Platelet activating factor-acetylhydrolase in insectsIdentification and partial characterization of a 1-alkyl-2-acetyl-sn-glycero-3-phosphocholine acetylhydrolase in a cell-free system of Heliothis.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 237-45 (1988) Platelet Activating Factor-AcetyIhydrolase in Insects: Identification and Partial Characterization of a 1-Alkyl-2-Acetyl-snG l ycero-3-Phosphocholine Acety Ihydrolase in a Cell-Free System of Heliothis Edward N. Lambremont, Boyd Malone, and Fred Snyder Nuclear Science Center, Louisiana State University, Baton Rouge (E. N . L.) and Biological Chemist y Laboratoy, Medical and Health Sciences Division, Oak Ridge Associated Universities, Oak Ridge, Tennessee (B.M., F.S.) A specific acetylhydrolase that inactivates platelet activating factor (PAF; 1alkyl-2-acetyl-sn-glycero-3-phosphocholine),a potent cellular mediator in mammalian cells, b y removal of the sn-2 acetyl moiety, has been found in the cytosolic fraction of several postemhryonir developmental stages and specific tissues of the corn eanvorm, Heliothis zea (Boddie). Effects of magnesium, calcium, EGTA, deoxycholate, dithiothreitol, diisopropylfluorophosphate, egg phosphatidylcholine, and an acylacetyl-glycerophosphocholine show that hydrolysis of the acetate moiety is due t o a specific acetylhydrolase for PAF. The activity does not appear to b e due t o a typical cellular phospholipase A2 that utilizes phospholipid substrates with a long-chain acyl group at position sn-2 o f glycerol. Specific activities and properties of the acetylhydrolase from this insect match closely with those described from tissues of vertebrate animals. Key words: lipid metabolism and enzymes, ether-linked lipids, inactivation of bio-active lipids and cell mediators INTRODUCTION The occurrence of lipids containing an ether bond has been studied extensively since the early 1920s in a wide variety of lower and higher animals [l]. Acknowledgments: The insects used in this study were kindly provided by Dr. J.B. Graves of the Department of Entomology, Louisiana State University, Baton Rouge, and Dr. Ed King, USDA, Stoneville, MS. We thank Dr. M.C. Cabot, Dr. T-c Lee, and M.L. Blank for procedural suggestions, support, and the gift of several lipid compounds, and Yvonne Thomas for technical support. This research was jointly supported by Louisiana State University and Oak Ridge Associated Universities’ Medical and Health Sciences Division and University Programs Division. Work at the Medical and Health Sciences Division was supported by the Office of Energy Research, US Department of Energy (Contract No. DE-AC05-7604000033). Received July 29,1987; accepted December 1,1987. Address reprint requests t o Dr. E.N. Lambremont, Nuclear Science Center, Louisiana State University, Baton Rouge, LA 70803-5820. 0 1988 Alan R. Liss, Inc. 38 Lambremont, Malone, and Snyder Until the late 1960s the biosynthesis of ether-linked lipids was unknown, although their structure had been established. It was not until the late 1960s and into the 1970s that ether-linked glycerolipids came under intense research when they were found to occur in elevated amounts in certain neoplasms and appeared to be associated with biomembranes . Alkyldihydroxyacetone-P synthase, the enzyme that catalyzes the formation of the ether bond in lipids, was discovered in 1969 . Then in 1979, a semisynthetic acetylated alkylglycerophosphocholine was shown to induce potent pharmacologic actions that were identical to an antihypertensive principle isolated from kidney medulla called "APRL" by Muirhead in Blank et al.  and "platelet activating factor," or PAF*, by Benveniste et al.  and Demopoullos et al. . Research interest in the metabolism and pharmacology of PAF has concentrated on higher animals. While this is quite understandable from a medical point of view, it is also of interest in the pursuit of comparative biochemical phylogeny to know whether other animal groups also contain and metabolize such bioactive lipids. Because insects are known to contain ether-linked lipids, including the alkyl-acyl type of glycerophosphocholines [6,q that are precursors