PROTEINS Structure, Function, and Genetics 9229-239 (1991)


Phospholipase A2 at the Bilayer Interface Fausto Ramirez' and Mahendra Kumar Jain' 'Department of Chemistry, SUNY, Stony Brook, New York 11 794, and 'Department

of Chemistry and

Biochemistry, University of Delaware, Newark, Delaware 19716

ABSTRACT Interfacial catalysis is a necessary consequence for all enzymes that act on amphipathic substrates with a strong tendency to form aggregates in aqueous dispersions. In such cases the catalytic event occurs at the interface of the aggregated substrate, the overall turnover at the interface is processive, and it is influenced the molecular organization and dynamics of the interface. Such enzymes can access the substrate only at the interface because the concentration of solitary monomers of the substrate in the aqueous phase is very low. Moreover, the microinterface between the bound enzyme and the organized substrate not only facilitates formation of the enzymesubstrate complex, but a longer residence time of the enzyme at the substrate interface also promotes high catalytic processivity. Binding of the enzyme to the substrate interface as an additional step in the overall catalytic turnover permits adaptation of the Michaelis-Menten formalism as a basis to account for the kinetics of interfacial catalysis. AS shown for the action of phospholipase A, on bilayer vesicles, binding equilibrium has two extreme kinetic consequences. During catalysis in the scooting mode the enzyme does not leave the surface of the vesicle to which it is bound. On the other hand, in the hopping mode the absorption and desorption steps are a part of the catalytic turnover. In this minireview we elaborate on the factors that control binding of pig pancreatic phospholipase A, to the bilayer interface. Binding of PLA2 to the interface occurs through ionic interactions and is further promoted by hydrophobic interactions which probably occur along a face of the enzyme, with a hydrophobic collar and a ring of cationic residues, through which the catalytic site is accessible to substrate molecules in the bilayer. An enzyme molecule binds to the surface occupied by about 35 lipid molecules with an apparent dissociation constant of less than 0.1 pM for the enzyme on anionic vesicles compared to 10 mM on zwitterionic vesicles. Results at hand also show that D


aggregation or acylation of the protein is not required for the high affinity binding or catalytic interaction at the interface. Key words: phospholipaseA,, interfacial catalysis, interfacial activation, resonance energy transfer, lipid-protein interaction INTRODUCTION Interfacial catalysis on biomembranes is an intriguing biophysical phenomenon. Recently it has attracted considerable attention because such processes are believed to be responsible for cellular regulatory mechanisms in which soluble proteins act on the membrane localized substrates. Several classes of proteins are known to be functionally active at interfaces; examples include lipolytic enzymes, acyltransferases, protein kinases, and glycosidases. Interfacial catalysis by lipolytic enzymes is involved in such diverse processes as digestion of fats to tailoring of membrane lipids and modulation of signal t r a n s d u ~ t i o nA . ~resurgence ~ of interest in phospholipase A, (PLAB)has been sparked by the possibility that the release of arachidonate and lysophospholipids from membrane phospholipids is the rate-limiting step in the biosynthesis of eicosanoids (prostaglandins, thromboxanes, leukotrienes, lipoxins) and platelet activating factor. These regulatory molecules have been implicated in the onset and control of a wide range of physiological and pathological states such as inflammation, asthma, ischaemia,

Received April 13, 1990; revision accepted September 2, 1990. Address reprint requests to Dr. Mahendra Kumar Jain, Department of Chemistry and Biochemistry, University of Delaware, Newark, DE 19716. Abbreviations: DMPC, dimyristoyl glycero-sn-3-phosphocholine; DTPC, ditetradecyl glycero-sn-3-phosphocholine; DMPM, dimyristoyl glycero-sn-3-phosphomethanol;DTPM, ditetraHDNS, dansyl-hexadecyldecyl glycero-sn-3-phosphomethanol; phosphoethanolamine; NK-529, 1,3,3,1',3',3'-hexamethylindocarbocyananine; PLA2, phospholipase A, from pig pancreas; proPLA2, prophospholipase A, from pig pancreas.








HOIJ 2 t

0Fig. 1. The hydrolytic reaction catalyzed by phosphciipase A,. DMPM is R = R' = C13H2,; R" = CH,.


toxic shock, psoriasis, pancreatitis, burn trauma, and rheumatoid arthritis.


