Purification of a Glycosyl-Phosphatidylinositol-specific Phospholipase ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY 1989 by The American Society for Biochemistry and Molecular Biology, Inc.

Val. 264, No. 23, Issue of August 15, pp. 13760-13764, 1989 Printed in U.S.A.

Q

Purification of a Glycosyl-Phosphatidylinositol-specific Phospholipase D from Human Plasma* (Received for publication, December 2,1988)

Michael A. DavitzS, Judy Hom, and SergioSchenkmang From

the Division of Immunology, Department of Pathology, New York University School of Medicine, New York,New York 10016

Mammalian plasma contains a phospholipase D, which is specific for the glycosyl-phosphatidylinositol anchor found on many eukaryotic cell surfaceproteins (Davitz, M. A., Hereld, D., Shak, S., Krakow, J., Englund, P. T., and Nussenzweig, V. (1987) Science 238, 81-84; Low, M. G., and Prasad, A. R. S. (1988)Proc. Natl. A c d . Sci. U. S. A. 85, 980-984; Cardoso de Almeida, M. L., Turner, M. J., Stambuk, B. V., and Schenkman, S. (1988) Biochem. Biophys. Res. Commun. 150,476-482). We have purified this phospholipase D to homogeneity by a four-step procedure involving a Mono Q and phenyl-5PW columns, followed by wheat germ lectin affinity chromatography and finally another Mono Q column. A 4,500-fold purification was achieved with a 5%yield. By sodium dodecyl sulfate-polyacrylamide gel electrophoresis, the homogeneous enzyme has a M , of 110,000 and appears to consist of a single polypeptide chain. It exhibits identical substrate specificity as compared with the crude preparation, is active over a broad pH range (4.0-8.5), inhibited by the thiol-blocking agent p-chloromercuriphenylsulfonic acid and by 1,lO-phenanthroline, and is partially heat-labile.

glycan structure consisting of (Man)sGlcN, with a branched side chain linked to one of the mannose sugars containing a variable number of galactose residues (4,5). Thisglycan is 0glycosidically linked to the 6-hydroxyl group on the inositol ring of dimyristoyl-PI (DMPI) (4, 5). The structure of the glycolipid anchors in both parasiticand mammalian proteins appears to be highly conserved (1-3). There aretwo proposed functions for the GPI anchor. One is facilitation of lateral mobility of the protein in the lipid bilayer (6-8); the other is that it may allow for the selective release of the protein through cleavage of the anchor by GPIspecific phospholipases (1-3). GPI-specific phospholipase C enzymes have been discovered in Trypanosoma brucei (9-11), Trypanosoma cruzi (12, 13), and mammalian liver cells (14). Recently, we and others have discovered a GPI-specific phospholipase Dinmammalian plasma (15-17). The only known substrate for this enzyme is the GPI membrane anchor present on mfVSG, decay accelerating factor (DAF), alkaline phosphatase, and 5‘-nucleotidase, as well as thebiosynthetic precursors of mfVSG, glycolipids A and C (18, 19). In this paper, we describe a method for purifying the plasma GPIspecific phospholipase D to homogeneity and characterize some of its biochemical properties. EXPERIMENTALPROCEDURES

Many eukaryotic proteins are anchoredto theplasma membrane by a glycosyl-phosphatidylinositol (GPI)’ anchor (1-3). The structureof this type of membrane anchor is best known for the membrane form of the variant surface glycoprotein of the African trypanosome (mfVSG). In mfVSG, the C-terminal amino acid is coupled via phosphoethanolamine to a core

