The co-chaperone HOP3 participates in jasmonic acid signaling by regulating CORONATINE-INSENSITIVE 1 activity.

HOPs (HSP70-HSP90 organizing proteins) are a highly conserved family of HSP70 and HSP90 co-chaperones whose role in assisting the folding of various hormonal receptors has been extensively studied in mammals. In plants, HOPs are mainly associated with stress response, but their potential involvement in hormonal networks remains completely unexplored. In this article we describe that a member of the HOP family, HOP3, is involved in the jasmonic acid (JA) pathway and is linked to plant defense responses not only to pathogens, but also to a generalist herbivore. The JA pathway regulates responses to Botrytis cinerea infection and to Tetranychus urticae feeding; our data demonstrate that the Arabidopsis (Arabidopsis thaliana) hop3-1 mutant shows an increased susceptibility to both. The hop3-1 mutant exhibits reduced sensitivity to JA derivatives in root growth assays and downregulation of different JA-responsive genes in response to methyl jasmonate, further revealing the relevance of HOP3 in the JA pathway. Interestingly, yeast two-hybrid assays and in planta co-immunoprecipitation assays found that HOP3 interacts with COI1, suggesting that COI1 is a target of HOP3. Consistent with this observation, COI1 activity is reduced in the hop3-1 mutant. All these data strongly suggest that, specifically among HOPs, HOP3 plays a relevant role in the JA pathway by regulating COI1 activity in response to JA and, consequently, participating in defense signaling to biotic stresses.

Introduction

Jasmonic acid (JA) is a signaling molecule that is involved in different developmental processes such as fertility, root elongation, and senescence (Huang et al., 2017). Furthermore, JA also plays an essential role in plant defense, especially against chewing-biting herbivores (such as Tetranychus urticae, an acarus also known as two spotted spider mite) and necrotrophic pathogens (such as Botrytis cinerea) (Gimenez-Ibanez et al., 2016). Upon T. urticae and B. cinerea attack, the microbe- and damage-associated molecular patterns and, in the case of the herbivore, the herbivore-associated molecular patterns are perceived by the specific pattern recognition receptors of the plant. This recognition triggers a transduction cascade that leads, among other processes, to the biosynthesis of different hormones including JA, which plays a critical role in the induction of defensive molecules and in plant defense against the spider mite and the necrotrophic fungus (Zhurov et al., 2014; Martel et al., 2015; AbuQamar et al., 2017; Santamaria et al., 2021).

After biosynthesis, the JA bioactive form (+) -7-iso-JA-Ile (JA-Ile) binds intracellularly to the CORONATINE INSENSITIVE 1 (COI1) protein, which acts as a JA co-receptor (Yan et al., 2007; Fonseca et al., 2009). In addition, COI1 forms part of the Skp, Cullin, F-box containing complex (SCFCOI1 complex), which, in the presence of JA-Ile, binds to the transcriptional repressors JASMONATE-ZIM DOMAIN PROTEIN (JAZs) and prompts JAZ degradation by the proteasome (Chini et al., 2007; Thines et al., 2007). Since JAZ proteins interact with and repress JA-related transcription factors, including the transcription factor MYC2 (Boter et al., 2004; Lorenzo et al., 2004), JAZ degradation upon JA signaling releases the transcriptional repression, triggering the expression of JA-regulated genes and the establishment of the JA response (Chico et al., 2008; Chini et al., 2009; Fonseca et al., 2009; Huang et al., 2017; Wasternack and Song, 2017).

Due to the central role of COI1 in JA perception and signaling, COI1 activity has been proven essential for the proper establishment of the JA-dependent processes. Indeed, COI1 was identified in a mutant screen for Arabidopsis plants insensitive to growth inhibition by the JA analog coronatine (Feys et al., 1994) and, subsequently, coi1 mutants were widely reported to be compromised in multiple JA responses, including inhibition of plant growth and defense against biotic stresses (Xie et al., 1998; Stintzi et al., 2001; Yan et al., 2007; Zhang and Turner, 2008). Remarkably, it has been specifically reported that coi1 mutants are highly susceptible to B. cinerea infection (Ferrari et al., 2007; Rowe et al., 2010), and that the mite T. urticae prefers to feed and lays more eggs on the tomato (Lycopersicon esculentum) coi1 mutant over wild-type plants (Li et al., 2004). These observations reinforce the relevant role of COI1 in plant defense against this specific fungus and herbivore, respectively.

The relevance of COI1 in the JA response has also been highlighted by the use of HSP90 and HSP70 inhibitors. In the presence of these inhibitors, COI1 turns unstable, which leads to the downregulation of JA-responsive genes and a reduced sensitivity to JA in root growth assays (Zhang et al., 2015). These data suggest that the HSP70-HSP90 complex plays a major role promoting COI1 stability maintenance, affecting COI1 activity and the establishment of the JA response.

