasse et al. 2015). In hypodermal passage cells above the root tips, PaPDR1 is localized in the outer-lateral membrane, consistent with its postulated role in transporting SLs from roots into the rhizosphere.

SLs and host specificity. Many parasitic plant species display host specificity, often with a narrow host range, such as O. cumana that mostly only parasitizes sunflower. For other species, the host range is wider, such as P. ramosa that can infect several members of both Solanaceae and Brassicaceae (Gibot-Leclerc et al. 2016; Perronne et al. 2017). Within the broad host range P. ramosa, however, there are ecotypes that seem to exhibit a narrower host specificty. For example, the germination of P. ramosa seeds harvested from different hosts, is differentially stimulated by the exudates of these different hosts (Perronne et al. 2017). Also in S. hermonthica, although it can infect several cereals (maize, sorghum, millet, rice), we find ecotypes that better parasitize millet than sorghum and maize and vice-versa (Kim, 1994; Mohemed et al. 2018). Whether this specificity is based on the recognition of germination stimulants is an intriguing question. Indeed, the sunflower parasite, O. cumana, germinates in response to dehydrocostuslactone, a molecule present in the exudate of sunflower, but not to strigolactones (Auger et al. 2012). In S. hermonthica, strigolactone composition of the host root exudate seems to play a role in host specificity. For example, a sorghum ecotype of S. hermonthica germinates much less well with a millet exudate and vice-versa (Mohemed et al. 2018). In sorghum, maize and rice, S. hermonthica susceptibility seems to correlate with 5-deoxystrigol production (Jamil et al. 2012; Xie et al. 2015; Mohemed et al. 2018). In contrast, orobanchol seems to correlate with reduced susceptibility to S. hermonthica (Gobena et al. 2017; Mohemed et al. 2018).

Similar to auxin and other plant hormones, strigolactones also need a receptor to exert their effects on biological processes. The branched phenotype of mutants has been instrumental in identifying components involved in strigolactone perception and downstream signaling. This includes the discovery of two canonical α/β hydrolase fold proteins: the strigolactone receptor, D14, and the homologous receptor, D14like1/KAI2/HTL (abbreviated as HTL) (Waters et al. 2017). D14 is a receptor with hydrolase activity and the strigolactone ligand is hydrolyzed upon binding (Seto et al. 2019). Although the crystallographic studies indicate that the D-ring is essential for the interaction between D14 and MAX2, it remains uncertain whether hydrolysis is a prerequisite for signaling or merely a result (Yao et al. 2016; Seto et al. 2019). When a strigolactone or the D-ring binds to D14, it causes a conformational change that allows D14 to recruit the F-box protein MAX2/D3. MAX2/D3 then targets repressor proteins such as DWARF53 and SMXLs for ubiquitination, leading to their proteasomal degradation. As a result, gene expression is induced and plant development is altered, including the inhibition of bud outgrowth (Al-Babili and Bouwmeester 2015). While D14 has been confirmed as the receptor responsible for regulating plant development through strigolactones, several impressive studies have demonstrated that HTL has evolved new ligand binding specificity in root parasitic broomrapes and witchweeds through duplication and adaptation (Conn et al. 2015; Toh et al. 2015). This facilitated germination of the parasites upon perception of strigolactones secreted by their host. Intriguingly, the function and ligand of HTL in non-parasitic plants is unknown (Waters et al. 2017). With the strigolactone receptors in hand we can now screen for ligands. Uraguchi et al. (2018), for example, used a S. hermonthica, HTL to find new synthetic suicidal germination stimulants (Uraguchi et al. 2018).

Testing individual SLs for their ability to induce parasitic plant seed germination has confirmed that seeds of different species differentially respond to specific SLs (Wang and Bouwmeester 2018). For example, germination of O. minor is induced with about 200-fold lower ent-2’-epi-orobanchol concentration than S. hermonthica (Ueno et al. 2011). Similarly, in the same concentration, 5-deoxystrigol only induced germination in S. hermonthica but not in of S. gesnerioides (Ueno et al. 2011).

