Signaling Lipids

(1)

Project-team INRIA-UPMC-CNRS REO Laboratoire Jacques-Louis Lions, CNRS UMR 7598, Université Pierre et Marie Curie, Place Jussieu 4, Paris Cedex 05, France

Abstract

Lipids include sterols, mono- and diglycerides, phospholipids, among others. They operate as structural components of cell membranes (Vol. 1 – Chap. 7. Plasma Membrane; Tables 2.1 and 2.2), contributors of energy metabolism, participants of intracellular transport (Vol. 1 – Chap. 9. Intracellular Transport), and signaling molecules.

Table 2.1

Cellular sites — plasma and organelle membranes — of main lipids involved in organelle recognition and signaling (Part 1; Source: [7]). The plasma membrane contains high levels of packed sphingolipids and sterols to resist mechanical stress, as well as signaling lipids. The endoplasmic reticulum produces cholesterol, main glycerophospholipids, steryl ester, triacylglycerol (or triglyceride), ceramide, and galactosylceramide. Among glycerophospholipids (or phosphoglycerides, i.e., glycerol-based phospholipids), phosphatidylcholine, phosphatidylethanolamine, and bisphosphatidylglycerol are also called lecithin, cephalin, and cardiolipin, respectively. The first 2 species are more common than other glycerophospholipids in most cellular membranes, except cardiolipin in the inner membrane of mitochondria. The endoplasmic reticulum has low concentrations of sterols and complex sphingolipids. The Golgi body synthesizes phosphatidylcholine, sphingolipids, and glycosphingolipids. Mitochondrion manufactures about half of its phospholipids. Early endosomes have a content similar to the plasma membrane, whereas in late endosomes concentrations of sterol and bis(monoacylglycero)phosphate decrease and increase, respectively.

Lipids	Cellular location
Cholesterol	All membranes
Glycerophospholipids (or phosphoglycerides)
Bis(monoacylglycero)phosphate	Late endosome
Diacylglycerol	Plasma membrane
Phosphatidic acid	Endoplasmic reticulum, mitochondrion
Phosphatidylcholine	Endoplasmic reticulum, Golgi body,
	mitochondrion, endosome, plasma membrane
Phosphatidylethanolamine	Endoplasmic reticulum, Golgi body,
	mitochondrion, endosome, plasma membrane
Phosphatidylglycerol	Mitochondrion
Phosphatidylinositol	Endoplasmic reticulum, Golgi body,
	mitochondrion, endosome, plasma membrane
Phosphatidylserine	Endoplasmic reticulum, Golgi body,
	mitochondrion, endosome, plasma membrane
Triacylglycerol	Endoplasmic reticulum, lipid droplet

Table 2.2

Cellular sites — plasma and organelle membranes — of main lipids that are involved in organelle recognition and signaling (Part 2; Source: [7]; PI: phosphatidylinositol). Kinases and phosphatases produce and hydrolyze specific phosphoinositides (PI(3)P on early endosomes and PI(3,5)P₂ on late endosomes). More complex sphingolipids require vesicular transport from the endoplasmic reticulum.

Lipids	Cellular location
Sphingolipids
Ceramide	Endoplasmic reticulum, plasma membrane
Galactosylceramide	Endoplasmic reticulum
Glycosphingolipids	Endoplasmic reticulum, Golgi body
Sphingomyelin	Golgi body, endosome,
	plasma membrane
Sphingosine	Plasma membrane
Sphingosine 1-phosphate	Plasma membrane
Phospholipids
Cardiolipin	Mitochondrion
PI(3)P	Early endosome
PI(4)P	Golgi body, plasma membrane
PI(3,4)P₂	Late endosome
PI(4,5)P₂	Plasma membrane
PI(3,4,5)P₃	Plasma membrane

Phospholipids are constituents of cellular membranes that are also used for emulsification in pharmaceuticals and food and preparation of liposomes for drug delivery. Phospholipids are characterized by a glycerol backbone to which a polar phosphodiester group is linked at the sn3 carbon.¹ Glycerol-based phospholipids are also named glycerophospholipids, or phosphoglycerides. Two fatty acid-derived acyl residues are connected at the sn1 and sn2 carbons. The type of polar head groups defines the class of phospholipids. Three distinct structural regions can be defined [8]: (1) a hydrophilic head at the lipid–water interface; (2) an interfacial region of intermediate polarity; and (3) a hydrophobic tail.

Lysophospholipids (LPL) are glycerophospholipids in which one acyl chain lacks.² Like phospholipids, these emulsifying and solubilizing agents lodge in small amounts in cellular membranes. These membrane-derived signaling molecules are generated by compartmentalized phospholipases, such as phospholipases PLA1, PLA2, and PLD, in addition to different lipase types [8]. Lysophospholipids and their receptors intervene in the reproduction, vascular development, and function of the nervous system. In particular, lysophosphatidylcholine, the most abundant lysophospholipids, participates in the regulation of gene transcription, cell division, monocyte chemotaxis, smooth muscle cell relaxation, and platelet activation. Lysophosphatidic acid is an intermediate mediator of transmembrane signal transduction.

Inositol-containing lipids are also strongly involved in signal transduction. Inositol lipids include phosphoinositides, i.e., phosphatidylinositol and its phosphorylated derivatives (in addition to inositol-containing ceramides in fungi and inositol glycolipids in pathogens). The metabolism of inositol-containing lipids is regulated by protein kinases and phosphatases.

Inositol phosphates are incorporated in this chapter (Sect. 2.3) because: (1) myoinositol ³ serves as a structural basis for inositol phosphates as well as lipids phosphatidylinositol (PI) and phosphatidylinositol phosphates (PIP), each category of molecules yielding numerous secondary messengers and (2) inositol phosphates, which constitute a group of more than 30 mono- and polyphosphorylated inositols, share some kinases and phosphatases with the 7 types of phosphoinositides.

Phosphoinositides and inositol phosphates modulate the recruitment and activation of regulators, thereby participating in the control of cell growth and proliferation, apoptosis, cytoskeletal dynamics, insulin signaling, vesicular transport, and nuclear function.

2.1 Lipids in Signal Transduction

Lipids serve as signaling effectors. Many signaling lipids, their enzymes, and effectors are common to multiple lipid signaling pathways. Numerous lipids act as intra- and extracellular messengers to control cell fate. As structural components of cell membranes, they are substrates for enzymes that generate second messengers involved in cell response to stimuli. Lipids can also modulate the activity of signaling proteic mediators. Moreover, they can contribute to the recruitment at effective sites of signaling effectors.

Signaling lipids include: (1) fatty acids;⁴ (2) eicosanoids and other products of arachidonic acid metabolism;⁵ (3) glycerolipid-derived regulators, such as phosphatidic acid (PA),⁶ lysophosphatidic acid (LPA),⁷ lysophosphatidylcholines (LPC), or lysolecithins,⁸ monoacylglycerol and diacylglycerol (DAG),⁹ anandamide, or ^Narachidonoylethanolamide¹⁰ and platelet-activating factor; as well as (4) phosphoinositides, or inositol phospholipids, such as the set of phosphatidylinositol mono-, bis-, and trisphosphates; (5) sphingolipids, such as sphingosine, sphingosylphosphorylcholine, sphingosine 1-phosphate, ceramide, and ceramide 1-phosphate.

Signaling lipids regulate cell metabolism, growth, proliferation, migration, senescence, and apoptosis. Growth factors and constituents of nutrients control the activity of enzyme-targeting lipids, such as phospholipases, prostaglandin synthases, 5-lipoxygenase (LOx), phosphoinositide 3-kinase (PI3K), sphingosine kinase (SphK), and sphingomyelinase (SMase).

Sphingomyelinase and sphingosine kinase are regulated enzymes that produce sphingosine 1-phosphate (S1P), a ligand for cognate G-protein-coupled receptors (Vol. 3 – Chap. 7. G-Protein-Coupled Receptors). The S1P pathway promotes, in particular, the survival of cells of the cardiovascular system.¹¹

Like sphingosine 1-phosphate, extracellular lysophospholipids ¹² interact with specific G-protein-coupled receptors¹³ to influence cell proliferation, differentiation, and motility. Lysophospholipids produced by phospholipase-A2 not only exert signaling functions, but also are processed into signaling mediators (e.g., lysophosphatidylcholine is converted to platelet-activating factor). Lysophosphatidic acid functions as an auto- and paracrine factor that binds to a set of receptors (LPARs; Vol. 3 – Chap. 7. G-Protein-Coupled Receptors). In addition, lysophosphatidic acid modulates gene expression by binding to lipid-sensing transcription activators such as peroxisome proliferator-activated receptors.

Phosphoinositides are synthesized or degraded using regulated enzymes excited by activated plasmalemmal receptors, such as phosphatidylinositol 3-kinases that generate the second messenger phosphatidylinositol trisphosphate and phospholipase-C that hydrolyzes the messenger phosphatidylinositol (4,5)-bisphosphate into inositol trisphosphate (IP₃), a Ca ${}^{++}$ mobilizer, and diacylglycerol, an activator of protein kinase-C, on the one hand, and activates protein kinase-B on the other (Sect. 5.2). Cardiac ion channel activity depends on PIP₂ density.

Membrane phospholipids are also substrates of phospholipase-A2 that generate arachidonic acid, the stem molecule for eicosanoids. Eicosanoids have a short half-life that ranges from seconds to minutes. The production of eicosanoids is initiated by the release of C20-polyunsaturated fatty acids such as arachidonic acid (rate-limiting stage) from phospholipids or diacylglycerol due to phospholipase-A2 in the presence of Ca ${}^{++}$ ions.

Among the prostaglandin synthases, prostaglandin-H2 synthase (PGH2S) and cyclooxygenases COx1 and COx2 lead to one of the 3 families of eicosanoids: prostaglandins, prostacyclins, thromboxanes. On the other hand, 5-lipoxygenase produces leukotrienes. Certain eicosanoids can bind to intracellular receptors (i.e., transcription factors), such as farnesoid X, liver X, and peroxisome proliferator-activated receptors (Vol. 3 – Chap. 6. Receptors).

Phosphatidylcholines are phospholipids that contain choline as a head group. They operate in membrane-mediated signaling. Phospholipase-D hydrolyzes phosphatidylcholine into phosphatidic acid, hence releasing the soluble choline head group into the cytosol. Lysophosphatidylcholines derive from partial hydrolysis of phosphatidylcholines, generally by phospholipase-A2, i.e., removal of one of the fatty acid groups.

2.2 Phosphoinositides

Although phosphoinositides account for a tiny fraction of cellular phospholipids (Table 2.3), they are required for signal transduction. They indeed recruit many signaling proteins in cellular membranes.¹⁴ The resulting signaling complexes participate not only in membrane transfer and cytoskeletal restructuring, but also cell survival, growth, and proliferation. In particular, phosphoinositides serve as mediators of calcium-mobilizing hormones and neurotransmitters.

Table 2.3

Approximate fractions of the total amount of phosphoinositides (PI) of various members of the families of mono- (PIP), bis- (PIP₂), and trisphosphate (PIP₃) derivatives (Source: [9]).

Contributors	Fraction (%) of
	total PI concentration
PI, PI(4)P, PI(4,5)₂	∼ 90
PI(3)P, PI(5)P	2–5
PI(3,5)₂	∼ 2 (fibroblast)
PI(3,4)₂, PI(3,4,5)₃	∼ 0 at rest
	↑ during stimulation
	(2–3)

Different phosphatidylinositol phosphates result from phosphorylation of single or multiple sites of the inositol ring of phosphatidylinositol: (1) phosphatidylinositol monophosphates PI(3)P, PI(4)P, and PI(5)P; (2) phosphatidylinositol bisphosphates PI(3,4)P₂, PI(3,5)P₂, and PI(4,5)P₂; and (3) phosphatidylinositol trisphosphate PI(3,4,5)P₃ (Tables 2.4 to 2.7).¹⁵ Multiple inositide kinases and phosphatases that interconvert phosphoinositides intervene in the sorting of molecules to specific organelles.

Table 2.4

Phosphatidylinositol (PI) and phosphatidylinositol phosphate (PIP) kinases (PIiK and PI(i)PjK); PI, PIP, and PIPP (phosphatidylinositol polyphosphate) phosphatases; and phospholipase-C (PLC) convert diverse types of phosphoinositides, which intervene in cell biology (MTMR: myotubularin-related phosphatase; PTen: phosphatase and tensin homolog deleted on chromosome 10; PTPmt: mitochondrial phosphoTyr phosphatase, previously named phospholipid-inositol phosphatase PLIP; SHIP: SH2-containing inositol polyphosphate 5-phosphatase; SKIP: skeletal muscle and kidney enriched inositol 5-phosphatase).

Phosphoinositide	Kinase	Phosphatase	Phospholipase
	(production)	(elimination)	and products
PI(3)P	PI3K	MTMR
PI(4)P	PI4K	4-Pase
PI(5)P	PI5K	PTPmt
PI(3,4)P₂	PI3K	PTen
	PI4K	PIP4Pase
PI(3,5)P₂	PI5K	PTPmt
		MTMR
PI(4,5)P₂	PI(5)P4K	IpgD	PLC: I(1,4,5)P₃ and DAG
	PI(4)P5K	5-Pase
PI(3,4,5)P₃	PI3K	PTen
	PI5K	SHIP, SKIP

Table 2.5

Phosphorylation and dephosphorylation (production and removal) of various types of phosphoinositides. (Part 1) Phosphoinositides (PI) and phosphatidylinositol trisphosphate (PI(3,4,5)P₃; PIKFYVE: FYVE finger-containing phosphoinositide kinase, or type-3 phosphatidylinositol 3-phosphate and phosphatidylinositol 5-kinase; IP(5)P: inositol polyphosphate 5-phosphatase PTen: phosphatase and tensin homolog deleted on chromosome 10 [ten], a phosphatidylinositol 3-phosphatase).

Substrate	Product	Enzyme
PI	PI(3)P	Class-3 PI3K
	PI(4)P	PI4K
	PI(5)P	PIP5K3 (PIKFYVE)
PI(3,4,5)P₃	PI(3,4)P₂	IP(5)P
	PI(4,5)P₂	PTen

Table 2.6

Phosphorylation and dephosphorylation (production and removal) of various types of phosphoinositides. (Part 2) Phosphatidylinositol monophosphates (PIP; PIPK: phosphatidylinositol phosphate kinase). The myotubularin superfamily of phosphatases — myotubularins (MTM) and myotubularin-related proteins (MTMR) — are inositol 3-phosphatases that dephosphorylate PI3P and PI(3,5)P₂.

Substrate	Product	Enzyme
PI(3)P	PI(3,4)P₂	PI4K
	PI(3,5)P₂	PIP5K3 (PIKFYVE)
		or PI(3)P5K (type-3 PIPK)
	PI	MTM, MTMR
PI(4)P	PI(3,4)P₂	Class-1 and -2 PI3Ks
	PI(4,5)P₂	PI(4)P5K (type-1 PIPK)
	PI
PI(5)P	PI(3,5)P₂	PI3K
	PI(4,5)P₂	PI(5)P4K (type-2 PIPK)

Table 2.7

Phosphorylation and dephosphorylation (production and removal) of various types of phosphoinositides. (Part 3) Phosphatidylinositol bisphosphates PIP₂ (PIC: phosphoinositidase-C [or phospholipase-C (PLC)]).

Substrate	Product	Enzyme
PI(3,4)P₂	PI(3,4,5)P₃	PI5K
	PI(3)P	4PTase
	PI(4)P	PTen
PI(3,5)P₂	PI(3)P	5PTase
	PI(5)P	MTM, MTMR
PI(4,5)P₂	PI(3,4,5)P₃	Class-1/2 PI3Ks
	PI(4)P	5PTase
	PI(5)P	4PTase
	IP₃, DAG	PLC
	IP₃, IP₆,	(PIC)
	IP₇, IP₈

Protein Tyr phosphatases usually dephosphorylate Tyr^P-containing proteins. Yet, some protein Tyr phosphatases dephosphorylate Ser^P– and Thr^P-containing proteins as well as phosphoinositides, such as phosphatase and tensin homolog deleted on chromosome 10 (PTen) and its homologs PTen2 and TPIP, as well as inositol polyphosphate 4-phosphatase-1 (IP(4)P1) and -2 (IP(4)P2) and the PTen-like phosphatase mitochondrial phosphoTyr phosphatase (PTPmt).

2.2.1 PI(4,5)P₂

The inositol region of phosphoinositides is used as a specific docking site for lipid-binding effectors of signaling cascades. Phosphatidylinositol (4,5)-bisphosphate is hydrolyzed by phospholipase-C into (1,2)-diacylglycerol and inositol (1,4,5)-trisphosphate. Both products act as second messengers that activate conventional protein kinase-C and release calcium ion from internal stores through IP₃R, respectively. Diacylglycerol that remains within the membrane to act as a PKC coactivator can also stimulate some of the canonical transient receptor potential channels (TRPC), independently of PKC enzyme.

Alternatively, PI(4,5)P₂ can be converted to phosphatidylinositol (3,4,5)-trisphosphate by phosphoinositide 3-kinase that is regulated by plasmalemmal receptors (GPCRs and RTKs; Vol. 3 – Chaps. 7. G-Protein-Coupled Receptors and 8. Receptor Kinases).

In addition, PIP₂ can also be hydrolyzed by phospholipases-D to generate phosphatidic acid, an activator of signaling mediators. Phosphatidic acid controls cell proliferation and survival via Son of sevenless and target of rapamycin, respectively. Last, but not least, PI(4,5)P₂ controls many ion channels and exchangers (Table 2.8).

Table 2.8

Activity of PI(4,5)P₂ (Sources: [10, 11]). Phospholipase-D (PLD) has 2 isoforms (PLD1–PLD2). The PLD1 subtype localizes primarily to vesicles; PLD2 mainly to the plasma membrane. Phospholipase-D interacts with PI(4)P 5-kinase-α(PI(4)P5Kα). Consequently, a local generation of PI(4,5)P₂ contributes to the regulation of PLD activity. Phospholipase-D hydrolyzes phosphatidylcholine to produce phosphatidic acid (PA), a lipidic second messenger. An intracellular effector of phosphatidic acid, the cytosolic enzyme sphingosine kinase (SphK), phosphorylates sphingosine to generate sphingosine 1-phosphate (S1P), another lipidic second messengers. This messenger stimulates cell proliferation. Transient receptor potential (TRP) channels open or close when they are bound to PI(4,5)P₂. Once receptors coupled to phospholipase-C are stimulated by agonist binding, PI(4,5)P₂ is degraded, and then TRPV1 opens and TRPM4, TRPM7, and TRPM8 close.

Effect	Mechanism of action
Generation of second messengers	Calcium release via IP₃
	Synthesis of PIP₃
	PLD–PA–SphK–S1P axis
Ion carrier activity	Inhibition of TRPV1
	Activation of TRPM4/7/8
	Inhibition of K⁺ channels
Intracellular transport	Endo- and exocytosis
	Phagocytosis

2.2.2 PI(3,4,5)P₃

The PI(3,4,5)P₃ concentration is negligible in resting cells, but rises transiently in response to activated GPCRs, growth factor RTKs, or cytokine receptors. Agent PI(3,4,5)P₃ modulates cell transport, growth, proliferation, and motility (Table 2.9). It indeed recruits guanine nucleotide-exchange factors and GTPase-activating proteins (Sect. 9.4) that regulate the actin cytoskeleton, as well as phosphoinositide-dependent protein kinase PDK1 (Sect. 5.2.2) and protein kinase-B (Sect. 5.2.6).

Table 2.9

Activity of PI(3,4,5)P₃ (Sources: [10]). This signaling mediator is, in particuler, an effector of insulin signaling to regulate energy uptake and storage. Agent PI(3,4,5)P₃ binds to various signaling mediators such as members of the TEC Tyr kinase family, Bruton Tyr kinase (BTK), and interleukin-2-inducible T-cell kinase (ITK). These kinases support the activity of phospholipase-Cγ (PLCγ), thereby fostering Ca ${}^{++}$ signaling. Agent PI(3,4,5)P₃ also stimulates phosphoinositide-dependent kinases PDK1 and PDK2 and protein kinase-B (PKB). The latter activates numerous effectors (B-cell lymphoma and leukemia [BCL2] antagonist of cell death [BAD], forkhead box transcription factor FoxO, glycogen synthase kinase GSK3, P70 ribosomal S6 kinase [S6K], and target of rapamycin [TOR]). It also activates monomeric CDC42, Rac, and Rho GTPases to remodel the cytoskeleton as well as to produce (via Rac) the free radical, superoxide anion (O2⁻).

Event	Mechanism of action
Cytoskeleton restructuring	Activation of CDC42, Rac, and Rho GTPases
Calcium signaling	Triggering of BTK (ITK)–PLCγ–IP₃ axis
Cell metabolism	Glycogen metabolism via GSK3
	Lipid synthesis
	Protein synthesis via S6K and TOR
Cell fate	Cell survival via BAD and TOR,
	hormonal modulation of apoptosis
	Cell growth via TOR
	Inflammation

Mediator PI(3,4,5)P₃ is dephosphorylated by 2 types of phosphatases. Phosphatase and tensin homolog deleted on chromosome 10 (PTen) regenerates PI(4,5)P₂ (Sects. 2.10.1 and 8.3.13.1). 5-Phosphatases, such as SH2 domain-containing inositol 5-phosphatases SHIP1 and SHIP2, or IP(5)Pd and IP(5)PL1, respectively, produce PI(3,4)P₂.

2.2.3 PI(5)P

Phosphoinositide PI(5)P exists at low concentrations. It is converted by PI(5)P4K2 into PI(4,5)P₂. However, the latter is mainly produced from PI(4)P. Conversion of PI(5)P into PI(4,5)P₂ by PIP4K2β impedes insulin signaling, as PIP4K2β activates PI(3,4,5)P₃ 5-phosphatases [12].

Bacterial IpgD 4-phosphatase can translocate into mammalian cells during bacterial invasion. It converts PI(4,5)P₂ into PI(5)P and provokes PKB activation.¹⁶

Agent PI(5)P enhances the activity of various myotubularins (MTM1, MTMR3, and MTMR6) toward their preferred substrate PI(3,5)P₂. The mitochondrial phosphoTyr phosphatase (PTPmt), a PTen-like phosphatase, produces PI(5)P by dephosphorylating PI(3,5)P₂ agent.

2.2.4 Nuclear Phosphoinositides

The nuclear envelope is a double membrane that contains phospholipids and proteins. Lipids also reside in the intranuclear space; phospholipids exist in the nucleus as proteolipid complexes. Polyphosphoinositol lipids, together with their synthesizing enzymes, form an intranuclear phospholipase-C signaling complex that generates diacylglycerol and inositol (1,4,5)-trisphosphate from phosphatidylinositol (4,5)-bisphosphate [28]. Diacylglycerol can recruit protein kinase-C to the nucleus to phosphorylate intranuclear proteins, whereas IP₃ can mobilize Ca ${}^{++}$ from the space between the 2 nuclear membranes, thereby increasing the nucleoplasmic Ca ${}^{++}$ level. Furthermore, PI(4,5)P₂ can participate in RNA splicing.

The PI–PLCβ1¹⁷ and PI–PLCδ4 complexes, diacylglycerol kinase, and PI and PIP kinases, as well as diacylglycerol, multiple PKC isoforms, phospholipase-A2, phosphoinositide 3-kinase, and sphingolipids, in addition to nuclear Ca ${}^{++}$ , contribute to nuclear lipid signaling. Enzymes of lipid synthesis localize to both the nuclear envelope and nucleoplasm [28]. Both type-1 PI(4)P5K and type-2 PI(5)P4K kinases reside in the nucleus. Phosphatidylcholine, phosphatidylethanolamine, and phosphatidylserine lodge within the nucleus. Nuclear phosphatidylglygerol contributes to PKC translocation and activation [28].

Phosphatidylcholine provides an additional source of diacylglycerol. The 2 distinct nuclear DAG pools generated from different phospholipids are assigned to subnuclear locations. Diacylglycerol kinase targets preferentially PI–PLC-derived DAG because of DAG accessibility rather than actual DAG specificity of DAG kinase [28]. The relative contributions of phosphatidylcholine and phosphoinositides to overall nuclear DAG concentration differ between distinct cell types.

Diacylglycerol kinase-ζ (DAGKζ) is a nuclear isoform following PKC-mediated phosphorylation (PKCα and/or PKCγ) [28]. Other isoforms of diacylglycerol kinase also translocate to the nucleus. Diacylglycerol kinase regulates nuclear PKC activity, as it removes the PKC activator diacylglycerol.

Phosphoinositide transfer protein-α (PITPα)¹⁸ that also localizes in the nucleus can have several functions, as it can bring phosphoinositides into the nucleus from the envelope, transport phosphoinositides to different sites within the nucleus, and/or serve to present more substrate to phosphoinositide kinases [28].

Insulin-like growth factor-1 (IGF1) stimulates nuclear inositide metabolism [28]. It transiently activates the nuclear PI–PLCβ1 complex and causes PKC translocation to the nucleus. The PI–PLC complex can also be phosphorylated (inhibited) by PKC (negative feedback loop). Extracellular signal-regulated kinases ERK1 or ERK2 can translocate into the nucleus and phosphorylate the PI–PLCβ1b complex.

In cardiomyocytes, whereas PKCα, PKCβ2, and PKCζ reside predominantly in the perinuclear region, PKCδ and PKCε shuttle between the cytosol and the nucleus. However, all PKC isoforms can translocate between these 2 main cellular compartments [28]. Isozyme PKCβ2 can phosphorylate lamin-B.

Nuclear PI3KR1 (P85 regulatory subunit) can be activated by nuclear guanosine triphosphatase PIKE (Sect. 9.3.28). The PI–PLCγ1 complex can act as a guanine nucleotide-exchange factor for PIKE agent. Besides, PI(3,4,5)P₃-binding protein (PIP₃BP) can translocate to the nucleus. In addition, type-2 PI3Ks — PI3KC2α and PI3KC2β— contain a nuclear localization sequence [28].

Phospholipase-A2 can translocate to the nucleus or nuclear envelope. Moreover, prostanoid receptors have a perinuclear distribution. Arachidonic acid (Sect. 2.6) can then serve near the nucleus. Phosphatidylcholine is a potential PLA2 substrate.

2.2.4.1 Cell Cycle and Nuclear Lipids

The concentration of nuclear DAG often increases close to the beginning of S phase. The PI–PLCβ1 complex is able to stimulate the cell cycle progression. A pulse of PI–PLC activation appears approximately at the G1–S transition. The PI–PLCβ1 complex yields nuclear DAG generation and PKC activation. It induces an increase in constitutive level of the CcnD3–CDK4 complex and retinoblastoma protein phosphorylation and its dissociation from E2F factor [28]. However, the absence of PI–PLCβ1 does not disturb the cell cycle progression, as most DAG derives from phosphatidylcholine.

