Research Collection

Doctoral Thesis

Effects of pro-inflammatory cytokines on central actions of leptinand insulin

Author(s): Mc Allister, Eugenia

Publication Date: 2014

Permanent Link: https://doi.org/10.3929/ethz-a-010262596

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

DISS. ETH NO. ETH 22082

EFFECTS OF PRO-INFLAMMATORY CYTOKINES ON CENTRAL ACTIONS

OF LEPTIN AND INSULIN

A thesis submitted to attain the degree of

DOCTOR OF SCIENCES of ETH ZURICH

(Dr. sc. ETH Zurich)

presented by

Eugenia Mc Allister

M. Sc. in Biology, University of Zurich

born on 24.02.1970

citizen of the United States of America

accepted on the recommendation of

Prof. Wolfgang Langhans, examiner

Prof. Stephen Woods, co-examiner

Prof. Gustavo Pacheco López, co-examiner

2014

Acknowledgements

My deepest gratitude is to my Professor Dr. Wolfgang Langhans for the life-changing

experience created by his lab, his exciting projects and the exchange of ideas with his team. He

gave me the opportunity to find intellectual development and to contribute to science, I can

only hope to make the world a little better. Thank you very much, Wolfgang, for your guidance

and help.

I would also like to thank Professor Dr. Steve Woods for finding the time for

mentoring me and for his thoughtful advice about science and life. I enjoyed our little talks

every time you visited our lab. I look forward to continue collaborating with you in the

future.

To Professor Dr. Gustavo Pacheco-Lopez also my sincere thanks. Gustavo guided me

early in my experimental work and showed me many techniques that proved very helpful in

my research.

Thanks also to Myrtha Arnold for her ever present help and support through the

more than the thousand days I spent in the lab. Your help in the everyday little things as well

as in the big surgical procedures has been invaluable.

Sandra Giovanoli taught me and supported me in my stereology studies and was

always available when I needed help. Thank you.

My appreciation also to Liz Weber for making my life so much easier by genotyping

and mounting and staining hundreds of the brain sections for me.

Of course without Mandy, Simone, Michelle and Daniela, our animal caretakers, there

would have been no animals for my experiments. Their care for animal is as patient and

dedicated as it was for me.

Very special thanks to Hawi who not only helped me in these last few months to

prepare my papers and thesis but gave me the emotional support I needed to climb the last

hill before my graduation.

Of course Richard, Dasha, and Blakie are always first in my heart and were always

supportive and loving during my doctorate as they have been in my life.

Abraham Hicks, Bashar, Ajahn Brahm, Eckhard Tolle, Siddhartha will probably never

read this thesis or even hear about it but their words were a constant support and kept me in

the Vortex often.

i

Table of Contents

Summary ..................................................................................................................................................... 1

Zusammenfassung ....................................................................................................................................... 3

Chapter 1. .................................................................................................................................................... 5

General Introduction ................................................................................................................................... 5

1.1. Obesity ......................................................................................................................................... 5

1.2. Metabolic syndrome .................................................................................................................... 5

1.3. TNF-α ........................................................................................................................................... 6

1.4. IL-6 ............................................................................................................................................... 7

1.5. Insulin .......................................................................................................................................... 7

1.6. Leptin ........................................................................................................................................... 8

1.7. Melanocortin system ................................................................................................................... 9

1.8. Aim of the thesis ........................................................................................................................ 10

1.9. References ................................................................................................................................. 11

Chapter 2. .................................................................................................................................................. 15

Order Effects Can Obscure the Eating-Inhibitory Effect of Intracerebroventricular Insulin in Mice .......... 15

2.1. Introduction ............................................................................................................................... 15

2.2. Materials and Methods .............................................................................................................. 16

2.2.1. Animals and housing .......................................................................................................... 16

2.2.2. Diet and food intake measurements.................................................................................. 17

2.2.3. Third ventricular (i3vt) cannulation ................................................................................... 17

2.2.4. Cannula patency verification ............................................................................................. 17

2.2.5. Experiment 1 ...................................................................................................................... 18

2.2.6. Experiment 2 ...................................................................................................................... 18

2.2.7. Experiment 3 ...................................................................................................................... 18

2.2.8. Experiment 4 ...................................................................................................................... 18

2.2.9. Statistical analysis .............................................................................................................. 18

2.3. Results ....................................................................................................................................... 19

2.3.1. Experiment 1 ...................................................................................................................... 19

2.3.2. Experiment 2 ...................................................................................................................... 20

ii

2.3.3. Experiment 3 ...................................................................................................................... 21

2.3.4. Experiment 4 ...................................................................................................................... 22

2.4. Discussion .................................................................................................................................. 22

2.5. References ................................................................................................................................. 27

Chapter 3. .................................................................................................................................................. 30

Effect of Central Leptin and TNF-α on Food Intake in Mice ....................................................................... 30

3.1. Introduction ............................................................................................................................... 30

3.2. Materials and methods .............................................................................................................. 32

3.2.1. Animals and housing .......................................................................................................... 32

3.2.2. Diet / food intake and body weight measurements .......................................................... 32

3.2.3. Third ventricular surgery (i3vt) .......................................................................................... 32

3.2.4. Intra-third-ventricular (i3vt) infusions ............................................................................... 33

3.2.5. Cannula placement verification: NPY functional test ........................................................ 33

3.2.6. Experiments ....................................................................................................................... 34

3.2.7. Statistical analysis .............................................................................................................. 36

3.3. Results ....................................................................................................................................... 37

3.4. Discussion .................................................................................................................................. 42

3.5. References ................................................................................................................................. 46

Chapter 4. .................................................................................................................................................. 52

Effect of IL-6 on Central Food Inhibitory Effect of Leptin in Mice .............................................................. 52

4.1. Introduction ............................................................................................................................... 52

4.2. Materials and methods .............................................................................................................. 53

4.2.1. Animals and housing .......................................................................................................... 53

4.2.2. Diet / food intake and body weight measurements .......................................................... 53

4.2.3. Third ventricular surgery (i3vt) .......................................................................................... 53

4.2.4. Intra-third-ventricular (i3vt) infusions ............................................................................... 54

4.2.5. Cannula placement verification: NPY functional test ......................................................... 54

4.2.6. Experimental design........................................................................................................... 55

4.2.7. Statistical analysis .............................................................................................................. 55

4.3. Results ....................................................................................................................................... 55

4.3.1. Experiment 1. Effect of i3vt administration of IL-6 on food intake in mice ........................ 55

iii

4.3.2. Experiment 2. Effect of i3vt co-administration of IL-6 and leptin on food intake in mice fed

chow 56

4.3.3. Experiment 3. Effect of i3vt administration of MT-II on food intake in mice ..................... 57

4.4. Discussion .................................................................................................................................. 58

4.5. References ................................................................................................................................. 61

Chapter 5. .................................................................................................................................................. 65

General Discussion .................................................................................................................................... 65

5.1. References ................................................................................................................................. 70

List of Abbreviations .................................................................................................................................. 73

iv

1

Summary

The peripheral expression and plasma levels of pro-inflammatory cytokines are known to be increased in

obesity and are implicated in the development of the metabolic syndrome. The metabolic syndrome is

characterized by insulin and leptin resistance. Tumor necrosis factor alpha (TNF-α) and interleukin-6 (IL-

6) are the major cytokines implicated in the induction of this resistance. TNF-α and IL-6 are also

expressed in the brain, and their role in the central regulation of energy homeostasis is unclear. The

cells that express TNF-α and IL-6 receptors are in the same brain areas that are involved in the

mediation of the central effects of the adiposity signals insulin and leptin on food intake and energy

expenditure, suggesting a potential interaction between these cytokines and adiposity signals. Hence,

the aim of the present thesis was to explore the potential effects and interactions of insulin and leptin

with TNF-α and IL-6 on food intake in the brain.

In the first study we investigated the central effects of insulin on food intake in mice. Previous

studies showed that centrally administrated insulin reduces food intake in laboratory animals and other

species. We found that the effect of insulin infused into the third ventricle (i3vt) can be ambiguous and

that the experimental design, in particular order effects, may further obscure it. Thus, combining mean

cumulative food intake of both cross-over trials in a within-subjects design did not reveal a significant

effect of insulin on food intake. Separate within-subject analysis of animals that received insulin or

vehicle on the first day, however, showed that insulin in fact reduced food intake substantially, i.e., that

the within-subject design obscured this effect. In another experiment in the same study, mice did not

reduce food intake in response to higher doses of insulin, but reduced food intake in response to the

melanocortin receptor agonist melanotan-II (MT-II). As insulin supposedly reduces food intake mainly

through activating the melanocortin system, this indicates that whatever obscured the insulin effect in

the crossover design must have occurred upstream of the melanocortin receptors. Regardless, our

findings indicate that the effect of the intra-third ventricular (i3vt) insulin infusion on food intake can be

variable and the specific causes of such variability are difficult to identify.

In the next study we investigated the potential central interaction between the effects of TNF-α

and leptin on food intake. We found that co-administration of TNF-α and leptin elicited a greater

inhibition of food intake than each compound alone in mice fed a low fat diet. The

immunohistochemical analysis of the activated signal transducer and activator of transcription-3

2

(pSTAT3) expression in the hypothalamus also revealed an additive effect of the combined

administration of TNF-α and leptin compared with both compounds alone. High-fat-diet (HFD)-fed mice

reduced food intake in response to the combined administration at an early time point of measurement

(2 h) and did not respond to leptin alone, indicating that HFD feeding induced leptin tolerance. TNF-α

knock-out mice (TNFKO) did not reduce food intake in response to i3vt TNF-α administration. These

animals also reduced food intake in response to intraperitoneal (IP) injections of bacterial

lipopolysaccharide (LPS) similar to wild type (wt) mice, indicating that other cytokines secreted in

response to LPS administration compensated for the lack of TNF-α. Overall, this study showed that the

central eating-inhibitory effects of TNF-α and leptin are additive rather than synergistic.

In the last study we evaluated the effects of IL-6 on food intake alone and in combination with

leptin. We found that IL-6 increased food intake later in the light cycle (22 h) when administered alone.

The combined administration of IL-6 and leptin produced an enhanced eating-inhibitory effect of leptin.

This study supported the previously described divergence of the IL-6 effects. We are the first, however,

to show that the central administration of IL-6 increases food intake and at the same time enhances

leptin’s eating-inhibitory effect. The mechanisms of this interesting phenomenon still need to be

identified.

In conclusion, our studies show that the pro-inflammatory cytokines TNF-α and IL-6, which are

implicated in peripheral insulin resistance, may have additive or even potentiating central effects with

respect to the eating-inhibitory actions of adiposity signals, in particular with respect to leptin. Further

studies are necessary to examine in detail the mechanisms of their action.

3

Zusammenfassung

Bei der Adipositas und beim Metabolische Syndrom sind sowohl die periphere Expression als auch die

Plasmakonzentrationen von pro-inflammatorischen Cytokinen erhöht. Das Metabolische Syndrom ist

durch Leptin- und Insulinresistenz charakterisiert. Der Tumor-Nekrosefaktor-alpha (TNF-a) und

Interleukin-6 (IL-6) sind wichtige Cytokine bei der Entwicklung dieser Resistenz. TNF-a und IL-6 sind auch

im Gehirn exprimiert, aber es ist unklar, welche Rolle sie dort bei der Regulation der Energiehomöostase

spielen. Die Rezeptoren für TNF-a und IL-6 befinden sich in den Regionen des Gehirns, die auch die

Effekte der Adipositassignale Insulin und Leptin auf die Nahrungsaufnahme und den Energieumsatz

vermitteln, was die Vermutung nahelegt, dass es diesbezüglich Interaktionen zwischen Cytokinen und

Adipositassignalen gibt. Daher war es das Ziel der vorliegenden Arbeit, mögliche Interaktionen von

Insulin und Leptin mit TNF-a und IL-6 im Gehirn bezüglich ihrer Effekte auf die Nahrungsaufnahme zu

überprüfen.

In einer ersten Studie untersuchten wir den zentralen Effekt von Insulin auf die

Nahrungsaufnahme bei Mäusen. Vorangegangene Studien zeigten, dass zentral verabreichtes Insulin bei

Labortieren und anderen Spezies die Nahrungsaufnahme hemmt. Wir fanden, dass der Effekt von Insulin

bei der Infusion in den dritten Hirnventrikel (i3vt) nicht eindeutig war und dass das experimentelle

Design, insbesondere „Order Effects“ das Ergebnis verfälschten. So konnten wir bei der Kombination der

mittleren kumulativen Nahrungsaufnahme bei beiden Cross-over Trials eines Experiments keinen

signifikanten Effekt von Insulin auf die Nahrungsaufnahme feststellen. Bei der separaten Analyse der

Tiere, die entweder Insulin oder Vehikel am ersten Tag bekamen, zeigte sich hingegen eine eindeutige

Hemmung der Nahrungsaufnahme durch Insulin.

Ein weiteres Experiment derselben Studie zeigte, dass auch höhere Insulin-Dosen die

Nahrungsaufnahme in einem Crossover Trial nicht reduzierten. Hingegen reduzierte der Melanocortin-

Receptoragonist Melanotan-II (MT-II) die Nahrungsaufnahme. Da Insulin die Nahrungsaufnahme

vermutlich primär über eine Aktivierung des Melanocortin-Systems reduziert, kann man daraus ableiten,

dass die Verschleierung des Insulin-Effekts bei dem „Crossover design“ „upstream“ der Melanocortin-

Rezeptoren auftrat. Insgesamt zeigen unsere Befunde, dass der Effekt von Insulin auf die

Nahrungsaufnahme nach i3vt-Infusion variieren kann. Die genaue Ursache dieser Variabilität ist schwer

zu identifizieren.

4

In der zweiten Studie untersuchten wir die potentiellen Interaktionen zwischen TNF-a und

Leptin bezüglich ihres Effekts auf die Nahrungsaufnahme. Bei Mäusen, die mit einer Niedrig-Fett-Diät

gefüttert wurden, fanden wir bei gemeinsamer Infusion von TNF-a und Leptin eine stärkere Hemmung

der Nahrungsaufnahme als bei separater Verabreichung der beiden Substanzen. Die

immunhistochemische Analyse des aktivierten “signal transducer and activator of transcription-3”

(pSTAT3) zeigte ebenfalls einen additiven Effekt der kombinierten Verabreichung von TNF-a und Leptin,

verglichen mit der separaten Verabreichung beider Substanzen. Bei Mäusen, die mit einer fettreichen

Diät gefüttert wurden, führte die kombinierte Infusion von Leptin und TNF-a ebenfalls zu einer

Verzehrsreduktion an einem frühen Messzeitpunkt (2 h), Leptin alleine hatte jedoch keinen Effekt, was

auf eine Leptin-Toleranz hinweist. TNF-a knockout Mäuse (TNFKO) reduzierten die Nahrungsaufnahme

nach i3vt-Infusion von TNF-a nicht. Diese Tiere zeigten auch eine vergleichbare Verzehrsdepression wie

Wild-Typ-Mäuse nach intraperitonealer Injektion von bakteriellem Lipopolysaccharid (LPS). Dies lässt

vermuten, dass andere Cytokine das fehlende TNF-a kompensieren können. Alles in allem zeigte die

zweite Studie, dass der zentrale verzehrshemmende Effekt von TNF-a und Leptin eher additiv als

synergistisch ist.

