Lipids as hormones and second messengers

Lipids as hormones and second messengers

Lipids as hormones and second messengers Alfred H. Merrill Jr and Dennis C. Liotta Emory University, Atlanta, Georgia, USA Some lipids act as hormones...

2MB Sizes 26 Downloads 109 Views

Lipids as hormones and second messengers Alfred H. Merrill Jr and Dennis C. Liotta Emory University, Atlanta, Georgia, USA Some lipids act as hormones to induce responses in target tissues whereas others influence the behavior of cell-surface receptors and/or serve as second messengers; i.e. they are formed as part of the cellular response to extracellular agonists, and directly affect protein kinases, ion channels and other systems that govern cell behavior. Our understanding of these functions on a structural level is improving, but this is still a relatively new, and an extremely complex, area of investigation. Current Opinion in Structural Biology 1991, 1:516-521

Introduction Lipids participate in cellular structure and function at essentiaUy all levels, which probably accounts for the diversity of the lipid classes and subtypes. As stated by Thudichum [1] over a century ago, 'Their physical properties are, viewed from a teleological point of standing, eminently adapted to their functions.' This has proven to be true at at least three levels. Firstly, the molecular features of lipids allow them to define membranes, lipoproteins and other biological structures, as discussed elsewhere in this issue by Reiss-Husson (pp 506-509). Secondly, many lipids also embody structural domains that interact with receptors, enzymes and other components of cell membranes as well as with extrinsic cytoskeletal proteins, antibodies, microbial toxins, etc. (some aspects of this topic are discussed in this issue by Ferguson, pp 522-529). Finally, a growing number of lipids can be converted to products that are highly bioactive as hormones and lipid second messengers. When one considers the iterative fine-tuning that proteins have undergone over the course of evolution, it should not be surprising that the lipid components of biological membranes have also evolved to exhibit similar sophisticated properties.

Overview of lipid hormones and second-messenger systems As shown in Fig. 1, there are a large number of compounds with proposed functions as lipid second messengers, and we probably know only a fraction of the intraceIIular systems that are controlled by these types of mechanisms. In most instances, the formation of the lipid hormones (e.g. platelet-activating factor, the eicosanoids, etc.) and lipid second messengers (e.g. diacylglycerols (DAGs), phosphatidic acid (PA), lysophospholipids, ceramides, etc.) proceeds from the cleavage of more complex membrane lipids by phospholipases

[2 oo] that have been activated by the relevant agonist. Nearly all of the possible products of these cleavage reactions (i.e. phospholipid headgroups, lysophospholipids, DAGs, PA and fatty acids) have been implicated in signal transduction. For example, release of lysophospholipids and fatty acids via phospholipase A2 provides compounds that serve as precursors for platelet-activating factor, prostaglandins, leukotrienes, thromboxanes and other eicosanoids, and as possible modulators of the activity of protein kinase C (PKC). Phospholipases C and D produce DAGs, alkylacylglycerols and PA from diverse phospholipid precursors (including phosphatidylinositols (PIs), phosphaticlylcholines (PCs) and PI glycanlinked proteins inter alia). These products can have multiple intracellular targets, such as. the activation of PKC by DAGs and the stimulation of Ca2 + release from intracellular stores by inositol triphosphate, as shown in Fig. 1. Bioactive lipids can be made in multiple intracellular sites; in Fig. 1, this is illustrated by the formation of steroid hormones but this can also occur, at least theoretically, for most of the other types of lipid second messengers. In addition to these phospholipases for glycerolipids, the hydrolysis of sphingomyelin to ceramides has recently been proposed as another lipid second-messenger cycle [2",3]. Other hydrolysis products of sphingolipids (e.g. sphingosine and some lysosphingolipids) are potent inhibitors of PKC [4°], Na+/K + -ATPase, PA phosphohydrolase, and other phospholipases, as well as activators of the epidermal growth factor receptor; hence, the hydrolysis of sphingolipids to bioactive constituents may prove to be widespread and important [5°]. The multiplicity of these pathways, and the fact that most of the enzymes, substrates and products are membraneassociated, make it difficult to understand these systems on a molecular level. Nonetheless, recent findings have provided a more detailed picture of the interactions between lipid second messengers and their targets, and some of these lessons will be discussed using PKC as a model.

