Activation of Phospholipase D by Phosphatidic Acid
1998; Elsevier BV; Volume: 273; Issue: 20 Linguagem: Inglês
10.1074/jbc.273.20.12195
ISSN1083-351X
AutoresDong Geng, Justin Chura, Mary F. Roberts,
Tópico(s)Growth Hormone and Insulin-like Growth Factors
ResumoThe activity of bacterial phospholipase D (PLD), a Ca2+-dependent enzyme, toward phosphatidylcholine bilayers was enhanced 7-fold by incorporation of 10 mol % phosphatidic acid (PA) in the vesicle bilayer. Addition of other negatively charged lipids such as phosphatidylinositol, phosphatidylmethanol, and oleic acid either inhibited or had no effect on enzyme activity. Only negatively charged lipids with a free phosphate group, phosphatidylinositol 4-phosphate and lyso-PA, had the same effect as PA on enzyme activity. Changes in vesicle curvature and fusion were not the reason for PA activation; rather, a metal ion-induced lateral segregation of PA in the vesicle bilayer correlated with PLD activation. Significant PA activation was also observed with monomer phosphatidylcholine substrate upon the addition of PA vesicles. The PA activation was caused by Ca2+·PA interacting with PLD at an allosteric site other than active site. The activity of bacterial phospholipase D (PLD), a Ca2+-dependent enzyme, toward phosphatidylcholine bilayers was enhanced 7-fold by incorporation of 10 mol % phosphatidic acid (PA) in the vesicle bilayer. Addition of other negatively charged lipids such as phosphatidylinositol, phosphatidylmethanol, and oleic acid either inhibited or had no effect on enzyme activity. Only negatively charged lipids with a free phosphate group, phosphatidylinositol 4-phosphate and lyso-PA, had the same effect as PA on enzyme activity. Changes in vesicle curvature and fusion were not the reason for PA activation; rather, a metal ion-induced lateral segregation of PA in the vesicle bilayer correlated with PLD activation. Significant PA activation was also observed with monomer phosphatidylcholine substrate upon the addition of PA vesicles. The PA activation was caused by Ca2+·PA interacting with PLD at an allosteric site other than active site. Phospholipase D (PLD) 1The abbreviations used are: PLD, phospholipase D; PA, phosphatidic acid; PLA2, phospholipase A2; lyso-PA, 1-acyl-2-hydroxyglycero-3-phosphate; DAG, diacylglycerol; PC, phosphatidylcholine; PIP, phosphatidylinositol monophosphate; PIP2, phosphatidylinositol bisphosphate; POPC, 1-palmitoyl-2-oleoylphosphatidylcholine; diC4PC, dibutyroylphosphatidylcholine; LPC, 2-lauroyl-2-hydroxyglycero-3-phosphocholine; POPA, 1-palmitoyl-2-oleoylphosphatidic acid; PMe, phosphatidylmethanol; PI, phosphatidylinositol; C16pyr-COOH, 1-pyrenehexadecanoic acid; C16C6pyr-PC, 1-hexadecanoyl-2(1-pyrenehexanoyl)-sn-glycero-3-phosphocholine; C16C6pyr-PMe, 1-hexadecanoyl-2(1-pyrene-hexanoyl)-sn-glycero-3-phosphomethanol; C16C6pyr-PA, 1-hexadecanoyl-2(1-pyrenehexanoyl)-sn-glycero-3-phosphate; diC4PA, dibutyroylphosphatidic acid; C16C6pyr-PA, 1-hexadecanoyl-2(1-pyrenehexanoyl)-sn-glycero-3-phosphate; SUV, small unilamellar vesicle; LUV, large unilamellar vesicle; LPA, 2-lauroyl-2-hydroxyglycero-3-phosphate; PLC, phospholipase C; CMC, critical micelle concentration; ARF, ADP-ribosylation factor.1The abbreviations used are: PLD, phospholipase D; PA, phosphatidic acid; PLA2, phospholipase A2; lyso-PA, 1-acyl-2-hydroxyglycero-3-phosphate; DAG, diacylglycerol; PC, phosphatidylcholine; PIP, phosphatidylinositol monophosphate; PIP2, phosphatidylinositol bisphosphate; POPC, 1-palmitoyl-2-oleoylphosphatidylcholine; diC4PC, dibutyroylphosphatidylcholine; LPC, 2-lauroyl-2-hydroxyglycero-3-phosphocholine; POPA, 1-palmitoyl-2-oleoylphosphatidic