The Function of Inositol Phosphatases in Plant Tolerance to Abiotic Stress

Inositol signaling is believed to play a crucial role in various aspects of plant growth and adaptation. As an important component in biosynthesis and degradation of myo-inositol and its derivatives, inositol phosphatases could hydrolyze the phosphate of the inositol ring, thus affecting inositol signaling. Until now, more than 30 members of inositol phosphatases have been identified in plants, which are classified intofive families, including inositol polyphosphate 5-phosphatases (5PTases), suppressor of actin (SAC) phosphatases, SAL1 phosphatases, inositol monophosphatase (IMP), and phosphatase and tensin homologue deleted on chromosome 10 (PTEN)-related phosphatases. The current knowledge was revised here in relation to their substrates and function in response to abiotic stress. The potential mechanisms were also concluded with the focus on their activities of inositol phosphatases. The general working model might be that inositol phosphatases would degrade the Ins(1,4,5)P3 or phosphoinositides, subsequently resulting in altering Ca2+ release, abscisic acid (ABA) signaling, vesicle trafficking or other cellular processes.


Introduction
Myo-inositol (Inositol, Ins) and its derivative metabolites are ubiquitous in all eukaryotes as both lipids and soluble compounds playing important roles in stress responses, development, and many other processes [1,2]. Upon environmental stresses, some of them are vital in various signal transduction in plants, especially inositol (1,4,5)triphosphate (Ins(1,4,5)P 3 ) and phosphatidylinositol (4,5)bisphosphate (PtdIns(4,5)P 2 ) [3][4][5][6][7]. They pass the cellular messages via addition or removal of lipids or phosphates to Ins and its derivatives, which could be mediated by synthases, kinases, phospholipases, and phosphatases [8,9]. Thus, those related enzymes are crucial in the regulation of these signaling pathways. In comparison to the other well-studied enzymes, limited information has been reviewed for the phosphatases in the Ins and phosphatidylinositol (PtdIns) signaling in plants. Here, we focus on these phosphatases and their function in abiotic tolerance.

Phosphatases in Inositol Signaling Pathways
Among the processes of inositol phosphate (IP) and the phosphoinositide (PI) signaling pathway, dephosphorylation is catalyzed by specific inositol phosphatases on various substrates ( Figure 1). Until now, dozens of enzymes have been identified, including inositol polyphosphate 5-phosphatases (5PTases), suppressor of actin (SAC) phosphatases, SAL1 phosphatase/FIERY1 (FRY1) and its homologs, inositol monophosphatase (IMP), and phosphatase and tensin homologue deleted on chromosome 10 (PTEN)-related phosphatases ( Figure 2). Most knowledge of them was obtained from the studies in the model plant Arabidopsis thaliana. These plant inositol phosphatases have a broad function in development and adaptation by altering the IP or PI signaling pathways. The general information of those Arabidopsis proteins was listed in Table 1. Interestingly, one certain inositol phosphatase could hydrolyze several substrates, even both inositol phosphate and phosphoinositide. One substrate could be degraded by more than one enzyme as well, suggesting their redundant roles in multiple aspects of life processes.
The 5PTases family is the biggest family of the mentioned inositol phosphatases, containing 15 members in Arabidopsis, 21 in rice, and 39 in soybean [17]. 5PTases hydrolyze the phosphate bond on the 5-position of the inositol ring from both inositol phosphate and phosphoinositide with the conserved inositol polyphosphate phosphatase catalytic (IPPc) domain. Due to the substrate specificity, mammalian 5PTases have been classified into four groups [18]. Group I, 5PTases hydrolyze only the water-soluble inositol polyphosphates (Ins(1,4,5)P 3 and Ins(1,3,4,5)P 4 ); group II the water-soluble inositol polyphosphates and the membrane-bound phosphoinositide; group III, Ins(1,3,4,5)P 4 and PtdIns(3,4,5)P 3 with a 3-position phosphate group; and group IV only phosphoinositide. Similar as the mammalian counterparts, plant 5PTases also have various substrate specificities. The substrates have been identified by biochemical evidences for twelve of the fifteen Arabidopsis 5PTases, including Group I, Group II, and Group IV 5PTases (Table 1). Since several 5PTases could hydrolyze Ins(1,4,5)P 3 to prevent its accumulation, it is believed to terminate the corresponding Ins(1,4,5)P 3 pathway and alter abscisic acid (ABA) signaling, Ca 2+ release, and reactive oxygen species (ROS) production [19][20][21].

