Lipid Matters

Subscribe to Lipid Matters Blog

You need an RSS Feed Reader to subscribe.

19 January 2026

Mechanistic snapshots of lipid-linked sugar transfer

I’m always drawn to articles that highlight the need for structural studies that AI approaches can’t resolve. That’s why I was drawn to a recent article in Nature Communications by Morgan et al titled “Mechanistic Snapshots of Lipid-linked Sugar Transfer". The authors used UV-photolysis of a chemically caged substrate with cryogenic time-resolved electron microscopy (cryo-TREM) to examine the catalytic mechanism of a membrane-bound glycosyltransferase, GtrB. They were able to visualize conformational changes during the catalytic cycle that moves each substrate, UDP-glucose and undecaprenyl phosphate, in proximity for catalysis. They were able to visualize the initial substrate-bound state, a catalytically poised intermediate, and the product-bound state involved in catalysis.They further supplemented their results with molecular dynamics simulations and biochemical analyses, to identify the conformations within the active site that drive catalysis. This in an intriguing studies that represents the power of structural biology approaches that provide an understanding of a catalytic that would be very difficult if not impossible to obtain with AI approaches alone. Structural biology is still alive and well.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

6 January 2026

The elusive origin of brain DHA, and the importance of publishing replication studies.

Supplementation with omega3 fatty acids derived from fish oil has been promoted as a neutraceutical approach towards maintaining health for at least 50 years now, driven in the early days by epidemiological data on diverse inflammatory conditions including cardiovascular disease and latterly, by studies on dementia risk. Of course, epidemiology does not prove cause and effect and in more recent times, several randomized clinical trials have conclusively failed to evidence the attractive idea that these lipids might be a panacea for any chronic diseases of ageing. One major question in this field relates to the bioavailability of omega3 fatty acids when orally administered. Dogma has been that dietary forms should be absorbed then taken up into cell membranes, particularly in the brain, where they’ll magically (through largely unknown but oft debated mechanisms) prevent cognitive decline. But does this idea hold up?

At the recent Society of Chemistry in Industry meeting in London on Lipids in Diet and Health, there was lively discussion of this exact question, initiated by a presentation from Richard Bazinet on his recent study in J Lipid Res. This seminal paper discussed how the brain is unable to make its own DHA, so it relies on other sources to maintain the high levels required for normal brain function. Here, Klievik et al set out to re-evaluate a series of studies by Sugasini et al, where supplementation with either LPC-DHA or di-DHA-PC was claimed to increase brain DHA content by up to 2-3 fold in mice. This was prompted by a recent study showing in contrast, that LPC-DHA supplementation didn’t increase brain DHA levels in Apoe3 and Apoe4 knock-in mice, and because as the authors state, they “are not aware of any direct evidence supporting the intact absorption of sn-1 DHA into the plasma”. In summary, in the recent study from Bazinet and colleagues, while oral supplementation of DHA either as PC or LPC led to significant enrichment in plasma and heart lipid pools, brain DHA levels were completely unchanged. This agrees with more recent data on this question cited by Bazinet, supporting the notion that the brain self-regulates its DHA levels rather than being influenced solely by what happens in the diet.

The importance of this study is twofold. First, it’s essential to understand where the brain derives DHA from, and how this could be regulated therapeutically. This study shows that in mice this is not from diet, leaving endogenous synthesis as the source. Indeed, in his seminal lecture at the SCI meeting, Richard presented new data beyond the published study, using natural abundance carbon isotope ratio analysis to demonstrate that DHA in the brain originates from endogenous synthesis. An earlier JLR paper describes this method.

The second reason that this paper is important is that it’s a replication study, performed in response to conflicting literature on the topic. Science depends on careful replication of key findings and all too often this doesn’t happen at all, or if it does, it occurs many years later having followed a long period of wasted time, funding and lost careers. Publication of negative data is difficult and hard work for all involved but it’s an essential part of science if we are to ensure the record is accurate. The field of omega3 fatty acids isn’t unique in facing this issue, nor are the issues addressed in this study by any means the only question marks hanging over fish oil neutraceutical claims. The editors of JLR should also be commended for supporting publication of replication studies, upholding standards in research through providing opportunities to correct and debate research respectfully through the process of peer review.

Valerie O'Donnell

Cardiff University

30 December 2025

Structural Insights into TMEM164-mediated Phospholipid Remodeling Involved in Ferroptosis

Ferroptosis is gaining increasing attention. This process is an iron-dependent cell death that involves the accumulation of lipid peroxides and the oxidation of polyunsaturated fatty acids (PUFAs) with a particular role for ether-linked PUFAs. TMEM164 is a transmembrane acyltransferase involved in ferroptosis catalyzing the formation of the ferroptotic C20:4 ether-linked phosphospholipids (ePLs). Consistent with this, cells lacking TMEM164 showed a selective reduction in ePLs, and genetic ablation of TMEM164 protects cells from ferroptosis. Recently, Ke et. al presented a cryo-EM study revealing a structural understanding of the role of TMEM164 in the lipid remodeling involved ferroptosis. The authors show that TMEM164 overall architecture is a dimer of two 7 transmembrane domain monomers with a metal ion bound catalytic center. Further, the authors identified a phospholipid substrate in a PUFA-bound in an intermediate state to cysteine 123 in the catalytic center. Interestingly, both loss and gain of function leads to a decline of PUFA-ePE and elevation of C16/18:1-ePE which confers resistance to the glutathione peroxidase 4 inhibitor RSL3-induced ferroptosis. Mutagenesis studies further validate critical residues for the catalytic center (C123) and the chelates center (E106, Y177 and H181). This work demonstrates structural feathers of TMEM164 as a membrane lipid remodeler which modulates ferroptosis.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

23 December 2025

A Lysolipid for Treating Obesity?

Holidays are upon us and thoughts of overeating and obesity occupy an increased amount of space in our minds.Given this, and the interest of the readers of this blog, I thought it would be interesting to bring some attention to a recent article suggesting a mechanism by which obesity may be combated by using lysophospholipid, 1-linoleoylglycerophosphocholine (1-LGPC). This mechanism capitalizes on a new potential connection between 1-LGPC and the KEAP1-Nrfs (Kelch-like ECH-associated protein 1 - Nuclear Factor Erythroid 2-Related Factor 2) axis. This axis is a recognized cellular system that defend against oxidative and electrophilic stress. In this axis, KEAP1 binds NRF2 and tags it for degradation, and when stress signals are present (e.g. ROSs), KEAP1 releases NRF2 upon which it to enters the nucleus, bind to DNA, and activate protective genes such as antioxidant enzymes (e.g. HMOX1), detoxification enzymes (e.g. GSTs), as well as metabolic enzymes and enzymes involved in autophagy. In a recent article by Wang et al (J Lipid Res. 2025 Nov;66(11):100914), the authors report a decline in 1-LGPC in the blood of obese patients. Interestingly, 1-LGPC reduced the high-fat diet-induced lipid accumulation in zebrafish larvae and in human adipocytes. Their data indicated that uncoupling protein 1-dependent thermogenesis and mitochondrial respiration were significantly boosted. Importantly, NRF2 expression and nuclear translocation were induced by 1-LGPC. Other data indicated the KEAP1-Nrf2 axis was involved in the 1-LGPC-induced energy expenditure. The authors suggest their results provide a new and interesting insight into a novel physiological role for 1-LGPC in obesity and points to a new target for treating obesity. I should note that there could approaches to confirm and strengthen their conclusions such as the use of another lysolipid, and an alternative to using brusatol such as an RNAi knockdown Nrf2. Nonetheless, their results are indeed intriguing.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

25 November 2025

Potential Treatment Avenue on the Horizon for Neimann-Pick Disease Type A?

