A Phosphatidylinositol 3-Kinase and Phosphatidylinositol Transfer Protein Act Synergistically in Formation of Constitutive Transport Vesicles from the Trans-Golgi Network*

Current evidence suggests that phosphatidylinositol (PI) kinases and phosphatidylinositol transfer protein (PITP) are involved in driving vesicular traffic from yeast and mammalian trans-Golgi network (TGN). We have tested the interaction between these cytosolic proteins in an assay that measures the formation of constitutive transport vesicles from the TGN in a hepatocyte cell-free system. This reaction is dependent on a novel PI 3-kinase, and we now report that, under conditions of limiting cytosol, purified PI 3-kinase and PITP functionally cooperate to drive exocytic vesicle formation. This synergy was observed with both yeast and mammalian PITPs, and it also extended to the formation of PI 3-phosphate. These collective findings indicate that the PI 3-kinase and PITP synergize to form a pool of PI 3-phosphate that is essential for formation of exocytic vesicles from the hepatocyte TGN.

Much effort has recently been focused on understanding the molecular mechanisms that underlie the various vesicular trafficking reactions that operate throughout the eukaryotic secretory pathway. The p62 cplx is a cytosolic complex required for the formation of polymeric IgA receptor (pIgA-R) 1 containing exocytic transport vesicle from the TGN of hepatocytes. The p62 cplx consists of a 62-kDa phosphoprotein and a 25-kDa GTPase and regulates the activity of a novel PI-specific PI 3-kinase (1,2). In cytosol, the p62 molecule is phosphorylated and is not associated with the PI 3-kinase catalytic subunit. Upon receipt of some unknown signal, p62 is dephosphorylated, the PI 3kinase regulatory p62 cplx and catalytic subunits assemble with the cytoplasmic domain of TGN38 (an integral membrane pro-tein of the TGN), and exocytic vesicle formation ensues. The PI 3-kinase activity is wortmannin-sensitive at micromolar concentrations and is stimulated by the activation of the p62 cplxassociated 25-kDa GTPase. Present evidence suggests that the essential function of the PI 3-kinase in exocytic vesicle formation from the TGN is the generation of a specific PI(3)P pool. Supporting evidence comes from the demonstration that both catalytic activity and vesicle formation are equally inhibited by wortmannin.
PITPs also have been demonstrated to function in vesicle formation from the TGN in yeast and mammalian systems (3)(4)(5). The yeast PITP (Sec14p) is required for the formation of yeast Golgi-derived secretory vesicles (4), and this essential Sec14p requirement can be bypassed by modulation of metabolic flux through specific phospholipid biosynthetic pathways (6 -8). In mammalian membrane trafficking reactions both the formation of TGN-derived transport vesicles of constitutive and regulated secretory pathways and the regulated fusion of secretory granules with the plasma membrane are stimulated by PITP (5,9). Whereas PITP cooperates with at least one other unidentified cytosolic factor to stimulate TGN-derived vesicle production, the mechanism of PITP function in that reaction remains unresolved (5). In the secretory granule fusion reaction, PITP synergizes with phospholipid kinases to generate PI 4,5-bisphosphate (10). One of the mechanisms by which PITP may stimulate phosphoinositide synthesis is by presenting PI to PI kinases (11)(12)(13). This concept remains controversial (14).
In this paper, PITP is shown to be an essential component required for the efficient, cell-free formation of pIgA-R containing exocytic vesicles from the hepatocyte TGN. PITP synergizes with the p62 cplx -associated PI 3-kinase in the formation of PI(3)P, and this synergy extends to formation of exocytic transport vesicles from the TGN.

Materials
Unless otherwise indicated, all chemicals were obtained from Sigma or Boehringer Mannheim. Phosphatidylinositol was purchased from Avanti Polar Lipids, (Alabaster, AL). Production of specific antibodies against the pIgA-R and PITP has been described (15,16).

