JXB Advance Access originally published online on December 5, 2006
Journal of Experimental Botany 2007 58(3):361-376; doi:10.1093/jxb/erl220
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Review Article |
Inositol trisphosphate receptor in higher plants: is it real?
ej Krinke1,3
1Department of Biochemistry and Microbiology, Institute of Chemical Technology Prague, Technická 3, 166 28 Prague 6, Czech Republic
2Institute of Experimental Botany, The Academy of Sciences of the Czech Republic, Rozvojová 135, 165 02 Prague 6, Czech Republic
3Université Pierre et Marie Curie-Paris 6 and Centre National de la Recherche Scientifique, Formation de Recherche en Evolution 2846, Laboratoire de Physiologie Cellulaire et Moléculaire des Plantes, Ivry-sur-Seine, F-94200 France
* To whom correspondence should be addressed. E-mail: martinec{at}ueb.cas.cz
Received 31 January 2006; Revised 22 September 2006 Accepted 2 October 2006
 |
Abstract |
|---|
Top
Abstract
IntroductionThe receptor for D-myo-inositol 1,4,5-trisphosphate (InsP3-R) has been well documented in animal cells. It constitutes an important component of the intracellular calcium signalling system. Today the corresponding genes in many species have been sequenced and the antibodies against some of the InsP3-Rs are available. In contrast, very little is known about its plant counterpart. Only a few published works have dealt directly with this topic. This review summarizes the available relevant data and determines some properties of putative plant receptor(s) including the in silico search for its gene in plant genomes, in vivo evidence, its electrophysiology, the parameters of InsP3-induced calcium release and InsP3 binding, immunological cross-reactivity, and subcellular localization. Future progress in this area seems to be inevitable as, despite the efforts, its gene in plants has not been identified yet.
Key words: Ca2+ signalling, higher plants, inositol trisphosphate receptor, ligand-gated Ca2+ channels
|
Introduction |
|---|
The activation of phospholipase C leads to the hydrolysis of phosphatidylinositol 4,5-bisphosphate (Fig. 1) and the production of D-myo-inositol 1,4,5-trisphosphate (InsP3). InsP3 as a second messenger in the cytoplasm acts on its specific receptor/Ca2+-permeable ion channel on endomembranes. Opening of the Ca2+-permeable channel temporarily increases the cytosolic concentration of Ca2+ ([Ca2+]cyt) near the channel mouth. The specific spatiotemporal pattern of the calcium signal is then the key to the physiological response of the cell. In the case of the inositol trisphosphate receptor (InsP3-R), this elevation is more like a trigger for opening other ion channels than the principal response to InsP3. The InsP3-induced Ca2+ release (IICR) is well documented in animal systems; it has also been reported in fungi (Cornelius et al., 1989; Belde et al., 1993) and in a number of plant species during the last two decades.
|
The properties of InsP3-R in plants are reviewed and discussed with respect to its animal counterpart. This further supports the idea that the phosphoinositide signalling system in plants and animals is quite similar. No matter how much evidence for the phosphoinositide signalling pathway in plants has been gathered, the final step has not been realized as the gene corresponding to plant InsP3-R has not yet been identified.
|
In silico search for the InsP3-R gene in plants |
|---|
The first step in characterizing the InsP3-R should be looking for its gene. From the time when the first plant genome sequencing project was completed, it has become clear that the search for the plant InsP3-R gene will not be an easy task. To date (September 2006), no plant gene has been annotated as InsP3-R. Lin et al. (2004), in an article dealing with the mapping of the inositide signalling pathways in plants by the DNA microarray technique, published two AGI annotations supposed to be the InsP3-R genes in Arabidopsis, but the first one (At3g10380) is now annotated as a probable component Sec8 of the exocyst complex and the second one (At5g27230) is a membrane electron transport protein according to its domain composition. A simple NCBI Blast search (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi) for the InsP3-R protein sequence found no plant protein bearing a significant homology. A more sensitive method for detection of remote protein homologies is based on the profile hidden Markov model algorithm (Eddy, 1998). The profile hidden Markov model turns a multiple sequence alignment into a position-specific scoring system suitable for searching databases for remotely homologous sequences. Profile hidden Markov model analyses complement standard pairwise comparison methods for large-scale sequence analysis. After aligning the input sequences, the algorithm generates a set of consensual motifs typical for the protein family. These sequence motifs are then compared with the library of already annotated protein domains. It is a way to run ‘Blast’ using only important traits of the studied protein family. Generating a homology profile for the three isoforms of the rat InsP3-R using a profile hidden Markov model algorithm and searching all the protein databases (http://motif.genome.jp/MOTIF2.html) with this profile did not give any significant hits for any plant proteins.
The mammalian InsP3-R is a surprisingly large molecule (
2700 amino acids) and has a well-defined domain structure, shown in Fig. 2. The so-called RIH [ryanodine receptor (RyR) and InsP3-R homology] domain is very specific for InsP3-R and for its close relatives, the RyRs. Based on the presence of this domain, no homologous protein in plants has been found in InterPro (http://www.ebi.ac.uk/interpro/) or the Conserved Domain Database (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). The presence of an ion transport domain is more general, and several proteins possessing a similar domain can be found in Arabidopsis and rice genomes, but those represent mostly the genes previously annotated as K+ channels with a quite different domain architecture from what would be expected for a true InsP3-R homologue. The lack of domain homology with any plant protein implies that IICR in plants is on a different molecular basis from that in animals. Despite the failure of the sequence database mining, it is very likely now that IICR exists in plants. The in vivo evidence for this mechanism is quite convincing.
