Calcium-independent phospholipases A₂ and their roles in biological processes and diseases

The Ca²⁺-independent phospholipases A₂ (iPLA₂s) are part of a diverse family of PLA₂s that hydrolyze the sn-2 substituent from membrane phospholipids to release a free fatty acid and a lysolipid (, ). These enzymes are ubiquitously expressed, and in contrast to secretory PLA₂s (sPLA₂s) and cytosolic PLA₂s (cPLA₂s), do not require Ca²⁺ for either translocation or activity. Some of the first descriptions of iPLA₂ activity were in the mid- to late-1980s with the identification of a plasmalogen-selective PLA₂ in the cytosol of canine myocardium () that migrated with a molecular mass of 40 kDa. Analogous activity was subsequently described in insulinoma cells () and renal proximal tubules (), as well as in a macrophage cell line (). The Ca²⁺-independent PLA₂s are designated as group VI iPLA₂s (, ) and now include seven members, as described in Table 1: iPLA₂β (VIA-1 and -2), iPLA₂γ (VIB), iPLA₂δ (VIC), iPLA₂ε (VID), iPLA₂ζ (VIE), and iPLA₂η (VIF). Three others (iPLA₂φ, iPLA₂ι, and iPLA₂κ) have been recognized, but very little is known about them and they are not yet assigned to group VI (, ). Due to their shared homology with patatin, the iPLA₂s are included in the patatin-like protein family and are also referred to as PNPLAs. The iPLA₂s also share a consensus GXSXG catalytic motif contained within a patatin-like lipase domain. This review discusses the current understanding of the various iPLA₂s, starting with the less-described iPLA₂δ, iPLA₂ε, iPLA₂ζ, and iPLA₂η, followed by emerging reports relating to iPLA₂γ, and ending with the most widely examined, iPLA₂β.

The group VIC iPLA₂δ (PNPLA6), also known as neuropathy target esterase (NTE), was recognized for manifesting iPLA₂ and lysophospholipase activities in 2002 (). The gene for iPLA₂δ is located at chromosome 19p13.3-13.2, and encodes a protein containing 1,366 amino acids with a molecular mass of 146 kDa and an active site at S¹⁰⁰⁵. Expressed predominantly in neurons, iPLA₂δ localizes to the endoplasmic reticulum (ER) and Golgi apparatus (), and its inhibition or deletion leads to axonal degeneration. It is in this context that NTE was discovered during studies for causes of organophosphorus ester-induced paralysis in the late 1960s (, ). Whereas conventional KO of NTE is embryonic lethal (), conditional KO of NTE in the CNS leads to neurodegeneration (, ), suggesting loss of function as a causative factor in the development of neurological diseases. However, mutations in the catalytic site of NTE lead to hereditary spastic paraplegia (, ), a symptom of NTE-motor neuron disorder. Clinical manifestations of NTE-motor neuron disorder are only evident when mutations are carried by both alleles, suggesting that the neurodegeneration results from production of abnormal NTE, rather than due to reduction in NTE activity (, ). Among the syndromes associated with mutations in PNPLA6 are: Gordon Holmes (, ) and Boucher-Neuhauser (), characterized by early-onset ataxia and hypogonadism; Oliver-McFarlane (), characterized by trichomegaly, congenital hypopituitarism, retinal degeneration, and choroidal atrophy; Laurence-Moon (), characterized by progressive spinocerebellar ataxia and spastic paraplegia; and photoreceptor degeneration and childhood blindness (). An NTE-related iPLA₂ (PNPLA7) awaits further characterization (, ).