Methods
Subcellular Fractionation Procedures-Stacked Golgi fractions (SGF) were isolated from rat liver according to Taylor et al. (17). Briefly, livers were removed, finely minced, and resuspended at 6 g/10 ml 0.5 M sucrose in 100 mM KPO 4 , pH 6.8, 5 mM MgCl 2 , and 1 g/ml each of a mixture of proteolytic inhibitors: chymstatin, leupeptin, antipain, and pepstatin. All sucrose solutions contained the same buffer and proteolytic inhibitors. The homogenate was centrifuged (1500 ϫ g for 10 min) to pellet unbroken cells, cell debris, and nuclei. This pellet contained at least 50% of the cell protein. The resulting supernatant (PNS) was loaded in the middle of a sucrose step gradient in an SW28 tube; steps of 1.3 and 0.86 M sucrose were overlaid with the PNS supernatant (0.5 M) followed by a 0.25 M layer and centrifuged for 1 h at 100,000 ϫ g (Beckman Instruments, Palo Alto, CA). The 0.5 M sucrose soluble fraction was collected and used for the preparation of cytosol. The SII fraction (0.5/0.86 M interface) was adjusted to 1.15 M sucrose with 2 M sucrose using a refractometer (Bausch & Lomb, Boston, MA). The adjusted SII was loaded into the bottom of an SW28 tube and overlaid with equal volumes of 1.0, 0.86, and 0.25 M sucrose and centrifuged for 3 h at 76,000 ϫ g. The resulting SGF floated to the 0.25/0.86 M sucrose interface. The two-dimensional gel mapping of the protein composition of the fraction is presented in Taylor et al. (18). To prepare cytosol the soluble 0.5 M sucrose fraction of the first gradient was adjusted to 0.25 M sucrose with 100 mM KPO 4 , pH 6.8, 5 mM MgCl 2 and centrifuged for 30 min at 100,000 ϫ g to remove any pelletable material. The remaining supernatant was concentrated using an Amicon fitted with a PM10 membrane to ϳ40 mg/ml (Amicon, Beverly MA). Protein assays (DC Protein Assay, Bio-Rad) were carried out on all fractions. Aliquots of these fractions were frozen in liquid nitrogen and stored at Ϫ70°C.
Gel Electrophoresis and Immunoblotting-SDS-polyacrylamide gel electrophoresis was carried out using a 5-15% acrylamide gradient and the buffer system of Maizel (19). SDS-polyacrylamide gel electrophoresis molecular weight standards were from Bio-Rad. For immunoblots, nitrocellulose filters (Schleicher & Schuell) were blocked for 1 h in 5% defatted milk/phosphate-buffered saline/0.02% sodium azide. The filters were incubated overnight in primary antibody and washed. When using a mouse primary antibody, the filters were incubated with rabbit antibodies against mouse IgG for 2 h before the blots were visualized using 125 I-protein A (ICN, Costa Mesa, CA) by autoradiography and quantitated by PhosphorImager analysis (Molecular Dynamics, Sunnyvale, CA).
Immunopurification of p62 cplx -associated PI 3-Kinase from SGF-Immunopurification and analysis of PI 3-kinase activity were as described (2). SGF (1 mg) was solubilized in 1 ml of CHAPS buffer (20 mM HEPES, pH 6.8, 100 mM KCl, 20 mM CHAPS and proteolytic inhibitors) for 1 h on ice with vortexing. Samples were centrifuged for 30 min at 14,000 ϫ g in a microfuge, and the soluble material was incubated overnight with an immunoaffinity column to which antibodies against p62 were covalently bound. The column was washed with three bed volumes of CHAPS buffer and seven bed volumes of phosphate-buffered saline, and the column was eluted with two bed volumes 0.2 M glycine, pH 2.8. The eluted fraction was neutralized and concentrated and dialyzed into PAN (20 mM PIPES, pH 7.0, 100 mM NaCl) for assay of enzymatic activity. The p62 cplx -associated PI 3-kinase immunopurified from SGF contains dimeric TGN38 (the transmembrane receptor for the PI 3-kinase), the p62 cplx (62-kDa regulatory subunit bound to a 25-kDa GTPase), and an ϳ100-kDa catalytic subunit. All experiments are carried out with p62 cplx -associated PI 3-kinase isolated from SGF. The p62 cplx in the cytoplasm is not associated with the PI 3-kinase catalytic subunit and has no PI 3-kinase activity.
