Interleukin-6 (IL-6) from different organs
Interleukin-6 (IL-6) was initially discovered as a pro-inflammatory cytokine, but recently it has been found to be related to metabolic functions, such as insulin resistance. Our group has investigated the regulation of body fat mass by IL6 with the aim to find new mechanisms for occurrence of obesity. Large quantities of IL-6 are produced by immune cells during inflammation. Circulating IL-6 is, to a large extent, produced by adipose tissue in sedentary and healthy individuals. After prolonged and heavy exercise, circulating IL-6 is mainly produced by skeletal muscle. Finally, there is evidence that IL-6 is produced in the CNS. As discussed below, our data suggest, however, that IL-6 produced by the CNS is most important in the regulation body fat.
Fig 1. Deficiency of interleukin-6 (IL-6) causes obesity. A normal wild type male mouse is shown to the left and a mouse with IL-6 gene knockout is shown to the right.
Interleukin-6 IL-6) prevents obesity in mice
We have found that mice lacking interleukin-6 (IL-6) due to IL-6 gene knockout developed mature onset obesity (see Fig 1). The IL-6 deficient mice also became insulin- and leptin-resistant. Peripheral treatment with a low dose of IL-6 could partly reverse the obesity in the IL-6 deficient mice, but had no effect in control mice. IL-6 deficient mice also had obesity associated metabolic changes, such as decreased glucose tolerance (Wallenius V et al 2002). The results of others have indicated that IL-6 in pharmacological doses can induce insulin resistance (Klover et al 2003). In general, there is a controversy regarding the effects of IL-6 on insulin resistance (Febbraio & Pedersen 2002). Factors such as dose, time span, target organ, and physiological context (e g exercise) may play a role.
The antiobesity effect of IL-6 could be exerted at the hypothalamus
To study the mechanism and site of action of the antiobesity effect of IL-6, we gave rats intracerebroventricular (ICV) IL-6 treatment and found that a single ICV IL-6 injection increased energy expenditure, while peripheral injection of the same dose had no effect (Wallenius V et al, 2002). Moreover, chronic treatment with IL-6 ICV during two weeks decreased body weight, total fat pad weight, and serum levels of the fat derived hormone leptin, but did not change the weight of several nonfat organs (Wallenius K et al, 2002).
These early results show that centrally acting IL-6 is an important regulator of body fat mass in rodents (see Fig 2). Recently, we obtained evidence that IL-6 deficiency in mice decreased the expression of several peptides found in the paraventricular nucleus (PVN) of the hypothalamus, which is a nucleus that has been attributed an adipostatic function. For example, corticotrophin-releasing hormone (CRH), which is reported to stimulate the sympathetic nervous system, was decreased by 40% in older IL-6(-/-) mice. Oxytocin, which is reported to prevent obesity, was also decreased in older IL-6(-/-) animals. The IL-6 receptor alpha was abundantly expressed in the PVN, but also in the supraoptic nucleus, and was shown to be co-expressed to a high extent with CRH, oxytocin and thyrotrophin-releasing hormone. These results are line with direct anti-obesity effects of IL-6 at the level of the PVN in the hypothalamus (Benrick et al 2009).
We investigated energy metabolism, core temperature, heart rate and activity in young pre-obese IL-6 knockout mice by indirect calorimetry together with telemetry. At 30 °C, the basal core temperature was lower in IL-6 -/- compared to wild type mice, while the oxygen consumption did not differ significantly. The respiratory exchange ratio at 20 °C was significantly higher and the calculated fat utilization rate was lower in IL-6 knockout mice. In response to new-cage stress, the increase in oxygen consumption at both 30 and 20 °C was lower in IL-6 knockout than in wild type mice. At 4 °C, both the oxygen consumption and core temperature were lower in IL-6 -/- mice, suggesting a lower cold induced thermogenesis. These results indicate that endogenous IL-6 is of importance for stress- and cold-induced energy expenditure in mice (Wernstedt et al, 2006).
Association between IL-6 production and obesity in humans
IL-6 levels in cerebrospinal fluid (CSF) are correlated negatively with total body fat. Moreover, in some individuals CSF IL-6 levels are higher than serum IL-6 levels. In contrast to CSF IL-6, CSF leptin levels correlated positively with total body fat. Therefore, CSF IL-6 differs in many ways from CSF leptin. CSF IL-6 may be locally produced in the CNS and not serum derived as assumed for leptin in CSF (Fig 2). Moreover, our data indicate that body fat-regulating regions in the CNS are not resistant to IL-6 in more severe obesity (Stenlöf et al, 2003).
