Salinity stress-induced modification of pectin activates stress signaling pathways and requires HERK / THE and FER to attenuate the response

Soil salinity is an increasing worldwide problem for agriculture, affecting plant growth and yield. In our attempt to understand the molecular mechanisms activated in response to salt in plants, we investigated the Catharanthus roseus Receptor like Kinase 1 Like (CrRLK1L) family, which contains well described sensors previously shown to be involved in maintaining and sensing the structural integrity of the cell walls. We have observed that herk1the1-4 double mutants, lacking the function of the Arabidopsis thaliana Receptor like Kinase HERKULES1 combined with a gain of function allele of THESEUS1, phenocopied the phenotypes previously reported in plants lacking FERONIA (FER) function. We report that both fer-4 and herk1the1-4 mutants respond strongly to salt application, resulting in a more intense activation of early and late stress responses. We also show that salt triggers de-methyl esterification of loosely bound pectins. These cell wall modifications might be partly responsible for the activation of the signaling pathways required to activate salt stress responses. In fact, by adding calcium chloride or by chemically inhibiting pectin methyl esterase (PME) activity we observed reduced activation of the early signaling protein Mitogen Activated Protein Kinase 6 (MPK6) as well as a reduced amplitude in salt-induced marker gene induction. We show that MPK6 is required for the full induction of the salt-induced gene expression markers we tested. However, the sodium specific root halotropism response is likely regulated by a different branch of the pathway being independent of MPK6 or calcium application but influenced by the cell wall sensors FER/HERK1/THE1-4 and PME activity. We hypothesize a model where salt-triggered modification of pectin requires the functionality of FER alone or the HERK1/THE1 combination to attenuate salt responses. Collectively, our results show the complexity of salt stress responses and salt sensing mechanisms and their connection to cell wall modifications, likely being in part responsible for the response phenotypes observed in salt treated plants. Introduction In plants, cell walls act as a barrier that protects plants to environmental threats and functions as structural support for the whole plant body [1]. In young Arabidopsis plants, primary cell walls (CWs) are made of complex polysaccharide networks where the main load bearing component is cellulose, a linear polymer composed of amorphous and crystalized ß 1-4 linked glucose monomers. Cellulose fibrils are intimately connected with pectin, another major component of the cell wall, described to be essential for regulating cell expansion and elasticity (reviewed in [2]). Contrary to cellulose, which is synthetized at the plasma membrane level, pectins are synthetized in Golgi bodies and released in the apoplast in a highly methylated form [3], [4]. Cell walls are subjected to controlled modifications that allow the cell to extend, driving the direction of cell growth (reviewed in [5]). Cell wall modifications also happen in response to both biotic and abiotic stresses and these structural changes are perceived and sensed by plants. Receptors/sensors are responsible for triggering intracellular downstream signals in response to turgor pressure driven cell wall-dependent mechanical distortions, a process called cell wall integrity (CWI) maintenance [6], [7]. Several members of the Catharanthus roseus Receptor Like Kinase1 Like (CrRLK1L) family have been described to play a role in mechanisms related to cell wall remodeling (e.g. cellulose impairment, pathogen infections, fertility) [8]. CrRLK1Ls contain 2 malectin domains, proposed to selectively bind complex carbohydrates, and an intracellular kinase domain responsible for transmitting signals intracellularly [9]. In Arabidopsis, 17 genes encoding for CrRLK1Ls have been identified and in some cases their functional roles have extensively been studied. For example FERONIA (FER), appears to be vital in multiple aspects of plant life (e.g. fertility, root hair development, mechanical stress, immune responses) while THESEUS1 (THE1), HERKULES 1/2 (HERK1/2) and CURVY (CVY1) are essential for cell elongation and cell morphology related processes [8], [10], [11]. Even if CrRLK1Ls are extensively studied, the mechanism through which they work is not fully understood. It seems that CrRLK1Ls might form complexes through transient hetero-oligomerization as found for the newly identified LETUM1/LETUM2 modulating plant autoimmunity and HERK1/ANJEA or ANXUR (ANX)/BUDDHA’S PAPER SEAL (BUPS) that control pollen tube related mechanisms [12], [13]. CrRLK1Ls also act as receptors for small extracellular peptides called Rapid Alkalinization Factors (RALFs). In Arabidopsis, 34 different RALF genes have been identified and RALF22, 23, 33, 34 and RALF1 have been reported to function as FER ligands, with the latter being required