Purification of Sec14p-Hexahistidine-tagged Sec14p was expressed in Escherichia coli and purified essentially as described previously (20). Briefly, cells were harvested, resuspended in ice-cold lysis buffer (50 mM sodium phosphate, pH 7.1, 300 mM sodium chloride, 10 mM 2-mercaptoethanol, 1 mM NaN 3 , 0.2 mM phenylmethylsulfonyl fluoride) and disrupted in a bead beater (Biospec Products). The homogenate was serially clarified by centrifugation at 5,000 ϫ g, 12,000 ϫ g, and 100,000 ϫ g. Sec14p was precipitated at 50% saturation with ammonium sulfate, and precipitates were dissolved in lysis buffer, dialyzed exhaustively against the same, loaded onto a column of Ni ϩ -nitrilotriacetic acid resin (Qiagen), and eluted with a linear gradient of imidazole (0 -200 mM) in lysis buffer. Peak fractions were collected and dialyzed against lysis buffer. The purified Sec14p does not contain either PI or phosphatidylcholine (PC) because these two lipids are not synthesized in E. coli. Sec14p was subsequently bound either with egg PC or PI (Avanti Polar Lipids) by incubation at room temperature for 3-4 h in the presence of at least a 100:1 molar ratio of phospholipid:Sec14p with phospholipid presented in the form of unilamellar vesicles. Protein was rebound to a Ni ϩ -nitrilotriacetic acid column and re-eluted as before. The preparation was dialyzed exhaustively against 10 mM HEPES, pH 7.0, 150 mM KCl, 10 mM 2-mercaptoethanol, 1 mM NaN 3 , 0.2 mM phenylmethylsulfonyl fluoride. Purified Sec14p (25 g) contained one unit of PI transter activity.
Generation of Rat PITP␣ E. coli Lysate-Hexahistidine-tagged rat PITP␣ was generated by cloning the rat PITP␣ structural gene into the pQE31 vector (Qiagen). E. coli expressing the His 6 -tagged PITP␣ were harvested, resuspended in ice cold lysis buffer, and disrupted as above. The homogenate was serially clarified at 5,000 ϫ g, 12,000 ϫ g, and 100,000 ϫ g, and the 100,000 ϫ g supernatant was used in the PI 3-kinase and transfer assays. The bacterial high speed supernatant (1 mg) contained one unit of PI transfer activity.
Phosphatidylinositol Transfer Assays-PI transfer assays have been described previously (20). Briefly, rat liver microsomes were employed as [ 3 H]PI donors in the transfer reaction, and unlabeled PC liposomes served as acceptor vesicles. Reaction mixtures (0.25 M sucrose, 1 mM EDTA, and 5 mM Tris-HCl, pH 7.4) were incubated with either purified Sec14p or lysates containing rat PITP␣ lysates at 37°C. After 30 min the reactions were centrifuged at 10,000 ϫ g for 10 min to pellet the donor microsomes, and 1 ml of the supernatant was collected for scintillation counting. Under these conditions, the PI transfer reaction is linear as long as the input concentrations of yeast or mammalian PITP sustain 20% transfer or less. One unit of activity is defined as the amount of transfer protein that catalyzes the transfer of 1% radiolabeled phospholipid in 1 min (21).
