Lipid metabolism in sickness and in health: emerging regulators of lipotoxicity

Summary

Lipids play crucial roles in signal transduction, contribute to the structural integrity of cellular membranes, and regulate energy metabolism. Questions remain as to which lipid species maintain metabolic homeostasis and which disrupt essential cellular functions leading to metabolic disorders. Here we discuss recent advances in understanding lipid metabolism with a focus on catabolism, synthesis, and signaling. Technical advances including functional genomics, metabolomics, lipidomics, lipid-protein interaction maps, and advances in mass spectrometry have uncovered new ways to prioritize molecular mechanisms mediating lipid function. By reviewing what is known about the distinct effects of specific lipid species in physiological pathways, we provide a framework for understanding newly identified targets regulating lipid homeostasis with implications for ameliorating metabolic diseases.

eTOC Blurb

Yoon et al., discuss recent advances in understanding lipid metabolism with a focus on catabolism, synthesis, and signaling. Through the lens of recent technical advances including functional genomics, metabolomics, lipidomics, lipid-protein interaction maps, and advances in mass spectrometry, the authors discuss new ways to prioritize molecular mechanisms mediating lipid function.

INTRODUCTION

Cellular lipids contain a diverse collection of individual molecular components that give rise to many tens of thousands of lipid species, the compendium of the cell, collectively called the lipidome (Yang et al., 2009). Lipid metabolism affects many cellular processes critical for homeostasis including membrane synthesis and the use of lipids (i.e. triglycerides) as an energy store. Fatty acids (FAs) are essential lipids that constitute the major structural components of membrane lipids (i.e., glycerophospholipids and sphingolipids) while also serving as an important energy source through mitochondria-mediated beta-oxidation and tricarboxylic acid (TCA) cycle catabolism.

Excessive levels of circulating lipids have been linked to metabolic diseases (Musunuru and Kathiresan, 2016, 2019) and cancer (Beloribi-Djefaflia et al., 2016). The harmful effects of prolonged exposure to excess lipids is referred to as “lipotoxicity” (Lytrivi et al., 2020; Sharma and Alonso, 2014) -- a term first coined by Roger Unger and colleagues to explain the inhibition of pancreatic β-cell function and the development of type 2 diabetes in the pancreatic islets of rats overloaded with lipids (Lee et al., 1994). The molecular mechanisms underlying lipotoxicity include endoplasmic reticulum (ER) stress, oxidative stress, mitochondrial dysfunction, impaired autophagy, and inflammation (Lytrivi et al., 2020). Specifically, in metabolic disorders where there is an imbalance between the uptake or synthesis and consumption of fatty acids (FAs), lipid intermediates accumulate intracellularly resulting in cellular dysfunction and death in diverse tissues including the kidney, brain, skeletal muscle, and heart (Goldberg et al., 2012). Effectively channeling free FAs to structural lipids, lipid droplets, or to the mitochondria for beta-oxidation has the potential to mitigate harmful effects of lipid accumulation, leading to new questions: (i) How do imbalances in the uptake or synthesis of lipids and their consumption or destruction affect downstream signaling pathways? (ii) How does the intracellular accumulation of lipid intermediates directly contribute to cellular dysfunction?

In this review, we highlight key roles for lipids across diverse cell types in order to provide a framework for understanding the mechanisms that link excess lipids and lipotoxicity to dysfunction in metabolic diseases including chronic kidney disease, fatty liver, heart failure, obesity, neurodegeneration and cancer. Understanding the mechanisms regulating the fate of lipids within cells will provide clues into tightly regulated mechanisms of homeostasis. We discuss fatty acid synthesis, uptake, degradation, and signaling in the context of homeostasis as well as in disease states (Figure 1). Finally, we highlight emerging technologies including functional genomics, lipid-protein interaction maps, and advances in mass spectrometry as tools to identify therapeutic targets for metabolic diseases.

Figure 1. Overview of Lipid Metabolism.

(A) A systematic approach is necessary to categorize lipids as beneficial or lipotoxic. Bioactive lipid species have different roles in cellular responses, including beneficial roles in lipid homeostasis through the regulation of proliferation, storage, energy production, cell signaling, and lipid membrane composition. Lipids play a lipotoxic role by influencing cell death, ER stress, and ROS production. It is important to understand which lipid species maintain metabolic homeostasis and which disrupt essential cellular functions leading to metabolic disorders. (B) The role and structure of lipids are determined by uptake, synthesis, storage and consumption across different cellular organelles. In the anabolic pathway, lipids are taken to the ER and cytosol for lipid synthesis. For catabolism, lipids are transmitted to the mitochondria and peroxisomes. These pathways generate fatty acids, storage lipids, such as cholesterol, and triglycerides, signaling lipids containing N-acetylethanolamines (oleoylethanolamide) and cholesterol-derived bile acids. Membrane lipids include glycerophospholipids and sphingolipids. Fatty acids are building blocks of all lipids, and palmitic acid, stearic acid, oleic acid, linoleic acid, arachidonic acid, docosahexaneoic acid are illustrated as examples for saturated, monounsaturated, and polyunsaturated FAs. ABCD1, ATP Binding Cassette Subfamily D Member 1; ACAT, acyl-CoA:cholesterol acyltransferase; ATGL, adipose triglyceride lipase; CPT, carnitine palmitoyl-transferase; DGAT1, diacylglycerol O-acyltransferase 1; NPC1, Niemann Pick type-C 1; TAG, triacylglycerol.

LIPID BIOLOGY

Over 40,000 lipids have been identified across the kingdoms of life (http://www.lipidmaps.org), yet we still have an incomplete understanding of the roles most of these lipids play in cell biology and physiology. By definition, lipids are complex molecules generated from simpler constituents through enzymatic reactions. Typically, each lipid consists of a head group with a unique chemical composition that is esterified to hydrophobic tails made up of fatty acyl chains or sphingoid bases (Raghu, 2020). The biological functions of different lipid classes are defined by the lipid head group. Fatty acids have diverse biological roles and serve as building blocks in cells, important biochemical intermediates, major determinants of membrane properties, modulators of cellular signaling pathways, and as a fuel source (Figure 2). For a deeper analysis of the chemical diversity that regulates lipid function, we direct readers to an excellent comprehensive review on this topic (Harayama and Riezman, 2018).

Figure 2. Major Processes of Lipid Metabolism: uptake, synthesis, and consumption.

Lipid metabolism includes catabolic processes that generate energy, and anabolic processes that create diverse lipid species. Lipids are transmitted into cells using FA transport or translocase proteins, including FATPs and CD36. Once cells take up lipids, fatty acids are transported into the mitochondria using membrane proteins. In the mitochondria, lipids are oxidized to produce acetyl-CoA, which is further used to make ATP. Glucose uptake through glucose transporters contributes to pyruvate and acetyl-CoA to support the TCA cycle in mitochondria. Acetate uptake using MCT is another source of acetyl-CoA in the cytosol. Acetyl-CoA can affect histone and protein acetylation for epigenetic alteration in the nucleus. In addition to fatty acid oxidation in mitochondria, VLCFAs and BRCFAs are oxidized in peroxisomes contributing to TCA metabolism. Citrate is synthesized during TCA cycling and exported from the mitochondria for de novo lipogenesis (FA and cholesterol synthesis). Fatty acids can be synthesized from malonyl-CoA by fatty acid synthase. Acetyl CoA and malonylCoA are used to produce palmitate and further elongate to MUFAs and PUFAs. Long-chain fatty acids are combined into triglyceride species and stored in lipid droplets. Palmitate is converted to CDP-DAG, DAG, and triglycerides. DAG is also used in the synthesis of phospholipids for membranes, with predominant species including PC, PE, PI, and PS in ER membrane, mitochondria, and cytosol. Lipid metabolism pathways intersect to coordinate cellular metabolic state. BRCFA, branched-chain fatty acid; CDP-DAG, cytidine diphosphate diacylglycerol; DAG, diacylglycerol; ETC, electron transport chain; FA, fatty acid; FATP, fatty acid transport protein; MUFAs, monounsaturated fatty acids; PA phosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PI, phosphatidylinositol; PS, phosphatidylserine; PUFA, polyunsaturated fatty acid; TCA cycle, tricarboxylic acid cycle; VLCFA, very-long-chain fatty acid.

While all lipids are insoluble in water, broad categories help classify lipids as fatty acids (FA), phospholipids, or neutral lipids (triglycerides and cholesteryl esters) (Mutlu et al., 2021). FAs, the building blocks of all lipids, serve as a primer for the synthesis of other lipids including glycerolipids, glycerophospholipids, sphingolipids, sterols, and saccharolipids (de Carvalho and Caramujo, 2018). Imbalances between FA uptake and oxidation lead to the accumulation of long-chain FAs that are incorporated into triglycerides (TG) and phospholipids as well as into other lipid species (Goldberg et al., 2012). Ceramides, diacylglycerols, and acylcarnitines, all regulators of intracellular signaling cascades and metabolism (Itani et al., 2002; Koves et al., 2008), are largely considered to be toxic signaling lipid species (Goldberg et al., 2012). Defective mitochondrial FA oxidation increases medium-chain acyl carnitines, another toxic species (Wajner and Amaral, 2015). Studying the diverse roles played by lipids, especially in the context of metabolic disease and cancer, offers an entry point for understanding lipid-mediated toxicity.

