Yck1 casein kinase I regulates the activity and phosphorylation of Pah1 phosphatidate phosphatase from Saccharomyces cerevisiae

The PAH1-encoded phosphatidate phosphatase in Saccharomyces cerevisiae plays a major role in triacylglycerol synthesis and the control of phospholipid synthesis. For its catalytic function on the nuclear/endoplasmic reticulum membrane, Pah1 translocates to the membrane through its phosphorylation/dephosphorylation. Pah1 phosphorylation on multiple serine/threonine residues is complex and catalyzed by diverse protein kinases. In this work, we demonstrate that Pah1 is phosphorylated by the YCK1-encoded casein kinase I (CKI), regulating Pah1 catalytic activity and phosphorylation. Phosphoamino acid analysis coupled with phosphopeptide mapping of the CKI-phosphorylated Pah1 indicated that it is phosphorylated mainly on multiple serine residues. Using site-directed mutagenesis and phosphorylation analysis of Pah1, we identified eight serine residues (i.e. Ser-114, Ser-475, Ser-511, Ser-602, Ser-677, Ser-705, Ser-748, and Ser-774) as the target sites of CKI. Of these residues, Ser-475 and Ser-511 were specific for CKI, whereas the others were shared by casein kinase II (Ser-705), Cdc28–cyclin B (Ser-602), Pho85–Pho80 (Ser-114, Ser-602, and Ser-748), protein kinase A (Ser-667 and Ser-774), and protein kinase C (Ser-677). CKI-mediated phosphorylation of Pah1 stimulated both its phosphatidate phosphatase activity and its subsequent phosphorylation by casein kinase II. However, the CKI-mediated phosphorylation of Pah1 strongly inhibited its subsequent phosphorylation by Pho85–Pho80, protein kinase A, and protein kinase C. In a reciprocal analysis, Pah1 phosphorylation by Pho85–Pho80 inhibited subsequent phosphorylation by CKI. CKI-mediated Pah1 phosphorylation was also inhibited by a peptide containing the Pah1 residues 506–517, including the kinase-specific Ser-511 residue. These findings advance our understanding of how Pah1 catalytic activity and phosphorylation are regulated by multiple protein kinases.

The PAP enzyme is conserved in eukaryotes, including humans (1,23), mice (24,25), flies (26,27), worms (28), fungi (29), and plants (30,31). All PAP enzymes have the haloacid dehalogenase (HAD)-like domain with a DXDX(T/V) catalytic motif and the N-LIP domain whose role in enzyme function is still unclear (Fig. 1B) (1,18,24). The critical roles that PAP plays in humans and mice are characterized by lipid-based syndromes associated with the loss or overexpression of the enzyme. For example, the deficiency of lipin 1 PAP in humans and mice causes rhabdomyolysis (32,33), and its deficiency in mice is also characterized by hepatic steatosis during the neonatal period, lipodystrophy, insulin resistance, and peripheral neuropathy (24,34). In contrast, the overexpression of lipin 1 PAP in mice causes an increase in lipogenesis and obesity (35). Moreover, the mutations of the human LPIN1 gene are associated with insulin resistance and the metabolic syndrome (36).
Clearly, the regulation of the PAP expression and its catalytic activity is crucial for lipid homeostasis and cell physiology (5,37).
Studies using yeast as a model organism have made great strides in understanding the mode of action and regulation of the PAP enzyme (5,6,8). Pah1 PAP is regulated by genetic and biochemicalmechanisms,anditsphosphorylation/dephosphorylation has emerged as a major biochemical regulation required to control its cellular function (3)(4)(5)(6)8). The posttranslational modification is essential for the subcellular localization of Pah1 and also affects its catalytic activity and protein stability (38 -48). Pah1 phosphorylated by Pho85-Pho80 (40) and Cdc28cyclin B (39) is less active due to its sequestration in the cytosol apart from the substrate PA in the nuclear/ER membrane (Fig.  1A). In addition, Pah1 PAP activity is diminished through its phosphorylation by the protein kinases at seven sites (Fig. 1B) distributed at the N and C termini of the protein (39,40). In conjunction with Pho85-Pho80 and Cdc28 -cyclin B, PKA functions to regulate the location and activity of Pah1 through its phosphorylation at five sites ( Fig. 1B) distributed at the beginning and end of the protein (41). Moreover, Pah1 phosphorylated by those protein kinases is protected against its degradation by the 20S proteasome (47,48). In contrast to its protection, Pah1 phosphorylated by PKC, which occurs on four sites at the C terminus (Fig. 1B), is susceptible to the proteasomal degradation (42). The phosphorylation of Pah1 by CKII, which occurs on six sites at the N and C termini, prevents its subsequent phosphorylation by PKA and stimulates its function in TAG synthesis (43). The regulatory effects of Pah1 phosphorylation by PKC (42) and CKII (43) on its stability and function are shown when it is not prephosphorylated at the seven sites by Pho85-Pho80/Cdc28 -cyclin B. The phosphorylation of Pah1 by Pho85-Pho80/Cdc28 -cyclin B exerts an inhibitory effect on its phosphorylation by PKC (42), but it has no effect on its phosphorylation by CKII (43). Unlike their effects on Pah1 phosphorylation, PKC (42) and CKII (43) have little effect on its PAP activity.
The phosphorylated form of Pah1, which is produced by multiple kinases (Fig. 1B), is dephosphorylated by a single protein phosphatase that is composed of the Nem1 (catalytic) and Spo7 (regulatory) subunits (44,49). The catalytic activity of the membrane-bound Nem1-Spo7 phosphatase complex serves in translocating Pah1 to the nuclear/ER membrane; the dephosphorylated form of Pah1, which is associated with the membrane, is more active but prone to proteasomal degradation unlike its phosphorylated form produced by Pho85-Pho80, Cdc28 -cyclin B, and PKA (38 -41, 44, 45, 48). Based on our current understanding, Fig. 1A depicts a model for the phosphorylation/dephosphorylation-mediated regulation of Pah1 localization for its PAP function in lipid synthesis.
In this communication, we further examined the complex regulation of Pah1 by the CKI-mediated phosphorylation and found that Pah1 is a bona fide substrate of CKI, a serine/threonine protein kinase that plays roles in nutrient-mediated cell morphogenesis, cytokinesis, secretion, and endocytosis (50 -54). Of eight serine residues of Pah1 phosphorylated by CK1, two residues (Ser-475 and Ser-511) were unique to the protein kinase and the others were common targets of phosphorylation by CKII, Pho85-Pho80, Cdc28 -cyclin B, PKA, and PKC (Fig.  1B). The CKI phosphorylation of Pah1 stimulates its PAP activity and affects its subsequent phosphorylation by CKII, Pho85-Pho80, PKA, and PKC.

