Genetic analysis of digestive physiology using fluorescent phospholipid reporters.

Zebrafish are a valuable model for mammalian lipid metabolism; larvae process lipids similarly through the intestine and hepatobiliary system and respond to drugs that block cholesterol synthesis in humans. After ingestion of fluorescently quenched phospholipids, endogenous lipase activity and rapid transport of cleavage products results in intense gall bladder fluorescence. Genetic screening identifies zebrafish mutants, such as fat free, that show normal digestive organ morphology but severely reduced phospholipid and cholesterol processing. Thus, fluorescent lipids provide a sensitive readout of lipid metabolism and are a powerful tool for identifying genes that mediate vertebrate digestive physiology.

Efficient synthesis of the cholinephosphate phospholipid headgroup.

In search of an efficient method to prepare cholinephosphate headgroups in phospholipids under mild conditions (where the diacylglycerol moiety is not subject to oxidation), a method was developed for phosphorylation using a trialkyl phosphite and I2. The active intermediate is a phosphoryl iodide formed by oxidation of the phosphite with I2. 2-Bromoethanol, dimethyl chlorophosphite, and an alcohol (diglyceride) are converted to a phosphate triester in a one-pot reaction with high yield. In the second reaction, the phosphate triester is demethylated, and the ethyl bromide group is converted to choline by treatment with aqueous trimethylamine. This procedure is applied to the synthesis of hexadecylphosphocholine, and 1,2-didecanoyl-1-deoxy-1-thio-sn-glyceryo-3-phosphocholine.

Intramolecularly quenched BODIPY-labeled phospholipid analogs in phospholipase A(2) and platelet-activating factor acetylhydrolase assays and in vivo fluorescence imaging.

Phospholipase substrate analogs containing both a fluorescent BODIPY group and a quenching 2,4-dinitrophenyl (DNP) group were synthesized. They showed little fluorescence, but upon hydrolysis became fluorescent as the quenching group was removed. Two substrates were phosphatidylethanolamine analogs with a BODIPY-pentanoyl group at the sn-2 position and DNP linked to the amino head group. The third was a phosphatidylcholine analog with a BODIPY-labeled alkyl ether at the sn-1 position and a N-(DNP)-8-amino-octanoyl group at the sn-2 position. These compounds were evaluated as substrates for cytosolic (85 kDa) phospholipase A(2) (cPLA(2)) and plasma platelet-activating factor acetylhydrolase (rPAF-AH). Two were good substrates for cPLA(2) (specific activities: 18 and 5 nmol min(-1) mg(-1)) and all were good for rPAF-AH (specific activities: 17, 11, and 6 micro mol min(-1) mg(-1)). The minimal amount of enzyme detectable was 50 ng for cPLA(2) and 0.1 ng for rPAF-AH. These substrates were active in assays of PLA(2) in zebrafish embryo extracts and one was well suited for imaging of PLA(2) activity in living zebrafish embryos. Embryos were injected with substrate at the one- to four-cell stage and allowed to develop until early somitogenesis when endogenous PLA(2) activity increases dramatically; substrate persisted (12 h) and specifically labeled cells of the developing notochord.

Binding of phosphatidylinositol-specific phospholipase C to phospholipid interfaces, determined by fluorescence resonance energy transfer.

Dissociation constants for binding of phosphatidylinositol-specific phospholipase C from Bacillus cereus (bcPI-PLC) and the mammalian phosphatidylinositol-specific phospholipase C-delta(1) to lipid interfaces containing phosphoinositol, phosphocholine, and phosphomethanol head groups were determined by fluorescence resonance energy transfer. Dansyl-labeled lipid probes were used as acceptors, with intrinsic tryptophan of the enzyme as the donor. Titration of protein into lipid provided data from which K(d) and N, the limiting number of lipid molecules per protein bound, were calculated by non-linear regression analysis of exact binding equations. These results were compared with apparent K(m) values from kinetic data. K(d) values in the low microM range in terms of lipid monomers or low nM range in terms of binding sites were calculated with good fits of experimental data to theoretical binding curves. bcPI-PLC binds with high affinity to PI interfaces, slightly lower affinity to PC interfaces, and much lower affinity to PM interfaces. The mammalian enzyme also binds with high affinity to PI interfaces, but shows little or no binding with PC interfaces under similar concentration conditions. These K(d) values correlate reasonably with apparent K(m) values from kinetic experiments.