Abbreviations DAG~cliacylglycerol; PA~hosphatidic acid; PC--phosphatidylcholine; PG~phosphatidylglycerol; PKC--protein kinase C; Pl--phosphatidylinositol; PMA--phorbol 12-myristate,13-acetate; PS~phosphoatidylserine.

516

(~) Current Biology Lid ISSN 0959-440X

tipids as hormones and second messengers Merrill Jr and Liotta

PI-glycanlinkedproteins

I(viaStimuli I /N~ DAG ) ) ~

~?EDAAG(~G

)) _.~_._

!O~

S~Y hidr~Og~j ipisd

I ,,o,o,o I VpAF ~

Arachidonic acid~ Prostagl.~ins'etc"~2"

C esters oe ,eo Cholesterol

Ceua responses

Fig. 1. Overview of hormone and signal-transduction pathways involving lipid mediators. Binding of stimuli (hormones, growth and differentiation factors, etc.) to membrane receptors triggers a sequence of events that usually involves cleavage of a membrane lipid to a bioactive fragment or derivative. Examples of the lipases that are involved in these events are as follows. Phospholipase A 2 (PLA 2) releases a (usually polyunsaturated) fatty acid (FA, which may proceed to other bioactive products such as prostaglandins, etc.) from the 2-position of a phosphoglycerolipid; in some cells, lysoalkylphosphatidylcholine (lysoalkyI-PC) undergoes reacylation to produce platelet-activating factors (PAF). Phospholipase C (PLC) cleaves phosphoglycerolipids (such as phosphatidylinositol 4,5-bisphosphate, PIP2) to diacylglycerol (DAG) or 1-alkyl- or 1-alkenyl-2-acylglycerols (EAG); DAG activates protein kinase C (PKC) and EAG may activate or inhibit this enzyme. When the substrate is PIP2, release of inositol triphosphate (IP3) increases cytosolic Ca2+. Phospholipase D (PLD) cleaves phosphoglycerolipids to phosphatidic acid (PA) which may act directly, or be converted to DAG to stimulate cell responses. Other classes of PLC and PLD act on phosphatidylinositol-glycan-linked proteins to release the protein and DAG or EAG. Membranes also contain sphingolipid (SL) hydrolases, such as sphingomyelinase and ceramidase that can liberate ceramide (Cer) and/or sphingosine (So). Other signal transduction elements in cells include the enzymes that hydrolyse cholesterol esters to cholesterol then.to various steroid hormones. In some instances, the lipids act on systems in the same cell (i.e., are second messengers); in others (which have not been distinguished for simplicity), they act on targets elsewhere and are termed hormones. Other abbreviations: PS, phosphatidylserine.

Protein kinase C as a model for lipid modulation of a signal-transduction system Protein ldnase C is actually a family of enzymes, therefore caution is to be recommended in making generalizing statements about this system. Nonetheless, the studies reported thus far illustrate many structural considerations that should be pertinent to this and other lipid-modulated signal-transduction systems [6°°]. Firstly, the activity is affected both by phospholipids (phosphatidylserine (PS)) and by phospholipase cleavage products acting as activators (DAGs and unsaturated fatty acids) and inhibitors. Secondly, the phorbol-ester-binding domain

(regarded as the lipid-activating domain of PKC) is located in the amino-terminal region - - hence, activation is thought to involve a conformation change that displaces a pseudosubstrate from the active site. Thirdly, PKC occurs in multiple intracellular sites so that the localized formation of activators (and inhibitors) and enzyme translocation may be important factors in the biological responses. Finally, PKC activity is also influenced by other factors such as the availability of Ca 2 +, the extent of oxidation of key sulfhydryl groups, the conformation of PKC upon insertion into membranes (some forms are less sensitive to lipid modulators), down-regulation by proteases, and many other factors.