of PAF, and because insects readily synthesize ether-linked lipids from other constitutents [8-lo], we decided to determine whether insects contained enzymes involved in the metabolism of PAF as do vertebrate animals. Of several possible choices to begin such a line of investigation, we elected first to determine if a cell-free enzyme system from an insect could inactivate PAF by removal of the acetate grouping. Hydrolysis of the acetate group from the sn-2 position of PAF by an acetylhydrolase produces l-alkyl2-lyso-GPC, which is an inactive molecule when compared with similar concentrations of PAF [ll]. The experiments described in this report demonstrate that insect tissues do indeed contain a PAF acetylhydrolase. MATERIALS AND METHODS Chemicals CHAPS, EGTA, deoxycholic acid (sodium salt), dithiothreitol (Cleland's reagent), DFP, and unlabeled l-hexadecyl-2-acetyl-sn-glycero-3-phosphocholine were from Sigma Chemical Co., St. Louis, MO. L-cd'hosphatidylcholine (egg yolk lecithin) was from Calbiochem-Behring Corp., LaJolla, CA. The 1palmitoyl-2-acetyl-sn-glycero-3-phosphocholine was provided by M. Blank. Its purity was determined just before use to be greater than 99% by TLC on silica gel H in a solvent system of ch1oroform:methanol:ammonium hydroxide (65:35:8). Lobster hemolymph was from Grand Island Biological Co., Grand Island, NY. All other chemicals were of reagent grade quality. Labeled PAF, l-hexadecyl-2-[3H]-acetyl-sn-glycero-3-phosphocholine, was prepared by the method of Blank et al. [ll]; it had a radiochemical purity in *Abbreviations: 3-[(3-cholarnidopropyI) dirnethyl-arnmoniol-I-propanesulfonate = CHAPS; diisopropylfluorophosphate = DFP; ethyleneglycol-bis-(beta-arninoethylether) N, "-tetraace- also shortened to alkylacetyl tic acid = EGTA; l-alkyl-2-acetyl-sn-glycero-3-phosphocholine, CPC = PAF, platelet activating factor. PAF-Acetylhydrolase in Heliothis 39 excess of 99% as shown by TLC on Silica gel H plates in a system of ch1oroform:methanol:glacial acetic acid:water (50:30:8:6). Insects and Enzyme Preparations Insect material was obtained from laboratory stocks reared on artificial diets. Cytosols were prepared from freshly dissected rat kidney and from third instar larvae, mature sixth instar larvae, newly molted (untanned) pupae, tanned pupae, pupal fat bodies, and adult flight muscles of Heliothis zea (Boddie) (Insecta, Lepidoptera: Noctuidae). To each gram of insect material, 9 ml of ice cold 0.25 M sucrose were added. After a brief mincing of the tissues with sterile scissors, the mascerate was homogenized with four to five strokes in a motor-driven Teflon-Pyrex tissue grinder in an ice bath. Cell debris, nuclei, and mitochondria were pelleted at 15,OOOg for 10 min in a refrigerated centrifuge; the pellet was discarded. Microsomes were then pelleted from the supernatant fraction at lO0,OOOg for 60 min, and the supernatant (cytosolic fraction) was used for all enzyme analyses. The cytosols were subdivided into 0.5-ml portions and transferred to ice-cold glass vials (with Teflon-lined screw caps) and immediately frozen (-20°C). In some initial experiments, freshly prepared cytosols were shown to have enzyme activities comparable with those obtained with the frozen samples. Pupal hemolymph was obtained by puncturing the cuticle at an intersegmental suture of a wing cover with a 50-pl glass capillary pipette. The clear hemolymph was easily drawn by capillary action into the pipette, and with very gentle pressure on the sides of the pupa, volumes in excess of 50 p1 per insect were obtainable. Pooled hemolymph samples were kept on ice, diluted with 10 volumes of 0.25 M sucrose, centrifuged very briefly in the cold to sediment fat body cells, and then frozen in 0.5-ml aliquots as above. Postmitochondrial supernatants of fat bodies were prepared after cutting off the head end of the pupa and gently extruding the fat body into cold 0.25 M sucrose. The tissue was gently homogenized as above and centrifuged in the cold at 12,SOOg for 10 min. The supernatant fluid was diluted with 20 volumes of cold 0.25 M sucrose and frozen as above. Protein concentrations in all cytosols and tissue