ABA-dependent amine oxidases-derived H2O2 affects stomata conductance

Recently we showed that ABA is at least partly responsible for the induction of the polyamine exodus pathway in Vitis vinifera plants. Both sensitive and tolerant plants employ this pathway to orchestrate stress responses, differing between stress adaptation and programmed cell death. Herein we show that ABA is an upstream signal for the induction of the polyamine catabolic pathway in Vitis vinifera. Thus, amine oxidases are producing H₂O₂ which signals stomata closure. Moreover, the previously proposed model for the polyamine catabolic pathway is updated and discussed.Key words: plant growth, abscissic acid, polyamines, amine oxidases, signaling, oxidative stress, programmed cell deathWe have shown that tobacco salinity induces an exodus of the polyamine (PA) spermidine (Spd) into the apoplast where it is oxidized by polyamine oxidase (PAO) generating hydrogen peroxide (H₂O₂). Depending on the size of H₂O₂, it signals either tolerance-effector genes or the programmed cell death syndrome¹ (PCD). PAs are ubiquitous and biologically active molecules. In the recent years remarkable progress has been accomplished regarding the regulation of PAs biosynthesis and catalysis, not only under normal physiological but also under stress conditions.¹ The most studied PAs are the diamine Putrescine (Put) and its derivatives the triamine Spd and the tetramine spermine (Spm). They are present in the cells in soluble form (S), or conjugated either to low molecular weight compounds (soluble hydrolyzed form, SH) or to “macro” molecules or cell walls (pellet hydrolyzed form, PH). In higher plants, Put is synthesized either directly from ornithine via ornithine decarboxylase (ODC; EC 4.1.1.17) or indirectly from arginine via arginine decarboxylase (ADC; EC 4.1.1.19). Spd and Spm are synthesized via Spd synthase (EC 2.5.1.16, SPDS) and Spm synthase (EC 2.5.1.22, SPMS), respectively, by sequential addition of aminopropyl groups to Put, catalyzed by S-adenosyl-L-methionine decarboxylase (SAMDC; EC 4.1.1.50).^2,3 In plants, PAs are present in the cytoplasm, as well as in cellular organelles.⁴ Recently it was shown that during stress, they are secreted into the apoplast where they are oxidized by amine oxidases (AOs), such as diamine oxidase for Put (DAO, E.C. 1.4.3.6) and polyamine oxidase (PAO, E.C. 1.4.3.4) for Spd and Spm.^1,5,6 Oxidation of PAs generates, amongst other products, H₂O₂^1,7,8 which is involved in cell signaling processes coordinated by abscissic acid (ABA),⁹ but also acts as efficient oxidant and, at high concentration, orchestrates the PCD syndrome.^6,10 Two types of PA catabolism by PAO are known in plants: the terminal and the back-conversion pathways. The terminal one takes place in the apoplast, produces except H₂O₂, 1,3-diaminopropane and an aldehyde depending on the species. On the other hand, the back-conversion pathway is intracellular (cytoplasm and peroxisomes) resulting to the production of H₂O₂ and the sequential production of Put by Spm via Spd.^1,7 Now we have shown that PA exodus also occurs in Vitis vinifera and this phenomenon is at least partially induced by abscissic acid (ABA).¹¹ Thus, exogenous application of ABA results to PA exodus into the apoplast of grapevine. PA is oxidized by an AO resulting to production of H₂O₂. When the titer of H₂O₂ is below a threshold, expression of tolerance-effector genes is induced, while when it exceeds this threshold the programmed cell death (PCD) syndrome is induced.     

Reactive oxygen species as transducers of sphinganine-mediated cell death pathway

Long chain bases or sphingoid bases are building blocks of complex sphingolipids that display a signaling role in programmed cell death in plants. So far, the type of programmed cell death in which these signaling lipids have been demonstrated to participate is the cell death that occurs in plant immunity, known as the hypersensitive response. The few links that have been described in this pathway are: MPK6 activation, increased calcium concentrations and reactive oxygen species (ROS) generation. The latter constitute one of the more elusive loops because of the chemical nature of ROS, the multiple possible cell sites where they can be formed and the ways in which they influence cell structure and function.Key words: hydrogen peroxide, long chain bases, programmed cell death, reactive oxygen species, sphinganine, sphingoid bases, superoxideA new transduction pathway that leads to programmed cell death (PCD) in plants has started to be unveiled.^1,2 Sphingoid bases or long chain bases (LCBs) are the distinctive elements in this PCD route that naturally operates in the entrance site of a pathogen as a way to contend its spread in the plant tissues.^2,3 This defense strategy has been known as the hypersensitive response (HR).^4,5As a lately discovered PCD signaling circuit, three connected transducers have been clearly identified in Arabidopsis: the LCB sphinganine (also named dihydrosphingosine or d18:0); MPK6, a mitogen activated kinase and superoxide and hydrogen peroxide as reactive oxygen species (ROS).^1,2 In addition, calcium transients have been recently allocated downstream of exogenously added sphinganine in tobacco cells.⁶Contrary to the signaling lipids derived from complex glycerolipid degradation, sphinganine, a metabolic precursor of complex sphingolipids, is raised by de novo synthesis in the endoplasmic reticulum to mediate PCD.^1,2 Our recent work demonstrated that only MPK6 and not MPK3 (commonly functionally redundant kinases) acts in this pathway and is positioned downstream of sphinganine elevation.² Although ROS have been identified downstream of LCBs in the route towards PCD,¹ the molecular system responsible for this ROS generation, their cellular site of formation and their precise role in the pathway have not been unequivocally identified. ROS are produced in practically all cell compartments as a result of energy transfer reactions, leaks from the electron transport chains, and oxidase and peroxidase catalysis.⁷Similar to what is observed in pathogen defense,³ increases in endogenous LCBs may be elicited by addition of fumonisin B1 (FB1) as well; FB1 is a mycotoxin that inhibits ceramide synthase. This inhibition results in an accumulation of its substrate, sphinganine and its modified forms, leading to the activation of PCD.^1,2,8 The application of FB1 is a commonly used approach for the study of PCD elicitation in Arabidopsis.^1,2,9–11An early production of ROS has been linked to an increase of LCBs. For example, an H₂O₂ burst is found in tobacco cells after 2–20 min of sphinganine supplementation,¹² and superoxide radical augmented in the medium 60 min after FB1 or sphinganine addition to Arabidopsis protoplasts (Fig. 1A). In consonance with this timing, both superoxide and H₂O₂ were detected in Arabidopsis leaves after 3–6 h exposure to FB1 or LCBs.¹ However, the source of ROS generation associated with sphinganine elevation seems to not be the same in both species: in tobacco cells, ROS formation is apparently dependent on a NADPH oxidase activity, a ROS source consistently implicated in the HR,^13,14 while in Arabidopsis, superoxide formation was unaffected by diphenyliodonium (DPI), a NADPH oxidase inhibitor (Fig. 1A). It is possible that the latter oxidative burst is due to an apoplastic peroxidase,¹⁵ or to intracellular ROS that diffuse outwards.^16,17 These results also suggest that both tobacco and Arabidopsis cells could produce ROS from different sources.Open in a separate windowFigure 1ROS are produced at early and long times in the FB1-induced PCD in Arabidopsis thaliana (Col-0). (A) Superoxide formation by Arabidopsis protoplasts is NADPH oxidase-independent and occurs 60 min after FB1 or sphinganine (d18:0) exposure. Protoplasts were obtained from a cell culture treated with cell wall lytic enzymes. Protoplasts were incubated with 10 µM FB1 or 10 µM sphinganine for 1 h. Then, cells were vacuum-filtered and the filtrate was used to determine XTT [2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide, disodium salt] reduction as described in references ²⁸ and ²⁹. DPI was used at 50 µM. (B) H₂O₂ formation in Arabidopsis wt and lcb2a-1 mutant in the presence and absence of FB1. Arabidopsis seedlings were exposed to 10 µM FB1 and after 48 h seedlings were treated with DA B (3,3-diaminobencidine) to detect H₂O₂ according to Thordal-Christensen et al.³⁰It has been suggested that the H₂O₂ burst associated with the sphinganine signaling pathway leads to the expression of defense-related genes but not to the PCD itself in tobacco cells.¹² It is possible that ROS are involved in the same way in Arabidopsis, since defense gene expression is also induced by FB1 in Arabidopsis.⁹ In this case, it will be important to define how the early ROS that are DPI-insensitive could contribute to the PCD manifestation mediated by sphinganine.The generation of ROS (4–60 min) found in Arabidopsis was associated to three conditions: the addition of sphinganine (Fig. 1A), FB1 (Fig. 1A) or pathogen elicitors.¹⁵ This is consistent with the MPK6 activation time, which is downstream of sphinganine elevation and occurs as early as 15 min of FB1 or sphinganine exposure.² All of them are events that appear as initial steps in the relay pathway that produces PCD.In order to explore a possible participation of ROS at more advanced times of PCD progression, we detected in situ H₂O₂ formation in Arabidopsis seedlings previously exposed to FB1 for 48 h. As shown in Figure 1B, formation of the brown-reddish precipitate corresponding to the reaction of H₂O₂ with 3,3′-diaminobenzidine (DAB) was only visible in the FB1-exposed wild type plants, as compared to the non-treated plants. However, when lcb2a-1 mutant seedlings were used, FB1 exposure had a subtle effect in ROS formation. This mutant has a T-DNA insertion in the gene encoding subunit LCB2a from serine palmitoyltransferase (SPT), which catalyzes the first step in sphingolipid synthesis¹⁸ and the mutant has a FB1-resistant phenotype.² These results indicate that mutations in the LCB1¹ and LCB2a² genes (coding for the subunits of the heterodimeric SPT) that lead to a non-PCD phenotype upon the FB1 treatment, are unable to produce H₂O₂. In addition, they suggest that high levels of hydrogen peroxide are produced at advanced times in the PCD mediated by LCBs in Arabidopsis.Exposure of Arabidopsis to an avirulent strain of Pseudomonas syringae produces an endogenous elevation of LCBs as a way to implement defense responses that include HR-PCD.³ In this condition, we clearly detected H₂O₂ formation inside chloroplasts (Fig. 2A). When ultrastructure of the seedlings tissues exposed to FB1 for 72 h was analyzed, integrity of the chloroplast membrane system was severely affected in Arabidopsis wild-type seedlings exposed to FB1.² Therefore, we suggest that ROS generation-LCB induced in the chloroplast could be responsible of the observed membrane alteration, as noted by Liu et al. who found impairment in chloroplast function as a result of H₂O₂ formation in this organelle from tobacco plants. Interestingly, these plants overexpressed a MAP kinase kinase that activated the kinase SIPK, which is the ortholog of the MPK6 from Arabidopsis, a transducer in the PCD instrumented by LCBs.²Open in a separate windowFigure 2Conditions of LCBs elevation produce H₂O₂ formation in the chloroplast and perturbation in the membrane morphology of mitochondria. (A) Exposure of Arabidopsis leaves to the avirulent strain Pseudomonas syringae pv. tomato DC3000 (avrRPM1) (or Pst avrRPM1) induces H₂O₂ formation in the chloroplast. Arabidopsis leaves were infiltrated with 1 × 10⁸ UFC/ml Pst avrRPM1 and after 18 h, samples were treated to visualize H₂O₂ formation with the DAB reaction. Controls were infiltrated with 10 mM MgCl₂ and then processed for DAB staining. Then, samples were analyzed in an optical photomicroscope Olympus Provis Model AX70. (B) Effect of FB1 on mitochondria ultrastructure. Wild type Arabidopsis seedlings were treated with FB1 for 72 h and tissues were processed and analyzed according to Saucedo et al.² Ch, chloroplast; M, mitochondria; PM, plasma membrane. Arrows show mitochondrial cisternae. Bars show the correspondent magnification.In addition, we have detected alterations in mitochondria ultrastructure as a result of 72 h of FB1 exposure (Fig. 2B). These alterations mainly consist in the reduced number of cristae, the membrane site of residence of the electron transport complexes. In this sense, it has been shown that factors that induce PCD such as the victorin toxin, methyl jasmonate and H₂O₂ produce alterations in mitochondrial morphology.^20–22 In fact, some of these studies propose that ROS are formed in the mitochondria and then diffuse to the chloroplasts.^22–24It is reasonable to envisage that damage of the membrane integrity of these two organelles reflects the effects of vast amounts of ROS produced by the electron transport chains.^25,26 Recent evidence supports the destruction of the photosynthetic apparatus associated to the generation of ROS in the HR.²⁶ At this time of PCD progression, ROS could be contributing to shut down the energy machinery in the cell, which ultimately would become the point of no-return of PCD²⁷ as part of the execution program of the cell death mediated by LCBs.In conclusion, we propose that ROS can display two different functional roles in the PCD process driven by LCBs. These roles depend on the time of ROS expression, the cellular site where they are generated, the enzymes that produce them, and the magnitude in which they are formed.     