517

518 Upids Structure-function studies of lipid modulators of PKC have adopted two broad approaches. The first has compared the structures of analogs of phorbol esters, DAGs and other potent activators (and inhibitors) of PKC. A major conclusion of these studies has been that DAGs overlay well with phorbol 12-myristate,13-acetate (PMA), with properly orientated oxygen atoms located at sites corresponding to C-4, C-9 and C-20 of PMA, and a hydrophobic component located in the proper region of space, i.e. R and R' of PMA and DAG respectively [7,8], as illustrated in Fig. 2. By reducing the alkyt-chain length of phorbol esters (to phorbol dibutyrate), it is possible to measure binding to PKC directly, and KdS of the nanomolar order are typically obtained. A similar strategy has led to the development of short-chain analogs of DAGs (1oleoyl,2-acetylglycerol and, more usefully, m,2-dioctanoylglycerol) as more water-soluble activators of PKC [9]. However, attempts to mimic the potent activation and binding by phorbol esters with glyceride-type molecules have, thus far, been fairly unsuccessful, indicating that the interactions between phorbol esters and PKC go beyond the molecular features of DAGs that are important for biological activity [10].

o

°'W ,3

H

o |

R'~

O O

O~ O

"R' OH

Fig. 2. Comparison of the presumed bioactive conformation of a diacylglycerol (DAG) with phorbol 12-myristate,13-acetate (PMA).

The second approach has been to elucidate the molecular details of PKC activation by lipid mediators encountered under more physiological conditions. This is much more complicated because it requires knowledge of not only the structural features of the lipids (e.g. stereochemistry, alkyt-chain composition, etc.) but also of the physical state of these molecules. The use of mixed micelles by Bell and colleagues [6"'] has simplified these analyses considerably. Figure 3 attempts to illustrate some of the

features that are important in understanding PKC activation at this level.

Structural requirements of lipid modulators of protein kinase C A series of elegant investigations have identified the structural specificity of PKC for DAGs and PSs (reviewed in [6"'] ). Activation requires ester-linked fatty acids at positions 1 and 2 of sn-l,2-diacylglycerols (1,3- and 2,3isomers are not active) and a free hydroxyl group at position 3. The nature of the alkyt chains is less important, and shorter-chain DAGs will activate if they are sufficiently hydrophobic to be membrane bound [9]; for this reason, a fairly close and specific interaction between PKC and the glycerol moiety has been depicted in Fig. 3. The major structural requirements for PS binding to PKC appears to be the L-serine headgroup as other naturally occurring phospholipids, and synthetic phospholipids made from serine analogs, are not as effective [6..,11]. In contrast, there is little specificity inherent to the glycerol backbone or the fatty acyl chains, and 1,3-diacyl-PS is able to activate PKC despite its considerable deviation from the usual 1,2-diacyt configuration [6..,11]. The exact stoichiometry of the interaction between PKC and PS is not known; however, on the basis of the co-operativity of activation in a mixed micelle system [12] and similar studies with vesicles [13], at least four PSs appear to be involved. Hence, several interactions between PKC and the headgroups of PSs have been depicted in Fig. 3. Flexibility has been left in the diagram in this region, nonetheless, because some activation can be achieved by other acidic lipids [6..,14.]. In the case of PKC activation by PI 4,5-bisphosphate, the effect was to reduce the concentration of PS required for activation [14°]. This may indicate that there are a few sites that are highly specific for PS and others that accommodate other acidic lipids more readily. Not depicted in Fig. 3 is the ability of certain lipids, especially unsaturated fatty acids [15 o] but perhaps other lipids as well [16°], to interact with cytosolic PKC.

