or-A review of phosphatidate phosphatase assays

Phosphatidate phosphatase (PAP) catalyzes the penultimate step in the synthesis of triacylglycerol and regulates the synthesis of membrane phospholipids. There is much interest in this enzyme because it controls the cellular levels of its substrate, phosphatidate (PA), and product, DAG; defects in the metabolism of these lipid intermediates are the basis for lipid-based diseases such as obesity, lipodystrophy, and inflammation. The measurement of PAP activity is required for studies aimed at understanding its mechanisms of action, how it is regulated, and for screening its activators and/or inhibitors. Enzyme activity is determined through the use of radioactive and nonradioactive assays that measure the product, DAG, or P i . However, sensitivity and ease of use are variable across these methods. This review summarizes approaches to synthesize radioactive PA, to analyze radioactive and nonradioactive products, DAG and P i , and discusses the advantages and disadvantages of each PAP assay.


INCORPORATION OF PA INTO DETERGENT MICELLES AND LIPOSOMES
Natural PA, which contains long-chain fatty acyl groups on positions 1 and 2 of the glycerol backbone, is a waterinsoluble substrate that should be presented to PAP in an aqueous environment that is buffered to the pH optimum (e.g., pH 7.5) for the enzyme reaction (2,3,12). The two most popular methods for delivery of PA to the assay system, which are discussed below, include its incorporation into detergent micelles and unilamellar phospholipid vesicles (liposomes). Water-soluble PA analogs containing short-chain (e.g., DiC8:0) fatty acyl groups are directly added to the aqueous assay system (63). Sodium, potassium, or ammonium salts of PA should be used in PAP assays; calcium should be avoided as it interferes with the reaction.

Incorporation of PA into detergent micelles
In the original work (1), PA derived from egg lecithin is added to the assay system as an undefined emulsified form. Sonicated mixtures of PA and PC are also utilized to measure PAP activity (11,64,65), but the size and uniformity of the substrate-containing aggregates/vesicles are unknown. PAP activity from undefined substrates in the assay system is not conducive to meaningful data for understanding the mode of enzyme action and its kinetic properties. PAP, like other interfacial enzymes, acts on the substrate through a three-dimensional interaction in solution as well as through a two-dimensional surface interaction, and accordingly both of the enzyme-substrate interactions must be taken into account in the assay (66). Triton X-100 forms a mixed micelle with PA, providing a surface for catalysis (67). The use of detergent/phospholipid-mixed micelles makes it possible to analyze PAP activity that is dependent on the molar and surface concentrations of PA (3,67,68) according to the "surface dilution kinetics" model (66). The detergent/phospholipid-mixed micelle system is also useful in assessing the regulation of activity by phospholipids (69), sphingolipids (70), and nucleotides (71), as well as by the enzyme's posttranslational modification by phosphorylation (72)(73)(74)(75).
Triton X-100 micelles containing PA are easily prepared by dissolving the dried phospholipid in the detergent solution (67), and the size and uniformity of Triton X-100/ PA-mixed micelles are determined by gel filtration chromatography or by glycerol density gradient centrifugation (67,69,70). Additional lipids may be incorporated into the Triton X-100/PA-mixed micelles to examine their effects on PAP activity as activators or inhibitors (69,70).

Incorporation of PA into liposomes
Although the Triton X-100/PA-mixed micelle is useful in measuring PAP activity to assess biochemical mechanisms of the enzyme regulation (3,67,69,(71)(72)(73)(74)(75)(76), it does not simulate the membrane phospholipid bilayer. Liposomes are a widely accepted mimic of the biological membrane (77)(78)(79), and they have been used to measure PAP activity (11,65,(80)(81)(82)(83). In the assay, the phospholipid vesicles resembling the complex composition of the nuclear/endoplasmic reticulum membrane (e.g., PC/PE/PI/PS/PA) are shown to be better substrates than those composed of only PC (and/or PE) and PA (83). The size of liposomes also affects PAP activity; greater activity is elicited with larger liposomes (e.g., 100 nm) when compared with smaller liposomes (e.g., 50 nm) (83). Of various methods to prepare the unilamellar vesicles (84), the most popular one is to extrude a phospholipid suspension through a membrane filter with a defined pore size (79). The size and distribution of the liposomes are assessed by light scattering (e.g., particle size analyzer) (85) or by gel filtration chromatography (86).  The  structures of CDP-DAG, PA, DAG, and TAG are shown with the C16:0 and C18:1 fatty acyl groups. PAP plays a major role in governing whether cells utilize PA for the synthesis of TAG via DAG or for the synthesis of membrane phospholipids via CDP-DAG. The DAG produced in the PAP reaction is also used for the synthesis of phospholipids. More detailed pathways for the synthesis of phospholipids in yeast (38) and mammals (19)  PAP activity on the PA-containing liposomes, like that on Triton X-100/PA-mixed micelles (3,67,68), is dependent on the molar and surface concentrations of PA when the substrate is incorporated into liposomes composed of complex phospholipid mixtures (83). The dependence of PAP activity on the molar concentration of PA indicates that the enzyme operates in the hopping mode (87). Kinetic analysis with these liposomes also demonstrates that PAP activity is dependent on the surface concentration of PA, and thus the enzyme also operates in the scooting mode (87).