Nitric oxide production is not required for dihydrosphingosine-induced cell death in tobacco BY-2 cells

Sphinganine or dihydrosphingosine (d18:0, DHS), one of the most abundant free sphingoid Long Chain Base (LCB) in plants, is known to induce a calcium-dependent programmed cell death (PCD) in tobacco BY-2 cells. We have recently shown that DHS triggers a production of H₂O₂, via the activation of NADPH oxidase(s). However, this production of H₂O₂ is not correlated with the DHS-induced cell death but would rather be associated with basal cell defense mechanisms. In the present study, we extend our current knowledge of the DHS signaling pathway, by demonstrating that DHS also promotes a production of nitric oxide (NO) in tobacco BY-2 cells. As for H₂O₂, this NO production is not necessary for cell death induction.Key words: tobacco BY-2 cells, sphingolipids, LCBs, dihydrosphingosine, sphinganine, apoptosis, programmed cell death (PCD), nitric oxide (NO)These last few years, it has been demonstrated in plants that long chain bases (LCBs), the sphingolipid precursors, are important regulators of different cellular processes including programmed cell death (PCD).^1–3 Indeed, plant treatment with fumonisin B1 or AAL toxin, two mycotoxins that disrupt sphingolipid metabolism, leads to an accumulation of the dihydrosphingosine (d18:0, DHS), one of the most abundant free LCB in plants and correlatively to the induction of cell death symptoms.^4,5 A more recent study shows a rapid and sustained increase of phytosphingosine (t18:0), due to a de novo synthesis from DHS, when Arabidopsis thaliana leaves are inoculated with the avirulent strain Pseudomonas syringae pv. tomato (avrRpm1), known to induce a localized PCD called hypersensitive response (HR).⁶ More direct evidences were obtained from experiments on Arabidopsis cells where external application of 100 µM C2-ceramide, a non-natural acylated LCB, induced PCD in a calcium (Ca²⁺)-dependent manner.⁷ Recently, we have shown that DHS elicited rapid Ca²⁺ increases both in the cytosol and the nucleus of tobacco BY-2 cells and correlatively induced apoptotic-like response. Interestingly, blocking nuclear Ca²⁺ changes without affecting the cytosolic Ca²⁺ increases prevented DHS-induced PCD.⁸Besides calcium ions, reactive oxygen species (ROS) have also been suggested to play an important role in the control of PCD induced by sphingolipids in plants.⁹ Thus, the C2-ceramide-induced PCD in Arabidopsis is preceded by an increase in H₂O₂.⁷ However, inhibition of ROS production by catalase, a ROS-scavenging enzyme, did not prevent C2-ceramide-induced cell death, suggesting that this PCD is independent of ROS generation. Moreover, we recently showed in tobacco BY-2 cells that DHS triggers a dose-dependent production of H₂O₂ via activation of a NADPH oxidase.¹⁰ The DHS-induced cytosolic Ca²⁺ transient is required for this H₂O₂ production while the nuclear calcium variation is not necessary. In agreement with the results of Townley et al. blocking the ROS production using diphenyleniodonium (DPI), a known inhibitor of NADPH oxidases, does not prevent DHS-induced cell death. Gene expression analysis of defense-related genes, using real-time quantitative PCR (RT-qPCR) experiments, rather indicates that H₂O₂ generation is likely associated with basal defense mechanisms.¹⁰In the present study, we further investigated the DHS signaling cascade leading to cell death in tobacco BY-2 cells, by evaluating the involvement of another key signaling molecule i.e., nitric oxide (NO). In plants, NO is known to play important roles in numerous physiological processes including germination, root growth, stomatal closing and adapative response to biotic and abiotic stresses (reviewed in ref. ^11–14). NO has also been shown to be implicated in the induction of PCD in animal cells,¹⁵ in yeast,¹⁶ as well as in plant cells, in which it is required for tracheid differentiation¹⁷ or HR activation.^18,19 Interestingly in the latter case, the balance between NO and H₂O₂ production appears to be crucial to induce cell death.²⁰ Here we show in tobacco BY-2 cells that although DHS elicits a production of NO, this production is not necessary for the induction of PCD.     

The potential physiological implications of polygalacturonic acid-mediated production of superoxide