In der dritten und letzten Studie werteten wir den Effekt der alleinigen i3vt-Infusion von IL-6 und

der Kombination von IL-6 und Leptin auf die Nahrungsaufnahme aus. Dabei beobachteten wir, dass IL-6

alleine die Nahrungsaufnahme steigerte, während die Kombination von IL-6 mit Leptin den

verzehrsreduzierenden Effekt von Leptin verstärkte. Ein derartiger Befund wurde bisher noch nicht

beschrieben, unterstützt aber die bereits früher beobachtete Divergenz des IL-6 Effekts auf die

Nahrungsaufnahme. Der Mechanismus dieses interessanten Phänomens bleibt Gegenstand weiterer

Forschung.

Unsere Studien zeigten, dass pro-inflammatorische Cytokine wie TNF-a und IL-6, welche zur

peripheren Insulinresistenz beitragen, vermutlich einen additiven Effekt bezüglich der

Nahrungsaufnahme im zentralen Nervensystem haben und möglicherweise den verzehrsreduzierenden

Effekt von Adipositassignalen wie Leptin verstärken. Weitere Studien sind notwendig, um den genauen

diesbezüglichen Wirkmechanismus der Cytokine zu untersuchen.

5

Chapter 1.

General Introduction

1.1. Obesity

According to the World Health Organization fact sheet adapted March 2013, obesity nearly doubled

since 1980. Sixty-five percent of the world's population lives in countries where overweight and obesity

kills more people than underweight. Among the consequences of obesity are cardio-vascular diseases,

type-2-diabetes (T2D), and osteoarthritis, to name but a few. Unlike life-style changes or most

pharmacological treatments, bariatric surgery, supposedly a long-term solution, provides an opportunity

for obese patients to maintain reduced body weight. The surgery, however, is invasive, costly,

recommended only for the morbidly obese patients, and may lead to side effects such as vitamin

deficiency, blood clots, kidney stones, leaking incisions, and others. Therefore, the search for a safe,

long-term pharmacological approach, along with an improvement of the surgical procedures, has been

the focus of the international medical and scientific community for years. Importantly, however, obesity

is preventable, but a deep understanding of the physiology and psychology of energy balance regulation

and nutrition will be essential to eventually reduce the globalization of obesity by prevention or

treatment.

1.2. Metabolic syndrome

Obesity is closely related to the metabolic syndrome, which is characterized by low grade chronic

inflammation (23, 25). The hypoxia-affected adipose tissue, as well as infiltrated macrophages, secrete

pro-inflammatory mediators (10, 21, 35, 45). The resulting increased levels of these inflammatory

mediators in plasma and adipose tissue are linked to obesity-related metabolic dysfunctions such as

peripheral insulin resistance (22, 55). In particular the visceral adipose tissue plays a major role in insulin

resistance (15). Obesity and energy surplus conditions also lead to central nervous system (CNS)

inflammation, which alters the regulation of energy homeostasis, i.e., modifies the function of the brain

areas involved in control of eating and energy expenditure such as the hypothalamus (6). Hypothalamic

inflammation causes a reduction of insulin and leptin signaling in the brain, thus decreasing the ability of

both hormones to reduce food intake and increase energy expenditure and, hence, protecting the

bodies energy stores (13, 37, 58). Among the adipocyte-derived pro-inflammatory factors are the

cytokines tumor-necrosis factor-alpha (TNF-α) and interleukin 6 (IL-6), key molecules in the induction of

6

peripheral inflammation (18, 59, 62). The peripheral cytokines can enter the brain through active

transport or via circumventricular organs and act directly on neurons (4, 40), but the origin of cytokines

in the CNS is controversially discussed, and it is unclear whether in obesity circulating levels of pro-

inflammatory cytokines are sufficiently increased to reach the brain or act there. Pro-inflammatory

cytokines are, however, also produced locally in the brain by microglia as well as neurons, and these

central cytokines can mediate the CNS inflammation in the metabolic syndrome (58).

1.3. TNF-α

Tumor necrosis factor alpha (TNF-α) is a pro-inflammatory cytokine that plays a key role in the acute

phase response to peripheral immune challenges (36, 42). It is secreted primarily by macrophages in

response to different noxious stimuli. Increased TNF-α levels in the plasma and adipose tissue of obese

humans and animals are implicated as a major factor in insulin resistance and the resulting T2D (38).

Peripheral infusion of TNF-α in rodents causes an impairment of insulin-stimulated uptake of glucose by

skeletal muscle (26) and general insulin resistance (24, 30, 32, 48) by inhibiting the insulin receptor

(INSR) (27). The deletion of TNF-α or its receptors in obese mice has the opposite effect and protects

from insulin resistance. In addition, neutralization of TNF-α with specific antibodies improves insulin

sensitivity in rats (28). The TNF-α acts through two receptors, type I (TNFR1) and type II (TNFR2). The

receptors form protein complexes with cytoplasmic adaptor proteins (1) that ultimately determine the

pro- or anti-apoptotic effects of TNF-α (14). TNFR1, containing the apoptotic domain, is associated with

its pro-inflammatory function (17).

The duality of the TNF-α function is clear in the CNS, where TNF-α may exert opposite effects

depending on several factors: the type of the activated receptor, the concentration of the cytokine, the

duration of the action, and the type of the target cell (17, 49). For example, TNF-α signaling via TNFR1 at

glutamatergic synapses increases neuronal excitability and may lead to excitotoxicity (3). Conversely,

TNFR2 knock-out mice are susceptible to seizures, suggesting an anticonvulsant action of TNFR2

signaling (2). “Healthy” central levels of TNF-α, secreted by astrocytes, microglia, and neurons (60),

mediate sleep and cognition, and TNF-α deficient mice exhibit sleep and cognitive dysfunctions (5, 33).

The central effects of TNF-α on eating and energy expenditure are also divergent. On one hand,

TNF-α is an essential mediator of anorexia, cachexia, and fever during the acute phase response, causing

a negative energy balance (34, 42). In line with this, centrally administered TNF-α suppresses food intake

7

in rats (43). On the other hand, HFD-induced hypothalamic inflammation with increased CNS levels of

TNF-α causes insulin and leptin resistance, promoting an increase in body weight (13, 58).

1.4. IL-6

IL-6 is another cytokine implicated in the mediation of the obesity-induced inflammation and insulin

resistance (41, 44). It has been shown that peripheral levels of IL-6 correlate with adiposity and fasting

insulin levels (29), and the chronic infusion of IL-6 in wt mice disrupted insulin signaling in the liver (31).

In contrast, mice overexpressing human IL-6 in the brain and lung, resulting in a sustained increase in

circulating IL-6, did not develop obesity when fed a HFD (47). In addition, IL-6 deficient mice develop late

onset obesity (7, 50, 51). This indicates that the actions of IL-6 in energy homeostasis are divergent.

Also, an anti-inflammatory property of muscle-derived IL-6, which is secreted in response to exercise

and enhances insulin sensitivity, adds to the complexity of this cytokine’s role in metabolism (16, 39).

Studies on the effect of IL-6 on food intake are scarce and the results are inconsistent. Some of

them show a food intake decrease in response to central administration of exogenous IL-6 in HFD-fed or

obese rats (46, 61), whereas others report no change in response to the central and peripheral

administration of IL-6 (50, 67). IL-6 receptor expression in several regions of the brain that are the site of

action of adiposity signals such as leptin and insulin (57) could provide the basis for an influence of IL-6

on food intake control.

1.5. Insulin

Insulin is a pancreatic hormone that plays a central role in metabolism and energy balance. It is secreted

by β-cells in the islets of Langerhans and enters the brain from the circulation acting on hypothalamic

neurons in the Arc, to reduce energy intake. In the periphery it regulates blood glucose, as it inhibits

hepatic gluconeogenesis and promotes glycogen synthesis and storage in liver and muscle, triglyceride

synthesis in liver and storage in adipose tissue, and amino acid storage in muscle (6). When insulin fails

to maintain normal blood glucose levels because of peripheral insulin resistance, diabetes arises (8).

Obesity, inactivity, and aging are common contributors to insulin resistance, a pathophysiological

condition in which the PI3K pathway of the insulin receptor is affected. This leads to a reduction in

GLUT4 translocation (9), impaired cellular glucose uptake and metabolism, and elevated glucose levels

as well as hyperinsulinemia (10). The long-term hyperinsulinemia leads to pancreatic beta-cell failure

and T2D (15).

8

In the brain insulin acts as an adiposity signal. Plasma and cerebrospinal fluid (CSF) insulin levels

are positively correlated with adipose tissue mass, and insulin has a long-term regulatory effect on body

fat stores and food intake (11, 18). How exactly the amount of adipose tissue affects insulin secretion is

not completely understood.

Insulin enters the CNS via a saturable, receptor-mediated transport (1). The activation of the

insulin receptors expressed on pro-opiomelanocortin (POMC) and neuropeptide Y (NPY) / agouti-related

protein (AgRP) neurons in the Arc initiates the central melanocortin system, leading to a decrease of

food intake and an increase of energy expenditure (2, 12). In experimental animals exogenous insulin

administered into the third ventricle of the brain has often been reported to reduce food intake and

increase energy expenditure (3, 4). The food intake reduction in response to insulin, however, is not

always reproducible. Several factors have been implicated in this phenomenon: learning or

conditioning, central inflammation as a result of intra-cerebro-ventricular cannulation, or various

features of the experimental paradigms (17).

HFD-feeding and obesity-induced hypothalamic inflammation are implicated in central tolerance

to insulin (7, 16). The increased levels of pro-inflammatory cytokines such as IL-6 and TNF-α in

hypothalamic inflammation may lead to an attenuation of the central effect of insulin (5, 13, 14).

1.6. Leptin

The adipocyte-derived hormone leptin is an important factor in the regulation of reproductive function,

bone metabolism, sympathetic nerve activity, neuroplasticity, food intake, and energy balance (1, 6, 14,

16).

The classic animal models used to study leptin function are ob/ob and db/db mice. ob/ob mice

are leptin-deficient due to a mutation in the leptin gene (19), and db/db mice have a mutation in the

leptin receptor gene leading to the absence of the trans-membrane sequence of the receptor and,

hence, of leptin signaling (11). Both animal models develop hyperphagia, obesity, insulin resistance and

T2D. In obese humans and animals, leptin levels are increased without a decrease in food intake,

indicating leptin resistance (8).

Among the possible mechanisms of leptin resistance are decreased leptin uptake into the CNS

(2) and increased expression of the suppressor of cytokine signaling 3 (SOCS3) (15) that mediates the

inhibition of downstream effects of the long-form of the leptin receptor LepRb (12). The LepRb binding

9

activates the Janus tyrosine kinase-2 (JAK2) and leads to phosphorylation of the signal transducer and

activator of transcription-3 (STAT3) (9) in the hypothalamus, which is associated with leptin’s role in the

control of food intake and energy balance. Leptin also activates several short forms of its receptor

(LepRa, LepRc-f), which mediate the uptake of leptin across the blood-brain-barrier (BBB) and, hence,

determine the bioavailability of leptin in the brain (3, 7).

Leptin has also been shown to be produced in the brain (13), but it is unknown whether brain-

derived leptin can act locally (6). Peripheral leptin is capable of reaching the brain. The peripheral leptin

levels are proportional to adiposity, more specifically to adipocyte size. Leptin exerts the function of an

adiposity signal by activating the pro-opiomelanocortin (POMC) and inhibiting the NPY/AgRP neurons in

the Arc, thus reducing food intake and increasing energy expenditure (5, 18). Lean or ob/ob mice

increase metabolic rate and decrease food intake in response to leptin (4, 8, 10).

Leptin is structurally similar to some cytokines and may exert pro-inflammatory function. For

example, it increases the peripheral production of monocyte TNF-α and IL-6 in humans (17). In the CNS

leptin shares action sites with other cytokines, indicating the possibility of interactions.

1.7. Melanocortin system

The central melanocortin system is a major component of the regulation of energy homeostasis.

Receptors for insulin, leptin and other cytokines are expressed in neurons of the melanocortin system.

The first order neurons of the central melanocortin system are located in the Arc of the hypothalamus.

Two subsets of Arc neurons initiate the downstream catabolic or anabolic pathways. The catabolic

neurons synthesize proopiomelanocortin (POMC), a precursor of the neurotransmitter α-melanocyte-

stimulating hormone (α-MSH). α-MSH acts at melanocortin 3 and melanocortin 4 receptors (MC3R and

MC4R) on neurons in other hypothalamic areas and in the hindbrain to reduce food intake. The

synthetic MC3R/MC4R agonist MT II causes hypophagia and weight loss in experimental animals (54).

The anabolic subset of ARC neurons synthesizes agouti-related peptide (AgRP) and neuropeptide Y

(NPY). AgRP acts as a long-term antagonist at MC3R and MC4R (36), counteracting the catabolic effects

of αMSH. NPY activates Y receptors to stimulate food intake (11, 12). AgRP or NPY administration into

the brain increases body weight and food intake (19, 20, 64, 65). Also, the optogenetic activation of

AgRP leads to hyperphagia (63).

The axons of both subsets of neurons send projections to several brain areas, such as the

ventromedial and dorsomedial hypothalamus (VMH and DMH, respectively), the paraventricular nucleus

10

(PVN), the lateral hypothalamic area (LHA), and back to the hindbrain to the Nucleus tractus solitarii

(NTS). The NTS is the primary neuroanatomical site receiving afferent information from the periphery

(52). It is also the main integrator of gastrointestinal eating-inhibitory signals and receives descending

catabolic projections from the Arc either directly or through the hypothalamic paraventricular nucleus

(PVN). PVN neurons activated by leptin send oxytocinergic projections to the NTS. Activation of oxytocin

receptors is involved in the integration of the adiposity signal leptin and the satiation signal CCK in the

brainstem NTS, and their effect on food intake (52, 56). Anabolic pathways reach the NTS from the LHA

(53). The area postrema, a circumventricular organ in close proximity to the NTS, has a leaky BBB and is

therefore able to sense blood-borne signals such as various peptide hormones (e.g. Amylin, GLP-1). The

hindbrain is important for the integration of peripheral signals in the control of food intake and for the

transmission of these signals to mid- and forebrain structures (8, 9).

1.8. Aim of the thesis

Considering the physiological relevance of the putative adiposity signals leptin and insulin and pro-

inflammatory cytokines TNF-α and IL-6 in energy homeostasis, we examined whether centrally

administered adiposity signals and pro-inflammatory cytokines interact with respect to their effects on

food intake and energy homeostasis. In a first series of experiments we assessed the effect of centrally

administrated insulin on food intake in mice. Then we assessed the central interaction between TNF-α

and leptin with regards to food intake. In the last series of our experiments we investigated the effect of

IL-6 on the eating-inhibitory action of leptin.

