**1. Introduction**

In nature, plants are usually under biotic stress caused by different pathogens, making their survival difficult [1]. To combat attackers, plants have developed a sophisticated immune system, whose activation usually produces a decrease in yield and fitness [2], and therefore it must be tightly regulated. In this immune system, the plant cell wall (CW) is the first physical barrier to prevent the pathogen from progressing into the cell, having the capacity for remodeling the architecture of its components and their physicochemical properties [3–5].

Plant CW is a dynamic structure which surrounds every plant cell, determining their shape and providing mechanical support to the protoplasts to counteract the turgor pressure. Elongating cells are walled by the primary CW, a thin layer in continuous change, mainly composed of a complex matrix of cellulose, pectins, and hemicelluloses, and structural glycoproteins [6,7]. Cellulose is the main CW scaffold, which is composed of numerous β-(1-4)-glucans tightly packed to form microfibrils, very resistant to enzymatic hydrolysis. The hemicelluloses and pectins, commonly named as matrix polysaccharides, are a group of several distinct polysaccharides [7]. Pectins are a group of complex acid polysaccharides that are divided in three main domains by their occurrence: homogalacturonan (HG), rhamnogalacturonan I (RG-I), and rhamnogalacturonan II (RG-II). The proportion of these pectin domains depends on species, tissue, and even cell type, with the presence of minor pectic domains such as apiogalacturonan or xylogalacturonan. The binding among these pectins results in complex macromolecular structures which form a

**Citation:** De la Rubia, A.G.; Mélida, H.; Centeno, M.L.; Encina, A.; García-Angulo, P. Immune Priming Triggers Cell Wall Remodeling and Increased Resistance to Halo Blight Disease in Common Bean. *Plants* **2021**, *10*, 1514. https://doi.org/ 10.3390/plants10081514

Academic Editor: William Underwood

Received: 30 June 2021 Accepted: 19 July 2021 Published: 23 July 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

hydrated gel phase where cellulose and hemicelluloses are embedded [8]. They are mostly extracted from the CW by means of calcium chelators or hot water treatments. Hemicelluloses usually bind to the surface of the cellulose microfibrils, often bearing short side branches, and are typically extracted under alkali incubation. Dominant hemicelluloses in primary CW are xyloglucan (XG), xylan, or arabinoxylan (AX) depending on plant species [9]. Accompanying polysaccharides, structural proteins such as hydroxyprolinerich glycoproteins (HRGP) or arabinogalactan proteins (AGPs) are widespread in the CW matrix [10,11]. When the cell stops growing, a strengthened secondary CW, mainly composed of cellulose, hemicelluloses, and lignin, is deposited between preexisting primary CW and the plasma membrane [12,13].

As plant morphogenesis ultimately depends on CW properties, this structure needs to adapt to any cellular change at any developmental stage along the entire life of a plant [10]. These CW processes are tightly regulated to control the metabolism responsible of CW remodeling. Firstly, the synthesis loci differ depending on the CW polymers. Cellulose is synthesized at the plasma membrane [14], while matrix polysaccharides are deposited throughout the secretory apparatus [15,16]. Secondly, the properties that these polymers initially could confer to the matrix can be changed once they are deposited, by homo and/or hetero-crosslinking and by modifying the pattern and/or degree of backbone substitution. For instance, HG is synthesized in Golgi apparatus with most of its acidic carboxyl groups methyl-esterified, however later on, by removing the methyl groups, ionic bridges can be formed by calcium ions, establishing a network among HG chains [8]. RG-I and RG-II, which are pectins decorated with neutral sugars chains, are complex polysaccharides whose abundance, cross-linking and substitutions are also regulated in muro, depending on the development stage [8,17]. To gain resistance, RG-I backbone can be enlarged, rendering a higher molecular weight of the polysaccharides. This can be also achieved by cross-linking throughout borate diesters between the RG-II domains [17]. On the other hand, typical modifications of dicotyledonous hemicellulosic glycans are the presence of fucose or galactose residues attached to the XG backbone, which is important during the cell elongation [18]. In the same way, extensins, which are HRGP, play a crucial role in the structural and compositional CW plasticity, by forming intra- and intermolecular cross-links [19].