Materials-FPLC Mono Q HR 10/10, Mono Q HR 5/5, TSK (33000 SW, and phenyl-5PWglass-pac, chromatographic columns were purchased from Pharmacia LKB Biotechnology Inc. Wheat germ lectinSepharose 6MB and phenyl-Sepharose CL-4B were purchased from Pharmacia. DEAE-cellulose was obtained from Whatman Biosystems. An FPLC system from Pharmacia utilizing two P-500 pumps together with a LC-500 controller and a UV-M monitor was used for * This work is supported in part by grants from the MacArthur purification. PI-specific phospholipase C was a kind gift of Dr. Martin Low Foundation and theLucille P. Markey Charitable Trust andNational (Columbia University). [9,10-3H]Myristic acid (47.5 Ci/mmol) was Institutes of Health Grant AI-08499. The costs of publication of this article were defrayed in part by the payment of page charges. This purchased from Amersham; [1,2-di(l-’4C)]myristoyl-~-3-phosphatiarticle must therefore be hereby marked “advertisement” in accord- dylcholine (DMPC) (108mCi/mmol) and EN3HANCE were purchased from Du Pont-New England Nuclear. ance with 18 U.S.C. Section 1734 solely to indicate this fact. Outdated humanplasma was obtained from the New York Univer$ Lucille P. Markey Scholar. To whom correspondence and reprint sity Hospital Blood Center. Serum was generated by addition of 1.25 requests should be addressed. units of thrombin/ml of plasma. § Supported by a grant from the Conselho Nacional de Desenvol[3H]Myristate-labeled n~fVSG-[~H]Myristate-labeled mfVSG vimento Cientifico 6 Tecnologico (CnPq, Brazil). ([3H]mfVSG) for the GPI-specific phospholipase D assay was preThe abbreviations used are: GPI, glycosyl-phosphatidylinositol; BSA, bovine serum albumin; CAPS, 3-(cyclohexylamine)-propane- pared by labeling trypanosomes (2 X lo9 cells) in vitro with (9,lO)[3H]myristic acid (47.5 Ci/mmol) according to the method of Hereld sulfonic acid; DAF, decay accelerating factor; DMPA, dimyristoylphosphatidic acid; DMPC, dimyristoylphosphatidylcholine; DMPI, et al. (9). T. brucei (Iltat 1.3; obtained from Dr. Paul Englund, The dimyristoylphosphatidylinositol; FPLC,fast protein,peptide, and Johns Hopkins University) were isolated from the blood of Swiss mice by DEAE-cellulose chromatography (20).The labeled substrate, polynucleotide liquid chromatography; Hepes, N-2-hydroxyethylpi[3H]mfVSG, was greater than 90% pure as judged by Coomassie perazine-N’-Z-ethanesulfonicacid; mfVSG, membrane form of the 90% variant surface glycoprotein; [3H]mfVSG, [3H]myristate-labeled staining of SDS-PAGE; fluorography indicated that greater than mfVSG; PI, phosphatidylinositol; SDS-PAGE, sodium dodecyl sul- of the countswere associated withmfVSG. This preparationtypically fate-polyacrylamide gel electrophoresis; Solvent A, CHCkCH30H: yielded [3H]mfVSG with a specific activity of 17,000 cpm/pg. GPZ-specific Phospholipase D Assay-In this assay, the release of 0.25% KC1 (55:45:10); Solvent B, CHC13:CH30H:CH3COCH3: acid (DMPA) was CH3COOH:H20 (30:10:40:10:5); EGTA, [ethylenebis(oxyethyleneni- [3H]myristate-labeleddimyristoylphosphatidic tri1o)Jtetraactic acid; bis-Tris, 2-[bis(2-hydroxyethyI)amino]-2-(hy- measured by liquid scintillation counting of the n-butyl alcohol extract of the reaction mixtures. This assay is similar to the ones used droxymethyl)propane-1,3-diol.

