The obesogenic effects of polyunsaturated fatty acids are ...

The obesogenic effects of polyunsaturated fatty

acids are dependent on background diets

Tao Ma

Supervisor:

Karsten Kristiansen

Lise Madsen

Department of Biology, University of Copenhagen, Copenhagen DK-

2200, Denmark.

The Graduate School of Science, Faculty of Science, University of

Copenhagen, Denmark.

July 2011, 家

2

Acknowledgement

I am grateful to my supervisor Karsten Kristiansen for giving me the opportunity to continue

my research life. It is full of excitement, as all I can ask for. Thanks to his guidance and

inspiration. I would as well give my thanks to my co-supervisor Lise Madsen for conceiving,

organizing and directing my project. That is precious experiences that I can benefit from for

my future work.

We can only be a part of a team in order to achieve. I would like to thanks all the group

members in Denmark and Norway for their helps and joys they bring along, especially Qin

Hao and Rasmus Koefoed Petersen for their patience and kindness. Also, my appreciation to

Alison Keenan for her great help with the language.

And finally, there is nothing can be accomplished without the continuous support and love

from my family.

3

Table of Contents Acknowledgement ............................................................................................................................................. 2

Abstract ............................................................................................................................................................. 4

Abstract in Danish .............................................................................................................................................. 5

Introduction ....................................................................................................................................................... 6

Obesity and food intake ................................................................................................................................ 6

Polyunsaturated fatty acids in the diet ......................................................................................................... 7

Adipocyte biology .......................................................................................................................................... 7

Adipogenesis is dependent on cAMP levels. ................................................................................................. 8

Results ............................................................................................................................................................... 9

Discussion, perspectives .................................................................................................................................. 11

Low carbohydrate, high protein diet ........................................................................................................... 11

Branched-chain amino acids and mTOR ...................................................................................................... 12

Regulation of hepatic lipogenesis ................................................................................................................ 12

Fructose ....................................................................................................................................................... 13

Fish oil contaminants ................................................................................................................................... 14

References ....................................................................................................................................................... 16

Annexes ........................................................................................................................................................... 21

4

Abstract

Polyunsaturated n-3 fatty acids (n-3 PUFAs) are reported to protect against high fat diet-induced

obesity and inflammation in adipose tissue. While it has previously been reported that the

adipogenic effects of n-6 PUFAs are dependent on the macronutrient composition of the diet,

whether n-3 PUFAs metabolism is subjected to similar impact by carbohydrates and proteins in the

diet is not well explored. In the present thesis two studies are described. Isocaloric high fat diets

enriched with protein and carbohydrate with different weight ratio or with various carbohydrate

sources were fed to male C57BL/6J mice. We show that increasing the amount of sucrose at the

expense of protein in the diet correlated with increased energy efficiency and fat mass irrespective

of fat sources. We propose that this effect is a result of reduced thermogenesis in fat tissues,

combined repression of gluconeogenesis and ureagenesis in the liver. Also, high fat diets induced

glucose intolerance regardless of the adiposity of the animals. Moreover, by using carbohydrates

with differing glycemic indices we provide evidence that insulin plays a central role in promoting

adiposity and inflammation in fat tissues. This idea was strengthened further by exploring

pharmaceutical drugs to modulate insulin secretion.

In summary, the ability of background diet, namely carbohydrates and proteins, in regulating insulin

secretion significantly modulates the beneficial effects of n-3 PUFAs in development of obesity,

glucose intolerance and adipose tissue inflammation.

5

Abstract in Danish

Polyumættede n-3 fedtsyrer (n-3 PUFA) er rapporteret at beskytte mod fedme og inflammation i

fedtvæv induceret af kost med højt fedtindhold. Mens det tidligere har været rapporteret, at de

adipogene effekter af n-6 PUFA er afhængige af kostens sammensætning af makronæringsstof, er

påvirkning af kulhydrater og protein i kosten på n-3 PUFA metabolisme ikke godt udforsket. I

nærværende afhandling er to undersøgelser beskrevet. C57BL/6J hanmus blev fodret med

isokaloriske dieter med højt fedtindhold beriget med protein og kulhydrat med varierende indbyrdes

vægt-forhold eller med forskellige kulhydratkilder. Vi viser, at øget sukker i kosten er korreleret

med øget energieffektivitet og fedtmasse uanset fedtkilder. Vores hypotese er, at disse observationer

er en følge af nedsat termogenese i fedtvæv, og gluconeogenese og ureagenese i leveren. Ligeledes

inducerer kost med højt fedtindhold glukose intolerance uafhængigt af dyrenes (over)vægt. Ved

brug af kulhydrater med forskellige glykæmiske indeks viser vi desuden, at insulin spiller en central

rolle for udviklingen af overvægt og inflammation i fedtvæv. Denne observation blev styrket

yderligere i forsøg med lægemidler, der direkte modulerer insulinsekretion.

Sammenfattende kan vi konkludere, at kostens indhold af kulhydrater og proteiner og evnen til at

regulere insulinsekretionen påvirker de gavnlige effekter af n-3 PUFA på udvikling af fedme,

glukose intolerance og inflammation i fedtvæv.

6

Introduction

Obesity and food intake

Obesity is when excess fat is accumulated in the body, mainly in adipose tissues, to an unhealthy

extent. An adult with a body mass index (BMI) between 25 and 30 kg/m2 is considered overweight,

while an adult with a BMI over 30 kg/m2 is regarded as obese

1; nonetheless, this classification is

debatable when applied to different races2,3

.

Utilizing the historic record provided by American obesity data4, we are able to track the increasing

incidence of overweight and obesity in the new world. Not trailing far behind, developing countries

are catching up with this trend, which is a direct consequence of overconsumption of energy-dense

food and engaging in a sedentary lifestyle5. Obesity is one of the biggest threats to public health in

the new century, as it increases the risk of developing type 2 diabetes6, cardiovascular disease

7 and

certain types of cancer8.

Consequently, considerable financial resources as well as social and scientific efforts have been

mobilized towards the war against obesity9. One focus point has been to promote a reduction in

dietary fat intake, as can be observed with the “light” labeled goods filling supermarket shelves.

This is mainly because fat contains more energy per unit weight than carbohydrate or protein,

9kcal/g, 4.5kcal/g and 4kcal/g respectively. By reducing the energy density of food, the average

energy intake of the general public should therefore be reduced. Besides the increase in average

energy intake over time10

, another major alteration over the thousands of years of dietary history is

the dramatic elevation of carbohydrate content in the food at the expense of protein11

. One recent

publication suggested that over a period of two years a low-carbohydrate diet had a similar weight

loss effect as a low-fat diet12

. Moreover, a body of evidence pointed to the beneficial effects of high

protein diet on whole body metabolism13

, implying that macronutrients are not simply different

with regard to their caloric value, but have other impacts on weight gain and maintenance that are

not fully elucidated.

7

Polyunsaturated fatty acids in the diet

Fat comes in the form of a blend of different fatty acids, namely saturated fatty acids (SFAs),

monounsaturated fatty acids (MUFAs) and polyunsaturated fatty acids (PUFAs). Multiple studies

suggest that the consumption of MUFAs in place of SFAs improves insulin sensitivity14,15

. The

increased intake of n-6 PUFAs, or higher dietary n-6 to n-3 ratio, has been implicated in promoting

many diseases including obesity and diabetes16,17

. As shown in the work of Massiera et al., in mice,

pups from mothers fed a diet enriched with n-6 linoleic acid became obese. However, this can be

prevented by inclusion of the n-3 PUFA α-linolenic acid in the diet. The authors attributed this

phenomenon to the ability of n-6 PUFAs to promote adipogenesis18

. In line with the animal work, a

similar outcome has been observed in a human study that revealed a higher n-6/n-3 PUFA ratio in

the umbilical cord plasma was associated with a higher incidence of obesity19

. Furthermore, work

from our group suggests that the adipogenic nature of n-6 PUFAs is dependent on the cellular

cAMP status20

.

The possible anti-obesity mechanisms of n-3 PUFAs include, but are not limited to: 1), competition

with n-6 PUFAs in enzymatic pathways, which may dampen the adipogenic and pro-inflammatory

effects of n-6 PUFAs21

, as one metabolic effect from obesity is chronic low-grade inflammation.

Anti-inflammatory intervention has been shown to be helpful against metabolic dysfunction22

. 2),

up regulation of β-oxidation and mitochondrial biogenesis. In both rodent and human settings

addition of n-3 PUFAs, particularly of marine source, increased lipid oxidation23,24

. 3), activation of

G protein-coupled receptor 120, which exerts its anti-inflammatory effects by inhibiting both Toll-

like receptor (TLR) and tissue necrosis factor- α (TNF-α) signaling pathways in the residing

macrophages in adipose tissue25

.

Adipocyte biology

Historically, adipose tissue was seen as a static storage unit when energy is in excess and to release

energy during food deprivation in the form of fatty acids and glycerol. When energy consumption

persistently exceeds energy expenditure, expansion of adipose tissue will occur, contributed to by

both hyperplasia and hypertrophy of adipocytes and finally culminate in obesity. Recently adipose

tissue has become appreciated as an important mediator of systemic metabolism for its role of

secreting regulatory proteins26

. The first adipokine being identified was adipsin (Factor D) from

8

198727

. Over the years, more and more adipokines were discovered including leptin, TNF-α,

adiponectin and so on, most of which bridge the links between obesity, inflammation and insulin

resistance28

. Furthermore, with the rediscovery of brown adipose tissue in adult humans, people

have begun exploring the possible therapeutic role of this special fat depot29

.

Insulin is an important driver for adipocyte differentiation and the primary anabolic hormone

promoting energy storage 30,31

. After ingesting a meal of high glycemic carbohydrate, the high

glycemia will induce pancreatic insulin secretion to promote glucose uptake in adipose tissue. As

shown in animal models, mice with adipose tissue-specific insulin receptor knock out (FIRKO mice)

are protected against obesity and remain glucose tolerant on a high-fat diet32

. In contrast, mice

lacking the insulin receptor in muscle (MIRKO mice) maintain normal plasma glucose levels by

increasing glucose utilization in white adipose tissue (WAT) and fat mass in order to compensate

the loss of muscular glucose transport activity33

. Furthermore, mice overexpressing facilitated

glucose transporter, member 4 (GLUT4)34

or insulin receptor substrate 1 (Irs1)35

in adipose tissue

are obese.

Adipogenesis is dependent on cAMP levels.

Our earlier works has shown the interplay of cAMP-elevating agents and arachidonic acid in

mediating adipogenesis20,36

. The inclusion of n-6 fatty acid arachidonic acid (AA) in the induction

cocktail inhibited adipogenesis of 3T3-L1 preadipocytes. This inhibitory effect requires the cAMP –

elevating agent 3-isobutyl-1-methylxanthine (IBMX) in the mixture and depends on protein kinase

A (PKA) and cycloxygenase (COX) activity. During the initial phase of adipogenic induction, the

combination of IBMX and AA dramatically increases the expression levels of both COX-1 and 2,

which in turn metabolize AA towards the inhibitory prostaglandin E2 (PGE2) and F2α (PGF2α) 37

.

Furthermore, when AA is present, even without an increase of cellular cAMP levels, overexpression

of COX-1 and 2 can still execute their anti-adipogenic power. On the other hand, in the absence of

IBMX, AA induces adipogenesis which was also reported previously by others, probably through

the production of adipogenic prostaglandins18

.

In order to create a similar scenario in vivo, on top of high corn oil, different proportions of

carbohydrate and protein were included in the diets trying to manipulate the cAMP levels by

changing the plasma insulin/glucagon ratio. Indeed as we reported38

, the insulin to glucagon ratio

9

was more than 3 times higher in the mice fed with high corn oil diet enriched with sucrose than with

protein. With the different diets, significant changes in body weight gains were observed both ad

libitum and pair-fed., and were due to the different degrees of adiposity. As expected, in multiple

adipose tissues, the expression of Crem (cAMP-responsive element modulator) and Pde4b (cAMP-

specific phosphodiesterase 4b) and the phosphorylation status of CREB (cAMP-responsive element

-binding protein) were differently regulated because of the diets, which revealed an altered cAMP

level. Accordingly, raised circulating PGE2 and PGF2 levels were detected in mice fed with a

protein supplemented diet that correlated with increased expressions of both Cox-1 and Cox-2 in

most of fat tissues.

Taken together, it was a clear demonstration of the cAMP-PKA-COX-prostaglandin axis in

regulating adipogenesis both in vitro and in vivo. The obesogenic effect of n-6 fatty acid is

dependent on the macronutrient content of the diets.

Results

As shown previously, cAMP-dependent signaling controls the obesogenic effect of n-6 PUFAs in

mice. We want to further investigate whether this phenomenon is restricted to n-6 PUFAs, or if the

beneficial effects of n-3 PUFAs can also be affected by background diets (Annex 1). To achieve

this, we prepared isocaloric diets enriched with either fish oil or corn oil (Table 1). After 9 weeks of

feeding, to our surprise, mice fed with sucrose-supplemented diets became obese regardless of the

fat sources (Fig.1 E-F). This was linked to the differential insulin to glucagon level in plasma that

we had previously identified (Fig.1 C-D), suggesting one of the determining effects background diet

has on obesity. Furthermore, the inflammation levels in the adipose tissues were correlated with

adiposity instead of dietary fat types assessed by gene expression of macrophage and inflammatory

markers (Fig.2A).

When compared to mice fed a standard chow diet, all individuals on high fat diets had impaired

glucose tolerance regardless they were fat or lean. But the separation of the HOMA index readout

implied different mechanisms underlying the observed glucose intolerance (Fig.2B). We observed

one of the recognized beneficial effects of taking fish oil in our settings. Hepatic lipid accumulation

10

was considerably less in the fish oil-fed mice than in the corn oil-fed mice (Fig.3A), but it was not

directly related to the lipogenic gene expression in their livers (Fig.3B).

Since the mice on high fat diets consumed equivalent amounts of food, the lean ones had markedly

reduced energy efficiency (Fig.3D, 4B). The following points could partly explain the differences,

1). protein-fed animals could have more brownish phenotype in inguinal WAT (iWAT) judged by

the up-regulation of Ucp1 expression, which dispenses energy as heat (Fig.3E, 4E); 2). elevated

gluconeogenesis in mice fed protein-enriched diets even in the fed state, indicated by the enhanced

glucose production during pyruvate tolerance test and hepatic gene expressions (Fig.6A and B); 3).

nitrogen metabolism was another potential contributor to the energy-wasting effect as protein

cannot be stored but must be processed immediately (Fig.6B).

In order to evaluate the role of insulin in modulating the obesogenic effects of fish oil, isocaloric

high fish oil diets with different sucrose to protein weight ratios were fed to the mice (Annex 2).

Supporting our previous findings, the fat mass development (Fig.1D), energy efficiency (Fig.1B),

inflammation levels in adipose tissues (Fig.1E) were all does-dependently correlated to sucrose

content of the diets. mRNA levels of the indicators pointing to the three energy consumption

processes, Ucp1 in the iWAT--- thermogenesis (Fig.1E); Pck1 in the live--- gluconeogenesis and

Agxt in the liver--- ureagenesis (Fig.1F), were inversely associated as earlier suggested.

By exchanging sucrose with glucose or fructose in a fish oil-supplemented diet, we further

pinpointed that the insulin stimulating agent glucose moiety in sucrose facilitated the obesity-

promoting effect of fish oil, as mice fed a fructose diet gained less WAT weights (Fig.2D) and had

lower plasma triglycerides levels (Fig.2E). However, in the same set of mice, hepatic genes related

to β-oxidation were down-regulated and the expression of lipogenic genes were increased (Fig.2F).

Our theory was further supported by including starches promote different insulin responses

(different glycemic index-GI). Although these animals did not differ in weight gain, the WAT

masses were significantly higher in high GI diet-fed mice (Fig.4A and C), and this was

accompanied by the increased expression of lipogenic genes Fasn and Scd1 (Fig.4G).

Furthermore, pharmaceutical regulators of insulin secretion were introduced to the feeding regime.

Inclusion of insulin secretagogue glybenclamide in the protein-enriched high fish oil diet did not

result in a separation of the body weight gains (Fig.5A). A tendency of fat mass differences was

observed (Fig.5B), and could be linked to observed changes in gene expression related to

11

gluconeogenesis and ureagenesis (Fig.5E). On the other hand, inhibition of insulin secretion by

nifedipine did achieve the expected fat mass-reducing effect (Fig.6C)

.

Discussion, perspectives

Low carbohydrate, high protein diet

It has been suggested that a modest increase in protein ingestion combined with a low glycemic

index diet could be protective against obesity in children39

. Further, as reported in a recent large

European cohort study, after an initial weight loss, a diet with lower glycemic index and higher

protein content is most ideal for maintaining body weight and preventing weight regain40

. When

consuming a diet very low in carbohydrates, such as in the case of our protein-enriched mouse feed

with only 8% of calories from carbohydrates, the animals need to mobilize liver glycogen storage

and increase gluconeogenesis in order to sustain blood glucose levels. After feeding for a long

period of time, glycogen will be exhausted, and the body will have to turn to other energy sources.

Even in the fed state, these protein-fed mice still kept high rates of β-oxidation, indicated by the

high circulating levels of 2-hydroxybutyrate (Annex 1Fig.5B (1.5B)), suggesting the need to

mobilize fat and/or protein for basal metabolism. This situation, to a certain degree, mimics what

has been observed in long term fasting. During starvation, following the decrease of circulating

glucose, insulin levels drop dramatically and glucagon secretion is elevated. Although

gluconeogenesis contributes partially to the blood glucose supply, it will be prioritized for central

nervous system and red blood cells functions.

Despite the protein-fed mice kept lean as their low fat diet-fed littermates, they showed the same

degree of glucose intolerance as the obese ones on sucrose-enriched diet. If our analogy to fasting is

accurate, the glucose intolerance we have observed in lean protein-fed mice can be delineated. As

fasting reduced insulin secretion in rats, this repression can only be rescued by refeeding with diets

high in carbohydrates41

. Although no difference was observed in the insulin tolerance test (Fig

1.s1B), there are reports both in rats and humans showing that consuming low carbohydrate, high

protein diet can attenuate the suppression of hepatic glucose output by insulin42,43

. In order to

12

resolve the interplay between insulin secretion and action, detailed studies are required to clarify the

mechanisms behind impaired glucose tolerance in these mice.

Branched-chain amino acids and mTOR

Dietary proteins are important modulators of metabolism and insulin sensitivity44

. It has been

suggested that a modest increase in protein intake could potentially combat obesity45

, owing at least

in part to an increased thermic effect. We have shown these processes include but are not limited to

thermogenesis in the iWAT (See annex 3), gluconeogenesis and ureagenesis in the liver. Casein has

been widely used as the main protein source in rodent feed with the supplement of L-Cystine46

.

Given its relative high percentage of branched-chain amino acids (BCAAs) and low content of

taurine and glycine compared to fish or soy protein, special consideration needs to be taken into

account when using high amount of casein. In a separate study, we have shown by replacing casein

with salmon protein hydrolysates in the diet, rats were protected from diet-induced obesity due to

raised plasma bile acid concentrations (Annex 4). Furthermore, there are links established between

excess intake of dietary protein, especially BCAAs and insulin resistance47,48

.

Activation of mTOR (mammalian target of rapamycin) by BCAAs, particularly leucine, could

regulate glucose uptake in skeletal muscle through the inhibitory phosphorylation of S6 kinase on

insulin receptor substrate 1 (IRS-1) 49

. Furthermore Newgard et al, had shown that by 50% increase

of BCAAs in a high fat diet, rats developed insulin resistance regardless of a low rate of weight gain,

which was connected to chronic phosphorylation of mTOR, JNK, and IRS1Ser307

and the

accumulation of multiple acylcarnitines in muscle50

. This observation relates to our reported results

that insulin resistance can be disconnected from adiposity. In a recent multicenter cohort

metabolomic study, elevated plasma BCAAs levels had been shown to correlate strongly with and

could predict future diabetes51

. If BCAAs did play a role contributing to the overall effect we have

presented, protein composition in the diet needs to be thought over in the future experiment design.

Regulation of hepatic lipogenesis

Hepatic glucose and lipid metabolism are cross-regulated by multiple transcriptional regulators with

sterol regulatory element binding protein-1c (SREBP-1c) taking the central stage52

. In the context of

13

high dietary PUFAs, the modulation is less clear. PUFAs have been shown to suppress hepatic

Srebp-1c transcription 53

, mRNA stability 54

and its proteolytic activation 55

through liver X receptor

(LXR)56

- and/or SREBP-1c55

- mediated pathways. In our study, the suppression on Srebp-1c

transcription was only sustained when diets were supplemented with protein but abolished when

combined with sucrose (Fig1.3B, 2.7A). The most straight forward explanation would be that

increased insulin levels played a role, as multiple lines of evidence suggested the importance of

insulin in controlling expression and activity of Srebp-1c 57

. But further in our study it was

elucidated otherwise, as modulating insulin secretion by pharmaceutical drugs did not have any

effect on either Srebp-1c mRNA abundance or its activity, judged by the transcription of its target

genes Scd1, Fasn and Acaca (Fig2.5E 2.7A).

The role of glucose in the context of this story needs also to be discussed. Besides insulin,

carbohydrate ingestion can very efficiently lead to the induction of lipogenic genes58

. However,the

main sensor of glucose carbohydrate-responsive element–binding protein (ChREBP)59

, like

SREBP-1c, was also suppressed in the presence of PUFAs60

. Mitro et al. has shown that LXRs

could also work as glucose sensors to regulate cholesterol metabolism genes and in some degree

lipogenic genes as well 61

. In the LXRα/β knockout model, high carbohydrate diet-induced hepatic

gene expression of Scd1 and Fasn were attenuated62

. As a whole, they suggest that in response to

glucose stimuli, LXR-dependent pathway is indispensable for the full activation of Scd1 and Fasn

expression, while the regulation of Acaca is mostly under the control of ChREBP60

. So in our high

fish oil feeding supplemented with different starches, the observed elevation of Scd1 and Fasn

expression with high GI diet (Fig2.4G) could be explained as under the impact of fish oil, only LXR

was still able to sense plasma glucose changes though mechanism that is still under debate63,64

. Of

course all speculations need supporting evidence from follow up studies. And with mTORC165,66

and AMP-activated protein kinase (AMPK)67,68

also in the picture now, the already intricate control

of SREBP-1c and lipogenesis is only going to be more complicated.

Fructose

The increased world consumption of fructose or high-fructose corn syrup has been associated with

the obesity epidemic69,70

. Fructose is taken up by the liver rapidly because of the low Km of

fructokinase towards fructose71

. Without the negative feedback on phosphofructokinase as in

glucose metabolism, the reactions continue to provide substrate for glucose and lipid production71

.

14

Although fructose has long been used to induce glucose intolerance in rats, in C57BL/6 mice

neither hyperinsulimia nor hyperglycemia is developed. It had been shown C57BL/6 mice are less-

responsive towards fructose-induced Srebf1 activation among different strains of mice due to a

single nucleotide polymorphism in the promoter region72

. However we failed to observe this

unresponsiveness. As others have shown in rats73, in our fructose-fed mice despite the presence of

fish oil in the diet, lipogenic genes were induced and β-oxidation-related targets were repressed

(Fig.2.2F). But when compared to sucrose- and glucose-fed mice, these animals gained least body

weight, had the lowest amounts of fat tissues and lowest circulating triglycerides (Fig2.2), so the

extra fat has to be either stored ectopically or burnt. The increased brown adipocyte phenotype in

the iWAT could, in part, assist this process. Although others have shown the development of

hepatic lipid accumulation in fructose-fed mice74

, it was not the case in our hands despite elevated

liver to body weight ratios (Fig2.3A). Interestingly, the authors also reported that a nonabsorbable

antibiotics treatment could attenuate lipid accumulation in liver, suggesting potential involvement

of gut microbes in fructose metabolism74

.

The metabolic effects of fructose do not limit to the listed above, as there have been reports

showing centrally-administrated fructose led to increased food intake in rodents75,76

by reducing

hypothalamic malonyl–CoA levels76

. Furthermore, the ability of fructose to cross the blood-brain

barrier indirectly supports this observation, as the orexigenic effect relies on CNS function77

. One

interesting study, by using hyperinsulinemic-euglycemic clamps, demonstrated that sucrose feeding

was not the same as with the combination of its hydrolysis products glucose and fructose in

promoting hepatic insulin resistance78

. This may suggest the involvement of gastric emptying or the

release of gut-secreted hormones.

Fish oil contaminants

Long chain n-3 fatty acids in fish oil are not synthesized de novo by fish themselves, but by algae

they consume in the wild or added to the feed in farmed fish colonies79

. Along with the beneficial

nutrients, through food chains, persistent organic pollutants (POPs) are also accumulated. Despite

the effort to limit their release, POPs still persist in the environment80

. POPs such as organochlorine

pesticides, polychlorinated biphenyls (PCBs) and dioxins are hydrophobic compounds, and

consequently they accumulate in fatty tissues, and therefore are inevitably retained in the process of

fish oil production.

15

In a pair of population studies, an association was reported between serum concentrations of POPs

and insulin resistance in spite of the subjects’ diabetic status81,82

. Furthermore, different

mechanisms were proposed in explaining this deteriorating effect, which include causing

mitochondrial dysfunction83

and impairment of insulin actions both in vivo and in vitro (ref84

,

Annex 5). In order not to weigh out the beneficial effect by pollutants, purity would be a major

concern when consuming very long chain marine n-3 fatty acids. This also applies to the use of fish

oil in animal feeds85

.

16

References

1. World Health Organization. Obesity: preventing and managing the global epidemic Report of a WHO

Consultation (WHO Technical Report Series 894). (2000).

2. Wang, J. et al. Comparisons for body mass index and body fat percent among Puerto Ricans, blacks, whites and

Asians living in the New York City area. Obesity research 4, 377-84(1996).

3. Deurenberg, P., Yap, M. & Staveren, W.A. van Body mass index and percent body fat: a meta analysis among

different ethnic groups. International journal of obesity and related metabolic disorders : journal of the

International Association for the Study of Obesity 22, 1164-71(1998).

4. Centers for Disease Control and Prevention U.S. Obesity Trends. at

<http://www.cdc.gov/obesity/data/trends.html>

5. Hossain, P., Kawar, B. & El Nahas, M. Obesity and diabetes in the developing world--a growing challenge. The

New England journal of medicine 356, 213-5(2007).

6. Colditz, G.A. et al. Weight gain as a risk factor for clinical diabetes mellitus in women. Annals of internal

medicine 122, 481-6(1995).

7. Willett, W.C. et al. Weight, weight change, and coronary heart disease in women. Risk within the “normal”

weight range. JAMA : the journal of the American Medical Association 273, 461-5(1995).

8. Calle, E.E. et al. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults.

The New England journal of medicine 348, 1625-38(2003).

9. NIH New strategic plan for NIH obesity research seeks to curb epidemic. NIH News (2011).at

<http://www.nih.gov/news/health/mar2011/niddk-31.htm>

10. Wright, J. et al. Trends in Intake of Energy and Macronutrients --- United States, 1971--2000. Morbidity and

Mortality Weekly Report (2004).at <http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5304a3.htm>

11. Cordain, L. et al. Origins and evolution of the Western diet: health implications for the 21st century. Am J Clin

Nutr 81, 341-354(2005).

12. Foster, G.D. et al. Weight and metabolic outcomes after 2 years on a low-carbohydrate versus low-fat diet: a

randomized trial. Annals of internal medicine 153, 147-57(2010).

13. Paddon-Jones, D. et al. Protein, weight management, and satiety. Am J Clin Nutr 87, 1558S-1561(2008).

14. Vessby, B. et al. Substituting dietary saturated for monounsaturated fat impairs insulin sensitivity in healthy

men and women: The KANWU Study. Diabetologia 44, 312-9(2001).

15. Gillingham, L.G., Harris-Janz, S. & Jones, P.J.H. Dietary Monounsaturated Fatty Acids Are Protective Against

Metabolic Syndrome and Cardiovascular Disease Risk Factors. Lipids (2011).doi:10.1007/s11745-010-3524-y

16. Ailhaud, G. et al. Temporal changes in dietary fats: role of n-6 polyunsaturated fatty acids in excessive adipose

tissue development and relationship to obesity. Progress in lipid research 45, 203-36(2006).

17. Ailhaud, G., Guesnet, P. & Cunnane, S.C. An emerging risk factor for obesity: does disequilibrium of

polyunsaturated fatty acid metabolism contribute to excessive adipose tissue development? The British journal

of nutrition 100, 461-70(2008).

17

18. Massiera, F. et al. Arachidonic acid and prostacyclin signaling promote adipose tissue development: a human

health concern? Journal of lipid research 44, 271-9(2003).

19. Donahue, S.M. et al. Prenatal fatty acid status and child adiposity at age 3 y: results from a US pregnancy

cohort. The American journal of clinical nutrition (2011).doi:10.3945/ajcn.110.005801

20. Madsen, L. et al. cAMP-dependent signaling regulates the adipogenic effect of n-6 polyunsaturated fatty acids.

The Journal of biological chemistry 283, 7196-205(2008).

21. Groeger, A.L. et al. Cyclooxygenase-2 generates anti-inflammatory mediators from omega-3 fatty acids. Nature

chemical biology 6, 433-41(2010).

22. Iyer, A. et al. Inflammatory lipid mediators in adipocyte function and obesity. Nature reviews. Endocrinology 6,

71-82(2010).

23. Javadi, M. et al. The effect of six different C18 fatty acids on body fat and energy metabolism in mice. The

British journal of nutrition 92, 391-9(2004).

24. Couet, C. et al. Effect of dietary fish oil on body fat mass and basal fat oxidation in healthy adults. International

journal of obesity and related metabolic disorders : journal of the International Association for the Study of

Obesity 21, 637-43(1997).

25. Oh, D.Y. et al. GPR120 Is an Omega-3 Fatty Acid Receptor Mediating Potent Anti-inflammatory and Insulin-

Sensitizing Effects. Cell 142, 687-698(2010).

26. Guerre-Millo, M. Adipose tissue and adipokines: for better or worse. Diabetes & metabolism 30, 13-9(2004).

27. Cook, K.S. et al. Adipsin: a circulating serine protease homolog secreted by adipose tissue and sciatic nerve.

Science (New York, N.Y.) 237, 402-5(1987).

28. Ouchi, N. et al. Adipokines in inflammation and metabolic disease. Nature reviews. Immunology 11, 85-

97(2011).

29. Nedergaard, J., Bengtsson, T. & Cannon, B. Unexpected evidence for active brown adipose tissue in adult

humans. American journal of physiology. Endocrinology and metabolism 293, E444-52(2007).

30. Saltiel, A.R. & Kahn, C.R. Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414,

799-806(2001).

31. Madsen, L. & Kristiansen, K. The importance of dietary modulation of cAMP and insulin signaling in adipose

tissue and the development of obesity. Annals of the New York Academy of Sciences 1190, 1-14(2010).

32. Blüher, M. et al. Adipose Tissue Selective Insulin Receptor Knockout Protects against Obesity and Obesity-

Related Glucose Intolerance. Developmental Cell 3, 25-38(2002).

33. Kim, J.K. et al. Redistribution of substrates to adipose tissue promotes obesity in mice with selective insulin

resistance in muscle. The Journal of clinical investigation 105, 1791-7(2000).

34. Shepherd, P.R. et al. Adipose cell hyperplasia and enhanced glucose disposal in transgenic mice overexpressing

GLUT4 selectively in adipose tissue. The Journal of biological chemistry 268, 22243-6(1993).

35. Murata, Y. et al. IRS-1 transgenic mice show increased epididymal fat mass and insulin resistance. Biochemical

and biophysical research communications 364, 301-7(2007).

18

36. Petersen, R.K. et al. Arachidonic acid-dependent inhibition of adipocyte differentiation requires PKA activity

and is associated with sustained expression of cyclooxygenases. Journal of lipid research 44, 2320-30(2003).

37. Reginato, M.J. Prostaglandins Promote and Block Adipogenesis through Opposing Effects on Peroxisome

Proliferator-activated Receptor gamma. Journal of Biological Chemistry 273, 1855-1858(1998).

38. Madsen, L. et al. cAMP-dependent signaling regulates the adipogenic effect of n-6 polyunsaturated fatty acids.

The Journal of biological chemistry 283, 7196-205(2008).

39. Papadaki, A. et al. The effect of protein and glycemic index on childrenʼs body composition: the DiOGenes

randomized study. Pediatrics 126, e1143-52(2010).

40. Larsen, T.M. et al. Diets with high or low protein content and glycemic index for weight-loss maintenance. The

New England journal of medicine 363, 2102-13(2010).

41. Grey, N.J., Goldring, S. & Kipnis, D.M. The effect of fasting, diet, and actinomycin D on insulin secretion in

the rat. The Journal of clinical investigation 49, 881-9(1970).

42. Rossetti, L. et al. Effect of dietary protein on in vivo insulin action and liver glycogen repletion. Am J Physiol

Endocrinol Metab 257, E212-219(1989).

43. Linn, T. et al. Effect of long-term dietary protein intake on glucose metabolism in humans. Diabetologia 43,

1257-65(2000).

44. Tremblay, F. et al. Role of dietary proteins and amino acids in the pathogenesis of insulin resistance. Annual

review of nutrition 27, 293-310(2007).

45. Westerterp-Plantenga, M.S. et al. Dietary protein, metabolism, and body-weight regulation: dose–response

effects. International Journal of Obesity 30, S16-S23(2006).

46. Reeves, P.G., Nielsen, F.H. & Fahey, G.C. AIN-93 purified diets for laboratory rodents: final report of the

American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet. The

Journal of nutrition 123, 1939-51(1993).

47. Patti, M.E. et al. Bidirectional modulation of insulin action by amino acids. The Journal of clinical investigation

101, 1519-29(1998).

48. Promintzer, M. & Krebs, M. Effects of dietary protein on glucose homeostasis. Current opinion in clinical

nutrition and metabolic care 9, 463-8(2006).

49. Krebs, M. et al. The Mammalian target of rapamycin pathway regulates nutrient-sensitive glucose uptake in

man. Diabetes 56, 1600-7(2007).

50. Newgard, C.B. et al. A branched-chain amino acid-related metabolic signature that differentiates obese and lean

humans and contributes to insulin resistance. Cell metabolism 9, 311-26(2009).

51. Wang, T.J. et al. Metabolite profiles and the risk of developing diabetes. Nature Medicine 17, 448-453(2011).

52. Ferré, P. & Foufelle, F. SREBP-1c transcription factor and lipid homeostasis: clinical perspective. Hormone

research 68, 72-82(2007).

53. Xu, J. Sterol Regulatory Element Binding Protein-1 Expression Is Suppressed by Dietary Polyunsaturated Fatty

Acids. A MECHANISM FOR THE COORDINATE SUPPRESSION OF LIPOGENIC GENES BY

POLYUNSATURATED FATS. Journal of Biological Chemistry 274, 23577-23583(1999).

19

54. Xu, J. et al. Polyunsaturated fatty acids suppress hepatic sterol regulatory element-binding protein-1 expression

by accelerating transcript decay. The Journal of biological chemistry 276, 9800-7(2001).

55. Takeuchi, Y. et al. Polyunsaturated fatty acids selectively suppress sterol regulatory element-binding protein-1

through proteolytic processing and autoloop regulatory circuit. The Journal of biological chemistry 285, 11681-

91(2010).

56. Ou, J. et al. Unsaturated fatty acids inhibit transcription of the sterol regulatory element-binding protein-1c

(SREBP-1c) gene by antagonizing ligand-dependent activation of the LXR. Proceedings of the National

Academy of Sciences of the United States of America 98, 6027-32(2001).

57. Horton, J.D., Goldstein, J.L. & Brown, M.S. SREBPs: activators of the complete program of cholesterol and

fatty acid synthesis in the liver. The Journal of clinical investigation 109, 1125-31(2002).

58. Matsuzaka, T. et al. Insulin-Independent Induction of Sterol Regulatory Element-Binding Protein-1c Expression

in the Livers of Streptozotocin-Treated Mice. Diabetes 53, 560-569(2004).

59. Kawaguchi, T. et al. Glucose and cAMP regulate the L-type pyruvate kinase gene by

phosphorylation/dephosphorylation of the carbohydrate response element binding protein. Proceedings of the

National Academy of Sciences of the United States of America 98, 13710-5(2001).

60. Dentin, R. et al. Polyunsaturated fatty acids suppress glycolytic and lipogenic genes through the inhibition of

ChREBP nuclear protein translocation. The Journal of clinical investigation 115, 2843-54(2005).

61. Mitro, N. et al. The nuclear receptor LXR is a glucose sensor. Nature 445, 219-23(2007).

62. Denechaud, P.-D. et al. ChREBP, but not LXRs, is required for the induction of glucose-regulated genes in

mouse liver. The Journal of clinical investigation 118, 956-64(2008).

63. Lazar, M.A. & Willson, T.M. Sweet dreams for LXR. Cell metabolism 5, 159-61(2007).

64. Anthonisen, E.H. et al. Nuclear receptor liver X receptor is O-GlcNAc-modified in response to glucose. The

Journal of biological chemistry 285, 1607-15(2010).

65. Porstmann, T. et al. SREBP activity is regulated by mTORC1 and contributes to Akt-dependent cell growth.

Cell metabolism 8, 224-36(2008).

66. Li, S., Brown, M.S. & Goldstein, J.L. Bifurcation of insulin signaling pathway in rat liver: mTORC1 required

for stimulation of lipogenesis, but not inhibition of gluconeogenesis. Proceedings of the National Academy of

Sciences of the United States of America 107, 3441-6(2010).

67. Kawaguchi, T. et al. Mechanism for fatty acid “sparing” effect on glucose-induced transcription: regulation of

carbohydrate-responsive element-binding protein by AMP-activated protein kinase. The Journal of biological

chemistry 277, 3829-35(2002).

68. Li, Y. et al. AMPK Phosphorylates and Inhibits SREBP Activity to Attenuate Hepatic Steatosis and

Atherosclerosis in Diet-Induced Insulin-Resistant Mice. Cell metabolism 13, 376-88(2011).

69. Basciano, H., Federico, L. & Adeli, K. Fructose, insulin resistance, and metabolic dyslipidemia. Nutrition &

metabolism 2, 5(2005).

70. Tappy, L. & Lê, K.-A. Metabolic effects of fructose and the worldwide increase in obesity. Physiological

reviews 90, 23-46(2010).

20

71. Mayes, P.A. Intermediary metabolism of fructose. The American journal of clinical nutrition 58, 754S-

765S(1993).

72. Nagata, R. et al. Single nucleotide polymorphism (-468 Gly to A) at the promoter region of SREBP-1c

associates with genetic defect of fructose-induced hepatic lipogenesis [corrected]. The Journal of biological

chemistry 279, 29031-42(2004).

73. Nagai, Y. et al. Amelioration of high fructose-induced metabolic derangements by activation of PPARalpha.

American journal of physiology. Endocrinology and metabolism 282, E1180-90(2002).

74. Bergheim, I. et al. Antibiotics protect against fructose-induced hepatic lipid accumulation in mice: role of

endotoxin. Journal of hepatology 48, 983-92(2008).

75. Miller, C.C. et al. Intracerebroventricular Injection of Fructose Stimulates Feeding in Rats. Nutritional

Neuroscience 5, 359-362(2002).

76. Cha, S.H. et al. Differential effects of central fructose and glucose on hypothalamic malonyl-CoA and food

intake. Proceedings of the National Academy of Sciences of the United States of America 105, 16871-5(2008).

77. Cha, S.H. & Lane, M.D. Central lactate metabolism suppresses food intake via the hypothalamic AMP

kinase/malonyl-CoA signaling pathway. Biochemical and biophysical research communications 386, 212-

6(2009).

78. Thresher, J.S. et al. Comparison of the effects of sucrose and fructose on insulin action and glucose tolerance.

Am J Physiol Regulatory Integrative Comp Physiol 279, R1334-1340(2000).

79. Doughman, S.D., Krupanidhi, S. & Sanjeevi, C.B. Omega-3 fatty acids for nutrition and medicine: considering

microalgae oil as a vegetarian source of EPA and DHA. Current diabetes reviews 3, 198-203(2007).

80. The International Programme on Chemical Safety PERSISTENT ORGANIC POLLUTANTS. (1996).

81. Lee, D.-H. et al. A strong dose-response relation between serum concentrations of persistent organic pollutants

and diabetes: results from the National Health and Examination Survey 1999-2002. Diabetes care 29, 1638-

44(2006).

82. Lee, D.-H. et al. Association between serum concentrations of persistent organic pollutants and insulin

resistance among nondiabetic adults: results from the National Health and Nutrition Examination Survey 1999-

2002. Diabetes care 30, 622-8(2007).

83. Lim, S. et al. Chronic exposure to the herbicide, atrazine, causes mitochondrial dysfunction and insulin

resistance. PloS one 4, e5186(2009).

84. Ruzzin, J. et al. Persistent organic pollutant exposure leads to insulin resistance syndrome. Environmental

health perspectives 118, 465-71(2010).

85. Berntssen, M.H.G. et al. Reducing persistent organic pollutants while maintaining long chain omega-3 fatty

acid in farmed Atlantic salmon using decontaminated fish oils for an entire production cycle. Chemosphere 81,

242-52(2010).

21

Annexes

1. Ma T, Liaset B, Hao Q, Petersen RK, Fjære E, et al. (2011) Sucrose Counteracts the Anti-nflammatory Effect of Fish Oil in Adipose Tissue and Increases Obesity Development in Mice. PLoS ONE 6(6): e21647. doi:10.1371/journal.pone.0021647

2. Hao Q, Lillefosse HH, Fjære E, … Ma T, et al. (2011). Carbohydrate Source and Insulin Secretion

Modulate the Obesity Promoting Effect of Fish Oil. Manuscript in revision.

3. Madsen L, Pedersen LM, Lillefosse HH, Fjære E, … Ma T, et al. (2010) UCP1 Induction during Recruitment of Brown Adipocytes in White Adipose Tissue Is Dependent on Cyclooxygenase Activity. PLoS ONE 5(6): e11391. doi:10.1371/journal.pone.0011391.

4. Liaset B, Hao Q, Jørgensen H, Hallenborg P, Du ZY, Ma T, et al. (2011). Nutritional regulation of bile acid metabolism improves pathological characteristics of the metabolic syndrome. J Biol Chem., in press.

5. Ruzzin J, Petersen RK, Meugnier E, Madsen L, … Ma T, et al. (2010). Persistent organic pollutant exposure leads to insulin resistance syndrome. Environ Health Perspect. 118(4):465-71.

Sucrose Counteracts the Anti-Inflammatory Effect of FishOil in Adipose Tissue and Increases ObesityDevelopment in MiceTao Ma1, Bjørn Liaset2, Qin Hao1, Rasmus Koefoed Petersen1, Even Fjære1,2, Ha Thi Ngo2¤, Haldis Haukas

Lillefosse1,2, Stine Ringholm1, Si Brask Sonne1, Jonas Thue Treebak3, Henriette Pilegaard1, Livar

Frøyland2, Karsten Kristiansen1*, Lise Madsen1,2*

1 Department of Biology, University of Copenhagen, Copenhagen, Denmark, 2 National Institute of Nutrition and Seafood Research (NIFES), Bergen, Norway,

3 Department of Exercise and Sport Sciences, University of Copenhagen, Copenhagen, Denmark

Abstract

Background: Polyunsaturated n-3 fatty acids (n-3 PUFAs) are reported to protect against high fat diet-induced obesity andinflammation in adipose tissue. Here we aimed to investigate if the amount of sucrose in the background diet influences theability of n-3 PUFAs to protect against diet-induced obesity, adipose tissue inflammation and glucose intolerance.