The group VID iPLA₂ε (PNPLA3), also referred to as adiponutrin, was described in 2001 (). The gene for iPLA₂ε is located at chromosome 22q13.31 and encodes a protein containing 481 amino acids with a molecular mass of 52 kDa and an active site at S⁴⁷. PNPLA3 is mainly expressed in intracellular membrane fractions in hepatocytes () and was originally described as a nutritionally-regulated adipose-specific transcript in 3T3-L1 adipocytes (). In addition to phospholipase activity, iPLA₂ε manifests TG lipase and acylglycerol transacylase activities (), leading to the suggestion that it facilitates energy/lipid mobilization and storage in adipocytes. In this regard, iPLA₂ε correlates highly with the development and progression of nonalcoholic fatty liver disease and has been identified as a genetic determinant of liver fibrosis (, ). Whereas the WT PNPLA3 exhibits lipolytic activity toward TGs, the rs738409 variant PNPLA3, where isoleucine¹⁴⁸ is replaced by methionine (L148M), reduces the access of substrates and activity of PNPLA3 toward glycerolipids. This leads to development of macrovesicular steatosis (, ), simple steatosis to steatohepatitis and progressive cirrhosis (), and hepatic fibrinogenesis by a sterol regulatory element-binding protein (SREBP)-1c-PNPLA3 pathway (). A greater impact of the L148M variant on hepatic lipid content is unmasked in the presence of other risk factors such as obesity (), visceral adiposity (), increased intake of sugars (), omega-6 PUFAs (), glucokinase regulatory protein gene variant (), chronic hepatitis B (), and hepatocellular carcinoma (). In contrast to these reports, PNPLA3 has been reported to restore lipid homeostasis () by mediating acylation of lysophospholipids and hydrolyzing TGs in the liver in a direct manner or by regulation by cofactors (, ).

The group VIE iPLA₂ζ (PNPLA2), also known as TST2.2, desnutrin, and adipose TG lipase (ATGL), was described in 2004 (, , ). The gene for iPLA₂ζ is located at chromosome 11p15.5 and encodes a protein containing 504 amino acids with a molecular mass of 55 kDa and an active site at S⁴⁷. Similar to iPLA₂ε, iPLA₂ζ also exhibits TG lipase and acylglycerol transacylase activities (). For optimal activity, ATGL requires the cofactor comparative gene identification-58 (CGI-58), which amplifies the hydrolase activity 20-fold (). Mutations in CGI-58, as in Chanarin-Dorfman syndrome (), lead to TGs in various tissues and decreases in both CGI-58 and ATGL have been reported to exacerbate myocardial steatosis and oxidative stress to promote cardiac apoptosis in a rodent T2D model (). Analogously, ATGL deficiency in mice promotes tissue accumulation of lipids and leads to premature death due to cardiomyopathy, as a consequence of reductions in fatty acid oxidative gene expression, mitochondrial fatty acid oxidation, and reduced oxygen consumption (). Macrophages with ATGL deficiency are more susceptible to ceramide-mediated mitochondrial dysfunction and programmed cell death (). β-Cell-specific ATGL-deficiency has been demonstrated to lead to hyperglycemia due to impaired insulin secretion, as a consequence of increased islet TG content with lower fatty acid levels. These mice also have decreased expression of PPARδ genes that encode enzymes required in mitochondrial oxidation, and this is reflected by impaired mitochondrial respiration and ATP production needed for glucose-stimulated insulin secretion (). While polymorphisms in PNPLA2 are reported to highly correlate with T2D (), the contribution of ATGL to insulin secretion and signaling has been challenged (, ). In addition to its links to CGI-58 and PPARδ, ATGL has been reported to interact with TNFα in adipocytes (), estrogen receptor α (ERα) in bone marrow (), fat-specific protein 27 (FSP27) in human adipocytes (), sirtuin 1 (SIRT1) during β-adrenergic signaling (), hepatic PPARα (), AMPK during thermogenesis (), and to be a candidate for transcriptional control by PPARγ-mediated signals ().