acid; PMe, phosphatidylmethanol; PI, phosphatidylinositol; C16pyr-COOH, 1-pyrenehexadecanoic acid; C16C6pyr-PC, 1-hexadecanoyl-2(1-pyrenehexanoyl)-sn-glycero-3-phosphocholine; C16C6pyr-PMe, 1-hexadecanoyl-2(1-pyrene-hexanoyl)-sn-glycero-3-phosphomethanol; C16C6pyr-PA, 1-hexadecanoyl-2(1-pyrenehexanoyl)-sn-glycero-3-phosphate; diC4PA, dibutyroylphosphatidic acid; C16C6pyr-PA, 1-hexadecanoyl-2(1-pyrenehexanoyl)-sn-glycero-3-phosphate; SUV, small unilamellar vesicle; LUV, large unilamellar vesicle; LPA, 2-lauroyl-2-hydroxyglycero-3-phosphate; PLC, phospholipase C; CMC, critical micelle concentration; ARF, ADP-ribosylation factor.enzymes cleave the distal phosphodiester bond of phospholipids generating phosphatidic acid (PA) and a free base (1Roberts M.F. Zhou C. Encyclopedia of Molecular Biology and Molecular Medicine. VCH Publishers, Weinheim, Germany1996: 415-432Google Scholar). In addition to hydrolytic activity, PLD enzymes also catalyze a transphosphatidylation reaction in the presence of a high concentration of primary alcohol (2Yang S.F. Freer S. Benson A.A. J. Biol. Chem. 1967; 242: 477-484Abstract Full Text PDF PubMed Google Scholar,3Eibl A. Kovatchev S. Methods Enzymol. 1981; 197: 493-499Google Scholar). This reaction, which is consistent with a phosphoryl-enzyme intermediate (4Holbrook P. Pannell L.K. Daly J.W. Biochim. Biophys. Acta. 1991; 1084: 155-158Crossref PubMed Scopus (29) Google Scholar), has been used to monitor the presence of PLD in a variety of cells. PLD activities, observed in both membrane and cytosolic fractions of mammalian cells, play key roles in membrane trafficking and regulation of mitosis as well as signal transduction (5Exton J.H. Physiol. Rev. 1997; 77: 303-320Crossref PubMed Scopus (385) Google Scholar). The lipophilic product of PLD cleavage, PA, and its PLA2 degradation product, lyso-PA, are second messengers and have activation roles in a wide variety of cells (6Salmon D.M. Honeyman T.W. Nature. 1980; 284: 344-347Crossref PubMed Scopus (158) Google Scholar, 7Murayama T. Ui M. J. Biol. Chem. 1987; 262: 5522-5529Abstract Full Text PDF PubMed Google Scholar, 8Jalink K. van Corven E.J. Moolenaar W.H. J. Biol. Chem. 1990; 265: 12232-12239Abstract Full Text PDF PubMed Google Scholar, 9Ferguson J.E. Hanley M.R. Arch. Biochem. Biophys. 1992; 297: 388-392Crossref PubMed Scopus (40) Google Scholar). Lyso-PA appears to be the more potent lipid mediator (10Moolenaar W.H. Trends Cell Biol. 1994; 4: 213-219Abstract Full Text PDF PubMed Scopus (129) Google Scholar); concentrations in the nm range elicit diverse biological actions,e.g. activation of DNA synthesis. PA can also be converted to a nonsignaling lipid via the PLD transferase activity and DAG to produce bis-PA (11Van Blitterswijk W.J. Hilkmann H. EMBO J. 1993; 12: 2655-2662Crossref PubMed Scopus (61) Google Scholar, 12Van Blitterswijk W.J. Hilkmann H. de Widt J. van der Bend R. J. Biol. Chem. 1991; 266: 10337-10343Abstract Full Text PDF PubMed Google Scholar). In mammalian cells, basal PLD activity is low, although it can be activated very rapidly. Given its physiological importance, determining the factors that activate or inhibit PLD is of considerable interest. PLD enzymes are also present in plants and various microorganisms. Many of these enzymes have been purified and well characterized kinetically. Plant and bacterial PLD enzymes share a number of kinetic characteristics with the recently purified mammalian PLD isozymes and so may be reasonable models for the latter enzymes. PLD fromStreptomyces chromofuscus is a water-soluble enzyme purified from the culture supernatant (13Imamura S. Horiuti Y. J. Biochem. 