Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2022 Oct 31;55(23):3303–3311. doi: 10.1021/acs.accounts.2c00497

Membrane Association Allosterically Regulates Phospholipase A2 Enzymes and Their Specificity

Varnavas D Mouchlis 1,*, Edward A Dennis 1,*
PMCID: PMC9730854  PMID: 36315840

Conspectus

graphic file with name ar2c00497_0005.jpg

Water-soluble proteins as well as membrane-bound proteins associate with membrane surfaces and bind specific lipid molecules in specific sites on the protein. Membrane surfaces include the traditional bilayer membranes of cells and subcellular organelles formed by phospholipids. Monolayer membranes include the outer monolayer phospholipid surface of intracellular lipid droplets of triglycerides and various lipoproteins including HDL, LDL, VLDL, and chylomicrons. These lipoproteins circulate in our blood and lymph systems and contain triglycerides, cholesterol, cholesterol esters, and proteins in their interior, and these are sometimes interspersed on their surfaces. Similar lipid–water interfaces also occur in mixed micelles of phospholipids and bile acids in our digestive system, which may also include internalized triglycerides and cholesterol esters. Diacyl phospholipids constitute the defining molecules of biological membranes. Phospholipase A1 (PLA1) hydrolyzes phospholipid acyl chains at the sn-1 position of membrane phospholipids, phospholipase A2 (PLA2) hydrolyzes acyl chains at the sn-2 position, phospholipase C (PLC) hydrolyzes the glycerol–phosphodiester bond, and phospholipase D (PLD) hydrolyzes the polar group–phosphodiester bond. Of the phospholipases, the PLA2s have been the most well studied at the mechanistic level. The PLA2 superfamily consists of 16 groups and numerous subgroups, and each is generally described as one of 6 types. The most well studied of the PLA2s include extensive genetic and mutational studies, complete lipidomics specificity characterization, and crystallographic structures. This Account will focus principally on results from deuterium exchange mass spectrometric (DXMS) studies of PLA2 interactions with membranes and extensive molecular dynamics (MD) simulations of their interactions with membranes and specific phospholipids bound in their catalytic and allosteric sites. These enzymes either are membrane-bound or are water-soluble and associate with membranes before extracting their phospholipid substrate molecule into their active site to carry out their enzymatic hydrolytic reaction. We present evidence that when a PLA2 associates with a membrane, the membrane association can result in a conformational change in the enzyme whereby the membrane association with an allosteric site on the enzyme stabilizes the enzyme in an active conformation on the membrane. We sometimes refer to this transition from a “closed” conformation in aqueous solution to an “open” conformation when associated with a membrane. The enzyme can then extract a single phospholipid substrate into its active site, and catalysis occurs. We have also employed DXMS and MD simulations to characterize how PLA2s interact with specific inhibitors that could lead to potential therapeutics. The PLA2s constitute a paradigm for how membranes interact allosterically with proteins, causing conformational changes and activation of the proteins to enable them to extract and bind a specific phospholipid from a membrane for catalysis, which is probably generalizable to intracellular and extracellular transport and phospholipid exchange processes as well as other specific biological functions. We will focus on the four main types of PLA2, namely, the secreted (sPLA2), cytosolic (cPLA2), calcium-independent (iPLA2), and lipoprotein-associated PLA2 (Lp-PLA2) also known as platelet-activating factor acetyl hydrolase (PAF-AH). Studies on a well-studied specific example of each of the four major types of the PLA2 superfamily demonstrate clearly that protein subsites can show precise specificity for one of the phospholipid hydrophobic acyl chains, often the one at the sn-2 position, including exquisite sensitivity to the number and position of double bonds.

Key References

  • Mouchlis V. D.; Bucher D.; McCammon J. A.; Dennis E. A.. Membranes serve as allosteric activators of phospholipase A2, enabling it to extract, bind, and hydrolyze phospholipid substrates. Proc. Natl. Acad. Sci. U S A 2015, 112, E516–E525 10.1073/pnas.1424651112.1 First demonstration that membrane association causes an allosteric change in the enzyme conformation of phospholipase A2 when it is associated with a membrane.

  • Mouchlis V. D.; Chen Y.; McCammon J. A.; Dennis E. A.. Membrane Allostery and Unique Hydrophobic Sites Promote Enzyme Substrate Specificity. J. Am. Chem. Soc. 2018, 140, 3285–3291 10.1021/jacs.7b12045.2 Broad lipidomics analysis of phospholipase A2 specificity is first determined and correlated with molecular dynamics simulations to explain the binding specificity at the active site of c-, i-, and s-phospholipase A2s.

  • Hayashi D.; Mouchlis V. D.; Dennis E. A.. Omega-3 versus Omega-6 fatty acid availability is controlled by hydrophobic site geometries of phospholipase A2s. J. Lipid Res. 2021, 62, 100113. 10.1016/j.jlr.2021.100113.3 Detailed comparison of c-, i-, and s-phospholipase A2’s specificity and differing hydrophobic binding site geometries for the carbon chain length and the degree of unsaturation in the fatty acid in the sn-2 position of omega-3 and omega-6 fatty acids in phospholipids.

  • Mouchlis V. D.; Hayashi D.; Vasquez A. M.; Cao J.; McCammon J. A.; Dennis E. A.. Lipoprotein-associated phospholipase A2: A paradigm for allosteric regulation by membranes. Proc. Natl. Acad. Sci. U.S.A 2022, 119, e2102953118. 10.1073/pnas.2102953118.4 Demonstration of the conformational change that results from the association of lipoprotein-associated phospholipase A2 with a membrane and the correlation of lipidomics-determined specificity with sn-2 hydrophobic subsite geometry.

