‘Fatty retina’ — A root cause of vision loss in diabetes?

By Clay F. Semenkovich & Rithwick Rajagopal

Vision loss in diabetes, one of the most feared complications of this disease, is caused by a progressive pathogenic process known as diabetic retinopathy, or DR. Elevated blood glucose is the predominant risk factor for DR, so many people believe that glucose toxicity is the major contributor to the development of this disease. Yet, to date, no pharmaceuticals specifically targeting glucose-dependent pathways exist for DR.

Diabetes is a disease of broadly disordered metabolism that affects how cells handle lipids, amino acids and signaling networks that regulate growth and proliferation, in addition to glucose. Accordingly, abnormalities of lipid metabolism are common in diabetes. For example, patients with diabetes often suffer from non-alcoholic fatty liver disease, which is characterized by chronic positive energy balance causing increased lipid synthesis and elevated levels of hepatic triglycerides. Thus, we reasoned that the retina might switch its lipid metabolic programming in response to an abundance of fuel in diabetes.

To test this possibility, our group studied the pathways that govern retinal lipid biogenesis (the process of synthesizing fatty acids from small precursors) during experimental diabetes in mice. In multiple models of diabetes, we observed a roughly 70% increase over controls in the synthesis of retinal palmitate — a ubiquitous saturated fatty acid that forms a basic building block for many lipids. This shift in lipid production was likely due to elevated glucose alone, as isolated retinal tissue exposed to high glucose showed the same increase in palmitate production.

Mechanistically, high glucose levels increased enzymatic activity of two regulatory enzymes: acetyl Co-A carboxylase and fatty acid synthase, or FAS. Mice with partial FAS loss-of-function in rod photoreceptors — the predominant cell type of the retina — were spared from vision loss due to diabetes, even though they developed severe systemic metabolic disease on par with control mice. Conversely, mice with FAS gain-of-function developed vision loss in half the time as wild-type mice after induction of diabetes. Taken together, our results implicate increased retinal FAS activity and elevated palmitate as root causes of vision loss in diabetes.

The mechanisms for palmitate toxicity in the retina remain elusive. Unlike in the liver, the diabetic retina does not develop intracellular lipid droplets and does not possess any significant triglyceride stores. Moreover, in comprehensive surveys of membrane lipids in the retina, we found only modest disease-associated changes. Instead, palmitate could elicit pathological signaling either through lipid second messengers or via lipidation of protein messengers. Our group is actively investigating these possibilities.

Our results shed some mechanistic light on a puzzling feature of human DR: Though glucose is the major risk factor for vision loss in diabetes, it only explains a fraction of the variability in disease progression. Differences among individuals in terms of their retinal lipid biosynthetic flux could account for some of the variance in glucose response.

Future pharmacotherapy to finely tune retinal lipid biogenesis in DR could offer a novel approach to the treatment of an increasingly common cause of visual disability.

Cardiolipin targeting remediates mitochondrial defects in cellular models of TAFAZZIN deficiency and clinical findings in individuals with Barth Syndrome

Barth Syndrome is a rare X-linked genetic disorder caused by pathogenic variants in TAFAZZIN. TAFAZZIN encodes for an enzyme involved in the remodeling of cardiolipin, a phospholipid localized to the inner mitochondrial membrane. Cardiolipin has a characteristic structure consisting of two phosphatidic acid groups connected by a glycerol backbone to form a dimeric structure. Deficient TAFAZZIN function results in an elevation of the remodeling intermediate, monolysocardiolipin, and a decrease in remodeled cardiolipin. In a paper entitled “Diverse mitochondrial abnormalities in a new cellular model of TAFFAZZIN deficiency are remediated by cardiolipin-interacting small molecules”, Anzmann et al.(2021) used shotgun proteomics to characterize effects of TAFAZZIN deficiency on mitochondrial function in a new CRISPR-edited TAFAZZIN-deficient HEK293 cell model. Anzmann et al. identified abnormalities in expression, assembly, and function of Complex I of the mitochondrial respiratory chain and dysfunction of the PARL-PGAM5 mitochondrial stress-response pathway. PARL, a rhomboid protease associated with the inner mitochondrial membrane, was found to have increased expression in the TAFAZZIN-deficient cells, which correlated with increased processing of the downstream target PGAM5, both at baseline and in response to mitochondrial stress. Moreover, these defects were partially remediated via targeting of cardiolipin metabolism with bromoenol lactone, which inhibits cardiolipin deacylation, and targeting of cardiolipin stability with SS-31 (elamipretide), a tetrapeptide shown to selectively bind to cardiolipin and stabilize cristae morphology among other effects.