We have known for some time that Niemann-Pick Disease (NPD) is a genetic disorder that leads to fat accumulation is various tissues such as liver, spleen, and brain. When I asked folks about NPD, it is common for me to hear that lysosomal cholesterol transporters are defective in this disease. Actually, there are three major types of NPD designated as Types A, B, and C, with Type C involving the defective lysosomal cholesterol transporters designated NPC1 and NPC2. The genetic defect in Types A and B, however, appears to involve dysfunctional acid sphingomyelinases (aSMases), encoded by SMPD1 gene, which hydrolyze ceramide as well as phosphatidylcholine. The difference between NPD-A and NPD-B is that the mutation leading to NPD-A leads to a near-complete loss of SMase activity, while enzyme mutations leading to NPD-B retains some modest activity. These mutations are significant to patient outcomes as those with NPD-A rarely live more than 3 years of age, while those with NPD-B may live to adolescence or early adulthood without serious neurological difficulties. In a recent paper by Beard et al. generated a mouse model in which mice harbored a S505A mutation (aSMase^S505A), which corresponds to the S507A mutation in humans retained L-SMase activity and tissue sphingomyelin levels in the brain. Interestingly however, these mice lacked S-SMase activity in serum. ASMase^S505A mice also also protected from NPD physiology and pathophysiology responses. Importantly, aSMase^−/− mice that showed significant decreased locomotor control, was prevented in aSMase^S505A mice. The authors suggest that their results suggest that expression of L-SMase may serve to pave the way for enzyme replacement therapy in humans with in NPD.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

10 November 2025

Neurolipid Atlas, a lipidomics resource for neurodegenerative diseases.

Our brains represent complex biochemical landscapes, with a lipidome that’s long been known to be different to that of other organs, for example being highly enriched in long chain PUFA such as DHA and EPA. While lipid changes in the brain are increasingly implicated in neurodegenerative disease, many gaps in our knowledge about how they respond specifically and precisely during health and disease remain. Aiming to facilitate research in this area, Feringa et al recently published a resource that uses lipidomics to describe which lipid species are present in specific cell types, also including several derived from iPSCs, which are increasingly used as a tool for mechanistic neurological and neuropsychiatric research.Importantly, the project provided an opportunity to screen for lipidomic differences across multiple cell types and also whole brain tissue from humans and mouse models of disease, resulting in several interesting discoveries. One notable finding relates to the Alzheimers disease (AD) risk allele, APOE4, which the study showed was associated with cholesterol ester accumulation in astrocytes and in brains of patients with AD. Altered cholesterol metabolism was also demonstrated during astrocyte immune regulation, potentially linking this process with AD pathology. The full dataset has been made available in an online repository which allows users to both browse, and importantly, to add further data from their own research.

Making available comparative data on lipid composition (at an organ, tissue, cellular or even sub-cellular level) is a topic that often comes up when we talk to users of LIPID MAPS. A data repository that describes what constitutes a brain or a liver lipidome is an attractive concept for the field, however, this is an area that’s difficult to support at scale for both methodological as well as theoretical reasons. First, lipidomes are not generally unique across tissues/organs/cells. More often than not (if we consider mammals) specific lipids will show relative differences in amounts, not an absolute presence or absence. This means that defining a lipid composition that’s characteristic of brain versus liver or kidney is extremely challenging and the best we could say is that one tissue is enriched or relatively-deficient in particular species. Focusing on one type of organ, and comparing different cells within that organ, or different diseases is one way around this, with the caveat being that a valid comparison requires either the same assay to be performed on the same instrument to generate all the data, or highly validated quantitative values need to be generated using the same assay (if the analysis is performed by more than one lab). In the case of Feringa et al, to overcome these issues, all the measurements were made using a single pipeline on one machine (Lipidyzer, on a Sciex QTrap) allowing all samples to be directly compared within this study. The repository provides a very useful model for focused questions to be posed while comparing large numbers of lipids at scale in a single organ or tissue type. It will be interesting to see how/if data deposited by other researchers in the future, but generated using different platforms, can be integrated with currently deposited data in a meaningful way. Wider than this study, this is a major challenge for the field of systems lipidomics in general, where data reuse is becoming a significant research endeavour in its own right.

Valerie O’Donnell, Cardiff University.

28 October 2025

Mitochondrial NADPH fuels mitochondrial fatty acid synthesis and lipoylation to power oxidative metabolism

Nicotinamide adenine dinucleotide phosphate (NADPH) is a well-known cofactor involved in a variety of biosynthetic pathways as well as participating in the prevention of oxidative stress. This cofactor is confined to the cytosol and mitochondria.It is known that the principal enzyme involved in the production of mitochondrial NADPH is NAD+ kinase 2 (NADK2) which catalyzes the phosphorylation of NAD+ and ATP or polyphosphates. A report by Kim et al. has shown that in addition to this role, NADK2 is essential for maintaining the level of protein lipoylation, the posttranslational modification of specific lysine residues with lipoic acid. This modification has been found on some of the TCA cycle enzymes such as pyruvate dehydrogenase and assembly of the electron transport chain. In addition to the role of NADPH as a cofactor in mitochondrial fatty acid synthesis (mtFAS), NADK2 is also involved in the translation of genes involved in this pathway. Overall, their data show that NADK2 plays a critical role in mtFAS activity, cellular respiration, and mitochondrial translation.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

30 September 2025

ABDH18: A Newly Identified Cardiolipin Deacylase

Cardiolipin (CL) is a unique phospholipid composed of two phospholipids, four fatty acids, linked together by a glycerol group resulting in a central hydroxyl group. This phospholipid is essential for a number of mitochondrial functions. Interest in this CL results largely from its clinical impact in that defects in the synthesis of this lipid leads to a rare but devasting syndrome referred to as Barth syndrome. CL is synthesized on the inner mitochondrial membrane. An important step in the synthesis of mature CL involves remodeling of the de novo CL species involving the exchange of nascent acyl chains with longer, unsaturated chains. The gene involved in Barth syndrome has been identified as a mutated TAFAZZIN (TAZ) transacylase which is the final enzyme involved in CL-remodeling. This mutation leads to an accumulation of monolysocardiolipin (MLCL). Interestingly, the enzyme responsible for the generation of MLCL has been known in yeast (Cld1), the homologue in plants, animals, and humans remained a mystery until the publication of two recent papers. Both of these reports identified ADH18, a deacylase which converts CL into MLCL Masud et al, and Ren et al . These studies used human HAP1 cells (Masud et al.) as well as mouse myocytes and Drosophilia (Ren et al.) model systems. In both studies, inhibition or suppression of ADH18 resulted in a suppression of the accumulation of MLCL and appears to rescue TAZ mutant phenotypes. Interestingly, suppression of ABHD18 led to more saturated acyl chains typical of CL in both studies. The Ren et al paper suggested that, in cells, ABHD18 did not deacylated CL species with more than five double bonds. This is interesting as it suggests potential regulatory mechanisms such as substrate specificity, or a modulation by membrane architecture or interaction with other membrane proteins. Both studies, however, support a critical role for ABHD18 in CL biosynthesis and offer an exciting and much needed potential therapeutic target for Barth syndrome. Both papers are interesting to read.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

2 September 2025

Quantitative imaging of lipid transport in mammalian cells

How Lipids Get to Where they Need to Be

We all have all been taught that intracellular membranes contain specific lipid species with specific headgroups and constituent fatty acids in specific proportions. How this distribution is established has long been a mystery for most lipids. Using a combination of pulse-chase fluorescent lipid probe imaging coupled with ultra-high-resolution Fourier-transform (FT) mass spectrometry (MS) and mathematical modelling the authors have quantitatively mapped the kinetics of -specific lipid transport and metabolism. Interestingly they propose they have identified the primary mechanism involved in lipid sorting to specific organelle (Iglesias-Artola et al. Nature 2025 Aug 20.- Online ahead of print). The authors investigated the movement of modified, phosphatidycholines, phosphatdic acids, and phosphatidylethanolamines. As endocytic trafficking of sphingomyelins has been well characterized, the authors included modified sphingomyelin in this study. Their studies suggest that fast, species-specific lipid sorting occurs via directional, non-vesicular lipid transport and non-vesicular lipid transport also dominates the organelle distribution of lipids. Interestingly, the authors provide data implicating flippases engaged in species-selective flipping in establishing in establishing and maintaining organelle membrane lipid composition. The authors further suggest that the species-specific metabolism of phospholipids controls neutral lipid metabolism. Refreshingly, the authors also point out important that their study can falls short of being able to mimic the behavior of fully saturated lipids and they are not able to lipid exchange between organelles far from the plasma membrane. Overall, their approach, however, is an interesting approach to study how lipid distribute within cells and their potential impact on cell biology.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

18 August 2025

Every 5-10 years or so, new innovations in mass spectrometry come along that drive a step-change in how we see lipids in biology. For example, benchtop instruments which were developed around 2005 kick-started the lipidomics revolution, while untargeted screening methods initially applied to metabolomics around 2012 led to a huge interest in surveying not only what we knew, but what we didn’t know. Along with each new MS approach, software has come along that allowed us to make sense of what we found, enabling us to mine and uncover new lipids through analysing the huge amounts of information contained in these massive experiments.