Role of SLs in other belowground interactions of the host. Plants utilize multiple adaptation strategies when exposed to stress such as phosphate or nitrogen deficiency. The most crucial ones include altering the architecture of roots and shoots, establishing beneficial interactions with microorganisms, and modifying the pH of the rhizosphere (Bouwmeester et al. 2007; Péret et al. 2011; Auger et al. 2012; Kumar et al. 2015). All of these mechanisms aim to enhance the availability of nutrients for plants. It is interesting to note that when plants are grown in media deficient in nitrate, and particularly phosphate, there is a stimulated increase in the production of SLs (López‐Ráez et al. 2008; Yoneyama et al. 2012; Marzec et al. 2013; Torres-Vera et al. 2014). There are various indications that the up-regulation of SL production is involved in the plant adaptation to low-nutrient conditions. For instance, in the absence of phosphate, plants tend to prioritize the production of lateral roots (Péret et al. 2011), which increases the surface area in contact with the soil. This adaptation is facilitated by the important role of auxin in initiating lateral root primordia and promoting the emergence of lateral roots (Chiou and Lin 2011; Sun et al. 2014), but SLs also play an important role in the adaptation of root architecture to phosphate deficiency (Ruyter-Spira et al. 2011; Sun et al. 2014; Kumar et al. 2015) and, in cross talk with auxin, SLs induce an increase in lateral root density (Ruyter-Spira et al. 2011). The alteration in root architecture that occurs in response to low-nutrient conditions may also impact the susceptibility of host plants to parasitic infections. This modification increases the likelihood of host roots encountering parasitic plant seeds. Furthermore, the increased production and release of SLs in response to nutrient deficiency not only promotes the colonization of plant roots by beneficial microorganisms but also stimulates the germination of parasitic plant seeds, thereby increasing the risk of infection (Jamil et al. 2012, 2014).

In addition to the interaction with AM fungi (Akiyama et al. 2005; Besserer et al. 2006), SLs were also shown to regulate nodulation. The SL deficient increased branching (ramosus) pea rms1 mutant displayed a reduced nodule number (Foo et al. 2013). Also in soybean, nodulation was decreased in the SL deficient GmMAX3b mutant line while overexpression resulted in an increased nodule number (Haq et al. 2017). The precise mechanism behind this impact of SLs on nodulation is not fully understood, including whether it is due to their signaling function or hormonal effect.

In addition to SLs, plants also exude a wide range of other chemicals into the rhizosphere. These molecules provide carbon for microorganisms and serve as signaling molecules that play a critical role in recruiting and selecting specific microorganisms. Among the most extensively studied examples are phenylpropanoids, which are involved in both symbiotic and pathogenic interactions in the rhizosphere (Abdel-Lateif et al. 2012; Liu and Murray 2016) and allelopathy (Bais et al. 2006).

Recent work on sorghum showed that the Striga resistant genotype SRN39, which exudes mostly the SL orobanchol (Gobena et al. 2017) recruited a different microbiome than other genotypes that mostly exude 5-deoxystrigol (Schlemper et al. 2017). Also in rice a significant correlation was detected between orobanchol and a number of micro-organisms, including AM fungi and Burkholderia–Caballeronia–Paraburkholderia and Acidobacteria that potentially solubilize phosphate (Kim et al. 2022). Observations suggest that these changes at the microbiome level can mitigate infection of the host by parasitic plants. From a Striga suppressive Kenyan soil, bacteria were isolated that induced up to 45% of decay in Striga seeds (Neondo et al. 2017). Other mechanisms by which soil micro-organisms could suppress parasitic plants include the production of germination inhibiting factors, inhibitors of radicle growth and haustorium formation, strengthening the vigor of the host plant by activating plant defense mechanisms or competitive utilization of signaling molecules inducing parasitic plant seed germination.

SLs and parasitic weed management. Controlling parasitic Orobanchaceae weeds poses several challenges. On the one hand, their parasitic nature makes it difficult to intervene without harming the host plant, due to the physical and physiological connection between the two. On the other hand, the properties of their seeds, such as their small size, fecundity, and easy dispersal, make it challenging to prevent the spread of the weed. These factors lead to a rapid increase in the soil seed bank. Therefore, preventing the distribution of seeds and containing infested areas should be a significant focus of any parasitic weed management strategy, in addition to direct control measures against the parasites (Rubiales et al. 2009; Rubiales and Fernández-Aparicio 2012). In this Chapter, we will not review all possible control and management strategies of parasitic weeds, but focus on methods—that are already used or can potentially be developed—that are based on the importance of SLs in the lifecycle of these parasites. Indeed, there are several strategies of weed management focusing on the SLs, trying to avoid the stimulation of germination, or conversely to favor it, in the absence of a host. These practices are briefly considered in the next sections.

Host resistance through low strigolactone exudation. As discussed above SLs are the main germination stimulants for root parasitic plants. In studies that evaluated the induction of parasitic plant seed germination by exudates from different genotypes and cultivars of several crop species, a positive correlation was demonstrated between the SL concentration in the root exudate and the germination rate (Jamil et al. 2011a, b; Fernández-Aparicio et al. 2014; Yoneyama et al. 2015; Mohemed et al. 2018).

An approach for the management of parasitic plants in agricultural crops could thus be to reduce germination of the parasitic plant seeds by reducing the exudation of the germination stimulants. Several studies explored natural variation in germination stimulant production, for example in the New Rice for Africa (NERICA) rice cultivars. This work showed that several NERICA cultivars (1, 2, 5, 10 and 17) displayed post-germination resistance to S. hermonthica and S. asiatica unlike NERICA 7, 8 and 11, which were susceptible (Rodenburg et al. 2015). In parallel, variation in SL production in the NERICA genotypes was demonstrated, resulting in differences in Striga germination induction (Jamil et al. 2011b). The combination, by breeding, of germination-related resistance with post-germination resistance could result in better durable Striga resistance (Cissoko et al. 2011; Jamil et al. 2011b).