MOLECULAR CHARACTERISTICS OF PLAQ PLA2s are ubiquitous, and the enzymes from venoms and pancreas are abundant, stable, and biochemically well characterized. These water-soluble proteins have about 120 amino acids in a single peptide chain with a rigid three-dimensional structure stabilized by several disulfide bridges. All PLA2s have an absolute requirement for calcium, and His48 is the catalytic site residue that participates in hydrolysis by a general base mechanism. The enzymes from venoms and pancreas have highly conserved regions and exhibit over 60%homology.36XRay ~ r y s t a l l o g r a p h y ~ ,has ~ , ~revealed ~ , ~ ~ a common three-dimensional architecture in which 5 of the 7 disulfide bridges are also conserved. Certain intracellular PLA2s expressed from cDNA also exhibit a sequence homology with secretory P L A ~ s ,al~~ though a significant departure in their catalytic behavior is apparent.

CATALYTIC CHARACTERISTICS AT THE INTERFACE In this review we focus on PLA2 from pig pancreas which has served as a prototype for the study of interfacial catalysis. It catalyzes the hydrolysis of the sn-2-ester bond in 1,2-diacyl-sn-3-phosphoglycerides (Fig. 1).Compared to a catalytic turnover number of less than 30 per min for PLAB with solitary monomeric phospholipid molecules with short acyl chains, under optimum conditions the turnover number exceeds 25,000 per min for aqueous dispersions of long chain substrates. In order to appreciate the increase in the catalytic turnover at the interface it is necessary to invoke the binding of PLA2 to the interface as shown in Figure 2. In this review we focus on the factors that govern the binding step (E to E*),whereas a detailed analysis of the catalytic steps in the interface will be published elsewhere. The molecular organization and dynamics of amphipathic molecules like phospholipids at an interface in aqueous dispersions depend on several factors,2,22and such factors ultimately control interfacial catalysis by PLA2. The hydrophobic effect provides the driving force for amphipathie molecules in water to form organized structures such as bilayers, monolayers, micelles, and emulsions. The cylindrial



I t

E Fig. 2. The minimal Scheme I for interfacial catalysis by PLA2. Additional steps can be incorporated to account for binding of the bound enzyme (E') to products or inhibitors. The species shown in the box plane are bound to the bilayer, where as the enzyme in the aqueous phase is shown as E. The values of the rate constants shown in the scheme are kb = 4 s ~ - ' ;kd = 99.99% in favor of E*,18 i.e., the dissociation constant for E* is ~ 0 . pM. 1 For example, hydrolysis of 1,2-dimyristoyl glycero-sn-3phosphomethanol (DMPM) vesicles by pig pancreatic PLAB exhibits kinetic characteristics which show that the enzyme binds rapidly and essentially irreversibly. As shown in Figure 3, hydrolysis of DMPM vesicles starts immediately after the addition of PLA2. In the presence of excess enzyme 50 to 70% of the total substrate is hydrolyzed, corresponding to the fraction of the substrate present on the outer monolayer of the vesicles depending on their size. On the other hand, in the presence of an excess of small sonicated vesicles (more than five vesicles per enzyme to assure that there is a t most one E per enzyme containing vesicle) the reaction stops when 4,400 substrate molecules have been hydrolyzed per enzyme. These observations are best interpreted in terms of the diagram shown in Figure 4. Here, only the substrate on the outer surface of the target vesicles is accessible to the enzyme, and the bound enzyme does not exchange with excess vesicles. Two kinds of experiments demonstrate that the enzyme is fully active at the end of the reaction progress curve. As shown in Figure 3, addition of salt promotes the desorption of the enzyme from vesicles so that the enzyme molecules hop from vesicle to vesicle, and ultimately hydrolyzes all substrate molecules in the outer monolayer of all vesicles in the

Fig. 4. A schematic drawing to illustrate key features of interfacial catalysis on phospholipid vesicles in the (top) scooting and (bottom) hopping mode. In the scooting mode, when the vesicle to enzyme ratio is more than 5, there is at most one enzyme per vesicle. With KO i0.1 pM for E' on DMPM vesicles, the bound enzyme does not leave the vesicle even when all of the substrate in the outer monolayer of the target vesicle is hydrolyzed. Therefore, excess vesicles are not hydrolyzed by the enzyme added initially unless the vesicles are allowed to fuse, or the bound enzyme undergoes intervesicle exchange, or the excess vesicles are hydrolyzed by adding excess enzyme so that there is at least one enzyme per vesicle. On the other hand during catalysis in the hopping mode, the enzyme desorbs from the vesicle surface between the catalytic cycles, and thus all vesicles are ultimately hydrolyzed even if the vesicle to enzyme ratio is >> 1.

