Human 1-D-myo-Inositol-3-phosphate Synthase Is Functional in Yeast*

We have cloned, sequenced, and expressed a human cDNA encoding 1-d-myo-inositol-3-phosphate (MIP) synthase (hINO1). The encoded 62-kDa human enzyme converted d-glucose 6-phosphate to 1-d-myo-inositol 3-phosphate, the rate-limiting step for de novo inositol biosynthesis. Activity of the recombinant human MIP synthase purified from Escherichia coli was optimal at pH 8.0 at 37 °C and exhibited Km values of 0.57 mm and 8 μm for glucose 6-phosphate and NAD+, respectively. \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(\mathrm{NH}_{4}^{+}\) \end{document} and K+ were better activators than other cations tested (Na+, Li+, Mg2+, Mn2+), and Zn2+ strongly inhibited activity. Expression of the protein in the yeast ino1Δ mutant lacking MIP synthase (ino1Δ/hINO1) complemented the inositol auxotrophy of the mutant and led to inositol excretion. MIP synthase activity and intracellular inositol were decreased about 35 and 25%, respectively, when ino1Δ/hINO1 was grown in the presence of a therapeutically relevant concentration of the anti-bipolar drug valproate (0.6 mm). However, in vitro activity of purified MIP synthase was not inhibited by valproate at this concentration, suggesting that inhibition by the drug is indirect. Because inositol metabolism may play a key role in the etiology and treatment of bipolar illness, functional conservation of the key enzyme in inositol biosynthesis underscores the power of the yeast model in studies of this disorder.

Inositol metabolism is implicated in the etiology and treatment of bipolar disorder, a severe psychiatric illness that affects 1-2% of the general population (10). Lithium and valproate (VPA) 1 are the two FDA-approved drugs for the treatment of bipolar disorder. The cause of this illness is un-known, and the therapeutic mechanisms of the anti-bipolar drugs have not been elucidated. The possibility that inositol metabolism is involved in the etiology of bipolar disorder is supported by findings of aberrant inositol metabolism in bipolar patients (11). Reduced frontal cortex inositol levels and reduced levels of phosphatidylinositol bisphosphate in platelet membrane from bipolar patients have been reported (12)(13)(14). In addition, inositol incorporation into membrane phosphoinositides of lymphoblastoid cell lines derived from bipolar patients is reduced (15).
The inositol depletion hypothesis has been proposed to explain the therapeutic mechanism of Li (16,17) based on the observed uncompetitive inhibition of inositol monophosphatase by Li (16,18) resulting in decreased inositol and down-regulation of the phosphoinositide cycle. An argument raised against the inositol depletion hypothesis is that VPA does not affect inositol monophosphatase (19), which is, thus, not a common target of the drugs. However, we have shown that VPA decreases intracellular inositol in yeast (20), although it does not inhibit inositol monophosphatase (21). Similar to what is found in yeast, a reduced level of inositol also was reported in rodent brain after acute 2 and chronic (23) administration of VPA.
Although a number of metabolic routes emanate from myoinositol, de novo synthesis of inositol is carried out by one set of enzymatic reactions in all organisms studied to date. D-Glucose 6-phosphate (G-6-P) is converted to 1-D-myo-inositol-3-phosphate (MIP), which is subsequently dephosphorylated to myoinositol. The rate-limiting step in inositol de novo biosynthesis is the conversion of G-6-P to MIP, a three-step reaction catalyzed by the enzyme 1-D-myo-inositol-3-phosphate synthase (EC 5.5.1.4), including an oxidation step with NAD ϩ serving as a hydrogen acceptor, an intramolecular aldol cyclization step, and finally a reduction step with NADH acting as the hydrogen donor regenerating NAD ϩ (1). MIP synthase has been purified or partially purified and characterized from a wide range of organisms, and its active form is a multimer of identical subunits ranging in molecular mass from 58 to 67 kDa (1-3, 24 -28). The structural gene (INO1) encoding the MIP synthase subunit was first identified and cloned in Saccharomyces cerevisiae (27). Homologs of this gene subsequently have been cloned from several prokaryotic and eukaryotic microorganisms and higher plants (1,3,29,30).