Methodology/Principal Findings: We fed C57BL/6J mice a protein- (casein) or sucrose-based high fat diet supplementedwith fish oil or corn oil for 9 weeks. Irrespective of the fatty acid source, mice fed diets rich in sucrose became obesewhereas mice fed high protein diets remained lean. Inclusion of sucrose in the diet also counteracted the well-known anti-inflammatory effect of fish oil in adipose tissue, but did not impair the ability of fish oil to prevent accumulation of fat in theliver. Calculation of HOMA-IR indicated that mice fed high levels of proteins remained insulin sensitive, whereas insulinsensitivity was reduced in the obese mice fed sucrose irrespectively of the fat source. We show that a high fat diet decreasedglucose tolerance in the mice independently of both obesity and dietary levels of n-3 PUFAs and sucrose. Of note,increasing the protein:sucrose ratio in high fat diets decreased energy efficiency irrespective of fat source. This wasaccompanied by increased expression of Ppargc1a (peroxisome proliferator-activated receptor, gamma, coactivator 1 alpha)and increased gluconeogenesis in the fed state.

Conclusions/Significance: The background diet influence the ability of n-3 PUFAs to protect against development ofobesity, glucose intolerance and adipose tissue inflammation. High levels of dietary sucrose counteract the anti-inflammatory effect of fish oil in adipose tissue and increases obesity development in mice.

Citation: Ma T, Liaset B, Hao Q, Petersen RK, Fjære E, et al. (2011) Sucrose Counteracts the Anti-Inflammatory Effect of Fish Oil in Adipose Tissue and IncreasesObesity Development in Mice. PLoS ONE 6(6): e21647. doi:10.1371/journal.pone.0021647

Editor: Aimin Xu, University of Hong Kong, China

Received March 28, 2011; Accepted June 4, 2011; Published June 28, 2011

Copyright: � 2011 Ma et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricteduse, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by the Danish Natural Science Research Council, the Novo Nordisk Foundation, the Carlsberg Foundation, and the NationalInstitute of Nutrition and Seafood Research, Norway. Part of the work was carried out as a part of the research program of the Danish Obesity Research Centre(DanORC). DanORC is supported by the Danish Council for Strategic Research (Grant NO 2101 06 0005). The funders had no role in study design, data collectionand analysis, decision to publish, or preparation of the manuscript.

Competing Interests: The authors have declared that no competing interests exist.

* E-mail: [email protected] (KK); [email protected] (LM)

¤ Current address: Department of Food Safety and Nutrition, Division of Environmental Medicine, Norwegian Institute of Public Health, Oslo, Norway

Introduction

Today it is recognized that the potentially harmful effects of

high fat diets relates to not only the amount, but also the type of

dietary fatty acids. Whereas a high intake of saturated and trans

fatty acids has been shown to be associated with increased risk of

cardiovascular diseases in several studies, intake of polyunsaturat-

ed fatty acids (PUFAs) has been associated with lower cardiovas-

cular risk [1,2]. Thus, increasing the relative amount of PUFAs,

both vegetable n-6 PUFAs and marine n-3 PUFA, at the expense

of saturated fat is recommended. It is important to note, however,

that more than 85% of the total dietary PUFA intake in Western

diets today is vegetable n-6 PUFAs, mainly linoleic acid [3]. This is

largely due to the high amount of linoleic acid in corn-, sunflower-,

and soybean-oil used in both home-cooking and in industrially

prepared food [4]. Moreover, animal feeds are enriched with n-6

PUFAs, and although meat production methods are diverse, meat

fatty acid profiles will always reflect that of the animal feed [4].

Thus, the dietary n-3:n-6 PUFA ratio has decreased [3,4]. Al-

though the exchange of saturated fat with vegetable n-6 PUFAs

may have some beneficial effects on human health, a low n-3:n-6

PUFAs ratio is associated with a high risk of several lipid-related

disorders [2,3]. A high intake of n-6 PUFAs has also been asso-

ciated with childhood obesity, [4,5]. Animal studies have shown

that feeding mice a diet containing the n-6 PUFA, linoleic acid,

during the pregnancy-lactation period leads to obesity in the

offspring [6]. This effect, however, is prevented by inclusion of the

n-3 PUFA a-linolenic acid in the diet [6]. These findings are in

line with several studies demonstrating that dietary n-3 PUFAs are

able to limit the development of diet-induced obesity [7–13].

PLoS ONE | www.plosone.org 1 June 2011 | Volume 6 | Issue 6 | e21647

Obesity may be considered as a state of chronic low-grade

inflammation [14,15]. Accumulated evidence strongly suggests

that low grade chronic inflammation plays a crucial role in de-

velopment of obesity related insulin resistance [16]. Furthermore,

it should be noted that continuous subcutaneous infusion of lipo-

polysaccharide (LPS) is sufficient to induce adipose tissue inflam-

mation, insulin resistance and obesity in mice [17]. It is also well

documented that n-3 PUFAs are able to limit high fat diet-induced

inflammation in adipose tissue in rodents [18–20]. Both n-3

PUFAs and n-6 PUFAs are substrats for cyclo- and lipoxygenases

and n-3 PUFAs are traditionally assumed to act anti-innflamma-

tory by competitive inhibition of the biosynthesis of arachidonic

acid-derived pro-inflammatory prostaglandins of the 2-series and

furthermore, n-3 PUFA-derived prostaglandins of the 3-series

are believed to be less inflammatory [21,22]. Recent research

furthermore demonstrate that n-3 PUFAs may be converted to

anti-inflammatory cyclooxygenase-2 derived electrophilic oxoder-

ivatives and resolvins [21,22]. Moreover, by activation of the fatty

acid receptor, GPR120, n-3 PUFAs repress LPS- and TNFa-

mediated inflammatory signalling responses, and thereby increase

insulin sensitivity by repressing macrophage-induced adipose tissue

inflammation [23]. Thus, consumption of n-6 PUFAs at the

expense of n-3 PUFAs may aggravate the metabolic consequences

of obesity. Increasing the dietary intake of n-3 PUFAs is therefore

currently recommended by several health authorities.

In order to curb the increasing obesity problem, nutritionists

and authorities have largely focused on reducing fat intake, as

dietary fat contains more energy per gram than proteins and car-

bohydrates. As an alternative to low energy diets, low carbohy-

drate diets are becoming increasingly popular although still con-

troversial. The mechanisms by which such diets induce weigh loss

are still not fully elucidated, but it has been documented that high

protein diets increase energy expenditure in part due to a thermic

effect [24]. We have previously shown that the protein:sucrose

ratio in the background diet determines the adipogenic potential

of dietary n-6 PUFAs in mice [25]. Mice fed n-6 PUFAs in

combination with sucrose became obese, and had a markedly

higher feed efficiency than mice pair-fed n-6 PUFAs in combination

with proteins [25]. In fact, the high-protein fed mice needed

almost 7 times more energy to achieve a weight gain of 1 g than

mice on the high-sucrose diet [25]. The high protein diet led to

an increased glucagon/insulin ratio, concomitant with elevated

cAMP-dependent signaling, induction of COX-mediated prosta-

glandin synthesis and increased expression of uncoupling protein-1

(UCP1) in inguinal subcutaneous white fat [25]. In the present

paper we aimed to investigate whether this phenomenon is

restricted to n-6 PUFAs or if the effects of dietary fats, such as fish

oils, which are considered beneficial to human health, also depend

on the background diets. Furthermore, we aimed to examine

whether the background diet exerts an influence on the ability of

n-3 PUFAs to protect against glucose intolerance and adipose

tissue inflammation.

Results

Sucrose counteracts the obesity-reducing effect of fishoil in ad libitum fed mice

It is a general notion that intake of fish oil rich in n-3 PUFAs

limits high fat diet-induced obesity in rodents, whereas diets rich in

n-6 PUFAs have been associated with an increased propensity to

develop obesity [6,26]. As we have demonstrated that the

obesogenic effect of n-6 PUFAs is determined by the content of

carbohydrates and protein in the feed [25], we speculated whether

the effect of dietary fats considered health-beneficial, such as fish

oil, might be modulated by different background diets. To answer

this question we fed C57BL/6J male mice isocaloric high fat diets

(Table 1 and 2) containing corn oil or fish oil supplemented with

either protein or sucrose or a conventional low fat diet ad libitum for

9 weeks. Contrasting the general notion that fish oil attenuates

high fat diet-induced obesity, the mice fed the fish oil in com-

bination with sucrose gained as much body weight as the mice fed

corn oil and sucrose (Fig. 1A and B). When combined with

sucrose, fish oil did not reduce the weights of neither epididymal

(eWAT) nor inguinal white adipose tissue (iWAT) mass compared

with corn oil (Fig. 1E and F). Moreover, morphological analyses

demonstrated that the adipocyte size was similar in the two sucrose

fed groups (Fig. 1G). Of note, weight gain in mice fed corn oil or

fish oil plus protein were indistinguishable from that of mice fed

the low fat diet (Fig. 1A). Compared with low fat fed mice, the

weights and adipocytes sizes of eWAT and iWAT in mice fed both

high fat diets in combination with protein tended to be smaller, but

the differences did not reach statistical significance (Fig. 1E, F

and G). Thus, when combined with a high intake of sucrose fish oil

did not prevent obesity. However, high dietary protein content

prevented weight gain and obesity when combined with either

corn or fish oil.

Sucrose, but not protein or fat, strongly stimulates pancreatic

insulin secretion, and accordingly, plasma levels of insulin were

consistently higher in mice fed the sucrose-based diets than in mice

fed the protein-based diets (Fig. 1C). Conversely, the levels of

plasma glucagon were lower and hence, the insulin:glucagon ratio

was about three times higher in mice fed high sucrose than in mice

fed high protein irrespective of whether the diets contained corn

oil or fish oil (Fig. 1D). Collectively, these results indicate that

intake of sucrose and hence increased insulin secretion, abrogates

the protective effects of fish oil in relation to adipocyte hyperplasia

and hypertrophy and thereby the development obesity.

Sucrose counteracts the anti-inflammatory effect of fishoil in adipose tissue

The ability of n-3 PUFAs to limit high fat diet-induced

inflammation in adipose tissue is well documented [18–20]. As

chronic low grade inflammation in adipose tissue is a characteristic

trait of obesity [14,15] and sucrose abrogates the anti-adipogenic

effect of fish oil, we asked whether the background diet also

attenuated the ability of n-3 PUFAs to protect against adipose

tissue inflammation. Gene expression analyses of eWAT and

iWAT revealed a striking correlation between macrophage- and

inflammatory markers and the intake of sucrose-based diets irres-

pectively of the fat source (Fig. 2A). Expressions of macrophage

marker genes Emr1 (EGF-like module containing, mucin-like,

hormone receptor-like sequence 1 or F4/80) and Cd68, as well as

markers of inflammation Serpine1 (Plasminogen activator inhibitor-

1) and Ccl2 (chemokine (C-C motif) ligand 2), were significantly

higher in adipose tissue from mice fed sucrose than in mice fed

high protein or a low fat diets (Fig. 2A). Moreover, we noticed

a significant increase in the expression of Pparg (peroxisome

proliferator-activated receptor c) in eWAT in the mice fed protein

supplemented with corn oil compared with mice fed protein

supplemented with fish oil (Fig. 2A).

A high fat diet impairs glucose tolerance independent ofmacronutrient composition and obesity

As adipose tissue inflammation is causally linked to development

of insulin resistance and glucose intolerance, we subjected mice fed

the different diets for 9 weeks to an intraperitoneal glucose

tolerance test (GTT). Surprisingly, the GTT demonstrated that

Fish Oil, Sucrose and Obesity


the glucose tolerance was impaired both in the mice fed the

protein-based diets and in the mice fed the sucrose-based diets

(Fig. 2B). Evidently, impaired glucose tolerance was dissociated

from the state of obesity, suggesting that intake of relatively high

amounts of fat reduces glucose tolerance even if weight gain and

expression of inflammatory markers were maintained at low levels.

However, fasting glucose and insulin levels were lower in mice fed

high protein than high sucrose. Thus, calculation of HOMA-IR

indicated that the mice fed proteins remained insulin sensitive,

whereas insulin sensitivity tended to be reduced in the obese mice

fed sucrose even though the difference between sucrose and

protein fed mice did not reach statistical significance (Fig. 2B).

Sucrose does not reduce the ability of fish oil to preventdiet-induced accumulation of fat in the liver

As the anti-inflammatory effect of n-3 PUFAs in adipose tissue is

well documented, we investigated if the high level of dietary

sucrose reduced uptake of n-3 PUFAs. Thus, GC-MS analyses

were performed to determine the fatty acid composition in red

blood cells, liver and adipose tissues. These analyses demonstrated

the expected enrichment of n-3 PUFA in lipids in red blood cells

and liver (Table 3 and 4). In adipose tissue, the enrichment of n-3

was actually higher in mice fed the sucrose-based fish oil diet than

in the protein-based fish oil diet group (Table 4). However,

inclusion of sucrose in the diet did not reduce the ability of fish oil

to prevent accumulation of fat in the liver (Fig. 3A). When sucrose

was included in the diet, lipid accumulation in livers from fish oil

fed mice was significantly lower than in livers from mice fed corn

oil (Fig. 3A). Moreover, expressions of lipogenic genes seem to be

determined by the sucrose: protein ratio independent of fat source

(Fig. 3B). Thus, the ability of fish oil, but not corn oil, to protect

against diet-induced lipid accumulation in the liver did not seem to

be directly related to the suppression of lipogenic gene expression.

Other hallmarks of n-3 PUFA actions are their ability to

increase fatty acid oxidation and to reduce plasma triacylglycerol

levels [27,28]. Plasma triacylglycerol levels were significantly

reduced in mice fed fish oil in combination with proteins, but

inclusion of sucrose abrogated this effect (Fig. 3C). The higher

plasma levels of b-hydroxybutyrate in mice fed fish oil in

combination with proteins indicated that hepatic fatty acid

oxidation was increased in these mice, and inclusion of sucrose

attenuated this effect (Fig. 3C). Together these results demonstrate

that the protein:sucrose ratio also affects the ability of fish oil to

reduce plasma levels of triacylglycerol and increase fatty acid

oxidation.

Increasing the protein:sucrose ratio in a high fat dietdecreases energy efficiency irrespective of corn or fish oilsupplementation

To verify that the obesity in mice fed fish oil in combination

with sucrose was simply not due to increased energy-intake, feed

intake was recorded and energy efficiency calculated. Obviously,

energy intake was significantly higher in mice fed high fat diets

than that of mice receiving the low fat diet (Fig. 3D). Energy intake

tended to be higher in mice receiving the sucrose diets than in

mice fed the protein-based diets, but this was not statistically

significant (Fig. 3D). Thus, energy efficiency was dramatically

increased in mice receiving sucrose compared to protein, indi-

cating difference in energy expenditure. A simple way to detect

differences in catabolic rate is to subject mice to fasting and

measure the resulting weight loss. Figure 3D shows that mice on

the protein-based diets lost significantly more weight during 18 h

of fasting. This supports the notion that energy expenditure is

higher in mice on a protein-based diet irrespective of whether the

diet is supplemented with corn oil or fish oil.

Expression and activation of UCP1 in brown and white adipose

tissue lead to dissipation of energy in the form of heat, and may

thus protect against diet induced obesity [29]. Gene expression

analyses of adipose tissues demonstrated that expression of Ucp1

(uncoupling protein-1), in mice fed high protein was higher in

iWAT, but not in eWAT or (iBAT) (Fig. 3E). Increased expres-

sion of Ucp1 in iWAT in mice fed the protein-based diets was

accompanied by increased expression of Cpt1b (carnitine palmi-

toyltransferase-1b), Ppargc1a (peroxisome proliferator-activated

receptor gamma coactivator 1 alpha) and Dio2 (deiodinase, iodothy-

ronine, type II), suggesting that iWAT adopted a more brown-like

phenotype (Fig. 3E). Thus, the lean phenotype in mice fed the high

protein diets, appears, at least in part, to result from increased

uncoupled respiration in iWAT.

The obesogenic effect of fish oil is determined by themacronutrient composition in pair-fed mice

Since energy intake was slightly higher in mice receiving fish oil

in combination with sucrose compared with protein, we decided to

demonstrate directly that this difference was insufficient to account

for the increased adipose tissue mass. Accordingly, mice were pair-

fed the isocaloric diets containing fish oil in combination with

sucrose or protein. To achieve identical energy intake we recorded

Table 1. Macronutrient composition in the diets.

High fish oil High corn oil

Low energy Sucrose Protein Sucrose Protein

Protein (g/kg) 200 200 540 200 540

Casein 200 200 540 200 540

L-Cysteine 3 3 3 3 3

Carbohydrate (g/kg) 619.5 439.5 99.5 439.5 99.5

Corn starch 529.5 9.5 9.5 9.5 9.5

Sucrose 90 430 90 430 90

Fat (g/kg) 70 250 250 250 250

Soybean oil 70 70 70 70 70

Corn oil - - - 180 180

Fish oil - 180 180 - -

doi:10.1371/journal.pone.0021647.t001

Table 2. Fatty acid composition in the diets.



SFA (mg/g) 9.6 53.3 50.3 31.4 31.4

MUFA (mg/g) 17.0 54.3 54.0 65.9 65.9

PUFA (mg/g) 36.2 100.6 99.5 120.8 121.2

n-6 (mg/g) 32.7 40.6 38.9 117.3 117.8

n-3 (mg/g) 3.5 60.0 60.6 3.5 3.4

n-3/n-6 0.11 1.48 1.56 0.03 0.03

Abbreviations: SFA, saturated fatty acids; MUFA, monounsaturated fatty acids;PUFA, polyunsaturated fatty acids.doi:10.1371/journal.pone.0021647.t002



the ad libitum feed intake of mice receiving the protein-based diet,

and restricted the amount of feed to mice receiving the sucrose-

based feed accordingly. Figure 4A demonstrates that even under

conditions of pair-feeding, the mice receiving high sucrose gained

dramatically more weight than those receiving a high protein diet.

Similarly, as observed in ad libitum fed mice, energy efficiency and

adipose tissue mass were significantly higher when mice were fed

sucrose (Fig. 4B and C). Moreover, energy content in the feces was

similar in both groups and the apparent digestibility was not

increased by increased sucrose amount in the diet (Fig. 4B). Plasma

levels of insulin were higher and glucagon lower in mice fed the

sucrose than protein (Fig. 4D). In iWAT, but not eWAT, we

observed a significant induction of Ppargc1a and Ucp1 expression

indicative of the transformation of iWAT into a more brown-like

depot in protein fed mice (Fig. 4E). In iBAT, expressions of Ucp1

and cyt COXII, (cytochrome c oxidase, subunit II) a marker of

mitochondrial content, were not significantly different in mice fed

protein or sucrose (Fig. 4F).

Lower expression levels of inflammatory markers in eWAT and

iWAT in protein fed mice were also confirmed (Fig. 4E). In

addition, as observed in ad libitum fed mice, glucose tolerance was

similarly affected in protein and sucrose fed mice (Fig. 5A). Fasting

levels of insulin were higher in sucrose+fish oil fed mice than the

two other groups, but an ITT test showed no significant difference

between the groups (Supp Fig. 1). Plasma levels of triacylglycerol

were lower and b-hydroxybutyrate were higher in plasma from

mice fed fish oil in combination with protein than in mice fed fish

oil in combination with sucrose (Fig. 5B). However, expression of

genes involved in fatty acid oxidation was not increased in liver

(Fig. 6B) or in muscle (not shown). Actually, expression of the

classical PPARa target Acox1 (acyl-CoA oxidase 1) was higher in

liver of sucrose fed mice (Fig. 6B). Thus, a possible increase in fatty

acid oxidation in protein fed mice as indicated by the elevated

levels of b-hydroxybutyrate did not appear to be due to increased

expression of genes involved in fatty acid oxidation. Of note,

however, higher expressions of Srebf1 (sterol regulatory element

binding transcription factor 1) as well as Acaca (acetyl-Coenzyme A

carboxylase alpha) and Fasn (fatty acid synthase), indicate that

sucrose overrides the suppressive effect of fish oil on lipogenic gene

expression.

Energy expenditure is reduced in mice fed fish oil incombination with sucrose

Mice fed fish oil in combination with sucrose exhibited

increased weight gain and an increased feed efficiency indicative

of decreased energy expenditure. Moreover, mice fed fish oil in

combination with sucrose lost significantly less weight during 18 h

of fasting (Fig. 3C). Therefore, we examined whether energy

Figure 1. Sucrose counteracts the obesity-reducing effect of fish oil in ad libitum fed mice. Male C57BL/6 mice (n = 8) were fed isocalorichigh fish oil or high corn oil diets with different carbohydrate and protein contents ad libitum for 9 weeks. A: Body weight development of ad libitiumfed mice. (B) Prior to termination the mice were photographed. C–D: Insulin and glucagon levels were measured in plasma in the fed state. E–G: Theweights of epididymal and inguinal white adipose tissues were recorded and sections were stained with hematoxylin and eosin. Data are presentedas means 6 SEM. Different small letters denote significant differences between the groups (P,0.05).doi:10.1371/journal.pone.0021647.g001



expenditure was reduced when fish oil was combined with sucrose.

Accordingly, O2 consumption and CO2 production were mea-

sured by indirect calorimetry. Figure 5C shows that O2 con-

sumption both in the light and the dark periods tended to be lower

in mice fed fish oil in combination with sucrose than with protein.

As expected, mice fed fish oil in combination with sucrose had a

higher CO2 production resulting in a statistically significant higher

RER of about 0.9 indicating a lower rate of fatty acid oxidation

(Fig. 5C).

A diet enriched with fish oil and proteins increasesgluconeogenesis

High circulating levels of insulin combined with a low level of

glucagon translate into reduced cAMP signalling in the liver.

Thus, the observed reduced expressions of Crem (cAMP responsive

element modulator), Pde4c (phosphodiesterase 4C, cAMP specific),

Ppargc1a and Pck1 (phosphoenolpyruvate carboxykinase 1, cyto-

solic) as well as reduced expressions of enzymes involved in amino

acid degradation in the liver of sucrose fed mice were anticipated

(Fig. 4D). In the liver PGC1a is induced in response to elevated

levels of cAMP and plays a central role in the control of hepatic

gluconeogenesis [30–32]. In keeping with the increased expres-

sions of Ppargc1a and Pck1 in liver from mice fed fish oil in

combination with protein compared to sucrose, we anticipated

that gluconeogenesis was induced in the fed state in the protein fed

mice. To measure gluconeogenesis in vivo mice fed fish oil in

combination with either protein or sucrose were intraperitoneally

injected with pyruvate both after overnight fasting and in the fed

Figure 2. A high fat diet impairs glucose tolerance independent of macronutrient composition and obesity. A: Expressions ofadipogenic and inflammatory marker genes (Pparg (peroxisome proliferator activated receptor c), Adipoq (adiponectin), Serpine1 (Plasminogenactivator inhibitor-1), Ccl2 (chemokine (C-C motif) ligand 2), Emr1 (EGF-like module containing, mucin-like, hormone receptor-like sequence 1 or F4/80) and Cd68 (CD68 antigen)) were measured in epididymal and inguinal white adipose tissue using RT-qPCR (n = 8). B: Intraperitoneal glucosetolerance test was performed in a separate set of mice (n = 10). Fasting glucose and insulin levels were measured to calculate HOMA-IR. Data arepresented as means 6 SEM. Different small letters denote significant differences between the groups, in 2A within the same tissue (P,0.05).doi:10.1371/journal.pone.0021647.g002



state, and blood glucose was measured in the following 60 minutes.

In the fasted state, mice fed sucrose or protein exhibited similar

excursions, indicating similar rates of gluconeogenesis (Fig. 6A). In

fed mice, however, the rise of blood glucose following the injection

of pyruvate was dramatically faster and reached much higher levels

after 15 and 30 min in the protein fed mice than in chow fed mice

(Fig. 6A). Compared with chow fed mice the rise in blood glucose

was also increased in mice fed fish oil in combination with sucrose,

but this was not statistically significant (Fig. 6A). The decline in

blood glucose in chow fed mice remains to be explained, but this

was observed consistently. Taken together these results strongly

support the assumption that gluconeogenesis is markedly induced in

mice fed the protein-based diet.

Discussion

It is well documented that inclusion of n-3 PUFAs in high fat

diets leads to reduced development of diet-induced obesity in

rodents [7–9,11–13,33]. Unfortunately, not all studies where the

anti-obesogenic effects of fish oils are studied provide a detailed

description of the macronutrient composition. However, in

standard commercial available high fat- and very high fat diets,

starch is the most abounded carbohydrate source and the amount

of sucrose is low or absent. Here we show that a high amount of

sucrose in the diet counteracts the obesity-reducing effect of fish oil

as well as the well described anti-inflammatory effect in adipose

tissue [19,23,34,35]. Irrespective of the fatty acid source, mice fed

high protein diets remained lean whereas mice fed diets enriched

in sucrose became obese and had higher expressions of inflam-

matory markers in adipose tissue. Collectively, our results demon-

strate that a high intake of sucrose abrogates the protective effects

of fish oil in development of obesity.

As dietary sucrose, but not protein or fat, stimulates secretion of

insulin from pancreatic b-cells, an increased dietary sucrose:pro-

tein ratio will translate into an increased insulin:glucagon ratio in

the fed state. In this respect the observed higher insulin:glucagon

ratio in mice fed the sucrose-based diets than in mice fed the

protein-based diets was expected. Increased levels of insulin in fed

mice were observed irrespectively of the type of fat in the diet.

Insulin is a powerful anabolic hormone that stimulates adipocyte

differentiation and adipose tissue expansion [36]. Activation of

insulin signaling is crucial for the development of obesity [37] and

insulin receptor substrate-1 (IRS-1) transgenic mice are obese [38].

Increased insulin signaling and glucose uptake in adipose tissue

in the fed state in sucrose fed mice may thus override the

protective effect of fish oil when it comes to protection against

Table 3. Fatty acid composition in red blood cells.



Sum total (mg/g) 3.41±0.04 3.04±0.07 2.87±0.05 3.52±0.11 3.34±0.10

SFA (mg/g) 1.28±0.02 1.24±0.03 1.17±0.02 1.31±0.05 1.21±0.04

MUFA (mg/g) 0.49±0.01 0.34±0.01 0.35±0.01 0.41±0.01 0.40±0.01

PUFA (mg/g) 1.39±0.02 1.28±0.03 1.19±0.02 1.53±0.05 1.44±0.04

n-3 (mg/g) 0.2560.00 0.7960.02 0.7360.01 0.2160.01 0.1160.00

n-6 (mg/g) 1.1460.02 0.4960.01 0.4760.01 1.3260.05 1.3460.04

n-3/n-6 0.2260.00 1.6260.01 1.5560.28 0.1660.00 0.0860.00

Data are presented as mean 6 SEM (n = 8).doi:10.1371/journal.pone.0021647.t003

Table 4. Fatty acid composition in organs.



Liver

Sum total(mg/g)

45±4 41±1 35±2 71±10 42±3

% 100 100 100 100 100

SFA (mg/g) 15±1 15±0 12±1 21±3 13±1

% 32.560.9 36.560.4 35.260.3 29.660.3 31.360.8

MUFA (mg/g) 13±2 6±1 4±0 19±4 7±1

% 29.161.6 14.962.3 11.960.7 25.661.9 17.261.3

PUFA (mg/g) 16±1 19±1 18±1 29±3 20±1

% 36.761.6 47.262.0 51.260.4 42.761.9 47.860.9

n-3 (mg/g) 3.2460.27 12.9460.51 10.8360.72 3.2860.15 1.7260.11

% 7.2560.45 31.4161.25 31.3460.35 4.9760.53 4.1060.28

n-6 (mg/g) 1361 660 760 2663 1861

% 29.461.2 15.860.8 19.960.5 37.761.4 43.760.7

n-3/n-6 0.2560.01 2.0060.03 1.5860.05 0.1360.01 0.0960.01

eWAT

Sum total(mg/g)

871±14 841±17 761±78 840±27 747±43

% 100 100 100 100 100

SFA (mg/g) 198±6 250±8 208±23 155±8 120±7

% 22.760.3 29.760.5 27.360.6 18.460.6 16.160.2

MUFA (mg/g) 372±9 269±6 268±26 305±7 279±12

% 42.760.4 32.160.7 35.461.1 36.460.6 37.560.8

PUFA (mg/g) 295±2 308±8 271±29 375±14 341±24

% 33.960.6 36.660.5 35.560.7 44.560.4 45.460.7

n-3 (mg/g) 18.1660.47 114.8966.64 87.07613.22 8.6860.32 6.2560.54

% 2.0860.04 13.6360.59 11.2060.89 1.0360.02 0.8360.04

n-6 (mg/g) 27762 19063 181618 366614 322622

% 31.860.6 22.660.5 23.960.6 43.560.4 44.660.7

n-3/n-6 0.0760.00 0.6060.03 0.4760.05 0.0260.00 0.0260.00

Data are presented as mean 6 SEM (n = 8).doi:10.1371/journal.pone.0021647.t004



obesity-development. It should also be mentioned that although

several studies have demonstrated a protective effect of fish oil in

obesity-development, it has been reported that inclusion of fish oil

increased the amount of adipose tissue mass in hyperinsulinemic

ob/ob mice [19].

Differences in the insulin:glucagon ratio and hence differences

in cAMP-dependent signaling may at least in part orchestrate the

observed differences in energy homeostasis between the sucrose-

and protein-based diets regardless of whether these diet are

supplemented with corn oil or fish oil. In the liver, Ppargc1a is

induced in response to elevated levels of cAMP and plays a central

role in the control of hepatic gluconeogenesis [30–32]. High

circulating levels of insulin combined with a low level of glucagon

translate into reduced cAMP signalling in the liver. Thus, the

observed increased gluconeogenesis in the fed state in protein fed

mice may result from cAMP-mediated stimulation of Ppargc1a and

Pck-1 expression. Increased gluconeogenesis in the fed state may

contribute to the observed lower energy efficiency in protein fed

Figure 3. Fish oil prevents diet-induced accumulation of fat in the liver. A: Total lipids were extracted from liver and separated using HPTLC.B: Expressions of lipogenic genes (Srebf1 (sterol regulatory element binding transcription factor 1) and Acaca (acetyl-Coenzyme A carboxylase alpha)were measured by RT-qPCR. C: Plasma triacylglycerol and b-hydroxybutyrate were measured in the fed state. D: Energy efficiency was calculatedbased on energy intake and body weight gain. E: Expression levels of brown adipose tissue marker genes (Ucp1 (Uncoupling protein-1), Ppargc1a(peroxisome proliferator-activated receptor gamma, coactivator 1 alpha), Cpt1b (carnitine palmitoyltransferase-1b) and Dio2 (deiodinase,iodothyronine, type II) were measured in white adipose tissues using RT-qPCR. Data are presented as means 6 SEM (n = 8). Different small lettersdenote significant differences between the groups, in 3E within the same tissue (P,0.05).doi:10.1371/journal.pone.0021647.g003



mice, as 6 ATP molecules are consumed per molecule of glucose

synthesized from pyruvate, rendering gluconeogenesis an energy-

consuming process. Moreover, concomitant increased expressions

of Gpt, Got1, Agxt and Cps1 suggest that energy consuming

processes such as amino acid degradation and ureagenesis are

higher in protein than sucrose fed mice. As mammals have no

direct storage capacity for protein it needs to be metabolically

processed immediately. The high cost of urea production and

gluconeogenesis is actually often cited reasons for the higher

thermic effect of protein than other macronutrients [39,40] and

this may partly explain why diets higher in protein exert a larger

effect on energy expenditure than diets lower in protein [24].

A second mechanism by which a low sucrose:protein ratio in the

diet leads to reduced energy efficiency may be related to the observed

expression of Ucp1 in iWAT. Increased cAMP-signaling is known to

induce adaptive thermogenesis by induction of Ppargc1a and Ucp1

expression and it is well known that the UCP1 protein allows

dissipation of energy in the form of heat [41]. Of note, acute or

chronic upregulation of fatty acid oxidation alone, that is increased

fatty acid oxidation without a concomitant uncoupling of mitochon-

dria, has no net effect on whole-body energy expenditure or adiposity

[42]. Although Ucp1 expression was unchanged in iBAT, whole body

energy homeostasis may be influenced by increased expression in

iWAT. In fact, increased occurrence of brown-like adipocytes within

WAT depots is a feature of mouse strains resistant to dietary obesity,

such as the A/J strain [43] and reduced adiposity associated with aP2-

transgenic expression of Ucp1 is linked to increased energy dissipation

in white, but not interscapular brown, adipose tissue [29].

Conversely, inhibition of diet-induced expression of Ucp1 in iWAT

in Sv129 mice by administration of a general cyclooxygenase

inhibitor accentuates obesity-development [44].

Our finding that inclusion of sucrose abolishes the anti-obesity

effect of fish oil seems to contradict a recent study from Sato et al.

[45], as these authors demonstrated that inclusion of 5% the n-3

PUFA EPA (eicosapentaenoic acid) into a high fat-high sucrose

diet reduced body weight gain in mice. The reason for this

discrepancy is not clear, but different dietary compositions as well

as doses and type of n-3 PUFAs may account for the different

Figure 4. Sucrose counteracts the obesity-reducing effect of fish oil in pair-fed mice. Male C57BL/6 mice (n = 8) were pair-fed isocaloric highfish oil diets with different carbohydrate and protein contents for 8 weeks. A: Body weight development was followed throughout the feedingregime. B: Energy efficiency was calculated based on energy intake, weight gain and apparent digestibility. C: The weights of different adipose tissuedepots were recorded. D: Insulin and glucagon levels were measured in plasma in the fed state. E: Inflammation and adipocyte marker genes (Pparg(peroxisome proliferator-activated receptor c), Adipoq (adiponectin), Serpine1 (Plasminogen activator inhibitor-1), Ccl2 (chemokine (C-C motif) ligand2), Emr1 (EGF-like module containing, mucin-like, hormone receptor-like sequence 1 or F4/80) and Cd68 (CD68 antigen) and F: thermogenesis-relatedgenes (Ucp1 (Uncoupling protein-1) and cyt COXII, (cytochrome c oxidase, subunit II) were measured by RT-qPCR in adipose tissues. Data arepresented as means 6 SEM. Different small letters denote significant differences between the groups, in 4E within the same tissue (P,0.05).doi:10.1371/journal.pone.0021647.g004



results obtained. The amount of n-3 PUFAs used in this study is

slightly higher (6% n-3 fatty acids) than the 5% EPA used by Sato

et al.. However, whereas Sato et al. used EPA, the n-3 PUFAs used

in our study comprise a mixture (thereof 3263 g/kg and 1863 g/

kg EPA and DHA (docosahexaenoic acid), respectively). More-

over, in fish oil as used in our study, the n-3 PUFAs are present in

the form of triacylglycerols, whereas Sato et al. used purified EPA

ethyl ester. It should also be mentioned that the main fat source in

the diets used in our study is corn-oil rich in n-6 fatty acids,

whereas Sato et al. used anhydrate milk fat containing more than

60% saturated fat. Last, the amount of sucrose used in our study is

higher than the dose used by Sato et al. It is worth noting, how-

ever, that both the study by Sato et al. and our study demonstrated

that sucrose did not reduce the ability of fish oil and/or EPA to

prevent diet induced accumulation of fat in the liver.

A strong association between obesity and adipose tissue inflam-

mation exists and obesity is characterized by chronic low-grade

inflammation in adipose tissues [14,15]. In light of this it may not

be surprising that expression of macrophage and inflammatory

marker genes was elevated in obese mice compared to lean mice.

Still, as the anti-inflammatory effect of fish oil in adipose tissue is

well described [19,23,34,35], it was unexpected that the expression

of inflammatory markers was similar in adipose tissue from obese

corn oil and the fish oil fed groups. In our study the state of obesity

rather than the n-3:n-6 PUFA ratio in both feed and adipose

tissues appeared to determine the expression levels of inflamma-

tory markers in adipose tissue.

Chronic low grade inflammation plays an important role in

development of insulin resistance [46,47]. Pioneering work by

Storlien et al. [48], later confirmed by several others [13,23,35,

Figure 5. Metabolic parameters in mice fed fish oil in combination with sucrose or protein. A. Intraperitoneal glucose tolerance test wasperformed in mice pair fed fish oil enriched diets (n = 10). B. b-hydroxybutyrate, triacylglycerol, glycerol and free fatty acids were measured in pair-fedmice in both fasted and fed state (n = 10). C. Oxygen consumption, carbon dioxide and respiratory exchange ratio were measured during a 24-hperiod with indirect calorimetry (n = 8). Data are presented as means 6 SEM. Different small letters denote significant differences between thegroups, in 5B between fasted or fed state (P,0.05).doi:10.1371/journal.pone.0021647.g005



49,50] has demonstrated that n-3 PUFAs can prevent develop-

ment of diet-induced insulin resistance in rodents. The insulin

sensitizing effect of n-3 PUFAs is generally accepted to be related

to the anti-inflammatory effect, recently demonstrated to be me-

diated by the GPR120 receptor [23]. Calculation of HOMA-IR

indicated that mice fed high levels of proteins were more

insulin sensitive than mice fed sucrose, but no significant dif-

ferences was observed in an ITT. Similar to expression levels of

inflammatory markers in adipose tissue, this was irrespective of

whether the diets were supplemented with corn oil or fish oil. It

was therefore unexpected that the GTT demonstrated that mice

fed corn oil or fish oil in combination with sucrose or protein

exhibited impaired glucose tolerance irrespective of whether or not

the mice remained lean. It is possible that different mechanisms

underlay the impaired glucose tolerance observed in the sucrose

and the protein fed mice. It is likely that impaired glucose

tolerance in sucrose fed mice is related to the obese state. Of note,

in fish oil and protein fed mice, glucose tolerance was impaired

even if weight gain and inflammation were maintained at low

levels. Further studies are required to elucidate the mechanisms

underlying the impaired glucose tolerance in these mice, but the

possibility that adaption to a low carbohydrate intake with con-

comitant high hepatic gluconeogenesis and glucose output should

be considered.

Seen as a whole, our study indicate that the sucrose:protein

ratio, rather than the n-6:n-3 PUFA ratio in the diet determines

development of obesity, adipose tissue inflammation and glucose

intolerance. Activation of the NF-kB system appears to represent a

link between obesity, inflammation of adipose tissue and insulin

resistance [51–53]. Insulin is able to activate NF-kB by phos-

porylation of IkBa in different cell systems [54], thus, high levels of

circulating insulin may activate the NF-kB system also in adipose

tissue. Whether increased insulin levels in sucrose fed mice

translated into activation of the NF-kB system in adipose tissue in

these mice will require further investigation.

Together our results demonstrate that the background diet

exerts a crucial influence on the ability of n-3 PUFAs to protect

against development of obesity, glucose intolerance and adipose

tissue inflammation. High levels of dietary sucrose counteract the

anti-inflammatory effect of fish oil in adipose tissue and promote

Figure 6. Gluconeogenesis is increased in fed state when animals are fed fish oil supplemented with protein. A. Pyruvate tolerancetests were performed on mice in 16 h fasted (n = 7) and fed states (n = 10). B. Hepatic gene expression (Crem (cAMP responsive element modulator),Pde4c (phosphodiesterase 4C, cAMP specific), Ppargc1a (peroxisome proliferator-activated receptor gamma, coactivator 1 alpha), Pck1(phosphoenolpyruvate carboxykinase 1, cytosolic), Gpt (glutamic pyruvic transaminase), Got1 (glutamate oxaloacetate transaminase 1), Agxt(alanine-glyoxylate aminotransferase), Cps1 (carbamoyl-phosphate synthetase 1), Acox1 (acyl-CoA oxidase 1), Cpt1a (carnitine palmitoyltransferase1a), Cpt2 (carnitine palmitoyltransferase 2), Hmgcs2 (3-hydroxy-3-methylglutaryl-Coenzyme A synthase 2), Srebf1 (sterol regulatory element bindingtranscription factor 1), scd1 (stearoyl-Coenzyme A desaturase 1), Acaca (acetyl-Coenzyme A carboxylase alpha), Fasn (fatty acid synthase), wasmeasured using RT-qPCR (n = 8). Data are presented as means 6 SEM. Different small letters denote significant differences between the differentgroups (P,0.05).doi:10.1371/journal.pone.0021647.g006



obesity development in mice. As the intake of sucrose in Western

societies is high and increasing dietary intake of n-3 PUFAs is

recommended by several health authorities it would be of impor-

tance to investigate whether the background diet influences the

effect of fish oil also in humans.

Materials and Methods

Ethics StatementAll animal experiments were approved by National Animal

Health Authorities (Norwegain approval identification: 1840 and

1841). Care and handling were in accordance with local insti-

tutional recommendations and rules. Adverse events were not

observed.

Animals and dietsMale C57BL/6JBomTac mice approximately 8 weeks of age

were obtained from Taconic Europe (Ejby, Denmark) and were

divided into groups (n = 6–10). The mice were kept at a 12 h light/

dark cycle at 28uC. After acclimatization the animals were fed ad

libitum or pair-fed experimental diets obtained from Ssniff

Spezialdiaten GmbH (Soest, Germany) described in Table 1 and

2 for 8–10 weeks. All diets were supplemented with 3 g/kg L-

cysteine, 10 g/kg choline bitartrate, 10 g/kg Vitamin mix AIN 76

A, 45 g/kg Mineral mix AIN 93 and 0.014 g/kg t-butylhydro-

quinone. Mice were euthanized by cardiac puncture under

anesthesia with Isoflurane (Isoba-vet, Schering-Plough, Denmark)

using the Univentor 400 Anaesthesia Unit (Univentor Limited,

Sweeden) in the fed state and plasma prepared from blood. Tissues

were dissected out, freeze-clamped and frozen at 280uC.

Indirect calorimetryThe metabolic rate of mice was measured by indirect calori-

metry in open circuit chambers of Labmaster system (TSE Systems

GmbH, Germany). The animals were acclimated in the chambers

for 24 hours and measured continuously for another 24 hours.

Glucose, insulin and pyruvate tolerance testing (GTT, ITTand PTT)

GTT: mice were fasted 6 hours before intraperitoneal injection

of 2 g/kg glucose in saline. ITT: mice were fasted 4 hours before

i.p. injection of 0.5 Unit/kg human recombinant insulin in saline.

PTT: mice in the fed state or mice that were fasted overnight were

injected i.p. with of 2 g/kg pyruvate in saline. Blood was collected

from the lateral tail vein at indicated time points and measured

with Bayer Contour glucometer (Bayer A/S, Denmark).

Plasma analysesInsulin, glucagon and glucose [25] and lipid metabolites [55]

were measured in plasma as earlier described.

Lipid analysesTotal lipids were extracted from diets, red blood cells, liver and

adipose tissue samples with chloroform: methanol, 2:1 (v/v). Lipid

classes were analyzed using an automated High-Performance Thin

Layer Chromatography (HPTLC) system (Camaq, Switzerland)

and separated on HPTLC plates coated with silica gel 60 F [55]

whereas fatty acid composition of total lipids was analyzed on a

capillary gas chromatograph with flame ionization detector (Perkin

Elmer, USA) [56].

