Background
Chronic inflammation is a central feature in the pathophysiology of insulin resistance (IR) and diabetes [
1‐
3], both of which are risk factors for the development of atherosclerosis [
4]. Indeed, IR and overt type-2 diabetes may emerge during human infections and sepsis [
5] via activation of toll-like receptor 4 (TLR-4) signaling. TLR4 also may be activated by endogenous ligands that are increased in diet induced obesity and IR [
6]. Further, experimental studies in mouse models of TLR4 deficiency demonstrate a reduction in diet induced obesity [
6] and atherosclerosis [
7].
The TLR4 pathway can be activated in humans by exogenous administration of standardized preparations of lipopolysaccharide (LPS) under Food and Drug Administration oversight. We and others have shown that insulin resistance [
8,
9], adipose tissue inflammation [
10] and atherogenic lipoprotein changes [
11,
12] observed acutely during this experimental stimulation resemble those observed chronically in obesity, insulin resistance and atherosclerosis. However, our prior experimental work, which used a moderate dose of LPS (3 ng/kg) [
8], was associated with clinical symptoms and modulation of counter-regulatory hormones which limit the potential broad application for the study of cardiometabolic diseases.
There is a great need for development of a feasible human model of low grade inflammation in which novel therapies and endogenous exposures, such as genotypic variation, can be tested for their modulation of inflammatory atherogenic stress. In this paper, we address the hypothesis that low dose endotoxemia (LPS 0.6 ng/kg) induces biologically relevant inflammatory metabolic changes in the absence of overt clinical responses. Our goal was to investigate the feasibility of this potential subclinical model in humans in order provide a tool for testing drugs, diets and genes that modulate inflammatory cardio-metabolic disease.
Discussion
In this double blind, placebo-controlled, random sequence crossover study of low-dose endotoxemia designed to minimize clinical symptoms, we observed several findings which support its potential use in the study of cardiometabolic diseases: 1) a robust biochemical systemic inflammatory response; 2) an almost complete lack of clinical responses; 3) minimal counter regulatory response in cortisol and growth hormone; 4) significant modulation of cytokine, chemokine in adipose tissue and insulin signaling pathways; and 5) induction of systemic insulin resistance without evidence of pancreatic beta cell dysfunction.
Experimental endotoxemia, which stimulates toll-like receptor 4 (TLR-4) signaling
in vivo, may be an informative model to study cardio-metabolic traits in humans [
9,
26]. Observational data show that sepsis and chronic infection [
27,
28] induce insulin resistance, glucose intolerance and lipid derangement resembling that observed in obesity, type-2 diabetes and atherosclerosis. In addition, we and others have shown that experimental endotoxemia induces insulin resistance [
8,
9], adipose tissue inflammation [
10] and atherogenic lipoprotein changes, including impaired reverse cholesterol transport [
11,
12,
26,
29,
30]. In humans, endogenous TLR4 antigens, including fatty acids [
6] and oxidized lipids [
31], are generated in obese adipose and atherosclerosis and may drive inflammatory cardio-metabolic dysfunction. Indeed, TLR4 is directly implicated in diet induced obesity [
6] and atherosclerosis through studies in TLR4 deficient mouse models. Thus, endotoxemia has strong biological plausibility and activates relevant pathways known to be perturbed in obesity, diabetes and atherosclerosis.
We previously reported systemic inflammatory changes in adipose tissue with development of a systemic insulin resistance state [
8] utilizing a moderate dose of endotoxin (3 ng/kg). Despite its utility in studying inflammatory effects on lipoproteins, metabolic function and adipose tissue in human, this proof-of-principle model is supra-physiologic and induces marked changes in systemic inflammatory markers compared to the current low-dose model and compared to that observed chronically in cardio-metabolic disease states. More importantly, moderate-dose endotoxemia is associated with overt clinical responses, including fever, tachycardia and flu-like aches [
8,
10,
15] which may limit wider application in large-scale human clinical research. In the current study, we demonstrate more subtle changes in inflammatory and metabolic responses with an absence of clinical symptoms during low-dose endotoxemia (LPS 0.6 ng/kg). Our findings support the use of this low-dose model in studying clinically relevant metabolic changes while providing a safe and scalable approach for testing novel therapeutics and genomic influences on cardio-metabolic disorders.