Association between a common IL-6 gene polymorphism and obesity in humans
We have found that the weak C allele of the common –174 C/G polymorphism in the IL-6 gene promoter is associated with increased body fat mass in two different populations (Wernstedt et al 2004). Interestingly, the weak –174 C allele was recently shown to be associated with decreased energy expenditure (Kubaszek et al, 2003).
However, the association with BMI did not reach the level of significance in two meta analysis on over 25’000 subjects each (Qi et al 2007); (Huth C et al 2009). Nevertheless, in a recent study on more than over 3000 individuals, an association between variants in IL-6 gene and BMI was shown (Qi et al 2007). Moreover, there was an association between the 174G>C polymorphism and total and regional fat masses as determined by dual-energy x-ray absorptiometry (DXA) in a sample with 3014 elderly men (Strandberg et al 2008). It is possible that there are yet unknown polymorphisms in the IL-6 gene that explain the so far partly contradictory results.
IL-6 and exercise
During prolonged muscular exercise circulating levels of IL-6 in humans and animals increase by up to 100-fold (to levels considerably higher than those of even severely obese sedentary individuals (see below), while the release of pro-inflammatory cytokines, such as tumor necrosis factor-α and interleukin-1b, are much lower). The high plasma levels of IL-6 observed during exercise are mainly due to an increased production of IL-6 in the working skeletal muscle, and is not a result of muscle cell damage or infiltration of immune cells as shown by Bente Pedersen´s group (Febbraio and Pedersen, 2002).
We found that young pre-obese IL-6 knockout mice have reduced endurance during exercise. The young pre-obese IL-6 knockout mice also had decreased oxygen consumption during exercise, in a similar way as we have shown for sedentary mice (Fig 2). Taken together, these results suggest that IL-6 is an important stimulator of energy expenditure during different physiological conditions. Our study is the first to provide evidence for a physiological role of exercise induced IL-6 production (Fäldt, Wernstedt et al, 2004).
Fig 2. It is hypothesized that IL-6 produced in the CNS, possibly the hypothalamus, and by skeletal muscle during exercise enhances energy expenditure via an effect in the CNS. Lipolysis may be stimulated by IL-6 produced locally in fat tissue or by the sympathetic nervous system.
Interleukin-1 as a negative regulator of obesity in mice.
We have recently found evidence that IL-1 receptor I activation exerts a tonic antiobesity effect in mice. We found that mice with IL-1 receptor I knockout develop mature onset obesity from about 6 months of age in a similar way as IL-6 knockout mice. Moreover, IL-1 RI knockout mice showed a slight increase in longitudinal bone growth and lean body mass, and serum insulin-like growth factor-I (IGF-I). This is similar to earlier observations in mice with decreased stimulation of melanocortin-4 receptors (MC-4 R), but different from the IL-6 knockout mice. The IL-1 RI knockout mice were insulin resistant, as evidenced by hyperinsulinemia, and decreased glucose tolerance and insulin sensitivity. To elucidate the mechanisms for the development of obesity, young pre-obese IL-1RI -/- mice were investigated. They showed decreased suppression of body weight and food intake in response to systemic leptin treatment. The decreased leptin responsiveness was even more pronounced in older obese animals. Moreover, spontaneous locomotor activity and fat utilisation, as measured by respiratory quotient, were decreased in pre-obese IL-1RI -/- mice (Garcia et al 2006). Interestingly, combined deficiency of IL-1 and IL-6 in mice with double knockout causes severe early onset obesity (Chida et al 2006).
As for IL-6, we have evidence that the antiobesity effect of IL-1 is exerted at the level of the CNS presumably the hypothalamus. It is also well known that IL-1 and IL-6 exert similar effects in regulation of the immune system. The interactions between IL-1 and IL-6 in regulation of body fat at the level of the hypothalamus is one of the aims of future research in our group.
Associations between interleukin-1 system gene variants and body fat in humans.