|
|
In vivo evidence for InsP3-induced Ca2+ release in plants |
|---|
Before considering the in vitro properties of the putative InsP3-R, it should be clear that InsP3 has some physiological relevance in vivo. The biochemistry of inositol phosphates in plants differs somewhat from that described in animal systems (van Leeuwen et al., 2004) and has been reviewed from the metabolic (Drøbak, 1992; Loewus and Murthy, 2000) and the signalling point of view (Coté and Crain, 1993; Stevenson et al., 2000). Metabolism of InsP3 in plants was also studied (Coté et al., 1987; Morse et al., 1987; Drøbak et al., 1991; Brearley and Hanke, 1996a, b; DePass et al., 2001). Changes in the intracellular InsP3 level have been reported in many species of higher plants in response to various stimuli such as light, cold, gravistimulation, oxidative stress, hyperosmotic stress, plant hormones, G-protein activation, and pathogenic elicitors (for a summary, see Table 1). The best documented system seems to be the abscisic acid (ABA)-stimulated stomatal closure which was thoroughly studied in this context (Staxen et al., 1999; Hunt et al., 2003). Release of caged InsP3 was shown to cause Ca2+ influx in growing Agapanthus pollen tubes (Monteiro et al., 2005). A good review on how the sustained increase of InsP3 and Ca2+ leads to gravitropic growth response can be found in Stevenson et al. (2000). Plants often show elevated InsP3 levels upon stimulation. This suggests InsP3 to be a positive regulator of many signalling pathways. Nevertheless it has been shown in in vivo studies that the phosphoinositide turnover in higher plants is quite rapid (van der Luit et al., 2000) and some steady-state level of InsP3 is thus present even in non-stimulated cells. This InsP3 level may serve as a constitutive repressor or enhancer of gene expression, and its decrease would also lead to altered gene expression, as was shown in mutants with a constitutively decreased InsP3 level (Perera et al., 2002, 2006; Burnette et al., 2003).
|
Changes in InsP3 levels have often been linked with changes in activity of phosphoinositide-specific phospholipase C (PI-PLC). The role of this enzyme in plants has been reviewed several times (Munnik et al., 1998; Mueller-Roeber and Pical, 2002). An exhaustive study of the tissue and stress expression pattern of all nine Arabidopsis PI-PLC isoforms with suggestions as to how they may intervene in various stress responses was recently published by Hunt et al. (2004). Its mode of catalysis, domain structure, and possible cellular functions were reviewed by Wang (2001) and later in a more general context of phospholipid signalling (Wang, 2004). The fact that the InsP3 concentration in vivo varies upon stimulation points to its significance in plant cell signalling.
It was also necessary to demonstrate that elevation of the cytoplasmic concentration of InsP3 triggers some physiological response in living cells. This was done mainly by the photolysis of the caged InsP3 or by microinjection of InsP3. Release of InsP3 from its caged form in stomatal guard cells of Commelina communis led to an increase of [Ca2+]cyt and to the closure of stomata (Gilroy et al., 1990, 1991). A similar experiment was done with guard cells of Vicia faba (Blatt et al., 1990). They showed that the application of exogenous InsP3 reversibly inactivates the plasma membrane-located inward rectifying K+ channel in guard cells by releasing calcium into the cytoplasm. Microinjections of free InsP3 in algal cells resulted in a change in the plasma membrane conductance (Förster, 1990; Thiel et al., 1990). Microinjection of InsP3 was used to prove the cell–cell communication through oscillations of [Ca2+]cyt (Tucker and Boss, 1996).
Additional data about the importance of InsP3 have been gathered thanks to the characterization of mutants with constitutively increased levels of InsP3 (Xiong et al., 2001; Carland and Nelson, 2004) or transgenic plants with constitutively decreased InsP3 levels due to overexpression of InsP3 5-phosphatase (Perera et al., 2002, 2006; Burnette et al., 2003). One of the mutants with a constitutively increased InsP3 level is fry1, a loss-of-function mutant in a bifunctional enzyme which exhibits both 3'(2'), 5'-bisphosphate nucleotidase and inositol polyphosphate 1-phosphatase activities (Xiong et al., 2001). The other mutant is cvp2 which is affected in inositol polyphosphate 5-phosphatase activity. These two inositol polyphosphate phosphatase activities probably act on different levels of InsP3 signal attenuation, as proposed by Xiong et al. (2002), where the inositol polyphosphate 5-phosphatase acts directly on InsP3 and FRY1 hydrolyses the resulting Ins(1,4)P2. Slowing down the InsP3 breakdown on each level causes stress hypersensitivity, especially hypersensitivity to ABA, in these mutants. Consistently, the inositol polyphosphate 5-phosphatase gain-of-function mutant showed insensitivity to ABA (Burnette et al., 2003). Both of these observations indicate a role for InsP3 in ABA signalling. Besides ABA signalling, a role in cold and hyperosmotic stress was proposed for FRY1, and premature vascular termination was observed in cvp2. The other phenotypes reported for these mutants may be caused by the other enzymatic activity of FRY1 in the case of fry1, by very specific tissue localization of CVP2 in the case of cvp2, and by their intervention on different levels of InsP3 catabolism. Based on these data, it is clear that InsP3 elicits a physiological response in plants probably by mobilizing Ca2+ from either intracellular or extracellular stores. However, a role for other inositol polyphosphates in Ca2+ mobilization cannot be excluded.
|
Potential role of InsP6 in intracellular calcium signalling |
|---|
In a very few cases, InsP6 (known as phytate) was shown to be involved in plant cell signalling as an ion channel agonist. Lemtiri-Chlieh et al. (2000) showed that InsP6 is active in inhibiting the inward rectifying K+ ion channel in a Ca2+-dependent manner when delivered through patch electrode to guard cells in submicromolar concentrations. This inhibitory effect is very specific to myo-InsP6 which is 100-fold more potent than InsP3. The effect was documented in the response of guard cells to ABA in at least two distinct species of higher plants (Solanum tuberosum and V. faba). Lemtiri-Chlieh et al. (2003) later confirmed these findings by the release of caged InsP6, and found that it also releases Ca2+ from internal stores and not from extracellular space through the plasma membrane.
