Members

Obesity-linked insulin resistance is the most common precursor to the development of type 2 diabetes. Our previous work has shown that phosphorylation of PPARy at serine 273 stimulates a pattern of gene expression in obese adipose tissues associated with insulin resistance and type 2 diabetes. Inhibition of this modification is a major therapeutic mechanism for anti-diabetic drugs acting on PPARy, such as the thiazolidinediones (TZDs) and partial/non-agonists. Despite being powerful insulin sensitizing agents, the side effect profile of the TZDs limits their clinical utility. We would like to understand how the insulin sensitizing properties of the TZDs are linked to their ability to block PPARy phosphorylation in order to rationally target this pathway using safer new approaches. Development of new tools for obesity research: The Banks lab also directs the Energy Balance Core to study metabolic alterations which lead to obesity in mice. This equipment includes an indirect calorimeter containing gas exchange sensors to measure rates of conversion O2 into CO2, and sensitive scales to measure food intake. The large datasets produced by indirect calorimetry previously produced a statistical bottleneck. Our group developed CalR, a free indirect calorimetry data analysis tool now used on more than 30,000 experiments. We are continuing to develop new software tools for metabolic analysis.

In recent studies from our laboratory, we discovered that liver-directed A20-based therapies in mouse models of T1D and T2D, restored or significantly improved glycemic control in ways that fulfilled safety criteria. Bioengineering mouse livers to overexpress A20, led to unexpected reversal of hyperglycemia (i.e., diabetes), or at the very least to significant improvement of glycemic control in diabetic mice. A20, also known as TNFAIP3, is a gene that we showed, over 2 decades ago, to be a critical component of our physiologic anti- inflammatory defense mechanisms1. Previously established effects of A20 in the liver, mostly reported by our group, related to its anti-inflammatory, anti-apoptotic, and pro-regenerative functions, but there was no indication that it could also improve glucose metabolism, until our recent discovery. Mechanistically, we documented that overexpression of A20 in the liver positively impacted local hepatic glucose metabolism, and also systemically improved the regulation of glucose metabolism in other organs and tissues, mostly skeletal muscle. Liver-expressed A20 restored euglycemia and normalized glucose tolerance test (GTT) by decreasing hepatic gluconeogenesis and increasing peripheral glucose uptake. Importantly, A20 restored glycemic control in an insulin-independent manner, and without causing hypoglycemia, even under fasting conditions, including in the NOD mouse model of auto-immune diabetes. Additional benefits of hepatic expression of A20 include A20’s positive impact on lipid metabolism, which led to improved non-alcoholic fatty liver disease (NAFLD) in a mouse model of type II diabetes The case for A20 is further supported by our previously published data showing that physiologic protein levels of endogenous A20, whether in blood vessels or in the liver, are significantly decreased in the presence of high glucose levels i.e. badly controlled diabetes, hence the need to restore its expression through gene therapy in these conditions. We are currently exploring means to develop modified A20 constructs that could resist glucose, and hence reduce the level of gene therapy necessary to achieve glycemic control.

Currently, our goals are geared towards better understanding of the molecular basis for A20’s effects on glycemic control and performing key experiments to facilitate clinical translation of this novel A20-based gene therapy. Specifically, we wish to
1) Characterize the molecular basis that support the ability of liver-expressed A20 to regulates glucose metabolism.
2) Document A20’s effect on glucose uptake, hyperinsulinemic euglycemic clamps, insulin signaling.
3) Optimize AAV-based gene therapies, specifically promoter selection and enhancer elements, to ensure safe and sustained expression of A20 in livers of diabetic animals.
4) Perform pre-clinical toxicology and biodistribution studies in rodents using the optimal AAV-based gene therapy vector identified
5) Engage in additional proof-of-concept studies using a large animal model of T1D, in prelude to clinical implementation.