1979; 85: 79-95Crossref PubMed Scopus (135) Google Scholar). It can mimic the effect of endogenous PLD when added to a variety of mammalian cells. For example, the addition of exogenous S. chromofuscus PLD induces an activity similar to that of endogenous PLD in ovarian granulosa cell culture (14Liscovitch M. Amsterdam A. J. Biol. Chem. 1989; 264: 11762-11767Abstract Full Text PDF PubMed Google Scholar). Addition of bacterial PLD to the medium of vascular smooth muscle cells induces DNA synthesis along with formation of choline and PA (15Kondo T. Inui H. Konishi F. Inagami T. J. Biol. Chem. 1992; 267: 23609-23616Abstract Full Text PDF PubMed Google Scholar). With this in mind, we have examined the effect of the lipophilic PLD product PA on PLD from S. chromofuscus. The inclusion of PA in a phosphatidylcholine (PC) bilayer, rather than inhibiting the enzyme (as might be expected for simple product inhibition), activates the enzyme allosterically and enhances activity significantly. This type of interaction may be relevant to signal transduction because phosphatidylinositol monophosphate (PIP) (and presumably phosphatidylinositol bisphosphate (PIP2)) can also activate the enzyme. Phospholipids including POPC, dimyristoyl-PC, diC4PC, POPA, LPA, PMe, PI, and oleic acid were purchased from Avanti in chloroform solutions and used without further purification. Triton X-100 and β-octyl glucoside were obtained from Sigma. Pyrene-labeled phospholipids (C16pyr-COOH, C16C6pyr-PC, and C16C6pyr-PMe) were purchased from Molecular Probes. PLD from S. chromofuscus, obtained form Sigma, gave rise to three major bands in SDS-polyacrylamide gel electrophoresis analysis. The enzyme was purified further with the following steps. The commercially available enzyme was dissolved in 1 m(NH4)2SO4 and 50 mmphosphate buffer, pH 7.0, and loaded onto a Hitrap HIC column (Amersham Pharmacia Biotech) preequilibrated with the same buffer. The column was then eluted with a (NH4)2SO4gradient in phosphate buffer. Two protein fractions with PLD activity were obtained. PLD1 (the first protein eluted) showed one single band with a subunit molecular mass of 57 kDa. This appears to be the same as the PLD enzyme reported previously (13Imamura S. Horiuti Y. J. Biochem. 1979; 85: 79-95Crossref PubMed Scopus (135) Google Scholar). The second protein fraction containing PLD activity (termed PLD2) showed two bands on SDS-polyacrylamide gel electrophoresis, at 42 and 19.7 kDa. Although both PLD1 and PLD2 showed PLD activities, only PLD1 was used in extensive kinetic studies of the effect of PA because its subunit molecular mass corresponded to that reported previously for PLD from this organism (13Imamura S. Horiuti Y. J. Biochem. 1979; 85: 79-95Crossref PubMed Scopus (135) Google Scholar). Commercially available PLD from cabbage (obtained from Sigma) was used without further purification. Enzyme concentrations were measured by the Lowry assay using bovine serum albumin as a standard (16Lowry O.H. Rosebrough N.J. Farr A.L. Randall R.J. J. Biol. Chem. 1951; 193: 265-275Abstract Full Text PDF PubMed Google Scholar). 100 mmdiC4PC dissolved in 50 mm ammonium formate containing 0.5 mm Ca2+, pH 7.5, was incubated overnight at room temperature with 100 μg of PLD from S. chromofuscus. DiC4PA was purified from unreacted PC by elution from a QAE-Sephadex A-25 column using 1 m ammonium formate. C16C6pyr-PA was generated similarly byS. chromofuscus PLD-catalyzed hydrolysis of C16C6pyr-PC (solubilized in an ether/water/borate buffer emulsion with 0.5 mmCa2+ at pH 7.5). The hydrolysis product, C16C6pyr-PA, was isolated by extraction of the aqueous reaction mixture several times with a chloroform/methanol (100:10) solution. The purity of C16C6pyr-PA and diC4PA was confirmed by 1H and31P NMR