Introduction to the Phospholipase A2 Superfamily

The most well characterized phospholipase A2 (PLA2) enzymes are water-soluble peripheral enzymes that associate with the surface of membranes to extract and bind their phospholipid substrates.5,6 However, there are some PLA2s that are membrane-bound but they have been less well studied, and this Account will focus on the most well studied water-soluble PLA2s. To access its substrate, a PLA2 enzyme needs to overcome the lipid/water interface barrier.5,6 As a result, PLA2 enzymes evolved to contain unique interfacial surfaces that contain amphipathic helices facilitating the association process. The association of a PLA2 with the membrane causes a conformational change in the protein at an allosteric site distinct from its active site.1 In addition, each enzyme has a unique active site that is either highly specific for a fatty acid at the sn-2 position or more permissive depending on the cellular processes with which the enzyme is involved.2 In fact, the PLA2s have evolved very specific geometries at their specific hydrophobic site, which binds the fatty acyl group at the sn-2 position to result in highly specific substrate specificity for their biological function as determined by lipidomics analysis.3,4

Since PLA2 enzymes catalyze hydrolysis at the sn-2 position, the type of fatty acid at the sn-1 position is not expected to play a significant role in their activity. However, in some cases it seems that the ester bond at the sn-1 position is preferred versus the ether bond of the platelet activating factor (PAF) or vinyl ethers of plasmalogens.7 Hydrogen–deuterium exchange mass spectrometry (HDX-MS) is a powerful biophysical technique that has been used extensively to study the interactions of PLA2 enzymes with membranes and vesicles.8 Molecular dynamics simulations guided by HDX-MS results have proven to be very useful in understanding the association and inhibition mechanisms of PLA2 enzymes and in generating enzyme–inhibitor complexes for designing new inhibitors with improved properties.9

The PLA2 superfamily consists of 16 groups and many subgroups of structurally and functionally diverse enzymes.10 The six types of PLA2 enzymes include the secreted (sPLA2), cytosolic (cPLA2), calcium-independent (iPLA2), and lipoprotein-associated PLA2 (Lp-PLA2) also known as platelet-activating factor acetylhydrolase (PAF-AH), lysosomal PLA2, and adipose-PLA (AdPLA).6,10 Numerous important functions are attributed to the PLA2 superfamily as summarized in Table 1.6 Chief among them is the important role of PLA2 in generating inflammatory mediators, especially cPLA2, for which the group IVA cPLA2 shows dramatic specificity for the release of free eicosatetraenoic acid known as arachidonic acid (AA) from the sn-2 position of membrane phospholipids. Arachidonic acid is an omega-6 fatty acid which is generally considered to be pro-inflammatory.11

Table 1. Functions of the Human PLA2 Superfamilya.

type major groups major functions
sPLA2 groups IB, IIA, IID, IIE, V, X, XIIA digestion, antibacterial, antiviral, tumorigenesis, pro-inflammatory, anti-inflammatory
cPLA2 group IVA (cPLA2α) pro-inflammatory
iPLA2 group VIA (iPLA2β) remodeling of sn-2 fatty acids, insulin sensitivity, mitochondrial function (Barth syndrome)
LpPLA2 group VIIA (PAF-AH) anti-inflammatory, pro-inflammatory
LPLA2   lysosomal phospholipid hydrolysis
AdPLA   phospholipid hydrolysis in adipose tissue
Open in a new tab
a

Adapted with permission from ref (6). Copyright 2011 American Chemical Society.

The AA is further converted by cyclooxygenases (COX) to various tissue-specific prostaglandins that play potent cell signaling and pro-inflammatory roles by binding to specific G-protein coupled receptors (GPCRs) and activating various peroxisome-proliferator activated receptors (PPARS).11 In certain tissues, the AA is converted enzymatically to various HETEs and leukotrienes by various lipoxygenases, which activate other GPCRs and PPARS. We will discuss the specificity of the various types/groups of PLA2 at the molecular level herein. Of special current interest is determining which PLA2 types (and specific groups and subgroups) are responsible for releasing the pro-inflammatory mediator arachidonic acid versus those PLA2s that favor the omega-3 fish oil-derived fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which are often precursors of anti-inflammatory or pro-resolution lipid mediators.11 We recently determined that while GIVA cPLA2 prefers the omega-6 pro-inflammatory AA as a substrate in the sn-2 position of its phospholipid substrates, omega-3 fatty acid docosahexaenoic acid (DHA), which is generally considered to be anti-inflammatory or pro-resolution, is preferred by group V sPLA2.3 The functional pro- and anti-inflammatory activities and other functions in specific tissues and cells of each of the sPLA2s are reviewed elsewhere, including pro-inflammatory properties of GIV cPLA2 and anti-inflammatory properties of GIID sPLA2.12

In this Account, we will summarize several studies that reveal the association mechanism of group IVA cytosolic (cPLA2), group VIA calcium-independent (iPLA2), group V secreted (sPLA2), and group VIIA lipoprotein-associated phospholipase A2 (GVIIA Lp-PLA2), also known as platelet-activating factor (PAF) acetylhydrolase (PAF-AH) with membranes. Our studies revealed that these enzymes are allosterically regulated by membranes through their interfacial surface. Substrate specificity studies showed that each enzyme contains a distinct binding pocket for the sn-2 fatty acid with amino acid side chains that make it specific for a type of fatty acid. Inhibitor studies suggested that these binding sites could be used to achieve inhibition selectivity.