Physical properties of lipids related to their effects on protein kinase C What is the actual nature of the interactions between PKC and these lipid activators, Ca 2+ , the protein substrate, and other factors that can influence the activity of this enzyme? One difficulty in envisioning the lipid modulation of PKC and other membrane-associated processes is the necessity for diagrammatically depicting them in two dimensions and in a static state. Membrane phospholipids typically have a high degree of mobility (inplane exchange rates for phospholipids are of the order of 107s -1 [17°']) and dipalmitoyl-PC, for example, has been estimated to have 436 755 rotational isomeric configurations [18"]. A 'snapshot' of the membrane such as that shown in Fig. 3 clearly poses some impediment to fully appreciating the behavior of these molecules. The membrane surface that modulates the activity of PKC should be viewed, to paraphrase Roux and Bloom [19"'],

Lipids as hormones and second messengers Merrill Jr and Liotta

Fig. 3. Hypothetical depiction of the

PS

PS

PS PS

DAG

as a three-dimensional region of space whose limits can be broadly defined by the polar moieties of the phospholipids, the contiguous water layers (particularly the first layer containing orientated water molecules) and associated metallic cations adsorbed at the membrane-solution interface. The requirement for PS and, in particular, for a fairly large number of PSs for activation of PKC [12,13] has a number of structural implications. The headgroups of PSs are thought to be somewhat more rigid than those of other phospholipids [20], whereas ordering of the fatty awl chain is slightly increased as a result of electrostatic repulsion [21o]. The ionization state of the carboxyl group of PS, and of any other ionizable groups in this environment (such as fatty acids [22], PA [23 °] or sphingosines [24], inter alia), may be different to what one might expect as it has been estimated that the local pH at the PS surface is approximately 3 pH units lower than the bulk pH [25] (interestingly, micelles seem to behave differently from membranes in this regard [26°]). The large proportions of negatively charged lipids on plasma membranes give them a significant electrostatic potential that might affect second-messenger systems [27,28°]. The electrostatic characteristics of the major anionic lipids (PS, phosphatidytglycerol (PG) and PI) can be described by the simplest form of Gouy-Chapman-Stem theory, which assumes that the charges are smeared uniformly over the surface [28°]. One possible implication of this behavior is that one component of the membrane binding by PKC, its substrates and other components of this signal-transduction pathway may be mediated by the specific molecular interactions implied by Fig. 3, whereas another (such as the inhibition of PKC by cationic compounds) may be described by electrostatic factors.

interactions of protein kinase C with Ca2+, phosphatidylserine (PS) and diacylglycerol (DAG) at the membrane interface. The lipid components are represented using space-filling models which were individually subjected to energy minimization calculations. The lighter shaded area at the membrane interface symbolizes waters of hydration.

It appears that the initial interaction of PKC with the membrane is mediated through its binding to PS and Ca2+ [29o,30o], and that the Ca2+-binding site(s) a r e generated at the interface between PKC and the membrane [30°]. Calcium is well known to form a complex with PS [31,32°]; hence, a Ca2 + -binding site may be created by a mechanism like that depicted in Fig. 3. It is thought that after the enzyme has bound PS and Ca2+ , it binds DAG (or phorbol esters) to become fully active [33]. Phosphatidylserine affects the specificity of PKC such that substrate phosphorylation is favored over autophosphorylation [34°°], presumably through reorientation of amino acid groups coupled to the catalytic site of the enzyme. The binding of PKC to membrane induces an increase in surface pressure [29 ° ] that can be interpreted as a reorientation of the phospholipids and/or possible insertion of a protein domain into the membrane (Fig. 3). The binding of PKC to PS is also expected to induce a reorientation of the lipid, as indicated by model studies with amphophilic peptides [35°]. Elegant studies of the physical behavior of the 1,2-DAG activators of PKC [36 °°] indicate that the sn-2 carbonyl experiences a significantly greater degree of hydrogenbonding than the sn-1 carbonyl at the aqueous interface, as indicated in Fig. 3. Furthermore, the transbilayer movement of 1,2-dilauroyt-sn-glycerol is extremely fast (tl/2 is 10 ms). In addition to these physical considerations, it should be remembered that biological membranes are undergoing constant metabolism (albeit under careful metabolic control) and vesicular movement (it is estimated that the equivalent of the entire surface of many cells is internalized every 20-30 min). There appear to be discrepancies in the ability of DAGs liberated from different phos-

519

520

Lipids phoglycerolipids to activate PKC, and some, of these may reflect intracenular compartmentalization [37°]. Furthermore, the asymmetry of PS in plasma membranes appears to be maintained by ATP-dependent 'aminophospholipid translocase(s)' [17..]. These issues are especially pertinent to PKC because many of the biological processes regulated by this enzyme involve some form of membrane movement and remodelling.