PAP ASSAYS
The PAP assay typically measures the formation of reaction products (DAG and/or P i ) over time (65,88,89). To be in a linear range of activity, the enzyme assay is conducted such that less than 15% of the substrate PA is converted to products. The optimum reaction conditions of the PAP assay (e.g., pH, temperature, Mg 2+ , and saturating concentrations of PA) need to be defined so that the catalytic activity is dependent on incubation time and the enzyme amount (i.e., zero order kinetics) (2,3,12). By convention (90), PAP activity is expressed as the unit of the enzyme amount that catalyzes the formation of 1 mol of product per minute, and its specific activity is defined as the enzyme activity (i.e., units) per milligram of protein (65,88). For low levels of PAP activity from cell extracts or crude enzyme preparations, it is often reported in nanomoles (or picomoles) per minure. The turnover number (molecular activity), which requires the molecular mass information of PAP, is defined as the mole product formed per mole of enzyme per second. PAP activity has been measured with different sizes and forms (e.g., native, radioactive, and fluorescent) of PA.
In the PAP assay, to measure the release of 32 P i from [ 32 P]PA, the enzyme reaction is terminated (e.g., with acid) and then subjected to a chloroform/methanol-water phase partition (95,96), and a portion of the aqueous phase is measured for radioactivity by scintillation counting (88). In the PAP assay, to measure the production of the [ 14 C]DAG from [ 14 C]PA, the radioactive substrate and product are extracted in chloroform (e.g., chloroform/methanol-water phase partition) and then separated by TLC on silica gel. The silica gel spot containing the [ 14 C]DAG is scraped off from the TLC plate and measured for radioactivity by scintillation counting (65,91,92). Alternatively, [ 14 C]DAG on the TLC plate is quantified by phosphorimaging analysis with 14 C standards. Without the chromatographic separation, [ 14 C]PA in the chloroform extract is precipitated by aluminum oxide, and the supernatant containing the [ 14 C] DAG is measured for radioactivity by scintillation counting (65,92).

Advantages of radioactive assays
The major advantages of the radioactive assays are specificity and sensitivity. Because of the radioactive nature of the substrate, the substrate-product relationship of the PAP reaction is defined even in a crude system (1). Moreover, studies to examine a reaction mechanism are facilitated by the use of radioactive substrates and products (97). The radioactive assay is particularly useful in measuring a low level of PAP activity. The sensitivity of the assay is limited only by the specific radioactivity of the precursor molecules used for the synthesis of radioactive PA. The measurement of the product 32 P i from [ 32 P]PA is simpler and more sensitive when compared with the measurement of the product [ 14 C]DAG from [ 14 C]PA. In the former assay, the watersoluble 32 P i is easily separated from the chloroform-soluble [ 32 P]PA by a chloroform/methanol-water phase partition. In contrast, the latter assay requires the extraction of both [ 14 C]DAG and [ 14 C]PA in chloroform, and then the separation of the radioactive substrate and product by TLC or by the precipitation of PA with aluminum oxide (65,92). Compared with [ 14 C]PA, [ 32 P]PA can be synthesized to be more radioactive because of the high specific radioactivity of [-32 P]ATP.