PGA/OGA/PF represent apoplastic signaling molecules implicated in the control of gene expression and the activity of enzymes involved in defense regulation. However, the underlying mechanisms behind such processes are lacking. Here we unequivocally show using EPR spectroscopy with DEPMPO spin-trap capable of differentiating between ^•OH and ^•O₂⁻ that PGA and PF can produce ^•O₂⁻ by transforming ^•OH. The potential physiological implications of this unique property are discussed. We propose that PGA/OGA/PF could represent the initiators of redox signaling cascades in stress response, with H₂O₂ being a downstream secondary messenger.Key words: polygalacturonic acid, pectin, superoxide, hydrogen peroxide, apoplast, redox signalingPGA/OGA/PF represent apoplastic signaling molecules involved in defense regulation.^1–5 For example, they induce de novo enzyme synthesis in the wound-inducible defense reaction⁶ and increase the resistance of plants to pathogens.⁷ However, the underlying mechanisms behind such processes are lacking. Aldington et al.⁸ have postulated that OGAs do not act through a receptor, but rather they owe their activity to some specific physical property. Pertinent to this is the fact that there is a broad range of active OGA structures.⁵ In addition, it has been reported that methylated OGAs are not able to trigger signaling pathways that are activated by OGAs possessing ‘free’ carboxyl groups.^9,10 In contrast to this concept, several research groups have showed that PGA/OGA/PF bind to wall-associated kinases (WAK1 and WAK2).^11–15 However, potential effects of PGA/OGA/PF on the activity of WAK1 or WAK2 have not been observed to date. We propose here that the specific property proposed by Aldington and co-workers,⁸ is in fact the ability of the polymers of galacturonic acid (PGA/OGA/PF) to produce ^•O₂⁻. By taking into account previously proposed mechanisms of reaction of PGA with ^•OH,^16,17 and thermodynamic properties of species potentially involved in the reaction,¹⁸ we hypothesized that PGA could transform the ^•OH radical into ^•O₂⁻. To test our hypothesis we investigated the effects of PGA and pectin on radical production in two different ^•OH-generating systems using EPR spectroscopy with the DEPMPO spin-trap capable of differentiating between ^•OH and ^•O₂⁻.¹⁹The results presented in Figure 1 document the ability of PGA to transform ^•OH to ^•O₂⁻. In addition, our experimental approach showed that pectin shares the ^•O₂⁻-producing ability of its constituent PGA. In the control Fenton system (Fe²⁺ + H₂O₂ → ^•OH + OH⁻ + Fe³⁺) only ^•OH radical was produced (Fig. 1A). However in the presence of PGA or pectin, a significant production of ^•O₂⁻ was detected. Haber-Weiss-like reaction (^•O₂⁻ + H₂O₂ → ^•OH + OH⁻ + O₂) generated ^•OH radical, accompanied by a low level of ^•O₂⁻ (Fig. 1B). The supplementation of PGA or pectin to this system led to sole or pronounced production of ^•O₂⁻, respectively. Under the same experimental settings, no ^•O₂⁻ production was observed for other two major extracellular carbohydrates—cellulose and mannan (Fig. 2).Open in a separate windowFigure 1The ability of PGA and pectin to transform ^•OH radical into ^•O₂⁻. Presented are characteristic EPR spectra of adducts of DEPMPO with the ^•OH radical (/OH) and the ^•O₂⁻ radical (/ooh) in two ^•OH-generating systems: (A) Fenton reaction; (B) Haber-Weiss-like reaction; in the absence (control) or presence of PGA or pectin (15 mgml⁻¹ final concentration). The downward triangle represents the characteristic line of the/OH adduct. The circular symbol represents the characteristic line of the/OOH adduct. The grey lines represent the spectral simulations based on signals of DEPMPO adducts contributing to each spectrum in specific percentages [mean values from four experiments (standard deviations were <5%)].Open in a separate windowFigure 2Characteristic EPR spectra of adducts of DEPMPO with the ^•OH radical (downward triangle) and the ^•O₂⁻ radical (circular symbol) in Haber-Weiss-like ^•OH generating system in the absence (control) or presence of cellulose or mannan (15 mgml⁻¹ final concentration). No ^•O₂⁻ production can be observed in the presence of cellulose or mannan.Presented results illustrate the ability of PGA and pectin to transform ^•OH radical into ^•O₂⁻. Other carbohydrates involved in plant metabolism, such as cellulose and mannan, but also glucose and fructose,²⁰ do not show such properties. In addition, methylated PGA do not produce ^•O₂⁻ in the reaction with ^•OH, but methane,²¹ probably with ^•CH₃ radical as an intermediate.²² This implies that carboxyl groups which are characteristic for PGA play a critical role in the production of ^•O₂⁻. Zegota¹⁶ has proposed that pectin and ^•OH react to produce pectin C(5) radical, which further reacts with molecular oxygen thus forming C(5) peroxyl radical. This radical is unstable, especially at physiological pH values,¹⁷ hence it is further decomposed to carbohydrate fragment(s) and superoxide, via ^•O₂⁻-elimination.^16,17Under in vivo settings, superoxide generated in the apoplast by PGA/OGA/PF can be further dismutated by SOD to H₂O₂, which represents a crucial signaling molecule in plants.^23,24 It is very interesting that signaling properties of H₂O₂ in the plant immune response remarkably overlap with the events initiated by OGA: (1) Similarly to the inverted H₂O₂ gradient across the plant plasma membrane,²⁴ OGA has been reported to activate calcium-dependent protein kinases,²⁵ to provoke membrane depolarization with H⁺ influx and K⁺ efflux,²⁶ and to induce activation of mitogen-activated protein kinases.²⁷ (2) Both, OGA^10,28 and H₂O₂^29–31 provoke an influx of Ca²⁺ from the apoplast into the intracellular compartment. (3) It has been documented that apoplastic generation of ^•O₂⁻ and H₂O₂ follows mechanical stress and the recognition of pathogens,^4,32–34 but also the supplementation of OGAs.^35–37 In addition, it has been reported that the supplementation of OGA to plant cells leads to the increase of apoplastic and total concentration of H₂O₂.^37,38 The enlisted results obtained by others and data presented here imply that PGA/OGA/PF could represent the initiators of redox signaling cascades in stress response, with H₂O₂ being a downstream secondary messenger.Hereby-proposed mechanism of apoplastic production of H₂O₂ by PGA/OGA/PF and SOD, depends on ^•OH radicals. Hence, the central question of our hypothesis is: “Where do apoplastic ^•OH radicals come from, under in vivo conditions?”. Hydrogen peroxide is continually generated in the apoplast by NAD(P) H oxidase/SOD, cell wall peroxidase and other sources during normal metabolism, as reviewed by Neill and co-workers.²⁴ The physiological concentrations of H₂O₂ in plants are not well established,⁴⁰ but it seems that apoplastic and total (fresh weight) concentrations are similar and maintained at around 1 µM.^38–40 In the extracellular compartment, continuous generation of H₂O₂ is balanced by its degradation in ^•OH-generating Fenton reaction which involves redox active metals, such as copper and iron.⁴¹ In principle, ^•OH radicals are further removed by apoplastic ascorbate or cell wall constituents (Fig. 3A).^41–43 However, mechanical wounding (e.g., provoked by cold⁴⁴), degradation of the cell wall by pathogenic enzymes (such as polygalacturonase or pectate lyase⁴⁵) or insect chewing could release PGA/OGA/PF from the cell wall into the apoplast. The presence of PGA/OGA/PF in the apoplast related with these events could drastically change apoplastic redox poise. Positively charged redox active metals readily bind to negatively charged polymers of galacturonic acid.⁴⁶ The close proximity of PGA/OGA/PF to the site of ^•OH production could change the fate of ^•OH. Instead of being scavenged, ^•OH could react with PGA/OGA/PF, which leads to ^•O₂⁻ production and subsequent H₂O₂ re-generation (Fig. 3B). Such ‘recycling’ of H₂O₂ could result in a higher steady-state H₂O₂ concentration in the apoplast and consequent H₂O₂ influx, as H₂O₂ is capable of passing the membrane via passive diffusion and specific aquaporins.⁴⁷ In the stress response, H₂O₂ can traverse the membrane, induce Ca²⁺ influx or diffuse into surrounding healthy tissue to modulate enzyme activity⁴⁸ and initiate gene expression,^23,24,49 crucial for subsequent phases of defense and adaptation. To conclude, PGA/OGA/PF could provide the cell with information about the status of the cell wall affected by stressors, via H₂O₂ signaling.Open in a separate windowFigure 3Schematic presentation of potential effects of PGA/OGA/PF released from stressed wall on apoplastic redox poise and H₂O₂ and Ca²⁺ signaling cascades. (A) Redox processes in apoplast under physiological settings. (B) Redox processes in the apoplast of plant cell exposed to stress. PGA/OGA/PF are released from the cell wall into the apoplast changing the redox poise by transforming ^•OH to ^•O₂⁻ and H₂O₂ (‘H₂O₂ recycling’). This could lead to H₂O₂ accumulation, H₂O₂ influx (via diffusion or peroxiporins) or the activation of Ca²⁺ influx, which leads to the activation of different intracellular responses.     

Prions: En route from structural models to structures

The prion hypothesis^1–3 states that the prion and non-prion form of a protein differ only in their 3D conformation and that different strains of a prion differ by their 3D structure.^4,5 Recent technical developments have enabled solid-state NMR to address the atomic-resolution structures of full-length prions, and a first comparative study of two of them, HET-s and Ure2p, in fibrillar form, has recently appeared as a pair of companion papers.^6,7 Interestingly, the two structures are rather different: HET-s features an exceedingly well-ordered prion domain and a partially disordered globular domain. Ure2p in contrast features a very well ordered globular domain with a conserved fold, and—most probably—a partially ordered prion domain.⁶ For HET-s, the structure of the prion domain is characterized at atomic-resolution. For Ure2p, structure determination is under way, but the highly resolved spectra clearly show that information at atomic resolution should be achievable.Key words: prion, NMR, solid-state NMR, MAS, structure, Ure2p, HET-sDespite the large interest in the basic mechanisms of fibril formation and prion propagation, little is known about the molecular structure of prions at atomic resolution and the mechanism of propagation. Prions with related properties to the ones responsible for mammalian diseases were also discovered in yeast and funghi^8,9 which provide convenient model system for their studies. Prion proteins described include the mammalian prion protein PrP, Ure2p,¹⁰ Rnq1p,¹¹ Sup35,¹² Swi1,¹³ and Cyc8,¹⁴ from bakers yeast (S. cervisiae) and HET-s from the filamentous fungus P. anserina. The soluble non-prion form of the proteins characterized in vitro is a globular protein with an unfolded, dynamically disordered N- or C-terminal tail.^15–18 In the prion form, the proteins form fibrillar aggregates, in which the tail adopts a different conformation and is thought to be the dominant structural element for fibril formation.Fibrills are difficult to structurally characterize at atomic resolution, as X-ray diffraction and liquid-state NMR cannot be applied because of the non-crystallinity and the mass of the fibrils. Solid-state NMR, in contrast, is nowadays well suited for this purpose. The size of the monomer, between 230 and 685 amino-acid residues for the prions of Figure 1, and therefore the number of resonances in the spectrum—that used to be large for structure determination—is now becoming tractable by this method.Open in a separate windowFigure 1Prions identified today and characterized as consisting of a prion domain (blue) and a globular domain (red).Prion proteins characterized so far were found to be usually constituted of two domains, namely the prion domain and the globular domain (see Fig. 1). This architecture suggests a divide-and-conquer approach to structure determination, in which the globular and prion domain are investigated separately. In isolation, the latter, or fragments thereof, were found to form β-sheet rich structures (e.g., Ure2p(1-89),^6,19 Rnq1p(153-405)²⁰ and HET-s(218-289)²¹). The same conclusion was reached by investigating Sup35(1-254).²² All these fragements have been characterized as amyloids, which we define in the sense that a significant part of the protein is involved in a cross-beta motif.²³ An atomic resolution structure however is available presently only for the HET-s prion domain, and was obtained from solid-state NMR²⁴ (vide infra). It contains mainly β-sheets, which form a triangular hydrophobic core. While this cross-beta structure can be classified as an amyloid, its triangular shape does deviate significantly from amyloid-like structures of smaller peptides.²³Regarding the globular domains, structures have been determined by x-ray crystallography (Ure2p^25,26 and HET-s²⁷), as well as NMR (mammal prions^15,28–30). All reveal a protein fold rich in α-helices, and dimeric structures for the Ure2 and HET-s proteins. The Ure2p fold resembles that of the β-class glutathione S-transferases (GST), but lacks GST activity.²⁵It is a central question for the structural biology of prions if the divide-and-conquer approach imposed by limitations in current structural approaches is valid. Or in other words: can the assembly of full-length prions simply be derived from the sum of the two folds observed for the isolated domains?     