FATTY ACID METABOLISM

FA Uptake

Cellular uptake of fatty acids is a key component of metabolic regulation. While FAs can diffuse across phospholipid bilayers, much of fatty acid uptake in mammalian cells is facilitated by integral or membrane associated proteins. Several transporters, across multiple classes, mediate cellular fatty acid uptake including the scavenger receptor CD36 (fatty acid translocase, FAT), plasma membrane fatty acid-binding protein (FABPpm), and six fatty acid transport proteins (FATPs, solute carrier family SLC27A1–6) (Kazantzis and Stahl, 2012; Schwenk et al., 2010; Stahl et al., 2001; Su and Abumrad, 2009). Once at the inner side of the membrane, fatty acids are bound by cytoplasmic FABP (FABPc) before entering metabolic or signaling pathways. Interestingly, a series of studies have shown that FABPpm and mitochondrial aspartate aminotransferase (mAspAt) are identical proteins involved in amino acid metabolism (Birsoy et al., 2015; Cechetto et al., 2002). Additionally, fatty acids are activated by a set of acyl coenzyme A (CoA) synthetase (ACS) enzymes, which catalyze the activation of free fatty acids (FAs) to CoA esters (Roche et al., 2013). CoA conjugation contributes to the maintenance of the concentration gradient by directly pulling fatty acids into the cell. Furthermore, the FATPs are a group of membrane proteins that facilitate the import of long-chain fatty acids (LCFAs), and use ACS activity to regulate intracellular polyunsaturated fatty acids (Coe et al., 1999). Among the six FATP/SLC27A family members, overexpression of FATP1 in 3T3-L1 cells, a mouse embryonic fibroblast cell line that can differentiate into adipocyte-like cells, results in the internalization of palmitic acid (PA), oleic acid (OA) and arachidonic acid (AA) without any selective preference for these fatty acids (Schaffer and Lodish, 1994). Subcellular fractionation indicates that FATP is localized to the plasma membrane and transports LCFAs into the cell for use as an energy substrate.

FA Synthesis

FA synthesis is an anabolic process that creates diverse lipid species. The multifunctional enzyme fatty acid synthase (FASN) directly converts dietary carbohydrates into long-chain saturated fatty acids, predominately the 16-carbon palmitate, by using acetyl-CoA as a primer (Figure 2). FASN is used to supply additional lipids, to support membrane structure, and for cytosolic signaling. Several metabolic enzymes are involved in the conversion of carbons from citrate in the citric acid cycle (TCA) to bioactive fatty acids. ATP citrate lyase (ACLY) generates acetyl-CoA, a precursor for FA synthesis, from mitochondrial TCA-generated citrate in the cytosol (Zaidi et al., 2012). Additional molecular components include acetyl-CoA carboxylases (ACCs) which generate malonyl-CoA. Malonyl-CoA decarboxylase (MCD) converts malonyl-CoA to acetyl-CoA, reversing the reaction catalyzed by ACC (Zhou et al., 2009). The serial condensation of seven malonyl-CoA molecules and one priming acetyl-CoA by FASN generates palmitate, the initial product of FA synthesis. This 16‑ carbon saturated FA (16:0) is then activated by fatty acid-CoA ligase (ACS), elongated by fatty acid protein 5 (ELOVL5), and desaturated by stearoyl-CoA desaturase (SCD) and fatty acid desaturase 2 (FADS2) to produce molecules of various lengths and degrees of saturation (Bogie et al., 2020; Jakobsson et al., 2006). Synthesized fat is stored as triglycerides in cells. Diglyceride acyltransferase, DGAT, involves the TG synthesis pathway to convert diacylglycerol (DAG) to triacylglycerols (TAG). DGAT enzymes catalyze the final step in the known pathways of triglyceride synthesis. Although the 2 enzymes are dissimilar in protein sequences, both enzymes use fatty acyl CoA substrates (Stone et al., 2006). TGs synthesized by DGAT enzymes are then either stored in cytosolic lipid droplets or in other organs such as the liver and small intestine where they are secreted as components of lipoproteins. Both DGAT enzymes are universally expressed in tissues, and highly expressed in organs associated with TG metabolism including adipose tissue and the liver (Cases et al., 1998).

FA synthesis enzymes are regulated at the transcriptional level by sterol regulatory element-binding protein 1 (SREBP-1) transcription factors (Dihingia et al., 2018). Recently, a genome-wide CRISPR screen systematically mapped genetic interactions (GIs) in human HAP1 cells (a near-haploid human cell line derived from chronic myelogenous leukemia (CML) to investigate how cells adapt to the loss of de novo fatty acid synthesis (Aregger et al., 2020). Cells carrying a loss-of-function mutation in FASN, whose product catalyses the formation of long-chain fatty acids, show a strong dependence on lipid uptake that is reflected in negative GIs with genes involved in the low-density lipoprotein receptor signaling pathway (Aregger et al., 2020). A previously unrecognized role emerged for C12orf49 in the regulation of exogenous lipid uptake through a sterol regulatory element binding protein, SREBF2. This study demonstrates how pooled genome-wide CRISPR screens can nominate new metabolic targets in human cells.

Whereas most normal cells preferentially use extracellular lipids for the synthesis of new structural lipids, cancer cells elevate de novo FA synthesis to sustain proliferation in a lipid-poor microenvironment without extracellular lipids (Röhrig and Schulze, 2016). SREBP increases phospholipid, TAG, and cholesterol synthesis to promote cancer cell survival and tumor growth (Griffiths et al., 2013; Lewis et al., 2015). Cancer progression is accelerated through SREBP-1 signaling where the RNA-binding protein LIN-28 accelerates de novo fatty acid synthesis and promotes the conversion from saturated to unsaturated fatty acids (Zhang et al., 2019). Together with essential FAs including linolenic acid taken up through the diet, they form a complex collection of substrates to synthesize FA-containing lipids (Figure 2). Working in concert with membrane receptor tyrosine kinase (RTKs) and serine/threonine kinase mTOR, FASN regulates survival signaling by providing second messenger signaling lipids (Röhrig and Schulze, 2016). As a consequence, de novo fatty acid synthesis generates diverse lipids involved in regulating cellular signaling and lipid homeostasis. For a comprehensive review of fatty acid dysregulation in cancer cells, we direct readers to a recent review on the subject (Broadfield et al., 2021).

FA Regulation

Fatty acids interact with diverse metabolic enzymes to become incorporated into complex lipid species, including DAGs and TAGs or to be converted into phosphoglycerides, such as phosphatidic acid (PA), phosphatidylethanolamine (PE), and phosphatidylserine (PS) (Fagone and Jackowski, 2009; Koundouros and Poulogiannis, 2020)(Figure 2). Acyl groups of fatty acids - predominantly stearoyl groups in mammalian cells - can determine the diversity of phosphatidylcholine (PtdCho) and phosphatidylinositol (PtdIns) (Anaokar et al., 2019; de Carvalho and Caramujo, 2018; Vance, 2014). PtdIns are among the best-characterized secondary messengers in signal transduction pathways (Cantley, 2002). PtdIns can be converted to several phosphoinositide species by phosphorylation, containing phosphatidylinositol 4,5-bisphosphate and phosphatidylinositol (3,4,5)-trisphosphate (PIP2/3) (Cantley, 2002). PIP3 activates AKT to induce pro-tumorigenic signaling, through phosphoinositide-dependent kinase 1 (PDK1), tuberous sclerosis complex (TSC) 1/2, and mTORC2 (Koundouros and Poulogiannis, 2020). Moreover, fatty acids can be used for ceramide de novo synthesis in the endoplasmic reticulum (ER). The initial step is the condensation of the activated C16 fatty acid palmitoyl-CoA and the amino acid L-serine, which is catalyzed by pyridoxal 5’-phosphate (PLP)-dependent serine palmitoyltransferase (SPT). This produces 3-ketosphinganine (3KS), which can be rapidly reduced to sphinganine (dihydrosphingosine, d18:0 Sph) by 3-ketosphinganine reductase (KDSR) in a NADPH dependent manner (Wigger et al., 2019).

Fatty acid synthesis is activated by hypoxia-inducible factor (HIF) signaling (Wagner et al., 2017). Carnitine palmitoyltransferase 1 (CPT1) is repressed by HIF, reducing fatty acid transport into the mitochondria, and directing fatty acids to lipid droplets for storage (Du et al., 2017). Both HIF-1α and HIF-2α are upregulated upon ER stress (Pereira et al., 2014), leading to the formation of lipid droplets in an attempt to decrease cytotoxic ER stress responses (Qiu et al., 2015). An increase in lipid droplets in cells is commonly associated with lipotoxicity and altered metabolism that contributes to cellular dysfunction. Lipid droplet composition and catabolism is a key regulatory node that integrates physiological inputs, such as dietary lipids and lipolytic stimuli, to coordinate cellular signaling and metabolism. Lipid droplets maintain lipid homoeostasis, prevent lipotoxicity, and generate ATP by breaking down lipids stored in droplets during conditions of metabolic stress (Olzmann and Carvalho, 2019). Moreover, repression of SREBP or limitation of FASN can also trigger the HIF-1α signaling pathway and the UPR (Griffiths et al., 2013). In the context of energy-deficiency mediated stress, HIF signaling pathways coordinate with AMP-activated protein kinase (AMPK) and mTOR to compensate for the limitation of FASN and activate lipid metabolism to rescue lipid-mediated ER stress.

FASN antagonists are increasingly being investigated as a therapeutic approach to treat cancer. The use of techniques like MALDI-MS/MS, liquid chromatography-matrix assisted laser desorption/ionization mass spectrometry, enable greater proteome analysis (Mueller et al., 2007). Applications of mass spectrometry indicate that FASN-inhibitors such as C75 and G28UCM increase polyunsaturated fatty acids and decrease signaling lipids like DAG and PIP3 in ovarian cancer cell lines (Wagner et al., 2017). FASN inhibition influences multiple downstream targets suggesting that a greater understanding of pathway cross-talk will enhance drug target efficacy by influencing activity across multiple pathways. Specifically, FASN inhibition affects ERBB-PI3K-mTORC1 activity by blocking phosphorylation of EGF-receptor/ERBB/HER, inhibiting GRB2-EGF-receptor recruitment, and suppressing PI3K-AKT signaling (Giró-Perafita et al., 2016; Kumar-Sinha et al., 2003; Menendez et al., 2004; Wagner et al., 2017). Moreover, fatty acid synthesis is elevated in metastatic breast cancer, especially in the brain (Ferraro et al., 2021). This phenotype is an adaptation to decreased lipid availability in the brain compared to other tissues, resulting in site-specific dependency on fatty acid synthesis for breast tumors growing in the brain. Inhibition of fatty acid synthase reduces human EGF receptor 2-positive breast tumor growth in the brain, pointing to the emergence of new cancer targets based on differential nutrient availability across metastatic sites (Ferraro et al., 2021). FABP5 is an intracellular chaperone that delivers cytosolic fatty acids to nuclear receptors to enhance metastasis. FASN and monoacylglycerol lipase (MAGL) promote nuclear receptor activation and PCa metastasis are critically dependent upon co-expression of FABP5 (Carbonetti et al., 2019). Moreover, the expression level of FASN affects the PI3/AKT signaling through phosphatase and tensin homolog deleted on chromosome 10 (PTEN) (Van de Sande et al., 2002). A greater understanding of how these pathways intersect will enhance the ability to design effective FASN inhibitors that regulate multiple, interconnected targets.