Disadvantages of radioactive assays
The major disadvantage of the radioactive PAP assay is that its substrate is not always readily available and must be synthesized and purified. The use of radioactive chemicals requires institute authorization and specialized equipment (e.g., scintillation counter and/or phosphorimager) that is not universally available. Of course, radioactive contamination (e.g., personal and laboratory space/equipment) is also a deterrent to performing radioactive assays. Additionally, the specific radioactivity of the 14 C-labeled precursors used for radiolabeling PA in the phosphatidyl moiety is lower when compared with that of the [-32 P]ATP used for radiolabeling the phosphate moiety. Accordingly, the sensitivity of the assay that measures [ 14 C]DAG is lower when compared with that of the assay measuring 32 (65). These competing activities must be taken into account. For example, when glycerol-3-P is produced from PA, its conversion to 32 P i may be minimized by the addition of racemic glycerol-3-P to the assay, and when DAG is hydrolyzed by lipases, their reactions may be prevented with DAG lipase inhibitors (65).

NONRADIOACTIVE ASSAYS
In the original work, Kennedy and coworkers (1) measured the production of P i from PA by a colorimetric assay with the molybdate reagent first described in 1925 by Fiske and Subbarow (98). The sensitivity and stability of the molybdate reagent for measuring PAP activity has been improved by its combination with malachite green (12,63,(99)(100)(101)(102). The P i produced in the PAP assay is measured after the removal of the enzyme by precipitation with trichloroacetic acid, which is also used to terminate the reaction (1). Alternatively, P i is separated from PA by a chloroform/methanolwater phase partition (12,95,96). When water-soluble C8:0 PA is used as a substrate, the P i released in the PAP reaction is directly complexed with the malachite green-molybdate reagent, which is also used to stop the enzyme reaction (63).
In the original work (1), the DAG produced in the PAP reaction is extracted with ether, purified by silica gel chromatography, and quantified by ester analysis (1); this method of DAG quantification is tedious and impractical for a routine PAP assay. As discussed above in the Radioactive Assays section, DAG is extracted from the reaction mixture by a simple chloroform/methanolwater phase partition (95,96) and then separated from the coextracted PA by TLC or by the precipitation of PA with aluminum oxide (92). The lipids on the silica gel plate are detected by charring or by staining with iodine vapor (103,104) or the fluorescent lipophilic dye, primulin (105). The chloroform-soluble DAG is also analyzed by HPLC with an evaporative light scattering detector (106) or by mass spectrometry (107).
The sensitivity of the nonradioactive assay in measuring PAP activity is enhanced by utilization of fluorescently labeled substrates such as NBD PA (108) and BODIPY PA (109). The fluorescent product (e.g., NBD DAG or BODIPY DAG) and substrate (NBD PA or BODIPY PA) are separated by TLC and then detected by fluorescence scanning (46) or analyzed by HPLC with a fluorescence detector (108,109).

Advantages of nonradioactive assays
The obvious advantages of nonradioactive PAP assays are in the availability of PA and the reagents to analyze the reaction products. If short chain PA (e.g., DiC8:0) is used as substrate, there is no need to perform a chloroform/ methanol-water phase partition after the reaction is completed (63). The sensitivities of the P i measurement by the colorimetric assay and the DAG measurement by TLC or HPLC are in the nanomolar range. These assays are best suited for PAP samples with high activity (e.g., purified or overexpressed enzyme) that are devoid of, or less affected by, competing lipolytic and phosphatase activities. Fluorescently labeled PA is readily available, and the quantification of fluorescent DAG by TLC or HPLC is sensitive in the nanomolar to picomolar range. The fluorescence PAP assay is useful for measuring activity in crude extracts (46,108,109) or with purified enzyme (14) owing to the fluorescent nature of the substrates and products of the reaction.