HOPs (HSP70–HSP90 organizing proteins) are a highly conserved family of HSP70 and HSP90 co-chaperones, whose role in assisting the folding of signaling proteins, such as transcription factors, kinases, and nuclear receptors, has been extensively studied in nonplant eukaryotes (Schopf et al., 2017). Specifically in plants, Arabidopsis HOP proteins interact in vivo with HSP70 and HSP90, and their role has been mainly associated with plant responses to stress (Fernandez-Bautista et al., 2017, 2018; Toribio et al., 2020). Indeed, the HOP family modulates Arabidopsis capacity to acclimate to high temperatures for long periods by regulating the heat-induced transcriptional response and the maintenance of protein quality control (QC; Fernandez-Bautista et al., 2018). In addition, it was demonstrated that one of the members of the HOP family in Arabidopsis, HOP3, plays an essential role in the alleviation of the ER stress associated with the high accumulation of misfolded proteins during specific developmental programs and in response to adverse environmental cues (Fernandez-Bautista et al., 2017). Probably related to its role in this cellular stress, the rice homolog of AtHOP3 seems to be required, along with HSP90, for the efficient maturation of the CHITIN ELICITOR RECEPTOR KINASE 1 (CERK1) and for its transport to the plasma membrane, where it participates in chitin recognition and the activation of the innate response to Magnaporthe grisea (Chen et al., 2010). In addition, HOP has been recently identified as a cellular determinant of Potato virus Y (PVY) symptom development in tobacco (Nicotiana tabacum; Lamm et al., 2017). In this case, although the virus is able to replicate, antiviral host defense response and symptom appearance are suppressed in the HOP-RNAi transgenic lines, indicating that HOP also facilitates PVY perception in tobacco plants.

Despite these data strongly suggesting that HOPs play a main role in multiple stress responses, including specific aspects of plant defense, it is largely unknown whether plant HOPs could be also involved in the regulation of the hormonal networks, which is the case with the chaperones HSP70 and HSP90 and with HOPs’ mammalian counterpart. In this article, we describe that HOP3 is involved in different JA-associated processes and plays a major role in JA pathway through the regulation of COI1 activity in Arabidopsis (Arabidopsis thaliana).

Results

In a previous study, we described that the triple hop1 hop2 hop3 mutant showed an altered gene expression in response to heat (Fernandez-Bautista et al., 2018). During that study, we observed that, among the missexpressed genes, some belonged to the GO category of “JA responsive”. Therefore, in order to analyze if this altered expression could uncover a role of HOP in JA pathway, we decided to assess the performance of hop mutants during JA-related processes. For this, we initially analyzed the defense response to the spider mite T. urticae and to the necrotrophic fungus B. cinerea.

hop mutants show an increased susceptibility to T. urticae feeding

As cited above, it is well established that JA pathway plays a major role in plants during T. urticae infestation (Zhurov et al., 2014; Martel et al., 2015; Santamaria et al., 2021); therefore, to evaluate the possible implication of HOP proteins in this JA-associated process, we analyzed the susceptibility of the hop1 hop2 hop3 triple mutant (hop-tm) to this sucking herbivore in terms of leaf damage . It was previously observed that, among the members of the HOP family, HOP3 is the only member of the family highly induced in response to spider mite feeding (Supplemental Figure S1A), and based on this, in these experiments, along with the Col-0 wild-type plants and the hop1 hop2 hop3 triple mutant, we also included the hop3-1 single mutant and a hop3-1 complemented line described in Fernandez-Bautista et al. (2017). As shown in Figure 1A, the hop1 hop2 hop3 triple mutants showed higher damage (~47%) than wild-type plants at 4 d after mite feeding. A comparable increase of damage was also observed when, instead of the triple mutant, the single hop3-1 mutant was assayed. This damage was reduced to wild-type levels in the hop3-1 complemented lines. These results suggest that hop mutants show an increased susceptibility to T. urticae feeding and that HOP3 is the main player in this response. Based on this, we focused for the rest of the study on the analysis of the role of HOP3 in JA-associated processes.

Figure 1

HOP mutants show an increased susceptibility to T. urticae infestation. A, Damage quantification on leaves from Col-0, hop1 hop2 hop3 triple mutant (hop_tm), hop3-1 mutant and hop3-1 mutant complemented line 3.3 (Comp 3.3) after T. urticae feeding for 4 d. Data, normalized to the value of Col-0 that was assigned value 100%, are shown as means ± se from 18 replicates. Statistically significant differences, using Kruskal–Wallis followed by Dunn’s multiple comparison test, are highlighted by asterisks ***(P < 0.001); ns: nonsignificant differences. B, Electrolyte leakage measurements upon T. urticae infestation. Data are shown as means ± se from three replicates, each replicate involving five leaf disks. Ion leakage analyses in the absence of stress are shown in Supplementary Figure S2, where no significant changes were observed among genotypes. C, Quantification of the number of mite eggs on leaves from Col-0 and the hop mutant genotypes after T. urticae feeding for 36 h. Data are means ± se from eight replicates. For B and C, statistically significant differences using one-way ANOVA with post hoc Tukey honestly significant difference test are highlighted by asterisks *(P < 0.05), **(P < 0.01), and ***(P < 0.001); ns, non significant differences.