Step by Step: Deciphering Ion Transport in the Root Xylem Parenchyma

Proton pumps produce electrical potential differences and differences in pH across the plasma membrane of cells which drive secondary ion transport through sym- and antiporters. We used the patch-clamp technique to characterize an H⁺-pump in the xylem parenchyma of barley roots. This cell type is of special interest with respect to xylem loading. Since it has been an ongoing debate whether xylem loading is a passive or an active process, the functional characterization of the H⁺-pump is of major interest in the context of previous work on ion channels through which passive salt efflux into the xylem vessels could occur. Cell-type specific features like its Ca²⁺ dependence were determined, that are important to interpret its physiological role and eventually to model xylem loading. We conclude that the electrogenic pump in the xylem parenchyma does not participate directly in the transfer of KCl and KNO₃ to the xylem but, in combination with short-circuiting conductances, plays a crucial role in controlling xylem unloading and loading through modulation of the voltage difference across the plasma membrane. Here, our recent results on the H⁺ pump are put in a larger context and open questions are highlighted.Key Words: plant nutrition, H⁺-ATPase, anion conductance, K⁺ channel, electrophysiology, signaling networkThe root xylem parenchyma is of major interest with respect to nutrient (and signal) traffic between root and shoot. One of its main functions appears to be xylem loading. However, the cell walls of the vascular tissue provide apoplastic paths between xylem and phloem that represent the upward and downward traffic lanes, allowing nutrient circulation¹ (Fig. 1). Therefore mechanisms for ion uptake and for ion release must exist side by side. In the last 15 years major progress has been made in the investigation of transport properties of xylem-parenchyma cells, and both uptake and release channels and transporters were identified. Today, we have good knowledge on the role of K⁺ and anion conductances in xylem loading with salts.² Note, that from the functionally well characterized conductances only the molecular structure of K⁺ channels is known. In contrast, many transporters are identified on the molecular level, but functional data are scarce.Open in a separate windowFigure 1Distribution of tissues in the periphery of the stele. The stippled area marks the region from which early metaxylem protoplasts originated. E, Endodermis with Casparian strip; eMX, ‘early’ metaxylem vessel; IMX, ‘late’ metaxylem vessel; Mph, metaphloem (sieve tube); Pph, protophloem (sieve tube); P, pericycle; Cx, cortex. Symplasmic and apoplasmic transport routes are indicated in red and black, respectively. The Casparian strip prevents apoplastic transport into the stele. Plasmodesmata are shown exemplarily for the indicated symplastic pathway. All cells of the symplast are connected via plasmodesmata. Sites of active uptake into the root symplast and of release into the stelar apoplast are indicated by a black and an orange arrow. Modified from Wegner and Raschke, 1994.³A challenging question to deal with was the dispute about xylem loading with ions being a passive or active process. While it is clear that energy through electrogenic H⁺ efflux is needed to take up nutrient ions from the soil against their electrochemical gradient into the cortical symplast, it has been a matter of debate if ion release into xylem vessels also is energy-linked or if the electrochemical potentials of ions are raised high enough to allow a thermodynamically passive flux.^2,3 The Casparian strip prohibits apoplastic transport of nutrients into the stele and electrically insulates the stelar from the cortical apoplast. Therefore the electrical potential difference of the cells in the xylem parenchyma could be independent from the cortical potential difference but be subject to control, for instance, from the shoot.⁴ Indeed, evidence points to xylem loading as a second control point in nutrient transfer to the shoot.^5,6 The identification and characterization of K⁺ and anion conductances clearly showed that release of KCl and KNO₃ into the xylem can be passive through voltage-dependent ion channels.^2,3,7–9 No need appeared for a pump energizing the transfer of salts to the xylem.However, H⁺ pumps are ubiquitous. H⁺-ATPases are encoded by a multigene family and heterologous expression in yeast showed that isoforms have distinct enzymatic properties.^10,11 As the example of the amino acid transporter AAP6 from the xylem parenchyma shows, a cell-type specific functional characterization of transporters is essential to draw conclusions on their physiological role. AAP6 is the only member of a multigene family with an affinity for aspartate in the physiologically relevant range. The actual apoplastic concentration of amino acids and the pH will determine what is transported in vivo.^12,13 Xylem-parenchyma cells of barley roots were strongly labelled by antibodies against the plasma membrane H⁺-ATPase.¹⁴ In a recent publication in Physiologia Plantarum we report the functional analysis of the electrogenic pump from the plasma membrane of xylem parenchyma from barley roots that was done with the patch-clamp technique after specific isolation of protoplasts from this cell type. It displayed characteristics of an H⁺-ATPase: current-voltage relationships were characteristic for a ‘rheogenic’ pump¹⁵ and currents were stimulated by fusicoccin or by an enlarged transmembrane pH gradient and inhibited by dicyclohexylcarbodiimide (DCCD). Importantly, it also showed distinct characteristics. Neither intracellular pH nor the intracellular Ca²⁺ concentration affected its activity. Noteworthy, K⁺ and anion conductances from the same cell type are controlled by intracellular [Ca²⁺]^7,9 (Fig. 2). It was proposed that the effect of abscisic acid (ABA) on anion conductances is mediated via an increase in the cytosolic Ca²⁺ concentration.¹⁶ Very likely stelar H⁺ pumps are stimulated by ABA.¹⁷ Thus, a Ca²⁺ independent control has to be hypothesized in this case.Open in a separate windowFigure 2Control of ion conductances in the plasma membrane of xylem-parenchyma cells. Arrowheads indicate stimulation and bars indicate inhibition by an increase in cytosolic [Ca²⁺],^7,9,16 by ABA,^16,17,21 by cytosolic and apoplastic acidification,^4,22 by G-proteins²³ and by an increase in apoplastic [K⁺]⁷ and [NO₃⁻].²⁴ Apoplastic [K⁺] and [NO₃⁻] modify the voltage dependence exerting negative feedback on K⁺ efflux and a positive feedback on NO₃⁻ efflux. Abscisic acid has an immediate effect on ion channel activity, most likely via [Ca²⁺], and causes a change in gene expression as indicated by circles (up) and bars (down). ABA perception is not clear. A Ca²⁺ influx could occur through a hyperpolarization activated cation conductance (HACC).^16,25 Cation transporters are NORC, nonselective cation conductance, KORC, K⁺-selective outwardly rectifying conductance (=SKOR⁸), and KIRC, K⁺-selective inwardly rectifying conductance, and anion conductances with different voltage-dependencies and gating characteristics are X-QUAC, quickly activating anion conductance, X-SLAC, slowly activating anion conductance, and X-IRAC, inwardly rectifying anion channel.^2,3,9,16,26 Transported ions and direction of flux are plotted.To date, we know that besides Ca²⁺ and abscisic acid also the pH, nonhydrolyzable GTP analogs and extracellular NO₃⁻ and K⁺ affect membrane transport capacities of root xylem-parenchyma cells (Fig. 2). Other control mechanisms by metabolites, the redox potential and phytohormones have to be included, especially if they represent signals in xylem loading or root-shoot communication. The composition of the xylem sap changes during the course of a day, depending on nutrient supply and various stresses, and the apoplastic ion concentration is considered to be an important factor in ion circulation.^6,18,19 ABA is such a signal. It is known to increase solute accumulation within the root by inhibiting release of ions into the xylem.¹⁷ Any change in transport activity has an impact on the membrane potential. This again determines whether salt release or uptake takes place. Passive salt release is restricted to a limited range of membrane potentials in which conductances for anions and cations are active simultaneously, that is with depolarization. Negative membrane voltages will be required for reabsorption of NO₃⁻ by a putative NO₃⁻/H⁺-symporter and for the uptake of K⁺ and amino acids.^3,13 As shown in our recent paper, the balance between the activities of the H⁺-pump and the anion conductances could affect the position between a depolarized and a hyperpolarized state of the parenchymal membrane. Thus, H⁺ pump activity is crucial in membrane voltage control. Furthermore, the simultaneous activities of H⁺ pumps and anion conductances make the generation of a high pH gradient possible, whilst maintaining electroneutrality. The proton gradient could be used for ion transport through cotransporters and antiporters as suggested for the loading of borate into the xylem through the boron transporter BOR1.²⁰ So we are on the way to decipher xylem loading in roots and this exciting field will also provide information about small-scale nutrient cycling and root-shoot communication. To determine how the activities of pumps, channels and transporters are adjusted among each other is the next challenge. Further insight has to be obtained by experimentation as well as by biophysical modeling.     

pH signature for the responses of arbuscular mycorrhizal fungi to external stimuli