Disadvantages of nonradioactive assays
The measurement of P i by colorimetric assay with the malachite green-molybdate reagent may produce high background due to nonspecific phosphatase activities in crude enzyme samples and/or residual phosphate from cleaning detergents left in test tubes. The colored complex is not stable for long periods of time, and although used in some systems (110), the assay is not conducive for measurement of crude enzyme preparations. Measurement of DAG from nonradioactive PA requires separation by TLC or HPLC, and is beset by nonspecific phospholipase, lysophospholipase, and neutral lipase activities in crude enzyme preparations. Short-chain PA, NBD PA, and BODIPY PA are not physiologically relevant substrates, and quantification of the fluorescent PAP enzyme products requires specialized equipment for detection. The advantages and disadvantages of radioactive and nonradioactive PAP assays are summarized in Table 1. -dependent enzyme whose activity is based on the DXDX(T/V) catalytic motif in the haloacid dehalogenase-like domain (10,13,14). However, there is a nonspecific PAP [previously known as PAP2 and now known as lipid phosphate phosphatase (LPP) (111)(112)(113) that does not require Mg 2+ or any other divalent cation (114)(115)(116), and its activity is directed by a three-domain catalytic motif consisting of the sequences KX 6 RP, PSGH, and SRX 5 HX 3 D (117). Most of the PAP2/ LPP activity in yeast is encoded by the DPP1 (118) and LPP1 (119) genes, whereas the activity is encoded by LPP/ Lpp genes in higher eukaryotes (113,120,121). Unlike the yeast and mammalian PAP enzymes, which are specific for PA (3,12), the LPP enzymes in yeast and mammalian cells have a broad substrate specificity. In addition to PA, other substrates of LPP enzymes include lysoPA, DAG pyrophosphate, sphingoid base phosphates, and isoprenoid phosphates (113,(118)(119)(120)(121)(122)(123)(124)(125)(126)(127), which are known to play signaling roles in cell physiology. Thus, PAP activity involved with de novo lipid synthesis needs to be differentiated from PAP2/ LPP activity involved in lipid signaling.

METHODS TO DIFFERENTIATE Mg
Mammalian PAP and PAP2/LPP enzymes are differentiated by their sensitivity to the thioreactive compound, N-ethylmaleimide (NEM) (65,111,112). However, in yeast, NEM sensitivity is not applicable to differentiate the Mg 2+ -dependent and -independent PAP activities (3,122,123,128,129). Thus, NEM sensitivity cannot be used as a general approach to differentiate PAP and PAP2/ LPP enzymes. To differentiate the two activities based on cofactor requirement, the total PAP activity is measured in the presence of 1 mM of MgCl 2 , and the Mg 2+ -independent activity is separately measured in the absence of MgCl 2 but in the presence of 1 mM of EDTA, which is added to chelate the residual divalent cation in the cell extract (53,58). The Mg 2+ -dependent activity is then determined by subtracting the Mg 2+ -independent activity from the total PAP activity (53,58). In cases where the Mg 2+ -dependent PAP is encoded by more than one gene, then the enzyme activity encoded by the gene of interest can be accurately measured in the absence of the unwanted activity [e.g., knockout or knockdown of the unwanted PAP-encoding gene(s)] (128).

PAP ASSAY RECOMMENDATIONS
For analysis of a pure PAP preparation with high specific activity, the most convenient is a nonradioactive assay that measures P i with the malachite green-molybdate reagent. Our laboratory has used the colorimetric assay to characterize the PAP activity of purified human lipin 1 with respect to PA substrate specificity in detergent/PAmixed micelles and to screen the enzyme inhibitors and activators (12). We have also used this assay to examine the hopping and scooting modes of purified yeast Pah1 PAP with liposomes (83). If the focus is to study the PAP reaction in a system that mimics the in vivo condition of a membrane phospholipid bilayer, then the incorporation of PA into liposomes is recommended. Otherwise, the detergent/PA-mixed micelle system is recommended. In either case, we recommend that saturating molar and surface concentrations of PA be used for routine assays (3,12,67) to ensure zero order kinetics. The colorimetric assay with purified PAP should be applicable to high throughput screening of activators/inhibitors in 96-well plates. Utilization of DiC8 PA in the assay obviates the need to separate the product P i from the substrate PA following the reaction (63). That the short-chain PA is not physiologically relevant may not be an issue in a screen when a purified enzyme is utilized. To our knowledge, a high throughput PAP assay has not yet been developed but would be of great use in the search of enzyme effector molecules. For analysis of crude PAP preparations with low specific activity, a radioactive assay measuring the release of 32 P i from [ 32 P]PA is recommended. This assay is very sensitive and limited only by the specific radioactivity of [-32 P]ATP used for the synthesis of [ 32 P]PA. If the radioactive assay is not available for crude PAP preparations, we recommend a nonradioactive assay with the fluorescent substrate NBD PA or BODIPY PA. As discussed above, however, the fluorescence assay requires the separation of the product from substrate by TLC or HPLC, which is a cumbersome component of the PAP assay that is shared by other nonradioactive and radioactive assays to measure the product DAG. For differentiating PAP and PAP2/LPP activities, we recommend the genetic approach to eliminate either form of the activity, and when not possible, utilize assays conducted with and without Mg 2+ ions.