Slide 1

Thank you Anders. Great so the topic of my talk will be about adipokines, fat tissue gives nephrologists a message.

Slide 2

So this slide shows that compared to the US general population American dialysis patients are getting fatter at a higher rate. The forecasted prevalence of total obesity among incident dialysis patients in the US shows that within 3 or 4 years about 40% of the US dialysis population would be considered obese. Also recently a number of studies like this Swedish one have shown that obesity is a risk factor for chronic kidney disease independently of diabetes and hypertension. So to me it’s clear that nephrologists now need to learn more about fat mass because adipose tissue is by far the largest endocrine organ and it’s an organ that promotes vascular disease.

Slide 3

This slide shows that we have a variety of adipokines produced in fat tissue such as TNF-α, leptin, PAI, IL-6 and resistin which may promote endothelial dysfunction, contribute to the initiation of the atherosclerotic plaque, play a part in the progression and lesion expansion and also in destabilization and rupture. On the other hand, they have also one good adipokine, adiponectin which I will discuss a little more in detail later in my talk.

Slide 4

Obesity is also associated with a state of aberrant immune activity. About 20% of the total circulating concentration of IL-6 does originate from fat tissue. Clinical studies in non-renal patients have shown that there are correlations between fat mass and levels of pro-inflammatory cytokines and also weight loss has repeatedly been shown to be associated with a reduction in circulating levels of inflammatory biomarkers. These are data from our group in Stockholm showing that the truncal fat mass is significantly associated to circulating IL-6 levels, whereas no correlation between the subcutaneous non-truncal fat mass and IL-6 levels were observed.

Slide 5

I find it fascinating that there is a striking overlap between the biology of adipocytes and that of macrophages. So, when there’s an increase in weight the adipocytes start to secrete TNF-α and it stimulates the production of MCP-1 from preadipocytes and endothelial cells. The macrophage recruitment to fat plays a pivotal role in the onset of associated metabolic diseases. It has also been speculated that TNF- α secretion from macrophages with increasing fat tissue is a main cause of inhibited adiponectin secretion.

Slide 6

So under normal conditions there will be a clear distinction between macrophages and adipocytes. Macrophages play a part in the inflammatory response, whereas adipocytes store lipids and regulate metabolic homeostasis but in obesity there seems to be an overlapping biology and function of macrophages and adipocytes.

So, when the adipocytes increase in size and number, there is an infiltration of adipose tissue by macrophages and I find it really interesting that the body seems to react to obesity as it reacts to an infection. Because now adipocytes produce inflammatory cytokines via nuclear receptors, there is an increased production of TNF- α, IL-6, cytokines that may promote the insulin resistance. Moreover, the increase in adipocyte size is speculated to contribute to low oxygen tension in fat tissue contributing to the induction of hypoxic inducible factor, which may promote angiogenesis and inflammation.

Slide 7

But we should remember that fat is very different and visceral and subcutaneous adipose tissue are biologically distant. There is a differential gene expression between visceral and subcutaneous fat and the visceral adipose tissue releases 2-3 times more IL-6 than the subcutaneous fat tissue and of course, it’s the visceral fat mass that should be the most metabolically active fat tissue depot that may play a key role in the development of insulin resistance, type II diabetes and atherosclerosis. So, clearly the expression of IL-6, TNF- α and visfatin is higher in visceral compared to subcutaneous fat. On the other hand, the expression of adiponectin, the good guy is higher in subcutaneous compared to visceral fat mass.

Slide 8

So, when there is an increase in adipose tissue, there will be an increased secretion of what I call “The gang of four ”, the evil ones, leptin resistin, TNF- α and IL-6. Unfortunately, there seems to be decreased secretion of “Mr nice guy”, adiponectin.

Slide 9

So now let’s discuss what happens when there is decreased clearance of these proteins. Well, clearly decreased renal function does obscure the relationships between adipose tissue and the circulating adipokine levels. This is one example from our group showing quite a strong inverse correlation between GFR and resistin. Also an old piece of work from Stockholm shows that leptin level is not only associated to body mass index in renal patients but also increases when the renal function decreases.