It is clear that this Ca2+-releasing pathway is unique to plants, as no data for InsP6-induced Ca2+ release in animal species have been published so far. Although the temptation exists to interpret InsP3 findings as hidden InsP6 action, considering the in vitro studies of IICR it is more likely that these two pathways exist in parallel and this dichotomy may have some physiological relevance only in specific cell types such as the guard cells.
|
In vitro measurements of electrophysiological properties of the InsP3-induced calcium release |
|---|
Once the relevance of the in vivo InsP3 action has been shown, the next step is to prove the existence of such an ion channel in vitro. The conductometric measurement of ion channels (patch-clamp technique) is the most reliable one, but it can be performed only with some limitations of the biological material, especially the need for large membrane patches, which can be achieved only with plasma membrane of protoplasts or with large plant vacuoles. That might be the reason why these experiments have been successfully accomplished with only one type of membrane, the tonoplast from Beta vulgaris storage root. Alexandre et al. (1990) found that the channel has a conductance of 30 pS (after activation by 1 µM InsP3) at –80 mV (referenced to the vacuolar lumen) and it can be opened only by depolarization (i.e. it is voltage dependent). The Ca2+ transport was oriented and required outside-out patches with a higher Ca2+ concentration inside the membrane vesicles. The density of Ca2+ channels was estimated to be
1200 per vacuole of an average diameter 45±5 µm. Alexandre and Lassalles (1990) showed that by the membrane depolarization the InsP3-released Ca2+ closes the non-specific ion channels on the tonoplast. They also speculated that the depolarization might be sometimes more physiologically important than the increase of the [Ca2+]cyt itself. Alexandre and Lassalles (1992) discovered a higher level of the single channel conductance of 50 pS. Allen and Sanders (1994) found three levels of the single channel conductance of 11, 51, and 182 pS at –80 mV. This discrepancy was not satisfactorily explained. The whole vacuole ion permeability ratio was estimated as 200:1 (PCa:PK). The permeability ratio of a single channel was estimated to be in the range from 100:1 to 800:1 (PCa:PK). The channel opening was shown to be independent of the [Ca2+]cyt and reversible, although the ligand dissociation was very slow. The Ca2+ current of the whole vacuole patch-clamp was also shown to be cytosol directed and appreciably enhanced after hyperosmotic pre-treatment of the tissue. This might originate in de novo synthesis or in functional phosphorylation of the InsP3-R, or simply in more efficient extraction of the large InsP3-responsive vacuoles from the shrunken cytoplasm. The need for hyperosmotic pretreatment could explain the negative findings published for the vacuoles of the B. vulgaris cell suspension culture (Gelli and Blumwald, 1993). In contrast to the finding of Allen and Sanders (1994) on B. vulgaris, the IICR in Acer pseudoplatanus was found to be stimulated by a decrease of the osmotic pressure. This stimulation was explained by the expansion of the vacuolar surface which made the InsP3-R molecules more accessible and/or changed the conformation of the receptor to the active form. No Ca2+ channel opening upon 2–8 µM InsP3 stimulation at –120 mV occurred in the patch-clamp experiments with the tonoplast from tobacco cells (Nicotiana tabacum L. cv. BY-2) carried out by Ping et al. (1992).
The positive findings for the InsP3-R were later reviewed in the context of the cyclic adenosine diphosphate-ribose (cADPR)-induced Ca2+ release. The release was shown to be additive in the whole vacuole patch-clamp of B. vulgaris. The additivity was confirmed by the radiometric measurements; both ligands could release 15% of the releasable Ca2+ (Allen et al., 1995). Both of the mechanisms are assumed to trigger the calcium-induced calcium release (CICR) in plants. The CICR itself is then performed by the slow vacuolar (SV) channels during the depolarization of the membrane caused by the opening of either the InsP3-R itself (Sanders and Johannes, 1990) or of the tonoplast Ca2+-activated K+ channels. The InsP3- and cADPR-induced Ca2+ currents were not spontaneously deactivated, which means that the closure of the channels is induced by the voltage change, and later the metabolic degradation of the ligand occurs, which prevents further Ca2+ channel reactivation (Allen et al., 1995).
Muir et al. (1997) failed to measure the IICR from the whole vacuole of guard cells and from the whole vacuole of cells from the inflorescence of Brassica oleracea L. using the patch-clamp technique under the same conditions used for B. vulgaris. The cause of the failure with the guard cells could be the small diameter of the vacuole. The increase in current under the given conditions would be far below the resolution of the technique. The failure in the case of vacuoles from the inflorescence of B. oleracae could originate in the different subcellular localization of InsP3-R in this plant tissue, as documented elsewhere (Muir and Sanders, 1997).
|
In vitro estimated parameters of the InsP3-induced Ca2+ release |
|---|
More detailed insight into the IICR can be obtained by looking at the effect on Ca2+ transport after application of some Ca2+ channel blockers or InsP3 isomers. The in vitro Ca2+ transport can be measured by the previously discussed patch-clamp approach or by monitoring Ca2+ transport using radioactive 45Ca2+ or using Ca2+-specific fluorescent dyes.