Despite the advantages of switching from GC/MS to LC/MS that took place around 2005 when benchtop instruments arrived, one clear disadvantage was the loss of structural information generated using this method. This is because ionisation and fragmentation methods used with LC/MS/MS are considered “softer” and don’t achieve the same level of fragmentation as electron impact or chemical ionisation used with GC/MS. While this is fine for simple quantification of well-characterized molecules it means there a significant loss of information and makes the working out of novel lipid structures far more challenging, as well as making it impossible to discriminate what could be many different molecules from each other in a single biological sample.

Addressing this issue, recent innovations over the last 5 years focused on “enhanced fragmentation modes”, of which there are several including UV photodissociation (UVPD), ozone-induced dissociation (OzID), electron activated dissociation (EAD), Paternò–Büchi (PB) and oxygen activated dissociation (OAD). Combined with today’s powerful high resolution MS instruments, these fragment either at C-C bonds, or selectively at C=C double bonds, providing richer information on structure that enables the diversity of FA containing molecules to be more comprehensively profiled in large scale cataloguing experiments in a way that GC/MS instruments could never have done.

Along with this, interest in using retention time to discriminate structures has emerged. The first papers on this appeared around 2014, when Choi et al developed a phosphatidylcholine retention time index to support identification, followed in 2016 by a study from the Holčapek group showing how relative carbon and double bond numbers could predict retention time behaviour for over 400 lipids in 14 classes (5 categories of lipids). Adding to this, a new paper in Nature Comms last week, from Lamp et al shows how retention time can be used to assign chain-specific C=C positions for lipids, in particular determining the ω-positions. To develop the database of verified identifications, the authors used a stable isotope labelling approach supplementing RAW cells with FA of varying w-composition to alter the endogenous lipid pool, enabling identification of specific lipids enriched with labelled FA, as well as their metabolites. A machine learning-based retention time mapping approach was then developed to extend the identification of lipids to over 2.4K. A tool was developed which allows other researchers to apply this method to their own data moving forward.

A central question with this approach is how well a predictive machine learning method maps to real life data, which can only be generated experimentally. In response to a reviewer request, the authors conducted EAD and benchmarked their method against published papers using PB or OzID showing close alignment, and supporting the validity of the approach. The full validation data for the EAD comparison is deposited online.

Applying the approach to a real-world biochemical question, it was then asked (using existing published data) whether C=C positional specificity at either Sn1 or Sn2 dictated cPLA₂’s substrate preference. Several new findings were revealed including that the enzyme has a similar specificity for mead acid (MA, 20:3(n−9)) as it does for AA, however that wasn’t the case for 20:3(n−7) and 20:3(n−6), which turned out to be poor substrates. Overall, the study demonstrates elegantly how this approach can be applied to existing studies, where full chromatographic data is available, to extend our knowledge of the biochemistry of complex lipid metabolism, without the need for new experiments. As we move forward the mining of rich data housed in publicly accessible databases using methods like this will undoubtedly lead to many further insights into lipid metabolism.

Valerie O’Donnell, Cardiff University.

5 August 2025

Boswellic Acid Anyone?

Boswellia carterii is probably most popular for its oil referred to as Frankincense. This oil has been used in soaps, skin care products, and aroma therapy. Therapeutically, it has also been suggested to be cytotoxic for some tumors. It now appears that another product from this tree may be effective for treating non-alcoholic fatty liver disease (NAFLD) and its associated complications. NAFLD affects approximately 30% of the world’s population and while it occurs in diverse populations, there is an unfortunate rising incidence of obesity and NAFLD in pediatric populations. Such an increase represents an alarming potential for further increases in pediatric insulin resistance. This recognition has led to increased attention on identifying effective pharmaceutical approaches for treating NAFLD and its related metabolic disorders. In a recent study by Luan et al in the Journal of Lipid Research it appears that another product from Boswellia carterii, 3-acetyl-11-keto-beta-boswellic acid (AKBA) may be an effect agent for the treatment of NAFLD including the associated weight gain and insulin resistance.Using a mouse model and cultured hepatocytes, the authors showed that the mechanism of boswellic acid effects involved a direct interaction with monoacylglycerol lipase (MGLL) in hepatocytes and their studies highlighted an important role for MGLL in NAFLD.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

21 July 2025

Schools out for summer

While many of the schools in the northern hemisphere are on ‘Summer break’, have you ever given thought to the link between lipids and the holidays?

A 2024 paper, by Eglitis. et al. demonstrates that summer holidays can put children at risk for weight gain and less healthy habits—but structured summer programs can help reverse that trend.

Without structured routines, children tend to: Have less active time, Spend more time on screens, Skip physical activities, & Possibly eat more junk food

The systematic review and meta-analysis looked at 10 controlled studies including 1,446 children aged 5–18. This data showed that those who attended summer camps, and thus were more active had a healthier weight (a small reduction in adiposity) at the end of the summer compared to those who hadn’t attended any camps.

This study did not measure lipid profiles (e.g., cholesterol or triglycerides), however other studies have established that increasing activity and reducing adiposity typically improve lipid profiles, especially in children who are overweight.

In a similar paper from 5years previous, a similar conclusion was also reached: Larose et al.,2019 mapped interventions in summer camps across North America and Europe targeting physical activity, sedentary behaviour, and diet in children aged 6–16. Their findings highlighted modest but meaningful improvements: increased activity, healthier eating, and reduced screen time.

While Larose et al.'s review focused on movement and diet, its positive outcomes form a strong bridge to lipid health. Increased activity and better nutrition in childhood are proven to improve cholesterol and triglycerides, setting the stage for lifelong cardiovascular wellness.

Summer programs act as a public health buffer, offering scaffolded days that replicate school-like structure with adult supervision, planned activities, healthy meals, and peer interaction—promoting healthier movement habits and body composition.

Lauren Cockayne

Cardiff University - LIPID MAPS

24 June 2025

Lipid peroxidation induces formation of aldehydes that are well known to react with primary amines to form novel structures. One class of these molecules is represented by covalent adducts of aldehydes with phosphatidylethanolamine (PE) headgroups, called N-aldehyde-modified PEs, NALPEs).Large numbers of these lipids are already known, including forms modified by malondialdehyde (MDA), 4-hydroxynonenal (HNE) and Isolevuglandins (isoLG), and some of these have been shown to have biological effects that are generally considered proinflammatory.

A recent study from the Davies lab at Vanderbilt (Fadaei et al) has expanded our knowledge of how these interesting lipids could be metabolized if/when they are formed in vivo.Specifically, they focused on N-acyl PE-hydrolyzing phospholipase D (NAPE-PLD), an enzyme already known to metabolize PEs that have been modified enzymatically by addition of acyl chains greater than 4 carbons in length, termed NAPEs. In this study, the authors first identify many previously undiscovered NALPEs that are formed in complex lipid oxidation mixtures. They then go on to show that NAPE-PLD can indeed hydrolyse these diverse NALPEs generating the corresponding phosphatidic acids (PA).