To further evaluate plant defense in this mutant, we also analyzed ion leakage and mite fecundity rate after infestation. In accordance with the quantification of damage, hop3-1 mutant showed an increased rate of ion leakage upon mite feeding (Figure 1B). The higher susceptibility to the mite was also highlighted by the higher rate of accumulation of mite eggs on the leaf surface of the hop3-1 mutant compared to wild-type plants (Figure 1C). Both parameters were reduced to wild-type levels in the hop3-1 complemented line, further suggesting that hop3-1 shows an increased susceptibility to spider mite feeding.

hop3-1 mutant displays a higher susceptibility to B. cinerea infection

It is well demonstrated that JA also plays a main role in the immune response against B. cinerea (AbuQamar et al., 2017). Since among the members of the AtHOP family, HOP3 expression is specifically induced during B. cinerea infection (Supplemental Figure S1B) and a preliminary analysis suggested that HOP3 could have an important role in this response (Supplemental Figure S2A), we decided to further characterize the involvement of HOP3 in the plant defense against this necrotrophic fungus. For this, leaves from wild-type plants (Col-0), the hop3-1 mutant and the hop3-1 complemented line were inoculated with B. cinerea spores and symptom development was analyzed at 1 and 4 d post-inoculation (dpi). As shown in Figure 2, A and B, leaves of the hop3-1 mutant exhibited bigger lesions and displayed more intense areas of trypan blue staining than those from Col-0 or the complemented line. These data also parallel with the increase in ion leakage in the hop3-1 mutant (1.95-fold) compared to the other two genotypes in response to the infection (Figure 2C), while no significant difference was observed in plants in the absence of stress (Supplemental Figure S2B). Finally, to provide molecular evidence for the higher susceptibility of hop3-1 mutant to this necrotrophic fungus, we also analyzed the level of expression of PLANT DEFENSIN 1.2 (PDF1.2), a gene highly induced in response to B. cinerea that has been used as a marker of defense in this plant-pathogen system (Dombrecht et al., 2007). As shown in Figure 2D, compared to the wild-type genotype, PDF1.2 induction level during the infection was significantly reduced (4.7-fold) in the hop3-1 mutant, while the induction of PDF1.2 upon B. cinerea inoculation was similar in wild-type plants and in the hop3-1 complemented genotype. Together, all these data strongly suggest that hop3-1 mutant shows an impaired establishment of the defense response and a higher susceptibility to this necrotrophic fungus.

Figure 2

HOP3 participates in plant defense response to B. cinerea infection. A, B, Representative pictures of (A) the visual lesions (marked with white arrows) observed in the leaves from WT (Col-0), hop3-1 mutant and the hop3-1 complemented line 3.3 (Comp 3.3) inoculated with B. cinerea spores at 1 and 4 dpi or (B) after trypan blue staining at 2 dpi (B, upper); quantification of the trypan blue intensity on leaves from each genotype after B. cinerea infection (B, lower). Histograms show the mean ± sd of n = 4 leaves. C, Ion leakage analyses upon B. cinerea inoculation at 2 dpi. D, RT-qPCR analysis of PLANT DEFENSIN 1.2 (AtPDF1.2) expression at 1 dpi. The values represent changes in transcript abundance (fold change related to mock-inoculated plants). C, D, Histograms show the mean ± se of n = 3 independent experiments. In all cases, statistically significant differences, using one-way ANOVA with post hoc Tukey honestly significant difference test, are highlighted by asterisks *(P < 0.05), **(P < 0.01), and ***(P < 0.001); ns, nonsignificant differences. Scale bars in (A) and (B) correspond to 1 cm and 0.25 cm, respectively.

hop3-1 mutant is partially insensitive to methyl jasmonate

The provided susceptibility data indicated that HOP3 is involved in plant defense against piercing–sucking herbivores and necrotrophic fungi, two processes highly regulated by the JA pathway. This opened the possibility that HOP3 may somehow participate in the response to this phytohormone. This hypothesis was also in agreement with the reduced expression of PDF1.2 in response to B. cinerea since, apart from being a marker for defense, PDF1.2 is a well-known JA-responsive gene (Dombrecht et al., 2007). Despite these data suggesting a role of HOP3 in the JA pathway, it has to be considered that defense responses are especially complex and that their output could be affected by intricate hormonal and regulatory pathways. Therefore, to provide evidence of the possible involvement of HOP3 in the JA pathway, we directly analyzed the sensitivity of the hop3-1 mutant to methyl jasmonate (MeJA) in root growth assays.