11

1.9. References

1. Aggarwal BB. Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol 3: 745-756, 2003. 2. Balosso S, Ravizza T, Perego C, Peschon J, Campbell IL, De Simoni MG, and Vezzani A. Tumor necrosis factor-alpha inhibits seizures in mice via p75 receptors. Annals of Neurology 57: 804-812, 2005. 3. Balosso S, Ravizza T, Pierucci M, Calcagno E, Invernizzi R, Di Giovanni G, Esposito E, and Vezzani A. Molecular and Functional Interactions between Tumor Necrosis Factor-Alpha Receptors and the Glutamatergic System in the Mouse Hippocampus: Implications for Seizure Susceptibility. Neuroscience 161: 293-300, 2009. 4. Banks WA, Kastin AJ, and Gutierrez EG. Penetration of interleukin-6 across the murine blood-brain barrier. Neuroscience Letters 179: 53-56, 1994. 5. Baune BT, Wiede F, Braun A, Golledge J, Arolt V, and Koerner H. Cognitive dysfunction in mice deficient for TNF- and its receptors. American journal of medical genetics Part B, Neuropsychiatric genetics : the official publication of the International Society of Psychiatric Genetics 147B: 1056-1064, 2008. 6. Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, and Woods SC. The catabolic action of insulin in the brain is mediated by melanocortins. Journal of Neuroscience 22: 9048-9052, 2002. 7. Benrick A, Schele E, Pinnock SB, Wernstedt-Asterholm I, Dickson SL, Karlsson-Lindahl L, and Jansson JO. Interleukin-6 gene knockout influences energy balance regulating peptides in the hypothalamic paraventricular and supraoptic nuclei. Journal of Neuroendocrinology 21: 620-628, 2009. 8. Berthoud HR. Multiple neural systems controlling food intake and body weight. Neuroscience and Biobehavioral Reviews 26: 393-428, 2002. 9. Berthoud HR, Earle T, Zheng H, Patterson LM, and Phifer C. Food-related gastrointestinal signals activate caudal brainstem neurons expressing both NMDA and AMPA receptors. Brain Research 915: 143-154, 2001. 10. Cinti S, Mitchell G, Barbatelli G, Murano I, Ceresi E, Faloia E, Wang SP, Fortier M, Greenberg AS, and Obin MS. Adipocyte death defines macrophage localization and function in adipose tissue of obese mice and humans. Journal of Lipid Research 46: 2347-2355, 2005. 11. Clark JT, Kalra PS, Crowley WR, and Kalra SP. Neuropeptide-Y and Human Pancreatic-Polypeptide Stimulate Feeding-Behavior in Rats. Endocrinology 115: 427-429, 1984. 12. Corp ES, Melville LD, Greenberg D, Gibbs J, and Smith GP. Effect of 4th Ventricular Neuropeptide-Y and Peptide-Yy on Ingestive and Other Behaviors. American Journal of Physiology 259: R317-R323, 1990. 13. De Souza CT, Araujo EP, Bordin S, Ashimine R, Zollner RL, Boschero AC, Saad MJ, and Velloso LA. Consumption of a fat-rich diet activates a proinflammatory response and induces insulin resistance in the hypothalamus. Endocrinology 146: 4192-4199, 2005. 14. Dirks AJ and Leeuwenburgh C. Tumor necrosis factor alpha signaling in skeletal muscle: effects of age and caloric restriction. J Nutr Biochem 17: 501-508, 2006. 15. Duvnjak L and Duvnjak M. The Metabolic Syndrome - an Ongoing Story. J Physiol Pharmacol 60: 19-24, 2009. 16. Ellingsgaard H, Hauselmann I, Schuler B, Habib AM, Baggio LL, Meier DT, Eppler E, Bouzakri K, Wueest S, Muller YD, Hansen AM, Reinecke M, Konrad D, Gassmann M, Reimann F, Halban PA, Gromada J, Drucker DJ, Gribble FM, Ehses JA, and Donath MY. Interleukin-6 enhances insulin secretion by increasing glucagon-like peptide-1 secretion from L cells and alpha cells. Nature Medicine 17: 1481-1489, 2011.

12

17. Figiel I. Pro-inflammatory cytokine TNF-alpha as a neuroprotective agent in the brain. Acta Neurobiologiae Experimentalis 68: 526-534, 2008. 18. Fried SK, Bunkin DA, and Greenberg AS. Omental and subcutaneous adipose tissues of obese subjects release interleukin-6: Depot difference and regulation by glucocorticoid. J Clin Endocr Metab 83: 847-850, 1998. 19. Hagan MM, Benoit SC, Rushing PA, Pritchard LM, Woods SC, and Seeley RJ. Immediate and prolonged patterns of Agouti-related peptide-(83-132)-induced c-Fos activation in hypothalamic and extrahypothalamic sites. Endocrinology 142: 1050-1056, 2001. 20. Hagan MM, Rushing PA, Pritchard LM, Schwartz MW, Strack AM, Van der Ploeg LHT, Woods SC, and Seeley RJ. Long-term orexigenic effects of AgRP-(83-132) involve mechanisms other than melanocortin receptor blockade. Am J Physiol-Reg I 279: R47-R52, 2000. 21. Hosogai N, Fukuhara A, Oshima K, Miyata Y, Tanaka S, Segawa K, Furukawa S, Tochino Y, Komuro R, Matsuda M, and Shimomura I. Adipose tissue hypoxia in obesity and its impact on adipocytokine dysregulation. Diabetes 56: 901-911, 2007. 22. Hotamisligil GS. Inflammation and metabolic disorders. Nature 444: 860-867, 2006. 23. Hotamisligil GS. Inflammation and metabolic disorders. Nature 444: 860-867, 2006. 24. Hotamisligil GS, Arner P, Caro JF, Atkinson RL, and Spiegelman BM. Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J Clin Invest 95: 2409-2415, 1995. 25. Hotamisligil GS and Erbay E. Nutrient sensing and inflammation in metabolic diseases. Nature Reviews Immunology 8: 923-934, 2008. 26. Hotamisligil GS, Murray DL, Choy LN, and Spiegelman BM. Tumor necrosis factor alpha inhibits signaling from the insulin receptor. Proceedings of the National Academy of Sciences of the United States of America 91: 4854-4858, 1994. 27. Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, and Spiegelman BM. IRS-1-mediated inhibition of insulin receptor tyrosine kinase activity in TNF-alpha- and obesity-induced insulin resistance. Science 271: 665-668, 1996. 28. Hotamisligil GS and Spiegelman BM. Tumor necrosis factor alpha: a key component of the obesity-diabetes link. Diabetes 43: 1271-1278, 1994. 29. Kaur J. A Comprehensive Review on Metabolic Syndrome. Cardiology Research and Practice 2014: 943162, 2014. 30. Kern PA, Saghizadeh M, Ong JM, Bosch RJ, Deem R, and Simsolo RB. The expression of tumor necrosis factor in human adipose tissue. Regulation by obesity, weight loss, and relationship to lipoprotein lipase. J Clin Invest 95: 2111-2119, 1995. 31. Klover PJ, Zimmers TA, Koniaris LG, and Mooney RA. Chronic exposure to interleukin-6 causes hepatic insulin resistance in mice. Diabetes 52: 2784-2789, 2003. 32. Krogh-Madsen R, Plomgaard P, Moller K, Mittendorfer B, and Pedersen BK. Influence of TNF-alpha and IL-6 infusions on insulin sensitivity and expression of IL-18 in humans. American Journal of Physiology Endocrinology and Metabolism 291: E108-114, 2006. 33. Krueger JM. The role of cytokines in sleep regulation. Current Pharmaceutical Design 14: 3408-3416, 2008. 34. Langhans W. Anorexia of infection: Current prospects. Nutrition 16: 996-1005, 2000. 35. Lazar MA. How obesity causes diabetes: Not a tall tale. Science 307: 373-375, 2005. 36. Mei J and Erlansonalbertsson C. Effect of Enterostatin Given Intravenously and Intracerebroventricularly on High-Fat Feeding in Rats. Regulatory Peptides 41: 209-218, 1992. 37. Milanski M, Degasperi G, Coope A, Morari J, Denis R, Cintra DE, Tsukumo DM, Anhe G, Amaral ME, Takahashi HK, Curi R, Oliveira HC, Carvalheira JB, Bordin S, Saad MJ, and Velloso LA. Saturated

13

fatty acids produce an inflammatory response predominantly through the activation of TLR4 signaling in hypothalamus: implications for the pathogenesis of obesity. J Neurosci 29: 359-370, 2009. 38. Nieto-Vazquez I, Fernandez-Veledo S, Kramer DK, Vila-Bedmar R, Garcia-Guerra L, and Lorenzo M. Insulin resistance associated to obesity: the link TNF-alpha. Arch Physiol Biochem 114: 183-194, 2008. 39. Pal M, Febbraio MA, and Whitham M. From cytokine to myokine: the emerging role of interleukin-6 in metabolic regulation. Immunology and Cell Biology 92: 331-339, 2014. 40. Pan W and Kastin AJ. TNF alpha transport across the blood-brain barrier is abolished in receptor knockout mice. Exp Neurol 174: 193-200, 2002. 41. Pickup JC, Chusney GD, Thomas SM, and Burt D. Plasma interleukin-6, tumour necrosis factor alpha and blood cytokine production in type 2 diabetes. Life Sciences 67: 291-300, 2000. 42. Plata-Salaman CR. Cytokines and feeding suppression: an integrative view from neurologic to molecular levels. Nutrition 11: 674-677, 1995. 43. Plata-Salaman CR. Food intake suppression by growth factors and platelet peptides by direct action in the central nervous system. Neuroscience Letters 94: 161-166, 1988. 44. Pradhan AD, Manson JE, Rifai N, Buring JE, and Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. Jama-J Am Med Assoc 286: 327-334, 2001. 45. Rajala MW and Scherer PE. Minireview: The adipocyte - At the crossroads of energy homeostasis, inflammation, and atherosclerosis. Endocrinology 144: 3765-3773, 2003. 46. Ropelle ER, Flores MB, Cintra DE, Rocha GZ, Pauli JR, Morari J, de Souza CT, Moraes JC, Prada PO, Guadagnini D, Marin RM, Oliveira AG, Augusto TM, Carvalho HF, Velloso LA, Saad MJA, and Carvalheira JBC. IL-6 and IL-10 Anti-Inflammatory Activity Links Exercise to Hypothalamic Insulin and Leptin Sensitivity through IKK beta and ER Stress Inhibition. Plos Biol 8, 2010. 47. Sadagurski M, Norquay L, Farhang J, D'Aquino K, Copps K, and White MF. Human IL6 enhances leptin action in mice. Diabetologia 53: 525-535, 2010. 48. Saghizadeh M, Ong JM, Garvey WT, Henry RR, and Kern PA. The expression of TNF alpha by human muscle. Relationship to insulin resistance. J Clin Invest 97: 1111-1116, 1996. 49. Santello M, Bezzi P, and Volterra A. TNFalpha controls glutamatergic gliotransmission in the hippocampal dentate gyrus. Neuron 69: 988-1001, 2011. 50. Schele E, Fekete C, Egri P, Fuzesi T, Palkovits M, Keller E, Liposits Z, Gereben B, Karlsson-Lindahl L, Shao R, and Jansson JO. Interleukin-6 Receptor alpha is Co-localised with Melanin-Concentrating Hormone in Human and Mouse Hypothalamus. Journal of Neuroendocrinology 24: 930-943, 2012. 51. Schobitz B, Dekloet ER, Sutanto W, and Holsboer F. Cellular-Localization of Interleukin-6 Messenger-Rna and Interleukin-6 Receptor Messenger-Rna in Rat-Brain. European Journal of Neuroscience 5: 1426-1435, 1993. 52. Schwartz GJ. Integrative capacity of the caudal brainstem in the control of food intake. Philosophical transactions of the Royal Society of London Series B, Biological Sciences 361: 1275-1280, 2006. 53. Schwartz MW, Woods SC, Porte D, Jr., Seeley RJ, and Baskin DG. Central nervous system control of food intake. Nature 404: 661-671, 2000. 54. Seeley RJ, Burklow ML, Wilmer KA, Matthews CC, Reizes O, McOsker CC, Trokhan DP, Gross MC, and Sheldon RJ. The effect of the melanocortin agonist, MT-II, on the defended level of body adiposity. Endocrinology 146: 3732-3738, 2005. 55. Shoelson SE, Lee J, and Goldfine AB. Inflammation and insulin resistance. Journal of Clinical Investigation 116: 1793-1801, 2006. 56. Sohn JW, Elmquist JK, and Williams KW. Neuronal circuits that regulate feeding behavior and metabolism. Trends in Neurosciences 36: 504-512, 2013.

14

57. Stenlof K, Wernstedt I, Fjallman T, Wallenius V, Wallenius K, and Jansson JO. Interleukin-6 levels in the central nervous system are negatively correlated with fat mass in overweight/obese subjects. The Journal of Clinical Endocrinology and Metabolism 88: 4379-4383, 2003. 58. Thaler JP, Choi SJ, Schwartz MW, and Wisse BE. Hypothalamic inflammation and energy homeostasis: resolving the paradox. Frontiers in Neuroendocrinology 31: 79-84, 2010. 59. Uysal KT, Wiesbrock SM, Marino MW, and Hotamisligil GS. Protection from obesity-induced insulin resistance in mice lacking TNF-alpha function. Nature 389: 610-614, 1997. 60. Viviani B, Bartesaghi S, Corsini E, Galli CL, and Marinovich M. Cytokines role in neurodegenerative events. Toxicol Lett 149: 85-89, 2004. 61. Wallenius K, Wallenius V, Sunter D, Dickson SL, and Jansson JO. Intracerebroventricular interleukin-6 treatment decreases body fat in rats. Biochemical and Biophysical Research Communications 293: 560-565, 2002. 62. Wellen KE and Hotamisligil GS. Inflammation, stress, and diabetes. The Journal of Clinical Investigation 115: 1111-1119, 2005. 63. Williams KW and Elmquist JK. Lighting up the hypothalamus: coordinated control of feeding behavior. Nature Neuroscience 14: 277-278, 2011. 64. Wilson BD, Ollmann MM, and Barsh GS. The role of agouti-related protein in regulating body weight. Mol Med Today 5: 250-256, 1999. 65. Zarjevski N, Cusin I, Vettor R, Rohnerjeanrenaud F, and Jeanrenaud B. Chronic Intracerebroventricular Neuropeptide-Y Administration to Normal Rats Mimics Hormonal and Metabolic Changes of Obesity. Endocrinology 133: 1753-1758, 1993.

15

Chapter 2.

Order Effects Can Obscure the Eating-Inhibitory Effect of Intracerebroventricular

Insulin in Mice

Eugenia Mc Allister, Gustavo Pacheco-Lopez, Stephen C. Woods*, Wolfgang Langhans

Physiology and Behavior Laboratory, ETH Zurich, Schwerzenbach, Switzerland

2.1. Introduction

Insulin is a putative adiposity signal and therefore one of the central regulators of energy homeostasis

that has often been shown to reduce food intake after acute and chronic intracerebroventricular (ICV)

administration (1, 5, 14). The action of insulin is mediated by the insulin receptor (IR) that is distributed

throughout the brain with high concentrations in the olfactory bulb, cerebral cortex, cerebellum,

hippocampus, and hypothalamus (12, 27). The central eating-inhibitory action of insulin is mainly

ascribed to the arcuate nucleus (Arc) of the hypothalamus and its neuropeptide system, where insulin

inhibits the anabolic neuropeptide Y (NPY)/Agouti-related peptide (AgRP) and activates catabolic pro-

opiomelanocortin (POMC) neurons. Together, these actions decrease food intake and increase energy

expenditure (3).

