OUP user menu

★ editor's choice ★

Lipid signalling in cardiovascular pathophysiology

Joel S. Karliner , Joan Heller Brown
DOI: http://dx.doi.org/10.1093/cvr/cvp096 171-174 First published online: 18 March 2009

1. Introduction

Although the role of lipids in the pathogenesis of cardiovascular disease has been appreciated for several decades, recent studies have advanced our understanding and revealed complex new functions and interactions for lipid molecules that are both beneficial and adverse. It is now well appreciated that lipids are not merely structural components of cell membranes but serve as substrates for enzymes that generate second messengers involved in cell signalling. In the case of the sphingolipids, sphingomyelinase and sphingosine kinase (SK) are regulated enzymes and the product, sphingosine-1-phosphate (S1P), is a ligand for G-protein-coupled receptors (GPCRs) that activate myriad cellular responses. Inositol phospholipids, or phosphoinositides, are synthesized or degraded through regulated enzymes including phosphatidylinositol 3′-kinases (PI3Ks), which generate the second messenger PIP3, and phospholipase C which generates inositol trisphosphate and diacylglycerol (DAG) as well as modulating membrane levels of phosphatidylinositol 4′5′-bisphosphate (PIP2).

Sphingolipid pathways, including responses involving sphingomyelinases, have been appreciated as signalling mediators in the cardiovascular system under both normal conditions and during acute and chronic stress. It is well known that activation of the SK/S1P pathway promotes survival in many cells and tissues, but this pathway has only recently been identified as having similar effects in the heart and vasculature. The delineation of multiple S1P receptor subtypes has added to the complexity of this regulation, as have the observations regarding the function of S1P as a protective ‘cargo’ carried by high-density lipoprotein (HDL). Experimental data are beginning to result in clinical studies designed to test the efficacy of sphingolipids in therapeutics. These include agonists and antagonists for S1P receptors, neutralizing antibodies for S1P, and inhibitors of SK.

Phosphatidylinositol 4′5′-bisphosphate is hydrolysed by phospholipase C, which is activated by GPCRs and other cell-surface receptors. One product, DAG, is the endogenous activator of protein kinase C (PKC). Studies of PKC isozymes have yielded a wealth of information regarding their function in cardiac injury and hypertrophy and have also led to the development of new therapeutic tools. PIP2 can itself serve as a second messenger, with changes in its levels altering the activity of cardiac ion channels and contributing to the development of arrhythmias.

Phosphatidylinositol 3′-kinase is regulated by growth factors, cytokines, and GPCR activation to yield the product PIP3, which in turn participates in the activation of the protein kinase Akt and its downstream targets. Akt has well-defined protective roles, many of which are recapitulated in the cardiovascular system, and studies using knockouts of specific PI3K isoforms as well as deletion and overexpression of Akt have both confirmed and expanded the potential for signalling through the PI3K pathways in the cardiovascular system.

Finally, phospholipase A2 acts upon membrane phospholipids to generate arachidonic acid. Considerable progress has been made in elucidating the highly complex mechanisms involved in the generation of eicosanoid metabolites and in understanding their role in cellular responses, but much remains to be learned regarding the beneficial and harmful effects of these lipid second messengers.

2. Effects of sphingolipids on the heart

Pavoine and Pecker review the three main types of sphingomyelinases (SMases) and their roles in normal and abnormal cardiovascular physiology.1 SMases hydrolyse the ubiquitous membrane protein sphingomyelin to release ceramide. The latter can induce apoptosis, but via conversion to sphingosine and then to S1P, can also exert cardioprotection. In addition, the authors review the actions of several SMase species: lysosomal and secreted SMases, which comprise the acid SMase isoforms, and neutral SMase. Effects on vascular tone, atherosclerosis, ischaemia/reperfusion injury, and heart failure are described, and the possibility of targeting SMases for therapeutic purposes is discussed. In an original article, Cogolludo et al.2 identify neutral SMase-derived ceramide and activation of ξPKC as necessary events in the signalling cascade leading to hypoxic pulmonary vasoconstriction.

Following SMase hydrolysis of sphingomyelin to ceramide, the latter is further metabolized by ceramidase to yield sphingosine, which can then be phosphorylated by SK to yield S1P. The review by Karliner summarizes the regulation and functional consequences of activating SK.3 Evidence is presented that gene-targeted mice null for the SK1 isoform whose hearts are subjected to ischaemia/reperfusion injury exhibit increased infarct size and respond poorly either to ischaemic pre- or postconditioning. Data showing that cardiac SK activity and S1P parallel these observations are described. It is notable that ischaemic postconditioning combined with sphingosine and S1P rescued the heart from prolonged ischaemia. The regulation of SK is considered in relation to therapeutic approaches for the treatment of acute myocardial injury.