In addition to HIF-dependent pathways, fatty acid synthesis is regulated by other metabolic enzymes. Lipid biosynthesis and oxidation is regulated by a master regulator of FA metabolism, acetyl-CoA carboxylase (ACC), which converts acetyl-CoA to malonyl-CoA. This conversion serves as a precursor for fat synthesis and inhibits fatty acid oxidation. ACC1 is localized in the cytosol and promotes the production of FAs, while ACC2 is localized to the mitochondrial outer membrane and generates malonyl-CoA to inhibit the fatty acid transport protein CPT1 (German et al., 2016). Cellular stress directly affects enzymatic actions related to ACC and regulates FA metabolism. In energy stress conditions, AMPK activates fat synthesis and catabolism by inhibiting both ACCs (Park et al., 2002). In contrast, under conditions of nutrient abundance, AMPK is downregulated and no longer represses ACC1 and ACC2. Moreover, ACC2 is regulated by post-translational modifications. For example, prolyl-hydroxylase 3 protein (PHD3) presents a metabolic barrier to fatty acid utilization by hydroxylating and activating ACC2 (German et al., 2016; Yoon et al., 2020). ACC is therefore a signaling node that senses nutrient abundance and adjusts anabolic versus catabolic FA metabolism accordingly. In the setting of acute myeloid leukemia, PHD3 levels are decreased, fueling a dependence on fats that can be targeted with fatty acid oxidation (FAO) inhibitors. In addition to AML, many cancer cells use fat synthesis to induce proliferation. Questions remain regarding the extent to which modulating PHD3 in metabolic disorders can confer a therapeutic benefit.

LIPID METABOLISM IN MITOCHONDRIA AND PEROXISOMES

Fatty acid metabolism includes catabolic processes that generate energy, and anabolic processes that create diverse lipid species (DeBerardinis et al., 2007; Yoon et al., 2020). Mitochondria play an important role in the regulation of FA metabolism, including anabolic and catabolic pathways. Fatty acids provide twice as much ATP as carbohydrates and six times more when comparing stored fatty acids to stored glycogen (Carracedo et al., 2013). FAO occurs through a series of reactions that result in the shortening of fatty acids by two carbons per cycle. Each round generates NADH, FADH2, and acetyl CoA until the last cycle when two acetyl-CoA molecules are generated (Figure 2). The NADH and FADH2 that are generated by FAO enter the electron transport chain (ETC) in order to generate ATP. Fatty acid availability is a key signal for adaptations in mitochondria-rich muscle cells and their specific enzymes involved in lipid metabolism. Numerous studies underscore that an efficient capacity to oxidize fatty acids, and the ability to adapt fatty acid utilization to fatty acid availability, is of great importance for both lipid and glucose homeostasis and insulin action (Matoba et al., 2017; Santoro et al., 2021; Zhou et al., 2019).

β-oxidation of stored lipids leads to the production of acetyl-CoA through oxidative degradation of FAs (Figure 2). The acetyl-CoA produced from each round of β-oxidation can subsequently enter the TCA cycle to generate NADH and FADH2 for the electron transport chain (Fritz and McEWEN, 1959). Once cells take up lipids, FAs are transported into the mitochondria and oxidized in a multi-step pathway known as beta-oxidation (McGarry and Foster, 1980). As long-chain fatty acids are prepared for the multi-step process of mitochondrial import, they are regulated by CPT1, which is a transferase that converts acyl-CoAs into acyl-carnitines for transport. These FAs are processed by acyl-CoA dehydrogenase to form long-chain acyl-CoA, enoyl-CoA hydratase to acyl-chain forming hydroxy-acyl-CoA, by hydroxy-acyl-CoA dehydrogenase to the substrate forming a second keto-group, and by thiolase to acetyl-CoA and a free CoA with the new substrate. Using these processes, FAs are oxidized into acetyl-CoA, which is subsequently used to make ATP (Huynh et al., 2014). Cells containing increased FAO through metabolic alteration use acyl-CoA for producing ATP through the TCA cycle. To compensate for the acyl-CoA, levels of acyl-carnitine are increased (Yoon et al., 2020). Cancer cells exhibit a decrease in ATP and NADPH, due to decreased flux through the pentose phosphate pathway (PPP), which inhibits glycolysis and leads to elevated levels of reactive oxygen species (ROS) that repress FAO activity (Schafer et al., 2009). In conclusion, this crosstalk between FAO and metabolic signaling pathways includes redox systems and directly impacts cell survival, but questions remain as to which lipid species are specifically participating in FAO and the detailed mechanism of how FAO induces cell survival.

Peroxisomes regulate shortening of long-chain and very-long-chain fatty acyl-CoAs, dicarboxylic fatty acids, 2-methyl-branched fatty acids, eicosanoid inflammatory mediators, prostaglandins, and bile acid intermediates. Moreover, fatty acid oxidation of very long chain fatty acids or branched fatty acids occurs in peroxisomes. This process generates hydrogen peroxide. Peroxisomal β-oxidation does not degrade fatty acids completely as it is not coupled to oxidative phosphorylation for ATP synthesis. Peroxisome proliferator–activated receptors (PPAR) are the most important transcriptional regulators of peroxisomal β-oxidation (Reddy and Hashimoto, 2001; Vanhove et al., 1993). Recently, acyl-CoA oxidase 1 (Acox1) has been reported to activate peroxisome-derived acetyl-CoA to increase peroxisomal β-oxidation. The induction of cytosolic acetyl-CoA levels activates mTORC1, inhibits autophagy, and induces hepatic triglycerides (He et al., 2020). To date, the physiological significance of peroxisomal beta-oxidation is still an open question and the subject of active investigation.

LIPID-MEDIATED MODULATION OF CHROMATIN STATE

Lipids modulate chromatin states through histone and protein acetylation by generating acetyl-CoAs, which are generated from mitochondria and peroxisomes (Galdieri et al., 2013; McDonnell et al., 2016). Lipids can provide up to 90% of acetyl-carbon for histone acetylation using lipid-derived acetyl-CoA (Ac-CoA). We hypothesize that metabolites like Ac-CoA directly affect histone modification by regulating gene expression specific to lipid homeostasis and control of lipotoxicity. Understanding how Ac-CoA regulates gene expression specific to lipid homeostasis will provide important insights into the genes important for homeostatic or lipotoxic programs. The Ac-CoA pool regulates glucose which in turn drives a gene expression program characterized by activating genes involved in its metabolism, in part by increasing glucose-derived histone acetylation. Lipid-derived acetyl-CoA is a major source of carbon for histone acetylation (Galdieri et al., 2013; McDonnell et al., 2016). Using ¹³C-carbon tracing combined with acetyl-proteomics, up to 90% of acetylation on certain histone lysines can be derived from fatty acid carbon, even in the presence of excess glucose (Galdieri et al., 2013; McDonnell et al., 2016). This suggests a new mechanism for how Acetyl-CoA fluxes could regulate genes important for homeostatic/lipotoxic programs.

Free acetate is converted to Ac-CoA by acetyl-CoA synthetase (ACSS) and promotes lipid synthesis under hypoxic conditions through epigenetic reprogramming (Gao et al., 2016). ATP citrate lyase (ACLY) is the primary enzyme responsible for the synthesis of cytosolic acetyl-CoA in many tissues, and is critical for histone acetylation (Wellen et al., 2009). ACLY generates nucleus and cytosolic Ac-CoA by cleaving citrate derived from TCA cycle intermediates released from mitochondria (Figure 2). Although acetyl-CoA provided by ACLY activity regulates histone acetylation in adipocytes, other metabolites shunted from the TCA cycle also regulate the epigenome (Felix et al., 2021). For histone modification, α-KG is used as cofactor for 2-oxoglutarate dependent dioxygenases, such as Jumonji C domain-containing lysine histone demethylases. During this modification process, succinate is converted from α-KG. The ways in which energy balance impacts lipid synthesis is an ongoing area of investigation This in combination with the immune effects of acetate, succinate, and α-KG, suggest that nutrient metabolism must also match demand across a diverse mixture of adipocytes and stromal cells in WAT (Felix et al., 2021). ACLY is phosphorylated by AKT to induce histone acetylation, and pAkt(Ser473) levels correlate significantly with histone acetylation markers in human gliomas and prostate tumors (Lee et al., 2014). Moreover, histone methylation modifiers influence mono-unsaturated fatty acids (MUFA) metabolism (Han et al., 2017). The MUFA oleic acid plays a key role in the longevity of H3K4me3 methyltransferase-deficient worms. The role of oleic acid in lifespan regulation in the context of histone modification suggests the importance of MUFAs and their downstream polyunsaturated fatty acids (PUFAs) in the regulation of lifespan under physiological conditions (Han et al., 2017). In sum, lipids take part in a complex, interconnected regulatory network that includes signaling, epigenetics, aging, and metabolism.

