Skip to main content

Targeting iNKT cell and adipocyte crosstalk for control of metabolism and body weight

Periodic Reporting for period 2 - FAT NKT (Targeting iNKT cell and adipocyte crosstalk for control of metabolism and body weight)

Reporting period: 2018-03-01 to 2019-08-31

The major goals of the project is to understand the effects of the immune system on systemic metabolism and weight, so that we may develop methods to target immunometabolic pathways for protection against obesity-induced diseases in the clinic. Obesity has reached epidemic proportions globally. At least 2.8 million people die each year as a result of being overweight or obese, the biggest burden being obesity-related diseases. Obesity is far more complex and touches more aspects of biology than previously appreciated. In the last 20 years, it is has become clear that inflammation, particularly in adipose tissue, interferes with insulin signaling and alters adipocyte function, causing insulin resistance, diabetes and worsening obesity. Over the past ten years, numerous studies have emphasized the role of adipose-resident immune cells in metabolic diseases such as obesity and diabetes. Recognition that the immune system can regulate metabolic pathways has prompted a new way of thinking about diabetes and weight management. Despite much recent progress, most immunometabolic pathways, and how to target them, are currently unknown. Thus, targeting this pathway has potential to reduce inflammation in obesity and metabolic disorder like diabetes. One such pathway is the cross-talk between invariant natural killer (iNKT) cells and neighboring adipocytes. iNKT cells are the innate lipid-sensing arm of the immune system. Since our discovery that mammalian adipose tissue is enriched for iNKT cells, we have identified a critical role for iNKT cells in regulating adipose inflammation and body weight. The goal of this project is to use a multi-disciplinary approach to identify key signals and molecules used by iNKT cells to induce metabolic control and weight loss in obesityWe recently found that invariant natural killer T (iNKT) cells, innate-like lipid-sensing T cells, are highly abundant in lean adipose tissue, limit adipose tissue inflammation, and protect against weight gain and diabetes in mice and humans. The significant of the results has allowed us to find that both the adipocytes themselves and the lipids present in adipose tissue control the unique anti-inflammatory features of adipose iNKT cells. The ultimate goal is to identify novel targets for immune-based treatments of patients with metabolic disorders.
Objective 1: Determine which cells present antigens to adipose iNKT cells.
iNKT cells are activated by lipid antigens presented by CD1d molecules on antigen presenting cells (APCs), and this lipid-CD1d complex is recognized by the invariant TCR. Surprisingly adipose tissue contains the highest level of CD1d in the body, but the significance of this is yet to be discovered. Thus, we investigated what happened if only adipocytes didn’t express CD1d. Given the importance of the microbiome in obesity, we generated Adipo1-CD1d-/- and littermate controls. Adipo1-CD1d-/- were significantly heavier than littermate controls. We found that iNKT cells in adipose tissue are uniquely anti-inflammatory and regulatory, unlike iNKT cells elsewhere., and that the transcription factor E4BP4 controls their regulatory IL-10 production, and E4BP4 is uniquely expressed in adipose iNKT cells. Adipose iNKT cells also express Nur77, indicative of chronic stimulation through their T cell receptor. This suggests that they see antigen in adipose tissue, which must be presented by CD1d on antigen presenting cells, in this case possibly adipocytes. We found that In the absence of CD1d on adipocytes, iNKT cells expressed significantly less Nur77, however, adipocyte express ion of CD1d did not impact iNKT cell levels in adipocyte tissue or their high proliferation rate. Thus, adipocyte expression of CD1d is not required for iNKT cell recruitment or survival in adipose tissue but is important for their Nur77 expression. We also repeated these studies in macrophage specific CD1d ko mice and it did not effect Nur77 expression.
In the absence of iNKT cells (CD1dko mice), obesity is more severe, and when iNKT cells are activated, mice lose weight through thermogenesis. We investigated if the increased obesity and more severe diabetes was a result of increased energy intake or decreased energy output. The ERC Stg grant provided funds for a metabolic cage system, the first of its kind in Ireland. We placed CD1dko and WT controls in metabolic chambers (CLAMS), and found that despite increased obesity, CD1dko mice ate less and had the same activity as WT controls. However, CD1dko mice had a lower metabolic rate, highlighting that the obese diabetic phenotype was due to decreased energy expenditure. The next step is to investigate whether CD1d expression on adipocytes is important for this phenotype. This work is in progress as we are currently generating enough adipocyte-specific CD1dko and littermate controls for these experiments.
Objective 2: Identify novel lipid antigens for iNKT cells in adipose tissue.