Membranes Induce Allosteric Conformational Changes in Associated Proteins

PLA2 enzymes need to associate with an aggregated form of their phospholipid substrate including bilayer membranes and phospholipid vesicles as well as phospholipid monolayers surrounding lipoproteins, lipid droplets, liposomes, and mixed micelles (the subject of a previous Account)13 in order to extract and bind a single phospholipid molecule.8 “Surface-dilution kinetics” was successfully employed to interpret the kinetics of PLA2s.14,15 It proposed that the enzymes first undergo surface association to the membranes depending on the bulk concentration of enzyme and membranes. In a second distinct step, the enzyme pulls a phospholipid substrate into its catalytic site, which depends on the substrate phospholipid concentration in the two-dimensional interface. Each enzyme contains an interfacial surface, which has unique topology and function, and it facilitates the association with the membrane or any other form of lipid aggregate. A combination of HDX-MS with molecular dynamics simulations has proven to be very successful in understanding the association mechanism of the pure, recombinant human form of specific group/subgroups GIVA cPLA2, GV sPLA2, GVIA iPLA2, and GVIIA Lp-PLA2. For simplicity, in this Account, we will refer to these enzymes as cPLA2, sPLA2, iPLA2, and Lp-PLA2. These studies showed that PLA2 enzymes exist in at least three conformations including the “closed” conformation in water (E), the “open” membrane associated “unbound” conformation (E·M), and the “open” membrane-associated “bound” conformation (ES·M) after the extraction of the phospholipid molecule in the active site (Figure 1).2 In this section, we will discuss the mechanism of the association of PLA2s with the membrane through their interfacial surface and its associated interactions.

Figure 1.

Figure 1

Open in a new tab

Catalytic cycle of PLA2 enzymes. The association of each enzyme with the membrane causes an allosteric conformational change that enables it to extract and bind a phospholipid molecule. Reproduced with permission from ref (2). Copyright 2018 American Chemical Society.

HDX-MS studies on the iPLA2 enzyme in the presence of phospholipid vesicles led to the identification of the peptide regions of the enzyme that interact with the membrane.16 Among them, peptide region 708–730 showed a significant decrease in deuteration levels in the presence of phospholipid vesicles. Coarse-grained and atomistic molecular dynamics were successful in understanding the association of iPLA2 with a lipid bilayer.17 The proposed model suggested that the peptide region 708–730, which is close to the active site, penetrates the lipid/water interface and forms an amphipathic helix with its hydrophilic residues (Arg710 and Lys719) interacting with the headgroups of the fatty acids and its hydrophobic residues (Pro711, Pro714, Trp715, Leu717, Val721, and Phe722) interacting with the acyl chains of the fatty acids (Figure 2A).1 It is worth mentioning that homology modeling and later the crystal structure revealed an ankyrin repeat region along with the catalytic domain of iPLA2 that did not show any interactions with the membrane bilayer.1,18

Figure 2.

Figure 2

Open in a new tab

Association of each PLA2 enzyme with the membrane bilayer through the interfacial surface. (A) iPLA2, (B) cPLA2, (C) sPLA2, and (D) Lp-PLA2. Adapted with permission from refs (2) and (4). Copyright 2018 American Chemical Society and copyright 2022 National Academy of Sciences, respectively.

Similar HDX-MS studies were performed on cPLA2 to identify the peptide regions of the interfacial surface that interact with the membrane. The cPLA2 structure contains a C2 domain along with the catalytic domain that was thought to assist the association of the enzyme with the membrane through its calcium binding site.19,20 The HDX-MS studies showed several peptide regions on the C2 domain and the catalytic domain exhibiting decreases in deuteration levels in the presence of phospholipid vesicles.21 According to our model, two regions of the C2 domain (residues 35–39 and 96–98) penetrate the membrane approximately 15 Å, bringing along the catalytic domain to near the membrane surface. Residues 268–279 and 466–470 are two peptide regions of the catalytic domain that are close to the surface of the membrane containing some hydrophobic residues (Trp464 and Met468) interacting with the fatty acid tails of the phospholipids and some hydrophilic residues (Lys273, Lys274, and Arg467) interacting with the headgroups of the phospholipids (Figure 2B).1

As with the iPLA2 and cPLA2 enzymes, sPLA2 also exhibits an increase in activity due to interfacial activation in the presence of a membrane or any other form of lipid aggregates.22,23 HDX-MS studies on group IA sPLA2 showed that peptide regions 3–8, 18–21, and 56–64 exhibited a significant decrease in the “on-exchange” rates in the presence of dimyristoylphosphatidylcholine phospholipid vesicles. These peptide regions have an amphipathic nature containing hydrophobic residues that penetrate the membrane and hydrophilic residues that interact with the headgroups on the surface of the membrane.24 Along with GIA sPLA2, we were also interested in the human GV sPLA2, and this later enzyme was used in our detailed substrate specificity studies. Since there is no crystal structure available for GV sPLA2, a homology model was generated based on the crystal structure of GIIA sPLA2 because these two enzymes have 47% identity and 61% homology. In general, there is high homology between sPLA2 enzymes, and that allowed us to use the HDX-MS peptide regions to generate a structural model for GV sPLA2 docked on the surface of the membrane (Figure 2C). This model was subjected to extensive molecular dynamics simulations in the presence of various phospholipid substrates. These studies allowed us to determine the structural features that contribute to substrate selectivity, and they will be discussed in detail in the next section of this Account.

Lp-PLA2 is also membrane-associated and undergoes interfacial activation like c-, i-, and sPLA2 enzymes. According to our HDX-MS studies, peptide regions 114–120 and 360–368 exhibited decreases in on-exchange rates in the presence of phospholipid vesicles (Figure 2D). These two regions constitute two amphipathic helices that play a significant role in protein–membrane binding because their properties allow them to overcome the water–lipid energy barrier.25−27 An enzyme-membrane model was generated based on the HDX-MS data and was subjected to extensive molecular dynamics simulations to study the interfacial activation mechanism of Lp-PLA2.4 The simulations showed that in the presence of the membrane, the volume of the Lp-PLA2 active site was increased from ∼900 to 2000 Å3 due to a conformational change caused by the region 100–130. When the open form of the enzyme was placed in water, a conformational change in the same region caused the enzyme to adopt a closed form.4 Similar conformational changes were confirmed for c-, i-, and sPLA2.2