As expected from previous studies, the exogenous application of JA derivatives (in this case MeJA) had an inhibitory effect on primary root growth in Arabidopsis wild-type Col-0 plants (Xie et al., 1998; Chen et al., 2011; Fernandez-Calvo et al., 2011). As shown in Figure 3, A and B, MeJA inhibited root elongation and this inhibition increased at higher concentrations of MeJA. However, this inhibition was only partial in the hop3-1 mutant, since the roots of this mutant responded but their growth was inhibited to a lesser extent (16.6% and 24.29%) than those of wild-type plants in the presence of 2.5 and 5 μM MeJA, respectively. In contrast, hop3-1 complemented plants showed a restored sensitivity to the chemical, displaying a reduction in root length similar to that in wild-type plants. Remarkably, when assayed in parallel, the hop3-1 mutant and the hop1 hop2 hop3 triple mutant showed a similar sensitivity to MeJA (Supplemental Figure S3A). In addition to MeJA, a partial inhibition of root growth in the hop3-1 (comparable to the response of the hop1 hop2 hop3 triple mutant) was also observed in response to coronatine (Supplemental Figure S3B). All these data indicated that HOP3 plays a main role in JA-mediated root growth inhibition.

Figure 3

hop3-1 mutant shows impaired response to MeJA. A, Representative pictures of 7-d-old seedlings from Col-0, hop3-1, and the hop3-1 complemented line 3.3 (Comp 3.3) grown under control conditions (0 µM MeJA) or in the presence of 2.5 and 5 µM MeJA. Scale bar = 1 cm. B, Percentage of root elongation inhibition of Col-0, hop3-1 mutant and hop3-1 complemented line 3.3 (Comp 3.3) by increasing concentrations of MeJA. Data represent mean and se of n = 3 independent experiments (each containing 28 seedlings for each genotype and condition). Statistically significant differences (P < 0.01), calculated using two-way ANOVA and Bonferroni multiple tests, were obtained for comparison between Col-0 and hop3-1 at both MeJA concentrations and for the comparison between hop3-1 and line 3.3 at 5 µM MeJA; no significant differences were observed for the rest of the multiple comparisons. C, D, RT-qPCR expression analysis of the JA-responsive genes MYC2 (C) and JAZ1 (D) in roots from 7-d-old seedlings from Col-0 and from the hop3-1 mutant. Expression values (fold change in relation to the expression of each gene in Col-0 under control conditions, which was arbitrarily assigned value 1 after normalization with the calibrator gene ACT7 (At5g09810)) are shown as mean ± sem from n = 3 independent samples with three technical replicates. Statistically significant differences were calculated using a GLM analysis through Wald Chi-square followed by Bonferroni multiple comparison test and are highlighted by asterisks *(P < 0.05) and ***(P < 0.001); ns, nonsignificant differences.

It is well established that JA signaling pathway induces in the presence of JA the expression of JA-responsive genes. Therefore, in order to gain insight into whether this reduced sensitivity to MeJA impinges on the transcriptional output of the JA pathway, we selected two well-known JA responsive genes (MYC2 and JAZ1), and evaluated the expression of these genes in the wild-type genotype and hop3-1 mutant under control conditions and after MeJA treatment (Pauwels et al., 2008). As shown in Figure 3, C and D, as expected, both genes were induced in the wild-type plants in response to MeJA. Despite the fact that both genes showed a similar level of expression under control conditions, their induction in the presence of MeJA was significantly reduced in the hop3-1 mutant compared to Col-0. Since this mutant seemed unable to fully accomplish the JA transcriptional response when triggered with MeJA, these results confirm, as was also suggested by the reduced response to the assayed JA derivatives in root length assays, that this mutant is partially insensitive to JA. In addition, all these results strongly suggest that HOP3 is involved in JA perception or signaling.

HOP3 interacts with COI1 and modulates JAZ degradation levels

It was previously reported that HSP70 and HSP90 complexes are involved in the stabilization of COI1 (Zhang et al., 2015). Since HOP3 interacts with HSP70 and HSP90 in Arabidopsis (Fernandez-Bautista et al., 2017), and, according to our results, HOP3 is involved in JA perception and signaling, we reasoned that HOP3 could be part of the complexes that modulate COI1 activity. To test this hypothesis, we firstly analyzed the possible interaction between COI1 and HOP3 by directed yeast two-hybrid assays. As shown in Figure 4A, yeast cells co-transformed with plasmids expressing COI1 (fused to the Gal4-BD) and HOP3 (fused to the Gal4-AD) were able to grow on selective media, while independent co-transformations of the same constructs with vectors expressing the bare Gal4-AD and Gal4-BD, respectively, showed impaired growth in the same conditions. These data indicate that HOP3 interacts with COI1 in the two-hybrid assay. To validate this interaction in planta, we expressed different combinations of COI1-myc and GFP-HOP3 in N. benthamiana leaves and we carried out a HOP3 immunoprecipitation analysis. As shown in Figure 4B, we observed a clear band corresponding to COI1 in the eluate from HOP3 immunoprecipitations (IP: anti-GFP), but not in the eluates lacking HOP3, demonstrating that HOP3 also interacts with COI1 in vivo. This in vivo interaction was also tested by similar co-immunoprecipitation analyses expressing a different set of HOP3 and COI1 fusion proteins (HA-HOP3 and COI1-IgG-myc or COI1-GFP) and swapping the immunoprecipitated protein (i.e. immunoprecipitating COI1). In these experiments (Supplemental Figure S4), as expected from the previous results, HOP3 was co-immunoprecipitated with COI1 but not with a nonrelated protein GFP (Supplemental Figure S4B), further verifying the specific interaction between HOP3 and COI1.