HistologySections of adipose tissue were fixed, dehydrated, embedded in

paraffin blocks, cut into 3 mm thick section and stained with eosin

and hematoxylin as previously described [57]. Sections were visually

examined using an Olympus BX 51 binocular microscope (System

microscope, Japan), fitted with an Olympus DP50 3.0 camera.

RT-qPCRTotal RNA was purified from mouse tissue using Trizol

(Invitrogen). Reverse transcription (RT) was performed and cDNA

was analyzed in duplicates by qPCR using the ABI PRISM 7700

Sequence Detection System (Applied Biosystems) as earlier

described [58]. Primers for RT-qPCR were designed using Primer

Express 2.0 (Applied Biosystems) and are available on request.

Energy in faeces and dietsEnergy content was determined in a bomb calorimeter fol-

lowing the manufacturer’s instruction (Parr Instruments, Moline,

IL, USA).

StatisticsData represent mean 6 SEM. ANOVA, post hoc pairwise

comparison: Student t-test (RT-qPCR analysis) or Tukey HSD test

(GTT, ITT and organ weights). Newman-Keuls test (nonpara-

metric due to non-homogenous variances) rest of data. Data were

considered statistical significant when P,0.05).

Supporting Information

Figure S1 A high fish oil diet does not impair insulin tolerance.

Male C57BL/6 mice (n = 7) were fed a low fat or isocaloric high

fish oil diets with different carbohydrate and protein contents ad

libitum for 7 weeks. A: Insulin levels were measured in the fasted

state. B: Intraperitoneal insulin tolerance test was performed. Data

are presented as means 6 SEM. Different small letters denote

significant differences between the groups.

(TIF)

Acknowledgments

The authors thank Ase Heltveit and Jan Idar Hjelle at NIFES for excellent

assistance with animal care and lipid analyses.

Author Contributions

Conceived and designed the experiments: BL KK LM. Performed the

experiments: TM BL QH RKP EF HTN HHL SR SBS JTT HP LF LM.

Analyzed the data: TM BL QH RKP EF HTN HHL SR SBS JTT HP LF

KK LM. Wrote the paper: KK LM.

References

1. Erkkila A, de Mello VDF, Riserus U, Laaksonen DE (2008) Dietary fatty acids

and cardiovascular disease: An epidemiological approach. Progress in Lipid

Research 47: 172–187.

2. Harris W (2010) The Omega-3 Index: Clinical Utility for Therapeutic

Intervention. Current Cardiology Reports 12: 503–508.

3. Simopoulos AP (2002) The importance of the ratio of omega-6/omega-3

essential fatty acids. Biomedecine & Pharmacotherapy 56: 365–379.

4. Ailhaud G, Massiera F, Weill P, Legrand P, Alessandri JM, et al. (2006)

Temporal changes in dietary fats: role of n-6 polyunsaturated fatty acids in

excessive adipose tissue development and relationship to obesity. Prog Lipid Res

45: 203–236.

5. Ailhaud G, Massiera F, Alessandri JM, Guesnet P (2007) Fatty acid composition

as an early determinant of childhood obesity. Genes and Nutrition 2: 39–

40.



6. Massiera F, Saint-Marc P, Seydoux J, Murata T, Kobayashi T, et al. (2003)

Arachidonic acid and prostacyclin signaling promote adipose tissue develop-ment: a human health concern? J Lipid Res 44: 271–279.

7. Belzung F, Raclot T, Groscolas R (1993) Fish oil n-3 fatty acids selectively limit

the hypertrophy of abdominal fat depots in growing rats fed high-fat diets.Am J Physiol 246: R1111–R1118.

8. Wang H, Storlien LH, Huang XF (2002) Effects of dietary fat types on bodyfatness, leptin, and ARC leptin receptor, NPY, and AgRP mRNA expression.

Am J Physiol Endocrinol Metab 282: E1352–E1359.

9. Flachs P, Horakova O, Brauner P, Rossmeisl M, Pecina P, et al. (2005)Polyunsaturated fatty acids of marine origin upregulate mitochondrial biogenesis

and induce beta-oxidation in white fat. Diabetologia 48: 2365–2375.10. Ruzickova J, Rossmeisl M, Prazak T, Flachs P, Sponarova J, et al. (2005)

Omega-3 PUFA of marine origin limit diet-induced obesity in mice by reducingcellularity of adipose tissue (vol 39, pg 1177, 2004). Lipids 40: 115.

11. Arai T, Kim HJ, Chiba H, Matsumoto A (2009) Anti-obesity effect of fish oil and fish

oil-fenofibrate combination in female KK mice. J Atheroscler Thromb 16: 675–683.12. Rokling-Andersen MH, Rustan AC, Wensaas AJ, Kaalhus O, Wergedahl H,

et al. (2009) Marine omega-3 fatty acids promote size reduction of visceraladipose depots, without altering body weight and composition, in male Wistar

rats fed a high-fat diet. British Journal of Nutrition 102: 995–1006.

13. Samane S, Christon R, Dombrowski L, Turcotte S, Charrouf Z, et al. (2009)Fish oil and argan oil intake differently modulate insulin resistance and glucose

intolerance in a rat model of dietary-induced obesity. Metabolism-Clinical andExperimental 58: 909–919.

14. de Luca C, Olefsky JM (2008) Inflammation and insulin resistance. FEBS Lett582: 97–105.

15. Hotamisligil GS (2006) Inflammation and metabolic disorders. Nature 444:

860–867.16. Donath MY, Shoelson SE (2011) Type 2 diabetes as an inflammatory disease.

Nat Rev Immunol 11: 98–107.17. Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, et al. (2007) Metabolic

Endotoxemia Initiates Obesity and Insulin Resistance. Diabetes 56: 1761–1772.

18. Huber J, Loffler M, Bilban M, Reimers M, Kadl A, et al. (2007) Prevention ofhigh-fat diet-induced adipose tissue remodeling in obese diabetic mice by n-3

polyunsaturated fatty acids. International Journal of Obesity 31: 1004–1013.19. Todoric J, Loffler M, Huber J, Bilban M, Reimers M, et al. (2006) Adipose tissue

inflammation induced by high-fat diet in obese diabetic mice is prevented by n-3polyunsaturated fatty acids. Diabetologia 49: 2109–2119.

20. Perez-Echarri N, Perez-Matute P, Marcos-Gomez B, Baena MJ, Marti A, et al.

(2008) Differential inflammatory status in rats susceptible or resistant to diet-induced obesity: effects of EPA ethyl ester treatment. Eur J Nutr 47: 380–386.

21. Calder PC (2009) Polyunsaturated fatty acids and inflammatory processes: Newtwists in an old tale. Biochimie 91: 791–795.

22. Groeger AL, Cipollina C, Cole MP, Woodcock SR, Bonacci G, et al. (2010)

Cyclooxygenase-2 generates anti-inflammatory mediators from omega-3 fattyacids. Nature Chemical Biology 6: 433–441.

23. Oh DY, Talukdar S, Bae EJ, Imamura T, Morinaga H, et al. (2010) GPR120 Isan Omega-3 Fatty Acid Receptor Mediating Potent Anti-inflammatory and

Insulin-Sensitizing Effects. Cell 142: 687–698.24. Halton TL, Hu FB (2004) The Effects of High Protein Diets on Thermogenesis,

Satiety and Weight Loss: A Critical Review. J Am Coll Nutr 23: 373–385.

25. Madsen L, Pedersen LM, Liaset B, Ma T, Petersen RK, et al. (2008) cAMP-dependent Signaling Regulates the Adipogenic Effect of n-6 Polyunsaturated

Fatty Acids. J Biol Chem 283: 7196–7205.26. Madsen L, Petersen RK, Kristiansen K (2005) Regulation of adipocyte

differentiation and function by polyunsaturated fatty acids. Biochimica et

Biophysica Acta Molecular Basis of Disease 1740: 266–286.27. Froyland L, Madsen L, Vaagenes H, Totland GK, Auwerx J, et al. (1997)

Mitochondrion is the principal target for nutritional and pharmacologicalcontrol of triglyceride metabolism. J Lipid Res 38: 1851–1858.

28. Madsen L, Rustan AC, Vaagenes H, Berge K, Dyroy E, et al. (1999)

Eicosapentaenoic and docosahexaenoic acid affect mitochondrial and peroxi-somal fatty acid oxidation in relation to substrate preference. Lipids 34:

951–963.29. Kopecky J, Hodny Z, Rossmeisl M, Syrovy I, Kozak LP (1996) Reduction of

dietary obesity in aP2-Ucp transgenic mice: physiology and adipose tissuedistribution. Am J Physiol 270: E768–E775.

30. Puigserver P, Rhee J, Donovan J, Walkey CJ, Yoon JC, et al. (2003) Insulin-

regulated hepatic gluconeogenesis through FOXO1-PGC-1[alpha] interaction.Nature 423: 550–555.

31. Yoon JC, Puigserver P, Chen G, Donovan J, Wu Z, et al. (2001) Control ofhepatic gluconeogenesis through the transcriptional coactivator PGC-1. Nature

413: 131–138.

32. Herzig S, Long F, Jhala US, Hedrick S, Quinn R, et al. (2001) CREB regulateshepatic gluconeogenesis through the coactivator PGC-1. Nature 413: 179–183.

33. Ruzickova J, Rossmeisl M, Prazak T, Flachs P, Sponarova J, et al. (2005)Omega-3 PUFA of marine origin limit diet-induced obesity in mice by reducing

cellularity of adipose tissue (vol 39, pg 1177, 2004). Lipids 40: 115.

34. Itoh M, Suganami T, Satoh N, Tanimoto-Koyama K, Yuan X, et al. (2007)

Increased adiponectin secretion by highly purified eicosapentaenoic acid inrodent models of obesity and human obese subjects. Arteriosclerosis Thrombosis

and Vascular Biology 27: 1918–1925.

35. Kalupahana NS, Claycombe K, Newman SJ, Stewart T, Siriwardhana N, et al.

(2010) Eicosapentaenoic Acid Prevents and Reverses Insulin Resistance in High-Fat Diet-Induced Obese Mice via Modulation of Adipose Tissue Inflammation.

The Journal of Nutrition 140: 1915–1922.

36. Madsen L, Kristiansen K (2010) The importance of dietary modulation of

cAMP and insulin signaling in adipose tissue and the development of obesity.Foods for Health in the 21St Century: A Road Map for the Future 1190: 1–14.

37. Bluher M, Michael MD, Peroni OD, Ueki K, Carter N, et al. (2002) AdiposeTissue Selective Insulin Receptor Knockout Protects against Obesity and

Obesity-Related Glucose Intolerance. Developmental Cell 3: 25–38.

38. Murata Y, Tsuruzoe K, Kawashima J, Furukawa N, Kondo T, et al. (2007) IRS-

1 transgenic mice show increased epididymal fat mass and insulin resistance.Biochem Biophys Res Commun 364: 301–307.

39. Mikkelsen PB, Toubro S, Astrup A (2000) Effect of fat-reduced diets on 24-henergy expenditure: comparisons between animal protein, vegetable protein,

and carbohydrate. Am J Clin Nutr 72: 1135–1141.

40. Robinson SM, Jaccard C, Persaud C, Jackson AA, Jequier E, et al. (1990)

Protein turnover and thermogenesis in response to high-protein and high-carbohydrate feeding in men. Am J Clin Nutr 52: 72–80.

41. Cannon B, Nedergaard J (2004) Brown Adipose Tissue: Function andPhysiological Significance. Physiol Rev 84: 277–359.

42. Hoehn KL, Turner N, Swarbrick MM, Wilks D, Preston E, et al. (2010) Acuteor chronic upregulation of mitochondrial fatty acid oxidation has no net effect

on whole-body energy expenditure or adiposity. Cell Metab 11: 70–76.

43. Guerra C, Koza RA, Yamashita H, Walsh K, Kozak LP (1998) Emergence of

Brown Adipocytes in White Fat in Mice Is Under Genetic Control. Effects onBody Weight and Adiposity. J Clin Invest 102: 412–420.

44. Madsen L, Pedersen LM, Lillefosse HH, Fjaere E, Bronstad I, et al. (2010)UCP1 Induction during Recruitment of Brown Adipocytes in White Adipose

Tissue Is Dependent on Cyclooxygenase Activity. PLoS ONE 5.

45. Sato A, Kawano H, Notsu T, Ohta M, Nakakuki M, et al. (2010) Antiobesity

Effect of Eicosapentaenoic Acid in High-Fat/High-Sucrose Diet InducedObesity Importance of Hepatic Lipogenesis. Diabetes 59: 2495–2504.

46. Xu H, Barnes GT, Yang Q, Tan G, Yang D, et al. (2003) Chronic inflammationin fat plays a crucial role in the development of obesity-related insulin resistance.

J Clin Invest 112: 1821–1830.

47. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, et al. (2003)

Obesity is associated with macrophage accumulation in adipose tissue. J ClinInvest 112: 1796–1808.

48. Storlien LH, Kraegen EW, Chisholm DJ, Ford GL, Bruce DG, et al. (1987) Fish

Oil Prevents Insulin Resistance Induced by High-Fat Feeding in Rats. Science

237: 885–888.

49. Jucker BM, Cline GW, Barucci N, Shulman GI (1999) Differential effects of

safflower oil versus fish oil feeding on insulin-stimulated glycogen synthesis,glycolysis, and pyruvate dehydrogenase flux in skeletal muscle - A C-13 nuclear

magnetic resonance study. Diabetes 48: 134–140.

50. Neschen S, Morino K, Dong JY, Wang-Fischer Y, Cline GW, et al. (2007) N-3

fatty acids preserve insulin sensitivity in vivo in a peroxisonte proliferator-activated receptor-alpha-dependent manner. Diabetes 56: 1034–1041.

51. Shoelson SE, Lee J, Yuan M (2003) Inflammation and the IKKb/IkB/NF-kBaxis in obesity- and diet-induced insulin resistance. Int J Obes Relat Metab

Disord 27: S49–S52.

52. Kim JK, Kim YJ, Fillmore JJ, Chen Y, Moore I, et al. (2001) Prevention of fat-

induced insulin resistance by salicylate. J Clin Invest 108: 437–446.

53. Arkan MC, Hevener AL, Greten FR, Maeda S, Li ZW, et al. (2005) IKK-b linksinflammation to obesity-induced insulin resistance. Nat Med 11: 191–198.

54. Pandey SK, He HJ, Chesley A, Juhaszova M, Crow MT, et al. (2002)Wortmannin-Sensitive Pathway Is Required for Insulin-Stimulated Phosphor-

ylation of InhibitorkBa. Endocrinology 143: 375–385.

55. Liaset B, Madsen L, Hao Q, Criales G, Mellgren G, et al. (2009) Fish protein

hydrolysate elevates plasma bile acids and reduces visceral adipose tissue mass inrats. Biochimica et Biophysica Acta 1791: 254–262.

56. Arslan G, Brunborg LA, Frøyland L, Brun JG, Valen M, et al. (2002) Effects ofduodenal seal oil administration in patients with inflammatory bowel disease.

Lipids 37: 935–940.

57. Barbatelli G, Murano I, Madsen L, Hao Q, Jimenez M, et al. (2010) The

emergence of cold-induced brown adipocytes in mouse white fat depots isdetermined predominantly by white to brown adipocyte transdifferentiation.

American Journal of Physiology-Endocrinology and Metabolism 298:

E1244–E1253.

58. Madsen L, Petersen RK, Sørensen MB, Jørgensen C, Hallenborg P, et al. (2003)

Adipocyte differentiation of 3T3-L1 preadipocytes is dependent on lipoxygenaseactivity during the initial stages of the differentiation process. Biochem J 375:

539–549.



Figure S1.

A high fish oil diet does not impair insulin tolerance. Male C57BL/6 mice (n = 7) were fed

a low fat or isocaloric high fish oil diets with different carbohydrate and protein

contents ad libitum for 7 weeks. A: Insulin levels were measured in the fasted state. B:

Intraperitoneal insulin tolerance test was performed. Data are presented as means ± SEM.

Different small letters denote significant differences between the groups.

1

Carbohydrate source and insulin secretion modulate the obesity promoting

effect of fish oil

Qin Hao1, $

, Haldis Haukås Lillefosse1, 2, $

, Even Fjære1, 2, $

, Lene Secher Myrmel1, 2

, Lisa Kolden

Midtbø1, 2

, Ragnhild Jarlsby2, Tao Ma

1, Bingbing Jia

1, Rasmus Koefoed Petersen

1, Si Brask

Sonne1, André Chwalibog

3, Livar Frøyland

2, Bjørn Liaset

2, Karsten Kristiansen

1, * and Lise

Madsen1, 2, *

.

1Department of Biology, University of Copenhagen, Denmark.

2National Institute of Nutrition and

Seafood Research, Bergen, Norway. 3Department of Basic Animal and Veterinary Sciences,

University of Copenhagen, Denmark. $Contributed equally.

*Correspondence to: Karsten Kristiansen, Department of Biology, University of Copenhagen, Ole

Maaløes Vej 5, DK 2200 Copenhagen, Denmark. Fax +45 3522 2128; Phone: +45 3532 4443; E

mail: [email protected] or Lise Madsen, National Institute of Nutrition and Seafood Research, P.O.

Box 2029 Nordnes, N 5817 Bergen, Norway. Fax +47 5590 5299; Phone: +47 4147 6177; E mail:

[email protected]

Number of figures: 7

Number of tables: 2

Running title: Interaction between carbohydrates and fish oil

FOOTNOTES: This work was supported by the Danish Natural Science Research Council, the

Novo Nordisk Foundation, the Carlsberg Foundation and NIFES. Part of the work was carried out

as a part of the research program of the Danish Obesity Research Centre (DanORC). DanORC is

supported by the Danish Council for Strategic Research (Grant NO 2101 06 0005). The authors

have no conflicting interests.

2

ABSTRACT 1

Polyunsaturated n-3 fatty acids (n-3 PUFAs) are known to attenuate diet-induced obesity and 2

adipose tissue inflammation in rodents. In this study we aimed to investigate whether inclusion of 3

different carbohydrate sources modulated the effects of n-3 PUFAs. By feeding C57BL/6J mice 4

isocaloric high fat diets enriched with fish oil for 6 weeks, we show that increasing amounts of 5

sucrose in the diets dose-dependently increased energy efficiency and white adipose tissue (WAT) 6

mass. Mice receiving fructose had about 50% less WAT mass than mice fed a high fish oil diet 7

supplemented with either glucose or sucrose, indicating that the glucose moiety of sucrose was 8

responsible for the obesity promoting effect of sucrose. To investigate if the obesogenic effect of 9

sucrose and glucose was related to stimulation of insulin secretion, we combined fish oil with high 10

and low glycemic index (GI) starch. Mice receiving the high fish oil diet containing the low GI 11

starch had significantly less WAT than mice fed high GI starch. Moreover, inhibition of insulin 12

secretion by administration of nifedipine significantly reduced WAT mass in mice fed a high fish 13

oil diet in combination with sucrose. Our data show that the macronutrient composition of the diet 14

modulates the effects of fish oil. Fish oil combined with sucrose, glucose or high-glycemic index 15

starch promotes obesity and the reported anti-inflammatory actions of fish oil are abrogated. 16

Collectively, our data indicate that glycemic control of insulin secretion regulates the obesity 17

promoting effect of fish oil in combination with carbohydrates. 18

19

3

Introduction 20

A considerable number of studies has demonstrated the beneficial effects of n-3 polyunsaturated 21

fatty acids (PUFAs) on lipid metabolism and lipid-related disorders in humans (1-3) and increasing 22

the relative amount of n-3 PUFAs is presently recommended. Moreover, diets enriched with n-3 23

PUFAs have been demonstrated to reduce the development of insulin resistance (4-9) and to reduce 24

the development of diet-induced obesity in rodents (7, 10-15). We have shown that the 25

macronutrient composition of the diet determines the adipogenic potential of dietary corn oil in 26

mice (16). Moreover, we recently demonstrated that sucrose counteracts the anti-inflammatory 27

effect of fish oil in adipose tissue and increases obesity development in mice (17). Mice fed a fish 28

oil enriched diet in combination with sucrose had markedly higher feed efficiency and required less 29

than 50% of the calories to achieve the same weight gain as mice pair-fed a fish oil enriched diet in 30

combination with protein (17). 31

A major difference between proteins and sucrose is the ability to cause a rise in blood 32

glucose and stimulate insulin secretion. Insulin is a powerful anabolic hormone that stimulates 33

adipocyte differentiation and adipose tissue expansion (18), and activation of insulin signaling is 34

crucial for the development of obesity (19). Moreover, increased insulin signaling by transgenic 35

expression of insulin receptor substrate-1 (IRS-1) is sufficient to induce obesity (20). Thus, it is 36

possible that increased insulin signaling and glucose uptake in adipose tissue in the fed state in 37

sucrose fed mice may override the anti-inflammatory and anti-obesity effects of fish oil. 38

The glucose moiety of sucrose is responsible for the rise in blood insulin upon intake 39

of sucrose, as fructose unlike glucose, is unable to stimulate insulin secretion (21). This in part 40

relates to the very low levels of Slc2a5 (solute carrier family 2 (facilitated glucose transporter), 41

member 5, GLUT5) in pancreatic beta-cells (22). Furthermore, fructose does not stimulate the 42

release of gastric inhibitory peptide which stimulates insulin secretion indirectly (23, 24). Thus, the 43

http://www.informatics.jax.org/javawi2/servlet/WIFetch?page=markerDetail&key=49112

4

ability of sucrose to counteract the beneficial effects of fish oil seems to relate to a glucose-44

dependent stimulation of insulin secretion. The fructose moiety of sucrose may further modulate the 45

effect of fish oil on the development of obesity. Thus, the increased consumption of fructose over 46

the past decades has been linked to development of metabolic disorders (25), and fructose is 47

routinely used to induce glucose intolerance in rats (26). 48

As different types of starch differ in their ability to increase postprandial blood 49

glucose and insulin secretion, different types of starch may also modulate the effect of fish oil. The 50

glycemic index (GI) is a measure of the ability of different types of carbohydrate based foods to 51

raise blood glucose levels within 2 hours (27). The interest in low GI diets as a tool in weight 52

management is increasing. Although reviews and meta-analyzes conclude that such diets may be 53

effective, their effectiveness in terms of lasting weight reduction is still a matter of debate (28-32). 54

Different types of starches with different GI are known to induce different responses in plasma 55

glucose and insulin in rodents (33). However, it is unknown whether different types of starch 56

modulate the effects of fish oil enriched diets. Here we have performed systematic analyses to 57

investigate the influence of different carbohydrate sources on the reported anti-obesity and anti-58

inflammatory effect of n-3 PUFAs in mice. By using isocaloric high fat diets enriched with n-3 59

PUFAs, we show that the amount of sucrose dose-dependently increased energy efficiency and 60

adiposity. Moreover, we show that nutritional and pharmacological control of insulin secretion play 61

a pivotal role determining the obesogenic effect of high fish oil-high carbohydrate diets. 62

EXPERIMENTAL PROCEDURES 63

Animals and diets. Male C57BL/6JBomTac mice approximately 8 weeks of age were obtained from 64

Taconic Europe (Ejby, Denmark) and were divided into groups (n=6-10). The mice were kept at a 65

12 h light/dark cycle at thermoneutrality. After acclimation the animals were fed ad libitum or pair-66

fed experimental diets obtained from Ssniff (Ssniff Spezialdiäten GmbH, Soest, Germany) 67

http://www.medterms.com/script/main/art.asp?articlekey=17467

5

described in Table 1 and 2. The fish oil enriched diets contained 62 ± 3 g/kg n-3 fatty acids, thereof 68

27 ± 3 g/kg and 18 ± 3 g/kg EPA and DHA, respectively. All diets were supplemented with 3g/kg 69

L-cysteine, 10g/kg choline bitartrate, 10g/kg vitamin mix AIN 76 A, 45 g/kg mineral mix AIN 93 70

and 0.014 g/kg t-butylhydroquinone. As indicated Nifedipine (N7634, Sigma) was included in the 71

diet at a dose (1g/kg) earlier demonstrated to reduce plasma insulin levels in agouti mice (34). The 72

sulfonylurea Glybenclamide (G0639, Sigma) was used as an insulin secretagogue and was 73

administrated daily by intraperitoneal (i.p.) injection at a dose of 2 μg/g body weight..Control mice 74

received placebo. by daily i.p. injection. Mice in the fed state were euthanized by cardiac puncture 75

under anesthesia with isoflurane (Isoba-vet, Schering-Plough, Denmark) using the Univentor 400 76

Anaesthesia Unit (Univentor Limited, Sweden) and plasma was prepared from blood. Tissues were 77

dissected out, freeze-clamped and frozen at -80oC. All animal experiments were approved by 78

National Animal Health Authorities (Denmark and Norway). Adverse events were not observed. 79

Energy measurements and digestibility. Male C57BL/6JBomTac mice (n=5 for each group) were 80

fed experimental diets ad libitum for 4 days. Faeces, urine and feed residue were collected during a 81

24h period before respiration measurements. The nitrogen content in feed, faeces, and urine was 82

determined using the Tecator-Kjeltec system 1026 (Tecator AB, Höganäs, Sweden). The gross 83

energy content of feed and faeces was determined using an adiabatic bomb calorimeter (System 84

C700, IKA Analysentechnic GmbH, Heitersheim, Germany). 85

Glucose and insulin tolerance testing (GTT and ITT). GTT: mice were fasted 6 hours before i.p 86

injection of 2 g/kg glucose in saline. ITT: mice were fasted 4 hours before i.p. injection of 0.5 87

Unit/kg human recombinant insulin in saline. Blood was collected from the lateral tail vein at 88

indicated time points and blood glucose was measured with a Bayer Contour glucometer (Bayer 89

A/S, Denmark). 90

6

Plasma analysis. Insulin, glucose (16) and lipid metabolites (35) were measured in plasma as earlier 91

described. 92

Tissue lipid analysis. Total lipids were extracted from diets and liver with chloroform: methanol, 93

2:1 (v/v). Lipid classes were analyzed in liver samples using an automated High-Performance Thin 94

Layer Chromatography (HPTLC) system (Camaq, Switzerland) and separated on HPTLC plates 95

coated with silica gel 60 F (35). Fatty acid composition of total lipids from diets was analyzed on a 96

capillary gas chromatograph with flame ionization detector (Perkin Elmer, USA) (36). 97

Histology. Sections of adipose tissue were fixed, dehydrated, embedded in paraffin blocks, cut into 98

3 μm thick sections and stained with eosin and hematoxylin as described (37). Sections were 99

visually examined using an Olympus BX 51 binocular microscope (Olympus Corporation, Japan), 100

fitted with an Olympus DP50 3.0 camera. 101

RT-qPCR. Total RNA was purified from mouse tissue using Trizol (Invitrogen). Reverse 102

transcription (RT) was performed and cDNA was analyzed in duplicates by qPCR using the ABI 103

PRISM 7700 Sequence Detection System (Applied Biosystems) as earlier described (38). Primers 104

for RT-qPCR were designed using Primer Express 2.0 (Applied Biosystems) and are available on 105

request. 106

Statistics. Data represent mean ± SEM. ANOVA, post hoc pairwise comparison: Student t-test (RT-107

qPCR analysis) or Tukey HSD test (GTT, ITT and organ weights). Newman-Keuls test 108

(nonparametric due to non-homogenous variances, remaining data sets. A value of P<0.05 was 109

considered statistical significant. 110

111

RESULTS 112

Sucrose dose-dependently counteracts the obesity reducing effect of fish oil. 113

7

To investigate whether sucrose dose-dependently counteracts the obesity protective effect of fish 114

oil, we fed for 6 weeks C57BL/6J mice high fat diets containing 25 w% fat, thereof 62 ± 3 g/kg n-3 115

fatty acids, in combination with four diets of different carbohydrate:protein ratios (Table 1). All 116

diets contained 25.12 kJ/g. Body weight gain and white adipose tissue (WAT) mass increased in 117

parallel with the increase in the sucrose:protein ratio in the feed (Fig 1A and D). The weight of 118

interscapular brown adipose tissue (iBAT), liver and tibialis anterior muscle were not changed (not 119

shown). Total energy intake was not statistically significantly different in the four groups, and thus, 120

energy efficiency increased dose-dependently in response to the increased amount of dietary 121

sucrose (Fig 1B). Moreover, expressions of inflammatory markers, such as Serpine1 (serine (or 122

cysteine) peptidase inhibitor, clade E, member 1 or plasminogen activator inhibitor-1, PAI-1), Ccl2 123

(Ccl2 chemokine (C-C motif) ligand 2, MCP1), Cd68 (CD68 antigen) and Emr1 (EGF-like module 124

containing, mucin-like, hormone receptor-like sequence 1, F4/80) were increased in both eWAT 125

and iWAT in the obese mice, suggesting that sucrose attenuated the anti-inflammatory effect of fish 126

oil (Fig 1E). The expressions of Pparg (Peroxisome proliferator-activated receptor gamma), Srebf1 127

(sterol regulatory element binding transcription factor 1) and Fasn (Fatty acid synthase) did not 128

change significantly (Fig 1E). Mice receiving the highest amount of sucrose also lost significantly 129

less body weight during a 24h fast (Fig 1C) than mice receiving the lowest amount of sucrose, 130

indicative of a lower metabolic rate in high-sucrose fed mice. 131

Energy may be lost as heat by the uncoupling activity of UCP1 (uncoupling protein 1 132

(mitochondrial, proton carrier)) expressed in brown adipocytes, and increased expression of Ucp1 133

can prevent diet-induced obesity. Therefore, we measured Ucp1 expression in both white and 134

brown adipose tissue depots. In iBAT Ucp1 expression was similar in all groups, but the expression 135

of Ucp1 was significantly reduced in iWAT in mice fed the high amount of sucrose (Fig 1C and E). 136

8

This indicates that adipocytes in iWAT from mice receiving a low amount of sucrose and a high 137

amount of protein had a more brownish phenotype. 138

We have previously demonstrated that a high fat diet in combination with a high 139

protein:sucrose ratio translated into a high glucagon:insulin ratio, increased cAMP-signaling and 140

expression of Ppargc1a (peroxisome proliferative activated receptor, gamma, coactivator 1 alpha) 141

leading to increased gluconeogenesis and amino acid degradation in liver (16, 17). Here we show 142

that sucrose dose-dependently reduced expression of Ppargc1a, Pck1 (phosphoenolpyruvate 143

carboxykinase 1, cytosolic) and Agxt (alanine-glyoxylate aminotransferase) in the liver (Fig 1F). 144

Expression of the lipogenic gene, Fasn, was increased whereas expressions of enzymes involved in 145

fatty acid oxidation were unchanged when the sucrose:protein ratio was increased (Fig 1F). 146

147

The glucose moiety of sucrose is responsible for the obesity promoting effect of sucrose. 148

To investigate whether the obesity promoting effect of sucrose fed in combination with fish oil was 149

depended on the glucose or fructose moiety of sucrose, we prepared diets where fish oil was 150

combined with sucrose, glucose or fructose (Table 2). C57BL/6J mice were fed these diets ad-151

libitum. After 7 weeks mice receiving fish oil in combination with fructose had gained less weight 152

than mice receiving fish oil in combination with sucrose or glucose (Fig 2A). Energy intake was 153

slightly, but not statistically significantly lower in mice fed the fructose supplemented diets and 154

hence, energy efficiency was significantly reduced (Fig 2B). Calculation of digestibility 155

demonstrated a minor, but not significantly reduced digestibility of protein and fat in fructose fed 156

mice (Fig 2C). Of note, the mice receiving fructose had about 50% less white adipose tissue mass, 157

iWAT, pWAT and eWAT, than mice receiving sucrose or glucose, combined with a tendency 158

towards a slight decrease in the weight of the tibialis anterior muscle (Fig 2D). 159

9

We have previously provided evidence that the different obesogenic effect of corn oil 160

fed in combination with either protein or sucrose is related to the effect of the macronutrient 161

composition on hormonal status (16). As sucrose and glucose, unlike fructose, stimulate insulin 162

secretion, we measured plasma levels of glucose and insulin in the fed state. As expected, plasma 163

glucose and insulin were lower in fructose fed mice (Fig 2E). As in mice fed a high fat diet in 164

combination with proteins, expressions of Ppargc1a and Pck1 in the liver were increased (Fig 2F). 165

However, unlike mice fed a high protein diet, expression of genes involved in amino acid 166

degradation was not increased in fructose fed mice (Fig 2F). Plasma levels of 2-hydroxybutyrate, a 167

marker for fatty acid β-oxidation, were similar in all groups, but the expression of genes involved in 168

hepatic fatty acid oxidation was reduced in livers from fructose fed mice (Fig 2F). Hepatic 169

expression of lipogenic genes was higher in fructose fed mice (Fig 2F). Importantly, however, 170

excess lipid accumulation was not seen in liver or tibialis anterior muscle (Fig 3). The finding that 171

Ucp1 expression was strongly induced in iWAT, but not iBAT in fructose fed mice, indicates that 172

energy, as observed in protein fed mice, may be dissipated in the form of heat. Indeed, a higher 173

expression of markers for brown adipocytes such as Ppargc1a and Cox2 (cytochrome c oxidase 174

subunit II) suggested a higher number of brown-like adipocytes in iWAT (Fig 3C). 175

Fructose feeding is frequently used to induce glucose intolerance in rats (26, 39) and 176

pro-inflammatory cytokines such as CCL2 produced by both adipocytes and infiltrating 177

macrophages are causally linked to the development of glucose intolerance (40, 41). As expression 178

of inflammatory markers was significantly higher in mice fed fish oil in combination with glucose 179

or sucrose than fructose (Fig 3C), a glucose tolerance test was performed in mice fed the sucrose, 180

glucose and fructose-based diets. Of note, blood glucose levels reached higher concentrations in 181

glucose compared to fructose fed mice and the area under the curve (AUC) was significantly higher 182

in glucose than both sucrose and fructose fed mice (Fig 3D). 183

10

184

High glycemic index starch increases the adipogenic potential of fish oil. 185

To further investigate whether other types of carbohydrates, such as starch with different capacity to 186

stimulate insulin secretion are able to modulate the adipogenic potential of fish oil, we prepared 187

isocaloric diets (Table 2) where fish oil was combined with low or high GI starch (100% 188

amylopectin vs 60% amylose/40% amylopectin) demonstrated to induce the expected differences in 189

postprandial blood glucose levels in a meal tolerance test when combined with a high fat diet (33, 190

42). After 7 weeks of feeding, body weights were similar in mice receiving fish oil in combination 191

with either high- or low GI starch (Fig 4A). Energy intake was also similar, and no statistical 192

significant difference in energy efficiency was found (Fig 4A). However, mice fed fish oil in 193

combination with the low GI starch had significantly lower levels of insulin and less white adipose 194

tissue (Fig 4 B and C). Furthermore, expressions of inflammatory markers tended to be higher in 195

adipose tissues from mice fed high GI starch (Fig 4E). The reduced adiposity observed in mice 196

receiving fish oil in combination with the low GI starch was not due to decreased digestibility, as 197

digestibility of the diet containing the low GI starch was slightly higher than digestibility of the diet 198

containing the high GI starch (Fig 4D). Expressions of genes involved in fatty acid oxidation, 199

gluconeogenesis and amino acid degradation in the liver were similar in the two groups of mice, but 200

expressions of the lipogenic genes Scd1 (stearoyl-Coenzyme A desaturase 1) and Fasn were 201

reduced in mice fed low GI starch (Fig 4F). Of note, expression of Ucp1 was similar in iBAT, but 202

significantly higher in iWAT in mice fed the low GI starch (Fig 4E and F). Altogether, results from 203

the experiments combining fish oil with glucose or fructose as well as high or low GI starch 204

suggested that stimulation of insulin secretion increases the obesogenic effect of a diet enriched 205

with dietary fish oil. 206

207

11

The obesity promoting effect of sucrose in combination with fish oil is associated to increased 208

insulin secretion. 209

To investigate if increased insulin secretion is able to increase the obesogenic effect of fish oil when 210

the carbohydrate load is low, we fed mice a diet enriched with fish oil in combination with proteins. 211

To increase insulin secretion, half of the mice were injected daily with the sulfonylurea class drug 212

glybenclamide (43, 44), whereas the second half were injected with placebo. Energy intake, body 213

weight gain (Fig 5A) and energy efficiency (not shown) were not affected by glybenclamide, but 214

the amount of white adipose tissue mass tended to increase although not statistically significantly 215

(Fig 5B). However, expression of Pparg was significantly increased in both eWAT and iWAT (Fig 216

5C). As insulin levels in the fed state only tended to increase, a glucose tolerance test was 217

performed to validate whether the dose of glybenclamide used was sufficient to increase insulin 218

levels. As expected, glucose tolerance was improved by daily injections of glybenclamide (Fig 5D). 219

This was not accompanied by reduced expressions of markers for adipose tissue inflammation (Fig 220

5C). In accordance with unchanged energy efficiency, Ucp1 expression was not significantly 221

changed. However, glybenclamide treatment reduced expression of Pck1 and genes involved in 222

amino acid degradation and ureagenesis (Fig 5E). Together, these results demonstrate that increased 223

insulin secretion is unable to significantly increase the obesogenic effect of fish oil when the 224

carbohydrate load is low. It is thus likely that a certain threshold level of carbohydrates is required 225

in order for insulin to ellicit the obesogenic potential of fish oil. 226

If insulin promotes the obesogenic potential of fish oil in presence of carbohydrates, 227

inhibition of insulin secretion should attenuate the adipogenic effect of sucrose in combination with 228

fish oil. Since insulin secretion can be reduced by treatment with nifedipine (34) we supplemented a 229

diet enriched with fish oil and sucrose with nifedipine. Inclusion of nifedipine did not influence 230

body weight gain (Fig 6A), or feed intake (not shown). Consequently, energy efficiency was not 231

12

different (not shown). As expected, inclusion of nifedipine in the sucrose enriched high-fish oil diet 232

resulted in lower levels of plasma insulin in the fed state (Fig 6B). Importantly, the mice receiving 233

the nifedipine-supplemented diet had lower adipose tissue mass and adipocytes had a normal 234

appearance (Fig 6C). When nifedipine was added to a standard low fat diet, no effect on adipose 235

tissue mass was observed (not shown), and expression of inflammatory markers in white adipose 236

tissue was, with the exception of Serpine1 in iWAT, not reduced (Fig 6D). In line with unchanged 237

energy efficiency, we did not detect any increased expression of Ucp1 in iWAT in nifedipine 238

treated mice (Fig 6D). Moreover, nifedipine did not increase hepatic expression of Pck1 or genes 239

involved in lipogenesis or amino acid degradation (Fig 7A). Thus, inhibition of insulin secretion 240

was able to partly attenuate the obesogenic effect of sucrose in combination with fish oil, but did 241

not reduce energy efficiency. Furthemore, nifedipine did not improve the reduced glucose tolerance 242

observed after feeding fish oil in combination with sucrose, but a insulin tolerance test indicated 243

that insulin sensitivity was improved (Fig 7B). Together, our data indicate that an inhibition of 244

insulin secretion attenuates the adipogenic effect of sucrose in combination with fish oil. However, 245

inhibition of insulin secretion is insufficient to reduce energy efficiency. 246

247

DISCUSSION 248

Due to the well described beneficial effects of n-3 PUFAs on lipid metabolism and lipid-related 249

disorders in humans, increasing the relative amount of n-3 PUFAs in diet or as supplement is 250

presently recommended by Health Authorities. The potential use of n-3 PUFAs in weight control in 251

humans is not fully explored, but the ability of these fatty acids to reduce the development of diet-252

induced obesity rodents is well described (7, 10-15). In this study we demonstrate that the 253

composition of the diet determines the adipogenic potential of fish oil. In particular, we demonstrate 254

that a high dose of fish oil promoted obesity when combined with sucrose, glucose or high GI 255

13

starch. Thus, many if not all reported beneficial effects of fish oil intake might be diminishes or 256

completely abrogated by a simultaneous intake of high GI carbohydrates. 257

Insulin is the primary anabolic hormone promoting energy storage in the fed state and 258

is an important driver for adipocyte differentiation (18). Of note, mice lacking insulin receptors in 259

adipose tissue (FIRKO mice), are protected against obesity and remain glucose tolerant on a high-260

fat diet (19). Further, white adipose tissue mass is dramatically reduced in Irs1-/-

Irs2-/-

double-261

knockout mice (45). In contrast, mice lacking the insulin receptor in muscle (MIRKO mice) have 262

increased glucose utilization in WAT (46), and exhibit a more than 50% increase in fat mass and 263

adipocyte number (47). Similarly, mice overexpressing Slc2a4 (solute carrier family 2 (facilitated 264

glucose transporter), member 4, GLUT4) (48) or Irs1 (20) in adipose tissue are obese. Finally, 265

treating normal rats with insulin via osmotic minipumps increases Slc2a4 expression, glucose 266

utilization and de novo fatty acid synthesis in adipose tissue, which are accompanied by weight gain 267

and increased adipose tissue mass (49-51). The importance of insulin secretion and insulin levels in 268

circulation is further supported by the finding that daily injections of insulin increases weight gain 269

in transgenic mice expressing agouti in adipose tissue (52). Of note, unlike the traditional agouti 270

mice, mice with transgenic expression of the agouti gene driven by the aP2 promoter do not get 271

obese unless triggered by insulin (52). Transgenic mice treated with insulin gained significantly 272

more weigh than their wild type littermates, and the authors suggested that the daily insulin 273

treatment mimicked the hyperinsulinemia normally observed in mice expressing the agouti gene 274

ubiquitously (52). Moreover, weight gain is a well recognized side effect of type 2 diabetic drugs 275

that increase insulin sensitivity (53, 54). Together these observations suggest that hyperinsulinemia 276

is a contributing factor to development of obesity and reducing hyperinsulinemia would thus 277

possible counteract obesity development. In keeping with this view, inhibition of insulin secretion 278

14

by administration of nifedipine attenuated the adipogenic effect of fish oil in combination with 279

sucrose. 280

Increasing insulin secretion alone, however, is not sufficient to promote obesity 281

development, as mice receiving glybenclamide in combination with proteins and fish oil did not 282

become obese. Furthermore, increased insulin secretion by glybenclamide in the presence of low 283

levels of carbohydrates in the feed was insufficient to increase the adipogenic potential of fish oil. It 284

has been demonstrated that a high-fat diet is unable to increase adipose tissue mass in the absence 285

of carbohydrates (55, 56). 286

In our study nifedipine did not reduce adipose tissue mass in low fat fed mice (not 287

shown). However, as emphasized above, inclusion of nifedipine attenuated hyperinsulinemia and 288

obesity induced by fish oil in combination with sucrose. These findings are in line with the study by 289

Kim et al (34) demonstrating that nifedipine attenuated agouti-induced hyperinsulinemia and 290

obesity. Hyperinsulinemia is an early event also in obese Zucker rats, and attenuation of 291

hyperinsulinemia using diazoxide reduces obesity development in obese- but not lean Zucker rats 292

independently of feed intake (57). Together our results suggest that the increased insulin secretion 293

contribute to the obesity promoting effect of fish oil when combined with sucrose. 294

The obesity promoting effect of increased insulin secretion is further supported by our 295

finding that glucose, but not fructose is the obesity promoting moiety of sucrose. Unlike sucrose 296

and glucose, fructose does not stimulate insulin secretion from pancreatic beta-cells (26) and mice 297

receiving fish oil in combination with fructose had less adipose tissue mass than mice receiving fish 298

oil in combination with glucose or sucrose. Mice receiving fructose in combination with fish oil 299

were also less glucose intolerant than mice fed fish oil in combination with sucrose or glucose. 300