In animal cells, the IICR was shown to be a quantal process which means that the submaximal dose of the ligand does not fully empty the Ca2+ store even after a longer period of time. This phenomenon can be explained by the voltage-dependent channel closing described above, by the relatively slow dissociation of InsP3 from InsP3-R, or by the [InsP3]-driven ratio of the InsP3-Rbound/InsP3-Rfree molecules. The result of these processes is simple Michaelis–Menten saturation kinetics of the IICR. An apparent saturation constant (IC50) that defines the concentration of InsP3 which gives half-maximal Ca2+ release is one of the basic features of IICR. The second one is the percentage of mobilizable Ca2+. These characteristics for various plant materials are summarized in Table 2.
|
Johannes et al. (1992a) found that the InsP3-regulated Ca2+ channels represent only a minor portion of the Ca2+ channels residing on the tonoplast of B. vulgaris, the majority of those Ca2+ channels being voltage gated and InsP3 independent. Alexandre et al. (1990) showed that the InsP3 binding in B. vulgaris exhibits no cooperativity, unlike the binding to its animal analogue. The values of the IC50 differ greatly, ranging from 0.2 to 15 µM (Table 2); the lower values are in accord with those reported for animal InsP3-R, but the high values for Zea mays L. coleoptiles, Daucus carota L. cell suspension culture, and Cucurbita pepo L. hypocotyls were neither reproduced later nor satisfactorily explained. However, Drøbak and Ferguson (1985) warned that their InsP3 preparation was probably contaminated by a considerable amount of Ins(2,4,5)P3. The amount of released Ca2+ was usually in the range of a few nanomoles per mg of protein. The percentage of the InsP3-releasable Ca2+ normally ranges from 10% to 20% for the microsomal fraction and from 20% to 40% for the tonoplast-enriched microsomal fraction. A relatively small part of the mobilizable Ca2+ can be released by InsP3. The density of InsP3-binding sites in the microsomal fraction mirrors to some extent the density of InsP3-R. This value, called Bmax, is generally expressed in pmoles of specifically bound InsP3 per mg of proteins in the studied sample. In animal cerebellum, this value ranges from 10 to 70 depending on the species and on the age of the animals (Simonyi et al., 1998; Vanlingen et al., 1999; Coquil et al., 2004). The relatively small portion of InsP3-releasable Ca2+ can be explained using the example of B. vulgaris (the corresponding value of Bmax can be found in Table 4). Its Bmax is at least 10 times lower than that of microsomes from animal brain. The relatively small number of InsP3-R molecules per vacuole leads, after the reconstitution of the tonoplast vesicles, to the fact that only one vesicle from about eight contains at least one InsP3-R molecule. This happens because the diameter of the reconstituted vesicle is
1% of that of the intact vacuole (Brosnan, 1990).
|
The animal InsP3-R is regulated by calmodulin (CaM), either free or bound to Ca2+, which inhibits InsP3 binding to the receptor; no direct regulation by Ca2+ was observed (Bultynck et al., 2003). The dependence of IICR on [Ca2+]free in plants was also examined in a few studies. In experiments with C. pepo L., the [Ca2+]free during the Ca2+ release was only 10 nM, indicating no regulation by calcium ions in this case. Allen and Sanders (1994), using the patch-clamp technique, found that the IICR in B. vulgaris is not affected whether the [Ca2+]free is 100 nM or 1 mM. However, the IICR in Z. mays was maximal at 100 nM [Ca2+]free. Even this finding is questionable because the dependence on [Ca2+]free was not bell-shaped as for the animal InsP3-R (Finch et al., 1991), and it was not measured for concentrations lower than 50 nM. This may mean that, in contrast to the animal InsP3-R, the plant InsP3-R might not be regulated by the [Ca2+]free.
A wide range of different inhibitors was used as a tool to study the IICR. The effect of some well-known inhibitors of Ca2+ release is reviewed in Table 3. Nifedipine, the mammalian L-type Ca2+ channel blocker, ryanodine, the sarcoplasmic reticulum Ca2+ channel blocker and the antagonist of RyR, and ruthenium red, the blocker of Ca2+ uptake by mitochondria and another antagonist of RyR, were all ineffective in inhibition of IICR in plants. Surprisingly caffeine, an activator of RyR, acted as an InsP3-R inhibitor. Thus it is not a good inhibitor either because of its dual specificity. EDC [1-ethyl-3-(3-dimethylaminopropyl)carbodiimide], a carboxyl amino acid modifier, was shown to abolish the IICR, thus confirming the need for free aspartate or glutamate residues for the proper function of the receptor (Samanta et al., 1993). Trifluoperazine, the CaM antagonist, was not a good inhibitor because of the existence of the Ca2+-ATPase in the tonoplast which is also sensitive to it (Pfeiffer and Hager, 1993; Lommel and Felle, 1997). To achieve inhibition comparable with animal InsP3-R, a relatively high concentration of TMB-8 [8-(N,N-diethylamino)octyl 3,4,5-trimethylbenzoate], an inhibitor of some endomembrane Ca2+ channels and protein kinase C, is required. This points to a lower specificity for plant Ca2+ channels than reported for those of animals. Verapamil, a Ca2+ channel blocker, was effective in B. vulgaris, but totally ineffective in A. pseudoplatanus. The reason could be the insufficient concentration used or the different nature of the receptor in A. pseudoplatanus. In A. pseudoplatanus, the dependence of the IICR on the presence of K+ was proved, as well as coupling of Ca2+ influx with K+ efflux (Canut et al., 1989). The coupling with K+ efflux was not reported for other plant species, which further supports the idea of the different nature of the IICR in A. pseudoplatanus compared with that of other higher plants. Taken together, none of the inhibitors mentioned above seems to be specific enough for plant InsP3-R, and their use in plants will probably not give meaningful results.