The nuclear DAG level also increases at the G2–M transition. Kinase PKCβ2 indeed intervenes in nuclear disassembly, as it phosphorylates lamin-B. Conversely, DAGKζ suppresses nuclear DAG in non-proliferating cells.

2.2.5 Phosphoinositide-Dependent Enzymatic Activity

Phosphatase PTen dephosphorylates only PIP₃ into PI(4,5)P2. Activation of PTen that depends on PI(4,5)P₂ thus yields a positive feedback.

Myotubularin that produces PI(5)P by dephosphorylating PI(3,5)P₂ is upregulated by PI(5)P agent. Phosphoinositides that act as signaling molecules are involved in the coupling of voltage sensing and enzymatic activity of voltage-sensing phosphoinositide phosphatase (VSP).

2.2.6 Phosphoinositide Influence on Ion Carriers

Localized PIP₂ depletion can occur close to receptor-activated phospholipase-C, thereby inactivating adjacentK_IR orK_V channels (Table 2.10). However, activated phospholipase-C can be associated with activated phosphoinositide (PIK) and phosphoinositide phosphate (PIPK) kinases that result from stimulation of PLC-coupled receptors. Consequently, the PIP₂ concentration can locally rise.

Table 2.10

Regulation mediated by PIP₂ of cardiac ion carriers (Source: [30]; GIRK: Gβγ-regulated inwardly rectifying K⁺ channel; HCN: hyperpolarization-activated, cyclic nucleotide-gated K⁺ channel; hERG: human ether-à-go-go related channel; K_IR: inward rectifier K⁺ channel; IKr: rapidly activating delayed rectifier K⁺ current; IKs: slowly activating delayed rectifier K⁺ current; If: hyperpolarization-activated cation current; minK: auxiliary, kinetics-slowing, inactivation-inhibiting, conductance-raising β subunit MiRP: MinK-related peptide; NCX: Na⁺–Ca ${}^{++}$ exchanger; NSCC: Ca ${}^{++}$ -activated non-selective cation current; ROC: receptor-operated Ca ${}^{++}$ current; SOC: store-operated Ca ${}^{++}$ current; SuR: sulfonylurea receptor; TRP: transient receptor potential current activated by several types of stimuli; TRPC: TRP canonical; TRPM: TRP melastatin).

Protein	PIP₂ effect	Function
Potassium channels
K_IR2.1	Channel opening	Resting membrane
		potential maintenance (i _K1)
K_IR3.1/4	Gβγ-mediated	Parasympathetic
(GIRK1/4)	regulation change	stimulation ( ${i}_{\mathrm{{K}_{ACh}}}$ )
K_IR6.2	ATP-binding	ATP-induced closing ( ${i}_{\mathrm{{K}_{ATP}}}$ )
and SUR2a	exclusion	Repolarization reserve
K_V11.1	Activation shift	Promote repolarization (i _Kr)
(hERG)	(hyperpolarization)	Maintenance of
minK,	Slow deactivation	pacemaker automaticity
and MiRP1
K_V7.1	Prevention of	Repolarization (i _Ks)
and minK	inhibition
Potassium and sodium channels
HCN2/4	Activation shift	Spontaneous diastolic
	(toward depolarization)	depolarization (If)
Calcium and sodium channels
NCX1	Prevention of	Ca ${}^{++}$ extrusion or uptake
	auto-inhibition	(i _Na∕Ca)
	(binding to NCX	Spontaneous diastolic
	inhibitor)	depolarization
Non-selective polymodal channels
TRPC1–C7	Inhibition?	Pacemaker current?
		Stretch activation?
TRPM4	Depletion-induced	Depolarization
	desensitization

2.2.6.1 Inwardly Rectifying Potassium Channels

In heart, 3 major families ofinwardly rectifying K⁺ channel — K_IR2, K_IR3, and K_IR6 — require PIP₂ for activity.¹⁹ Channels of the K_IR class open after PIP₂ binding, whereas they close under PIP₂ depletion [30]. Most K_IR channels respond relatively specifically to PI(4,5)P₂ lipid. However, K_ATP channels are sensitive to PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂, and PIP₃ molecules.

2.2.6.2 Repolarizing Potassium Channels

Voltage-gated K⁺ channels (K_V) are responsible for cardiomyocyte repolarization. Both K_V11.1²⁰ and K_V7.1²¹ that are involved in rapid and slow phase of repolarization are activated by PIP₂ [30].

2.2.6.3 Pacemaker Channels

Hyperpolarization-activated, cyclic nucleotide-gated channel (HCN) is regulated by PIP₂ lipid [30]. The latter shifts HCN voltage dependence toward depolarized potentials, hence increasing the spontaneous firing rate.

2.2.6.4 Sodium–Calcium Exchanger

Sodium–calcium exchanger (NCX) contains an inhibitory domain that auto-inhibits the channel. Lipid PIP₂ binds to the inhibitory sequence of NCX1 exchanger, thereby preventing its auto-inhibition and activating the carrier [30].

2.2.6.5 Transient Receptor Potential Channels

Canonical transient receptor potential, low conductance, relatively non-selective cation channels, are activated by receptors coupled to phospholipase-C. They are also regulated by stretch,²² diacylglycerol, and IP₃R-induced Ca ${}^{++}$ store depletion. The TRPC3 channel can yield a pacemaker current modulated by IP₃ or PIP₂.²³ Channel TRPC4 can be inhibited by PIP₂ binding, whereas TRPC6²⁴ and TRPC7 are activated by PIP₂. In the sinoatrial node, TRPM4 channel is a calcium-activated, non-selective cation channel activated by IP₃ [30].

2.3 (Non-Lipid) Inositol Phosphates

Inositol (1,4,5)-trisphosphate (IP₃), a Ca ${}^{++}$ -mobilizing second messenger, undergoes a catabolism that terminates its action. Its metabolism is governed by a complicated inositol phosphate pathway (Tables 2.11 to 2.14). The inositol phosphate metabolism produces free inositol that enables synthesis of phosphoinositides, which can serve in further signaling. In addition, it generates a collection of inositol polyphosphates that also contribute to intracellular signaling.

Table 2.11

Phosphorylation and dephosphorylation (production and removal) of various types of inositol phosphates. (Part 1) Inositol monophosphates (Source: [10]). The free inositol synthesized by inositol monophosphatase (IMPase), which is inhibited by the lithium ion (Li⁺), is reused to resynthesize phosphorylated phosphatidylinositol, which can relocalize to the plasma membrane for phosphoinositide signaling.

Substrate	Product	Enzyme
I(1)P	Inositol	IMPase
I(3)P	Inositol	IMPase
I(4)P	Inositol	IMPase

Table 2.12

Phosphorylation and dephosphorylation of various types of inositol phosphates. (Part 2) Inositol bisphosphates (Source: [10]; IPMK: inositol phosphate multikinase [a.k.a. inositol (1,3,4,6)-tetrakisphosphate 5-kinase]).

Substrate	Product	Enzyme
I(1,3)P₂	I(1)P	IP(3)Pase
I(1,4)P₂	I(4)P	IP(1)Pase
I(3,4)P₂	I(3)P	IP(4)Pase
I(4,5)P₂	I(1,4,5)P₃	IPMK

Table 2.13

Phosphorylation and dephosphorylation of various types of inositol phosphates. (Part 3) Inositol trisphosphates (Source: [10]; IP₃6K: I(1,3,4)P₃ 6-kinase; IPMK: inositol phosphate multikinase; OCRL: oculocerebrorenal syndrome of Lowe [or inositol polyphosphate 5-phosphatase IP(5)Pf encoded by the INPP5F gene]).

Substrate	Product	Enzyme
I(1,3,4)P₃	I(1,3,4,6)P₄	IP₃(5,6)K (I(1,3,4)P₃5K/6K)
	I(1,3,4,5)P₄	IP₃(5,6)K (I(1,3,4)P₃5K/6K)
	I(1,3)P₂	IP(4)Pase
	I(3,4)P₂	IP(1)Pase
I(1,4,5)P₃	I(1,3,4,5)P₄	IMPK, IP₃3K
(IP₃)	I(1,4)P₂	5PTase1 (IP(5)Pa), OCRL (IP(5)Pf)

Table 2.14

Phosphorylation and dephosphorylation of various types of inositol phosphates. (Part 4) Inositol tetrakisphosphates and higher phosphorylated derivatives, i.e., inositol penta- (IP₅), hexa- (IP₆; phytic acid or phytate; concentration 10–100 μmol in mammalian cells), hepta- (IP₇), and octakisphosphatekisphosphate (IP₈; a.k.a. bispyrophosphorylated inositol phosphate and bisdiphosphoinositol tetrakisphosphate [(PP)₂IP₄]; Source: [10]; DIPP: diphosphoinositol phosphate phosphohydrolase; IP₄5K: I(1,3,4,6)P₄ 5-kinase; IPMK: inositol phosphate multikinase; MIPP: multiple inositol phosphate phosphatase [I(1,3,4,5,6)P₅/IP₆ 3-phosphatase]; OCRL: oculocerebrorenal syndrome of Lowe). Enzyme IP₆K, an IP₅/IP₆/IP₇ 5-kinase, generates diphosphoinositol phosphates (PPIP). Diphosphate is either at C4, C5, or C6, hence the designation (4)PPIP₅ (or IP ${}_{{7}_{(4)}}$ ], (5)PPIP₅ (or IP ${}_{{7}_{(5)}}$ ), and (6)PPIP₅ (or IP ${}_{{7}_{(6)}}$ ), respectively. Inositol hexakisphosphate (IP₆) and diphosphoinositol pentakisphosphate (PPIP₅) kinase-1, or diphosphoinositol pentakisphosphate kinase-1 (PPIP₅K1; a.k.a. histidine acid phosphatase domain-containing protein HisPPD2a, VIP1 homolog), is a bifunctional IP₆/IP₇ kinase that yields (i)PPIP₅ (IP ${}_{{7}_{(i)}}$ ) and (PP)₂IP₄ (IP₈).

Substrate	Product	Enzyme
I(1,3,4,5)P₄	I(1,3,4,5,6)P₅	IMPK
	I(1,3,4)P₃	5PTase1 (IP(5)Pa), OCRL (IP(5)Pf)
		IP(5)Pd (SHIP1)
I(1,3,4,6)P₄	I(1,3,4,5,6)P₅	IMPK (I(1,3,4,6)P₄5K I(1,3,4,5,6)P₅1Pase)
I(1,4,5,6)P₄	I(1,3,4,5,6)P₅	I(1,4,5,6)P₄3K
I(3,4,5,6)P₄	I(1,3,4,5,6)P₅	I(3,4,5,6)P₄1K (or I(1,3,4,5,6)P₅1Pase)
IP5	IP₆	IMPK, IP(1,3,4,5,6)P₅2K (IPK1)
	PPIP₄	IP₆K
	I(1,4,5,6)P₄	PTen
	I(3,4,5,6)P₄	I(1,3,4,5,6)P₅1Pase
IP6	IP7	IP₆K
	IP5	MIPP
IP7	IP8	PPIP5K
(PPIP₅)	IP6	DIPP
IP8	IP7	DIPP
([PP]₂IP₄)

Inositol (1,4,5)-trisphosphate is converted to numerous inositol phosphates, such as inositol tetrakisphosphate (IP₄), pentakisphosphate (IP₅), hexakisphosphate (IP₆), and pyrophosphate (diphosphoinositol phosphate [PPIP]) owing to inositol phosphate kinases (IPK; Fig. 2.1).

Fig. 2.1

Inositol phosphates and inositol phosphate kinases (IP₃: inositol trisphosphate; IP₄: inositol tetrakisphosphate; IP₅: inositol pentakisphosphate; IP₆: inositol hexakisphosphate; IP₆/IP₇K: inositol hexakisphosphate and heptakisphosphate kinase of the IP₆K/Kcs1 and IP₆K/VIP1–VIP2 categories; IP₇: inositol heptakisphosphate [monopyrophosphorylated inositol phosphate or diphosphoinositol pentakisphosphate (PPIP₅)]; IP₈: inositol octakisphosphate [bispyrophosphorylated inositol phosphate or bisdiphosphoinositol tetrakisphosphate ([PP]₂IP₄)]; IPK2: inositol polyphosphate 6-/3-/5-kinase; IPPK: inositol (1,3,4,5,6)-pentakisphosphate 2-kinase [a.k.a. IP₅2K and IPK1]; PI3K: phosphatidylinositol 3-kinase; PIP₂: phosphatidylinositol bisphosphate; PIP₃: phosphatidylinositol trisphosphate).

In addition to PI3K kinases, inositol polyphosphate multikinase (IPMK)²⁵ generates the second messenger phosphatidylinositol (3,4,5)-trisphosphate as well as water soluble inositol phosphates. Enzyme P110_PI3K phosphorylates (activates) IPMK kinase [13]. Therefore, growth factors stimulate the PI3K–PKB axis and generate PIP₃ via the sequential activations of P110_PI3K and IPMK kinases. As inositol phosphates prevent PKB signaling, IPMK either inhibits or stimulates PKB via its inositol phosphate kinase or PI3K activity, respectively.

Inositol hexakisphosphate is a coactivator of nucleoporin Gle1 that is involved in mRNA export from the nucleus to the cytoplasm. Protein Gle1 interacts with the nuclear-pore complex NuP155 and activates Dead-box ATPase DBP5 in nuclear-pore complex to export mRNA into the cytoplasm [14, 15]. Agent IP₆ binds Gle1 to potentiate Gle1 stimulation of DBP5. It contributes to the conformation of RNA-editing adenosine deaminase ADAR2 [16]. Inositol hexakisphosphate also acts as a cofactor for adenosine deaminase that acts on transfer RNA (i.e., that edits tRNA) ADAT1.

Inositol phosphates and pyrophosphates participate in various cellular activities. Inositol trisphosphate (IP₃) is phosphorylated into inositol tetrakisphosphate that is required for the activation of ITK kinase after T-cell receptor stimulation in CD4 + , CD8 + thymocytes [17]. Inositol tetrakisphosphate (IP₄), pentakisphosphate (IP₅), and hexakisphosphate (IP₆) modulate the activities of several chromatin-remodeling complexes, either preventing or favoring nucleosome mobilization [18].

Inositol pyrophosphate synthases target IP₆ and IP₇ kinases to regulate the actin-related protein ARP2–ARP3 complexes [19].²⁶ Inositol pyrophosphates diphosphoinositol pentakisphosphate (IP₇) and bisdiphosphoinositol tetrakisphosphate (IP₈) contribute to the regulation of the phosphorylation of many proteins [21]. Heat shock protein HSP90 binds to inositol hexakisphosphate kinase IP₆K2 that produces IP₇ and mediates apoptosis to inhibit IP₆K2 and promote cell survival [22]. In addition, inositol pyrophosphates are involved in vesicular transport.²⁷

Protein kinase-B and calcium released through inositol trisphosphate receptors are involved in pathways that lead to either cell survival or death. Once it is activated by survival signals, PKB phosphorylates IP₃Rs to reduce Ca ${}^{++}$ flux from the endoplasmic reticulum to mitochondria [23]. Large IP₃-induced Ca ${}^{++}$ release can indeed promote apoptosis. Kinases PKA and CamK2 also target IP₃Rs, but cause low levels of IP₃R phosphorylation.

2.3.1 I(1,4,5)P₃

Inositol (1,4,5)-trisphosphate (IP₃) is dephosphorylated by type-1 inositol polyphosphate 5-phosphatase to form I(1,4)P₂ or phosphorylated by IP₃ 3-kinase to generate I(1,3,4,5)P₄. Inositol trisphosphate triggers Ca ${}^{++}$ influx, especially from its intracellular stores, to launch a set of cellular processes (Table 2.15). Three isoforms of IP₃ 3-kinase exist (IP₃3Ka–IP₃3Kc). Both IP₃3Ka and IP₃3Kb are activated by Ca ${}^{++}$ –calmodulin, IP₃3Kb being the most sensitive. Protein kinase-A activates IP₃3Ka, but inhibits IP₃3Kb [10].

Table 2.15

Role of the IP₃–Ca ${}^{++}$ pathway (Source: [10]).

Opening of Ca ${}^{++}$ -sensitive ion channels
Cell contraction ( stress fibers, sarcomeres)
Exocytosis (fluid secretion, aldosterone release)
Synapse remodeling
Metabolism (liver)
Cell differentiation
Chemotaxis
Platelet aggregation
Fertilization

2.3.2 Main Enzymes of the Metabolism of Inositol Polyphosphates

2.3.2.1 Kinases of Inositol Polyphosphates

Inositol phosphate kinases include: (1) IPK1 (a.k.a. IP52K and IPPK), an IP₅ 2-kinase that synthesizes IP₆; (2) inositol polyphosphate 3-/5-/6-kinase IPK2 that produces I(1,3,4,5,6)P₅; (3) IP₃ 3-kinase (IP₃3K) that manufactures I(1,3,4,5)P₄ from I(1,4,5)P₃; (4) IP₃(5/6)K, an I(1,3,4)P₃ 5- and 6-kinase that forms I(1,3,4,5)P₄ and I(1,3,4,6)P₄; (5) IP₆K, an IP₅, IP₆, and IP₇ 5-kinase that generates diphosphoinositol phosphates (i.e., PPIPs); and (6) inositol hexakisphosphate (IP₆) and diphosphoinositol pentakisphosphate (PPIP₅) kinase-1 (IP₆K or PPIP₅K1),²⁸ which is a bifunctional IP₆ and IP₇ kinase that yields (i)PPIP₅ (IP ${}_{{7}_{(i)}}$ )²⁹ and (PP)₂IP₄ (IP₈). In addition to PPIP5K1, another PPIP₅K2 subtype exist.³⁰

Kinase IP₆K possesses 3 isoforms (IP₆K1–IP₆K3). The I(3,4,5,6)P₄ 1-kinase, or I(1,3,4,5,6)P₅ 1-phosphatase, convert I(3,4,5,6)P₄, a regulator of chloride channels. Inositol phosphate multikinase (IPMK) transforms I(4,5)P₂ into I(1,4,5)P₃, I(1,4,5)P₃ into I(1,3,4,5)P₄, I(1,3,4,5)P4 into IP₅, and IP₅ into PPIP₄.

The PPIP₅ kinase targets PPIP₅ (IP₇) to form (PP)₂IP₄ (IP₈). On the other hand, diphosphoinositol phosphate phosphohydrolase (DIPP) dephosphorylates both PPIP₅ (IP₇) and (PP)₂IP₄ (IP₈).

Inositol phosphate kinases can transduce cues received by GPCRs [24]. They are required for tissue development and cell adaptation. Activated Gα_q stimulates the formation of metabolites downstream from IP₃ (down to IP₈). Messengers IP and PPIP, which depend on IPK, can be involved in RNA production, mRNA export, chromatin remodeling, protein phosphorylation, and immunity.

Inositol (1,3,4)-trisphosphate [I(1,3,4)P₃] 5/6-kinase (ITPK1) synthesizes inositol tetrakisphosphates I(1,3,4,5)P₄ and I(1,3,4,6)P₄. It is not only phosphorylated, but also acetylated (Lys340, Lys383, and Lys410) by acetyltransferases CBP or P300 that reduce the ITPK1 half-life and enzyme activity [25]. On the other hand, ITPK1 is deacetylated by silent information regulator-2 (SIRT1).

2.3.2.2 Inositol Phosphate Phosphatases

Inositol phosphate phosphatases encompass, in particular: (1) IP(1)Pase, encoded by the INPP1 gene, which is an I(1,4)P₂ and I(1,3,4)P₃ 1-phosphatase; (2) 5PTase, an I(1,4,5)P₃ and I(1,3,4,5)P₄ 5-phosphatase; (3) PTen, a 3-phosphatase that also dephosphorylates IP₅; and (4) DIPP that degrades PPIPs.

5-Phosphatase-1 hydrolyzes I(1,4,5)P₃ and I(1,3,4,5)P₄ inositol phosphates. This enzyme is phosphorylated (inhibited) by Ca ${}^{++}$ –calmodulin-dependent protein kinase CamK2 [10]. 5-Phosphatase-2 hydrolyzes inositol phospholipids.

2.3.2.3 Inositol Monophosphatases

Inositol monophosphatase (IMPase) dephosphorylates inositol monophosphate to generate free inositol, a precursor of membrane phosphatidylinositols and its phosphorylated derivatives, the phosphatidylinositol phosphates. Cells also produce inositol using glucose 6-phosphate that is isomerized by inositol 1-phosphate synthase-A1 (ISynA1)³¹ to produce inositol 1-phosphate, which is dephosphorylated by inositol monophosphatase to generate free inositol.

Three subtypes of inositol 1 (or 4)-monophosphatases are encoded by 3 distinct genes: lithium-sensitive IMPase1 (IMP1 or isoform-A1 [IMPa1]); IMPase2 (IMP2 or isoform-A2 [IMPa2]); and IMPase3 (IMP3 or isoform-A3 [IMPa3]).³² Subtype IMPa2 homodimerizes, but does not heterodimerize with IMPa1 isoform [26]. In addition, IMPa2 has a significantly lower activity on inositol monophosphate than IMPa1 (Table 2.16). Calbindin-1, a Ca ${}^{++}$ -binding protein, interacts with IMPase (apparent equilibrium dissociation constant 0.9 μmol) and activates IMPase up to 250-fold, according to H⁺ and substrate concentrations [27].

Table 2.16

Substrate specificity of Mg ${}^{++}$ -dependent inositol monophosphatase IMPase1 and IMPase2. Relative catalytic activities are scaled between 1 and 10 (Source: [26]).

Type	IMPase1	IMPase2
I(1)P	10	2
I(2)P	2	1
I(3)P	10	2
I(4)P	10	2
I(5)P	10	2
I(6)P	10	2

2.3.2.4 Inositol Polyphosphate Phosphatases

Inositol polyphosphate 1-phosphatase (IP(1)Pase) dephosphorylates both I(1,4)P₂ and I(1,3,4)P₃. Inositol polyphosphate 3-phosphatase (IP(3)Pase) dephosphorylates I(1,3)P₂ and PI3P. Inositol polyphosphate 4-phosphatase (IP(4)Pase; Sect. 2.10.3) dephosphorylates both I(3,4)P₂ and I(1,3,4)P₃, as well as PI(3,4)P₂ lipid.

Inositol polyphosphate 5-phosphatases (IP(5)Pase; Sect. 2.10.4) constitute a family of 10 mammalian members. These enzymes regulate many important cellular functions, such as hematopoietic cell proliferation and activation, insulin signaling, endocytosis, and actin polymerization.

Among members of the IP(5)Pase family, inositol polyphosphate 5-phosphatase OCRL (or IP(5)Pf) hydrolyzes both I(1,4,5)P₃ and I(1,3,4,5)P₄ inositol phosphates and PI(4,5)P₂ and PI(3,4,5)P₃ phosphoinositides. Src homology-2 (SH2) domain-containing inositol phosphatases (SHIP1 and SHIP2), i.e., IP(5)Pd and IP(5)PL1, target I(1,3,4,5)P₄ and PI(3,4,5)P₃.

2.4 Choline-Containing Phospholipids

Choline-containing phospholipids (phosphatidylcholine, sphingomyelin, and ether phospholipids³³ with a choline head group) are the most abundant phospholipids ( ∼ 50% of the phospholipid pool) in cellular membranes [31]. They serve as structural components of membranes as well as signaling molecules and precursors of secondary messengers. Choline enters cells through specific membrane transporters and can be phosphorylated by choline kinase.

2.5 Sphingolipids

Sphingolipids constitute a class of lipids that derive from sphingosine, an amino alcohol with an unsaturated hydrocarbon chain that can be phosphorylated. Ceramide is the structural unit common to all sphingolipids. Sphingolipids participate in signal transmission and cell recognition, as they can serve as surface markers.

Two types of sphingolipids exist according to head group type: (1) sphingomyelins (or ceramide phosphorylcholines) that reside in cell membranes and have a phosphorylcholine or phosphoethanolamine linked to sphingosine bonded to a fatty acid (ceramide) and (2) glycosphingolipids that are ceramides with at least one oligosaccharide. Glycosphingolipids comprise: (1) cerebrosides such as myelin that possess a single glucose or galactose, sulfatides being sulfated cerebrosides, and (2) gangliosides with at least 3 oligosaccharide glucids, including one (monosialogangliosides) or more (polysialogangliosides) sialic acids. Glycosphingolipid sulfates³⁴ are glycosphingolipids that contain a sulfate ester group attached to the carbohydrate moiety.

The ganglioside family encompasses mono- (GM1–GM3), di- (GD1a/b and GD2–GD3), tri- (GT1b), and quadrisialoganglioside (GQ1b). Major signaling sphingolipids include ceramide, ceramide 1-phosphate, sphingosine, sphingosine 1-phosphate, sphingosine phosphorylcholine (SPC), and lysosulfatides (sulfogalactosylsphingosine) that have lost their fatty acid constituent.

Members of the orosomucoid (OrM)³⁵ family are encoded by the ORM and ORMDL genes. Proteins of the OrM1-like subfamily include 3 identified members (OrMdL1, OrMdL2 [adoplin-2], and OrMdL3).³⁶ Members of the ORM gene family encode transmembrane proteins located in the endoplasmic reticulum, where the synthesis of sphingolipids begins. Proteins of OrM1-like family are inhibitors of sphingolipid synthesis, as they form a complex with serine palmitoyltransferase, the first and rate-limiting enzyme in sphingolipid production in the endoplasmic reticulum.³⁷ Phosphorylation of OrM1-like proteins relieves their inhibitory activity [32].

2.5.1 Ceramide

Ceramide not only exerts biological effects, but also can be processed into bioactive sphingolipids. Ceramide synthesized by sphingomyelinases is indeed deacylated by neutral ceramidase into sphingosine. The latter is then phosphorylated into sphingosine 1-phosphate by sphingosine kinases. Ceramide and sphingosine are primarily antiproliferative and pro-apoptotic, whereas S1P promotes cell proliferation and impedes apoptosis.

2.5.2 Sphingomyelinases

Sphingomyelinases hydrolyze ubiquitous membrane-associated sphingomyelin into ceramide and phosphocholine. In the cardiovascular system, sphingomyelinases act in cardiomyocytes and vascular endothelial and smooth muscle cells not only to contribute to the regulation of cell proliferation and death, but also contraction of cardiac and vasculomyocytes [33].