A pea genotype with reduced SL exudation displayed partial field resistance to O. crenata infection (Pavan et al. 2016). Further in sorghum, the Striga resistant genotype SRN39 produced much less 5-deoxystrigol than the susceptible Tabat (Yoneyama et al. 2010). Later, SRN39 was shown to produce more orobanchol, instead of 5-deoxystrigol, due to a tentative modification in the SL biosynthetic pathway (Gobena et al. 2017). This mechanism was also observed in a number of other Striga resistant sorghum genotypes (Mohemed et al. 2018). Interestingly, the differences in SL composition in maize and sorghum did not affect the level of AM colonization (Gobena et al. 2017).

Besides exploiting natural variation, biotechnological approaches aiming to generate low SL exuding plants could be a strategy to reduce infestation by parasitic plants (Torres-Vera et al. 2014). Indeed, it was demonstrated in tomato that by knocking down SL biosynthesis through genetic modification, the mutant was more resistant to O. ramosa infection (Kohlen et al. 2012). A reduction in SL production to obtain parasitic weed resistance was also achieved unintentionally. The broomrape resistant tomato mutant Sl-ORT1 developed by Dor and coworkers was later found out to be SL-deficient and the resistance thus associated to the low amount of exuded SLs (Dor et al. 2010). Breeding, through conventional or biotechnological approaches, for a reduction in exuded SLs potentially also has negative consequences given their importance for the control of shoot and root architecture and the acquisition of nutrients through AM fungi (López‐Ráez et al. 2008). This could possibly be prevented by approaches that reduce transport of SLs into the rhizosphere, which is facilitated by an ABC transporter, PDR1 (Borghi et al. 2015). However, under certain abiotic stress conditions, this could still negatively affect the adaptive capacity of plants by hampering mycorrhization. Particularly the example of sorghum shows that solutions in which the composition rather than the level of the SLs is changed may be the best solution (Gobena et al. 2017). Nevertheless, several examples show that a reduction in SL production results in an acceptable level of resistance without large consequences for the plant phenotype (Jamil et al. 2011a, b; Pavan et al. 2016). In order to prevent that this partial germination-based resistance is overcome by parasitic plant populations, a combination of pre-and post-attachment resistance mechanisms is necessary.

In this chapter we describe how to analyse the level of germination stimulant produced by a Striga host, with an emphasis on the SLs as the most important class of Striga germination stimulants. With the methods described here it is possible to screen collections of germplasm and/or mutants suspected to induce lower Striga germination, to find genotypes that display (partial) Striga resistance through a lower (or no) induction of Striga germination. We describe how to reproducibly collect the root exudate of hosts, how to process these exudates, partially purify them and analyse them using a Striga germination bioassay, for which we will also describe a method to analyse the results through high-throughput image analysis, or using advanced Multiple Reaction Monitoring—Liquid Chromatography Coupled to Mass Spectrometry (MRM-LC-MS).

Protocols

Root exudate collection. One of the of plant adaptive characteristic features is their capacity to constantly communicate with their surrounding environment through metabolites. These chemical compounds are widely diverse and involved in a broad range of responses: attraction of beneficial or harmful micro-organisms, iron chelation, rhizosphere pH modulation, nutrient uptake, etc. It has been shown that to face a specific biotic or abiotic stress plants modulate the composition of their root exudates to attract the right beneficial micro-organisms. Due to coevolution, harmful organisms are often able to hijack these signals. A well-studied dual effect compounds example are the strigolactones (SLs) that the plant uses to attract arbuscular mycorrhizal fungi and that also induce parasitic plant seed germination. In their attempt to mitigate phosphate deficiency through mycorrhization, SL levels in roots and root exudates increase under phosphate stress. Therefore, characterizing the SL profiles of Striga hosts is often done under phosphate starvation. In addition to manipulating nutrition, temperature and light to optimize SL exudation, other aspects such as SL stability need to be considered while collecting root exudates. The following steps have been successfully used in our lab to collect root exudates from cereal Striga hosts like maize, millet, rice and sorghum.

Place the desired number of deglumed seeds in a 15 ml Falcon tube. Add 10 mL of 4% sodium hypochlorite with 0.2% Tween-20 and shake the tube for 45 min on a rotating mixer. After removing the bleach, soak the seeds for 30 s in 10 mL of 70% ethanol and wash with 10 mL of sterile MilliQ water (repeat this step three times). Wash the seeds an additional four times with 10 mL of sterile MilliQ water.