Data presented by Doctor Yilmaz yesterday at this meeting, data from Turkey shows clearly that in different chronic kidney disease stages there is a marked increase in visfatin, as well as adiponectin levels.

Slide 10

So decreased clearance by kidney disease causes hyperadipokinemia. This may have effects on the vasculature via stimulation of NF-kB. It promotes endothelial activation, inflammation, oxidative stress and also vascular calcification. We have free fatty acids, hyperadipokinemia may promote insulin resistance, lipid accumulation, muscle breakdown and as at least part of these adipokines may pass the blood brain barrier, it may promote central nervous effects such as anorexia, depression, increased resting energy expenditure and sympathetic overeactivity.

Slide 11

This could be a piece of modern art but it is not. This is actually a polarised light image of leptin which is one of the adipokines that has attracted most interest in nephrology because it has been speculated to promote uremic anorexia. In this regard, I find this maybe the most interesting recent paper. It is Doctor Marc from US that in my study showed that elevated levels of leptin maybe an important cause of uraemia associated cachexia via signalling through the central melanocortin system. In accordance, in a small study in Stockholm we found that PD patients that were followed for 12 months, those who lost lean body mass during this period also had an increase in leptin levels. However, others have not been able to find that leptin promotes wasting and uremic anorexia rather stating that increased leptin levels are associated with better nutritional status and this could be an indication of leptin resistance in the uremic milieu.

Slide 12

Therefore, I’d shortly like to discuss this recent paper in Nature evaluating leptin resistance. In this study they found that physiological concentrations of leptin can stimulate the expression of CRP. But more interestingly they showed that whereas leptin here as expected decreased the daily food intake and decreased the body weight in these mice, the addition of low and high concentrations of CRP abrogated these effects here.
So, this study suggests the potential mechanisms contributing to leptin resistance by which subcirculating CRP binds to leptin and attenuates its physiological functions.

Slide 13

There are a number of putative peripheral actions of leptin, which is a pleiotropic hormone. I will have no time to discuss all this but I will say a couple of words about leptin and vascular calcification. Because this paper in Cell shows that leptin inhibits bone formation through a hypothalamic relay. Moreover, studies like this one have shown that leptin enhances the classification of vascular cells and there are clinical studies in other patient groups, such as in type II diabetes showing that plasma leptin levels are associated with coronary artery calcification score even after controlling for adiposity and CRP. So, it could be hypothesised that hyperleptinaemia may be one link between the vascular calcification and low bone turnover observed in ESRD.

Slide 14

So, now let’s move on to the “good guy” adiponectin which is a protector of vascular function and a natural anti-TNF. So whereas, TNF- α simulates NF-kB, adiponectin inhibits the expression of course, resistin and IL-6 are other factors that may stimulate NF-kB. We know that by stimulation of NF-kB we will have oxidative stress, inflammation and these are factors that may promote atherogenesis.
This is a list of described anti-atherogenic functions of adiponectin and I have no time to go through this list. I’ll just mention that adiponectin has been shown to attenuate proliferation of vascular smooth muscle cells, it’s involved in stabilisation of plaques, inhibits growth factors etc. I think this is the most recent contribution, it was a paper in JCI 1-2 weeks ago showing that in the adiponectin knockout mice loss of adiponectin induced a primary state of endothelial dysfunction with increased leukocyte endothelium adhesiveness.

Slide 15

So, as I mentioned previously in renal disease you will find markedly elevated adiponectin levels as shown in this data from Doctor Yilmaz that with decreasing GFR there is a marked increase. But also other factors in our patient may contribute to elevated adipoenctin levels. This data from Stockholm where we compared adiponectin levels in 3 groups of patients with similar GFR it is quite clear that type I diabetics had higher levels than non-diabetics and type II diabetics.