Pah1 is a bona fide substrate of CKI
Pah1 is phosphorylated on at least 40 serine/threonine residues ( Fig. 1B) (38,(55)(56)(57)(58)(59)(60)(61)(62)(63)(64)(65), but protein kinases that target the sites are not all known. The proteome-wide analysis of protein phosphorylation in vitro by Ptacek et al. (66) has indicated that Pah1 is also a target of protein kinases other than those already Figure1.ModelfortheregulationofPah1PAPbyphosphorylation,dephosphorylation, and proteasomal degradation, and domains/regions and phosphorylation sites in Pah1. A, Pah1 is phosphorylated by multiple protein kinases. The phosphorylated Pah1 (small blue circles) is recruited to the ER membrane through its association and dephosphorylation by the Nem1-Spo7 protein phosphatase complex (45,49). Dephosphorylated Pah1 that is associated with the ER membrane via its amphipathic helix (45) catalyzes the conversion of PA to DAG (1), which is then acylated to form TAG. Dephosphorylated Pah1 or PKC-phosphorylated Pah1 that is not phosphorylated at the seven target sites for the Pho85-Pho80 protein kinase (42) is degraded by the 20S proteasome (indicated by the dashed line arrows) (48). PA is also utilized for the synthesis of membrane phospholipids via CDP-DAG (74,107). When the CDP-DAG-dependent pathway for phospholipid synthesis is blocked, phosphatidylcholine or phosphatidylethanolamine may be synthesized from the DAG derived from the PAP reaction if cells are supplemented with choline or ethanolamine via the Kennedy pathway (74,107). B, diagram shows the positions of the amphipathic helix (AH, pink) required for ER membrane interaction known (Fig. 1B). Among the protein kinases are Yck1 and Yck2, which are redundant yeast CKI isoforms (50,51). Yck1 and Yck2, which associate with the plasma membrane via palmitoylation (67,68), play redundant roles in cell physiology (50 -53). In this work, we examined the phosphorylation of Pah1 by Yck1, which is more active on the substrate than Yck2 (66). To examine the CKI phosphorylation of Pah1, we utilized Yck1 purified from yeast and Pah1 purified from heterologous expression in Escherichia coli. The rationale for using the E. coli-expressed Pah1 was to examine its phosphorylation in the absence of endogenous phosphorylation in yeast (38). The affinity-purified Yck1 using a TAP tag was confirmed by immunoblot analysis with anti-protein A antibody ( Fig. 2A). The phos-phorylation of Pah1 by Yck1 was monitored by following the incorporation of the radioactive phosphate from [␥-32 P]ATP into the substrate. In this analysis, the 32 P-labeled Pah1 was resolved from [␥-32 P]ATP by SDS-PAGE and then subjected to phosphorimaging. As shown in Fig. 2B (right), the E. coli-expressed Pah1 was phosphorylated by Yck1. Phosphoamino acid analysis of the 32 P-labeled Pah1 showed that the protein is phosphorylated on serine (75%) and threonine (25%) residues with serine being a major target site (Fig. 2C). Phosphopeptide mapping of the 32 P-labeled Pah1 indicates that the protein is phosphorylated by the CKI enzyme on multiple residues (Fig. 2D).
Using Pah1 as substrate, the CKI activity depended on reaction time and the amount of Yck1, indicating that the protein kinase follows zero-order kinetics (Fig. 3, A and B). Additionally, the CKI activity depended on the concentrations of Pah1 and ATP (Fig. 3, C and D). The V max and K m values for Pah1 are 16 fmol/min and 0.21 M, respectively, and for ATP, the V max and K m values are 5.2 fmol/min and 2.4 M, respectively. For both Pah1 (n ϭ 1.8) and ATP (n ϭ 2.7), the CKI activity followed positive cooperative kinetics, suggesting that the phosphorylation of one site enhances the phosphorylation of another site (69). Overall, these enzymological properties demonstrate that Pah1 is a bona fide substrate of Yck1 CKI.  (59,63,65,72). B, Pah1 (25 g/ml) was incubated for 30 min at 30°C in the presence (ϩ) or absence (Ϫ) of 4 fmol/min Yck1 and 100 M [␥-32 P]ATP (3,000 cpm/pmol). The reaction mixtures were resolved by SDS-PAGE and subjected to phosphorimaging, followed by staining with Coomassie InstantBlue. The amount of Yck1 is too low for detection by protein staining of the polyacrylamide gel. The positions of Pah1 and the molecular mass standards are indicated. C, 32 P-labeled Pah1 phosphorylated by CKI was incubated with 6 N HCl for 3 h at 100°C. The acid hydrolysate was mixed with standard phosphoamino acids and resolved by two-dimensional electrophoresis on a cellulose TLC plate, which was subjected to phosphorimaging and staining with ninhydrin. The positions of phosphoserine (p-Ser), phosphothreonine (p-Thr), and phosphotyrosine (p-Tyr) as detected by ninhydrin staining are indicated. The dashed ellipse designates the absence of phosphotyrosine. D, 32 P-labeled Pah1 phosphorylated by CKI was digested with TPCK-treated trypsin. The phosphopeptides produced by the proteolytic digestion were separated on cellulose TLC plates by electrophoresis (from left to right) in the first dimension and by chromatography (from bottom to top) in the second dimension; the TLC plate was subjected to phosphorimaging. The most heavily phosphorylated phosphopeptides are indicated by the white arrows. The data shown in all panels are representative of three experiments.