Moisten autoclaved 90 mm Whatman^® glass microfiber filters (GF/A Grade/qualitative filter paper) with 3 mL sterile MilliQ water in a sterile 90 mm petri dish, then place 10–15 seeds on each filter paper. Seal the plates with parafilm and wrap them with aluminum foil, then incubate them for two days at 30 °C. Select the best seedlings and transfer them to 50 mL Falcon tubes filled with 45 mL of washed sand. Sand can be prewashed by placing a bulk quantity in a sieve and placing under running water. Move the planted tubes to a greenhouse or growth chamber set at 28 °C/25 °C, 70% relative humidity, 16 h photoperiod, 450 μmol. photons m⁻² s⁻¹. This light intensity is ideal but a range of 200–450 μmol. photons m⁻² s⁻¹ could be used. Select the better growing plants after two days. Water the plants every two days with 5 mL of modified half strength Hoagland’s solution, containing 10% of the normal amount of phosphate. The use of sand for plant growth gives a relatively clean exudate and facilitates the low nutrient treatment. However, it has the disadvantage of drying very quickly. Daily watering is therefore very important for the first three days after the transfer of the seedlings. Grow the plants under these conditions for two weeks.

Elute the exudates with 30 mL of 5% ethanol–water and collect 20 mL. The analyte enrichment can be improved by addition of a suitable organic modifier at 5% to the root exudate collection solvent; also see below (Floková et al. 2020). Keep the exudates on ice for the whole exudate collection process. During the root exudate collection, the stability of SLs can be an issue. Keep the exudates therefore refrigerated until they are combined with the solid-phase extraction (SPE) sorbent. Centrifuge the exudates (4 °C, 3000 rpm for 20 min) and transfer supernatant to a second Falcon tube.

Processing the exudates, preserving and storing samples. The process of sample extraction and further purification significantly influences the overall sensitivity of final UHPLC-MS/MS analysis. The universal solid phase polymer-based sorbent for neutral compounds Oasis HLB^® (Waters), combining both hydrophilic and lipophilic retention characteristics, was successfully applied in pre-concentration of broad spectrum of strigolactones (Floková et al. 2020). The additional selective purification step using polar interactions with silica sorbent eliminates the co-extraction of lipophilic contaminants, greatly improves the sample purity and decreases detection limits of mass analysis on the sample background. The complete sample preparation procedure is shown in Fig. 2.

Fig. 2

Analysis of Striga germination stimulants. Root exudate collection and concentration for strigolactones measurement using UPLC-MRM-MS/MS (left panel) and image analysis pipeline for analysis of Striga germination rate (right panel)

Filter ~ 100 mL of collected exudates using a Büchner vacuum funnel¹ and transfer 20 mL of filtered exudates to a 50 mL Falcon tube. Add the mixture of internal standards (authentic or stable deuterium-labelled counterparts of strigolactones) to the sample and vortex.²

Before sample concentration with SPE column, activate the sorbent of Oasis HLB^® with 100% acetonitrile (3 mL) and subsequently equilibrate with 3 mL of 5% ethanol/water (v/v). Solvents and sample extracts are passed through the SPE cartridge Oasis HLB^® at a consistent flow rate 1–2 drops per second.

Load the extract of root exudates onto conditioned SPE sorbent (~ 50 mL). Discard the flow-through fraction. Wash sorbent with deionized water (3 mL) to remove salts and unretained compounds. Discard the flow-through fraction. Place 4 mL borosilicate glass vials in the rack of a solid-phase extraction (SPE) glass block vacuum manifold (e.g., 24-port manifold lid) and perform the elution with 80% acetone (v/v, 3 mL). Evaporate samples to dryness using vacuum sample concentrator. Keep dry samples at − 20 °C until the following purification step.

To purify the sample, dissolve dry sample in 150 μL of 100% ethyl acetate and vortex. Fill the reconstituted sample up to 3 mL with 100% hexane (2.85 mL). Before sample purification with SPE column, activate the sorbent of Strata^® SI-1 Silica with 100% ethyl acetate (3 mL) and subsequently equilibrate with 100% hexane (3 mL). Solvents and sample extracts are passed through the SPE cartridge Strata^® SI-1 Silica at a consistent flow rate 1–2 drops per second. Avoid drying out the silica-based SPE sorbent bed and loss of column capacity. Load dissolved sample onto conditioned SPE sorbent (3 mL). Discard the flow-through fraction. Wash the sorbent with 100% hexane (3 mL). Discard the flow-through fraction. Place 4 mL borosilicate glass vials in the rack of vacuum manifold and perform the elution with 80% ethyl acetate/hexane (v/v, 3 mL). Evaporate samples to dryness using vacuum sample concentrator. Keep dry samples at − 20 °C until the final LC–MS analysis.