Three main types of sphingomyelinases exist according to optimal pH for their activity [33]: (1) alkaline sphingomyelinases that are synthesized in intestinal mucosa and liver [33]; (2) acidic sphingomyelinases; and (3) Mg ${}^{++}$ -dependent, membrane-associated, neutral sphingomyelinases (nSMase; optimal pH 7.4). Acidic sphingomyelinases that are encoded by the single sphingomyelin phosphodiesterase SMPD1 gene and generated from a single precursor are subdivided into 2 functionally distinct forms: (1) Zn ${}^{++}$ -bound, lysosomal, acidic (_l,acSMase; 4.5 < pH optimum < 5) and (2) Zn ${}^{++}$ -dependent, secreted, acidic (_s,acSMase).

These types of enzymes have common activators, but differ in their enzymatic properties and subcellular locations [33]. Neutral sphingomyelinases reside in the endoplasmic reticulum and Golgi body. They also hydrolyze sphingomyelin in the inner leaflet of the plasma membrane. Lysosomal acidic sphingomyelinase can relocate to the outer leaflet of the plasma membrane. In human lymphocytes, members of the TNFR superfamily (Vol. 3 – Chap. 11. Receptors of the Immune System) actually trigger _l,acSMase translocation from lysosomes to the extracellular surface of the plasma membrane. Consequently, several types of sphingomyelinases act on both sides of the plasma membrane, i.e., in the cytosol and extracellular matrix.

In humans, once stimulated by inflammatory cytokines, such as interferon-γ and interleukin-1β, endothelial cells secrete large amounts of _s,acSMase with reduced _l,acSMase amount. Cytokine-induced s,acSMase release actually antagonizes lysosomal axis with a common precursor [33].

Three genes (SMPD2–SMPD4) encode neutral sphingomyelinases (nSMase1–nSMase3). Paralog nSMase2 (encoded by the SMPD3 gene) is ubiquitous. Isoform nSMase3, highly expressed in the heart, is targeted by the complex formed by TNFR1 receptor and adaptor factor associated with neutral sphingomyelinase activation (FAN) [33]. Neutral sphingomyelinase-derived ceramide and activated PKCξ can trigger hypoxic pulmonary vasoconstriction.

Translocated _l,acSMase localizes to sphingolipid-rich membrane rafts and releases extracellularly oriented ceramide. Ceramide-enriched plasmalemmal platforms cluster receptors for apoptosis initiation as well as, in vascular endothelial and smooth muscle cells, contribution of TNFSF6-induced impairment of the vasodilator response and muscarinic M₁ receptor-mediated constriction [33].

Neutral sphingomyelinase, but not acid sphingomyelinase, that produces ceramide and activation of PKCζ is an early signaling event in acute hypoxia-induced pulmonary vasoconstriction [34].

2.5.3 Sphingosine Kinases

Sphingosine kinases SphK1 and SphK2 control cell fate, especially in the cardiovascular system (Table 2.17). They are phosphorylated (activated) by extracellular signal-regulated protein kinases ERK1 and -2 (Sect. 6.7.2.2). Sphingosine kinase SphK2 resides in the nucleus. Its nuclear export happens after phosphorylation by protein kinase-D [35]. Enzyme S1P phosphohydrolase-1 intervenes in the recycling of sphingosine into ceramide.

Table 2.17

Effect of sphingosine kinases SphK1 and SphK2 via S1P in various cell types of the cardiovascular system (Sources: [35, 36]; − : negative effect; SMC: smooth muscle cell). Sphingosine 1-phosphate also acts in an auto- and paracrine manner via its cognate receptors.

Type	Endothelial cell	Cardiomyocyte	Fibroblast	SMC
SphK1	Survival	Survival	Survival	Survival
(S1P)	Proliferation	Hypertrophy	Proliferation	Contraction
	Migration	Inotropy −	Migration	Migration inhibition
	Angiogenesis	Chronotropy −
	Permeability
SphK2	Survival	Survival	Survival	Survival
	Apoptosis	Apoptosis	Apoptosis	Apoptosis

Paralogs SphK1 and SphK2 contain several splice variants. Isoforms SphK1 and SphK2 might have distinct subcellular locations, hence directing spatially restricted S1P production. In addition, SphK2 partly antagonizes SphK1, as SphK2 also contributes to conversion of sphingosine into ceramide [35].

In addition to PKCε, sphingosine kinases are stimulated by G-protein-coupled receptor ligands (e.g., acetylcholine, histamine, and S1P), receptor Tyr and Ser/Thr kinase agonists (e.g., EGF, PDGF, VEGF, TGFα, and TGF-β), immunoglobulin receptors, interleukins, and estrogens [35]. Moreover, sphingosine kinase-1 is stimulated by HIF1α and HIF2α, but reactive oxygen species lead to SphK1 degradation [35].

Sphingosine kinases are also stimulated by δ1-catenin–δ2-catenin, aminocyclase-1, and eukaryotic elongation factor-1A [35]. On the other hand, they are inhibited by SphK1-inteacting protein (SKIP), platelet endothelial adhesion molecule PECAM1, and LIM-only factor FHL2 (or SLIM3). Both SphK1 and SphK2 bind calmodulin that allows their translocation from the cytosol to the plasma membrane.

After injury, fibroblasts transform into myofibroblasts and produce extracellular matrix. The TGFβ-stimulated production of collagen by cardiac fibroblasts is mediated by S1P produced by SphK1 and released to act as an auto- and paracrine agent via S1P₂ receptors [37].

2.5.4 Sphingosine 1-Phosphate

Sphingosine 1-phosphate is a bioactive lipid metabolite that causes intracellular calcium mobilization and intervenes in cell survival, proliferation, migration, and contraction (Table 2.17). In general, it operates via ubiquitous S1P G-protein-coupled receptors S1P₁ to S1P₃ (Vol. 3 – Chap. 7. G-Protein-Coupled Receptors); Table 2.18). Sphingosine 1-phosphate can be dephosphorylated by S1P phosphatase into sphingosine or degraded by S1P lyase into hexadecanal and phosphoethanolamine.

Table 2.18

Signaling pathways from S1P receptors in heart and blood vessels (Sources: [36, 38]; ACase: adenylate cyclase; CMC: cardiomyocyte; GSK: glycogen synthase kinase; K_ACh: acetylcholine-activated inwardly rectifying potassium channel; NOS: nitric oxide synthase; PI3K: phosphatidylinositol 3-kinase; PKA, PKB, PKC: protein kinases-A, -B, and C; PTen: phosphatase and tensin homolog deleted on chromosome 10; VSMC: vascular smooth muscle cell). Receptors of S1P include S1P₁ to S1P₃. Receptor S1P₁ couples to Gi protein and S1P₂ and S1P₃ to Gi, Gq, and G12/13 proteins. Prolonged stimulation of Gq-coupled receptors (e.g. α1-adrenergic and endothelin-1 receptors) provokes hypertrophy. Subunits G12 and G13 also inhibit JNK and cause complementary hypertrophy.

Type	Pathway	Effect
S1P₁	Gi–ACAse	Negative inotropy via cAMP and PKA
		inhibition of Ca_V1 channel
	Gβγ–K_ACh	Bradycardia
	PI3K–PKB–GSK3β	Cardioprotection by GSK3 inhibition
	Gi–Rac1–NOS3–NO	Endothelium-dependent vasodilation
	Gi–PI3K	Inhibition of endothelium permeability
S1P₂	G12/13–RhoA	Fibroblast proliferation
	Gβγ–PKB–NOS	Cardioprotection
	RhoA–RoCK–PKC	Basal vasomotor tone (VSMC)
	Rho–PTen	PI3K inhibition; endothelium permeability
S1P₃	Gq–PLC	CMC hypertrophy
	G12/13–JNK	CMC concomitant hypertrophy
	Gβγ–PKB–NOS	Cardioprotection
	Rac1–NOS3–NO	Endothelium-dependent vasodilation
	RhoA–RoCK–PKC	VSMC contraction

2.5.4.1 HDL-Associated Sphingosine 1-Phosphate

High-density lipoproteins carry not only cholesterol in blood, but also sphingosine 1-phosphate. Both HDL-associated and free S1Ps differ in their signaling properties. Circulating S1P (100–1000 nmol) is detected in leukocytes, platelets, and erythrocytes as well as bound to albumin, LDLs, VLDLs, and mainly HDLs [39]. High-density lipoproteins actually have a high affinity for S1P. Sources of plasmatic S1P are hematopoietic (erythrocytes more than platelets) [40]) and endothelial cells. Sphingosine 1-phosphate attached to HDLs mediates, partly or entirely, indirectly or directly, some functions of HDL, such as endothelial nitric oxide production and vasodilation.³⁸

2.5.4.2 HDL–S1P Association in Endothelium

Sphingosine 1-phosphate particularly targets endothelial cells. Whereas free S1P can have vasoconstrictive and pro-inflammatory effects, HDL–S1P promotes vasorelaxation via nitric oxide synthesized by NOS3 as well as prostacyclin (PGI2) in both vascular endothelial and smooth muscle cells, as it stimulates cyclooxygenase-2 and P38MAPK.³⁹ The HDL–S1P complex also influences endothelial barrier function, angiogenesis, and endothelial precursor cell responsiveness.

High-density lipoproteins are able to induce capillary formation via the Ras–Raf–ERK and PKB–ERK–NOS3 pathways. Receptors of S1P targeted by HDL-associated S1P augment both endothelial cell motility and endothelial barrier integrity [39]. Between-endothelial cell junctions are strengthened by S1P₁ and S1P₃, but weakened by S1P₂ receptor. Moreover, S1P-mediated Ca ${}^{++}$ influx stabilizes the endothelial barrier, as it activates NOS3 and produced NO favors the sealing of endothelial cells. Protection against apoptosis also promotes endothelial integrity. Furthermore, HDLs stimulate endothelial progenitor cells for endothelial repair and differentiation of human peripheral mononuclear cells into endothelial progenitor cells to enhance ischemia-induced angiogenesis.

High-density lipoproteins reduce oxidative stress, as they carry anti-oxidative enzymes, such as paraoxonase-1 and -3 and platelet-activating factor acetylhydrolase (PAFAH), that counteract protein oxidation, especially that of LDLs [39]. In addition, HDL-associated S1P, particularly small dense HDL3 that has the highest S1P/sphingomyelin ratio, inhibits ROS production via S1P₃-dependent NADPH oxidase inhibition and antagonizes oxidized LDL-induced apoptosis.

High-density lipoproteins restrain leukocyte adhesion to the vascular endothelium, as they reduce the expression of endothelial adhesion molecules, such as VCAM1, ICAM1, and E-selectin [39]. They also lower the density of α_M-integrin on monocytes and chemokine CCL2 in vascular smooth muscle cells. Consequently, HDLs repress vascular inflammation and promote atheroma stabilization. They hamper cell transmigration, as they cause NO production mediated by SRb1 as well as PGI2 in addition to anti-adhesive, HDL-associated S1P that targets S1P₃ receptor. On the other hand, free S1P generated by sphingosine kinase-1 activated by TNFα stimulates the expression of VCAM1 and ICAM1 adhesion molecules.

2.5.4.3 HDL–S1P in Cardiomyocytes

In cardiomyocytes, HDL-associated S1P activates STAT3 via S1P₂ receptors. Activation of STAT3 follows stimulation of S1P₂ receptor, extracellular signal-regulated kinases ERK1 and ERK2, and Src kinase [41]. Prostacyclin, the production of which by COx2 is heightened by HDL-associated S1P, targets IP receptor to activate cAMP and preclude cardiomyocyte hypertrophy as well as EP receptor to protect cardiomyocytes from damage caused by oxidative stress via mitochondrial ATP-dependent potassium channels [39]. On the other hand, free S1P favors production by COx2 of inflammatory prostaglandins such as PGE2.

Some HDL effects in the cardiovascular system depend completely or partially on S1P, whereas others are independent of their S1P content [39]. Cholesterol efflux in macrophages, a major element of reverse cholesterol transport in arterial walls, does not depend on S1P. Vasodilation dependent on NOS3 mediated by HDL partly relies on S1P agent. Last, but not least, HDL can counteract S1P. High-density lipoproteins can indeed prevent S1P-induced adhesion molecule expression.

2.5.5 Cardiac Effects of Sphingolipids

2.5.5.1 Cardiomyocyte Fate: Apoptosis or Survival

Elevated ceramide content is correlated with cardiomyocyte death after ischemia–reperfusion injury. Cardioprotective pre- and postconditionings that consist of applied transient episodes of ischemia–reperfusion before and after sustained ischemia limit apoptosis. Limited accumulation of ceramide with increased S1P content can mediate protection during ischemic preconditioning, but a large amount that renders NOS3 unavailable favors apoptosis after relatively prolonged hypoxia–reoxygenation [33].

Early nSMase activation occurs in cardiac glutathione deficiency, as glutathione inhibits sphingomyelinase. Reduced glutathione is a cofactor of glutathione peroxidase to reduce intracellular reactive oxygen species, as it is oxidized to a disulfide-linked dimer [33]. In sustained oxidative stress, this dimer cannot be recycled and exit out of the cell, hence causing a glutathione deficiency.

In chronic heart failure, pro-inflammatory cytokines (e.g., TNFα and IL1β) trigger _S,AcSMase secretion from endothelial cells that combines with stimulation of reactive oxygen species on enzyme activity [33].

Activation of sphingosine kinase produces cardioprotective sphingosine 1-phosphate, especially in response to acute ischemia–reperfusion injury. Auto- and paracrine regulator S1P activates prosurvival PI3K pathway that activates PKB and inactivates GSK3β. In addition, nitric oxide synthase contributes to S1P-mediated cardioprotection. Ligand-bound S1P receptors trigger Gi- and RhoA-mediated hypertrophy, but more slowly and less efficiently than Gq–PLC signaling primed by norepinephrine and endothelin [36]. Sphingosine 1-phosphate also influences the electrophysiological and contractile behavior of cardiomyocytes. Transforming growth factor-β stimulates sphingosine kinase SphK1. Released S1P elicits autocrine and paracrine activation of S1P₂ receptors and subsequent collagen production.

2.5.5.2 Regulation of Cardiac Pacemaker and Myocyte Activity

Acetylcholine released during parasympathetic stimulation slows heart rate via activation of Gi-coupled muscarinic receptors M2 of sinoatrial node cells. Subunits Gβγ directly activate atrial muscarinicK⁺ channel (K_IR3; K_ACh current; Vol. 3 – Chap. 3. Main Classes of Ion Channels and Pumps). Members of the G-protein-gated inwardly rectifying K⁺ channel (GIRK1–GIRK5 or K_IR3.1–K_IR3.5) function as highly active heteromultimers or low to moderately active homomultimers responsible for acetylcholine-induced bradycardia during vagal activity.

β-Adrenergic signaling increases the heart frequency. β-Adrenoceptors are coupled with the stimulatory Gs subunit that stimulates the activity of adenylate cyclase to produce cAMP messenger. This mediator fosters the activity of protein kinase-A that phosphorylates different cardiac regulators with positive inotropic and/or lusitropic effects. Kinase PKA phosphorylatesCa_V1 channels and ryanodine receptors to prime Ca ${}^{++}$ influx from the extracellular space and intracellular stores. Among other substrates, phospholamban stimulates Ca ${}^{++}$ reuptake into the sarcoplasmic reticulum via SERCA2a pump. Troponin-I facilitates actin–myosin detachment. In the sinoatrial node, PKA enhances the pacemaker activity, as it phosphorylates Ca_V1 and delayed rectifierK⁺ channels (K_V).

On the other hand, cardiac protein phosphatases PP1 and PP2 (Sect. 8.3) contribute to the regulation of Ca ${}^{++}$ handling and myofilament activity. Both kinases and phosphatases control activities of Ca_V1 and delayed rectifier K⁺ channels in nodal cells. Phosphatase PP1 is regulated by subunitsof the G12/13 subclass as well as Rho and its effector RoCK (Sects. 5.2.14 and 9.3). In sinoatrial cells as well as atrial and ventricular myocytes, PP2 phosphatase, which dephosphorylates PKA, myosin-binding protein-C, and troponin-I, connects to P21-activated kinase PAK1 to form a regulatory complex [42] (Sect. 5.2.13.1). Activation of PP2 by PAK1 increases myofilament sensitivity to Ca ${}^{++}$ in cardiomyocytes. Moreover, PAK1 has a negative chronotropic effect. In addition to CDC42 and Rac1, sphingosine directly causes autophosphorylation (activation) of both PAK1 and PAK2 kinases.

2.5.6 Sphingolipid Effects on Smooth Muscle Cells

2.5.6.1 Smooth Muscle Cell Fate

Receptors S1P₁ and S1P₃ promote proliferation of smooth muscle cells in balloon injury models, whereas S1P₂ antagonizes this effect. In cultured vascular smooth muscle cells, apolipoprotein-C1-enriched high-density lipoproteins stimulate nSMase and prime apoptosis via the release of cytochrome-C from mitochondria and caspase-3 activation [33]. On the other hand, oxidized low-density lipoproteins and tumor-necrosis factor-α that activate nSMase2 in these cells provokes cell proliferation owing to the conversion of ceramide into S1P and activation of ERK1 and ERK2 kinases.

2.5.6.2 Contraction–Relaxation

Sphingosine 1-phosphate generates contraction of vascular smooth muscle cells mainly via S1P₃ receptors. However, S1P₂ is responsible for the maintenance of the basal vasomotor tone in resistance arteries of the mesenteric and renal trees. In human airway smooth muscle cells, S1P also elicits constriction via S1P₂ receptors.

Sphingosine 1-phosphate receptors are involved in cytoskeleton rearrangement. Receptor S1P₁ couples to Rac GTPase via Gi and S1P₃ to RhoA via G12/13 subunit. Both RhoA and Rac1 intervene in cardiac remodeling. Effectors of RhoA comprise RoCK kinase and diaphanous (Dia; Vol. 1 – Chap. 6. Cell Cytoskeleton). Kinase RoCK phosphorylates myosin light-chain phosphatase to promote contraction of vascular smooth muscle cells. Activated Rac1 causes cytoskeleton reorganization via WASP, WaVe, IqGAP, and PAK1, as well as ROS production.

Smooth muscle cell contraction is mainly triggered by a rise in cytosolic Ca ${}^{++}$ concentration that promotes binding of Ca ${}^{++}$ to calmodulin. The Ca ${}^{++}$ –calmodulin complex activates myosin light-chain kinase. Phosphorylation of myosin light chain provokes interaction of myosin-2 with actin, hence smooth muscle contraction. In smooth muscle cells, ligand-bound G-protein-coupled receptors activate the phosphoinositide cascade that causes Ca ${}^{++}$ release from the sarcoplasmic reticulum. In addition, activation of various Ca ${}^{++}$ channels prime Ca ${}^{++}$ influx from the extracellular medium. In vascular smooth muscle cells, once GTPase RhoA is activated by S1P receptors, RoCK phosphorylates (inactivates) myosin light-chain phosphatase and sensitizes smooth muscle cell to Ca ${}^{++}$ .

2.5.7 Sphingolipid Effects on Endothelial Cells

2.5.7.1 Endothelium Integrity and Permeability

Whereas S1P₁ activates the Gi–PI3K pathway to prevent permeability of vascular endothelium, S1P₂ precludes the PI3K pathway via the Rho–RoCK–PTen–Rac axis, hence raising endothelial permeability by disrupting adherens junction [38]. The balance between S1P₁ and S1P₂ in endothelial cells contributes to the regulation of blood vessel wall permeability.

2.5.7.2 Vasomotor Tone Regulation

When smooth muscle cells are exposed to NO, cGMP-dependent protein kinase PKG is activated. Subsequently, PKG activates myosin light-chain phosphatase, thereby relaxing smooth muscle cells.

Ceramide

Mechanical stimulation can generate ceramide in caveolae and NOS3 phosphorylation (activation), hence vasorelaxation. On the other hand, in aging rat arteries, activated nSMase produces ceramide that excites protein phosphatase-2A, thereby diminishing NOS3 phosphorylation [33].

Sphingosine 1-Phosphate

A given blood vessel responds to S1P with vasodilation or vasoconstriction according to the context, i.e., S1P concentration, targeted receptor subtypes, vascular bed type, etc. Moreover, like numerous vasoactive substances, S1P regulates vascular tone by modulating stress fiber state in smooth muscle cells or by stimulating endothelial cell release of agents that regulate SMC contraction level. Factor S1P can cause both vasorelaxation from adjoining endothelial cells via the Rac1–NOS3 pathway and nitric oxide release and vasoconstriction in smooth muscle cells via the RhoA–RoCK pathway. Resulting effect depends on the context (vascular bed, environmental conditions, S1P concentration, S1P receptor subtype expression pattern, etc.). Agent S1P activates the NOS3–NO axis mainly via S1P₁ receptors [40].

Vascular endothelial growth factor can regulate S1P₁ density and magnitude of NOS3 response to S1P via protein kinase-C.⁴⁰ Reactive oxygen species also raise S1P₁ expression, and hence S1P-dependent NOS3 activation.

Both endothelial nitric oxide synthase that is dually acylated by saturated fatty acids myristate and palmitate⁴¹ and S1P₁ reside mainly in caveolae, invaginated domains relatively enriched in cholesterol and sphingolipids of the plasma membrane. Enzyme NOS3 is phosphorylated by phosphoinositide 3-kinase-β and protein kinase-B (Ser117) downstream from a pathway that incorporates Gβγ, small GTPase Rac1, and AMP-activated protein kinase [40].⁴²

2.6 Arachidonic Acid and Eicosanoids

Phospholipids are hydrolyzed by phospholipase-A2 (Sect. 2.8.1) into arachidonic acid, a precursor of eicosanoids (Tables 2.19 and 2.20).

Table 2.19

Production of arachidonic acid (AA) and its derivatives from membrane phospholipids by phospholipase-A2 (PLA2) and eicosanoid-producing enzymes (Source: [43]). Arachidonate-containing plasmalogen phospholipids are hydrolyzed by PLA2 to generate free arachidonic acid. The latter is metabolized by cyclooxygenases (COx), lipoxygenases (LOx), and cytochrome-P450 monooxygenases (CyP450) to produce prostaglandins (PG), leukotrienes (Lkt), and epoxyeicosatrienoic (EET) and hydroxyeicosatrienoic (HETE) acids. Arachidonate-containing diacyl phospholipids are processed by PLA2α or PLA2β that release arachidonic acid. On the other hand, PLA2γ mainly catalyzes the production of 2-arachidonoyl lysolipids (2ALL). Subsequently, lysophospholipase-D (lysoPLD) and lysophospholipase (LPLase) synthesize 2-arachidonoyl glycerol (2AG) and arachidonic acid from 2-arachidonoyl phospholipids, respectively. 2-Arachidonoyl-glycerol is targeted by cyclooxygenases, lipoxygenases, and monoacylglycerol lipase (MAGL) to produce prostaglandin glycerol esters (PGE), leukotriene glycerol esters (LGE), and arachidonic acid, respectively.

Substrate	Reaction cascade
Plasmalogen phospholipids	PLA2β/PLA2γ–AA
Diacyl phospholipids	PLA2α/PLAβ–AA
	PLA2γ–2ALL–lysoPLD–2AG–COx–PGE
	PLA2γ–2ALL–lysoPLD–2AG–LOx–LGE
	PLA2γ–2ALL–lysoPLD–2AG–MAGL–AA
	PLA2γ–2ALL–LPL–AA
Arachidonic acid	AA–COx–prostaglandins
	AA–LOx–leukotrienes
	AA–CyP450–EETs/HETEs

Table 2.20

Eicosanoid synthesis from arachidonic acid (AA; Source: [43]). Prostaglandins PGG2 and PGH2 are synthesized by COx1 and COx2. Prostaglandin PGH2 is further metabolized to PGE2, PGF2α, and PGI2 by corresponding synthases. Arachidonic acid is oxidized into 12-hydroperoxyeicosatrienoic acid (12HpETE) by 12-lipoxygenase (12LOx) that is then converted into 12-hydroxyeicosatrienoic acid (12HETE). Cytochrome-P450 epoxigenase and hydroxylase form (14,15)-epoxyeicosatrienoic acid ((14,15)EET), and 16- and 20-HETEs, respectively. Epoxide hydrolase (EH) catalyzes the conversion of (14,15)EET to (14,15)-dihydroxyeicosatrienoic acid ((14,15)DHET).

Upstream enzymes	Pathways
Cyclooxygenases	AA–COx–PGG2–PGH2–PGE2
	AA–COx–PGG2–PGH2–PGF2α
	AA–COx–PGG2–PGH2–PGI2
Lipoxygenases	AA–12LOx–12HpETE–12HETE
Cytochrome-P450	AA–(14,15)EET–(14,15)DHET
epoxygenase
Cytochrome-P450	AA–16HETE
hydroxylase	AA–20HETE

Eicosanoids are generated by cyclo- and lipoxygenases as well as cytochrome-P450-based monooxygenases. Eicosanoids interact with receptors and ion channels and pumps [43] (Table 2.21).

Table 2.21

Effect of arachidonic acid and epoxyeicosatrienoic and hydroxyeicosatrienoic acids on ion channels (Source: [43]).

Molecule	Effect
Arachidonic acid	K_V1.1 kinetics increase
Epoxyeicosatrienoic acids	Ca_V1 activity enhancement
	K_ATP sensitivity reduction [(11,12)EET]
	Na⁺ channel-gating attenuation
Hydroxyeicosatrienoic acids	BK channel inhibition (20HETE)

2.6.1 Arachidonic Acid

In heart, arachidonic acid metabolism is mainly associated with cardiomyocytes, and vascular endothelial and smooth muscle cells that are interconnected by intercellular communications. In inflammation, neutrophils and macrophages can infiltrate myocardium and contribute to the production of eicosanoids and other lipidic messengers, as well as platelets and fibroblasts. Arachidonic acid increases both voltage-dependent activation and inactivation kinetics of delayed rectifier K⁺ channel K_V1.1 (Table 2.21).

2.6.2 Cyclooxygenases

Cells such as cardiomyocytes produce 2 types of cyclooxygenases, constitutively expressed COx1 and inducible COx2, that have distinct functions. Cyclooxygenases form prostaglandin-G2 and then prostaglandin-H2 (Table 2.20).

Enzyme COx2 is a mediator of inflammation that can ensure cardioprotection. It is upregulated during ischemic preconditioning via kinases PKC and Src, transcription factor NFκB, as well as NOS2 [43]. PGF2α activates its Gq-coupled prostanoid FP receptor that stimulates ERK2, JNK1, and cytosolic Tyr kinases and provokes cardiac hypertrophy. Kinase JNK1 can suppress hypertrophic signaling via phosphorylation of nuclear factor of activated T-cell that hinders its nuclear translocation [43]. Agent PGF2α also raises glucose transporter GluT1 expression and thus glucose uptake. PGI2 binds to its Gs-coupled prostanoid IP receptor and represses maladaptive cardiac hypertrophy in response to pressure overload.