LIPOTOXICITY ACROSS DIVERSE ORGAN SYSTEMS

Impairments in fatty acid metabolism have significant consequences for a range of human diseases. The application of CRISPR-based genetic screens and unbiased lipidomics has identified a new approach to studying the enzymes responsible for regulating how fatty acids incorporate into membrane and storage glycerolipids. Lipid accumulation in tissues is increasingly recognized as a contributor to cellular dysfunction. Many cells across organ systems are not equipped to handle large lipid loads, and the mechanism by which excess lipids cause cellular injury, or lipotoxicity, is an area of investigation across the kidney, liver, heart, skeletal muscle, bone, pancreas, and brain (Figure 3). Saturated fatty acids are thought to be particularly harmful to cells invoking a diverse array of harmful cellular responses: apoptosis, inflammation, ceramides, reactive oxygen species (ROS), small nucleolar RNAs (Michel et al., 2011), and ER stress. An ongoing priority in the field is nominating bioactive lipid species that modulate the lipotoxic cellular response.

Figure 3. Lipid homeostasis and disease.

Under conditions of excess nutrients, there is less lipolysis, conversion of triglycerides to fatty acids, in favor of increased lipogenesis or triglyceride storage in adipose tissue. In muscle cells, glucose oxidation is increased. In fasting conditions, lipolysis is increased in adipose tissue and fatty acid oxidation increases in muscle cells to provide energy. Changes in metabolism occur in the brain, heart, liver, pancreas, kidney, and adipose tissue.

Lipid Accumulation in Adipose Tissue

Adipose tissue is a major regulator of energy homeostasis, and its dysregulation results in an imbalance in energy homeostasis due to inappropriate loads in peripheral tissues. Adipocytes act as a reservoir for energy storage, but also sense energy demands and secrete paracrine factors to regulate other metabolic tissues. Mammals have two types of adipose tissue: white adipose tissue (WAT) which stores excess energy, and brown adipose tissue (BAT) which releases excess energy as heat. A major function of WAT is the release of nonesterified fatty acids (NEFAs) into the bloodstream during periods of energy-demand. Recent studies have investigated which byproducts of lipid metabolism affect the function in adipocytes. Specifically, there is an emerging role for short-chain fatty acids (SCFAs) and TCA cycle metabolites that connect lipogenesis to WAT energy balance (Felix et al., 2021). SCFAs including acetate, butyrate, and propionate inhibit lipolysis and promote adipogenesis in WAT, and provide substrates for glucose and lipid synthesis. SCFAs act on G protein-coupled receptors (GPR41 and GPR43) to inhibit lipolysis and decrease plasma levels of FFAs (Felix et al., 2021). Applying highly-sensitive, mass-spectrometry-based proteomics to human adipocytes identified 471 secreted proteins including hormones, growth factors, extracellular matrix proteins that are differentially regulated between brown and white adipose tissue (Deshmukh et al., 2019). Interestingly, brown and white adipocytes have distinct secretory profiles and metabolic functions. Mammalian ependymin-related protein 1 (EPDR1) is selectively secreted from brown adipocytes where it plays a vital role in promoting the development into functional thermogenic adipocytes by activating UCP1 expression. Thus, this recent profiling of the secretome of human white adipocytes and energy-burning brown adipocytes identified important regulators of human metabolism. Since this secretome analysis was done in mature adipocytes, questions remain about the dynamic regulation of lipid metabolism and secretome throughout the differentiation process. Studies conducted at different differentiation stages in fat cells would provide additional insights into the mechanism responsible for preserved metabolic health in people with obesity, severe insulin resistance, and type 2 diabetes.

Lipid Accumulation in Kidney

Extensive work in animal models has demonstrated a link between kidney dysfunction and lipid accumulation in models of metabolic disease including obesity, metabolic syndrome, diabetes mellitus, chronic kidney disease, and acute kidney injury (Jiang et al., 2005; Kume et al., 2007; Wang et al., 2005). Lipid accumulation and kidney dysfunction have been widely documented in human clinical studies including focal segmental glomerulosclerosis (FSGS), minimal change disease, Fabry’s disease, and lipoprotein glomerulopathy (Bobulescu, 2010). The kidney can use multiple substrates as fuel, depending on availability (Elhamri et al., 1993; Guder et al., 1986; Klein et al., 1981). Substrate use varies across regions on the basis of energy demand (Bobulescu, 2010). The proximal tubules have a high energy demand, second only to cardiac myocytes. Therefore, they have relatively little glycolytic capacity and rely instead on mitochondrial β-oxidation of FFAs to maximize ATP production (Balaban and Mandel, 1988; Gullans et al., 1984; Uchida and Endou, 1988). Recent evidence indicates that during diabetic kidney disease, the proximal tubule expresses kidney injury molecule (KIM)-1. KIM-1 mediates uptake of palmitic acid leading to enhanced tubule injury characterized by DNA damage, interstitial inflammation, and fibrosis (Mori et al., 2021). A small molecule inhibitor of KIM-1, TW-37, ameliorates kidney inflammation and fibrosis. These studies highlight that small molecule targets upstream of FAO can be a novel therapeutic strategy in kidney injury.

The kidney responds to lipid toxicity through the upregulation of regulators involved in lipid peroxidation and the accumulation of toxic metabolites including fatty acyl CoA, diacylglycerol, and ceramides. Lipotoxic cellular dysfunction results in the generation of reactive oxygen species, organelle damage, disruption of intracellular signaling pathways, release of proinflammatory and pro-fibrotic factors, and lipid-induced apoptosis. Lipid peroxidation occurs when oxidants such as free radicals attack lipids containing carbon-carbon double bonds, especially PUFAs. Even though many studies have shown that PUFAs reduce kidney disease by decreasing triglycerides and inflammation, it is hypothesized that PUFAs are converted to oxidized lipid by lipoxygenases (LOX), cyclooxygenases (COX), and cytochrome P450 (CYP) (Hajeyah et al., 2020). The kidney, in particular, is susceptible to changes in gene expression in sterol regulatory element-binding proteins (SREBPs) in response to diabetes, and results in TG accumulation, mesangial expansion, and glomerulosclerosis (Sun et al., 2002). This suggests that activation of renal SREBP-1 results in alterations in renal lipid metabolism and renal lipid accumulation plays an important role in the pathogenesis of diabetic nephropathy.

Lipid accumulation is a major contributor to diabetic kidney disease, the most rapidly growing cause of kidney failure worldwide (Alicic et al., 2017). As essential components of the kidney filter, podocytes are post-mitotic, highly-differentiated epithelial cells that are particularly vulnerable to lipid accumulation and toxicity (D’Agati et al., 2016). Coenzyme Q10 (CoQ), a ubiquitous lipid present in all cellular membranes, protects against polyunsaturated fatty acid-mediated (PUFA-mediated) lipid peroxidation (Sidhom et al., 2021). With mitochondrial dysfunction, the absence of protection from lipid peroxidation sensitizes cells to death (To et al., 2019). Elevation of lipid peroxidation, shown by upregulated glutathione peroxidase 4 (GPX4), has been reported in podocytes of CoQ deficient mice (Sidhom et al., 2021). Moreover, loss of GPX4 triggers ferroptosis death in the kidney resulting in renal degeneration (Angeli et al., 2014). Recent efforts to understand the connection between kidney disease, PUFAs, and dysregulated pathways in podocytes used single-nucleus RNA-Seq (sNuc-Seq) and integrated metabolomics and transcriptomics to identify a therapeutically relevant Braf/MAPK pathway (Sidhom et al., 2021). In addition, JAML (junctional adhesion molecule-like protein) is expressed in podocytes and induced under diabetic conditions (Fu et al., 2020). Podocyte-specific deletion of Jaml ameliorates podocyte injury and proteinuria in two different models of diabetic mice (Fu et al., 2020). Junctional adhesion molecules, members of an immunoglobulin subfamily, play an emerging role in lipid metabolism. Specifically, Jam-A knockout mice fed a high-saturated fat, fructose, and cholesterol diet (HFCD) develop severe non-alcoholic steatohepatitis (Rahman et al., 2016). Deploying LC/MS-based lipidomics analysis revealed that JAML deletion in podocytes reduces levels of lipids including free fatty acids, cholesteryl esters, and phosphatidylcholines (Fu et al., 2020). JAML regulates podocyte lipid metabolism through SIRT1-mediated SREBP1 signaling, and is higher in the glomeruli of patients with kidney disease. In the future, clinical studies aimed at preventing lipid accumulation and preserving glomerular function may be an attractive therapeutic target for diabetic kidney disease and other types of proteinuric kidney diseases.

Roles of Liver, Bone, and Skeletal Muscle in Lipotoxicity

Fatty acids are delivered to the liver through the blood following lipolysis of triglycerides in adipose tissue. Fatty acids in the liver bind to FABP-1 and are metabolized by mitochondrial β-oxidation. Obesity and type 2 diabetes are frequently complicated by excess fat accumulation in the liver, which is known as nonalcoholic fatty liver disease (NAFLD). The major genetic determinants of NAFLD are PNPLA3, HSD17B13, and TM6SF2, and liver steatosis develops due to the dysregulation of pathways controlling de novo lipogenesis and fat catabolism. Recent evidence suggests that reduction in the activity of lysosomal acid lipase (LAL), which is a key enzyme for intracellular fat disposal, is of clinical relevance for patients with NAFLD (Baratta et al., 2019). With the advance of high-throughput sequencing technology, liver transcriptome sequencing results have identified potential gene candidates affecting fat deposition. FABP1 is a liver-specific FABP that plays important roles in intracellular lipid metabolism in the liver. Knockdown of FABP1 blocks lipid accumulation in hepatocytes (Mukai et al., 2017). FABP1 affects the regulation of fat deposition through PPAR signaling and biosynthesis of fatty acids (Wang et al., 2019). Palmitic acid hydroxystearic acids (PAHSAs) are endogenous lipids with anti-diabetic and anti-inflammatory effects. Chronic PAHSA treatment augments insulin-stimulated glucose uptake in glycolytic muscle and heart in high-fat diet-fed mice by enhancing hepatic insulin sensitivity and inhibiting lipolysis in adipose tissue (Zhou et al., 2019). Moreover, PAHSAs mediate GPR40 receptors to regulate improvements in glucose tolerance and insulin sensitivity (Syed et al., 2018).