The best studied MIP synthase is the enzyme from S. cerevisiae in which expression has been characterized at the molecular level. INO1 expression is regulated by precursors of phos-pholipid biosynthesis, inositol and choline (31)(32)(33)(34)(35)(36). INO1 is fully repressed in the presence of inositol and choline and derepressed more than 10-fold in their absence. Derepression of INO1 when inositol is limiting is brought about by the positive transcriptional activators Ino2p and Ino4p. In the presence of inositol, the negative regulator Opi1p represses transcription of INO1 (37). Because inositol metabolism is so well characterized in yeast, we have utilized this model system to elucidate the effects of anti-bipolar drugs on this pathway. Employing genetic and molecular approaches to identify common targets of Li and VPA in S. cerevisiae, we have shown that both Li and VPA affect the inositol metabolic pathway. Both drugs cause a decrease in intracellular myo-inositol and an increase in expression of INO1 as well as the regulatory gene INO2 required for inositol biosynthesis (20). In addition, both drugs lead to a decrease in the phosphatidylinositol/phosphatidylcholine ratio (38). Furthermore, VPA abrogates the normal response to inositol depletion of inositol-responsive genes and leads to aberrant synthesis of phospholipids (39).
Given the essential role of inositol in cellular function, surprisingly little is known about the regulation of MIP synthase in human cells. Despite some studies on the biochemistry of the enzyme from various animal systems (40 -43), the INO1 gene has not been cloned from these sources. Recently, Guan et al. (29) reported the molecular cloning of the putative human MIP synthase cDNA. However, functional expression of the cDNA and characterization of the enzyme was not carried out.
In this study, we cloned the human MIP synthase cDNA and characterized the recombinant enzyme. We show for the first time that the human enzyme is functional in yeast. Furthermore, human MIP synthase activity was decreased in cells grown in the presence of VPA. Because inositol metabolism is thought to play a key role in the etiology and treatment of bipolar illness, conservation of function of this key enzyme underscores the power of the yeast model in studies of this illness.

EXPERIMENTAL PROCEDURES
Materials-All chemicals used were reagent grade or better. Glucose, yeast extract, and peptone were purchased from Difco. Amino acids, NAD ϩ , myo-inositol, glucose 6-phosphate, imidazole, lysozyme, valproate, ampicillin, and chloramphenicol were obtained from Sigma. [ 14 C]Glucose 6-phosphate, [ 14 C]inositol, and [ 32 P]UTP were bought from PerkinElmer Life Sciences. Protease inhibitor tablets were from Roche Applied Science. pRSETA vector, anti-Xpress antibody, and Ni 2ϩ -chelated resin were purchased from Invitrogen. Dialysis tubing was made by Spectrum Laboratories Inc.
Isolation of Human MIP Synthase cDNA-A post-mortem human prefrontal cortex specimen derived from a brain collection described previously (44) was used for mRNA purification. mRNA was purified using the Quick-Prep micro mRNA purification kit (Amersham Biosciences) according to the manual. Amplification of the human MIP synthase cDNA was performed as follows. First strand cDNA synthesis was carried out using the first strand cDNA synthesis kit (Amersham Biosciences) with the reverse primer 5Ј-GGGGGCCTTCAAGGTAGG-3Ј. The cDNA was amplified using the forward (5Ј-GCCGCCGCTGC-CTGAGTCGAC-3Ј) and reverse (5Ј-GGGGGCCTTCAAGGTAGG-3Ј) specific primers. Amplification of the open reading frame was performed using the PCR product from the first reaction as a template along with a forward primer supplemented with a BamHI restriction site (5Ј-GCGCCGGGATCCACGATGGAGGCCGCCGCCCAG-3Ј) and a reverse primer supplemented with an XhoI site (5Ј-CTAGACTC-GAGGGTGGTGGGCATTGGG-3Ј). The PCR product was cut by BamHI and XhoI and inserted into pRS426GPD and pRSETA vectors. All constructs were confirmed by sequencing.
Measurement of Intracellular Inositol-Intracellular inositol was measured as described (39). Briefly, cells were washed three times and resuspended in water (ϳ1 ml/g cells), and glass beads were added to ϳ50% of the volume of the suspension. Each sample was vortexed for 10 min at 2-min intervals alternating with 2-min incubations on ice. The cell extracts were clarified by centrifugation for 2 min at 2,000 ϫ g, and the supernatants were transferred to Eppendorf tubes and centrifuged for 15 min at 14,000 ϫ g. The supernatants were collected and frozen at Ϫ80°C. Intracellular inositol mass per 100 g of protein was determined by the enzyme-coupled fluorescence assay (45).