13760

Glycosyl-Phosphatidylinositol-specific Phospholipase

D

13761

by Hereld et al. (9) and Bulow and Overath (IO) for the VSG lipase. products was performed as previously described (151, using either Reactions (100 pl) contained 0.15 pg (approximately 2000 cpm) of CHCl3:CHaOH:0.25%KC1(55:45:10) (solvent A) or CHC13:CH30H: (3010:40:10:5) (solvent B) as the solvent [3H]mfVSG,0.01% Nonidet P-40, 0.01 M NaCl, 50 mM Tris-HCI, pH CH3COCH3:CHaCOOH:H20 7.4, and 2.6 mM CaC12.After addition of the enzyme and incubation systems for development of the TLCplates. for 30 min at 37 "C, the reaction was thoroughly mixed with 1 ml of H20-saturated n-butyl alcohol. The phases were separated by brief RESULTS centrifugation in an Eppendorf Microfuge (30 S ; 15,000 X g); 500 pl The various steps of the purification of the GPI-specific of the upper (organic) phase weremixed with 10 mlof complete counting mixture (3a70b, Research Products International, Inc.) and phospholipase D are summarized in Fig. 1 and Table I. analyzed for radioactivity by liquid scintillation counting. All assays Step 1: FPLC Mono Q (HR 10/10)-Human serum (30 ml) were done in duplicate. The assay was linear with respect to time and was dialyzed overnight at 4 "C against 20 mM bis-Tris-HC1, enzyme concentration for reactions in which up to 0.075 pg of [3HJ mfVSG was hydrolyzed. For the purposes of following the purifica- pH 6.5, 0.01 M NaCl, and 2.6 mM CaC12 and then chromatotion, 1 unit of enzyme was defined as the amount of GPI-specific graphed on a FPLC Mono Q column. Typically, 1000 mg of protein were injected during each chromatographic run. The phospholipase D activity required to completely hydrolyze 0.5 pg of [3H]mfVSG(9) in 30 min a t 37 "C in the presence of excess substrate. column was eluted with a linear gradient from 0.01 M NaCl p H Dependence-Because of wide pH range assayed, pH 3.5-11.0, to 0.5 M NaCl. GPI-specific phospholipase D activity eluted a mixed buffer system was prepared containing: 12.5 mM NaCH3CO0, in a broad peak from 0.15 M to 0.25 M NaC1; active fractions 12.5 mM bis-Tris-HC1, 12.5 mM Tris-HC1, 12.5 mM CAPS, 0.01 M NaCl, and 2.6 mM CaC12. The pH of this buffer system was adjusted were pooledand designated fraction I. No flow-through activto between pH 3.5 and 11.0 at 0.5 pH unit intervals by addition of ity was observed. Recovery of activity in fraction I was 60% HCl. The GPI-specific phospholipase D was diluted in this buffer at with a purification of 18-fold. the appropriate pH, 0.15 pg of [3H]mfVSGwas added, the reaction Step 2: FPLC Phenyl-5PW-Fraction I was directly chromixtures were incubated for 30 min at 37 "C, and enzymatic activity matographed on a FPLC phenyl-5PW column, and thebound was assayed as described above. The results of these assays were protein eluted with a decreasing NaCl gradient (0.25 M NaCl/ expressed as specific activity of the GPI-specific phospholipase D, defined as thenanomoles of 13H]mNSGhydrolyzed/min/mg of pro- H20). Typically, 20 mg of protein were injected during each chromatographic run. Enzymatic activity eluted in a single tein. Inhibitors-The effect of various inhibitors onthe cleavage of [3H] peak (fraction 11, Fig. lA). Recovery of activity was approximfVSG by the GPI-specific phospholipase D was examined. The mately loo%, with a 40-fold purification. purified enzyme was incubated with [3H]mfVSG together with Step 3: Wheat Germ-Fraction I1was chromatographed phenylmethylsulfonyl fluoride, p-chloromercuriphenylsulfonic acid, onto a wheat germ lectin column, and thecolumn eluted with or lJ0-phenanthroline a t the indicated concentrations for 30 min at 37 "C. The heat lability of the enzyme was determined by treating buffer containing 0.45 M n-acetylglucosamine (GlcNAc). Enbroad peak after addition of the GPI-specific phospholipase D for 30 min at 56 "C prior to incu- zymatic activityelutedina bation with [3H]mfVSG.Samples were analyzed for enzymatic activ- GlcNAc (Fig. 1B); activity was pooled as indicated (fraction ity as described above. 