2.6.3 Lipoxygenases

Lipoxygenases are iron-containing enzymes that catalyze dioxygenation of polyunsaturated fatty acids in lipids. Numerous lipoxygenase isozymes are involved in the metabolism of prostaglandins and leukotrienes. Well known types includes arachidonate 5-⁴³ and 12-lipoxygenase, in addition to erythroid cell-specific arachidonate 15-lipoxygenase.⁴⁴

Lipoxygenases catalyze oxidation of arachidonic acid to produce hydroperoxyeicosatetraenoic acids that are then reduced to form hydroxyeicosatrienoic acids (HETE). In addition, lipoxygenases promote insulin-stimulated glucose uptake via GluT4 transporter, as it favors suitable actin rearrangement [43].

Lipoxygenase-mediated production of HETEs participates in the maintenance of the myocardial function. However, overexpression of 12-lipoxygenase, the most abundant type in myocardium, leads to cardiac hypertrophy and fibrosis [43].

Enzyme 15-lipoxygenase-1 oxidizes unsaturated fatty acids such as arachidonic acid at the cell membrane, to generate active hydroperoxy and epoxy metabolites. It is produced by macrophages, eosinophils, and mastocytes, among others. Epithelial 15-lipoxygenase-1 generates intracellular 15-hydroxyeicosatetraenoic acid. In particular, 15HETE acts in interleukin-13-induced mucin Muc5ac production by human airway epithelial cells. Product 15HETE can bind to phosphatidylethanolamine to form the 15HETE–PE complex. This complex as well as 15LOx1 interact with phosphatidylethanolamine-binding protein PEBP1,⁴⁵ thereby dissociating PEBP1, a Raf inhibitor,⁴⁶ from cRaf kinase.⁴⁷ Both the 15LOx1–PEBP1 and the 15HETE–PE–PEBP1 complexes then activate the cRaf–ERK pathway, especially in asthma [44].⁴⁸

2.6.4 Cytochrome-P450 Monooxygenases

Membrane-bound, heme-containing, cytochrome-P450-based monooxygenases catalyze the insertion of a single atom of oxygen as a part of multicomponent electron transfer complexes. Cytochrome-P450-containing monooxygenases primarily reside in the inner membrane of mitochondria and endoplasmic reticulum.

They oxidize several endogenous lipids, such as arachidonic, retinoic, and linoleic acids. From arachidonic acid, CyP450 forms a series of fatty acid epoxides, such as (5,6)-, (8,9)-, (11,12)-, and (14,15)-EETs, as well as alcohols such as midchain and ω-terminal HETEs. They also catalyze oxygen- and NADPH-dependent metabolism (oxidation, peroxidation, and/or reduction) of xenobiotics.

Multiple cytochrome-P450 types exist.⁴⁹ Cytochrome-P450 monooxygenases require O₂, CyP450 reductase, and NADPH cofactor. Some CyP450 types, such as CyP450-2b, CyP450-2c, and CyP450-2j, are primarily arachidonic acid epoxygenases. Others are chiefly arachidonic acid hydroxylases, such as CyP450-4a and CyP450-4f family members.

Most cytochrome-P450-based enzymes are predominantly synthesized in the liver. However, some cytochrome-P450-containing enzymes, such as members of the CyP450-2j and -4a subfamilies are predominantly detected in the heart, vasculature, kidney, lung, and gastrointestinal tract. Some cytochrome-P450 monooxygenases are expressed constitutively, whereas others are induced.

CyP450 ω-hydroxylases (CyP450-4a1, -4a2, and -4f) produce 20HETE, a potent vasoconstrictor that inhibits Ca ${}^{++}$ -sensitive K⁺ channels [43]. Product 20HETE thus aggravates ischemia damage.

On the other hand, epoxyeicosatrienoic acids ensure cardioprotection. Epoxygenase CyP450-2j2 attenuates ischemia–reperfusion injury, as produced EETs modulate K_ATP channel activity and MAPK signaling [43] (Table 2.22).

Table 2.22

Effects of epoxyeicosatrienoic acids (Source: Wikipedia). Epoxyeicosatrienoic acids activate large-conductance, calcium-activated potassium channels in vascular smooth muscle cells, thereby causing hyperpolarization of the membrane potential and vasodilation. They also influence Ca_V1 channel activity in cardiomyocytes. On the other hand, 20HETE, the activity of which is impeded by nitric oxide, inhibits calcium-activated potassium channels and activates Ca_V1 channels in vascular smooth muscle cells.

Calcium release from intracellular stores
Increased sodium–hydrogen antiporter activity
Increased cell proliferation
Decreased cyclooxygenase activity
Decreased release of insulin and glucagon
Vasodilation
Increased risk of tumor adhesion on endothelial cells
Decreased platelet aggregation
Cardioprotection after ischemia–reperfusion

Cytochrome-P450 epoxygenases produce (5,6)EET and (11,12)EET that raise cAMP and intracellular Ca ${}^{++}$ concentration, as Ca ${}^{++}$ influx from the extracellular medium throughCa_V1 channels that are stimulated by EETs activates ryanodine receptor [43]. Agent (11,12)EET lowersK_ATP channel sensitivity to ATP (Table 2.21). Several EETs attenuate Na⁺ channel gating. Agent 20HETE inhibitshigh-conductance, Ca ${}^{++}$ -activated, voltage-gated K⁺ channel (BK).

2.6.5 Arachidonic Acid Metabolites in Lungs

Arachidonic acid metabolites derived from action of cyclo-, lipo-, and CyP450 monooxygenases achieve various functions in lungs, particularly pulmonary vascular and bronchial smooth muscle tone as well as airway epithelial ion transport. Cytochrome P450-derived arachidonate metabolites contribute to hypoxic pulmonary vasoconstriction, regulation of bronchomotor tone, control of the composition of airway lining fluid, and limitation of pulmonary inflammation [45].

Lung cells produce arachidonic acid from membrane phospholipid stores via phospholipases (e.g., cytosolic PLA2), both constitutively and on chemical or mechanical stimuli. Free arachidonic acid can then enter into one of the 3 following metabolic pathways [45]: (1) the prostaglandin-H synthase pathway that produces thromboxane and prostacyclin; (2) the lipoxygenase pathway that synthesizes leukotrienes, midchain hydroxyeicosatetraenoic acids, and lipoxins; and (3) the cytochrome P450 monooxygenase pathway that manufactures midchain and ω-terminal HETEs and cis-epoxyeicosatrienoic acids.

2.6.6 Prostaglandins

Prostaglandin-E2 is synthesized in the airway epithelium. It is a potent bronchodilator. It also has anti-inflammatory effects. Prostacyclin (PGI2) is produced in the pulmonary artery endothelium. It has vasodilatory, bronchodilatory, and platelet anti-aggregatory effects. On the other hand, prostaglandin-F2α and thromboxane (TXA2) are potent bronchoconstrictors. These vasoconstrictors also enhance platelet aggregation.

Widespread prostaglandin-F2α that causes renin secretion and vasoconstriction, among other functions, is synthesized by prostaglandin-F synthase of the aldoketo reductase (AKR) family.⁵⁰

Products of prostaglandin-H synthases PGHS1 and PGHS2 limit lung inflammation in response to inhaled allergens and lipopolysaccharides [45]. On the other hand, lipoxygenase products of arachidonic acid contribute to airway inflammation, bronchoconstriction, increased mucous secretion, and vascular permeability in asthma.

2.6.7 Hydroxy- and Epoxyeicosatrienoic Acids

Epoxyeicosatrienoic acids participate in the relaxation of vascular smooth muscle cells by the so-called endothelium-derived hyperpolarizing factor, especially in the coronary bed. On the other hand, 20HETE, the dominant arachidonic acid metabolite in the renal cortex that promotes natriuresis, is one of the most potent vasoconstrictors.

Human peripheral lung microsomes that convert arachidonic acid to 20HETE are enriched in CyP450-4a enzymes [45]. In human lungs, CyP450-2j enzymes are detected in both ciliated and non-ciliated airway epithelial cells, bronchial and vascular smooth muscle cells, endothelial cells, and alveolar macrophages. Members of the CyP450-1a, -2b, -2e, -2f, -3a, and -4b subfamilies have also been observed in human lungs [45].

Arachidonic acid precludes Cl⁻ secretion in human airways [45]. Epoxyeicosatrienoic acids hamper tracheal Ca ${}^{++}$ -sensitive Cl⁻ flux. They thus modulate the composition of the airway lining fluid and gel layer (mucus).

Both (5,6)EET and (11,12)EET cause hyperpolarization of the plasmalemmal potential and relaxation of rabbit tracheal smooth muscle cells [45]. Epoxyeicosatrienoic acids activatelarge-conductance, Ca ${}^{++}$ -activated K⁺ channels.

Lipid 20HETE decreases the vasomotor tone of human pulmonary arteries [45]. The CyP450-4a–20HETE axis reduces subacute hypoxic pulmonary vasoconstriction. Conversely, a subacute exposure to hypoxia hinders 20HETE formation. On the other hand, the production of prostacyclin and nitric oxide rises during subacute or chronic hypoxia.

2.7 Lipases

Lipase activity can be achieved by numerous proteins, such as hormone-sensitive, lipoprotein, endothelial, gastric, hepatic, pancreatic, carboxyl ester, lysosomal acid, and monoacylglycerol lipase, as well as triacylglycerol hydrolases TGH1 and TGH2. Several other types of lipases exist, such as phospholipases and sphingomyelinases, but they are usually not considered as conventional lipases.

Triglyceride lipases hydrolyze linkages of triglycerides. Lipases produce free fatty acids used as an energy source. In particular, fatty acids are mobilized from triglyceride stores during exercise to supply working muscle with energy, especially during prolonged exercise, when carbohydrate reserves get depleted.

Nutritional glucids and lipids in excess are efficiently converted into triglycerides. Most of the body’s energy reserves are stored in white adipose tissue (Vol. 2 – Chap. 1. Remote Control Cells). In normal conditions, an equilibrium exists between lipid synthesis and degradation. Catabolism of triglyceride storage depots and mobilization of free fatty acids in adipocytes and other cell types depend on lipases. Two lipases operate successively: hormone-sensitive (HSL) and adipose triglyceride lipase (ATGL).

Mobilization of fatty acids from adipose tissue is controlled by hormones, cytokines, and adipokines. Catecholamines (β-adrenergic stimuli), atrial natriuretic peptide, growth hormone, and leptin activate lipases to raise fatty acid release into the blood circulation. On the other hand, insulin inhibits hormone-sensitive lipase.

2.7.1 Hormone-Sensitive Lipase

Hormone-sensitive lipase (HSL) is synthesized in multiple tissue and cell types, such as white and brown adipose tissue, skeletal muscle, heart, steroidogenic tissues, and intestine, as well as pancreatic β cells and macrophages (Table 2.23).

Table 2.23

Hormone-sensitive lipase and its products in different cell types (Source: [47]).

Tissue	Major product	Function
Cell
Adipose tissue	Fatty acids	Export for oxidation
Muscle, heart	Fatty acids	Oxidation
β Cells	Fatty acids
Adrenals	Cholesterol	Substrate for steroidogenesis
Ovaries, testes	Cholesterol	Substrate for steroidogenesis
Ovaries, placenta	Steroids	Transcriptional control
Mammary gland	Cholesterol	Milk component, membranogenesis
Macrophage	Cholesterol	Export (via high-density lipoprotein)

Hormone-sensitive lipase is able to hydrolyze tri-, di- and monoacylglycerols as well as cholesterol and retinyl esters. It most efficiently hydrolyzes diglycerides. It is phosphorylated (activated) by protein kinase-A and AMP-activated protein kinase. Conversely, it is inactivated by PP2 and PPM1 phosphatases. Phosphorylation by protein kinase-A of HSL and lipid droplet-associated perilipin trigger the translocation of HSL from the cytoplasm to the lipid droplet.⁵¹

2.7.1.1 HSL in Macrophages

Cholesterol of lipoproteins is stored in lipid droplets as cholesterol esters after re-esterification by acyl-CoA:cholesterol acyltransferase ACAT1. Hydrolysis of cholesterol esters is the initial step toward elimination of cholesterol. Hormone-sensitive lipase and cholesteryl ester hydrolase (CEH) hydrolyze cholesterol esters in macrophages.

Cholesterol homeostasis, i.e., balance between synthesis and hydrolysis, is actually achieved by an equilibrium between the activities of ACAT1 that catalyzes the formation of cholesteryl esters and cytosolic neutral cholesterol ester hydrolase (nCEH) that supports the remobilization of free cholesterol.

Neutral cholesterol ester hydrolase removes cholesterol from macrophages in cooperation with hormone-sensitive lipase that also has a neutral cholesteryl ester hydrolase activity [48].

2.7.2 Adipose Triglyceride Lipase

Adipose triglyceride lipase (ATGL) promotes the catabolism of stored lipids in adipose and non-adipose tissues. This triacylglycerol hydrolase specifically performs the first step of lipolysis. It indeed generates diglycerides and fatty acids. Therefore, efficient lipolysis depends on the coordinated action of lipases that operate sequentially: ATGL forms diglycerides that are subsequently hydrolyzed by hormone-sensitive lipase.

2.7.3 Hepatic Lipase

Hepatic lipase (LipC)⁵² is expressed not only in the liver, but also adrenal glands. It aims at converting intermediate-density lipoproteins to LDL particles. It possesses the dual function of triglyceride hydrolase and ligand-bridging factor for receptor-mediated lipoprotein uptake.

Hepatic lipase, like LPase, facilitates the interaction of lipoproteins with cell-surface proteoglycans and receptors, thereby enhancing binding and uptake of different lipoproteins as well as cholesteryl esters.

2.7.4 Endothelial Lipase

Endothelial lipase (EL), or endothelial cell-derived lipase (EDL), encoded by the LIPG gene, is synthesized, secreted, then either anchored to the plasma membrane or released by the endothelial cells [49]. It is also detected in hepatocytes and macrophages. It has high-sequence homology with lipoprotein lipase (LPase). In fact, endothelial lipase is an additional member of the triacylglyerol lipase family with lipoprotein lipase (LPase) and hepatic lipase (HL). They intervene in metabolism of high-density lipoproteins.

Endothelial lipase is produced in various organs, such as the liver, lung, kidney, and placenta, but not skeletal muscle [49]. Expression of endothelial lipase rises during inflammation.

Unlike LPase and HL, endothelial lipase is mainly a phospholipase with a slight triacylglycerol lipase activity. It is aimed at binding HDLs and taking up HDL-carried cholesterol esters. It reduces the plasma concentration of HDL–cholesterol (HDL–C) and its major protein apolipoprotein-A1 [49]. A low concentration of plasma HDL–C is major risk factor for the development of arterial diseases (Vol. 6 – Chap. 7. Vascular Diseases). Endothelial lipase also contributes to the cellular uptake of lipoproteins, independent of its catalytic activity.

Administration of endothelial lipase-coding adenovirus to cells raises HDL binding and uptake 1.5 fold as well as HDL–CE uptake 1.8 fold [50]. On the other hand, lipoprotein lipase has a less pronounced effect (1.1-fold and 1.3-fold increase in HDL and HDL–CE uptake, respectively).⁵³ Inhibition of EL enzymic activity by tetrahydrolipstatin (THL) enhances its effect on lipid uptake (5.2-fold, 2.6-fold, 1.1-fold increase in HDL binding and HDL and CE uptake, respectively). In the presence of THL, plasmalemmal EL concentration is augmented. Conversely, HDL and free fatty acids reduce the amount of cell surface-bound endothelial lipase [50].

Like in macrophages, on which lipoprotein lipase mediates HDL–CE uptake, in addition to ScaRs (Vol. 3 – Chap. 4. Membrane Compound Carriers), endothelial lipase alone is able to stimulate HDL–CE uptake, independently of ScaRb1 scavenger receptor.

Statins that lower LDL–C level, hence used to treat hypercholesterolemia, also increase plasma HDL–C concentration. They selectively inhibit 3-hydroxy 3-methyl glutaryl coenzyme-A reductase (HMGCoAR), a rate-limiting enzyme of the cholesterol synthesis. They also suppress the synthesis of isoprenoid intermediates, such as geranylgeranyl and farnesyl pyrophosphate, which serve to attach lipids to various signaling molecules [51]. In macrophages, statins increased apoA1 level, as they prevent membrane translocations of small GTPases CDC42, Rac, Ras, and RhoA and activates extracellular signal-regulated kinases ERK1 and ERK2 as well as P38MAPK and PPARγ (nuclear receptor NR1c3) [52]. They stimulate NR1c3 via COx2-dependent increase in prostaglandin PGJ2 level via the RhoA–P38MAPK and CDC42–P38MAPK pathways and RhoA- and CDC42-independent ERK1/2 pathway. Moreover, statins also provoke the synthesis of fatty acid translocase CD36,⁵⁴ ATP-binding cassette transporter ABCa1, as well as scavenger receptor ScaRb1 (or SRB1; a.k.a. CD36L1). The latter has high affinity for HDL–C and mediates cholesterol efflux and uptake. Transporter ABCa1 causes lipid efflux from cells to lipid-poor apolipoproteins. Translocation of ABCa1 to the plasma membrane thus participates in HDL plasma level, independently of ScaRb1 [53]. Furthermore, statins suppress the basal and TNFα-induced expression of endothelial lipase in human umbilical vein endothelial cells [51].

2.7.5 Lipoprotein Lipase

Lipoprotein lipase hydrolyzes lipids in lipoproteins, such as those in chylomicrons and very-low-density lipoproteins, into free fatty acids and monoacylglycerol. It requires ApoC2 cofactor. Lipoprotein lipase specifically lodges in endothelial cells of capillaries.

Insulin causes LPase synthesis in adipocytes. It also favors its insertion in the capillary endothelium. Different LPase isozymes exist in separate tissues, with distinct regulation modes. Lipoprotein lipase in adipocytes is activated by insulin, but not that in striated myocytes.

Lipoprotein lipase is a homodimer that has the dual function of triglyceride hydrolase and ligand-bridging factor for receptor-mediated lipoprotein uptake. Lipoprotein lipase actually mediates HDL–CE uptake by macrophages and hepatocytes. Lipoprotein lipase links to lipoproteins and anchors them to the plasma membrane. It indeed has a high affinity for cell-surface proteoglycans, especially heparan sulfate proteoglycans that tether molecules. Therefore, LPase recruits lipoproteins to the cell membrane and promotes cholesterol ester uptake from lipoproteins. In addition, complexes composed of HSPG, LPase, and lipoproteins can be internalized.

2.7.6 Lipase-H

The secreted enzyme lipase-H (LipH)⁵⁵ hydrolyzes phosphatidic acid into lysophosphatidic acid. This lipid mediator operates in platelet aggregation, smooth muscle contraction, and cell proliferation and motility.

2.8 Phospholipases

Inositol phosphates are intracellular second messengers that regulate various cell processes, from calcium signaling to chromatin remodeling (e.g., digestion, inflammation, membrane remodeling, and intercellular signaling via the degradation of phospholipids). Phospholipases target phospholipids of cellular membranes. Phospholipases are involved with phospholipids in the transmission of ligand-bound receptor signaling.

Four main types of phospholipases exist (PLA–PLD). Activated phospholipases particularly stimulate protein kinase-C that is maximally active in the presence of diacylglycerol and calcium ions released from its intracellular stores by inositol trisphosphate.

2.8.1 Phospholipases-A

Phospholipase-A hydrolyzes one of the acyl groups of phosphoglycerides or glycerophosphatidates. Phospholipases-A1 and -A2 target the acyl group at the 1- and 2-position, respectively.

2.8.2 Phospholipases-A1

Phospholipase-A1 (PLA1) is also called phosphatidylcholine 1-acylhydrolase. It produces 2-acyl lysophospholipids and fatty acids. Extracellular PLA1s that belong to the pancreatic lipase gene family include: (1) phosphatidylserine-specific PLA1 (psPLA1); (2) membrane-associated phosphatidic acid-selective PLA1s (_mpaPLA1α and _mpaPLA1β); (3) hepatic lipase (Sect. 2.7.1); (4) endothelial lipase (Sect. 2.7.4); and (5) pancreatic lipase-related protein PLRP2 [54].

The 3 first-mentioned PLA1s differ from other members by their structures and substrate specificities. They indeed exhibit only a PLA1 activity, whereas hepatic and endothelial lipases and PLRP2 are triacylglycerol-hydrolases with a PLA1 activity [54]. Both psPLA1 and _mpaPLA1s specifically hydrolyze phosphatidylserine and phosphatidic acid, respectively, producing lysophosphatidylserine and lysophosphatidic acid that serve as lipid mediators.

Activity of phospholipase-A1 is detected in many types of tissues and cells. Phosphatidylserine-specific PLA1 is a secreted enzyme that targets phosphatidylserine, which is normally located in the inner leaflet of the lipid bilayer, hence of reduced accessibility. Nevertheless, psPLA1 processes phosphatidylserine exposed on the surface of cells, such as apoptotic cells and activated platelets [55]. Generated 2-acyl lysophosphatidylserine is a lipid mediator for mastocytes, T lymphocytes, and neurons.

2.8.2.1 Pancreatic Lipase-Related Protein PLRP2

Dietary triglycerides, the predominant nutritional lipids, yield a major nutritional energy source. The intestine produces very-low-density lipoproteins and chylomicrons to transport lipids and lipid-soluble vitamins into blood.⁵⁶ Vitamin A-rich chylomicron remnants result from the selective removal of triglyceride catalyzed by lipoprotein lipase. Dietary triglycerides are precursors of cellular membrane components, prostaglandins, thromboxanes, and leukotrienes. They improve the palatability of food.⁵⁷ Dietary fats are particularly important for newborns and infants during the first 6 months of life. Efficient digestion of dietary triglycerides relies on their cleavage into fatty acids and monoacylglycerols.

Adult humans require pancreatic triglyceride lipase and colipase for efficient dietary lipid digestion. On the other hand, in newborns and infants, pancreatic triglyceride lipase does not contribute to nutritional lipid digestion. Newborns and suckling infants express colipase, but not pancreatic triglyceride lipase until the suckling-weanling transition. Therefore, colipase interacts with another lipase in newborns. Efficient lipid digestion in newborns actually requires colipase that interacts with pancreatic lipase-related protein-2 (PLRP2), a homolog of pancreatic triglyceride lipase [57].

2.8.3 Superfamily of Phospholipases-A2

In mammals, the genome encodes more than 30 phospholipases-A2 (PLA2) or related enzymes. Phospholipase-A2 hydrolyzes glycerophospholipids to yield fatty acids and lysophospholipids. In particular, processing of phosphatidylcholine produces free fatty acids and lysophosphatidylcholine, which potentiate diacylglycerol activity.

In most mammalian cells, PLA2 catalyzes the breakdown of membrane phospholipids into arachidonic acid, which is involved in signaling. Arachidonic acid is the precursor of eicosanoids (Vol. 5 – Chaps. 8. Smooth Muscle Cells and 9. Endothelium), such as leukotrienes and prostaglandins. Eicosanoids can also be synthesized from diacylglycerol formed by phospholipase-C.

About one third of enzymes that possess PLA2 or related activity are secreted subtypes with distinct localizations and enzymatic properties. They act on extracellular phospholipids. Intracellular PLA2s intervene in signal transduction and membrane homoeostasis.

Numerous categories (giPLA2) and subcategories of PLA2s have been defined within the PLA2 superfamily.⁵⁸ In humans, the phospholipases-A2 superfamily can be split into the following groups (Table 2.24): group 1 encoded by the gene PLA2G1B; group 2 by the genes PLA2G2A and PLA2G2C (possible pseudogene) to PLA2G2F; group 3 by the PLA2G3 gene; group 4 by the genes PLA2G4A to PLA2G4F; group 5 by the PLA2G5 gene; group 6 by the PLA2G6 gene; group 7 by the PLA2G7 gene; group 10 by the PLA2G10 gene; and group 12 by the PLA2G12A and PLA2G12B genes. Enzymes g4PLA2s and g6PLA2 operate in the perinuclear membranes, where arachidonic acid-metabolizing enzymes (cyclooxygenases and prostaglandin synthases) reside.

Table 2.24

Groups of phospholipases-A2 (Source: [60]; cPLA2: intracellular, calcium-dependent PLA2; iPLA2: intracellular, calcium-independent PLA2; _SPLA2: secreted PLA2).

Group	Sources
Small secreted phospholipases-A2
1A	Old world snake
1B	Human, pig (pancreas)
2A	Human (synovium), new world snake
2B	New world snake
2C	Rodent
2D	Human, Murinae (pancreas, spleen)
2E	Human, Murinae (brain, heart, uterus)
2F	Human, Murinae (testis, embryo)
3	Human, Murinae, lizard, bee
5	Human, Murinae (heart, lung, macrophage)
9	Marine snail Conus venom (conodipine-M)
10	Human (spleen, thymus, leukocyte)
11A	Green rice shoots (PLA2-1)
11B	Green rice shoots (PLA2-2)
12	Human, Murinae
13	Parvovirus
14	Symbiotic fungus, bacteria
G7a	PAFAH1 (lpPLA2)
Intracellular phospholipases-A2
4	Human (cPLA2)
(4A–4F)
6	Human (iPLA2)
(6A1/2, 6B–6H)
G7b	Human (PAFAH2)
G8a/b	Human (PAFAH1b [PAF1b] subunits)
15	Human (lPLA2)

Four main families of PLA2s include (Table 2.25): (1) low-molecular-weight, secretory PLA2s (_SPLA2); (2) cytosolic, calcium-dependent PLA2s (cPLA2) such as g4PLA2s; (3) intracellular, calcium-independent PLA2s (iPLA2) such as g6PLA2; and (4) lipoprotein-associated phospholipase-A2 (lpPLA2), or platelet-activating factor acid hydrolases (PAFAH), such as g7PLA2. Enzyme iPLA2 is observed in endothelial cells and contributes to endothelium-dependent vasoconstriction induced by acetylcholine.

Table 2.25

Phospholipases-A2, their subcellular localization, and Ca ${}^{++}$ -dependent activity (Source: [61]; cPLA2: intracellular, calcium-dependent PLA2; iPLA2: intracellular, calcium-independent PLA2; lPLA2: lysosomal PLA2; lpPLA2: lipoprotein-associated PLA2; PAFAH: platelet-activating factor acid hydrolase; _SPLA2: secreted PLA2; ER: endoplasmic reticulum). Both arachidonic acid (AA) and docosahexaenoic acid (DHA) increase entry of extracellular Ca ${}^{++}$ into neurons via NMDA-type glutamate receptors. Arachidonic acid and docosahexaenoic acid increases and inhibits glutamate-induced prostaglandin release from astrocytes, respectively. Unesterified DHA can promote Ca ${}^{++}$ release from the endoplasmic reticulum. Neuroprotectin-D1 is a metabolite of docosahexaenoic acid. Cell-specific and agonist-dependent events coordinate translocation of cPLA2 to the nuclear envelope and membranes of the endoplasmic reticulum and Golgi body. Calcium ion attenuates iPLA2β activity that promotes iPLA2β–calmodulin interaction, which impedes iPLA2 activity.