NAPE-PEs are formed endogenously by enzymatic acylation reactions and have been demonstrated to form in vivo in mammalian tissues for many years already, for example, as part of the biosynthetic pathway leading to formation of N-acylethanolamines (NAE). On the other hand, there seems to be only few reports of NALPE-PEs being formed in vivo, for example, a paper from 2009 from the Salomon group, that used LC/MS/MS to show the presence of LG and isoLG adducts that form with PE, then ultimately further oxidize to form stable lactams and hydroxylactams. Interestingly, in that paper they did not find the intact PE, with both acyl chains attached, and instead, their detection required prior treatment with PLA_2. This released the Sn2 FA, with the result being that all species with the same Sn1 FA when totalled together, rose above the instrument limit of detection and could be identified. Levels of these lipids were increased in livers from ethanol fed mice and human plasma from patients with age related macular degeneration. Considering the work was done several years ago, and MS instrument sensitivity for quantitation is probably around 100+ fold higher than it was back then, it would be interesting to revisit this work without PLA₂ hydrolysis and try to map the specific lipids as intact species. Also, additional structures now identified by Fadaei et al might also be identifiable using the newer generation instruments.

Valerie O'Donnell

Cardiff University

6 June 2025

Implication of CDP-DAG synthases in plant growth and disease resistance

While many of us focus on mammalian lipids and lipid metabolism, there are often reports that remind us of the interesting aspects of plant lipid metabolism. Such is the case with a recent report by Tan et al. In addition to phosphatidic acid (PtdOH) playing important signaling roles in mammals and plants it is subject to multiple routes of metabolism. One of its metabolic routes is its conversion into cytidine diphosphate diacylglycerol (CDP-DAG) by CDP-DAG synthases (CDSs). In the report by Tan et al, while the knock-down of the CDS genes, cds1 and cds2, suppressed the growth of Arabidopsis thaliana, it also provided resistance to multiple pathogens.

The level of reactive oxygen species (ROS) production induced was significantly higher cds mutants than that in wild-type (WT) leaves. Additionally, phosphorylation of mitogen-activated protein kinases (MAPKs) in the cds mutant was increased compared to the WT. The authors then employed a lipidomic, transcriptomic, and metabolomic approach, the authors provide evidence that an accumulation of PtdOH in the cds mutant, led to the activation of the jasmonic acid (JA) and salicylic acid (SA) signaling pathway, and increased transcript levels of known plant defense-related genes. Interestingly, downstream metabolites involved in in plant immunity also increased.

It is not surprising that levels of the non-amine phospholipids, such as phosphoatidylinositol, was decreased as CDP-DAG is involved in the synthesis of this lipid. It is curious, however, that the levels of amine phospholipids, such as phosphatidylethanolamine and phosphatidylcholilne, as well as their lysolipid counterparts were elevated perhaps as a compensatory mechanism. Nonetheless, the authors suggestion that their data provides evidence that CDSs may play role(s) in metabolic regulation and disease resistance in Arabidopsis is intriguing. Indeed, their data may suggest there are some unappreciated roles for CDS genes in mammalian cells as well.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

27 May 2025

This week, I’m blogging about a recent article published in Science Signaling, which represents the culmination of around a year and a half’s work by the International Lipidomics Society Oxylipin Interest Group, on community-agreed recommendations for oxylipin analysis. A preprint link is also here.

A relatively short focus article introduces the main supplement which then provides more in-depth information on how the project came about, as well as a wealth of information for researchers, on the analytical factors that need to be accounted for when measuring and reporting on these lipids.

Of course, most of the recommendations are not specific to oxylipins, and are more broadly relevant, for example describing how to make a reasonably educated judgement about whether a molecule is or is not actually present in a sample is not just specific to one category of lipids but applies more broadly to analysis of all molecules using chromatographic methods.

The story behind this article is familiar to many of us working in the field; there was an acknowledged need for consistent approaches to ensure data quality when measuring and reporting these lipids. Where levels of oxylipins are low or below limit of detection of a method, over-interpretation of poor data could lead to errors, and since much of our research into oxylipins could either indirectly or directly impact patients at some point down the line, then we have a duty to ensure we’re working in line with best practice.

One consideration we had when approaching this project is that most of us are not working with clinical grade assays in our labs, and the sort of validation required for a precise estimation of actual amounts in human samples for diagnostic purposes wasn’t going to be feasible for the majority of our studies, or necessary. So, we needed to take a pragmatic approach. Taking into account the approaches being used in the clinical domain was essential, and to that end, we had expert advice from colleagues working at the Centre for Disease Control and the Clinical and Laboratory Standards Institute, in the USA, who described in detail how they apply formal guidelines, e.g. from CLSI, FDA, EMA and others, to their routine quantitative pipelines. This was enormously helpful, and it also revealed how easy it can be to make mistakes by adopting selected aspects of clinical recommendations into research grade workflows, mis-using them and drastically overestimating the level of confidence in detecting a molecule.

The process for developing this recommendation is worth a mention as it serves as an excellent model for this type of work. ILS have an Interest Group format where the community are all invited to take part and input, through open forums. This was a perfect way to bring the community together and invite open discussion.Following a series of initial webinars, led by a small core group (chaired by myself), a draft was written (by Nils Schebb) and eventually almost 100 researchers became co-authors of the final article. To complement this, LIPID MAPS have just released an oxylipin standard spectral library, already with contributions from 3 laboratories already, and a 4^th shortly to be added. The aim here is to have a curated set of spectra generated using several different platforms for most or all of the available standards of oxylipins, that is searchable and downloadable and can be used to aid in identification of lipids in biological samples (https://lipidmaps.org/databases/oxylipin/browse).

Valerie O’Donnell, Cardiff University

5 May 2025

The Antidepressive Drug Sertraline Inhibits Phosphatidic Acid Phosphatases

Sertraline, commercially known as Zoloft, is a well-recognized selective serotonin reuptake inhibitor (SSRI) widely used in the treatment of depression. SSRIs are also known to be useful anti-fungal agents although the mechanism of this activity remains unclear. While this mystery still prevails, new evidence from the Carman lab suggests an intriguing player may be a phosphatidic acid (PA) phosphatase (PAP). This enzyme catalyzes the Mg⁺² -dependent dephosphorylation of PA leading to the generation of diacylglycerol. Stukey et al, recently reported the ability of sertraline to noncompetitively inhibit the Saccharomyces cerevisiae PAP enzyme termed Pah1 which led to a decrease in triacylglycerol synthesis. The noncompetitive inhibition mechanism was substantiated by molecular docking of sertraline, as well as the well-established PAP inhibitor propranolol, to non-catalytic sites in the haloacid dehalogenase-like domain of Pah1. Consistent with the potential importance of this drug in humans, the authors showed the ability of sertraline to inhibit the human PA phosphatases α, β, and γ which are lipin 1 orthologs of Pah1. These studies should inspire future studies to examine the role of lipin-mediated lipid metabolism in modulating depressive symptoms in humans.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

15 April 2025

Today’s blog comes from the end of the Keystone Symposia Lipids in Cellular Function and disease, which took place over the last week in Breckenridge, Colorado.

I’ve been honoured to act as co-chair with Junken Aoki and Vytas Bankaitis at this wonderful meeting which hosted research from across the globe into the roles of our favourite molecules, lipids, in driving underpinning mechanisms of health as well as how they contribute to causing human illness. A major strength of this meeting has always been the focus on basic biochemical processes, discovering new lipids and metabolic pathways and then mapping these onto the wider context of whole-body regulation and this meeting was no exception. Indeed, among the many highlights were new metabolic pathways and therapeutic applications for oxylipins (glucuronidation, clinical trials, clinical assays), discovery of new enzymes involved in phospholipid metabolism (Lands cycle, BMP synthesis), a huge focus on sphingolipids which included new molecules discovered, and their roles in metabolic disease, and novel insights into ferroptosis, a cell death process now reaching maturity with emerging clinical applications. There were so many great discussions, with robust and open Q&A sessions, and engaging and enthusiastic poster conversations, all signs of people having fun discussing science, learning and exploring as it should be. Working with Keystone also deserves a mention, they are truly a great organisation, supporting science and scientists with passion, professionalism and a sense of fun, exemplified by the new entertainment programme of music bingo on the last night hosted by Debbie and David.