5PTases and Plant Responses to Abiotic Stress
The capacity of 5PTases hydrolyzing IP 3 is believed to be vital in the termination of IP 3 , consequently altering Ca 2+ oscillations, ABA signaling, and other stress-related pathways. The transgenic Arabidopsis plants overexpressing mammalian type I (group I) inositol polyphosphate 5-phosphatase (InsP 5-ptase) exhibited increased drought tolerance with less water loss [70]. The contents of IP 3 and IP 6 were decreased in the transgenic lines as expected, thus attenuating ABA induction and Ca 2+ signal transduction. The stomata were less responsive to the inhibition of opening by ABA and more sensitive to ABA-induced closure. Furthermore, the microarray data showed that dehydration-responsive element-binding protein 2A (DREB2A), encoding a drought-inducible ABA-independent transcription factor, and the DREB2A-regulated genes were induced in the InsP 5-ptase transgenic plants, suggesting the drought tolerance is mediated via the DREB2A-dependent pathway [70].
For plant 5PTases, it is common to take a role in the degradation process of inositol phosphate or phosphoinositide, terminating the IP 3 signaling, thus altering of ABA pathway and Ca 2+ release, which is believed to be vital in stress tolerance [19,21,34,39]. However, only three of the 15 At5PTases have been identified to play important roles in abiotic stress with genetic and biochemical evidences until now. At5PTase7 and At5PTase9 function in salt tolerance, and At5PTase13 in low nutrient and sugar stress [35,36,41].
The T-DNA insertion mutants of At5PTase7 or At5PTase9 increased salt sensitivity and the overexpression plants increased salt tolerance [35,36]. Mutation in either At5PTase7 or At5PTase9 reduced ROS production in the Arabidopsis roots after 10 to 15 min after salt treatment. Additionally, the expression of salt-responsive genes, such as RD29A and RD22, was not induced highly in both mutants as in the wild-type under salt stress [35,36]. It suggested that the defect in At5PTase7 or At5PTase9 would disturb ROS production and salt-responsive gene expression, probably hampering the subsequent rescue signal transduction. Interestingly, the At5PTase9 mutants appeared to have a better ability to resistant osmotic stress. Meanwhile, the At5PTase9 mutants decreased Ca 2+ influx and fluid-phase endocytosis [36]. Though the At5PTase7 and At5PTase9 isomers take non-redundant roles in regulating plant responses to salt stress, they share the same substrates, membrane-bound phosphoinositide, indicating that phosphoinositide would be important in salt tolerance [36].
At5TPase13 is one of the four At5TPases (At5TPase12-15), which contain the plant specific WD40 repeats [38,44]. The T-DNA insertion mutants of At5TPase13 showed a reduction of root growth under limited nutrient conditions and germination rates in response to sugar stress, along with ABA insensitivity [41]. The yeast two-hybrid analyses suggested that its WD40 repeat domain interacts with the sucrose nonfermenting-1-related kinase (SnRK1.1), which is an energy/stress sensor [41]. The genetic and biochemical evidences indicated that At5TPase13 acts as a positive regulator of SnRK1.1 under low-nutrient or low-sugar conditions, as a negative regulator under severe starvation conditions through affecting the proteasomal degradation of SnRK1.1. Strangely, the At5ptase13 mutants accumulate less IP 3 in response to sugar stress [41]. Again, At5PTase13 could alter cytosolic Ca 2+ to regulate PHOYOTROPIN1 signaling under blue light [40].
Besides, several transcriptional analyses showed that the expression of multiple At5PTases is greatly up-or down-regulated in response to a series of abiotic stresses, such as cold, osmotic, salt, drought, oxidative, and heat [35,36,71]. Considering the general function of the known 5PTases in the inositol pathway, Ca 2+ signaling, ABA responses, ROS generation, vesicle trafficking, and possible connection with other phytohormones [43,71], it could imply their potential roles in plant responses to abiotic stress.