Family	Calcium effect	Distribution	Fatty acid
			preference
_SPLA2	Stimulation	Ubiquitous	None
cPLA2	Translocation	Ubiquitous	AA
	(membrane binding)
iPLA2	Activation by	Brain, heart,	DHA
	Ca ${}^{++}$ release	skeletal muscle,
	from ER (but not	pancreas, testis,
	Ca ${}^{++}$ binding)	placenta
lPLA2		Brain, Heart, lung,	Oleate,
		liver, kidney,	linoleate
		spleen, thymus,
		macrophages
PAFAH		Ubiquitous	AA, DHA
(lpPLA2)

These families can be subdivided into subfamilies. For example, the family of secreted PLA2 can contain 5 groups: group-1 (g1_SPLA2), 2 (g2_SPLA2), 3 (g3_SPLA2), 5 (g5_SPLA2), and 10 (g10_SPLA2).

Several members of the cPLA2 and iPLA2 families also possess PLA1, lysophospholipase, or triglyceride lipase activity, whereas _SPLA2s have only a PLA2 activity [62]. The conserved catalytic center of mammalian extracellular _SPLA2s contains a His–Asp dyad and a Ca ${}^{++}$ -binding loop, whereas that of intracellular cPLA2s and iPLA2s is characterized by a Ser–Asp dyad.⁵⁹

Whereas cPLA2s preferentially release arachidonic acid, iPLA2s liberate docosahexaenoic acid (DHA), a polyunsaturated fatty acid [63]. Arachidonic acid is a precursor of eicosanoids (Table 2.26). Docosahexaenoic acid is a precursor of neuroprotectin-D, a docosanoid.⁶⁰ It also contributes to prostanoid production.

Table 2.26

Signaling axes activators of PLA2 enzymes and effects (DAG: diacylglycerol; ERK: extracellular signal-regulated protein kinase; PDK: phosphoinositide-dependent kinase; PG: prostaglandin; PI3K: phosphatidylinositol 3-kinase; PIP₃: phosphatidylinositol triphosphate; PKB/C: protein kinase B/C; PLA2/C: phospholipase A2/C; Tx: thromboxane).

PLC–DAG–PKCα–cRaf–ERK1/2–cPLA2–AA–PGs/Tx
PLC–DAG–PKCα–iPLA2–AA
PI3K–PIP₃–PDK1–PKB–PKCα–iPLA2–AA

2.8.3.1 Secreted Phospholipases-A2

Many secreted forms of phospholipases-A2 (Table 2.27) hydrolyze phosphatidylcholine and lysophosphatidylcholine. Extracellular forms of phospholipases-A2 require Ca ${}^{++}$ ion. A secretory PLA2-binding protein that has a modest affinity for _SPLA2-1b with respect to _SPLA2-2a may possess a clearance function for circulating _SPLA2 as well as signaling functions [58]. These enzymes operate in digestion, reproduction, skin homoeostasis, and host defense, as well as in inflammation, tissue injury, and atherosclerosis.

Table 2.27

Secreted forms of phospholipase-A2 (Sources: [58, 251]; npsPLA2: non-pancreatic secretory PLA2; PLA2S: synovial PLA2; PPLA2: pancreatic PLA2; SPLASH: secretory-type PLA2, stroma-associated homolog). Pancreatic, group-1b phospholipase-A2 (PLA2-1b) interacts with PLA2-2a. Membrane-associated, group-2a phospholipase-A2 is detected in platelets and synovial fluid. Element PLA2G2C is a possible pseudogene.

Type	Gene	Other names
Conventional _SPLA2
_SPLA2-1b	PLA2G1B	G1b-PLA2, PLA2a, PPLA2
Conventional _SPLA2
_SPLA2-2a	PLA2G2A	G2a-PLA2, G2c-sPLA2, npsPLA2, sPLA2,
		PLA2b, PLA2L, PLA2S, PLAS1, MOM1
		(inflammatory sPLA2)
_SPLA2-2d	PLA2G2D	G2d-PLA2, G2d-sPLA2, sPLA2S, SPLASH
_SPLA2-2e	PLA2G2E	G2e-PLA2, G2e-sPLA2
_SPLA2-2f	PLA2G2F	G2f-PLA2, G2f-sPLA2
Atypical sPLA2
_SPLA2-3	PLA2G3	G3-PLA2, G3-sPLA2
Conventional sPLA2 without group-1 and -2 properties
_SPLA2-5	PLA2G5	G5-PLA2, G5-sPLA2
Conventional sPLA2 with group-1 and -2 properties
_SPLA2-10	PLA2G10	G10-PLA2, G10-sPLA2
Atypical sPLA2
_SPLA2-12a	PLA2G12A	G12a-PLA2, G12-sPLA2, ROSSY
_SPLA2-12b	PLA2G12B	G12b-sPLA2, G13-PLA2, G13-sPLA2, FKSG71

2.8.3.2 Family of Secreted Phospholipases-A2

Eleven _SPLA2s (secreted phospholipase-A2-1B, -2A, -2C–2F, -3, -5, -10, and -12A–12B) have been identified in mammals. They are subdivided into conventional _SPLA2 groups 1, 2, 5, and 10 and atypical groups 3 and 12. Subtype _SPLA2-2c is absent in humans (pseudogene) [62].

2.8.3.3 Effects of Secreted Phospholipases-A2

Secreted phospholipases-A2 can act as auto- or paracrine enzymes on plasma membranes to release not only arachidonic acid,⁶¹ but also various types of saturated and mono- and polyunsaturated fatty acids, such as ω3-eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which are precursors of anti-inflammatory lipid mediators [62]. However, the action of _SPLA2s on lipid mediator synthesis depends on the coordinated activation of cPLA2α. The other products, i.e., lysophospholipids, such as lysophosphatidylcholine and lysophosphatidic acid, have their own activities.

In fact, in humans, group-2A and -10 PLA2 release arachidonic acid mainly during their secretion on the inside rather than on the outside of the cell membrane [60]. Moreover, they may promote the exocytosis of substances during their own release, in particular the degranulation of mastocytes.

Moreover, _SPLA2s can also act on phospholipids of extracellular vesicles, pulmonary surfactant,⁶² and lipoproteins, as well as foreign phospholipids, such as dietary phospholipids and microbial membranes.

In mammals, G2A, G2D, and G5 PLA2s inhibit prothrombinase activity, as they bind to clotting factor Xa (potent anticoagulant activity) [60].

2.8.3.4 Receptors of Secreted Phospholipases-A2

In addition to their enzymatic action, _SPLA2s have a catalysis-independent function that relies on their ligand-like action, as they bind to numerous plasmalemmal receptors and soluble interactors, especially the catalytically weak or inactive group-12 _SPLA2s, which are widespread.

Different types of plasmalemmal receptors exist for _SPLA2s, such as N- and M-type receptors [64]. N-Type receptors of _SPLA2s exist in the nervous system. M-Type receptors of _SPLA2s that belong to the C-type lectin subclass⁶³ are expressed in various tissues.

Receptor PLA2R1 (CLec13c) interacts with _SPLA2 with high to moderate affinity, thereby hindering enzymatic activity of _SPLA2s [62]. Furthermore, the clearance receptor PLA2R1 internalizes bound _SPLA2 into phagolysosomes, in which the enzyme is rapidly degraded. However, in humans, the PLA2R1–_SPLA2 interaction specificity is not conserved.⁶⁴

Secreted PLA2s can tether not only to the PLA2R1 member of the mannose receptor family of C-type lectins, but also to glycosaminoglycans to possibly release cytokines, leukotrienes, and platelet-activating factor. In humans, _SPLA2 receptor transcripts can generate membrane-bound _SPLA2 receptors as well as secreted, soluble _SPLA2 receptors [65].

2.8.3.5 Secreted Phospholipases-A2 in the Cardiovascular Apparatus

Secreted phospholipase-A2-2A that requires calcium ion (millimolar concentration) for its enzymatic activity markedly increases in acute phase of inflammation, especially in cardiovascular diseases. Inflammatory cytokines, such as interleukin-1β and -6, interferon-γ, and tumor-necrosis factor-α, stimulate its synthesis in vascular smooth muscle cells and hepatocytes. Enzyme _SPLA2-2a hydrolyzes ester bonds in glyceroacyl phospholipids in cell membranes as well as both low- and high-density lipoproteins to release fatty acids and lysophospholipids. In addition, it contributes to an augmented production of interleukin-6 and cyclooxygenase-2 by lung macrophages [66] and mastocytes [67], respectively. Moreover, it stimulates differentiation of monocytes into dendritic cells [68].

Among the mammalian secretory _SPLA2s that hydrolyze lipoprotein phosphatidylcholine to lysophosphatidylcholine and free fatty acids in the arterial wall from the phosphatidylcholine-rich outer leaflet of the plasma membrane, _SPLA2-2f, -3, -5, and -10 have much more potent phosphatidylcholine-hydrolyzing activity than that of the others, including pro-inflammatory _SPLA2-2a, thereby contributing to the development of atherosclerosis [69].⁶⁵

2.8.3.6 Cytosolic Phospholipase-A2

Cytosolic PLA2 types participate in cell signaling (Tables 2.28 and 2.29). Two categories of cytosolic PLA2s exist whether they depend on Ca ${}^{++}$ ions (cPLA2), like secreted PLA2s (but are larger with different 3D structures than _SPLA2s), or not (Ca ${}^{++}$ -independent PLA2s [iPLA2]).

Table 2.28

Cytosolic, calcium-dependent types of phospholipase-A2 (Source: [58]). Group-4, cytosolic cPLA2s comprise calcium-dependent and -independent isoforms.

Type	Gene	Other names
cPLA2-4a	PLA2G4A	G4a-PLA2, cPLA2α
cPLA2-4b	PLA2G4B	G4b-PLA2, cPLA2β
cPLA2-4c	PLA2G4C	G4c-PLA2, cPLA2γ
cPLA2-4d	PLA2G4D	G4d-PLA2, cPLA2δ
cPLA2-4e	PLA2G4E	G4e-PLA2, cPLA2ε
cPLA2-4f	PLA2G4F	G4f-PLA2, cPLA2ζ

Table 2.29

Cytosolic, calcium-independent types of phospholipase-A2 (Sources: [60, 62, 71]; ATGL: adipose triglyceride lipase; GS2: gene sequence-2; GS2L: GS2-like phospholipase; INAD: infantile neuroaxonal dystrophy; kREH: keratinocyte retinyl ester hydrolase; NBIA: neurodegeneration with brain iron accumulation; NRE: neuropathy target esterase-related esterase; NTE: neuropathy target esterase; NTEL1: NTE-like esterase-1; NTER1: liver NTE-related protein-1; Park: parkinsonism; PEDFR: pigment epithelium-derived factor [serpin-F1] receptor; PnPLA: patatin-like phospholipase-A2 domain-containing protein; TTS: transport-secretion protein). Membrane-bound phospholipase iPLA2-6b, a.k.a. calcium-independent phospholipase-A2γ and intracellular, membrane-associated, Ca ${}^{++}$ -independent phospholipase-A2γ, possesses mitochondrial and peroxisomal localization signals. Phospholipases PnPLA6 and PnPLA7 contain a transmembrane domain.

Type	Gene	Group	Other names
PnPLA1	PNPLA1	G6a1	iPLA2 (iPLA2α)
PnPLA2	PNPLA2	G6e	iPLA2ζ, ATGL, desnutrin, PEDFR, TTS2, TTS2.2
PnPLA3	PNPLA3	G6d	iPLA2ε, adiponutrin (Adpn),
			acylglycerol ^Oacyltransferase
PnPLA4	PNPLA4	G6f	iPLA2η, GS2, kREH
PnPLA5	PNPLA5	G6g	GS2L
PnPLA6	PNPLA6	G6c	iPLA2δ, NTE
PnPLA7	PNPLA7	G6h	NRE, NTEL1, NTER1
PnPLA8, iPLA2-6b	PNPLA8	G6b	iPLA2γ
PnPLA9, iPLA2-6a	PLA2G6	G6a2	iPLA2β, INAD1, NBIA2a, NBIA2b, Park14

Group-4, cytosolic phospholipase-A2 types (g4aPLA2–g4fPLA2 or cPLA2-4a–cPLA2-4f) hydrolyze membrane phospholipids to release arachidonic acid that is subsequently metabolized into eicosanoids. Lysophospholipids that are also produced can be converted into platelet-activating factor. They are activated by an increased intracellular Ca ${}^{++}$ concentration and on phosphorylation.

Group-4 cPLA2 is constituted by: (1) calcium-independent PLA2s such as cPLA2-4c that is constitutively associated with membranes and functions for PGE2 production and (2) calcium-dependent cPLA2 such as cPLA2-4a enzyme. The first discovered cytosolic PLA2 was 85 kDa cPLA2-4a. It bears Ca ${}^{++}$ -directed translocation to perinuclear membranes.

Calcium-dependent phospholipases-A2 intervene in excitotoxic functions in neurons, whereas _SPLA2-2a operate in response to inflammation in glial cells [63]. In addition, cPLA2 and _SPLA2 influence cell membrane properties and dynamics.

2.8.3.7 Intracellular Calcium-Independent Phospholipases-A2 – Patatin-like Phospholipases

Patatins constitute a family of soluble glycoproteins of mature potato tubers. These storage proteins operate as non-specific lipid acyl hydrolases that have the serine lipase Gly-X-Ser-X-Gly motif. Members of the family of patatins cleave fatty acids from membrane lipids and act as defense against plant parasites.

Nine patatin-like phospholipases (PnPLA1–PnPLA9) characterized by a catalytic Ser–Asp dyad are encoded by the human genome [72, 73]. Mammalian PnPLAs target diverse substrates, such as triacylglycerols, phospholipids, and retinol esters.⁶⁶

In humans, patatin-like hydrolases that do not have phospholipase activity participate in the regulation of adipocyte differentiation. In human preadipocytes, several members are indeed differentially regulated during cell differentiation [72]. The expression of PnPLA2, PnPLA3, and PnPLA8 is upregulated during adipocyte differentiation.

Three patatin-like proteins (PnPLA2–PnPLA4) link to the plasma membrane. They have catabolic (triacylglycerol) and anabolic (transacylation) activities [72]. Isoforms PnPLA4, PnPLA7, and PnPLA8 are widespread. Structure, enzymatic activity, or function of PnPLA1 have not been deeply investigated.

PnPLA2

Subtype PnPLA2 acts coordinately with hormone-sensitive lipase in the catabolism of triglycerides. The highest expression levels of this widely distributed protein are found in white and brown adipose tissue [73]. In the cell, it localizes to lipid droplets, where it catalyzes the initial step in triacylglycerol hydrolysis, thereby yielding diacylglycerol and fatty acids.

The activator of PnPLA2, Comparative gene identification CGI58, or α/β-hydrolase domain-containing protein α/βHD5, enhances PnPLA2-mediated triacylglycerol hydrolysis without affecting hormone-sensitive lipase activity.⁶⁷ This activator binds to perilipin, a lipid droplet protein. When protein kinase-A phosphorylates perilipin, CGI58 dissociates from perilipin and colocalizes with PnPLA2 [73].

Subtypes PnPLA2, PnPLA3, and PnPLA4 possess high triacylglycerol lipase and acylglycerol transacylase activities, in addition to PLA2 activity [60].

PnPLA3

Patatin-like phospholipase-A domain-containing protein PnPLA3, or adiponutrin, localizes to cellular membranes of adipocytes. It serves as an indicator of the nutritional state. Its concentration rises during adipocyte differentiation as well as upon insulin stimulation, but lowers during fasting [73]. In vitro, human PnPLA3 has triacylglycerol hydrolase, acylglycerol transacylase, and modest PLA2 activity. In vivo, these activities are minor.

PnPLA4

Despite its wide distribution, PnPLA4 is predominantly expressed in both muscle, heart, and adipose tissue. It may function as acylglycerol and retinol transacylase, triacylglycerol hydrolase, and PLA2 [73].

PnPLA5

Isoform PnPLA5 is produced at low levels in the brain, lung, brown and white adipose tissues, and hypophysis (pituitary gland). In vitro, it may operate as a triacylglycerol hydrolase [73]. Like PnPLA3, it is regulated by the nutritional status.

PnPLA6

Patatin-like phospholipase-A domain-containing protein PnPLA6 is an essential lysophospholipase in the brain [73].⁶⁸

It localizes to the endoplasmic reticulum and Golgi body. In vitro, it acts as an esterase and hydrolase against membrane lipids, preferentially lysophospholipids.

In lymphocytes, PnPLA6 hydrolyzes numerous lysophospholipids and is involved in membrane trafficking [72]. Both PnPLA6 and PnPLA7 are integral membrane proteins that can be regulated by cAMP or cGMP nucleotides [72].

PnPLA7

Subtype PnPLA7 is an insulin-regulated lysophospholipase in muscle and adipose tissue [73]. In humans, PnPLA7 is primarily synthesized in white adipose tissue, pancreas, and prostate. Its expression is strongly induced by fasting.

PnPLA8

Subtype PnPLA8 is a myocardial phospholipase maintaining mitochondrial integrity [73]. Its expression is prominent in the human heart, but also, to a lower extent, in other tissues. It localizes to mitochondria and peroxisomes. It exhibits both PLA1 and PLA2 activity for saturated or monounsaturated fatty acids of phospholipids.

PnPLA9

Subtype PnPLA9 is activated during apoptosis [72]. It hydrolyzes (as an acetyl hydrolase) platelet-activating factor. It mainly localizes to the cytoplasm; upon stimulation, it translocates to membranes of the perinuclear area [73].

It may intervene in the differentiation of bone marrow stromal cells, as, in the absence of PnPLA9, undifferentiated bone marrow stromal cells evolve toward adipogenesis rather than osteogenesis [73].

Subtype PnPLA9 operates in phospholipid processing, eicosanoid formation upon arachidonic acid release, protein expression, acetylcholine-mediated endothelium-dependent relaxation of the vasculature, secretion, and lymphocyte proliferation, in addition to apoptosis [60].

2.8.3.8 Intracellular Calcium-Dependent Phospholipases-A2

Enzyme cPLA2 was first detected in platelets and macrophages. The cPLA2 family, or group-4 PLA2, includes at least 6 isoforms. Subtype cPLA2α is the most ubiquitous enzyme in this group. Its regulation involves subcellular localization, intracellular Ca ${}^{++}$ content (Ca ${}^{++}$ -dependent lipid binding), phosphorylation, proteic interaction, and cleavage. Activated cPLA2 specifically cleaves the acyl ester bond of phosphatidylcholine, thereby generating lysophosphatidylcholine and release free fatty acid such as arachidonic acid.

In non-neural cells, cPLA2 targets preferentially phospholipids in endoplasmic reticulum and Golgi membranes [63]. It can also process other subcellular membranes, such as mitochondrial, nuclear, and plasma membranes. Processing of membrane phospholipids changes physical properties of cellular membranes. Enzyme PLA2 decreases the liquid-disordered and increases the liquid-ordered phase [63].

In neurons and glial cells, intracellular Ca ${}^{++}$ -dependent phospholipases-A2 (cPLA2) are coupled to Ca ${}^{++}$ -mobilizing receptors and oxidative signaling [63]. In astrocytes, cPLA2 can be activated via G-protein-coupled P2Y₂ receptor that stimulates MAPK and PKC kinases, as well as angiotensin-2 AT₁ receptor. In neurons, activated cPLA2 stimulates the NMDA-type glutamate receptor and its signaling via NADPH oxidase that resides in particular in synapses, hence the production of reactive oxygen species. The latter activate the signaling axis that causes phosphorylation of extracellular signal-regulated kinases. Reactive oxygen species may modulate synaptic function via cPLA2 and the release of arachidonic acid preferentially from phosphatidylcholine. Arachidonic acid can inhibit both pre- and postsynaptic potassium channels [63].

Activation of neuronal receptors can stimulate cPLA2 though Ca ${}^{++}$ influx from the extracellular space or via the coupling to Gi-associated Gβγ subunit [61]. On the other hand, Ca ${}^{++}$ ion attenuates iPLA2β activity that promotes iPLA2β–calmodulin interaction, which impedes iPLA2 activity. In unstimulated cells, SERCA pump works continuously to fill the endoplasmic reticulum with Ca ${}^{++}$ ions and calmodulin inhibits iPLA2 protein. When receptors of the neuronal plasma membrane (purinergic, muscarinic, or metabotropic glutamatergic receptors) are stimulated, Ca ${}^{++}$ can be released from the endoplasmic reticulum through inositol trisphosphate and ryanodine receptors (Ca ${}^{++}$ -induced Ca ${}^{++}$ release).

cPLA2-4a – Phospholipase-A2α

Cytosolic, Ca ${}^{++}$ -binding, phospholipase-A2α contributes to limit maladaptive hypertrophy of cardiomyocytes that results from pressure overload [43].⁶⁹ Elevated intracellular Ca ${}^{++}$ concentration causes translocation of PLA2α to cellular membranes. Calmodulin-dependent kinase CamK2 and mitogen-activated protein kinase phosphorylate PLA2α to modulate the release of arachidonic acid [43]. In addition, PLA2α precludes activation of adenylate cyclase and protein kinase-A induced by β2-adrenoceptors.

The cPLA2α activity is regulated by phosphorylation and nitrosylation. Enzyme cPLA2α is phosphorylated by extracellular signal-regulated kinases ERK1 and ERK2, P38MAPK, Ca ${}^{++}$ –calmodulin-dependent protein kinase CamK2, and MAPK-interacting kinase MNK1 [74]. Messengers ATP and UTP stimulate cPLA2α phosphorylation via PKC-dependent and -independent ERK pathways. In noradrenaline-stimulated vascular smooth muscles cells, phosphorylation of cPLA2α (Ser515) by CamK2 is required for the phosphorylation by ERK1 and ERK2 (Ser505). Enzyme cPLA2α can colocalize with cyclooxygenase-2 in the plasma membrane [74]. In macrophages, production of prostaglandins via the cPLA2–COx2 axis regulates expression of the Nos2 gene. In human epithelial cells, nitric oxide produced by inducible nitric oxide synthase (NOS2) enhances cPLA2α activity via S-nitrosylation.

cPLA2-4b – Phospholipase-A2β

In the myocardium, the predominant PLA2 isoform is Ca ${}^{++}$ -independent PLA2β that possesses calmodulin- and ATP-binding domains, hence is sensitive to local Ca ${}^{++}$ concentration and energetic status [43]. Ca ${}^{++}$ –calmodulin inhibits PLA2β. Enzyme PLA2β also hydrolyzes fatty acylCoA.⁷⁰ Hypoxia primes PLA2β activity in cardiomyocytes.

cPLA2-4c – Phospholipase-A2γ

A second major PLA2 isoform in myocardium corresponds to Ca ${}^{++}$ -independent PLA2γ. Multiple splice variants of PLA2γ have been detected. Enzyme PLA2γ sequentially hydrolyzes diacyl, arachidonate-containing phospholipids [43]. Afterward, PLA2α or lysophospholipase hydrolyzes 2-arachidonoyl lysophosphatidylcholine to generate a large amount of arachidonic acid for eicosanoid production in the myocardium. On the other hand, 2-arachidonoyl lysophosphatidylcholine is targeted by the nucleotide pyrophosphatase–phosphodiesterase to generate the endocannabinoid 2-arachidonoyl glycerol, a substrate for cyclooxygenase-2 (but not COx1), and 12- and 15-lipoxygenases.

cPLA2-4d – Phospholipase-A2δ

The PLA2G4D mRNA can be identified in the placenta [75]. Upon stimulation, cPLA2δ that localizes in the cytoplasm in the resting state translocates to the perinuclear region, like cPLA2α, but the motion takes a longer time (5 mn instead of 1 mn for cPLA2α) [75].⁷¹

cPLA2-4e – Phospholipase-A2ε

A transcript of the PLA2G4E gene is observed predominantly in the heart, skeletal muscle, thyroid, and testis, and, at low expression levels, in the brain and stomach [75]. Enzyme cPLA2ε is partly associated with lysosomes, but neither with the endoplasmic reticulum, Golgi body, nor mitochondria. Stimulation does not cause redistribution of cPLA2ε and cPLA2ζ until at least 10 mn [75].⁷²

cPLA2-4f – Phospholipase-A2ζ

The PLA2G4F mRNA is detected strongly in thyroid, moderately in stomach, and very weakly in large intestine and prostate [75].⁷³

2.8.3.9 Platelet-Activating Factor Acid Hydrolases

Platelet-activating factor acid hydrolases (PAFAH), or lipoprotein-associated phospholipases-A2 (lpPLA2), are classified as [61]: g7aPLA2, or lipoprotein-associated PLA2;⁷⁴ g7bPLA2, or PAFAH2; g8aPLA2, or PAF1b; and g8bPLA2, which heterodimerizes with g8aPLA2 (Table 2.30).

Table 2.30

Platelet-activating factor acid hydrolases and lysosomal group-15 phospholipase-A2 (Source: [60] LDL: low-density lipoprotein; lPLA2: lysosomal PLA2; lpPLA2: lipoprotein-associated PLA2; PL: phospholipid).

Group	Aliases	Features
G7a	lpPLA2	Secreted, plasma PAFAH,
		catalytic Ser–His–Asp triad,
		hydrolysis of short- and medium fatty acids
		from diacylglycerols and triacylglycerols,
		hydrolysis of oxidized PL in LDLs,
		PLA1 activity
G7b	PAFAH2	Intracellular, myristoylated
G8a	PAFAH1bα1	Intracellular, Ser–His–Asp triad,
		homodimer or g8aPLA2–g8bPLA2 heterodimer
G8b	PAFAH1bα2	Intracellular, Ser–His–Asp triad,
		homodimer or g8aPLA2–g8bPLA2 heterodimer
G15	lPLA2,	Ser–His–Asp triad,
	LLPL	glycosylated

Enzyme PAFAH cleaves the acetyl group from the sn2 position of platelet-activating factor. Furthermore, this enzyme can cleave oxidized lipids in the sn2 position of 9-carbon-long chains [61]. Unlike other PLA2s, PAFAH access substrate in the aqueous phase in a Ca ${}^{++}$ -independent fashion.

Type-1 PAFAH (PAFAH1b) is a complex of 2 catalytic subunits (PAFAH1bα1–PAFAH1bα2), a regulatory β-subunit, and a γ-subunit. In humans, these subunits are encoded by the PAFAH1B1 to PAFAH1B3 genes that produce platelet-activating factor acetylhydrolase isoform-1B subunit-1 to -3, or PAFAH α-, β-, and γ-subunit, respectively [76]. The catalytic subunits G8aPLA2 and G8bPLA2 form catalytically active homo- and heterodimers. The g8bPLA2–g8bPLA2 homodimer preferentially targets platelet-activating factor (1-^Oalkyl 2-acetyl glycero 3-phosphocholine) and 1-^Oalkyl 2-acetyl glycero 3-phosphoethanolamine; the g8aPLA2–g8aPLA2 homodimer and the g8aPLA2–g8bPLA2 heterodimer have a higher activity toward 1-^Oalkyl 2-acetyl glycero 3-phosphatidic acid [60].