Our laboratory focuses on the nexus of cardiac and metabolic diseases with a particular interest in exercise biology and inter-organ communication. To expand the novelty and clinical impact of our studies, we have developed and incorporated emerging metabolomics and proteomic technologies towards the discovery of new biomarkers and pathways. We make observations in humans and then turn to cell and animal-based systems to test for causal relationships. Because metabolites and proteins are downstream of genetic variation and transcriptional changes, they serve as “proximal reporters” of physiology and may be highly relevant biomarkers for human diseases. At the same time, we leverage human genetics to understand the genetic architecture of circulating factors for pathway elucidation. Our multi-disciplinary research incorporates basic molecular and cell biology, genetics, chemistry, mass spectrometry and bioinformatics, all with a foundation in clinical medicine.

The overall goal of research in my lab is to determine the cellular and molecular mechanisms for insulin resistance in obesity and type 2 diabetes. Major research areas include:

1) discovering the mechanisms by which novel adipocyte-associated molecules alter insulin action and fuel metabolism in other tissues;
2) determining the molecular mechanisms that render obesity a risk factor for type 2 diabetes;
3) understanding the regulation and biological activities of a novel class of anti-diabetic and anti-inflammatory lipids that we discovered and named branched Fatty Acid Esters of Hydroxy Fatty Acids.
4) determining the therapeutic potential of Fatty Acid Esters of Hydroxy Fatty Acids for both Type 1 and Type 2 diabetes and immune mediated diseases.

Our work has had a major impact on understanding the important role of the adipocyte as an endocrine organ and as a metabolic “factory” consuming nutrients and substrates and producing metabolites that have systemic effects on insulin action, energy balance and inflammation. We use genomic and metabolomics approaches in mouse models we have genetically engineered mice to discover novel adipocyte-associated molecules which have provided important markers and mechanisms for insulin resistance and diabetes in humans. For example, we demonstrated that retinol binding protein 4 levels are elevated in insulin-resistant people and that elevated retinol binding protein 4 causes insulin resistance by activating both the innate and adaptive immune systems in adipose tissue. Lab members also showed that de novo lipogenesis in adipocytes has a major role in regulating systemic insulin sensitivity. These studies focused our interest on identifying novel metabolites which regulate glucose homeostasis.

Glut4 is the major insulin-regulated glucose transporter and is expressed at highest levels in skeletal and cardiac muscle and brown and white adipocytes. Reduced levels of Glut4 in adipocytes is associated with insulin resistance in humans and is a risk factor for developing type 2 diabetes. To understand the role of Glut4 and glucose transport specifically in adipocytes on glucose homeostasis, we engineered mice to have adipose-specific overexpression of Glut4. These mice have markedly enhanced glucose tolerance in spite of obesity. This improved glucose tolerance depends on increased lipogenesis in adipose tissue driven Carbohydrate Response Element Binding Protein, a transcription factor that regulates de novo lipogenesis and glycolysis. Knocking out Carbohydrate Response Element Binding Protein from adipose-specific Glut4- overexpressing mice reverses their enhanced glucose tolerance.

Since plasma fatty acids were elevated in the adipose-specific Glut4-overexpressing mice, we sought to determine whether specific lipids were being synthesized that have beneficial metabolic effects. With Dr. Alan Saghatelian, we used a global lipidomic platform and discovered a novel class of lipids that are made in human tissues, correlate highly with insulin sensitivity in humans, and have anti-diabetic and anti-inflammatory effects. These lipids lower glycemia and improve insulin sensitivity in insulin-resistant obese mice and protect against inflammatory diseases including autoimmune Type 1 diabetes in a mouse model. We are investigating the biology and pharmacokinetics of these lipids and the pathways that regulate their levels and mediate their effects. Because of the constellation of beneficial effects of these lipids and their favorable safety profile, they could lead to new therapeutic agents to prevent and treat diabetes and immune-mediated diseases.

Lab website

The Kajimura laboratory focuses on the molecular mechanisms of metabolic adaptation to stress. In this regard, fat cells (adipocytes) serve as a unique model because adipose tissue comprises a dynamic organ that remodels its cellular size and composition in response to a variety of hormonal cues, nutritional changes (e.g., overeating or fasting), and temperatures. Such metabolic adaptation, involving lipolysis, lipogenesis, adipogenesis, mitochondrial biogenesis/clearance, and thermogenesis, plays a central role in the regulation of energy homeostasis. We apply the most cutting-edge technologies and multidisciplinary approaches (biochemistry, genetics, bioinformatics, molecular biology, engineering, etc.) to generate a blueprint for engineering regulatory circuits of adaptive responses and restoring metabolic health by defined factors. This approach will have a profound impact on the prevention and treatment of metabolic disorders, cancer, aging, and beyond.