The group VIF iPLA₂η (PNPLA4), also known as gene sequence-2 (GS2), was described in 1994 (). The gene for iPLA₂η is located at xp22.3 and encodes a protein containing 253 amino acids with a molecular mass of 27 kDa and an active site at S⁴³. Similar to iPLA₂ε and iPLA₂ζ, iPLA₂η exhibits TG lipase and acylglycerol transacylase activities (). Though expression of iPLA₂η in a variety of tissues (liver, brain, skeletal muscle, lung, placenta, kidney, and pancreas) was identified in 1994, and more recently in adipose tissue (), to date, very little is known about its biology or its role in metabolic diseases. Similar to iPLA₂ε and iPLA₂ζ, iPLA₂η activation is proposed to contribute to regulation of anabolic and catabolic fluxes of acyl equivalents in tissues. It has been suggested that the TG lipase activity of iPLA₂ε, iPLA₂ζ, and iPLA₂η play roles in serum fatty acid accumulations associated with metabolic syndrome and T2D. A related GS2-like iPLA₂ (PNPLA5) has yet to be characterized (, ).

The group VIB iPLA₂γ (PNPLA8) genomic organization and mRNA sequence were first described in a variety of tissues (skeletal muscle, heart, placenta, brain, liver, and pancreas) in 2000 () and later in the same year in lymphocytes (). The gene for iPLA₂γ is located at 7q31 and encodes a protein containing 782 amino acids with a molecular mass of 90 kDa and an active site at S⁴⁸³. Recognition of the similarity in the catalytic domain between human iPLA₂γ, cPLA₂, and plant PLA₂ patatin and conservation of sequence surrounding Asp⁶²⁷, and noting that substitution of alanine for either Ser⁴⁸³ of Asp⁶²⁷ caused loss of iPLA₂γ activity, led to the suggestion that the Ser-Asp dyad constitutes the active site in human iPLA₂γ (). Initially recognized as membrane associated (, ), dual-competing subcellular localization signals have been identified in discrete isoforms of iPLA₂γ () that promote its accumulation and expression of activity in the peroxisomes and mitochondria (), leading to the suggestion that iPLA₂γ plays a role in integration of lipid and energy metabolism. Further, iPLA₂ activity in the ER of rabbit and rat kidney () and ventricular myocyte membranes () has been demonstrated to be due to iPLA₂γ.

The iPLA₂γ protein contains four methionine residues that can act as potential translational initiation sites (, ) to generate the full-length (∼88 kDa) and three truncated products (77, 74, and 63 kDa). Attempts at expression of the truncated products in HEK293 cells, however, led to the predominant expression of the 63 kDa product (), the isoform reported earlier to be expressed in peroxisomes (). Further examination of parental cells revealed that the 63 kDa isoform was much more abundant than the full-length iPLA₂γ in HEK293 and human colorectal cancer cell lines, HCA-7 and WiDr, while in human bronchial epithelial (BEAS-2B) and rat fibroblastic (3YI) cells, the full-length iPLA₂γ was the predominant isoform (). These authors suggested that iPLA₂γ potentiates arachidonic acid (AA) release from various subclasses of phosphatidylethanolamine (PE) and phosphatidylcholine (PC) to increase prostaglandin E₂ (PGE₂) production via cyclooxygenase (COX)-1 and -2, and this contributes to cell growth and tumorigenesis. In contrast, comparative substrate preference studies revealed that unlike cPLA₂, which generates predominantly 1-palmitoyl lysophosphatidylcholine (LPC) and AA from 1-palmitoyl-2-arachidonoyl-sn-phosphatidylcholine hydrolysis, and iPLA₂β, which exhibits mixed PLA₁/PLA₂ activities and generates 1-palmitoyl LPC at an initial 3-fold rate greater than 2-arachidonoyl LPC, iPLA₂γ overexpressed in and purified from Sf9 cells hydrolyzed saturated and monounsaturated fatty acids at equal rates from the sn-1 or sn-2 position in diacyl PC substrates. However, it was less effective in releasing PUFAs from the sn-2 position, as reflected by generation of 2-arachidonoyl LPC at a 10-fold faster rate than 1-palmitoyl LPC ().