111).Recovery was 26% with a 4-fold purification. Free calcium ion concentrations from 0.5to 1000p M were achieved Step 4: FPLC Mono Q (HR 5/5)-Fraction 111 was injected using CaC12/EGTA buffers (21,22); the numerical calculations of free onto a FPLC Mono Q column, and thebound material eluted Ca2+were kindly provided by Dr. Boon Chock, National Institutes of Health. GPI-specific phospholipase D (0.2 unit) was diluted in 50 mM with a linear NaCl gradient from 0.045 M to 0.5 M. Enzymatic Hepes, pH 7.0, 0.1 M NaCl. Reactions (100 pl) contained 0.15 pg activity eluted in a broad peak (Fig. IC). SDS-PAGE of the (approximately 2000 cpm) of 13H]mfVSG, 0.01% Nonidet P-40, 50 active fractions revealed that only one fraction contained a mM Hepes, pH 7.0,O.l M NaCl, and 2 mM EGTA. In order to achieve single band (see inset, Fig. lC, fraction 17). This was desigthe appropriate free Caz+ion concentration, increasing amounts of nated fraction IV; recovery was 30%, with a %fold purificaCaC12were added from a 0.1 M stock solution to the mixture; the tion. reaction mixtures were incubated for 30 min a t 37 "C; and enzymatic Following these steps, 5% of the total enzymatic activity activity was assayed as described above. The results of these assays were expressed as specific activity of the GPI-specific phospholipase was recovered, and theenzyme was purified 4,500-fold. Silver D, defined as the nanomoles of [3H]mfVSG hydrolyzed/min/mg of staining of the SDS-PAGE revealed that fraction IV conprotein. tained a single band with an M , of 110,000 under reducing Decay AcceleratingFactor (DAF)-The ability of the purified GPI- conditions (Fig. 2, lane 4),as well as nonreducing conditions specific phospholipase D to cleave DAF was examined using a phenyl(data notshown). Sepharose binding assay that discriminated between hydrophobic To furtherdemonstratethat enzymaticactivitycorre(GPI-anchored) and hydrophilic (lacking GPI anchor) forms of DAF (15). 220 ng of purified DAF was incubated for 1 h at 37 "C with 1pl sponded with the band seen in Fig. 2, lane 4, the purified (0.85 unit) of purified GPI-specific phospholipase D (total volume, enzyme was chromatographed over a TSK-3000 SW gel fil100 PI). Following incubation, 975 pl of phosphate-buffered saline tration column. The fractions were assayed both for activity containing 75 pl of the phenyl-Sepharose beads were added to the and analyzed by SDS-PAGE. Enzymatic activityco-fractionreaction mixture. After incubation, the unbound DAF was assayed by ated with the 110-kDa band (Fig. 3, A and B ) . The retention two-site immunoradiometric assay (23). To determine whether the GPI-specific phospholipase D could time for enzymatic activity,which was between IgG and BSA release DAF from cell membranes, 75 p1 (7.5 X lo7 cells) of washed (Fig. 3A), is consistent with a molecule of 110 kDa. Enzymatic human erythrocytes, obtained from normal volunteers, were incu- activity in whole serum exhibited similar chromatographic bated with 1 r l (0.85 unit) of the purified enzyme for 1 h a t 37 "C. behavior (data not shown). After incubation, the cells were pelleted by centrifugation, and the DAF-The ability of the enzyme to cleave a mammalian supernatants were assayed for DAF as previously described (23). As controls, human erythrocytes were incubated with buffer alone. ,411 substrate was studied using the GPI-anchored complement regulatory protein, DAF. Treatment of purified membrane assays were done in duplicate. DAF with the enzyme converted 72.3 f 4.0% ( n = 4) of the Polyacrylamide Gel Electrophoresis (SDS-PAGE), Protein Determination, and Thin Layer Chromatography (TLC)-SDS-PAGE was DAF to a hydrophilic form. In the presence of EGTA, only performed using a 7% reducing polyacrylamide gel. "Rainbow" pre15 k 0.4% ( n = 4) of the DAF was hydrophilic. The purified stained molecular weight markers were obtained from Amersham. enzyme, however, failed to release DAF from intact erythroGelswere stained with a Bio-Rad Silver Stain kit, and protein cytes (data not shown). determinations were done using the Bio-Rad protein determination Cleavage Product andSpecificity-To establish that a phostype 11 kit, with bovine serum albumin (BSA) as the standard @iopholipase Dtype enzyme had been purified, the cleavage Rad). Thin layer chromatography of the n-butyl alcohol-soluble reaction product of [3H]mfVSGwas examined by TLC. Purified GPI-