preparations were measured by the Lowry et al. method . Enzyme Assay Enzyme activity was determined by adaptations of the methods of Blank et al. [ll] and Cabot et al. . The complete enzyme reaction system contained l-he~adecyl-2-[~H]-acetyl GPC (10 nmol and 0.25 pCi 3H) and enzyme protein in a final volume of 0.5 ml of 0.1 M phosphate buffer, pH 8.0. Duplicate incubations were done in screw-capped polypropylene tubes for 5 min at 37°C. Protein concentrations of the enzyme preparations in each incubation ranged from 50-350 pg. Labeled acetate released by the active enzyme was compared with 3Hradioactivities in water blanks or heat-denatured cytosols or tissue preparations prepared in sealed screw-capped vials placed for 10 min in a boiling water bath. The precipitated proteins in the boiled preparations were resuspended 40 Larnbremont, Malone, and Snyder by gentle homogenization in a Teflon-Pyrex tissue grinder before use in the assays as blanks. RESULTS AND DISCUSSION Presence of an Alklyacetylhydrolase in Insects Alkylacetyl-GPC:acetylhydrolase activity was found in all insect stages examined and, except for hemolymph, was easily detectable in two specific tissues, namely the fat body and flight muscle (Table 1).While typical values for the insect preparations were approximately 3 to 4 times less than that observed in rat kidneys, it is clear that an enzyme capable of inactivating PAF is present. Typical values for the acetylhydrolase in insects ranged between 0.3 to 1.3 nmol of acetate released per min per mg protein, with mean values of 0.7 for most larval and pupal cytosols. Of three hemolymph preparations tested, two showed no enzyme activity at all, and the other had a very low activity, possibly due to the presence of contaminating fat body cells. Assays of two samples of frozen lobster hemolymph also showed no acetylhydrolase activity. To our knowledge, these findings constitute the first available information on the presence of any enzyme capable of metabolizing PAF for any insect or invertebrate animal. Optimum activity of acetylhydrolase from Heliothis occurred over a broad pH range (6.0 to S.9), with no sharp peak noted. Increases in activity above pH 8.0 were less steep than those at the lower pH levels, which indicates there is a leveling off of activity at higher pH. Activities were linear with TABLE 1. Alkylacetyl-GPC Acetylhydrolase Activities in Cystosolic Fractions at Various Developmental Stages in Tissues of Heliothis as Compared With the Enzyme Activity in Rat Kidnev* Enzyme source Rat Kidney (5) Heliot his Larva Third instar (2) Sixth instar (13) Pupa Untanned (2) Tanned Fat body Hemolymph Adult flight muscle Acetate liberated (nmol acetate min-l protein-*) 2.52 0.93 1.03 f 0.98 0.72 f 0.16 0.57 rt 0.25 0.61 f 0.39 0.93 0.0 f 0.18 0.97 *All preparations were of cytosolic origin except postmitchrondrial supernatants from the fat body and hemolymph; see Materials and Methods. The number of experiments are in parentheses; each experiment represents duplicate incubations. Values are means of the incubations f SD. PAF-Acetylhydrolase in Heliothis 41 protein concentration up to about 370 pg and with incubation time up to 5 min. All subsequent incubations, therefore, were done at pH 8.0 for 5 min and at protein concentrations equal to or below 350 pg per incubation. Acetylhydrolase appears to be quite stable during long-term storage. Activity in cytosolic fractions stored at -20°C for more than 2 years was the same as fresh or freshly frozen preparations. Characterization of Properties of Acetylhydrolase in Insects While finding alky1acetyLGPC:acetylhydrolase in Heliothis is of itself both interesting and significant, it was next necessary to characterize the enzyme's properties in greater detail. Although activity was detected in a variety of developmental stages, most of our work was done with the easily obtainable cytosol from the mature sixth larval instar. A key question was first to determine whether the enzyme was specific for alkylacetyl-GPC or whether it was essentially a phospholipase A2 that could utilize both diacyl phospholipids as well as PAF. To determine such