Germination stimulant analysis using UPLC-ESI–MS/MS: strigolactones. Ultra-high performance liquid chromatography (UHPLC) coupled to fast scanning tandem mass spectrometry (MS/MS) is routinely applied for simultaneous profiling and quantitation of strigolactones, isolated from different sample matrices (Boutet-Mercey et al. 2018; Floková et al. 2020; Rial et al. 2019, p. 20; Sun et al. 2014). After analyte enrichment and additional purification using SPE-based sorbents, special attention has to be paid to suitable chromatography system and mobile phase composition. The sufficient retention of strigolactones is achieved on reversed-phase ethylene-bridged hybrid polymer-based columns (BEH) packed with sub-2-micron particles, improving both separation efficiency and speed of analysis (Fig. 3). The presence of eluent additive formic acid in dual-component mobile phase (water/acetonitrile) facilitates the chromatographic resolution between strigolactone stereoisomers with the same MS-fragmentation pattern, such as 5-deoxystrigol and 4-deoxyorobanchol (Floková et al. 2020). However, the main reason of mobile phase acidification is to support protonation of the strigolactone molecules [M + H]⁺ in the electrospray (ESI) source, operating in positive mode, and thus improve analyte extraction by the mass analyzer. Compounds are analysed in multiple reaction monitoring mode (MRM) using two specific precursor-to-product ion transitions, listed in Table 1.

Fig. 3

The chromatographic separation of five authentic strigolactones coupled to internal standards with similar retention time: strigol, orobanchol and sorgomol to GR24 (red peak line); sorgolactone and 5-deoxystrigol (5-DS) to [²H₆]-5-deoxystrigol ([²H₆]-5-DS, blue peak line). Separation is performed on Acquity UPLC™ BEH C18 column

To analyse SLs, reconstitute the dried sample with 100 μL of 25% acetonitrile/water (v/v) and vortex. Transfer the dissolved sample onto the membrane of a 750 μL micro-centrifugal nylon filter with pore size 0.2 μm and spin in a pre-cooled (4 °C) centrifuge at 10,000 g for one minute. Transfer the eluent to a 250 μL glass insert in a 1.5 mL LC/MS autosampler vial. Place the vials in the autosampler of the UPLC System,³ maintained at 10 °C. Inject 5 μL of the sample onto an analytical Acquity UPLC™ BEH C18 column (2.1 × 100 mm, 1.7 μm particle size). Separate analytes with a 12 min binary gradient elution, consisting of 15 mM formic acid in both water (eluent A) and acetonitrile (eluent B), at flow rate 0.45 mL/min and constant temperature of 45 °C. Set the elution profile as follows: 0–0.4 min initial isocratic elution at 15% B; linear gradient increase to 27, 40 and 65% B in 0.65, 5 and 8 min, respectively; isocratic elution at 65% until 8.7 min. Finally wash the column with 95% B for 1.6 min and equilibrate back to initial conditions (15% B) for 1.7 min. The retention time of individual compounds is listed in Table 1.

Introduce the eluate in-line into the electrospray ion source (ESI) of the tandem mass analyzer, extract positively charged parent mass [M + H]⁺ and determine strigolactones in multiple reaction monitoring mode (MRM) using characteristic quantifying and confirming fragment of individual analytes. Settings of the mass spectrometer are: capillary voltage 1.2 kV; ion source/desolvation temperature 120/550 °C; desolvation/cone gas flow 1000/150 L/h; collision gas flow 0.15 mL/min. Optimized conditions for each compound, such as cone voltage, collision energy and dwell time, are listed in Table 1.

To calibrate, inject a standard mixture with varying concentration levels of authentic analytes and quantify strigolactones using internal standardization method. A calibration curve is constructed by serial dilutions of authentic standards in the range from 5 × 10^–7 to 2.5 × 10⁻¹⁰ M, containing a fixed amount of internal standard. Make a plot of known concentration of authentic standards to the ratio of peak areas for analyte vs. peak area for internal standard (response, Table 1). The internal standardization is a quantitative method of analysis, used to compensate for analyte losses during the sample preparation procedure and loss of signal during the ESI ionization process. The analyte is quantified based on solid signal of the appropriate reference standard, added to sample at the known concentration. Individual strigolactones are coupled to synthetic structural analogues with similar chromatographic behavior, listed in Table 1.

Striga germination bioassay. Seeds of parasitic Orobanchaceae will only germinate if they perceive the presence of SLs, which in the field means they are within the host rhizosphere and thus after germination they have a better chance to rapidly attach to the host root. A simple bioassay based with variation on that developed by Mangnus et al. (1992) for studying SLs has been used extensively since the discovery of their role as germination stimulants. It is based on in vitro reproduction of the steps necessary for parasitic seeds to germinate. The conditioning requirement, as described previously, is met by keeping seeds in warm moist conditions (e.g., wet filter paper in a petri dish at 25 °C) for several days. Conditioned seeds are then exposed to the stimulant at proper concentration (usually at ppm or ppb levels) to induce germination which generally occurs within a day or two. Several observations can then be performed, e.g., percentage of germination, shape and length of the germination tubes and seed viability. More recently, high-throughput germination bioassays have been developed based on a standardized 96-well plate test coupled with spectrophotometric reading of tetrazolium salt (MTT) reduction (Pouvreau et al. 2013). These bioassays can be useful for guiding the purification steps for the identification of novel stimulants, testing dose–response effectiveness of SLs or their derivatives and analogues, evaluating SL selectivity/specificity in parasitic species/strains, assaying germination inhibitors or studying the physiology of the first stages of the parasitism. The complete Striga germination bioassay procedure is shown in Fig. 2.