SAL1 and Plant Responses to Abiotic Stress
AtSAL1, identified as a homologue of the yeast HAL2 in Arabidopsis and also well-known as FIERY1 (FRY1), has dual enzymatic activity of inositol phosphatase and nucleotidase, which play a role in both inositol signaling and nucleotide metabolite [47,48]. AtSAL1 functions broadly in responses to abiotic stresses, including salt, cold, lithium, drought, cadmium, high light, and oxidative, probably with the contributions of both enzymatic activity [48,49,[51][52][53]63,72,73]. Here we will focus on its activity of inositol polyphosphate 1-phosphatase. Remarkably, it can hydrolyze the signaling molecular IP 3 , thus affecting the subsequent steps in a similar pattern of 5PTases, which we have discussed above.
It seems the effects of AtSAL1 on stress responses are controversial. Ectopic expression of AtSAL1 could increase lithium tolerance in yeast by modifying Na + and Li + effluxes [47]. Ectopic expression of its homologue in soybean, GmSAL1, could alleviate salinity stress in tobacco BY-2 cells [74]. Mutation in AtSAL1 would cause more sensitivity to salt, osmotic, and cold stress in Arabidopsis [48,72]. However, another Atsal1 mutant, hos2 with a single amino acid substitution exhibited as more resistant to lithium and salt stress [72]. Moreover, overexpression of AtSAL1 or ectopic expression of GmSAL1 could not enhance salt tolerance in Arabidopsis [49,74]. Loss function in AtSAL1 would enhance drought and cadmium resistance in Arabidopsis, suggesting it would be a negative regulator of stress tolerance [51,63]. Expressing the modified SAL1, by inserting the META motif from black yeast Aureobasidium pullulans, ApHal2, improved salt and drought tolerance in Arabidopsis [73]. It seems the presence of the META motif should be responsible for its ability on the stress tolerance, but the mechanism is still obscure.
The molecular mechanism of AtSAL1 in stress responses seems to be complicated for its multiple effects in various cellular processes. First, AtSAL1 would regulate stress tolerance and ABA responses via IP 3 signaling. The Atsal1 mutants increase IP 3 accumulation and the expression levels of ABA and stress genes, including RD29A, cold-specific CRT-binding factor 2 (CBF2), and CBF3 [48]. On the contrary, ectopic expression of GmSAL1 leads to a reduction of IP 3 accumulation and suppression of the ABA-induced stomatal closure [74]. Furthermore, it also showed AtSAL1 could regulate Ca 2+ release and modulate the auxin pathway by IP 3 signaling in plant development [54,55]. It seems a similar consequence of AtSAL1 in the IP 3 signaling as for 5PTases. Maybe further investigation will supply evidence that AtSAL1 takes a role in Ca 2+ release and its downstream signaling in response to abiotic stress as well. Secondly, AtSAL1 also regulates the ion homeostasis via the IP 3 pathway. Ectopic expression of AtSAL1 could modify Na + and Li + effluxes in yeast for lithium and salt tolerance [47]. GmSAL1-transgenic BY-2 cells could compartmentalize more Na + in vacuolar for protection from salt stress [74]. Additionally, AtSAL1 takes a role more likely as a phosphoadenosine phosphatase under drought, high light, and oxidative stress, for only 3 -phosphoadenosine 5 -phosphate (PAP), not IP 3 , accumulated in the Atsal1 mutants, when suffering stresses [49,52,53,65]. The genetic evidences indicated that PAP accumulation could also affect the ABA pathway, relying on, rather, the negative regulator ABH1 in the branched ABA pathway, than ABI1 in the core ABA pathway [49]. AtSAL1 could protect 5 to 3 exoribonucleases (XRNs) by degrading PAP and subsequently modulate the expression of the corresponding nuclear genes, supposed as the chloroplast retrograde pathway [52,53,65]. Besides, the AtSAL1-deficient mutants have been found to attenuate endoplasmic reticulum (ER) stress under cadmium stress [63]. But no exploration has been made to determine its connection with the IP3 signaling or SAL1-PAP pathway. This would provide a new insight on the mechanism of AtSAL1 in various stress tolerance [63].