Continuous spectrophotometric assay of mammalian phosphoinositide-specific phospholipase Cdelta1 with a thiophosphate substrate analog.

1,2-Dimyristoyloxypropane-3-thiophospho(1D-1-myo-inositol) (D-thio-DMPI) was used as a substrate for the continuous assay of phosphoinositide-specific phospholipase C (PI-PLC). Its activity with a Delta(1-132) deletion mutant of mammalian PI-PLCdelta1 is about one-fourth that with PI under similar conditions. Optimal conditions for the assay include 0.2 mM substrate, 0.2 mM Ca2+, and a mole ratio of hexadecylphosphocholine detergent to substrate of 2.0. A minimum of about 60 ng of pure enzyme can be detected. The apparent bulk Km for PI-PLC with D-thio-DMPI under these conditions is about 6 microM. Enzyme activity as a function of surface concentration of substrate shows no sign of saturation up to the maximum mole fraction.

A thiophosphate analog of dimyristoylphosphatidyl-inositol-4-phosphate is a substrate for mammalian phosphoinositide-specific phospholipase C.

1,2-Dimyristoyloxypropane-3-thiophosphate(rac-1-myo-inositol-4- phosphate), a thiophosphate analog of dimyristoyl phosphatidylinositol-4-phosphate was synthesized as a substrate for mammalian phosphoinositide-specific phospholipase C. Its activity with delta(1-132)-PI-PLC-delta 1 (a deletion mutant with the N-terminal pleckstrin homology domain removed) was studied in sonicated dispersions, with and without added Triton X-100. It had an initial activity of about 30 mumol min-1 mg-1, which rapidly decreased due to substrate depletion in the vesicle or micelle. The slower rate of hydrolysis appeared limited by enzyme hopping or exchange of substrate between vesicles or micelles, which was more rapid in the presence of detergent.

Activity of phosphatidylinositol-specific phospholipase C from Bacillus cereus with thiophosphate analogs of dimyristoylphosphatidylinositol.

Phosphatidylinositol-specific phospholipase C (PI-PLC) was studied with sonicated dispersions of a thiophosphate analog of phosphatidylinositol, 1, 2-dimyristoyloxypropane-3-thiophospho(1D-1-myo-inositol) (D-thio-DMPI). Kinetic parameters were derived from the rate as a function of bulk lipid concentration at constant saturating surface concentration of substrate (case I), and as a function of surface concentration of substrate at a constant saturating bulk concentration of lipid (case II). The substrate, D-thio-DMPI, was diluted with L-thio-DMPI or dimyristoyl phosphatidylmethanol (DMPM). In the presence of L-thio-DMPI, values for Vmax = 133 mumol min-1 mg-1, Ks' (the apparent dissociation constant for the enzyme-interface complex) = 0.097 mM, and Km* (the apparent interfacial Michaelis constant) = 0.22 mol fraction were obtained. DMPM caused enzyme inhibition in case I but no inhibition in case II. L-Thio-DMPI is an ideal neutral diluent with which to study the kinetics of PI-PLC.

Kinetics of phosphatidylinositol-specific phospholipase C with vesicles of a thiophosphate analogue of phosphatidylinositol.

1,2-Dimyristoyloxypropane-3-thiophospho(1D-1-myo-inositol) (D-thio-DMPI) was synthesized as a substrate for the continuous spectrophotometric assay of phosphatidylinositol-specific phospholipase C (PI-PLC) from Bacillus cereus. Release of thio-diglyceride is followed by a coupled reaction with 4,4'-dithiopyridine to produce a chromophore, 4-thiopyridine, measured by its absorption at 324 nm. Sonicated vesicles of D-thio-DMPI gave sigmoidal Michaelis-Menten kinetics with PI-PLC as a function of bulk concentration of substrate (Hill plot: Vmax = 132 mumol min-1 mg-1, apparent Km = 0.115 mM, h = 1.8). Addition of dimyristoyl phosphatidylcholine (DMPC) or dimyristoyl phosphatidylmethanol to vesicles of D-thio-DMPI resulted in an initial increase in rate followed by a decrease at higher concentrations of non-substrate lipid. Binding of PI-PLC to vesicles of DMPC with 10 mol% of N-dansyl phosphatidylethanolamine was demonstrated by fluorescence resonance energy transfer from tryptophan in the enzyme to the dansyl lipid probe.