Environmental and developmental signals can elicit differential activation of membrane proton (H⁺) fluxes as one of the primary responses of plant and fungal cells. In recent work,¹ we could determine that during the presymbiotic growth of arbuscular mycorrhizal (AM) fungi specific domains of H⁺ flux are activated by clover root factors, namely host root exudates or whole root system. Consequently, activation on hyphal growth and branching were observed and the role of plasma membrane H⁺-ATPase was investigated. The specific inhibitors differentially abolished most of hyphal H⁺ effluxes and fungal growth. As this enzyme can act in signal transduction pathways, we believe that spatial and temporal oscillations of the hyphal H⁺ fluxes could represent a pH signature for both early events of the AM symbiosis and fungal ontogeny.Key words: H⁺-specific vibrating probe, pH signatures, arbuscular mycorrhiza, pH signalling, Gigaspora margaritaThe 450-million-year-old symbiosis between the majority of land plants and arbuscular mycorrhizal (AM) fungi is one of the most ancient, abundant and ecologically important symbiosis on Earth.^2,3The development of AM interaction starts before the physical contact between the host plant roots and the AM fungus. The hyphal growth and branching are induced by the root factors exudated by host plants, followed by the formation of appressorium leading to the hyphal penetration in the root system. These root factors seems to be specifically synthesized by host plants, since exudates from non-host plants are not able to promote neither hyphal differentiation nor appressorium formation.^4,5 Most root exudates contain several host signals or better, active compounds including flavonoids^6,19 and strigolactones,^7,8 however many of them are not yet known.Protons (H⁺) may have an important role on the fungal growth and host signal perception.¹ In plant and fungal cells, H⁺ can be pumped out through two different mechanisms: (1) the activity of the P-type plasma membrane (PM) H⁺-ATPase⁹ and (2) PM redox reactions.¹⁰ The proportional contribution from both mechanisms is not known, but in most plant cells the PM H⁺-ATPase seems to be the major responsible by the H⁺ efflux across plasma membrane. AM Fungal cells also energize their PM using P-type H⁺-pumps quite similar to the plant ones. Indeed, some genes codifying isoforms of P-type H⁺-ATPase have been isolated of AM fungi,^11–13 and AM fungal ATP hydrolysis activity was shown by cytochemistry, localized mainly in the first 70 µm from the germ tube tip.¹⁴ This structural evidence correlates with data obtained by H⁺-specific vibrating probe (Fig. 1A and B), which indicates that the H⁺ efflux in Gigaspora margarita is more intense in the subapical region of the lateral hyphae¹ (Fig. 1A). Furthermore, the correlation between the cytosolic pH profile previously obtained by Jolicoeur et al.,¹⁵ with the H⁺ efflux pattern (erythrosine-dependent), seems to clearly indicate that an active PM H⁺-ATPase takes place at the subapical hyphal region. Using orthovanadate, we could show that those H⁺ effluxes are susceptible mainly in the subapical region, but no effect in the apical was found.¹ Recently, a method to use fluorescent marker expression in an AM fungus driven by arbuscular mycorrhizal promoters was published.³¹ It could be adjusted as an alternative to measure “in vivo” PM H⁺-ATPase expression in AM fungal hyphae and their responses to root factors.³¹Open in a separate windowFigure 1(A) H⁺ flux profile along growing secondary hyphae of G. margarita in the presence (open squares) or absence (closed squares) of erythrosin B and its correlation with cytosolic pH (pHc) data described by Jolicoeur et al.,¹⁵ (dotted line). Dotted area depicts the region with higher susceptibility to erythrosin B. (B) ion-selective electrode near to AM fungal hyphae. (C) Stimulation on hyphal H⁺ efflux after incubation with root factors or whole root system. R, roots; RE, root exudates; CO₂, carbon dioxide; CWP, cell wall proteins; GR24, synthetic strigolactone. The medium pH in all treatment was monitored and remained about 5.7, including with prior CO₂ incubation. Means followed by the same letter are statistically equal by Duncan''s test at p < 5%.The H⁺ electrochemical gradient generated by PM H⁺-ATPases provides not only driving force for nutrient uptake,^9,16 but also can act as an intermediate in signal transduction pathways.¹⁸ The participation of these H⁺ pumps in cell polarity and tip growth of plant cells was recently reported,²⁷ addressing their crucial role on apical growth.²⁸ Naturally, in the absence of root factors the AM fungi have basal metabolic^8,21–23 and respiratory activity.²⁴ However when root signals are recognized and processed by AM fungal cells they might become activated.²² We thus searched for pH signatures that could reflect the alterations on fungal metabolism in response to external stimuli. In fact, preliminary analyses from our group demonstrate that AM fungal hyphae increase their H⁺ efflux in response not only to root exudates recognition, but also to other root factors (Fig. 1C). The incubation for 30 min of AM fungal hyphae with several root factors induces hyphal H⁺ efflux similar to the response to intact root system (5 days of incubation). The major increases were found with 1% CO₂ (750%) followed by root cell wall proteins (221%), root exudates (130%) and synthetic strigolactone (5%) (Fig. 1C). Those stimulations could define the transition from the state without root signals to the presymbiotic developmental stage (Fig. 1C). In the case of CO₂, the incorporation of additional carbon could represent a new source of energy, since CO₂ dark fixation takes place in Glomus intraradices germ tubes.^22,25Interestingly, after the treatment with synthetic strigolactone (10⁻⁵ M GR24), no significant stimulation was found compared to the remaining factors (Fig. 1C). It opens the question if the real effect of strigolactone is restrict to hyphal branching and does not intervene in very fast response pathways. Likewise, strigolactones need additional time to exhibit an effect, as recently discussed by Steinkellner et al.,²⁶ However, at the moment, no comprehensive electrophysiological analyses are presently available separating the effects of strigolactone and some flavonoids in AM fungal hyphae.The next target of our work is the study of ionic responses of single germ tubes or primary hyphae to root factors (Fig. 2). As reported by Ramos et al.,¹ we have been observing that the pattern of ion fluxes at the apical zone of primary hyphae is differentiated from secondary or lateral hyphae. In the primary, two interesting responses were detected in the absence of root factors: (1) a “dormant Ca²⁺ flux” and (2) Cl⁻ or anion fluxes at the same direction of H⁺ ions, suggesting a possible presence of H⁺/Cl⁻ symporters at the apex, similarly to what occurs in root hairs (Fig. 2).³⁰ In the presence of root factors such as root exudates the stimulated influxes of Cl⁻ (anion), H⁺, Na⁺ and effluxes of K⁺ and Ca²⁺ are activated. It can explain why the AM fungi hyphal tips are depolarized^20,29 during the period without root signals—“asymbiosis”—as long as K⁺ efflux and H⁺ influx occur simultaneously. Indeed, H⁺ as well as Ca²⁺ ions may act as second messengers, where extra and intracellular transient pH changes are preconditions for a number of processes, including gravity responses and possibly in plant-microbe interactions.^17,30Open in a separate windowFigure 2Ion dynamics in the apex of primary hyphae of arbuscular mycorrhizal fungi. It represents the Stage 1 described in Ramos et al.¹ After treatment with root factors, an activation of Ca²⁺ efflux is observed at the hyphal apex.Clearly, further data on the mechanism of action of signaling molecules such as strigolactones over the signal transduction and ion dynamics in AM fungi will be very important to improve our understanding of the molecular bases of the mycorrhization process. Future studies are necessary in order to provide basic knowledge of the ion signaling mechanisms and their role on the response of very important molecules playing at the early events of AM symbiosis.     

Voltage,reactive oxygen species and the influx of calcium

The apical plasma membrane of young Arabidopsis root hairs has recently been found to contain a depolarisation-activated Ca²⁺ channel, in addition to one activated by hyperpolarisation. The depolarisation-activated Ca²⁺ channel may function in signalling but the possibility that the root hair apical plasma membrane voltage may oscillate between a hyperpolarized and depolarized state suggests a role in growth control. Plant NADPH oxidase activity has yet to be considered in models of oscillatory voltage or ionic flux despite its predicted electrogenicity and voltage dependence. Activity of root NADPH oxidase was found to be stimulated by restricting Ca²⁺ influx, suggesting that these enzymes are involved in sensing Ca²⁺ entry into cells.Key words: calcium, channel, NADPH oxidase, oscillation, root hairElevation of cytosolic free Ca²⁺ ([Ca²⁺]_cyt) encodes plant cell signals.¹ Reactive oxygen species (ROS) are potent regulators of the PM Ca²⁺ channels implicated in signalling and developmental increases in [Ca²⁺]_cyt.^1,2 Plasma membrane (PM) voltage (V_m) also plays a significant part in generating specific [Ca²⁺]_cyt elevations through the opening of voltage-gated Ca²⁺-permeable channels, allowing Ca²⁺ influx.^1,3 Patch clamp electrophysiological studies on the root hair apical PM of Arabidopsis have revealed co-localisation of hyperpolarisation-activated Ca²⁺ channels (HACCs),⁴ ROS-activated HACCs⁵ and depolarisation-activated Ca²⁺ channels (DACCs).⁶ The DACC characterisation pointed to the presence of a Cl⁻-permeable conductance that was activated by moderate hyperpolarisation (−160 mV) but rapidly inactivated when the voltage was maintained at such negative values.⁶ This may be the R-type anion efflux conductance previously described in Arabidopsis root hair and root epidermal PM.⁷ Previous studies have shown that root hair PM also harbors K⁺ channels (mediating inward or outward flux)^8–10 and a H⁺-ATPase.¹¹ A key problem to address now is how these transporters interact to generate and be influenced by PM V_m, thus gating and in turn being regulated by their companion Ca²⁺ channels to encode developmental and environmental signals at the hair apex.A seminal study on the relationship between V_m and ionic fluxes in wheat root protoplasts not only confirmed oscillatory events but also determined that the PM can exist in three distinct states.¹² In the “pump state” the H⁺-ATPase predominates, there is net H⁺ efflux and the hyperpolarized V_m is negative of the equilibrium potential for K⁺ (E_K). In the “K state”, K⁺ permeability predominates but there is still net H⁺ efflux and V_m = E_K. In the third state, there is net H⁺ influx and V_m > E_K. In this depolarized H⁺-influx state, the H⁺-ATPase is thought to be inactive. Oscillations in PM V_m and H⁺ flux may be more profound in growing cells^13,14 and oscillations between these states may explain the temporal changes in H⁺ flux recently observed at the apex of growing Arabidopsis root hairs.¹⁵ Peaks of H⁺ influx may reflect a depolarized V_m that could activate DACC, suggesting that DACC would play a significant role in growth regulation. The view has arisen that the HACC would be the main driver of growth, primarily because in patch clamp assays its current is greater than DACC^4–6 and because resting V_m is usually found to be hyperpolarized. In a growing cell, with a V_m oscillating between a hyperpolarized and depolarized state, a DACC could just as well be a driver of growth given that the Ca²⁺ influx it permits could be amplified through intracellular release.The PM H⁺-ATPase traditionally lies at the core of models of voltage and ionic flux^14,16 but in terms of [Ca²⁺]_cyt regulation, the activity of PM NADPH oxidases must also now be considered. The Arabidopsis root hair apical PM also contains an NADPH oxidase (AtrbohC) that catalyses extracellular superoxide production.⁵ AtrbohC is implicated in the transition to polar growth at normal extracellular pH⁵ and also osmoregulation.¹⁷ NADPH oxidases catalyse the transport of electrons out of the cell and thus, in common with PM redox e⁻ efflux systems,¹⁸ their activity would depolarize the membrane voltage unless countered by cation efflux or anion influx.¹⁹ Two H⁺ would also be released into the cytosol for every NADPH used. The voltage-dependence of plant NADPH oxidases is unknown but e⁻ efflux by animal NADPH oxidases is fairly constant over negative V_m and decreases at very depolarized V_m.²⁰ AtrbohC is implicated in generating oscillatory ROS at the root hair apex and loss of function affects magnitude and duration of apical H⁺ flux oscillations.¹⁵ The latter suggests that AtrbohC function does in some way affect V_m, a situation extending to other root cell types (such as the epidermis) expressing NADPH oxidases.²¹NADPH oxidase activity in roots is under developmental control but also responds to anoxia and nutrient deficiency^22,23 to signal stress conditions. Blockade of PM Ca²⁺ channels by lanthanides increases superoxide production in tobacco suspension cells.²⁴ This suggests that NADPH oxidases are involved in sensing the cell''s Ca²⁺ status and the prediction would be that extracellular Ca²⁺ chelation would increase their activity. To test this, superoxide anion production by excised Arabidopsis roots was measured using reduction of the tetrazolium dye XTT (Sodium, 3′-[1-[phenylamino-carbonyl]-3,4-tetrazolium]-bis(4-methoxy-6-nitro) benzene-sulphonic acid).^25,26 Lowering extracellular Ca²⁺ from 0.5 mM to 1.4 µM by addition of 10 mM EGTA caused a mean 95% increase in diphenyliodinium-sensitive superoxide production (Fig. 1; n = 9), implicating NADPH oxidases as the source of this ROS. Stimulation of NADPH oxidase activity by decreasing Ca²⁺ influx at first appears contradictory as NADPH oxidases are stimulated by increased [Ca²⁺]_cyt²⁷ (Fig. 1). However, reduction of Ca²⁺ influx should promote voltage hyperpolarisation (just as block of K⁺ influx causes hyperpolarisation in root hairs²⁸) and this could feasibly cause increased NADPH oxidase activity. Production of superoxide could then result in ROS-activated HACC activity⁵ to increase Ca²⁺ influx.Open in a separate windowFigure 1Superoxide anion production by Arabidopsis roots. Assay medium comprised 10 mM phosphate buffer with 0.5 mM CaCl₂, 500 µM XTT, pH 6.0. Production was linear over the 30 min incubation period. Control, mean ± standard error, n = 9. Test additions were: 20 µM of the NADPH oxidase inhibitor diphenylene iodonium (DPI; n = 6); 100 µM of the Ca²⁺ ionophore {"type":"entrez-nucleotide","attrs":{"text":"A23187","term_id":"833253","term_text":"A23187"}}A23187,³⁰ to increase [Ca²⁺]_cyt (n = 9); 10 mM of the chelator EGTA (n = 9). Dimethyl sulphoxide [DMSO; 1% (v/v)] was used as a carrier for XTT and DPI and a separate control for this is shown (n = 9).In addition to V_m, activities of PM transporters in vivo will be subject to other levels of regulation such as phosphorylation, nitrosylation and the action of [Ca²⁺]_cyt itself. Distinct spatial separation of transporters will undoubtedly play a significant role in governing V_m and [Ca²⁺]_cyt dynamics, particularly in growing cells. An NADPH oxidase has already been found sequestered in a potential PM microdomain in Medicago.²⁹ While there is still much to do on the “inventory” of PM transporters involved in Ca²⁺ signalling in any given cell, placing them in context not only requires knowledge of their genetic identity but also modelling of their concerted action.     