spectroscopy. Lipid solutions in chloroform were dried under argon, lyophilized, and then suspended in 50 mmimidazole-D2O buffer, pH 7.2 (meter reading). For preparation of small unilamellar vesicles (SUVs), the aqueous lipid suspensions were sonicated in 3-min intervals (using a Branson W-350 sonicator) until maximum optical clarity was achieved. LUVs were prepared by extrusion using a LiposoFast Basic extruder with a 100-nm pore filter. The diameter of 6 mm PC LUVs prepared in this fashion was 80 ± 27 nm (17MacDonald R.C. MacDonald R.I. Menco B.M. Takeshita K. Subbarao N.K. Hu L. Biochim. Biophys. Acta. 1991; 1062: 297-303Crossref Scopus (1366) Google Scholar). 1H NMR (500 MHz) spectra, monitoring choline production, were acquired with a Varian Unity 500 spectrometer using an indirect probe and temperature of 30 °C. The following parameters were used in acquiring spectra: 1.2-s acquisition time, 1.0-s relaxation delay, and 6.2-μs pulse width (90°). Chemical shifts were referenced to the residual water resonance (before presaturation) at 4.75 ppm. The total volume of each assay sample was 400 μl. An initial spectrum was acquired before adding the enzyme and calcium solutions; this served as the zero time control. After the addition of PLD, an arrayed experiment was carried out for about 1 h. Initial rates were obtained from the progress curve for 10–20% PC hydrolysis as monitored by the increase in choline N(CH3)3 intensity. Errors in determining rates with this method were typically ≤15%. For kinetic studies involving diC4PC, a radiometer pH-stat model VIT90 was used to monitor generation of the soluble PA. The pK a2 for short chain PA (both monomer and micelle) is 6.8 (18Garigapati V. Bian J. Roberts M.F. J. Coll. Int. Sci. 1995; 169: 486-492Crossref Scopus (12) Google Scholar), so that with an end point of 8.0 essentially all of the product can be titrated. PLD activity at each PC concentration was measured in duplicate or triplicate using 4 mm NaOH as the titrant. Steady-state fluorescence measurements of vesicle mixing were carried out on a Shimadzu RF5000V spectrofluorometer at 30 °C (19Soltys C.E. Roberts M.F. Biochemistry. 1994; 33: 11608-11617Crossref PubMed Scopus (21) Google Scholar). The pyrene-labeled lipids were excited at 350 nm with both excitation and emission slit widths set at 1.5 nm. The emission spectrum was monitored from 360 to 550 nm; almost all of the fluorescence information for both monomer and excimer bands was included in this range. The final concentration of labeled probe was 20 μm, the volume of sample was 400 μl, and total lipid concentration after dilution was 10 mm. These conditions mimic the reaction conditions used in the NMR experiments. PLD enzymes, including the crude enzyme from S. chromofuscus, have an absolute requirement for Ca2+because enzyme activity was abolished in the presence of 2 mm EDTA. DiC4PC is a soluble phospholipid with a very high CMC. More critical for kinetics where PA is the product, diC4PA does not form a precipitate with mmCa2+. With this soluble substrate/soluble product assay system, the K D for Ca2+ was found to be 0.075 mm. There was no PLD activity without Ca2+ (e.g. in the presence of EDTA). Neither Zn2+ nor Ba2+ could substitute for the Ca2+ requirement. However, low PLD activity was observed in the absence of Ca2+ but with 1 mmMg2+ added. The PLD specific activity with Mg2+was about half of the activity of PLD without the addition of any metal ions (but without added EDTA). Enzyme activity did not increase with increasing Mg2+, suggesting that the observed activity was from the low level of contaminating Ca2+ in the assay solution and that the Mg2+ did not compete well for the Ca2+ site