On the other hand, it’s impossible not to mention the wider context in which this meeting took place. Many heartbreaking stories of the damage the current administration is doing to science abounded. From the loss of federal teams and committees charged with maintaining and improving provision of healthcare to the US population, to the wholesale cancelling of training grants because they used terminology the administration don’t approve of, to the many investigators waiting day by day to find out if they even have funding to pay their staff it was as bad as we have been hearing, and worse. Decisions aren’t made logically but arbitrarily which means they are unpredictable and impossible to fight. Damage limitation one day doesn’t work the next as the goalposts shift unexpectedly making the horizon move away. Not only researchers but associated industries that support research are all badly affected with all this compounded by the impact of tariffs. The carnage being inflicted on the epicentre of the world’s global science base is truly shocking. We can only hope that there’s a seismic shift soon before it’s too late, to return some sense of sanity and prevent this jewel in the crown from being irreversibly lost. We wonder how we can help, but it’s not obvious other than to support our US colleagues with a shoulder to cry on and practical help if that’s at all possible.It’s not much but having come here (at least, without any problems crossing the border) I can see how important it is to my colleagues to continue to maintain some semblance of control, a view of normality and a sense that despite all this, there will be a way out the other side.

We in Europe have looked to US science to provide a marker in the sand for global innovation, cutting edge progress, major discovery, collaboration and passion for research for almost a century. Seeing the erosion of this is profoundly depressing. These shifting sands don’t make for stability, they sow division and lead to conflict. Although this may seem obvious, it’s not something that the powers that be over here seem to have any care about as they go thrashing through the infrastructure with no thought for what’s being lost. As someone who grew up in the era before MMR, but never having had measles, I had my first dose of this lifesaving vaccine two weeks ago, just before coming to Keystone. NHS eligibility includes lack of being vaccinated and travel to an area with active outbreaks of disease. This is the first time I have been vaccinated for any communicable disease before travelling to the US and serves as a sign of where we’re headed if things don’t start to improve soon. Yesterday, in some sort of turn around, RFJ said “The federal government’s position, my position, is that people should get the measles vaccine,….”. Let’s hope we see a reversal of other ill-informed health related decisions soon, before more people die unnecessarily. If you want to learn how universities are dealing with these issues, listen to the President of Princeton University talk with bravery and passion about the impact of the cuts on his university, as published in the New York Times and available here.

Valerie O'Donnell, Cardiff University

19 March 2025

A Positive Feedback Loop of PIP5K-mediated production of PI(4,5)P₂

The role of phosphatidylinositol-4,5-bisphosphate (PtdIns(4,5)P₂) in signaling cascades is now recognized as a paradigm of lipid signaling. Indeed, it’s difficult to read any review of lipid signaling pathways and not see a reference to the “PI Cycle” in which PtdIns(4,5)P₂ plays a central role. As a result, my posts often try to focus on other lipids to highlight the important role of other members in this family of biological molecules. Recently, however, a paper by Duewell et al. caught my attention given its focus on increasing our understanding of the structural elements of an important interfacial lipid metabolizing enzyme, phosphatidylinositol-4-phosphate 5-kinase (PIP5K). This enzyme is a major enzyme involved in the generation of PtdIns(4,5)P₂ by catalyzing the phosphorylation of PtdIns(4)P at the plasma membrane. In a previous publication, the authors showed that PIP5K displays a positive feedback loop which involves membrane-mediated dimerization and cooperative binding to its PtdIns(4,5)P₂ product. The recent report by Duewell et al. identified structural motifs involved in PIP5K recognition of PtdIns(4,5)P₂ and dimerization. Using TIRF microscopy and kinetic analyses, they provide a model whereby PIP5K cooperatively engages with PtdIns(4,5)P₂ mediated by an N-terminal region termed the specificity loop. Further, after orienting the enzyme on the membrane, the enzyme binds PI(4,5)P₂ near the active site through a motif previously referred to as the substrate or PIP-binding motif (PIPBM). Their data supports an intriguing model in which the specificity loop and PIPBM act in concert to orient PIP5K on the membrane and modulate its catalytic activity resulting in a positive feedback loop during PI(4,5)P₂ production.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

5 March 2025

New phospholipid remodelling enzymes that generate fatty acyl thiamine esters

Cells dynamically remodel their phospholipid membranes to ensure that their molecular species compositions are in line with the requirements of their specific tissues. For example, immune cells contain many plasmalogens, while brain has a high proportion of longer chain n3 PUFA. Membranes are also remodelled in response to acute challenge, e.g. following agonist activation of platelets or white cells. The classic pathway for this is the Lands cycle, described by Bill Lands in the 1950s, which utilises families of enzymes from the ACSL and LPLAT/MBOAT families. These show strong cell and tissue specific expression patterns.  Over the last several decades, many of these enzymes with differing PL and FA specificities have been discovered and characterised and as they increased in number and complexity, a new nomenclature was proposed in 2022 by Shindou et al.

Recently, a new study in Science Advances from the Petkevicius lab at the MRC Mitochondrial Biology Unit in Cambridge has identified that a family of poorly understood TRAM-LAG-CLN8 domain (TLCD) containing proteins also can act as phospholipid remodelling enzymes, regulating cellular lipid composition and generating novel esters as biproducts. 

What’s also very interesting about this study is that it’s the first identification of new lipids comprised of fatty acyls attached to thiamine, which the authors proved using labelling studies, revealing palmitic, stearic and oleic-thiamine esters. These were identified in Hela cells and were dependent on expression of TLCD1 (and this was conserved in both yeast and worm).It will be interesting to see how this develops, in particular whether these novel products display unexpected biological roles.

Importantly, Sheokand et al also show that one of these proteins is a lysoPG acyltransferase, participating in lysosome function. Specifically, CLN8 is involved in generation of bis(monoacylglycero)phosphate (BMP), identifying a new way to form this family of lipids, of direct relevance to Battens disease where mutations in the protein are directly implicated.

Valerie O’Donnell, Cardiff University

19 February 2025

Lro1 Regulates Endoplasmic Reticulum Biogenesis.

One of the ever present and interesting biological mysteries is how cellular organelles regulate their size.  In a recent study by Lysyganicz et al, the authors present some intriguing evidence regarding the role of a phospholipid diacylglycerol acyltransferases (PDAT) termed Lro1 plays a key role modulating endoplasmic reticulum biogenesis in yeast. This is interesting as we all learned that triglycerides are classically generated via the hydrolysis of phosphatic acid to diacylglycerol which is then acylated using a fatty-acyl-Coa catalyzed by diacylglycerol O-acyltransferases (DGATs). Using a catalytically depressed Lro1 variant, however, the authors provide evidence that Lro1 modulates ER membrane expansion driven by phospholipid synthesis.  Interestingly, the authors show that the subcellular distribution, and subsequent membrane turnover, of Lro1 are controlled by production of diacylglycerol via the phosphatidic acid phosphatase Pah1. The authors suggest that their data implicates a lipid-metabolic network involved in regulating endoplasmic reticulum biogenesis that involves converting phospholipids into storage lipids.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

23 January 2025

De novo lipid synthesis and polarized prenylation are involved in cell invasion

As has been known, during development, immune surveillance, and in cancer metastasis, cells must breach and clear basement membrane barriers.  It appears that large, transient, specialized lipid-rich membrane protrusions are used in this process.  One popular model system used to examine this process is the anchor cell invasion of Caenorhabditis elegans. In a study published in July, Park et al. used live imaging, endogenous protein tagging, and cell-specific RNAi show that lipogenesis and a polarized lipid prenylation drives for formation of the invasive protrusion of the anchor cells.  The SREBP-dependent expression of fatty acid synthesis enzymes POD-2 (acetyl-CoA carboxylase, ACC), FASN-1 (a fatty acid synthase), the ZMP-1 matrix metalloproteinase and the endoplasmic reticulum localized HMG-CoA reductase HMGR-1, which generates isoprenoids for polarized prenylation, are all involved.  This study highlights the importance of coordinated lipid synthesis in the formation of cellular protrusions that are essential for basement membrane invasion.