To surface sterilize and condition S. hermonthica seeds, place the desired amount in a 15 mL Falcon tube. Add 10 mL of 2% sodium hypochlorite with 0.02% Tween-20 and invert the tube for five minutes. After removing the bleach, wash with 10 mL sterile MilliQ water 4–8 times until the water becomes clear. Moisten autoclaved 90 mm glass fiber filter papers (GF/A) set in a sterile petri dish with 3 mL sterile MilliQ water, then spread up to 100 mg of Striga seeds on each filter paper. Seal the plates with parafilm and wrap them with aluminum foil, then incubate them at 30 °C for 5–12 days.

To test germination, prepare 0.1% agarose containing 50–100 times diluted exudate or containing another compound of interest (e.g., 1 ppm GR24) in 2 mL Eppendorf tubes. It is not necessary to use GR24 as positive control. One may use any strigolactone that is available to them, and can choose a different concentration as well. Add preconditioned Striga seeds to the solution, and shake the tubes. Transfer 650 µL of Striga in solution into each well of a 12-well plate. Seal the plates with parafilm and wrap them with aluminum foil, then incubate them at 30 °C for two days.

Take a picture of each well using a binocular microscope with 10× magnification, or using a camera with a macro lens with 10× magnification at a set distance to have exactly one well in each picture. Regardless of with what instrument the picture is taken, a light background is necessary. A binocular microscope is the most obvious choice, but other options may be possible. Also, a relatively specific white balance (causing a bluish tint) is needed to obtain the contrast necessary for the automated image analysis.

Striga germination assay automated image analysis. Germination counts from the images can be automated using ImageJ (v1.53, available for free download from the United States National Institutes of Health at https://imagej.nih.gov/ij/) and Rstudio (v.1.1.453, a free, open source integrated development environment for R available for download at https://www.rstudio.com/products/rstudio/download/) software. A computer to run the software should have a minimum of 64-bit operating system, Windows 7 or greater, Intel^® Core™ i5 and 8 GB RAM. Save the pictures to be analyzed in one folder. Save the pictures as TIFF files, or use another format that is supported by ImageJ. Open ImageJ, and go to Analyze > Set Measurements. Select Area, Shape descriptors, Perimeter, Feret’s diameter, and Display label. Next, go to Process > Batch > Macro. Select the input folder containing the photos to be analyzed. Select the output folder, which is where altered/analyzed pictures will be created (if the input and output folder are the same, the original picture will be written over). Batch process the folder with the pictures using the supplied ImageJ macro. Save the result table as an Excel file. If one is only interested in the germination rates, and not in the rate of haustorium formation, the macro can be slightly adjusted so that the results table will not be formed (by removing ‘display’ when using ‘analyze particles’ to count the number of germinated Striga seed. Open the result file in Rstudio, and add necessary columns to the data (e.g. treatment and concentration). Run the provided R script to obtain the results in tables, graphs, and statistical tests.

• ImageJ Macro

• *Altering file names*

• a=getTitle();

• dotIndex=indexOf(a, ".");

• b=substring(a, 0, dotIndex)

• rename(a+"_total_seeds");

• run("Duplicate…", " ");

• rename(a+"_Germinated");

• selectWindow(a +"_total_seeds")

• *Thresholding the Hue slice (these are settings used for a specific combination of camera settings, and may require adjustments for other users)*

• run("HSB Stack");

• run("Convert Stack to Images");

• selectWindow("Saturation");

• close();

• selectWindow("Brightness");

• close();

• selectWindow("Hue");

• setThreshold(0, 120);

• run("Make Binary");

• *Selecting the area of interest within the photo (needs to be consistent within an experiment to give accurate results, and may need readjusting between experiments)*

• makeOval(100, 100, 2750, 2750);

• run("Clear Outside");

• run("Despeckle");

• rename(b+"_Seeds");

• *Set the scale of the picture in order to convert pixels into length in mm and count the number of seeds*

• run("Set Scale…", "distance=2822 known=21.6 pixel=1 unit=mm");

• run("Analyze Particles…", "size=0.012–0.2 circularity=0.2–1 display include add");

• close();

• *Selecting the area of interest within the Blue slice, which should be the same as for the Hue slice (needs to be consistent within an experiment to give accurate results, and may need readjusting between experiments)*

• run("RGB Stack");

• run("Convert Stack to Images");

• selectWindow("Red");

• close();

• selectWindow("Green");

• close();

• selectWindow("Blue");

• makeOval(100, 100, 2750, 2750);

• run("Clear Outside");

• *Thresholding the Blue slice (these are settings used for a specific combination of camera settings, and may require adjustments for other users)*

• setThreshold(0, 100);

• run("Convert to Mask");

• run("Make Binary");

• run("Despeckle");

• rename(b+"_Germinated");

• *Set the scale of the picture in order to convert pixels into length in mm (same as for hue slice) and count the number of germinated seeds*

• run("Set Scale…", "distance=2822 known=21.6 pixel=1 unit=mm");

• run("Analyze Particles…", "size=0.052–0.5 circularity=0.05–1 display include add");

• close();

• R Script

• *General notes about naming before starting analysis*

• - This analysis only works when pic_info file contains 2 columns (picture name/number & treatment): Beware of how you name the treatments!