|
Low molecular weight heparin is a potent inhibitor of the IICR in plants and also competes more strongly than other inositol phosphates. The IC50 values are in close quantitative agreement with those reported for rat cerebellum (Worley et al., 1987; Challiss et al., 1991) although the affinity of InsP3-R from animal peripheral tissues for heparin is somewhat weaker (Guillemette et al., 1989; Tones et al., 1989). Heparin of higher molecular weight is a markedly less potent inhibitor of IICR, both in animals (Chopra et al., 1989) and in plants (Sanders et al., 1990; Johannes et al., 1992b). Some difficulties were observed in the case of heparin and the tonoplast of B. vulgaris because its competitive inhibitory effect on the Ca2+ release observed with 45Ca2+ (Brosnan and Sanders, 1990; Johannes et al., 1992b) was not confirmed by the patch-clamp techniques (Alexandre and Lassalles, 1992) as the inhibitory effect there was rather non-specific. In that work, it might have been a problem of inappropriate molecular weight of the heparin preparation used; unfortunately this information was missing. Nevertheless, the use of low molecular weight heparin may be a good tool for the study of plant IICR.
The IICR in Vigna radiata was also enabled by Ins(2,4,5)P3 (EC50=0.8 µM) but the maximal release thus gained was about three times lower than for InsP3, which is consistent with the findings of Alexandre et al. (1990) and with the data obtained for the animal InsP3-R. The IICR was stimulated when InsP3 was complexed with phytase (Samanta et al., 1993). Dasgupta et al. (1996) found that this enhancement comes from the presence of the allosteric site on phytase with high affinity for InsP3. This allosteric site changes the conformation of the phytase upon binding of InsP3, thus enabling its effective interaction with the InsP3-R.
|
In vitro estimated parameters of InsP3 binding |
|---|
The receptor part of the InsP3-R can also be characterized by the affinity of the ligand (InsP3) for the InsP3-binding site. However, unless the gene for the receptor is known, it cannot be excluded that the studied binding site does not belong to the InsP3-R molecule. Just the expression of the receptor part of the InsP3-R and a clear demonstration that its binding properties are the same as those of the natural samples can prove that it is in fact the receptor part of the InsP3-R that is studied under the given conditions.
The experiments are usually carried out as the [3H]InsP3 competitive displacement assay. It is supposed that, under certain conditions, only the specifically bound [3H]InsP3 can be displaced by the excess of the other ligand and the non-specifically bound [3H]InsP3 cannot as the non-specific binding is not saturable. The data for various plant materials are summarized in Table 4 together with the pH values used and with the densities of the binding sites per mg of protein (Bmax) where available.
These types of experiments are difficult to interpret because they are usually performed under non-physiological conditions (alkaline pH, high [Ca2+], absence of ATP, solubilized InsP3-R molecules) and their results cannot be interpreted in terms of the function of InsP3-R; they only give us limited information about the receptor part of the ion channel. However, this approach is very suitable for the evaluation of the purification process as it does not require maintaining the InsP3-R in its native conformation.
The IC50 values for InsP3 can be compared with the EC50 only in the case of B. vulgaris. The IC50 is lower perhaps because of no ATP in the InsP3 binding assay. It was previously shown that ATP, at the concentration used for the Ca2+ transport assay, has a considerable affinity for the InsP3-binding site in this plant (IC50=980 µM; Johannes et al., 1992b). The influence of ATP is further confirmed by the lower EC50 value obtained by the patch-clamp technique where ATP is also not present (Alexandre et al., 1990).
The affinity of plant InsP3-R for ATP observed in some studies is a property reflecting that of its rat counterpart where the IC50 for ATP ranges from 0.35 to 0.79 mM (Nunn and Taylor, 1990; Challiss et al., 1991). As ATP is present in millimolar concentrations in living cells, this affinity may also have some physiological consequences in vivo. This concentration of ATP anticipates the InsP3 binding to its receptor, and this might account for the higher than expected concentrations of InsP3 in animal cells (Nunn and Taylor, 1990). However, the higher InsP3 concentration in plants compared with animals may just be a simple consequence of the lower affinity of plant InsP3-R for InsP3 combined with the need for a comparable physiological impact of phosphoinositide signalling in plant cells. Another explanation could be that the true binding ligand is InsP6 and that the InsP3 serves only as a precursor for its synthesis, thus being more abundant in the plant cell and less efficient in receptor binding. Further studies where InsP3 and InsP6 would be studied in parallel could substantially contribute to resolving this dilemma. If both these ligands have a physiological role in plant cells, their mutual interaction and interpretation of the two different signals by the plant cell are definitely points of future interest. Plant InsP3 and more generally phosphoinositide signalling is one of the current topics under discussion, and more and more researchers tend to abandon the idea of simple transformation of the animal model of phosphoinositide signalling to plants (van Leeuwen et al., 2004).