In a companion Spotlight review, Means and Brown summarize the literature describing the functions of S1P receptors in the heart.4 The authors focus on the differential distribution and downstream signalling cascades that are selectively or promiscuously coupled to the S1P1, SIP2, and S1P3 receptor subtypes. The role of S1P receptor activation in cardioprotection is discussed, with particular emphasis on the receptor subtypes and biochemical mediators as revealed both by pharmacological inhibitor studies and by investigations using S1P2 and S1P3 receptor gene-targeted mice. In contrast to the role of S1P in protecting cardiomyocytes, there is limited evidence for the involvement of S1P receptors in cardiomyocyte hypertrophic growth. S1P does, however, have well-established effects on electrophysiological and contractile responses of cardiomyocytes, and these can increasingly be attributed to particular receptor subtypes and receptor compartmentalization within the myocyte.

In an original article, Lowe et al.5 report that in cultured adult mouse cardiac fibroblasts, transforming growth factor-β stimulates SK1 activity, causing ‘inside-out’ S1P release that results in autocrine/paracrine activation of S1P2 receptors and subsequent collagen production. Taken together, these reviews and original papers illuminate recent progress in unravelling the mysteries of these ‘enigmatic’ lipids in the heart, and identify areas that as yet remain hidden.

3. Sphingolipid effects on the vasculature

Three review articles cover different aspects of S1P effects on the cardiovascular system with emphasis on their effects on the vasculature. The contribution from the Levkau group considers the functions of HDL-associated S1P,6 that from Igarishi and Michel concerns the effect of S1P on vascular tone,7 and the contribution by the Hla laboratory focus on S1P2 receptor-mediated regulation of vascular physiology and pathology.8

High-density lipoprotein is best known for its role in cholesterol transport but somewhat surprisingly turns out to be the major carrier of S1P in the plasma. It has become increasingly clear that HDL has many salutary effects, and the hypothesis that some of these are mediated through S1P signalling is presented in the scholarly review by Sattler and Levkau.6 The authors present a broad spectrum of experimental findings demonstrating both commonality and difference in HDL and S1P effects on the cardiovascular system, with the goal of elucidating the extent to which the effects of HDL can be explained by its being a carrier for S1P and inducing S1P-mediated responses. Regulation of endothelial functions by HDL and S1P are discussed, including their effects on endothelial NO synthase (eNOS)-mediated NO production and prostacyclin production (promoting vasorelaxation) as well as the effects of S1P and HDL on endothelial barrier integrity, angiogenesis, and endothelial precursor cell responsiveness. Inflammatory cell adhesion regulation by HDL and S1P involves induction of NO and prostacyclin as well as inhibition of ROS generation, and in this, as in other systems, the authors delineate how free S1P and HDL-associated S1P may differ in their signalling properties. The intriguing possibility that HDL-associated S1P has largely beneficial effects, whereas free S1P can, in contrast, exert pro-inflammatory, vasoconstrictive, and other adverse effects, provides compelling rationale for all of us to increase our HDL levels!

An interesting extension of these studies comes from the original article by Frias et al.9 in which HDL and reconstituted HDL, which can be modulated to contain different components including S1P, were examined for their effects on cardiomyocytes. The authors demonstrate for the first time that HDL and S1P activate Stat 3, apparently through S1P2 receptors, and suggest that reconstituted HDL could have therapeutic cardioprotective functions.

Igarashi and Michel,7 in their Spotlight review, provide insight into mechanisms by which S1P affects the vasculature and discuss the complexities in delineating the ultimate effects of S1P on vascular tone. The authors' seminal studies and those of others demonstrate that endothelial cells are major targets for S1P action, that eNOS is enriched in endothelial cell caveolar domains that also contain S1P1 receptors, and that S1P activates eNOS through both Ca2+/calmodulin-mediated release of inhibition by caveolin and through eNOS phosphorylation by Akt. Both mechanisms are enlisted by S1P for its effects on endothelial cells, resulting in NO release and subsequent relaxation of vascular smooth muscle cells. S1P1 receptor expression in endothelial cells is also dynamically regulated. A potentially counteracting mechanism is the direct effect of S1P on vascular smooth muscle. Mechanistically, S1P receptor activation of RhoA and ROCK phosphorylate MYPT and sensitize smooth muscle to Ca2+, but an increase in cytosolic [Ca2+] is also required for contraction. As the authors make clear, whether a given blood vessel responds to S1P with vasodilatation or vasoconstriction can be context dependent, determined by the available concentration of S1P, the receptor subtypes that are activated, the particular vascular bed, and other factors. Nonetheless, mechanistic understanding can lead to informed predictions on which vascular response would occur and why.