1) Metabolite compartmentalization via mitochondrial transporters: A notable metabolic change during cold adaptation is fuel utilization from glucose to fatty acids and amino acids. We recently found that, besides glucose and fatty acids, brown/beige fat cells actively uptake branched-chain amino acids (BCAA) in the mitochondria, thereby enhances systemic BCAA clearance. This is highly significant because increased BCAA levels – due to impaired BCAA oxidation in metabolic organs – are tightly associated with human diabetes. By studying the fuel switch mechanisms, we identified SLC25A44 as the first mitochondrial transporter for BCAA (Yoneshiro et al. Nature 2019). We aim to explore the biological roles of this newly identified mitochondrial BCAA transporter SLC25A44 as well as other uncharacterized transporters in health and disease.

2) Cellular and functional heterogeneity in adipose tissues: Historically, it has been considered that mammals possess “two types” of adipose cells – brown and white fat cells. However, emerging evidence suggests that adipose cell origins and composition are far more complicated than merely two types. In fact, we and others showed that beige adipocytes- an inducible form of thermogenic fat cells – exist in mice and humans (e.g., Shinoda et al. Nature Med 2015). More recently, we found that myogenic progenitors in the subcutaneous WAT give rise to a glycolytic form of beige fat (termed “g-beige” fat) in the absence of �-adrenergic receptor signaling (Chen et al. Nature 2019). It is conceivable that adipose tissues contain diverse progenitors that differentially respond to external and hormonal stimuli (e.g., exercise, tissue injury, cancer cachexia, and intermittent fasting), and each of them gives rise to developmentally and functionally distinct mitochondria- enriched adipocytes. We aim to generate a complete lineage/functional map of adipose cells in mice and humans.

3) Metabolic engineering to improve metabolic health: The “browning” of white fat – enhanced beige fat biogenesis – is accompanied by a substantial improvement in metabolic health, including improved glucose tolerance, insulin sensitivity, lipid profile, and cardiovascular health. The conventional dogma was that these metabolic effects are through UCP1-mediated thermogenesis; however, surprisingly, we demonstrated that a large part, if not all, of the anti-diabetic actions of beige fat is UCP1-independent (Ikeda et al. Nature Medicine 2017; Haseawa et al. Cell Metabolism 2017). We aim to explore this unexpected observation by 1) uncovering the mechanisms of UCP1-independent anti-diabetic actions, and 2) reconstitution of such anti-diabetic effects of beige fat in adipose tissues, i.e., fat-specific “cold mimetics.”

Identifying ROCK1 as a novel regulator of insulin signaling, glucose homeostasis, and lipogenesis: My works suggested that Rho-kinase positively regulates insulin-stimulated glucose transport and signaling via either IRS-1 serine phosphorylation or active polymerization, establishing a new mechanism for the regulation of glucose transport. My works also demonstrated that inhibition of Rho-kinase causes insulin resistance in vivo. Our studies also revealed that adipose ROCK1 isoform plays an inhibitory role for the regulation of insulin sensitivity in diet-induced obesity in vivo. Recent our work established a ROCK1-AMPK signaling axis that regulates de novo lipogenesis, providing a unique target for treating obesity-related metabolic disorders such as NAFLD. Identification of ROCK1 as a key player of glucose and lipid metabolism has a major impact on the understanding of the pathogenesis of diabetes and has significantly advanced the diabetes field.