References and recommended reading Papers of special interest, published within the annual period of review, have been highlighted as: • of interest •• of outstanding interest 1.

THUDICHUMJLW: A Treatise on the Chemical Constitution of the Brain~ London: Bailliere, Tindall and Cox, 1884.

DENNISEA, RHEE SG, BILIAH MM, HANNUNYA: Role of Phospholipase in Generating Lipid Second Messengers in Signal Transduction. FASEBJ 1991, 5:2068-2077. A condse up-to-date review of the properties of phospholipase A2, C and D, sphingomyelinase and their products (arachidonic acid, alkytetherlysophosphatidylcholine, DAG, PA and ceramide). 2. ••

3.

HANNUNYA, BELLRM: Functions of Sphingolipids and Sphin. golipid Breakdown Products in Cellular Regulation. Science 1989, 243:500-507.

4. •

O ~ T, BmtAWSKAA, BELL RM, HANNUN YA: Role of Ceramide as a Lipid Mediator of lct-25-Dihydroxyvitamin D3-Induced HL-60 Cell Differentiation. J Biol Ox, m 1990, 265:15823-15831. A study of the effects of la-25-dihydroxyvitamin D3 on sphingomyelin turnover to ceramide, and the induction of hi-60 cell differentiation by a ceramide analog. These studies indicate that ceramides may be formed from sphingomyelin and act as another class of lipid second messenger to mediate the action of this, and possibly other, hormones. MERRILLAH JR: Cell Regulation by Sphingosine and More Complex Sphingolipids. J Bioenerg Biomembr 1991, 23:83-104. A review of the properties of sphingolipids, sphingolipid metabolism and possible roles of sphingosine and other more complex sphingolipids in regulating PKC and various aspects of cell behavior. 5. •

6.

BELL RM, BURNSDJ: Lipid Activation of Protein Kinase C. J Biol Chem 1991, 266:4661-4664. A•minireview of the structure-function relationships between PKC and lipid activators and inhibitors. A scheme for the possible interactions between the enzyme, its pseudosubstrate domain and the phospholipid, DAG and Ca2+-binding sites is proposed. 7.

8.

WENDERPA, KOEI-EER KF, SHARKLEYNA, DELL'AQUILLAML, BLUMBERG P: Analysis of the Phorbol Ester Pharmacophore on Protein Kinase C as a Guide to the Rational Design of New Classes of Analogs. Proc Nail Acad Sci USA 1986, 83:4214--4218. NAKAMURA H, KISrUY, PAJARESMA, RANDOR_R:Structural Basis of Protein Kinasc C Activation by Tumor Promoters. Proc Naa Acad Sci USA 1989, 86:9672-9676.

9.

GANONGB, IL~M1S C, HANNUNYA, BELLRM: Specificity and Mechanism of Protein Kinase C Activation by sn-l,2-Diacylglycerols. Proc Natl Acad Sci USA 1986, 83:1184-1188.

10.

LAUGHTONCA, BRADSHAWTD, GESCHER & Sterically Hindered Analogues of Diacylglycerols. Synthesis, Binding to the Phorbol Ester Receptor and Metabolism in A549 Human Lung Carcinoma Cells. lnt J Cancer 1989, 44:320-324.

11.

LEE M-H, BELL RM: Phospholipid Functional Groups Involved in Protein Kinase C Activation, Phorbol Ester Binding, and Binding to Mixed Micelles. J Biol Claem 1989, 264:14797-14805.

12.