Pah1 phosphorylation by CKI increases its catalytic efficiency and relieves its cooperative dependence on PA
We questioned what effect the CKI phosphorylation of Pah1 has on its PAP activity. The CKI-mediated phosphorylation stimulated the PAP activity in a dose-dependent manner (Fig.  5A). Compared with unphosphorylated Pah1, the CKI-phosphorylated one exhibited ϳ2-fold higher activity at its maximum stimulation. To gain insight into a mechanism for the stimulatory effect of the phosphorylation, we examined the unphosphorylated and phosphorylated forms of Pah1 as to their dependence of PAP activity on the surface concentration of PA in the Triton X-100/PA mixed micelle (1,73). As expected (1), unphosphorylated Pah1 displayed positive cooperative kinetics (n ϭ 2.7) with respect to PA (Fig. 5B). The phosphorylation by CKI relieved the cooperativity of the enzyme for its substrate; the phosphorylation reduced the Hill number for the reaction from 2.7 to 1.3. (Fig. 5B). Thus, the phosphorylated enzyme is more active at lower concentrations of PA when  CKI regulates phosphorylation of Pah1 PA phosphatase compared with the unphosphorylated enzyme. The V max values for the phosphorylated and unphosphorylated forms of Pah1 were 3.2 and 1.6 mol/min/mg, respectively. The K m values for PA of the phosphorylated (K m ϭ 4.1 mol %) and unphosphorylated (K m ϭ 3.4 mol %) forms of Pah1 were not majorly different.
The effect of the CKI-mediated phosphorylation on PAP activity was also examined with Pah1 mutant enzymes with the alanine substitutions for the unique CKI sites (Ser-475 and Ser-511) and the seven sites phosphorylated by Pho85-Pho80 (Fig.  5C). These experiments showed that the stimulation of PAP activity by CKI was not due to the phosphorylations at Ser-475, Ser-511, or the sites common to CKI and Pho85-Pho80. In fact, the phosphorylation of Pah1 with the S475A, S511A, and 7A mutations resulted in a greater stimulation (3-3.5-fold) of PAP activity when compared with the stimulation of activity caused by the phosphorylation of WT Pah1 (Fig. 5C).