Fructose is commonly used to induce glucose intolerance in rats and some mice strains also develop 301

15

metabolic syndrome in response to high fructose feeding (58). However, in C57BL/6 mice neither 302

hyperinsulinemia nor hyperglycemia developed in response to high fructose feeding (58). 303

The obesity promoting effect of fish oil can also be affected by different type of starch 304

as mice receiving fish oil in combination with a high GI starch had higher adipose tissue mass than 305

mice receiving fish oil in combination with low GI starch. These results may also relate to the 306

effect of the different type of starch on postprandial glucose levels and insulin secretion (33, 42, 307

59). Low GI diets are becoming popular in weight management also in humans although their 308

effectiveness in terms of lasting weight reduction is not commonly accepted (28-32). The lack of 309

acceptance may, however, partly be due to insufficient power of early studies as a meta-analysis has 310

indicated that diets in which there was a reduction in the glycemic index produced moderately more 311

weight loss than control low fat diets in humans (32). However, a recent large European study 312

demonstrated that intake of low glycemic index carbohydrates combined with a modest increase in 313

protein content improved maintenance of weight loss (60). 314

Obviously, increased adipose tissue mass is related to energy-intake. However, as 315

demonstrated here, the macronutrient composition can influence energy efficiency, and mice 316

consuming the same amount of calories end up with quite different amounts of adipose tissue. Of 317

note, increasing the amount of sucrose from 13 to 43% led to an approximately 5-fold higher energy 318

efficiency. Circulating insulin levels may also indirectly influence energy efficiency. Hepatic PGC-319

1α is a central target of the insulin/glucagon axis regulating activation of the entire gluconeogenesis 320

cascade in liver (61-63) and a dose-dependent decrease in Ppargc1a and Pck1 expression was 321

observed in response to increasing intake of dietary sucrose. This might be of importance regarding 322

energy efficiency as gluconeogenesis requires ATP and activation of gluconeogenesis reduces feed 323

efficiency as protein is converted to glucose at a cost of 16-20 kJ/g protein (64). Moreover, 324

increased catabolism of amino acids requires ATP for disposal of nitrogen as urea at an energy-cost 325

16

of ca 5.4 kJ/g urea. Consistent with this notion, we have observed that fish oil combined with a high 326

level of protein and low levels of sucrose induces hepatic expression of Ppargc1a that was 327

accompanied by an induction of mRNAs encoding enzymes involved in catabolism of glucogenic 328

amino. Thus, increased energy cost by increased gluconeogenesis and ureagenesis probably 329

contribute to the reduced feed-efficiency observed by a high protein diet. 330

In mice fed increasing amounts of protein at the expense of sucrose, we observed a 331

dose-dependent increase in Ucp1 expression in iWAT. Others and we have recently demonstrated 332

that cyclooxygenase-dependent induction of UCP1 expression in WAT counteracts diet-induced 333

obesity (65, 66). The importance of UCP1 expressing brown-like adipocyte in WAT for regulation 334

of energy expenditure is further underscored by aP2-UCP1 transgenic mice in which endogenous 335

Ucp1 expression and respiration were reduced in iBAT, whereas UCP1 expression, respiration and 336

total oxidative capacity were induced in WAT, and this was sufficient to account for the observed 337

changes of total energy balance (67). Thus, UCP1-dependent uncoupled respiration in iWAT in 338

combination with increased energy cost from gluconeogenesis may account for the reduced energy-339

efficiency observed when sucrose levels in the diet were low. 340

If the background diet determines the adipogenic potential of fish oil also in humans, 341

this is of great concern, as the intake of refined sugars from sources such as soft-drinks has 342

increased dramatically during the last several decades (68). Moreover, n-3 supplements are often 343

taken in combination with morning meals containing high-glycemic index carbohydrates such as 344

cereal, bread and orange juice. Thus, comprehensive studies of the interaction between dietary 345

macronutrients and fish oil in humans seem warranted. 346

347

348

349

17

350

FIGURE LEGENDS 351

352

FIGURE 1. Macronutrient composition determines the obesogenic effect of fish oil. Male 353

C57BL/6 mice (n=8) were pair-fed isocaloric high fat diets with different carbohydrate/protein 354

ratios for 6 weeks. (A) Body weight development in pair-fed mice is shown as relative increase. (B) 355

Energy efficiency was calculated as body weight gain divided by feed energy intake. (C) Weight 356

reduction during 24h fasting was measured and Ucp1 (Uncoupling protein-1) and Dio2 (deiodinase, 357

iodothyronine, type II) expressions were measured by RT-qPCR. (D-F) Adipose tissue and liver 358

were dissected out and the weights were recorded. Expressions of Pparg (peroxisome proliferator 359

activated receptor gamma), Srebf1 (sterol regulatory element binding transcription factor 1), Fasn 360

(fatty acid synthase), Ucp1 (Uncoupling protein-1), Serpine1 (Plasminogen activator inhibitor-1), 361

Ccl2 (chemokine (C-C motif) ligand 2), Cd68 (CD68 antigen) and Emr1 (EGF-like module 362

containing, mucin-like, hormone receptor-like sequence 1 or F4/80) were measured in epididymal 363

and inguinal white adipose tissue and expressions of Acox1 (acyl-CoA oxidase 1), Cpt1a (carnitine 364

palmitoyltransferase 1a), Fasn (fatty acid synthase), Ppargc1a (peroxisome proliferator-activated 365

receptor gamma, coactivator 1 alpha), Pck1 (phosphoenolpyruvate carboxykinase 1, cytosolic) and 366

Agxt (alanine-glyoxylate aminotransferase) were measured in liver using RT-qPCR and normalized 367

to Tbp (TATA-box binding protein). Data are presented as mean ± SEM. Different small letters 368

denote significant differences (P<0.05). Statistics were performed separately for each individual 369

tissue. 370

371

FIGURE 2. The glucose moiety of sucrose is responsible for the obesity promoting effect of 372

sucrose in combination with fish oil. Male C57BL/6 mice (n=6) were fed isocaloric high fat diets 373

18

with different carbohydrate sources ad-libitum for 8 weeks. (A) Body weight development is shown 374

as relative increase. (B) Energy efficiency was calculated as body weight gain divided by feed 375

intake. (C) Digestibility of energy, nitrogen, organic material and fat were calculated in a separate 376

set of mice (n=5). (D-F) Blood was collected, organs were dissected out and the weights were 377

recorded. Hormones and lipid parameters were measured in plasma and expressions of Acox1 (acyl-378

CoA oxidase 1), Cpt1a (carnitine palmitoyltransferase 1a), Cpt2 (carnitine palmitoyltransferase 2), 379

Hmgcs2 (3-hydroxy-3-methylglutaryl-Coenzyme A synthase 2), Srebf1 (sterol regulatory element 380

binding transcription factor 1), Scd1 (stearoyl-Coenzyme A desaturase 1), Acaca (acetyl-Coenzyme 381

A carboxylase alpha), Fasn (fatty acid synthase), Crem (cAMP responsive element modulator), 382

Pde4c (phosphodiesterase 4C, cAMP specific), Ppargc1a (peroxisome proliferator-activated 383

receptor gamma, coactivator 1 alpha), Pck1 (phosphoenolpyruvate carboxykinase 1, cytosolic), 384

Gpt1a (glutamic pyruvic transaminase 1a), Got1 (glutamate oxaloacetate transaminase 1, soluble), 385

Agxt (alanine-glyoxylate aminotransferase) and Cps1 (carbamoyl-phosphate synthetase 1) were 386

measured in liver using RT-qPCR and normalized to Tbp (TATA-box binding protein). Data are 387

presented as mean ± SEM. Different small letters denote significant differences (P<0.05). Statistics 388

were performed separately for each individual tissue. 389

390

FIGURE 3. A diet enriched in fructose increases expression of brown adipocyte marker genes 391

in white adipose tissue. Male C57BL/6 mice (n=6) were fed isocaloric high fat diets with different 392

carbohydrate sources ad libitum for 8 weeks. (A) Total lipids were extracted from liver and tibialis 393

anterior muscle and separated using HPTLC. (B) Expression of Ucp1 (Uncoupling protein-1) was 394

measured in interscapular brown adipose tissue and (C) expressions of Pparg (peroxisome 395

proliferator-activated receptor gamma), Ppargc1a (peroxisome proliferator-activated receptor 396

gamma, coactivator 1 alpha), Ucp1 (Uncoupling protein-1), cyt COXII, (cytochrome c oxidase, 397

19

subunit II), Serpine1 (Plasminogen activator inhibitor-1), Ccl2 (chemokine (C-C motif) ligand 2), 398

Cd68 (CD68 antigen) and Emr1 (EGF-like module containing, mucin-like, hormone receptor-like 399

sequence 1 or F4/80) were measured in epididymal and inguinal white adipose tissue by RT-qPCR 400

and normalized to Tbp. (D) In a separate set of mice (n=9), an intraperitoneal glucose tolerance and 401

insulin tolerance test were performed. Data are presented as mean ± SEM. Different small letters 402

denote significant differences (P<0.05). Statistics were performed separately for each individual 403

tissue. 404

405

FIGURE 4. High-glycemic index starch increases the adipogenic potential of fish oil. Male 406

C57BL/6 mice (n=6) were fed isocaloric high fat diets with different carbohydrate sources ad 407

libitum for 8 weeks. (A) Energy efficiency was calculated as body weight gain divided by feed 408

energy intake. (B) Fed state glucose and insulin were measured in plasma. (C) Adipose tissue 409

weights were recorded. (D) Digestibility of energy, nitrogen, organic material and fat were 410

calculated in a separate set of mice (n=5). (E-F). Expressions of Pparg (peroxisome proliferator-411

activated receptor gamma), Srebf1 (sterol regulatory element binding transcription factor 1), Fasn 412

(fatty acid synthase), Ucp1 (Uncoupling protein-1), Serpine1 (Plasminogen activator inhibitor-1), 413

Ccl2 (chemokine (C-C motif) ligand 2), Cd68 (CD68 antigen) and Emr1 (EGF-like module 414

containing, mucin-like, hormone receptor-like sequence 1 or F4/80) were measured in epididymal 415

and inguinal white adipose tissue, Ucp1 (Uncoupling protein-1) and cyt COXII, (cytochrome c 416

oxidase, subunit II) were measured in interscapular brown adipose tissue and expressions of Acox1 417

(acyl-CoA oxidase 1), Hmgcs2 (3-hydroxy-3-methylglutaryl-Coenzyme A synthase 2), Srebf1 418

(sterol regulatory element binding transcription factor 1), Scd1 (stearoyl-Coenzyme A desaturase 1), 419

Acaca (acetyl-Coenzyme A carboxylase alpha), Fasn (fatty acid synthase), Ppargc1a (peroxisome 420

proliferator-activated receptor gamma, coactivator 1 alpha), Pck1 (phosphoenolpyruvate 421

20

carboxykinase 1, cytosolic), Got1 (glutamate oxaloacetate transaminase 1, soluble), Agxt (alanine-422

glyoxylate aminotransferase) and Cps1 (carbamoyl-phosphate synthetase 1) were measured in liver 423

using RT-qPCR and normalized to Tbp (TATA-box binding protein). Data are presented as mean ± 424

SEM. Different small letters denote significant differences (P<0.05). Statistics were performed 425

separately for each individual tissue. 426

427

FIGURE 5. Increased insulin secretion by glybenclamide is not sufficient to increase the 428

obesogenic effect of fish oil when the carbohydrate intake is low. Male C57BL/6 mice (n=6) 429

were fed a standard low fat diet or high fat-low carbohydrate diet for 4 weeks. The mice fed the 430

high fat diet were injected daily with glybenclamid (2 µg/g body weight) or placebo. (A) Body 431

weight development in ad libitum fed mice is shown as relative increase. (B) The weights of 432

epididymal and inguinal white adipose tissue were recorded and sections were stained with 433

hematoxylin and eosin. (C) Expressions of Pparg (peroxisome proliferator-activated receptor 434

gamma), Srebf1 (sterol regulatory element binding transcription factor 1), Fasn (fatty acid 435

synthase), Ucp1 (Uncoupling protein-1), Serpine1 (Plasminogen activator inhibitor-1), Ccl2 436

(chemokine (C-C motif) ligand 2), Cd68 (CD68 antigen) and Emr1 (EGF-like module containing, 437

mucin-like, hormone receptor-like sequence 1 or F4/80) were measured in epididymal and inguinal 438

white adipose tissue by RT-qPCR and normalized to Tbp. (D) In a separate set of mice (n=6), an 439

intraperitoneal glucose tolerance test was performed. (E) Expressions of Acox1 (acyl-CoA oxidase 440

1), Cpt1a (carnitine palmitoyltransferase 1a), Cpt2 (carnitine palmitoyltransferase 2), Hmgcs2 (3-441

hydroxy-3-methylglutaryl-Coenzyme A synthase 2), Srebf1 (sterol regulatory element binding 442

transcription factor 1), Scd1 (stearoyl-Coenzyme A desaturase 1), Acaca (acetyl-Coenzyme A 443

carboxylase alpha), Fasn (fatty acid synthase), Crem (cAMP responsive element modulator), Pde4c 444

(phosphodiesterase 4C, cAMP specific), Ppargc1a (peroxisome proliferator-activated receptor 445

21

gamma, coactivator 1 alpha), Pck1 (phosphoenolpyruvate carboxykinase 1, cytosolic), Gpt1a 446

(glutamic pyruvic transaminase 1a), Got1 (glutamate oxaloacetate transaminase 1, soluble), Agxt 447

(alanine-glyoxylate aminotransferase) and Cps1 (carbamoyl-phosphate synthetase 1) were measured 448

in liver using RT-qPCR and normalized to Tbp (TATA-box binding protein). Data are presented as 449

mean ± SEM. Different small letters denote significant differences (P<0.05). Statistics were 450

performed separately for each individual tissue. 451

452

FIGURE 6. The obesity promoting effect of sucrose in combination with fish oil is partly due 453

to increased insulin secretion. Male C57BL/6 mice (n=6) were fed a standard low fat diet or high 454

fat-high carbohydrate with or without nifedipine supplementation (1 g/kg) diet for 4 weeks. (A) 455

Body weight development in ad libitum fed mice is shown as relative increase. (B) Fed state levels 456

of hormones and lipid parameters were measured in plasma. (C) The weights of epididymal and 457

inguinal white adipose tissue were recorded and sections were stained with hematoxylin and eosin. 458

(D) Expressions of Pparg (peroxisome proliferator-activated receptor gamma), Srebf1 (sterol 459

regulatory element binding transcription factor 1), Fasn (fatty acid synthase), Ucp1 (Uncoupling 460

protein-1), Serpine1 (Plasminogen activator inhibitor-1), Ccl2 (chemokine (C-C motif) ligand 2), 461

Cd68 (CD68 antigen) and Emr1 (EGF-like module containing, mucin-like, hormone receptor-like 462

sequence 1 or F4/80) were measured in epididymal and inguinal white adipose tissue by RT-qPCR 463

and normalized to Tbp (TATA-box binding protein). Data are presented as mean ± SEM. Different 464

small letters denote significant differences (P<0.05). Statistics were performed separately for each 465

individual tissue. 466

467

FIGURE 7. Nifedipine treatment attenuates reduced insulin, but not glucose tolerance. Male 468

C57BL/6 mice (n=9) were fed a standard low fat diet or high fat-high carbohydrate with or without 469

22

nifedipine supplementation (1 g/kg) diet for 4 weeks. (A) Expressions of Acox1 (acyl-CoA oxidase 470

1), Cpt1a (carnitine palmitoyltransferase 1a), Cpt2 (carnitine palmitoyltransferase 2), Hmgcs2 (3-471

hydroxy-3-methylglutaryl-Coenzyme A synthase 2), Srebf1 (sterol regulatory element binding 472

transcription factor 1), Scd1 (stearoyl-Coenzyme A desaturase 1), Acaca (acetyl-Coenzyme A 473

carboxylase alpha), Fasn (fatty acid synthase), Crem (cAMP responsive element modulator), Pde4c 474

(phosphodiesterase 4C, cAMP specific), Ppargc1a (peroxisome proliferator-activated receptor 475

gamma, coactivator 1 alpha), Pck1 (phosphoenolpyruvate carboxykinase 1, cytosolic), Gpt1a 476

(glutamic pyruvic transaminase 1a), Got1 (glutamate oxaloacetate transaminase 1, soluble), Agxt 477

(alanine-glyoxylate aminotransferase) and Cps1 (carbamoyl-phosphate synthetase 1) were measured 478

in liver using RT-qPCR and normalized to Tbp (TATA-box binding protein). (B) Intraperitoneal 479

glucose tolerance test and insulin glucose tests were performed. Data are presented as mean ± SEM. 480

Different small letters denote significant differences (P<0.05). Statistics were performed separately 481

for each individual tissue. 482

483

23

TABLE 1. 484

Macronutrient composition in the diets 485

Data represent weight % (g /100g) 486

High fish oil

Sucrose

13% 23% 33% 43%

Protein 500 400 300 200

Casein 500 400 300 200

L-Cysteine 3 3 3 3

Carbohydrate 189,5 289,5 389,5 489,5

Corn starch 9,5 9,5 9,5 9,5

Cellulose 50 50 50 50

Sucrose 130 230 330 430

Fat 250 250 250 250

Soybean oil 70 70 70 70

Fish oil 180 180 180 180

Energy (kJ/g) 25.12 25.12 25.12 25.13

487

488

24

489

TABLE 2. 490

Macronutrient composition in the diets 491

Data represent weight % (g /100g) 492

493

High fish oil

Low

energy Sucrose Fructose Glucose Low GI High GI

Protein 200 200 200 200 200 200

Casein 200 200 200 200 200 200

L-Cysteine 3 3 3 3 3 3

Carbohydrate 669,5 489,5 489,5 489,5 489,5 489,5

Corn starch 529,5 9,5 9,5 9,5 9,5 9,5

Cellulose 50 50 50 50 50 50

Amylose 204

Amylopectin 136 340

Sucrose 90 430 90 90 90 90

Fructose 340

Glucose 340

Fat 70 250 250 250 250 250

Soybean oil 70 70 70 70 70 70

Fish oil 180 180 180 180 180

25

Energy (kJ/g) 17.16 25.12 25.12 25.12 25.12 25.12

494

495

496

497

Acknowledgments: 498

We thank Alison Keenan for the helpful comments and suggestions during preparation of the 499

manuscript. Moreover, we thank Åse Heltveit and Jan Idar Hjelle at NIFES for their excellent 500

assistance with animal care and lipid analyses. 501

502

Authors´ contribution: 503

B. L., K. K. and L. M. designed research. Q. H., H. H. L., E. F., L. S. M., L. K. M., R. J., T. M., B. 504

J., R. K. P., S. B. S., A. C., L. F., B. L. and L. M. conducted research and analyzed data. K. K. and 505

L. M. had primary responsibility for the final content. All authors read and approved the final 506

manuscript. 507

508

509

Literature Cited 510

511

1. Marik, V. & Varon, J. (2009) Omega-3 dietary supplements and the risk of cardiovascular 512

events: a systematic review. Clin Cardiol 32: 365-372. 513

2. Harris, W. S. (1997) n-3 Fatty acids and serum lipoproteins: Human studies. Am J Clin Nutr 514

65: S1645-S1654. 515

3. Ruxton, C. H. S., Reed, S. C., Simpson, M. J. A. & Millington, K. J. (2004) The health 516

benefits of omega-3 polyunsaturated fatty acids: a review of the evidence. Journal of Human 517

Nutrition and Dietetics 17: 449-459. 518

4. Storlien, L. H., Kraegen, E. W., Chisholm, D. J., Ford, G. L., Bruce, D. G. & Pascoe, W. S. 519

(1987) Fish Oil Prevents Insulin Resistance Induced by High-Fat Feeding in Rats. Science 520

237: 885-888. 521

26

5. Jucker, B. M., Cline, G. W., Barucci, N. & Shulman, G. I. (1999) Differential effects of 522

safflower oil versus fish oil feeding on insulin-stimulated glycogen synthesis, glycolysis, 523

and pyruvate dehydrogenase flux in skeletal muscle - A C-13 nuclear magnetic resonance 524

study. Diabetes 48: 134-140. 525

6. Neschen, S., Morino, K., Dong, J. Y., Wang-Fischer, Y., Cline, G. W., Romanelli, A. J., 526

Rossbacher, J. C., Moore, I. K., Regittnig, W. et al. (2007) N-3 fatty acids preserve insulin 527

sensitivity in vivo in a peroxisonte proliferator-activated receptor-alpha-dependent manner. 528

Diabetes 56: 1034-1041. 529

7. Samane, S., Christon, R., Dombrowski, L., Turcotte, S., Charrouf, Z., Lavigne, C., Levy, E., 530

Bachelard, H., Amarouch, H. et al. (2009) Fish oil and argan oil intake differently modulate 531

insulin resistance and glucose intolerance in a rat model of dietary-induced obesity. 532

Metabolism-Clinical and Experimental 58: 909-919. 533

8. Kalupahana, N. S., Claycombe, K., Newman, S. J., Stewart, T., Siriwardhana, N., Matthan, 534

N., Lichtenstein, A. H. & Moustaid-moussa, N. (2010) Eicosapentaenoic Acid Prevents and 535

Reverses Insulin Resistance in High-Fat Diet-Induced Obese Mice via Modulation of 536

Adipose Tissue Inflammation. The Journal of Nutrition 140: 1915-1922. 537

9. Oh, D. Y., Talukdar, S., Bae, E. J., Imamura, T., Morinaga, H., Fan, W. Q., Li, P. P., Lu, W. 538

J., Watkins, S. M. & Olefsky, J. M. (2010) GPR120 Is an Omega-3 Fatty Acid Receptor 539

Mediating Potent Anti-inflammatory and Insulin-Sensitizing Effects. Cell 142: 687-698. 540

10. Belzung, F., Raclot, T. & Groscolas, R. (1993) Fish oil n-3 fatty acids selectively limit the 541

hypertrophy of abdominal fat depots in growing rats fed high-fat diets. Am J Physiol. 246: 542

R1111-R1118. 543

11. Wang, H., Storlien, LH. & Huang, XF. (2002) Effects of dietary fat types on body fatness, 544

leptin, and ARC leptin receptor, NPY, and AgRP mRNA expression. Am J Physiol 545

Endocrinol Metab. 282: E1352-E1359. 546

12. Ruzickova, J., Rossmeisl, M., Prazak, T., Flachs, P., Sponarova, J., Vecka, M., Tvrzicka, E., 547

Bryhn, M. & Kopecky, J. (2004) Omega-3 PUFA of marine origin limit diet-induced obesity 548

in mice by reducing cellularity of adipose tissue. Lipids 39, pg 1177, 2004): 1177-1185. 549

13. Flachs, P., Horakova, O., Brauner, P., Rossmeisl, M., Pecina, P., Franssen-van Hal, N., 550

Ruzickova, J., Sponarova, J., Drahota, Z. et al. (2005) Polyunsaturated fatty acids of marine 551

origin upregulate mitochondrial biogenesis and induce beta-oxidation in white fat. 552

Diabetologia 48: 2365-2375. 553

14. Arai, T., Kim, H. J., Chiba, H. & Matsumoto, A. (2009) Anti-obesity effect of fish oil and 554

fish oil-fenofibrate combination in female KK mice. J Atheroscler Thromb. 16: 675-683. 555

15. Rokling-Andersen, M. H., Rustan, A. C., Wensaas, A. J., Kaalhus, O., Wergedahl, H., Røst, 556

T. H., Jensen, J., Graff, B. A., Caesar, R. & Drevon, C. A. (2009) Marine omega-3 fatty 557

acids promote size reduction of visceral adipose depots, without altering body weight and 558

composition, in male Wistar rats fed a high-fat diet. British Journal of Nutrition 102: 995-559

1006. 560

27

16. Madsen, L., Pedersen, L. M., Liaset, B., Ma, T., Petersen, R. K., van den Berg, S., Pan, J., 561

Muller-Decker, K., Dulsner, E. D. et al. (2008) cAMP-dependent Signaling Regulates the 562

Adipogenic Effect of n-6 Polyunsaturated Fatty Acids. J. Biol. Chem. 283: 7196-7205. 563

17. Ma, T., Liaset, B., Hao, Q., Pedersen, R. K., Fjære, E., Ngo, H. T., Lillefosse, H. H., 564

Jørgensem, S. R., Wojtaszewski, J. et al. (2011) Sucrose counteracts the anti-inflammatory 565

effect of fish oil in adipose tissue and increases obesity development in mice. PLoS ONE, In 566

press. 567

18. Madsen, L. & Kristiansen, K. (2010) The importance of dietary modulation of cAMP and 568

insulin signaling in adipose tissue and the development of obesity. Foods for Health in the 569

21St Century: A Road Map for the Future 1190: 1-14. 570

19. Bluher, M., Michael, M. D., Peroni, O. D., Ueki, K., Carter, N., Kahn, B. B. & Kahn, C. R. 571

(2002) Adipose Tissue Selective Insulin Receptor Knockout Protects against Obesity and 572

Obesity-Related Glucose Intolerance. Developmental Cell 3: 25-38. 573

20. Murata, Y., Tsuruzoe, K., Kawashima, J., Furukawa, N., Kondo, T., Motoshima, H., Igata, 574

M., Taketa, K., Sasaki, K. et al. (2007) IRS-1 transgenic mice show increased epididymal fat 575

mass and insulin resistance. Biochem Biophys Res Commun. 364: 301-307. 576

21. Curry, D. L. (1989) Effects of mannose and fructose on the synthesis and secretion of 577

insulin. Pancreas 4: 2-9. 578

22. Sato, Y., Ito, T., Udaka, N., Kanisawa, M., Noguchi, Y., Cushman, S. W. & Satoh, S. (1996) 579

Immunohistochemical localization of facilitated-diffusion glucose transporters in rat 580

pancreatic islets. Tissue Cell. 28: 637-643. 581

23. Ganda, O. P., Soelder, J., Gleason, R. E., Cleator, I. G. & Reynolds, C. (1979) Metabolic 582

Effects of Glucose, Mannose, Galactose, and Fructose in Man. J Clin Endocrinol Metab 49: 583

616-622. 584

24. Sykes, S., Morgan, L., english, J. & Marks, V. (1980) Evidence for preferential stimulation 585

of gastric inhibitory polypeptide secretion in the rat by actively transported carbohydrates 586

and their analogues. J Endocrinol 85: 201-207. 587

25. Rizkalla, S. (2010) Health implications of fructose consumption: A review of recent data. 588

Nutrition & Metabolism 7: 82. 589

26. Tran, L. T., Yuen, V. G. & McNeill, J. H. (2009) The fructose-fed rat: a review on the 590

mechanisms of fructose-induced insulin resistance and hypertension. Molecular and Cellular 591

Biochemistry 332: 145-159. 592

27. Jenkins, D. J., Wolever, T. M., Taylor, R. H., Barker, H., Fielden, H., Baldwin, J. M., 593

Bowling, A. C., Newman, H. C., Jenkins, A. L. & Goff, D. V. (1981) Glycemic index of 594

foods: a physiological basis for carbohydrate exchange. The American Journal of Clinical 595

Nutrition 34: 362-366. 596

28

28. Esfahani, A., Wong, J. M. W., Mirrahimi, A., Villa, C. R. & Kendall, C. W. C. (2011) The 597

application of the glycemic index and glycemic load in weight loss: A review of the clinical 598

evidence. IUBMB Life 63: 7-13. 599

29. Barclay, A. W., Petocz, P., Millan-Price, J., Flood, V. M., Prvan, T., Mitchell, P. & Brand-600

Miller, J. C. (2008) Glycemic index, glycemic load, and chronic disease risk-a meta-analysis 601

of observational studies. The American Journal of Clinical Nutrition 87: 627-637. 602

30. Lara-Castro, C. & Garvey, W. T. (2004) Diet, Insulin Resistance, and Obesity: Zoning in on 603

Data for Atkins Dieters Living in South Beach. J Clin Endocrinol Metab 89: 4197-4205. 604

31. Astrup, A., Larsen, T. M. & Harper, A. (2004) Atkins and other low-carbohydrate diets: 605

hoax or an effective tool for weight loss? The Lancet 364: 897-899. 606

32. Thomas, D. E., Elliott, E. J. & Baur, L. (2007) Low glycaemic index or low glycaemic load 607

diets for overweight and obesity. Cochrane Database of Systematic Reviews. 608

33. Pawlak, D. B., Kushner, J. A. & Ludwig, D. S. (2004) Effects of dietary glycaemic index on 609

adiposity, glucose homoeostasis, and plasma lipids in animals. The Lancet 364: 778-785. 610

34. Kim, J. H., Mynatt, R. L., Moore, J. W., Woychik, R. P., Moustaid, N. & Zemel, M. B. 611

(1996) The effects of calcium channel blockade on agouti-induced obesity. FASEB J. 10: 612

1646-1652. 613

35. Liaset, B., Madsen, L., Hao, Q., Criales, G., Mellgren, G., Marschall, H.-U., Hallenborg, P., 614

Espe, M., Frøyland, L. & Kristiansen, K. (2009) Fish protein hydrolysate elevates plasma 615

bile acids and reduces visceral adipose tissue mass in rats. Biochimica et Biophysica Acta 616

1791: 254-262. 617

36. Arslan, G., Brunborg, L. A., Frøyland, L., Brun, J. G., Valen, M. & Berstad, A. (2002) 618

Effects of duodenal seal oil administration in patients with inflammatory bowel disease. 619

Lipids 37: 935-940. 620

37. Barbatelli, G., Murano, I., Madsen, L., Hao, Q., Jimenez, M., Kristiansen, K., Giacobino, J. 621

P., De Matteis, R. & Cinti, S. (2010) The emergence of cold-induced brown adipocytes in 622

mouse white fat depots is determined predominantly by white to brown adipocyte 623

transdifferentiation. American Journal of Physiology-Endocrinology and Metabolism 298: 624

E1244-E1253. 625

38. Madsen, L., Petersen, R. K., Sørensen, M. B., Jørgensen, C., Hallenborg, P., Pridal, L., 626

Fleckner, J., Amri, E.-Z., Krieg, P. et al. (2003) Adipocyte differentiation of 3T3-L1 627

preadipocytes is dependent on lipoxygenase activity during the initial stages of the 628

differentiation process. Biochem. J. 375: 539-549. 629

39. Hwang, I. S., Ho, H., Hoffman, B. B. & Reaven, G. M. (1987) Fructose-Induced Insulin 630

Resistance and Hypertension in Rats. Hypertension 10: 512-516. 631

40. Cancello, R. & Clement, K. (2006) Is obesity an inflammatory illness? Role of low-grade 632

inflammation and macrophage infiltration in human white adipose tissue. BJOG: An 633

International Journal of Obstetrics and Gynaecology 113: 1141-1147. 634

29

41. Guilherme, A., Virbasius, J. V., Puri, V. & Czech, M. P. (2008) Adipocyte dysfunctions 635

linking obesity to insulin resistance and type 2 diabetes. Nat Rev Mol Cell Biol 9: 367-377. 636

42. Coate, K. C. & Huggins, K. W. (2010) Consumption of a high glycemic index diet increases 637

abdominal adiposity but does not influence adipose tissue pro-oxidant and antioxidant gene 638

expression in C57BL/6 mice. Nutr Res 30(2): 141-150 (abs.). 639

43. Yamazaki, K., Yasuda, N., Inoue, T., Yamamoto, E., Sugaya, Y., Nagakura, T., Shinoda, 640

M., Clark, R., Saeki, T. & Tanaka, I. (2007) Effects of the Combination of a Dipeptidyl 641

Peptidase IV Inhibitor and an Insulin Secretagogue on Glucose and Insulin Levels in Mice 642

and Rats. J Pharmacol Exp Ther 320: 738-746. 643

44. Fagerholm, V., Scheinin, M. & Haaparanta, M. (2008) alpha2A-adrenoceptor antagonism 644

increases insulin secretion and synergistically augments the insulinotropic effect of 645

glibenclamide in mice. Br J Pharmacol 154: 1287-1296. 646

45. Miki, H., Yamauchi, T., Suzuki, R., Komeda, K., Tsuchida, A., Kubota, N., Terauchi, Y., 647

Kamon, J., Kaburagi, Y. et al. (2001) Essential Role of Insulin Receptor Substrate 1 (IRS-1) 648

and IRS-2 in Adipocyte Differentiation. Mol. Cell. Biol. 21: 2521-2532. 649

46. Kim, J. K., Michael, M. D., Previs, S. F., Peroni, O. D., Mauvais-Jarvis, F., Neschen, S., 650

Kahn, B. B., Kahn, C. R. & Shulman, G. I. (2000) Redistribution of substrates to adipose 651

tissue promotes obesity in mice with selective insulin resistance in muscle. The Journal of 652

Clinical Investigation 105: 1791-1797. 653

47. Cariou, B., Postic, C., Boudou, P., Burcelin, R., Kahn, C. R., Girard, J., Burnol, A. F. & 654

Mauvais-Jarvis, F. (2004) Cellular and Molecular Mechanisms of Adipose Tissue Plasticity 655

in Muscle Insulin Receptor Knockout Mice. Endocrinology 145: 1926-1932. 656

48. Shepherd, P. R., Gnudi, L., Tozzo, E., Yang, H., Leach, F. & Kahn, B. B. (1993) Adipose 657

cell hyperplasia and enhanced glucose disposal in transgenic mice overexpressing GLUT4 658

selectively in adipose tissue. J. Biol. Chem. 268: 22243-22246. 659

49. Cusin, I., Rohner-Jeanrenaud, F., Terrettaz, J. & Jeanrenaud, B. (1992) Hyperinsulinemia 660

and its impact on obesity and insulin resistance. Int J Obes Relat Metab Disord. 16: S1-S11. 661

50. Cusin, I., Terrettaz, J., Rohner-Jeanrenaud, F. & Jeanrenaud, B. (1990) Metabolic 662

consequences of hyperinsulinaemia imposed on normal rats on glucose handling by white 663

adipose tissue, muscles and liver. Biochem J. 267: 99-103. 664

51. Cusin, I., Terrettaz, J., Rohner-Jeanrenaud, F., Zarjevski, N., Assimacopoulos-Jeannet, F. & 665

Jeanrenaud, B. (1990) Hyperinsulinemia increases the amount of GLUT4 mRNA in white 666

adipose tissue and decreases that of muscles: a clue for increased fat depot and insulin 667

resistance. Endocrinology. 127: 3246-3248. 668

52. Mynatt, R. L., Miltenberger, R. J., Klebig, M. L., Zemel, M. B., Wilkinson, J. E., Wilkinson, 669

W. O. & Woychik, R. P. (1997) Combined effects of insulin treatment and adipose tissue-670

specific agouti expression on the development of obesity. Proc Natl Acad Sci U S A. 94: 671

919-922. 672

30

53. Mitri, J. & Hamdy, O. (2009) Diabetes medications and body weight. Expert Opin Drug Saf. 673

8: 573-584. 674

54. Fonseca V (2003) Effect of thiazolidinediones on body weight in patients with diabetes 675

mellitus. Am J Med. 115: 42s-48s. 676

55. Marsset-Baglieri, A., Fromentin, G., Tome, D., Bensaid, A., Makkarios, L. & Even, P. C. 677

(2004) Increasing the Protein Content in a Carbohydrate-Free Diet Enhances Fat Loss 678

during 35% but Not 75% Energy Restriction in Rats. J. Nutr. 134: 2646-2652. 679

56. Morris, K. L., Namey, T. C. & Zemel, M. B. (2003) Effects of dietary carbohydrate on the 680

development of obesity in heterozygous Zucker rats. The Journal of Nutritional 681

Biochemistry 14: 32-39. 682

57. Alemzadeh, R., Jacobs, W. & Pitukcheewanont, P. (1996) Antiobesity effect of diazoxide in 683

obese zucker rats. Metabolism 45: 334-341. 684

58. Nagata, R., Nishio, Y., Sekine, O., Nagai, Y., Maeno, Y., Ugi, S., Maegawa, H. & 685

Kashiwagi, A. (2004) Single Nucleotide Polymorphism (Çô468 Gly to Ala) at the 686

Promoter Region of Sterol Regulatory Element-binding Protein-1c Associates with Genetic 687

Defect of Fructose-induced Hepatic Lipogenesis. J. Biol. Chem. 279: 29031-29042. 688

59. Pawlak, D. B., Bryson, J. M., Denyer, G. S. & Brand-Miller, J. C. (2001) High Glycemic 689

Index Starch Promotes Hypersecretion of Insulin and Higher Body Fat in Rats without 690

Affecting Insulin Sensitivity. J. Nutr. 131: 99-104. 691

60. Larsen, T. M., Dalskov, S. M., van Baak, M., Jebb, S. A., Papadaki, A., Pfeiffer, A. F. H., 692

Martinez, J. A., Handjieva-Darlenska, T., Kunesova, M. et al. (2010) Diets with High or 693

Low Protein Content and Glycemic Index for Weight-Loss Maintenance. New England 694

Journal of Medicine 363: 2102-2113. 695

61. Puigserver, P., Rhee, J., Donovan, J., Walkey, C. J., Yoon, J. C., Oriente, F., Kitamura, Y., 696

Altomonte, J., Dong, H. et al. (2003) Insulin-regulated hepatic gluconeogenesis through 697

FOXO1-PGC-1[alpha] interaction. Nature 423: 550-555. 698

62. Yoon, J. C., Puigserver, P., Chen, G., Donovan, J., Wu, Z., Rhee, J., Adelmant, G., Stafford, 699

J., Kahn, C. R. et al. (2001) Control of hepatic gluconeogenesis through the transcriptional 700

coactivator PGC-1. Nature 413: 131-138. 701

63. Herzig, S., Long, F., Jhala, U. S., Hedrick, S., Quinn, R., Bauer, A., Rudolph, D., Schutz, 702

G., Yoon, C. et al. (2001) CREB regulates hepatic gluconeogenesis through the coactivator 703

PGC-1. Nature 413: 179-183. 704

64. Fine, E. J. & Feinman, R. D. (2004) Thermodynamics of weight loss diets. Nutrition and 705

metabolism 1: 15-23. 706

65. Madsen, L., Pedersen, L. M., Lillefosse, H. H., Fjaere, E., Bronstad, I., Hao, Q., Petersen, R. 707

K., Hallenborg, P., Ma, T. et al. (2010) UCP1 Induction during Recruitment of Brown 708

Adipocytes in White Adipose Tissue Is Dependent on Cyclooxygenase Activity. PLoS ONE 709

5. 710

31

66. Vegiopoulos, A., Muller-Decker, K., Strzoda, D., Schmitt, I., Chichelnitskiy, E., Ostertag, 711

A., Diaz, M. B., Rozman, J., Hrabe de Angelis, M. et al. (2010) Cyclooxygenase-2 Controls 712

Energy Homeostasis in Mice by de Novo Recruitment of Brown Adipocytes. Science 328: 713

1158-1161. 714

67. Kopecky, J., Hodny, Z., Rossmeisl, M., Syrovy, I. & Kozak, L. P. (1996) Reduction of 715

dietary obesity in aP2-Ucp transgenic mice: physiology and adipose tissue distribution. Am 716

J Physiol. 270: E768-E775. 717

68. Cordain, L., Eaton, S. B., Sebastian, A., Mann, N., Lindeberg, S., Watkins, B. A., O'Keefe, 718

J. H. & Brand-Miller, J. (2005) Origins and evolution of the Western diet: health 719

implications for the 21st century. Am J Clin Nutr 81: 341-354. 720

721

722

UCP1 Induction during Recruitment of Brown Adipocytesin White Adipose Tissue Is Dependent onCyclooxygenase ActivityLise Madsen1,2*, Lone M. Pedersen3, Haldis Haukaas Lillefosse1,2, Even Fjære1,2, Ingeborg Bronstad4, Qin

Hao1, Rasmus K. Petersen1, Philip Hallenborg1, Tao Ma1, Rita De Matteis5, Pedro Araujo2, Josep

Mercader6, M. Luisa Bonet6, Jacob B. Hansen7, Barbara Cannon8, Jan Nedergaard8, Jun Wang1,9, Saverio

Cinti10, Peter Voshol11, Stein Ove Døskeland4, Karsten Kristiansen1,9*

1 Department of Biology, University of Copenhagen, Copenhagen, Denmark, 2 National Institute of Nutrition and Seafood Research, Bergen, Norway, 3 Department of

Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark, 4 Department of Biomedicine, University of Bergen, Bergen, Norway,

5 Department of Biomolecular Sciences, University of Urbino, Urbino, Italy, 6 Laboratory of Molecular Biology, Nutrition and Biotechnology, Universitat de les Illes Balears,

and CIBER de Fisiopatologıa de la Obesidad y Nutricion (CIBERobn), Palma de Mallorca, Spain, 7 Department of Biomedical Sciences, University of Copenhagen,

Copenhagen, Denmark, 8 The Wenner-Gren Institute, Stockholm University, Stockholm, Sweden, 9 BGI-Shenzhen, Shenzhen, China, 10 Department of Molecular

Pathology and Innovative Therapies, University of Ancona, Ancona, Italy, 11 Metabolic Research Laboratories, University of Cambridge, Cambridge, United Kingdom

Abstract

Background: The uncoupling protein 1 (UCP1) is a hallmark of brown adipocytes and pivotal for cold- and diet-inducedthermogenesis.

Methodology/Principal Findings: Here we report that cyclooxygenase (COX) activity and prostaglandin E2 (PGE2) arecrucially involved in induction of UCP1 expression in inguinal white adipocytes, but not in classic interscapular brownadipocytes. Cold-induced expression of UCP1 in inguinal white adipocytes was repressed in COX2 knockout (KO) mice andby administration of the COX inhibitor indomethacin in wild-type mice. Indomethacin repressed b-adrenergic induction ofUCP1 expression in primary inguinal adipocytes. The use of PGE2 receptor antagonists implicated EP4 as a main PGE2

receptor, and injection of the stable PGE2 analog (EP3/4 agonist) 16,16 dm PGE2 induced UCP1 expression in inguinal whiteadipose tissue. Inhibition of COX activity attenuated diet-induced UCP1 expression and increased energy efficiency andadipose tissue mass in obesity-resistant mice kept at thermoneutrality.

Conclusions/Significance: Our findings provide evidence that induction of UCP1 expression in white adipose tissue, but notin classic interscapular brown adipose tissue is dependent on cyclooxygenase activity. Our results indicate thatcyclooxygenase-dependent induction of UCP1 expression in white adipose tissues is important for diet-inducedthermogenesis providing support for a surprising role of COX activity in the control of energy balance and obesitydevelopment.

Citation: Madsen L, Pedersen LM, Lillefosse HH, Fjære E, Bronstad I, et al. (2010) UCP1 Induction during Recruitment of Brown Adipocytes in White Adipose TissueIs Dependent on Cyclooxygenase Activity. PLoS ONE 5(6): e11391. doi:10.1371/journal.pone.0011391

Editor: Aimin Xu, University of Hong Kong, China

Received May 19, 2010; Accepted May 30, 2010; Published June 30, 2010

Copyright: � 2010 Madsen et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by the Danish Natural Science Research Council, the Novo Nordisk Foundation and the Carlsberg Foundation. Part of thework was carried out as a part of the research program of the Danish Obesity Research Centre (DanORC). DanORC is supported by The Danish Council for StrategicResearch (Grant No 2101 06 0005). CIBER de Fisiopatologia de la Obesidad y Nutricion is an initiative of the ISCIII. The funders had no role in study design, datacollection and analysis, decision to publish, or preparation of the manuscript.