13762

Glycosyl-Phosphatidylinositol-specificPhospholipase D TABLE I Purification of t h e plasma GPI-specific phospholipase D Fraction

Total protein

PFb

%R'

mR

H

Serum 1785 34,550 20,360 18 60 60 Mono Q (HR10/10) (I) 21,060 700 60 1.59 Phenyl-5PW (11) 5,600 1660 16 0.176 Wheat germ (111) 0.020 1,700 4500 5 Mono Q (HR5/5) (IV) Thetotal number of units chromatographed a t each step is indicated by the total number of units in the previous step. The total number of units recovered at each step isindicated by the total number of units at thatstep. * P F purification factor = (total units/total protein)~,,i,./(total units/total protein),,,. %R: % recovery = total units~,,i../34,550.

0 . U Y GlcN*C)

1

2

3

4

FIG.2. SDS-PAGE of the purification. The pooled fractions from each step were subjected to SDS-PAGE (7%) under reducing conditions and then silver-stained. Fractions I-IV are in lunes 2 - 4 , respectively.

specific phospholipase D generated DMPA (data not shown) confirming the identify of the enzyme. The purified enzyme exhibited identical substrate specificity ascompared with the crude enzyme (15). The enzyme was also capable of cleaving glycolipids A and C, the glycolipid precursors of mfVSG (19 and data not shown). However, it did not cleave DMPC, nor did it cleave DMPI, which had been generated by HNOz cleavage of mfVSG (4). Frw(lan Numbor p H Dependence-The enzyme was active over a broad pH FIG. 1. Purification of the plasma GPI-specific phospholi- range, 4.0-8.5. The specific activity of the GPI-specific phospase D. To screen for activity during the purification, 1 pl of each fraction was assayed for enzymatic activity as described under "Ex- pholipase D increased sharply from 0 to 5.0 when the pHwas perimental Procedures." It should be noted that because activity may raised from pH 3.5 to 4.5, remained a t approximately the he underestimated for some fractions, cpm as opposed to units are same level from pH 4.5 to 8.0, and decreased to 0 at pH9.0. shown. A, FPLC phenyl-5PW:fraction I (58 ml) was chromatoInhibitors-The GPI-specific phospholipase D activitywas graphed on a FPLC phenyl-5PW column which had previously been equilibrated in 20 mM bis-Tris-HCI, pH 6.5,0.25 M NaC1, and 2.6 not affected by an inhibitor of serine esterases phenylmethmM CaC12 (flow rate: 1.2 ml/min); 1.2-ml fractions were collected. ylsulfonyl fluoride (1mM). The thiol-blocking agent, p-chloThe bound protein was eluted in 10 min with a decreasing NaCl romercuriphenylsulfonic acid, showed a 47% inhibition of gradient (0.25 M NaC1-H20). The pooled fraction was designated enzymatic activity at 1 mM and an 88%inhibition at 10 mM. fraction 11. R,wheat germ-Sepharose: fractionI1 (15 ml) was adjusted to 50 mM Tris-HCI, pH 7.4, 0.2 M NaCI, and 2.6 mM CaCl, by the Incubation with 1,lO-phenanthroline resulted in a 45% inhiaddition of concentrated Tris-HC1 (1 M), pH 7.4, NaCl (5 M), and bition of enzymatic activity at 50 p~ and a 68% inhibition of CaCI2 (0.1 M). It was then loaded on a 6.4- X 0.5-cm wheat germ activity at 100 p ~ Purified . enzyme was partially heat-labile lectin column (flow rate: 5-10 ml/min). The column was washed with with a 42% loss in activity after heat treatmentfor 30 min at eight column volumes of 50 mM Tris-HCI, pH 7.4, 0.5 M NaCl, and 2.6 mM CaC12,until the absorbance had dropped to below 0.05. The 56 "C. Preliminary experimentsusing unfractionated serumas the column was eluted with 50 mM Tris-HC1, pH 7.4, 0.5 M NaCl, and 2.6 mM CaC12,which contained 0.45 M GlcNAc; 1.5-ml fractions were source of enzyme suggested that the enzyme was Ca2+ ioncollected. The pooled fraction was designated fraction 111. C, FPLC dependent (15,16). Toclarify this issue, purified enzyme was Mono Q (HR5/5): fraction 111 (11ml) was equilibrated in 20 mM bisassayed in CaC1,:EGTA buffers with known free Ca2+ ion Tris-HCI, pH 6.5, 0.045 M NaCI, and 2.6 mM CaC12 by gel filtration over a PD-10 Sephadex G-25M column (Pharmacia). Thesample was concentrations. The specific activity of the enzyme increased chromatographed on a FPLC Mono Q (HR5/5) column, which had with increasing Ca2+.From 1-100 p~ Ca2+,there was a slow been equilibrated in the samebuffer (flow rate: 2ml/min). Thebound increase in specific activity (specific activity 0-25). This was material was eluted in 30 min with a linear gradient from 0.045 M NaCl to 0.5 M NaCI; 2-ml fractions were collected. The inset shows followed bya sharp increase, with specific activity reaching a maximum of 75 at greater than 200 pM Ca2+. the SDS-PAGE analysis of the active fractions.