The specific binding is usually calculated as the difference between the total and non-specifically bound [3H]InsP3. The non-specific binding usually accounted for 35–50% for the tonoplast-enriched microsomal fraction from B. vulgaris (Johannes et al., 1992b) and 10–50% for the microsomal fraction from Chenopodium rubrum (Scanlon et al., 1996) which is higher than that reported for the animal membrane preparations where it represents 5–30% (Wilcox et al., 1994; Vanlingen et al., 1999). By comparing the specific [3H]InsP3 binding to the microsomal fraction from the inflorescence of B. oleracea L., from the coleoptiles of Z. mays L., and from the storage root of B. vulgaris, Muir et al. (1997) found the inflorescence of B. oleracea L. to be the best material for further studies because it has the highest specific InsP3 binding under the given conditions. The polyethylene glycol (PEG) precipitation technique was criticized by Scanlon et al. (1996) due to the possible underestimation of Bmax values, and the rapid filtration method was suggested instead. However, the degree of this possible underestimation was not specified.
[Ca2+] was either not controlled in the above experiments or it was set to high non-physiological values (e.g. 10 mM in the case of C. rubrum; Scanlon et al., 1995, 1996). It was also found that EGTA acts as the antagonist of InsP3 binding either by decreasing the [Ca2+]free or by direct interaction with the InsP3-binding site (Richardson and Taylor, 1993; Scanlon et al., 1996).
The sulphydryl reagents 1 mM N-ethylmaleimide and 50 µM p-chloromercuribenzoic acid (PCMB) were shown to reduce the InsP3 binding in B. vulgaris (Brosnan and Sanders, 1993). This was further confirmed by the finding that 1 mM pCMBS (p-chloromercuribenzoylsulphonate), another sulphydryl reagent, abolishes the InsP3 binding in V. radiata (Biswas et al., 1995). Scanlon et al. (1996) found that it is critical to add at least 5 mM DTT (dithiothreitol) to ensure the reproducibility of the results with the microsomal fraction of C. rubrum, suggesting some role for the cysteine residues in the binding site of InsP3-R. In contrast, 400 µM TMB-8, an inhibitor of the IICR, did not alter the InsP3 binding in B. vulgaris (Brosnan and Sanders, 1993) which means that it does not act as a competitive inhibitor of InsP3 but rather acts directly on the Ca2+ channel part of the plant InsP3-R.
Samanta et al. (1993) described that phytase increased the affinity of inositol trisphosphates for the InsP3-R (it decreased their IC50). This finding was further confirmed by Dasgupta et al. (1996) who found that the phytase from V. radiata cotyledons has an allosteric site for InsP3/Ins(2,4,5)P3 and that the phytase with InsP3 bound to its allosteric site has a higher affinity for the InsP3-R than the InsP3 itself (the IC50 value for both InsP3 and Ins(2,4,5)P3 was 75±10 nM using the [3H]InsP3 competitive displacement assay). However, the physiological relevance of this finding remains unclear.
Employing the InsP3 binding assay, Biswas et al. (1995) found that the Kd(InsP3)=1.5 nM in V. radiata. This significant difference from the IC50 value might be explained by the fact that the [3H]InsP3 concentration used in this experiment (25 nM) was relatively high (the IC50 value approximately matches the Kd only when the [3H]InsP3 concentration is lower than 0.1xKd). The pellet after Triton X-100 solubilization was shown to inhibit the InsP3-specific binding by 30%, suggesting the modulation by some non-identified factor (perhaps present in the membrane microdomains; Peskan et al., 2000). Biswas et al. (1995) were also the first to report the purification of the InsP3-R to apparent homogeneity using affinity chromatography with heparin-modified agarose. Analysis of the purified InsP3-R revealed that the receptor is a homotetramer of 110 kDa subunits. The native molecular mass of the animal InsP3-R is 310 kDa, with the equivalent band of 260 kDa on SDS–PAGE. In vivo the animal receptor also functions as a homotetramer (Ferris and Snyder, 1992; Pozzan et al., 1994; Taylor and Traynor, 1995). The reported molecular mass of that plant InsP3-R is rather in the range of an InsP3-R from the olfactory cilia of the fish Ictalurus punctatus (Kalinoski et al., 1992).
The interaction of InsP3 with the InsP3-R in V. radiata was further studied by Dasgupta et al. (1997). He found that Ins(1,3,4)P3 and Ins(1,5,6)P3 do not significantly bind to the InsP3-binding site and that both InsP3 and Ins(2,4,5)P3 bind to this site, but interestingly each of them causes a different conformational change of the InsP3-R, leading to a different functional response (only InsP3 affects the membrane-spanning helical domain in the InsP3-R). The binding stoichiometry for InsP3 was ascertained to be one molecule of InsP3 per one subunit of InsP3-R.
The animal InsP3-R has supramicromolar affinity for InsP2, InsP4, and InsP6 (Worley et al., 1987; Challiss et al., 1991). However, this affinity has not been confirmed for the commercially available InsP2 (Maeda et al., 1990), pointing to the necessity for highly purified batches of inositol phosphates for this type of study. This is consistent with the finding that InsP2 failed to bind to the InsP3-R from B. vulgaris even at millimolar concentrations (Brosnan and Sanders, 1993), thus emphasizing its high specificity for InsP3.
|
Immunological cross-reactivity between the animal and putative plant InsP3-R |
|---|
The immunological cross-reactivity of the putative plant InsP3-R might point to some structural similarities with its animal counterpart and thus might serve as a valuable tool for purification of the corresponding plant homologue. Two studies examined this possibility. Muir and Sanders (1997) studied the cross-reactivity in the inflorescence of B. oleracea L. Using two different antibodies (T210, the polyclonal antibody raised against the C-terminal part of animal InsP3-R1; and NT, the polyclonal antibody raised against the N-terminal part of animal InsP3-R1), they have identified a 200 kDa protein in the microsomal fraction but numerous other cross-reactive smaller fragments were detected with both antibodies. The 200 kDa protein band on SDS–PAGE can be correlated with the size of the animal InsP3-R subunit of 260 kDa. However, the effort to perform a subcellular localization using these antibodies failed and only putative proteolytic fragments were detected in the tested membrane fractions.
