A Divergent Clade KAI2 Protein in the Root Parasitic Plant Orobanche minor Is a Highly Sensitive Strigolactone Receptor and Is Involved in the Perception of Sesquiterpene Lactones

Abstract

Strigolactones (SLs) were initially discovered as germination inducers for root parasitic plants. In 2015, three groups independently reported the characterization of the SL receptor in the root parasitic plant Striga hermonthica, which causes significant damage to crop production, particularly in sub-Saharan Africa. The characterized receptors belong to HYPOSENSITIVE TO LIGHT/KARRIKIN INSENSITIVE2 (HTL/KAI2), which is a member of the α/β-hydrolase protein superfamily. In non-parasitic plants, HTL/KAI2 perceives the smoke-derived germination inducer karrikin and a yet-unidentified endogenous ligand. However, root parasitic plants evolved a specific clade of HTL/KAI2 that has diverged from the KAI2 clade of non-parasitic plants. The S. hermonthica SL receptors are included in this specific clade, which is called KAI2 divergent (KAI2d). Orobanche minor is an obligate root holoparasitic plant that grows completely dependent on the host for water and nutrients because of a lack of photosynthetic ability. Previous phylogenetic analysis of KAI2 proteins in O. minor has demonstrated the presence of at least five KAI2d clade genes. Here, we report that KAI2d3 and KAI2d4 in O. minor have the ability to act as the SL receptors. They directly interact with SLs in vitro, and when expressed in Arabidopsis, they rescue thermo-inhibited germination in response to the synthetic SL analog GR24. In particular, KAI2d3 showed high sensitivity to GR24 when expressed in Arabidopsis, suggesting that this receptor enables highly sensitive SL recognition in O. minor. Furthermore, we provide evidence that these KAI2d receptors are involved in the perception of sesquiterpene lactones, non-strigolactone-type germination inducers.

Introduction

Root parasitic plants from the Orobanchaceae family cause significant damage to crop production by parasitizing the host plants, including important crops such as rice, maize and sorghum (Parker 2017). They have a unique germination system that senses host root-derived strigolactone (SL) molecules to germinate only in the presence of the nearby host. Orobanchaceae can be divided into three groups according to the degree of dependence on the host (Mutuku et al. 2021). Facultative parasites such as Phtheirospermum japonicum can survive without parasitizing the host; thus, the germination of P. japonicum does not require SL recognition. Very recently, it was reported that the naturally occurring SL molecule strigol induces P. japonicum germination only under nitrate-limited conditions (Ogawa and Shirasu 2022). However, under normal growth conditions, it can germinate in the absence of SLs. On the other hand, obligate hemiparasites such as Striga hermonthica have the ability to photosynthesize, but they cannot obtain sufficient energy to survive from photosynthesis alone (Graves et al. 1989). Moreover, Orobanche species are holoparasites, which do not photosynthesize and whose growth is completely dependent on the hosts for water and nutrient (Mutuku et al. 2021). Notably, the seeds of the obligate parasitic plants in both Striga and Orobanche species are highly sensitive to exogenously applied SLs; they germinate in response to pico- to nano-molar range concentrations of certain SLs (Xie et al. 2007, 2008).

Although SLs have activity as seed germination stimulants for root parasitic plants, SLs are now well recognized to be the endogenous hormones that regulate shoot branching and diverse aspects of plant growth (Gomez-Roldan et al. 2008, Umehara et al. 2008, Seto et al. 2012, Brewer et al. 2013). Moreover, SLs also function as the rhizosphere signal to establish a symbiotic relationship with arbuscular mycorrhizal fungi that supply inorganic phosphate to the host plant (Akiyama et al. 2005). When SLs act as plant hormones, they are perceived by an α/β-hydrolase fold receptor, DWARF14 (D14) (Seto et al. 2019, Mashiguchi et al. 2021). D14 has a closely related homolog that is called HYPOSENSITIVE TO LIGHT/KARRIKIN INSENSITIVE2 (HTL/KAI2). HTL/KAI2 functions in the perception step of the smoke-derived germination inducer, karrikin (KAR) (Waters et al. 2012). Moreover, the presence of the endogenous ligand for HTL/KAI2 has been proposed, yet its chemical identity has not been discovered (Conn and Nelson 2015). In 2015, three groups independently reported that the proteins in the HTL/KAI2 family function as the SL receptors in a root parasitic plant, S. hermonthica, for their SL-dependent germination (Conn et al. 2015, Toh et al. 2015, Tsuchiya et al. 2015). Interestingly, there are at least 11 HTL/KAI2 genes in the S. hermonthica genome, whereas seed plants usually have one or two copies of HTL/KAI2. In addition, other parasitic plants also possess many numbers of HTL/KAI2, and phylogenetic analyses revealed that most of them formed a specific clade that was diverged from the HTL/KAI2 family in the non-parasitic seed plant species (Conn et al. 2015). Conn et al. called the genes belonging to this specific clade KAI2 divergent (KAI2d). Moreover, most genes in the KAI2d clade in a root parasitic plant, S. hermonthica, were revealed to function as the receptors for SL, but not KAR. Interestingly, one of these KAI2d receptors in S. hermonthica (ShHTL7) showed extremely high sensitivity to SL when expressed in Arabidopsis; thermo-inhibition of the germination of the transgenic Arabidopsis seed expressing ShHTL7 was recovered by quite a low concentration of SL such as pico-molar range (Toh et al. 2015). Therefore, ShHTL7 would be involved in the highly sensitive SL perception by S. hermonthica. More recently, in another root parasitic plant, Phelipanche ramosa, one of the divergent clade KAI2d proteins, PrKAI2d3, was reported to act as an SL receptor with high sensitivity to SL (de Saint Germain et al. 2021).