Discovery of Rho-kinase as a key regulator of leptin action: Our recent works also have established the critical roles of the serine threonine kinase ROCK1 on regulation of leptin signaling and action in hypothalamus in vivo, using transgenic mouse models. Our study establishes a new hypothesis that ROCK1 regulates energy balance by targeting leptin receptor signaling, suggesting ROCK1 as a key regulator of leptin action. In fact, genetic disruption of ROCK1 in either POMC or AgRP neurons increases body weight and adiposity. Mice lacking ROCK1 in POMC neurons show hyperphagia and hypoactivity. However, AgRP neuron-specific ROCK1-deficient mice display lower oxygen consumption and locomotor activity. The molecular mechanism for this is involved in ROCK1-mediated JAK2 phosphorylation, which promotes downstream signaling pathways of leptin, including STAT3 and PI3K signaling, ultimately leading to the control of energy balance. This model provides a new mechanism that advances our understanding of central leptin action and energy homeostasis.

ApoJ is a new metabolic signal controlling energy homeostasis and glucose metabolism: Identification of a new molecule that cures potentially eating disorder has been a key subject of obesity field. Our recent work discovered ApoJ as a novel anorexigenic molecule that regulates appetite and energy balance. Like leptin, ApoJ treatment caused anorexia, weight loss, and hypothalamic Stat3 activation. These effects were most likely mediated by a cellular mechanism that was dependent on a physical interaction between functional leptin receptors and LRP (low-density lipoprotein receptor-related protein). However, peripheral actions of ApoJ in the context of glucose metabolism are unknown. Our studies demonstrated that ApoJ functions as a hepatokine targeting insulin signaling and glucose metabolism in skeletal muscle, which is mediated via the LRP signaling cascade. Taken together, our works identify the ApoJ ® LRP signaling axis as a novel metabolic signaling pathway that is central for the maintenance of normal glucose homeostasis and energy balance.

We utilize genetic engineering techniques in mice, electrophysiology, optogenetics, chemogenetics, rabies mapping, ChR2-assisted circuit mapping, in vivo assessments of neuronal activity, and single neuron transcriptomics to elucidate neural circuits controlling hunger and energy balance, neuroendocrine regulation and autonomic function. Neuron-specific recombinase driver mice are used in conjunction with recombinase-dependent AAVs expressing various genetically encoded “tools” to selectively (in a neuron cell type-specific fashion): a) map connectivity between neurons to establish their “wiring diagrams”, b) manipulate neuron firing rates in vivo to determine their roles in regulating behavior and physiology, and c) measure neuron activity in vivo to establish their responses to discrete behavioral and physiologic perturbations. These three parameters are key to understanding how neurons and their circuits control of behavior and physiology.

The above-mentioned approaches are powerful. However, their impact is sometimes limited by the fact that we do not know the different neurons that make up each “homeostatic” brain region (i.e., the “parts list”), and related to that, we do not have the “enabling” recombinase mice that would provide experimental access to these important, but presently unknown, neurons. We are addressing this by: 1) performing single-neuron transcriptomics on “homeostatic” brain regions to reveal: i) the different neurons that exist in each of these brain sites (i.e., the “parts list”), and ii) the genetic markers that specify each of the unique “parts” (i.e., the neurons). 2) We then use these genetic markers and CRISPR genetic engineering to rapidly generate mice expressing recombinases in each of the interesting, novel neurons. 3) Finally, we use these mice to uncover the wiring diagrams and determine the role these novel neurons play in regulating energy homeostasis and autonomic function.

Lab Website

The Rosen lab strives to understand how gene expression networks control the development, function, and pathophysiology of metabolically relevant tissues, with an emphasis on adipose tissue. We employ genetic, genomic, computational, cell biological, and in vivo modeling approaches to questions related to adipogenesis, thermogenesis, insulin action, and inflammation. Current projects include:
1. Single cell approaches to human and murine adipose tissue biology. We have generated large atlases of mouse and human adipose tissue at single cell resolution from different depots and under different nutritional conditions. These atlases are being used to identify unique subpopulations of adipocytes and other cells and to generate hypotheses about cell-cell interactions within the adipose niche.

2. Understanding the role of interferon regulatory factors at the intersection of inflammation and metabolism. We have identified IRF family members as critical transcriptional regulators of adipogenesis, lipolysis, thermogenesis, hepatic glucose metabolism and fatty liver. Studies are also underway to assess the role of IRF3 in mediating leptin’s actions in the hypothalamus. Furthermore, we have identified ISGylation as a downstream effector of inflammation in adipose tissue, and we are pursuing studies to identify ISGylated target proteins.