Figure 4

HOP3 interacts with COI1 in yeast two-hybrid assays and in vivo. HOP3 interaction with COI1 was tested using the yeast-two hybrid system (A) or by co-immunoprecipitation in N. benthamiana leaves (B). A, The proteins fused to the Gal4-BD and Gal4-AD that were co-expressed in the AH109 strain are shown on the left of the panel. The constructs expressing the bare Gal4-BD and Gal4-AD were used as controls (−). Independent co-transformants were tested for growth in nonselective medium (-Leu-Trp) or prototrophy-selective medium (-Leu, -Trp, -His). B, Protein extracts (crude extracts) from N. benthamiana leaves transiently expressing, under the control of the 35S promoter, different combinations of COI1-myc and GFP-HOP3 were subjected to immunoprecipitation using anti-GFP beads. The presence of the different proteins in the crude extracts and in the eluted fractions from HOP3 immunoprecipitations (IP:α-GFP) was analyzed by western-blot using anti-myc and anti-GFP antibodies.

Once the role of HOP3 in the JA pathway and its interaction with COI1 was confirmed, we wondered whether HOP3’s function in JA signaling could be exerted through the regulation of COI1 activity, which triggers the JA-dependent degradation of the JAZ repressors (Chini et al., 2007; Fonseca et al., 2009). In Arabidopsis, JAZ proteins are encoded by 12 genes designated JAZ1 to JAZ12. Among these gene products, JAZ10 has been shown to interact with COI1 in the presence of coronatine in yeast two-hybrid and pull-down assays, and to be degraded in the presence of MeJA (Shyu et al., 2012; Moreno et al., 2013; Garrido-Bigotes et al., 2020). In addition, JAZ10 was shown to directly participate in JA-dependent defense response and in the inhibition of root elongation (Moreno et al., 2013; de Torres Zabala et al., 2016), which makes JAZ10 a good candidate to analyze COI1 activity in our assays. Therefore, to gain insight into the possible role of HOP3 in COI1 activity, we introgressed a JAZ10 reporter line in the hop3-1 mutant by crossing hop3-1 with the previously reported line p35S:JAZ10-GUS in Col-0 (Chico et al., 2014). Homozygous plants expressing similar levels of the reporter gene in the hop3-1 and Col-0 backgrounds were selected (Figure 5A). These lines, as expected by the expression levels, showed a similar level of JAZ10 (measured as GUS activity) under control conditions (i.e. in the absence of the JA derivative; Figure 5B). However, in the presence of MeJA, JAZ10 degradation was significantly reduced to 70% in the hop3-1 mutant compared to the wild-type background (Figure 5C), demonstrating that COI1 activity is reduced in the hop3-1 mutant and highlighting the role of HOP3 in COI1 regulation.

Figure 5

HOP3 modulates COI1 activity. A, RT-qPCR analysis of GUS expression in 7-d-old seedlings expressing the p35S:JAZ10-GUS reporter construct in the Col-0 or hop3-1 mutant backgrounds. Values are means and se, n = 3 independent experiments. B, Quantification of β-glucuronidase activity under control conditions in the cited lines. C, Quantification of COI1 activity in seedlings expressing the p35S:JAZ10-GUS reporter construct in the Col-0 and hop3-1 mutant backgrounds upon treatment with 1µM MeJA for 1 h. For B and C, data represent mean and se of n = 5 independent experiments (each containing 40 plants per condition and genotype). In all cases, statistically significant differences were analyzed using t test; *(P = 0.029); ns, non significant differences.

Discussion

HOP3 is involved in different JA-mediated processes

JA plays a main role in the regulation of many physiological processes including inhibition of root elongation and plant defense, especially against wounding herbivores and necrotrophic fungi (Gimenez-Ibanez et al., 2016). Our results show that the hop3-1 mutant is more susceptible to T. urticae feeding and to infection by the necrotrophic fungus B. cinerea. In addition, the hop3-1 mutant also shows a reduced inhibition of root elongation in response to MeJA and coronatine and an altered expression of JA-responsive genes in the presence of MeJA. All these data highlight that HOP3 is involved in different processes associated with the JA pathway. This involvement is reinforced by the in vivo interaction of HOP3 with COI1 and the reduced activity of COI1 in the hop3-1 lines in response to MeJA. Since it has been previously described that COI1 activity is required for the proper establishment of the cited JA-associated processes (Li et al., 2004; Ferrari et al., 2007; Rowe et al., 2010; Moreno et al., 2013), the reduced activity of COI1 contributes to explaining the impaired defense and the root elongation inhibition phenotypes observed in the hop3-1 mutant.