HANNUNYA, LOOMISC, BELLRM: Protein Kinase C Activation in Mixed Micelles. J Biol CtJem 1986, 261:7184-7190.

13.

NEWTONA, KOSrlLANDD: High Cooperativity, Specificity and Multiplicity in the Protein Kinase C-Lipid Interaction. J Biol C/~m 1989, 264:14909-14915.

14. •

LEE MH, BELLRM: Mechanism of Protein Kinase C Activation by Phosphatidylinositol 4,5-Bisphosphate. Biochemistry 1991, 30:1041-1049. PKC activation is investigated using a Triton X-100 mixed-micelle assay. Phosphatidylinositol 4,5-bisphosphate activates PKC in a Ca2+- and PSdependent manner, reducing the concentration of PS required for activation. This may be important in vivo if the availability of PS for PKC is limiting. 15. •

EL TOUNYSE, KHANW, HANNUNY: Regulation of Platelet Protein Kinase C by Oleic Acid. Kinetic Analysis of Allosteric Regulation and Effects on Autophosphorylation, Phorbol Ester Binding, and Susceptibility to Inhibition. J Biol C)~em 1990, 265:16437-16443. Sodium oleate preferentially activates soluble PKC relative to membrane-associated enzyme by a mechanism that is distinct from activation by DAGs or phorbol esters. The activation exhibits mild cooperative behaviour which suggests that Ca2+(oleate)2 may be the active species. 16. •

BOSCOBOINIKD, ST_EWEZYKA, HENSEYC, AZZIA: Inhibition of Cell Proliferation by ct-Tocopherol. Role of Protein Kinase C. J Biol Chem 1991, 266:6188--6194. a-Tocopherol stimulates phorbol dibutyrate binding by smooth muscle cells and inhibits the translocation, activation, and the phorbol-esterinduced down-regulation of PKC. The results indicate that at-tocopherol may interact with the cytosolic form of PKC, and that this may account for the effects of vitamin E on cell proliferation. 17. DEVAUXPF: Static and Dynamic Lipid Asymmetry in Cell •• Membranes. Biochemistry 1991, 30:1163-1173. A review of the asymmetric distribution of phospholipids in eukaryotic cell membranes, the ways that transbilayer asymmetry can be established, and the possible biological functions of these processes. 18. •

PASTORRW, VENABLERM, KARPLUSM: Model for the Structure of the Lipid Bilayer. Proc Nail Acad Sci USA 1991, 88:892-896. A detailed model for the structure and dynanaics of the interior of the lipid bilayer. 19. ••

ROUXR, BLOOM M: Ca 2+, Mg2+, Li +, Na +, and K + Distribution in the Headgroup Region of Binary Membranes of Phosphatidylchollne and Phosphatidylserine as Seen by Deuterium NMR. Biochemistry 1990, 29:7077-7089. An analysis of cation binding to l-palmitoyt-2-oleoyI-PC and -PS. The 2H-NMR spectra suggest that Ca2 + is bound in at least two steps, the first occurring at a stoichiometry of 0.5 Ca2+ per PS. Roux and Bloom interpret the data as reflecting conformational changes in the phospholipid headgroups induced by an electric field resulting from the charges of the membrane-bound cations, with Ca2+ and Mg2+ more deeply buried in the membrane than Na + or K + . 20.

BROWNINGJIz Motions and Interactions of Phospholipid Head Groups at the Membrane Surface. 3. Dynamic Properties of Amine-Containing Head Groups. B i ~ i s t r y 1981, 20:7144-7151.

21. •

DEKROOMAIPM, TIMMERMANSJ~V, "KILLIANJA, DEKRUIJFFB: The pH Dependence of Headgroup and Acyl Chain Structure and Dynamics of Phosphatidytserine, Studied by 2H-NMIZ O~em Pbys L~ot2/s 1990, 54:33-42. At neutral pH, dioleoyI-PS adopts a bilayer configuration with ordering of the fatty acyi chain that is only slightly increased over that for dioleoyl-PC. At lower pH ( < 6), electrostatic repulsion between the PS headgroups is reduced, intermolecular hydrogen-bonding interactions are stronger, and the behavior of PS begins to resemble that of phosphatidyiethanolamine. 22.