Interrelationship of Pah1 phosphorylation between CKI and other protein kinases
We questioned whether the prephosphorylation of Pah1 by CKI affects its subsequent phosphorylation by CKII, Pho85-Pho80, PKA, and PKC. We did not examine the phosphorylation interrelationships with Cdc28 -cyclin B because its three target sites are included in the seven sites phosphorylated by Pho85-Pho80 (Fig. 1B). Unphosphorylated Pah1, which had been prepared through heterologous expression in E. coli, was first phosphorylated by CKI with nonradioactive ATP and then phosphorylated by the other protein kinases with [␥-32 P]ATP. Compared with a lack of prephosphorylation, the CKI-mediated prephosphorylation resulted in a dose-dependent stimulation of Pah1 phosphorylation by CKII (Fig. 6A). The prephosphorylation of Pah1 by 10 fmol/min CKI showed an 80-fold increase in its phosphorylation by CKII. The major sites of Pah1 phosphorylated by CKII include Thr-170, Ser-313, Ser-705, and Ser-818 (43). Pah-4A, which contains alanine substitutions for the four residues, did not exhibit the CKI-mediated stimulation of its phosphorylation by CKII (Fig. 5A), confirming that the increased phosphorylation was mediated by CKII. The alanine substitutions for Ser-475 and Ser-511, which are the CKI-specific target residues (Fig. 4), had major effects on phosphorylation by CKII (Fig. 6A). The S511A mutation reduced the CKI stimulatory effect by 80%, whereas the S475A mutation eliminated the stimulatory effect of prephosphorylation (Fig. 5A). These data support the conclusion that the CKI phosphorylation of Pah1 at Ser-475 and Ser-511 primes the subsequent phosphorylation by CKII. The CKI target sites of Pah1 that are also phosphorylated by Pho85-Pho80, PKA, and PKC are not expected to have a stimulatory effect on subsequent phosphorylation by CKII because the common site of prephosphorylation by the other kinases does not exhibit the stimulatory effect (43).
In contrast to the stimulatory effect, Pah1 prephosphorylation by CKI showed a great inhibitory effect on its phosphorylations by Pho85-Pho80 (95%), PKA (67%), and PKC (95%) (Fig.  6, B-D). The inhibitory effect of prephosphorylation on the phosphorylation by Pho85-Pho80 could not be attributed to the CKI-specific phosphorylation sites (i.e. Ser-475 and Ser-511) because alanine substations for the serine residues did not alleviate the inhibitory effect (Fig. 6B). The inhibitory effect of theCKI-mediatedPah1phosphorylationonthesubsequentphosphorylation by Pho85-Pho80 might be explained by the fact that the two protein kinases share common phosphorylation sites (Fig. 1B). In addition, the amount of CKI needed to inhibit the subsequent phosphorylation by Pho85-Pho80 was 10-fold less than that needed to inhibit the subsequent phosphorylation by PKA and PKC or to stimulate the CKII-mediated phosphorylation. The CKI-mediated inhibition of Pah1 phosphorylation by PKA and PKC could be attributed to the phosphorylation at Ser-475 and Ser-511; the S475A and S511A mutations alleviated the inhibition caused by CKI (Fig. 6, C and D).
Inreciprocalexperiments,weexaminedwhethertheprephosphorylation of Pah1 by CKII, Pho85-Pho80, PKA, and PKC affects the subsequent phosphorylation by CKI. The prephosphorylation of Pah1 by Pho85-Pho80 caused a dose-dependent inhibition (75%) of its phosphorylation by CKI (Fig. 7B). However, the prephosphorylation of Pah1 by the other protein kinases had no or little effect on the subsequent phosphoryla- tion of Pah1 by CKI (Fig. 7, A, C, and D). As discussed above, CKI and Pho85-Pho80 share common target sites (Fig. 1B), and thus the inhibitory effect of Pho85-Pho80 on the CKI activity was considered to be caused by a reduction in available target residues. To test this possibility, we analyzed the CKI phosphorylation of Pah1-7A, which has alanine substitutions for the seven target sites of Pho85-Pho80 (40). The 7A mutations reduced (40%) the inhibitory effect of the prephosphorylation by Pho85-Pho80 (Fig. 7B). Thus, the inhibition of the CKImediated phosphorylation is due to the phosphorylation by Pho85-Pho80 of sites common to both protein kinases. However, the data also indicate that phosphorylation by Pho85-Pho80 at sites that are not common to CKI affect the phosphorylation by CKI.