In moderate dose endotoxemia [
8], we observed strong induction of subcutaneous adipose TNF and IL-6 as well as MCP-1, which is known to recruit CCR-2 expressing monocytes, increase inflammatory-M1 adipose tissue macrophage (ATM) and promote insulin resistance [
22]. In support of this concept, we also observed increased mRNA levels of the macrophage marker, EMR1-F4/80 [
32], in adipose tissue. Thus, endotoxemia may promote adipose recruitment of macrophages, a characteristic of adipose tissue in obese, insulin resistant humans
33. In our current study, we demonstrate that low dose endotoxemia produced a more subtle adipose tissue inflammation. However, the increases in adipose cytokines and chemokines as well as induction of SOCS were similar to those observed in diet and obesity-related insulin resistance [
22,
32‐
37]. Further, adipose changes coincided with subsequent emergence of modest insulin resistance which was less than that with higher dose endotoxin but consistent with that in metabolic syndrome [
38] and diabetes [
19]. Thus, low dose endotoxemia also provides a model of inducible adipose tissue inflammation permitting specific interrogation of factors that modulate this important tissue component of insulin resistance.
Epidemiological studies have demonstrated a consistent relationship between chronic low grade inflammation and states of obesity, insulin resistance, diabetes and atherosclerosis. A challenge in understanding the mechanism of these associations in humans however remains hampered by a lack of a reliable
in vivo model. Here we show that an evoked inflammatory model to simulate these states can be fruitful in identifying genes and pathways activated in cardio-metabolic disease. We acknowledge that the low dose endotoxemia model does not reproduce the chronic pathophysiology of complex cardio-metabolic diseases. It is, however, associated with minimal clinical response and approximates acutely the inflammatory and metabolic responses of the chronic disease states of interest. Furthermore, low-dose experimental endotoxin induction of toll-like receptor 4 (TLR-4) signaling
in vivo is one well established model of inflammation-induced metabolic disturbances in humans. Sepsis and chronic infections in humans induce insulin resistance, glucose intolerance and lipoprotein changes similar to the metabolic profile observed in obesity, type-2 diabetes and established coronary artery disease. The insulin resistance, adipose inflammation and lipoprotein changes observed acutely during experimental endotoxemia resemble those observed chronically in cardio-metabolic disease states. Finally, the role of TLR4 (the endotoxin signaling receptor) is further suggested by studies demonstrating reduced diet induced obesity and atherosclerosis in TLR4 deficient mouse models. Indeed, in addition to endotoxin, TLR4 may be activated by endogenous ligands that are increased in diet induced obesity and insulin resistance. Of further relevance to obesity and metabolic disease
in vivo, rodents raised in germ-free conditions are protected from diet induced obesity [
39]. Therefore, our model of induced inflammation from acute bacterial exposure is one of several models of inflammation evoked metabolic disturbance
in vivo and may not exactly mimic all means of generating subclinical inflammation in obesity, diabetes and cardiovascular disease. However, this model also offers the additional advantage in permitting direct assessment of the directional impact of induced inflammation on metabolic parameters such as insulin resistance. This avoids confounding and reverse causation that are features of observational studies where inflammatory changes may result from risk factors and disease rather than be causal.
Competing interests
The authors declare that they have no competing interests.
Authors’ contribution
NNM and MPR designed and executed the study and wrote the manuscript. SPH, JEF, RDS, CH, KT, RS collected data and revised the manuscript. PNP, PK and MRR analyzed data and revised the manuscript and JTM collected the data. All authors have read and approved the final manuscript.