Although some of the insulin in the brain may be produced locally (6, 33), most of it originates

from the periphery (39). Because the IRs are also expressed on blood-brain barrier (BBB) endothelial

cells (24) peripheral insulin can reach the central nervous system (CNS) by receptor-mediated transport

across the BBB (2). The cerebrospinal fluid insulin concentration is typically multifold higher than the

peripheral concentration (10, 25). Because of insulin’s presumed role in energy homeostasis regulation,

its central effect on food intake has been studied in several species, under different experimental

paradigms, and using a wide range of doses (1, 14, 26, 44). For instance, insulin doses below 1 µU

produced a significant food intake reduction after infusion into the third cerebral ventricle (i3vt) in mice

(5, 14, 41). Also higher insulin doses (4 µU – 5000 µU) have been reported to have physiological effects

after intracerebroventricular infusions in mice (14, 20, 22), and much higher insulin doses (1, 2, 4, 8 mU)

showed an eating-inhibitory effect in rats (1), indicating that insulin can inhibit food intake under

16

different experimental conditions and in different species. Most of these studies were performed in

animals kept on regular lab chow, with relatively low contents of fat and mono- or disaccharides. Air et

al. assessed the acute effects of bolus i3vt injections of insulin on a sucrose test meal and on 24 h food

intake in rats. Offering a 15% sucrose meal during the light phase in addition to regular chow ad lib did

not change the eating-inhibitory effect of insulin, and insulin infusions decreased both sucrose and chow

intake in a dose-dependent manner (1) suggesting that insulin can also inhibit the consumption of

palatable food.

Despite some differences in the insulin doses, the mode of administration (acute and chronic)

and discrepancies in the magnitude of the observed insulin effects, all these studies showed a reliable

eating-inhibitory effect of insulin in several species under different paradigms in a more or less dose-

dependent manner. Yet, other studies failed to find a decrease in total daily food intake in rats when

insulin was chronically or acutely administered at various doses into the third or lateral cerebral

ventricle (15, 23). Central as well as peripheral insulin sensitivity may be impaired in cases of energy

surplus, contributing to metabolic disorders such as obesity and type-2-diabetes mellitus (T2D) (13, 17,

34). In these situations, insulin resistance may be related to metaflammation, i.e., to low grade,

chronically increased levels of pro-inflammatory cytokines derived from adipose tissue and infiltrating

macrophages (40). Nevertheless, similar to the eating-inhibitor effects of several other peptides, the

effect of i3vt insulin on food intake seems to be quite variable, and often it is difficult or even impossible

to identify a specific cause for this variability (43). One factor that may contribute to some of the

variability in the effect of i3vt insulin on food intake is the experimental design, i.e., between vs. within

subject designs and, in the latter case, the order of treatments. We here tested whether such factors

can modulate the effect of insulin on eating after intra third cerebral ventricular (i3vt) administration in

mice.

2.2. Materials and Methods

2.2.1. Animals and housing

Male C57BL/6J mice (13-14 weeks old) bred locally were used for the study. Breeding pairs were

originally purchased from Jackson Laboratories (Charles River Inc., Germany). The mice were single

housed in Makrolon® type III cages (Indulab, Gems, Switzerland) on pine wood chip bedding (Lignocel

hygienic animal bedding, IRS, Rosenberg, Germany). The animal room was kept under a controlled

temperature (22.5°C) with 12 h light/12 h dark cycle, 40-60% humidity, and 10/h air exchanges. In the

17

pre-surgical period mice were adapted to the handling and food restriction procedures. At the age of 14-

16 weeks (average body weight 28 g) the mice underwent the third ventricular (i3vt) cannulation

surgery. All procedures were approved by the Veterinary Office of the Canton of Zurich.

2.2.2. Diet and food intake measurements

Mice were fed a low fat diet (12% of the energy from fat; extruded KLIBA 3436, Provimi Kliba SA,

Switzerland) with a caloric density of 13.1 kJ/g = 3.1 kcal/g. For food intake measurements pre-weighed

food (3 pellets, ≈ 9g) was offered in food containers (50 ml plastic tubes; BD) and intake was calculated

by weighing the remaining food to the nearest 0.1 g using a digital balance (Mettler PM 460 Delta

Range, Switzerland) at several time points (see below). Individual food containers were assigned to each

animal. The food pellets were manipulated with forceps to avoid contamination of food with any odor.

Water was available ad libitum.

2.2.3. Third ventricular (i3vt) cannulation

Mice received antibiotics (20 mg/kg, s.c. 4µl/g, Borgal 24%, Intervet, Switzerland) and analgesics (5

mg/kg, s.c. 4µl/g, Rimadyl, Pfizer, Switzerland) approximately 30 min prior to the inhalation anesthesia

(Isoflurane; induction = 5.0% / 300 cc/min O2; maintenance = 1.5-2.5% / 200 cc/min O2). A digital

stereotaxic frame (940 Kopf Instruments, USA) was used to determine the coordinates for the

implantation of the guide cannula and the supporting screw. The guide cannula (C315GS-4-SPC; Plastic

one Inc., USA) was implanted at coordinates: A-P: -0.83, M-L: 0.0, D-V: -4.80. The sagittal sinus was

displaced laterally prior to lowering the guide cannula. Dental acrylic and cyanoacrylate glue were used

to secure the cannula to the scull.

2.2.4. Cannula patency verification

Cannula patency was examined one week after i3vt surgery and during the week after treatments in all

experiments by testing for the presence of the orexigenic effect of i3vt administerd NPY (4, 5). On test

days, mice were food deprived for 2 hours prior the i3vt NPY (1 µg / µl, Bachem AG, Switzerland) or

saline (1 µl, NaCl 0.9%, B. Braun, Switzerland) administration. Pre-weighed food was returned

immediately after the infusion and food intake was measured 2 and 22 hours later. Treatments were

given in counterbalanced order with one day between treatments. Mice underwent the adaptation

handling to the experimental conditions two days prior the experiment as well as on the intervening day

between treatments. NPY (1 µg/mouse) elicited a strong orexigenic effect (+525%) (2 h post infusion:

NPY 1.26 ± 0.10 g / Sal 0.24 ± 0.02g) (t (15) = 9.603, p ≤ 0.05). This effect was still present at 22 h (NPY

18

4.40 ± 0.17 g / Sal 3.68 ± 0.11g (t (15) = 4.826, p ≤ 0.05). The inclusion criterion was a minimum food

intake of 0.5 g at 2 h after i3vt NPY infusion (5).

2.2.5. Experiment 1

Fifteen mice were weighed and food was removed. The animals received i3vt infusions of insulin

(Actrapid, NovoNordisk Pharma AG, Switzerland, 0.4 µU / 1 µl) or an equivalent volume of vehicle

(saline; NaCl 0.9%) at 10:00, 4 h prior to dark onset. An infusion took about 30 s. The injector was kept in

the cannula for 30 s after the administration to allow enough time for the drug to diffuse. Pre-weighed

food containers were offered after 4 h of food deprivation at dark onset, and cumulative food intake

was measured after 2, 4 and 20 h. Every animal received each treatment in a counterbalanced design,

with two intervening days. Mice were adapted to the experimental procedure for two days prior to the

experiments and on the intervening days.

2.2.6. Experiment 2

In this experiment mice (n = 7) received i3vt infusions of 0.2 or 0.4 µU insulin or an equivalent volume of

vehicle in randomized order as described for Experiment 1, except that there were 7 intervening days

between infusions, with the idea to avoid order or carry-over effects.

2.2.7. Experiment 3

In this experiment we tested whether increasing the insulin dose might produce a clearer effect of

insulin on food intake that might perhaps even override any order effect. Six mice received 4.0 or 40.0

U insulin or vehicle in a within subjects design with one day between trials.

2.2.8. Experiment 4

To test whether mice that were non-responsive to insulin would still react to another anorexic agent,

the same six mice as in Experiment 3 received i3vt infusions of MTII (H-3902, Bachem AG, Switzerland, 1

µg / 1 µl), an agonist of the melanocortin receptors (MC3-R and MC4-R), or artificial cerebrospinal fluid

(aCSF, 1 µl) in a within subject design with one intervening day between injections.

2.2.9. Statistical analysis

SPSS for Windows (ver. 17.0) was used for all analyses. Data were tested for normality using the

Shapiro-Wilcoxon test. Means were compared using a paired t-test or a repeated measures ANOVA, as

19

required by the test design. If data were not normally distributed, we used the Wilcoxon Matched-Pairs

test to analyze the data. P-values ≤ 0.05 were considered significant.

2.3. Results

2.3.1. Experiment 1

The combined mean cumulative food intake of both days of the cross-over trial did not reveal a

significant effect of insulin on cumulative food intake at 2, 4, and 20 h (Fig. 1). Separate within subjects

analysis of the animals that received either insulin or vehicle on the first day, however, revealed an

order effect, i.e., a significant eating-inhibitory effect of insulin was evident in those animals that

received vehicle on the first and insulin on the second day (n= 8, Fig. 2. left). Food intake was reduced at

2 h (44% reduction), t (7) = 3.523, p ≤ 0.05, 4 h (38% reduction), t (7) = 3.029, p ≤ 0.05, and 20 h 23%

reduction), t (7) = 3.921, p ≤ 0.05. This effect was not evident in mice that received insulin on the first

and vehicle on the second day (n=7): 2 h, t (6) = -2.069, p = 0.084, 4 h, t (6) = -2.240, p = 0.066, and 20 h,

t (6) = -1.771, p = 0.127 (Fig. 2, right). Performing a between subjects comparison only on the first day

also showed a significant effect of insulin at 2 and 4 hours when compared to the food intakes of the

mice that received vehicle on the first day. In other words, in a between subjects design insulin would

have had a substantial overall effect, and only the within subjects cross-over design eliminated it.

2h 4h 20h0

1

2

3

4 Veh

Ins 0.4

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

Figure 1. I3vt insulin infusion (0.4 µU/mouse) failed to reduce food intake in mice in a within subjects cross-over design with two intervening days between trials. The graph shows the means ± SEM of all 15 mice.

20

Figure 2. I3vt insulin infusion (0.4 µU/mouse) reduced 2, 4 and 20 h food intake if the data were analyzed separately for animals that received insulin or vehicle first, i.e., there was a clear order effect. Bars represent the means ± SEM of 7 and 8 mice respectively.

2.3.2. Experiment 2

Results are reported for the 5 mice that completed all treatments. Data from 2 mice had to be excluded

because they developed cannula problems and did not receive all three treatments. Repeated measures

ANOVA indicated a significant main effect F (2, 8) = 4.828, p ≤ 0.05 for 2h cumulative food intake (6 h

post infusion). Subsequent planned contrasts revealed a significant eating-inhibitory effect for the 0.2

µU dose, F (1, 4) = 12.689, p ≤ 0.05 when compared to vehicle (24 % reduction; Fig. 3). There was no

significant effect for the 0.4 µU dose of insulin at 2 h, although a tendency (p = 0.073) was observed.

Repeated measurements ANOVA indicated no effect, neither at 4 h (p = 0.158) nor at 20 h (p = 0.650).

Veh first Ins first

21

Figure 3. I3vt insulin infusion (0.2 µU/mouse) reduced 2 h food intake in mice. The dose of 0.4 µU/mouse failed to reduce 2, 4 and 20 h food intake. Bars represent the means ± SEM of 5 animals.

2.3.3. Experiment 3

Four U insulin/mouse did not reduce food intake (Fig. 4) at 2 (Mdn = 1.01; z = - 1.60, r = - 0.92), 4 (Mdn

= 1.77; z = - 1.07, r = - 0.61), or 20 h (Mdn = 3.58; z = - 0.45, r = - 0.26) compared to vehicle (Mdn = 0.84,

1.52, 3.04 respectively). Neither did the 40.0 U insulin/mouse at 2 (Mdn = 0.90; z = - 0.54, r = -0.31), 4

(Mdn = 1.75; z = - 0.54, r = - 0.31), or 20 h (Mdn = 3.87; z = - 1.60, r = - 0.92) compared to vehicle (Mdn =

1.12, 1.68 or 2.64 respectively).

Figure 4. I3vt insulin infusion (4.0 and 40.0 µU/mouse) failed to reduce 2, 4 and 20 h food intake. Bars represent the means ± SEM of 3 animals.

2h 4h 20h0

1

2

3

4

5

Cu

mu

lati

ve f

oo

d in

take (

g)

Veh

Ins 4 U

2h 4h 20h0

1

2

3

4

5

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

Veh

Ins 40 U

2h 4h 20h0

1

2

3

4

Cu

mula

tive f

ood

inta

ke (

g) Veh

Ins 0.2 U

Ins 0.4 U

*

22

2.3.4. Experiment 4

In the same mice that did not respond to insulin (Experiment 3), MTII drastically inhibited food intake at

4 h after injection from 2.02 ± 0.11 (Control) to 0.50 ± 0.13 g (MTII) (mean ± SEM, t (5) = 8.23, p < 0.05, r

= 0.96. The paired t-test was used to analyze these data.

Figure 5. I3vt MTII (1 µg/mouse) reduced 4 h food intake in mice. Bars represent the means ± SEM of 6 animals.

2.4. Discussion

In the last decades, significant progress has been made in understanding insulin’s role in energy

homeostasis, but the results with respect to the effects of exogenous insulin on eating after central

administration are conflicting (see Table 1). It is a long recognized phenomenon that peptide effects on

eating can be variable, often without an easily discernable reason (43). Most studies so far failed to

convincingly explain this phenomenon. Also other eating-inhibitory peptides have effects of such a

probabilistic nature. Thus, the absence or presence of eating-inhibitory effects after exogenous

administration has also been reported for cholecystokinin (CCK), leptin, peptide YY (PYY) and other

peptides (18, 36). Moreover, possible interactions among peptides need to be considered. For

example, CCK’s anorectic effect may critically depend on insulin’s and leptin’s actions to increase the

central sensitivity to CCK (21, 31). In sum, the probabilistic nature of the eating-inhibitory effect is not

specific to insulin, and may depend on various different factors.

Conventionally, insulin is quantified in International Units (IU or U), a unit of measurement for

the amount of a substance based on biological activity or effect (i.e. insulin’s blood-glucose lowering

Veh MTII0.0

0.5

1.0

1.5

2.0

2.5

Cu

mu

lati

ve f

oo

d in

take (

g)

*

23

activity). Despite its name, IU is not part of the International System of Units (SI) and is difficult to

convert into SI units because conversion factors differ from one type of insulin to another (11, 35). This

complicates the comparison of insulin data from different sources and also from different publications,

and this difficulty may contribute to some of the discrepancies with respect to the effects of insulin on

food intake and body weight in the literature. Clearly, however, this is not the only reason for the

discrepant results.

Interestingly, in Experiment 1 we observed that when insulin was administrated prior to saline,

within two intervening days, the food intake after the saline infusion was lower in comparison to other

animals which received first saline infusion and then insulin. One possible explanation for this

observation might be a carryover effect (7), i.e., it is conceivable that the insulin we infused had a long-

lasting effect and, hence, still reduced food intake after vehicle administration two days later. But we

used Act-rapid® HM insulin, which supposedly acts only for about 8 hours in the periphery according to

the description by the manufacturer (NovoNordisk Pharma AG, Switzerland), i.e., it should not have such

a long-lasting carryover effect. Nevertheless, a sufficient number of intervening days between trials

should always be factored in to allow for the complete washout of any administered compound.

Another and probably more likely explanation for the order effect observed in Experiment 1 is

associative learning. We tried to mimic as closely as possible the design of Brown and colleagues (5),

which is one of the few studies in which insulin was administered into the third ventricle in mice, but the

one detail that we did not emulate was the 1 week intervening time between trials. Brown and

colleagues reported a substantial decrease in food intake in response to insulin doses above 0.05 µU /

mouse (5). We did see a strong effect of 0.4 µU insulin after accounting for the order effect. Several

factors, alone or together, might play a role for associative learning in this context (30).