Type-2 PAFAH, or serine-dependent phospholipase-A2 (sdPLA2), is a single polypeptide encoded by the Pafah2 gene. It is myristoylated at its N-terminus. It resides both in the cytosol and membranes. It indeed translocates from the cytosol to membranes in the presence of oxidants and from membranes to the cytosol in the presence of anti-oxidants [60].

Plasma PAFAH, or lpPLA2 (g7aPLA2), has a structure close to that of PAFAH2. It associates with plasma lipoproteins. In the blood, it travels mainly with low-density lipoproteins; less than 20% is connected to high-density lipoproteins. It is produced by inflammatory cells. This secreted enzyme catalyzes the degradation of PAF to inactive products lysoPAF and acetate. It also hydrolyzes oxidized phospholipids in LDL particles, hence its name oxidized lipid lipoprotein-associated PLA2 subtype.

2.8.3.10 Lysosomal Phospholipase-A2

Lysosomal PLA2, or group-15 PLA2, is a calcium-independent enzyme that has its maximal activity in an acidic medium (optimal enzymatic activity at pH 4.5), hence its other label acidic iPLA2 enzyme (Table 2.30). This enzyme was first named 1-^Oacylceramide synthase (ACS). It possesses Ca ${}^{++}$ -independent PLA2 and transacylase activities [60]. It preferentially hydrolyzes phosphatidylcholine and phosphatidylethanolamine.

2.8.3.11 Phospholipases-A2 in Endothelial Cells

In endothelial cells, phospholipases-A2 can be involved in blood–brain barrier function, adhesion and transmigration of inflammatory cells, and angiogenesis (Table 2.31). In the vasculature and blood, PLA2s are active in leukocytes, macrophages, platelets, smooth muscle cells, preadipocytes, and fibroblasts.

Table 2.31

Intracellular PLA2 isoforms in endothelial cells in humans (EC; Source: [77]). Group-2 sPLA2s are not detected in quiescent endothelial cells; thay are produced upon cytokine stimulation. Group-6A iPLA2s contribute to regulation of endothelial cell proliferation.

Endothelial cell type	Group(s)	PLA2 isoforms
Aortic ECs	4, 5	cPLA2, _SPLA2
Bone marrow ECs	2	_SPLA2
Brain microvascular ECs	4	cPLA2
Coronary artery ECs	2, 4–6	cPLA2, iPLA2, _SPLA2
Umbilical vein ECs	4–6	cPLA2, iPLA2, _SPLA2

Kinases ERK1 and ERK2 phosphorylate (activate) cPLA2 concomitantly with Ca ${}^{++}$ influx. The latter fosters cPLA2 translocation to cellular membranes (endoplasmic reticulum, Golgi body, and nuclear envelope) [77]. Ceramide 1-phosphate is an allosteric activator of cPLA2α. Protein kinase Cα phosphorylates (activates) either iPLA2 or its regulator. This kinase can be activated by diacylglycerol and the PI3K–PDK1–PKB axis.

Free arachidonic acid is oxidized to prostaglandins by cyclooxygenases COx1 and COx2,⁷⁵ hydroxyeicosanoic acid by lipoxygenases, or epoxyeicosatrienoic acids by P450-dependent epoxygenase. Lysophospholipid, the other product of PLA2, can generate the precursor of platelet-activating factor, another potent inflammatory mediator.

2.8.4 Phospholipase-B

Phospholipase-B (PLB) attacks lysolecithin, releasing glycerylphosphorylcholine and fatty acid.⁷⁶ Certain phospholipase B types hydrolyze phosphatidylcholine, phosphatidylinositol, and phosphatidylethanolamine.

2.8.5 Phospholipase-C

Phospholipase-C (PLC) cleaves phosphatidylinositol (4,5)-bisphosphate at the plasma membrane to produce 2 second messengers: inositol (1,4,5)-trisphosphate and diacylglycerol. The latter remains bound to the membrane, where it can activate protein kinase-C to phosphorylate its substrates. The former is released as a soluble molecule into the cytosol, where it targets IP₃ receptors, particular those in the endoplasmic reticulum to raise the cytosolic concentration of calcium.

Messenger IP₃ can be phosphorylated by kinases to produce inositol hexakisphosphate IP₆, mono- and bispyrophosphorylated inositol phosphates PPIP₅ and (PP)₂IP₄. Agent PPIP₅ is implicated in gene expression and protein phosphorylation, among other functions. Inositol (1,3,4,5)-tetrakisphosphate (IP₄) is the polar head group of phosphatidylinositol (3,4,5)-trisphosphate, which allows the PIP₃ effector to distinguish PIP₃ from its PIP₂ precursor. Phosphorylation of IP₃ into IP₄ promotes PLCγ phosphorylation [17].

Thirteen types of mammalian phospholipases-C are classified into 6 groups (PLCβ–PLCη) according to the structure and activation mechanisms. Subtype-specific domains indeed contribute to specific regulations. Each isoform type (PLCβ1–PLCβ4, PLCγ1–PLCγ2, PLCδ1–PLCδ3, PLCε1, PLCζ1, or PLCη1–PLCη2) require calcium for catalysis.

Isoform PLCβ is activated by G proteins of the Gq subclass (Gq, G11, and G14–G16) as well as Gβγ liberated from Gi subunit; PLCγ by receptor Tyr kinases,⁷⁷ PLCδ by high calcium levels; and PLCε by small GTPase Ras;⁷⁸ PLCη has a high sensitivity to the calcium concentration (PLCη2 being activated at calcium level $\mathcal{O}$ [10 nmol]). Sperm PLCζ can cause a large, transient increase in free Ca ${}^{++}$ concentration in an ovum (or unfertilized oocyte) upon encountering a spermatozoon (i.e., a motile sperm cell). Activity of isozymes PLCβ and PLCγ are stimulated by bradykinin and platelet-derived growth factor, respectively.

Isoform PLC-β1 is highly expressed in the central nervous system; PLC-β2 slightly in hematopoietic cells (but at greater value than in other tissues); PLCβ3 in the brain, liver, and parotid gland; PLCβ4 particularly in the cerebellum and retina [78] (Table 2.32). Subtype PLCγ1 abounds in the embryon cerebral cortex; PLCγ2 localizes primarily to hematopoietic cells. Isozyme PLCδ1 abounds in the brain, heart, lung, skeletal muscle, and testis and PLCδ3 in the brain, heart, and skeletal muscle; PLCδ4 resides especially in the brain, kidney, skeletal muscle, and testis. PLCε is detected in the brain, lung, and colon, with the highest expression in heart. PLCζ1 is only found in the testis. The highest level of PLCη1 is observed in the brain and kidney and, at smaller levels in the lung, spleen, intestine, thymus, and pancreas; PLCη2 is identified in the brain and intestine.

Table 2.32

Tissue distribution of PLC isoforms in the cardiovascular and ventilatory systems as well as related organs (Source: [78]).

Type	Localization
PLCβ1	Brain, nerve, kidney, lung, adipose tissue
PLCβ2	Brain, ganglia, kidney, lung, blood, bone marrow, lymph node
PLCβ3	Brain, nerve, kidney, lymph node, adipose tissue
PLCβ4	Brain, nerve, ganglia, heart, blood, kidney, lung
PLCγ1	Brain, ganglia, heart, blood, bone marrow, lymph node,
	kidney, lung, adipose tissue
PLCγ2	Brain, ganglia, nerve, heart, blood, bone marrow, lymph node,
	kidney, lung, adipose tissue
PLCδ1	Brain, ganglia, heart, blood, bone marrow, kidney, lung
PLCδ3	Brain, heart, blood, lymph node, kidney, lung
PLCδ4	Brain, heart, kidney, lung
PLCε	Brain, heart, kidney, lung, blood
PLCζ	Brain
PLCη1	Brain, lung
PLCη2	Brain, kidney, lung, lymph node

2.8.5.1 Phospholipase-Cβ

Phospholipases-Cβ constitute a family that includes PLCβ1 to PLCβ4. They possess an elongated C-terminus. These isoforms are regulated by heterotrimeric guanine nucleotide-binding (G) proteins. They have a high GTPase-stimulating activity. Dimer Gβγ interacts with PLCβ1 to PLCβ3, but exhibits a high affinity only for PLCβ2 [78]. In platelets, various signals activate PLCβ isozymes via Gq (but not Gβγ).

Many signaling cascades relies on inositol lipid hydrolysis catalyzed by phospholipase-Cβ⁷⁹ that is activated by Gq subunit of heterotrimeric G proteins, and, in turn, inactivates Gq, thus limiting the signaling duration, a procedure that is required when signal transduction necessitates high amplification and rapid activation–deactivation cycles [79].⁸⁰

Isoform PLCβ2 can be activated by Gα_q and Gβγ subunits of heterotrimeric G proteins as well as small, monomeric guanosine triphosphatase Rac. These 3 activators use overlapping and distinct mechanisms to recruit PLCβ2 to the plasma membrane and activate signaling [80]. Activation of PLCβ2 by Gq protomer causes PLCβ2 association with fast-diffusing plasmalemmal constituents (PLCβ2 movement within the membrane similar to that of lipids), thereby searching for relatively localized substrates. PLCβ2 substrate sampling is carried out over mid-size zone prior to dissociation. Stimulation of PLCβ2 by Gβγ dimer generates pure lateral diffusion 3- to 5-fold faster than lipids, hence transient linkage to the membrane that allow PLCβ2 to scan the plasma membrane over large regions to interact with dispersed substrates. Excitation of PLCβ2 by Rac2 induces slow lateral diffusion and much faster PLCβ2 interaction with slowly diffusing molecules in the membrane, hence creating spatially restricted signaling.

2.8.5.2 Phospholipase-Cγ

Phospholipase-Cγ hydrolyzes PI(4,5)P₂ into IP₃ and DAG messengers. Two mammalian PLCγ isoforms (PLCγ1 and PLCγ2) have been identified. They are activated by receptor Tyr kinases as well as other types of receptors. Isoform PLCγ1 can be stimulated by GPCRs, such as angiotensin-2 and bradykinin receptors, cytokine receptors, and immunoreceptors such as T-cell receptors [78]. Isozyme PLCγ2 is activated by immunoglobulin and adhesion receptors on B lymphocytes, platelets, and mastocytes.

During cell proliferation caused by growth factors, i.e., during phenotypic change from a differentiated contractile phenotype to a proliferative phenotype, marker genes of vascular smooth muscle cells are repressed, partly following translocation of extracellular signal-regulated kinases ERK1 and ERK2 to the nucleus that phosphorylate (activate) ELk1 transcription factor. In particular, upon stimulation by platelet-derived growth factor and subsequent activation of PLCγ1, ERK-binding, 15-kDa phosphoprotein enriched in astrocytes PEA15 is phosphorylated and, then, supports the nuclear translocation of ERK1 and ERK2 [81].⁸¹

In humans, naive T lymphocytes⁸² do not express vitamin-D receptor. Moreover, they produce a very low amount of phospholipase-Cγ1, thereby displaying a low T-cell antigen receptor responsiveness. They are thus not able to deliver a specific antigenic response. Vitamin-D controls T-cell receptor signaling that primes the synthesis of vitamin-D receptor via mitogen-activated protein kinase P38MAPK, upregulates PLCγ1 expression ( ∼ 75 fold) and causes a greater TCR responsiveness required for subsequent classical T-lymphocyte activation [82].

2.8.5.3 Phospholipase-Cδ

Three isozymes have been identified in the PLCδ family (PLCδ1, PLCδ3, and PLCδ4). Phosphatidylserine activates PLCδ1, but not PLCδ4 [83]. PLCδ1 and PLCδ4 have similar activities and dependence on Ca ${}^{++}$ cofactor. Although all PLC isozymes require calcium for activity, PLCδ isozymes are among the most sensitive to calcium [78].

Rho GTPase-activating protein RhoGAP7 (or P122RhoGAP) binds and activates PLCδ1 to suppress the formation of stress fibers and focal adhesions [78]. In addition, PLCδ1 shuttles between the cytoplasm and nucleus. Isozyme PLCδ1 is involved in osmotic response.

Differences exist between the C2 domain of PLCδ4 and those of PLCδ1 and PLCδ3. The C2 domain of the former that has Ca ${}^{++}$ -independent membrane affinity allows localization to various cellular membranes, including the nuclear envelope and plasma membrane and engages in between-protein interaction [83]. It interacts with glutamate receptor-interacting protein GRIP1. The C2 domain of PLCδ1 and PLCδ3 support linkage specifically to the plasma membrane with a strong Ca ${}^{++}$ -dependent membrane affinity.

In the regenerating liver, PLCδ4 is detected in the nucleus during the S phase of the cell cycle, whereas PLCδ1 mostly resides in the cytosol [83].

Phospholipase-Cδ4 is expressed in many tissues. Its transcription is primed by growth factors as well as bradykinin and lysophosphatidic acid. Unlike other PLCδ isozymes, PLCδ4 has 3 alternative splice variants (PLCδ4_alt1–PLCδ4_alt3) [83]. Variant PLCδ4_alt3 is catalytically inactive. It may act as an inhibitor of PLC activity. The alternative splice variants PLCδ4_alt1 and PLCδ4_alt2 are expressed mainly in the testis and PLCδ4_alt3 primarily in some neuron types.

2.8.5.4 Phospholipase-Cε

Phospholipase-Cε (or 1-phosphatidylinositol (4,5)-bisphosphate phosphodiesterase-ε1) is the largest identified PLC isozyme. Small GTPases hRas, Rap (Rap1a and Rap2b upon stimulation by growth factor RTKs and GPCRs, respectively [78]), and Rho activates PLCε isozyme. Subtype PLCε is activated by Gβγ subunit released upon activation by Gα_{12 ∕ 13}-, but not Gα_q-coupled receptors [84].

Dermal fibroblasts and epidermal keratinocytes expressed pro-inflammatory cytokines via PLCε once stimulated with cytokines (IL4, IL17, Ifnγ, and TNFα) released by infiltrated, CD4 + T lymphocytes [85].

Two spliced variants have been identified (PLCε1a–PLCε1b). The PLCε1a transcript is expressed in various tissues, but not in blood leukocytes; PLCε1b has a limited expression in the lung, spleen, and placenta.

2.8.5.5 Phospholipase-Cη

Phospholipase-Cη has been identified as a sperm-specific isozyme. It contains a nuclear localization signal that promotes its accumulation in the nucleus [78]. On the other hand, phospholipase-Cη and adaptor B-cell linker cause preB-cell differentiation.

The family of PLCη consists of 2 enzymes (PLCη1 and PLCη2). They are more sensitive to Ca ${}^{++}$ than other PLC isozymes. They mediate G-protein-coupled receptor signaling. Both enzymes are expressed in neurons and neuroendocrine cell lines [86].

2.8.5.6 Phospholipase-Cζ

Two PLCζ isozymes (PLCζ1 and PLCζ2) exist in humans. Both PLCζ isoforms are highly expressed in cerebral neurons. They may then be involved in the regulation of neural or neuroendocrine systems [78].

Isozyme PLCζ1, the smallest known mammalian PLC isotype, elicits Ca ${}^{++}$ oscillations following phosphatidylinositol (4,5)-bisphosphate hydrolysis and activation of embryogenesis. It is approximately 100 times more sensitive to Ca ${}^{++}$ than the closely related PLCδ1 isoform [87].

2.8.6 Phospholipase-D

Phospholipase-D hydrolyzes phosphatidylcholine and releases phosphatidic acid. Phosphatidic acid controls various cellular functions, such as cell proliferation and survival via signaling mediated by Son of sevenless and target of rapamycin, respectively, as well as cell migration. Phosphatidic acid is hydrolyzed into diacylglycerol by phosphatidic acid phosphomonoesterase.⁸³

Phosphatidic acid activates some guanine nucleotide-exchange factors such as Dedicator of cytokinesis-2 (DOCK2). Activator DOCK2 is a RacGEF that mediates neutrophil chemotaxis. Upon PIP₃ formation, DOCK2 is recruited to the plasma membrane. Subsequently, PLD is activated and produces phosphatidic acid. Moreover, phosphatidic acid binds directly to SOS protein.

Phospholipase-D also concurs to Rac recruitment to the plasma membrane. It binds to GRB2 that recruits Son of sevenless. In addition, phosphatidylinositol triphosphate helps to unmask the GEF activity of Son of sevenless for Rac.

Platelets contain protein kinase-C and phospholipase-C and -D. Phospholipase-D is rapidly activated once platelets are stimulated to aggregate and secrete their materials. Phospholipase-D activation requires protein kinase-C, but not α_2Bβ₃-integrin and fibrinogen binding [88]. Yet, both α_2Bβ₃-integrin and fibrinogen are necessary for full platelet aggregation. The main adhesion receptor, α_2Bβ₃-integrin, can shift from a low- to high-affinity state for its ligands to enable platelet adhesion and aggregation. In response to stimulation of G-protein-coupled- and immunoreceptors, phospholipases cleave membrane phospholipids to generate soluble second messengers. In adherent platelets, whereas phosphatidylinositol 3-kinase activation is controlled by both α_2Bβ₃-integrin engagement and released ADP, phospholipase-C is stimulated only by α_2Bβ₃-integrin–fibrinogen interaction [89].

Two PLD isoforms exist (PLD1–PLD2). In addition to phospholipase-C, phospholipase-D1 and its product phosphatidic acid intervene in α_2Bβ₃-integrin activation in response to agonists and platelet aggregation under high shear mediated by glycoprotein-1B [90]. Glycoprotein-1B serves as a receptor for von Willebrand factor and a component of the GP1b–GP5–GP9 complex on platelets.⁸⁴

Phospholipase-D2 binds directly to Rac2 GTPase, a regulator of actin cytoskeletal remodeling, and functions as a guanine nucleotide-exchange factor [91]. In leukocytes, PLD2 thus intervene in cell polarization and adhesion, chemotaxis, and phagocytosis.

2.9 Phosphoinositide Kinases

Phosphatidylinositol phosphate kinases (PIPK) phosphorylate 3 hydroxyl groups (3–5) of phosphatidylinositol among 5 because of hindrance. These kinases phosphorylate PIP and PIP₂ phosphoinositides. The set of PIPKs is subdivided into 3 groups with significant sequence homology, but distinct substrate specificities, subcellular locations, and functions.

Kinases and phosphatases that regulate phosphoinositide pathways participate in cell fate. Both PIP₂ and PIP₃ control actin polymerization. In particular, phosphoinositides and phosphoinositide kinases and phosphatases are involved in cell transport (Vol. 1 – Chap. 9. Intracellular Transport). Moreover, plasmalemmal gradients of PIP₃ organized by phosphoinositide kinases and phosphatases are achieved during cell migration (Vol. 2 – Chap. 6. Cell Motility).

2.9.1 Phosphatidylinositol 3-Kinases

Phosphoinositide 3-kinases (PI3K) phosphorylate different phosphoinositides to produce PI(3)P, PI(3,4)P₂, PI(3,5)P₂, and PI(3,4,5)P₃ once stimulated by their activators.⁸⁵ They operate early in intracellular signal transduction. They contribute to the recruitment and activation of several protein kinases, such as phosphoinositide-dependent protein kinase PDK1 that phosphorylates (activates) AGC protein kinases (Sect. 5.2), especially protein kinase-B (Table 2.33). Second messenger PI(3,4,5)P₃ interacts with kinases of the TEC family (Sect. 4.12.12), GTPase-activating proteins, guanine nucleotide-exchange factors, and scaffold proteins. In addition to their lipid kinase activity, several PI3Ks also have protein kinase activity.

Table 2.33

Targets of antagonists phosphatidylinositol 3-kinase (PI3K) and phosphatase and tensin homolog deleted on chromosome 10 (phosphatidylinositol 3-phosphatase [PTen]). Phosphoinositide 3-kinases phosphorylate PI, PI(4)P, and PI(4,5)P₂ to produce PI(3)P, PI(3,4)P₂, and PI(3,4,5)P₃. Lipid PI(3,5)P₂ is synthesized by PIP5K3 (PIKfyve). These lipidic mediators interact with multiple proteic effectors (kinases, adaptors, guanine nucleotide-exchange factors, and GTPase-activating proteins). Lipid phosphatases (PTen, inositol polyphosphate 4-phosphatase INPP4, phosphatidylinositol polyphosphate 5-phosphatase INPP5e, SH2 domain-containing inositol 5-phosphatase SHIP2, myotubularin phosphatases, a dual phosphatidylinositol 3-phosphatase and (3,5)-bisphosphatase) antagonize PI3K kinases (BAD: BCL2 antagonist of cell death; BTK: Bruton Tyr kinase; eIF4eBP1: eukaryotic translation initiation factor-4e-binding protein; FoxO: type-O forkhead box transcription factor; IKK: IκB kinase; GSK: glycogen synthase kinase; GyS: glycogen synthase; ITK: interleukin-2-inducible T-cell kinase; NFAT: nuclear factor of activated T cells; NOS: nitric oxide synthase; PDK: phosphoinositide-dependent kinase; PFK: phosphofructokinase; PKB(C): protein kinase-B(C); PKB1S1: PKB1 substrate-1; RLK: resting lymphocyte kinase; SGK: serum and glucocorticoid-regulated kinase S6K: p70 ribosomal S6 kinase; TOR: target of rapamycin; TEC: Tyr kinase expressed in hepatocellular carcinoma; ( + ): activation; ( − ): inactivation).

Target	Activity
PDK1	Phosphorylation ( + ) of PKB, PKC, SGK1, S6K
GSK3	Phosphorylation ( − ) of β-catenin, cyclin-D, GyS, Jun, NFAT
PKB	Phosphorylation of BAD, FoxO3α, DM2, Rho, PKB1S1, cRaf,
	NOS3, IKK, GSK3β ( − ), PFK2, PKC, TOR,
	TEC (TEC, BTK, ITK, RLK)
TOR	Phosphorylation of eIF4eBP1 ( − ), S6K ( + )

Heterodimeric PI3K contains a subunit that has enzymatic activity. The superclass of phosphatidylinositol 3-kinases includes 8 PI3K isoforms divided into 5 categories (1A–1B and 2–4) and regrouped into 3 classes (C1–C3). PI3K-related kinases are actually grouped in a class 4 that includes target of rapamycin (TOR), ataxia telangiectasia mutated (ATMK), ataxia telangiectasia mutated-related (ATRK), and DNA-dependent (DNAPK) protein kinase.

Signaling via PI3K is implicated in metabolic control, actin remodeling, intracellular vesicular transport, and cell growth, survival, proliferation, and migration, hence, immunity, angiogenesis, and cardiovascular homeostasis. Furthermore, it is one of the most frequently deregulated pathways in cancer.

2.9.1.1 PI3K Classes

Phosphatidylinositol 3-kinases correspond to main nodes in growth factor signaling from receptor Tyr kinases to effectors. They activate distinct signaling pathways that govern various cellular processes, such as glucose homeostasis and protein synthesis, in addition to cell fate.

The superclass of 8 PI3K isoforms is subdivided into 3 functional classes based on protein domain structure, lipid substrate specificity, and associated regulatory subunits (Table 2.34): class-1 PI3K enzymes with catalytic subunits PI3K_c1α to PI3K_c1δ; class-2 PI3K enzymes (PI3K_c2α–PI3K_c2γ); and class-3 PI3K enzyme that corresponds to the single, ubiquitous, Ca ${}^{++}$ –calmodulin-activated vacuolar protein sorting VPS34 that connects to regulatory VPS15 (or PI3K_r4) subunit. It can be better named type-3 PI3K_c3 catalytic subunit.

Table 2.34

Classes and types of phosphoinositide 3-kinases and their genes. The 4 class-1 PI3K isoforms (PI3K_c1α–PI3K_c1γ) are synthesized in all mammalian cell types at various concentrations. Class-2 PI3Ks do not have regulatory subunits.

Gene	Subunit type	Name	Other alias
Class-1A PI3K subunits
PIK3CA	Catalytic	PI3K_c1α	P110α, PI3Kα
PIK3CB	Catalytic	PI3K_c1β,	P110β, PI3Kβ
PIK3CD	Catalytic	PI3K_c1δ	P110δ, PI3Kδ
PIK3R1	Regulatory	PI3K_r1	P85α, P55α, P50α
PIK3R2	Regulatory	PI3K_r2	P85β
PIK3R3	Regulatory	PI3K_r3	P55γ
Class-1B PI3K subunits
PIK3CG	Catalytic	PI3K_c1γ	P110γ, PI3Kγ
PIK3R5	Regulatory	PI3K_r5	P101
PIK3R6	Regulatory	PI3K_r6	P87 (P84)
Class-2 PI3K subunits
PIK3C2A	Catalytic	PI3K_c2α
PIK3C2B	Catalytic	PI3K_c2β
PIK3C2G	Catalytic	PI3K_c2γ
Class-3 PI3K subunits
PIK3C3	Catalytic	PI3K_c3	VPS34
PIK3R4	Regulatory	PI3K_r4	VPS15 (P150)
Class-4 PI3K-related kinases
Target of rapamycin kinase (TOR
Ataxia telangiectasia mutated kinase (ATMK)
Ataxia telangiectasia mutated-related kinase (ATRK)
DNA-Dependent protein kinase (DNAPK)

Some PI3Ks are ubiquitous or widespread (PI3K_c1α, PI3K_c1β, PI3K_c2α, PI3K_c2β, and PI3K_c3), whereas others have a restricted distribution (PI3K_c1γ, PI3K_c1δ, and PI3K_c2γ; Table 2.35).

Table 2.35

Distribution and effects of PI3Ks (Source: [92]). Both PI3K_c1δ (class 1A) and PI3K_c1γ (class 1B) are synthesized predominantly (but not exclusively) in leukocytes.

Type	Distribution	Effects
PI3K_c1α	Ubiquitous	Development, myocardial contractility
PI3K_c1β	Ubiquitous	Development, insulin signaling, cell motility,
		phagocytosis, blood coagulation
PI3K_c1γ	Leukocytes	T-cell development and migration,
		neutrophil, macrophage, dendritic cell migration,
		mastocyte degranulation, neutrophil burst,
		insulin secretion, myocardial contractility
PI3K_c1δ	Leukocytes	Immunity, cytokine receptor signaling,
		B-cell receptor-mediated antigen presentation
PI3K_c2α	Widespread	Insulin signaling, vesicular transport,
		cell survival,
		vascular smooth muscle cell contraction
PI3K_c2β	Widespread	Cell migration, liver growth,
		vesicular transport
PI3K_c2γ	Hepatocyte	Liver regeneration
PI3K_c3	Ubiquitous	Toll-like receptor signaling,
		receptor-independent membrane trafficking

Three genes encode regulatory subunits PI3K_r1 to PI3K_r3 [93]. The PIK3R1 gene encodes via distinct promoter usage not only ^P85αPI3K_r1, but also ^P55αPI3K_r1 and ^P50αPI3K_r1 [94].