on the enzyme. PLD activity toward monomeric PC increased with increasing PC acyl chain length: for diC4PC,V max = 29.0 ± 1.2 μmol min−1 mg−1 and K m = 0.36 ± 0.06 mm; for monomeric diC6PC,V max = 61.5 ± 4.7 μmol min−1 mg−1 and K m = 0.05 ± 0.04 mm. The PLD specific activity showed no dependence on micellization of a monomeric substrate. PLD specific activity was constant, 61.4 ± 1.8 μmol min−1mg−1, for five concentrations of diC6PC ranging from 2 to 30 mm. This phospholipid has a CMC of 14 mm; therefore, this bacterial PLD exhibits no interfacial activation. The two-dimensional concentration of the substrate in the interface was also not an important parameter for this phospholipase. With Triton X-100/POPC mixed micelles as the substrate, no decrease in activity was observed when the Triton concentration was increased 3-fold at a fixed PC concentration. The lack of a surface dilution effect as well as interfacial activation under these conditions indicated that this bacterial PLD behaved more like an esterase than a typical lipase. The absolute requirement of PLD for calcium complicates kinetics because Ca2+ forms a precipitate with PA, the PLD hydrolysis product. The Ca2+·PA complexes also cause massive particle growth in short chain (except for diC4PC) micelle systems when PA is generated. The relatively large size of vesicles (so that substrate depletion is not a significant problem for <20% hydrolysis) and tolerance of higher Ca2+ concentrations before precipitation or fusion occur make POPC/POPA vesicles ideal for examining the effects of product on PLD catalysis. However, the high pK a2 of PA in vesicles (pK a2 = 7.6 in predominantly PC bilayers (20Swairjo M. Seaton B.A. Roberts M.F. Biochim. Biophys. Acta. 1994; 1191: 354-361Crossref PubMed Scopus (49) Google Scholar)) makes pH-stat assays of PLD action problematic because at an end point of 8.0 only part of the PA is titratable. Furthermore, the PA pK a2 increases as more of the bilayer surface is occupied by PA (20Swairjo M. Seaton B.A. Roberts M.F. Biochim. Biophys. Acta. 1994; 1191: 354-361Crossref PubMed Scopus (49) Google Scholar). Therefore, 1H NMR spectroscopy was used to monitor PLD activity by measuring the intensity of the water-soluble choline N-methyl resonance. The N-methyl region of the SUVs of POPC exhibits resonances for inner and outer leaflet PC at 3.15 and 3.18 ppm. The water-soluble choline N-methyl resonance is observed as a sharp resonance at 3.09 ppm, upfield of the inner PC (Fig.1). The integral of the choline resonance as a function of time provides a sensitive measure of PLD activity. Calcium was added with enzyme to avoid any vesicle fusion before PLD generation of PA. There is a small lag in the reaction progress curve (typically around 5 min) under these conditions (Fig.2 A, filled circles). The lag is independent of the POPC concentration. This is reminiscent of the lag phase toward vesicle substrates observed for 14-kDa PLA2 enzymes acting on PC bilayer substrate (21Jain M.K. Gelb M.H. Methods Enzymol. 1991; 197: 112-125Crossref PubMed Scopus (82) Google Scholar). For PLA2, the binding affinity of enzyme to the PC vesicle surface was low and represented a slow step in the enzyme reaction. Any factor that can facilitate enzyme surface binding abolished the lag and increased the observable enzyme activity (22Ghomashchi F. Yu B. Berg O. Jain M.K. Gelb M.H. Biochemistry. 