Competing Interests: The authors have declared that no competing interests exist.

* E-mail: [email protected] (KK); [email protected] (LM)

Introduction

The two types of adipose tissues, white (WAT) and brown (BAT),

have opposite functions in whole body energy homeostasis. Whereas

white adipocytes store excess energy as fat, brown adipocytes

contain a large number of mitochondria dedicated to convert fat

into heat through uncoupled respiration. The uncoupling of

respiration and the resulting heat dissipation depend on the

expression of the uncoupling protein 1 (UCP1). UCP1 is an integral

membrane protein unique to brown adipocyte mitochondria, where

it acts as a proton channel to uncouple oxidative phosphorylation by

dissipating the proton gradient across the inner mitochondrial

membrane [1]. In mice, an increased content of UCP1 in adipose

tissue mitochondria is strongly linked to protection against diet-

induced obesity. This is true whether increased UCP1 expression is

induced by transgenic expression of UCP1 itself [2;3], of forkhead

box 2 (FOXC2) [4], of PR domain containing 16 (PRDM16) [5] or

by disruption of the RIIb subunit of protein kinase A [6;7],

eukaryotic translation initiation factor E4-binding protein 1 (4E-

BP1) [8], cell death inducing DFFA like effector A and C (Cidea and

Cidec/Fsp27) [9], the p160 coregulator TIF2 [10] or retinoblasto-

ma Rb [11–13].


Although it has been estimated that 50 g of brown adipocytes

would be sufficient to burn 20% of the daily energy intake [14],

BAT has traditionally been considered to be virtually absent and

of no physiological relevance in adult humans. This view has

recently changed dramatically with the demonstration of func-

tional BAT in adult humans [15–19] adding to the observation of

brown-like multilocular adipocytes expressing UCP1 interspersed

within human WAT [20–22]. Actually, UCP1 mRNA has been

detected in all adipose tissues in adult humans, and it has been

estimated that 1 in 100–200 adipocytes in human intraperitoneal

adipose tissue expresses UCP1 [23].

Classic interscapular brown adipocytes and brown-like adipo-

cytes found in WAT depots appear to originate from distinct

lineages. Brown pre-adipocytes derived from the interscapular

region (iBAT) demonstrate myogenic gene expression [24] and

classic brown adipocytes arise from Myf5 expressing progenitors

[25]. In contrast, brown-like adipocytes appearing in white

adipose tissue by b-adrenergic stimulation (‘‘brite adipocytes’’)

appear to originate from another lineage, much closer to white

adipocytes [26–29] and display different molecular markers [30].

Several lines of evidence suggest that the number of brown-like

adipocytes in WAT depots might influence whole body energy

balance. Increased occurrence of brown-like adipocytes within

WAT depots is a feature of mouse strains resistant to dietary

obesity, such as the A/J strain [31;32], and reduced adiposity

associated with aP2-transgenic expression of UCP1 is linked to

increased energy dissipation in white, but not interscapular brown,

adipose tissue [33]. Human obesity is associated with a reduced

expression of UCP1 and other thermogenesis related genes in

WAT depots [34;35]. Thus, identification of factors controlling

induction of UCP1 expression and an increase in the number of

brown-like adipocytes in white depots obviously deserves further

attention.

It is intriguing that the cold-induced occurrence of brown-like

adipocytes and UCP1 requires the presence of the b3-adrenocep-

tor in previously white adipose tissue, but not in interscapular

brown adipose tissue [36]. Furthermore, the presence of the b3-

adrenoceptor is required for full stimulation of energy expenditure

and oxygen consumption in white adipose tissue [37].

Adipocytes from lean rats have higher isoprenalin-stimulated

prostaglandin E2 (PGE2) synthesis, than adipocytes from obese

Zucker rats [38]. We therefore hypothesized that prostaglandins

or related products synthesized by cyclooxygenases (COXs)

might be involved in the recruitment of brown adipocytes in

white depots. The COXs have previously been implicated in

adipogenesis [39–41], but no specific role has been assigned. Here,

we demonstrate that COX activity is crucially involved in the

induction of UCP1 expression in WAT providing further evidence

for a role of COXs in the control of energy balance and obesity

development. In view of the worldwide epidemic of obesity and

associated metabolic disorders it is obviously of importance to

identify pathways that can be manipulated genetically or

pharmacologically and regulate the induction of UCP1 expression

and recruitment of brown-like adipocytes in white adipose tissues.

Results

COX1 and COX2 protein expression is upregulated iniWAT during cold treatment

When mice are kept at 28uC, close to thermoneutrality, the

majority of the adipocytes in the inguinal white adipose depot

(iWAT) – the major subcutaneous depot in the mouse – appear as

UCP1-negative, spherical unilocular adipocytes [42]. In iWAT

from warm-acclimated mice, only endothelial cells and macro-

phages stained positive for COX1 and COX2, respectively

(Figure 1A and B). However, when mice were transferred to a

cold environment, the emerging multilocular adipocytes stained

positive for COX1 and COX2. In particular, cells that appeared

to be in a transition state between uni- and multilocular cells

stained strongly (Figure 1A and B). Western blotting demonstrated

increased expression of COX1 and COX2 in iWAT, and also in

iBAT, after cold exposure (Figure 1C). Real time qPCR analysis

verified that genes preferentially expressed in brown vs. white

adipose tissue, such as UCP1, peroxisome proliferator activated

receptor gamma coactivator 1a (PGC1a), type II thyroxine

deiodinase (Dio2), cytochrome C oxidase subunit 8b (Cox8b),

epithelial V like antigen 1 (Eva1) and Cidea were all highly

induced in iWAT upon cold exposure, whereas expression of 4E-

BP1 and of nuclear receptor interacting protein 140 (Nrip1/

RIP140) was reduced (Figure 1D). Immunohistochemical staining

of iBAT from cold-treated mice demonstrated that adipocytes

stained positive for COX2, whereas only endothelial cells stained

positive for COX1 (Figure S1A). The lack of COX1 and COX2

expression in adipocytes from iBAT in warm-acclimated mice was

verified by analysis of protein and RNA isolated from fractionated

adipose tissue, in which COX1 and COX2 was detected solely in

the stromal vascular fraction (Figure S1B).

Inhibition of COX activity represses induction of UCP1expression

Differentiated mouse embryo fibroblasts (MEFs) lacking the

retinoblastioma (Rb) gene, resemble brown or brown-like

adipocytes in demonstrating b-adrenergic induction of UCP1

expression [43]. To achieve a robust induction of UCP1

expression, differentiated Rb2/2 MEFs were treated with a

combination of isoproterenol and 9-cis retinoic acid [44]. Just as

cold exposure increased COX1 and COX2 mRNA and protein

levels in brown-like adipocytes (Figure 1), isoproterenol/9-cis

retinoic acid treatment increased COX1 and COX2 mRNA and

protein expression (Figure 2A and B) in this model system.

Upregulation of COX1 and COX2 expression in Rb2/2

adipocytes was accompanied by increased production of PGE2,

the primary prostaglandin produced by mature adipocytes

[45;46], but not of PGF2a and 6-keto-PGF1a (Figure 2C). This

indicates that Rb2/2 adipocytes resemble mature adipocytes in

producing PGE2 as the major prostaglandin species.

To investigate the importance of COX activity for induction of

UCP1 expression, we treated differentiated Rb2/2 adipocytes

with isoproterenol/9-cis retinoic acid in the absence or presence of

the general COX inhibitor indomethacin. Indomethacin prevent-

ed induction of UCP1 mRNA and protein expression (Figure 2D

and E), thus suggesting the intriguing possibility that COX activity

is required for induction of UCP1.

To examine if COX activity was required also in primary

adipocytes, we induced cells from the stromal vascular fraction of

iBAT and iWAT to differentiate and then treated the mature

adipocytes with the b-adrenergic agonist isoproterenol in the

absence and presence of indomethacin. Interestingly, indometh-

acin inhibited isoproterenol-induced UCP1 expression in cells

derived from iWAT but not from iBAT (Figure 3A), indicating

that COX activity is required for b-adrenergic induction of UCP1

expression in adipocytes from iWAT, but not in iBAT adipocytes.

In keeping with this notion, indomethacin only marginally

attenuated induction of UCP1 expression in the WT-1 cell model

representing interscapular brown adipocytes (Text S1, Figure S2)

[47].

To investigate the role of COX activity during induction of

UCP1 expression in iWAT and iBAT in vivo, we treated warm-

Cyclooxygenases and UCP1


acclimated mice with the COX inhibitor indomethacin and

transferred the mice to 4uC. Measurements of rectal temperature

revealed that mice treated with indomethacin had slightly, but

significantly lower body temperature (Figure 3B). As expected,

UCP1 expression was induced in both iBAT and iWAT in

vehicle-treated mice (Figure 3C and D). While indomethacin

treatment only slightly attenuated cold-induced UCP1 expression

in iBAT, it almost completely prevented the induction of UCP1

expression in iWAT (Figure 3C and D). Thus, COX activity

appeared to be necessary for cold-induced UCP1 expression in

iWAT, but not in iBAT. In addition, indomethacin treatment

attenuated cold-induced enhancement of PGC1a, Dio2, Cox8b,

Eva1 and Cidea expression in iWAT, while preventing cold-

induced repression of RIP140 and 4E-BP1 expression in iWAT

(Figure 3D).

Forced expression of COX2 induces UCP1 expression inRb2/2 adipocytes

Since indomethacin attenuated b-adrenergically stimulated

UCP1 expression in Rb2/2 adipocytes and primary inguinal

Figure 1. Cold exposure induces COX1 and COX2 expression in iWAT and iBAT. Sv129 mice were warm-acclimated at 28–30uC for 6 daysand then transferred to 4–6uC. Samples for cryosections, RNA and protein extractions were prepared from iBAT and iWAT after 2, 4 and 6 days at 4–6uC. A–B. Representative COX1 (A) and COX2 (B) immunoreactivity in iWAT from mice kept at 28–30uC for 6 days and after 6 days of cold exposure. C.Proteins were isolated from warm-acclimated mice (lane 1) and after 2, 4 and 6 days of cold exposure. COX1 and COX2 expression were determinedby Western blotting. D. RNA was harvested from iBAT and iWAT from individual mice (n = 4 in each group) that were warm-acclimated or cold-exposed for 6 days. Expressions of genes were measured by RT-qPCR in duplicates and normalized to TBP (TATA box binding protein). The barsrepresent mean 6 standard error. * indicates statistical difference (p,0.05) compared to expression in warm acclimated mice.doi:10.1371/journal.pone.0011391.g001



adipocytes, but not in WT-1 cells and primary interscapular

brown adipocytes, we again used Rb2/2 adipocytes as a model

system for ‘‘brite’’ adipocytes. To investigate the relative

importance of COX1 and COX2 activities in mediating induction

of UCP1 expression in such cells, we treated Rb2/2 adipocytes

with isoproterenol/9-cis retinoic acid in the absence and presence

of selective COX1 and COX2 inhibitors. As shown in Figure 4A,

selective inhibition of COX1 and COX2 with SC560 or NS398,

respectively, partially prevented UCP1 induction, whereas a

combination of these inhibitors or treatment with the non-selective

inhibitor indomethacin fully prevented UCP1 induction. Accord-

ingly, activities of both COX1 and COX2 seem necessary for full

UCP1 induction.

To further examine the relative importance of COX1 and

COX2 for prostaglandin synthesis and UCP1 expression, these

enzymes were retrovirally expressed both singly and in combi-

nation in Rb2/2 MEFs (Figure 4B). The cells were induced to

differentiate, and on day 8, the medium was replaced by fresh

medium, which was harvested 24 h later and analyzed for PGE2

content. The level of PGE2 was higher when the cells were

transduced with COX2 alone or in combination with COX1,

than with COX1 alone (Figure 4C). These results, together with

the fact that PGE2 formation in adipose tissue in COX2 KO

mice is significantly lower than in COX1 KO mice [48], point to

COX2 expression as being of major importance for PGE2

production. In accordance with this, forced expression of COX1

alone was unable to induce UCP1 expression (Figure 4D).

However, UCP1 expression was significantly induced by forced

expression of COX2 alone or in combination with COX1

(Figure 4D). Increased expression of UCP1 was accompanied

with increased expression of PGC1a, Dio2, Cox8b, Eva1 and

Cidea, as well as reduced expression of RIP140, but not 4E-BP1

(Figure 4D).

Cold-induced UCP1 expression is attenuated in iWAT inCOX2 KO mice

To confirm the importance of COX2 for UCP1 induction in

iWAT, wild-type and COX2 KO mice were challenged with a

cold environment after warm acclimation. The wild-type mice

defended their body temperature better than the COX2 KO mice

(Figure 5A). The COX2 KO mice develop severe nephropathy

and are susceptible to peritonitis in early life [49]; therefore, KO

and wild-type littermates 6 weeks of age were used in this

experiment. Unfortunately, we were unable to collect sufficient

amounts of iWAT from these young mice to detect UCP1 or COX

by Western blotting. However, as expected, cold-induced UCP1

mRNA expression was attenuated in iWAT in COX2 KO mice

(Figure 5B). Cold-induced expression of Dio2 and Cidea was also

attenuated in iWAT in the COX2 KO mice and PGC1aexpression also tended to be attenuated (Figure 5B). Moreover,

the cold-induced reduction of RIP140 expression was prevented in

the COX2 KO mice (Figure 5B). Expression of Cox8b, Eva1 and

4E BP1 was, however, not significantly different in iWAT from

wild-type and COX2 KO mice, suggesting that inhibition of both

COX1 and COX2 might be necessary to attenuate cold-induced

changes in the expression of these genes. As expected, we observed

no differences in UCP1 expression in iBAT in COX2 KO and

wild-type mice, and surprisingly, cold-treated COX2 KO mice

had significantly higher expression of PGC1a in iBAT than did

wild-type mice (Figure 5B).

PGE2 induces UCP1 expression via activation of the EP3/EP4 receptors

PGE2 is reported to mediate its action by interacting with four

subtypes of PGE receptors, the EP1, EP2, EP3 and EP4 receptors

[50], but may also bind to the prostaglandin F (FP) receptor with an

Figure 2. Indomethacin prevents isoproterenol-induced UCP1 expression in Rb-negative adipocytes. Rb-negative mouse embryofibroblasts were induced to differentiate as described in experimental procedures. Differentiated cells were treated with vehicle or isoproterenol(100 nM) and 9-cis-retinoic acid (1 mM) for 24 h. Indomethacin (1 mM) was included when indicated in the figure. A and D. RNA was isolated andexpressions of genes were measured by RT-qPCR in duplicates and normalized to TBP. C. The levels of prostaglandin E2, F2a and 6-keto-prostaglandinF1a were determined in cell medium using ELISA kits after 24 h. B and E. Proteins were harvested and expressions of COX1, COX2 and UCP1 weremeasured by Western blotting. The bars represent mean 6 standard error. The experiments were performed in triplicates and performed 3–5 times.* indicates statistical significant difference (p,0.05).doi:10.1371/journal.pone.0011391.g002



affinity that is only 10–30 fold lower than that of PGF2a [51]. In

order to probe the relative importance of these receptors in

mediating the possible effect of PGE2 on induction of UCP1,

expression of the EP and FP receptors was measured in adipose

tissue and in Rb2/2 adipocytes. All receptors were expressed in

both white and brown adipose tissue, whereas no expression of the

EP3 receptor could be detected in Rb2/2 adipocytes (Figure 6A),

implying a minor if any role of this receptor in mediating the PGE2

response in those cells. Thus, Rb2/2 adipocytes were treated with

isoproterenol and 9-cis retinoic in the absence or presence of

AL8810, AH6809, or AH23848, that are FP-, EP1/EP2 and EP4

receptor antagonists, respectively. Isoproterenol-stimulated UCP1

expression was not affected by the FP receptor antagonist, but

slightly attenuated by the EP1/EP2 receptor antagonist and strongly

attenuated by the EP4 antagonist (Figure 6B). Reduced expression

of UCP1 was accompanied by reduced expression of PGC1a and

Cidea (Figure 6B). To verify the importance of PGE2 signaling via

the EP4 receptor with a possible minor contribution by the EP3

receptor, mice were injected with an EP3/EP4 receptor agonist [52],

the stable PGE2 analogue 16,16-dimethyl-PGE2 As predicted,

qPCR analysis revealed that UCP1 expression was induced in

iWAT, but not in iBAT (Figure 6C). Together, the in vitro and in vivo

results suggest that PGE2-induced UCP1 expression at least in part

is mediated via the EP3/EP4 receptors with EP4 being the

predominant receptor involved.

Inhibition of COX activity increases adiposity and energyefficiency in obesity resistant Sv129 mice

Diet-induced thermogenesis protects several mouse strains

against obesity [53–55]. Since it appears that the protection

against diet-induced obesity is related to increased occurrence of

brown-like adipocytes in white depots [56;57], we aimed to

investigate the hypothesis that indomethacin could also attenuate

diet-induced UCP1 expression and thereby increase the propen-

sity for diet-induced obesity in Sv129 mice. Since it was recently

demonstrated that UCP1-deficient mice become obese when

housed at thermoneutrality [58], we predicted that the most

pronounced effect of COX inhibition would be observed for mice

kept under thermoneutral conditions. Accordingly, we fed Sv129

mice a very high-fat diet with or without indomethacin

Figure 3. Indomethacin prevents cold-induced UCP1 expression in iWAT. A. Stromal vascular fractions (SVFs) were isolated from mouseiWAT and iBAT, cultured and induced to differentiate as described in experimental procedures. Differentiated adipocytes were treated with vehicle orisoproterenol (100 nM) in the absence or presence of indomethacin (1 mM) for 24 h. Expression of UCP1 was measured by RT-qPCR in duplicates andnormalized to PPARc. The bars represent mean 6 standard deviation. The experiment was performed in triplicates and repeated 3 times. B–D. C57Bl/6 mice were warm-acclimated at 28–30uC for 10 days and then transferred to 4–6uC for 48 h. Mice were injected with vehicle or indomethacin(2.5 mg/kg) 2 h prior transfer to 4uC and thereafter every 12 h. Rectal temperature was measured before the mice were transferred and after 24 and48 h (B). Protein and RNA extractions were isolated after 48 h. UCP1 expression was measured by Western blotting (C) and expressions of genes weremeasured by RT-qPCR and normalized to TBP (D). The bars represent mean 6 error (n = 5–6). * indicates statistical significant difference (p,0.05).doi:10.1371/journal.pone.0011391.g003



Figure 4. UCP1 expression is induced by forced expression of COX2 alone or in combination with COX1 in cultured cells. A. Rb2/2

MEFs were induced to differentiate as described in experimental procedures. Differentiated adipocytes were treated with vehicle or isoproterenol(100 nM) and 9-cis-retinoic acid (1 mM) in the absence and presence of the COX1 inhibitor SC560 (50 nM) or the COX2 inhibitor NS398 (5 mM), aloneor in combination, or with indomethacin (1 mM), for 24 h. Expression of UCP1 was measured by RT-qPCR. The bars represent mean 6 standard error.The experiment was performed in triplicates and repeated 2 times. B–D. Rb2/2 MEFs were retrovirally transduced with empty vector, vectorencoding COX1 or COX2, or both. The transduced cells were selected and induced to differentiate and analyzed for COX1 and COX2 expression byWestern blotting (B). PGE2 levels were measured in cell media (C). RNA was isolated on day 8 and expressions of genes were measured by RT-qPCR(D). The bars represent mean 6 standard error. Different letters indicate statistically significant difference (p,0.05). The experiments were performedin triplicates.doi:10.1371/journal.pone.0011391.g004



supplementation for 4 weeks while keeping the mice at 28–30uC.

As demonstrated in Figure 7A, indomethacin supplementation led

to a higher weight gain. Energy intake was slightly, but not

statistically significantly lower (data not shown). However, energy

intake relative to body weight gain was significantly lower in mice

that received the indomethacin-supplemented high-fat diet

(Figure 7A). Mice fed the diet supplemented with indomethacin

also had more WAT in different depots, but not iBAT (Figure 7B).

Histological analysis revealed that the adipocytes in both iWAT

and iBAT appeared normal, but adipocytes in iWAT in mice fed

the high-fat diet supplemented with indomethacin were slightly

larger (Figure 7C and D). As expected, feeding mice a high-fat diet

lead to augmented expression of UCP1 in both iBAT and iWAT

in vehicle-treated mice (Figure 7E). Importantly, diet-induced

UCP1 expression was prevented in iWAT, but not in iBAT in the

indomethacin-treated mice (Figure 7E). Reduced expression of

UCP1 in iWAT in mice fed a high-fat diet supplemented with

indomethacin was accompanied with reduced expression of

Cox8b. Expression of PGC1a, Dio2, Eva1, Cidea, 4E-BP1 and

RIP140 was not affected by inclusion of indomethacin with the

high-fat diet (Figure 7E). Collectively, these results underscore the

notion that inhibition of COX activity attenuates the acquisition of

‘‘brite’’ adipocytes in white adipose depots with an accompanying

increase in feed efficiency leading to accumulation of more adipose

tissue. Obviously, other mechanisms may contribute to the

increase in feed efficiency, but the lack of ‘‘brite’’ adipocyte

recruitment seems a key player.

Discussion

The unique energy-dissipating ability of UCP1 makes control of

its expression and activation potential targets for the development

of novel drugs for the treatment of obesity and obesity-associated

diseases. Here we present evidence that COX activity and COX-

derived PGE2 are intimately linked to induction of UCP1

expression in iWAT, but not in iBAT. Thus, cold-induced

expression of UCP1 in iWAT was repressed in mice treated with

the general COX inhibitor indomethacin, and in COX2 KO

Figure 5. Cold-induced UCP1 expression is attenuated in iWAT in COX2 KO mice. A B COX2 KO mice and wild-type littermates were warm-acclimated at 28–30uC for 10 days and then transferred to 4–6uC for 48 h. Rectal temperature was measured before transfer and after 24 and 48 h (A).RNA was extracted and expressions of genes were measured by RT-qPCR (B). The bars represent mean 6 standard error (n = 5–6). * indicatesstatistical significant difference between wild-type and KO mice (p,0.05).doi:10.1371/journal.pone.0011391.g005



mice. Also, injection of a stable analog of the COX2 downstream

product PGE2, 16,16-dimethyl-PGE2, induced UCP1 expression

in iWAT. Forced expression of COX2, alone or in combination

with COX1, induced UCP1 expression in a cell model resembling

inguinal adipocytes. Finally, the inhibition of COX activity not

only attenuated diet-induced UCP1 expression in iWAT, but also

increased weight gain in Sv129 mice kept at thermoneutrality.

The association between diet-induced thermogenesis and the

recruitment of brown adipose tissue was first noted more than 30

years ago and believed to involve the classical brown adipose tissue

located in the interscapular region [59]. The anti-obesity role of

UCP1 was challenged by the finding that UCP1 KO mice were

not obese [60]. However, the recent demonstration that UCP1

ablation per se induced obesity when the mice were kept at

thermoneutrality [61] clearly indicates that UCP1 is important in

diet-induced energy dissipation at thermoneutrality. Our data

indicate that inhibition of COX activity increased weight gain and

concomitantly attenuated diet-induced UCP1 expression in

iWAT, but not iBAT in Sv129 mice kept at thermoneutrality,

pointing to a novel role of COX activity in the control of energy

balance and the development of obesity. These results are in line

with our earlier observation that enhanced cAMP signaling in

response to an increased glucagon/insulin ratio led to an increased

COX-mediated PGE2 production. This was accompanied by

increased expression of UCP1 in iWAT, but not in iBAT, and

decreased feed efficiency [62]. Although neither COX1 KO nor

COX2 KO mice are obese, COX2+/2 KO mice have more

adipose tissue than wild-type littermates when fed an obesogenic

diet [48]. The reason why COX2+/2, but not COX2 KO mice

were reported to be prone to obesity is not clear. However, these

studies were not performed at thermoneutrality [48]. A similar

phenomenon is actually seen in GLUT4 KO mice, where the

majority of GLUT42/+, but not GLUT42/2 mice develops

diabetes [63]. Moreover, release of PGE2 from adipose tissue in

COX2+/2 mice was reported to be reduced compared to adipose

tissue from wild-type mice [48] and adipose tissue cultures

obtained from obese rats have lower PGE2 release rates than

cultures from lean rats [38]. In addition, microsomal prostaglandin

Figure 6. UCP1 expression is attenuated by an EP4 receptor antagonist in Rb2/2 adipocytes and induced by the EP4 receptoragonist 16,16dmPGE2 in vivo. A. Expressions of EP1, EP2, EP3, EP4 and FP receptors were measured by RT-qPCR in iBAT and iWAT isolated fromwarm- and cold-acclimated mice, and in Rb2/2 adipocytes treated with vehicle or isoproterenol (100 nM) and 9-cis-retinoic acid (1 mM). B. Rb2/2

adipocytes were treated with vehicle or isoproterenol (100 nM) and 9-cis-retinoic acid (1 mM) by RT-qPCR in absence and presence of AL8810,AH6809, or AH23848, which are FP, EP1/EP2 and EP4 receptor antagonists, respectively. Expressions of genes were measured by RT-qPCR. The barsrepresent mean 6 standard error. Different letters indicate statistically significant differences (p,0.05). C. C57BL/6J mice were subcutaneouslyinjected with vehicle or 16,16dmPGE2 (50 mM/kg) every 12 h for 48 h. Expressions of genes were measured by RT-qPCR. The bars represent mean 6standard error (n = 5). * indicates statistical significant difference between vehicle and 16,16 dmPGE2 treated mice (p,0.05).doi:10.1371/journal.pone.0011391.g006

α

Figure 7. Indomethacin prevents high-fat diet-induced UCP1 expression in iWAT but not iBAT in the obesity-resistant Sv129 mousestrain. Mice were fed a very high-fat diet (VHF) with or without indomethacin supplementation (16 ppm) for 4 weeks at a temperature of 28–30uC.One group of mice was killed before the experiment started. A. Body weight gain and energy intake relative to body weight gain. B. Differentadipose tissue depots were dissected and weighed. C and D. Representative paraffin-embedded representative sections from iWAT and iBAT werestained with hematoxylin and eosin. The scale bars represent 50 mM. E. Expressions of genes in iWAT and iBAT were measured by RT-qPCR. The barsrepresent mean 6 standard error (n = 6). * indicates statistical significant difference (p,0.05) between different groups.doi:10.1371/journal.pone.0011391.g007



E synthase1 (mPGES1) expression is reported to be downregulated

in iWAT and eWAT in obese mice [64], suggesting a

dysregulation of prostaglandin synthesis in obesity.

PGE2 mediates its action by interacting with four subtypes of

PGE2 receptors, the EP1, EP2, EP3 and EP4 receptors [50]. Using

the Rb2/2 brown-like adipocytes, we show that b-adrenergic

stimulation of UCP1 expression is attenuated by an EP4 receptor

antagonist. This combined with our finding that injection of the

EP3/EP4 receptor agonist 16,16-dimethyl-PGE2 increased expres-

sion of UCP1 in iWAT indicates that the action of PGE2 is

predominantly mediated via the EP4 receptor with a possible minor

contribution by the EP3 receptor. Most EP4 KO mice die shortly

after birth, and no adipose tissue phenotype has been reported for

the few surviving pups [65]. Similarly, to our knowledge, no adipose

tissue phenotype has been reported for the EP3 KO mice.

Collectively, our findings strongly indicate that both cold- and

diet-induced expression of UCP1 in iWAT, but not in iBAT,

requires COX activity and most likely PGE2 formation.

Furthermore, our results point to differential roles of induction

of UCP1 expression in iWAT and iBAT in the context of diet- and

cold-induced thermogenesis. We suggest that whereas UCP1 in

iWAT plays an important role in protection against obesity, UCP1

in iBAT is essential for temperature adaptation. Upon cold

challenge, body temperatures were only slightly lower in wild-type

mice treated with indomethacin and in COX2 KO mice

compared to non-treated wild-type mice. This is in line with the

earlier finding that UCP1 expression is blunted in iWAT, but not

in iBAT, in cold-adapted b3 adrenoreceptor KO mice [66].

The importance of COX in diet-induced expression of UCP1 in

iWAT and diet-induced thermogenesis is underscored by our

demonstration that inclusion of the general COX inhibitor

indomethacin in the diet augmented high-fat diet-induced obesity

in Sv129 mice kept at thermoneutrality irrespectively of UCP1

expression in iBAT. Increased expression of UCP1 in WAT with

accompanying increased thermogenic activity coupled with un-

changed or even reduced BAT activity has been observed in several

genetically modified lean mouse models such as RIP140 [67],

Caveolin 1 [68], Fsp27 [69], hormone sensitive lipase [70] and

vitamin D receptor [71] KO mice. Further, in RIIb mice [72], pRb-

deficient mice [11;73] and in mice overexpressing FOXC2 [74],

protection against diet-induced obesity is accompanied by an

increased RIa/RIIb ratio, rendering PKA somewhat more sensitive

to cAMP, which is accompanied by an increased occurrence of

brown adipocytes in WAT. Last, it should be recalled that in aP2-

UCP1 transgenic mice, both endogenous UCP1 expression and

respiration are actually reduced in iBAT [75]. UCP1 expression,

respiration and total oxidative capacity are, however, strongly

induced in WAT and the oxidative capacity of WAT is sufficient for

the changes of total energy balance induced by the transgene [76].

In keeping with the earlier notion that i) mouse strains that have

more UCP1-expressing adipocytes in their WAT depots are

protected against diet-induced obesity [77;78] and ii) brown-like

multilocular adipocytes expressing UCP1 are detected interspersed

within white adipose tissue in humans [20;21;79], we suggest that

factors influencing UCP1 expression in white adipose tissue are of

particular importance for the regulation of energy balance and the

development of obesity also in humans.

Materials and Methods

Ethics StatementThe animal experiments were approved by the Norwegian

Animal Health Authorities, ID 819 and 888. Care and handling

were in accordance with local institutional recommendations.

Cell culture, transduction and differentiationMouse embryo fibroblasts (MEFs) were prepared from wild-type

and Rb2/2 embryos [80]. The cells were grown and differentiated

in AmnioMax Medium as described earlier [81]. Retrovira

expressing pLXSN-hygro, pBabe-puro, pLXSN-COX1 or

pBabe-COX2 were harvested from Phoenix–Eco cells, plated at

30–40% confluence in DMEM supplemented with 10% fetal

bovine serum, and transductions performed as described [82].

Isolation of the stromal vascular fraction and adipocytesfrom mice

The stromal vascular fraction and adipocytes were obtained

from iWAT and iBAT dissected from 8-week old C57BL/6J mice

as earlier described [83]. Contaminating erythrocytes were

eliminated from the stromal-vascular fraction by a wash with

sterile distilled water. Cells were plated and induced to

differentiate as described [83].

Cold acclimation experimentsGroups (n = 5–8) of 10-week old male mice were acclimated at a

temperature of 28–30uC for at least 1 week and transferred to 4uCfor 1, 2, 3 or 6 days. Where relevant, mice were injected with

indomethacin (2.5 mg/kg) 2 h prior transfer to 4uC. The mice

received a dose of indomethacin every 12 h. Injections were

performed subcutaneously from a 0.75 mg/ml solution. Final dose

was 5 mg/kg/day. Control mice received vehicle. Animals were

housed individually with a 12 h light/dark cycle and free access to

pellet food and water. Mice used for immunohistochemical

analyses were immediately perfused intracardially with 4%

paraformaldehyde. iBAT, iWAT, lung and skin were dissected

and frozen for immunohistochemistry on cryosections. For

morphology experiments, the mice were immediately perfused

with 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 for

5 min. COX2 KO mice (B6;129P2 Ptgs2 tm1Unc) and correspond-

ing wild-type littermates were obtained from Taconic. C57BL/6J

used in indomethacin experiments were obtained from Møllegard

breeding laboratories.

16,16dmPGE2 injectionsMale, C57BL/6J approx 10-weeks old from Møllegard

breeding laboratories, Denmark were divided into two groups

(n = 5). The mice received a dose of 50 mg/kg 16,16dmPGE2 every

12 h for 48 h. Injections were performed subcutaneously and the

total dose was 0.1 mg/kg/day. Control mice received vehicle.

Animals were housed individually with a 12 h light/dark cycle and

free access to pellet food and water.

High-fat feedingMale Sv129 mice, 11 weeks old, were obtained from Taconic.

The mice were acclimated for 1 week at a temperature of 28–30uCand divided into three groups (n = 6 in each). One group of mice

was sacrificed before dietary intervention while the remaining

mice were fed a very high-fat diet (ssniff EF R/M acc D12492)

with or without indomethacin supplementation (16 ppm) for 4

weeks at a temperature of 28–30uC. Body weight and feed intake

were recorded twice a week. Mice were anesthetized using

isoflurane, cardiac puncture was performed and mice were killed

by cervical dissociation. Tissues were immediately frozen in liquid

N2.

Real time qPCRTotal RNA was extracted from cultured cells or mouse tissue

using TRIzol (Invitrogen). Reverse transcription and qPCR were



performed in duplicates as described earlier [83]. Primer

sequences are available on request.

Western blottingPreparation of extracts from mouse tissues or whole cell dishes,

electrophoresis, blotting, visualization and stripping of membranes

were performed as described [84]. Primary antibodies used were

goat anti-COX1, goat anti-COX2, rabbit anti-UCP1 and rabbit

anti-TFIIB antibodies (Santa Cruz Biotechnology). Secondary

antibodies were horseradish peroxidase-conjugated anti-mouse,

anti-goat or anti-rabbit antibodies obtained from DAKO.

ImmunohistochemistryCOX1 (M-20; sc 1754) and COX2 (M-19; sc 1747) antibodies

were obtained from Santa Cruz Biotechnology, diluted 1:300 on

cryosections and 1:100 on paraffin-embedded sections (for

COX1). Lung [85] and skin [86] were used as positive control

for both COX1 and COX2 antibodies.

Histological analysesParts of adipose tissue were fixed in 4% formaldehyde in PB

buffer for 24 h, washed in PB, dehydrated in ethanol, embedded

in paraffin after 2610 min xylen treatment. Sections (8 mm thick)

of the embedded tissue sections were subjected to standard

hematoxylin and eosin staining.

Supporting Information

Text S1 Experimental.

Found at: doi:10.1371/journal.pone.0011391.s001 (0.04 MB

DOC)

Figure S1 COX1 and COX2 are mainly expressed in the

stromal-vascular fraction of iBAT in warm-acclimated mice. A.

Sv129 mice were warm-acclimated at 28–30uC for 6 days and

then transferred to 4–6uC. Samples for cryosections were prepared

from iBAT and iWAT after four days of cold exposure.

Representative COX2 immunoblots in iBAT B. iBAT from

warm-acclimated mice was fractionated into SVF and adipocyte

fractions, respectively. Expression levels of COX1 and COX2

were determined by RT-qPCR and Western blotting.

Found at: doi:10.1371/journal.pone.0011391.s002 (15.01 MB

EPS)

Figure S2 Inhibition of COX does not prevent isoproterenol/9-

cis-retinoic acid-induced UCP1 expression in WT-1 cells.

Differentiated WT-1 cells were treated with a combination of

isoproterenol (100 nM) and 9-cis-retinoic acid (1 mM) for 24 h.

When included, indomethacin (1 mM) were added 2 h before

isoproterenol and 9-cis retinoic acid. UCP1 and PGE1aexpressions were determined by RT-qPCR.

Found at: doi:10.1371/journal.pone.0011391.s003 (2.44 MB EPS)

Author Contributions

Conceived and designed the experiments: LM KK. Performed the

experiments: LM LMP HHL EF IB QH RKP PH TM RDM PA JM

MLB JBH BC JN JW SC PV SOD. Analyzed the data: LM LMP HHL EF

RKP PH TM PA JM MLB JBH BC JN JW SC PV SOD KK. Wrote the

paper: LM LMP BC JN SOD KK.

References

1. Cannon B, Nedergaard J (2004) Brown Adipose Tissue: Function and

Physiological Significance. Physiol Rev 84: 277–359.

2. Kopecky J, Hodny Z, Rossmeisl M, Syrovy I, Kozak LP (1996) Reduction of

dietary obesity in aP2-Ucp transgenic mice: physiology and adipose tissue

distribution. Am J Physiol 270: E768–E775.

3. Kopecky J, Rossmeisl M, Hodny Z, Syrovy I, Horakova M, et al. (1996)

Reduction of dietary obesity in aP2-Ucp transgenic mice: mechanism and

adipose tissue morphology. Am J Physiol Endocrinol Metab 270: E776–E786.

4. Cederberg A, Gronning LM, Ahren B, Tasken K, Carlsson P, et al. (2001)

FOXC2 Is a Winged Helix Gene that Counteracts Obesity, Hypertriglyceride-

mia, and Diet-Induced Insulin Resistance. Cell 106: 563–573.

5. Seale P, Kajimura S, Yang W, Chin S, Rohas LM, et al. (2007) Transcriptional

control of brown fat determination by PRDM16. Cell Metab 6: 38–54.

6. Cummings DE, Brandon EP, Planas JV, Motamed K, Idzerda RL, et al. (1996)

Genetically lean mice result from targeted disruption of the RII[beta] subunit of

protein kinase A. Nature 382: 622–626.

7. Nolan MA, Sikorski MA, McKnight GS (2004) The role of uncoupling protein 1

in the metabolism and adiposity of RII beta-protein kinase A-deficient mice. Mol

Endocrinol 18: 2302–2311.

8. Tsukiyama-Kohara K, Poulin F, Kohara M, DeMaria CT, Cheng A, et al.

(2001) Adipose tissue reduction in mice lacking the translational inhibitor 4E-

BP1. Nat Med 7: 1128–1132.

9. Zhou Z, Yon Toh S, Chen Z, Guo K, Peng Ng C, et al. (2003) Cidea-deficient

mice have lean phenotype and are resistant to obesity. Nat Genet 35: 49–56.

10. Picard F, Gehin M, Annicotte JS, Rocchi S, Champy MF, et al. (2002) SRC-1

and TIF2 Control Energy Balance between White and Brown Adipose Tissues.

Cell 111: 931–941.

11. Dali-Youcef N, Mataki C, Coste A, Messaddeq N, Giroud S, et al. (2007)

Adipose tissue-specific inactivation of the retinoblastoma protein protects against

diabesity because of increased energy expenditure. Proc Natl Acad Sci U S A

104: 10703–10708.

12. Mercader J, Ribot J, Murano I, Feddersen S, Cinti S, et al. (2009)

Haploinsufficiency of the retinoblastoma protein gene reduces diet-induced

obesity, insulin resistance and hepatosteatosis in mice. Am J Physiol 297:

E184–E193.

13. Hansen JB, Jorgensen C, Petersen RK, Hallenborg P, De Matteis R, et al. (2004)

Retinoblastoma protein functions as a molecular switch determining white

versus brown adipocyte differentiation. Proc Natl Acad Sci U S A 101:

4112–4117.

14. Cannon B, Nedergaard J (2004) Brown Adipose Tissue: Function and

Physiological Significance. Physiol Rev 84: 277–359.

15. Cypess AM, Lehman S, Williams G, Tal I, Rodman D, et al. (2009)

Identification and Importance of Brown Adipose Tissue in Adult Humans.

N Engl J Med 360: 1509–1517.

16. van Marken Lichtenbelt WD, Vanhommerig JW, Smulders NM, Drossaerts JMAF,

Kemerink GJ, et al. (2009) Cold-Activated Brown Adipose Tissue in Healthy Men.

N Engl J Med 360: 1500–1508.

17. Virtanen KA, Lidell ME, Orava J, Heglind M, Westergren R, et al. (2009)

Functional Brown Adipose Tissue in Healthy Adults. N Engl J Med 360:

1518–1525.

18. Zingaretti MC, Crosta F, Vitali A, Guerrieri M, Frontini A, et al. (2009) The

presence of UCP1 demonstrates that metabolically active adipose tissue in the

neck of adult humans truly represents brown adipose tissue. FASEB J 23:

3113–3120.

19. Saito M, Okamatsu-Ogura Y, Matsushita M, Watanabe K, Yoneshiro T, et al.

(2009) High incidence of metabolically active brown adipose tissue in healthy

adult humans: effects of cold exposure and adiposity. Diabetes 58: 1526–

1531.

20. Garruti G, Ricquier D (1992) Analysis of uncoupling protein and its mRNA in

adipose tissue deposits of adult humans. Int J Obes Relat Metab Disord 16:

383–390.

21. Lean ME, James WP, Jennings G, Trayhurn P (1986) Brown adipose tissue in

patients with phaeochromocytoma. Int J Obes 10: 219–227.

22. Oberkofler H, Dallinger G, Liu YM, Hell E, Krempler F, et al. (1997)

Uncoupling protein gene: quantification of expression levels in adipose tissues of

obese and non-obese humans. J Lipid Res 38: 2125–2133.




24. Timmons JA, Wennmalm K, Larsson O, Walden TB, Lassmann T, et al. (2007)

Myogenic gene expression signature establishes that brown and white adipocytes

originate from distinct cell lineages. PNAS 104: 4401–4406.

25. Seale P, Bjork B, Yang W, Kajimura S, Chin S, et al. (2008) PRDM16 controls a

brown fat/skeletal muscle switch. Nature 454: 961–967.

26. Himms-Hagen J, Melnyk A, Zingaretti MC, Ceresi E, Barbatelli G, et al. (2000)

Multilocular fat cells in WAT of CL-316243-treated rats derive directly from

white adipocytes. Am J Physiol Cell Physiol 279: C670–C681.

27. Granneman JG, Burnazi M, Zhu Z, Schwamb LA (2003) White adipose tissue

contributes to UCP1-independent thermogenesis. Am J Physiol Endocrinol

Metab 285: E1230–E1236.

28. Cinti S (2009) Transdifferentiation properties of adipocytes in the Adipose

Organ. Am J Physiol Endocrinol Metab: 00183.



29. Granneman JG, Li P, Zhu Z, Lu Y (2005) Metabolic and cellular plasticity in

white adipose tissue I: effects of beta3-adrenergic receptor activation.

Am J Physiol Endocrinol Metab 289: E608–E616.

30. Petrovic N, Walden TB, Shabalina IG, Timmons JA, Cannon B, et al. (2009)

Chronic PPARc Activation of Epididymally Derived White Adipocyte Cultures

Reveals a Population of Thermogenically Competent, UCP1-containing

Adipocytes Molecularly Distinct From Classical Brown Adipocytes. J Biol Chem

Dec 22. [Epub ahead of print].


Brown Adipocytes in White Fat in Mice Is Under Genetic Control. Effects on

Body Weight and Adiposity. J Clin Invest 102: 412–420.

32. Watson PM, Commins SP, Beiler RJ, Hatcher HC, Gettys TW (2000)

Differential regulation of leptin expression and function in A/J vs. C57BL/6J

mice during diet-induced obesity. Am J Physiol Endocrinol Metab 279:

E356–E365.







35. Semple RK, Crowley VC, Sewter CP, Laudes M, Christodoulides C, et al.

(2003) Expression of the thermogenic nuclear hormone receptor coactivator

PGC-1[alpha] is reduced in the adipose tissue of morbidly obese subjects.

Int J Obes Relat Metab Disord 28: 176–179.

36. Jimenez M, Barbatelli G, Allevi R, Cinti S, Seydoux J, et al. (2003) b3 -

Adrenoceptor knockout in C57BL/6J mice depresses the occurrence of brown

adipocytes in white. European Journal of Biochemistry 270: 699–705.