Pah1 phosphorylation by CKI is inhibited by a peptide containing Ser-511
Based on the highest reduction (90%) of Pah1 phosphorylation caused by the S511A mutation, Ser-511 was considered a major site of phosphorylation by CKI (Fig. 4). A peptide consisting of the Pah1 residues 506 -517 was synthesized and examined as a substrate for CKI. In this analysis, the Pah1 pep-tide was phosphorylated by CKI with [␥-32 P]ATP and captured on P81 phosphocellulose for radioactivity measurement by scintillation counting. The peptide served as a substrate of CKI, but the peptide phosphorylation was abolished by alanine substitution for Ser-511 (Fig. 8A). This result confirmed that Ser-511 of the peptide is the site of phosphorylation. Next, we questioned whether the peptide as a substrate competes with the full-length Pah1 for phosphorylation by CKI. The mixture of Pah1 and its peptide was phosphorylated by CKI with [␥-32 P]ATP, and the phosphorylated Pah1 was resolved by SDS-PAGE and analyzed by phosphorimaging. The CKI phosphorylation of Pah1 was inhibited by its peptide (IC 50 ϭ 80 M) in a dose-dependent manner (Fig. 8B). In contrast to the reduction of Pah1 phosphorylation, the peptide phosphorylation was reciprocally increased. These results indicate that the peptide is a competitive inhibitor of Pah1 phosphorylation by CKI (Fig.  8B). The V max and K m values for the Pah1 peptide were calculated to be 49 pmol/min/ml and 104 M, respectively.