Fluorescence-based assays of lipases, phospholipases, and other lipolytic enzymes.

In choosing an assay, one needs to consider the following questions: What level of sensitivity is required? Must the assay be continuous? Is the substrate readily available; can it be purchased or must it be custom synthesized? How specific is the substrate? How convenient is the method? How compatible is it with monomolecular, micellar, or vesicular substrates? How tolerant is it of added detergents and proteins that may be present? What is the cost of substrates, fluorescent probes, and instrumentation? Of the many methods described in this review, discontinuous assays using natural substrates and derivatization of the products with fluorescent probes are probably the most reliable and most tolerant of reaction conditions. A drawback is the involvement of tedious and time-consuming steps which limit the number of trials that can be performed. Continuous assays, in which changes in fluorescent properties of the probe are monitored, are most convenient for kinetic studies, although they are also most sensitive to reaction conditions and intolerant of added detergents and proteins. One has to carefully consider all of these issues and choose a method best suited to the enzyme, the particular information one wants to obtain, and the availability of substrates, probes, and instrumentation. Hopefully, increased commercial availability of fluorescent substrates and probes will make these choices easier. Nevertheless, the search goes on for better, more sensitive and convenient fluorescent assays.

Synthesis of optically-active hexadecyl thiophosphoryl-1-D-myo-inositol: a thiophosphate analog of phosphatidylinositol.

The synthesis of optically-active hexadecyl thiophosphoryl-1-D-myo-inositol 11 was accomplished from 2,3-O-(D-1',7',7'-trimethyl[2.2.1]bicyclohept-2'-ylidene)-4,5,6-O- tris(methoxymethyl)-D-myo-inositol 6 or 2,3,4,5,6-O-pentakis(methoxymethyl)-D-myo-inositol 14, using the Arbusov reaction of their dimethyl phosphite derivatives 7 and 15 with N-hexadecyl thiophthalimide 8. This product was a substrate for phosphatidylinositol-specific phospholipase C from Bacillus cereus.

The chemical step is not rate-limiting during the hydrolysis by phospholipase A2 of mixed micelles of phospholipid and detergent.

The effect of detergents on the overall catalytic turnover by secreted phospholipase A2 (PLA2) on codispersions of the substrate phospholipid is characterized. The overall rate of interfacial catalytic turnover depends on the effective substrate "concentration" (mole fraction) that the bound enzyme "sees" at the interface. Therefore, besides the intrinsic catalytic turnover rate determined by the Michaelis-Menten cycle in the interface [Berg et al. (1991) Biochemistry 30, 7283], two other interfacial processes significantly alter the overall effective rate of hydrolysis: first, the fraction of the total enzyme at the interface; second, the rate of replenishment of the substrate. At low mole fractions (< 0.3), bile salts promote the binding of pig pancreatic PLA2 to zwitterionic vesicles, and the rate of hydrolysis increases with the fraction of the enzyme in the interface. At higher (> 0.3) mole fractions of the detergent, the bilayer is disrupted, and the rate of hydrolysis decreases by more than a factor of 10. The detergent-dependent decrease in the rate of hydrolysis of the sn-2-oxyphospholipids is much larger than that of sn-2-thiophospholipid, and therefore the element effect (O/S ratio) decreases from about 10 in bilayers to less than 2 in mixed micelles. This loss of the element effect in mixed micelles shows that the chemical step is no longer rate-limiting during the hydrolysis of mixed micelles formed by the disruption of vesicles by the detergent. Such effects were observed with phospholipase A2 from several sources acting on substrates dispersed in a variety of detergents including bile salts, 2-deoxylysophosphatidylcholine, and Triton X-100.(ABSTRACT TRUNCATED AT 250 WORDS)

Kinetics of Bacillus cereus phosphatidylinositol-specific phospholipase C with thiophosphate and fluorescent analogs of phosphatidylinositol.