1) The overall handling/experimental procedure: The animals in our study were adapted to the

i3vt infusions by prior vehicle administrations, but they were food deprived and regained access to food

at the onset of the dark phase. Thus, the food deprivation with the subsequent access to food might

have served as a conditioned stimulus for whatever the animals experienced after the i3vt insulin

infusion. Therefore, the eating-inhibitory effect of insulin might have caused the animals to eat less

after the control infusion, and this phenomenon might have disguised insulin's eating-inhibitory effect.

2) The smell of the m-cresol (3-methylfenol), the aromatic organic compound commercially used

as solvent (preservative and stabilizer) of the insulin, might have contributed to the conditioning of low

24

food intake in the control animals. In our within-subject counterbalanced experiments, mice were

exposed to the smell of m-cresol during the administration of the vehicle as well as during the insulin

infusions, but the animals that received insulin might have associated the smell of m-cresol with the

eating-inhibitory effect and this may have prompted them to eat less after the subsequent vehicle

infusion, resulting in lower food intake measures than normal and, hence, contributing to the

diminished effect of insulin. Human diabetes patients also report the smell of insulin, and in some

cases, parabens-based insulin was substituted for diabetic patients developed dermatitis as an allergic

reaction to m-cresol (28). Also, previous studies (29, 42) described the conditioned secretion of insulin

in response to external cues, such as odor (m-cresol). Thus, after pairing pharmacological doses of

exogenous insulin with unique cues these same cues elicited hypoglycemia in rats in the absence of

exogenous insulin. The magnitude of the hypoglycemia in such experiments appears to be sensitive to

the number of conditioning trials, the dose of insulin given in these trials, and the nature of the cues

(42). So, perhaps the low baseline food intake in response to saline infusion in the second trial of our

experiment was caused by a conditioned behavioral response to insulin. Considering these possibilities,

we used a longer intervening period for any further experiments we conducted with insulin, consistent

with the study done by Brown (5).

3) Another factor that could modify the eating-inhibitory effect of exogenous insulin is the meal-

anticipatory release of peptides such as GLP-1. GLP-1, released in response to the meal, interacts with

GLP-1 receptors located at the pancreatic B-cells, stimulating insulin secretion depending on the level of

glycemia (32). Exogenous GLP-1 reduces food intake in healthy and obese diabetic subjects (9). One

study, however, showed that the meal-anticipatory cephalic release of GLP-1 increased subsequent

meal size (37). Such an effect could also antagonize any eating-inhibitory effect of i3vt insulin, but as

the major reason for this failure of insulin to reduce food intake appeared to be the low food intake in

control animals (and not a high food intake in previously insulin injected animals) it appears unlikely that

this phenomenon was crucial for the lack of an eating-inhibitory effect of insulin in our experiments.

In Experiment 3, not even 10-100 times higher doses of insulin reduced food intake, suggesting

that the conditioning effects which presumably prevented the eating-inhibitory effect of insulin were

fairly strong. Additional experiments that are not reported in detail showed that the insulin used for all

experiments was biologically active because it reliably and substantially decreased blood glucose levels

after peripheral administration. Interestingly, the MT-II clearly reduced food intake after i3vt infusion in

animals that were non-responsive to i3vt insulin. MT-II is a synthetic analog of α-melanocyte-

25

stimulating hormone (α-MSH). Insulin stimulates the expression of its precursor, pro-opiomelanocortin

(POMC), in one population of neurons in the hypothalamic Arc that mediate the catabolic action of

insulin (3). That the mice still reduced food intake in response to MT-II, but did not in response to

insulin, suggests that the function of the hypothalamic melanocortin system was not impaired by

whatever antagonized the insulin effect. It also indicates that the impediment of insulin action took

place at the insulin receptors of the hypothalamic POMC neurons or in these neurons, i.e., upstream of

the MC4 receptors activated by exogenous MT-II. The significant stimulation of food intake after the

i3vt administration of NPY also indicated that the energy homeostasis regulating hypothalamic

neuropeptide systems, and in particular the anabolic pathway downstream of the pertinent NPY

receptors, was functioning properly in our animals. Moreover, it confirmed that the mice were capable

to increase food intake under our conditions.

In sum, we identified an order effect as one possible mechanism that contributed to the

obliteration of the eating-inhibitory effect of i3vt administered insulin in a regular within subject cross-

over trial. Further studies are necessary to determine the exact conditioned stimulus that caused this

order effect under our conditions.

26

Table 1. Central anorectic effects of insulin icv infusions

Insulin dose

icv Food deprivati

on (h)

Specie (strain)

[sex]

Insulin source (type)

Effects References

1 / 2 / 4 / 8 mU [single]

3V 4 Rat (Long-Evans)

[M]

Iletin II (rp)

4 and 8 mU reduced 4h food intake (1)

0.05 / 0.1 / 0.4 µU

[single]

3V 4

Mice (C57BL/6J)

[M]

Novolin (rh)

Reduced 2, 4 and 24 h food intake dose-dependently Body weight reduction

(5)

0.4µ U [single]

3V 5

Mice (C57BL/6J)

[M]

Actrapid (rh)

Reduced 12 and 24 h food intake

(14)

0.01 / 0.1 U [single]

3V 6 Rat (Dawley)

[M/F]

Eli Lilly Decrease up to 24 h food intake Increase of meal latency and size Increase of hypothalamic POMC

(16)

1.0/2.0/4.5 mU [single]

3V none Rat (Wistar)

[M]

Novo Actrapid

(rp)

2.0 mU reduced 23.5 h food intake 1.0 and 4.5 no effect

(26)

2µU in Arc/PVN 10 µU in 3V

[single]

Arc PVN 3V

none Rat (Wistar)

[M]

Novo Actrapid

2 µU in Arc reduced HFD caloric intake No effect on protein- and carbohydrate-reach diets intake

(38)

8 mU 3V 4 Rat (Long-Evans)

[M]

insulin Decreased 4 h food intake (3)

5.0/10.0/20.0 mU

[single]

3V 6 Rat (Wistar)

[M]

Eli Lilly (hr)

Decrease 12 h food intake

(8)

4, 8, 13, 16, 32, and 10 mU

3V

1, 6 Rat (Long-Evans)

(Dawley) [M]

Novo N. Actrapid

(h) Eli Lilly

(h)

No effect on food intake (15)

45/85/170/600 ng/day

[chronic]

LV none Rat (Dawley)

[M]

Actrapid (mp)

No effect on food intake and body weight

(19)

LV: Lateral ventricle, 3V: 3rd ventricle, mp: monocomponent porcine, rp: regular pork, rh: recombinant human, h: regular human, M: male, Underline: minimal effective dose, HFD: high-fat diet.

27

2.5. References

1. Air EL, Benoit SC, Blake Smith KA, Clegg DJ, and Woods SC. Acute third ventricular administration of insulin decreases food intake in two paradigms. Pharmacology, Biochemistry, and Behavior 72: 423-429, 2002. 2. Baura GD, Foster DM, Porte D, Jr., Kahn SE, Bergman RN, Cobelli C, and Schwartz MW. Saturable transport of insulin from plasma into the central nervous system of dogs in vivo. A mechanism for regulated insulin delivery to the brain. The Journal of Clinical Investigation 92: 1824-1830, 1993. 3. Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, and Woods SC. The catabolic action of insulin in the brain is mediated by melanocortins. Journal of Neuroscience 22: 9048-9052, 2002. 4. Bowen H, Mitchell TD, and Harris RB. Method of leptin dosing, strain, and group housing influence leptin sensitivity in high-fat-fed weanling mice. Am J Physiol Regul Integr Comp Physiol 284: R87-100, 2003. 5. Brown LM, Clegg DJ, Benoit SC, and Woods SC. Intraventricular insulin and leptin reduce food intake and body weight in C57BL/6J mice. Physiology & Behavior 89: 687-691, 2006. 6. Clarke DW, Mudd L, Boyd FT, Jr., Fields M, and Raizada MK. Insulin is released from rat brain neuronal cells in culture. Journal of Neurochemistry 47: 831-836, 1986. 7. Diaz-Uriarte R. Incorrect analysis of crossover trials in animal behaviour research. Anim Behav 63: 815-822, 2002. 8. Flores MB, Fernandes MF, Ropelle ER, Faria MC, Ueno M, Velloso LA, Saad MJ, and Carvalheira JB. Exercise improves insulin and leptin sensitivity in hypothalamus of Wistar rats. Diabetes 55: 2554-2561, 2006. 9. Gutzwiller JP, Goke B, Drewe J, Hildebrand P, Ketterer S, Handschin D, Winterhalder R, Conen D, and Beglinger C. Glucagon-like peptide-1: a potent regulator of food intake in humans. Gut 44: 81-86, 1999. 10. Havrankova J, Roth J, and Brownstein MJ. Concentrations of insulin and insulin receptors in the brain are independent of peripheral insulin levels. Studies of obese and streptozotocin-treated rodents. The Journal of Clinical Investigation 64: 636-642, 1979. 11. Heinemann L. Insulin assay standardization: leading to measures of insulin sensitivity and secretion for practical clinical care: response to Staten et al. Diabetes Care 33: e83; author reply e84, 2010. 12. Hopkins DF and Williams G. Insulin receptors are widely distributed in human brain and bind human and porcine insulin with equal affinity. Diabetic Medicine : a Journal of the British Diabetic Association 14: 1044-1050, 1997. 13. Hotamisligil GS. Inflammation and metabolic disorders. Nature 444: 860-867, 2006. 14. Jaillard T, Roger M, Galinier A, Guillou P, Benani A, Leloup C, Casteilla L, Penicaud L, and Lorsignol A. Hypothalamic reactive oxygen species are required for insulin-induced food intake inhibition: an NADPH oxidase-dependent mechanism. Diabetes 58: 1544-1549, 2009. 15. Jessen L, Clegg DJ, and Bouman SD. Evaluation of the lack of anorectic effect of intracerebroventricular insulin in rats. Am J Physiol Regul Integr Comp Physiol 298: R43-50, 2010. 16. Keen-Rhinehart E, Desai M, and Ross MG. Central insulin sensitivity in male and female juvenile rats. Hormones and Behavior 56: 275-280, 2009. 17. Kim F, Pham M, Maloney E, Rizzo NO, Morton GJ, Wisse BE, Kirk EA, Chait A, and Schwartz MW. Vascular inflammation, insulin resistance, and reduced nitric oxide production precede the onset of peripheral insulin resistance. Arteriosclerosis, Thrombosis, and Vascular Biology 28: 1982-1988, 2008.

28

18. le Roux CW, Aylwin SJ, Batterham RL, Borg CM, Coyle F, Prasad V, Shurey S, Ghatei MA, Patel AG, and Bloom SR. Gut hormone profiles following bariatric surgery favor an anorectic state, facilitate weight loss, and improve metabolic parameters. Annals of Surgery 243: 108-114, 2006. 19. Manin M, Balage M, Larue-Achagiotis C, and Grizard J. Chronic intracerebroventricular infusion of insulin failed to alter brain insulin-binding sites, food intake, and body weight. Journal of Neurochemistry 51: 1689-1695, 1988. 20. Morgan DA and Rahmouni K. Differential effects of insulin on sympathetic nerve activity in agouti obese mice. Journal of Hypertension 28: 1913-1919, 2010. 21. Morton GJ, Blevins JE, Williams DL, Niswender KD, Gelling RW, Rhodes CJ, Baskin DG, and Schwartz MW. Leptin action in the forebrain regulates the hindbrain response to satiety signals. The Journal of Clinical Investigation 115: 703-710, 2005. 22. Muller AP, Gnoatto J, Moreira JD, Zimmer ER, Haas CB, Lulhier F, Perry ML, Souza DO, Torres-Aleman I, and Portela LV. Exercise increases insulin signaling in the hippocampus: physiological effects and pharmacological impact of intracerebroventricular insulin administration in mice. Hippocampus 21: 1082-1092, 2011. 23. Nagaraja TN, Patel P, Gorski M, Gorevic PD, Patlak CS, and Fenstermacher JD. In normal rat, intraventricularly administered insulin-like growth factor-1 is rapidly cleared from CSF with limited distribution into brain. Cerebrospinal Fluid Research 2: 5, 2005. 24. Pardridge WM, Eisenberg J, and Yang J. Human blood-brain barrier insulin receptor. Journal of Neurochemistry 44: 1771-1778, 1985. 25. Plata-Salaman CR. Insulin in the cerebrospinal fluid. Neuroscience and Biobehavioral Reviews 15: 243-258, 1991. 26. Plata-Salaman CR, Oomura Y, and Shimizu N. Dependence of food intake on acute and chronic ventricular administration of insulin. Physiology & Behavior 37: 717-734, 1986. 27. Plum L, Schubert M, and Bruning JC. The role of insulin receptor signaling in the brain. Trends in Endocrinology and Metabolism: TEM 16: 59-65, 2005. 28. Rajpar SF, Foulds IS, Abdullah A, and Maheshwari M. Severe adverse cutaneous reaction to insulin due to cresol sensitivity. Contact Dermatitis 55: 119-120, 2006. 29. Reiss WJ. Conditioning of a hyper-insulin type of behavior in the white rat. Journal of Comparative and Pphysiological Psychology 51: 301-303, 1958. 30. Rescorla RA. Contemporary study of Pavlovian conditioning. The Spanish Journal of Psychology 6: 185-195, 2003. 31. Riedy CA, Chavez M, Figlewicz DP, and Woods SC. Central insulin enhances sensitivity to cholecystokinin. Physiology & Behavior 58: 755-760, 1995. 32. Schirra J, Katschinski M, Weidmann C, Schafer T, Wank U, Arnold R, and Goke B. Gastric emptying and release of incretin hormones after glucose ingestion in humans. The Journal of Clinical Investigation 97: 92-103, 1996. 33. Schwartz MW, Figlewicz DP, Baskin DG, Woods SC, and Porte D, Jr. Insulin in the brain: a hormonal regulator of energy balance. Endocrine Reviews 13: 387-414, 1992. 34. Shoelson SE, Lee J, and Goldfine AB. Inflammation and insulin resistance. Journal of Clinical Investigation 116: 1793-1801, 2006. 35. Staten MA, Stern MP, Miller WG, Steffes MW, and Campbell SE. Insulin assay standardization: leading to measures of insulin sensitivity and secretion for practical clinical care. Diabetes Care 33: 205-206, 2010. 36. Tschop M, Castaneda TR, Joost HG, Thone-Reineke C, Ortmann S, Klaus S, Hagan MM, Chandler PC, Oswald KD, Benoit SC, Seeley RJ, Kinzig KP, Moran TH, Beck-sickinger AG, Koglin N, Rodgers RJ, Blundell JE, Ishii Y, Beattie AH, Holch P, Allison DB, Raun K, Madsen K, Wulff BS, Stidsen CE, Birringer M, Kreuzer OJ, Schindler M, Arndt K, Rudolf K, Mark M, Deng XY, Whitcomb DC, Halem H,

29

Taylor J, Dong J, Datta R, Culler M, Craney S, Flora D, Smiley D, and Heiman ML. Physiology: does gut hormone PYY3-36 decrease food intake in rodents? Nature 430: 1 p following 165; discussion 162 p following 165, 2004. 37. Vahl TP, Drazen DL, Seeley RJ, D'Alessio DA, and Woods SC. Meal-anticipatory glucagon-like peptide-1 secretion in rats. Endocrinology 151: 569-575, 2010. 38. van Dijk G, de Groote C, Chavez M, van der Werf Y, Steffens AB, and Strubbe JH. Insulin in the arcuate nucleus of the hypothalamus reduces fat consumption in rats. Brain Research 777: 147-152, 1997. 39. van Houten M, Posner BI, Kopriwa BM, and Brawer JR. Insulin-binding sites in the rat brain: in vivo localization to the circumventricular organs by quantitative radioautography. Endocrinology 105: 666-673, 1979. 40. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, and Ferrante AW, Jr. Obesity is associated with macrophage accumulation in adipose tissue. The Journal ofClinical Investigation 112: 1796-1808, 2003. 41. Won JC, Jang PG, Namkoong C, Koh EH, Kim SK, Park JY, Lee KU, and Kim MS. Central administration of an endoplasmic reticulum stress inducer inhibits the anorexigenic effects of leptin and insulin. Obesity (Silver Spring) 17: 1861-1865, 2009. 42. Woods S. Insulin and the brain: a mutual dependency. Progress in Psychobiology and Physiological Psychology: 53 - 81, 1996. 43. Woods SC and Langhans W. Inconsistencies in the assessment of food intake. American Journal of Physiology Endocrinology and Metabolism 303: E1408-1418, 2012. 44. Woods SC, Lotter EC, McKay LD, and Porte D, Jr. Chronic intracerebroventricular infusion of insulin reduces food intake and body weight of baboons. Nature 282: 503-505, 1979.