To elucidate the clinical relevance of our results in mice, we investigated the associations between genetic variations in the IL-1 system and DXA determined fat mass in a homogenous and well characterized cohort of 18-20-year old men of the so called GOOD study in Gothenburg. The results showed that the IL-1beta gene +3953 C variant, which before has been associated with decreased IL-1 activity, is associated with increased fat mass (Strandberg et al 2006). We also found that the IL-1 receptor antagonist (IL-1RN) gene variant IL-1RN*2 (two 86-bp repeats, in complete LD with +2018 T>C or rs419598) was associated with both obesity and increases in the levels of circulating IL-1RN and L-1RN production by leukocytes ex vivo. Moreover, these two variants are in linkage disequilibrium and the haplotype is also associated with obesity (Strandberg et al 2006; Strandberg et al 2008). Recently, we have found that the rs4252041 SNP of the IL-1 receptor antagonist is associated with body fat mass (Andersson et al 2009). In summary, these results fit well with the hypothesis that IL-1RI stimulation suppresses fat mass (Garcia et al 2006). Finally, we have found that a third cytokine, leukemia inhibitory factor (LIF), also suppresses body fat mass in mice. The effect of LIF, unlike that of IL-1 and IL-6, seem to be exerted outside the CNS. (Berndtsson et al, 2006).
Side project: Lack of liver derived insulin-like growth factor-1 I (GF-1) inhibits obesity
In collaboration with the group of Professor Claes Ohlsson (http://www.ceross.sahlgrenska.gu.se/CEROSS/Senior_Researchers/Claes_Ohlsson/), we have used an unique mouse model for inducible, hepatocyte-specific IGF-I knockout (LI-IGF-I-/- mice) (Sjögren et al 1999) to study the role of liver derived IGF-I in regulation of body fat. The LI-IGF-I-/- mice did not get the age dependent increase in body fat seen in wildtype mice. The LI-IGF-I-/- mice also had elevated levels of insulin during basal conditions and after an intravenous glucose charge(Sjögren et al 2001) The LI-IGF-I-/- mice had 85% decreased serum levels of IGF-I and increased serum levels of GH, due to lack of IGF-I feedback inhibition. Moreover, the LI-IGF-I-/- mice had increased number of GH-releasing factor (GRF) and ghrelin receptors in the pituitary, and enhanced responsiveness to GRF and ghrelin analogues (Wallenius K et al 2001). Moreover, lack of liver IGF-1 increases blood pressure, peripheral resistance as well as endothelin-1 expression in the aorta (Tivesten et al 2002). In summary, liver IGF-1 seems to have associations with obesity, insulin resistance, and hypertension, i e several components of the metabolic syndrome.
In addition, liver derived IGF-1 seems to affect longitudinal bone growth (Sjögren er al 1999), bone mass (Sjögren et al 2002) as well as special learning and memory functions and prostatic growth in mice (Svensson et al 2006; Svensson et al 2008). Professor Ohlssons group and we have recently summed up these results in a comprehensive review (Ohlsson C et al 2009).
Side project 2: The lipogenic effect of ghrelin
It has been known for some time that substances acting on the hypothalamic ghrelin receptors can stimulate release of growth hormone (GH). In a collaboration with Professor Suzanne Dickson, we found that ghrelin analogues also can induce obesity via a GH independent mechanism (Lall et al, 2000). Tschöp et al report similar results and found that increased respiratory quotient, i e glucose/lipid metabolism ratio was a possible mechanism (Tschöp et al, 2000). The lipogenic effect of ghrelin is quite potent and overrides the lipolytic effect of ghrelin induced GH secretion. Ghrelin replacement therapy can reverse loss of body weight, body fat and lean body mass in gastrectomized mice (Domonville de la Cour et al 2005), opening up a potential use of Ghrelin analogues to reverse the malnutrition caused by clinical gastrectomies. We exploring the therapeutic potential of ghrelin in collaboration with Rose Pharma A/S (http://www.gastrotechpharma.com/).
The publication list of Prof Jansson includes articles in Science (one as first author, one as 1 of 3 authors), J Clin Invest (first author) , Proc Natl Acad Sci (2 articles, 1 as 2nd last author), Diabetes (1 as last and 1 as 2nd last author), Endocrine Reviews (1 as first author, 1 as last author of 2 authors, 1 as second last author), Nature Medicine (last and corresponding author). Of these, 15 articles have been cited more than 100 times, and 1 article more than 600 times, the Nature Medicine article ha been cited more than 300 times (Web of Science April 2009). The H-index is 38 meaning that 38 published articles are cited more than 38 times.
Dept. Physiology/Endocrinology, Box 434, SE-405 30 Göteborg
+46 31 741 1713, +46 31 786 3526, +46 739 01 1122
+46 31 773 3512