Discussion
Regulation of PAP activity is important for the control of lipid synthesis in eukaryotic organisms, including yeast (5,37). The enzyme, which may govern whether cells utilize PA for the synthesis of the neutral lipid TAG or membrane phospholipids, is regulated on a biochemical level by the posttranslational modification of phosphorylation/dephosphorylation (3-6, 8, 74). The regulation of Pah1 by phosphorylation is complex; 40 phosphorylation sites have been identified, but only half of the sites can be ascribed to a specific protein kinase (Fig. 1B). Some Pah1 phosphorylation sites are common to multiple protein kinases (Fig. 1B), indicating that different signaling networks overlap to control the PAP function. This study focused on Pah1 phosphorylation by CKI, a constitutively active serine/ threonine protein kinase that plays a key role in nutrient-mediated cell morphogenesis, cytokinesis, secretion, and endocytosis (50 -54). The work here indicates that CKI also plays a role in lipid synthesis through the phosphorylation of Pah1. The enzymological studies demonstrated that Pah1 is a bona fide substrate of CKI; the phosphorylation reaction depended on time and the amounts of CKI, ATP, and Pah1. The K m value for Pah1 was essentially the same as that of CKII, Cdc28 -cyclin B, and Pho85-Pho80 but was lower (2-3.5-fold) that that of PKA and PKC (Table 1). The K m value for ATP of the CKI reaction was lower (1.5-2.4-fold) than that of the other protein kinases known to phosphorylate Pah1 ( Table 1). The stoichiometry of the CKI-mediated phosphorylation was 2.8 mol of phosphate/ mol of Pah1, a value lower than that for the phosphorylation by Pho85-Pho80, but greater than that for the phosphorylation by other protein kinases ( Table 1).
The CKI phosphorylation of Pah1 had two major effects. First, Pah1 phosphorylated by the protein kinase was catalytically more active than the unphosphorylated form. The enzyme stimulation is shown to occur by a mechanism that increases the V max but reduces the positive cooperativity of the enzyme for PA. CKI is the first protein kinase that exhibits a stimulatory effect on the catalytic function of Pah1; other protein kinases (e.g. Pho85-Pho80 and PKA) inhibit its PAP activity (40,41). Second, the CKI phosphorylation of Pah1 stimulated its phosphorylation by CKII but inhibited its phosphorylation by Pho85-Pho80, PKA, and PKC. The interrelationships of the phosphorylations are summarized in Fig. 9. The phosphorylation analysis of mutant Pah1 enzymes containing alanine substitutions for putative target serine residues led to the identification of eight CKI target sites with the major sites (based on Ն50% reduction in phosphorylation) being Ser-475, Ser-511, Ser-602, Ser-748, and Ser-774 (Fig. 4). The threonine residue(s) that is phosphorylated by CKI was not identified.
Six of the CKI sites are phosphorylated by other protein kinases. Being a constitutively-active protein kinase, CKI has the potential to phosphorylate a site(s) when another protein kinase is not active. For example, Pho85-Pho80 and PKA are most active during logarithmic growth when phospholipid synthesis is favored over TAG synthesis (75). Thus, at any point during the life cycle, lipid synthesis might be regulated via the CKI-mediated phosphorylation of Pah1 at a site(s) that is phosphorylated by another protein kinase that may not be active. This notion, however, has yet to be established in vivo.
Two serine residues (Ser-475 and Ser-511) of Pah1, which are located within the HAD-like domain (Fig. 1B), were identified as CKI-specific phosphorylation sites. The phosphorylation of these sites was shown to play roles in the subsequent phosphorylations by other protein kinases. On the one hand, the CKI phosphorylation of Ser-475 and Ser-511 was required for the stimulation of Pah1 phosphorylation by CKII. Previous work showed that the CKII phosphorylation of Pah1 stimulates the synthesis of TAG and lipid droplet formation, but only when it is not phosphorylated at the seven sites by Pho85-Pho80 (43). Thus, the CKI-mediated phosphorylation of Ser-475 and Ser-511 would be expected to have a positive impact on Pah1 function as mediated by CKII. On the other hand, the phosphorylation of Ser-475 and Ser-511 was required for the CKI-mediated inhibition of the phosphorylation by PKA and PKC. PKA works in conjunction with Pho85-Pho80 to attenuate Pah1 interaction with the ER membrane and inhibit PAP activity but at the same time to stabilize the enzyme to proteasomal degradation (41,48). PKC facilitates the proteasomal degradation of Pah1, but only when the seven Pho85-Pho80 sites are not phosphorylated (42,48). Accordingly, the phosphorylation of Ser-475 and Ser-511 might be expected to sustain the PKA-and PKCmediated regulations of Pah1. The phosphorylation of Pah1 by CKI inhibited the subsequent phosphorylation by Pho85-Pho80, but the phosphorylation of Ser-475 and Ser-511 was not involved.
The phosphorylation of Ser-475 and Ser-511 was not involved in the stimulatory effect of CKI on PAP activity. In fact, the S475A and S511A mutations enhanced the stimulation of PAP by the CKI-mediated phosphorylation, suggesting that the  CKI regulates phosphorylation of Pah1 PA phosphatase phosphorylation of these sites inhibits the phosphorylation of another site(s) that must be responsible for the stimulation of activity. The phosphorylation of Pah1 by Pho85-Pho80 inhibits PAP activity, and the 7A mutation augmented the stimulatory effect CKI has on PAP activity. Clearly, the stimulation of the PAP activity by CKI is complex. The PAP activity stimulation observed with WT Pah1 is masked by inhibitory effects caused by the phosphorylation of some sites. Additional studies are needed to identify the phosphorylatable residue(s) responsible for the CKI-mediated stimulation of PAP activity. The protein kinases (e.g. CKII, Pho85-Pho80, Cdc28 -cyclin B, PKA, and PKC) previously known to phosphorylate Pah1 are associated with the soluble fraction (e.g. cytosol) of the cell (76). Accordingly, we presume that the phosphorylation of Pah1 by most protein kinases occurs in cytosol as depicted in Fig. 1A. However, the Yck1/2 isoforms of CKI associate with the plasma membrane through the posttranslational modification of palmitoylation (68), and their association with this membrane is required for their known physiological functions (77). Whether a soluble unmodified form of CKI phosphorylates Pah1 in the cytosol is unknown. Contact sites between the ER membrane and plasma membrane exist (78), and so CKI associated with the plasma membrane could phosphorylate Pah1 associated with the ER membrane. Additional studies are warranted to address this important question.
The lack of PAP activity in yeast, as well as in human, leads to aberrant lipid metabolism and cell physiology (5,37). Too much PAP activity is also detrimental (e.g. obesity in the mouse model) (35). Thus, understanding the phosphorylation of PAP enzymes may lead to the identification of effector molecules to fine-tune activity and/or the enzyme cellular location. Peptidebased protein kinase inhibitors are attractive because they may block phosphorylation or disrupt protein-protein interactions (87,88). Most peptide inhibitors of protein kinases are based on the consensus phosphorylation motif (87,88). However, those inhibitors have a drawback of not having specificity for a particular substrate target of phosphorylation. A peptide sequence unique to a specific target like PAP would provide specificity without off-target side effects. The Pah1 peptide LYFEDSD-NEVDT, which contains the major phosphorylation site Ser-511, inhibited the phosphorylation of Pah1 by CKI. This sequence is specific to Pah1; it does not align with any other proteins in the yeast database. The mechanism of inhibition was competitive with an IC 50 value (80 M) similar to the K m value (104 M) for the peptide as a CKI substrate. Understandably, an effective inhibitor should work in the nanomolar range, and thus, the peptide tested in this work would not be useful. Yet, the data provide an initial proof-of-concept. Moreover, strategies (peptide cyclization and modification of backbone structure) are available that could enhance the inhibitory activity and effectiveness in vivo of the peptide (89). Studies with the Ser-511-containing peptide, as well as peptides that target other protein kinase target sites in Pah1, will be the subject of future work.
In summary, this work advances the understanding of the complex regulation of Pah1 PAP by phosphorylation. The CKImediated phosphorylation of Pah1 stimulates the PAP activity and regulates subsequent phosphorylations by CKII, Pho85-Pho80, Cdc28 -cyclin B, PKA, and PKC. This study not only advances our knowledge of the multisite phosphorylations of Pah1, it also sheds new light on the CKI-mediated regulation of cellular processes.