Slide 16

Dyslipidemia and insulin resistance are also associated with lower adiponectin and this was first described by Doctor Zoccali’s group. Here you find the associations between HDL cholesterol, triglycerides and adiponectin levels and similarly low adiponectin levels seem to be associated with insulin resistance. Here you see an inverse correlation between plasma insulin and here HOMA index and the adiponectin levels. But we should be aware that adiponectin does circulate as both a hexamer of low molecular weight and a larger multimeric structure of high molecular weight. It’s the high molecular weight adiponectin complex that is the active form of the protein. So it’s the low molecular weight and high molecular weight ratio rather than the absolute amounts of adiponectin that is critical in determining insulin sensitivity.
So, in renal patients we now need to evaluate the respective impact of how high and low molecular weight adiponectins.

Slide 17

Another factor that may contribute to relatively lower adiponectin levels in renal patients is inflammation. Because adiponectin modulates endothelial inflammatory response in vitro and inhibits endothelial NF-kB signalling. Our data from Stockholm shows that surrogate markers of inflammation such as CRP and fibrinogen are inversely related to adiponectin levels. Also other groups have shown similar results. This is a Korean study showing inverse correlation between CRP and adiponectin.

Slide 18

An inflammation may be one factor that contributes to low adiponectin gene expression. This data shows the adipoenctin mRNA to β actin mRNA ratio and it’s to me clear that compared to the controls the addition of TNF- α and IL-6 plus its receptor down regulates adiponectin gene expression. This maybe one factor explaining why we found lower adiponectin gene expression in patients with ESRD in fat tissue compared to age and gender matched healthy controls.

Slide 19

So by now we have numerous studies that have evaluated the effects of adiponectin and its ability to predict cardiovascular disease in renal patients.
First, Dr Zoccali in 227 haemodialysis patients found that low adiponectin predicted cardiovascular events but not all-cause mortality. A Japanese study in 150 CKD 1-5 patients show that low adiponectin predicted cardiovascular events. Moreover, Ignacy showed that in 80 haemodialysis patients low adiponectin predicted survival. Another Japanese study in haemodialysis patients showed that low adiponectin predicted coronary and stent restenosis.
So we were all very surprised when this large study based on the HEMO material of 820 CKD 3-4 patients, follow up for ten years showed that high adiponectin predicted increased mortality. If you look in the literature, you’ll find that there was an earlier study published in Circulation in 195 patients with congestive heart failure also showing that high adiponectin predicted increased mortality.

Another study in 325 patients with coronary artery disease followed for 2 years also showed that high adipoectin predicted cardiac and all cause mortality.

Slide 20

So it’s important that we now elucidate the potential detrimental effects of high adiponectin levels. It’s clear that the clinical application of adiponectin as a prognostic marker is complex and also context sensitive. A number of factors you see in the renal milieu increase adiponectin levels. I have mentioned a decline in GFR.  Zoccali’s group showed that also an increase in urinary protein excretion maybe associated with elevated adiponectin. Type-1 diabetes seems to be associated with higher adiponectin and also of course, some drugs as PPAR-γ agonists and ACE inhibitors may increase adiponectin. But on the other hand, several features may decrease adiponectin levels. Increased fat mass, which interestingly has been associated with better outcome, is associated with lower adiponectin. Inflammation, insulin resistance and dyslipidemia are also associated with lower adiponectin levels.
More recently there have been discussions that also protein energy wasting maybe associated with relatively higher adiponectin levels.

Slide 21

I’d like to show you some unpublished data from Stockholm by Doctor Carrero. These are 228 prevalent haemodialysis patients and here we have divided the patients into adiponectin groups according to gender specific tertiles. We could confirm that the all ischemic vascular disease and ischemic heart disease, the prevalence is higher in the low adiponectin group. However, we were surprised that when we looked at nutritional markers, we found that actually that MAMC seems to be lower in the high adipoentcin tertile group. The handgrip strength seems to be lower in the high adiponectin tertile group. Also biochemical markers such as IGF 1 were significantly lower in the high adiponectin tertile group.