30

Chapter 3.

Effect of Central Leptin and TNF-α on Food Intake in Mice

Eugenia Mc Allister, Gustavo Pacheco-Lopez, Wolfgang Langhans

Physiology and Behavior Laboratory, ETH Zurich, Schwerzenbach, Switzerland

3.1. Introduction

Adipose tissue functions as an energy storage and also as an endocrine organ that secretes hormones

and adipokines (64). Leptin, secreted by adipocytes, is one of the two best known adiposity signals

(insulin is the other one) (11). Leptin controls food intake and energy expenditure to maintain energy

homeostasis (18, 56, 62); its peripheral level positively correlates with adiposity, in particular with

adipocyte size. Leptin crosses the blood-brain barrier through a receptor-mediated transport

mechanism (7, 29) and acts on specific populations of neurons in the hypothalamus, the midbrain, and

the brain stem (27, 28) to inhibit food intake and to stimulate energy expenditure (36). Obese humans

and animals, however, are hyperphagic despite high circulating leptin levels. In addition to the

inefficiency of endogenous leptin to inhibit eating and to stimulate energy expenditure, the catabolic

effect of exogenous leptin is also blunted in these animals, reflecting leptin resistance (13).

A chronic positive energy balance causes obesity, i.e., it promotes growth of adipose tissue and

macrophage infiltration, resulting in an increase of peripheral levels of pro-inflammatory cytokines that

may play a role in metabolic disorders (47, 90). The inflammatory responses disrupt metabolic functions,

leading to peripheral insulin desensitization and ultimately to type-2-diabetes (7). Tumor necrosis factor-

alpha (TNF-α), is one of the pro-inflammatory cytokines that is overexpressed in the adipose tissue of

obese mice and humans. TNF-α has profound metabolic effects and contributes to insulin resistance (42,

48). Interestingly, the chronic metabolic effects of TNF-α in obesity differ from its systemic effects during

an acute phase response. Rats exposed to bacterial lipopolysaccharide (LPS), an inflammatory

component of the gram-negative bacterial cell wall, show sickness symptoms including fever,

somnolence, and a reduction in food intake (23). TNF-α is supposed to be a major mediator of these

acute phase responses (66). This duality of TNF-α action may be related to different concentrations, i.e.,

high and low circulating levels during the acute phase response and during obesity, respectively.

31

Unlike leptin, however, increased circulating levels of endogenous TNF-α during obesity may not

be sufficient to reach the central nervous system (CNS) (16, 91). Yet, TNF-α is also synthesized by

astrocytes, microglia, and neurons in the brain (2, 8, 34, 35, 45, 49, 72, 83). As in the periphery, the

function of central TNF-α is dual. Depending on the type of the predominantly activated receptor,

duration of action, the targeted cell type, and the concentration of the cytokine, TNF-α function may be

regulatory or noxious. For example, the TNF-α concentration can determine the expression of α-amino-

3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and γ-aminobutyric acid (GABAa) receptors, and

potentially lead to changes in the excitatory-inhibitory balance of synaptic networks (9, 32, 55),

including the one regulating energy homeostasis.

The central melanocortin system is being implicated in the mediation of the hypophagic effects

of leptin (30, 76). The long form of the leptin receptor (LepRb) that is associated with the central

catabolic actions of leptin is expressed in the hypothalamic arcuate nucleus (Arc). At least two subsets of

Arc neurons (POMC/CART and NPY/AgRP neurons) are activated or inhibited as a result of the leptin

binding, leading to catabolic downstream signaling (75, 92). The TNF-α type two receptor (TNFR2)

associated with the pro-inflammatory action of TNF-α is also expressed in these neurons (44, 69, 71). In

addition, the hypothalamic TNF-α levels in high fat diet (HFD) feeding and obesity (25, 46) may activate

POMC neurons similar to leptin (10).

The expression of the POMC and inhibition of the NPY genes in the Arc is regulated by signal

transduction and activator of transcription-3 (STAT3) (78). STAT3 is a common signaling molecule for

cytokine JAK/STAT type of receptors such as LepRb and TNFR2. The deletion of STAT3 specifically in the

CNS produces hyperphagia and obesity, which indicates a crucial role of STAT3 pathway signaling in the

regulation of energy homeostasis (62, 88).

In summary, (i) the increase in the CNS concentration of TNF-α in response to high fat diet (HFD)

feeding and during obesity, (ii) the leptin resistance of obese animals and individuals, (iii) the co-

expression of LepRb and TNFR2 in the brain areas associated with the regulation of energy homeostasis,

and (iv) the overlapping intracellular signaling pathways of the LepRb and the TNFR2 prompted us to

examine whether TNF-α and leptin interact with respect to their effects on food intake. To do so we

measured food intake and body weight in C57/B6 mice after intra-third-ventricular (i3vt) administration

of leptin, TNF-α, and their combination. We also examined STAT3 activation in the hypothalamus after

32

administration of these compounds, and we measured food intake in normal wild type (WT) mice and

TNF- knock-out mice in response to intraperitoneal (ip) LPS injections.

3.2. Materials and methods

3.2.1. Animals and housing

Male 11-12 week old C57BL/6J mice were purchased from Jackson Laboratories (Charles River Inc.,

Germany) or bred locally with breeding pairs that were originally purchased from Jackson Laboratories

(Charles River Inc., Germany). The mice were single housed in Makrolon® type III cages (Indulab, Gems,

Switzerland) on pine wood chip bedding (Lignocel hygienic animal bedding, IRS, Rosenberg, Germany).

The animal room was kept under a controlled 12 h light/dark cycle (lights on at 01:00 AM) with 220 lux

(approx. 60-80 lux inside the animal cage) during the light phase and 1 lux during the dark phase. The

temperature in the room was 23.0 ± 2 °C with 55% ± 5% humidity, and 10/h air exchanges. The mice

were fed chow (12% of the energy from fat; extruded KLIBA 3436, Provimi Kliba SA, Switzerland) with a

caloric density of 3.1 kcal/g, or a high fat diet (HFD) (60% of the energy from fat; extruded KLIBA 2127,

Provimi Kliba SA, Switzerland) with a caloric density of 5.3 kcal/g. Prior to surgery the mice were

adapted to handling and the experimental procedures. At the age of 14-16 weeks (average body weight

26 g) the mice underwent i3vt surgery (see below). All procedures were approved by the Veterinary

Office of the Canton Zurich.

3.2.2. Diet / food intake and body weight measurements

For food intake measurements pre-weighed food (4 pellets, ~ 13g) was offered and food intake was

calculated by weighing the remaining food to the nearest 0.01 g using a digital balance (Mettler PM 460

Delta Range, Switzerland) at several time points. Individual food containers (50 ml Falcon; Sarstedt,

Germany) were assigned to each animal. Water was constantly available ad libitum. Body weight was

measured to the nearest 0.1 g using a digital balance (Mettler PM 3000 Delta Range, Switzerland).

3.2.3. Third ventricular surgery (i3vt)

A few hours before surgery, the mice received intraperitoneal (ip) injections of

trimethoprim/sulfadoxine (20 mg/kg, Borgal 24%, Intervet, Zurich, Switzerland) and carprofen (5 mg/kg,

Norocarp, Ufamed, Sursee, Switzerland). Inhalation anesthesia was induced with 5.0% / 600 cc/min O2

and maintained with 1.5 - 2.5% / 200 cc/min O2. The mice were placed in a digital stereotaxic frame (940

Kopf Instruments, USA), and a small incision was made to expose the sagittal, coronal, and lambdoid

33

sutures of the scull. The soft tissues were removed from the skull surface with a scraper and ethanol

(70%) soaked swabs. A guide cannula (C315GS-4-SPC; Plastic one Inc., USA) was implanted into the third

cerebral ventricle using the following coordinates from Bregma: - 0.82 mm posterior, 0.00 mm lateral, +

4.80 mm ventral. A screw (4.0 mm length / 0.85 mm diameter; FST, Germany) was implanted at the

following coordinates: - 2.8 mm posterior and + 2.1 mm lateral to support the guide cannula.

Cyanoacrylate glue and dental acrylic was used to secure the cannula and the screw to the scull surface.

The wound was closed with a skin suture (6-0 Vicryl; Ethicon, Germany). Trimethoprim/sulfadoxine (20

mg/kg, Borgal 24%, Intervet, Switzerland) and carprofen (5 mg/kg, Norocarp, Ufamed, Switzerland) were

ip injected one day after the surgery. The post-surgery recovery period was 7 days.

3.2.4. Intra-third-ventricular (i3vt) infusions

I3vt infusions were done in awake mice using a micro-syringe pump controller (Micro4TM, WPI, USA)

operating a 10 µl micro-syringe (Microliter TM #701, Hamilton Bonaduz, Switzerland) attached to a silicon

tube (0.212 × 0.024 mm; Connectors, Switzerland) with an injector cannula (IC315IS-4-SPC, 33GA; Plastic

one Inc., USA). The injector cannula protruded 1.5 mm beyond the tip of the guide cannula. The infusion

volume was always 1µl with an infusion rate of 1µl/30 sec.

We infused leptin (rec Leptin (mouse), Bachem AG, Switzerland), TNF-α (Recombinant Mouse

TNF-α, Thermo Scientific, USA; former Pierce Biotechnology, Inc.), Neuropeptide Y (NPY, Bachem AG,

Switzerland), artificial cerebro-spinal fluid (aCSF) saline (NaCl 0.9%; B. Braun) based, or bovine serum

albumin (BSA, 0.01%; A-2153, Sigma-Aldrich, Switzerland) dissolved in saline (2.0 µg BSA/10 µl). TNF-α

was dissolved in saline (NaCl 0.9%) containing 2.0 µg BSA/10 µl. The injector remained in the cannula for

30 s after the infusion to allow a drug to defuse. All infusions were done during the light phase, starting

3 or 4 h before the onset of the dark phase.

3.2.5. Cannula placement verification: NPY functional test

One week after the i3vt cannula implantation, mice were adapted to the test regimen and i3vt infusion

procedure. On the test day, the mice received an i3vt infusion of NPY (1µg/1µl) or aCSF (1µl) in the

middle of the light phase after 2 h food restriction. The animals received food immediately after the

drug administration, and 2 h cumulative food intake was measured. The test was done in a counter-

balanced manner during the light phase with two intervening days between infusions. Animals whose

food intake was ≥ 0.5 g were considered to have correct cannula placement and were included in the

study as previously described (15). The experiments were done 4 - 5 days after the NPY test.

34

3.2.6. Experiments

Experiments 1-3: In these experiments food was removed 4 h prior to dark phase onset and body weight

was measured. The guide cannula cap was removed, washed in 70% ethanol and then in saline (NaCl

0.9%), and then placed back into the guide cannula. On some days the mice received i3vt infusions of

1µl of aCSF before the guide cannula cap was placed back. Pre-weighed food was distributed 5 min

before the onset of the dark phase. Food intake at was measured at 2, 4, and 20 h. The same procedures

were followed during the intervening days when only aCSF was infused. In Experiment 1, mice (n = 7)

received three doses of TNF-α (0.1, 1.0, 10.0 ng) and vehicle (0.01% BSA in saline) in a within subjects

design with 2 intervening days between trials. Experiment 2 determined whether the repetitive i3vt

administrations of TNF-α elicited desensitization, a phenomenon that has often been described for pro-

inflammatory cytokines and other inflammatory mediators (89). Mice (n = 8) received vehicle (0.01%

BSA in saline) and three repetitive infusions of 10.0 ng of TNF-α with 2 intervening days between

infusions. In Experiment 3, separate groups of mice (n = 11, 7, or 6) received four doses of leptin (0.2,

0.4, 1.0, 5.0 µg) and vehicle (aCSF). Each animal received one leptin dose and vehicle in a crossover

design. Leptin doses of 0.2, 1.0, and 5.0 µg were administered in a counterbalanced manner with 6

intervening days between trials. The dose of 0.4 µg leptin and vehicle were administered on consecutive

days.

Experiments 4-7: In these experiments, on the experimental day the food was removed from the

cages 3 h prior to the onset of the dark phase, body weight was measured, and the animals received the

i3vt infusions. One h after the onset of the dark phase, pre-weighed food was distributed and

cumulative food intake was measured at 2, 4, and 20 h. Experiment 4: mice (n = 9) received TNF-α (10.0

ng), leptin (0.4 µg), a combination of the same doses of TNF-α and leptin, and vehicle (aCSF, BSA) in a

counterbalanced manner with 6 intervening days between trails. Experiment 5: 26 mice (n = 5 or 7 per

group) fed HFD (60% kcal % fat) for 3 weeks received TNF-α (10.0 ng), leptin (0.4 µg), a combination of

the same doses of TNF-α and leptin or vehicle (aCSF, BSA) in a between subjects design. Experiment 6:

TNFKO (n = 6) and wt (n = 4) mice received two doses of TNF-α (1.0, 10.0 ng) and vehicle (0.01% BSA in

saline) in a within subjects design with 2 intervening days between trials. Experiment 7: TNFKO mice (n =

6) received three doses of leptin (0.2, 1.0, 5.0 µg) and vehicle (aCSF) in a within subjects design with 5-6

intervening days between the trials.

35

Experiment 8: To determine the effect of LPS on food intake in TNFKO mice, about 20 min

before the onset of the dark phase, mice were ip injected (0.3 ml/mouse) with saline. Pre-weighed food

was offered 5 min before dark phase onset. Cumulative food intake was measured at 2 and 4 h. The

same regimen was followed on the experimental days. The C57B6 (n = 4) and TNFKO (n = 6) mice

received ip injections (0.01ml/g) of LPS (lipopolysaccharides from E.coli, γ-irradiated, 100 µg/kg; Sigma-

Aldrich, USA) or saline in a counterbalanced manner with one intervening day between infusions. Food

was then returned and cumulative food intake was measured at 2 and 4 h.