Beyond the systems associated with metabolism, bone has an important role in the clearance of circulating lipoproteins and non-esterified fatty acids (Kushwaha et al., 2018). Interestingly, long-chain fatty acid oxidation affects postnatal bone development by altering fatty acid utilization. Eicosapentaenoic acid (EPA, long-chain polyunsaturated n-3 fatty acids) affects substrate cycling in human skeletal muscle cells by altering lipolysis rate of intracellular triacylglycerol and re-esterification of fatty acids by increasing fatty acid turnover (Løvsletten et al., 2018). In the future, further studies of how ER stress and the UPR pathways underlie lipotoxicity in peripheral tissues may provide an important point of therapeutic intervention for tissue damage.

Lipids and Cardiovascular Function

The heart has both the greatest caloric need and the most extensive oxidation of fatty acids (Goldberg et al., 2012) and adeptly acquires lipids from circulating, non-esterified fatty acids and esterified FAs bound to lipoproteins. Lipid energy metabolism is an important factor for heart disease including heart failure and ischemia. Specifically, extensive clinical evidence links lipid oxidation and the inflammatory response to cardiovascular diseases. Polyunsaturated fatty acids (PUFAs) affect the levels of phospholipids and cholesterol esters in lipoproteins during the development of atherosclerosis (Berliner et al., 2009). Free radical lipids and modified lipoproteins generated from oxidized lipid peroxidation play a key role in modulating inflammatory responses (Binder et al., 2016; Hansson and Hermansson, 2011). Higher lipid availability promotes ischemia-induced cardiac dysfunction and decreases myocardial mitochondrial efficiency. Myocardial fatty acid-linked respiration and oxidative stress are increased, whereas mitochondrial efficiency is decreased. Increased lipid availability favors susceptibility to ischemia-induced cardiac dysfunction (Jelenik et al., 2018).

Cell death occurs through different mechanisms, and ferroptosis, a programmed iron-dependent cell death, is driven by damage to the lipid membrane in ischemia/reperfusion-induced cardiomyopathy as well as peroxidation of lipids. Recent studies have suggested that free iron accumulates in mitochondria to cause oxidative stress and ferroptosis-induced heart damage (Fang et al., 2019). The glutathione metabolic pathway and reactive oxygen species (ROS) pathway are significantly downregulated during myocardial infarction (Park et al., 2019). Interestingly, GPX4, which protects cells from ferroptosis, is downregulated in myocardial infarction. Moreover, oxidized phospholipids promote inflammation in global myocardial ischemia/reperfusion injury. Cytokine IL-10 plays an anti-inflammatory role and modulates the production of oxidized phosphatidylcholines in cardiomyocytes thereby mitigating inflammation and cell death (Bagchi et al., 2020).

Lipid levels and composition in patient blood during myocardial infarction have been shown to predict the risk of complications (Meeusen et al., 2017). Specifically, sphingolipids serve as a biomarker for both recurrence and mortality after myocardial infarction (MI) (Hadas Yoav et al., 2020). Ceramides are simple membrane sphingolipids that form the backbone of all complex sphingolipids and can trigger programmed cell death upon reaching high cellular levels (Arana et al., 2010). Studies have shown that ceramide levels are high in the heart tissues of humans during acute MI (Hadas Yoav et al., 2020). 24 hours post-MI, 30% of sphingolipid metabolism genes are significantly upregulated and the levels of C16-ceramide, C20-ceramide, C20:1-ceramide, and C24-ceramide are significantly higher (Hadas Yoav et al., 2020). In hypoxic conditions that mimic myocardial infarction, several inhibitors limit ceramide degradation including the pan-ceramidase inhibitor B13 and the acid ceramidase (AC) specific inhibitor ARN14974. Additionally, a pan-sphingosine kinase inhibitor SK1-II significantly increases cardiomyocyte cell death levels (Hadas Yoav et al., 2020). Alterations in sphingolipid metabolism by ceramidase, which hydrolyzes proapoptotic ceramide and generates sphingosine, is necessary for regulating ceramide levels and cell survival in ischemic heart disease (Hadas Yoav et al., 2020). Transcriptomic and protein analyses reveal that altering ceramide metabolism through chemical inhibitor modulation of sphingolipid metabolism can induce cardioprotection after MI. Furthermore, the expression of microsomal triglyceride transport protein (MTTP) is associated with structural and perfusion abnormalities in patients with ischemic heart disease, suggesting that triglycerides play an important role in cardiac function as it relates to ischemic events (Klevstig et al., 2019).

Decreases in the level of FA oxidation are commonly associated with heart failure (Neubauer, 2007). Genetic overexpression of PPARα in the heart recapitulates the phenotype of lipotoxic cardiomyopathy (Finck et al., 2002) whereas knockdown attenuates the phenotype (Finck et al., 2003). Questions related to PPARα activity are emerging including (1) how is endogenous PPARα regulated in the context of diabetes and obesity, and (2) how does this modification contribute to lipotoxicity? PPARα is a nuclear receptor transcription factor known to play a role in controlling the expression of genes involved in FA metabolism, specifically, uptake, storage, and oxidation. Since saturated FAs including palmitic acid, monounsaturated FAs (such as oleic acid), and polyunsaturated FAs (such as linoleic acid) have distinct effects on metabolic diseases (Roberts et al., 2014), the effect of different FAs on the GSK-3α-PPARα signaling pathways is a current area of interest. Specifically, palmitic acid increases PPARα activity; however, knockdown of GSK-3α abolishes the palmitic-acid induced increase (Nakamura et al., 2019). The functional role of PPARα phosphorylation is an effect on energy metabolism. Specifically, PPARα-S280D increased expression of genes related to FA uptake and pyruvate dehydrogenase PDH kinase 4 (PDK4) which inactivates the PDH complex and inhibits glucose oxidation. Overall, this study suggests that, in the heart, FA exposure increases GSK-3α activity as part of a feedforward axis with PPARα that induces lipotoxic cardiomyopathy in obesity. Importantly, constitutively active GSK-3α and GSK-3β exert opposite effects on genes involved in FA uptake and transport. As a result, the therapeutic potential for GSK-3α inhibitors will depend on an isoform specific small-molecule inhibitor as the beneficial effects of GSK-3α inhibition could be negated by unintended GSK-3β inhibition.

Aberrant Lipid Metabolism in Neurodegenerative Diseases

The central nervous system regulates systemic metabolism and lipid balance. Highly coordinated interactions between the brain and metabolic organs maintain energy and glucose homeostasis. The energy state of the body is assessed by sensing regions of the brain including key nuclei within the hypothalamus, including the ventromedial nucleus (VMH), arcuate nucleus (ARC), dorsomedial hypothalamic nucleus (DMH), and the paraventricular nucleus. Alterations in plasma levels of key nutrients, including glucose, fatty acids, and amino acids, provide information about nutrient availability. Increasing malonyl CoA content in hypothalamic neurons acts as a fuel gauge: adding the fatty acid synthase inhibitor, C75, to hypothalamic neurons, decreases food intake. Additionally, increasing long-chain fatty acyl-CoA (LCFA-CoA) content in hypothalamic neurons decreases food intake through hypothalamic inhibition of carnitine palmitoyltransferase-1 (Gao et al., 2013; Obici et al., 2003).

Neurons do not typically make lipid droplets and have a low capacity for fatty acid consumption in mitochondria for energy production (Schönfeld and Reiser, 2013). When neurons undergo periods of sustained activity, high levels of reactive oxygen species induce peroxidation of FAs (Reynolds and Hastings, 1995). Highly active neurons are particularly susceptible to peroxidated FAs as they do not have the ability to divert them into lipid droplets. In contrast, lipid droplet accumulation in glia has been demonstrated to protect neurons (Bailey et al., 2015). In the brain, hyperactivity in neurons results in the production of toxic fatty acids. The transfer of these toxic fatty acids via lipid particles to astrocytes provides a mechanism of detoxification, especially during periods of enhanced activity (Ioannou et al., 2019). Astrocytes endocytose neuron-derived lipid particles to deliver the fatty acid to lipid droplets. Thus, activity-dependent stimulation of lipid metabolism in astrocytes represents a dynamic process that regulates prevention of FA toxicity in neurons.

In contrast to neurons, astrocytes make LDs and produce antioxidants. Oxidative stress in neurons induces lipid droplet formation in neighboring astrocytes. Glial and neuronal monocarboxylate transporters (MCTs), fatty acid transport proteins (FATPs), and apolipoproteins are important for this type of lipid droplet formation (Liu et al., 2017a). MCTs enable glia to secrete lactate which is converted to pyruvate and acetyl-CoA in neurons. Lactate metabolites provide a substrate for fatty acid synthesis. In the presence of elevated levels of ROS, inhibiting lactate transfer or lowering FATP or apolipoprotein levels has been shown to decrease glial lipid droplet accumulation (Liu et al., 2017a).

Dysregulation of lipids has recently emerged as a key factor in neurodegenerative diseases including Alzheimer’s disease (AD). Many lipid species have been used as markers for early diagnosis of AD and identified as playing a role in neurotoxicity. The most validated genetic risk factor for late-onset Alzheimer’s disease is the 4 allele of the APOE gene (APOE4). The presence of APOE4 lowers the age of AD onset. Apolipoprotein E (APOE) is a component of many lipoprotein particles and acts as a ligand for membrane receptors that mediate lipoprotein uptake. Recent efforts to shed light on how APOE4 alters the composition of lipids have demonstrated that human iPSC-derived APOE4 astrocytes accumulate unsaturated triacylglycerols stored in lipid droplets to a greater extent than isogenic APOE3 counterparts. Using liquid chromatography-mass spectrometry (LC-MS), the authors compared the lipid composition of APOE3- and APOE4-expressing human iPSC-derived astrocytes and observed that the degree of unsaturation of the fatty acids attached to triacylglycerides in APOE4-expressing astrocytes was higher (Sienski et al., 2021). To determine if the iPSC-derived human astrocyte cultures reflected APOE4-related dysfunction in human brains, the authors examined transcriptomic data from postmortem human brain samples using the Genotype-Tissue Expression (GTEx) project. Genes that emerged as significantly differentially expressed and of relevance to lipid metabolism included up-regulated genes involved in the metabolism of neutral lipids (FA2H, ACSL1, SQLE, HMCGR, and MVK) and downregulated genes involved in the metabolism of fatty acids and neutral lipids (OLAH, CNEP1R1, and GPAM). It remains to be seen whether a clinical intervention that takes advantage of genotype-specific dietary supplementation might mitigate disease progression.