We have shown here that a HOP member is involved in the defense against leaf-sucking herbivores. Nevertheless, in rice, it has been previously reported that OsHOP participates in the defense against Magnaporthe grisea by promoting the maturation and transport of the chitin receptor CERK1 (Chen et al., 2010). This makes it possible to speculate that in the specific case of necrotrophic fungi, HOP could have a dual role in defense: on the one hand, by facilitating chitin perception through its role in CERK1 maturation and, on the other hand, by assuring the proper activation of the JA pathway through the regulation of COI1 activity. This may not be the case for the defense against spider mite, since in contrast to the important role of chitin perception in the response to B. cinerea, a relevant role of chitin signaling in the defense against T. urticae has not been reported (e.g. possible changes in susceptibility to T. urticae associated with the chitin receptor mutants has not been described). Based on this, it might be possible to speculate that the enhanced susceptibility of the hop3-1 mutant to T. urticae, as is the case in the inhibition of root length, could mainly reflect the role of HOP3 in the JA pathway.

hop3-1 mutant is partially insensitive to JA derivatives

Compared to wild-type plants, hop3-1 mutant shows a partial insensitivity to MeJA and coronatine in root length assays (i.e., inhibition of root elongation is not fully abolished in the hop3-1 mutant but is partially blocked). This implies that this mutant is partially impaired in JA response. This observation is consistent with the reduced (but not absent) activity of COI1 in the hop3-1 mutant and its reduced expression of JA-responsive genes in response to the hormone. In mammals, the function of HOP in hormone signaling has been deeply studied in the context of the folding of the glucocorticoid receptor (GR). In these eukaryotes, HOP is not strictly required for GR folding, but substantially increases the yield in the acquisition of GR’s native conformation (Morishima et al., 2000). Based on our data, we speculate that this could also be the case in plants, where HOP3 could facilitate the acquisition of COI1 native conformation and stability.

HOP3 modulates COI1 activity

Protein activity depends on protein conformation and, in this sense, protein folding is an essential step to achieve high activity yields. Folding also affects protein stability, since, in general terms, misfolded proteins are usually recognized by the QC machinery and targeted for degradation (Diaz-Villanueva et al., 2015).

During our study, we tried to test whether HOP3, through the folding of COI1, directly modulates COI1 stability. For this reason, we introgressed the p35S:COI1-FLAG constructs into the hop3-1 background. Unfortunately, the transgene p35S:COI1-FLAG got silenced in the mutant background. Furthermore, the lack of good commercial antibodies against COI1 also precluded the direct assessment of COI1 stability in the hop3-1 lines. Despite these technical problems, our data clearly demonstrate that COI1 activity (i.e. JAZ degradation) is reduced in the hop3-1 mutant. This observation is highlighted by the reduced expression of JA-responsive genes upon MeJA treatment in the mutant background. All these results demonstrate that HOP3, probably through the folding and stabilization of COI1, modulates COI1 activity.

HOP3 may participate with other co-chaperones in JA signaling

It is well known that the HSP70–HSP90 folding cycle is assisted by a cohort of different co-chaperones that act in cooperation to promote client proper folding and activity (Li et al., 2012; Prodromou, 2012; Schopf et al., 2017; Bohush et al., 2019). These co-chaperones modulate different aspects of chaperone function such as substrate selection, ATPase activity or their capacity to form multiprotein complexes. In addition, it is well established that some co-chaperones indirectly regulate the binding of other co-chaperones to the HSP70–HSP90 complex. This is the case, for example, for the FK506-BINDING PROTEINS, the CYCLOPHILIN 40 (CyP40) or the H(+)-ATPASE 1 (AHA1), whose interactions with HSP90 are inhibited in the presence of HOP (Owens-Grillo et al., 1996; Harst et al., 2005; Ebong et al., 2016). All these observations suggest that different sets of co-chaperones could mediate quality control (including the folding and degradation) of specific substrates, which makes the identification of the precise set of co-chaperones that regulate important signaling proteins, such as, COI1 extremely interesting.

In this sense, it is worth mentioning that, although AtHOP3 belongs to a highly conserved family of proteins in Arabidopsis that is composed of two other constitutively expressed members (AtHOP1 and AtHOP2), the single hop3-1 mutant displays, presumably in the presence of HOP1 and HOP2, an increased susceptibility to T. urticae feeding and to B. cinerea infection and a decreased sensitivity to JA derivates and COI1 activity. All these data suggest that HOP3 plays a prevalent role that does not fully overlap with HOP1 and HOP2 in these JA-associated processes. This prevalent role is also reinforced by the comparable responses of hop3-1 and hop1 hop2 hop3 mutants in the processes we analyzed. Although our data do not fully discard a role for HOP1 and HOP2, these data highlight the possible specificity of the different members of the HOP family in plants (Toribio et al., 2020).