1991; 30: 7318-7329Crossref PubMed Scopus (94) Google Scholar). The dependence of PLD activity (measured after the lag phase) toward POPC vesicles on calcium concentration was hyperbolic with a K D for Ca2+ of 3.9 mm, considerably higher than the value for short chain PC monomers. Ca2+ induces fusion in SUVs containing negatively charge phospholipids. Therefore, to optimize PLD activity and minimize fusion when PA (or other negatively charged lipids) was included in vesicles, 5 mm calcium was used in the assays unless otherwise noted. As shown in Fig.3, the dependence of PLD specific activity on POPC concentration was hyperbolic with an apparentK m = 6.5 ± 1.7 mm andV max = 13.5 ± 1.5 μmol min−1 mg−1. Although PLD does not exhibit interfacial activation or surface dilution in micelle systems, perturbations of the bilayer could alter the observed specific activity of PLD in a vesicle system. Incorporation of 5–10 mol % DAG, a lipid that destabilizes bilayers to fusion, into POPC SUVs had little effect on the PLD specific activity. Furthermore, the presence of the DAG had no effect on the lag phase (e.g. see Fig. 2 B, where the 5-min lag is still observed).Figure 2Reaction progress curve for hydrolysis of different POPC SUVs. The amount of hydrolysis product choline was calculated from the 1H integrated intensity of the choline resonance compared with an internal standard. SUVs were composed of (panel A) 10 mm POPC (•) and 9 mmPOPC with 1 mm LPA (○); (panel B) 9 mm POPC with 1 mm DAG (⋄) or 1 mmPI (▿); (panel C) 9 mm POPC with 1 mm PMe (♦) or 1 mm oleic acid (▵). Reaction conditions include 50 mm imidazole, pH 7.2, 5 mm Ca2+, and 1.2 μg of PLD.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 3Specific activity of S. chromofuscus PLD toward different concentrations of POPC SUVs in the absence (○) and presence of (•) 10 mol % POPA. Reaction conditions are the same as in Fig. 2.View Large Image Figure ViewerDownload Hi-res image Download (PPT) POPA was deliberately incorporated into POPC SUVs to determine if the product of PLD had any effect on the initial rates of reaction. The addition of 0.5–5 mm Ca2+ caused no massive aggregation or precipitation of POPC/POPA vesicles as long as the PA content was <30%. In studies of PLA2 activity toward PC vesicles, it was found that incorporation into the bilayer of a threshold level of product or other anionic lipids diminished the lag phase and increased the enzyme activities (23Jain M.K. Cordes E.H. J. Membr. Biol. 1973; 14: 119-134Crossref PubMed Scopus (27) Google Scholar). With that phospholipase, negatively charged lipids facilitate the binding of enzyme to the bilayer interface (24Jain M.K. Berg O. Biochim. Biophys. Acta. 1989; 1002: 127-156Crossref PubMed Scopus (292) Google Scholar). Products also alter the polarity in the interface region, and this facilitates PLA2 action (25Sheffield M.J. Baker B.L. Owen N.L. Baker M.L. Bell J.D. Biochemistry. 1995; 34: 7796-7806Crossref PubMed Scopus (46) Google Scholar). If POPA promotes binding of PLD to the vesicle surfaces, there might be an increase in PLD specific activity as well as the disappearance of the lag phase. On the other hand, PA is a PLD product and might give rise to product inhibition. PLD activity toward POPC/POPA SUVs was examined as a function of calcium concentration and mole fraction POPA. The relative activities of PLD toward POPC and POPC/POPA SUVs with various Ca2+concentrations are shown in Table I. The incorporation of POPA in PC SUVs increased the enzyme specific activity about 5–8-fold. The presence of POPA also diminished the lag phase, presumably because it facilitated binding of the PLD to the vesicle. The activation of PLD by PA was calcium-dependent. At low Ca2+ (0.5 mm), incorporation of POPA in the vesicles inhibited PLD; at higher Ca2+ concentrations, the same concentration of POPA activated PLD. The inhibition at low calcium concentrations may be caused by the competition of Ca2+preferential binding to PA instead of