37. Grujic D, Susulic VS, Harper ME, Himms-Hagen J, Cunningham BA, et al.

(1997) beta 3-Adrenergic Receptors on White and Brown Adipocytes Mediate

beta 3-Selective Agonist-induced Effects on Energy Expenditure, Insulin

Secretion, and Food Intake. J Biol Chem 272: 17686–17693.

38. Gaskins HD, Hausman DB, Martin RJ, Hausman GJ (1989) Evidence for

abnormal prostaglandin synthesis in obese Zucker rat adipose cell cultures. J Nutr

119: 458–462.

39. Madsen L, Pedersen LM, Liaset B, Ma T, Petersen RK, et al. (2008) cAMP-

dependent Signaling Regulates the Adipogenic Effect of n-6 Polyunsaturated

Fatty Acids. J Biol Chem 283: 7196–7205.

40. Petersen RK, Jorgensen C, Rustan AC, Froyland L, Muller-Decker K, et al.

(2003) Arachidonic acid-dependent inhibition of adipocyte differentiation

requires PKA activity and is associated with sustained expression of

cyclooxygenases. J Lipid Res 44: 2320–2330.

41. Vassaux G, Gaillard D, Darimont C, Ailhaud G, Negrel R (1992) Differential

response of preadipocytes and adipocytes to prostacyclin and prostaglandin E2:

physiological implications. Endocrinology 131: 2393–2398.

42. Cinti S (2009) Transdifferentiation properties of adipocytes in the Adipose

Organ. Am J Physiol Endocrinol Metab: 00183.




4112–4117.




4112–4117.

45. Hyman BT, Stoll LL, Spector AA (1982) Prostaglandin production by 3T3-L1

cells in culture. Biochim Biophys Acta 713: 375–385.




47. Tseng YH, Kokkotou E, Schulz TJ, Huang TL, Winnay JN, et al. (2008) New

role of bone morphogenetic protein 7 in brown adipogenesis and energy

expenditure. Nature 454: 1000–1004.

48. Fain JN, Ballou LR, Bahouth SW (2001) Obesity is induced in mice

heterozygous for cyclooxygenase-2. Prostaglandins Other Lipid Mediat 65:

199–209.

49. Morham SG, Langenbach R, Loftin CD, Tiano HF, Vouloumanos N, et al.

(1995) Prostaglandin synthase 2 gene disruption causes severe renal pathology in

the mouse. Cell 83: 473–482.

50. Ushikubi F, Hirata M, Narumiya S (1995) Molecular biology of prostanoid

receptors: an overview. J Lipin Mediat 12: 343–359.

51. Abramovitz M, Adam M, Boie Y, Carriere MC, Denis D, et al. (2000) The

utilization of recombinant prostanoid receptors to determine the affinities and

selectivities of prostaglandins and related analogs. Biochimica et Biophysica Acta

1483: 285–293.

52. Kunikata T, Tanaka A, Miyazawa T, Kato S, Takeuchi K (2002) 16,16-

Dimethyl Prostaglandin E2 Inhibits Indomethacin-Induced Small Intestinal

Lesions Through EP3 and EP4 Receptors. Digestive Diseases and Sciences 47:

894–904.

53. Koza RA, Hohmann SM, Guerra C, Rossmeisl M, Kozak LP (2000) Synergistic

Gene Interactions Control the Induction of the Mitochondrial Uncoupling

Protein (Ucp1) Gene in White Fat Tissue. J Biol Chem 275: 34486–34492.




55. Collins S, Daniel KW, Petro AE, Surwit RS (1997) Strain-Specific Response to

b3-Adrenergic Receptor Agonist Treatment of Diet-Induced Obesity in Mice.

Endocrinology 138: 405–413.







E356–E365.

58. Feldmann HM, Golozoubova V, Cannon B, Nedergaard J (2009) UCP1

Ablation Induces Obesity and Abolishes Diet-Induced Thermogenesis in Mice

Exempt from Thermal Stress by Living at Thermoneutrality. Cell Metabolism 9:

203–209.

59. Rothwell NJ, Stock MJ (1979) A role for brown adipose tissue in diet-induced

thermogenesis. Nature 281: 31–35.

60. Enerback S, Jacobsson A, Simpson EM, Guerra C, Yamashita H, et al. (1997)

Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese.

Nature 387: 90–94.

61. Feldmann HM, Golozoubova V, Cannon B, Nedergaard J (2009) UCP1

Ablation Induces Obesity and Abolishes Diet-Induced Thermogenesis in Mice

Exempt from Thermal Stress by Living at Thermoneutrality. Cell Metabolism 9:

203–209.




63. Stenbit AE, Tsao TS, Li J, Burcelin R, Geenen DL, et al. (1997) GLUT4

heterozygous knockout mice develop muscle insulin resistance and diabetes. Nat

Med 3: 1096–1101.

64. Hetu PO, Riendeau D (2007) Down-regulation of Microsomal Prostaglandin E2

Synthase-1 in Adipose Tissue by High-fat Feeding. Obesity Res 15: 60–68.

65. Sakuma Y, Tanaka K, Suda M, Komatsu Y, Yasoda A, et al. (2000) Impaired

Bone Resorption by Lipopolysaccharide In Vivo in Mice Deficient in the

Prostaglandin E Receptor EP4 Subtype. Infect Immun 68: 6819–6825.

66. Jimenez M, Barbatelli G, Allevi R, Cinti S, Seydoux J, et al. (2003) b3 -

Adrenoceptor knockout in C57BL/6J mice depresses the occurrence of brown

adipocytes in white. European Journal of Biochemistry 270: 699–705.

67. Leonardsson G, Steel JH, Christian M, Pocock V, Milligan S, et al. (2004)

Nuclear receptor corepressor RIP140 regulates fat accumulation. Proc Natl

Acad Sci U S A 101: 8437–8442.

68. Razani B, Combs TP, Wang XB, Frank PG, Park DS, et al. (2002) Caveolin-1-

deficient mice are lean, resistant to diet-induced obesity, and show hypertri-

glyceridemia with adipocyte abnormalities. J Biol Chem 277: 8635–8647.

69. Toh SY, Gong J, Du G, Li JZ, Yang S, et al. (2008) Up-Regulation of

Mitochondrial Activity and Acquirement of Brown Adipose Tissue-Like

Property in the White Adipose Tissue of Fsp27 Deficient Mice. PLoS ONE 3:

e2890.

70. Strom K, Hansson O, Lucas S, Nevsten P, Fernandez C, et al. (2008)

Attainment of Brown Adipocyte Features in White Adipocytes of Hormone-

Sensitive Lipase Null Mice. PLoS ONE 3: e1793.

71. Narvaez CJ, Matthews D, Broun E, Chan M, Welsh J (2009) Lean Phenotype

and Resistance to Diet-Induced Obesity in Vitamin D Receptor Knockout Mice

Correlates with Induction of Uncoupling Protein-1 in White Adipose Tissue.

Endocrinology 150: 651–661.

72. Cummings DE, Brandon EP, Planas JV, Motamed K, Idzerda RL, et al. (1996)

Genetically lean mice result from targeted disruption of the RII[beta] subunit of

protein kinase A. Nature 382: 622–626.




4112–4117.

74. Cederberg A, Gronning LM, Ahren B, Tasken K, Carlsson P, et al. (2001)

FOXC2 Is a Winged Helix Gene that Counteracts Obesity, Hypertriglyceride-

mia, and Diet-Induced Insulin Resistance. Cell 106: 563–573.













E356–E365.







Retinoblastoma protein functions as a molecular switch determining whiteversus brown adipocyte differentiation. Proc Natl Acad Sci U S A 101:

4112–4117.

81. Hansen JB, Jorgensen C, Petersen RK, Hallenborg P, De Matteis R, et al. (2004)Retinoblastoma protein functions as a molecular switch determining white

versus brown adipocyte differentiation. Proc Natl Acad Sci U S A 101:4112–4117.


dependent Signaling Regulates the Adipogenic Effect of n-6 PolyunsaturatedFatty Acids. J Biol Chem 283: 7196–7205.

83. Madsen L, Petersen RK, Sørensen MB, Jørgensen C, Hallenborg P, et al. (2003)Adipocyte differentiation of 3T3-L1 preadipocytes is dependent on lipoxygenase

activity during the initial stages of the differentiation process. Biochem J 375:

539–549.




4112–4117.

85. Bauer AK, Dwyer-Nield LD, Malkinson AM (2000) High cyclooxygenase 1

(COX-1) and cyclooxygenase 2 (COX-2) contents in mouse lung tumors.

Carcinogenesis 21: 543–550.

86. Muller-Decker K, Hirschner W, Marks F, Furstenberger G (2002) The Effects of

Cyclooxygenase Isozyme Inhibition onIncisional Wound Healing in Mouse

Skin. 119: 1189–1195.



1

Nutritional Regulation of Bile Acid Metabolism is Associated with Improved Pathological Characteristics of the Metabolic Syndrome

Bjørn Liaset1*, Qin Hao2, Henry Jørgensen3, Philip Hallenborg4, Zhen-Yu Du1, Tao Ma2, Hanns-Ulrich Marschall 5, Mogens Kruhøffer6, Ruiqiang Li7, Qibin Li 7, Christian Clement Yde3, Gabriel Criales1, Hanne C. Bertram8, Gunnar Mellgren 9, 10, Erik Snorre Øfjord 11, Erik-Jan Lock1, Marit Espe1, Livar Frøyland 1, Lise Madsen1,2, Karsten Kristiansen2*

1National Institute of Nutrition and Seafood Research, Bergen, Norway, 2Department of Biology, University of Copenhagen, Denmark, 3Department of Animal Health, Welfare and Nutrition, Aarhus University , Denmark, 4Department of Biochemistry and Molecular Biology, University of Southern Denmark, Denmark, 5Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Sweden, 6AROS Applied Biotechnology, Aarhus, Denmark, 7Beijing Genomic Institute, Shenzhen, Peoples Republic of China, 8Department of Food Science, Aarhus University, Denmark, 9Institute of Medicine, University of Bergen, Norway, 10Hormone Laboratory, Haukeland University Hospital, Norway, 11Center for Clinical Trials, Bergen, Norway.

*Correspondence to: Bjørn Liaset, National Institute of Nutrition and Seafood Research, P.O. Box 2029 Nordnes, N-5817 Bergen, Norway. Fax: +47 55 90 52 99; Phone: +47 46 81 12 97; E-mail: [email protected] or Karsten Kristiansen, Department of Biology, University of Copenhagen, Ole Maaløes Vej 5, DK-2200 Copenhagen, Denmark. Fax +45 3522 2128; Phone +45 6011 2408; E mail: [email protected]

Running title: Bile acid metabolism is regulated by dietary protein source

Bile acids (BAs) are powerful regulators of metabolism, and mice treated orally with cholic acid are protected from diet-induced obesity, hepatic lipid accumulation, and increased plasma triacyl glycerol (TAG) and glucose levels. Here, we show that plasma BA concentration in rats was elevated by exchanging the dietary protein source from casein to salmon protein hydrolysate (SPH). Importantly, the SPH-treated rats were resistant to diet-induced obesity. SPH-treated rats had reduced fed state plasma glucose and TAG levels, and lower TAG in liver. The elevated plasma BA concentration was associated with induction of genes involved in energy metabolism and uncoupling, Dio2, Pgc-1αααα and Ucp1, in interscapular brown adipose tissue. Interestingly, the same transcriptional pattern was found in white adipose tissue depots of both abdominal and

subcutaneous origin. Accordingly, rats fed SPH-based diet exhibited increased whole-body energy expenditure and heat dissipation. In skeletal muscle, expressions of the PPARββββ/δδδδ target genes (Cpt-1b, Angptl4, Adrp, and Ucp3) were induced. Pharmacological removal of BAs by inclusion of 0.5wt% cholestyramine to the high-fat SPH diet attenuated the reduction in abdominal obesity, the reduction in liver TAG and the decrease in non-fasted plasma TAG and glucose levels. Induction of Ucp3 gene expression in muscle by SPH treatment was completely abolished by cholestyramine inclusion. Taken together, our data provide evidence that bile acid metabolism can be modulated by diet, and that such modulation may prevent/ameliorate characteristic features of the metabolic syndrome.

http://www.jbc.org/cgi/doi/10.1074/jbc.M111.234732The latest version is at JBC Papers in Press. Published on June 16, 2011 as Manuscript M111.234732

Copyright 2011 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest, on July 7, 2011w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/cgi/doi/10.1074/jbc.M111.234732

http://www.jbc.org/

2

Bile acids (BAs) are synthesized in the liver from cholesterol. After their synthesis, they are conjugated to the amino acids taurine or glycine in a species-dependent manner (1). Conjugation of bile acids increases their solubility and facilitates their secretion into bile (2). In the intestine, the bile acids aid in the absorption of lipophilic nutrients (3). Normally, the bile acids are efficiently taken up by the enterocytes of the small intestine, transported back to the liver, and are thus conserved through the enterohepatic circulation (4). Bile acids are activators of the nuclear receptor farnesoid X receptor α (Fxr/Nr1h4) (5-7). The main physiological role of Fxr is to maintain bile acid homeostasis and to regulate genes encoding enzymes involved in bile acid synthesis and transport. In addition to bile acid homeostasis, Fxr is important for normal lipid metabolism. Fxr-/- mice display higher triacylglycerol (TAG) concentrations in serum and liver (8,9) and have an increased synthesis of apolipoprotein (Apo) B-containing lipoproteins (9). In line with these findings, bile acid treatment has been reported, in a Fxr-dependent manner, to induce liver very-low-density lipoprotein receptor (Vldlr) transcription (10) and to prevent hepatic TAG accumulation, VLDL secretion and elevated serum TAG concentration in mice (11). Thus, bile acids and Fxr play central roles in VLDL lipoprotein metabolism and in the control of TAG levels.

Glucose metabolism is also regulated by bile acids and mice treated with cholic acid are protected from diet-induced hyperglycemia (12) and have down-regulated liver expression of the gluconeogenic phosphoenolpyruvate carboxykinase (Pck1) gene (13,14). Furthermore, Fxr-/- mice are glucose intolerant and insulin resistant, and insulin signaling is blunted in several tissues (15-17). Activation of Fxr with a synthetic ligand, GW4064, or hepatic over-expression of constitutively active Fxr lowered blood glucose in both diabetic db/db and wild-type mice (15). It is therefore evident that bile acid signaling and Fxr function are essential for normal glucose regulation.

Bile acids are also endogenous activators of the G-protein coupled receptor Tgr5 (also referred to as GB37/ M-BAR/ Gpbar1) (18,19). Stimulation of Tgr5 by bile

acids has been shown to increase energy expenditure and to protect mice from diet-induced obesity and glucose intolerance (20). Consistent with the role of Tgr5 in the control of energy expenditure, female Tgr5-/- mice show increased adiposity when challenged with a high fat diet (21). Also, Tgr5 activation by bile acids has been reported to promote production and release of glucagon-like peptide 1 (GLP-1) in enteroendocrine cell lines (22) and in mice (23), adding another aspect of bile acid treatment for the prevention of diet-induced glucose intolerance and insulin resistance. Thus, bile acid signaling through Tgr5 might be important for the maintenance of normal energy homeostasis and insulin sensitivity.

The benefits of activating Tgr5 or Fxr for the prevention of the metabolic syndrome have stimulated the development of synthetic ligands for these receptors (24,25). Another strategy to increase Tgr5 and/or Fxr signaling would be to alter endogenous BA metabolism. As hepatic bile acid conjugation is important for secretion of BAs into bile (2), and rats conjugate BAs to both taurine and glycine with high efficiency (26), the dietary levels of these amino acids might be crucial for BA conjugation and secretion (27). Feeding rats casein-based diets supplemented with either glycine or taurine led to reduced plasma cholesterol and liver TAG concentrations, whereas only taurine supplementation reduced plasma TAG levels (28). Furthermore, taurine administration has been reported to induce hepatic protein expression of the canalicular transporter proteins ATP-binding-cassette b11, Abcb11 (also called Bsep) and Abcc2 (also called Mrp2). Concomitantly, bile flow and taurocholate excretion was induced in experiments with perfused rat livers (29). In rats fed low-fat diets, we have previously shown that BA metabolism can be modulated by dietary proteins with different endogenous glycine and taurine contents (30).

Thus, evidence is accumulating that BA metabolism can be modulated by dietary levels of glycine and taurine in normal-energy diets. However, development of the metabolic syndrome is tightly associated with intake of energy-dense diets. Therefore, the present study was undertaken in order to test the hypothesisis that rats treated with a taurine- and glycine-rich protein source, salmon protein


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

3

hydrolysate (SPH), would be protected from developing high-fat diet-induced pathological characteristics of the metabolic syndrome. We demonstrate that bile acid metabolism can be modulated by the protein source in high-fat fed rats, and that such a modulation may protect against development of the metabolic syndrome.

EXPERIMENTAL PROCEDURES

Animals - Male Wistar Hannover GALAS (HanTac:WH) were obtained from Taconic Europe (Ejby, Denmark) and divided into experimental groups (n=6). The rats were kept at a 12-h light/ dark cycle in a temperature-controlled room at 22 °C. After acclimatization, the animals were fed experimental diets with either SPH or casein as the sole protein source (Supplemental Table 1). The composition of the protein sources has been described elsewhere (31). Feed intake and body weight were recorded throughout the experiments and faeces were collected the last 5 days. The rats were killed by cardiac puncture under anaesthesia (0.23 mg/ kg BW Fentanyl (Janssen) and 0.45 mg/ kg BW Dormitor Vet (Orion Pharma)). Heparin-plasma and EDTA-plasma containing aprotinin were prepared from blood. Tissues were dissected out and weighted. A portion of the liver was used for sub-cellular fractionation and measurement of mitochondrial carnitine palmitoyl transferase-1 capacity, and portions of interscapular brown adipose tissue (iBAT) and epididymal white adipose tissue (eWAT) were homogenized and used for determination of palmitoyl-Coenzyme A oxidation capacity. The rest of the tissues were freeze-clamped and frozen at -80oC. All animal experiments were approved by the National State Board of Biological Experiments with Living Animals (Norway and Denmark).

Whole body composition – After 40 days of feeding, rats were anaesthetized (0.4 mg Dormitor Vet (Medetomidin Hydrochlorid)/ kg BW rat (Orion Pharma, Espoo, Finland), and whole body composition was determined by a dual X-ray absorptiometry scanner equipped with a small animal option (Discovery QDR Series, Hologic, Bedford, MA, USA).

Indirect calorimetric measurements – A separate set of rats (n=6/ group) was fed experimental diets for 17 days. Heat production was calculated from gas exchange measurements, as previously described (32). Gas exchange was determined twice, each time for 22 hours. From the average gas exchange measurements, heat production was calculated by the respiratory quotient method, and reported per 24 h.

Plasma metabolomics - A separate set of rats (n=5/ group) was fed experimental diets for 25 days. After termination in the fed state, 200 µl heparin plasma of each sample was mixed with a solution of 400 µl 0.9% saline and 20% D2O. Measurements were performed at 310 K on a on a Bruker Avance III 600 spectrometer, operating at a 1H frequency of 600.13 MHz, and equipped with a 5-mm 1H TXI probe (Bruker BioSpin, Rheinstetten, Germany). 1H NMR spectra were acquired using the Carr-Purcell-Meiboom-Gill (CPMG) spin-echo pulse sequence with water suppression. All spectra were referenced to the lactate doublet signal at 1.33 ppm. The spectra were segmented into 0.013 ppm bins and each of the bins was integrated. The reduced spectra excluding the residual water signal were normalized to the whole spectrum. Principal component analysis (PCA) and partial least squares regression discriminate analysis (PLS-DA) was performed using the Unscrambler software version 9.8 (Camo, Oslo, Norway) to elucidate biochemical differences between pre-defined classes. Martens' uncertainty test was applied to find significant variables on the full cross-validated data (33,34).

Bile acid measurements - For total bile acid measurements in feces, bile acids were extracted according to Suckling et al (35). Amounts of total bile acids in fecal extracts and in non-fasted EDTA-plasma were determined enzymatically by the 3-α-hydroxysteroid dehydrogenase reaction (Dialab, Vienna, Austria). Bile acids in liver samples were extracted and analyzed using gas-chromatography-mass spectrometry and electrospray-mass spectrometry as previously described in detail (36).

Real time RT-qPCR - Total RNA was purified tissues using Trizol, and cDNA was


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

4

synthesized from individual rats. Gene expression was determined in individual samples by real-time qPCR using ABI PRISM 7700 Sequence Detection System (Applied Biosystems) carried out in 96-well plates and in duplicate as earlier described (37). Primers for real-time PCR (Supplemental Table 2) were designed using Primer Express 2.0 (Applied Biosystems).

Liver microarray analysis- The following procedures were all performed according to Affymetrix standard procedures. Briefly, equal amounts of RNA isolated from livers were pooled (n=6) and 5 µg of total RNA was used as starting material for the target preparation. First and second strand cDNA synthesis were performed using the SuperScript II System (Invitrogen) according to the manufacturers’ instructions except using an oligo-dT primer containing a T7 RNA polymerase promoter site. Labeled aRNA was prepared using the BioArray High Yield RNA Transcript Labeling Kit (Enzo) using Biotin labeled CTP and UTP (Enzo) in the reaction together with unlabeled NTP´s. Unincorporated nucleotides were removed using RNeasy columns (Qiagen). Fifteen µg of cRNA were fragmented, loading onto the Affymetrix Rat Genome RAE 230 2.0 probe array cartridge and hybridized for 16h. The arrays were washed and stained in the Affymetrix Fluidics Station and scanned using a confocal laser-scanning microscope (GeneChip® Scanner 3000 System with Workstation and AutoLoader). The raw images files from the quantitative scanning were analyzed by the Affymetrix Gene Expression Analysis Software (MAS 5.0) resulting in cell files containing background corrected probe values.

Liver microarray KEGG pathway analysis – Liver microarray data was used to identify metabolic pathways regulated by treatments, using the KEGG resource (38,39), release 43. Genes with at least two-fold change in expression level between the two treatment groups were taken as differentially expressed genes (DEGs). The enrichment of these DEGs among KEGG pathways was measured by two-tailed Fisher’s exact test. The pathways with P-value less than 0.05 were considered as statistically significant and were further

studied.

Histological analyses – Lipids in cryosections of frozen liver samples were stained by the standard Oil Red-O method. Parts of the epididymal white adipose tissue (eWAT) were subjected to standard hematoxylin and eosin staining as previously described (40). Plasma measurements - Lipids, metabolites and enzyme activities in plasma were determined by commercially available enzymatic kits: alanine aminotransferase (ALAT), total cholesterol, HDL-cholesterol, LDL-cholesterol (Dialab, Vienna, Austria), glucose, TAG (MaxMat, Montpellier, France), OH-butyrate (Randox, Crumlin, United Kingdom). Hormones were measured in EDTA plasma containing aprotonin: insulin by an ELISA kit (DRG Diagnostics, Germany) and glucagon by a Radioimmunoassay (RIA) Kit (LINCO Research, ST. Charles, MO, USA).

Liver glycogen content – Glycogen in liver samples was converted to glucose by treatment with amyloglucosidase as described elsewhere (41). Glucose content after the treatment was determined using a commercial glucoses quantification kit (Dialab, Vienna, Austria).

Liver lipid analyses - Total lipids were extracted from liver samples with chloroform: methanol, 2:1 (v/v). The lipid classes (TAG, cholesterol and sterol esters) were analyzed on an automated Camaq HPTLC system and separated on HPTLC silica gel 60 F plates as previously described (42).

Liver mitochondrial preparation and CPT I assay – Liver mitochondria-enriched fractions were prepared as previously described (30). Freshly prepared mitochondrial enriched fractions were used for determination of carnitine-palmitoyl transferase capacity, as previously described (43), with or without addition of the CPTI inhibitor malonyl-CoA (5µM).

Liver cytosolic preparation and total glutathione determination – After removal of the mitochondrial enriched fraction, the homogenates were further centrifuged at 100 000gav for 90 minutes at 4oC. The supernatant collected after this centrifugation


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

5

was used as the liver cytosolic fraction and total glutathione (GSH) content was determined by a colorimetric assay, Bioxytech GSH-400 (OXISResarchTM, Portland, OR, USA).

Adipose tissue fatty acid oxidation capacity – samples from eWAT and iBAT was homogenized in a glycine-glycine buffer as previously described (44). The homogenates were centrifuged at 1000 g for 5 minutes and the supernatant was collected (post-nuclear fraction). Fatty acid oxidation capacity was measured in the post-nuclear fractions by the acid-soluble product method using radio-labeled palmitoyl-Coenzyme A as the substrate, as previously described in detail (30).

In vitro adipocyte differentiation and fatty acid oxidation – The stromal vascular fraction was isolated from eWAT dissected out from rats as earlier described (37). Contaminating erythrocytes were eliminated from the stromal-vascular fraction by a wash with sterile distilled water. Cells were plated and induced to differentiate as described (37). Differentiated cells were treated with increasing concentrations of taurocholic acid in the medium for 24h. Afterwards, the cells were scraped and homogenized in a buffer containing 0.25 M sucrose, 2 mM EGTA and 10 mM Tris/HCl, pH 7.4. Fatty acid oxidation capacity was evaluated by the total amount of acid-soluble products using radio-labeled palmitoyl-Coenzyme A as the substrate, as previously described in detail (30).

Energy in faeces and diets - The gross energy content was determined in a bomb calorimeter following the manufacturer’s instruction (Parr Instruments, Moline, IL, USA).

Statistical analysis – Data are presented as mean+S.E. Analysis of variance was performed by post-hoc pairwise comparison Student’s t-test or Tukey’s HSD test. Data that failed to show homogeneity in variance by Levenes’ test were tested by the non-parametric tests Mann-Whitney U test or Kruskal-Wallis test. Data were considered statistically different at P<0.05.

RESULTS

Nutritional regulation of endogenous BA metabolism alters energy expenditure and diet-induced adiposity. To demonstrate that endogenous BA metabolism could be altered by dietary protein source, rats were fed SPH- or casein-based high-fat diets for 46 days. As expected, and in agreement with our previous findings (30), plasma BA concentrations were higher in SPH treated rats, whereas total liver BA amounts were not statistically different (Fig. 1A, B). The SPH treated rats had significantly higher liver α-muricholic acid (a-MCA), and tended to have lower cholic acid, but higher β-MCA levels (both P=0.06), relative to the casein-fed rats (Fig. 1B). Body weight gain was reduced and at the end of the experiment mean body weight was 110 g lower in SPH-treated rats (Fig 1C). The accumulated energy intake was slightly but significant lower (-11%, P<0.01) in the SPH-treated rats. Still, after correcting for the difference in energy intake, energy efficiency was reduced by 44% (Fig 1D). Rats fed the SPH high-fat diet exhibited reduced white adipose depot weights, epididymal (eWAT, -56%) and mesenteric (MeWAT, -46%), while no difference was observed in interscapular brown adipose tissue (iBAT) (Fig 1E). No difference was observed in apparent energy digestibility (Fig. 1F). Thus, our data supported the hypothesis that endogenous BA levels could be altered by diet, and that this might be important for the observed reduced adiposity. To further test this hypothesis, we chose to pharmacologically remove BAs from circulation. Thus, we fed a new set of rats the experimental diets, including a group that received a small amount, 0.5 weight%, of the BA-binding resin cholestyramine (c’am) added to the SPH diet. In order to secure equal energy intake, the rats were pair-fed. As predicted, inclusion of cholestyramine successfully lowered plasma BA concentrations in the SPH-based diet, whereas liver BA levels remained unaltered (Fig 1G, H). Quantification of whole-body fat content by dual energy X-ray absorptiometry (DEXA) revealed that the reduced plasma BA concentration was associated with increased body fat content (Fig 1J). Despite lower plasma BA concentrations, and increased adiposity, no significant difference in body weight gain was found between the SPH and


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

6

the SPH+c’am treated groups (Fig 1I). However, the lean body mass including bones was borderline reduced in the SPH-treated (231±4 g, P = 0.052) but was significantly reduced in the SPH+cholestyramine-treated (223±1 g, P = 0.005), relative to the casein-fed rats (259±7 g). Since bile acids are known to increase energy expenditure and energy dissipation in the form of heat (20), we considered it likely that the reduced fat content in the SPH-treated rats was accompanied by higher energy dissipation. To demonstrate this we determined energy expenditure in the rats by indirect calorimetry. Indeed, SPH-treated rats had significantly higher O2 consumption and heat production, as well as CO2 elimination relative to the casein treated rats (Fig 1K-M). Importantly, and providing further evidence for a role of the increased plasma BA concentration with regard to the increased energy expenditure, pharmacological lowering of plasma BAs attenuated heat production, and led to higher body fat content at an equal energy intake (Fig 1J-M). The higher CO2 production in the SPH-treated rats was independent of the reduced bodyweight in these animals, as total CO2 production (Liter/24h) was higher in the SPH, relative to the casein fed rats, 8.69 ± 0.15 and 8.09 ± 0.11, respectively (P = 0.03). The corresponding value for the SPH+c’am treated rats was 8.31 ± 0.18, which was not statistically different from neither the SPH nor the casein treated rats.

Nutritional regulation of endogenous

BA metabolism modulates fat oxidation capacity in brown and white adipose tissues. Mice ingesting cholic acid supplemented diets have increased fat oxidation and energy dissipation as heat in iBAT, which is important to prevent high-fat diet-induced adiposity (20). To elucidate whether nutritional modulation of plasma BAs could induce fat oxidation in iBAT, we measured ex vivo fatty acid oxidation capacity in iBAT. In agreement with the increased whole-body heat production, fatty acid oxidation capacity was significantly higher in iBAT from SPH fed rats compared with rats fed casein (Fig 2A). The metabolic effect of bile acids on energy expenditure is critically dependent on the cAMP-inducible thyroid activating enzyme type 2

iodothyronine deiodinase (Dio2) and is lost in Dio2-/- mice (20). Expression of Dio2 has been linked to expression of the thyroid-responsive gene, uncoupling protein 1 (Ucp1) (45). Exogenously added bile acids have also been shown to increase the expression of peroxisome proliferator-activated receptor γ coactivator 1α (Pgc-1α) in brown adipose tissue (20), a key regulator of mitochondrial biogenesis and energy expenditure (46,47). Here we show that the higher heat production and fatty acid oxidation capacity in iBAT was accompanied by induction of genes encoding Ucp1 and Dio2 whereas Pgc-1α expression was not significantly increased (Fig. 2B-D). Even though brown adipose tissue is considered the major adipose tissue to dissipate chemical energy in the form of heat (48), white adipose tissue is plastic and under certain conditions the expression of Pgc-1α and Ucp1 can be induced (49-51). Under conditions with increased intracellular cAMP signaling, studies in mice suggest that induced expression of Ucp1 in white adipose tissues is associated with a lean and healthy phenotype (40,52-54). As BAs are known to increase intracellular cAMP levels, we speculated whether the increased plasma BA concentrations could also alter the phenotype of WAT. Indeed, the SPH-treated rats displayed higher expressions of Ucp1, Dio2 and Pgc-1α in white adipose tissues of subcutaneous origin (inguinal, iWAT) and of abdominal origin, eWAT and MeWAT (Fig. 2E-G). These data supported the idea that not only iBAT, but also white adipose tissues contributed to the lean phenotype in the SPH-treated rats. To further corroborate the hypothesis that nutritional regulation of plasma BA levels was of importance for the WAT phenotype, we investigated WAT from the rats treated with SPH and SPH+c’am. Pharmacological removal of BAs from circulation attenuated the reduction in abdominal WAT mass (eWAT + MeWAT + perirenal/retroperitoneal WAT) (Fig. 2H). Furthermore, ex vivo fat oxidation capacity and expression of Ucp1, Dio2 and Pgc-1α were lower in eWAT from SPH+c’am, relative to the SPH treated rats, even though the difference did not reach significant levels (Fig. 2I, J). The lower plasma BA concentrations were also accompanied by larger adipocyte cell sizes in eWAT (Fig. 2K).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

7

In order to demonstrate that bile acids could induce fat oxidation in white adipocytes, we isolated preadipocytes from the eWAT depot, and treated differentiated adipocytes with increasing concentrations of taurocholic acid for 24hrs. Compared to untreated cells, a physiologic relevant dose of 25 µM taurocholic acid significantly raised fat oxidation capacity in the adipocytes (Fig 2L). Nutritional regulation of endogenous BA metabolism alters TAG concentrations in liver and plasma. Oral bile acid treatment lowers TAG concentrations in mouse liver (11) and blood (11,55). Moreover, BA treatment has been reported to down-regulate fatty acid de novo synthesis via repression of Sterol regulatory-element binding protein 1c (Srebp-1c) and its downstream lipogenic target genes in mouse primary hepatocytes and liver (11,56). We found no significant differences in hepatic gene-expressions of Srebp-1c (24±4 vs 34±4, P=0.18), Acetyl-CoA carboxylase 1 (Acc1) (36±2 vs 43±2, P=0.10), fatty acid synthase (Fas) (3.8±0.8 vs 5.7±1.3, P=0.10) between the high-fat SPH and high-fat casein treated rats, respectively. Others have reported that in HepG2 cells, BA treatment represses transcription of microsomal triglyceride transfer protein (MTP) and APO B, both important for hepatic secretion of VLDL (57). However, the high-fat SPH-treated rats had higher liver expression of ApoB (6.3±0.6 vs 4.1±0.2, P=0.009) and Mtp (6.6±0.5 vs 5.0±0.2, P=0.14), compared to high-fat casein treated rats, respectively. Our results obtained using rat therefore deviate from the reported findings in mice and HepG2 cells. To gain further information on the mechanisms by which nutritional regulation of BA metabolism modulates hepatic TAG metabolism we used a microarray approach. To identify functional differences between treatment groups, we performed a Kegg pathway analysis. Six Kegg metabolic pathways were significantly altered, and four tended (P<0.1) to be altered (Fig. 3A). SPH treated rats had induced liver expression of genes encoding class I proteins of the rat major histocompatibility complex, such as RT-1-Aw2, RT1-A3, RT1-T24-1, RT1-CE2, RT1-CE15 and RT1-CE10. These genes were annotated to the Kegg pathways Type I diabetes mellitus, Cell adhesion molecules and

Antigen processing and presentation. Transcription of these genes might be involved in the immune response, but their functions are little described in the literature. Interestingly, high-fat SPH treatment also altered the hepatic PPAR signaling pathway, indicating that these rats had increased hepatic fat metabolism (Fig. 3A). Indeed, our microarray data further supported the hypothesis that hepatic fatty acid uptake, binding, activation and oxidation were induced in rats given the high-fat SPH diet (Fig. 3B). Cross-talk between bile acid metabolism and PPAR α signaling appears to take place in a species-dependent manner. Incubation of human hepatoma HepG2 cells with chenodeoxycholic acid (CDCA), significantly induced PPARα transcription, whereas liver Pparα was not regulated in taurocholic acid-fed mice (58). However, despite no regulation of Pparα itself at the transcriptional level, down-regulation of Ppar α target genes was observed in cholic acid-fed mice, a finding partly explained by inhibition of Ppar α co-activator recruitment by bile acids (59). To determine whether nutritional regulation of BA metabolism also affected Ppar signaling in rats, as suggested by our microarray analysis, we measured liver gene expression by RT-qPCR. In contrast to previous reports from bile acid-fed mice (58,59), we observed a down-regulation of liver Pparα in the SPH-treated rats (Fig. 4A). Concomitantly, liver acyl-CoA oxidase 1 (Acox1) expression was reduced, whereas carnitine palmitoyl-transferase 1a (Cpt-1a) mRNA level was unaltered (Fig. 4B). Interestingly, mRNA levels of both Pparβ/δ and its downstream target gene Adipocyte differentiation–related protein (Adrp) were robustly increased in the livers of the SPH treated rats (Fig. 4C, D). As activation of PPARβ/δ increases fat catabolism and energy uncoupling and thereby prevents lipid accumulation (60), we further investigated hepatic lipid catabolism. Despite unaltered Cpt-1a mRNA levels, the SPH-treated rats had significantly higher liver mitochondrial CPT-1 capacity (Fig. 4E). CPT-1 activity is regulated by malonyl-CoA. Of note, the expression of Acc2, which encodes the enzyme catalyzing the formation of the CPT-1-inhibitor malonyl-CoA, was down-regulated (Fig. 4F).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

8

Accordingly, when we measured mitochondrial CPT-1 capacity in the presence of 5 µM malonyl-CoA, we observed a stronger inhibition in SPH-treated rats than in casein treated rats, even though the difference not was significant (Fig. 4G). Furthermore, the fasting plasma levels of the ketone-body β-hydroxybutyrate, a marker of liver mitochondrial fatty acid oxidation was significantly elevated (Fig. 4H) and liver TAG deposition was lower in the SPH-treated rats (Fig. 4I, J), further supporting that the SPH-treated rats had higher liver fat catabolism. Exogenous bile acid supplementation (11), as well as the activation of the PPAR β/δ receptor (61) prevents increased plasma TAG through inhibition of hepatic VLDL secretion. Higher liver Vldlr expression might be one underlying mechanism to the lower plasma VLDL amounts, as both BA-treatment (10) and expression and activation of the PPARβ/δ receptor (62) might induce Vldlr transcription. As the rats given the high-fat SPH diet had both elevated plasma BA concentrations and induction of Pparβ/δ gene expression in liver, we hypothesized that plasma TAG would be decreased in these animals. As predicted, the plasma TAG concentrations were reduced in the non-fasted state (Fig. 4K), and this decrease was accompanied by a pronounced elevation in liver Vldlr expression (Fig. 4L). To verify that plasma VLDL concentration actually was reduced, we analyzed plasma samples by nuclear magnetic resonance (NMR), which confirmed that SPH-treated rats had decreased VLDL amounts (Fig. 4M). The NMR analysis also suggested that plasma LDL- or HDL-cholesterol concentrations were elevated in the SPH-treated rats, but no significant differences were observed in plasma cholesterol levels by standard clinical chemistry methods (Fig. N, O). Our data supported the hypothesis that nutritional regulation of endogenous BA metabolism could modulate hepatic fatty acid oxidation capacity, liver TAG storage and plasma non-fasted TAG-rich VLDL concentrations. To investigate further the potential role of BAs in relation to the observed effects on TAG metabolism, we examined the rats treated with SPH+cholestyramine. Indeed, pharmacological

removal of BAs by the cholestyramine treatment attenuated the reduction in liver and non-fasted plasma TAG concentrations (Fig. 4P, Q), and attenuated the elevation in liver mitochondrial fat oxidation capacity and non-fasted plasma β-hydroxybutyrate (Fig. 4R, S). We conclude that nutritional regulation of endogenous BA metabolism contributes significantly to the observed modulation of lipid metabolism introduced by SPH treatment. Nutritional regulation of endogenous BA metabolism modulates skeletal muscle Ucp3 expression. PPARβ/δ is abundantly expressed in skeletal muscle, a peripheral tissue that accounts for approximately 40% of total body mass. Activation of PPARβ/δ leads to increased expression of genes involved in lipid catabolism and energy uncoupling in skeletal muscle (63-65). One of the genes robustly induced by PPARβ/δ activation is Ucp3, and mice over-expressing the human UCP3 in skeletal muscle are protected from diet-induced adiposity (66). In keeping with the strong hepatic Pparβ/δ induction combined with the lean phenotype of the SPH treated rats, we hypothesized that PPARβ/δ signaling was altered also in skeletal muscle of the SPH treated rats. As in liver, skeletal muscle expressions of Pparα and its target genes Acox1 were down-regulated in the SPH treated rats (Fig. 5A). In contrast, expression of Pparβ/δ , and its downstream target genes Adrp and angiopoietin-like 4 (Angptl4, also called fasting-induced adipose factor, Fiaf) were significantly induced (Fig. 5B). Concomitantly, Cpt-1b, Ucp2 and in particular Ucp3 expressions were elevated (Fig. 5C). To further corroborate a regulatory role for BAs on skeletal muscle Ucp3 expression, we measured the Ucp3 expression in the SPH+c’am treated rats. Of note, skeletal muscle Ucp3 induction by SPH treatment was completely abolished by reducing plasma BA concentrations with cholestyramine (Fig. 5D). Our data indicate that besides the known regulatory effects of BAs on liver and adipose tissue metabolism, skeletal muscle fatty acid oxidation and uncoupling may also be regulated through altered bile acid metabolism.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

9

Nutritional regulation of endogenous BA metabolism modulates high-fat diet-induced hyperglycemia. Oral BA treatment reduces hyperglycemia in mice (12), possibly through reduced hepatic glucose output by repression of the Pck1 gene expression (13,14). In addition, PPARβ/δ activation reduces hepatic glucose output and improves peripheral glucose disposal (67). In the present study, the SPH-treated rats both had elevated plasma BA levels and induction of Pparβ/δ target gene expression in liver and skeletal muscle. Therefore, we decided to examine whether glucose metabolism was improved in these animals. The non-fasted plasma insulin and liver Pck1 expression were lower in the SPH-treated rats (Fig. 6A, B). Hepatic glycogen concentrations were significantly higher (Fig. 6 C), whereas non-fasted glucose was significantly lower in the SPH-treated rats (Fig. 6D). No difference was observed in fasting plasma glucose levels (Fig. 6D). The lower fed-state plasma insulin and glucose concentrations, as well as the reduced hepatic Pck-1 expression in the SPH fed rats might indicate that these animals were more insulin-sensitive, relative to those fed casein. This notion is in agreement with a recent report in which rats fed a high-fat SPH-diet had increased whole-body insulin-sensitivity determined by the hyperinsulinemic-euglycemic clamp technique (68). In the present study, pharmacological removal of BAs attenuated the reduction in non-fasted blood glucose (Fig. 6E). Therefore, our data support the hypothesis that nutritional modulation of endogenous BA metabolism may regulate blood glucose concentrations.

Possible determinants of the elevated plasma BA levels in the SPH treated rats. In the present study, the SPH treated rats had elevated plasma BA levels. Increased circulating BAs might be due to cholestasis or increased bile acid synthesis. The plasma levels of alanine aminotransferase, although slightly elevated, did not indicate hepatocellular damage in the SPH fed rats (Fig. 7A). Furthermore, liver Ucp2 gene expression, recently shown to be up-regulated in bile duct obstructed rats (69), was borderline (P=0.053) down-regulated in the SPH fed rats (Fig. 7B). Hence, it is unlikely that the elevated plasma BA level was due to cholestasis in the present

study. Supplementing taurine (70,71) or glycine (70) at doses of 50g/ kg diet to casein-based, atherogenic diets (containing 1wt% cholesterol and 0.25 wt% cholic acid) have previously been shown to increase fecal BA excretion in rats. Furthermore, the rats treated with the taurine-supplemented diet had higher liver GSH content, decreased liver cholesterol concentration and higher cholesterol-7a hydroxylase (Cyp7a1) mRNA levels and enzyme activity (71). In the present study, the SPH diet provided taurine (1.9g/kg diet), whereas the casein diet was free of taurine. Also, the glycine level differed and was 23 and 4 g/ kg diet in the SPH and casein diets, respectively (Supplemental Fig. 1). The SPH-treated rats had significantly reduced liver cholesterol and sterol ester concentrations (Fig. 7C). However, the SPH fed rats had lower liver expression of genes involved in the de novo bile acid synthesis, Cyp7a1, and in bile acid conjugation, Bile acid Coenzyme A: amino acid N-acyltransferase, (Baat) (Fig. 7D). Therefore, our data does not support that a higher BA synthesis was the underlying factor for the elevated plasma BA concentrations. Secretion of bile acids into bile is partly dependent on bile acid availability and transport, and partly determined by the availability and transport of phospholipids, cholesterol, glutathione and bicarbonate, as reviewed in (72). In the present study, the liver expression of genes involved in canalicular bile flow generation was either increased (Atp-binding cassette b4, Abcb4, also called Mdr2), tended (P=0.09) to be increased (Abcb11/ Bsep), or was statistically unaltered (Abcb1b/Mdr1 and Abcc2/ Mrp2) in the SPH-treated rats (Fig. 7E). Furthermore, the liver gene expression of Glutathione synthase (Gss), as well as the liver concentration of glutathione (GSH), a primary osmotic driving force in hepatic bile formation (73), was increased by SPH treatment (Fig. 7F, G). However, fecal BA excretion measured over 5 days did not differ between the treatments (Fig. 7H). Thus, our liver data indicate that the SPH fed rats had increased biliary BA secretion, yet the fecal BA excretion was unaltered. If this was the case, the SPH-fed rats must have had an efficient intestinal BA re-uptake. In humans with a normal hepatic function, the major determinant of circulating bile acid


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

10

concentration is their rate of intestinal absorption (74). In the present study, the hepatic amount of α-muricholic acid, a bile acid of intestinal origin (75) was significantly higher in the SPH-fed rats (Fig. 1B). Based on the higher hepatic presence of intestinal-derived BA species, as well as the fact that the SPH + cholestyramine treated rats exhibited reduced plasma BA levels, we conclude that the increased plasma BA level in the SPH-fed rats was likely due to higher intestinal influx.