Cell-free Assay of pIgA-R Containing Exocytic Vesicle Formation from the TGN-The cell-free assay of budding from an immobilized SGF was carried out as described (22). Each assay contains 2.5 mg of magnetic core and shell beads with approximately 50 g of SGF immobilized. The immobilized fraction is characterized in Ref. 23. For the budding reaction the immobilized fraction was incubated in 2.5 ml containing 0.70 mg/ml cytosol, 25 mM HEPES, pH 6.7, 25 mM KCl, 1.5 mM magnesium acetate, 1.0 mM ATP, an ATP regenerating system (8.0 mM creatine phosphate, 0.043 mg/ml creatine phosphokinase), and 5 mg/ml bovine serum albumin (final concentrations). After 10 min at 37°C the Golgi fraction remaining on the beads was retrieved with a magnet, and the budded vesicles remained in the supernatant. The high concentration of soluble protein made it impractical to carry out gel analysis on the total budded fraction. Therefore, the budded fraction was pelleted through a 0.25 M sucrose cushion (for 1 h at 100,000 ϫ g) to reduce the large amounts of cytosolic protein and 5 mg/ml bovine serum albumin present in the budding reaction. The pellet was resuspended in gel sample buffer and resolved by SDS-polyacrylamide gel electrophoresis. The amount of exocytic vesicle budding was determined by quantitative immunoblotting using the mature, sialylated pIgA-R (116 kDa) as the marker, and budding efficiency is calculated by determining the percentage of the 116-kDa form of the pIgA-R that is present in the budded fraction with reference to the total amount in the starting immoblized SGF. The pIgA-R is a PM receptor synthesized in relatively high amounts in rat liver (15) and is used to define a specific population of exocytic vesicles (22). Budding of this marker in the presence of the complete cell-free system is ϳ70% efficient. When the ATP regenerating system and cytosol are omitted, the background budding is ϳ5%. There is no detectable PITP on the SGF, therefore, in antibody inhibition studies the antibodies against PITP were added only to cytosol for 30 min on ice before addition of cytosol to the assay. Chromatography; Union City, CA). The TLC plates had been pretreated with 60 mM EDTA, 2% sodium tartrate, and 50% EtOH and dried in a 100°C oven overnight. Development of the TLC plates was in CHCl 3 : MeOH:4 N NH 4 OH (9:7:2) for approximately 2 h, and then the plates were dried for PhosphorImager analysis and subsequently exposed to film for autoradiography.

RESULTS AND DISCUSSION
PITP Plays an Essential Role in Budding of Exocytic Vesicles from the Hepatocyte TGN-PITP is a cytosolic factor that plays an essential role in secretory vesicle formation from the yeast Golgi complex and constitutive and regulated secretory granules in neuroendocrine cells (4 -6). We have used two independent approaches to examine whether PITP is required for cellfree formation of pIgA-R containing vesicles from the rat hepatocyte TGN. First, the cell-free assay was challenged with a polyclonal antibody that recognizes both rat PITP␣ and PITP␤ isoforms (16). A concentration-dependent inhibition of vesicle formation was observed (Fig. 1). The ϳ65% budding efficiency of the control assay was reduced to ϳ40, 25, and 10% by 5, 10, and 15 g of antiserum, respectively. This inhibition was specific as in the presence of an equivalent amount of preimmune serum the budding efficiency remained at ϳ65%.
As a second approach to test for a role for PITP in formation of exocytic vesicles from the TGN, we examined whether PITP was capable of stimulating vesicle formation when the assay was carried out in the presence of limiting concentrations of cytosol. Cytosol titration experiments demonstrated that formation of pIgA-R containing vesicles was linear at cytosol concentrations ranging from 0.25 to 1.25 mg/ml ( Fig. 2A). At higher cytosol concentrations, pIgA-R vesicle formation plateaued at an efficiency of approximately 70%. For subsequent experiments, cytosol was limited to 0.25 mg/ml, a concentration that provides an ϳ12% efficiency of pIgA-R vesicle formation, and increasing concentrations of PITP were added. Yeast PITP (Sec14p) was employed because the recombinant protein was quite stable and could be readily purified to homogeneity. Previous studies had demonstrated the interchangability of Sec14p and mammalian PITP in phosphoinositide-dependent systems reconstituted from mammalian cells (5,10,12). At the highest concentration of PITP, vesicle formation was restored to an efficiency of approximately 40% (Fig. 2B). Similarly, supplementation of the budding assay with p62 cplx restored vesicle budding to an efficiency of ϳ40%. Neither component alone was sufficient to restore the assay to the full budding efficiency of ϳ70%. Strikingly, co-addition of PITP and p62 cplx elicited a dramatic stimulation in vesicle formation relative to that achieved by addition of either component alone. The coaddition of both components at concentrations that individually restored vesicle budding activity to efficiencies of ϳ25% effected a cooperative restoration of vesicle formation to maximal efficiency (ϳ70%).