Reagents
All chemicals used were reagent grade or better. Avanti Polar Lipids was the source of all lipids. Coomassie Blue R-250, DNA size ladders, molecular mass protein standards, and reagents for electrophoresis and immunoblotting were from Bio-Rad. Cayman Chemical was the source of leupeptin and pepstatin. Difco Laboratories was the supplier of all growth media. The InstantBlue protein stain was purchased from Expedeon. The Pah1 peptide was synthesized by EZBiolab. GE Healthcare was the source of IgG-Sepharose, Q-Sepharose, PVDF membrane, and the enhanced chemifluorescence Western blotting detection kit. MilliporeSigma was the source of ATP, L-1-tosylamido-2-phenylethyl chloromethyl ketone (TPCK)-trypsin, protease, and phosphatase inhibitors, phosphoamino acid standards, Triton X-100, Ponceau S stain, BSA, rabbit anti-protein A antibody (product no. P3775, lot no. 053M4806V), and cellulose and Silica Gel 60 TLC plates. Scintillation counting supplies and radiochemicals, respectively, were purchased from National Diagnostics and PerkinElmer Life Sciences. New England Biolabs was the source of enzyme reagents for DNA manipulations and human CKII. Promega was the source of bovine heart PKA catalytic subunit and rat brain PKC. Plasmid DNA purification kit and nickel-nitrilotriacetic acid-agarose resin were from Qiagen. The S. cerevisiae strain that expresses TAP-tagged Yck1 and the alkaline phosphatase-conjugated goat anti-rabbit IgG antibody (product no. 31340, lot no. NJ178812) were purchased from Thermo Fisher Scientific. Whatman supplied the P81 phosphocellulose paper.

Strains, plasmids, and growth conditions
The strains and plasmids used in this study are listed in Tables 2 and 3, respectively. E. coli strain DH5␣ was used for the propagation of plasmids. E. coli strains BL21(DE3)pLysS and BL21(DE3) were used for the induced expressions of Pah1 and Pho85-Pho80, respectively. Plasmid pGH313 directs the expression of His 6 -tagged Pah1 (1), and plasmids EB1164 and CKI regulates phosphorylation of Pah1 PA phosphatase EB1076 direct the expressions of His 6 -tagged Pho85 and Pho80, respectively (90). The derivatives of pGH313 that contain threonine-to-alanine mutations were constructed by the QuikChange site-directed mutagenesis with appropriate templates and primers as described by Choi et al. (39). All mutations were confirmed by DNA sequencing. The isolation of plasmid DNA and its transformation into E. coli were performed by standard methods (91,92). The bacterial cells were grown at 37°C in lysogeny broth medium (1% tryptone, 0.5% yeast extract, 1% NaCl, pH 7.0). The growth medium was supplemented with ampicillin (100 g/ml) and chloramphenicol (34 g/ml) to select for cells carrying plasmids for the expression of proteins (93). The expression of proteins in E. coli cells was induced at 30°C with 1 mM isopropyl ␤-D-1-thiogalactopyranoside. S. cerevisiae cells expressing TAP-tagged Yck1 were grown at 30°C in YEPD medium (1% yeast extract, 2% peptone, and 2% glucose) (94). Cell numbers in liquid cultures were estimated by measuring absorbance at 600 nm.