Specific Phospholipase A2 Binding Sites Extract and Bind a Specific Phospholipid Molecule and Determine the Phospholipid Substrate Specificity

Phospholipids are highly flexible molecules and are often very challenging to cocrystallize with PLA2 enzymes. Thus, there are no available crystal structures to help understand the binding mode and interactions of phospholipids with the active site of PLA2 enzymes. Computational methods such as molecular docking and induced-fit docking could not handle the high flexibility of phospholipids.28 Atomistic molecular dynamics at a microsecond scale have proven successful in developing enzyme–substrate models. These models help understand the binding mode and interactions of phospholipids with each PLA2 active site and explain each enzyme’s specificity for various types of phospholipid molecules. Simulations were carried out for c-, i-, s-, and Lp-PLA2 enzymes with phospholipids containing various types of fatty acids at the sn-2 position.1,2,4 Arachidonic (20:4), linoleic (18:2), myristic (14:0), and azelaoyl (9:0, COOH) esterified at the sn-2 position of membrane phospholipids will be discussed in this Account because they are the optimum substrates for cPLA2, i and sPLA2, and Lp-PLA2, respectively.

The active site of cPLA2 contains a deep channel containing a large hydrophobic binding pocket that accommodates the two acyl chains of a phospholipid molecule. Enzymatic specificity studies on cPLA2 showed that this enzyme is highly specific for arachidonic acid at the sn-2 position. The simulations showed that cPLA2 contains a hydrophobic pocket that is rich with aromatic residues including Phe199, Phe291, Phe295, Trp232, Phe397, Phe401, Phe681, Phe683, and Tyr685 that act as a fingerprint recognizing and interacting with the double bonds of arachidonic acid (20:4) through π–π stacking (Figure 3A).2 iPLA2 exhibited optimum activity for phospholipids containing linoleic (18:2) and myristic (14:0) acid at the sn-2 position. The simulations showed that iPLA2 contains two hydrophobic binding pockets for the sn-2 fatty acid tail: one that could accommodate myristic (14:0) (not shown) and a second that could accommodate linoleic (18:2) (shown) (Figure 3B). The first hydrophobic pocket contains residues such as Leu491, Ile494, Ile523, Leu524, Leu564, Met537, and Leu560, and it has a suitable site to accommodate the acyl chain of the myristic (14:0) acid. The second hydrophobic pocket contains residues such as Tyr541, Tyr544, and Phe644, which interact with the double bonds of the linoleic (18:2) fatty acid.2

Figure 3.

Figure 3

Open in a new tab

Molecular dynamics simulations revealed the structural features that contribute to substrate specificity. (A) cPLA2, (B) iPLA2, (C) sPLA2, and (D) Lp-PLA2. Adapted with permission from refs (2) and (4). Copyright 2018 American Chemical Society and copyright 2022 National Academy of Sciences, respectively.

sPLA2 is a small 14 kDa enzyme which exhibits specificity toward linoleic (18:2) and myristic (14:0) acid at the sn-2 position, similarly to iPLA2.2,9 The simulations showed that sPLA2 also contains two hydrophobic pockets that could accommodate either the linoleic (18:2) or the myristic (14:0) fatty acid tail at the sn-2 position (Figure 3C). The first pocket contains residues such as Val12, Leu102, Leu98, and Leu94, and it accommodates the shorter saturated myristic (14:0) acyl chain. The second pocket contains residues such as Tyr21, Tyr24, and Tyr105 that interact with the double bonds of the linoleic (18:2) acyl chain through π–π stacking interactions. Lp-PLA2 exhibits greater enzymatic activity toward oxidized-PC (phosphocholine) substrates or PAF (platelet activating factor) compared to normal phospholipid substrates.4,5 The simulations showed that Gln352 forms a stronger hydrogen bond with the oxygen atom of the sn-1 carbonyl group of the PC analogues versus the ether oxygen of the PAF analogues (Figure 3D). A smaller semihydrophilic pocket was also identified in the Lp-PLA2 active site containing residuesTyr160, His151, and His272 forming hydrogen bonds with the carbonyl group of the oxidized phospholipids.

Therapeutic Targeting through Inhibitor Interactions with Highly Evolved Phospholipid Hydrophobic Acyl Chain Subsites

Phospholipid substrate specificity is tightly connected to the hydrophobic pockets of the PLA2 active site. Thus, inhibitor selectivity is also driven by the hydrophobic pockets of the PLA2 active site. Molecular dynamics simulations guided by HDX-MS were extensively used by our group to understand inhibitor interactions of PLA2 enzymes, especially for c- and iPLA2 since no cocrystallized structures are available.29,30 Fluoroketone inhibitors are a class of potent and selective inhibitors developed for iPLA2 that are a clear example of how substrate specificity could be translated into inhibitor selectivity.31−33 The fluoroketone group of these inhibitors interacts with the “oxyanion hole”, which consists of two glycine residues in both c- and iPLA2 and is in close proximity to the catalytic serine which is also common in both enzymes. Despite the common interaction pattern of the fluoroketone moiety, our structure–activity relationships revealed that fluoroketone compounds with specific hydrophobic tails could be highly selective toward iPLA2.

Based on the simulations of phospholipid substrates, it was found that cPLA2 contains a deep hydrophobic pocket that accommodates the sn-2 arachidonic tail. This pocket contains many aromatic residues that interact with the double bonds of the arachidonic tail. There are two factors that govern tight binding in this hydrophobic pocket: the size of the ligand and its aromaticity. Fluoroketone compounds that contain longer aromatic hydrophobic tails tend to be potent inhibitors for cPLA2 (Figure 4A). Such compounds contain large enough tails to complement the hydrophobic pocket of cPLA2 and at the same time form π–π stacking with aromatic residues such as Phe199, Phe291, Phe295, Trp232, Phe397, Phe401, Phe681, Phe683, and Tyr685. Larger compounds such as oxoamides containing long aliphatic or aromatic tails and are also potent and selective inhibitors of cPLA2.34−36

Figure 4.