Structural basis of transmembrane domain interactions in integrin signaling

Cell surface receptors of the integrin family are pivotal to cell adhesion and migration. The activation state of heterodimeric αβ integrins is correlated to the association state of the single-pass α and β transmembrane domains. The association of integrin αIIbβ3 transmembrane domains, resulting in an inactive receptor, is characterized by the asymmetric arrangement of a straight (αIIb) and tilted (β3) helix relative to the membrane in congruence to the dissociated structures. This allows for a continuous association interface centered on helix-helix glycine-packing and an unusual αIIb(GFF) structural motif that packs the conserved Phe-Phe residues against the β3 transmembrane helix, enabling αIIb(D723)β3(R995) electrostatic interactions. The transmembrane complex is further stabilized by the inactive ectodomain, thereby coupling its association state to the ectodomain conformation. In combination with recently determined structures of an inactive integrin ectodomain and an activating talin/β complex that overlap with the αβ transmembrane complex, a comprehensive picture of integrin bi-directional transmembrane signaling has emerged.Key words: cell adhesion, membrane protein, integrin, platelet, transmembrane complex, transmembrane signalingThe communication of biological signals across the plasma membrane is fundamental to cellular function. The ubiquitous family of integrin adhesion receptors exhibits the unusual ability to convey signals bi-directionally (outside-in and inside-out signaling), thereby controlling cell adhesion, migration and differentiation.^1–5 Integrins are Type I heterodimeric receptors that consist of large extracellular domains (>700 residues), single-pass transmembrane (TM) domains, and mostly short cytosolic tails (<70 residues). The activation state of heterodimeric integrins is correlated to the association state of the TM domains of their α and β subunits.^6–10 TM dissociation initiated from the outside results in the transmittal of a signal into the cell, whereas dissociation originating on the inside results in activation of the integrin to bind ligands such as extracellular matrix proteins. The elucidation of the role of the TM domains in integrin-mediated adhesion and signaling has been the subject of extensive research efforts, perhaps commencing with the demonstration that the highly conserved GFFKR sequence motif of α subunits (Fig. 1), which closely follows the first charged residue on the intracellular face, αIIb(K989), constrains the receptor to a default low affinity state.¹¹ Despite these efforts, an understanding of this sequence motif had not been reached until such time as the structure of the αIIb TM segment was determined.¹² In combination with the structure of the β3 TM segment¹³ and available mutagenesis data,^6,9,10,14,15 this has allowed the first correct prediction of the overall association of an integrin αβ TM complex.¹² The predicted association was subsequently confirmed by the αIIbβ3 complex structure determined in phospholipid bicelles,¹⁶ as well as by the report of a similar structure based on molecular modeling using disulfide-based structural constraints.¹⁷ In addition to the structures of the dissociated and associated αβ TM domains, their membrane embedding was defined^{12,13,16,18,19} and it was experimentally recognized that, in the context of the native receptor, the TM complex is stabilized by the inactive, resting ectodomain.¹⁶ These advances in integrin membrane structural biology are complemented by the recent structures of a resting integrin ectodomain and an activating talin/β cytosolic tail complex that overlap with the αβ TM complex,^20,21 allowing detailed insight into integrin bi-directional TM signaling.Open in a separate windowFigure 1Amino acid sequence of integrin αIIb and β3 transmembrane segments and flanking regions. Membrane-embedded residues^{12,13,16,18,19} are enclosed by a gray box. Residues 991–995 constitute the highly conserved GFFKR sequence motif of integrin α subunits.     

Involvement of glucose-6-phosphate dehydrogenase in reduced glutathione maintenance and hydrogen peroxide signal under salt stress

Cellular redox homeostasis is essential for plant growth, development as well as for the resistance to biotic and abiotic stresses, which is governed by the complex network of prooxidant and antioxidant systems. Recently, new evidence has been published that NADPH, produced by glucose-6-phosephate dehydrogenase enzyme (G6PDH), not only acted as the reducing potential for the output of reduced glutathione (GSH), but was involved in the activity of plasma membrane (PM) NADPH oxidase under salt stress, which resulted in hydrogen peroxide (H₂O₂) accumulation. H₂O₂ acts as a signal in regulating G6PDH activity and expression, and the activities of the enzymes in the glutathione cycle as well, through which the ability of GSH regeneration was increased under salt stress. Thus, G6PDH plays a critical role in maintaining cellular GSH levels under long-term salt stress. In this addendum, a hypothetical model for the roles of G6PDH in modulating the intracellular redox homeostasis under salt stress is presented.Key words: glucose-6-phosphate dehydrogenase, hydrogen peroxide, reduced glutathione, redox homeostasis, salt stressEnvironmental stresses inevitably induce the production of reactive oxygen species (ROS).¹ Reduced glutathione (GSH) is a key substance in the network of antioxidants that include ascorbate, glutathione, α-tocopherol and a serial of antioxidant enzymes,² which metabolizes H₂O₂ mainly via the ascorbate-glutathione cycle, the most important detoxifying system in plants.³ Thus, the regulatory ability to maintain the cellular GSH balance is crucial to confer the resistance to oxidative stress in plants. However, to our knowledge, the regulatory mechanism on the intracellular GSH-pool equilibrium under environmental stresses has been largely unknown in plants.A main source of GSH is regenerated from its oxidative form (GSSG) via glutathione cycling, which uses NADPH as the reductant.⁴ G6PDH is the key enzyme of pentose phosphate pathway that is responsible for the generation of NADPH.⁵ G6PDH has been shown to play a protective role against ROS in human and animal cells,^6,7 and the enhanced expression of G6PDH could enhance the GSH levels and the ability of resistance to oxidative stress.^5,8 In plants, it has been reported that oxidative stress induced by the elicitor stimulated G6PDH activity in tobacco cells,^9,10 and the GSH-biosynthesis inhibitor or GSH precursor could increase or suppressed G6PDH activity, respectively.¹⁰ Interestingly, after G6PDH activity was inhibited, not only GSH levels dramatically decreased, but the elicitor-induced H₂O₂ accumulation was also completely counteracted.^9,10 Thus, the functions of G6PDH under oxidative stress seem to be involved in these two contradictory courses in cells: the regeneration of GSH as well as H₂O₂ accumulation. The role of G6PDH under environmental stresses remained limited to clarify this, so we studied the G6PDH functions with a series of inhibitor or donor of GSH, H₂O₂ and G6PDH in reed calli under salt stress. Our recent studied clearly demonstrated that G6PDH activity was also simultaneously involved in intracellular GSH maintenance and H₂O₂ accumulation in salt stress. Further studies revealed that a plasma membrane (PM) NADPH oxidase, using NADPH as substrate mainly produced by G6PDH, was mainly responsible for the generation of H₂O₂. And H₂O₂, produced under salt stress, induced the increased G6PDH activity and the enzymes of glutathione cycle, which concomitantly resulted in an increased GSH contents. Foyer and Noctor (2005) suggested that the cellular “oxidative signaling” was made possibly by homeostatic regulation by antioxidant redox buffer.¹¹ Based on these, it can be speculated that G6PDH might play an important role in maintaining the cellular redox signals under salt stress in plants.Our recent work provides a new insight into G6PDH functions under environmental stresses in plants. Growing evidences suggest that PM NADPH oxidase is responsible for H₂O₂ accumulation under stresses,^12,13 and H₂O₂ is involved in various signaling pathways in plants, such as defense gene expression, stomatal closure, root growth, programmed cell death (PCD) and so on.¹¹ In addition, GSH, as a key antioxidant, also influences gene expression associated with biotic and abiotic stress responses to maximum defense.² Recent study also reported that G6PDH was involved in NR-dependent NO production, and thus played a pivotal role in establishing tolerance of red kidney bean roots to salt stress.¹⁴ Therefore, the research work is required to further clarify the regulatory mechanism underlying the roles of G6PDH in the cellular redox homeostasis as well as the related signals under environmental stresses in plants.Based on the results obtained so far, a model for G6PDH functions under salt stress is proposed (Fig. 1). In our model, the increased G6PDH activity is tightly correlated with GSH maintenance and H₂O₂ accumulation through PM NADPH oxidase under salt stress in plants. Under salt stress, H₂O₂ activities the activities of G6PDH and the enzymes in glutathione recycle, which finally result in the enhanced glutathione cycling rate and thus the increased GSH levels. This enhanced antioxidant ability can facilitate to maintain a steady-state level of H₂O₂. Eventually, the properly intracellular redox state is established under salt stress and forms a metabolic interface for signals. Thus, we suggest that G6PDH plays a crucial role in establishing this cellular redox homeostasis under salt stress.Open in a separate windowFigure 1Hypothetical model for the roles of G6PDH under salt stress. Under salt stress, G6PDH activity is involved in both GSH maintenance and H₂O₂ accumulation through PM NADPH oxidase. H₂O₂, as a signal, increases the activities of G6PDH, glutathione (GR) and glutathione peroxidase (GPX), which finally enhance glutathione cycle rate and result in the increased GSH levels. This enhanced antioxidant ability could facilitate to keep H₂O₂ in a steady state for signal in salt stress.     