DISCUSSION

Bile acids strongly affect metabolism and energy-expenditure in mice (11,12,20). From studies in rats, it is known that supplementing low-fat, casein-based diets with rather high doses of glycine and/or taurine affects bile acid metabolism (29,70,71). Previously, we have reported that exchanging casein with a glycine and taurine rich protein-source modulated BA metabolism in low-fat fed rats (30). Here we show that choice of dietary protein source is sufficient to increase plasma BA levels also in high-fat fed rats. Importantly, the elevated plasma BA level was accompanied with attenuated diet-induced obesity and ameliorated characteristics of the metabolic syndrome. Little is known about the role(s) that endogenous BA metabolism may play in the development of the metabolic syndrome. However, a few recent publications support such a role for endogenous bile acid metabolism. Bile secretion is impaired in both the Zucker (fa/fa) rat (76) and in the ob/ob mice (77), two animal models extensively studied as they both develop metabolic features resembling the metabolic syndrome. Furthermore, in lean and fat mouse lines developed from the same founder population by long-term divergent selection for low- or high body fat %, the lean mice exhibited higher hepatic Cyp8b1 and Abcb11 gene expressions, and had increased blood BA concentrations, relative to the obese mice (78). Even though the underlying mechanisms need to be further elucidated, our data strongly indicate a regulatory role for endogenous bile acid metabolism on the development of the metabolic syndrome. We used diets in which the casein was fully exchanged with SPH. Obviously,

differences in dietary amino acids other than taurine and glycine, such as the level of the branched-chain amino acids might have affected the outcome in the present study (Supplemetal Fig. 1). Moreover, we recognize that taurine has metabolic effects beyond bile acid metabolism, such as anti-oxidant capacity (79). However, the taurine tissue level in the rats (Supplemetal Fig 1) excludes the possibility that the observed differences were due to increased tissue taurine concentrations. Since addition of cholestyramine reduced plasma BA levels and attenuated the beneficial effects on adiposity, TAG metabolism and hyperglycemia, we conclude that parts of the beneficial effects found by SPH treatment was caused by nutritional regulation of endogenous BA metabolism. Our results provide evidence that nutritional regulation of bile acid metabolism may attenuate characteristics of the metabolic syndrome in high-fat fed rats. The relevance of our findings for man remains to be elucidated. However, it has been reported that obese subjects had a lower post-prandial bile-acid response, relative to normal weight subjects (80). Furthermore, subjects that had previously undergone gastric bypass, showed higher circulating bile acid levels, relative to both obese and severely obese objects. Also, total bile acids were inversely correlated with 2hr post meal-glucose and fasting TAG levels (81). In another study with objects subjected to bariatric surgery, circulating bile acid and GLP-1 levels increased post surgically, relative to pre-surgical baseline levels (82). Finally, it was reported in a human intervention study with cross-over design that subjects on a high fat, high protein diet had increased fasting plasma bile acid concentrations as compared to subjects on a high-fat diet alone. The elevation in plasma BA levels was accompanied by reduced hepatic lipids and increased fasting plasma concentrations of β-hydroxybutyrate (83). Thus, nutritional regulation of endogenous BA metabolism may also be related to development of the metabolic syndrome in man. In conclusion, we provide compelling evidence that plasma bile acid levels can be modulated by the dietary protein source in high-fat treated rats. Increased levels of plasma BAs were associated with a significant reduction in diet-induced obesity and resulted


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

11

in increased whole body energy expenditure and dissipation of energy in the form of heat. Concomitantly, fed-state plasma glucose and TAG concentrations were reduced. Thus, protein source dependent increase in plasma BA levels appears to have the potential to attenuate pathological characteristics of the metabolic syndrome in the high-fat fed rats.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

12

REFERENCES

1. Shonsey EM, Sfakianos M, Johnson M, et al. (2005) Meth. Enzymol. 400: 374-394

2. Vessey DA, Whitney J, and Gollan JL. (1983) Biochem. J. 214: 923-927.

3. Hofmann AF. (1999) News Physiol. Sci. 14: 24-29.

4. Pauli-Magnus C, Stieger B, Meier Y, Kullak-Ublick GA, and Meier PJ. (2005) J. Hepatol. 43:

342-357.

5. Parks DJ, Blanchard SG, Bledsoe RK, et al. (1999) Science 284: 1365-1368.

6. Makishima M, Okamoto AY, Repa JJ, et al. (1999) Science 284: 1362-1365.

7. Wang H, Chen J, Hollister K, Sowers LC, and Forman BM. (1999) Mol. Cell 3: 543-553.

8. Sinal CJ, Tohkin M, Miyata M, Ward JM, Lambert G, and Gonzalez FJ. (2000) Cell 102: 731-

744.

9. Lambert G, Amar MJA, Guo G, Brewer HB, Gonzalez FJ, and Sinal CJ. (2003) J. Biol. Chem. 278:

2563-2570.

10. Sirvent A, Claudel T, Martin G, et al. (2004) Febs Letters 566: 173-177.

11. Watanabe M, Houten SM, Wang L, et al. (2004) J. Clin. Invest. 113: 1408-1418.

12. Ikemoto S, Takahashi M, Tsunoda N, et al. (1997) Am. J. Physiol. Endocrinol. Metab. 36: E37-

E45.

13. De Fabiani E, Mitro N, Gilardi F, Caruso D, Galli G, and Crestani M. (2003) J. Biol. Chem. 278:

39124-39132.

14. Cao RS, Cronk ZX, Zha WB, et al. (2010) J. Lipid Res. 51: 2234-2244.

15. Zhang YQ, Lee FY, Barrera G, et al. (2006) Proc. Natl. Acad. Sci. USA 103: 1006-1011.

16. Ma K, Saha PK, Chan L, and Moore DD. (2006) J. Clin. Invest. 116: 1102-1109.

17. Cariou B, van Harmelen K, Duran-Sandoval D, et al. (2006) J. Biol. Chem. 281: 11039-11049.

18. Maruyama T, Miyamoto Y, Nakamura T, et al. (2002) Biochem. Biophys. Res. Commun. 298:

714-719.

19. Kawamata Y, Fujii R, Hosoya M, et al. (2003) J. Biol. Chem. 278: 9435-9440.

20. Watanabe M, Houten SM, Mataki C, et al. (2006) Nature 439: 484-489.

21. Maruyama T, Tanaka K, Suzuki J, et al. (2006) J. Endocrinol. 191: 197-205.

22. Katsuma S, Hirasawa A, and Tsujimoto G. (2005) Biochem. Biophys. Res. Commun. 329: 386-

390.

23. Thomas C, Gioiello A, Noriega L, et al. (2009) Cell Metab. 10: 167-177.

24. Sato H, Macchiarulo A, Thomas C, et al. (2008) J. Med. Chem. 51: 1831-1841.

25. Evans MJ, Mahaney PE, Borges-Marcucci L, et al. (2009) Am. J. Physiol. Gastrointest. Liver

Physiol. 296: G543-G552.

26. Zhang RW, Barnes S, and Diasio RB. (1992) Am. J. Physiol. 262: G351-G358.

27. Hardison WGM. (1978) Gastroenterology 75: 71-75.

28. Park T, Oh J, and Lee K. (1999) Nutr. Res. 19: 1777-1789.

29. Mühlfeld A, Kubitz R, Dransfeld O, Haussinger D, and Wettstein M. (2003) Arch. Biochem.

Biophys. 413: 32-40.

30. Liaset B, Madsen L, Hao Q, et al. (2009) Biochim. Biophys. Acta (BBA) - Mol. Cell Biol. Lip.

1791: 254-262.

31. Liaset B, and Espe M. (2008) Proc. Biochem. 43: 42-48.

32. Lauridsen C, Yong C, Halekoh U, et al. (2008) J. Sci. Food Agric. 88: 720-732.

33. Martens HA, and Dardenne P. (1998) Chemometr. Intelligent Lab. Syst. 44: 99-121.

34. Martens H, and Martens M. (2000) Food Qual. Pref. 11: 5-16.

35. Suckling KE, Benson GM, Bond B, et al. (1991) Atherosclerosis 89: 183-190.

36. Marschall H-U, Wagner M, Bodin K, et al. (2006) J. Lipid Res. 47: 582-592.

37. Madsen L, Petersen RK, Sorensen MB, et al. (2003) Biochem. J. 375: 539-549.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

13

38. Kanehisa M, and Goto S. (2000) Nucl. Acids Res. 28: 27-30.

39. Kanehisa M, Goto S, Hattori M, et al. (2006) Nucl. Acids Res. 34: D354-D357.

40. Madsen L, Pedersen LM, Lillefosse HH, et al. (2010) PLoS ONE 5: e11391

41. Bär A, Lina BAR, de Groot DMG, de Bie B, and Appel MJ. (1999) Regul. Toxicol. Pharmacol. 29:

S11-S28.

42. Liaset B, Julshamn K, and Espe M. (2003). Proc. Biochem. 38: 1747-1759.

43. Madsen L, and Berge RK. (1999) Lipids 34: 447-456.

44. Bas S, and Giacobino JP. (1983) Arch. Biochem. Biophys. 222: 416-423.

45. Xue BZ, Coulter A, Rim JS, Koza RA, and Kozak LP. (2005) Mol. Cell. Biol. 25: 8311-8322.

46. Wu Z, Puigserver P, Andersson U, et al. (1999) Cell 98: 115-124.

47. Lin J, Handschin C, and Spiegelman BM. (2005) Cell Metab. 1: 361-370.

48. Cannon B, and Nedergaard J. (2004) Physiol. Rev. 84: 277-359.

49. Hansen JB, Jorgensen C, Petersen RK, et al. (2004) Proc. Natl. Acad. Sci. USA 101: 4112-4117.

50. Frontini A, and Cinti S. (2010) Cell Metab. 11: 253-256.

51. Barbatelli G, Murano I, Madsen L, et al. (2010) Am. J. Physiol. Endocrinol. Metab. 298: E1244-

E1253.

52. Ghorbani M, Claus TH, and Himms-Hagen J. (1997) Biochem. Pharmacol. 54: 121-131.

53. Cederberg A, Gronning LM, Ahren B, Tasken K, Carlsson P, and Enerback S. (2001) Cell 106:

563-573.

54. Madsen L, Pedersen LM, Liaset B, et al. (2008) J. Biol. Chem. 283: 7196-7205.

55. Kast HR, Nguyen CM, Sinal CJ, et al. (2001) Mol. Endocrinol. 15: 1720-1728.

56. Zhang Y, Castellani LW, Sinal CJ, Gonzalez FJ, and Edwards PA. (2004) Genes Dev. 18: 157-169.

57. Hirokane H, Nakahara M, Tachibana S, Shimizu M, and Sato R. (2004) J. Biol. Chem. 279:

45685-45692.

58. Pineda Torra I, Claudel T, Duval C, Kosykh V, Fruchart J-C, and Staels B. (2003) Mol.

Endocrinol. 17: 259-272.

59. Sinal CJ, Yoon M, and Gonzalez FJ. (2001) J. Biol. Chem. 276: 47154-47162.

60. Wang Y-X, Lee C-H, Tiep S, et al. (2003) Cell 113: 159-170.

61. Akiyama TE, Lambert G, Nicol CJ, et al. (2004) J. Biol. Chem. 279: 20874-20881.

62. Sanderson LM, Boekschoten MV, Desvergne B, Muller M, and Kersten S. (2010) Physiol.

Genom. 41: 42-52.

63. Tanaka T, Yamamoto J, Iwasaki S, et al. (2003) Proc. Natl. Acad. Sci. USA 100: 15924-15929.

64. Dressel U, Allen TL, Pippal JB, Rohde PR, Lau P, and Muscat GEO. (2003) Mol. Endocrinol. 17:

2477-2493.

65. Kleiner S, Nguyen-Tran V, Bare O, Huang XM, Spiegelman B, and Wu ZD. (2009) J. Biol. Chem.

284: 18624-18633.

66. Clapham JC, Arch JRS, Chapman H, et al. (2000) Nature 406: 415-418.

67. Lee C-H, Olson P, Hevener A, et al. (2006) Proc. Natl. Acad. Sci. USA 103: 3444-3449.

68. Pilon G, Ruzzin J, Rioux L-E, et al. (2011) Metab. Clin. Exp. doi:10.1016/j.metabol.2010.12.005

69. Enochsson L, Isaksson B, Strömmer L, Erlanson-Albertsson C, and Permert J. (2010) Nutrition

26: 405-410.

70. Sugiyama K, Ohishi A, Ohnuma Y, and Muramatsu K. (1989) Agricult. Biol. Chem. 53: 1647-

1652.

71. Yokogoshi H, Mochizuki H, Nanami K, Hida Y, Miyachi F, and Oda H. (1999) J. Nutr. 129: 1705-

1712.

72. Esteller A. (2008) World J. Gastroenterol. 14: 5641-5649.

73. Ballatori N, and Truong AT. (1992) Am. J. Physiol. Gastrointest. Liver Physiol. 263: G617-624.

74. Larusso NF, Hoffman NE, Korman MG, Hofmann AF, and Cowen AE. (1978) Am. J. Digest. Dis.

23: 385-391.

75. Hofmann AF, and Hagey LR. (2008) Cell. Mol. Life Sci. 65: 2461-2483.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

14

76. Pizarro M, Balasubramaniyan N, Solis N, et al. (2004) Gut 53: 1837-1843.

77. Martin IV, Schmitt J, Minkenberg A, et al. (2010) Biol. Chem. 391: 1441-1449.

78. Simoncic M, Rezen T, Juvan P, et al. (2011) BMC Genomics 12: 96.

79. Bouckenooghe T, Remacle C, and Reusens B. (2006) Curr. Opin. Clin. Nutr. Metab. Care 9:

728-733.

80. Glicksman C, Pournaras DJ, Wright M, et al. (2010) Annals Clin. Biochem. 47: 482-484.

81. Patti M-E, Houten SM, Bianco AC, et al. (2009) Obesity 17: 1671-1677.

82. Nakatani H, Kasama K, Oshiro T, Watanabe M, Hirose H, and Itoh H. (2009) Metab. Clin. Exp.

58: 1400-1407.

83. Bortolotti M, Kreis R, Debard C, et al. (2009) Am. J. Clin. Nutr. 90: 1002-1010.


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

15

FOOTNOTES This work was carried out as a part of the ‘DOCMAR’ research program funded by Innovation Norway and RUBIN. This work was supported by the Danish Natural Science Research Council, The Novo Nordisk Foundation, The Carlsberg Foundation and performed as part of the research program of the Danish Obesity Research Centre, supported by the Danish Council for Strategic Research, Grant 2101-06-0005. Financial support was also received from the University of Bergen, Program Committee on Nutrition, from the Eckbo Foundation and from the Norwegian Research Council, Grant 200515-I30.

ACKNOWLEDGMENTS

We thank Aase Heltveit and Lars Erik Pindard for help during the animal studies. We also thank Guro K. Christensen and Jacop Wessels for valuable technical assistance.

FIGURE LEGENDS

FIGURE 1. Elevated plasma bile acid concentration is an underlying factor for the increased energy expenditure and decreased adiposity elicited by salmon protein hydrolysate (SPH) feeding.(A-F) Male Wistar rats (n=6) were fed high-fat diets (45 kcal% fat) ad libitum for 46 days with either SPH or casein as the sole protein source. (A-B) SPH-fed rats had elevated plasma bile acids (BAs), whereas total liver BAs were unchanged. (C-E) The SPH-fed rats showed reduced body weight gain. Energy efficiency, calculated as bodyweight gain per energy intake, and white adipose tissue masses were reduced by SPH treatment. (F) Energy digestibility was equal in SPH and casein fed rats. (G-M) Three groups of rats (n=6) were pair-fed the SPH-diet, the casein diet and the SPH-diet with 0.5wt% cholestyramine (c’am). (G-I) Inclusion of cholestyramine to the SPH-diet attenuated the increase in plasma BA concentrations, without modulating liver total BAs or growth. (J) Inclusion of cholestyramine attenuated the reduction in body fat mass determined by dual-X ray absorptiometry (DEXA). (K-L) Three groups of rats (n=6) were pair-fed the SPH-diet, the casein diet and the SPH-diet with 0.5wt% cholestyramine (c’am) and energy expenditure was calculated by indirect calorimetry. Cholestyramine treatment attenuated the increase in O2 consumption, CO2 elimination and heat production. Data are presented as mean + S.E. Significant differences from casein-fed rats are denoted by * (P<0.05) and ** (P<0.01). FIGURE 2. Bile acids induce fat oxidation in brown and white adipose tissues. (A-G) Male Wistar rats (n=6) were fed high-fat diets (45 kcal% fat) ad libitum for 46 days with either SPH or casein as the sole protein source. (A) SPH-treatment increased ex vivo palmitate-CoA oxidation capacity in interscapular brown adipose tissue (iBAT). (B-D) mRNA levels of Uncoupling protein1 (Ucp1) and Type 2 iodothyronine deiodinase (Dio2) were increased by SPH treatment, whereas the level of Peroxisome proliferator-activated receptor γ coactivator-1α (Pgc-1α) mRNA was not significantly increased. (E-G) mRNA levels of Ucp1, Dio2 and Pgc-1α in white adipose tissue depots of subcutaneous (inguinal, iWAT) and abdominal (epididymal and mesenteric, eWAT and MeWAT, respectively). (H-K) Three groups of rats were pair-fed the same diets, plus the SPH-diet with 0.5wt% cholestyramine (c’am) for 46 days. The reduction in abdominal (mesenteric + epididymal + perirenal and retroperitoneal) fat mass was attenuated by inclusion of cholestyramine in the SPH-diet. (I, J) Epididymal WAT ex vivo palmitate-coA oxidation capacity and mRNA levels. (K) Adipocyte cell size in eWAT was increased by inclusion of cholestyramine to the SPH diet. (L) Increased in vitro palmitate-CoA oxidation capactity in adipocytes of eWAT origin after 24 hours pre-incubation with taurocholic acid. mRNA levels are normalized to General transcription factor IIB (Gtf2b). Data are presented as mean + S.E (A-J: n= 4-6, L: n=3). Significant differences from casein-fed rats are denoted by * (P<0.05) and ** (P<0.01).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

16

FIGURE 3. Liver microarray analysis. Male Wistar rats (n=6) were fed high-fat diets (45 kcal% fat) ad libitum for 46 days with either SPH or casein as the sole protein source. RNA from individual rats was pooled within experimental groups, and the pooled samples were analyzed by Affymetrix arrays. (A) Data were analyzed and significantly altered KEGG pathways (P<0.05) are highlighted in darker blue, whereas pathways in lighter blue tended to be altered (P<0.01). (B) Heat map of selected genes involved in liver fatty acid import and oxidation regulated by SPH treatment. FIGURE 4. Reduced fed-state plasma TAG concentrations by elevated plasma bile acid levels. (A-L) Male Wistar rats (n=6) were fed high-fat diets (45 kcal% fat) ad libitum for 46 days with either SPH or casein as the sole protein source. (A-D) In the liver, SPH-fed rats exhibited reduced expression of Peroxisome proliferator-activated receptor α (Ppar α) and its target genes Carnitine palmitoyltransferase-1a (Cpt-1a) and acyl-CoA oxidase (Acox), but increased expression of Peroxisome proliferator-activated receptor β/δ (Ppar β/δ) and its target gene Adipocyte differentiating-related protein (Adrp). (E-G) Ex vivo liver mitochondrial CPT-1 capacity was enhanced, and liver mRNA level of the CPT-1 regulator Acetyl-CoA carboxylase 2 (Acc2) was reduced without significantly altering liver mitochondrial CPT-1 capacity in the presence of the CPT-1 inhibitor malonyl-CoA (5µM). (H) SPH-fed rats had elevated fasting plasma hydroxybutyrate. (I-L) Liver TAG and fed-state plasma TAG concentrations were reduced, whereas liver Very-low density lipoprotein receptor (Vldlr) gene expression was increased in the SPH-treated rats. (M-S) Two groups of rats (n=5) were pair-fed the SPH- and the casein-diets for 25 days. (M) Plasma metabolomic analysis by nuclear magnetic resonance (NMR) revealed that VLDL concentration was significantly reduced by SPH-treatment. Peaks on the positive y-axis scale are significantly elevated by SPH feeding and peaks on the negative y-axis are significantly higher in casein feed fed rats. R2 is the full cross-validated regression coefficient. (N, O) No significant alterations were found in plasma cholesterol levels measured by biochemical assay. (P-S) Three groups of rats (n=6) were pair-fed the SPH-diet, the casein diet and the SPH-diet with 0.5wt% cholestyramine (c’am)for 46 days. Pharmacological removal of bile acids by 0.5 wt% cholestyramine attenuated the lipid lowering effects of SPH in liver and fed-state plasma TAG. mRNA levels are normalized to General transcription factor IIB (Gtf2b). Data are presented as mean + S.E. Significant differences from casein-fed rats are denoted by * (P<0.05) and ** (P<0.01). FIGURE 5. Induction of skeletal muscle Ucp3 gene expression is abolished by pharmacological removal of bile acids. (A-C) Male Wistar rats (n=6) were fed high-fat diets (45 kcal% fat) ad libitum for 46 days with either SPH or casein as the sole protein source. (A) In skeletal muscle, SPH-fed rats exhibited reduced expression of Peroxisome proliferator-activated receptor α (Ppar α) and its target genes acyl-CoA oxidase (Acox) and Medium-chain acyl-CoA dehydrogenase (Mcad). (B) Muscle gene expression of Peroxisome proliferator-activated receptor β/δ (Ppar β/δ) and its target gene Adipocyte differentiating-related protein (Adrp) and angiopoietin-like 4 (Angptl4) was enhanced by SPH treatment. (C) Expression of genes related to fatty acid oxidation Carnitine palmitoyl transferase 1b (Cpt-1b) and Uncoupling proteins (Ucp2 and 3) was enhanced in the SPH-treated rats. (D) Three groups of rats (n=6) were pair-fed the same diets, plus the SPH-diet with 0.5wt% cholestyramine (c’am) for 46 days. Pharmacological removal of bile acids by 0.5 wt% cholestyramine completely abolished Ucp3 induction. mRNA levels are normalized to General transcription factor IIB (Gtf2b), and relative to the expression in the casein-fed rats. Data are presented as mean + S.E. Significant differences from casein-fed rats are denoted by * (P<0.05) and ** (P<0.01). FIGURE 6. Reduction in fed-state plasma glucose is attenuated by pharmacological removal of bile acids. (A-D) Male Wistar rats (n=6) were fed high-fat diets (45 kcal% fat) ad libitum for 46 days with either SPH or casein as the sole protein source. (A) Fed state plasma insulin level was decreased, whereas glucagon levels tended (P=0.06) to be lower in the SPH-treated rats. (B-C) Liver gene expression of the rate limiting gluconeogenic enzyme Phosphoenolpyruvate carboxy kinase 1 (Pck1) was decreased, and liver glycogen concentration was increased by SPH-feeding. (D) Fed state, but not fasted plasma glucose level was lower in the SPH-fed rats. (E) Three groups of rats (n=6) were pair-


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

17

fed the same diets, plus the SPH-diet with 0.5wt% cholestyramine (c’am) for 46 days. Pharmacological removal of bile acids by 0.5 wt% cholestyramine attenuated the glucose lowering effect of SPH. Pck1 mRNA levels are normalized to General transcription factor IIB (Gtf2b). Data are presented as mean + S.E. Significant difference from casein-fed rats are denoted by * (P<0.05) and ** (P<0.01). Significant differences between SPH- and SPH+cholestyramine-fed rats are denoted by ## (P<0.01). FIGURE 7. Repressed de novo bile acid synthesis and indications of increased biliary BA secretion. (A-H) Male Wistar rats (n=6) were fed high-fat diets (45 kcal% fat) ad libitum for 46 days with either SPH or casein as the sole protein source. (A) Plasma alanine aminotransferase (ALAT) was increased in the SPH fed rats, but the elevation did not indicate hepatocellular damage. (B) Liver gene expression of Uncoupling protein 2 (Ucp2), a marker of oxidative stress and linked to bile obstruction, was borderline (P=0.053) reduced by SPH treatment. (C) Hepatic concentrations of cholesterol and sterol esters were markedly decreased in the SPH treated rats. (D) The elevated plasma bile acid concentrations in SPH treated rats were accompanied with lower liver expression of genes involved in de novo bile acid synthesis, cholesterol-7a hydroxylase, Cyp7a1, and in bile acid conjugation, Bile acid Coenzyme A: amino acid N-acyltransferase, Baat. (E) Liver expression of genes encoding for apical transporters involved in bile generation Atp-binding cassette b4 (Abcb4, also called Mdr2) was strongly induced, whereas expression of the Abcb11 (also called Bsep) tended (P=0.09) to be higher. Expressions of Abcc2 (also called Mrp2) and Abcb1b (also called Mdr1) were not altered by SPH treatment. (F) Liver gene expressions of genes involved in glutathione (GSH) synthesis, glutamate-cysteine ligase, catalytic unit (Gclc) and modulator unit (Gclm) were unaltered, whereas glutathione synthase (Gss) was induced by SPH treatment. (G) Liver cytolocis total glutathione (GSH) was increased in the SPH treated rats. (H) No significant difference was observed in 5 days fecal bile acid excretion. mRNA levels are normalized to General transcription factor IIB (Gtf2b). Data are presented as mean + S.E. Significant difference from casein-fed rats are denoted by * (P<0.05) and ** (P<0.01).


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

18

Figure 1

0,0

0,5

1,0

1,5

µµ µµm

ol/

tissu

e

Liver BAs

SPH Casein

B

**

0

10

20

30

SPH Casein

µµ µµm

ol/ L

Plasma BA

**

A

0

100

200

300

0 10 20 30 40 50

Gra

m B

W

Days of feeding

Growth chart

Casein

SPH

C

0

7

14

21

SPH Casein

Gra

m B

W/

MJ

inta

ke

Energy efficiency

**

D

0

32

64

96

SPH Casein

% o

f ing

este

d en

ergy

Energy digestibilityF

0

2

4

6

eWAT MeWAT iBAT

g tis

sue

Adipose tissue masses

SPH

Casein

**

**

E

0

30

60

90

SPH SPH + c'am

Casein

Gra

m

Body fat

**

J

0

1

2

3

SPH SPH + c'am

Casein

µµ µµmol

/ tis

sue

Liver total BAs H

0

18

36

54

SPH SPH + c'am

Casein

µµ µµmol

/ L

Plasma BA

**

G

150

200

250

300

350

0 10 20 30 40 50

Gra

m B

W

Days of feeding

Growth chart

Casein

SPH

SPH + c'am

I

40

50

60

SPH SPH + c'am

Casein

Lite

r O

2/ k

g B

W/

24h

O2 consumed

**

K

*

35

45

55

SPH SPH + c'am

Casein

Lite

r C

O2/

kg

BW

/ 24

h

CO2 eliminated

*

**

L

600

660

720

780

SPH SPH + c'am

Casein

kJ/ 2

4h/

kg B

W

0.75

Heat production**

M

*


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

19

Figure 2

0

3

6

9

Ucp1 Dio2 Pgc-1a

Rel

exp

ress

ion

eWAT gene-expression

SPH

SPH + c'am

Casein

J

0

10

20

30

SPH SPH + c'am

Casein

Gra

m

Abdominal fatH

**

0,8

1,0

1,2

1,4

SPH SPH + c'am

Casein

Rel

ativ

e to

cas

ein

eWAT Palm-CoA oxI

0

0,6

1,2

1,8

0 25 75 200

Rel

ativ

e to

con

trol

Taurocholic acid, µµµµmol/L

In vitro adipocyte Palm-CoA oxidation

* ** *

L

0

120

240

360

SPH CaseinU

cp1/

Gtf2

b

iBAT Ucp1**

B

0

2

4

6

SPH Casein

Dio

2/ G

tf2b

iBAT Dio2

*

C

0,0

0,8

1,6

2,4

SPH Casein

nmol

/ m

in/ m

g pr

ot

iBAT Palm-CoA ox

*

A

0,0

1,3

2,6

3,9

SPH Casein

Pgc

-1αα αα

/ Gtf2

b

iBAT Pgc-1ααααD

0

1,2

2,4

3,6

iWAT eWAT MeWAT

Rel

ativ

e to

cas

ein

WAT Ucp1

SPH Casein

**

E

0,0

2,5

5,0

7,5

iWAT eWAT MeWAT

Rel

ativ

e to

cas

ein

WAT Dio2

SPH Casein

*

*

*

F

0

0,8

1,6

2,4

iWAT eWAT MeWAT

Rel

ativ

e to

cas

ein

WAT Pgc-1αααα

SPH Casein

* *

G

K eWAT adipocyte cell size

SPH SPH + c'am Casein


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

20

Figure 3

A

KEGG Pathway>2 fold Changed in this pathway

Total expressed genes in this

pathwayFisher P-

value Genes induced by SPH treatmentGenes repressed by SPH

treatment

Type I diabetes mellitus 6 19 0,015 RT1-Aw2, RT1-A3, RT1-T24-1, -

RT1-CE2, RT1-CE15, RT1-CE10

PPAR signaling pathway 10 49 0,024 Cd36, Fabp5, Fabp2, Cpt1a, Angptl4, Cyp7a1, Apoa2, Scd1

Fads2, Me1

Cell adhesion molecules (CAMs) 10 50 0,027 RT1-Aw2, RT1-A3, RT1-T24-1, RT1-CE2, Ptprm, Igsf4a, Jam3

RT1-CE15, RT1-CE10, Cldn1

Nicotinate and nicotinamide metabolism 4 11 0,031 - Enpp3, Enpp2, Pbef1, Aox1

Bile acid biosynthesis 5 18 0,038 Hadhb Adh7, Adh1, Srd5a1, Cyp7a1

Sulfur metabolism 3 7 0,044 - Smp2a, Sult2a1, Sult2a2_predicted

Antigen processing and presentation 7 35 0,058 RT1-Aw2,RT1-A3,RT1-T24-1,RT1-CE2, Klrd1

RT1-CE15,RT1-CE10

Terpenoid biosynthesis 2 3 0,058 Fdps, Fdft1 -

Jak-STAT signaling pathway 8 47 0,085 Akt2, Socs2, Stat3, Stat5b, Stam2, Ghr Il6ra, RGD1560373_predicted

Hematopoietic cell lineage 5 25 0,100 Itga5, Itga5_mapped, Cd36, Thpo Il6ra, Cd1d1

B Signal log ratio < -2 -2,0 -1,5 -1,0 -0,5 0.5 1.0 1,5 2,0 > 2

Function Gene

Cellular lipid import cd36 antigenlipase, hepaticlow density lipoprotein receptorlow density lipoprotein receptor-related protein 3low density lipoprotein receptor-related protein 4low density lipoprotein receptor-related protein associated protein 1very low density lipoprotein receptor

Fatty acid binding fatty acid binding protein 2, intestinalfatty acid binding protein 4, adipocytefatty acid binding protein 5, epidermalfatty acid binding protein 6, ileal (gastrotropin)

Fatty acid activation acyl-CoA synthetase long-chain family member 3acyl-CoA synthetase long-chain family member 4acyl-CoA synthetase long-chain family member 5acyl-CoA synthetase long-chain family member 6

Mitochondrial fatty acid oxidation carnitine palmitoyltransferase 1, livercarnitine palmitoyltransferase 1bsolute carrier family 25 (carnitine/acylcarnitine translocase), member 20carnitine palmitoyltransferase 2acyl-Coenzyme A dehydrogenase, short/branched chainacyl-Coenzyme A dehydrogenase, very long chainenoyl coenzyme A hydratase 1, peroxisomal/mitochondrialHADH (trifunctional protein), alpha subunitHADH (trifunctional protein), beta subunitmitochondrial acyl-CoA thioesterase 1dodecenoyl-coenzyme A delta isomerase

Krebs/ TCA-cycle citrate synthaseaconitase 2, mitochondrialisocitrate dehydrogenase 3, gammamalate dehydrogenase, mitochondrialpyruvate dehydrogenase kinase, isoenzyme 2pyruvate dehydrogenase kinase, isoenzyme 4

Ketone body metabolism acetyl-coenzyme A acetyltransferase 13-hydroxy-3-methylglutaryl CoA lyase


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

21

Figure 4

0,0

0,3

0,6

0,9

SPH Casein

Acc

2/ G

tf2b

Liver Acc2

**

F

0,0

1,7

3,4

5,1

SPH Casein

nmol

/ mg

prot

/ m

in

Liver mit CPT-1

*

E

-45

-30

-15

0

SPH Casein

% in

hibi

tion

CPT-1 inhibition by malonyl-CoA

G

0,0

0,5

1,0

1,5

Fasted Fed

mm

ol/ L

Plasma OH-butyrate

SPH

Casein

*

H

0

90

180

270

SPH Casein

mg/

tiss

ue

Liver TAGI

**

0

1

2

3

Fasted Fed

mm

ol/ L

Plasma TAG

SPH

Casein

K

*

0

12

24

36

SPH CaseinV

ldlr/

Gtf2

b

Liver Vldlr

*

L

1,5

1,7

1,9

2,1

SPH SPH + c'am

Casein

nmol

/ mg

prot

/ m

in

Liver mit Palm-CoA oxidation

R

0,00

0,17

0,34

0,51

SPH SPH + c'am

Casein

mm

ol/ L

Fed plasma OH-butyrate

S

0

2

4

6

SPH Casein

Ppa

r αα αα

/ Gtf2

b

Liver Pparαααα

*

A

0

7

14

21

Cpt-1a Acox1R

el e

xpre

ssio

n

Liver mRNA levels

SPH

Casein

B

*

0,0

0,7

1,4

2,1

SPH Casein

Ppa

rββ ββ

/ δδ δδ/ G

tf2b

Liver Ppar ββββ/δδδδ

**

C

0

11

22

33

SPH Casein

Adr

p/ G

tf2b

Liver Adrp

**

D

0,0

1,6

3,2

4,8

SPH SPH + c'am

Casein

mm

ol/ L

Fed plasma TAG

**

Q

Liver Oil red-O staining

SPH Casein

J

Fed plasma metabolomic by NMRM

0,0

0,8

1,6

2,4

Total HDL

mm

ol/ L

Fed plasma cholesterol

SPH

Casein

N

0,0

0,3

0,6

0,9

LDL HDL/total

mm

ol/ L

Fed plasma cholesterol

SPH

Casein

O

0

80

160

240

SPH SPH + c'am

Casein

mg/

tiss

ue

Liver TAGP

-40

-30

-20

-10

0

10

20

30

40

9 8 7 6 5 4 3 2 1

Reg

ress

ion

coe

ffic

ient

Chemical shift (ppm)

(R2=0.92)Choline

Unassigned peaks (not in casein)

Creatine

Lipid (unsaturated)

Lipid (mainly VLDL)

Lipid (mainly HDL or LDL)

TyrPyruvate or oxalacetate

Glycoprotein


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

22

Figure 5

Figure 6

0

0,6

1,2

1,8

Ppar a Acox Mcad

Rel

ativ

e ex

pres

sion

Ppar αααα related

SPH

Casein

A

* *

0

0,9

1,8

2,7

Ppar b/d Adrp Angptl4

Rel

ativ

e ex

pres

sion

Ppar ββββ/δ δ δ δ related

SPH

Casein

B

**

** *

0,0

1,7

3,4

5,1

Cpt-1b Ucp2 Ucp3

Rel

ativ

e ex

pres

sion

Fatty acid oxidation and uncoupling

SPH

Casein

C

***

*

0

0,8

1,6

2,4

Ppar b/d Ucp3

Rel

ativ

e ex

pres

sion

Ppar β/δβ/δβ/δβ/δ and Ucp3

SPH

SPH+c'am

Casein

*

##

D

0

27

54

81

Insulin Glucagon

pmol

/ L

Fed state hormones

SPH

Casein

A

*

0

6

12

18

Fasted Fed

mm

ol/ L

Plasma glucose

SPH

Casein

**

D

0,0

0,5

1,0

1,5

SPH Casein

Pck

1/ G

tf2b

Liver Pck1

*

B

0

35

70

105

SPH Casein

mg/

g li

ver

Liver glycogen

**

C

0,0

4,2

8,4

12,6

SPH SPH + c'am Casein

mm

ol/ L

Fed plasma glucose

SPH

SPH + c'am

Casein

E

*


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

23

Figure 7

0

0,4

0,8

1,2

Cyp7a1 Baat

Rel

ativ

e ex

pres

sion

Bile acid synthesis

SPH

Casein

D

**

0

0,9

1,8

2,7

Abcb11/ Bsep

Abcc2 / Mrp2

Abcb4/ Mdr 2

Abcb1b/ Mdr1

Rel

ativ

e ex

pres

sion

Apical hepatocyte transporters

SPH

Casein

E**

0

17

34

51

SPH Casein

µµ µµmol

/ 5 d

ays

Fecal bile acidsH

0

13

26

39

SPH Casein

U/ L

Plasma ALATA**

0,00

0,06

0,12

0,18

SPH Casein

µµ µµmol

/ mg

prot

ein

Liver GSHG*

0,0

0,6

1,2

1,8

Gclc Gclm GssRel

ativ

e ex

pres

sion

Liver expression GSH synthezising genes

SPH

Casein

F

**

0

12

24

36

Cholesterol Sterol esters

mg/

tiss

ue

Liver cholesterol

SPH

Casein

C

**

**

0

0,4

0,8

1,2

SPH CaseinR

elat

ive

expr

essi

on

Liver Ucp2B


ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

Environmental Health Perspectives • volume 118 | number 4 | April 2010 465

Research

Despite international agreements intended to limit the release of persistent organic pollu-tants (POPs) such as organochlorine pesti-cides, polychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins (PCDDs), and polychlorinated dibenzofurans (PCDFs), POPs still persist in the environment and food chains (Atlas and Giam 1981; Dougherty et al. 2000; Fisher 1999; Jorgenson 2001; Schafer and Kegley 2002; Van den Berg 2009). Most human populations are exposed to POPs through consumption of fat-containing food such as fish, dairy products, and meat (Fisher 1999). Humans bioaccumulate these lipo-philic and hydrophobic pollutants in fatty tis-sues for many years because POPs are highly resistant to metabolic degradation (Fisher 1999; Kiviranta et al. 2005). The physiologi-cal impact associated with chronic exposure to low doses of different mixtures of POPs is poorly understood, but epidemiological studies have reported that Americans, Europeans, and Asian patients with type 2 diabetes accumu-lated greater body burdens of POPs, including 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD),

2,2´,4,4´,5,5´-hexachlorobiphenyl (PCB153), coplanar PCBs (PCB congeners 77, 81, 126, and 169), p,p´-diphenyldichloroethene (DDE), oxychlordane, and trans-nonachlor (Fierens et al. 2003; Henriksen et al. 1997; Lee et al. 2006; Rignell-Hydbom et al. 2007; Turyk et al. 2009; Wang et al. 2008).

The incidences of type 2 diabetes and the insulin resistance syndrome have increased at a globally alarming rate, and > 25% of adults in the United States have been estimated to be affected by metabolic abnormalities associ-ated with insulin resistance (Ford et al. 2004). Impaired insulin action is a central dysfunc-tion of the insulin resistance syndrome charac-terized by abdominal obesity and defects in both lipid and glucose homeostasis, increas-ing the risk for developing type 2 diabetes, cardio vascular diseases, non alcoholic fatty liver disease, polycystic ovarian disease, and certain types of cancer (Biddinger and Kahn 2006; Reaven 2005). Although a sedentary lifestyle and consumption of high-fat food are considered major contributors to insulin resistance and obesity, these conventional risk

factors can only partly explain the worldwide explosive prevalence of insulin resistance–associated metabolic diseases. We therefore sought to elucidate whether the exposure to POPs present in a food matrix could con-tribute to insulin resistance and metabolic disorders.

POPs accumulate in the lipid fraction of fish, and fish consumption represents a source of POP exposure to humans (Dougherty et al. 2000; Hites et al. 2004; Schafer and Kegley 2002). Therefore, certain European countries have dietary recommendations to limit the consumption of fatty fish per week (Scientific Advisory Committee on Nutrition 2004). On the other hand, n-3 polyunsaturated fatty acids present in fish oil have a wide range of beneficial effects (Jump 2002), including pro-tection against high-fat (HF) diet– induced insulin resistance (Storlien et al. 1987). Accordingly, we fed rats an HF diet contain-ing either crude (HFC) or refined (HFR) fish oil obtained from farmed Atlantic salmon and

Address correspondence to J. Ruzzin, National Institute of Nutrition and Seafood Research, N-5817 Bergen, Norway. Telephone: (0047) 41450448. Fax: (0047) 559052 99. E-mail: [email protected]

Supplemental Material is available online (doi:10.1289/ehp.0901321 via http://dx.doi.org/).

We thank J. Burén for technical assistance on pri-mary adipocyte investigations, Å. Heltveit for ani-mal care, J.I. Hjelle for analysis of hepatic lipids, J. Wessel for assistance on tissue harvest, K. Heggstad for his expertise on contaminant analysis, J. Rieusset for advice on histology investigation, and C. Debard for real-time PCR analysis. We thank T. Fenchel, P. Grandjean, and E. Lund for discussions and com-ments on the manuscript.

This work was supported by grants from the European Research Council, the Research Council of Norway, the Danish Natural Science Research Council, The Diabetes Association, and the Novo Nordisk Foundation.

C.L.B., employed at Novo Nordisk (a leading manu facturer of insulin analogs and diabetes manage-ment care products), holds shares in the company, and contributed independently with the clamp stud-ies. The Novo Nordisk Foundation supports basic sci-ence independently of the company interests of Novo Nordisk A/S. The remaining authors declare they have no competing financial interest.

Received 11 August 2009; accepted 19 November 2009.