We have made progress in the goal of identifying lipid antigens for iNKT cells in adipose tissue. We found that splenic iNKT cells cultured with adipose tissue or with adipose-conditioned media become activated and upregulate E4BP4 and IL-10. We first took an unbiased approach and fractioned the adipose-conditioned media into several fractions including a protein and lipid fraction. To date, we have designed several controls and we found that it was only the lipid fraction that induced iNKT cell activation and E4BP4 expression. We are currently further fractioning the lipid fraction into free fatty acids, TAGS, DAGs, MAGs, cholesterol etc. However, in addition to cognate ligand stimulation of adipose iNKT cells, we also found that the regulatory nature of adipose iNKT cells was independent of CD1d expression.

Objective 3: Identify soluble ‘weight-loss/ thermogenic’ factor produced by adipose iNKT cells.
Adipose iNKT cells can regulate body weight, although the mechanism is not fully understood. One major goal was to publish the data on iNKT cell induced weight loss. We identified that iNKT cells induced weight loss through browning of white fat, which turned on b-oxidation and generated heat. We identified FGF21 as a critical step in this process and found that in the absence of FGF21, iNKT cell induced weight loss was significantly less. This was published in Cell Metabolism. In addition, this publication was selected to appear in BioCentury Innovations magazine. Work in progress includes sequencing iNKT cells in adipose tissue after aGalCer injection to identify what soluble factor they produce that induces We are currently analyzing the data, and to date, we have found several soluble factors produced by adipose iNKT cells after aGalCer, that are not produced by splenic iNKT cells, and which may play an important role in insulin sensitization and in induction of thermogenesis. These include meteorin-like which is exciting as The Bruce Spiegelman Lab have shown that meteorin-like can induce browning of white adipose, but it is not known that it can be produced by cells of the immune system, and in this case, specifically adipose iNKT cells.
However, during the course of this project we discovered that adipose iNKT cells are not a homogeneous population, and while the majority produce IL-10 ad little IFNg, there are subsets that produce mutually exclusive cytokines (IFNg, IL-17, IL4 and IL-10). Thus we have performed single cell sequencing on adipose tissue iNKT cells, compared to splenic iNKT cells for the first time. We have also repeated the sequencing in different physiological settings such as after exercise and cold exposure to determine which subsets respond and how. My PhD student has recently learned bioinformatics to analyse the single cell sequencing data which we received this week.
Firstly we have established the first metabolic core in Ireland. Through this ERC grant, I purchased a complete lab animal monitoring system (CLAMS) to study whole body metabolism in vivo. We also purchased light and temperature cabinets to manipulate the system to study circadian rhythm and response to changing temperatures, for which adipose tissue is critical. Using the CLAMS we are able to monitor metabolic parameters including oxygen consumption (VO2), carbon dioxide production (VCO2), respiratory exchange ratio (RER), food consumption, locomotor activity levels, and (using implantable transmitters) core body temperature and heart rate. Combining these measurements with mice fed HFD or SFD we will be able to greatly understand the in vivo metabolic and cellular changes during obesity and cold.
In addition we performed state of the art single cell sequencing on adipose iNKT cells for the first time. We have also performed this in response to these perturbations, which we predict will highlight key roles of adipose resident iNKT cells in different physiological but non pathological settings.