A complementary original paper by Choi et al.10 compares contractile mechanisms by which S1P and sphingosylphosphorylcholine lead to constriction of rabbit coronary artery. In this system, contractility induced by increased Ca2+ vs. Rho/ROCK/MYPT-mediated Ca2+ sensitization pathways are shown to be agonist specific, with S1P acting largely through the elevation of Ca2+ rather than through the Rho pathway.

The Spotlight review by Skoura and Hla8 considers S1P effects on vascular development and pathophysiology. The review provides a novel focus by considering the role of the S1P2 receptor, far less well examined than S1P1 for its role in vascular pathology. The authors detail the molecular events in coupling of S1P2 receptors to multiple G-proteins and signalling pathways, emphasizing that the S1P2 receptor is particularly efficient in activating RhoA signalling and indirectly causing negative regulation of Rac. Reports in the literature describing S1P receptor deletion are reviewed to demonstrate that formation and maintenance of mouse vascular networks during normal development in the mouse is dependent on S1P2 receptors. Haemorrhage and embryonic lethality as well as hearing defects and pathological changes in inner ear vascular structure accompany S1P2 receptor gene deletion, most evidently when it occurs along with loss of either S1P1 or S1P3 receptors, which can provide redundant signalling. The authors discuss work demonstrating that S1P2 receptor activation increases vascular permeability through the disruption of VE-cadherin-based junctions, and that S1P2 receptors are involved in inflammatory cell infiltration, pathological angiogenesis, and neointimal lesion formation. Although much more needs to be learned about the distribution as well as physiological function of S1P2 receptors, there is evidence that this uniquely coupled receptor subtype may be a useful pathophysiological therapeutic target.

4. Cardiac effects of other lipid-related mediators

In addition to the well-recognized effects of ϵ- and δPKC isozyme activation during ischaemia/reperfusion injury and ischaemic preconditioning, these and other PKC isozymes may mediate different functions in acute vs. chronic heart disease. Palaniyandi et al.11 identify these functions in cardiac remodelling and heart failure. After describing the mechanisms of PKC peptide activators and inhibitors, they review the roles of PKC isozymes in cellular and animal models of cardiac hypertrophy with a focus on βPKC. Discrepancies between results in different rodent species are described, and a convenient and comprehensive reference table is provided. The role of PKC isozymes in cardiac fibroblast proliferation, fibrosis, and inflammation is reviewed. Finally, downstream targets of PKC isozymes in cardiac remodelling are described. The observations summarized in this review form the basis for testing activation or inhibition of selected PKC isozymes as potential therapeutic targets in the treatment of heart failure.

An even more complex series of biochemical transformations is described by Jenkins et al.12 in their review of eicosanoids and their signalling pathways in the heart. These pathways may be acutely adaptive, but may become maladaptive when they are chronically activated. The effects of calcium-independent phospholipase A2β (iPLA2β), which is the primary mediator of arachidonic acid release from cellular phospholipids during ischaemia, are described in detail. Other forms of phospholipase A2 are covered as well. The importance of calmodulin as an inhibitor of iPLA2β is emphasized. The subsequent stereospecific oxidation of arachidonic acid via cyclo-oxygenases, lipoxygenases, and cytochrome P450 mono-oxygenases is described. The specific targets of each enzymatic transformation (prostaglandins, leukotrienes, and eicosatrienoic acids) may variously have either detrimental and/or beneficial effects on myocardial inflammation, vasomotion, viability, and ion channel function, resulting in effects on cardiac hypertrophy, preconditioning, infarction, and arrhythmogenesis. Nadtochiy et al.13 note that the activation of both lipoxygenases and mitochondrial PLA2 favour the generation of NO-derived electrophilic nitroalkene derivatives such as nitro-linoleate. These authors report that nitro-linoleate induced by ischaemic preconditioning causes mild mitochondrial uncoupling that results in resistance to ischaemia/reperfusion injury.

5. Phosphoinositide signalling in the cardiovascular system

Membrane phosphoinositides (PtdIns) that are phosphorylated at their 3′ position, in particular PIP3, play a central role in signal transduction by virtue of their ability to recruit downstream effectors containing pleckstrin homology domains. Accordingly, the enzymes that regulate PIP3 levels are critical determinants of cellular responses. Phosphoinositide 3-kinases control the synthesis of PIP3 while PTEN is the major phosphatase that degrades this lipid. The Spotlight review by Oudit and Penninger14 provides a comprehensive review of the molecular targets of PI3K and PTEN and the physiological and pathological functions that are regulated through changes in PI3K and PTEN activity. These include alterations in cell survival, myocardial hypertrophy, cardiomyocyte contractile function, channel activity, and cell metabolism. The complexities of the existence of distinct isoforms of class 1 PI3Ks is highlighted by a number of examples of differences in the effects of knockdown or overexpression of PI3Kγ vs. PI3Kα/β, e.g. on myocardial contractility. Additional insight is provided into the effects of PI3K/PTEN on preconditioning, ischaemia/reperfusion injury, and heart failure development, paradigms that in several cases reveal unexpected dissociations between data from cellular and in vivo models.