PLD. At 5 mmCa2+, the apparent V max of PLD toward POPC vesicles with 10 mol % POPA in the bilayer was 96.0 ± 8.9 μmol min−1 mg−1, and the apparentK m was 4.8 ± 1.4 mm. The 7-fold increase in V max could reflect an increase in PLD adsorption to the vesicle surface as well as an increase ink cat. The slight reduction in the apparentK m for substrate (from 6.5 to 4.8 mm) was within the error of K m determination. At a fixed total phospholipid concentration (10 mm) and 5 mm Ca2+, the mol % of POPA incorporated into the vesicles was also varied. As seen in Fig.4, PLD activity increased with the mol % of POPA to a maximum ∼20 mol % PA. At greater than 30 mol % PA, the presence of the Ca2+ caused vesicle aggregation and fusion; PLD activity also decreased somewhat. However, under these conditions (PC + PA = 10 mm), it is also possible that the substrate (PC) concentration decreased to a level near the apparentK m, leading to the decrease in observed activity.Table IRelative activity of S. chromofuscus PLD toward small unilamellar vesicles of POPC and POPC/POPACa2+Relative PLD activity1-aSpecific activities are divided by the value of 3.1 μmol min−1 mg−1 for PLD acting on 10 mmPOPC vesicles in the presence of 0.5 mm Ca2+. Typical errors in measuring PLD activity by 1H NMR methods are less than 15%. Thus, changes in relative activity greater than 0.3 are significant.POPCPOPC/POPA (9:1)POPC/POPA (8:2)POPC/POPA (7:3)mm0.51.00.4911.252.6321.505.7633.5116.753.5017.831.329.11-a Specific activities are divided by the value of 3.1 μmol min−1 mg−1 for PLD acting on 10 mmPOPC vesicles in the presence of 0.5 mm Ca2+. Typical errors in measuring PLD activity by 1H NMR methods are less than 15%. Thus, changes in relative activity greater than 0.3 are significant. Open table in a new tab PA is dianionic, and the negative charges of PA may play a major role in the binding of enzyme to the vesicle surface and thus activating PLD. Other lipids and amphiphiles with negative charges, PMe, oleic acid, PI, LPA, and PIP, were incorporated into POPC vesicles to see if they could activate PLD. The relative PLD activities toward POPC vesicles with 10 mol % anionic lipids are shown in TableII. Most of the negatively charged lipids did not activate PLD, nor did they reduce the lag phase (Fig. 2,B and C). PMe had almost no effect, whereas the incorporation of oleate and PI inhibited the enzyme activity. Along with the inhibited PLD activity was an increased lag time (compare Fig.2, B and C, where the presence of 10 mol % PI and oleate leads to an increased lag phase). However, both PIP and lyso-PA activated PLD. The lyso-PA activation was comparable to that of PA, whereas the PIP activation was 3-fold less. Again, when PLD activation was observed, the lag phase was reduced or abolished (e.g. see Fig. 2 A, open circles, for the effect LPA). Activating lipids all have phosphate monoester groups, suggesting that the activation was specific for a membrane-localized phosphate moiety.Table IIRelative activity of S. chromofuscus PLD toward POPC/X (9 mm:1 mm) vesiclesXRelative activity2-aSpecific activities are divided by the value of 11.5 μmol min−1 mg−1 for PLD acting on 10 mmPOPC vesicles in the presence of 5 mm Ca2+. Typical errors in measuring PLD activity by 1H NMR methods are less than 15%. Thus, changes in relative activity greater than 0.3 are significant.SUVs —2-bVesicles made from 10 mm POPC used as a control.1.0 POPA5.0 Lyso-PA5.7 PI0.63 PMe1.1 DAG1.1 Oleic acid0.22 PIP1.8LUVs2-cLUVs prepared by extrusion. —2-bVesicles made from 10 mm POPC used as a control.0.12 POPA1.82-a Specific activities are divided by the value of 11.5 μmol min−1 mg−1 for PLD acting on 10 mmPOPC vesicles in the