In summary, antibodies against PITP␣ specifically inactivated the vesicle formation reaction (Fig. 1). When the cell-free assay is carried out in limiting amounts of cytosol, introduction of either yeast Sec14p or p62 cplx -associated PI 3-kinase to the assay stimulated vesicle formation (Fig. 2B). Importantly, PITP and p62 cplx -associated PI 3-kinase synergized to drive pIgA-R vesicle formation to the full efficiency achieved with an optimal amount of cytosol (Fig. 2B). These data indicate that PITP and p62 cplx -associated PI 3-kinase functionally cooperate in the formation of exocytic vesicles from the TGN.

Synergy between PITP␣ and the p62 cplx -associated PI 3-Kinase Extends to Synthesis of PI(3)P-
The similar wortmannin sensitivities of p62 cplx -associated PI 3-kinase activity and vesicle formation from the TGN suggested that PI(3)P formation underlies the p62 cplx -associated PI 3-kinase requirement for vesicle formation. If PI 3-kinase activity and PI(3)P production are prerequisites for exocytic vesicle formation, the functional synergy between the p62 cplx -associated PI 3-kinase and PITP should extend to PI(3)P production. Moreover, the production of PI(3)P should be stimulated by GTP because activation of the p62-associated small GTPase results in activation of PI 3kinase activity (2).
To establish that the mammalian PITP␣ exhibits the capability to stimulate p62 cplx -associated PI 3-kinase activity, recombinant PITP␣ (in the form of a bacterial high speed supernatant) was added to the immunopurified p62 cplx -associated PI 3-kinase. As shown in Fig. 3A, a bacterial high speed supernatant from E. coli not expressing PITP␣ sustained only a low level of PI 3-kinase activity. By contrast, addition of bacterial high speed supernatant that contained PITP␣ supported a FIG. 1. PITP is essential for exocytic vesicle formation. The cell-free assay was carried out as described under "Methods" in the presence of increasing concentrations of an antiserum (5, 10, and 15 g) that recognizes rat PITP␣ and PITP␤. Controls assays were as follows: Control(-ATP), the absence of cytosol and ATP; Control(ϩATP), the presence of cytosol and ATP; and PreImmune 15 g, in the presence of preimmune antiserum. The budded fraction was resolved by SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, immunoblotted with antibodies against the pIgA-R, detected with 125 I-protein A, visualized by autoradiography, and quantitated with a Phosphor-Imager. The budding efficiency of formation of pIgA-R containing vesicles (amount of the 116-kDa form of the pIgA-R in budded fraction/total amount 116-kDa form in immobilized SGF) was calculated and is plotted for each condition.
FIG. 2. PITP and p62 cplx -associated PI 3-kinase are synergistic in restoration of vesicle formation at limiting concentrations of cytosol. The cell-free assay was carried out as described under "Methods" in the presence of increasing amounts of cytosol (0 -2.5 mg/ml). The efficiency of formation of pIgA-R containing vesicles is plotted versus the concentration of cytosol in each assay (A). Cell-free assays were carried out in the presence of limiting amounts of cytosol (0.25 mg/ml) with the addition of increasing concentrations of purified Sec14p (0 -100 g) (B, diamonds) or the immunopurified p62 cplx -associated PI 3-kinase (0 -60 g) (squares). To test for additive effects of using the two components in combination, increasing concentrations of Sec14p were added in the presence of 20 g of p62 cplx -associated PI 3-kinase (B, circles).
3-5-fold concentration-dependent stimulation of the p62 cplxassociated PI 3-kinase activity (Fig. 3A). This stimulation was amplified further by activation of the 25-kDa GTPase bound to p62 (Fig. 3B). Whereas p62 cplx -associated PI 3-kinase was individually stimulated 3-and 5-fold by a fixed concentration of GTP␥S (100 nM) and PITP␣, respectively, the simultaneous addition of both reagents resulted in ϳ30-fold stimulation (Fig.  3B). These data revealed a powerful synergy between the GTPase bound to p62 and PITP␣ in stimulation of the p62 cplxassociated PI 3-kinase activity.
p62 cplx -associated PI 3-Kinase Can Utilize PITP-bound Phospholipid as Substrate-There are presently two general views for how PITP might stimulate p62 cplx -associated PI 3-kinase activity. The "substrate presentation" model posits that PITP acts as a co-factor that presents PI to PI kinases and thereby stimulates the initial rate of the headgroup phosphorylation reaction (11)(12)(13). The "lipid transfer" model proposes that PITP merely sustains PI kinase activity by effecting transfer of PI down a chemical gradient that is itself created by depletion of PI by metabolic enzymes such as PI kinases and PI phospholipases (14).
