The Isolation and Characterization of Inositol Polyphosphate 4Phosphatase*

We previously identified an alternative pathway for the metabolism of inositol 1,3,4-trisphosphate (Ins(1,3,4)P3) in calf brain. The enzyme responsible for the degradation of Ins(1,3,4)P3 was designated as inositol polyphosphate 4-phosphatase (Bansal, V. S., Inhorn, R. C., and Majerus, P. W. (1987) J. Biol. Chem. 262, 9644-9647). We have now purified this enzyme 3390-fold from calf brain-soluble fraction. The isolated enzyme has an apparent molecular mass of 110 kDa as determined by gel filtration. On sodium dodecyl sulfate-polyacrylamide gel electrophoresis, the enzyme migrates as a protein of 105 kDa, suggesting that it is monomeric. Among various 4-phosphate-containing inositol polyphosphates, the enzyme hydrolyzes only Ins(1,3,4)P3 and inositol 3,4-bisphosphate (Ins(3,4)P2), yielding inositol 1,3-bisphosphate and inositol 3-phosphate as products. The inositol polyphosphate 4-phosphatase has apparent Km values of 40 and 25 microM for Ins(1,3,4)P3 and Ins(3,4)P2, respectively. The maximum velocities for these two substrates are 15-20 mumol of product/min/mg protein. Ins(1,3,4)P3 is a competitive inhibitor of Ins(3,4)P2 hydrolysis with an apparent Ki of 27 microM implying that the same active site is involved in hydrolysis of both substrates. The final enzyme preparation retained a small inositol polyphosphate 3-phosphatase activity (less than 2% of rate of inositol polyphosphate 4-phosphatase activity) which most likely reflects a contaminant. The enzyme displays maximum activity between pH 6.5 and 7.5. It is not inhibited by Li+, Ca2+, or Mg2+ except at 10 mM divalent ions. Mn2+ inhibits enzyme at high concentrations IC50 = 1.5 mM.

A widely distributed response of cells to hormones, growth factors, and other extracellular agonists is a rapid receptormediated hydrolysis of inositol phospholipids to produce several messenger molecules (l-4).
The metabolism of these phospholipids is initiated by inositol phospholipid-specific phospholipase C, which liberates a mixture of cyclic and noncyclic inositol phosphates (4,5).
Initially, the metabolism of inositol  (20). This enzyme is Meindependent and is not inhibited by Li'. The relative activities of these two pathways varies widely in different organs and tissues; in calf brain, the 4-phosphatase pathway predominates (23). We