In a root holoparasitic plant, such as O. minor, the presence of genes in the KAI2d clade was demonstrated (Conn et al. 2015); however, their biochemical function has not yet been investigated. Intriguingly, the KAI2d genes in the Orobanche species are classified into a different sub-clade from those in the Striga species. Therefore, functional analysis of this sub-clade KAI2d genes would be an important subject to understanding the SL perception mechanism by root parasitic plants. Since Orobanche species also show high sensitivity for SL perception (Kim et al. 2010), identification of the SL receptors that are responsible for the highly sensitive SL recognition is of particular interest. In addition, it was reported that the germination of O. cumana and O. minor is induced not only by SLs but also by sesquiterpene lactones (STLs), such as dehydrocostus lactone (Perez de Luque et al. 2000, Raupp and Spring 2013, Ueno et al. 2014). If KAI2d in Orobanche species act as SL receptors, it would be very interesting to see if these receptors also function as the receptors for STLs. The possibility of STLs binding to KAI2 was discussed in a previous commentary paper (Rahimi and Bouwmeester 2021).

In this paper, we analyzed the function of KAI2d proteins in O. minor and found that some of them act as SL receptors. In particular, KAI2d3 from O. minor showed high sensitivity to the synthetic SL analog rac-GR24 (Fig. 1) when expressed in Arabidopsis. We also discovered that STLs can directly, albeit weakly, interact with some of the OmKAI2d receptors. Moreover, the thermo-inhibited seed germination of the transgenic Arabidopsis expressing OmKAI2d3/OmKAI2d4 was weakly rescued by some STLs. We solved the crystal structure of one of the O. minor SL receptors, OmKAI2d4. The OmKAI2d4 structure demonstrates that the protein has a large ligand-binding pocket, possibly allowing it to recognize a broad range of SL molecules.

Fig. 1.

Chemical structure of naturally existing SLs and those synthetic analogs.

Results

In vitro biochemical function of the Orobanche minor KAI2d proteins

To evaluate the biochemical properties of the O. minor KAI2d proteins as SL receptors, we heterologously expressed each KAI2d protein in Escherichia coli. Previous phylogenetic analysis has revealed that there are at least five genes in the divergent clade of KAI2 in O. minor, namely, OmKAI2d1 to OmKAI2d5 (Conn et al. 2015). Using the recombinant protein of each OmKAI2d, we performed a differential scanning fluorimetry (DSF) assay, which can evaluate the protein–ligand interaction by detecting the chemically inducible melting temperature shift of the protein (Hamiaux et al. 2012). We found that the melting temperature of OmKAI2d3 or OmKAI2d4 was lowered in the presence of rac-GR24, while that of OmKAI2d1, OmKAI2d2 or OmKAI2d5 was not affected by rac-GR24 (Fig. 2). A previous report on the analysis of the SL receptor D14 has suggested that the SL-inducible lower melting temperature shift of D14 seems to reflect the transition of the D14 to the active signaling state (Seto et al. 2019). Thus, the above-mentioned results suggest that OmKAI2d3 and OmKAI2d4 function as the SL receptors. To evaluate the ligand specificity of OmKAI2d3 and OmKAI2d4, we performed a DSF assay using various naturally occurring or synthetic SLs. Interestingly, in the presence of almost all the tested SLs, we detected a lower melting temperature shift (Supplementary Fig. S1), supporting the fact that O. minor germination is induced by structurally diverse SL molecules (Kim et al. 2010).

Fig. 2.

DSF analysis of each OmKAI2d in the presence of rac-GR24. Melting temperature curves of OmKAI2d, which was incubated with the indicated concentration of rac-GR24, are shown. Data are the means (n = 3).