SPOONERPJR, CLARKSB, GANTZ DL, HAMILTONJA, SMALLDM: The Ionization and Distribution Behavior of Oleic Acid in Chylomicrons and Chyiomicron-like Emulsion Particles

Lipids as h o r m o n e s a n d s e c o n d m e s s e n g e r s Merrill Jr a n d kiotta and the Influence of Serum Albumin. J Biol Chem 1988, 263:1444-1455. 23. •

EASTMANSJ, HOPE MJ, CULLIS PR: Transbllayer Transport of Phosphatidic Acid in Response to Transmembrane pH Gra. dients. Biochemistry 1991, 30:1740-1745. The transbilayer transport of PA is measured using large unilamellar vesicles prepared from dioleoyI-PC and PA and with detection of the acidic lipid based on its effect on fluorescence of 2-(p-toluidinyL naphthalene-6-sulfonic acid). Phosphatidic acid crossed the bilayer most rapidly in the neutral (protonated) form with a half-time of 4.1 rain at 45 *C and an activation energy of 28 kcal mol - I. 24.

25.

MERRILLAll, NIMKARS, MENALDINOD, HANNUN YA, LOOMIS C, BElL RM, TYAGI SR, LAMBETHJD, STEVENSVL, HUNTER R, DO'VIA DC: Structural Requirements for Long-Chain (Sphingoid) Base Inhibition of Protein Kinase C in vitro and for the Cellular Effects of These Compounds. Biochemistry 1989, 28:3138-3145. MACDONALDRC, SIMON SA, BAER E: Ionic Influences on the Phase Transition of Dipalmitoylphosphatidylserine. Biochemistry 1986, 15:885-891.

SCARLATASE, ROSENBERG M: Effect of Increased Lipid Packing on the Surface Charge of Micelles and Membranes. Biochemistry 1990, 29:10233-10240. Two fluorescent pH-sensitive probes are used to measure changes in lipid packing and the sensitivity of the surface pH of micelles and membranes to changes in bulk pH. The surface pH of micelles is very sensitive to bulk pH, but the pH of PC or PA bilayers changes little. Highpressure spectroscopy is used to determine the effect of increased chain packing. Pressure causes proton dissociation from the micelles, but has less of an effect in the case of bilayers. 26. ,

27.

McLAUGHLINS: The Electrostatic Properties of Membranes.

Annu Rev Biophys Biophys Chem 1989, 18:113-136. 28. .

LANGNERM, CAFISO D, MARCELJA S, MCLAUGHLIN S: Electrostatics of Phosphoinositide Bilayer Membranes. Theoretical and Experimental Results. Biophys J 1990, 57:335-349. Fluorescence, electron paramagnetic resonance, electrophoretic mobility and ionizing electrode measurements are used to study the effect of PI and PI 4,5-bisphosphate on the electrostatic potential adjacent to bilayer membranes. 29. •

SOUV1GNETC, PELOSIN J, DANIEL S, CHAMBAZ EM, RANSAC S VERGERR: Activation of Protein Kinase C in Lipid Monolayers. J Biol Chem 1991, 266:40-44. The association of PKC with membranes is assessed by the increase in monolayer surface pressure using a mixed phospholipid film. The addition of Ca 2+ results in an increase in the penetration of the kinase (DAG and phorbol esters have no effect); hence, Ca 2 + may play a major role in the translocation of PKC to membranes. 30. •

BAZZIMD, NELSESTUENGk Protein Kinase C Interaction with Calcium: a Phospholipid-Dependent Process. Biochemistry 1990, 29:7624-7630. An analysis of Ca 2+ binding by PKC using equilibrium dialysis. Free PKC binds virtually no Ca 2+ , but at least eight Ca 2+ ions are bound in the presence of acidic phospholipids, after correction for Ca 2+ binding by the phospholipids alone. These findings suggest that the Ca z +binding sites 'may be generated at the interface between PKC and the membrane'. 31.