presence of 5 mm Ca2+. Typical errors in measuring PLD activity by 1H NMR methods are less than 15%. Thus, changes in relative activity greater than 0.3 are significant.2-b Vesicles made from 10 mm POPC used as a control.2-c LUVs prepared by extrusion. Open table in a new tab Vesicle fusion is always a concern in enzymatic reactions involving vesicles containing anionic lipids and Ca2+. Enzyme can be transferred from one vesicle to another through the fusion, potentially complicating kinetics. High concentrations of Ca2+ enhance fusion of vesicles with negatively charged amphiphiles such as PA and fatty acid (26Yamamoto I. Konto A. Handa T. Miyajima K. Biochim. Biophys. Acta. 1995; 1233: 21-26Crossref PubMed Scopus (32) Google Scholar, 27Blackwood R.A. Smolen J.E. Transue A. Hessler R.J. Harsh D.M. Brower R.C. French S. Am. J. Physiol. 1997; 272: 1279-1285Crossref PubMed Google Scholar). To investigate whether PA activation of PLD was caused by calcium-related vesicle fusion, the effect of other divalent cations on the PLD hydrolysis of POPC/POPA vesicles was investigated. TableIII shows the effect of Mg2+and Ba2+ on PA activation. Because there is no activity if Ca2+ is absent, 0.5 or 1 mm Ca2+was used with 4.5 and 4 mm Mg2+ or Ba2+ added. The addition of the other two metal ions had small effects on PLD activity toward POPC vesicles, with Mg2+ more effective in replacing the excess Ca2+ than Ba2+. A more significant activation was seen with PA-containing vesicles. Neither Mg2+ nor Ba2+ led to amounts of activation comparable to Ca2+, although the addition of Mg2+ enhanced PLD activity to roughly half of that produced by Ca2+ at a comparable concentration. The added Mg2+ caused some precipitation in the POPC/POPA vesicle solution (even without the addition of PLD). In contrast, the POPC/POPA vesicle solution was opalescent with 5 mm Ba2+ and 1 mmCa2+. POPC/POPA (9:1) vesicle solutions appeared opalescent for several hours after the addition of up to 10 mmCa2+, indicating that they were reasonably stable in the absence of PLD under these conditions. Thus, an excess of divalent ion along with a minimum amount of Ca2+ for catalysis is critical for optimal PLD activity.Table IIIEffect of divalent metal ions on the specific activity of S. chromofuscus PLD toward POPC and POPC/POPA vesiclesM2+Specific activity3-aTypical errors in measuring PLD specific activity by1H NMR methods are less than 15%.Ca2+Mg2+Ba2+POPC (10 mm)POPC/POPA (9 mm:1 mm)POPC/POPA (7 mm:3 mm)μmol min −1 mg −10.53.11.513.98.7511.563.160.4147.435.630.20.54.53.43.0143.622.325.53-a Typical errors in measuring PLD specific activity by1H NMR methods are less than 15%. Open table in a new tab The effect of PA on the specific activity of PLD from cabbage was also examined. Vesicles with 1 mm POPA and 9 mm POPC (with 5 mmCa2+) exhibited a 26-fold enhanced PLD specific activity compared with pure POPC (10 mm) vesicles. Cabbage PLD is different from the bacterial enzyme in that it appears to show some interfacial activation (28Lewis, K. (1992) Short-chain Phospholipids as Probes of Phospholipase Activity. Ph.D. dissertation, Boston College.Google Scholar) using short chain PCs as substrates. That PA activates the enzyme might suggest that PA activation is a property for many PLD enzymes. The incorporation of dianionic POPA into SUVs stabilizes the curvature of the small vesicles. PLD may bind more effectively to a curved surface than to a flat surface. The effect of vesicle curvature on enzyme activity can be assessed by comparing specific activities toward large and small unilamellar POPC and POPC/POPA vesicles (Table II). The specific activity of PLD toward POPC LUVs with 5 mm Ca2+w
Referência(s)