specificity, we evaluated divalent metal ion requirements, chelators, and other factors known to influence the activity of phospholipase A2 from other animal sources [lo, 121. The results are summarized in Table 2. The addition of either Ca2+ or Mg2+ had a slight stimulatory effect. However EGTA, which chelates Ca2+and which would have depressed the activity of a typical cellular phospholipase A2, actually showed a slight acetylhydrostimulation of the l-alkyl-2-acetyl-sn-glycero-3-phosphocholine lase in the cytosol of Heliothis larvae. Dithiothreitol had no effect on the enzyme activity, whereas there appeared to be a slight inhibition (82% of the control) by deoxycholate, which agrees very closely with 80% of control reported by Blank et al. [ll]for the acetylhydrolase from rat liver. Diisopropylfluorophosphate (1mM), a known inhibitor of the acetylhydrolase from rat liver [ll], reduced activity in the Heliothis cytosol to about one-fourth of control values. These combined observations strongly suggest that the properties of the enzyme that hydrolyzes the acetate moiety of PAF are not the same as a typical cellular phospholipase A2. TABLE 2. Characteristics of Alkylacetyl-GPC Acetylhydrolase Activity in the Cvtosolic Fraction of Heliothis (Last Instar Larvae)* Addition None Mg2+ (10 mM) Ca2+ (10m ~ ) Deoxycholate (0.1 mM) Dithiothrietol (1mM) EGTA (2 mM) DFP (1 mM) DFP (10 mh4) Acetate liberated (nmol acetate min-l protein-') 0.78 0.91 0.77 0.64 0.76 0.94 0.21 0.20 % of Control f 0.18 - k 0.08 117 99 82 97 120 27 26 f 0.11 f 0.13 0.09 f 0.05 k 0.16 + 0.11 + *Values are based on two separate experiments of duplicate incubations and represent the mean of the four incubations f SD. Each incubation contained 350 pg protein. 42 Lambremont, Malone, and Snyder Effects of Other Lipids as Potential Substrates for the Acetylhydrolase To confirm the specificity of the enzyme activity further, the effect of egg lecithin (L-alpha-phosphatidylcholine)in the presence of several solvents and solubilizing agents was used as a possible competitive inhibitor of the alkylacetyl-GPC:acetylhydrolase in the cytosol of Heliothis larvae (Table 3). Three of the solubilizing agents (ethanol, acetone, and CHAPS) depressed the enzyme activity, but no further reduction of activity was noted when either 2 or 14 mM egg phosphatidylcholine was included. In fact, slight but probably insignificant stimulation appeared to occur in some instances when egg lecithin was added to the standard incubations. Propyleneglycol, used by Snyder and Pfleger  to solubilize fatty acids, showed a slight stimulation of acetylhydrolase activity. However, when propyleneglycol was used in conjunction with 2 or 14 mh4 egg phosphatidylcholine, the enzyme was unaffected. On the other hand, when a phosphatidylcholine (14 mM) with an acetyl group in the sn-2 position (l-acyl-2-acetyl-sn-glycero-3-phosphocholine) was incubated under the same conditions, the acetylhydrolase activity decreased to approximately one-third of control values. This observation, and the fact that egg phosphatidylcholine did not compete for the enzyme TABLE 3. Effects of Phosphatidylcholine (PC) and Various Solubilizing Agents on the Activity of Alkylacetyl-GPC Acetylhydrolase From Heliothis (Last Larval Instar)* Addition Egg phosphatidylcholine None Ethanol (10 pl) +2 mM PC +14 mM PC Acetone (10 pl) +2 mM PC +14 mM PC None CHAPS (0.05%) +2 mh4 PC +14 mM PC Propylenegly col (0.05%) +2 mM PC +14 mM PC 1-Palmitoyl-2-acet ylglycerophosphocholine None Ethanol (10 pl) +2 mM acylacetyl GPC +14 mM acvlacetvl GPC Acetate liberated (nmol acetate m i d protein-*) 0.81 0.43 0.42 0.46 0.46 0.53 0.50 0.48 0.33 0.40 0.34 0.64 0.44 0.46 f 0.20 O h of Control f 0.03 f 0.07 f 0.08 k 0.07 i 0.01 f 0.01 f 0.13 100 53 52 57 57 65 62 100 69 83 71 133 92 96 1.18 k 0.20 1.06 + 0.44 1.02 f 0.22 0.73 k 0.20 100 90 86 62 0.01 f 0.11 k 0.20 *Values are based on three separate enzyme preparations; results in each subsection of the table relate to control readings for each particular preparation appropriate with additive indicated; each value represents the mean of four incubations 5 SD when shown. Protein