The PI3K signalling pathway also plays a prominent role in the vascular system by virtue of its regulation and signalling role in vascular endothelial and smooth muscle cells as well as in inflammatory cells and platelets. An informed understanding of how this occurs is provided in the review by Morello et al.15 Functionally, the PI3K pathway, in large part through the activation of the PIP3 target Akt and its myriad downstream signal transducers including eNOS, mTOR, and FOXO, effects changes in vascular tone as well as endothelial cell differentiation, endothelial/leukocyte interactions, and angiogenesis. The translational implications for understanding these pathways are provided by the discussion of the role of PI3K signalling in many events contributing to atherosclerosis and of the role of PI3K in platelets and thrombus formation. Information on the role of eNOS activation complements that presented for S1P signalling in the Igarashi and Michel review,7 and the issue of PI3K isoforms complements that provided by Oudit and Penninger14 by discussing the potential utility of isoform-selective inhibitors as therapeutic modalities for the treatment of cardiovascular disease.

The formation of PIP3 occurs secondarily to the activation of PI3Ks by membrane-delimited GPCRs, growth factor, or cytokine receptors. Accordingly, the general conception of this signalling pathway as depicted in most schematics is of a process that begins at the plasma membrane and generates messages in the cytosol. The review by Miyamoto et al.16 reveals relatively new evidence that in cardiomyocytes, as in other cells, there is significant signalling through Akt at the level of two intracellular organelles, the nucleus, and the mitochondria. Indeed, nuclear expression of Akt mimics many effects of hormonal activation in cardiomyocytes, and enzymes involved in Akt activation can be demonstrated in the nuclear compartment. By virtue of its actions in the nucleus, Akt increases gene expression, particularly of a protein kinase called Pim-1. There is abundant evidence concerning regulation and effects of Pim-1 kinase in non-cardiac systems and, as documented in this review, many of these appear to be replicated in cardiomyocytes where nuclear expression of Akt has profound effects on cardiomyocyte survival and cardiac function. Mitochondria-associated Akt is increased in cardiomyocytes following agonist treatments known to activate PI3K, and there is other evidence for translocation of Akt to this compartment. Miyamoto et al. describe many potential mitochondrial targets of Akt including Bcl-2 family proteins, GSK3, and hexokinase. Phosphorylation of these proteins contributes to the ability of Akt to protect mitochondria against Ca2+ overload and subsequent dysfunction. Hexokinase is a particularly interesting target, since, as discussed in detail, it is a putative component of the mitochondrial permeability transition pore as well as being an enzyme involved in energy production and metabolism; thus, it is a molecule that may integrate the well-known metabolic and survival functions of Akt in cardiomyocytes.

PIP2 is a minor phospholipid component of the plasma membrane, but one that is in fact considerably more abundant than PIP3. Although PIP3 has seen a stellar rise in popularity as the product of PI3K and a regulator of Akt (see above), its parent, PIP2, had a far earlier debut as a key player in cell signalling. PIP2 plays an important role as the substrate for phospholipase C and hence for InsP3/Ca2+ and DAG/PKC second messenger formation. Equally important, to the function of excitable tissues such as the heart and brain is the role of PIP2 as a regulator of the activity of other proteins, in particular the ion channels that are embedded in the plasma membrane sea of phospholipids. The review by Woodcock et al.17 presents an overview of the role of PIP2 in cardiac arrhythmias, focusing on the multitude of K+ channels as well as Na+ and Ca2+ currents that can change their activity in response to alterations in PIP2 levels. Because these channels regulate resting membrane potential, repolarization, automaticity, and pacemaker activity as well as electrical conduction, their involvement in cardiac pathophysiology can be profound. The possibility that the second messenger, InsP3, participates in Ca2+ release in defined compartments in the heart and, under pathological conditions, is associated with arrhythmias, is also discussed. A range of evidence implicating PIP2 or InsP3 in arrhythmias associated with heart failure, ischaemia/reperfusion, and atrial fibrillation is discussed, along with the need to approach this pathway through highly targeted therapeutics.

6. Summary

The review articles and original contributions clearly document the importance of lipid mediators and their signalling pathways in cardiac pathophysiology. We hope that this issue will serve not only to update the knowledge of cardiovascular scientists in this rapidly moving and complex field but also to stimulate new lines of investigation that may enlighten therapeutic approaches to the treatment of cardiovascular disease.

Conflict of interest: none declared.


  • The opinions expressed in this article are not necessarily those of the Editors of the Cardiovascular Research or of the European Society of Cardiology.