Figure 4

Open in a new tab

Specificity of each phospholipase A2 hydrophobic binding site for specific substrates and inhibitors. Binding mode of fluoroketone inhibitors in the active sites of (A) cPLA2 and (B) iPLA2. (C) Binding mode of a potent specific inhibitor in the active site of Lp-PLA2. (D) Binding mode of a phospholipid containing myristic acid at the sn-2 position in the active site of iPLA2. Adapted with permission from refs (2), (4), and (31). Copyright 2018 American Chemical Society, copyright 2022 National Academy of Sciences, and copyright 2016 American Chemical Society, respectively.

The hydrophobic part of the iPLA2 active site consist of two pockets: one that contains aromatic residues and accommodates the sn-2 linoleic chain and one that contains aliphatic residues and accommodates the sn-2 myristic chain.2 Fluoroketone compounds that contain shorter aromatic tails tend to be potent and selective inhibitors for iPLA2. The aromatic pocket of iPLA2 has a significantly smaller volume than the one in cPLA2 because the sn-2 site in iPLA2 is much more specific for linoleic acid at the sn-2 position. In addition, the active site of iPLA2 is more flexible, and the inhibitor locks the enzyme in the closed conformation in which the volume of the hydrophobic pocket is significantly smaller (Figure 4B).31 Interestingly, very potent inhibitors of Lp-PLA2 contain aromatic residues that specifically associate with aromatic residues in the protein active site (Figure 4C).4 It is also worth noting that fluoroketone compounds containing only short aliphatic tails are also potent and selective inhibitors for iPLA2. Molecular dynamics simulations combined with HDX-MS studies showed that the aliphatic tail binds to the smaller aliphatic pocket of iPLA2 that accommodates the sn-2 myristic chain in myristoyl-containing phospholipids (Figure 4D).32

Our recently developed lipidomic assay for c-, i-, s-, and Lp-PLA2 enzymes, using existing potent inhibitors for each enzyme, provided a semihigh throughput setup for performing virtual and experimental screening to identify new chemical matter for PLA2 inhibition.37 Having in mind the above structural characteristics of the c- and iPLA2 hydrophobic pockets, a virtual screening workflow was developed, including clustering analysis, docking in multiple enzyme conformations, and interaction criteria to select compounds from libraries to be tested in vitro.38 This work led to the identification of novel micromolar hits that are currently under development. sPLA2 also contains an aromatic and an aliphatic hydrophobic pocket, and it is inhibited by indole compounds that contain short aliphatic and aromatic chains.37 Lp-PLA2 contains a large hydrophobic pocket that accommodates the sn-1 acyl chain and a smaller amphipathic pocket that accommodates short-chain oxidized sn-2 chains. Compounds that contain a long chain that occupies the sn-1 pocket and a shorter chain that binds the sn-2 chain are potent inhibitors for this enzyme (Figure 4C).4

Members of the phospholipase A2 superfamily are responsible for numerous important biological functions, but they have also been implicated in numerous deleterious physiological effects on lipid metabolism with significant disease implications. This includes primary roles for cPLA2 in arthritis and a multitude of inflammatory diseases, sPLA2 and Lp-PLA2 in various cardiovascular diseases, and iPLA2 in diabetes and a variety of neurological diseases. Over the years, numerous inhibitors with excellent potency and specificity for a particular PLA2 were synthesized, and the pharmaceutical industry has pursued the development of several as potential therapeutic moieties. This has resulted in many clinical trials, including several large phase 3 trials. Kokotos and co-workers39,40 have recently reviewed the patent and clinical trials literature and report that, surprisingly, to date no approved drug has emerged. Perhaps the insights into the importance of the sn-2 site for binding potency and specificity discussed in this Account will provide new avenues of focus for the development of successful therapeutic agents.

Conclusions

Receptors, phospholipid transfer and exchange proteins, and enzymes often exhibit remarkable specificity for specific phospholipids in each of their sn-1 acyl chain, sn-2 acyl chain, and polar group subsites. It is likely that membranes induce allosteric conformational changes in proteins associated with bilayer membranes both intra- and extracellularly as well as when associated with monolayers surrounding lipoproteins and lipid droplets. These interactions occur at the lipid–water interface of membranes. The interaction of PLA2 with bilayer membranes and with its specific phospholipid substrate in its active site and especially the specificity for the sn-2 fatty acyl chain serve as a general paradigm for the allosteric regulation of proteins by membranes

A recent review41 focused on the early development of the detailed kinetic concepts, experimental approaches, and results of molecular dynamics simulations including links to the resulting movies, which led to the concept of allosteric regulation by membranes and the demonstration of the importance of the sn-2 site for the specificity of PLA2 enzymes. In the current Account, we have presented an expanded comprehensive general picture of allosteric regulation by membranes and the specificity of hydrophobic subsites in enzymes that act in or on membranes.

This was possible because of the recent demonstration4 of the allosteric regulation and specificity of Lp-PLA2, which complements the earlier work on the three traditional cellular PLA2s. Of special note is the recent results3 comparing the differential selectivities of c-, s-, and iPLA2 for phospholipids containing in the sn-2 position the omega-3 fatty acid EPA or DHA compared to the omega-6 fatty acid AA. In total, these experimental and computational approaches allow us to articulate the structural features of each enzyme that contribute to its substrate potency and specificity. By connecting substrate specificity with inhibitor structure–activity relationships, we can now extend the substrate specificity determinations to determine the precise factors contributing to inhibitor potency and selectivity. This information should aid in the development of a general approach to designing novel chemical moieties for PLA2-selective inhibition.

Acknowledgments

Work on this manuscript and research in our laboratory on phospholipase A2 have been supported by NIGMS MIRA grant R35 GM139641, which is a renewal of RO1 GM20501-44.