contents of the cytosolic fractions were 370, 300, and 230 pg, respectively. PAF-Acetylhydrolase in Heliothis 43 activity, coupled with the failure of EGTA to inhibit and calcium ions to stimulate activity, strongly imply that a phospholipase A2 is not responsible for the release of [3H]-acetatefrom the substrate. Thus the acetylhydrolase in the insect preparations appears to be specific for alkylacetyl GPC (or acylacetyl GPC). Comparison With the Acetylhydrolase of Higher Animals Since these observations constitute the first report of an alkylacetylGPC:acetylhydrolase from insects, comparisons can only be drawn from what is known about the properties of this enzyme obtained from higher animals. Such a comparison reveals a striking similarity between the acetylhydrolase of Heliothis and that found in vertebrates [11,13,15]. For example, a broad pH range was reported for the acetylhydrolase from rat plasma and rat kidney cortex with indistinct optima near 8.0, as is shown here for Heliothis. EGTA was shown to be slightly stimulatory with the acetylhydrolase from rat tissues [ll], and EGTA similarlly stimulated the acetylhydrolase in insects to the same extent, as reported here. Ca2+ and Mg2+ ions had essentially no effect on the enzyme activities in rats, although with insect preparations we observed a very slight stimulation of the acetylhydrolase with Mg2+ only. We also noted that inhibition of acetylhydrolase activity by deoxycholate in the insect system was essentially equal to that reported for the rat enzyme preparations. Similarly, DFP, which inhibits the acetylhydrolase in rat liver to about 9% of control (1.1mM) [ll],inhibited the acetylhydrolase of Heliothis to 27% of control (1.0 mM).Phosphatidylcholine does not appear to compete as a substrate with the alkylacetyl-GPC, because the acetylhydrolase activities for the variety of insect tissues and at different developmental stages reported here overlap broadly with those reported in the literature for sera and plasma from rat [ll], reptile , or fish  and for cytosolic fractions from liver and kidney of rats [ll], rabbit platelets and rat alviolar macrophages , and human neutrophils and eosinophils [17,18]. Our results with insects differed from those reported for acetylhydrolase in mammals only in three ways, namely, the slight Mg2+ ion stimulation, the lack of any stimulatory effect by dithiothreitol, and the slight inhibitory effect of ethanol. These differences are very minor and probably not of any significance. We must conclude that Heliothis has a specific alkylacetyl-GPCacetylhydrolase (EC 126.96.36.199) in a variety of its cellular systems similar to the enzyme found in higher animal sources . Finding a specific alkylacetyl-GPC:acetylhydrolase in Heliothis therefore suggests that insects might contain a compound similar or identical to the potent alkylacetyl glycerophosphocholine (or PAF) mediator found in higher animals . What role such a biologically active ether-linked lipid might have in insects if found is, of course, unclear. Because "platelet activating factor" is known to possess a diverse number of potent biologic activities in higher animals  (e.g., the induction of platelet and neutrophil aggregation, degranulation reactions, hypotension, hypersensitivity, and the cellular release of other bioactive agents), it is conceivable that PAF has an even more widespread role as a cellular mediator throughout the animal kingdom. 44 Lambremont, Malone, and Snyder While the presence of the acetylhydrolase in Heliothis merely infers the presence of PAF in this species, it is known that this and other insect species contain ether-linked phosphoglycerides and that they possess metabolic pathways necessary to provide at least the ether-lipid intermediates required for PAF synthesis. For example, Heliothis has been shown to contain an endogenous pool of alkylacetyl-GPC in all of its postembryonic developmental stages [6,7l. Heliothis is also capable of incorporating various radiolabeled precursors, such as acetate, glycerol, long-chain fatty acids, and longchain fatty alcohols into the alkylacetyl-GPC pool 18/91;in addition, it is able