Measurement of hINO1 Expression-Ino1⌬/hINO1 cells (the ino1⌬ mutant transformed with the hINO1 gene on pRS426GPD) were grown in Ura Ϫ synthetic medium in the presence or absence of 0.6 mM VPA to the early stationary phase of growth. Northern analysis was performed as described (39). RNA probes for Northern analysis were synthesized from plasmids linearized with restriction enzymes as follows. The plasmid, restriction enzyme, and RNA polymerase for hINO1 are pGEM-hINO1, StuI, and SP6, respectively, and for TCM1 are pAB309, EcoRI, and SP6, respectively.
Overexpression of Human MIP Synthase in E. coli-The recombinant constructs were transformed into E. coli BL21(DE3)pLysS for expression of the protein. A single colony of BL21(DE3)pLysS containing the recombinant pRSETA/hINO1 was inoculated into 5 ml of LB medium containing 100 g/ml ampicillin and 35 g/ml chloramphenicol. The culture was incubated overnight and used to inoculate 200 ml of LB medium with the same concentration of ampicillin and chloramphenicol. Cell pellets from 200-ml cultures were used to inoculate 8 liters of fresh LB medium containing 25 g/ml ampicillin and 35 g/ml chloramphenicol. These cultures were grown at 37°C to an A 550 of 0.4. Production of recombinant protein was induced by the addition of isopropyl-1-thio-␤-D-galactopyranoside to a final concentration of 1 mM followed by incubation for 3 h. Cells were harvested by centrifugation and stored at Ϫ80°C until needed.
Purification of Recombinant Human MIP Synthase-All the purification steps were carried out at 4°C. Frozen cells (10 g from 8 liters of culture) were thawed and resuspended in 150 ml of binding buffer (50 mM NaPO 4 , 0.5 M NaCl, 10 mM imidazole, 0.1ϫ protease inhibitor, pH 8.0). The cell suspension was incubated with 100 mg of lysozyme on ice for 30 min and then lysed by sonication 10 times for 30 s on ice. DNase (5 g/ml) and RNase (10 g/ml) were added, and this was followed by incubation on ice for another 15 min. The supernatant was separated from cell debris by centrifugation (10,000 rpm for 15 min, SS34 rotor), loaded onto a Ni 2ϩ column (Invitrogen), and allowed to bind with resin for 60 min using gentle agitation to keep the resin suspended. The purification column was then washed five times with 200 ml of washing buffer (50 mM NaPO 4 , 0.5 M NaCl, 20 mM imidazole, pH 7.0) and eluted with 60 ml of eluting buffer (50 mM NaPO 4 , 0.5 M NaCl, 250 mM imidazole, pH 8.0). One-ml fractions were collected and analyzed by SDS-PAGE. The peak fractions containing MIP synthase were combined and loaded onto another Ni 2ϩ column. The column was washed and the enzyme was eluted with imidazole.
Dialysis-Spectrum Spectra/Por molecularporous membrane dialysis tubing (MWCO 50 kDa, Rancho Dominguez, CA) was prepared according to the suggested protocol. 10-cm strips were cut and soaked in sterile distilled H 2 O for 10 min. The tubing was filled to 60% of capacity, leaving a small air pocket at one end of the tube. Both ends of the tube were clamped, and the tube was soaked in 4 liters of precooled buffer (1 mM Tris acetate, pH 8.0, 0.05 mM dithiothreitol, 0.025ϫ protease inhibitor). Dialysis was carried out for 36 h at 4°C with the buffer changed every 12 h.
Protein Assay-Protein concentration was determined by the method of Bradford (46) using bovine serum albumin as the standard.
Electrophoresis and Western Blot-SDS-PAGE was performed on Ready Gel (4 -20% gradient) from Bio-Rad. Electrophoresis was carried out using the procedure described by Laemmli (47). All gels were stained with Coomassie Brilliant Blue R-250 (Bio-Rad). Western blots were performed by standard procedures using a mouse-derived monoclonal antibody against Xpress (Invitrogen) for protein expressed in E. coli, and alkaline phosphatase goat anti-mouse IgG (Promega) as a secondary antibody.