Persistent Organic Pollutant Exposure Leads to Insulin Resistance SyndromeJérôme Ruzzin,1 Rasmus Petersen,2,3 Emmanuelle Meugnier,4 Lise Madsen,1,3 Erik-Jan Lock,1 Haldis Lillefosse,1 Tao Ma,2,3 Sandra Pesenti,4 Si Brask Sonne,3 Troels Torben Marstrand,5 Marian Kjellevold Malde,1 Zhen-Yu Du,1 Carine Chavey,6 Lluis Fajas,6 Anne-Katrine Lundebye,1 Christian Lehn Brand,7 Hubert Vidal,4 Karsten Kristiansen,3 and Livar Frøyland1

1National Institute of Nutrition and Seafood Research (NIFES), Bergen, Norway; 2Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark; 3Department of Biology, University of Copenhagen, Copenhagen, Denmark; 4INSERM U-870, INRA (Institut national de la recherche agronomique) U-1235, Lyon 1 University, Institut National des Sciences Appliquées de Lyon and and Hospices Civils de Lyon, Oullins, France; 5The Bioinformatics Centre, Department of Biology and Biotech Research and Innovation Centre, University of Copenhagen, Copenhagen, Denmark; 6Institut de Recherche en Cancérologie de Montpellier, INSERM U896 Metabolism and Cancer Laboratory, Montpellier, France; 7Department of Insulin Pharmacology, Histology and Delivery, Novo Nordisk A/S, Maaloev, Denmark

Background: The incidence of the insulin resistance syndrome has increased at an alarming rate worldwide, creating a serious challenge to public health care in the 21st century. Recently, epide-miological studies have associated the prevalence of type 2 diabetes with elevated body burdens of persistent organic pollutants (POPs). However, experimental evidence demonstrating a causal link between POPs and the development of insulin resistance is lacking.

oBjective: We investigated whether exposure to POPs contributes to insulin resistance and meta-bolic disorders.

Methods: Sprague-Dawley rats were exposed for 28 days to lipophilic POPs through the con-sumption of a high-fat diet containing either refined or crude fish oil obtained from farmed Atlantic salmon. In addition, differentiated adipocytes were exposed to several POP mixtures that mimicked the relative abundance of organic pollutants present in crude salmon oil. We measured body weight, whole-body insulin sensitivity, POP accumulation, lipid and glucose homeostasis, and gene expres-sion and we performed micro array analysis.

results: Adult male rats exposed to crude, but not refined, salmon oil developed insulin resis-tance, abdominal obesity, and hepatosteatosis. The contribution of POPs to insulin resistance was confirmed in cultured adipocytes where POPs, especially organochlorine pesticides, led to robust inhibition of insulin action. Moreover, POPs induced down-regulation of insulin-induced gene-1 (Insig-1) and Lpin1, two master regulators of lipid homeostasis.

conclusion: Our findings provide evidence that exposure to POPs commonly present in food chains leads to insulin resistance and associated metabolic disorders.

key words: contaminants, farmed salmon, metabolic syndrome, nonalcoholic fatty liver, obesity, pollution, public health, type 2 diabetes. Environ Health Perspect 118:465–471 (2010). doi:10.1289/ehp.0901321 [Online 19 November 2009]

Ruzzin et al.

466 volume 118 | number 4 | April 2010 • Environmental Health Perspectives

investigated the metabolic impacts of POPs and their ability to interfere with n-3 poly-unsaturated fatty acids.

Materials and MethodsTissue RNA from liver of rats fed HFC and HFR was extracted using Trizol, and micro-array analysis was performed using the Operon Rat Oparray. Levels of specific mRNA were quantified using real-time polymerase chain reaction (PCR) as described previously (Rome et al. 2008). 3T3-L1 cells were exposed to different POP mixtures, and we measured insulin-stimulated glucose uptake and mRNA expression of target genes. Details of the methods are available in the Supplemental Material (doi:10.1289/ehp.0901321).

Animals. All experimental protocols described below were approved by the Norwegian State Board of Biological Experiments with Living Animals, and the animals were treated humanely and with regard for alleviation of suffering. Male Sprague-Dawley rats (Taconic, Ry, Denmark) weighing 200–250 g were housed with a 12-hr light/dark cycle and with free access to food and tap water. Animals were fed a standard diet (chow; 17% fat-derived calo-ries, 3.4 kcal/g) or an HF diet (65% fat-derived calories, 5.5 kcal/g) for 28 days (Lavigne et al. 2001). Two additional HF diets were made by substituting corn oil (20% wt/wt) with either crude or refined salmon oil. Crude salmon oil was obtained by heat-ing the rest raw material of farmed Atlantic salmon to 92°C and separating oil from water and solid material. Refined salmon oil was obtained by bleaching, carbon filtering, and deodorizing the crude oil. HF, HFC, and HFR diets were supplemented with cellu-lose (50 g/kg), choline bitartrate (2 g/kg), American Institute of Nutrition (AIN) vita-min mixture 76 (14 g/kg), and AIN min-eral mixture 76 (67 g/kg) (MP Biochemicals, Inc, Illrich, France) to meet the daily nutrient requirement levels of adult rats (Reeves et al. 1993). Fatty acid composition of HF, HFC, and HFR diets was analyzed as previously described (Jordal et al. 2007).

Hepatic lipids. We determined levels of triacylglycerol, diacyl glycerol, and total choles-terol in frozen liver samples of overnight-fasted rats using high-performance thin-layer chro-matography as described previously (Berntssen et al. 2005). Frozen (O.C.T. compound; Sakura Finetek Europe, Zoeterwoude, the Netherlands) and fixed (paraffin-embedded) liver sections were stained with Oil red O and hematoxylin and eosin (H&E), respectively.

Determination of POP levels. We meas-ured levels of POPs as described previously (Berntssen et al. 2005; Julshamn et al. 2004).

Determination of insulin action in periph-eral tissues. We used soleus muscles and

epididymal fat of overnight-fasted animals to assess insulin-stimulated glucose uptake as described previously (Buren et al. 2002; Ruzzin et al. 2005).

Hyperinsulinemic–euglycemic clamps. Animals were catheterized, and hyper-insulinemic–euglycemic clamps were per-formed 7 days later (Brand et al. 2003; Ruzzin et al. 2007). After a 6-hr fasting period, conscious unrestrained catheterized animals were infused with a prime (6 µCi) continu-ous (0.1 µCi/min for basal; 0.17 µCi/min for clamp) infusion of [3-3H]glucose from –90 to 120 min for assessment of whole-body glucose disappearance (Rd) and appearance (Ra) using Steele’s non–steady-state equations (Steele et al. 1956). The hyperinsulinemic–euglyce-mic clamp was performed (0–120 min) by a continuous infusion of human insulin (3 mU/kg/min) (Actrapid, Novo Nordisk, Bagsvaerd, Denmark), and euglycemia (~ 115 mg/dL) was maintained by variable infusion rates of a 20% non labeled glucose solution [glucose infusion rate (GIR)]. At the end of the clamp, rats were given a lethal dose of pento barbital sodium; liver, epididymal fat, and gastroc-nemius muscles were removed, frozen in liquid nitrogen, and stored at –80°C for determina-tion of POP levels. Plasma glucose and insulin levels were analyzed by the glucose oxidase method (YSI 2300 STAT Plus glucose ana-lyzer; YSI Incorporated, Yellow Spring, OH, USA) and an enzyme-linked immunosor-bent assay kit (DRG Instruments, Marburg, Germany), respectively. To determine plasma [3-3H]glucose, plasma was deproteinized, dried to remove tritiated water, resuspended in water, and counted in biodegradable scin-tillation fluid (Nerliens Meszansky, Oslo, Norway) on a beta scintillation counter (Tri-Carb 1900TR; Packard, Meriden, CT, USA). All samples were run in duplicate. Hepatic glucose production (HGP) was calculated as tracer-determined Ra minus GIR.

Insulin resistance was further assessed by the homeostasis model assessment of insulin resistance (HOMA-IR) index as described by Lee et al. (2008).

Cultured adipocyte studies. We used cul-tured and differentiated 3T3-L1 cells (Petersen et al. 2008) to assess insulin-stimulated glucose uptake and mRNA expression of target genes. On day 8 of the differentiation program, cells were exposed to vehicle (dimethyl sulfoxide) or POP mixtures for 48 hr, and glucose uptake was assessed.

Cytotoxicity. Membrane integrity of POP-treated adipocytes was determined by the release of lactate dehydrogenase into cell medium by a Tox7 kit (Sigma-Aldrich, Leirdal, Norway).

Statistical analysis. We examined differ-ences between groups for statistical signifi-cance using analysis of variance (ANOVA)

with the least-square difference post hoc test. We used one-class statistical analysis of microarray to identify differentially expressed genes (Tusher et al. 2001) between HFC- and HFR-fed rats. We determined statistical sig-nificance of the real-time PCR results using the Student’s t-test, and the threshold for sig-nificance was set at p ≤ 0.05.

ResultsCharacteristics of animals exposed to POPs. As we expected, concentrations of POPs were consistently much higher in the HFC diet than in the HFR diet [Supplemental Material, Table 1 (doi:10.1289/ehp.0901321)], whereas the contents of n-3 polyunsaturated fatty acids and other fatty acids were similar in the two diets because both the crude and the refined fish oils were obtained from the same batch of farmed salmon (Supplemental Material, Table 2 (doi:10.1289/ehp.0901321).

After 28 days, rats fed the HFC diet appeared normal, although they tended to gain more weight than rats fed the HFR diet despite similar daily energy intake (Figure 1A,B). Intake of the HFC diet, but not HFR diet, enhanced the accumulation of vis-ceral adipose tissue induced by HF consump-tion (Figure 1C,D). Profound dys regulation in lipid homeo stasis was further observed in livers of HFC-fed rats, which exhibited ele-vated levels of triacyl glycerol, diacyl glycerol, and total cholesterol compared with HF-fed rats; livers of HFR-fed rats tended to exhibit a reduced lipid accumulation (Figure 1E–G). Histological examinations highlighted the development of severe hepatosteatosis in rats fed HFC (Figure 1H) and confirmed that the presence of POPs in salmon oil provokes sig-nificant impairment of lipid metabolism.

To gain further insight into the pheno-typical changes of animals exposed to POPs, we performed a comparison of gene expres-sion profiles in the liver of rats fed the HFC and HFR diets, using oligonucleo tide micro-arrays. The expression of genes involved in drug metabo lism was affected, indicating dietary POP exposure [Supplemental Material, Table 3 (doi:10.1289/ehp.0901321)]. We also observed major differences for genes involved in lipid metabo lism and for several genes linked to lipid deposition (Supplemental Material, Table 3). Of interest, POPs induced robust down-regulation of insulin-induced gene-1 (Insig-1) and Lpin1, two master reg-ulators of lipo genesis and synthesis of trig-lyceride and cholesterol (Croce et al. 2007; Engelking et al. 2004; Finck et al. 2006; Lee and Ye 2004). Real-time PCR analysis con-firmed the strong repression of Lpin 1 and Insig-1 genes in the liver of rats consuming the HFC diet (Table 1). Similarly, in adipose tis-sue of HFC-fed rats, expression of Lpin1 and Insig-1 genes was repressed compared with

POP exposure provokes insulin resistance


HFR-fed animals [mean ± SE, 78 ± 8 vs. 55 ± 5 (n = 9, p = 0.02) for Insig-1 and 98 ± 11 vs. 64 ± 8 (n = 9, p = 0.03) for Lpin1 for HFR- and HFC-fed rats, respectively]. Furthermore, POPs induced a significant increase in the expression level of SREBP1C (sterol regula-tory element-binding protein 1C), the mas-ter regulator of the lipogenic pathway, and FAS (fatty acid synthase), a well-known target gene of SREBP1C (Table 1). Interestingly, the

hepatic expression of LXRα (liver X receptor alpha) was not affected, suggesting that the oxy sterol pathway was not modified by POP exposure (Table 1). Altogether, these results demonstrate that POP exposure signifi cantly affects the expression of critical genes involved in the regulation of lipid homeostasis. Gene set enrichment analysis further revealed sig-nificant effects on several biological pathways [Supplemental Material, Table 4 (doi:10.1289/

ehp.0901321)]. This analysis demonstrated a highly significant up-regulation of path-ways designated “pathogenic Escherichia coli infection” (EPEC/EHEC). The core genes up-regulated in the pathways include TLR5, ROCK2, CD14, and YWHAZ, a gene encod-ing a member of the 14-3-3 family of proteins reported to interact with insulin receptor sub-strate-1 and thereby regulating insulin signal-ing. Similarly, the roles of toll-like receptors,

Figure 1. Characteristics of rats fed salmon oil containing POPs. Body weight gain (A) and daily energy intake (B) in rats fed chow or the HF, HFR, or HFC diets over a 4-week period. (C) Exposed ventral view of a representative rat from each diet group showing increased visceral adipose tissue after consumption of the HFC diet. (D) Quantification of visceral fat (epididymal and perirenal fat pads). (E–G) Levels of hepatic triacylglycerol (E), diacyl glycerol (F), and total choles-terol (G). (H) Representative histological sections of liver stained with Oil red O (top) or H&E at low (middle) and high (bottom) magnifications; the three sections for each treatment group are from the same liver sample. All data are shown as mean ± SE; n = 8–9. *p < 0.02 compared with control. **p < 0.04 compared with HF.

200

160

120

80

40

0

200

160

120

80

40

0

30

25

20

15

10

5

0

120

100

80

60

40

20

0

1.2

0.8

0.4

0

20

15

10

5

0

** * *

Bod

y w

eigh

t gai

n (g

)

Ener

gy in

take

(kca

l/day

)

Visc

eral

fat (

g)

Tria

cylg

lyce

rol

(mg/

g)

Dia

cylg

lyce

rol (

mg/

g)

Tota

l cho

lest

erol

(mg/

g)

Chow

HFR HFC

HF

**

*

** *

*

*

****

** **

Chow

100 µM

200 µM

50 µM 50 µM 50 µM 50 µM

200 µM 200 µM 200 µM

100 µM 100 µM 100 µM

Oil r

ed O

H&E

HF HFR HFC

ChowHFHFRHFC

Ruzzin et al.


CD14, and rho kinases in regulating insulin signaling and establishment of insulin resis-tance in response to chronic low-grade inflam-mation are well documented (Begum et al. 2002; Cani et al. 2007; Furukawa et al. 2005; Petersen et al. 2008; Tzivion et al. 2001).

Effects of POPs on insulin action in vivo. Next, we assessed the impacts of POPs on

whole-body insulin action. In the basal state, intake of the HFC diet exacerbated the hyper-insulinemia induced by HF consumption, whereas animals fed HFR and control diets had similar plasma insulin levels (Figure 2A). Basal plasma glucose levels were similar in all groups (Figure 2B), but the HOMA-IR index was significantly increased in rats fed

the HFC diet (7.1 for control rats, 11.2 for rats fed HF, 8.4 for rats fed HFR, and 15.5 for rats fed HFC; p < 0.04).

The performance of hyperinsulinemic– euglycemic clamp, the gold standard for investigating and quantifying insulin resis-tance (Kraegen et al. 1983), revealed that the consumption of the HFC diet aggravated HF-induced reduced GIR, whereas HFR-fed rats showed no impairment of insulin action compared with control rats (Figure 2C). Reduced GIR reflects decreased insulin- mediated suppression of HGP, reduced insulin- stimulated peripheral glucose disposal rates, or both. Analysis of these parameters revealed that basal HGP was similar in all groups (Figure 2D), whereas suppression of HGP by insulin was impaired in animals consum-ing both HFC and HF diets (Figure 2E). Moreover, intake of HFC led to impaired insulin-mediated glucose disposal in peripheral tissues, which mainly include skeletal muscles and adipose tissue (Figure 2F). To investigate this further, we determined the rates of glucose uptake in isolated soleus muscles and primary adipocytes. We found that insulin-stimulated glucose uptake was reduced to a similar extent in skeletal muscle of animals fed HFC and HF diets (Figure 2G). In contrast, rats fed the HFR diet were protected against muscle insulin resistance (Figure 2G). In adipose tis-sue, the ability of insulin to stimulate glucose uptake was impaired in both the HFR and HF groups, and this metabolic defect was wors-ened by the consumption of the HFC diet (Figure 2H). Thus, exposure to POPs present in HFC exacerbated the deleterious metabolic effects of HF and attenuated the protective effects of n-3 polyunsaturated fatty acids, which indicates that the presence of environ-mental organic contaminants may influence the outcomes of food and dietary products.

There is growing evidence that mitochon-drial dysfunction contributes to insulin resis-tance (Lowell and Shulman 2005). To assess the impact of POPs on hepatic mitochon-drial content, we measured mitochondrial DNA levels by quantitative polymerase chain reaction (qPCR), using primers specific for the COXII gene, and determined the ratio between mitochondrial DNA and nuclear DNA as previously validated (Bonnard et al. 2008). We found no apparent modification of the amount of mitochondrial DNA in the liver of the animals fed HFC (ratio COXII/PPIA, 1.1 ± 0.2 (mean ± SE) for rats fed HFR and 0.9 ± 0.1 for rats fed HFC, p = 0.189). However, despite this apparent lack of change in mitochondrial content, we observed sig-nificant reduction in the expression of sev-eral genes related to mitochondrial function, such as PGC1α (peroxisome proliferator- activated receptor gamma-coactivator-1 alpha), citrate synthase, medium-chain acyl

Table 1. Real-time PCR determination of mRNA expression of a set of relevant genes in the liver of rats fed HFR or HFC diets (n = 9 per group).

HFR HFC p-ValueGenes related to mitochondrial functionPGC1α 0.73 ± 0.3 0.05 ± 0.02 0.043PPARα (peroxisome proliferator-activated receptor α) 76 ± 7 75 ± 18 0.988CS (citrate synthase) 316 ± 19 214 ± 10 0.002SDHA (succinate dehydrogenase) 74 ± 2 63 ± 4 0.038MCAD (medium chain acyl CoA dehydrogenase) 332 ± 30 170 ± 18 0.003Genes related to lipogenesisSREBP1C 3.0 ± 0.3 4.6 ± 0.6 0.021LXRα 50 ± 3 51 ± 7 0.932FAS 1.1 ± 0.1 1.9 ± 0.2 0.01Lpin 1 96 ± 17 22 ± 10 0.0017Insig-1 123 ± 23 43 ± 12 0.0071

Figure 2. Effects of salmon oil and POPs on insulin action and glucose metabolism evaluated by hyperin-sulinemic–euglycemic clamps performed in rats fed chow or HF, HFR, or HFC diets over a 4-week period. (A) Basal insulinemia. (B) Basal glycemia. (C) GIR. (D) Basal HGP. (E) HGP during the clamps. (F) Glucose disposal rate (Rd). (G) Insulin-stimulated glucose uptake in soleus muscles. (H) Insulin-stimulated glucose uptake in primary adipocytes. All data are shown as mean ± SE; n = 6–9. *p < 0.04 compared with chow control. **p < 0.04 compared with HF. #p < 0.05 compared with HFR. ##p < 0.03 compared with HF.

3

2

1

0

15

10

5

0

15

10

5

0

5

4

3

2

1

0

15

10

5

0

15

10

5

0

8

6

4

2

0

160

120

80

40

0

*

#

# #

#

##

##

##

**

Plas

ma

insu

lin

(ng/

mL)

GIR

(mg/

kg/m

in)

Clam

p H

GP

(mg/

kg/m

in)

Glu

cose

upt

ake

(mm

ol/k

g w

w/3

0 m

in)

Glu

cose

upt

ake

(fmoL

/1,0

00 c

ells

/min

)Cl

amp R d

(mg/

kg/m

in)

Bas

al H

GP

(mg/

kg/m

in)

Plas

ma

gluc

ose

(mg/

dL)

0 2,000

Insulin (µU/mL) Insulin (µU/mL)

2,0000

ChowHFHFRHFC



CoA dehydrogenase, and SDHA (succinate dehydrogenase) (Table 1), indicating the pres-ence of alterations in mitochondrial function and oxidative capacities in the liver of the rats exposed to POPs.

Analysis of POPs distribution in these animals revealed that whereas both liver and adipose tissue stored organochlorine pesti-cides, indicator PCBs, mono-ortho-substituted PCBs, and non–ortho-substituted PCBs, the liver preferentially retained PCDDs or PCDFs [Supplemental Material, Table 5 (doi:10.1289/ehp.0901321)].

Effects of POPs on insulin action in vitro. To further demonstrate the contribution of lipophilic POPs to the development of insu-lin resistance–associated metabolic distur-bances, we exposed differentiated adipocytes to a POP mixture that mimicked the relative abundance of organic contaminants found in crude salmon oil. Incubation of adipo-cytes with this POP mixture impaired the ability of insulin to stimulate glucose uptake (Figure 3A), which is in agreement with the reduced insulin–stimulated glucose uptake observed in adipose tissue of rats fed the HFC diet (Figure 2H). We then determined whether POP exposure, as observed in rats fed the HFC diet, could affect the expres-sion of Lpin1 and Insig-1 mRNA in cultured adipo cytes. After 48-hr treatment with the POP mixture, Lpin1 and Insig-1 mRNA lev-els were dose-dependently reduced in adipo-cytes [Supplemental Material, Figure 1 (doi:10.1289/ehp.0901321)], which con-firms the ability of POPs to interfere with key regulators of lipid metabolism. Importantly, the metabolic defects observed in adipocytes exposed to POPs were independent of cyto-toxicity, as demonstrated by the absence of an increased release of lactate dehydrogenase into the cell culture media (Supplemental Material, Figure 2). Altogether, these find-ings clearly establish the capacity of POPs to impair insulin action and associated metabolic abnormalities in a cell-autonomous manner.

Humans and other organisms are chroni-cally exposed to a variety of organic pollutants. To investigate which POPs contributed sig-nificantly to the impairment of insulin action, we incubated adipocytes with different POP mixtures. Although adipocytes exposed to a PCDD or PCDF mixture showed normal insulin action (Figure 3B,C), those exposed to non-ortho- substituted and mono-ortho- substituted PCB mixtures had reduced insulin action (Figure 3D,E). Impaired insulin action was independent of the total toxic equivalent (TEQ) concentration (Van den Berg et al. 2006) of the mixtures; up to 6.027ng WHO 2005 TEQ/mL for the PCDF mixture com-pared with 0.0016ng WHO 2005 TEQ/mL for the mono-ortho-PCB mixture. These find-ings demon strate that risk assessment based

on TEQ assigned to dioxins and dioxin-like PCBs (Van den Berg et al. 2006) is unlikely to reflect the risk of insulin resistance. Further investigations showed that insulin-stimulated glucose uptake was dramatically reduced in adipocytes treated with both the mixture of organochlorine pesticides (Figure 3F) and dichloro diphenyl trichloroethanes (DDTs) (Figure 3G), whereas the mixture of indica-tor PCBs had less inhibitory effects on insulin action (Figure 3H).

DiscussionIn this study, we demonstrate for the first time a causal relationship between POPs and insulin resistance in rats. In vivo, chronic exposure to low doses of POPs commonly found in food chains induced severe impairment of whole-body insulin action and contributed to the development of abdominal obesity and hepato-steatosis. Treatment in vitro of differentiated adipocytes with nano molar concentrations of POP mixtures mimicking those found in crude

Figure 3. Effects of POPs on insulin action in adipocytes shown as the ability of differentiated 3T3-L1 adipocytes to take up radioactive-labeled glucose in response to insulin measured after 48 hr exposure to several POP mixtures found in crude oil from farmed Atlantic salmon. (A) POP mixture, (B) PCDD mixture, (C) PCDF mixture, (D) non–ortho-substituted PCB mixture, (E) mono–ortho-substituted PCB mixture, (F) Pesticide mixture, (G) DDT mixture, or (H) PCB mixture. Concentrations of POP mixtures are shown according to the highest contaminant compound present in the mixture, as well as the World Health Organization (WHO) 2005 TEQ for dioxins and dioxin-like PCBs (Van den Berg et al. 2006). Glucose uptake was determined in eight parallel wells for each mixture and for each concentration. Data are expressed as relative glucose uptake and presented as mean ± SE. *p < 0.05 compared with vehicle (dimethyl sulfoxide)-treated cells.

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

Rela

tive

gluc

ose

upta

ke(fo

ld in

crea

se)

Rela

tive

gluc

ose

upta

ke(fo

ld in

crea

se)

Rela

tive

gluc

ose

upta

ke(fo

ld in

crea

se)

Rela

tive

gluc

ose

upta

ke(fo

ld in

crea

se)

Rela

tive

gluc

ose

upta

ke(fo

ld in

crea

se)

Rela

tive

gluc

ose

upta

ke(fo

ld in

crea

se)

Rela

tive

gluc

ose

upta

ke(fo

ld in

crea

se)

Rela

tive

gluc

ose

upta

ke(fo

ld in

crea

se)

0 1 3 10

Insulin (nM) Insulin (nM)

30

DDTs PCBs

Mono-ortho-PCBs Organochlorine pesticides

PCDFs Non-ortho-PCBs

POPs PCDDs

TEQ (ng WHO 2005 TEQ/mL)

0.0040.0430.425


0.0600.6036.027


0.0100.1031.025


0.00020.00161.0158

Vehicle1 nM10 nM100 nM



Vehicle10 nM100 nM1 µM

0 1 3 10 30

* * ***

*

* *

** *

** *

*

* *

*

**

**

*

*

**

*

* *

* *

**

*

**

* **

* *

*

** *

**

*

*

**

*

**

*

*

* *

* **

** **

* *

* *

****

Ruzzin et al.


salmon oil induced a significant inhibition of insulin-dependent glucose uptake. These data provide compelling evidence that exposure to POPs increases the risk of developing insulin resistance and metabolic disorders.

Despite intense investigations and estab-lishment of both preventive and therapeutic strategies, insulin resistance–associated meta bo-lic diseases such as type 2 diabetes, obesity, and non alcoholic fatty liver disease have reached alarming proportions worldwide (Angulo 2002; Ford et al. 2004; Zimmet et al. 2001). By 2015, the World Health Organization (WHO) estimates that > 1.5 billion people will be overweight and that 338 million people will die from chronic diseases such as diabe-tes and heart disease (WHO 2005). Although physical inactivity and regular intake of high-energy diets are recognized contributors (Hill and Peters 1998; Roberts and Barnard 2005), these lifestyle factors can only partially explain the explosive and uncontrolled global increase in metabolic diseases. Recently, the develop-ment of insulin resistance and inflammation was found to be exacerbated in humans and animals exposed to air pollution (Kelishadi et al. 2009; Sun et al. 2009). Furthermore, the widespread environmental contaminant bisphenol A was reported to impair pancreatic beta cells and trigger insulin resistance (Alonso-Magdalena et al. 2006). Our data, together with the finding that type 2 diabetics accu-mulate significant body burdens of POPs (Lee et al. 2006), provide additional evidence that global environmental pollution contributes to the epidemic of insulin resistance–associated metabolic diseases.

Although rats chronically fed the HFC diet for 28 days were exposed to a rela-tively high intake of organic pollutants, the

concentrations of PCDDs/PCDFs and indi-cator PCBs in adipose tissue of these animals did not exceed those observed in Northern Europeans 40–50 years of age (Kiviranta et al. 2005), thereby indicating that doses of POP exposure sufficient to induce detrimental health effects were not excessive. Whether the exposure to lower levels of POPs would induce similar detrimental effects as those observed in the present study remains to be investigated.

Dietary interventions are current strate-gies to prevent or treat metabolic diseases, and nutritional guidelines are usually based on energy density and glycemic index of the diet; however, the levels of POPs present in food has received less attention. Given that POPs are ubiquitous in food chains (Fisher 1999), such under estimation may interfere with the expected beneficial effects of some dietary recommendations and lead to poor outcomes. For instance, the presence of POPs in food products may, to some extent, explain the conflicting results regarding the protec-tive effects of n-3 polyunsaturated fatty acids against the incidence of myocardial infarction (Guallar et al. 1999; Rissanen et al. 2000). Overall, better understanding of the inter-actions between POPs and nutrients will help improve nutritional education of patients with insulin resistance syndrome.

To protect consumer health, the presence of contaminants in food is internationally regulated. In the European Union legislation, certain POPs including dioxins and dioxin-like PCBs are regulated in foodstuffs (European Union 2006). Risk assessment of these organic pollutants is based on the ability of individual compounds to produce hetero geneous toxic and biological effects through the binding of the aryl hydrocarbon receptor. Interestingly,

we found that cultured adipo cytes exposed to a PCDF or PCDD mixture have normal insulin action, even though the TEQ of these mixtures could be up to 3,500 times higher than the TEQ of the non-ortho-substituted and mono-ortho-substituted PCB mixtures that impaired insulin action. These findings demonstrate that risk assessment based on WHO TEQs assigned to dioxins and dioxin-like PCBs is unlikely to reflect the risk of insulin resistance and the pos-sible development of metabolic disorders.

Although the production of organo-chlorine pesticides has been restricted since the 1970s, the global production and use of pesticides are poorly controlled (Jorgenson 2001; Nweke and Sanders 2009), and the presence of these environmental chemicals in seafood still remains unregulated in European countries (European Union 2008). Of the POP mixtures tested in vitro, organochlorine pesticides were the most potent disruptors of insulin action. This powerful inhibitory effect of pesticides on insulin action likely explains the common finding emerging from several independent cross-sectional studies reporting an association between type 2 diabetes and the body burdens of p,p´-DDE, oxychlordane, or trans-nonachlor (Lee et al. 2006; Rignell-Hydbom et al. 2007; Turyk et al. 2009). Therefore, widespread pesticide exposure to humans appears to be of particular global con-cern in relation to public health.

We draw two main conclusions from these observations. First, exposure to POPs present in the environment and food chains are capable of causing insulin resistance and impair both lipid and glucose metabolism, thus supporting the notion that these chemi-cals are potential contributors to the rise in prevalence of insulin resistance and associated disorders (Figure 4). Second, although benefi-cial, the presence of n-3 polyunsaturated fatty acids in crude salmon oil (in the HFC diet) could not counteract the deleterious metabolic effects induced by POP exposure. Altogether, our data provide novel insights regarding the ability of POPs to mediate insulin resistance– associated metabolic abnormalities and pro-vide solid evidence reinforcing the importance of international agreements to limit the release of POPs to minimize public health risks.

RefeRences

Alonso-Magdalena P, Morimoto S, Ripoll C, Fuentes E, Nadal A. 2006. The estrogenic effect of bisphenol A disrupts pan-creatic β-cell function in vivo and induces insulin resis-tance. Environ Health Perspect 114:106–112.

Angulo P. 2002. Nonalcoholic fatty liver disease. N Engl J Med 346(16):1221–1231.

Atlas E, Giam CS. 1981. Global transport of organic pollutants: ambient concentrations in the remote marine atmosphere. Science 211(4478):163–165.

Begum N, Sandu OA, Ito M, Lohmann SM, Smolenski A. 2002. Active Rho kinase (ROK-α) associates with insulin receptor substrate-1 and inhibits insulin signaling in vascular smooth muscle cells. J Biol Chem 277(8):6214–6222.

Figure 4. Schematic representation of the possible mechanisms behind the development of the insulin resistance syndrome induced by POP exposure. POPs may activate nuclear receptors including aryl hydrocarbon receptor (AhR), pregnane X receptor (PXR), constitutive androstane receptor (CAR), or yet unknown receptors. POP exposure may induce the regulation of genes involved in the inflammatory path-way, mitochondrial function, lipid oxidation, and lipogenesis, thereby contributing to the development of the insulin resistance syndrome.

POPs

TLR5, ROCK2,CD14, and YWHAZ

Chronic low-gradeinflammation

Lipogenesis

Mitochondrial functionFatty acid oxidation

PGC1α

Insig-1

Lpin1

SREBP1cFAS

Insulin resistancesyndrome

Nuclearreceptors

(AhR, CAR, PXR, ?)



Berntssen MHG, Lundebye AK, Torstensen BE. 2005. Reducing the levels of dioxins and dioxin-like PCBs in farmed Atlantic salmon by substitution of fish oil with vegetable oil in the feed. Aquac Nutr 11(3):219–231.

Biddinger SB, Kahn CR. 2006. From mice to men: insights into the insulin resistance syndromes. Annu Rev Physiol 68:123–158.

Bonnard C, Durand A, Peyrol S, Chanseaume E, Chauvin MA, Morio B, et al. 2008. Mitochondrial dysfunction results from oxidative stress in the skeletal muscle of diet-induced insulin-resistant mice. J Clin Invest 118(2):789–800.

Brand CL, Sturis J, Gotfredsen CF, Fleckner J, Fledelius C, Hansen BF, et al. 2003. Dual PPARalpha/gamma activation provides enhanced improvement of insulin sensitivity and glycemic control in ZDF rats. Am J Physiol 284(4):E841–854.

Buren J, Liu HX, Jensen J, Eriksson JW. 2002. Dexamethasone impairs insulin signalling and glucose transport by deple-tion of insulin receptor substrate-1, phosphatidylinositol 3-kinase and protein kinase B in primary cultured rat adipo-cytes. Eur J Endocrinol 146(3):419–429.

Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, et al. 2007. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes 56(7):1761–1772.

Croce MA, Eagon JC, LaRiviere LL, Korenblat KM, Klein S, Finck BN. 2007. Hepatic lipin 1β expression is diminished in insulin-resistant obese subjects and is reactivated by marked weight loss. Diabetes 56(9):2395–2399.

Dougherty CP, Holtz SH, Reinert JC, Panyacosit L, Axelrad DA, Woodruff TJ. 2000. Dietary exposures to food contaminants across the United States. Environ Res 84(2):170–185.

Engelking LJ, Kuriyama H, Hammer RE, Horton JD, Brown MS, Goldstein JL, et al. 2004. Overexpression of Insig-1 in the livers of transgenic mice inhibits SREBP processing and reduces insulin-stimulated lipogenesis. J Clin Invest 113(8):1168–1175.

European Union. 2006. Amending regulation (EC) No 466/2001 setting maximum levels for certain contaminants in food-stuffs as regards dioxins and dioxin-like PCBs. Commission Regulation (EC) No 199/2006. Official Journal of the European Union L32:34–38. Available: http://www.health.gov.mt/fsc/fsc_euleg_files/RegEC199_2006e.pdf [accessed 19 February 2010].

European Union. 2008. Amending regulation (EC) No 396/2005 on maximum residue levels of pesticides in or on food and feed of plant and animal origin, as regards the imple-menting powers conferred on the Commission. Regulation (EC) No 299/2008. Official Journal of the European Union L97:37–71. Available: http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2008:097:0067:0071:EN:PDF [accessed 19 February 2010].

Fierens S, Mairesse H, Heilier JF, De Burbure C, Focant JF, Eppe G, et al. 2003. Dioxin/polychlorinated biphenyl body bur-den, diabetes and endometriosis: findings in a population- based study in Belgium. Biomarkers 8(6):529–534.

Finck BN, Gropler MC, Chen Z, Leone TC, Croce MA, Harris TE, et al. 2006. Lipin 1 is an inducible amplifier of the hepatic PGC-1alpha/PPARalpha regulatory pathway. Cell Metab 4(3):199–210.

Fisher BE. 1999. Most unwanted. Environ Health Perspect 107:A18–23.

Ford ES, Giles WH, Mokdad AH. 2004. Increasing prevalence of the metabolic syndrome among U.S adults. Diabetes Care 27(10):2444–2449.

Furukawa N, Ongusaha P, Jahng WJ, Araki K, Choi CS, Kim HJ, et al. 2005. Role of Rho-kinase in regulation of insulin action and glucose homeostasis. Cell Metab 2(2):119–129.

Guallar E, Aro A, Jimenez FJ, Martin-Moreno JM, Salminen I, van’t Veer P, et al. 1999. Omega-3 fatty acids in adipose

tissue and risk of myocardial infarction—The EURAMIC study. Arterioscler Thromb 19(4):1111–1118.

Henriksen GL, Ketchum NS, Michalek JE, Swaby JA. 1997. Serum dioxin and diabetes mellitus in veterans of Operation Ranch Hand. Epidemiology 8(3):252–258.

Hill JO, Peters JC. 1998. Environmental contributions to the obesity epidemic. Science 280(5368):1371–1374.

Hites RA, Foran JA, Carpenter DO, Hamilton MC, Knuth BA, Schwager SJ. 2004. Global assessment of organic contami-nants in farmed salmon. Science 303(5655):226–229.

Jordal AEO, Lie O, Torstensen BE. 2007. Complete replacement of dietary fish oil with a vegetable oil blend affect liver lipid and plasma lipoprotein levels in Atlantic salmon (Salmo salar L.). Aquac Nutr 13(2):114–130.

Jorgenson JL. 2001. Aldrin and dieldrin: a review of research on their production, environmental deposition and fate, bio-accumulation, toxicology and epidemiology in the United States. Environ Health Perspect 109(suppl 1):113–139.

Julshamn K, Lundebye AK, Heggstad K, Berntssen MHG, Boe B. 2004. Norwegian monitoring programme on the inorganic and organic contaminants in fish caught in the Barents Sea, Norwegian Sea and North Sea, 1994–2001. Food Addit Contam 21(4):365–376.

Jump DB. 2002. The biochemistry of n-3 polyunsaturated fatty acids. J Biol Chem 277(11):8755–8758.

Kelishadi R, Mirghaffari N, Poursafa P, Gidding SS. 2009. Lifestyle and environmental factors associated with inflamma-tion, oxidative stress and insulin resistance in children. Atherosclerosis 203(1):311–319.

Kiviranta H, Tuomisto JT, Tuomisto J, Tukiainen E, Vartiainen T. 2005. Polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls in the general population in Finland. Chemosphere 60(7):854–869.

Kraegen EW, James DE, Bennett SP, Chisholm DJ. 1983. In vivo insulin sensitivity in the rat determined by euglycemic clamp. Am J Physiol 245(1):E1–E7.

Lavigne C, Tremblay F, Asselin G, Jacques H, Marette A. 2001. Prevention of skeletal muscle insulin resistance by dietary cod protein in high fat-fed rats. Am J Physiol 281(1):E62–E71.

Lee DH, Lee IK, Song K, Steffes M, Toscano W, Baker BA, et al. 2006. A strong dose-response relation between serum con-centrations of persistent organic pollutants and diabetes: results from the National Health and Examination Survey 1999–2002. Diabetes Care 29(7):1638–1644.

Lee JN, Ye J. 2004. Proteolytic activation of sterol regulatory element-binding protein induced by cellular stress through depletion of Insig-1. J Biol Chem 279(43):45257–45265.

Lee S, Muniyappa R, Yan X, Chen H, Yue LQ, Hong EG, et al. 2008. Comparison between surrogate indexes of insulin sensitiv-ity and resistance and hyperinsulinemic euglycemic clamp estimates in mice. Am J Physiol 294(2):261–270.

Lowell BB, Shulman GI. 2005. Mitochondrial dysfunction and type 2 diabetes. Science 307(5708):384–387.

Nweke OC, Sanders WH III. 2009. Modern environmental health hazards: a public health issue of increasing significance in Africa. Environ Health Perspect 117:863–870.

Petersen RK, Madsen L, Pedersen LM, Hallenborg P, Hagland H, Viste K, et al. 2008. Cyclic AMP (cAMP)-mediated stimula-tion of adipocyte differentiation requires the synergistic action of Epac- and cAMP-dependent protein kinase-de-pendent processes. Mol Cell Biol 28(11):3804–3816.

Reaven GM. 2005. Why Syndrome X? From Harold Himsworth to the insulin resistance syndrome. Cell Metab 1(1):9–14.

Reeves PG, Nielsen FH, Fahey GC Jr. 1993. AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet. J Nutr 123(11):1939–1951.

Rignell-Hydbom A, Rylander L, Hagmar L. 2007. Exposure to persistent organochlorine pollutants and type 2 diabetes mellitus. Hum Exp Toxicol 26(5):447–452.

Rissanen T, Voutilainen S, Nyyssonen K, Lakka TA, Salonen JT. 2000. Fish oil-derived fatty acids, docosahexaenoic acid and docosapentaenoic acid, and the risk of acute coronary events—the Kuopio Ischaemic Heart Disease Risk Factor Study. Circulation 102(22):2677–2679.

Roberts CK, Barnard RJ. 2005. Effects of exercise and diet on chronic disease. J Appl Physiol 98(1):3–30.

Rome S, Lecomte V, Meugnier E, Rieusset J, Debard C, Euthine V, et al. 2008. Microarray analyses of SREBP-1a and SREBP-1c target genes identify new regulatory pathways in muscle. Physiol Genomics 34(3):327–337.

Ruzzin J, Turcotte S, Leblanc V, Lavigne C, Froyland L, Remman TR, et al. 2007. Dietary salmon protein improves insulin sensitivity in high-fat fed obese rats [Abstract]. Ann Nutr Metab 51:347.

Ruzzin J, Wagman AS, Jensen J. 2005. Glucocorticoid-induced insulin resistance in skeletal muscles: defects in insulin signalling and the effects of a selective glycogen synthase kinase-3 inhibitor. Diabetologia 48(10):2119–2130.

Schafer KS, Kegley SE. 2002. Persistent toxic chemicals in the US food supply. J Epidemiol Community Health 56(11):813–817.

Scientific Advisory Committee on Nutrition. 2004. Advice on Fish Consumption: Benefits and Risks. Available: http://cot.food.gov.uk/pdfs/fishreport2004full.pdf [accessed 19 February 2010].

Steele R, Wall JS, De Bodo RC, Altszuler N. 1956. Measurement of size and turnover rate of body glucose pool by the iso-tope dilution method. Am J Physiol 187(1):15–24.

Storlien LH, Kraegen E.W, Chisholm DJ, Ford GL, Bruce DG, Pascoe WS. 1987. Fish oil prevents insulin resistance induced by high-fat feeding in rats. Science 237(4817):885–888.

Sun Q, Yue P, Deiuliis JA, Lumeng CN, Kampfrath T, Mikolaj MB, et al. 2009. Ambient air pollution exaggerates adipose inflammation and insulin resistance in a mouse model of diet-induced obesity. Circulation 119(4):538–546.

Turyk M, Anderson HA, Knobeloch L, Imm P, Persky VW. 2009. Prevalence of diabetes and body burdens of poly-chlorinated biphenyls, polybrominated diphenyl ethers, and p,p´-diphenyldichloroethene in Great Lakes sport fish consumers. Chemosphere 75(5):674–679.

Tusher VG, Tibshirani R, Chu G. 2001. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA 98(9):5116–5121.

Tzivion G, Shen YH, Zhu J. 2001. 14-3-3 Proteins; bringing new definitions to scaffolding. Oncogene 20(44):6331–6338.

Van den Berg H. 2009. Global status of DDT and its alternatives for use in vector control to prevent disease. Environ Health Perspect 17:1656–1663.

Van den Berg M, Birnbaum LS, Denison M, De Vito M, Farland W, Feeley M, et al. 2006. The 2005 World Health Organization reevaluation of human and mammalian toxic equivalency factors for dioxins and dioxin-like compounds. Toxicol Sci 93(2):223–241.

Wang SL, Tsai PC, Yang CY, Guo YL. 2008. Increased risk of diabetes and polychlorinated biphenyls and dioxins—a 24-year follow-up study of the Yucheng cohort. Diabetes Care 31(8):1574–1579.

WHO. 2005. Chronic Diseases: a Vital Investment. Geneva:World Health Organization. Available: http://www.who.int/chp/chronic_disease_report/contents/en/index.html [accessed 19 February 2010].

Zimmet P, Alberti KG, Shaw J. 2001. Global and societal implica-tions of the diabetes epidemic. Nature 414(6865):782–787.

The obesogenic effects of polyunsaturated fatty acids are ...

Documents

Transcript of The obesogenic effects of polyunsaturated fatty acids are ...