to synthesize from fatty acids the fatty alcohol precursor needed to form the ether bond at the sn-1 position of glycerolipids [lo]. Heliothis and other insect species could be interesting experimental models to investigate the presence and synthesis of PAF compounds in nonvertebrate animals and to determine if they have physiologic or pharmacologic roles in these species. Even if such acetylated phospholipids do not occur, the significance of a specific alkylacetyl-GPC acetylhydrolase as described by us raises intriguing questions about the possible role of this enzyme in Heliothis. Obviously the phylogenic significance of ether-linked lipids and their metabolism in insects warrant further study. LITERATURE CITED 1. Snyder F (editor): Ether Lipids: Chemistry and Biology. Academic Press, New York, 433 PP (1972). 2. Snyder F, Lee T-c, Wykle RL: Ether-linked glycerolipids and their bioactive species: Enzymes and metabolic regulation. In: The Enzymes of Biological Membranes. Martonosi, AN, ed. Plenum Publishing Co., New York, Vol. 2, pp 1-58 (1985). 3. Blank ML, Snyder F, Byers LW, Brooks B, Muirhead EE: Antihypertensive activity of an alkyl ether analog of phosphatidylcholine. Biochem Biophys Res Commun 90, 1194 (1974). 4. Benveniste J, Henson PM, Cochrane CG: Leucocyte-dependent histamine release from rabbit platelets. The role of IgE, basophils, and platelet-activating factor. J Exp Med, 136, 1356 (1972). 5. Demopoullos CA, Pinckard RN, Hanahan DJ: Platelet-activating factor: Evidence for 1-0alkyl-2-acetyl-sn-glycero-3-phosphocholine as the active component (a new class of lipid chemical mediators). J Biol Chem, 254, 9355 (1979). 6. Lambremont EN: Ether-bonded lipids of insects: A quantitative comparison of the glyceryl ethers associated with ethanolamine and choline phosphoglycerides. Comp Biochem Physiol, 41B, 337 (1972). 7. Lambremont EN, Wood R: Glyceryl ethers in insects: Identification of alkyl and alk-1-enyl glyceryl ether phospholipids. Lipids 3, 503 (1968). 8. Lambremont EN: The in vivo synthesis of acyl- and ether-bonded phospholipids in Heliothis virescens. J Insect Physiol, 18, 581 (1972). 9. Lambremont EN: Glyceryl ethers in insects: Biosynthesis of ethanolamine phosphoglycerides containing alkyl and alk-1-enyl glyceryl ether linkages. Lipids 7, 528 (1972). 10. Lambremont EN: Lipid metabolism of insects: Interconversion of fatty acids and fatty alcohols. Insect Biochem, 2, 197 (1972). 11. Blank ML, Lee T-c, Fitzgerald V, Snyder F: A specific acetylhydrolase for 1-alkyl-2-acetylsn-glycero-3-phosphocholine(a hypotensive platelet activating lipid). J Biol Chem, 256, 175 (1981). 12. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem, 293, 265 (1951). 13. Cabot MC, Faulkner LA, Lackey RJ, Snyder F: Vertebrate class distribution of l-alkyl-2acetyl-sn-glycero-3-phosphocholineacetylhydrolase in serum. Comp Biochem Physiol, 78B, 37 (1984). PAF-Acetylhydrolasein Heliothis 45 14. Snyder F, Pfleger RC: Metabolism of alpha-alkoxy glyceryl monoethers in rat liver, in vivo and in vitro. Lipids I, 328 (1966). 15. Snyder F, Lee T-c, Blank ML, Cabot MC, Malone B, Albert DH: Enzymatic pathways for platelet activating factor. In: Platelet-Activating Factor and Structurally Related Ether Lipids. Benveniste J, Arnoux B, eds. Amsterdam, The Netherlands, Elsevier Scientific Publs. 253 pp (1983). 16. Lenihan DL, Greenberg D, Lee T-c: Involvement of platelet activating factor in physiological stress in the lizard, AnoZis curolinesis. Comp Biochem Physiol, 81C, 81 (1986). 17. Lee T-c, Malone B, Wasserman SI, Fitzgerald V, Snyder F: Activities of enzymes that metabolize platelet-activating factor (l-alkyl-2-acetyl-sn-glycero-3-phosphocholine) in neutrophils and eosinophils from human and the effect of a calcium ionophore. Biochem Biophys Res Commun, 105, 1303 (1982). 18. Lee T-c, Snyder F: Function, metabolism and regulation of platelet activating factor and related ether lipids. In: Phospholipids and Cellular Regulation. Kuo FJ, ed. CRC Press, Inc., Boca Raton, FL, Vol. 2, pp 1-39 (1985). Platelet activating factor: A biologically active phosphoglyceride. Annu Rev 19. Hanahan DJ: Biochem 55, 583 (1986).