MIP Synthase Assay-MIP synthase was assayed in crude extracts by the chromatographic method of Chen and Charalampous (48). Purified MIP synthase activity was determined by the rapid colorimetric method of Barnett et al. (49) with minor modification. Purified protein was incubated in the reaction buffer in a final volume of 150 l (100 mM   FIG. 2. Expression of hINO1 functionally complements yeast  ino1⌬. A, growth of the ino1⌬ mutant harboring the human MIP synthase cDNA (ino1⌬/hINO1) or empty vector (ino1⌬/vector) on Iϩ and IϪ medium. B, cells of ino1⌬/hINO1 were streaked on synthetic Iϩ medium containing 1 mg/ml 5Ј-fluoro-orotic acid to select for loss of the plasmid. These cells were then plated (along with controls ino1⌬, ino1⌬/ hINO1, and ino1⌬/vector) on the indicated synthetic media. C, opi phenotype of ino1⌬/hINO1. Cells from ino1⌬/hINO1 (top four) or ino1⌬/ vector (bottom four) were spotted on indicator plates containing the inositol-requiring strain AID-1. Inositol excretion is indicated by growth of the AID-1 lawn around ino1⌬/hINO1 cells.
Tris acetate, pH 8.0, 5 mM G-6-P, 0.8 mM NAD ϩ , 2 mM dithiothreitol, if not otherwise indicated) for 1 h at 37°C. 14 mM NH 4 ϩ was included in the assays for determining optimal pH, temperature, and K m . The reaction was terminated by the addition of 50 l of 20% (w/v) trichlo- FIG. 3. Expression of hINO1 restores inositol synthesis to ino1⌬ mutant. Cells from strain ino1⌬/hINO1 and ino1⌬/vector were grown in medium containing 75 M inositol to the middle logarithmic phase of growth and were shifted at time zero to IϪ medium. Cells were harvested at indicated times after the shift. A, intracellular inositol per 100 g of protein was determined by the enzymecoupled fluorescent assay of Maslanski and Busa (45). The inositol values are represented relative to that observed at time zero (100%). f, ino1⌬/hINO1; Ⅺ, ino1⌬/vector. B, viable cells were counted at 0 and 24 h after the shift. C, MIP synthase was assayed in crude extracts of cells grown in Iϩ medium to early stationary phase. D, intracellular inositol was assayed in ino1⌬/hINO1, and wild type (WT) cells were grown in IϪ medium to the early stationary phase. The values shown are the average of two independent experiments.  roacetic acid and kept on ice for 10 min. The precipitated protein was removed by centrifugation. 200 l of the supernatant was incubated with 200 l of NaIO 4 for 1 h. 200 l of 1 M Na 2 SO 3 then was added to the supernatant to remove the excess NaIO 4 . For the measurement of phosphate, a 600-l reagent mixture (240 l of H 2 O, 120 l of 2.5% ammonium molybdate, 120 l of 10% ascorbic acid, and 120 l of 6 N sulfuric acid) was added and incubated for 1 h at 37°C. The absorbance was measured at 820 nm, and specific activity was defined as units per mg of protein where 1 unit is the amount of enzyme catalyzing the formation of 1 nmol of product per min at 37°C. For each assay, a second aliquot of the sample was measured for phosphates not released by periodate to control for phosphatase activity. This value was subtracted from the experimental sample to obtain synthase activity.

Isolation of the Human MIP Synthase cDNA
A full-length human MIP synthase cDNA (1677 bp) was isolated from post-mortem brain as described under "Experimental Procedures." The cDNA encodes a protein of 558 amino acids with a molecular mass of ϳ62 kDa. The deduced protein is 50% identical and 69% similar to MIP synthase from S. cerevisiae (Fig. 1).

Functional Expression of Human MIP Synthase cDNA
A yeast-based functional assay was employed to determine whether the cDNA encoded a functional MIP synthase. To perform the assay, hINO1 was cloned into the yeast expression vector pRS426GPD, which is constitutively expressed in high copy (50). This vector was transformed into the yeast ino1⌬ mutant, which cannot synthesize inositol because of the disrupted INO1 gene. As shown in Fig. 2A, the ino1⌬ mutant transformed with hINO1 (ino1⌬/hINO1) grew on a plate lacking inositol (IϪ), in contrast to mutant cells transformed with empty vector (ino1⌬/vector). To confirm that complementation of the yeast ino1⌬ mutant was caused by the recombinant construct, 5Ј-fluoro-orotic acid was used to select for the loss of the plasmid. As shown in Fig. 2B, the ino1⌬ cells containing the construct before 5Ј-fluoro-orotic acid selection grew on IϪ plates, whereas cells that lost the plasmid after 5Ј-fluoro-orotic acid selection did not. Yeast cells that overexpress the yeast INO1 gene excrete inositol into the medium, creating a phenotype known as opi (overproduction of inositol), which is detected by the ability of the excreted inositol to support growth of an inositol-requiring indicator strain (51). As shown in Fig.  2C, the ino1⌬ mutant harboring the hINO1 gene exhibited the opi phenotype. These results indicate that human MIP synthase is functional in yeast.