Glycosyl-Phosphatidylinositol-specific A

Froellon Numbw

B

- GPI-PLD 46

-

3-

t3

w

1s

10

17

21

22

Fmctknr

FIG. 3. Gel filtration of the GPI-specific phospholipase D. 5.4 pg of purified GPI-specific phospholipase D in 200 pI (1780 units) were chromatographed on a TSK-3000 S W column (flow rate: 0.4 ml/min); the column was equilibrated in 20 mM bis-Tris-HCI, pH 6.5.0.15 M NaCI, and 2.6 mM CaCI2. Fractions (0.4 ml) were collected and 1 pl of each fraction assayedfor activity as described under “Experimental Procedures.” Fractions were also subjected to SDSPAGE (7%) under reducing conditions and then silver-stained. A monoclonal IgGl from mouse and BSA were used as gel filtration standards. A, activityandabsorbance profile; B, SDS-PAGE of selected fractions.

Phospholipase D

13763

of enzymatic activity during this purification step suggested the presenceof an inhibitor(s) in theflow through. However, reconstitution experiments of the purified enzyme with flowthrough fractions from the phenyl-5PW column which contained no enzymatic activity failed to indicate thepresence of any significant inhibitor of enzymatic activity. Although all fractions eluted from the phenyl column by H 2 0 contained activity, we cannot exclude the possibility that an inhibitory molecule remained bound to the column under these elution conditions. Since it binds toa wheat germ lectin column and isspecifically eluted with GlcNAc, the GPI-specific phospholipase D is a glycoprotein, probably containing N-acetylglucosaminyl residues. The flow-through activity observed in step 3 (Fig. 1B) is due to saturation of the column, because in other of enzymatic activitywas achieved experiments, 100% binding by increasing thebed volume of the lectincolumn. The physiological role of the plasmaGPI-specific phospholipase D in the metabolism of GPI-anchored membrane proteins remains tobe determined. The GPI-specific phospholipase D does not release DAF or alkaline phosphatase’ from intact blood cells; that is, the enzyme does not act on the proteins when they are anchored in the externalleaflet of the cell membrane. Another possibility is that the GPI-specific phospholipase D is a cellular enzyme and that the plasma product represents shed material. The broad range of pH over which the GPI-specific phospholipase D is active would allow the enzyme to operate in an intracellular compartment such as an endosome where the pH is between 5.0 and 6.0 (24). Intracellular processing of GPI-anchored proteins has been proposed for the trypanosoma1 mfVSG, where the GPI-specific phospholipase C appears to belocated in specialized vesicles below the plasma membrane(25). The purification of the plasma GPI-specific phospholipase D should greatly facilitate thedevelopment of the necessary tools, e.g. polyclonal and monoclonal antibodies aswell as the cDNA encodingfor the gene, with which to study the enzyme’s role in the metabolismof GPI-anchored proteins.