In vivo functional analysis of O. minor HTL/KAI2 using Arabidopsis

To further evaluate the biological function of each KAI2d in O. minor as a potential SL receptor, we prepared transgenic Arabidopsis plants expressing each OmKAI2d in the kai2 knockout mutant background. The germination of Arabidopsis seeds is inhibited at a high temperature of approximately >30°C. However, SL application was reported to rescue this thermo-inhibition in a manner dependent on the KAI2 pathway (Toh et al. 2012, 2015). Therefore, the Arabidopsis kai2 mutant is insensitive to the exogenously applied SL, while the heterologous expression of KAI2d from S. hermonthica complemented the thermo-inhibited germination defect of the kai2 mutant in an SL-dependent manner. The above-mentioned report on the SL receptor characterization in S. hermonthica was demonstrated using this assay. We used the same system to evaluate the function of OmKAI2d. We also evaluated the function of OmKAI2c to understand whether its function is distinct from that of OmKAI2d or not. We generated transgenic Arabidopsis, in which each OmKAI2d or OmKAI2c was expressed under the control of a KAI2 native promotor in the kai2 knockout mutant background. As a result, in agreement with the biochemical analysis, the thermo-inhibition of germination in the transgenic Arabidopsis expressing OmKAI2d3 or OmKAI2d4 was rescued by rac-GR24 treatment. In particular, the germination of OmKAI2d3-expressing plants was highly sensitive to the exogenously applied rac-GR24 with a moderate rescue at a low concentration of rac-GR24 such as 500 pM to 50 nM (Fig. 3A). We also generated ShHTL7-expressing transgenic plants. In a previous study, the cauliflower mosaic virus 35S promoter was used for ShHTL7 expression, whereas we used the KAI2 native promoter to reduce differences in the transgene expression levels between different lines. As a result, under our germination assay conditions, the sensitivity of the ShHTL7-expressing plants was not as high as the reported data (Toh et al. 2015). A previous report suggested that the native KAI2 expression in Arabidopsis is upregulated during seed germination at normal temperature. However, its expression remains low in the presence of a gibberellin biosynthetic inhibitor, paclobutrazol (Bunsick et al. 2020). This result suggests that under a high-temperature condition, which inhibits GA biosynthesis (Toh et al. 2008), it is likely that the KAI2 expression is downregulated. To evaluate this, we examined the KAI2 expression levels under normal (22°C) and high (31°C) temperature conditions by quantitative RT-PCR (qRT-PCR) analysis and found that the KAI2 transcript level is significantly lower at 31°C compared with 22°C (Supplementary Fig. S2). Therefore, this might be the reason for lower sensitivity under our experimental conditions. We then compared the sensitivity of OmKAI2d3 and ShHTL7 using each transgenic line and found that OmKAI2d3-expressing plants showed slightly weaker sensitivity to rac-GR24 compared with the ShHTL7-expressing plants (Fig. 3B).

Fig. 3.

(A) Thermo-inhibition germination assay results of transgenic Arabidopsis expressing each OmKAI2d. Thermo-inhibited germination of Arabidopsis was monitored in the presence of rac-GR24 at indicated concentrations. Data are the means ± SD (n = 3). * = significantly increased compared with the control (Student’s t-test, P < 0.05). (B) Comparison of the sensitivity of OmKAI2d3 and ShHTL7 to rac-GR24 using the transgenic Arabidopsis expressing each one of them. Thermo-inhibited germination of Arabidopsis was monitored in the presence of rac-GR24 at indicated concentrations. Data are the means ± SD (n = 3). * = significantly induced compared with the control (Student’s t-test, P < 0.05).

We also performed a yeast two-hybrid (Y2H) assay using OmKAI2d3/OmKAI2d4 and an Arabidopsis signaling component, SUPPRESSOR OF MAX2 1 (SMAX1). In Arabidopsis, SMAX1 functions as the signaling repressor in the KAI2 pathway, and an F-box protein, MAX2, is involved in the proteasome-dependent degradation of SMAX1 (Waters et al. 2014). We found that the interaction between OmKAI2d3/OmKAI2d4 and SMAX1 was induced in the presence of rac-GR24 (Supplementary Fig. S3). Moreover, the interaction between OmKAI2d3 and SMAX1 was induced at a lower concentration of rac-GR24 compared with the case with OmKAI2d4, supporting that OmKAI2d3 showed higher sensitivity in the germination assay compared with OmKAI2d4.

These results demonstrate that both OmKAI2d3 and OmKAI2d4 act as SL receptors, with OmKAI2d3 likely involved in the highly sensitive SL perception by O. minor seeds.