reaction mixture. In other experiments, fusion of vesicles induced (e.g., by calcium, polymyxin B) at the end of the reaction progress curve (Fig. 3) promotes hydrolysis of excess vesicles by making them accessible to the bound enzyme. The fact that only the substrate in the outer monolayer of the target vesicles is hydrolyzed shows that the binding and catalysis by PLAP does not require solubilization of phospholipids, or the formation of nonbilayer phases, e.g., hexagonal or micellar. Other factors that influence the time course of interfacial catalysis on anionic vesicles also merit consideration because in this system the E to E* equilibrium is essentially completely in favor of E*: (1) The number of substrate molecules hydrolyzed by an enzyme molecule is the number of substrate molecules in the outer monolayer of the target vesicle, and the extent of hydrolysis per enzyme increases with the size of vesicles. (2) As expected, the polydispersity in the size of vesicles influences the shape of the reaction progress curve. (3) The reaction progress curves of type shown in Figure 3 have been observed with well over 50 PLA2s from differ-



ysis. A primary consideration is that binding be ent sources, their isozymes, and mutant forms. (4) observed only under conditions that support catalyAs expected, in such cases, the extent of hydrolysis sis, although catalysis is not necessary for the bindper enzyme (in the presence of excess vesicles) reing. Thus the catalytic significance of binding of mains essentially the same for enzymes from differPLA2 to vesicles of DTPM is demonstrated by obserent sources, which shows that a single PLAB molevations such as binding requires calcium;12binding cule on a vesicle is catalytically active. (5) It may be is observed only with anionic but not with zwitteremphasized here that the catalysis in the scooting ionic bilayers unless the products of hydrolysis are mode does not exhibit any anomalous kinetic behavalso present;" the binding characteristics of proior at the gel-fluid thermotropic transition,21v23or PLAB are different;21 the bilayer organization reduring isothermal phase transition induced by limains intact on binding of PLA2; only phospholipid pophilic solutes.27Such observations show that the molecules in the outer monolayer are accessible for anomalous effects observed under comparable conand h y d r o l y ~ i s ; ~the ~,'~ rate of intervesicle ditions with zwitterionic v e s i ~ l e s ~ (also , ~ ~ see , ~ ~ , ~binding ~ exchange of the enzyme bound to DTPM or DMPM below) are due to a shift in the E to E* equilibvesicles is very and the kinetic rate conrium.16 stants and the equilibrium dissociation constant for Characteristic features of interfacial catalysis are the E to E* step obtained from direct equilibrium best observed in the scooting mode where an enzyme binding, stopped-flow, and the kinetic measuremolecule bound to a vesicle "scoots" a t the interface ments are comparable. As elaborated below such oband hydrolyzes all the available substrate in the servations provide information about the enzymeouter monolayer of the target vesicle (Fig. 4). As bilayer microinterface, i.e., the surface of PLAB that shown elsewhere25 the reaction progress curves of interacts with the substrate interface (Fig. 5). Bindtype shown in Figure 3 are a direct consequence of ing of PLAB to DTPM vesicles causes a 2-fold inthe high affinity binding of PLAB to vesicles where crease in the quantum yield of the fluorescence the enzyme, substrate, and products do not exhibit emission from the only tryptophan residue (Trp-3) any intervesicle exchange even when all the subon the protein. The resonance energy transfer to a strate in the outer monolayer of the target vesicle dansyl fluorophore (HDNS) in the polar head group has been completely hydrolyzed. We have obtained region of the bilayer occurs with >90% efficiency, values of interfacial equilibrium and kinetic rate suggesting that the Trp-3 to dansyl distance is less constants (Fig. 2), and through these it is possible to than 10 A.21 As elaborated below, such an increase appreciate some of the otherwise inaccessible feain the fluorescence emission from Trp-3and dansyl tures of interfacial catalysis, such as the shape of the fluorophores is useful for obtaining kinetic and therreaction progress curve, activation by calcium and modynamic information about the E to E* step. other agents, substrate specificity,26specific comStopped-flow show that the association petitive i n h i b i t i ~ nand , ~ ~the anomalous kinetics a t rate constant, kb, for the E to E* step has two comthermotropic gel-fluid or isothermal phase transiponents: a second-order diffusion limited step, foltion in zwitterionic bilayers.1,12'15,16,24,42 lowed by a first order rate constant of 4 sec-l. The BINDING OF PLAQTO BILAYER VESICLES value of kb for the micellar interface are the same as for anionic b i l a y e r ~The . ~ ~upper limit for the dissoBinding of PLAB to bilayers of nonhydrolyzable ciation rate constant, kd, is estimated as

Phospholipase A2 at the bilayer interface.

Interfacial catalysis is a necessary consequence for all enzymes that act on amphipathic substrates with a strong tendency to form aggregates in aqueo...
1MB Sizes 0 Downloads 0 Views