DISCUSSION

Acknowledgments-We would like to thank Drs. Stephanie DiWe report here the purification tohomogeneity of a GPIment, Paul Englund, Jessica Krakow,Kyoko Iida, Victor Nussenspecific phospholipase D from humanplasma. The conclusion zweig, Jose Rosales, and Victoria Werth for their excellent suggesthat the110-kDa band is theGPI-specific phospholipase D is tions and comments. based on several linesof evidence. The intensity of staining REFERENCES of the 110-kDa band was proportional to the degree of purification of the GPI-specific phospholipase D (Fig. 2, lanes 11. Low, M. G. (1987) Biochem. J. 2 4 4 , 1-13 2. Low, M. G., and Saltiel, A. R. (1988) Science 239, 268-275 4 and Table I). The 110-kDa band co-fractionated with activity in the final Mono Q column (Fig. IC) and with activity in 3. Ferguson, M. A. J., and Williams, A. F. (1988) Annu. Rev. Biochem. 5 7 , 286-320 the gel filtration column (Fig. 3). Both gel filtration and SDS- 4. Ferguson, M.A. J., Homans, S. W., Dwek, R. A., and Rademacher, PAGE are consistent with a monomeric molecule of 110 kDa. T. W. (19M) Science 2 3 9 , 753-759 Low and Prasad (16) stated that the GPI-specificphospholi5. Ferguson, M. A. J., Low, M. G., and Cross, G . A. M. (1985) J . Biol. Chem. 260,14547-14555 pase D from rabbit and dog plasma had a M , of 500,000, 6. Thomas, J., Webb, W., Davitz, M. A., and Nussenzweig, V. (1987) however, they suggested that this may have been an aggregate Riophys. J. 51,522a form of the enzyme (16). The GPI-specific phospholipase D 7. Ishihara, A., Hou, Y., and Jacobson, K. (1987) Proc. Natl. Acad. is active over a broad p H range, 4.0-8.5, with significantly Sci. U. S. A. 84, 1290-1293 greater levels of activity between pH values 4.5 and 8.0. 8. Noda, N., Yoon, K., Rodan, G . A., and Koppel, D. E. (1987) J. Cell Biol. 1 0 5 , 1671-1677 Although the GPI-specific phospholipase D does appear tobe dependent, in part, onCa2+ foractivity, the fact that activity 9. Hereld, D., Krakow, J. L., Bangs, J. D., Hart, G. W., and Englund, P. T.(1986) J. Biol. Chem. 2 6 1 , 13813-13819 could be inhibited by 1,lO-phenanthroline suggests that other 10. Bulow, R., and Overath, P. (1986) J . Biol. Chem. 2 6 1 , 11918metal ions are also important for activity. The GPI-specific 11923 phospholipaseD exhibitsidenticalsubstrate specificity as 11. Fox, J. A., Duszenko, M., Ferguson, M. A. J., Low, M. G., and Cross, G. A. M. (1986) J. Biol. Chem. 2 6 1 , 15767-15771 compared with the crude enzyme. Purified membrane DAF 12. Andrews, N. W., Hong, K. S., Robbins, E. S., and Nussenzweig, also serves asa good substrate. V. (1987) Exp. Parasitol. 64,474-484 The flow-through activity observed in the phenyl step is 13. Schenkman, S., Yoshida, N., and Cardoso de Almeida, M. L. due to overloading of the column, since complete bindingof (1988) Mol. Biochem. Parasitol. 2 9 , 141-152 activity could be achieved by decreasing the total amountof protein loaded onto thecolumn. The observed 100% recovery * M. G. Low, personal communication.

13764 14.

D

Glycosyl-Phosphatidylinositol-specific Phospholipase

FOX,J. A,, Soliz, N. M., and Saltiel, A. (1987) Proc. Natl. Acad. Sci. U. S. A . 84, 521-534 2663-2667

s.,

1.5. Davitz, M. A,, Hereld, D., Shak, Krakow, J., Englund, p. T., and Nussenzweig, V. (1987) Science 2 3 8 , 8 1 - 8 4 16. Low, M. G., and Prasad, A. R. (1988) Proc. Natl. Amd. Sci. S. A . 85,980-984 17. Cardoso de Almeida, M. L., Turner, M. J., Stambuk, B. V., and Schenkman, S. (1988) Biochem. Biophys. Res. Commun. 1 5 0 , 476-482 18. Krakow, J. L., Hereld, D., Bangs, J. D., Hart, G. W., and Englund, P. T. (1986) J . Biol. Chem. 261,12147-12153 19. Masterson, W. J., Doering, T. L., Hart, G. W., and Englund, P. T . (1989) Cell, in press

s.

u.

20. Lanham, S . M., and Godfrey, D. G. (1970) Exp. Par&tol. 2 8 , 21. Ryu, S. H., Cho, K. S.,Lee, K. Y., Suh, P. G., and Rhee, S . G. (1987) J . Biol. Chem. 2 6 2 , 12511-12518 22. H ~ c. y.,~ chau,~ v., chock, ~ , p. B., wang, J. H., and sharma, R. K. (1981) Proc. Nutl. Acad. Sci. U. S. A . 78,871-874 23, Kinoshita, T,, Medof, M. E,, Silber, R., and Nussenzweig, v. (1985) J. Exp. Med. 1 6 2 , 7 5 - 9 2 24. R. G. and Orci9 L. (1988) J . BioL. 539543 25. Bulow, R., Griffiths, G., Wehster, P., Stierhof, Y.-D., Opperdoes, F. R., and Overath, P. (1989) J. Cell. Sci., in press

w.*