Biographies

Varnavas D. Mouchlis is a senior computational chemist at Bristol Myers Squibb working on developing next-generation therapeutics using novel modalities including targeted protein degradation. He was born in Nicosia, Cyprus, and he received his Ph.D. in organic and computational chemistry from the University of Athens, Greece in 2010 with Prof. George Kokotos. He carried out his postdoctoral studies at the University of California, San Diego with Prof. Edward A. Dennis, during which time he also studied with Prof. J. Andrew McCammon. His postdoctoral research focused on understanding protein–membrane, protein–lipid, and protein–inhibitor interactions using computational and experimental methods including molecular dynamics, hydrogen–deuterium exchange mass spectrometry, homology modeling, lipidomics assays, and virtual screening. He successfully developed structural models to carry out virtual and experimental screenings of compound libraries to identify new chemical matter for phospholipase A2 inhibition. During his postdoctoral studies, he received several awards including the Wiley Computers in Chemistry Outstanding Postdoc Award by the American Chemical Society, Division of Computers in Chemistry. He then moved to Nova Mechanics Ltd. to work on the development of therapeutics for targets involved in CNS diseases including Alzheimer’s and Parkinson’s. He applied computational methods including virtual screening and molecular dynamics to develop antiaggregation inhibitors for intrinsically disordered proteins. Subsequently at UCB Biosciences, he worked on small-molecule drug discovery projects and new modalities using computer-aided drug design methods including ternary complex modeling, mixed-solvent molecular dynamics to identify cryptic pockets, molecular dynamics on ternary complexes, scaffold hopping, free-energy perturbation, and virtual screening. His goal is to continue to learn and grow into an experienced drug hunter and to develop new therapeutics that will positively impact the lives of patients.

Edward A. Dennis is the Chancellor I Endowed Chair and Distinguished Professor of Chemistry, Biochemistry, Pharmacology, and the Graduate Division at the University of California, San Diego (UCSD). He was born in Chicago, Illinois, received his B.A. from Yale University (1963) and his Ph.D. from Harvard University (1968), and carried out postdoctoral studies at Harvard Medical School (1967–1969). He received a doctorate in medicine (honorary, 2008) from Goethe University in Frankfurt, Germany and a doctorate (honorary, 2014) from the University of Lyon INSA in Lyon, France. Dr. Dennis started (1970) as an assistant professor at UCSD and served two terms as Chair of the Department of Chemistry and Biochemistry (1999–2002 and 2017–2019). He received a Guggenheim Fellowship for 1983 to 1984, serving as a visiting professor at Harvard Medical School and a visiting scientist at Brandeis University. Dr. Dennis served as a visiting foreign professor at the Collège de France in Paris (2012), as an adjunct professor at The Scripps Research Institute since 1999, and as a visiting research professor at the Université Pierre et Marie Curie since 2010. He also served as Chair and President of the Keystone Symposia (1996–2004), Editor-in-Chief of the Journal of Lipid Research (2003–2018), and Director of the LIPID MAPS Lipidomics Consortium (2003–2014). He is an inaugural fellow of the American Association for the Advancement of Science (AAAS) (1984) and an inaugural fellow of the American Society of Biochemistry and Molecular Biology (ASBMB) (2021). He was the recipient of the ASBMB’s Avanti Award in Lipid Enzymology (2000), the European Federation for Lipid Science and Technology’s European Lipid Science Award (2008), the Yale Medal (2008), the American Chemical Society, San Diego Section, Distinguished Scientist Award (2016), and the ASBMB Bert Vallee Award in Biomedical Science (2021). Prof. Dennis’ career research focus has been on the structure, function, mechanism, and inhibition of the enzyme phospholipase A2 as well as on protein–membrane and protein–lipid interactions, micelle formation, signal transduction, inflammation, lipid metabolism, eicosanoid action, and especially developing the lipidomics field.

Author Contributions

This manuscript was written and edited by both authors, and they have given approval to the final version of the manuscript.

The authors declare no competing financial interest.

Special Issue

Published as part of the Accounts of Chemical Research special issue “Chemical Biology of Lipids”.