Putting the brakes on cancer cell migration: JAM-A restrains integrin activation

Junctional Adhesion Molecule A (JAM-A) is a member of the Ig superfamily of membrane proteins expressed in platelets, leukocytes, endothelial cells and epithelial cells. We have previously shown that in endothelial cells, JAM-A regulates basic fibroblast growth factor, (FGF-2)-induced angiogenesis via augmenting endothelial cell migration. Recently, we have revealed that in breast cancer cells, downregulation of JAM-A enhances cancer cell migration and invasion. Further, ectopic expression of JAM-A in highly metastatic MDA-MB-231 cells attenuates cell migration, and downregulation of JAM-A in low-metastatic T47D cells enhance migration. Interestingly, JAM-A expression is greatly diminished as breast cancer disease progresses. The molecular mechanism of this function of JAM-A is beyond its well-characterized barrier function at the tight junction. Our results point out that JAM-A differentially regulates migration of endothelial and cancer cells.Key words: JAM-A, integrin, α_vβ₃, FGF-2, breast cancer, cell migration and invasion, T47D, MDA-MB-231, siRNAEndothelial and epithelial cells exhibit cell polarity and have characteristic tight junctions (TJs) that separate apical and basal surfaces. TJs are composed of both transmembrane and cytoplasmic proteins. The three major families of transmembrane proteins include claudins, occludin and JAM family members.^1–3 Additionally, interaction between the peripheral proteins such as PDS-95/Discs large/ZO family (PDZ) domain-containing proteins in TJs plays an important role in maintaining the junctional integrity.^2,4,5JAMs are type I membrane proteins (Fig. 1) predominately expressed in endothelial and epithelial cell TJs, platelets and some leukocytes.^6–8 The classical JAMs are JAM-A, JAM-B and JAM-C, which can all regulate leukocyte-endothelial cell interaction through their ability to undergo heterophilic binding with integrins α_Lβ₂ or α_vβ₃, α₄β₁ and α_Mβ₂ respectively. The cytoplasmic tail of JAMs contains a type II PDZ-domain-binding motif (Fig. 1) that can interact with the PDZ domain containing cytoplasmic molecules such as ZO-1, ASIP/PAR-3 or AF-6.^9,10 Additionally, consistant with their junctional localization and their tendency to be involved in homophilic interactions, JAMs have been shown to modulate paracellular permeability and thus may play an important role in regulating the epithelial and endothelial barrier.^11,12 In addition, ectopic expression of JAM-A in CHO cells promotes localization of ZO-1 and occludin at points of cell contacts, which suggests a role for JAM-A in TJ assembly.^10,13,14 Recently, it has been shown that JAM-A regulates epithelial cell morphology by modulating the activity of small GTPase Rap1 suggesting a role for JAM-A in intracellular signaling.¹⁵Open in a separate windowFigure 1Schematic representation of the domain structure of JAM family proteins. V, variable Ig domain; C2, constant type 2 Ig domain; TM, transmembrane domain; T-II, Type II PDZ-domain binding motif.We have previously shown that JAM-A is a positive regulator of fibroblast growth factor-2 (FGF-2) induced angiogenesis.¹⁶ Evidence was provided to support the notion that JAM-A forms a complex with integrin α_vβ₃ at the cell-cell junction in quiescent human umbilical cord vein endothelial cells (HUVECs) and FGF-2 dissociates this complex.¹⁶ It was further established that inhibition of JAM-A using a function-blocking antibody also inhibits FGF-2 induced HUVECs migration in vitro and angiogenesis in vivo. Overexpression of JAM-A induced a change in HUVECs morphology similar to that observed when treated with FGF-2.¹⁷ Furthermore, overexpression of JAM-A, but not its cytoplasmic domain deletion mutant, augmented cell migration in the absence of FGF-2.¹⁷ In addition, downregulation of JAM-A in HUVECs using specific siRNA, resulted in reduced FGF-2-induced cell migration and inhibition of mitogen activated protein (MAP) kinase activation.¹⁸ These findings clearly suggested that JAM-A positively regulates FGF-2-induced endothelial cell migration. This was further confirmed in vivo by using JAM-A null mouse in which FGF-2 failed to support angiogenesis.¹⁹It is known that JAM-C, a JAM family member, is involved in the process of tumor cell metastasis.²⁰ However, little is known about JAM-A''s role in cancer progression. We recently found that JAM-A is expressed in breast cancer tissues and cell lines.²¹ Based on our studies with endothelial cells it was felt that JAM-A expression in breast cancer cells may also enhance the migratory ability of these cells. Surprisingly, we found an inverse relation between the expression of JAM-A and the metastatic ability of breast cancer cells. T47D cells, which express high levels of JAM-A, are the least migratory; whereas MDA-MB-231 cells, which are highly migratory, are found to express the least amount of JAM-A.²¹ We also found that overexpression of JAM-A in MDA-MB-231 cells caused a change in cell morphology from spindle-like to rounded shape and formed cobblestone-like clusters.²¹ This is consistent with the previous report, that downregulation of JAM-A expression from epithelial cells using siRNA results in the change of epithelial cell morphology.¹⁵ This change in cell morphology by knockdown of JAM-A was attributed to the disruption of epithelial cell barrier function.¹⁵ It was further shown that knockdown of JAM-A affects epithelial cell morphology through reduction of β₁integrin expression due to decreased Rap1 activity.¹⁵ Our observed effect of JAM-A downregulation in T47D cells, however, is not due to downregulation of β₁integrin, since the level of this integrin was not affected in these cells. Interestingly, overexpression of JAM-A significantly affected both the cell migration and invasion of MDA-MB-231 cells. Furthermore, knockdown of JAM-A using siRNA enhanced invasiveness of MDA-MB-231 cells, as well as T47D cells.²¹ The ability of JAM-A to attenuate cell invasion was found to be due to the formation of functional tight junctions as observed by distinct accumulation of JAM-A and ZO-1 at the TJs and increased transepithelial resistance. These results identify, for the first time, a tight junctional cell adhesion protein as a key negative regulator of breast cancer cell migration and invasion.²¹JAM-A has been shown to be important in maintaining TJ integrity.^15,22–25 Disruption of TJs has been implicated to play a role in cancer cell metastasis by inducing epithelial mesenchymal transition.²⁶ Several laboratories, including ours, have shown that cytokines and growth factors redistribute JAM-A from TJs.^16,27,28 Consistent with this finding, it has been shown that hepatocyte growth factor (HGF) disrupts TJs in human breast cancer cells and downregulates expression of several TJ proteins.²⁹ It is therefore conceivable that the loss of JAM-A in highly metastatic cells is a consequence of disruption of TJs. This was further supported by the findings that overexpression of JAM-A forms functional TJs in MDA-MB-231 cells and attenuates their migratory behavior. Our result is the first report correlating an inverse relationship of JAM-A expression in breast cancer cells to their invasive ability.²¹Using cDNA microarray technology, it has been revealed how genes involved in cell-cell adhesion, including those of the TJ, are under or overexpressed in different carcinomas.^15,30 Cell-cell adhesion molecules have been well documented to regulate cancer cell motility and invasion. Of these, the cadherin family have been studied the most.^31,32 It was proposed that a cadherin switch, that is, the loss of E-cadherin and subsequent expression of N-cadherin, may be responsible for breast cancer cell invasion.^33,34 Although the role of cadherins is well-documented, it remains controversial since some breast cancer cell lines that do not express these proteins still posses highly invasive characteristics.^33,34 However, the observed effect of overexpression of JAM-A does not appear to be simply due to the formation of TJs, since individual cells that express increased JAM-A show reduced migration.²¹ This is not surprising, considering the fact that JAM-A in addition to its function of regulating TJ integrity is also shown to participate in intracellular signaling. JAM-A is capable of interacting homotypically as well as heterotypically on the cell surface.^35,36 It has also been shown that it interacts with several cytoplasmic proteins through its PDZ domain-binding motif and recruits signaling proteins at the TJs.³⁷ Recent findings using site-directed mutagenesis suggest that cis-dimerization of JAM-A is necessary for it to carry out its biological functions.³⁸ Our own observations suggest that a JAM-A function-blocking antibody inhibits focal adhesion formation in endothelial cells (unpublished data), whereas overexpresion of JAM-A in MDA-MB-231 cells show increased and stable focal adhesions.²¹ It is therefore conceivable that in quiescent endothelial/epithelial cells JAM-A associates with integrin to form an inactive complex at the TJ (Fig. 2). Growth factors such as FGF-2 signaling dissociates this complex thus allowing dimerization of JAM-A and activation of integrin augmenting cell migration (Fig. 2). On the contrary, in MDA-MB-231 cancer cells, which express low levels of JAM-A and do not form tight junctions, there may not be efficient inactive complex formation between JAM-A and integrin. Overexpression of JAM-A in these cells however, may promote such inactive complex formation leading to inhibition of integrin activation and JAM-A dimerization, both necessary events for cell migration. We are currently in the process of determining the specificity of interaction of JAM-A with integrins. Further experimentation is ongoing to determine the contribution of JAM-A dependent signaling in cell migration.Open in a separate windowFigure 2Schematic representation of JAM-A regulation of cell migration. JAM-A forms an inactive complex with the integrin and sequesters it at the TJs. Growth factor signaling dissociates this complex, promoting integrin activation and JAM-A dimerization leading to cell migration via MAP kinase activation. Ectopic expression of JAM-A in cancer cells may induce its association with integrin, forming an inactive complex and hence attenuation of migration.JAM-A differentially regulates cell migration in endothelial and cancer cells due to its ability to form inactive complex with integrin, making it a metastasis suppressor. The downregulation of JAM-A in carcinoma cells may be detrimental to the survival of breast cancer patients. It is therefore very important to determine the molecular determinants that are responsible for the downregulation of JAM-A during cancer progression. Thus, JAM-A, a molecule that dictates breast cancer cell invasion, could be used as a prognostic marker for metastatic breast cancer.     