Thiophosphate analogs (C-S-P bond) of phosphatidylinositol (Cn-thio-PI: racemic hexadecyl-, dodecyl-, and octylthiophosphoryl-1-myo-inositol) and a fluorescent analog (pyrene-PI: rac-4-(1-pyreno)-butylphosphoryl-1-myo-inositol) were all substrates for phosphatidylinositol-specific phospholipase C from Bacillus cereus. Hydrolysis of thio-PI was followed by coupling the production of alkylthiol to a disulfide interchange reaction with dithiobispyridine. Hydrolysis of pyrene-PI was followed using a HPLC-based assay with fluorescence detection. The activity of PI-PLC with thio-PI analogs showed an interfacial effect. C16-Thio-PI, which had a critical micelle concentration (CMC) of 7 microM, gave a hyperbolic activity versus concentration curve between 0 and 2 mM, while C8-thio-PI, which had a CMC above 10 mM, showed very low activity which increased greatly upon introduction of an interface in mixed micelles with hexadecylphosphocholine (HDPC). Pyrene-PI, which aggregates above 0.3 mM, gave a sigmoidal activity curve with much higher activity above the CMC. All three thio-PI homologs as mixed micelles with HDPC gave hyperbolic activity curves with PI-PLC that were a function of bulk concentration of substrate at constant surface concentration and surface concentration of substrate at constant bulk concentration. The maximal activity of PI-PLC with pure C16-thio-PI micelles was 6.25 mumol min-1 mg-1, while that with pyrene-PI was estimated to be 68 mumol min-1 mg-1. With pure C16-thio-PI micelles, 0.022 mM substrate gave half Vmax, similar to that in mixed micelles with HDPC.

Interfacial catalysis by phospholipase A2: the rate-limiting step for enzymatic turnover.

The kinetics of the phospholipase A2-catalyzed hydrolysis of bilayer vesicles and mixed micelles of several oxyglycero and thioglycero analogues of phospholipids have been studied. The results with vesicles show that, depending on the source of the enzyme, the rates of hydrolysis of the oxy-containing long-chain phosphatidylmethanols are 2.5- to 28-fold higher compared to the rates of hydrolysis of the analogous thio substrates. The oxygen to sulfur substitution does not significantly alter the affinities of the enzymes for the reaction products or calcium. Since it is unlikely that sulfur substitution changes the rate constants for the formation and dissociation of the enzyme-product complex by the same factor, the element effects seen in the rates of hydrolysis of the oxy- and thioester phospholipids in vesicles are primarily due to a change in the rate constant for the chemical step of the catalytic turnover cycle. For bovine pancreatic phospholipase A2, various mutants with lower catalytic activity were used to show that the value of the element effect does not increase in the mutants. These results establish that, for the pancreatic phospholipase A2, the element effect is fully expressed, and the chemical step is fully rate-limiting for both oxyglycero and thioglycero phospholipids in vesicles. It was found that the element effect decreases from 7 to 1 when long-chain phosphatidylmethanols are present in micelles of a neutral diluent. This result suggests that the chemical step is not rate-limiting during the hydrolysis of these mixed micelle substrates.

Cholesterol esterase catalyzed hydrolysis of mixed micellar thiophosphatidylcholines: a possible charge-relay mechanism.

Mechanistic features of cholesterol esterase catalyzed hydrolysis of two thiophospholipids, rac-1-(hexanoylthio)-2-hexanoyl-3-glycerophosphorylcholine (6TPC) and rac-1-(decanoylthio)-2-decano-yl-3-glycerophosphorylcholine (10TPC), have been characterized. The hydrolysis of 10TPC that is contained in mixed micelles with Triton X-100 occurs strictly at the micellar interface, since the reaction rate is independent of the micelle concentration but depends hyperbolically on the mole fraction of the substrate in the micelles. This latter observation allows one to calculate the interfacial kinetic parameters V*max and K*m. The hydrolyses of 10TPC and p-nitrophenyl butyrate are similarly inhibited by the transition state analogue inhibitor phenyl-n-butylborinic acid, and therefore, physiological and nonphysiological substrates are processed at the same active site. The similarity of k*cat values for the acyl-similar substrates 10TPC and p-nitrophenyl decanoate indicates that the phospholipase A1 activity of cholesterol esterase is partially rate limited by turnover of a decanoyl-enzyme intermediate. Solvent isotope effects on V*max and V*max/K*m (which monitors acylation only) are approximately 2-3 and are consistent with transition states that are stabilized by general acid-base proton transfers. Proton inventories of V*max/K*m indicate that simultaneous proton transfers stabilize the acylation transition state, which requires a multifunctional acid-base machinery (perhaps a charge-relay system) in the cholesterol esterase active site. Similar results are obtained for the 6TPC reaction, both in the presence and absence of Triton X-100 micelles.