Crystal structure of OmKAI2d4

To obtain more information about the possible reasons for OmKAI2d4’s low substrate specificity, we solved its crystal structure at a resolution of 2.3 Å (Table 1). As expected, OmKAI2d4 folded into an α/β-hydrolase architecture (Fig. 4A). A Distance-matrix ALIgnment search (Holm 2020) revealed ShHTL7 as the closest homolog in the Protein Data Bank (PDB; accession code 5Z7Y), with a root mean square deviation of 0.9 Å over 267 amino acids. SL receptors feature a four-helix lid that covers a substrate-binding pocket with a serine/histidine/aspartate catalytic triad located at the bottom (Hamiaux et al. 2012). We used Computed Atlas of Surface Topography of proteins (CASTp) (Tian et al. 2018) to analyze the substrate-binding pocket in OmKAI2d4 and found that it had a volume of 861 ų (Fig. 4A). In comparison, the pocket volume of ShHTL7 has previously been reported to be 1,111 ų (Xu et al. 2018), and when we analyzed ShHTL7 (PDB code 5Z7Y) with CASTp, we obtained a value of 1,158 ų. However, we noticed that a tunnel-shaped sub-pocket was connected to the main pocket through a narrow conduit around amino acid I193 (Supplementary Fig. S4A). Because this tunnel was located below the protein’s active site and did not correspond to the shape of any known SL molecule, we decided to limit the volume determination to the main ligand-binding pocket. We therefore used a probe radius of 1.5 Å instead of the 1.4 Å CASTp default value in our analyses of OmKAI2d4 and ShHTL7. In doing so, we obtained a volume of 726 ų for the pocket in ShHTL7 (Supplementary Fig. S4B). We further compared the biophysical properties of the binding pockets in OmKAI2d4 and ShHTL7. Except for the active site, the pocket in OmKAI2d4 is entirely hydrophobic (Fig. 4C). In comparison, ShHTL7’s pocket contains six hydrophobic and three polar residues (Fig. 4D). When we analyzed the secondary structure elements of OmKAI2d4 and ShHTL7 based on the Define Secondary Structure of Proteins annotations (Kabsch and Sander 1983, Touw et al. 2015), we noticed that the first alpha helix in the lid domain (αT1) is shorter in ShHTL7, terminating at amino acid S145 (which corresponds to A146 in OmKAI2d4) and continuing as a loop (Fig. 4E). We wondered if the longer loop between alpha helices αT1 and αT2 could potentially cause a more flexible lid structure in ShHTL7 compared to OmKAI2d4. We thus checked whether the atomic B-factors in helices αT1 and αT2 were elevated compared to the average protein B-factor. While we saw marginally elevated B-factors in in helices αT1 and αT2 in OmKAI2d4 (Fig. 4F), the difference was greater in ShHTL7 (Fig. 4G), suggesting that OmKAI2d4’s first two helices of the lid domain are more rigid compared to ShHTL7. Taken together, OmKAI2d4’s crystal structure revealed a substrate-binding pocket with a large volume, highly hydrophobic properties, and rather rigid structural elements forming its entrance.

Table 1.

OmKAI2d4 data collection and refinement statistics

Wavelength (Å)	1.000
Resolution range (Å)	46.27–2.3 (2.382–2.3)
Space group	P 43 21 2
Unit cell	97.4478 97.4478 147.668 90 90 90
Total reflections	697,558 (65,669)
Unique reflections	32,317 (3,165)
Multiplicity	21.6 (20.7)
Completeness (%)	99.96 (100.00)
Mean I/sigma(I)	8.03 (2.55)
Wilson B-factor	20.69
R-merge	0.4397 (1.561)
R-meas	0.4502 (1.6)
R-pim	0.09582 (0.3487)
CC1/2	0.989 (0.744)
CC*	0.997 (0.924)
Reflections used in refinement	32,317 (3,165)
Reflections used for R-free	1,688 (163)
R-work	0.1761 (0.2150)
R-free	0.2320 (0.2765)
CC (work)	0.961 (0.883)
CC (free)	0.928 (0.804)
Number of non-hydrogen atoms	4,645
Macromolecules	4,236
Ligands	2
Solvent	407
Protein residues	535
RMS (bonds)	0.004
RMS (angles)	0.65
Ramachandran favored (%)	97.92
Ramachandran allowed (%)	2.08
Ramachandran outliers (%)	0.00
Rotamer outliers (%)	1.05
Clashscore	2.96
Average B-factor	22.95
Macromolecules	22.50
Ligands	39.12
Solvent	27.49

Statistics for the highest-resolution shell are shown in parentheses. CC_1/2: half-dataset correlation coefficient. CC*: estimate of correlation between true and observed intensities.

Fig. 4.

Structural comparison between OmKAI2d4 and ShHTL7. (A, B) Comparison of the size of the ligand-binding pocket between OmKAI2d4 and ShHTL7. (C, D) Comparison of the amino acids on the ligand-binding pocket surface of OmKAI2d4 and ShHTL7. (E) Comparison of the length of the αT1 helix between OmKAI2d4 and ShHTL7. (F, G) The temperature factors of the αT1 helix region of OmKAI2d4 and ShHTL7.