From a clinical perspective Barth Syndrome is characterized by childhood-onset cardiomyopathy, neutropenia, skeletal muscle defects and growth defects. The clinical applications of targeting cardiolipin in TAFAZZIN deficiency were shown by Thompson et al. (2021) in a recent paper entitled “A phase 2/3 randomized clinical trial followed by an open-label extension to evaluate the effectiveness of elamipretide in Barth syndrome, a genetic disorder of mitochondrial cardiolipin metabolism”. Thompson et al. (2021) demonstrated that after 36 weeks of treatment with elamipretide, patients had improvements in muscle strength, exercise tolerance and some cardiac parameters. Thus, on both cellular and clinical levels, modification of cardiolipin can remediate some of the effects of TAFAZZIN deficiency.

Identification of ER scramblases with crucial roles in lipid biology

Several recent papers have provided evidence that the DedA family proteins TMEM41B and VMP1 are ER-localized phospholipid scramblases. Phospholipid synthesis at the ER occurs at the cytosolic leaflet and therefore a scramblase that allows rapid, bidirectional flip-flop should be required to achieve balanced expansion of both leaflets.  These papers represent an important advance because a scramblase activity was found in ER membranes several decades ago but the proteins catalyzing this activity have been elusive.  In addition, TMEM41B and VMP1 were known to play an important roles in the production of autophagic membranes, lipoprotein secretion, regulation of lipid droplet formation, survival of neurons and as host factors for coronavirus and flavivirus infection. Papers from the Chen, Reinisch and Yang labs provide convincing biochemical evidence that the purified and reconstituted VMP1 and TMEM41B catalyze energy-independent flip-flop of phospholipid in proteoliposomes. Moreover, Huang et al. found that Tmem41b is required for lipoprotein production and lipid homeostasis in the mouse. Liver-specific inactivation of Tmem41b causes a substantial loss of plasma lipids and reduction of lipoprotein production in the ER. Lipids presumably bound for export by the hepatocytes are instead stored in cytosolic lipid droplets wrapped by distorted ER membranes, leading to a fatty liver and rapid progression to nonalcoholic steatohepatitis (NASH). This phenotype is exacerbated by the failure of these Tmem41b-deficient cells to properly downregulate SREBP to suppress sterol and lipid synthesis.  These ER scramblases appear to be acutely needed to support large fluxes of lipid from the ER membrane to either luminal structures in the case of lipoproteins, or cytosolic structures like lipid droplets and autophagosomes. In the latter process, Ghanbarpour et al. provide a compelling model for how a tandem protein array of the ER scramblase (TMEM41B and VMP1), a lipid transfer protein (ATG2) and an autophagosomal membrane scramblase (ATG9) collaborate to move phospholipid from the site of synthesis at the ER to the growing autophagosome membrane. The role of the scramblases is to provide balanced extraction of phospholipid from both leaflets of the ER, and balanced growth of both leaflets of the autophagosome. These papers provide an exciting new dimension in the study of membrane biogenesis.

TMEM41B acts as an ER scramblase required for lipoprotein biogenesis and lipid homeostasis. Huang et al.

https://www.sciencedirect.com/science/article/pii/S1550413121002230?via%3Dihub

TMEM41B and VMP1 are scramblases and regulate the distribution of cholesterol and phosphatidylserine. Li et al.

https://rupress.org/jcb/article/220/6/e202103105/212020/TMEM41B-and-VMP1-are-scramblases-and-regulate-the

A model for a partnership of lipid transfer proteins and scramblases in membrane expansion and organelle biogenesis. Ghanbarpour et al.