To examine the possibility that PITP presents substrate to the p62 cplx -associated PI 3-kinase, PI 3-kinase assays were performed in a membrane-free system using purified Sec14p loaded with either PC (Sec14p-PC) or PI (Sec14p-PI). These experiments allowed the interaction between a lipid kinase and PITP to be examined in a purified system. The source of PI in the kinase assays was limited to that which was stoichiometrically bound to Sec14p (20 g/ml, and this concentration of PI was an order of magnitude lower than that present in the standard assay (i.e. 200 g/ml) (Fig. 4A). Under these assay conditions, the p62 cplx -associated PI 3-kinase alone had minimal PI 3-kinase activity. Introduction of Sec14p-PC in the assay reduced that activity to base-line levels. By contrast, Sec14p-PI stimulated the p62 cplx -associated activity 5-fold. These data suggest that the p62 cplx -associated PI 3-kinase is capable of directly phosphorylating PI directly bound to PITP and that this presentation of PI enhances kinase activity.
To further characterize these interactions, additional PI was added to a parallel set of kinase assays to bring the total PI concentration (Sec14p-bound and free) to the levels present in the standard assay (200 g/ml). A stimulation of PI(3)P formation was observed under all assay conditions (Fig. 4B). Addition of either Sec14p-PC or Sec14p-PI to the assay increased PI 3-kinase activity 6 -7-fold. This stimulatory effect of Sec14p-PC is attributed to its ability to rapidly exchange bound ligand for the excess PI present in the reaction mixture, effectively converting Sec14p-PC to Sec14p-PI during the course of the reaction.
Activation of the GTPase bound to p62 provided an additional enhancement of PI 3-kinase activity as measured by PI(3)P production (Fig. 4C). Sec14p-PI and GTP␥S individually stimulated this activity some 4-fold, and co-addition of both Sec14p-PI and GTP␥S effected a synergistic 10 -12-fold stimulation of the PI 3-kinase activity.
A Role for PITP in PI 3-Kinase Activity-The collective data reported herein demonstrate that the p62 cplx -associated PI 3-kinase is stimulated both by Sec14p and mammalian PITP␣. More dramatically, activation of the small GTPase bound to p62 in the presence of PITP␣ supported a synergistic activation of the p62 cplx -associated PI 3-kinase of up to 30-fold. In addition to activating the generation of PI(3)P, the cooperation of PITP␣ and the p62 cplx -associated PI 3-kinase are essential for pIgA-R vesicle formation from the TGN. We propose that a specific pool of PI93)P is generated "on demand" at the site at which it will be utilized. In this regard, we emphasize that the p62 cplx -FIG. 3. Mammalian PITP␣ and activation of the GTPase bound to p62 synergistically enhance the rate of PI(3)P production by the p62 cplxassociated PI 3-kinase. PI 3-kinase assay were carried out for increasing amounts of time (0 -20 min) using the immunopurified p62 cplx -associated PI 3-kinase (50 ng) (squares). Mammalian PITP␣ was added to the assay at 2 and 5 g of bacterial high speed supernatant (2 g, diamonds and 5 g, circles) (A). To test if activation of the small GTPase influences the PI 3-kinase activity, assays were carried out in the presence or absence of GTP␥S (100 nM) under the following conditions: p62 cplx -associated PI 3-kinase (open squares); p62 cplx -associated PI 3kinase and GTP␥S (squares with ϫ); p62 cplx -associated PI 3-kinase and PITP␣ (circles); p62 cplx -associated PI 3-kinase and PITP␣ and GTP␥S (triangles) (B). Note that the PhosphorImager units change increase from A (350,000 maximum) to B (4,500,000 maximum). associated PI 3-kinase assembles with the cytosolic domain of an integral membrane protein of the TGN (TGN38) (2). This organization suggests that trans-Golgi proteins (perhaps even cargo proteins) provide positional cues for assembly of this specific PI 3-kinase.