for Arabidopsis ovule fertilization [12], [14]–[16]. On the other hand, THESEUS1 (THE1) seems to be the receptor of RALF34, and both are essential for regulating lateral root development in Arabidopsis [16]. Some CrRLK1Ls have been defined being responsible for sensing cell wall alterations and mechanical stress responses with THESEUS1 (THE1) being essential to fully activate at least one branch of the CWI signaling pathways [17], [18]. In Arabidopsis several other proteins have been characterized to play a role in CWI signaling, including stretch-activated plasma membrane-localized calcium channels (MCAs) and the Nitrate Reductase 1 and 2 (NIA1/2), the latter being recently described to be play an essential role in the cell wall damage dependent cell cycle regulation [2], [5], [17], [18]. Interestingly, many of the CWI players have been found to overlap with salt stress response regulators. This is the case for Leucin Rich Repeat Receptor Like Kinases MIK2/KISS (MALE DISCOVERER 1-INTERACTING), FEI1/2, FERONIA, but also for NIAs and MCAs [19], [20, p. 5], [21]–[23]. Intriguingly, while loss of function mutants for mik2, fei1/2, nia1/2 and mcas (mca1,2) were reported to be highly sensitive to salt application, their responses to cellulose induced cell wall damage varied, with fei1/2 being extremely sensitive and mik2, nia1/2 and mca1, being partly or almost completely insensitive to cellulose inhibition [17], [18], [20]–[23]. Yet how and why CWI and salt induced signaling overlap has been poorly investigated. For decades, researchers have tried to discover how plants sense salt. Salt application simultaneously induces ionic and osmotic stress, which both trigger subsequential general and specific signaling responses. Within seconds to minutes after salt treatment, early signaling molecules are induced, such as calcium spikes, reactive oxygen species (ROS) accumulation and mitogen protein kinase (MAPK) cascade phosphorylation (reviewed in [24], [25]). Only recently, the identification of MONOCATION-INDUCED [Ca2+]i INCREASES 1 (MOCA1) has proposed that salt is sensed at the plasma membrane. MOCA produces glycosyl inositol phosphorylceramide (GIPC) sphingolipids, able to bind to monovalent cations (such as Na2+), triggering the opening of unknown plasma membrane localized Ca2+ channels involved in calcium spikes burst [26]. The accumulation of intracellular calcium is necessary to activate the salt-overlysensitive (SOS) pathway, which consists of the Na+/H+ exchanger SOS1 and SOS2/3 signaling components, all required for limiting the damaging effects of salt on cells by increasing Na+ efflux [27]–[29]. On the other hand, the change in main root direction observed in high salt concentrations, called halotropism, was actually stronger in the sos mutants, and was shown to be dependent on auxin distribution [30], [31]. Interestingly, salt-stress induced calcium spikes depend on FER, as fer-4 loss of function mutants are unable to induce calcium bursts as wild type plants [19]. FERONIA seems to be required to maintain cell wall architecture in response to salt stress since loss of function fer-4 mutants display cell bursting and severe cell swelling in the root in the presence of high salt concentrations. It has been proposed that the salt oversensitive phenotypes observed in FER mutants depend on the enhanced cell wall softening due to the presence of cell wall localized sodium ions. These results are corroborated by supporting data showing that FER’s malectin domain can bind pectin macromolecules in vitro [19]. Exogenous calcium and boron application, can complement, to some extent, FER-dependent salt-induced phenotypes suggesting that salt application might affect pectin architecture [19]. No available information exists for a role of other CrRLK1L members in sensing salt induced cell wall damage. However, other cell wall localized proteins such as the Leucine-rich repeat extensins (LRX) 3/4/5 have been reported be involved in helping the physical association between RALF22/23 and FERONIA causing FER inactivation through its internalization upon salt stress [32]. Here we report that salt changes cell wall architecture by de-methyl esterification of pectin, which in turn is proposed to be responsible for a range of early cellular signaling, gene expression and phenotypic responses. In our attempt to expand our current knowledge on the role of other CrRLK1L in the abiotic field, we have analyzed a complete set of salt stress responses to shed some lights on the insight of early and late salt stress signaling events in previously characterized mutants. We also show that while loss of function mutants lacking THE1 or HERK1 alone show wild type like salt stress responses, plants lacking both HERK1 in combination with a truncated, gain of function allele version of THESEUS1 (herk1the1-4 mutants [33]) show severe salt sensitivity, similarly to that of fer-4 loss of function mutants. FER and HERK1/THE1-4 are required for dampening the salt-induced Mitogen Activated protein Kinase 6 (MPK6)