• -> for more indepth comparisons, you may want to add more columns to your pic_info file (e.g. concentration, compound, sample type, rep, …), but you'll need to adjust the script accordingly

• *Set working directory & open imageJ results file (original csv file), and pic info file (excel converted to txt file)*

• setwd("C:/Users ")

• Results <- read.csv("Results.csv", sep=";")

• Pic_info <- read.table("Pic_info.txt", header=T, colClasses=‘character’)

• *Open the needed libraries*

• library(tidyr)

• library(reshape2)

• library(ggplot2)

• library(car)

• *Merge ImageJ output with Pic_info file*

• Results <- separate(data=Results, col=Label, into=c("del", "Image", "Type"), sep="_")

• Results <- merge(Results, Pic_info, by="Image", all.y=TRUE)

• *OPTION A: Remove extra columns (if second-forth column are the separated label)*

• Results <- Results[c(1,15,3,4,5,7,11,12,14)]

• *OPTION B: Remove extra columns (if second column is nameless with numbers, and the third-fifth column are the separated label)*

• Results <- Results[c(1,16,4,5,6,8,12,13,15)]

• *Create subset tables for seeds and for germinated counts*

• Seed_Results <- subset(Results, Type=="Seeds")

• Seed_Results <- Seed_Results[c(1,2,4)]

• Seed_Results["Seed_Count"] <- as.integer(ifelse(Seed_Results$Area < 0.1, "1", ifelse(0.1 <=Seed_Results$Area & Seed_Results$Area < 0.15, "2", ifelse(0.15 <=Seed_Results$Area & Seed_Results$Area < 0.2, "3", ifelse(Seed_Results$Area > =0.2, "4", NA)))))

• Germinated_Results <- subset(Results, Type=="Germinated")

• Germinated_Results["FM"] <- Germinated_Results$Feret / Germinated_Results$MinFeret

• Germinated_Results["PFS"] <- Germinated_Results$Perim. / (Germinated_Results$Feret / Germinated_Results$Solidity)

• Germinated_Results["PAR"] <- Germinated_Results$Perim. * Germinated_Results$AR

• Germinated_Results["PFS.FM"] <- Germinated_Results$PFS/Germinated_Results$FM

• Germinated_Results["Haust_Count"] <- as.integer(Germinated_Results$Feret < 0.6 & Germinated_Results$FM < 1.8 & Germinated_Results$PFS > =2.3 | Germinated_Results$Feret > =0.6 & Germinated_Results$FM < 2 & Germinated_Results$PAR < 4.8 & Germinated_Results$PFS > 2)

• Germinated_Results["Int_Count"] <- as.integer(Germinated_Results$Feret > =0.6 & Germinated_Results$PFS > =2 & Germinated_Results$FM < 2 & Germinated_Results$PAR > =4.8 | Germinated_Results$Feret > =0.6 & Germinated_Results$PFS > =2 & Germinated_Results$FM > 2 & Germinated_Results$PFS.FM > =0.9 & Germinated_Results$PAR > =4.8)

• Germinated_Results["Rad_Count"] <- as.integer(Germinated_Results$Feret < 0.6 & Germinated_Results$FM > =1.8 & Germinated_Results$PFS > =2.3 | Germinated_Results$Feret < 0.6 & Germinated_Results$PFS < 2.3 | Germinated_Results$Feret > =0.6 & Germinated_Results$PFS < 2 | Germinated_Results$Feret > =0.6 & Germinated_Results$PFS > =2 & Germinated_Results$FM > =2 & Germinated_Results$PFS.FM < 0.9 | Germinated_Results$Feret > =0.6 & Germinated_Results$PFS > =2 & Germinated_Results$FM > =2 & Germinated_Results$PFS.FM > =0.9 & Germinated_Results$PAR < 4.8)

• *Summarize all results per picture*

• Summary_per_Pic <- aggregate(Seed_Results[, 4], by=list(Treatment=Seed_Results$Treatment, Image=Seed_Results$Image), sum)

• Germinated_per_Pic <- aggregate(Germinated_Results[, c(14:16)], by=list(Treatment=Germinated_Results$Treatment, Image=Germinated_Results$Image), sum)

• Summary_per_Pic <- merge(Summary_per_Pic, Germinated_per_Pic, all.x=TRUE)