CW remodeling serves to strengthen the CW, maintaining the plasticity to react to environmental changes, being especially important under a pathogen attack. The role of the CW components in defense has been attested in different works carried out with several mutants mainly of the model species *Arabidopsis thaliana* (Arabidopsis), in which a specific CW alteration translated into resistance or vulnerability against different diseases [20]. For example, the impairment of cellulose synthesis in an isoxaben-resistant mutant (*ixr1*, also known as *cev1*) resulted in an enhanced resistance to fungal pathogens by means of constitutive activation of some defensive pathways [21]. Similarly, Arabidopsis mutants unable to synthesize cellulose in the secondary CW, as *irx1-6*, showed resistance against the fungus *Plectosphaerella cucumerina* and the bacterium *Ralstonia pseudosolanacearum* [22]. The impairment in the callose deposition, observed in constitutively activated cell-death (*cad1-3*) Arabidopsis mutants, produced an increment to *Pseudomonas syringae* susceptibility [23]. In the case of lignin, the overexpression of *WRKY1* gene in rice, which increased lignification and resulted in major resistance against *Verticillium dahlia* [24]. By contrast, downregulation of the genes involved in lignin biosynthesis (hydroxycinnamoyl-CoA shikimate quinate hydroxycinnamoyl transferase -HCT- or cinnamoyl-CoA reductase 1 -CCR1-), with alteration in lignin content and composition, triggered defense responses in Arabidopsis [25]. Interestingly, the role of polysaccharide feruloylation in monocot *Brachypodium* and dicot Arabidopsis was also demonstrated by the transgenic expression of a fungal feruloyl esterase in both species, which reduced mono and/or dimeric ferulic acids, resulting in major susceptibility to *Bipolaris sorokiniana* and *Botrytis cinerea*, respectively [26].

Similar effects were found when matrix polysaccharides were altered. Arabidopsis mutants repressed in glucuronate 4-epimerases genes (*gae1* and *gae6*) had an impairment in galacturonic acid and showed less resistance to *P. syringae* pv. *maculicola* and *Botrytis cinerea* [27]. In addition, Arabidopsis response regulator 6 mutant (*arr6*) was enriched in pectins and showed more resistance to *P. cucumerina* but more susceptibility to *R. pseudosolanacearum* [28]. Reductions in the content of galactomannans and in the decoration of xyloglucan, observed in immutans (*im*) variegation of Arabidopsis, involved susceptibility to *P. syringae* [29]. Finally, with respect to CW proteins, the reduced residual arabinose-2 (*rra2*) mutant, defective in extensin arabinosylation, could limit oomycete colonization in Arabidopsis roots [19]. Other CW enzymes known to participate in pathogenesis are pectin modified enzymes, such as pectin methylesterases (PME), which take part in the demethylesterification of HG. An example is the infection of *P. syringae pv. macuolicola* in Arabidopsis, which leads to an increment of PME activity [27]. Thus, *pme17* Arabidopsis mutants showed an increase in susceptibility to *B. cinerea* [30]. In the same way, PME inhibitor (PMEI) mutants such as *pmei10*, *pmei11*, and *pmei12* lead to immunity against *B. cinerea* [31].

Apart from mutants, another strategy consists in comparing resistant and susceptible wheat breeding lines to *Fusarium graminearum*. It was revealed that resistance could be due to CW changes such as a higher content of lignin, a higher degree of arabinosylation on xylans, and a higher degree and different pattern of methylesterification on pectins [32].

All these works were based on CW modified mutants or breeding lines that showed susceptibility or resistance against pathogens, but their CW was already modified on these genotypes, so the level of susceptibility was determined by a preformed structure that in some mutants involved defensive pathways constitutively activated [20]. However, much less is known about CW remodeling at a physiological stage after pathogen infection. In this sense, most studies have focused on interactions with necrotrophic fungi or insects, which produces mechanical damage in the CW by the release of CW degrading enzymes (CWDEs) [26,32]. The main documented responses so far are the local deposition of CW materials, known as papillae or the cross-linking among components of the matrix, in order to hinder the access of the pathogens to the protoplast [22,31,32]. Recently, several transcriptomic and proteomic analysis would indicate that CW remodeling is a widely extended process in nature [33–36], however the detailed characterization of the changes that take place in muro after bacteria attack remain poorly characterized.

Among other defense layers, it has been proposed that cellular defense responses can be activated, but stronger, faster, and more efficiently, by means of the application of different treatments, a phenomenon known as immune priming [37]. However, it is not known whether, somehow, this immune priming conducts to the strengthening of the CW, or the immunity triggers a remodeled CW capable to cope with a second attack from pathogens. Over the past decades, several stimuli, compounds, or abiotic stresses have been studied in order to produce priming [37–41]. This new state is reached by changes at physiological, transcriptional, metabolic, and epigenetic levels, which can hold priming throughout the entire life of a plant and can even be inherited by further generations [42,43]. Among other compounds proposed to produce this long-term effect, it is the functional salycilic acid (SA) analogue 2,6-dichloroisonicotinic acid (INA), the first synthetic compound to induce defense responses in laboratory [39]. Recent studies have proposed that INA enhances basal disease resistance in common bean (*Phaseolus vulgaris* L.) by epigenetic changes that can even be inherited [41,43]. However, although it has been described that the priming activator (R)-β-homoserine increased Arabidopsis CW defense against *H. arabidopsidis* by enhanced callose deposition [41], there are no studies that focus on CW remodeling after INA priming.