In plants, another HSP70-HSP90 co-chaperone from the Suppressor of G2 allele SKP1 (SGT1) family, SGT1b, has been involved in disease resistance R protein accumulation and in JA pathway through the binding and stabilization of COI1 (Zhang et al., 2015). Since HOP3 and SGT1b associate with the HSP70 and HSP90 chaperones and share the same client protein (COI1), it is tempting to speculate that HOP3 could participate, along with SGT1b, in the folding of COI1. In this case, a partial folding of COI1 may be achieved by SGT1b in the absence of HOP3. This partial folding may lead to a reduced but not fully abolished COI1 activity, as observed in the hop3-1 mutant. This hypothesis is also consistent with the partial reduction of the levels of COI1 and the moderate alteration (compared to the coi1 mutants) of the expression of different JA-responsive genes in the sgt1b mutants (Zhang et al., 2015). SGT1b was also involved in the regulation of TIR1 stability (Gray et al., 2003; Zhang et al., 2015; Wang et al., 2016). Furthermore, COI1 and TIR1 share similar structures, where both are E3 ligases that contain an F-box and different leucine-rich repeats motifs. These parallelisms open the possibility that HOP3 could also regulate TIR1 stability. Nevertheless, whether HOP proteins (or HOP3) show an intrinsic specificity for proteins with special structures or motifs is an aspect that should be further investigated.

Taken together, this study uncovers the specific role of HOP3 as a regulator of COI1 activity and points out its relevant role in the JA signaling pathway and in plant defense. Furthermore, our results demonstrate that HOP3 forms part of the specific set of HSP70 and HSP90 co-chaperones that regulate the activation of the JA pathway.

Materials and methods

Plant material and growth conditions

Arabidopsis T-DNA insertion hop3-1 mutant (SALK_00794) was acquired from the Arabidopsis Biological Resource Center. The hop3-1 mutant and the hop3-1 line complemented with the construct pHOP3:HOP3-HA (line 3.3) were previously characterized (Fernandez-Bautista et al., 2017). The hop1 hop2 hop3 triple mutant was described in Fernandez-Bautista et al. (2018). Arabidopsis thaliana ecotype Columbia-0 (Col-0) was used as wild-type genetic background control. Unless otherwise stated, all seeds were surface-sterilized, stratified at 4°C for 48 h and grown at 22°C using a 16-h-light photoperiod. For growth on plates, Murashige and Skoog (MS) medium supplemented with 1% (w/v) sucrose was used in all cases, unless otherwise stated.

Phytopathogen strains and growth conditions

Tetranychus urticae, London strain (Acari: Tetranychidae) was provided by Dr Miodrag Grbic (University of Western Ontario, Canada). Spider mite maintenance was previously reported in Santamaria et al. (2019). Botrytis cinerea strain was kindly provided by Plant Response Biotech S. L. This fungal pathogen was grown on potato dextrose agar medium at 28°C for 8 d as previously described in Berrocal-Lobo et al. (2002) and Windram et al. (2012). Spores were collected in sterile water, filtered, quantified with a Neubauer chamber and stored in 20% (v/v) glycerol at −80°C until use.

Plant damage, ion leakage, and fecundity assays upon T. urticae feeding

Quantification of plant damage after arthropod feeding was done as described in Santamaria et al. (2019). Briefly, 3-week-old plants were infested with 20 T. urticae adults, which were carefully transferred with a brush to the leaf surface. After 4 d of feeding, leaf damage was assessed by scanning the entire rosette using a scanner (HP Scanjet 5590 Digital Flatbed Scanner series) according to Cazaux et al. (2014). Six rosettes from each genotype were included for each of the three assayed experiments. Leaf damage was calculated in mm2 using Adobe Photoshop CS software.

For ion leakage analyses, leaf disks (1-cm diameter) were infested with 10 mites for 24 h. In each case, ion leakage was determined as described in Santamaria et al. (2017). Three replicates were assayed, each of them including five disks per genotype.

Fecundity assays were performed on detached leaves from 3-week-old plants. The newest emerged leaf (about 1 cm long) from each plant was fit in special dishes and infested with 12 adult synchronized females. After 36 h of infestation, the number of eggs was counted. Eight replicates were analyzed per plant genotype.

Botrytis cinerea inoculation and phenotypic assessment

Either 2- or 3-week-old Arabidopsis plants were used for pathogen inoculation experiments. Plant leaf surfaces were inoculated with 5 μL of B. cinerea fungal inoculum (2 × 105 spores·mL−1) in potato dextrose broth medium. Inoculated plants were placed in a growth chamber, allowing fungal growth until harvesting. Infection progress, necrosis, and cell death development in infected leaves were monitored for 1 week. Ion leakage analysis, conductivity measurements and determination of cell death by trypan blue staining were performed as described in Fernández-Bautista et al. (2016). Four independent replicates were assayed, each of them including five leaves per genotype. Intensities of trypan blue staining (obtained from intensity profile plots) were estimated using ImageJ software. Four independent biological replicates were used for quantification.