2.9.1.2 Class-1 PI3K Enzymes

Class-1 PI3Ks are involved in cell size. All class-1 PI3Ks are able to phosphorylate PI to PI(3)P, PI(4)P to PI(3,4)P₂, and PI(4,5)P₂ to PI(3,4,5)P₃, but PI(4,5)P₂ is the preferred lipid substrate in vivo (Table 2.36).

Table 2.36

Products of members of the PI3K superclass (Source: [92]).

Type	Products (in vitro)
PI3K_c1α	PI(3)P, PI(3,4)P₂, PI(3,4,5)P₃
PI3K_c1β	PI(3)P, PI(3,4)P₂, PI(3,4,5)P₃
PI3K_c1γ	PI(3)P, PI(3,4)P₂, PI(3,4,5)P₃
PI3K_c1δ	PI(3)P, PI(3,4)P₂, PI(3,4,5)P₃
PI3K_c2α	PI(3)P, PI(3,4)P₂
PI3K_c2β	PI(3)P, PI(3,4)P₂
PI3K_c2γ	PI(3)P, PI(3,4)P₂
PI3K_c3	PI(3)P

In resting cells, class-1 PI3Ks reside mainly in the cytosol. Upon stimulation, they are recruited to cellular membranes via interactions with receptors or adaptors. Once receptors bind to cognate ligands, class-1 PI3Ks mainly generate PI(3,4,5)P₃ that recruits and activates various effectors, such as protein kinase-B, 3-phosphoinositide-dependent kinase PDK1, monomeric GTPases. Therefore, numerous targets are phosphorylated, such as glycogen synthase kinase GSK3β, target of rapamycin, S6 ribosomal kinase (^P70RSK or S6K), nitric oxide synthase NOS3, and several anti-apoptotic effectors.

Three class-1A catalytic subunits exist with isoforms: PI3K_c1α, PI3K_c1β, and PI3K_c1δ, which are encoded by the PIK3CA, PIK3CB, and PIK3CD genes, respectively.⁸⁶

Phosphoinositide 3-kinases of the class-1A are composed of: (1) a regulatory subunit among 5 isoforms (^P50αPI3K_r1, ^P55αPI3K_r1, ^P85αPI3K_r1, PI3K_r2, and PI3K_r3) that potentially generate to 15 distinct subunit combinations) that contains SH2 and SH3 domains, which interact with other signaling proteins; and (2) a catalytic subunit among 3 isoforms (PI3K_c1α, PI3K_c1β, and PI3K_c1δ) that phosphorylates inositol.

Isoform PI3K_c1α, but not PI3K_c1β, acts via insulin receptor substrate. Catalytic subunits PI3K_c1α and PI3K_c1β are ubiquitous, whereas PI3K_c1γ is synthesized primarily in hematopoietic cells, myocytes, and pancreatic cells. Both PI3K_c1δ and PI3K_c1γ are highly enriched in leukocytes.

The regulatory subunit yields at least 3 functions to PI3Ks: (1) stabilization; (2) inactivation of kinase activity in the basal state, and (3) recruitment to Tyr^Presidues of receptors and adaptors that relieve inhibition mediated by the regulatory subunit. Class-1A PI3K enzymes, in fact, consist of an inhibitory (among 5 distinct P85 isoforms) and a catalytic subunit. Binding of the regulatory subunit to activated receptor Tyr kinases or adaptor proteins, such as insulin receptor substrates IRS1 and IRS2, relieves the basal repression of the catalytic subunit.

Class-1A PI3Ks are stimulated by activated receptor Tyr kinases and cytokine receptors. Class-1A PI3Ks are involved in cell growth, division, and survival in response to activated growth factor receptors.⁸⁷

A small fraction (10%) of total class-1 PI3K activity suffices to maintain cell survival and proliferation at least in hematopoietic progenitors and mouse embryonic fibroblasts [96]. In the case of persistent selective or combined inhibition of PI3K isoforms, the remaining isoforms, even at very low concentrations, ensure signaling downstream of Tyr kinases and small GTPase Ras due to functional redundancy of class-1A PI3K isoforms.

Whereas class-1A enzymes are preferentially activated by protein Tyr kinases, the class-1B enzyme is linked to G-protein-coupled receptors. However, most class-1 PI3K subunits may be activated by GPCRs, either directly via Gβγ subunits, such as PI3K_c1β and PI3K_c1γ, or indirectly via Ras GTPase, in particular.⁸⁸ Enzyme PI3Kβ can actually effectively couple to G-protein-coupled receptors.

The catalytic PI3K_c1γ subunit, which is encoded by the PIK3CG gene is the sole class-1B catalytic member. Kinase PI3Kγ consists of a catalytic and a regulatory subunit: (1) PI3K regulatory subunit-5 (PI3K_r5)⁸⁹ or (2) PI3K regulatory subunit-6 (PI3K_r6).⁹⁰ Therefore, 2 PI3Kγ heterodimers exist: PI3K_c1γ–PI3K_r5 and PI3K_c1γ–PI3K_r6. Kinase PI3Kγ elicits cardiovascular and immunological responses.⁹¹

The regulatory subunits PI3K_r5 and PI3K_r6 have a distinct tissue distribution. They relay signals from Gβγ dimer and Ras GTPase to PI3K_c1γ (Table 2.37). Lipid PI(3,4,5)P₃ generated by PI3K_c1γ–PI3K_r5, unlike that generated by PI3K_c1γ–PI3K_r6, is rapidly endocytosed with vesicles [94].

Table 2.37

Activators of class-1 PI3Ks (Sources: [92, 94]; Gq: type-q Gα subunit of heterotrimer Gαβγ coupled to GPCR [G-protein-coupled receptor]; RTK: receptor Tyr kinase; − : inhibition). Kinase PI3K is activated by different Toll-like receptors (TLR) in eosinophils, macrophages, neutrophils, and dendritic cells. Ligands bind to immune, cytokine, and G-protein-coupled receptors to prime signal transduction via adaptors, subunits of heterotrimeric G proteins, monomeric Ras GTPases, and protein Tyr kinases that recruit PI3Ks to the plasma membrane and activate it. Once activated, PI3Ks phospharylate their lipidic substrates, hence transforming them into potent signaling mediators.

Catalyzer	Regulators
PI3K_c1α	RTK, Ras; Gq ( − )
PI3K_c1β	RTK, Gβγ, Rab5 (endosomes)
PI3K_c1δ	RTK, TLR, rRas2
PI3K_c1γ	Gβγ, TLR, Ras

Class-1B PI3Ks link G-protein-coupled receptors to signaling phospholipids for cell growth, division, survival, and migration (Table 2.38). Activation of PI3K_c1γ by stimulated GPCRs is amplified by simultaneous binding of Gβγ subunit of G protein to PI3K_r5 subunit. Interaction of PI3K_r6 with Gβγ subunit differs from that of PI3K_r5 subunit. Subunit PI3K_r5 strongly interacts with Gβγ, whereas PI3K_r6 only weakly interacts, hence failing to sensitize PI3K_c1γ for its activation by Gβγ and limiting its membrane recruitment. On the other hand, Ras^GTP connects to PI3K_c1γ, thereby causing translocation of the PI3K_c1γ–PI3K_r6 dimer and activation [97]. Small GTPase Ras, in cooperation with Gβγ, hence yields PI3Kγ signaling specificity, because Ras: (1) can recruit PI3K_c1γ (as well as PI3K_c1γ–PI3K_r6 and PI3K_c1γ–PI3K_r5) to membranes and (2) can act as a costimulator of PI3K_c1γ–PI3K_r6, whereas Gβγ suffices to activate PI3K_c1γ–PI3K_r5.

Table 2.38

Effectors of PI3Ks, signal transduction pathways, and resulting effects (Source: [92]; BCLxL: B-cell lymphoma-extra large protein; BTK: Bruton Tyr kinase; GAP: GTPase-activating protein; GEF: guanine nucleotide-exchange factor; ITK: interleukin-2-inducible T-cell kinase; NFκB: nuclear factor κ light chain-enhancer of activated B cells; PDK: phosphoinositide-dependent kinase; PKB: protein kinase-B; PKC: protein kinase-C; PLC: phospholipase-C; PREx: PIP₃-dependent Rac exchanger; PTK: protein Tyr kinase; SwAP70: switch-associated protein-70 (SWAP switching B-cell complex 70-kDa subunit); S6K: P70 ribosomal S6 kinase).

PI^P effector	Mediators	Effects
Centaurins	CDC42	Cytoskeleton organization
(ArfGAPs)	Rac
PREx, SWAP70	CDC42	Cytoskeleton organization
(RacGEFs)	Rac
TEC family (PTK)	CDC42, Rac	Cytoskeleton organization
(BTK, ITK)	Ca ${}^{++}$	Degranulation
PDK	PLCγ–Ca ${}^{++}$	Degranulation
	PLCγ–PKC	Microfilament and -tubule assembly
		Cell adhesion
	PKB–S6K	Cell proliferation
	PKB–BCLxL	Cell survival
	PKB–NFκB	Inflammation

Class-1 phosphoinositide 3-kinases target not only lipids, but also proteins. They indeed allow autophosphorylation of both catalytic and regulatory subunits as well as activation of mitogen-activated protein kinase (Chap. 6).

Small GTPase Ras can activate PI3Kα and PI3Kγ. Moreover, rRas2 may contribute to the activation of PI3K_c1δ [94]. In addition to Ras, other small GTPases can be recruited to PI3K. Protein PI3Kβ can be activated by active Rab5 in clathrin-coated vesicles. Class-1 PI3Ks can interact with small CDC42, Rac, and Rho GTPases. The PI3K kinases can act both downstream from monomeric GTPases as well as upstream from them via guanine nucleotide-exchange factors and GTPase-activating proteins. These links create possible feedback loops [94].

Phosphoinositides PI(3,4,5)P₃ and PI(3,4)P₂ coordinate the recruitment and function of many effectors, such as protein kinases (e.g., PKB and BTK), adaptors (e.g., GAB2), and regulators of small GTPases (GAPs and GEFs). The PI3K kinases can regulate ARF, Rac, and Ras GTPases, as they can control their guanine nucleotide-exchange factors and GTPase-activating proteins. All PI3K isoforms promote Rac activation. On the other hand, RhoA is inhibited by PI3Kδ, but stimulated by PI3Kα subtype.

Membrane ruffling caused by actin remodeling upon growth factor stimulation (e.g., epidermal and platelet-derived growth factors) is mediated by various signaling molecules, such as phosphatidylinositol (3,4,5)-trisphosphate, ^Factin- and PI(3,4,5)P₃-binding guanine nucleotide-exchange factor SWAP70 (70-kDa switch-associated protein), and Rac1 GTPase [98, 99]. Activator SWAP70 is a RacGEF that cooperates with activated Rac1 GTPase.

Members of the DBL epifamily of Rho guanine nucleotide-exchange factors can serve as direct effectors of heterotrimeric G proteins. Subunits G12/13, Gq, and Gβγ directly bind and regulate RhoGEF regulators. Guanine nucleotide-exchange factors of the PIP₃-dependent Rac exchanger (PREX) family activate Rac GTPase. Protein PREx1 is widely expressed in the central nervous system, whereas PREx2 is specifically produced in Purkinje neurons of the cerebellum [100].

Class-1 PI3Ks can serve as scaffolds. Subunit PI3K_c1 binds to phosphodiesterase PDE3b and protein kinase-C [94]. Enzymes PI3Kα, PI3Kβ, and PI3Kγ, as well as PTen phosphatase, are synthesized in cells of the cardiovascular system (cardiomyocytes, fibroblasts, and vascular endothelial and smooth muscle cells), in which they modulate metabolism, survival, hypertrophy, contractility, and mechanotransduction [101].

Active calpain small subunit-1 (CapnS1) of calpain heterodimers interacts with class-1A PI3Ks in stimulated cells [109]. Calpain-1 and -2 cleave PI3K_c1α isoform, thereby terminating PI3K signaling.

2.9.1.3 Class-2 PI3K Enzymes

Class-2 phosphoinositide 3-kinase is made of a catalytic subunit among 3 isoforms (PI3K_c2α–PI3K_c2γ) without regulatory subunit. Class-2 PI3Ks are encoded by the PIK3C2A, PIK3C2B, and PIK3C2G genes. They are mainly associated with cellular membranes and nucleus. Isozymes PI3K_c2α and PI3K_c2β have a broad, but not ubiquitous, tissue distribution, whereas the expression pattern of PI3K_c2γ is more restricted [94]. Class-2 PI3Ks do not have regulatory subunits.

2.9.1.4 Class-3 PI3K Enzyme

Class-3 PI3K can phosphorylate only PI substrate. It complexes to a single regulatory subunit encoded by the PIK3R4 gene. Class-3 PI3K is actually represented by a single known heterodimer with a catalytic subunit (PI3K_c3) encoded by the PIK3C3 gene and a regulatory (PI3K_r4) subunit.

Class-3 PI3K heterodimer assembly at membranes is promoted by small Rab5 and Rab7 GTPases that connect to PI3K_r4 protomer. It is antagonized by phosphoinositide 3-phosphatase myotubularin-6 [94].

2.9.1.5 PI3Kα Enzyme (Class 1A)

Subunit PI3K_c1α with Tyr kinase activity that targets Ras signaling has specific functions in growth factor, insulin, and leptin signaling [102]. This function relies on the selective recruitment of adaptors, insulin receptor substrates, transmitting signals from receptors of insulin and insulin-like growth factor-1 mainly to IRS-bound PI3K to activate the PKB and ERK pathways.⁹² Isoform PI3K_c1α is the primary insulin-responsive PI3K, at least in cultured cells, whereas PI3K_c1β is dispensable, but sets a threshold for PI3K_c1α activity [103].

2.9.1.6 PI3Kβ Enzyme (Class 1A)

Ubiquitous class-1A PI3Kβ isoform has a PI3K_c1β subunit with both kinase-dependent and -independent functions. Synthesis of PI3K_c1β in the liver is required for insulin sensitivity and glucose homeostasis, but does not signicantly affect protein kinase-B phosphorylation in response to receptor activation by insulin and epidermal growth factor [104].

Subunit PI3K_c1β also regulates cell proliferation independently of its kinase activity. However, its kinase activity is required for the signaling triggered by lysophosphatidic acid attached to its G-protein-coupled receptor. In addition, PI3K_c1β, but not PI3K_c1α, intervenes in tumorigenesis with a concomitant increase in PKB phosphorylation.

Lipid PTen phosphatase opposes kinase activity of class-1A PI3Ks. The basal pool of PI(3,4,5)P₃ catalyzed by PI3K_c1β that is not disturbed by stimulation by insulin and growth factors supports oncogenic transformation in the absence of PTen, at least in some tumor models.

Among its kinase-independent functions, PI3Kβ participates in sensing of double-strand breaks. Enzyme PI3Kβ regulates binding of nibrin, a sensor of damaged DNA and a member of the double-strand DNA break-repair complex [105].

2.9.1.7 PI3Kγ (Class 1B)

Phosphoinositide 3-kinase-γ is a heterodimer composed of PI3K_r5 or PI3K_r6 regulatory and PI3K_c1γ catalytic subunit. It converts PI(4,5)P₂ into PI(3,4,5)P₃ at the inner leaflet of the plasma membrane. The lipid kinase activity is stimulated by the combined interaction of regulatory and catalytic subunit with activated Gβγ dimer and Ras GTPase.

Kinase PI3Kγ is activated by Gi-coupled receptors stimulated by chemokines (e.g., chemokines CCL2,⁹³ CCL3,⁹⁴ CCL5,⁹⁵ CXCL2,⁹⁶ and CXCL8),⁹⁷ pro-inflammatory lipids (PAF and LTb4), bacterial products, and vasoactive molecules (C5a, ADP, and angiotensin-2). Activated PI3Kγ acts cooperatively with other signaling pathways. These signaling pathways involve Src kinases, class-1A PI3Ks, MAPKs, etc.

Agent PIP₃ and its dephosphorylation product PI(3,4)P₂ are signaling lipids that support the plasmalemmal localization and regulation of several effectors (Table 2.39). The transduction magnitude and duration via these effectors is regulated by inositol polyphosphate 5-phosphatases such as SHIP, 3-phosphatases such as PTen, and inositol polyphosphate 4-phosphatases. The PI3Kγ pathway regulates the function of: (1) immunocytes (macrophages, monocytes, mastocytes, neutrophils, dendritic cells, and T lymphocytes); (2) vascular endothelial cells and smooth muscle cells that determine the vasomotor tone; and (3) platelets involved in blood coagulation.

Table 2.39

Phosphoinositide 3-kinase-γ effectors downstream from PI(3,4,5)P₃ and its dephosphorylation product PI(3,4)P₂ (GEF: guanine nucleotide-exchange factors; GAP: GTPase-activating protein; Source: [106]).

Direct	Indirect
Ser/Thr kinases	Protein kinases
Tyr kinases	Lipid messengers
RacGEFs	Transcription factors
ArfGEFs	Integrins
ArfGAPs	Small GTPases
Adaptors	NADPH oxidase
	cAMP

2.9.1.8 PI3Kδ (Class 1A)

Isozymes PI3Kγ and PI3Kδ that are relatively specific to leukocytes orchestrate innate and adaptive immunity, especially in respiratory diseases (allergic asthma and chronic obstructive pulmonary diseases [COPD]) and cardiovascular pathologies ( atherosclerosis and myocardial infarction).

Phosphoinositide 3-kinases contribute to the genesis of asthma, as they regulate inflammatory mediator activation, inflammatory cell recruitment, and immunocyte function. In particular, they control the expression of interleukin-17 that is an important cytokine in airway inflammation. Signaling mediated by PI3Kδ promotes IL17 expression via nuclear factor-κB [107].

Glucocorticoid function is markedly impaired in lungs of patients with chronic obstructive pulmonary disease. Oxidative stress causes phosphorylation of PKB in monocytes and macrophages that is associated with a selective upregulation of PI3Kδ [108].

2.9.1.9 Class-2 Phosphatidylinositol 3-Kinases

Class-2 PI3Ks are stimulated by extracellular signals, such as integrin engagement, growth factors, and chemokines [93]. They can phosphorylate PI and PI(4)P, but not PI(4,5)P₂. Class-2 PI3Ks can also be stimulated by protein kinases and GPCRs, after recruitment from the small cytosolic pool to the plasma membrane via membrane-associated adaptor proteins [94]. Class-2 PI3Ks may also be recruited and activated by membrane-bound GTPases such as monomeric RhoJ GTPase.

2.9.1.10 Class-3 Phosphatidylinositol 3-Kinase

Class-3 phosphatidylinositol 3-kinase (PI3KC3) produces phosphatidylinositol 3-phosphate, thereby intervening in membrane trafficking, endocytosis, phagocytosis, autophagy, nutrient sensing by the target of rapamycin pathway, and cell signaling, especially downstream from G-protein-coupled receptors. It acts in the signaling primed by activated Gq-coupled receptors such as muscarinic M₁ receptor [94]. It can control amino acid-dependent activation of S6K1 kinase.

It is implicated in endosome fusion during intracellular transport. Kinase PI3KC3 participates in transport to lysosomes via multivesicular bodies, from endosomes to the trans-Golgi network via retromers (proteic complexes that contribute to the recycling of plasmalemmal receptors), and phagosome and autophagosome maturation. In addition, PI3KC3 is involved in nutrient sensing in the target of rapamycin pathway.

Enzyme PI3KC3 alternates between a closed cytosolic form and an open form on membranes. It is, in fact, located mainly on intracellular membranes [93, 110]. It possesses a constricted adenine-binding pocket. Both the phosphoinositide-binding loop and C-terminus of PI3KC3 mediate catalysis on membranes [110].

Kinase PI3KC3 synthesizes PI(3)P that is specifically recognized by proteins that contain FYVE or PX binding domains and initiate the assembly of complexes at membranes of endosomes, phagosomes, or autophagosomes. Regulators, such as Rab5 and Rab7, bind to PI3K_r4 and enable activation of the PI3K_r4–PI3K_c3 complex at membranes [110].

Proteins associated with PI3K_c3 regulate, directly or not, PI3K_c3 activity (Table 2.40). Agent BAX-interacting factor-1 (BIF1) stimulates PI3K_c3 in the beclin-1–UVRAG complex. Subunit PI3K_c3 actually forms different functional complexes [94]. A complex devoted to autophagy contains beclin-1 (coiled-coil, moesin-like, B-cell lymphoma protein-2 [BCL2]-interacting protein), a protein essential for autophagy, either subunit of autophagy-specific PI3K complex, autophagy-related protein-like agent Atg14L (or beclin-1-associated autophagy-related key regulator [barkor]) or ultraviolet radiation resistance-associated gene product (UVRAG), BAX-interacting factor BIF1, activating molecule in beclin-1-regulated autophagy protein AMBRA1, and RUN domain and Cys-rich domain-containing, beclin-1-interacting protein Rubicon. The composition of this complex can vary, as the presence of some components excludes the presence of others. Components Atg14L and UVRAG bind to the beclin-1–PI3K_c3 complex in a mutually exclusive manner, whereas Rubicon binds only to a subpopulation of UVRAG complexes [111]. Effectors include WD repeat domain-containing phosphoinositide-interacting protein WIPI1, autophagy-linked FYVE protein (ALFY), and double FYVE-containing protein-1 (DFCP1).

Table 2.40

The PI3KC3 complexes and their subcellular localization (Source: [94]). Enzyme PI3KC3 participates in various vesicular transfer types that always deliver cargos to lysosomes. Cup-shaped omegasome is a double-membrane structure that forms upon amino acid starvation, expands, sequesters cytoplasmic material, and eventually forms an autophagosome (AMBRA: activating molecule in beclin-1-regulated autophagy protein; Atg14L: autophagy-specific PI3K complex Atg14-like protein; beclin: BCL2-interacting protein; BIF: BAX-interacting factor; EEA: early endosomal antigen; Rubicon: RUN domain and Cys-rich domain-containing, beclin-1-interacting protein; UVRAG: ultraviolet radiation resistance-associated gene product).

Role	Complex components	Locus
Autophagy	Beclin-1	Endoplasmic reticulum
	UVRAG (or Atg14L)	(omegasome)
	BIF1, AMBRA1, Rubicon
Endocytosis	Rab5, EEA1	Cell cortex (early endosome)
Phagocytosis	Dynamin, Rab5	Cell cortex (early phagosome)

The PI3K_r4–PI3K_c3 molecule linked to endosomes, which acts as a regulator of vesicular transfer, can complex with MTM1 myotubularin, a phosphatidylinositol 3-phosphatase, and (or) UVRAG [94]. The latter increases PI3K_c3–VPS15 binding [112]. Effectors comprise early endosome antigen EEA1, endosomal sorting complexes required for transport ESCRT2 and ESCRT3, hepatocyte growth factor-regulated Tyr kinase substrate (HRS), sorting nexins, and Rab5 effector rabenosyn-5.⁹⁸ The degradative pathway comprises early Rab5 + endosomes, multivesicular bodies, and late Rab7 + endosomes before fusion with lysosomes. In early endosomes, Rab5 and PI(3)P bind and assemble effectors, such as EEA1, HRS, and ESCRT proteins. In late endosomes, Rab5 is replaced by Rab7 that recruits its effector Rab7-interacting lysosomal protein (RILP) for acquisition of the lysosomal marker lysosome-associated membrane protein LAMP1 and fusion with lysosomes.

The PI3K_r4–PI3K_c3 complex intervenes in phagocytosis. Several waftures of PI(3)P accumulation occur during phagocytosis with an early transient peak immediately after particle internalization and a following wave during phagosome maturation [94]. Production of PI3P depends on class-1A PI3Ks; phagosome maturation on PI3KC3 kinase. In nascent phagosomes, PI3KC3 interacts with large GTPase dynamin that recruits small GTPase Rab5 during phagosome maturation. Synthesized PI3P is also required for activation of NADPH oxidase assembled on phagosomes.

2.9.1.11 Nuclear PI3K

Although PI3K and its effector protein kinase-B are predominantly located in the cytoplasm, they also operate in the nucleus. Nuclear PI3K and its effectors intervene in various activities, such as cell survival, differentiation, and proliferation, as well as pre-mRNA export from the nucleus and splicing. Splicing of pre-mRNA and export of mRNA are coupled.

Most enzymes of the phosphoinositol lipid metabolism (phospholipase-C, phosphatidylinositol phosphate kinases, phosphoinositide-dependent kinase-1, PTen, and protein kinase-B) reside in the nucleus. Phosphoinositides associate with nuclear speckles. These nuclear subcompartments contain small ribonucleoproteins, splicing factors, and elements of nuclear phosphoinositide metabolism.

Nuclear protein THO complex subunit THOC4⁹⁹ is required for cell cycle progression owing to phosphorylation and phosphoinositide binding. The mRNA export factor and transcriptional regulator THOC4 is a protein of nuclear speckles that interacts with nuclear PI(4,5)P₂ and PI(3,4,5)P₃ [113]. During the cell division cycle, PI3K is involved in both G1–S transition and S-phase progression. Molecular chaperone THOC4 regulates PI3K subnuclear residency. Nuclear PKB phosphorylates THOC4, and hence controls THOC4 binding to phosphoinositides.

2.9.1.12 PI3K Signaling

Phosphatidylinositol 3-kinase is phosphorylated (activated) by various receptor and receptor-associated protein Tyr kinases. Enzyme PI3K is stimulated by receptors of hormones (e.g., insulin) and growth factors (EGF, HGF, IGF1, NGF, PDGF, and SCF; Vols. 2 – Chap. 3. Growth Factors and 3 – Chap. 8. Receptor Kinases). Kinase PI3K phosphorylates phosphatidylinositols at position 3 of the inositol ring. In response to various stimuli, phosphatidylinositol 3-kinase phosphorylates phosphatidylinositol (4,5)-bisphosphate to generate phosphatidylinositol (3,4,5)-trisphosphate that binds to pleckstrin homology (PH) domains of signaling proteins as well as phox homology (PX) domains of certain proteins.