To further characterize the function of hINO1 in yeast, cells were shifted from Iϩ to IϪ medium, and intracellular inositol, MIP synthase activity, and cell viability were assayed. As shown in Fig. 3A, following the shift to IϪ, intracellular inositol decreased markedly in ino1⌬ cells lacking hINO1 but remained constant in cells containing the human gene. Consistent with these data, ino1⌬/hINO1 cells continued to grow, but ino1⌬/ vector cells lost viability after 24 h (Fig. 3B). MIP synthase activity was high in ino1⌬ cells containing hINO1 but undetectable in ino1⌬ cells transformed with empty vector (Fig. 3C). Interestingly, intracellular inositol in ino1⌬/hINO1 was 30% FIG. 6. Effect of pH and temperature on enzyme activity. A, enzyme activity was measured at the indicated pH values with 100 mM Tris acetate buffer. B, enzyme activity was measured at the indicated temperature. C, the enzyme was heated at the indicated temperature for 20 min before the reaction. Activity of the heated enzyme was measured at optimal pH (8.0) and temperature (37°C). Data represent the average of two independent experiments. higher than in isogenic wild type cells lacking the hINO1 gene (Fig. 3D), which most likely accounts for the inositol excretion phenotype of ino1⌬/hINO1 observed in Fig. 2C. This can be explained by the constant high level of expression of the hINO1-containing plasmid. In contrast, expression of the native yeast INO1 gene in wild type cells is regulated and expressed highly only in logarithmic phase cells (32).

Effect of VPA on Inositol and MIP Synthase Activity
We have shown that both Li and VPA cause a decrease in intracellular inositol in yeast (20). This is consistent with the observed decrease in brain inositol levels following Li (52) and VPA (23) treatment. The Li-induced decrease can be explained by inhibition of inositol monophosphatase (19,21). However, a decrease in MIP levels was observed during growth in the presence of VPA, suggesting that VPA may inhibit MIP synthase (20). To address this possibility, inositol and MIP synthase activities were assayed in crude extracts of ino1⌬/hINO1 grown in the presence of 0.6 mM VPA (the concentration used in therapeutic treatment). As shown in Fig. 4, in the presence of VPA, intracellular inositol was decreased 25% (Fig. 4A), and a 35% decrease in MIP synthase activity was observed (Fig. 4B). The gene was expressed at high levels in all growth phases (data not shown), and expression was not affected by VPA (Fig.  4C), indicating that the inhibitory effect of the drug was at the level of the protein.
The purified protein analyzed by SDS-PAGE migrated with an apparent molecular mass of ϳ65 kDa (Fig. 5A, lane 6) which is consistent with the expected size (62-kDa protein plus a 3-kDa tag). Western blot analysis using anti-Xpress antibody confirmed that this was the recombinant protein (Fig. 5B, lanes 2-6). The specific activity of the purified enzyme was measured by the rapid colorimetric method of Barnett et al. (49) and confirmed by the chromatographic assay of Chen and Charalampous (48) using [ 14 C]glucose 6-phosphate as a substrate.

Properties of Human MIP Synthase
Effect of pH and Temperature on Enzyme Activity-Human MIP synthase activity was measured at pH 6.0 -10.0 using 100 mM Tris acetate buffer. All other components in the reaction mixture were kept constant, and assays were performed using standard conditions as described under "Experimental Procedures." The optimal pH of human MIP synthase was 8.0 (Fig. 6A).
Enzyme activity was measured in a controlled temperature water bath from 20 to 60°C under standard assay conditions. The temperature profile suggested that maximum activity was obtained at 37°C, and the enzyme still was 90% active at 45°C (Fig. 6B). Thermal stability of the enzyme was determined by assaying samples heated for 20 min at 30 -60°C. The enzyme was stable to heating up to 45°C. However, ϳ60% activity was lost upon heating at 60°C (Fig. 6C).