References

  1. Mouchlis V. D.; Bucher D.; McCammon J. A.; Dennis E. A. Membranes serve as allosteric activators of phospholipase A2, enabling it to extract, bind, and hydrolyze phospholipid substrates. Proc. Natl. Acad. Sci. U S A 2015, 112, E516–E525. 10.1073/pnas.1424651112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Mouchlis V. D.; Chen Y.; McCammon J. A.; Dennis E. A. Membrane Allostery and Unique Hydrophobic Sites Promote Enzyme Substrate Specificity. J. Am. Chem. Soc. 2018, 140, 3285–3291. 10.1021/jacs.7b12045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Hayashi D.; Mouchlis V. D.; Dennis E. A. Omega-3 versus Omega-6 fatty acid availability is controlled by hydrophobic site geometries of phospholipase A2s. J. Lipid Res. 2021, 62, 100113. 10.1016/j.jlr.2021.100113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Mouchlis V. D.; Hayashi D.; Vasquez A. M.; Cao J.; McCammon J. A.; Dennis E. A. Lipoprotein-associated phospholipase A2: A paradigm for allosteric regulation by membranes. Proc. Natl. Acad. Sci. U.S.A 2022, 119, e2102953118. 10.1073/pnas.2102953118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Mouchlis V. D.; Dennis E. A. Phospholipase A2 catalysis and lipid mediator lipidomics. Biochim. Biophys. Acta Mol. Cell Biol. Lipids 2019, 1864, 766–771. 10.1016/j.bbalip.2018.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dennis E.; Cao J.; Hsu Y.-H.; Magrioti V.; Kokotos G. Phospholipase A2 enzymes: physical structure, biological function, disease implication, chemical inhibition, and therapeutic intervention. Chem. Rev. 2011, 111, 6130–6185. 10.1021/cr200085w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Hayashi D.; Mouchlis V. D.; Dennis E. A. Each phospholipase A2 type exhibits distinct selectivity toward sn-1 ester, alkyl ether, and vinyl ether phospholipids. Biochim. Biophys. Acta Mol. Cell. Biol. Lipids 2022, 1867, 159067. 10.1016/j.bbalip.2021.159067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cao J.; Burke J.; Dennis E. Using hydrogen/deuterium exchange mass spectrometry to define the specific interactions of the phospholipase A2 superfamily with lipid substrates, inhibitors, and membranes. J. Biol. Chem. 2013, 288, 1806–1813. 10.1074/jbc.R112.421909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Vasquez A. M.; Mouchlis V. D.; Dennis E. A. Review of four major distinct types of human phospholipase A2. Adv. Biol. Regul 2018, 67, 212–218. 10.1016/j.jbior.2017.10.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Mouchlis V. D.; Dennis E. A. Membrane and inhibitor interactions of intracellular phospholipases A2. Adv. Biol. Regul 2016, 61, 17–24. 10.1016/j.jbior.2015.11.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dennis E. A.; Norris P. C. Eicosanoid storm in infection and inflammation. Nat. Rev. Immunol 2015, 15, 511–523. 10.1038/nri3859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Murakami M.; Sato H.; Miki Y.; Yamamoto K.; Taketomi Y. A new era of secreted phospholipase A2. J. Lipid Res. 2015, 56, 1248–1261. 10.1194/jlr.R058123.1015.1.test. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Robson R. J.; Dennis E. A. Micelles of nonionic detergents and mixed micelles with phospholipids. Acc. Chem. Res. 1983, 16, 251–258. 10.1021/ar00091a003. [DOI] [Google Scholar]
  14. Dennis E. A. Phospholipase A2 activity towards phosphatidylcholine in mixed micelles: Surface dilution kinetics and the effect of thermotropic phase transitions. Arch. Biochem. Biophys. 1973, 158, 485–493. 10.1016/0003-9861(73)90540-7. [DOI] [PubMed] [Google Scholar]
  15. Carman G.; Deems R.; Dennis E. Lipid signaling enzymes and surface dilution kinetics. J. Biol. Chem. 1995, 270, 18711–18714. 10.1074/jbc.270.32.18711. [DOI] [PubMed] [Google Scholar]
  16. Hsu Y.-H.; Burke J.; Li S.; Woods V.; Dennis E. Localizing the membrane binding region of Group VIA Ca2+-independent phospholipase A2 using peptide amide hydrogen/deuterium exchange mass spectrometry. J. Biol. Chem. 2009, 284, 23652–23661. 10.1074/jbc.M109.021857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Bucher D.; Hsu Y. H.; Mouchlis V. D.; Dennis E. A.; McCammon J. A. Insertion of the Ca2+-independent phospholipase A2 into a phospholipid bilayer via coarse-grained and atomistic molecular dynamics simulations. PLoS Comput. Biol. 2013, 9, e1003156 10.1371/journal.pcbi.1003156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Malley K. R.; Koroleva O.; Miller I.; Sanishvili R.; Jenkins C. M.; Gross R. W.; Korolev S. The structure of iPLA2β reveals dimeric active sites and suggests mechanisms of regulation and localization. Nat. Commun. 2018, 9, 765. 10.1038/s41467-018-03193-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Dessen A.; Tang J.; Schmidt H.; Stahl M.; Clark J.; Seehra J.; Somers W. Crystal structure of human cytosolic phospholipase A2 reveals a novel topology and catalytic mechanism. Cell 1999, 97, 349–360. 10.1016/S0092-8674(00)80744-8. [DOI] [PubMed] [Google Scholar]
  20. Málková S.; Long F.; Stahelin R. V.; Pingali S. V.; Murray D.; Cho W.; Schlossman M. L. X-ray reflectivity studies of cPLA2α-C2 domains adsorbed onto Langmuir monolayers of SOPC. Biophysical journal 2005, 89, 1861–1873. 10.1529/biophysj.105.061515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Burke J.; Hsu Y.-H.; Deems R.; Li S.; Woods V.; Dennis E. A phospholipid substrate molecule residing in the membrane surface mediates opening of the lid region in group IVA cytosolic phospholipase A2. J. Biol. Chem. 2008, 283, 31227–31236. 10.1074/jbc.M804492200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gelb M. H.; Jain M. K.; Hanel A. M.; Berg O. G. Interfacial enzymology of glycerolipid hydrolases: Lessons from secreted phospholipases A2. Annu. Rev. Biochem. 1995, 64, 653–688. 10.1146/annurev.bi.64.070195.003253. [DOI] [PubMed] [Google Scholar]