Enzyme preparations
All steps were performed at 4°C. E. coli-expressed His 6tagged WT and mutant forms of yeast Pah1 (1) and His 6 -tagged Pho85-Pho80 protein kinase complex (90) were purified from bacterial cell extracts by affinity chromatography with nickelnitrilotriacetic acid-agarose. The protein content of the enzyme preparations was estimated by the method of Bradford (95) using BSA as a standard. SDS-PAGE (96) analysis indicated that the Pah1 and Pho85-Pho80 preparations were highly purified. The TAP-tagged version of Yck1 was partially purified from the Triton X-100 -solubilized extract (97) by affinity chromatography with IgG-Sepharose (98,99). The cleavage with tobacco etch virus protease (98,99) was omitted in the IgG-Sepharose chromatography step to retain the protein A tag on Yck1. Accordingly, the Yck1-TAP was eluted from the column with 50 mM glycine (pH 3.0) containing 0.1% Triton X-100 followed by the neutralization of the fractions by mixing with  PAH1 (S110A/S114A/S168A/S602A/T723A/S744A/S748A) derivative of pGH313 40 EB1164 PHO85-His 6 derivative of pQE-60 90 EB1076 PHO80 derivative of pSBETA 90 CKI regulates phosphorylation of Pah1 PA phosphatase 0.2 volume of 1 M Tris-HCl (pH 8.0). The isolation of Yck1 was confirmed by immunoblotting with the anti-protein A antibody ( Fig. 2A).

Protein kinase assays
Protein kinase activity was measured at 30°C by following the incorporation of radiolabeled phosphate from [␥-32 P]ATP into Pah1 in a total volume of 20 l (39). The phosphorylation reactions were terminated by the addition of 6.7 l of 4ϫ Laemmli sample buffer (96). Phosphorylated Pah1 was resolved from labeled ATP by SDS-PAGE (96) and visualized by phosphorimaging. The extent of phosphorylation was quantified by ImageQuant software. For the phosphorylation of Pah1 peptide, the reaction was terminated by spotting the reaction mixture onto a P81 phosphocellulose paper, followed by phosphoric acid washing and scintillation counting. The reaction mixture for the protein kinase reactions contained 50 mM Tris-HCl (pH 7.5), 10 mM MgCl 2 , 100 M [␥-32 P]ATP (ϳ3,000 cpm/ pmol), 40 g/ml Pah1, 2 mM DTT, and the indicated amounts of the protein kinase. The reaction mixture for PKC also contained 1.7 mM CaCl 2 , 500 M phosphatidylserine, and 156 M DAG. The units of CKI, CKII, Pho85-Pho80, PKA, and PKC were defined as femtomoles/min, nanomoles/min, picomoles/ min, picomoles/min, and nanomoles/min, respectively.

Phosphoamino acid analysis and phosphopeptide mapping
The 32 P-labeled Pah1 transferred to PVDF membrane was hydrolyzed with 6 N HCl at 100°C for phosphoamino acid analysis and digested with TPCK-trypsin for phosphopeptide mapping (100 -102). The acid hydrolysates were mixed with standard phosphoamino acids and separated by two-dimensional electrophoresis on the cellulose plates. The tryptic digests were separated on the cellulose plates first by electrophoresis and then by TLC (100 -102). Radioactive phosphoamino acids and phosphopeptides were visualized by phosphorimaging analysis. Nonradioactive phosphoamino acid standards were visualized by ninhydrin staining.

Immunoblotting
Proteins were separated by SDS-PAGE (96) using 8% slab gels. Immunoblotting with PVDF membrane was performed as described previously (103)(104)(105). Protein transfer from SDSpolyacrylamide gels to PVDF membranes was monitored by Ponceau S staining. Rabbit anti-protein A or rabbit anti-Pah1 (39) was used at a final concentration of 2 g/ml. The secondary goat anti-rabbit IgG antibodies conjugated with alkaline phosphatase were used at a dilution of 1:5,000. Immune complexes were detected using the enhanced chemifluorescence immunoblotting substrate. Fluorimaging was used to acquire fluorescence signals from immunoblots, and the intensities of the images were analyzed by ImageQuant software. A standard curve was used to ensure that the immunoblot signals were in the linear range of detection.

PAP assay
PAP activity was measured by following the release of watersoluble P i from chloroform-soluble PA using the Triton X-100/ PA-mixed micellar assay as described by Han and Carman (23).
The reaction mixture contained 50 mM Tris-HCl (pH 7.5), 1 mM MgCl 2 , 0.2 mM PA, 2 mM Triton X-100, and Pah1 protein in a total volume of 100 l. Water-soluble P i was measured with the malachite green-molybdate reagent at A 650 nm (23,106). Enzyme assays were conducted in triplicate, and the average standard deviation of the assays was Ϯ5%. The enzyme reactions were linear with time and protein concentration. Protein concentration was estimated by the Coomassie Blue-based assay of Bradford (95) using BSA as standard. A unit of PAP activity was defined as micromoles/min.

Data analysis
Microsoft Excel software was used for the statistical analysis of data. The p values Ͻ0.05 were taken as a significant difference. The enzyme kinetics module of SigmaPlot software was used to analyze kinetic data according to Michaelis-Menten and Hill equations.