There are multiple steps in formation of vesicles from the TGN, and the cell-free assay more than likely measures all of these steps. These include sorting of molecules, tubule formation, and vesicle budding (22,25,26). We propose that PI(3)P is a key regulatory molecule in the vesicle formation and that its generation may provide signaling molecule required for activation/integration of any one of these steps.
Recently others have shown a similar PI 3-kinase activity associated with TGN46, the human orthologue of TGN38 (27). Those studies failed to reveal either a GTPase or a PITP␣mediated activation of the associated PI 3-kinase activity. Unfortunately, the relevant experiments performed in that study are difficult to evaluate because the specific activities for PI transfer of the PITP␣, and the concentrations of GTP␥S employed are not provided, so it is difficult to relate these two bodies of work in a meaningful manner.
Finally, our data speak to how PITPs might cooperate with PI 3-kinase in exocytic vesicle formation from the TGN. The demonstration that p62 cplx -associated PI 3-kinase activity is stimulated when the sole source of PI in the kinase assay was PITP-bound lipid strongly suggests that p62 cplx -associated PI 3-kinase is at least capable of utilizing the PI presented by PITP as an effective (and perhaps even optimal) substrate for phosphorylation (Fig. 4A). These findings are consistent with the notion that the PITP-bound PI pool represents the physiologically relevant substrate for the p62 cplx -associated PI 3kinase. Moreover, the finding that PC-bound PITP was inactive in the stimulation of PI 3-kinase argues against a role for the PC-bound PITP effecting some allosteric stimulation of the kinase. An important caveat to these membrane-free PI 3kinase experiments is that Sec14p was the source of PITP. Our choice of Sec14p for these studies was driven by its stability and our ability to rapidly purify it to homogeneity. The mammalian PITP␣ was unstable after purification, which made these experiments impossible. It is important to note that present evidence indicates that the in vivo function of Sec14p is directed at maintaining a pool of Golgi membrane diacylglycerol in a manner that employs converging functions for the PCand PI-bound forms of Sec14p (8). There is no evidence that Sec14p synergizes with a yeast PI 3-kinase in a physiologically relevant way. Nonetheless, both mammalian PITP␣ and FIG. 4. Sec14p and an activated GTPase synergistically enhance the rate of PI(3)P production by p62 cplxassociated PI 3-kinase. PI 3-kinase assay were carried out for increasing amounts of time (0 -20 min) using the immunopurified p62 cplx -associated PI 3-kinase (50 ng) (no membrane was present in the assay). The substrate was PI (squares), PI bound to Sec14p (diamonds), or PC bound to Sec14p (circles). The total lipid concentration of the assay was 20 g/ml, the amount that is stoichiometrically bound to Sec14p (A). PI 3-kinase assays were carried out as in A with the addition of PI to adjust its final concentration to 200 g/ml. The symbols are the same as in A. To test whether activation of the small GTPase bound to p62 influences the PI 3-kinase activity, assays were carried out using 200 g/ml PI in the presence or absence of GTP␥S (100 nM) under the following conditions: p62 cplxassociated PI 3-kinase (filled squares); p62 cplx -associated PI 3-kinase and GTP␥S (half-filled squares); p62 cplx -associated PI 3-kinase and Sec14p (closed diamonds); and p62 cplx -associated PI 3-kinase and Sec14p and GTP␥S (half-filled diamonds) (C). Note that the PhosphorImager units change increase from A (150,000 maximum), B (200,000 maximum), to C (600,000 maximum).
Sec14p behaved similarly in the cell-free assay of pIgA-R vesicle formation from the TGN with regard to synergy with p62 cplx -associated PI 3-kinase activity. The synergy observed between the p62 cplx -associated PI 3-kinase activity and both mammalian PITP␣ and Sec14p indicates that a functional interaction between these factors is critical for PI(3)P generation. Further, a spatially and temporally regulated burst of PI(3)P synthesis is a prerequisite for exocytic vesicle formation from the TGN.