STL-dependent germination of O. minor

Among Orobanche species, the germination of O. cumana and O. minor was reported to be induced not only by SLs but also by STLs, such as dehydrocostus lactone (Perez de Luque et al. 2000, Joel et al. 2011, Raupp and Spring 2013, Ueno et al. 2014). We tested the germination-inducing activity of several commercially available STLs using O. minor seeds (Fig. 5A). We found that O. minor germination is induced by almost all tested STLs, and the most active compound, costunolide, induced the germination even at the 0.1 nM concentration (Fig. 5B). We next evaluated the direct interaction of OmKAI2d3/OmKAI2d4 with STLs using isothermal titration calorimetry (ITC) experiment. Interestingly, we saw moderate binding between OmKAI2d3/OmKAI2d4 and some certain STLs at relatively low K_D values (Supplementary Fig. S5). OmKAI2d3 exhibited a 1–10-μM K_D value for binding to costunolide, parthenolide or α-santonin. Moreover, OmKAI2d4 exhibited a 2.9-μM K_D value for binding to alantolactone. We also performed a DSF assay using OmKAI2d3 and OmKAI2d4 in the presence of each STL. We found that some STLs, at a high concentration such as 500 μM, affected the melting temperature of OmKAI2d3. Particularly, the effect of costunolide was the strongest among the tested STLs, yet the effect was still much weaker compared with that of rac-GR24 (Fig. 6). In the case of OmKAI2d4, the melting temperature was very slightly affected by the tested STLs, yet there was no favorable correlation with the ITC results.

Fig. 5.

(A) Chemical structures of STLs. (B) Germination-inducing activity of STLs toward O. minor seed. Data are the means ± SD (n = 3). Control means only acetone at 0.1%. * = significant difference compared with the control (Student’s t-test, P < 0.05).

Fig. 6.

DSF analysis of each OmKAI2d in the presence of rac-GR24 or each STL. Melting temperature curves of AtD14, which was incubated with the indicated concentration of each STL, are shown.

These results suggest that OmKAId3 and OmKAI2d4 can directly interact not only with SLs but also with STLs. However, it appears that the direct interaction with STLs is much weaker than that with SLs.

To further evaluate whether STLs can transmit the biological signal via the OmKAI2d receptors, we assessed the activity of STLs using Arabidopsis transgenic plants expressing each OmKAI2d. We found that costunolide weakly induced the thermo-inhibited germination of the Arabidopsis seeds expressing OmKAI2d3 (Fig. 7A), demonstrating that OmKAId3 is involved in the perception of STLs. In addition, we found that costunolide tended to slightly induce the germination of some of the transgenic lines expressing other OmKAI2d genes. However, the concentration at which the activity was observed (100 μM) was much higher than the concentration at which rac-GR24 induces the germination (5 μM). Moreover, O. minor seed germination was induced at a much lower concentration (∼1 nM) of costunolide (Fig. 5B) (Ueno et al. 2014).

Fig. 7.

(A) Thermo-inhibition germination assay results of transgenic Arabidopsis expressing each OmKAI2d. Thermo-inhibited germination of Arabidopsis was monitored in the presence of rac-GR24 or costunolide at indicated concentrations. Data are the means ± SD (n = 3). * = significantly induced compared with the control (Student’s t-test, P < 0.05). (B) Analysis of the OmKAI2d3/OmKAI2d4–SMAX1 interaction in the presence of costunolide using the Y2H method. Yeast transformants were spotted onto the control medium [SD–Leu/–Trp (–LT)] and the selective medium [SD–Leu/–Trp/–His (–LTH) + 0.1 mM 3-AT] in the absence or presence of costunolide at indicated concentrations.

We next performed a Y2H experiment to see the STL-inducible interaction between OmKAI2d3/OmKAI2d4 and the Arabidopsis SMAX1. In agreement with the germination assay results, the interactions between OmKAI2d3 and SMAX1 were weakly induced in the presence of costunolide at a relatively high concentration, 10 μM (Fig. 7B).

These results suggest that the SL receptors in the KAI2d clade are involved in the perception step of STLs as well. However, considering that the STL function is weaker compared with SL in both in vitro and in vivo experiments, the full activation of STL function in root parasitic plant seeds may require some other factors.

Discussion

Orobanche minor is an obligate root holoparasite that exhibits highly sensitive SL-dependent germination. In the obligate root hemiparasite S. hermonthica, KAI2d family proteins were identified to be the SL receptors (Conn et al. 2015, Toh et al. 2015, Tsuchiya et al. 2015). It was revealed that O. minor also possesses KAI2d genes, whose functions, however, are not yet reported. Here, we show that among five OmKAI2d proteins, at least OmKAI2d3 and OmKAI3d4 function as SL receptors. DSF assays demonstrate that only OmKAI2d3 and OmKAI2d4 underwent SL-induced protein destabilization in the presence of rac-GR24. Moreover, OmKAI2d3 and OmKAI2d4 were able to function as SL receptors in transgenic Arabidopsis, in which each KAI2d was expressed in the kai2 knockout mutant. In particular, OmKAI2d3-expressing transgenic Arabidopsis seeds showed high sensitivity to rac-GR24, suggesting that this receptor is involved in the highly sensitive SL perception in O. minor. As for KAI2d1, KAI2d2 and KAI2d5, SL-inducible melting temperature shift was not detected in our DSF assay. Moreover, the transgenic Arabidopsis expressing each one of them was almost insensitive to rac-GR24 treatment; the germination of only one transgenic line expressing OmKAI2d1 was slightly induced by a high concentration of rac-GR24 (5 μM) (Fig. 3A). These results imply that these KAI2 proteins are not involved in the SL-dependent germination in O. minor. However, we cannot perform a genetic experiment such as gene knockout using root parasitic plant species including O. minor. Thus, we cannot conclude about the contribution of these KAI2d proteins in SL-dependent germination. Very recently, it was reported that some members of KAI2d in a facultative root parasite P. japonicum are involved in the chemoattracting root tropism phenomenon, which is induced by host-derived SLs (Ogawa et al. 2022). It would be possible that KAI2d1, KAI2d2 or KAI2d5 is involved in such a process in O. minor.