Cramer et al. (1998) tested two different plant materials, leaves from V. faba L. and Z. mays L. Polyclonal antibody raised against the C-terminus of the animal InsP3-R (Calbiochem) interacted with a 260 kDa protein in the microsomal fraction and tonoplast of V. faba L., but not in the plasma membrane. A 252 kDa protein was also found in the microsomal fraction of Z. mays L. However, smaller cross-reactive fragments were also detected in both plant species.
Immunological cross-reactivity experiments have also been done in our laboratory (Feltlová, unpublished data) with C. rubrum as the plant material, but their outcome was not interpretable. Unfortunately, in no case were these cross-reactive proteins shown to co-purify with the IICR activity. This more or less general failure of efforts to immunolocalize the plant InsP3-R by antibodies raised against the animal receptor is most probably a good example of non-specific binding of antibodies raised against animal proteins. Antibodies raised against animal InsP3-R will certainly not be useful for purification of its plant homologue.
|
Subcellular localization of the putative plant InsP3-R |
|---|
The subcellular localization of InsP3-R was studied in a considerable number of publications, and various separation techniques were applied to reach this goal (the results are summarized in Table 5). It is clear that the InsP3-R is localized on the tonoplast in the majority of the plants; its presence on this membrane was proved by various techniques. It is not a very surprising finding as the vacuole is the largest intracellular Ca2+ store and, moreover, its volume represents >90% of the cytoplasm. In addition to this, evidence is emerging that there are other InsP3-sensitive Ca2+ stores in the plant cell. Some studies have pointed to the plasma membrane (Muir and Sanders 1997; Cramer et al., 1998), but these results may be artefacts to some extent because the plasma membrane is in close appositional contact with the membrane system of the cortical endoplasmic reticulum (ER) and this contact is further stabilized by the structures of the cytoskeleton (Hepler et al., 1990). This artefact can be removed by using the microfilament-disruptive drug cytochalasin B (Lièvremont et al., 1994). By preparing the ER-enriched fraction from the microsomes of C. rubrum, Martinec et al. (2000) showed that the specific [3H]InsP3 binding is co-localized with this fraction. It seems clear that in many plants there is another InsP3-sensitive Ca2+ store besides the central vacuole. Only the ER and plasma membrane bear a sufficient Ca2+ gradient (Bush, 1995); the role of the ER seems to be proved, but the possible involvement of the plasma membrane remains to be elucidated in the future.
|
The InsP3-R is probably localized in the membrane microdomains (Muir and Sanders, 1997) or its function is at least modified by some factor present in the microdomains (Biswas et al., 1995). In summary, although the IICR seems to be a well established Ca2+-releasing mechanism in plants, its subcellular localization has not been satisfactorily unravelled so far.
|
Conclusions and future perspectives |
|---|
In vivo evidence for IICR is strong and it should encourage researchers to persevere in their efforts in identifying the plant protein(s) and corresponding gene(s) responsible for it. The overwhelming majority of the experiments presented in this review were carried out with InsP3; nevertheless, a physiological role also undoubtedly exists for InsP6, at least in guard cells. In the light of this finding, some in vivo experiments (Gilroy et al., 1990, 1991) should be re-evaluated with consideration given to the potential conversion of InsP3 into InsP6. One can imagine engineering mutants compromised in InsP6 synthesis and their crossing with cvp2 to show that the cvp2 phenotype in these hybrids remains conserved and thus it is really a result of an elevated InsP3 level and not a result of a consecutively increased InsP6 level.
The electrophysiological approach has been used only by two laboratories and it gave some positive results only with B. vulgaris vacuoles. Even though this approach is the most reliable one, such limitations prevent some general conclusions about plant cell signalling to be made. The lack of published positive experimental data may point to a number of negative unpublished results which makes the whole issue even hazier. The parameters of IICR obtained by in vitro Ca2+ transport assay are quite reliable too as they measure the function of the putative InsP3-R, unlike the InsP3 binding assay which may mix up the InsP3-R with other InsP3-binding proteins (e.g. InsP3-metabolizing enzymes).
Immunological data are quite confusing, reporting fragments of various molecular masses which makes one doubt about any relevant structural similarities between the animal and plant InsP3-R. This heterogeneity is further supported by the failure of the in silico approach. Subcellular localization studies point mainly to the vacuole as the prevalent InsP3-sensitive Ca2+ store, but even at this point there is some reliable evidence for other localizations such as the ER. The observed discrepancies may originate in the diversity of the studied plant materials or in their different developmental stages.
All presented studies are based on those performed earlier on the animal InsP3-R. It is clear that the animal receptor has higher affinity for InsP3 (between 1 and 100 nM) and it is present in microsomes at at least 10 times higher relative abundance (Simonyi et al., 1998; Vanlingen et al., 1999; Coquil et al., 2004) than the plant receptor, with the exception of C. rubrum (Scanlon et al., 1996). Also the InsP3-releasable Ca2+ pool is much larger in animal cells where it represents up to 90% of the Ca2+ mobilizable from the ER (Prentki et al., 1985). This finding points to the existence of another mechanism for Ca2+ mobilization in plants and implies IICR as its potential trigger.