Dan M. Raben

The John Hopkins University School of Medicine, Baltimore, MD, USA

11 December 2023

Effect of Lipid Saturation on Nuclear Envelope Function

Since it became obvious that lipids do more than provide a platform to support membrane proteins, there has been a long-standing and growing interest among many lipid and membrane investigators regarding the role of lipids in the regulation of membrane protein function.  While most of the attention has been given to plasma membrane lipids, there has been some interest in the role of lipid residing in intracellular membranes.  Initially, much of the attention was given to the endoplasmic reticulum (ER) given its role in secretion and lipid metabolism.  In a recent article by Romanauska and Kohler (Nature Cell Biology 25:1290-1302, 2023) have provided compelling evidence that lipid acyl chain structure is linked to the structure of the double bilayer membranes that surrounds the nucleus. This double membrane is continuous with the ER.  Interestingly, their data shows that increased lipid saturation is detrimental to maintaining the homeostasis between the nuclear pore complex and the ER.  The authors argue that increases in saturated lipids leads to micron-scale lipid phase separation of the NE/ER into rigid and more elastic domains resulting in the anomalous segregation of NPCs into the elastic phase.  Importantly, the provide evidence that these phase separation makes nuclei more susceptible to rupture leading to the leakage and exposure of chromosomal DNA.  Surprisingly, lipid droplets appear to preserve the integrity of the nuclear membrane including the nuclear pore complex.  Further work on the role of lipids on the integrity and function of the nuclear envelop will certainly uncover some fundamental aspects driving the relationship between lipids and nuclear functions.

Daniel M. Raben,

The Johns Hopkins University School of Medicine, Baltimore, MD, USA

29 November 2023

Last week, I read with interest a new paper from Anaisa Ferreira, Martin Giera and colleagues, published in Nat Comms (https://www.nature.com/articles/s41467-023-43315-x#Sec2) , showing that BCG vaccination leads to increased levels of LOX products in monocytes of healthy individuals. Importantly, this study also showed that this increase, along with changes in long-chain PUFA biosynthesis was required for the trained immunity response of these cells.  This follows on from other seminal studies showing that LOXs in monocytes and macrophages are not only involved in innate, but also adaptive immunity, however there’s still a lot we don’t know about this and what the lipids are specifically doing in this context from a signaling point of view.  Ferreira et al showed that pharmacological inhibition of FADS2, LXR, 5- or 12-LOX all reduced trained immunity in vitro, while SNPs in desaturases and LOX genes influenced trained immunity in human volunteers.  Several mono-hydroxy oxylipins were implicated in this process.  The question then becomes what are the lipids doing?  Several years ago, Stefan Uderhardt and Gerhard Krönke showed that 12/15-LOX, which is the most likely candidate for formation of many of these lipids in monocytes, was involved in immunologic tolerance (https://www.sciencedirect.com/science/article/pii/S1074761312001288?via%3Dihub). They showed how 12/15-LOX is required for removal of apoptotic cells via a mechanism involving oxidized phospholipids acting as a specific signal on the cell surface.  Although this sounds very different to the new study, both address the role of monocytic LOX in adaptive immunity and it’s tempting to speculate on how these findings maybe related. How the mono-hydroxy oxylipins detected in Ferreira et al are acting is so far unknown, and could include direct receptor dependent signaling or their esterification into complex lipid pools, as seen in Uderhardt.  Follow on work to address these questions will be very interesting.

Valerie O'Donnell, Cardiff University.

13 November 2023

Sphingolipids are a major class of membrane lipids that play vital physiological roles, especially in the nervous system.  It has been long-established that gangliosides, the glycosphingolipids found in numerous membranes, serve as receptors for several bacterial toxins and viruses and interact with and modulate the function of other cellular receptors. The peptide (sequence CGSPGWVRC) binds to the surface of pulmonary primary vascular endothelial cells contributes to emphysema-like changes.  In a recent article by Staquicini et al. (Proc Natl Acad Sci U S A. 2023 Aug 22;120(34)) they have identified the receptor to be C16-ceramide.  These authors show that CGSPGWVRC activates acid sphingomyelinase and ceramide production, in the absence of apoptotic signaling, leading to the formation of ceramide-rich platforms.  Interestingly, these authors further showed that the targeting of the peptide to C16-ceramide can be used as a bioinorganic hydrogel for pulmonary imaging as well as a ligand-directed lung immunization tool against COVID-19. This study has provided evidence for yet another new role for sphingolipids.

Daniel M. Raben,

The Johns Hopkins University School of Medicine, Baltimore, MD, USA

30 October 2023

Two weeks ago, via the magic of Zoom, I sat in on a lab meeting at Bruce Hammock’s lab, which was being presented by Nils Schebb, over visiting from Germany.  A large part of his talk concerned an under-researched but very interesting area…. The occurrence of oxylipins in food.  Those of us working biomedical applications of oxylipins generally think only of them as being produced endogenously. We don’t consider how we might be exposed to them through other routes, and food, especially that containing fats that have undergone some oxidation (heat, age, etc) or express lipoxygenases and other PUFA oxidizing enzymes, are an obvious potential source.  Nils focused in two recent papers, Koch et al, available here (10.1021/acs.jafc.2c04987, 10.1021/acs.jafc.3c02724) on the occurrence of C18 oxFA in flaxseed, sunflower and rapeseed oil.  He and his team identified a large number of new mono-OH lipids derived from linoleic (LA) or a-linolenic (ALA) acids, and present in higher abundance than the most often studied 9- and 13- species from either LA or ALA.  In several oils, they were present at up to 0.1% of the oil content, which is pretty significant.  We have to wonder what they might be doing there and how this might impact human health.

The work raises many new and interesting questions. Oxylipins generated enzymatically and non-enzymatically don’t only include mono-hydroxy-FA, they also include truncated reactive aldehyde species, and for PUFA with more double bonds, others with complex PG-like ring structures may also form, so what else is there? Analysis was carried out after hydrolysis to remove FA from complex lipids, so are these free oxylipins or mainly esterified to glycerides or phospholipids? Considering that refined oil consists of >99% triglycerides (fat), the oxylipins in vegetable oils should be largely bound to this species. Schebb et al also expertly used GC/MS to work out double bond position of these new lipids, so it will also be interesting to see how new LC/MS methods like UVPD might be useful in future work of this kind.

Last, what are the biological consequences once these lipids are first exposed to high acid levels in the stomach then hydrolysed to release the free oxylipins in the gut by lipases during absorption. Oxylipins have diverse biological effects, including both pro- (e.g. EP, DP, TP receptors during inflammation) and anti-inflammatory (e.g. PPARg signaling) and the likely impact of these will of course be context dependent. Much work remains to understand the implications of this work and to further our knowledge of oxylipin environmental exposure on human health.

Val O'Donnell

Cardiff University

17 October 2023

The biophysical properties of membranes is critically important to membrane function.  This architecture can influence a number of physiologically important functions including transport, signaling, cell recognition, fusion, and interactions with the cytoskeleton.  Such biophysical properties depend on the fluidity membranes which depends on the length and degree of unsaturated fatty acid components of constituent phospholipids.  It has been recognized that the degree of saturation in eukaryotes depends on desaturases that require molecular oxygen.  This raised an interesting question as to how certain organisms, such as Schizosaccharomyces japonicus can grow in both aerobic and anaerobic conditions.  In a recent study, Panconi et al J. Biol. Chem. (2023), use microscopic, lipidomic, and molecular dynamic approaches to show that these organisms modulate membrane fluidity, at 24⁰C, but interestingly not 37⁰C, by increasing the amount of asymmetric tail phospholipids, such as 18:0 and 10:0, allowing for increased fluidity, in response to anoxic conditions.  It is noteworthy that this was not seen in a related species of yeast, Schizosaccharomyces pombe.  This gives S. japonicus a growth advantage under anoxic conditions showing once again lipids come to the rescue!