LIPID SIGNALING AND THE IMMUNE SYSTEM

Lipid-Immune Interactions

Different fatty acids and lipids differentially influence immune cell subsets. Omega-3 PUFAs possess potent immunomodulatory activities and play a role in regulating inflammatory and autoimmune diseases including arthritis, Crohn’s disease, ulcerative colitis and lupus erythematosus (Simopoulos, 2002). PUFAs suppress the production of interleukin 1 (IL-1) and the expression of Cyclooxygenase (COX) 2 mRNA that is induced by IL-1. Moreover, α-linolenic acid (ALA), the precursor to the omega-3 family compounds, increases pro-inflammatory cytokine secretion (IL-1, IL-2, and tumor necrosis factor α). In addition to these mediators with pro-inflammatory effects, ALA has been known to suppress prostaglandins and leukotrienes and induce anti-inflammatory and resolvin functions (Marton et al., 2019; Simopoulos, 2002). PUFAs and related FAs represent a potential target for diseases of inflammation.

The immune system plays a critical role in removing cancerous cells. The cross-talk between immune cells and cancer is demonstrated by the ways in which tumor-infiltrating T lymphocytes (TILs) adapt to the metabolic constraints within the tumor microenvironment (TME) introducing a route to combat tumor progression. T cells can exhibit some degree of metabolic flexibility. Since CD36, in the plasma membrane, is essential for facilitating exogenous FA uptake, the upregulation of CD36 in intratumoral T_reg cells has an important role in tumor progression and T cell function (Wang et al., 2020). In TMEs with low levels of glucose, CD8⁺ T cells enhance PPAR-alpha signaling and fatty acid catabolism under hypoglycemic and hypoxic conditions to partially preserve effector functions. Importantly, metabolic reprogramming of T cells using a PPAR-alpha agonist inhibits tumor growth, an effect that is enhanced in combination with PD-1 inhibition (Zhang et al., 2017). As FAO is important for the differentiation of Tregs, FAO inhibition could prevent the accumulation of this immunosuppressive T cell population (Bader et al., 2020). These studies suggest that lipid metabolism provides intriguing opportunities to modulate the TME and specific immune cell populations.

Beyond T cells, metabolic rewiring has been linked to a protumoural phenotype in tumor associated macrophages. Alternative (M2) activation of macrophages is dependent on fat oxidation. The triacylglycerol substrates are taken up through the scavenger receptor CD36, and their lipolysis by lysosomal acid lipase (LAL) leads to prolonged survival and M2 activation in ovarian cancer, suggesting that CD36 inhibition is an important strategy for combating cancer (Huang et al., 2014). Moreover, the production of a-ketoglutarate from glutaminolysis promotes fatty acid oxidation and epigenetic activation in alternative (M2) activation of macrophages (Liu et al., 2017b). This study suggests that macrophage responses are fine-tuned through FA metabolism and epigenetic reprogramming. Metabolic adaptations in tumor and immune cells within the tumor microenvironment likely occur in response to changes in local nutrient levels, and we will discuss how this lipid metabolism affects cancer biology.

Fat Metabolism in Cancer

Different lipid species have opposing effects on cancer proliferation and death requiring careful mechanistic interrogation. Cellular proliferation is a common feature of all cancers which require fatty acids in order to synthesize membranes and signaling molecules. Tumor cells acquire abundant lipids for rapid cell growth, but despite this seeming overload of FAs, tumors avoid toxicity. Indeed, highly proliferative cancer cells have been found to upregulate enzymes involved in lipid and cholesterol biosynthesis leading to increased aggressiveness in certain cancers (Bader et al., 2020). How cancers metabolize fat is an emerging area of investigation. Numerous tumors forgo the use of glucose and glutamine in favor of fatty acid oxidation. In many tissues, fatty acids are not the predominant fuel choice, rather, fatty acid metabolism is reserved for conditions of stress or nutrient depletion as a means to restore metabolic homeostasis. Metabolic rewiring in fatty acid metabolism occurs in mouse hepatocellular carcinomas, primary human liver, and in lung carcinomas. Cancer cells desaturate palmitate to the unusual fatty acid sapienate to support membrane biosynthesis (Vriens et al., 2019). Moreover, plasma membrane remodeling is a necessary component of oncogenic signaling. Lipids are the main component of cellular membranes and play a crucial role in creating a barrier between subcellular compartments. Subtle changes in the structure, composition, and interactions of lipids in cellular membranes can dramatically alter biological functions. Advances in lipidomics have identified the lipid species that make up mammalian membranes. The major membrane lipids are glycerophospholipids (GPL), sphingolipids, and sterols (Harayama and Riezman, 2018). Membrane lipid composition, levels of saturation and cellular distribution of lipids are underexplored aspects that are crucial for organelle homeostasis, cell signaling, and the management of nutrient and oxidative stress (Röhrig and Schulze, 2016; Rysman et al 2010; Young et al. 2013). Lysophosphatidylcholine acyltransferase (LPCAT1) enhances saturated phosphatidylycholine content in the composition of the plasma membrane and also drives tumor growth by activating oncogenic signals (Bi et al., 2019).

Interestingly, high levels of MUFAs suppress ferroptosis by competing with PUFAs for insertion into the membrane (Magtanong et al, 2019). This membrane lipid composition is related to suppression of ROS at the plasma membrane and decreased levels of phospholipids containing oxidizable polyunsaturated fatty acids. This effect requires MUFA activation by acyl-coenzyme A synthetase long-chain family member (ACSL) 3 (Magtanong et al, 2019). Like ACSL3, lysophosphatidylcholine acyltransferase 3 (LPCAT3) has an important role in PUFA incorporation into phospholipids. Lack of LPCAT3 leads to modulation of membrane phospholipid composition by drastic reductions of AA, which is a polyunsaturated fatty acid present in phospholipids (Hashidate-Yoshida et al., 2015). ACSL4 is also implicated in the localized release of AA in the mitochondria by catalyzing the conversion of long-chain fatty acids to their active form, acyl-CoA, for synthesis of cellular lipids (Kuwata and Hara, 2019). ACSL4 limits the cytotoxicity associated with elevated cellular pools of unesterified AA by producing arachidonoyl-CoA, thereby increasing the apoptotic threshold and survival of castration-resistant prostate cancer (Kuwata and Hara, 2019). This suggests that the localized accumulation of PUFAs in the mitochondria can contribute to membrane depolarization and electron transport chain uncoupling, leading to increased ROS production. Therefore, specific fatty acids trigger ROS generation, ER stress and ferroptosis, suggesting the need for studies that define the function of lipid properties in cancer (Figure 4).

Figure 4. Lipid signaling in cancer.

Lipid metabolism directly affects cancer signaling. AKT phosphorylation, the major cancer signaling pathway, activates de novo lipid biosynthesis through ACLY. This phosphorylation cascade coordinates with PA, amino acids, mTOR signaling to boost tumorigenesis, and increase SREBP transcription, resulting in fatty acid synthesis. Phospholipids and TAGs from de novo lipid biosynthesis activate proliferation and rewire membrane lipid composition by increasing saturated phospholipids. Fatty acid uptake into the nucleus directly activates cancer metastasis by nuclear receptor and the transcription factor PPAR. PUFAs induce ROS-dependent ferroptosis. Different lipid signaling employs different adaptive responses in cancer -- either survival or death. ACC2, acetyl-CoA carboxylase 2; ACLY, ATP-citrate lyase; ACSL3, acyl-CoA synthetase long-chain family member 3/4; AKT, RAC-alpha serine/threonine-protein kinase; AMPK, AMP-activated protein kinase; DAG, Diacylglycerol; DGAT1, diacylglycerol O-acyltransferase 1; FAs fatty acids; FABP5, fatty-acid binding proteins 5; FASN, fatty acid synthase; LPC, Lysophosphatidylcholine; RTK, Receptor tyrosine kinase.

Lipid Toxicity in Cancer and Immune Cells

Tumors hijack pathways that protect normal cells from lipotoxicity as a mechanism to circumvent the harsh tumor microenvironment. Cancer cells prevent the accumulation of toxic cellular lipids and waste products using autophagy, which targets intracellular products and organelles to the lysosomal compartment for degradation (Poillet-Perez and White, 2019). Palmitate and other saturated FAs induce apoptosis in breast cancer, while unsaturated FAs, such as oleate, are non-toxic for cancer cells (Hardy et al. 2003; Yang et al. 2018). Cancer cells also protect against lipid toxicity by converting potentially toxic lipids, including fatty acids, DAG, cholesterol and ceramide to triglycerides, cholesterol esters and acylceramides that can be stored in lipid droplets (Senkal et al., 2017). Moreover, DGAT1 converts excess fatty acids into triglycerides and lipid droplets to protect glioblastoma cells from oxidative damage (Cheng et al. 2020). Therefore, inhibiting DGAT1 results in an excess of fatty acids moving into the mitochondria for oxidation, leading to the generation of high levels of ROS, mitochondrial damage, cytochrome c release, and apoptosis. Moreover, blocking DGAT1 can also channel fatty acids into phospholipids and increase ferroptosis (Dierge et al., 2021).