Phosphatidylinositol 3-kinase generates substrates for phospholipase-Cγ that produces second messengers diacylglycerol and inositol trisphosphate. Effectors also include Ser/Thr kinases, protein kinase-B, and phosphatidylinositol-dependent protein kinase-1 (PDK1), various isoforms of protein kinase-C (PKC), and members of the TEC family of Tyr kinases, such as TEC (Tyr kinase expressed in hepatocellular carcinoma), BTK (Bruton Tyr kinase), ITK (interleukin-2-inducible T-cell kinase), and RLK (resting lymphocyte kinase, or TXK) [114].

The PI3K–PKB pathway is involved in cell growth, survival, metabolism, and migration, as well as tumorigenesis. Kinase PKB is recruited to the plasma membrane, as it binds to PI3K products PI(3,4)P₂ and PI(3,4,5)P₃ owing to its PH domain. After PKB translocation to the plasma membrane, the formation of a complex with its upstream kinases at the plasma membrane is facilitated by scaffold proteins. Activation of PKB can require a sequential phosphorylation by 2 phosphoinositide-dependent kinases, PDK1 and PDK2 (or TOR complex-2). For full activation, PKB indeed needs to be phosphorylated by PDK1 and PDK2 (Thr308 and Ser473, respectively).

Phosphatidylinositol 3-kinase stimulated by growth factors and chemoattractants intervenes in cell migration, as it activates Rac and CDC42 (Sect. 9.3) at the leading edge of migrating cells (Vol. 2 – Chap. 6. Cell Motility).¹⁰⁰ Small Rac and CDC42 GTPases act via their effectors of group-1 P21-activated protein kinases. Kinase PI3K activates a signaling pathway that includes Rac1, PDK1, and PKB effectors. Small Rac1 GTPase not only provokes phosphorylation by PAK1s, but also causes an additional kinase-independent PAK1 activity: PAK1 operates as a scaffold for PKB recruitment to the plasma membrane and its activation by PDK1 [115].

In T-lymphocytes, the PI3K pathway is implicated in growth, proliferation, cytokine secretion, and survival. T lymphocytes produce all 3 class-1A PI3K isoforms, which are regulated by receptor Tyr kinases, and class-1B PI3K isoform, which is activated by G-protein-coupled receptors. Chemokines binds to cognate GPCRs. T-cell receptor, costimulatory receptors of the CD28 family, and cytokine receptors activate class-1A PI3Ks.

Among class-1 PI3K heterodimeric lipid kinases ( with regulatory and catalytic subunits), class-1A PI3Ks with a P85 regulatory subunit (^P85αPI3K_r1 or ^P85βPI3K_r2 adaptor) is activated by receptor Tyr kinases, whereas class-1B enzyme (PI3Kγ) is stimulated by Gβγ subunits of heterotrimeric G proteins.

In addition to serve as a regulatory subunit for PI3K_c1 protomer of PI3Ks, P85_PI3K subunit also interacts with other proteins, such as the small CDC42 GTPase, nuclear receptor corepressor, receptor Tyr protein phosphatase PTPRj,¹⁰¹ as well as the short splice variant of the transcription factor X-box-binding protein XBP1_S of the ATF–CRE (activating transcription factor–cAMP response element)-binding protein family [116].¹⁰²

2.9.1.13 PI3K Signaling in the Cardiovascular System

Signaling via PI3K plays a prominent role in the cardiovascular system. It regulates the activity of vascular endothelial and smooth muscle cells as well as platelets and leukocytes. Its effectors include mainly protein kinase-B (Sect. 5.2.6), nitric oxide synthase NOS3 (Sect. 10.3.1), target of rapamycin (Vol. 2 – Chap. 2. Cell Growth and Proliferation and Sect. 5.5.3), and forkhead transcription factor FoxO (Sect. 10.9.3.3). It thus controls various processes, such as vascular tone, angiogenesis, endothelial cell differentiation, and endothelial cell–leukocyte interactions. Protein kinase-B operates in the nucleus to increase the gene transcription.

2.9.1.14 Cardiac Effects of PI3K

Cardiac effects of PI3K are summarized in Table 2.41. Anti-apoptotic action of growth factors relies almost exclusively on the PI3K–PKB pathway. Cardioprotection use also this pathway downstream from GPCRs of adrenomedullin, ghrelin, and urocortin, as well as β2-adrenoceptors.

Table 2.41

Cardiac effects of PI3K (Source: [101]).

Effect	Mediators
Cell survival	PKB, BAD, caspase-3, FoxO, IκB kinase
Glucose uptake	GluT
CMC hypertrophy	PKB1, S6K
CMC contractility	PDE4
Mechanotransduction	PKB, GSK3β
Angiogenesis	PKB

Cardiomyocyte adaptive and maladaptive hypertrophy requires PI3K. Kinase PI3Kα controls adaptive growth, whereas PI3Kγ is involved in maladaptive hypertrophy.

The cAMP–PKA pathway has positive chronotropic, inotropic, and lusitropic effects. Kinase PI3Kγ targets phosphodiesterase that restricts cAMP activity, whereas PI3Kα favors cardiomyocyte contractility.

Kinase PI3K regulates several ion channels and exchangers, such asvoltage-gated Ca_V1 andinward (K_IR) anddelayed rectifier (K_V) K⁺ channels. Kinase PKB2, but not PKB1, is implicated in insulin-stimulated glucose uptake and metabolism. In addition, stretch activates PKB and GSK3β. Lastly, PI3Kγ contributes to the recruitment of endothelial progenitor cells.

2.9.1.15 Vascular Effects of PI3K

Protein and lipid PI3K kinases operate also in vascular compartments, upon receptor Tyr kinase, G-protein-coupled receptor, or Ras activation. Ubiquitous PI3Kα and PI3Kβ abound in the vasculature, whereas PI3Kδ and PI3Kγ are mainly restricted to leukocytes.

Endothelial Cells

The PI3K–PKB pathway intervenes in several endothelial functions, such as vascular tone regulation, angiogenesis, and extravasation (Table 2.42). Kinase PI3K indeed promotes nitric oxide release, endothelial progenitor cell recruitment, and cell migration [117]. Endothelial PI3Kγ hinders and PI3Kδ favors neutrophil adhesion to endothelium. Subtype PI3Kα operates in VEGFa-dependent migration of endothelial cells. Sphingosine 1-phosphate-dependent endothelial migration requires both PI3Kβ and PI3Kγ.

Table 2.42

Effects of PI3K on vascular endothelium (Source: [117]; EPC: endothelial progenitor cell).

Process	Targets
Angiogenesis	RhoA (cell migration)
Endothelial permeability	hRas
NO release	PKB (NOS3 phosphorylation)
Leukocyte extravasation	Adhesion molecules
Cell survival	BAD, FoxO
EPC differentiation	HDAC3

Vascular Smooth Muscle Cells

Kinase PI3Kγ regulates contractility and proliferation of vascular smooth muscle cells, thereby causing vasoconstriction and neointimal hyperplasia [117] (Table 2.43).

Table 2.43

PI3K Effects on vascular smooth muscle cells (Source: [117]).

Process	Effects
Vascular tone	Ca ${}^{++}$ influx
	Ca_Vβ2a subunit phosphorylation
	Ca ${}^{++}$ -permeable, non-selective cation channel NSCC2 gating
	Store-operated Ca ${}^{++}$ channel opening
Remodeling	Activation of target of rapamycin

2.9.1.16 PI3K Effects on Platelets

Different class-1 PI3Ks (PI3Kα–PI3Kγ) are synthesized in thrombocytes (Vol. 5 – Chap. 4. Blood Cells). Phosphoinositide 3-kinases promote thrombosis, as it enhances calcium release and activation of α_2Bβ₃-integrins that promote platelet anchoring to the vessel wall despite blood flow (Vol. 5 – Chap. 10. Endothelium). Isoform PI3Kβ is the main contributor to formation and stability of integrin bonds. Class-2 PI3Ks intervene in signaling from platelet environment. The ADP–P2Y₁₂–PI3Kβ axis, in cooperation with PI3Kγ, enhances long-term Ca ${}^{++}$ mobilization induced by Gs-coupled thrombin receptor. In addition, pro-aggregant IGF1 signals via PI3Kα.

2.9.2 Other Phosphoinositide Kinases

Type-1 phosphatidylinositol phosphate kinases correspond to PI(4)P5Ks that phosphorylate PI(4)P to form PI(4,5)P₂. Type-2 PIPKs are PI(5)P4Ks that phosphorylate PI(5)P. Type-3 PIPKs are PI(3)P5Ks that phosphorylate PI(3)P to PI(3,5)P₂.

Lipid kinases, such as phosphatidylinositol 3-kinase, phosphatidylinositol 4-kinase, and phosphatidylinositol 4-phosphate 5-kinase, are implicated in membrane trafficking. Phosphatidylinositol 4- and 5-kinases convert PI(5)P and PI(4)P to PI(4,5)P₂. The latter is a major substrate for activated phospholipase-C that hydrolyzes PI(4,5)P₂ into inositol (1,4,5)-trisphosphate and diacylglycerol. Lipid PI(4,5)P₂ is also a substrate for several other phosphoinositide kinases, such as PI(4)P5K and PI3K that produce PI(4,5)P₂ and PI(3,4)P₂, respectively. Signaling molecule PI(4,5)P₂ controls cytoskeletal organization as well as activity of enzymes (e.g., phospholipase-D) and ion transporters and channels (e.g., inwardly rectifying potassium channelsK_IR1.1,¹⁰³ K_IR2.1,¹⁰⁴ and K_IR3.2,¹⁰⁵voltage-gated potassium channels K_V7.2¹⁰⁶ and K_V7.3¹⁰⁷, cyclic nucleotide-gated [CNG] ion channel, and transient receptor potential channels) [118]. Channel TRPM4 is implicated in PI(4,5)P₂ sensing.

Signaling lipid PI(3,5)P₂ in endosomes regulates retrograde transport to the trans-Golgi network. The PI3P 5-kinase (PI(3)P5K)¹⁰⁸ produce PI(3,5)P₂. Scaffold Vac14¹⁰⁹ recruits PI(3,5)P₂ regulators: the lipid kinase PI(3)P5K, lipid 5-phosphatase suppressor of actin mutations-like phosphatase SAc3,¹¹⁰ and PI(3)P5K activator Vac7 and inhibitor Atg18 for fast and transient conversion of PI(3)P into PI(3,5)P₂ [119].

2.9.2.1 Phosphatidylinositol 4-Kinase

Ubiquitous phosphatidylinositol 4-kinase (PI4K) catalyzes the phosphorylation of membrane phosphatidylinositol to generate phosphatidylinositol 4-monophosphate (PI4P). The latter operates in various signaling events as well as vesicular trafficking and lipid transport. In particular, it serves as a precursor in the phosphoinositide 3-kinase–phospholipase-C pathway. In addition, PI4P recruits PH domain-containing proteins involved in intracellular transport and clathrin AP1 and AP3 adaptor complexes.

Multiple PI4K isoforms exist in eukaryotic cells (Table 2.44). Types PI4K can be distinguished by their distinct phosphatidylinositol kinase activities: (1) type-1 corresponds to phosphoinositide 3-kinase, whereas (2) types-2 and –3 exclusively use phosphatidylinositol and phosphorylate only at the 4-OH position of the inositol ring to produce phosphatidylinositol 4-monophosphate. Type-2 PI4Ks are membrane-bound kinases owing to palmitoylation. However, a significantly larger fraction of PI4K2β than that of PI4K2α is cytosolic.

Table 2.44

The PI4K isozymes in mammalian cells (Source: [120]; ER: endoplasmic reticulum; MVB: multivesicular body; TGN: trans-Golgi network). They localize to distinct membrane compartments and have specific roles in vesicular trafficking and Golgi body function, where PI(4)P rather than PI(4,5)P₂ is the lipid regulator. The PI4K enzymes are divided into structurally and biochemically distinct PI4K2 and PI4K3 subfamilies. Type-2 kinases are relatively sensitive to Ca ${}^{++}$ and adenosine. Type-3 isoforms share significant amino acid sequence similarity with the catalytic domains of members of the protein- and phosphoinositide 3-kinase superclass.

Type	Location	Ki	Km	Km
		(adenosine; mmol)	(ATP; mmol)	(PI; mmol)
PI4K2α	Plasma membrane	10–70	10–50	20–60
	TGN, endosomes
PI4K2β	Plasma membrane	10–70	10–50	20–60
	TGN, endosomes
PI4K3α	Plasma membrane,	O[1]	∼ 700	∼ 100
	ER, Golgi, MVB,
	mitochondrion,
	nucleolus
	nervous system
PI4K3β	Golgi, nucleus,	O[1]	∼ 400	∼ 100
	plasma membrane

Isoforms PI4K2α and PI4K2β that constitute the type-2 PI4K family are structurally and chemically distinct from type-3 PI4Ks that are members of the PI3K superclass. Type-2 PI4Ks localize to organelle and plasma membranes.

Type-3 phosphatidylinositol 4-kinases favor synthesis of plasmalemmal phosphoinositides during phospholipase-C activation and Ca ${}^{++}$ signaling. Phosphatidylinositol 4-kinase-3α resides at the endoplasmic reticulum and Golgi body, whereas phosphatidylinositol 4-kinase-3β localizes primarily to the Golgi body and in the pericentriolar zone.

2.9.2.2 Phosphatidylinositol 4-Kinase-2α

Epidermal growth factor and Gi/o protein enhances PI4K2α activity. In response to EGF, PI4K2α indeed connects to epidermal growth factor receptor. In addition, it also links to receptor HER2 as well as activated T-cell coreceptor CD4 [121]. It can form a complex with: (1) scaffold tetraspanins of the transmembrane 4 superfamily (TM4SF) — Tspan24 (CD151), Tspan28 (CD81), and Tspan30 (CD63) — that anchor and regulate other receptors;¹¹¹ (2) protein kinase-Cμ and type-1 phosphatidylinositol 4-phosphate 5-kinase, as PKCμ autophosphorylates and serves as a scaffold for recruitment of phosphoinositide kinases to membrane;¹¹² and (3) epidermal growth factor receptor, phosphatidylinositol transfer protein PITPα, and phospholipase-Cγ1.¹¹³ Kinase PI4K2α is required in recruitment of heterotetrameric adaptor AP1 to the Golgi body and AP3 to endosomes [120].

2.9.2.3 Phosphatidylinositol 4-Kinase-2β

Unlike PI4K2α, a significant proportion of PI4K2β is cytosolic and substantially less active than the membrane-associated fraction. Cytosolic PI4K2β is recruited to the plasma membrane in response to platelet-derived growth factor and active small GTPase Rac1 [125]. Calcium inhibits both type-2 PI4Ks.

2.9.2.4 Phosphatidylinositol 4-Kinase-3α

Phosphatidylinositol 4-kinase-3α is mainly located in the endoplasmic reticulum of mammalian cells, but also resides in the plasma membrane as well as pericentriolar area over the Golgi body, mitochondria, multivesicular bodies, and nucleolus. Phosphatidylinositol 4-kinase-3α actually possesses nuclear localization signals. In the nucleolus, it can participate in DNase- and RNase-sensitive complexes [126].

Phosphatidylinositol 4-kinase-3α is responsible for the generation of the plasmalemmal pool of PI(4)P [127]. In particular, PI4K3α is required for the production of PI(4)P, PI(4,5)P₂, and Ca ${}^{++}$ signaling during angiotensin-2 stimulation [128]. Phosphatidylinositol 4-kinase-3α forms a signaling complex with P2X₇ ion channels [120].

Phosphatidylinositol 4-kinase-3α can function with membrane-associated phosphatidylinositol transfer protein PITPm that lodges in the endoplasmic reticulum and Golgi body.¹¹⁴

2.9.2.5 Phosphatidylinositol 4-Kinase-3β

Phosphatidylinositol 4-kinase-3β is primarily located in the Golgi body, but can be detected in the nucleus owing to its nuclear localization and export signals [120]. Recruitment of PI4K3β to the Golgi body is regulated by small GTPase ARF1 that is regulated by calcium-binding protein frequenin (Freq or neuronal calcium sensor NCS1). Active Rab11^GTP also binds PI4K3β. Kinase PI4K3β is phosphorylated (Ser268) by protein kinase-D to ensure its recruitment to Golgi body and activity.

Phosphatidylinositol 4-kinase-3β intervenes in the endoplasmic reticulum-to-Golgi body transport of ceramide [128] as well as Golgi-to-plasma membrane trafficking. Its lipid products, together with small GTPases Rab11 and ARFs, recruit clathrin adaptors and other effectors to promote budding and cleavage of Golgi-derived transport vesicles. Among proteins that possess PH domains, which specifically recognize PI(4)P, there are lipid transport proteins, such as oxysterol-binding proteins, phosphoinositol 4-phosphate adaptor protein FAPP2, and ceramide transport protein. Kinase PI4K3β also regulates exocytosis at the plasma membrane. In addition, both PI4K3α and PI4K3β isoforms operate during hepatitis C virus infection [129].

2.9.2.6 Phosphatidylinositol 5-Kinase

In polarized cells, phosphatidylinositol 5-kinase (PI5K) stimulates delivery from the trans-Golgi network to apical membrane of raft-associated proteins, but not that of non-raft-associated apical or basolateral proteins [130]. This transport is carried out via Arp2/3 complex, i.e., using both actin filaments and microtubules. The Golgi body indeed contains a pool of phosphatidylinositol 4-phosphate that limits delivery to apical membrane of raft-associated proteins. Increased phosphatidylinositol (4,5)-bisphosphate level caused by PI5Ks results in generation of nucleating branches of actin filaments, the so-called actin comets, that transport vesicles in the cytoplasm.

2.9.2.7 Phosphatidylinositol Phosphate 4-Kinase

Phosphatidylinositol 5-phosphate 4-kinases regulate the cellular concentration of phosphatidylinositol 5-phosphate (PI(5)P). Three PIP4K (or PI(5)P4K) isoform exist (PIP4Kα–PIP4Kγ). Isoform PIP4Kα resides predominantly in the cytosol and can be recruited to the plasma membrane. Subtype PIP4Kβ is located in the nucleus. Isozyme PIP4Kγ is the predominant PIP4K in kidney [131].

In the brain, all 3 isoforms are produced with different spatial distribution. In spleen, PIP4Kα is the most abundant isoform. In heart and skeletal muscles, PIP4Kβ is synthesized at high levels. Isoform PIP4Kγ is highly expressed in kidneys, brain, heart, ovary, and testis. It contributes to actin remodeling during endocytosis. In kidneys, PIP4Kγ is primarily distributed in epithelial cells in the thick ascending limb and intercalated cells of the collecting duct, where it regulates vesicular transport [131].

2.9.2.8 Phosphatidylinositol Phosphate 5-Kinase

Phosphatidylinositol 4-phosphate 5-kinase (PI(4)P5K) synthesizes phosphatidylinositol (4,5)-bisphosphate, a precursor in phosphoinositide signaling.¹¹⁵ In addition, it controls actin polymerization and operates in cell adhesion and migration independently of its catalytic activity.

Phosphatidylinositol 4-phosphate 5-kinases constitute a superfamily with several kinase types: type-1 (PIP5K1), -2 (PIP5K2), and -3 (PIP5K3) PI(4)P 5-kinases. In addition, PIP5K1 possesses isoforms, such as PIP5K1α and PIP5K1β [132]. In erythrocytes, both PIP5K1α and PIP5K2 exist. Lipid PIP5K1, but not PIP5K2, can be stimulated by phosphatidic acid and can use PI4P in membranes as a substrate.

Different phosphatidylinositol phosphate 5-kinase generate functionally distinct PIP₂ pools. Isokinase PIP5K1, but not PIP5K2, is implicated in the secretion of neurotransmitters and regulation of monomeric GTPases [132]. Isoform PIP5K1α tethers to Rac, PIP5K1β serves in endocytosis, and PIP5K1γ targets focal adhesions.

Isoform PI(4)P5K1α, but not PI(4)P5K1β, serves as a scaffold protein that localizes Rac1 to the plasma membrane at sites of integrin activation [133]. Small Rac1 GTPase links activated integrins to the regulation of cell migration.

In the brain, the major PI(4,5)P₂-producing enzyme corresponds to phosphatidylinositol 4-phosphate 5-kinase-γ₆₆₁ [134]. Lipid PI(4,5)P₂ recruits the components of clathrin-mediated endocytosis that retrieve synaptic vesicles at nerve terminals. The AP2 complex binds to and activate PIP5Kγ₆₆₁.

2.10 Phosphoinositide Phosphatases

Phosphoinositide phosphatases (Table 2.45) dephosphorylate diverse positions (1, 3, 4, and 5) of the inositol ring (1-, 3-, 4-, and 5-phosphatases). They belong to 2 families of enzymes: protein Tyr phosphatases and inositol 5-phosphatase isoenzymes. There are multiple phosphoinositide lipid phosphatases in humans.

Table 2.45

Phosphoinositide phosphatases (Source: [118]; MTM: myotubularin; MTMR: myotubularin-related phosphatase; OCRL: oculocerebrorenal syndrome of Lowe phosphatase; PTen: phosphatase and tensin homolog deleted on chromosome 10; SHIP: SH-containing inositol phosphatase; SKIP skeletal muscle- and kidney-enriched inositol phosphatase; Synj: synaptojanin).

Type	Function	Substrate(s)
PTen	3-Phosphatase	PI(3,4)P₂, PI(3,4,5)P₃
MTM1	3-Phosphatase	PI(3)P, PI(3,5)P₂
MTMR1–8	3-Phosphatase	PI(3)P, PI(3,5)P₂
4-Phosphatase type 1	4-Phosphatase	I(3,4)P₂, I(1,3,4)P₃, PI(3,4)P₂
4-Phosphatase type 2	4-Phosphatase
OCRL1	5-Phosphatase	PI(4,5)P₂
SHIP1	5-Phosphatase	I(1,3,4,5)P₄, PI(3,4,5)P₃
SHIP2	5-Phosphatase	I(1,3,4,5)P₄, PI(3,4,5)P₃
Synj1	5-Phosphatase	PI(4,5)P₂, PI(3,4,5)P₃
SKIP	5-Phosphatase	PI(4,5)P₂, PI(3,4,5)P₃

Lipid phosphatases — 3′-phosphatase PTen and 5′-phosphatase SHIP — control PI3K-dependent T lymphocyte signaling, i.e., immune system development and responsiveness as well as prevention of lymphocyte proliferation and autoimmunity. Phosphatases PTen and SHIP convert PI(3,4,5)P₃ into PI(4,5)P₂ and PI(3,4)P₂, respectively.

2.10.1 Phosphatase and Tensin Homolog

Phosphatase and tensin homolog deleted in chromosome 10 (PTen) is a dual-specificity protein–lipid phosphatase that has protein Tyr phosphatase activity and hydrolyzes phosphorylated position 3 of PI(3)P, PI(3,4)P₂, and PI(3,4,5)P₃, as well as inositol (1,3,4,5)-tetrakisphosphate.¹¹⁶ Substrate PI(3,4,5)P₃ counteracts phosphoinositide 3-kinase activity. Phosphatase PTen can indeed be considered essentially as a PI(3,4,5)P₃ phosphatase that produces PI(4,5)P₂. The balance between PTen and PI3K controls PI(3,4,5)P₃ concentration in the plasma membrane that regulates cell survival and proliferation.

Although PTen functions predominantly as a lipid phosphatase, it dephosphorylates focal adhesion kinase and adaptor SHC to inhibit cell migration [101].

Binding of PTen to the plasma membrane is needed to maintain an appropriate level of phosphatidylinositol (3,4,5)-trisphosphate. Phosphatase PTen binds to negatively charged phosphatidylserine in the inner plasmalemmal leaflet and phosphatidylinositol (4,5)-bisphosphate. Phosphatase PTen actually switches between 2 states that regulate its location (membrane attachment and detachment), as well as its function and degradation. Phosphorylation of PTen regulates its membrane binding [135].¹¹⁷

Ubiquitous phosphatase PTen is rapidly degraded (half-life 2–4 h according to the cell type). Phosphatase PTen is involved in many pathways that can stimulate cell growth or apoptosis. It particularly intervenes in the development of the nervous system. Phosphatase PTen is an important regulator of class-1 PI3Ks. Leptin effect in pancreatic β cells requires coincident PI(3,4,5)P₃ generation and actin depolymerization owing to both the lipid and protein phosphatase PTen activities. Phosphatase PTen suppresses many tumor types. It functions as a tumor suppressor by precluding the protein kinase-B pathway. It indeed stops cell division and causes apoptosis when necessary.

Enzyme PTen is activated by transcription factors early growth response EGR1 and P53.¹¹⁸ Upon activation by IGF1, PKB phosphorylates EGR1. Afterward, EGR1 is sumoylated by alternate reading frame protein (ARF) in the nucleolus. The PKB–EGR1–ARF pathway of PTen transcription is characterized by inhibitory feedback, as PTen inhibits its own production [136].

Nucleus–cytosol partitioning of PTen results from PTen ubiquitinination and deubiquitinination by ubiquitin-specific peptidase (deubiquitinase) USP7 [137].¹¹⁹ Promyelocytic leukemia protein (PML) opposes HAUSP activity via adaptor death domain-associated protein and excludes PTen from the cell nucleus. Conversely, PML function is disrupted by the PML–RARα fusion oncoprotein that restores PTen nuclear location that is required for repression of tumor initiation and development.

2.10.2 Myotubularins and Myotubularin-Related Phosphoinositide 3-Phosphatases

Myotubularin MTM1¹²⁰ (Sect. 8.3.13.7) is either considered as a potent phosphoinositide 3-phosphatase [118] or an enzymatically inactive homolog of phosphoinositide phosphatase ( pseudophosphatases; Sect. 8.1) that can still bind 3-phosphorylated inositol lipids and acts as an adaptor [138].

The myotubularin-related (MTMR) family of phosphoinositide 3-phosphatases includes 14 members in humans (MTM1, MTMR1–MTMR13) [118]. These phosphatases use inositol phospholipids rather than phosphoproteins (Tyr^P) as substrates.

Myotubularin MTM1 and MTMR phosphatases dephosphorylate specifically PI(3)P and PI(3,5)P₂ that are involved in endocytosis to generate PI and PI(5)P, respectively. However, the myotubularin superfamily contains a subset of 6 proteins (MTMR5, MTMR9–MTMR13) that are catalytically inactive and function as adaptors for the active members as well as regulators.

Phosphatase MTMR6 interacts with Ca ${}^{++}$ -activated K⁺ channel K_Ca3.1 that can be indirectly activated by PI(3)P lipid. Calcium-activated potassium channels maintain the membrane potential that generates Ca ${}^{++}$ influx in activated naive and memory CD4 + T lymphocytes. Phosphatase MTMR6 regulates T-cell activation by dephosphorylating PI(3)P agent.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Tags: Intracellular Signaling Mediators in the Circulatory and Ventilatory Systems

Jun 3, 2017 | Posted by admin in CARDIOLOGY | Comments Off

Thoracic Key

Fastest Thoracic Insight Engine