Slide 22

So we have to discuss whether or not adiponectin is a promoter of wasting. It’s clear that weight reduction is associated with elevated levels of adiponectin. This data clearly shows that when you lose weight and body mass index decreases, adiponectin levels increase. So high adiponectin levels could be a marker of an ongoing wasting process in our patients. On the other hand, it’s also been suggested in animal studies that adiponectin acts in the brain to decrease body weight. This is a very nice study. They have injected adiponectin intra-cerebroventricularly to mice and they found that this was associated with lowering of body weight, lowering of body fat, lower food intake and lower respiratory quotient and then increase in the uncoupling protein gene expression. So this suggests that intra-cerebroventricular administration of adiponectin decreases body weight mainly by stimulating energy expenditure. As Agouti mice did not respond to adiponectin injections this suggests that a melanocortin pathway is the target here.

Slide 23

In the last part of my talk I will discuss a little bit about being fat. Is it good for dialysis patients? We’re all aware that numerous data from American studies show that increased body mass index is associated with a survival advantage. However, this data were challenged recently by a Dutch study by Doctor Mutsert showing that after 7 years follow up there is no protective effect of elevated BMI compared to the general population.

Slide 24

There maybe a number of reasons for these discrepant results observed between US and Europe. Of course, the distribution and prevalence of obesity is different in USA compared to Europe. Also the obesity paradox may be much stronger in African-Americans than it is in Caucasians. It could also be discussed that statistical fallacies, such as residual confounding and differences in observation time and age between the general population and dialysis patients in the USA may contribute to the results. As 9 out of 10 patients in the USA with chronic kidney disease do not reach the dialysis facility chair, it could also be speculated that obese patients starting dialysis in USA may be a selected group of survivors.
Finally, we should be aware that body mass index is a very poor marker of fat mass. So the absurd effect maybe rather due to increased muscle mass than fat mass.

Slide 25

I think this is a nice example of this. This is a Brazilian study of 344 incident haemodialysis patients that were followed for I think 20 months and we have this group that did the worst and this group where the one with BMI above 25 but low MAMC suggesting that the loss of muscle tissue maybe the most important.

Slide 26

I’d also like to draw your attention to another problem when doing these studies and I will use some recent data from Stockholm that is in press in the American Journal of Clinical Nutrition and in this study we evaluated IL-6 levels in patients starting dialysis, we grouped them according to the presence of wasting or not and this is the non-wasted group according to SGA and you see that with increasing BMI there is an increase in IL-6 levels.
If you do the same in wasted patients with SGA above 1, you’ll find a similar relationship but on a different level. I think this is the most interesting connection to make because you will find similar IL-6 levels in well-nourished patients with obesity than in the malnourished patients with low BMI. So I think the lesson to be learned here that cross-sectional studies are indeed insufficient to reveal the true associations between body composition, inflammation and outcome.

Slide 27

So, there are clearly detrimental effects of increased fat mass associated with increased systemic inflammation, dyslipidemia, insulin resistance, obstructive sleep apnoea and of course, psychosocial effects. But we should be aware that also there could be beneficial effects of increased fat mass. Of course, it indicates well preserved energy stores and better appetite.
But there are also some interesting beneficial aspects of increased fat mass. It may be associated to genetic traits with beneficial effects. Of course, increase in fat mass is associated with higher bone mineral density. It could be more efficient disposal of lipophilic toxins such as plasma cresol or pentosidin and it could also be speculated that increased fat mass is associated to a better stem cell mobilisation.

Slide 28

So, I’d like to sum up what I’ve tried to tell you and it’s time for you to wake up now. I’ve told you that there’s a striking overlap between the biology of adipocytes and that of macrophages. Further studies are needed to understand the reasons why the body reacts to obesity as it reacts to infections. The clinical application of adiponectin as a prognostic marker is complex and context sensitive. More studies are needed to determine if we measure active adiponectins or just inactive fragments. Finally, as BMI does not differentiate between fat and muscle and visceral fat is metabolically active than subcutaneous fat I think it’s now important to relate muscle mass to different fat tissue depots to reveal the effects on outcome.

Slide 29

Thank you.