Experiment 9: The effect of the LPS challenge on plasma levels of TNF-α was assessed in TNFKO

(n = 9) and C57BL/6J mice (n = 11). In the middle of the light phase the mice were ip injected with 100

µg/kg of LPS. Just prior to the LPS injection and 90 min thereafter, about 140 µl/mouse blood was

collected by tail vein puncture using hematocrit tubes (Na-Heparinized, Milian, Switzerland). Blood was

immediately transferred into Eppendorf tubes (0.2 ml, Thermo Scientific, UK), kept on wet ice, and

centrifuged (10 min, 10.000 rpm, 4 °C). Plasma was stored at –80 °C for later analysis of TNF-α using an

Ultra-Sensitive Kit Mouse Cytokine Assay (Meso Scale Discovery, USA). Samples were defrosted on wet

ice and centrifuged (30 sec, 10.000 rpm, 4 °C). The assay was performed according to the

manufacturer’s recommendations. Fifty-eight % of the analyses were done in duplicates, yielding a

mean intra-assay coefficient of variation of 4.9%.

Experiment 10: To determine the pSTAT3 expression in the brain in response to TNF-α, leptin,

and their combination, mice received TNF-α (10.0 ng; n = 5), leptin (0.4 µg; n = 5), a combination of the

same doses of TNF-α and leptin (n = 6), and vehicle (aCSF, BSA; n = 6). Fifteen minutes later the mice

were deeply anesthetized with an overdose of Nembutal (Abbott Laboratories, USA) and perfused

transcardially with 0.9% NaCl, followed by 4% phosphate-buffered paraformaldehyde solution

containing 15% picric acid. The brains were dissected for further immunohistochemical analysis. The

dissected brains were post-fixed in the same fixative for 6 h and processed for antigen retrieval involving

overnight incubation in citric acid buffer (pH = 4.5) followed by a 90 s microwave treatment at 480 W

according to protocols established before (84). The brains were then cryo-protected using 30% sucrose

in PBS, frozen with powdered dry ice, and stored at −80 °C until further processing. Perfused brain

samples were cut coronally at 30 μm from frozen blocks with a sliding microtome. Eight serial sections

were prepared for each animal. The sections were rinsed in PBS, and stored at −20 °C in antifreeze

solution until further processing. For immunohistochemical staining, the slices were rinsed three times

for 10 min in PBS, and blocking was done in PBS, 0.3% Triton X-100, 10% normal serum for 1 h at room

36

temperature. The following primary antibodies were used: rabbit anti-Stat3 (Cat.No. # 4904; diluted

1:2000; Cell Signaling, USA), rabbit anti phosphor Stat3 (Cat.No. # 9145; diluted 1:1000; Cell Signaling,

USA). All antibodies were diluted in PBS containing 0.3% Triton X-100 and 2% normal serum, and the

sections were incubated free-floating overnight at room temperature. After three washes with PBS (10

min each), the sections were incubated for 1 h with the biotinylated secondary antibodies diluted 1:500

in PBS containing 2% NGS and 0.3% Triton X-100. Sections were washed again three times for 10 min in

PBS and incubated with Vectastain kit (Vector Laboratories) diluted in PBS for 1 h. After three rinses in

0.1 M Tris-HCl, pH 7.4, the sections were stained with 1.25% 3,3-diaminobenzidine and 0.08% H2O2 for

10–15 min, rinsed again four times in PBS, dehydrated, and coverslipped with Eukitt (Kindler, Germany).

The numbers of STAT3 immunoreactive cells were determined by unbiased stereological

estimations using the optical fractionator method (38). Every section of a one-in-eight series was

measured with the image analysis computer software Stereo Investigator (version 6.50.1;

MicroBrightField), resulting in 3 sections per brain sample as previously described (63, 84). The following

sampling parameters were used: a fixed counting frame with a width of 30 μm and a length of 30 μm;

and a sampling grid size of 70 × 100 μm. The counting frames were placed randomly at the intersections

of the grid within the outlined structure of interest by the software. The cells were counted following

the unbiased sampling rule using the 20× oil lens (numerical aperture (NA), 0.75) and included in the

measurement when they came into focus within the optical dissector (5). All immunohistochemical

preparations were quantified in selected hypothalamic areas.

3.2.7. Statistical analysis

All data were analyzed using SPSS for Windows (ver. 17.0; SPSS, Inc., US) and tested for normality using

the Shapiro-Wilcoxon test. Extreme values (standard scores with absolute values > 1.96, i.e., p < 0.05)

were excluded. The data are expressed as means ± SEM. Means were compared using a Repeated

Measures ANOVA or Independent ANOVAs, followed by the Bonferroni correction. Datasets that were

not normally distributed were analyzed using Wilcoxon Matched-Pairs, Mann-Whitney, Kruskal-Wallis

tests or Friedman’s ANOVA. In this case the data are expressed as median with 25th/75th percentiles.

Differences were considered significant when p ≤ 0.05.

37

3.3. Results

Experiment 1. Effect of i3vt administration of TNF-α on food intake in mice. Ten ng TNF-α reduced

cumulative food intake at 2 and 4 h (Fig. 1A). Doses of 0.1 and 1.0 ng did not significantly affect food

intake.

Experiment 2. Effect of repeated i3vt administration of TNF-α on food intake in mice. Ten ng of TNF-α

reduced 2 h food intake on all three injection days, indicating the absence of desensitization (Fig. 1B).

Four h cumulative food intake was reduced on the 2nd day. On the first and third infusion days 4 h food

intake was not affected, and neither was 20 h food intake on any day, indicating that TNF-α had rather

short lasting effect.

A B

Figure 1. Effect of i3vt TNF-a infusion on mean (± SEM) chow intake. A. Mice (n = 7) received 0.1, 1.0, and 10.0 ng TNF-α with 2 intervening days between trials. The vehicle values represent the mean of two vehicle administrations before and after the TNF-α administration. * lower than 0.1 ng of TNF-α (p < 0.05). B. Mice (n = 8) received 3 repetitive infusions of 10.0 ng TNF-α (1st, 2nd, 3rd) and vehicle (BSA) with 2 intervening days between trials. * lower than Vehicle on all three infusion days (p < 0.05); & lower than Vehicle on infusion day 2 (p < 0.05).

2 h 4 h 2 0 h

0

1

2

3

4

5C

um

ula

tiv

e f

oo

d i

nta

ke

(g

)

2 n d

veh

1 st

3 rd

*

&

2 h 4 h 2 0 h

0

1

2

3

4

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

*

T N F 0 .1

veh

T N F 1 .0

T N F 1 0 .0

*

38

Experiment 3. Effect of i3vt administration of leptin on food intake in mice. Doses of 0.4, 1.0, and 5.0

µg/mouse reduced 2, 4, and 20 h cumulative food intake. The 0.2 µg dose reduced food intake at 20, but

not at 2 and 4 h (Fig. 2).

Figure 2. Effect of i3vt leptin infusion on mean (± SEM) chow intake. Mice (n = 6, 7, or 11) received 0.2, 0.4, 1.0, and 5.0 µg leptin and vehicle (aCSF) with 6 intervening days or a day apart between trials. * lower than Vehicle (p < 0.05, ** p < 0.01, *** p < 0.00005).

Experiment 4. Effect of i3vt co-administration of TNF-α and leptin on food intake in mice fed chow.

Leptin reduced 2 and 4 h cumulative food intake (Fig. 3A) compared to vehicle. TNF-α reduced 4 and 20

h cumulative food intake compared with vehicle. The combined i3vt administration of leptin and TNF-α

elicited a greater inhibition of eating than TNF-α or leptin alone at all time points of measurement.

Experiment 5. Effect of i3vt co-administration of TNF-α and leptin on food intake in mice fed HFD. TNF-α

and the combination of TNF-α and leptin reduced 2 h (Fig. 3B), but not 4 and 20 h cumulative food

intake. Leptin alone did not reduce food intake compared to vehicle at any time point. Body weight was

measured before and after 3 weeks of HFD feeding (Fig. 3C). The average increase of initial body weight

was 22.7% (Fig. 4).

0 .2 0 .4 1 .0 5 .0

0 .0

1 .0

2 .0

3 .0

4 .0

5 .0

le p t in ( g )

2 0 h

veh

lep*

*

** *

**

0 .2 0 .4 1 .0 5 .0

0 .0

0 .2

0 .4

0 .6

0 .8

1 .0

le p t in ( g )

2 h

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

*

**

** *

0 .2 0 .4 1 .0 5 .0

0 .0

0 .5

1 .0

1 .5

2 .0

le p t in ( g )

4 h

*

**

** *

39

A B

Figure 3. Effect of i3vt infusion of TNF-α, leptin, and their combination on mean (± SEM) chow or HFD intake. A. Mice (n = 9) received 10.0 ng TNF-α, 0.4 µg leptin, a combination (comb) of the same doses of TNF-α and leptin, and vehicle (aCSF, BSA) with 6 intervening days between trials. * lower than Vehicle (p ≤ 0.05); & higher than a

combination (p < 0.05). B. HFD fed mice received 10.0 ng TNF-α (n = 5), 0.4 µg leptin (n = 7), a combination (comb) of the same doses of TNF-α and leptin (n = 7) or vehicle (aCSF; n = 7) in between subject design. * lower than Vehicle (p ≤ 0.05); & higher than a combination (p < 0.05).

Figure 4. Three weeks of HFD feeding increased the mean (± SEM) body weight in mice. Mice (n = 27) were fed HFD (60% kcal % fat), and body weight was measured before (pre-HFD) and after (3w-HFD) feeding with HFD. * higher than pre-HFD body weight. (p < 0.0001).

2 h 4 h 2 0 h

0

1

2

3

4

5

veh

lep

TNF

c o m b

**

&

2 h 4 h 2 0 h

0

1

2

3

4

5

Cu

mu

lativ

e f

oo

d i

nta

ke

(g

)

*&

*&*

&

*

*

*

*

&

C h o w H F D

p r e -H F D 3 w -H F D

0

1 0

2 0

3 0

4 0

Bo

dy

we

igh

t (g

)

*

40

Experiment 6. Effect of i3vt administration of TNF-α on food intake in TNFKO mice. Ten ng TNF-α reduced

cumulative food intake in wt mice at 2 and 20 h (Fig. 5). One ng did not significantly affect food intake.

TNF-α did not reduce food intake in TNFKO mice. The control 2 h cumulative food intake was lower in

TNFKO mice than in wt mice.

Figure 5. Effect of i3vt TNF-a infusion on mean (± SEM) chow intake in TNFKO mice. TNFKO (n = 6) and wt (n = 4)

mice received 1.0 and 10.0 ng TNF-α, and vehicle (BSA) with 2 intervening days between trials. * lower than

Vehicle (p ≤ 0.05); & lower than wt (p < 0.05).

Experiment 7. Effect of i3vt administration of leptin on food intake in TNFKO mice. TNFKO mice (n = 6)

received three doses of leptin (0.2, 1.0, 5.0 µg) and vehicle (aCSF) in a within subjects design with 5-6

intervening days between the trials (Fig. 6). Five µg leptin reduced 2 and 4 h cumulative food intake.

Doses of 0.2, 1.0 µg did not significantly affect food intake.

Figure 6. Effect of i3vt leptin infusion on mean (± SEM) chow intake in TNFKO mice. TNFKO mice (n = 6) received 0.2, 1.0, and 5.0 µg leptin, and vehicle (aCSF) with 5 - 6 intervening days between trials. * lower than Vehicle (p < 0.05).

v e h tn f1 tn f1 0

0 .0

0 .5

1 .0

1 .5

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

2 h

*

&

v e h tn f1 tn f1 0

0 .0

0 .5

1 .0

1 .5

2 .0

2 .5

4 h

v e h tn f1 tn f1 0

0 .0

1 .0

2 .0

3 .0

4 .0

5 .0

wt

T N FK O

2 0 h

*

2 h 4 h 2 0 h

0

1

2

3

4

5

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

1 .0 g

veh

0 .2 g

5 .0 g

**

41

Experiment 8. LPS effect on food intake in wt and TNFKO mice. Four wt and 6 TNFKO mice received ip

injections of 0.01 ml / g saline and 0.1 µg / g LPS in counterbalanced manner with one intervening day

between injection days (Fig. 7). LPS reduced 2 and 4 h food intake in both genotypes relative to saline-

treated mice by 74% and 83%, respectively. There was no difference in food intake in response to LPS

administration between TNFKO and wt mice.

Figure 7. Effect of ip LPS injections on mean (± SEM) chow intake. TNFKO (n = 6) and wt (n = 4) mice received 0.1 µg / g LPS and 0.01 ml / g saline injections with 1 intervening day. * lower than corresponding saline-treated mice of the same genotype (p < 0.05).

Experiment 9. Effect of ip LPS on TNF-α serum concentration in TNFKO and wt mice. TNFKO and wt mice

(n = 9, 11) received ip injections of 0.1 µg / g LPS. TNF-α serum concentration was measured 90 min

after the injection. Plasma TNF-α was increased in wt animals (427.95 ± 61.66 pg / ml). Plasma TNF-α

was below the limit of detection in TNFKO mice. The basal concentration of the serum TNF-α was below

the limit of detection in both groups of animals (data are not shown).

Experiment 10. Effect of TNF-α, leptin, and their combination on pSTAT3 immunoreactivity in the brain of

wt mice. The combination of leptin and TNF-α (n = 7) increased the number of pSTAT3 expressing

neurons in the Arc (Fig. 8A) more than any other treatment. Leptin (0.4 µg; n = 5) alone also increased

s a l L P S s a l L P S

0 .0

0 .5

1 .0

1 .5

2 .0

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

wt

T N F K O

2 h 4 h

**

**

42

the number pSTAT3 expressing neurons in the Arc compared to vehicle (aCSF, BSA; n = 6), but TNF-α (10

ng; n = 5) alone did not. The combination of leptin and TNF-α also increased the number of pSTAT3

expressing cells in the VMH (Fig. 8B). Leptin and TNF-α alone did not significantly affect pSTAT3

expression in the VMH.

A B

Figure 8. Mean (± SEM) of pSTAT immunoreactive cell number in Arc and VMH in response to TNF-α and leptin co-administration. A. STAT3 activated cells in the Arc; B. STAT3 activated cells in the VMH. * higher than Vehicle (p < 0.05); & lower than a combination (p < 0.05).

3.4. Discussion

Our findings demonstrate that the i3vt co-administration of TNF-α and leptin additively inhibits food

intake in chow-fed mice (Exp. 4), whereas in HFD-fed animals, the observed food intake inhibition of the

combined administration of both compounds was primarily due to the TNF-α effect because leptin itself

did not reduce food intake (Exp. 5). HFD-induced hypothalamic inflammation is characterized by

increased levels of TNF-α and reduced leptin sensitivity (25, 62), which is consistent with the lack of a

leptin effect on food intake that we observed. The i3vt administration allowed us to deliver TNF-α and

leptin in close proximity to the neurons in the hypothalamic Arc that initiate central nervous system

anabolic or catabolic responses (14, 27, 28, 79). These neurons express TNFR2 and LepRb, which are

associated with the eating-inhibitory effects of TNF-α and leptin, respectively, suggesting that leptin and

TNF-α can interact through their effects on these neurons.

v e h le p T N F c o m b

0

2 0 0 0 0

4 0 0 0 0

6 0 0 0 0

8 0 0 0 0

pS

TA

T3

po

sit

ive

ce

lls

(m

m3

)

*

* & &

v e h le p T N F c o m b

0

1 0 0 0 0

2 0 0 0 0

3 0 0 0 0

4 0 0 0 0

&

*

A rc V M H

43

LepRb, the only isoform of the leptin receptor with signal transduction capabilities (26, 65), is

highly expressed in the CNS, specifically in the hypothalamus (17, 70). The activation of the LepRb

expressed in the Arc POMC neurons activates the catabolic pathway of the central melanocortin system.