Gene expression analysis by RT-qPCR

For the analysis of gene expression during B. cinerea infection, four leaves per genotype from four different mock- or B. cinerea-inoculated plants were harvested at 1 dpi. RNA isolation was carried out using the TRIzol reagent (GIBCO-Invitrogen-LifeTechnologies). Reverse transcription polymerase chain reaction (RT-qPCR) analysis was performed as described in Berrocal-Lobo et al. (2010), using 1 µg of RNA for cDNA synthesis and β-ACTIN (At3g18780) for normalization. Three biological replicates, each of them including three technical replicates, were analyzed. Primer sequences for the RT-qPCR experiments are shown in Supplemental Table S1.

For the analysis of gene expression upon MeJA treatment, 7-d-old seedlings were grown in a vertical position in MS medium in the absence or in the presence of 5 μM MeJA. For each treatment, the roots of 25 seedlings were collected. RNA isolation and RT-qPCR analyses were performed as described in Fernandez-Bautista et al. (2017) using ACT7 (At5g09810) for normalization.

For quantification of JAZ10-GUS expression, approximately 20 seedlings from the p35S:JAZ10-GUS lines in Col-0 or hop3-1 backgrounds were grown for 7 d on MS in a vertical position. RT-qPCR were performed as described in Echevarria-Zomeno et al. (2015), using PP2AA3 (At1g13320) for normalization. Each experiment was conducted in three technical replicates with three biological replicates.

Root growth inhibition assays

Arabidopsis seedlings from Col-0, hop3-1 mutant or hop3-1 mutant complemented line 3.3 seedlings were grown in a vertical position side-by-side on MS in the absence or in the presence of 2.5 and 5 μM MeJA. After 7 d of growth, pictures were taken and root length was calculated with ImageJ as described in Reveglia et al. (2018). From each experiment, at least 20 seedlings were included. Each experiment was repeated three times, obtaining similar results.

Interaction assays

The yeast two-hybrid constructs pGBK:COI1 and pGADT7:HOP3 were previously reported in Chini et al. (2007) and Fernandez-Bautista et al. (2017), respectively. Yeast transformation into AH109 and growth in selective media were carried out as described in Castellano and Sablowski (2008). Co-immunoprecipitation analyses in N. benthamiana leaves were carried out as described in Fernandez-Bautista et al. (2017) and Munoz and Castellano (2018). The assay was repeated three times obtaining similar results.

JAZ degradation assays

The p35S:JAZ10-GUS line in the wild-type background was previously described (Chico et al., 2014). This marker line was introgressed into the hop3-1 background by crossing, and a double homozygous line was used for further analyses. For quantification of JAZ10-GUS protein degradation, approximately 40 seedlings from the p35S:JAZ-GUS lines in Col-0 or hop3-1 backgrounds were grown for 7 d on MS in a vertical position and subsequently treated for 1 h with 1 µM MeJA (Reveglia et al., 2018). After the treatment, roots were collected and frozen in liquid nitrogen. GUS and total protein quantifications from the root extracts were carried out as described in Chini et al. (2018) with minor modifications. COI1 activity was calculated as the difference between JAZ10 initial quantity and final quantity per time unit and per microgram of total protein. Statistically significant differences of n = 5 independent experiments were analyzed by t test.

Statistical analyses

Statistical analyses were performed using GraphPadPrismv6.01. In all cases, data sets were subjected to normality and homoscedasticity tests in order to select the proper statistical analyses. The statistical analyses used for each set of experiments were the following: in the case of calculations of electrolyte leakage, T. urtiace egg laying, trypan blue intensity and PDF1.2 expression, one-way ANOVA with post hoc Tukey honestly significant difference test; for the analyses of leaf damage, Kruskal–Wallis followed by Dunn's multiple comparisons tests; for root growth inhibition assays, two-way ANOVA and Bonferroni multiple tests; for gene expression upon MeJA treatment, GLM analysis through Wald Chi-square followed by Bonferroni multiple comparison test; finally, for the analysis of β-glucuronidase activity in the Col-0 and hop3-1 mutant backgrounds, t test.

Accession numbers

Sequence data from this article can be found in the GenBank/EMBL data libraries under the following accession numbers: HOP1 (AT1G12270), HOP2 (AT1G62740), HOP3 (AT4G12400), COI1 (AT2G39940) and JAZ10 (AT5G13220).

Supplemental data

The following materials are available in the online version of this article.

Expression analyses of the different members of the AtHOP family in response to T. urticae infestation or to B. cinerea infection.

Preliminary analysis of hop mutants’ susceptibility to B. cinerea infection and ion leakage analyses in the absence of stress (controls for Figure 2C).

Percentage of root elongation inhibition of different hop mutants by increasing concentrations of MeJA or coronatine (COR).

HOP3 specifically interacts with COI1 by co-immunoprecipitation analyses in N. benthamiana leaves.

List of primers used in the study.

Click here for additional data file.