In this study, *Riñón* common bean was used. It is a widely cultivated variety in León (Spain), under the protected geographic identification (PGI) "Alubia de la Bañeza-León". This variety is attacked by the biotrophic gamma-proteobacteria *Pseudomonas syringae* pv. *phaseolicola* (Pph), causing halo blight disease [44–46], which causes yield losses of up to 45%, with its main symptoms being general chlorosis, stunting and distortion of growth [45]. Until now, disease control has mainly consisted in growing healthy seeds

every season or replacing the susceptible varieties with others more resistant. This latter alternative is not suitable when the susceptible variety cultivated has gastronomical and economic interest, such as *Riñón* [45]. *Riñón* common bean has been previously described as a Pph-susceptible variety [44], although the reasons behind this susceptibility are unknown. A recent work from our group revealed that this bean variety is able to perceive the presence of Pph but it is unable to produce an effective and fast defense response, at least in part because of the lack of a quick SA peak production [47]. This is the reason why the use of SA analogue, such INA, could help in the defense process against Pph. On the other hand, there is no knowledge about CW remodeling after the Pph infection and it becomes relevant because Pph is a CWDEs producer [48].

Taking all of this into account, the aims of the present work were to know: (1), if a susceptible bean variety can remodel its CW after Pph inoculation, (2) if the application of a priming compound, such as INA, has effects in CW remodeling, and finally, (3) if these CW changes increased bean resistance against Pph. For these purposes, 15-day-old bean plants were pretreated or not with INA, and 1 week after they were inoculated or not with Pph. One-week post infection, the plants were collected and their CWs were extracted for the subsequent analyses.

#### **2. Results**

#### *2.1. INA Reduced Bean Susceptibility to Pph and Increased Flg22-Triggered Response*

To determine whether the INA application was able to prime bean plants defense as it was described before [42,49], the phenotypical symptoms after Pph inoculation were recorded. Plants were grown in vitro, a system in which infection can be forced through high pathogen concentration and humidity and without other environmental stresses [50,51]. When plants were 2 weeks old, the experimental conditions were established as follows: bean plants pre-treated with INA (INA) or not (Mock) were inoculated with Pph (INA + Pph, and Mock + Pph, respectively) (Supplementary Figure S1A). The observed disease symptoms were similar to those previously described [45]. Mock plants grew without symptoms (Figure 1A), as their foliar color and development corresponded with the normal growth of non-stressed plants. By contrast, Mock + Pph-inoculation triggered general chlorosis, greasy appearance on leaves, and the development of necrotic tissue areas, as shown in Figure 1B, indicating the susceptibility of these plants against the pathogen. INA did not cause visible stress symptoms compared to Mock (Figure 1C). Interestingly, foliar damage produced by the Pph inoculation was reduced considerably in INA + Pph plants (Figure 1D). In this case, chlorosis was restricted to foliar margins, and no greasy leaves appeared. Therefore, compared to Mock + Pph, INA + Pph plants were more resistant to Pph, and showed a statistically smaller lesion area (Figure 1E), and interestingly showed small necrotic spots in interveinal areas, which could be the result of a hypersensitive response (highlighted area in Figure 1D), as has been observed in other bean varieties resistant to Pph [52].

**Figure 1.** Phenotypic damage caused by the virulent bacteria *P. syringae* pv. *phaseolicola* (Pph) in common bean plants. Entire plants and extended leaves are shown to compare symptoms in Mock (**A**,**C**) and INA-pre-treated plants (**B**,**D**), non- (**A**,**B**) or Pph-inoculated (**C**,**D**). In panel D, a detail of the observed necrotic spots indicative of hypersensitive response (pointed by an arrow) is shown. (**E**) Quantification of the lesion area. Data represent mean ± SE (*n* = 6) and statistically differences, indicated by letters, were achieved according to *t*-Student test (*p* < 0.05). Pictures correspond to one experiment representative of three independent ones with similar results.

In view of the results obtained, whether INA priming was able to modify the ability of Riñón variety to trigger a general defense response was investigated. For this purpose, reactive oxygen species (ROS) production in bean leaf disks was monitored by a peroxidase/luminol-based assay (See Supplementary Figure S1B). This method is commonly used to reveal whether pathogen- or host-derived ligands are able to trigger early immune hallmarks [53]. In our experimental system, the INA application did not trigger a ROS production compared to mock (Figure 2). However, INA pre-treatment produced the highest ROS peak after the addition of the bacterial-derived peptide flg22 (which may mimic Pph inoculation). This suggested that INA did not induce a typical immuneassociated fast response such as ROS production but primed the bean cells for more intense defense responses.

**Figure 2.** Reactive oxygen species (ROS) production in bean leaf-disks subjected to different treatments: water (Mock, green line), 100 μM INA (INA, blue line), 1 μM flg22 (flg22, red line), or 1 μM flg22 after pre-treatment with 100 μM of INA (preINA + flg22, yellow line), and measured as relative light units (RLU) produced by Luminol reaction over the time. Total areas under the curves were integrated, and resultant values are represented at the right side of the panel. Data represent mean ± SE (*n* = 8) from one experiment representative of three independent ones with similar results. Statistically significant differences, indicated by letters, were achieved according to one-way ANOVA (*p* < 0.05), by post hoc Tukey test.