Daniel M. Raben,

The Johns Hopkins University School of Medicine, Baltimore, MD, USA

10 October 2023

I was very interested to read Bill Christie’s recent blog on the challenges of dealing with big data in lipid research, which is nowadays exemplified by the application of informatics approaches to lipidomics.  It’s inspired me to add my own thoughts, relating to the task of assessing and ensuring data quality at the level of raw data.  The launch of new software is often accompanied by claims of its superiority in relation to existing tools.  However, it’s not till the community have time to evaluate software, that we can agree on how robust, accurate and reliable a tool really is and this can take time. Software is increasingly being asked to “adjudicate” on whether a lipid is present in a sample based on appearance of chromatographic and MS/MS data.  In the past, this was performed by visual/manual inspection which when done correctly, works very well. However, when we move to large datasets, we often need to use software to automate and increase throughput.  Great care is needed with this because computers only do as good a job as they are programmed to do, and they don’t replace common sense.  Different software tools will use distinct algorithms for the “same” job, and comparing their outputs can show wildly different results. How can we deal with this?  We have to exercise caution by sanity checking our data using our own eyes, especially where we are analysing low abundance lipids.  Understanding how a tool works “under the bonnet” is also important. Algorithms making spurious claims from raw MS data isn’t a new problem as those familiar with the story about proteomics MS in ovarian cancer biomarkers may remember.  If you haven’t heard of this, check out this study from 20 years ago, which showed that processing noise had allowed cancer patients to be distinguished from controls, using SELDI-ToF MS: https://academic.oup.com/bioinformatics/article/20/5/777/214156?login=false.  

Val O'Donnell

Cardiff University

26 September 2023

DIESL: A New DGAT Regulated by TMX1 for the Synthesis of TAGs in Mammalian Cells

A paradigm in metabolism is the recognition that triacylglycerols (TAGs) represent a major source of stored energy in a variety of organisms from bacteria to humans.  TAGs are composed of three fatty acids esterified to the three carbons of glycerol.  There is great interest in these neutral lipids as their synthesis and metabolism play roles in a variety of physiological and pathophysiological processes.  It has been long recognized that TAGs are synthesized via in humans by the condensation of coenzyme A-conjugated fatty acids to the carbons on glycerol  This reaction has long been recognized to be catalyzed by two diacylglycerol O-acyltransferases (DGATs): DGAT1 and DGAT2.  Interestingly, other organisms possess additional enzymes for the generation of TAGs but has not been clear whether alternative pathways also exist in humans.

In a recent report by McLelland et al Nature 621:171–178 (2023) and see review by Schaffer, JC in Nature 621:47-48 (2023) has solved this mystery by identifying a novel pathway for the synthesis of TAGs in mammalian cells.  The authors performed a loss of function CRISPR screen using haploid human cells in which both DGAT1 and DGAT2 had been knocked out. In this screen, they discovered that the elimination of a transmembrane thioredoxin (TMX1) led to an increase in TAG synthesis.  As TMX1 does not catalyze the synthesis of TAGs the authors screened for genes that led to a decrease in TAG synthesis when disrupted in the haploid cells also lacking TMX1. They identified a protein they called DAG1/2-independent enzyme synthesizing storage lipids (DIESL).  Further analyses identified DIESL as another DGAT but one that is inhibited by TMX1.  Interestingly, DIESL appears to use phospholipids, or phospholipid precursors, as a source of TAG fatty acids.  This is an exciting discovery and future work to uncover the regulation and roles of this enzyme is sure to lead to other important discoveries.

Daniel M. Raben,

The Johns Hopkins University School of Medicine, Baltimore, MD, USA

11 September 2023

As I have mentioned on several occasions, I feel that I have missed out in never having had access to modern mass spectrometry methods for analysis of intact lipids. However, I am sure that I am not alone in finding the vast amount of data that is now produced completely indigestible. I know that moves are underway to improve comparisons of data between labs, but as an independent observer I find that I can use very little of what I read in my web pages other than general conclusions. For example, in comparing lipid compositions from different species, organelles, etc, I have tabulated data from publications that are more than 50 years old, simply because I can find little comparable that is more recent.

I have just been reading an excellent review on lipidomics information, but in 27 pages and 260 references, there are no tables of compositional data or even graphical illustrations (Sarmento, M.J. et al. The expanding organelle lipidomes: current knowledge and challenges. Cell. Mol. Life Sci., 80, 237 (2023); DOI). This is not a criticism of the authors, as I understand the problem – there are simply far too many data points from each lipid in every study, especially when positional distributions in glycerolipids are taken into account. I would like to see a table in each publication (or in the supplemental information) in which the data are simplified by aggregating molecular species to give the total amount for each lipid class. Then, within each lipid class, molecular species should be tabulate as the total, saturated, monoenes, dienes, etc. Finally, the fatty acid positional distributions (and/or totals) within each glycerolipid class should be determined by aggregating the results for each molecular species. It should not be a major task to devise a computer programme to do this.

I am not advocating that the individual data points are ignored, and data must continue to be expressed as at present. My suggestion is for an additional way to present the information – not an alternative. My concern is for the external observer, who wants the big picture as well as the minutiae.

Bill Christie

The LipidWeb, Dundee, Scotland

5 September 2023

Last week, I attended the annual meeting of the International Lipidomics Society (ILS) in Vienna. A key part of harmonization and standardization within lipidomics is the correct lipid identification and quantification by mass spectrometry. To support this, ILS is now developing minimal lipid analysis guidelines, which will describe how to properly identify and quantify lipids from raw mass spectrometry signals.  The focus will be on low abundant lipids such as oxylipins, that includes octadecanoids, eicosanoids (e.g., prostaglandins, leukotrienes, thromboxanes) and docosanoids as well as oxygenation products of n3 and n6 fatty acids termed specialized pro-resolving mediators.

I’m delighted to serve as the Chair of the new Interest Group on Oxylipin Analysis Guidelines that the International Lipidomics Society have set up. I’m grateful especially to Kim Ekroos and Gerhard Liebisch for productive and positive discussions on this topic, and for those who have agreed to act as our initial core group to set out the principles under which we will operate. These are: Makoto Arita (Riken, Japan), Craig Wheelock (Karolinska Institute, Sweden), Nils Schebb (Uni Wuppertal, Germany), Hubert Vesper (Centre for Disease Control and Prevention, USA) and Miguel Gijon (Cayman Chemical, USA). Once a plan is in place, the Interest Group will be opened up and all who are interested can join us to be part of guideline development. We encourage you to get involved by joining us on zoom and developing a published guideline.

Valerie O'Donnell

Cardiff University

18 August 2023

Interesting Chemistry Underlying the Synthesis of GDGT in Archaea

Understanding the chemistry of lipid metabolizing enzymes is often challenging, but insights can be fascinating. Take, for example, the synthesis of isoprenoid-based ether-linked membrane lipids in Archaea.  These lipids are important because they enable these organisms to withstand extreme environmental conditions. In some archaea, like Methanocaldococcus jannaschii, it has been shown that these lipids may form macrocyclic diether lipids or macrocyclic glycerol dibiphytanyl glycerol tetraethers (GDGT). Interestingly, GDGT is a membrane-spanning lipid that contains covalent bonds between the terminal carbons of the inner and outer leaflets of the membrane, providing enhanced membrane stability. While the mechanism underlying the formation of these unique lipids was obscure for many decades, work from Squire Booker’s laboratory at The Pennsylvania State University has elucidated it (Lloyd et al. Nature. 2022 Sep;609(7925):197-203 DOI). The reaction is unique in that it involves coupling two inert sp³-hybridized carbon centers. In vitro mechanistic studies indicate that C(sp³)–C(sp³) bond formation occurs on fully saturated archaeal lipid substrates and proceeds through an intermediate containing a bond between the substrate carbon and a sulfur ion of an auxiliary [Fe₄S₄] cluster to stabilize a transient carbon-centered radical. This work finalizes the biosynthetic route for GDGT formation and reveals the first instance of C(sp³)–C(sp³) coupling in nature.