We solved the crystal structure of one of O. minor’s SL receptors, OmKAI2d4. The results demonstrate that this receptor has a large ligand-binding pocket, whose volume exceeds that of the S. hermonthica hypersensitive receptor, ShHTL7. The structural data are in agreement with the DSF assay results using various SLs, in which the melting temperature of OmKAI2d3/OmKAI2d4 was lowered in the presence of almost all tested SLs. Thus, OmKAI2d4’s enlarged substrate-binding pocket may enable O. minor to germinate in response to various types of SLs.

STLs were reported to induce the germination of O. cumana and O. minor. ITC analysis along with the DSF assay showed that OmKAI2d3 and OmKAI2d4 directly interact with some STLs. Moreover, germination assays using transgenic Arabidopsis expressing each OmKAI2d revealed that costunolide can induce the germination of these transgenic plant seeds. However, the effective concentration was much higher than that of GR24. Considering that costunolide induces the germination of O. minor seeds at a 1-nM concentration, it might not be reasonable to conclude that costunolide is directly recognized by OmKAI2d3/d4 when it induces the germination. It would be possible that STLs are metabolized into active substances that can more strongly interact with the OmKAI2d receptors in O. minor seeds, and such a metabolic pathway does not present in Arabidopsis. In the case of KAR, the direct interaction between KAR and KAI2 was reported (Guo et al. 2013, Kagiyama et al. 2013). However, the interaction is not strong, and a recent report suggested that KAR is metabolized into more active substances in vivo before the perception by KAI2 (Sepulveda et al. 2022). According to our results, which show clear, but weak, activity of STLs detected in DSF assays and transgenic Arabidopsis germination assays, it would be possible that STLs induce the germination of parasitic plants by a similar mechanism. On the other hand, we cannot rule out the possibility that some proteins other than the KAI2d family proteins are involved in STL perception in O. minor. Although further experiments would be needed to understand the complete molecular mechanism underlying the STL-dependent germination in O. minor, our results support the idea that the KAI2d receptors are involved in the perception step of non-SL-type germination inducers, STLs (Larose et al. 2022). In addition to the germination stimulating function of STLs, it has also been reported that the endogenous STLs play a role in controlling plant development such as hypocotyl elongation, particularly in sunflowers (Spring and Hager 1982). Moreover, a possible interaction between sunflower HTL/KAI2 and 8-epixanthatin, an endogenous STL in sunflower, was suggested by molecular docking (Rahimi and Bouwmeester 2021). Therefore, it would be possible that there is a structural similarity between STLs and the as-yet-unidentified endogenous ligand of non-parasitic plants HTL/KAI2.

Here, we used a combination of biochemical approaches and cross-species complementation assays to identify SL receptors in O. minor. One of them, OmKAI2d3, showed high sensitivity when expressed in Arabidopsis, suggesting that at least this receptor is involved in the highly sensitive SL perception by O. minor. In addition, we show high SL promiscuity for both OmKAI2d3 and OmKAI2d4, suggesting that these proteins contribute to O. minor’s broad host range recognition. These results pave the way for developing chemical tools that can effectively control O. minor germination. Although highly sensitive SL receptors in other root parasitic plant species have been identified, the mechanism that determines their sensitivity is not yet fully understood. For ShHTL7, it was suggested that the high affinity of ShHTL7 to MAX2 seems to be responsible for high sensitivity (Wang et al. 2021). A detailed comparison of the amino acid sequence between the high- and low-sensitivity receptors would provide clues to understanding this important remaining issue.

Materials and Methods

Plant materials

We used Arabidopsis ecotype Col-0 and the kai2-4 (CSHL-GT6185) mutant (Umehara et al. 2015).

Arabidopsis germination assay

Arabidopsis seeds were sterilized in a 1% sodium hypochlorite solution for 5 min and rinsed with sterile water at least five times. Sterilized seeds (30–50 seeds/well) were imbibed in sterilized distilled water containing the tested chemicals (a 1,000× dilution from the DMSO stock solution) or 0.1% DMSO as a control and incubated at 31–32°C and under a continuous light condition (10–15 μMol/m²/s). Germinated seeds (radicle emergence) were counted after 4 or 5 d.

Orobanche minor germination assay

Germination assay of O. minor was performed as described previously (Suzuki et al. 2022).