IMPs and Plant Responses to Abiotic Stress
IMPs were first identified in tomato to play a role in inositol synthesis with high sensitivity to lithium [10]. Their homologues in Arabidopsis have also been characterized as multi-functional enzymes involved in inositol, ascorbate, and histone biosynthesis [57,59,66], so do their homologues in other plants, such as rice (Oryza sativa L.), chickpea (Cicer arietinum L.), soybean (Glycine max), barley (Hordeum vulgare), and Medicago truncatula [68,69,75]. The genetic studies showed that IMPs play a role in seed development in Arabidopsis [11]. Chickpea IMP could also influence seed size/weight [76]. But few explorations have been made with Arabidopsis IMP on stress tolerance yet. Only some authors have tried assays in chickpea and rice suggesting that IMPs also function in response to abiotic stress [67][68][69]. But it is still unclear how IMPs influence the inositol pathway to confer stress.
Biochemical evidence demonstrated that CaIMP contains the same enzyme activity as Arabidopsis IMP and IMP activity is increased in chickpea seedlings under abiotic stresses, including salt, cold, heat, dehydration, and paraquat. It is consistent with the results of the transcript analyses by qRT-PCR, which showed that CaIMP is induced under abiotic stress and ABA treatment [69]. The CaIMP-transgenic Arabidopsis plants exhibited enhanced tolerance to abiotic stress, whereas the IMP-deficient Arabidopsis mutants increased the sensitivity to stress during seed germination and seedling growth. The inositol content and ascorbate content of the CaIMP-overexpressing lines are higher than the wild-type and the vector control, suggesting CaIMP would improve the plant tolerance to stress through both metabolic pathways [69]. Association analyses performed with 60 chickpea germplasm accessions showed that NCPGR90, a simple sequence repeat marker for phytic acid content and drought tolerance, is located to the 5'UTR of CaIMP [68]. The transcript lengths of CaIMP are different between the drought-tolerant and drought-susceptible accessions, suggesting this variation might regulate phytic acid contents in plants, thus conferring drought tolerance in chickpea [68]. In another study, this variation also causes the differential protein level and enzymatic activity of CaIMP [76].
Rice OsIMP is significantly upregulated by cold and ABA treatment by transcript analyses [67]. The promoter analyses on sequence also identified several important stress-responding cis-acting elements, including ABRE-element (abscisic acid responsiveness), LTR (low-temperature responsiveness), TCA-element (salicylic acid responsiveness), GARE-motif (gibberellins responsive), and MBS (MYB binding site). Ectopic expression of OsIMP in tobacco improved cold tolerance. The transgenic plants contained more inositol content, less hydrogen peroxide (H 2 O 2 ), and less malondialdehyde (MDA), with increased antioxidant enzyme activities under normal and cold stress conditions [67]. It suggested that the accumulation of inositol by expressing OsIMP would modulate the antioxidant enzymes' activities to conquer cold stress.

Conclusions
Substantial evidences demonstrate inositol phosphates, phosphoinositides, and the related inositol signaling play a crucial role in various life processes of development and environmental adaptation in plants [1,4,6,7,12]. When plants suffer abiotic stress from the environment, a membrane receptor would accept the stimulus and the membrane-associated phosphoinositides would pass the cellular message by producing second messages, lipid-bound DAG, and soluble IP 3 . Components involved in the inositol pathways have been noted for their general roles in stress tolerance. This article focused on the knowledge about inositol phosphatases, which are considered to be more important in the degradation pathway of IP 3 signaling, and their function in plant responses to abiotic stress.
Around 30 members of inositol phosphatases from five families have been identified. Their functions and mechanisms are still largely unknown. Biochemical and physiological data, especially those from analytical techniques, have delineated their substrates and the affecting signals. Moreover, the genetic evidences elucidate the genes' function and how to pass the signals. In general, loss-in-function of inositol phosphatases usually cause the accumulation of IP 3 or phosphoinositides, thus facilitating Ca 2+ release from cellular stores and affecting ABA or other phytohormones' pathways. For their effects on lipid-bound phosphoinositides, several enzymes have been proved to be involved in vesicle trafficking. For most of the inositol phosphatases, the existed evidences could only support part of the model. There are also some other puzzles. Since phytic acid (InsP 6 ) could also serve as a signaling molecule to regulate Ca 2+ release [6], what is the role of inositol phosphatases in this process? There are multiple genes in the same family, especially 5PTases and SAC phosphatases. How do plants coordinate their function? Most of the knowledge about these enzymes is obtained from the mode plant Arabidopsis. Study from other plants is relatively rare. Do these inositol phosphatases take a universal role in all plants under abiotic stress? Hopefully, more exploration will expand our understanding about inositol phosphatases.

Conflicts of Interest:
The authors declare no conflicts of interest.