Myosin light chain kinases: Division of work in cell migration

Cell motility is a highly coordinated multistep process. Uncovering the mechanism of myosin II (MYO2) activation responsible for the contractility underlying cell protrusion and retraction provides clues on how these complementary activities are coordinated. Several protein kinases have been shown to activate MYO2 by phosphorylating the associated myosin light chain (MLC). Recent work suggests that these MLC kinases are strategically localized to various cellular regions during cell migration in a polarized manner. This localization of the kinases together with their specificity in MLC phosphorylation, their distinct enzymatic properties and the distribution of the myosin isoforms generate the specific contractile activities that separately promote the cell protrusion or retraction essential for cell motility.Key words: myosin, MLCK, ROK, MRCK, phosphorylation, cell migrationCell movement is a fundamental activity underlying many important biological events ranging from embryological development to immunological responses in the adult. A typical cell movement cycle entails polarization, membrane protrusion, formation of new adhesions, cell body translocation and finally rear retraction.¹ A precise temporal and spatial coordination of these separate steps that take place in different parts of the cell is important for rapid and efficient movement.²One major event during eukaryotic cell migration is the myosin II (MYO2)-mediated contraction that underlies cell protrusion, traction and retraction.^1,3 An emerging theme from collective findings is that there are distinct myosin contractile modules responsible for the different functions which are separately regulated by local myosin regulatory light chain (MLC) kinases. These kinases contribute to contractile forces that connect adhesion, protrusion and actin organization.² Unraveling the regulation of these contractile modules is therefore pivotal to a better understanding of the coordination mechanism.At the lamellipodium, the conventional calcium/calmodulin-dependent myosin light chain kinase (MLCK) has been shown to play an essential role in a Rac-dependent lamellipodial extension.⁴ Inhibition of calmodulin or MLCK activity by specific photoactivatable peptides in motile eosinophils effectively blocks lamellipodia extension and net movement.⁵ Furthermore, there is a strong correlation between activated MLCK and phosphorylated MLC within the lamellipodia of Ptk-2 cells as revealed by fluorescence resonance energy transfer (FRET) analysis.⁶ More recent studies showed MLCK to regulate the formation of focal complexes during lamellipodia extension.^7,8 Functionally, MLCK is thought to play a critical role in the environment-sensing mechanism that serves to guide membrane protrusion. It mediates contraction that exerts tension on integrin-extracellular matrix (ECM) interaction, which, depending on the rigidity of the substratum, will lead to either stabilization of adhesion resulting in protrusion or destabilization of attachment seen as membrane ruffling on non-permissive surfaces.^8,9As a Rho effector, Rho-associated kinase (ROK/ROCK/Rho-kinase) has been shown to regulate stress fibers and focal adhesion formation by activating myosin, an effect that can be blocked by the specific ROK inhibitor Y-27632.^10,11 Myosin activation by ROK is the effect of two phosphorylation events: the direct phosphorylation on MLC and the inhibition of myosin phosphatase through phosphorylation of its associated myosin-binding subunit (MBS).¹¹ Consistent with this notion of a localization-function relationship, ROK and MBS, which can interact simultaneously with activated RhoA,¹¹ have been shown to colocalize on stress fibers.^12,13 In migrating cells, Rho and ROK activities have been mostly associated with the regulation of tail retraction, as inhibition of their activities often results in trailing tails due to the loss of contractility specifically confined to the cell rear.^14,15 Tail retraction requires high contractile forces to overcome the strong integrin-mediated adhesion established at the rear end, an event which coincides with the strategic accumulation of highly stable and contractile stress fibers that assemble at the posterior region of migrating cells.MRCK was previously shown to phosphorylate MLC and promote Cdc42-mediated cell protrusion.¹⁶ More recently, it was found to colocalize extensively with and regulate the dynamics of a specific actomyosin network located in the lamella and cell center, in a Cdc42-dependent manner but independent of MLCK and ROK.¹⁷ The lamellar actomyosin network physically overlaps with, but is biochemically distinct from the lamellipodial actin meshwork.^9,18 The former network consists of an array of filaments assembled in an arrangement parallel to the leading edge, undergoing continuous retrograde flow across the lamella, with their disassembly occurring at the border of the cell body zone sitting in a deeper region.^17–19 Retrograde flow of the lamellar network plays a significant role in cell migration as it is responsible for generating contractile forces that support sustained membrane protrusion and cell body advancement.^17–19It is therefore conceivable that these three known MLC kinases are regulated by different signaling mechanisms at different locations and on different actomyosin contractile modules. The coordination of the various modules will ensure persistent directional migration (Figure 1). Phosphorylation of MLC by PAK and ZIP kinase has also been reported, but their exact roles in this event have yet to be determined.^20,21 It is also noteworthy that individual kinases can work independently of each other, as amply shown by evidence from inhibitor treatments. This is particularly true for MRCK in the lamella, whose activity on lamellar actomyosin flow is not affected by ML7 and Y-27632, the inhibitors of MLCK and ROK respectively.¹⁷ These findings further indicate that although both ROK and MRCK have been shown to upregulate phosphorylated MLC levels by inhibiting the myosins phosphatases,^11,22 they are likely to act as genuine MLC kinases themselves, without the need of MLCK as previously suggested.¹¹Open in a separate windowFigure 1Upper panel depicts a model for the specific activation of the different MLC kinases at various locations in the cell. In response to upstream signals, MLC kinases MLCK, MRCK and ROK are activated and localized to different regions. In the case of MRCK and ROK, the interaction of the GTP-bound Rho GTPase binding domain will determine the specific action of the downstream kinase, resulting in actomyosin contractility at different locations. The coordination of these signalling events is crucial for directional cell migration. Lower panel shows a typical front-rear location for Myosin 2A and 2B in a migrating U2OS cell.In conjunction with their differences in localization, the three MLC kinases show apparent individual preferences and specificity towards the MYO2 isoforms that they associate with. The two major MYO2 isoforms MYO2A and 2B are known to have distinct intracellular distributions that are linked to their individual functions (Figure 1).^23,24 In motile cells, MYO2A localization that is skewed towards the protruding cell front is consistent with it being the major myosin 2 component of the lamellar filaments regulated by MRCK as well as its regulation by MLCK in lamellipodial contraction.^8,17,19 In contrast, the enrichment of MYO2B at retracting cell rear conforms well with the requirement of thick and stable stress fibers capable of causing tail contraction and prevention of protrusion under the control of Rho/ROK signaling.^23,25 The selection for MYO2B filaments in the cell rear stems from their more contractile and stable nature compared with MYO2A, a consequence of their higher time-averaged association with actin.^26,27 Conversely, the lower tension property of MYO2A filaments suggests that they are more dynamic in nature,^26,27 a characteristic which fits well with the dynamic actomyosin activities at the leading edge and lamella that regulate protrusion.It deserves special mention that the three MLC kinases display subtle differences in their specificity towards MLC. While MLCK and MRCK phosphorylate only a single Ser19 site (monophosphorylation),^18,28 ROK is able to act on both Thr18 and Ser19 residues causing diphosphorylation of MLC,²⁹ MLCK only causes diphosphorylation when present at higher concentrations.³⁰ By further increasing its actin-activated ATPase activity, diphosphorylation of MLC has been shown to induce a higher myosin activation and filament stability.^30–32 The use of specific antibodies that can differentiate between the two populations of phosphorylated MLC has been instrumental in revealing their localization and correlation with the activity of the MLC kinases. The emerging picture from these experiments is that mono and diphosphorylated MLC exhibit distinct distributions in migrating cells, with the monophosphorylated MLC localized more towards the protrusive region, while the diphosphorylated form is more enriched at the posterior end.^21,33 Taking into account their biochemical properties, the polarized distributions of these differentially phosphorylated MLC coincide functionally with the segregation of the MYO2 isoforms and their corresponding regulators. These findings provide further support for the existence of segregated contractile modules in migrating cell and their distinctive regulation.The mechanisms that determine the specific segregation of the contractile modules and their regulation are unclear. However, some clues have emerged from recent studies. It has been shown that the C-terminal coiled-coil region of MYO2B is important for determining its localization in cell rear²⁵ and which requires Rho/ROK activity as their inhibition resulted in the loss of this specific localization.²³ Correspondingly, the inhibition of MRCK activity resulted in the loss of lamella-localized MYO2A.¹⁷ These findings suggest that activation of MYO2 filaments by their upstream regulators is important for their functional segregation and maintenance. It is noteworthy that both ROK and MRCK have distinct regulatory domains including the pleckstrin homology domains which have been shown to be essential for their localization, a process which may involve myosin interaction and lipid-dependent targeting as has been respectively shown for ROK and MRCK.^11,13,16 Further, the specificity of MRCK for lamellar actomyosin is believed to be largely determined by the two proteins it forms a complex with: the adaptor LRAP35a, and the MYO2-related MYO18A. Activation of MYO18A by MRCK, a process bridged by LRAP35a, is a crucial step which facilitates MRCK regulation on lamellar MYO2A.¹⁷The mechanisms responsible for segregating the contractile modules and their regulators may also comprise a pathway that parallels the microtubule-modulatory Par6/aPKC/GSK3β signalling pathway which regulates cellular polarization. This notion is supported by both Cdc42 and Rho being common upstream regulators of these two pathways.³⁴ GTPase activation may determine the localized activities of the separate contractile modules and create an actomyosin-based asymmetry across the cell body, which together with the microtubule-based activities, result in the formation of a front-back axis important for directional movement. The involvement of MRCK in MTOC reorientation and nuclear translocation events,³⁵ and our unpublished observation that LRAP35a has a GSK3β-dependent microtubule stabilizing function are supportive of a possible cross-talk between these two pathways.In conclusion, the complex regulation of contractility in cell migration emphasizes the importance of the localization, specificity and enzymatic properties of the different MLC kinases and myosin isoforms involved. The initial excitement and confusion caused by the emergence of the different MLC kinases are fading, being now overtaken by the curiosity about how they cooperate and are coordinated while promoting cell motility.