  23. Singer A. G.; Ghomashchi F.; Le Calvez C.; Bollinger J.; Bezzine S.; Rouault M.; Sadilek M.; Nguyen E.; Lazdunski M.; Lambeau G.; Gelb M. H. Interfacial Kinetic and Binding Properties of the Complete Set of Human and Mouse Groups I, II, V, X, and XII Secreted Phospholipases A2. J. Biol. Chem. 2002, 277, 48535–48549. 10.1074/jbc.M205855200. [DOI] [PubMed] [Google Scholar]
  24. Burke J. E.; Karbarz M. J.; Deems R. A.; Li S.; Woods V. L. Jr.; Dennis E. A. Interaction of group IA phospholipase A2 with metal ions and phospholipid vesicles probed with deuterium exchange mass spectrometry. Biochemistry 2008, 47, 6451–6459. 10.1021/bi8000962. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Cao J.; Hsu Y. H.; Li S.; Woods V. L. Jr.; Dennis E. A. Structural basis of specific interactions of Lp-PLA2 with HDL revealed by hydrogen deuterium exchange mass spectrometry. J. Lipid Res. 2013, 54, 127–133. 10.1194/jlr.M030221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Cao J.; Hsu Y.-H.; Li S.; Woods V. L.; Dennis E. A. Lipoprotein-associated phospholipase A2 interacts with phospholipid vesicles via a surface-disposed hydrophobic α-helix. Biochemistry 2011, 50, 5314–5321. 10.1021/bi101916w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Drin G.; Antonny B. Amphipathic helices and membrane curvature. FEBS Lett. 2010, 584, 1840–1847. 10.1016/j.febslet.2009.10.022. [DOI] [PubMed] [Google Scholar]
  28. Mouchlis V. D.; Barbayianni E.; Mavromoustakos T. M.; Kokotos G. The application of rational design on phospholipase A2 inhibitors. Curr. Med. Chem. 2011, 18, 2566–2582. 10.2174/092986711795933678. [DOI] [PubMed] [Google Scholar]
  29. Burke J.; Babakhani A.; Gorfe A.; Kokotos G.; Li S.; Woods V.; McCammon J.; Dennis E. Location of inhibitors bound to group IVA phospholipase A2 determined by molecular dynamics and deuterium exchange mass spectrometry. J. Am. Chem. Soc. 2009, 131, 8083–8091. 10.1021/ja900098y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Hsu Y.-H.; Bucher D.; Cao J.; Li S.; Yang S.-W.; Kokotos G.; Woods V.; McCammon J.; Dennis E. Fluoroketone inhibition of Ca2+-independent phospholipase A2 through pinding pocket association defined by hydrogen/deuterium exchange and molecular dynamics. J. Am. Chem. Soc. 2013, 135, 1330–1337. 10.1021/ja306490g. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mouchlis V. D.; Limnios D.; Kokotou M. G.; Barbayianni E.; Kokotos G.; McCammon J. A.; Dennis E. A. Development of potent and selective inhibitors for group VIA calcium-independent phospholipase A2 guided by molecular dynamics and structure-activity relationships. J. Med. Chem. 2016, 59, 4403–4414. 10.1021/acs.jmedchem.6b00377. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mouchlis V. D.; Morisseau C.; Hammock B. D.; Li S.; McCammon J. A.; Dennis E. A. Computer-aided drug design guided by hydrogen/deuterium exchange mass spectrometry: A powerful combination for the development of potent and selective inhibitors of Group VIA calcium-independent phospholipase A2. Bioorg. Med. Chem. 2016, 24, 4801–4811. 10.1016/j.bmc.2016.05.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Magrioti V.; Nikolaou A.; Smyrniotou A.; Shah I.; Constantinou-Kokotou V.; Dennis E.; Kokotos G. New potent and selective polyfluoroalkyl ketone inhibitors of GVIA calcium-independent phospholipase A2. Bioorg. Med. Chem. 2013, 21, 5823–5829. 10.1016/j.bmc.2013.07.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Antonopoulou G.; Magrioti V.; Kokotou M. G.; Nikolaou A.; Barbayianni E.; Mouchlis V. D.; Dennis E. A.; Kokotos G. 2-Oxoamide inhibitors of cytosolic group IVA phospholipase A2 with reduced lipophilicity. Bioorg. Med. Chem. 2016, 24, 4544–4554. 10.1016/j.bmc.2016.07.057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Mouchlis V. D.; Michopoulou V.; Constantinou-Kokotou V.; Mavromoustakos T.; Dennis E. A.; Kokotos G. Binding conformation of 2-oxoamide inhibitors to group IVA cytosolic phospholipase A2 determined by molecular docking combined with molecular dynamics. J. Chem. Inf. Model 2012, 52, 243–254. 10.1021/ci2005093. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Kokotos G.; Six D. A.; Loukas V.; Smith T.; Constantinou-Kokotou V.; Hadjipavlou-Litina D.; Kotsovolou S.; Chiou A.; Beltzner C. C.; Dennis E. A. Inhibition of group IVA cytosolic phospholipase A2 by novel 2-oxoamides in vitro, in cells, and in vivo. J. Med. Chem. 2004, 47, 3615–3628. 10.1021/jm030485c. [DOI] [PubMed] [Google Scholar]
  37. Mouchlis V. D.; Armando A.; Dennis E. A. Substrate-Specific Inhibition Constants for Phospholipase A2 Acting on Unique Phospholipid Substrates in Mixed Micelles and Membranes Using Lipidomics. J. Med. Chem. 2019, 62, 1999–2007. 10.1021/acs.jmedchem.8b01568. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Mouchlis V. D.; Mu C.; Hammons R.; Dennis E. A. Lipidomics-based assays coupled with computational approaches can identify novel phospholipase A2 inhibitors. Adv. Biol. Regul 2020, 76, 100719. 10.1016/j.jbior.2020.100719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kokotou M. G.; Limnios D.; Nikolaou A.; Psarra A.; Kokotos G. Inhibitors of phospholipase A2 and their therapeutic potential: an update on patents (2012–2016). Expert Opinion on Therapeutic Patents 2017, 27, 217–225. 10.1080/13543776.2017.1246540. [DOI] [PubMed] [Google Scholar]
  40. Batsika C. S.; Gerogiannopoulou A. D.; Mantzourani C.; Vasilakaki S.; Kokotos G. The design and discovery of phospholipase A2 inhibitors for the treatment of inflammatory diseases. Expert Opinion on Drug Discovery 2021, 16, 1287–1305. 10.1080/17460441.2021.1942835. [DOI] [PubMed] [Google Scholar]
  41. Dennis E. A. Allosteric regulation by membranes and hydrophobic subsites in phospholipase A2 enzymes determine their substrate specificity.. J. Biol. Chem. 2022, 298, 101873. 10.1016/j.jbc.2022.101873. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Accounts of Chemical Research are provided here courtesy of American Chemical Society

RESOURCES