Protein expression

The coding sequences of each OmKAI2 were codon-optimized for E. coli, and they were chemically synthesized with attB sites for BP reaction by Genewiz and were cloned into pUC19 vector. Each OmKAI2d gene was amplified by PCR using the primer sets as described in Supplementary Table S1. Each fragment was introduced into the pMALHis vector that has both MBP-tag and His₆-tag. The vector was transformed into E. coli BL21 star (DE3), and the cell was precultured in a lysogeny broth (LB) medium containing 50 μg/ml ampicillin. Overnight cultures (10 ml) were added to a fresh LB medium (1 l) containing 50 μg/ml ampicillin, and it was cultured at 37°C. After the optical density at 600 nm (OD₆₀₀) reached 0.8, the cultures were cooled at 16°C for 1 h, and then 0.1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) was added. The culture was further incubated at 16°C for 20 h. The culture medium was centrifuged at 3,700×g, and the pellet was stored at −20°C until use. The pellet was resuspended and sonicated in lysis buffer [50 mM Tris buffer (pH 8.0) containing 500 mM NaCl, 10 mM 2-mercaptoethanol and 10% glycerol]. The supernatant was purified by Ni Sepharose™ 6 Fast Flow (500 μl, Cytiva, Marlborough, MA, United States) or TALON Superflow (500 μl, Cytiva). As for the Ni Sepharose purification, after washing with the washing buffer [50 mM Tris buffer (pH 8.0) containing 500 mM NaCl and 20 mM imidazole], the bound protein was eluted with elution buffer [50 mM Tris buffer (pH 8.0) containing 500 mM NaCl and 200 mM imidazole]. As for the TALON purification, after washing with the washing buffer [50 mM Tris buffer (pH 8.0) containing 500 mM NaCl and 2 mM imidazole], the bound protein was eluted with elution buffer [50 mM Tris buffer (pH 8.0) containing 500 mM NaCl and 50 mM imidazole]. The eluate was concentrated using VIVASPIN Turbo 15 (Sartorius, Göttingen, Germany), and the concentration was adjusted to 5 mg/ml. The purified protein was aliquoted to the appropriate volume, immediately frozen in liquid nitrogen and stored at −80°C until use. For high-yield protein production to crystallize OmKAI2d4, E. coli BL21 (DE3) cells were transformed with an OmKAI2d4 gene codon optimized for E. coli and cloned into a pGEX-T3 vector. Cells were grown over night in LB medium, and the next day, they were used to start a new culture in a terrific broth medium with a 1:100 dilution. Cells were then grown at 23°C until an OD₆₀₀ of 0.6 and induced with 0.1 mM IPTG at 18°C overnight. Cells were harvested and lysed using sonication, cell debris was removed by centrifugation at 75,000×g for 45 min and the supernatant was loaded onto a glutathione affinity column. The column was washed with 50 mM TRIS-HCl, 150 mM NaCl, 5% glycerol, 1 mM TCEP and final pH 7.7 until no protein flow-through was found by UV detection. Human Rhinovirus 3C (Takara, Kusatsu, Japan) protease was added onto the column overnight. The cleaved target protein was eluted using the same buffer and further purified to homogeneity by size exclusion chromatography using a GE Healthcare HiLoad 16/60 Superdex 75 column in 20 mM TRIS-HCl, 30 mM NaCl, 1 mM TCEP-HCl and final pH 7.7. Proteins were concentrated to at least 10 mg/ml and flash frozen in liquid nitrogen.

DSF assay

DSF experiments were carried out using two different machines [Light Cycler 480 (Roche, Basel, Switzerland) for Figs. 2, 6B or CFX384 system (Bio-Rad) for Supplementary Fig. S1]. Sypro Orange (Ex/Em: 350, 470/570 nm, Thermo Fisher Scientific, Waltham, MA, United States) was used as the reporter dye. When using Light Cycler 480, reaction mixtures were prepared in 96-well plates, and each reaction was carried out on a 20-μl scale in PBS buffer containing 10 μg of proteins (each OmKAI2d), 0.02 μl of Sypro Orange and SLs in acetone so that the final acetone concentration was 5%. The program parameters were excitation at 483 nm, emission at 610 nm and a ramp rate of 0.02°C/s from 25°C to 95°C. The denaturation curve was obtained using Light Cycler 480 software. When using CFX384, 10 μg of protein was heat-denatured using a linear 25–95°C gradient at a rate of 1°C/min. The denaturation curve and its derivative were obtained using the CFX manager software. For Supplementary Fig. S1, final reaction mixtures were prepared in 20 μl volumes in triplicates in 384-well white microplates. Reactions were carried out in 20 mM Tris-HCl, 30 mM NaCl, 1 mM TCEP-HCl and final pH of 7.7. A final 3× concentration of Sypro Orange was used.