• Summary_per_Pic["Germinated_Count"] <- Summary_per_Pic$Haust_Count+Summary_per_Pic$Int_Count+Summary_per_Pic$Rad_Count

• Summary_per_Pic[is.na(Summary_per_Pic)] <- 0

• names(Summary_per_Pic) <- c("Treatment", "Image", "Seed.Count", "Haust_Count", "Int_Count", "Rad_Count", "Germinated.Count")

• Summary_per_Pic["Germination.rate"] <- Summary_per_Pic$Germinated.Count/Summary_per_Pic$Seed.Count*100

• Summary_per_Pic["Haust.Percentage"] <- Summary_per_Pic$Haust_Count/Summary_per_Pic$Germinated.Count*100

• Summary_per_Pic["Int.Percentage"] <- Summary_per_Pic$Int_Count/Summary_per_Pic$Germinated.Count*100

• Summary_per_Pic["Rad.Percentage"] <- Summary_per_Pic$Rad_Count/Summary_per_Pic$Germinated.Count*100

• Summary_per_Pic <- Summary_per_Pic[c(2,1,3,7,8,4,5,6,9,10,11)]

• Germination_per_Treatment <- do.call(data.frame, aggregate(Summary_per_Pic[, 5], by=list(Summary_per_Pic$Treatment), function(Summary_per_Pic) c(mean=mean(Summary_per_Pic), sd=sd(Summary_per_Pic))))

• names(Germination_per_Treatment) <- c("Treatment", "Germination.rate", "Stdev")

• Structures_per_Treatment <- do.call(data.frame, aggregate(Summary_per_Pic[, c(9:11)], by=list(Treatment=Summary_per_Pic$Treatment), function(Summary_per_Pic) c(mean=mean(Summary_per_Pic, na.rm=TRUE), sd=sd(Summary_per_Pic, na.rm=TRUE))))

• names(Structures_per_Treatment) <- c("Treatment", "Percentage_Haust", "Stdev_Haust", "Percentage_Int", "Stdev_Int", "Percentage_Rad", "Stdev_Rad")

• Structures_per_Treatment <- reshape(Structures_per_Treatment, idvar=c("Treatment"), varying=2:7, sep="_", direction="long")

• names(Structures_per_Treatment) <- c("Treatment", "Structure.Type", "Structure.Percentage", "Stdev.of.Percentage")

• Structures_per_Treatment <- Structures_per_Treatment[order(Structures_per_Treatment$Treatment, Structures_per_Treatment$Structure.Type),]

• rownames(Structures_per_Treatment) <- NULL

• *Create germination and structure type graphs (of all treatments!)*

• Germination_Graph <- ggplot(data=Germination_per_Treatment, aes(x=Germination_per_Treatment$Treatment, y=Germination_per_Treatment$Germination.rate))+geom_bar(position="dodge", stat="identity", fill="lightskyblue")+geom_errorbar(aes(ymin=Germination_per_Treatment$Germination.rate - Germination_per_Treatment$Stdev, ymax=Germination_per_Treatment$Germination.rate+Germination_per_Treatment$Stdev), width=0.2, position=position_dodge(0.9))+theme(panel.grid.major.x=element_blank(), axis.text.x=element_text(angle=-45, hjust=0, size=8))+xlab("Treatment")+ylab("Average germination (in %)")+scale_y_continuous(breaks=seq(0,100,10), limits=c(0,100))

• print(Germination_Graph)

• Structure_Graph <- ggplot(data=Structures_per_Treatment, aes(y=Structures_per_Treatment$Structure.Percentage, x=Structures_per_Treatment$Treatment, fill=Structures_per_Treatment$Structure.Type))+geom_bar(stat="identity")+scale_y_continuous(breaks=seq(0,100,10))+theme(axis.text.x=element_text(angle=-20, hjust=0, size=8))+xlab("Treatment")+ylab("Average structure frequency (in %)")+labs(fill="Structure Type")+scale_fill_brewer(palette="Blues")

• print(Structure_Graph)

Conclusion

Strigolactone research in parasitic plants has surged since discovery of their roles as signals for AM fungi and plant hormones. Technological advances have led to high-throughput bioassays that expedite SL purification, structure determination, and analysis. Metabolomics approaches have enhanced our understanding of the involvement of stimulants and inhibitors in host/parasite interactions. Given their complex biochemistry and crucial role as germination stimulants, it is likely that some Striga resistance gained through mutagenesis may involve SLs. The effectiveness of reduced Striga germination stimulant activity as a resistance mechanism through altered SL profiles in host root exudates must be balanced with potential fitness costs such that a net productivity gain is realized in Striga prone environments. The methods described in this chapter and others in this book can help determine the underlying mechanisms of Striga resistance gained through mutation breeding. Informed deployment in combination with good management can avoid the emergence of virulence in parasite populations, a risk of reliance on single-gene based resistance.