RESULTS
We purified inositol polyphosphate 4-phosphatase from the soluble fraction of calf brain which contains 60-75% of the total 4-phosphatase activity in this tissue. After the initial DEAE-Sepharose chromatography the enzyme was chromatographed on a DEAE-HPLC column as shown in Fig. 1. The active fractions from this column were pooled, further purified by ammonium sulfate precipitation, chromatographed on phosphocellulose, and then on Mono S as shown in Fig. 2. We obtained a sharp peak of inositol polyphosphate 4-phosphatase activity; however, this peak did not correspond to the protein peak. The inositol polyphosphate 4-phosphatase was further purified by gel filtration as shown in Fig. 3. The gel filtration profile shows a major peak of contaminating protein eluting just before inositol polyphosphate 4-phosphatase activity, but there was a small peak of protein that did correspond to the enzyme-containing fractions. The inositol polyphosphate 4-phosphatase had an apparent molecular mass of 110 kDa. A sodium dodecyl sulfate-polyacrylamide gel of fractions containing enzyme activity is shown in Fig. 4 proximately 105 kDa and the profile of enzyme activity in each fraction. Another protein migrating at 79 kDa partially correlated with enzyme activity but seemed unlikely to be the inositol polyphosphate 4-phosphatase, since it appeared to be absent in fractions 82 and 83. In order to establish further that the protein migrating at approximately 105 kDa represents the enzyme, we repeated the purification and added hydroxylapatite chromatography before the gel filtration step. Although this additional procedure did not yield material of higher specific activity, it did remove the 79-kDa protein as shown in Fig. 4B. In this case, the major contaminant is a 70-kDa protein not present in enzyme-containing fractions of the preparation shown in Fig. 4A. We have carried out this purification five times, and in each case the protein of 105 kDa corresponds to enzyme activity.
The molecular weight determined by gel filtration also agrees with that determined by sodium dodecyl sulfate gel  1,3,4)P3 as substrate, we observed both the formation of Ins(l,3)PP as well as Ins(l)P. This suggests inositol polyphosphate 3-phosphatase activity on the product of inositol polyphosphate 4phosphatase.
When assays were carried out using Ins(1,3)Ps as substrate, we found 0.29 pmol of Ins(1,3)Pz hydrolyzed/ min/mg protein. The activity of inositol polyphosphate-3phosphatase in crude calf brain supernatant is 2 nmol of Ins(l,3)Pp hydrolyzed/min/mg protein. Thus inositol polyphosphate-3-phosphatase is purified about 150-fold. The ratio of inositol polyphosphate 4-phosphatase to inositol polyphosphate 3-phosphatase increases from 1.6 in the crude to 60 in the final preparation implying that inositol polyphosphate 3phosphatase activity is most likely due to a contaminant protein and not an activity intrinsic to the inositol polyphosphate 4-phosphatase.
This conclusion is supported further by heat inactivation experiments, where after 10 min at 50 "C 90% of inositol polyphosphate 3-phosphatase activity is lost while 50% of inositol polyphosphate 4-phosphatase remains. Properties of Inositol Polyphosphate I-Phosphatase-We assayed inositol polyphosphate 4-phosphatase for activity with the other 4-phosphate containing substrates as shown in Table II. Only Ins(1,3,4)P3 and Ins(3,4)P2 were hydrolyzed. The enzyme was not inhibited by 2 mM of other phosphate esters as shown in Table II.
The enzyme has optimum activity between pH 6.5 and 7.5. The effect of varying Ins(1,3,4)P3 concentration on inositol polyphosphate I-phosphatase is shown in Fig. 5. In the experiment shown, the apparent K,,, was 40 pM Ins(1,3,4)   Inositol polyphosphate 4-phosphatase does not require divalent metal ions (e.g. Ca'+, MF, Mn*+) for its activity. Li+ up to 50 mM has no effect on inositol polyphosphate 4phosphatase activity. Phosphate and ammonium sulfate strongly inhibit enzyme activity having an IC, = 10 and 2 mM, respectively. Mn*+ ions inhibited the enzyme as shown in Fig. 7 although significant inhibition was achieved only at concentrations over 1 mM. In contrast inositol polyphosphate 1-phosphatase is inhibited by lower concentrations of Mn*+ with 50% inhibition at 0.01 mM. There was also inhibition of inositol polyphosphate 4-phosphatase at 10 mM Mg2' and Ca2+ as seen in Fig. 7. EDTA had no effect on enzyme activity although, when enzyme was diluted into buffers without EDTA, enzyme activity was lost rapidly (80% in 2 h). [3H]Ins(3,4)P2 (5-100 j.tM) was incubated in the presence (0) and absence (0) (27) reported a 4-phosphatase activity in rat brain which had a K,,, of 50 pM for Ins (3,4)P2. This crude preparation also hydrolyzed Ins(1,3,4)P3. Takimoto et al. (28) purified an enzyme from rat brain which they designated as inositol 1,4bisphosphate 4-phosphohydrolase. This enzyme was Mpdependent and inhibited by Li'. Thus it displayed the prop-erties of inositol polyphosphate 1-phosphatase (21)(22)(23). In conflict with their observations, inositol polyphosphate 4phosphatase from calf brain is neither inhibited by lithium ions nor does it hydrolyze Ins(1,4)Pz. The most likely explanation for the apparent discrepancy is that Takimoto et al. (28) actually purified inositol polyphosphate 1-phosphatase and mis-identified the product as Ins(l)P rather than Ins(4)P. The relative activities of inositol polyphosphate l-phosphatase uersus inositol polyphosphate 4-phosphatase may determine the levels of various intermediates produced from the metabolism of Ins (1,3,4)Ps. In human umbilical vein endothelial cells cultured in the presence of Li+, the levels of Ins( 1,3)P2 versus Ins (3,4)Pz varied considerably in response to different agonists (29). The relative activities of inositol polyphosphate 1-phosphatase and inositol polyphosphate 4-phosphatase toward Ins(1,3,4)P3 have been measured in crude extracts of various bovine tissues (23). The proportion that was utilized by inositol polyphosphate 4-phosphatase ranged from 5 to 20% in various tissues except in brain where it was -70%. These observations suggest that Ins(1,3)Pz may play some role in neuronal function. Sherman et al. (30) have directly measured the levels of Ins(l)P and Ins(4)P in rat brain and found Ins(l)P as the predominant inositol monophosphate in both control and lithium-treated animals. In many other tissues, Ins(4)P is the predominant inositol monophosphate (8). Ins(1,3)Pz is a potential source for the large amount of Ins( l)P in brain.