Generation of transgenic Arabidopsis

Each OmKAI2 gene in the pUC19 vector was cloned into pDONR221 (Invitrogen) by BP reaction and then shuttled into the pGWB1 vector, which has the KAI2 promoter region (KAI2Pro-pGWB1) (Burger et al. 2019) by an ligation reaction clonase (Invitrogen). The coding sequence of ShHTL7 was chemically synthesized by Integrated DNA Technologies (IDT (Coralville, IA, United States) gBlock service. The gene fragment was first cloned into the pENTR/D-TOPO vector (Invitrogen) and then shuttled into KAI2Pro-pGWB1. Arabidopsis kai2-4 plants were transformed with the resulting constructs by the floral dipping method using Agrobacterium tumefaciens. T3 homozygous line seeds were used for germination assay. To examine the mRNA expression levels in each homozygous line, sterilized seeds were put on the half-strength Murashige and Skoog medium containing 1% sucrose and 0.8% agar (pH 5.7) and cultured at 22°C under fluorescent white light with a 16-h-light/8-h-dark photoperiod for 15 d. Total RNA was extracted from the whole plants using an ISOSPIN Plant RNA extraction kit (NIPPON GENE, Toyama, Japan). First-strand cDNA was synthesized using ReverTra Ace (TOYOBO), and the expression levels were examined by PCR (Supplementary Fig. S6) using the primer sets as described in Supplementary Table S1.

Yeast two-hybrid experiment

The coding sequence of each OmKAI2d was cloned into pGADT7. The coding sequence of SMAX1 was cloned into pGBKT7. The resulting constructs were co-transformed with the yeast strain Y2H Gold (Takara, Kusatsu, Japan), and the transformants were grown on SD–Trp/–Leu for 2 d at 30°C. Interactions between the two proteins were examined on SD–Trp/–Leu/–His containing a 10,000× dilution of the tested chemicals in acetone (0.01% acetone was used as a control) and 0.1 mM 3-amino-1,2,4-triazole (3-AT). The plates were kept for 7 d at 30°C.

Crystallization

OmKAI2d4 crystals were grown under the following conditions in 2 μl hanging drops using a 1:1 protein:reservoir ratio: 0.1 M sodium acetate pH 4.5, 0.8 M di-ammonium phosphate. Na malonate (1.9 M) was used as cryo-protectant. X-ray data were collected at the Advanced Light Source at Lawrence Berkeley National Laboratory (Berkeley, CA, United States) at beamline 8.2.1. X-ray data were processed with X-ray Detector Software (XDS) (Kabsch 2010). The OmKAI2d4 structure was solved by molecular replacement using chain A of PDB structure 4IH1 (AtKAI2). Five percent of the data were flagged for R-free, and initial models were build using AutoBuild (Terwilliger et al. 2008) as part of Phenix (Adams et al. 2010), manually corrected and finalized with Coot (Emsley et al. 2010), refined with phenix.refine (Afonine et al. 2012) and validated with MolProbity (Chen et al. 2010).

Isothermal titration calorimetry

ITC experiments were performed in a MicroCal iTC200 MicroCalorimeter. Ligand solutions of 1 μM were titrated into 40 μM protein solutions in 20 steps of 2 μl and in 240-s intervals. Ligand solutions were diluted from 1 mM DMSO stock solutions; therefore, the DMSO content in the protein solution was adjusted to 1%. Thermodynamic parameters were calculated using the MicroCal ITC software as part of Origin (OriginLab, Northampton, MA, United States).

Quantitative RT-PCR analysis of the Arabidopsis HTL/KAI2 expression

Sterilized seeds were imbibed in sterilized distilled water for 4 d. Then, the seeds were further incubated under continuous light conditions at two different temperatures (22°C or 30°C). Total RNA was extracted from the Arabidopsis seeds using an RNA extraction kit (ISOSPIN Plant RNA; NIPPON GENE). First-strand cDNA was synthesized from 100 ng of total RNA by using ReverTra Ace qPCR RT Master Mix (TOYOBO). Light cycler 480 (Roche) was used to perform qRT-PCR by using the KOD SYBR pPCR Mix (TOYOBO).

Supplementary Data

Supplementary data are available at PCP online.

Data Availability

The structural coordinates and diffraction data of OmKAI2d4 have been deposited in the PDB under accession code 7UOC. The data underlying this article are available in the article and in its online supplementary materials.

Funding

Ministry of Education, Culture, Sports, Science and Technology KAKENHI (Kagaku Kenkyūhi, Grants-in-Aid for Scientific Research) (19K05852 to Y.S., 20H05684 to Y.S.); Japan Science and Technology Agency, Fusion Oriented Research for Disruptive Science and Technology (JST FOREST) Program (JPMJFR211S to Y.S.); Mitsubishi Foundation (to Y.S.); Kato Memorial Bioscience Foundation (to Y.S.); National Institutes of Health (NIH) grant R35 (GM122604). J.C. is an investigator of the Howard Hughes Medical Institute.

Author Contributions

S.T. and Y.U. performed functional analysis of OmKAI2d with guidance from Y.S. M.B. performed crystallography, ITC experiment and a part of the DSF assay with guidance from J.C. S.O. performed a part of Y2H experiment with guidance from Y.S. T.S. performed a part of qRT-PCR analysis. Y.S. and M.B. designed research. All authors wrote the manuscript.

Disclosures

The authors have no conflicts of interest to declare.