https://www.pnas.org/content/118/16/e2101562118

Plasma

Plasma membranes are asymmetric in lipid unsaturation, packing and protein shape

The manuscript "Plasma membranes are asymmetric in lipid unsaturation, packing and protein shape" (https://www.nature.com/articles/s41589-020-0529-6) integrates lipidomic, biophysical, and bioinformatic approaches to comprehensively describe the compositional and structural asymmetry of the mammalian plasma membrane. The authors conclude that the unique lipid compositions of the two leaflets of the plasma membrane bilayer impart distinct physical properties that affect the structure of protein transmembrane domains. Although it has been known for decades that lipids are non-randomly distributed in many living membranes, the precise lipid compositions and physical properties of the two plasma membrane leaflets remain open questions. Combining mass spectrometric lipidomics and classical enzymatic approaches, the authors define the comprehensive, quantitative compositions of the two leaflets of human red blood cell membranes. These compositions are intended to ground accurate and realistic models of both surfaces of the plasma membrane. While generally consistent with classical measurements, these measurements also reveal a robust asymmetry in lipid acyl chains, with inner leaflet lipids being much more unsaturated than outer. A combination of computational simulations and cellular experiments then showed that two leaflets of live cell plasma membranes have quantitatively different lipid packing and diffusivity. This structural asymmetry persists long after membrane internalization, suggesting asymmetry in endocytic membranes. Finally, these biophysical asymmetries appear to be reflected in structural asymmetries of protein transmembrane domains, whose physical characteristics reflect the biophysical profile of their membrane matrix. These protein asymmetries are observed only for residents of plasma membranes and endosomes, but not of biosynthetic organelles (i.e. ER and Golgi). Remarkably, protein transmembrane domains are asymmetric throughout Eukarya, implying that the lipidomic and biophysical asymmetries that experimentally detailed in this work may be conserved throughout the eukaryotic domain of life.

Lipidomic and biophysical homeostasis of mammalian membranes counteracts dietary lipid perturbations to maintain cellular fitness

A paper entitled "Lipidomic and biophysical homeostasis of mammalian membranes counteracts dietary lipid perturbations to maintain cellular fitness" (https://www.nature.com/articles/s41467-020-15203-1) discusses mechanisms of mammalian cellular homeostasis, in particular of membrane biophysical properties. The authors show that mammalian cells possess an autonomous, essential response designed to restore membrane physical properties following perturbations from the diet, and that this response is mediated by cholesterol. Membrane adaptiveness is a fundamental and ubiquitous response of non-thermoregulating organisms across the tree of life, from prokaryotes to cold-blooded animals. In such organisms, changes in body temperature affect membrane fluidity, which must in turn be compensated by modulation of lipid composition to maintain functional membrane phenotypes. In mammals, which do not experience large-scale changes in body temperature, the existence of such membrane responsiveness has not been widely investigated. However, it is a well-established but under-appreciated fact that mammalian membranes are challenged by dietary inputs, as dietary fatty acids and cholesterol are directly and robustly incorporated into cell membranes. The paper reports that such perturbations induce near-simultaneous, ubiquitous remodeling of cellular lipidomes in various mammalian cells and that this lipid remodeling leads to re-normalization of membrane properties. The response is mediated in part by a member of the SREBP family of lipid master regulators, with cholesterol regulation playing an important role. Finally, inhibition of this response causes cytotoxicity when membrane homeostasis is challenged by dietary lipids. These results reveal an essential mechanism of mammalian cellular homeostasis - analogous to prokaryotic homeoviscous adaptation - wherein cells remodel their lipidomes in response to dietary lipid inputs in order to preserve functional membrane phenotypes.

Major Advance in Understanding the Regulation of Sphingolipid Biosynthesis

The enzyme responsible for sphingolipid biosynthesis, serine palmitoyl transferase (SPT) which condenses serine and palmitic acid to generate sphingosine. This enzyme is inhibited by ceramide and this ceramide-mediated regulation is mediated by ORMDL proteins. While this has been known for some time, the mechanisms of SPT-mediated sphingosine biosynthesis and its regulation by ORMDLs have remained unresolved. Two back-to-back papers in Nature Structural and Molecular Biology by Wang et al and Li et. al. present cyro-EM structures of human serine palmitoyl transferase (SPT) in a complex with its ORMDL3 regulator. This exciting research, which garnered a News and Views comment (https://www.nature.com/articles/s41594-021-00562-0) provides a molecular and mechanistic understanding of the synthesis and regulation of sphingolipids. They show the organization of the ORMDL3 complex with the SPT and how this organization affects sphingolipid biosynthesis. These articles are sure to attract a great deal of attention and will have a major impact on our understanding of sphingolipid biochemistry and biology. (Note: an ASBMB Lipid News article highlighting this paper is scheduled to appear in the October, 2021 issue of A-Today).

Structural insights into the regulation of human serine palmitoyltransferase complexes Wang et al.
https://doi.org/10.1038/s41594-020-00551-9 (2021))

Structural insights into the assembly and substrate selectivity of human SPT–ORMDL3 complex Li et al
https://doi.org/10.1038/s41594-020-00553-7 (2021)

Suggest a Trend

We are currently looking for article suggestions to be featured here on Lipid Trends. If you have a suggestion, please let us know.