Enzyme Kinetics-Kinetic analysis of purified recombinant enzyme is shown in Fig. 7. When NAD ϩ was held constant at 0.8 mM and the G-6-P concentration varied, saturation kinetics were shown by the enzyme. The apparent K m for G-6-P was 0.57 mM (Fig. 7A). Saturation kinetics also were shown when G-6-P was held constant at 5 mM and the NAD ϩ concentration varied. The apparent K m for NAD ϩ was 8 M (Fig. 7B).
Effect of Cations on Enzyme Activity-MIP synthase activity was determined in the presence of NH 4 ϩ , Na ϩ , K ϩ , Mg 2ϩ , Mn 2ϩ , Ca 2ϩ , and Zn 2ϩ . Like MIP synthase from all other organisms except Archaeoglobus fulgidus (2), human MIP synthase was strongly stimulated by NH 4 ϩ . The activity was increased 5-fold at 5 mM NH 4 ϩ (Fig. 8A). 10 mM K ϩ stimulated activity 5-fold. Mg 2ϩ and Mn 2ϩ increased activity 2-3-fold at 1 or 10 mM (Fig.  8B). Activity was decreased 5-fold in the presence of 1 mM Zn 2ϩ and was totally inactivated in the presence of 10 mM Zn 2ϩ (Fig.  8B). Na ϩ and Ca 2ϩ had no effect on the enzyme at 1 mM and increased activity by about 2-fold at 10 mM (Fig. 8B).
Effect of VPA on Purified Human MIP Synthase-To determine whether the decrease in MIP synthase activity observed during growth in the presence of VPA (Fig. 4B) is caused by direct inhibition of the enzyme, purified human MIP synthase activity was determined in the presence of different concentrations of the drug. The enzyme was not inhibited by the therapeutic concentration of VPA and was only slightly (10%) inhib- ited by VPA at 10 mM (Fig. 9). This inhibition was not seen when 14 mM NH 4 ϩ was included in the reaction mixture (data not shown). DISCUSSION We report for the first time that human MIP synthase is functional in yeast and supports growth of the yeast ino1 mutant in the absence of inositol. In the presence of a therapeutically relevant concentration of the anti-bipolar drug VPA, yeast ino1 mutant cells containing the human enzyme exhibit a decrease in MIP synthase activity and intracellular inositol. We observed previously that growth of yeast cells in the presence of VPA leads to decreased inositol monophosphate and intracellular inositol and increased INO1 expression consistent with the inhibition of yeast MIP synthase (20). Similar to the observations in yeast cells, a decrease in MIP synthase activity also was observed in crude extracts from post-mortem human brain. 2 The purified recombinant human enzyme was not inhibited by therapeutic concentrations of VPA (Fig. 9). Purified yeast MIP synthase was similarly not affected by VPA. 3 These findings suggest that a metabolite of VPA or an inhibitor that accumulates in the presence of VPA may be responsible for MIP synthase inhibition. The current findings support the use of the yeast model to elucidate the effects of VPA on inositol biosynthesis.
The properties of the recombinant MIP synthase were similar mostly to those reported for native enzymes. The enzyme was active from pH 7.5 to 10.0. MIP synthase that was partially purified and characterized from human fetal brain was optimal at pH 7.5 (53). The K m of the recombinant enzyme for NAD ϩ was 8 M, much lower than 452 M reported for the partially purified human fetal brain enzyme but consistent with values from other sources, e.g. 11 M for the bovine testis enzyme (42), 17.9 M for the rat testis enzyme (40), and 8 M for the yeast enzyme (27). NH 4 ϩ strongly stimulated both the partially purified fetal brain enzyme and the recombinant enzyme. Other mono-and divalent cations also stimulated the enzyme to different extents, which may implicate a general ionic effect. The recombinant enzyme also was stimulated 1.5-2.5-fold by Li ϩ at 1-50 mM, consistent both with the general ionic effect and with similar results reported previously (40).
Although the therapeutic mechanism of action of Li and VPA are not known, both anti-bipolar drugs cause a decrease in intracellular inositol. In yeast, inositol is a major metabolic sensor for the secretory and unfolded protein response pathways (54). 2 If this function is conserved in human cells, inositol depletion by Li and VPA is a first common step that may lead to perturbation of these pathways, one of which (the secretory pathway) plays a fundamental role in neurotransmitter release. Functional conservation of MIP synthase from yeast to humans underscores the power of the yeast model to study the mechanism of action of anti-bipolar drugs in inositol metabolism.