Dr. Squire Booker
Evan Pugh University Professor of Chemistry and of Biochemistry and Molecular Biology
Pennsylvania State University
Eberly College of Science
Department of Chemistry

Mr. Cody Lloyd
Pennsylvania State University
Eberly College of Science
Department of Chemistry

7 August 2023

A few weeks ago, Matt Conroy and I had the pleasure of attending the Annual EpilipidNET meeting hosted by Justine Bertrand-Michel in Toulouse, France (https://www.epilipid.net/).  This was the penultimate annual meeting of this pan-European EU COST Network, led by Maria Fedorova and Rosario Domingues, which has over the last 3 years brought together over 390 researchers from 47 countries, far beyond Europe, to facilitate a huge range of lipidomics activities.  Huge credit is due to the leadership of EpilipidNET, for driving this initiative, which has transformed the profile of lipid research in Europe in a very short time, bringing together interest groups across diverse remits including:  Plant and Algal Lipidomics, Lipidomes of Common Model Organisms, Bacterial Lipidomes and Subcellular Lipidomics.  All these areas are of direct relevance to LIPID MAPS, since provision of data on structures, reactions pathways and lipid metadata and making this information freely available to the community is a common goal of both initiatives.  

If you are a young (or not so young) researcher, either experienced or new to the field, be sure to check out EpilipidNETs activities.  The last annual meeting will be in Dresden in 2024, but before this, there are several other meetings and workshops including a hackathon on curation of model organisms lipidomes.  All events and activities are free of charge to attend, and bursaries are often available too.

Valerie O'Donnell,

School of Medicine, Cardiff University, UK

24 July 2023

A family of proteins referred to as the ORMs/ORMDLs serve as regulatory subunits of the rate-liming enzyme in the synthesis of sphingolipids, serine palmitoyltransferase (SPT) complex.  This complex is known to be homeostatically regulated by cellular sphingolipid levels, but how cells sense these levels has been a matter of controversy.  This controversy seems to now be resolved.  In a recent article by Xie et al (Nat. Commun. 2023, Jun 13;14(1) 3475;  DOI) the authors show that purified human SPT-ORMDL complexes are directly inhibited by ceramide.  This was accomplished by solving the cryo-EM structure of the SPT-ORMDL3 complex in a ceramide-bound state, demonstrating a specific ceramide-binding site within the complex.  Furthermore, structure-guided mutational analyses demonstrated that this ceramide binding induces and locks the N-terminus of ORMDL3 into an inhibitory conformation. Interestingly, the authors note that childhood amyotrophic lateral sclerosis (ALS) variants in the SPTLC1 subunit cause impaired ceramide sensing in the SPT-ORMDL3 mutants.  This exciting work reveals the molecular basis of ceramide sensing by SPT-ORMDL as well as the functional consequences of this interaction and suggests an important role of impaired ceramide sensing in disease.

Daniel M. Raben,

The Johns Hopkins University School of Medicine, Baltimore, MD, USA

7 June 2023

There have been numerous studies focused on the generation and metabolism of lipid droplets. Despite these studies, our understanding of how these droplets develop from the endoplasmic reticulum (ER) has been incomplete. For example, triacylglycerol and cholesterol esters are two of the most abundant neutral lipids in these structures but they are very different molecules with very different physical properties. This is highlighted by the fact that triacylglycerols melt at -4°C while cholesterol esters melt at a much lower temperature of -44°C. A recent article from Abdou Rachid Thiam’s laboratory, and in collaboration with Ilp Vattulainen and Elina Ikonen, data are presented that indicate cholesterol esters can form supercooled lipid droplets in the presence of triacylglycerols (Nat. Commun., 14, 915 (2023);  DOI). These authors demonstrate that cholesterol esters form supercooled lipid droplets above 20 mol% with respect to triacylglycerol levels, and liquid-crystalline phases when the level increases to above 90 mol% at 37°C. They further show that at physiological temperatures, seipin-mediated triacylglycerol clusters catalyze the nucleation of cholesterol esters in the ER bilayer to initiate the formation of lipid droplets. Their data are particularly interesting given, as suggested by their melting temperatures, cholesterol esters would be expected to form a crystalline phase at physiological temperatures, but in the presence of triacylglycerols these lipids are condensed into nascent lipid droplets. Their data not only provides insights into the formation of lipid droplets, it suggests how macrophages generate cholesterol ester-rich lipid droplets leading to foam cells, as well as how spatially distinct other lipid droplets can form for other physiological processes such as steroid hormone synthesis.

Daniel M. Raben,
The Johns Hopkins University School of Medicine, Baltimore, MD, USA

24 May 2023

Each animal cell can contain up to 1000 distinct molecular species, with each lipid class containing multiple combinations of the fatty acid components. It seems likely that a high proportion of these are simply present to provide the correct blend of physical properties required for the structural function of a lipid in membranes, but it is surprising how many individual molecular forms have been recognized as having unique biological roles within tissues. One good example of this is 1-palmitoyl-2-oleoyl-phosphatidyl-sn-glycerol of lung surfactant, which attenuates inflammation by antagonizing the cognate ligand activation of the toll-like receptors (TLR2/1, TLR3, TLR4, and TLR2/6), while it disrupts the binding of virus particles to the plasma membrane receptors required for viral uptake in host cells, including influenza and SARS-CoV-2 viruses (Numata, M. et al. The anti-inflammatory and antiviral properties of anionic pulmonary surfactant phospholipids. Immun. Rev., in press (2023);  DOI).

It has been recognised for some time that the 18:0-18:1 species of phosphatidylserine has a distinctive role in membranes probably through physical interaction with sphingolipids (Skotland, T. and Sandvig, K. The role of PS 18:0/18:1 in membrane function. Nature Commun., 10, 2752 (2019);  DOI). More surprising is a recent publication demonstrating that a bacterial species from human gut, produces a phosphatidylethanolamine species with two different branched chain components (anteiso-15:0 and iso-15:0) that has remarkable specificity for immune signalling in its host via a toll-like receptor TLR2-TLR1 heterodimer; no other combination of acyl groups works (Bae, M. et al. Akkermansia muciniphila phospholipid induces homeostatic immune responses. Nature, 608, 168-173 (2022);  DOI).

The LipidWeb, Dundee, Scotland

17 May 2023

Some biochemical terms can have needlessly complex or obscure meanings, so it is always pleasing to find terms that are immediately understandable and useful, such as ‘flippases’ and ‘scramblases’ for proteins that mediate the movement of phospholipids between the leaflets of membrane bilayers. Flippases direct phosphatidylethanolamine and phosphatidylserine to the cytoplasmic leaflet (floppases work in the opposite direction), while scramblases as the name suggests randomly scramble phospholipids between leaflets across the membrane and collapse the membrane asymmetry. In particular, the latter can transfer phosphatidylserine to the outer leaflet where its exposure on the cell surface is an ‘eat-me’ signal to macrophages, another memorable term (we Scots would call them ‘couthy’). A new review on the topic is worth a read (Sakuragi, T. and Nagata, S. Regulation of phospholipid distribution in the lipid bilayer by flippases and scramblases. Nature Rev. Mol. Cell Biol., in press (2023);  DOI).

Pick up any newspaper and you will see that artificial intelligence (AI) is giving concern for any number of reasons. I understand that one such programme passed a US bar exam with flying colours, and universities world-wide are concerned with their use to cheat in essays. There are also worries that they are being used to create bogus scientific publications, and I have seen so many review articles on ferroptosis especially lately – I will say no more! There are several commercial programmes available that purport to improve the standard of written English - a worthy objective, and I am sure that they could be of legitimate value especially for those who have difficulty with the language. Are they too open to abuse?

Bill Christie,
The LipidWeb, Dundee, Scotland