Ferroptosis (Stockwell et al. 2017), also operative in cancer cells, is the result of lipid peroxidation of PUFAs present in phospholipids to generate various lipid hydroperoxides (Kuhn et al. 2015). The expression of ELOVL5 and fatty acid desaturase 1 (FADS1) is upregulated in gastric cancer cells, leading to ferroptosis sensitization. AA supplementation restores sensitivity to ferroptosis in gastric cancer cells (Lee et al. 2020). The enzyme GPX4 is a central regulator of ferroptosis, and protects cells by neutralizing lipid peroxides. Ferroptosis suppression by ferroptosis suppressor protein 1 (FSP1) reduces CoQ, which acts as a lipophilic radical-trapping antioxidant that inhibits the propagation of lipid peroxides in several cancer cells (Bersuker et al., 2019; Doll et al., 2019). A recent study showed that immunotherapy-activated CD8⁺ T cells enhance ferroptosis-specific lipid peroxidation in tumour cells, and that increased ferroptosis contributes to the anti-tumour efficacy of immunotherapy (Wang et al. 2019). Therefore, modulation of biosynthetic and peroxisomal oxidation pathways may offer new opportunities for ferroptosis-mediated cancer therapy.

LIPID METABOLISM AND THE MICROBIOME

Interactions between the gut microbiome and host lipid homeostasis are highly relevant to metabolic disease susceptibility. Microbiota generate monounsaturated fatty acids by stearoyl-CoA desaturase 1 and polyunsaturated fatty acids via elongation by fatty acid elongase 5, leading to significant alterations in glycerophospholipid acyl-chain profiles. Interestingly, gut microbiota also generate acetate from dietary fiber, which serves as a precursor for hepatic long-chain fatty acids and their related glycerophospholipid species (Kindt et al., 2018). Also, Streptococcus pneumoniae, Gram-positive spherical bacteria, respond to exogenous fatty acids by suppressing de novo biosynthetic pathways and exclusively utilizing extracellular fatty acids for membrane phospholipid synthesis. This suggests that Streptococcus pneumoniae permits the utilization of the entire spectrum of mammalian fatty acid structures to construct its membrane (Gullett et al., 2019). For a more detailed discussion of lipids and the microbiome, we refer the reader to the following comprehensive review on the subject (Schoeler and Caesar, 2019).

TARGETING LIPID METABOLISM: DRUG DISCOVERY

Fatty acid biology may illuminate new targets for the treatment of metabolic diseases. While there are currently no direct fatty acid targets, there are drugs that inhibit long-chain acyl carnitine import into mitochondria, including fatty acid oxidation inhibitors such as Etomoxir (German et al., 2016; Pike et al., 2011), Ranolazine (Samudio et al., 2010), Soraphen-A (Beckers et al., 2007), TOFA (5-(tetradecyloxy)-2-furoic acid) (Guo et al., 2009; Pizer et al., 2000), and A-769662 (Göransson et al., 2007). Ranolazine is FDA approved as an anticancer drug, based on its targeting of FAO. While the inhibition of de novo fatty acid synthesis may cause a multitude of side effects, TVB-2640, a FASN-antagonist, is currently studied in Phase II trials as another putative cancer therapy (Brenner et al., 2015). Metformin (Pollak, 2012) and AICAR are AMPK activators that increase both FAO and FAS with anti-diabetic properties (Jose et al., 2011; Swinnen et al., 2005). Drugs that target the FAS pathway include SB-204990 (Hatzivassiliou et al., 2005; Ros et al., 2012) and LY294002 (Migita et al., 2008) modulators of the PI3K signaling pathway. A more comprehensive list of FAS inhibitors is included in Table 1.

Table 1. Fatty acid targeting drug development.

Summary of drugs developed to target fatty acid oxidation, fatty acid synthesis, and fatty acid uptake.

Mechanism of Action	Target	Inhibitor	Selected Reference	Target Disease or Condition	Clinical trials (https://clinicaltrials.gov)
Fatty acid oxidation
Inhibition of fatty acid β-oxidation and activation of pyruvate dehydrogenase	ACC	Ranolazine	Zacharowski et al., 2001; Samudio et al., 2010	Chronic angina (FDA approval); acute myocardial infarction; leukemia	FDA approved (NDA #021526)
ACC (acetyl-CoA carboxylase) inhibitor, a PPAR-α agonist	ACC, PPARα	TOFA (5-(tetradecyloxy)-2-furoic acid)	Ottemann Abbamonte et al., 2021; Pizer et al., 2000; Guo et al., 2009	Cutaneous lupus; systemic lupus erythematosus, glioblastoma; breast cancer	Phase 2 clinical trials (NCT03288324)
FAO activator	AMPK	A-769662	Kemmerer et al., 2015; Goransson et al., 2007	Type 2 diabetes (T2D); macrophages	NA
Inhibition of long-chain fatty acid import into mitochondria	CPT1	Etomoxir	German et al., 2016; Pike et al., 2011	Leukemia; glioblastoma	NA
Blocking FAO	Long-chain 3-ketoacyl-CoA thiolase (LCTH)	Trimetazidine	Gatta et al., 2017	Precapillary pulmonary hypertension; muscle wasting (cachexia)	Phase 2 clinical trials (NCT03273387)
Fatty acid synthesis
Fatty acid elongation	ACC	Soraphen-A	Beckers et al., 2007	Prostate cancer; high fat diet-induced insulin resistance; hepatic steatosis	NA
Block lipigenesis	ACC	Firsocostat (GS-0976, NDI-010976, ND-630	Alkhouri et al., 2020	Non-alcoholic Steatohepatitis (NASH)	Phase 1 clinical trials (NCT02891408)
Inhibit TAG accumulation into lipid droplets	Acyl-CoA synthetases (ACS)	Triacscin C	Mashima et al., 2005	Lung cancer; colon cancer; stomach cancer; brain cancer; and breast cancer	NA
FAS inhibitor	Acylglycerolphosphate acyltransferase (AGPAT)	CT-32501	Takeuchi and Reue, 2009	Prostate cancer	NA
ACLY inhibitor and AMPK activator in liver	ATP-citrate lyase (ACLY) and AMPK	ETC-1002	Chen et al., 2020	Hyperlipidemia	Phase 2 clinical trials (NCT02659397)
FAS activator	AMPK	Metformin	Bhansali et al., 2020	Type 2 diabetes; solid cancer	FDA approved, Phase 4 clinical trials
FAS activator	AMPK	Aminoimidazole Carboxamide Riboside (AICAR)	Jose et al., 2011; Swinnen et al., 2005	Lesch-Nyhan Syndrome; cancer	Phase 2 clinical trials (NCT00004314)
Blocking generation of acetyl-CoA for cholesterol and de novo fatty acid synthesis	ATP citrate lyase (ACLY)	SB-204990	Hatzivassili ou et al., 2005; Ros et al., 2012	Non-small cell lung cancer; solid cancer	NA
Blocking phosphatidylcholine (PC) biosynthesis	Choline kinase alpha (CKα)	TCD-717	Glunde et al., 2011	Solid cancer	Phase 1 clinical trials (NCT01215864)
Blocking fatty acid synthase and glutamate dehydrogenase activity	DNA methyltransferase, EGF receptors, HER-2 receptors, FASN	Epigallocatechin-3-gallate (EGCG)	Humbert et al., 2021	Lung neoplasms; esophagitis	Phase 2 clinical trials (NCT02577393)
Blocking triglyceride (TG) synthesis	Diacylglycerol acyltransferase 1 (DGAT1)	A922500	Zhao et al., 2008	Obesity; dyslipidemia	NA
Suppressing triacylglyceride (TAG) plasma excursion and adipose tissue TAG synthesis	DGAT	AZD3988	McCoull et al., 2012	Type 2 diabetes (T2D); obesity	NA
Blocking TG synthesis	DGAT	AZD7687	Morentin et al., 2019	Obesity; overweight	Phase 1 clinical trials (NCT01119352)
Blocking TG synthesis	DGAT	JNJ DGAT2-A	Irshad et al., 2016	Type 2 diabetes (T2D); solid cancer	NA
Blocking hydrolysis of triglycerides and the absorption of FFA	FASN	Orlistat	Schcolnik-Cabrera et al, 2018; Zhi et al., 1995	Obesity; overweight	FDA approved
FASN antagonistic drug	FASN	TVB-2640	Lolkema et al., 2015	Non-alcoholic fatty liver disease	Phase 2 clinical trials (NCT049064210)
FAS inhibitor	Fatty acid synthase (FASN) and HMG-CoA reductase (HMGR)	Cerulenin	Currie et al., 2013; Lupu and Menendez, 2006; Ros et al., 2012	Hepatic steatosis; solid cancer	NA
FAS inhibitor	Mitochondrial citrate transporter (CIC)	Benzene-tricarboxylate analog (BTA)	Catalina-Rodriguez et al., 2012	Solid cancer	NA
Fatty acid amide hydrolase inhibitor	Monoacylglycerol lipase (MAGL)	JZL184	Walenna et al., 2020; Taïb et al., 2019	Type 2 diabetes (T2D); glioblastoma	NA
Blocking biosynthesis of monounsaturated fatty acids	Stearoyl-coA desaturase (SCD)	A939572	von Roemeling et al., 2013	Renal cell carcinoma	NA
Alters membrane fatty acid composition	SCD	BZ36	Fritz et al., 2010	Prostate cancer	NA
Blocking the conversion of saturated, long-chain fatty acyl-CoAs to monounsaturated	SCD	CAY10566	Liu et al., 2007	Breast cancer; lung cancer; colorectal cancer	NA
FAS inhibitor in liver	SCD	MK-8245	Oballa et al., 2011	Type 2 diabetes (T2D)	Phase 1 clinical trials (NCT00790556)
Inhibits the ER-Golgi translocation of SREBPs	SREBP1/2	Fatostatin	Williams et al., 2013	Prostate cancer	NA
Blocks biosynthesis and accumulation of fat	SREBP1/2	FGH10019	Williams et al., 2013, Kamisuki et al., 2011	Prostate cancer	NA
Blocking SREBP activity	PI3Kα/β/δ	LY294002	Migita et al., 2008	Neuroblastoma; solid cancer	Phase 1 clinical trials
PPAR agonists drug for T2DM	PPARδ	Thiazolidinediones (TZDs)	Kim et al., 2001	Type 2 diabetes (T2D)	FDA approved
Fat uptake
Peptide mimetics of thrombospondin-1 (TSP-1)	CD36	ABT-510	Markovic et al., 2007	Melanoma, renal cell carcinoma, lymphoma	Phase 2 clinical trials (NCT00602199)