3. Neuropeptide regulation of adipose biology. We have identified two neuropeptides, oxytocin (OXT) and neurotensin (NTS), with profound effects on adipose lipolysis and thermogenesis. Interestingly, these neuropeptides appear to be released by cells within the adipose niche. In the case of OXT, the contributing cell is a subset of sympathetic neurons, while NTS is derived from local lymphatic vessels.

Lab website

The Saper laboratory works on neural circuitry that underlies several basic and related hypothalamic functions, including circadian rhythms; wake-sleep cycles; thermoregulation; and feeding and metabolic control. We study brain circuitry that regulates these functions using a combination of cutting-edge neuroscience methods, including opto- and chemogenetics; fiber photometry; and use of animals with loxP sites inserted in key genes, which can then be disabled by injection of adeno-associated viral vectors that express Cre recombinase. This is combined with measuring EEG/EMG, body temperature, locomotor activity, feeding, brown adipose activation, and other physiological and behavioral measures.

Broadly, my laboratory seeks to utilize next generation sequencing technologies and analysis to better understand the genetic, epigenetic, and tissue/cell type bases of metabolic disease. In particular, we study the biological and neural circuits and gene regulatory networks underlying obesity and type 2 Diabetes in the hypothalamus and adipose tissue of human and rodent models. Research areas include:

1. Molecular taxonomy of hypothalamic neurons. With Brad Lowell and John Campbell groups, we are systematically defining the hypothalamic neuron types underlying weight regulation, insulin resistance, and glucose homeostasis using single cell transcriptional profiling of key hypothalamic regions involved in energy balance. Starting with unbiased single cell profiling, we define a “parts” list of each region’s constituent neuron types and the genetic markers that specify each. We integrate these profiles with human GWAS data to suggest function and validate hypotheses using specific recombinase lines (many engineered from above profiling) to monitor or manipulate each’s activity and connectivity using genetic, optogenetic or pharmacogenetic approaches. My initial work on the arcuate nucleus was supported by a BADERC P&F award (2016) and resulted in publication of a seminal manuscript in the field (PMC5323293). Similar manuscripts on DMV, PVH, and LPBN have or will be submitted within the year.

2. Role of median eminence cell types in barrier function and energy balance. The arcuate-median eminence ooccupies a unique anatomical position at the nexus between the peripheral circulation, CSF of the third ventricle, and the brain parenchyma. Using single cell transcriptional profiling, we have molecularly characterized the cell types regulating communication between these compartments including: 1) specialized fenestrated endothelial cells of the ME that allow passage of signals to and from the periphery,2) β1 and β2 tanycytes that form the junction between ME, arcuate, and 3rd ventricle, and 3) neuroendocrine cells of the pars tuberalis, which lines the ME and sends signals to central brain. Using tools to induce cell-type specific activation or loss of function, we are characterizing the molecular pathways by which these cells regulate energy intake and insulin sensitivity.

3. Reconstructing the functional gene regulatory circuitry of adipocytes. In collaboration with Evan Rosen, we are systematically identifying the functional components of the human adipocyte epigenome, using histone modification ChIP-seq, ATAC-seq, RNA-seq and Hi-C to profile regulatory elements controlling gene expression across a spectrum of obesity and insulin resistance states. Bioinformatically, we reconstruct the molecular circuits underlying variation in transcription, determine adipocyte-specific eQTLS, and prioritize causal GWAS variants mediating adipocyte roles in obesity, insulin resistance, and diabetes.

4. Atlas of cell types in human and mouse adipose tissue. Adipose tissue plays a central role in the pathophysiology of obesity and diabetes, modulating satiety, glucose, and lipid homeostasis. Adipose tissue cell type content is heterogeneous, dynamic, and altered by obesity. Using single nucleus RNA-seq we assess functional cell type interactions across a variety of metabolic perturbations, including fasting, weight loss, high fat diet exposure, gastric bypass, and aging to determine how individual cell types interact to produce the adipose tissue dysfunction.