Several strategies in identifying the responsible gene(s) can be proposed. Provided that the signalling molecule is really InsP3, one can imagine a screening of a ethylmethylsulphonate-mutated population of Arabidopsis plants with stable expression of aequorin. The first level of such screening should reveal mutants with lower Ca2+ influx upon ABA treatment. Among these mutants, a second screening should be carried out to select only those which have the same basal level and the same degree of InsP3 induction in the first seconds upon ABA treatment as the original non-mutated aequorin transformants. This second round of selection should exclude mutations in InsP3-metabolizing enzymes. The point mutations in these selected mutants can then be mapped to genes and these genes can be studied in detail, for example by complementation in yeast. This approach would certainly be very costly and time-consuming but has a good chance to succeed and finally identify the InsP3-R gene(s). It is quite probable that the responsible protein is not a canonical receptor/ion channel molecule as in animals and it would not be surprising if the IICR would be assured by a heteromeric subunit complex. Some exciting strategies using heterologous expression systems have been demonstrated recently such as the one for identification of a novel plant cell surface calcium-sensing receptor (Han et al., 2003). These strategies would require a system where the whole InsP3-mobilizing machinery would work and where the IICR would be absent or minor. Even if such a system existed, this approach would fail if the InsP3-R was a subunit complex, which is quite likely. Purification of the whole receptor/ion channel monitored by its function is not a good idea as this would be experimentally extremely costly with an uncertain outcome. One would need to reconstitute the membrane vesicles with the purified fraction in each step and still the whole complex could require some cytosolic factor which can be easily lost during the purification process. Another strategy would be to look for high-affinity binding sites for InsP3 in the plant microsomal fraction and then try to purify these sites. This would certainly reveal some ordinary InsP3-metabolizing enzymes associated with membranes, but one of these sites could be the receptor subunit or directly a part of the putative InsP3-R. Then, by means of the classical proteomics approach, the gene could be identified. If this turns out to be only one subunit of the whole complex, it would be necessary to look for its binding partners by pull-down experiments or yeast two-hybrid screens. Even though investigators have been very close to reaching this goal (Biswas et al., 1995), this final step has not been done yet. As soon as the putative gene(s) is found by any of the approaches, Arabidopsis T-DNA knock-out or RNA interference (RNAi) mutants should confirm that deletion of the corresponding protein(s) diminishes or even abolishes the in vivo observed IICR. Plant InsP3-R is a missing piece in the plant phosphoinositide signalling puzzle and its identification awaits those willing to accept the challenge.
|
Acknowledgements |
|---|
Our research is supported by Ministry of Education, Youth and Sports (grants No. LC 06034 and No. MSM 6046137305). We also gratefully acknowledge the perusing and remarks of reviewers that contributed considerably to improving the quality of the review.
|
Abbreviations |
|---|
ABA, abscisic acid; ER, endoplasmic reticulum; IICR, InsP3-induced calcium release; InsP3, D-myo-inositol 1,4,5-trisphosphate; InsP3-R, InsP3 receptor; RyR, ryanodine receptor.
|
References |
|---|
Acharya MK, Durejamunjal I, Guhamukherjee S. (1991) Light-induced rapid changes in inositolphospholipids and phosphatidylcholine in Brassica seedlings. Phytochemistry 30 2895–2897.[CrossRef]
Aducci P and Marra M. (1990) IP3 levels and their modulation FY fusicoccin measured by a novel [3H]IP3 binding assay. Biochemical and Biophysical Research Communications 168 1041–1046.[CrossRef][ISI][Medline]
Allen GJ, Muir SR, Sanders D. (1995) Release of Ca2+ from individual plant vacuoles by both InsP3 and cyclic ADP-ribose. Science 268 735–737.
Allen GJ and Sanders D. (1994) Osmotic stress enhances the competence of Beta vulgaris vacuoles to respond to inositol 1,4,5-trisphosphate. The Plant Journal 6 687–695.[CrossRef][ISI]
Alexandre J and Lassalles JP. (1990) Effect of D-myo-inositol 1,4,5-trisphosphate on the electrical properties of the red beet vacuole membrane. Plant Physiology 93 837–840.
Alexandre J and Lassalles JP. (1992) Intracellular Ca2+ release by InsP3 in plants and effect of buffers on Ca2+ diffusion. Philosophical Transactions of the Royal Society B: Biological Sciences 338 53–61.[CrossRef]
Alexandre J, Lassalles JP, Kado RT. (1990) Opening of Ca2+ channels in isolated red beet root vacuole membrane by inositol 1,4,5-trisphosphate. Nature 343 567–570.[CrossRef]
Apone F, Alyeshmerni N, Wiens K, Chalmers D, Chrispeels MJ, Colucci G. (2003) The G-protein-coupled receptor GCR1 regulates DNA synthesis through activation of phosphatidylinositol-specific phospholipase C. Plant Physiology 133 571–579.
Belde PJM, Vossen JH, Borst-Pauwels GWFH, Theuvenet APR. (1993) Inositol 1,4,5-trisphosphate releases Ca2+ from vacuolar membrane vesicles of Saccharomyces cerevisiae. FEBS Letters 323 113–118.[CrossRef][ISI][Medline]
Biswas S, Dalal B, Sen M, Biswas BB. (1995) Receptor for myo-inositol trisphosphate from the microsomal fraction of Vigna radiata. The Biochemical Journal 306 631–636.
Blatt MR, Thiel G, Trentham DR. (1990) Reversible inactivation of K+ channels of Vicia stomatal guard cells following the photolysis of caged inositol 1,4,5-trisphosphate. Nature 346 766–769.[CrossRef][Medline]
Brearley CA and Hanke DE. (1996a) Metabolic evidence for the order of addition of individual phosphate esters to the myo-inositol moiety of inositol hexakisphosphate in the duckweed Spirodela polyrhiza L. The Biochemical Journal 314 227–233.