A facile asymmetric synthesis of glycerol phospholipids via tritylglycidol prepared by the asymmetric epoxidation of allyl alcohol. Thiolester and thioether analogs of phosphatidylcholine.

Trityl-glycidol was synthesized by in situ derivatization of glycidol, which was prepared by the catalytic asymmetric epoxidation of allyl alcohol. Depending on the enantiomer of diisopropyl tartrate used with the titanium catalyst, either (R)- or (S)-trityl-glycidol was obtained in a "one pot" synthesis in about 50% overall yield. The optical purity, determined by NMR spectroscopy of a Mosher ester, was greater than 98% ee. Nucleophilic opening of the chiral epoxide with dodecyl mercaptan gave optically active 1-S-dodecyl-3-O-trityl-1-thio-glycerol, which was used to synthesize 1-S-dodecyl-2-O-decanoyl-thio-sn-glycero-3-phosphocholine. Opening of the epoxide with methyl xanthate gave a 1,2-trithiocarbonate derivative of trityl glycerol which can be used to synthesize 1,2-bis(S-decanoyl)-1,2-dithio-sn-glycero-3-phosphocholine. Opening of the epoxide with thiodecanoic acid gave 1-S-decanoyl-3-O-trityl-1-thio-glycerol which was used to synthesize 1-S-decanoyl-2-O-decanoyl-1-thio-sn-glycero-3-phosphocholine.

Evaluation of fluorescent and colored phosphatidylcholine analogs as substrates for the assay of phospholipase A2.

Two fluorescent phospholipid analogs, 1-O-[12-(2-naphthyl)-dodec-11-enyl]-2-O-decanoyl-sn-glycero-3-phos phocholine and dansyl-PC [rac 1-O-(N-dansyl-11- amino-1-undecyl)-2-O-decanoyl-glycero-3-phosphocholine], and a red-colored analog, dabsyl-PC [rac 1-O-(N-dabsyl-11-amino-1-undecyl)-2-O-decanoyl-glycero-3- phosphocholine], were evaluated as substrates for the assay of phospholipase A2 (PLA2) (Crotalus adamanteus). The assay reaction was monitored by separation of the fluorescent or colored substrate and product by TLC and quantitation by fluorescence or absorption scanning. All three substrates gave similar (within an order of magnitude) activities with PLA2. Dansyl-PC was best suited to this TLC assay. Dabsyl-PC was less sensitive to detection by absorbance, but had the advantage of being red colored and readily detected by the unaided eye.

Fluorophore-labeled ether lipids: substrates for enzymes of the platelet-activating factor cycle in peritoneal polymorphonuclear leukocytes.

Cell-free preparations of ionophore-stimulated peritoneal rat polymorphonuclear neutrophils (PMNs) incubated with 1-(N-dansyl-11-amino-1-undecyl)-sn-glycerol-3-phosphorylcholine (dansyllyso-PAF) converted this fluorescent lyso ether lipid into two different classes of products. In the absence of acetyl-CoA 1-(N-dansyl-11-amino-1-undecyl)-2-long chain acyl-sn-glycerol-3-phosphorylcholine (dansylalkyl-2-acyl-GPC) was the only identified new fluorescent phospholipid. In the presence of acetyl-CoA an additional new product, 1-(N-dansyl-11-amino-1-undecyl)-2-acetyl-sn-glycerol-3-phosphorylcholine (dansyl-PAF), was formed. The formation of dansyl-PAF in PMN homogenates was only transient with a maximum after about 4 min. When PMN homogenates were incubated with dansyl-PAF the formation of dansyllyso-PAF was observed prior to the formation of dansyl-2-acyl-GPC. Thus, our data indicate that enzymatically formed dansyl-PAF is completely remodeled into dansylalkyl-2-acyl-GPC by the sequential action of PAF acetylhydrolase and CoA-independent transacylase. These results demonstrate that peritoneal rat PMNs contain lyso-PAF acetyltransferase, PAF acetylhydrolase, and CoA-independent transacylase and that fluorophore-labeled ether lipids provide an easy means to assay enzymes which catalyze important enzymatic reactions involved in the biosynthesis and remodeling of platelet-activating factor.