Emerging evidence suggests that lipid transporters may represent a new therapeutic target in cancer and metabolic disorders. CD36 is one such transporter that plays an important role in facilitating intracellular FFA uptake and trafficking (Coburn et al., 2001). Importantly, CD36 membrane levels and turnover rates are disrupted in diabetes leading to dysfunctional FA utilization, and variants in the CD36 gene influence susceptibility for metabolic syndrome (Love-Gregory et al., 2008). In the context of cancer, neutralizing antibodies used to block CD36 cause complete inhibition of metastasis in mouse models of human oral cancers and encouragingly impair metastasis in human melanoma- and breast cancer derived-tumors. Palmitic acid or a high-fat diet specifically enhance the metastatic potential of CD36⁺ metastasis-initiating cells suggesting that metastasis-initiating cells respond to dietary lipids (Pascual et al., 2017). The therapeutic relevance of CD36 is not restricted to oral cancers since CD36 is upregulated in intratumoral T_reg cells where it acts as a central metabolic modulator. Furthermore, genetic ablation of CD36 in T_reg cells suppresses tumor growth and enhances antitumor reactivity in tumor-infiltrating lymphocytes while preserving immune homeostasis (Wang et al., 2020). Selectively disrupting intratumoral T_reg cells is a sought-after approach for cancer immunotherapy.

TECHNICAL ADVANCES

Historically, cellular lipids were detected based on their biochemical and morphologic features using BODIPY and oil red O staining (French et al., 1993; Mehlem et al., 2013). Moreover, certain FFAs have been used as markers of diabetes (Boden, 2008; Reaven et al., 1988). With advances in technology, our ability to identify lipid species with greater granularity grows, directly contributing to a greater understanding of lipid homeostasis in physiology as well as in metabolic diseases. Analytical chemistry methods applied to understanding the entire lipid content of a cell have illuminated the lipidome and have revealed exciting new roles for lipids in cell biology and physiology (Han, 2016). Advances in mass spectrometry have expanded analytical sensitivity and specificity in lipidomic analyses, including (1) the characterization of lipids in relevant cellular compartments and structures by cellular fractionation, (2) the measurement of the physical properties of lipids, and (3) the description of the phenotypic and functional consequences of specific lipid perturbations. Chemical imaging, including coherent Raman scattering (CRS) microscopy, enables label-free visualization of lipid molecules in live cells allowing unprecedented visualization of the distribution and heterogeneity of lipids (Chen et al., 2020b). Addressing the challenge of tracing metabolic reactions within the complex network of cellular lipid metabolism, click-chemistry mass spectrometry reporter strategies now enable parallel quantitative monitoring of as many as 120 distinct, labeled lipid species to trace alkyne-labeled lipids (Thiele et al., 2019). Advances in metabolic tracing continue to enable a deeper investigation into how fatty acids are incorporated into membrane lipids. Applying this method to follow de novo fatty acid synthesis or degradation by beta oxidation will be an interesting future application.

A substantial number of drug targets are lipid-binding proteins and a map of lipid-protein interactions could uncover new modes of signaling of relevance to pharmacological perturbation (Niphakis et al., 2015). Lipid-based chemical proteomic probes identify the proteins that participate in lipid pathways in cells (Niphakis et al., 2015). Ligand-receptor, substrate-enzyme, and client-carrier relationships are just some of the interactions regulated by lipid-protein interactions. To identify proteins that interact with fatty acid-derived lipids, a set of probes containing binding groups that resembled common fatty acids, including arachidonic (C20:4), oleic (C18:1), palmitic (C16:0), and stearic (C18:0) was used to demonstrate that arachidonoyl lipids preferentially interact with proteins. In situ drug profiling with arachidonoyl lipid probes revealed that the lipid-interaction proteome is enriched in known drug targets. Combining a lipid probe approach with high-throughput drug screening identified NUCB1 as a previously unknown protein involved in facilitating the intracellular transfer of FA-derived lipid messengers, fatty acyl ethanolamides (NAEs)/NATs, for delivery to metabolic enzymes, such as FAAH and PTGS2 (Niphakis et al., 2015). Applying this approach to structurally distinct drugs has the potential to uncover additional ligand-protein interactions. The current gold standard for visualizing lipids in complex with membrane proteins is to use cryo-electron microscopy and X-ray crystallography. In the past, extraction-analysis coupled with LC-MS-based quantitative lipidomics was used to compare the lipid profile that co-purifies with a protein of interest with that of the native membrane to provide indirect evidence of interaction. The limitation of this approach is that it reports on lipids that are co-purified with a particular protein complex without differentiating regulatory lipids that are linked to a specific biological function. A recently published approach combines high-energy native mass spectrometry (HE-nMS) and solution-phase lipid profiling to determine the identity of lipids that directly interact with a protein of interest (Gupta et al., 2018). This approach offers a way to identify lipids that interact with a membrane protein and has been applied to understand how specific lipids maintain oligomeric states.

Studies combining CRISPR-based genetic screens, unbiased lipidomics, and transcriptomics have the potential to identify novel regulators of lipotoxicity (Piccolis et al., 2019; Zhu et al., 2019). Recent investigations using the known toxic saturated fatty acid, palmitate (C16:0), identified genetic modifiers of lipotoxicity in a cellular model using human leukemia cells. The authors nominated two novel targets for follow-up studies: a putative E3 ligase and ER-localized glycerol-3-phosphate acyltransferase (GPAT) enzymes (Piccolis et al., 2019). In the context of this study, one lipid class emerged as central to lipotoxicity: di-saturated glycerolipids (Piccolis et al., 2019). Palmitic acid exposure increases the relative amounts and saturation levels of glycerolipids including TG, DAG, phosphatidic acid, and lysophospholipids. A parallel study used a combination of a CRISPR-based genetic screen using palmitate and unbiased lipidomics to identify calcineurin B homologous protein 1 (CHP1) as a major regulator of ER glycerolipid synthesis (Zhu et al., 2019). CHP1 binds and activates GPAT4, which catalyzes the initial rate-limiting step in glycerolipid synthesis. Knockout of CHP1 in Jurkat cells, a human leukemic T-cell line, surprisingly resulted in cell proliferation due to partial compensation for this CHP1-knockout-mediated loss of ER lipid synthesis by GNPAT, a peroxisomal enzyme (Zhu et al., 2019). This compensation indicates a degree of plasticity and dynamic regulation of glycerolipid metabolism and lipid synthesis of proliferating cells. Furthermore, ongoing efforts to use systematic and unbiased approaches to identify alterations in lipid droplets include the Lipid Droplet Knowledge Portal (LD-Portal, lipiddroplet.org), a platform that integrates transcriptional profiles of lipid storage, organelle proteomics, genome-wide screens, and human genetics to characterize determinants of lipid storage (Mejhert et al.). Systematic studies of lipid biology utilizing new technologies will help identify targets for therapeutic intervention for diseases related to disrupted lipid metabolism.

A compelling future direction for the field of lipid biology is the use of spatial information to investigate changes in metabolites using advanced imaging techniques. Mass spectrometry imaging (MSI) is a technique used to simultaneously visualize the spatial distribution of molecules in a biological sample with implications for pharmacological target screening. For example, in the aftermath of drug treatment, lipid disturbance analysis can be performed using liquid chromatography-mass spectrometry (LC-MS) based lipidomics combined with matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) to map the location of biomolecules within tissue (Djambazova et al., 2020). Fully characterizing the structural diversity of lipids is an ongoing challenge, but recent efforts showed separation and imaging of lipid isomers with distinct spatial distributions from a whole-body mouse pup using MALDI trapped ion mobility separation (TIMS) MS (Djambazova et al., 2020). Spatial information can have profound implications for understanding biological function. In the mouse liver, investigators analyzed lipid composition in the nucleus and mitochondria as well as the temporal changes in lipid composition throughout the day and in response to time restricted feeding. Applying shotgun lipidomics (Han et al., 2012) revealed a total of 222 individual lipid species of which 147 were present in both organelles, 5 were exclusive to the nucleus, and 70 were unique to the mitochondria (Aviram et al., 2016). Of interest, restricting food availability and intake to nighttime induced dramatic changes in the lipid composition of different organelles.

Conclusions and Future Perspectives

In this review, we investigate the role of lipids as mediators of the progression of metabolic disorders. Increasing evidence suggests that the fatty acid composition of fats and the individual concentrations of FAs (i.e. monounsaturated vs. polyunsaturated, ratio of saturated to unsaturated, odd or even chain saturated FAs, omega position, etc.) have distinct effects on metabolism at the cellular level. Currently, high-throughput methods that take advantage of transcriptomics and functional genomics can nominate genes that are most relevant for metabolic disorders; however, in the future, there is a need to develop comprehensive platforms that can probe the biological effects of a diversity of lipid species in an effort to tease out the mechanisms by which structurally distinct lipids contribute to disease progression. Future work to understand lipid metabolism will rely on advances in chemical imaging, functional genomics, and biochemical characterization of the spatiotemporal dynamics of lipids. Emerging efforts to understand lipid-protein interactions and spatial transcriptomics in metabolic disorders and cancer will identify drug targets with the potential to promote beneficial lipid regulatory mechanisms. Advances in lipidomics have enabled the identification and characterization of lipid species; however, in comparison to gene expression or protein interaction networks, the subcellular distribution of lipids is an ongoing challenge. Spatiotemporal analyses provide an opportunity to move beyond snapshots of lipids within the whole cell or tissue at a single time point.