Understanding the role of iPLA₂γ-derived lipid signals has substantially advanced following the generation of mice with tissue-specific overexpression or global KO of iPLA₂γ. Cardiac-specific overexpression of iPLA₂γ presented multiple phenotypes that included reductions in myocardial phospholipid mass in fasted and fed states, accu­mu­lation of TGs with caloric restriction, acute fasting-induced hemodynamic dysfunction (that was accompanied by loosely packed and disorganized mitochondrial cristae), and elevated levels of 2-arachidonoyl-sn-glycero-3-phosphocholine and 2-docosahexaenoyl-sn-glycero-3-phosphocholine (). These findings were associated with increased expression not of the full-length, but of the 70 and 63 kDa iPLA₂γ isoforms. Impairment in mitochondrial function is also evidenced in iPLA₂γ-null mice, which exhibit growth retardation, cold intolerance, and increased mortality due to aortic stress that were associated with decreased myocardial function and O₂ consumption (). The iPLA₂γ-null mice also become resistant to Western diet-induced increases in body weight, adiposity, circulating levels of cholesterol, glucose, and insulin; and insulin resistance, and glucose intolerance (). Ability to utilize fat and carbohydrates is also affected in these mice in association with severe impairment in skeletal muscle mitochondrial oxidation of fatty acids. Subsequent studies revealed that marked decreases in cardiolipin molecular species containing 22:6 were an underlying cause for the mitochondrial uncoupling evident in the iPLA₂γ-null mice (). Similarly, hippocampal phospholipid metabolism was found to be severely compromised with iPLA₂γ-deficiency leading to a mitochondrial neurodegenerative disorder characterized by degenerating mitochondria, autophagy, and cognitive dysfunction, that was associated with alterations in the compositions and content of PCs, PEs, oxidized PEs, and ceramides and a shift in cardiolipins to shorter chain molecular species (). Increased lipid peroxidation was also evident in the skeletal muscles of iPLA₂γ-null mice, which exhibited abnormal mitochondrial function, oxidative stress, growth retardation, and loss of skeletal muscle structure and function (). These findings are consistent with a previous report of the protective effects of iPLA₂γ, deduced using a selective inhibitor of iPLA₂γ against oxidant-induced lipid peroxidation and necrosis of renal proximal tubular cells (). In contrast to the earlier report in the heart (), the full-length iPLA₂γ was the major isoform detected in the last two studies (, ).

The identification of iPLA₂γ in the mitochondria and ER paved the way for studies that demonstrated a protective role for iPLA₂γ against cell death. Utilizing knockdown protocols, global iPLA₂γ-null mouse model, or by selective inhibition with bromoenol lactone (BEL) (suicide substrate) of iPLA₂γ (R-BEL) versus iPLA₂β (S-BEL), several studies reported protective effects of iPLA₂γ against oxidant- and cytokine-induced cell death. Human astrocytes exposed to hydrogen peroxide or tert-butyl hydroperoxide exhibited cell death, and pretreatment with S-BEL, but not R-BEL, amplified loss of ATP levels and cell necrosis (). Similarly, knockdown of iPLA₂γ in renal proximal tubular cells resulted in increased susceptibility to oxidant-induced cell death, elevations in lipid peroxidation, and uncoupled oxygen consumption (, ). Analogous findings were reported in INS-1 insulinoma cells, where knockdown of iPLA₂γ promoted increases in cytokine- and oxidant-induced membrane peroxidation and apoptosis (). Cytoprotective effects of iPLA₂γ were also demonstrated in glomerular epithelial cells, and these were attributed to iPLA₂γ-mediated upregulation of ER chaperones (). A general conclusion derived from these studies is that the lack of iPLA₂γ decreases substrate availability for reacylation, leading to increases in lipid peroxidation. In apparent contrast to these reports, genetic ablation of iPLA₂γ or its inhibition with R-BEL attenuated calcium-, reactive oxygen species (ROS)-, or oxidized lipid-mediated increases in liver mitochondrial swelling, mitochondrial permeability transition pore opening, and cytochrome c release from mitochondria, which trigger the intrinsic apoptotic pathway (, ).