NELSESTUENGL, LIM TK: Equilibria Involved in Prothrombin and Blood-Clotting Factor X-Membrane Binding. Biochemistry 1977, 16:4164---4171.

32. •

SWANSONJE, FEIGENSON GW: Thermodynamics of Mixing of Phosphatidylserine/Phosphatidylcholine from Measurements of High-Affinity Calcium Binding. Biochemistry 1990, 29:8291--8297. An analysis of high-alTmity Ca 2+ binding by PS in PS/PC multilayers. From these data, the authors calculate the excess partial molar free energy to transfer 1-palmityol-2-oleoyt-PS from PS to PC as - 0 . 7 kcal m o l - 1, and suggest that this arises in part from the transfer of negatively charged PS from an environment of negative charges to one of zwitterions. 33.

BAz.~ MD, NELSESTUEN GL: Properties o f the Protein Kinase C-phorbol Ester Interaction. Biochemistry 1989, 28:3577-3585.

34. ••

NEWTONAC, KOSHLANDDE: Phosphatidylserine Affects Specificity of Protein Kinase C Substrate Phosphorylation and Autophosphorylation. Biocheraistry 1990, 29:6656--6661. An analysis of PKC activity using small unilamellar vesicles of varying composition. Increasing the PS concentration (and the negative surface charge) favors substrate phosphorylation over autophosphorytation, whereas DAG increases both activities equally. Hence, PS affects the specificity of PKC toward different substrates. 35. •

DE KROON AIPM, KILLI.MqJA, DE GIER J, DE KRUIJFF B: The Membrane Headgroup and Acyl Chain Order and Dynamics. Application of the 'Phospholipid Headgroup Electrometer' Concept to Phosphatidylserine. Biochemistry 1991, 30:1155-1162. Amphiphilic peptides reduce the order parameter of both the dioleoylPS headgroup and acyl chains. The response of PS to the divalent peptides fits the electometer concept, i.e. the effective charge experienced by PS is related to the distance between the charges. A divalent peptide with a distance between the charges of ~ 10A. will have an effect similar to NaCI, which has a Debye length of 100 mM. Disruption of the phospholipid acyLchain order appears to occur with cations that are localized in the lipid-water interface and, in certain cases, where there is a collapse of the integrity of the vesicles. 36. ••

HAMILTONJA, BHAMIDIPATISP, KODALIDR, SMALLDM: The Interfacial Conformation and Transbilayer Movement of Diacylglycerois in Phospholipid BRayers. J Biol Cbem 1991, 266:1177-1186. The behavior of 1,2-dilauroyt-sn-gtycerol in PC is analysed by 31p and 13C NMR spectroscopy and other physical techniques. The NMR results indicate that the sn-2 carbonyl experiences a significantly greater degree of hydrogen-binding than the sn-1 carbonyl at the aqueous interface. Hence, the orientation of the glycerol moiety is more like that of PC, with the sn-2 carbonyl more accessible to solvent interactions. The transbilayer movement of 1,2-dilauroyl-sn-glycerol is extremely fast (t]/2 of ~ 10 ms). 37. LEACHKL, RUFF VA, WRIGHT TM, PESSlN MS, RABEN DM: Dis. sociation of Protein Kinase C Activation and sn-l,2-Diacylglycerol Formation. Comparison of Phosphatidylinositoland Phosphatidylcholine-Derived Diglycerides in at-Throm bin-Stimulated Fibroblasts. J Biol Chem 1991, 266:3215-3221. Treatment of IIC9 fibroblasts with different amounts of ct-thrombin stimulates differential formation of DAGs from P14,5-bisphosphate and PC. The results indicate that DAGs from the former, but not the latter, are able to activate PKC in this system.

All Merrill Jr, Department of Biochemistry, Emory University, Atlanta, Georgia 30322, USA. DC Liotta, Department of Chemistry, Emory University, Adanta, Georgia 30322, USA.

52"1