The more recent description of iPLA₂γ, to date, has limited wide studies of its role in clinical diseases, but a few reports suggest a role for iPLA₂γ in certain clinical-related disorders. Chagas' disease is caused by protozoan parasite Trypanosoma cruzi, which infects cardiac myocytes promoting release of inflammatory mediators such as eicosanoids. Inhibition of iPLA₂γ attenuated AA and PGE₂ release and platelet-activating factor (PAF) production from HL-1 cardiac myocytes infected with T. cruzi (), and these effects were alleviated by pretreatment with R-BEL. Consistent with a protective role of iPLA₂γ in this process, the survival rates were lowered and tissue parasitism amplified in T. cruzi-infected iPLA₂γ-null mice, suggesting that iPLA₂γ activity affords protection against acute state cardiomyopathy in Chagas' disease (). In contrast, iPLA₂γ-deficiency has been shown to increase bleeding time and provide resistance to thromboembolism (), raising the possibility of targeting iPLA₂γ for antithrombotic drug development. To date, the only reported clinical manifestation relating to iPLA₂γ is a report that its absence is associated with myocardial dysfunction, cognitive defects, and mitochondrial degeneration () in a case study that closely parallels the phenotype in iPLA₂γ-null mice ().

The group VIA iPLA₂β (PNPLA9) is the most widely described of the iPLA₂s and expression of its activity was first described in P388D1 macrophage-like cells in 1994 () and later shown to be the same enzyme () as that cloned from Chinese hamster ovary cells in 1997 (, , ). Unlike cPLA₂, which exhibits preference for hydrolysis of AA from the sn-2 position (), the iPLA₂s demonstrate no substrate specificity and manifest PLA₂/PLA₁, lysophospholipase (, ), transacylase (, ), and thioesterase (, ) activities. The extensively studied iPLA₂β was cloned from hamster (), mouse (), and rat (), and they represent species homologs that are 85 kDa proteins (752 amino acids) with a serine lipase consensus sequence (GTSGT), preceded by eight N-terminal ankyrin (Ank) repeats (, ).

A homologous 88 kDa iPLA₂β was cloned from human lymphocyte lines and testis () that contains a 54-amino acid insert interrupting the eighth Ank repeat. Subsequent analyses with human pancreatic islet mRNA identified cDNA species that encoded two distinct 85 kDa (VIA-1) and 88 kDa (VIA-2) human iPLA₂β isoforms (). Analogous transcripts were also identified in human promonocytic U937 cells. The human iPLA₂β gene resides on chromosome 22 in region q13.1 and contains 16 exons in the VIA-2 transcript. Exon 8 is not present in the VIA-1 transcript, indicating that it arises by an exon-skipping mechanism of alternative splicing.

Additional alternate splicing events generate iPLA₂β variants that differ in their subcellular localization, catalytic activity, and likely cellular function (). Splice variants (Ank-1 and Ank-2) encode premature stop codons due to alternatively spliced exon 10a. The proteins encoded by these splice variants, VIA Ank-1 (53 kDa, ∼479 amino acids) and VIA Ank-2 (47 kDa, ∼427 amino acids), terminate after the seventh Ank repeat domain and before the active site, whereas VIA-3 (∼70 kDa, 640 amino acids) terminates just after the lipase active site. Two additional active iPLA₂β isoforms have been recognized to arise from proteolytic cleavage: a ∼63 kDa isoform (VIA-4, 623 amino acids) arising from caspase-3-catalyzed cleavage at the N terminal (, ) and a ∼70 kDa (VIA-5, ∼640 amino acids) isoform arising from C-terminal cleavage (). Proteomic analyses by mass spectrometry further reveal that the iPLA₂β is a candidate for numerous truncations at the N-terminal end (, ), but the activities and biological roles manifested by these products have not yet been discerned.