Many studies document the resulting decrease in food intake (1, 75, 93). Accordingly, in our

experiments i3vt administered leptin dose-dependently reduced food intake in wt and TNFKO mice fed a

low fat chow diet (Exp. 3, 7). This fits leptin’s presumed role as an adiposity signal in the non-obese

condition (11). The i3vt administered leptin presumably simulated the peripheral leptin that reaches the

brain and exerts an eating-inhibitory effect. In HFD-fed animals, however, the eating-inhibitory effect of

leptin was attenuated, suggesting that HFD-fed animals were leptin resistant. SOCS3, the negative

regulator of cytokine receptor signaling and inhibitor of the LepRb signal transduction, is a factor that is

thought to play a role in the attenuated sensitivity to leptin in HFD feeding (60). Several studies show

that SOCS3 deletion in the medio-basal hypothalamus of mice fed regular chow diet led to enhanced

leptin receptor signaling, lower food intake, body weight gain, and adiposity than in controls (43). The

HFD-induced hypothalamic inflammation with the TNF-α mRNA increase (25, 87) is accompanied by an

activation of the JAK-STAT3 pathway and by an overexpression of SOCS3 (70). Therefore, an increased

hypothalamic expression of SOCS3 may have been one of the factors contributed to the inability of

leptin to reduce food intake in HFD-fed mice in our experiments (Exp. 5).

Considering the supposed overexpression of SOCS3, the negative regulator of TNF-α receptor

signaling, in HFD-fed mice, a difference in the eating-inhibitory effect of exogenous TNF-α might be

expected with feeding of different diets. Although effects should not be compared between different

experiments, our results (Exp. 4 vs. 5) suggest that TNF-α reduced food intake in chow and HFD-fed mice

similarly. A possible explanation is that the chronically increased TNF-α levels in response to HFD feeding

are still much lower than those produced by the acute supra-physiological doses used in our study. The

mice in our experiments received a dose of 10 ng, which presumably produced about 1000 times higher

local concentrations, considering that a mouse brain weighs about 4 g (85), and the estimated

physiological concentration of TNF-α in the healthy rat brain is about 25 pg/g tissue in the dark phase

(34). Compared to the amount of TNF-α administered, the endogenous levels of TNF-α produced by the

HFD exposure are presumably negligible.

Similarly, activation of LepRb and TNFR2 induced SOCS3 expression (1, 4, 8), suggesting an

additive increase in SOCS3 expression in response to the combined administration in our experiment

and perhaps an additive inhibition of the anorectic effect. However, we observed an additive inhibitory

44

effect of leptin and TNF-α on food intake. Perhaps the increased expression of SOCS3 prevented a

positive synergistic effect of leptin and TNF-α on food intake.

The eating-inhibitory effect of centrally administered TNF-α is well documented (22, 24, 53).

Accordingly, in our experiments, TNF-α dose-dependently reduced food intake in wt mice fed low-fat

chow as well as HFD (Exp. 1, 5). Paradoxically, TNF-α did not inhibit eating in TNFKO mice (Exp. 7),

indicating that these mice were resistant to the eating-inhibitory effect of centrally administrated TNF-α.

Unlike the TNFKO mice in our experiments, genetically deficient animal models usually exhibit increased

sensitivity to the deleted or missing factor. For example, leptin deficient ob/ob mice show increased

sensitivity to peripherally administered leptin’s effects on food intake and body weight (39, 40)

compared with wt animals. On the other hand, in line with our study, ip injections of IL-6 did not

decrease food intake in IL-6 knock-out mice compared with wt mice (81), suggesting a similar

phenomenon exists for IL-6, although peripheral and central administration of IL-6 may lead to different

phenomena (52). Hypothalamic NPY mRNA levels are increased in mice deficient of IL-1 and IL-6 (73),

the immune modulating cytokines that also affect metabolism. For example, IL-1 and IL-6 deficient mice

develop late-onset obesity and disturbed glucose metabolism (37, 57, 58, 86). Similar to TNF-α, IL-1 and

IL-6 are produced in the CNS, and their receptors are expressed in the hypothalamus (12, 19-21, 80, 94).

Perhaps the TNFKO mice in our study also overexpress NPY, a potent inducer of eating (33). The eating-

stimulatory effect of NPY might not be potent enough to increase eating under base line conditions, but

could diminish the eating-inhibitory action of TNF-α.

The hypophagia induced by peripheral inflammation is mediated in large part by TNF-α. TNF-α is

essential for the hypophagic effect of LPS, and TNF-α tolerant mice are unable to reduce food intake in

response to the LPS challenge (67, 74). Therefore, mice deficient of TNF-α should be unlikely to exhibit

LPS-induced hypophagia. We did, however, not observe a difference in the food intake reducing effect

of LPS between wt and TNFKO mice (Exp. 8). This was reported before (3, 31, 54) and presumably

reflects the redundant and overlapping effects of pro-inflammatory cytokines. For example, IL-1β and IL-

6 are secreted in TNF-a tolerant mice in response to LPS (5), and these cytokines can also induce a

hypophagic effect (41, 59) and therefore compensate for the lack of TNF-a. Also, IL-1β, IL-8, and TNF-α

act additively or synergistically to induce anorexia (77, 82, 95). Therefore, the LPS-induced food intake

reduction in TNFKO mice in our experiment indicates that none of the pro-inflammatory cytokines alone

is necessary for the anorectic effect of LPS; rather, different pro-inflammatory cytokines can substitute

for each-other with respect to the anorectic action in response to LPS.

45

Many cytokines, including TNF-α and leptin, a member of the cytokine superfamily, exert their

effects via the Janus kinase-signal transducer and activator of transcription (JAK-STAT) pathway (9).

Energy homeostasis is one of the functions that the JAK-STAT pathway is involved in (50). Several studies

show that LepRb activation induces STAT3 phosphorylation in the hypothalamus (51, 61), and s/s mice,

carrying a leptin receptor gene mutation with abrogated leptin-induced STAT3 signaling, are

hyperphagic (4, 6). Furthermore, the CNS specific deletion of STAT3 in mice leads to obesity and

hyperphagia, indicating a physiological role of STAT3 in the regulation of energy balance (68). Also, the

hypothalamic STAT3 activation by TNF-α (19) may contribute to the food intake suppression during an

immune response (85). Accordingly, in our study, the greater food intake inhibition by the combined

compounds perhaps reflected an additive increase in Arc pSTAT3 in response to the combined

administration of both compounds, even though TNF-α alone did not elicit a significant STAT3 activation.

This may have been due to the small number of animals in the group (Exp. 10). In general, however, the

pattern of the pSTAT3 expression in the Arc reflected the eating inhibition in response to the central

administration of leptin and TNF-α, suggesting that in the Arc STAT3 is a major intracellular mediator of

the observed eating-inhibitory effects.

In the VMH, the overall number of cells expressing pSTAT3 in response to leptin and its

combination with TNF-α was lower than in the Arc, and TNF-α virtually did not elicit pSTAT3 expression,

possibly due to the fact that the Arc and VMH tissue were of course fixed at the same time, i.e., at 15

min after compound administration. The VMH presumably lies downstream of the Arc with respect to

STAT3 activation and it may therefore require somewhat more time for the increase in pSTAT3

expression there than in the Arc. More specific time course studies would be needed to examine this

possibility.

In conclusion, we demonstrated that the centrally administered combination of TNF-α and

leptin additively inhibits food intake in mice fed a low-fat chow diet. This was consistent with the

immunohistochemical analysis of pSTAT3, the common signaling molecule for the receptors of both

cytokines, suggesting a cross-talk between central TNF-α and leptin in food intake control. This

interaction appears to be generally additive rather than synergistic.

46

3.5. References

1. Akassoglou K, Bauer J, Kassiotis G, Pasparakis M, Lassmann H, Kollias G, and Probert L.

Oligodendrocyte apoptosis and primary demyelination induced by local TNF/p55TNF receptor signaling

in the central nervous system of transgenic mice: models for multiple sclerosis with primary

oligodendrogliopathy. The American Journal of Pathology 153: 801-813, 1998.

2. Allan SM and Rothwell NJ. Cytokines and acute neurodegeneration. Nature reviews Neuroscience 2: 734-744, 2001. 3. Arsenijevic D, Garcia I, Vesin C, Vesin D, Arsenijevic Y, Seydoux J, Girardier L, Ryffel B, Dulloo A, and Richard D. Differential roles of tumor necrosis factor-alpha and interferon-gamma in mouse hypermetabolic and anorectic responses induced by LPS. Eur Cytokine Netw 11: 662-668, 2000. 4. Banks AS, Davis SM, Bates SH, and Myers MG, Jr. Activation of downstream signals by the long form of the leptin receptor. The Journal of Biological Chemistry 275: 14563-14572, 2000. 5. Banks WA, Kastin AJ, and Gutierrez EG. Penetration of interleukin-6 across the murine blood-brain barrier. Neuroscience Letters 179: 53-56, 1994. 6. Bates SH, Dundon TA, Seifert M, Carlson M, Maratos-Flier E, and Myers MG, Jr. LRb-STAT3 signaling is required for the neuroendocrine regulation of energy expenditure by leptin. Diabetes 53: 3067-3073, 2004. 7. Baumann H, Morella KK, White DW, Dembski M, Bailon PS, Kim H, Lai CF, and Tartaglia LA. The full-length leptin receptor has signaling capabilities of interleukin 6-type cytokine receptors. Proceedings of the National Academy of Sciences of the United States of America 93: 8374-8378, 1996. 8. Baune BT, Wiede F, Braun A, Golledge J, Arolt V, and Koerner H. Cognitive dysfunction in mice deficient for TNF- and its receptors. American journal of medical genetics Part B, Neuropsychiatric genetics : the official publication of the International Society of Psychiatric Genetics 147B: 1056-1064, 2008. 9. Beattie MS, Ferguson AR, and Bresnahan JC. AMPA-receptor trafficking and injury-induced cell death. The European Journal ofN 32: 290-297, 2010. 10. Begg DP and Woods SC. The endocrinology of food intake. Nat Rev Endocrinol 9: 584-597, 2013. 11. Benoit SC, Clegg DJ, Seeley RJ, and Woods SC. Insulin and leptin as adiposity signals. Recent Progress in Hormone Research 59: 267-285, 2004. 12. Benrick A, Schele E, Pinnock SB, Wernstedt-Asterholm I, Dickson SL, Karlsson-Lindahl L, and Jansson JO. Interleukin-6 gene knockout influences energy balance regulating peptides in the hypothalamic paraventricular and supraoptic nuclei. Journal of Neuroendocrinology 21: 620-628, 2009. 13. Bjorbaek C. Central leptin receptor action and resistance in obesity. Journal of investigative medicine : the official publication of the American Federation for Clinical Research 57: 789-794, 2009. 14. Boka G, Anglade P, Wallach D, Javoyagid F, Agid Y, and Hirsch EC. Immunocytochemical Analysis of Tumor-Necrosis-Factor and Its Receptors in Parkinsons-Disease. Neuroscience Letters 172: 151-154, 1994. 15. Brown LM, Clegg DJ, Benoit SC, and Woods SC. Intraventricular insulin and leptin reduce food intake and body weight in C57BL/6J mice. Physiology & Behavior 89: 687-691, 2006. 16. Chakravarty S and Herkenham M. Toll-like receptor 4 on nonhematopoietic cells sustains CNS inflammation during endotoxemia, independent of systemic cytokines. J Neurosci 25: 1788-1796, 2005. 17. Chen SC, Kochan JP, Campfield LA, Burn P, and Smeyne RJ. Splice variants of the OB receptor gene are differentially expressed in brain and peripheral tissues of mice. Journal of Receptor and Signal Transduction Research 19: 245-266, 1999.

47

18. Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephens TW, Nyce MR, Ohannesian JP, Marco CC, McKee LJ, Bauer TL, and et al. Serum immunoreactive-leptin concentrations in normal-weight and obese humans. The New England Journal of Medicine 334: 292-295, 1996. 19. Conti B, Tabarean I, Andrei C, and Bartfai T. Cytokines and fever. Frontiers in Bioscience : a Journal and Virtual Library 9: 1433-1449, 2004. 20. Cunningham ET, Jr., Wada E, Carter DB, Tracey DE, Battey JF, and De Souza EB. Distribution of type I interleukin-1 receptor messenger RNA in testis: an in situ histochemical study in the mouse. Neuroendocrinology 56: 94-99, 1992. 21. Cunningham ET, Jr., Wada E, Carter DB, Tracey DE, Battey JF, and De Souza EB. In situ histochemical localization of type I interleukin-1 receptor messenger RNA in the central nervous system, pituitary, and adrenal gland of the mouse. J Neurosci 12: 1101-1114, 1992. 22. Darling G, Fraker DL, Jensen JC, Gorschboth CM, and Norton JA. Cachectic effects of recombinant human tumor necrosis factor in rats. Cancer Research 50: 4008-4013, 1990. 23. de Carvalho Borges B, Rorato R, Uchoa ET, Marangon P, da Silva GS, de Paula FJ, Branco LG, Antunes-Rodrigues J, and Elias LL. High-fat diet induces site-specific unresponsiveness to LPS-stimulated STAT3 activation in the hypothalamus. Am J Physiol Regul Integr Comp Physiol 306: R34-44, 2014. 24. de Kloet AD, Pacheco-Lopez G, Langhans W, and Brown LM. The effect of TNFalpha on food intake and central insulin sensitivity in rats. Physiology & Behavior 103: 17-20, 2011. 25. De Souza CT, Araujo EP, Bordin S, Ashimine R, Zollner RL, Boschero AC, Saad MJ, and Velloso LA. Consumption of a fat-rich diet activates a proinflammatory response and induces insulin resistance in the hypothalamus. Endocrinology 146: 4192-4199, 2005. 26. El-Haschimi K, Pierroz DD, Hileman SM, Bjorbaek C, and Flier JS. Two defects contribute to hypothalamic leptin resistance in mice with diet-induced obesity. The Journal of Clinical Investigation 105: 1827-1832, 2000. 27. Elias CF, Kelly JF, Lee CE, Ahima RS, Drucker DJ, Saper CB, and Elmquist JK. Chemical characterization of leptin-activated neurons in the rat brain. The Journal of Comparative Neurology 423: 261-281, 2000. 28. Elmquist JK, Ahima RS, Maratos-Flier E, Flier JS, and Saper CB. Leptin activates neurons in ventrobasal hypothalamus and brainstem. Endocrinology 138: 839-842, 1997. 29. Erickson JC, Hollopeter G, and Palmiter RD. Attenuation of the obesity syndrome of ob/ob mice by the loss of neuropeptide Y. Science 274: 1704-1707, 1996. 30. Fan W, Boston BA, Kesterson RA, Hruby VJ, and Cone RD. Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature 385: 165-168, 1997. 31. Fantuzzi G, Hui Z, Faggioni R, Benigni F, Ghezzi P, Sipe JD, Shaw AR, and Dinarello CA. Effect of endotoxin in IL-1 beta-deficient mice. Journal of Immunology 157: 291-296, 1996. 32. Ferguson AR, Christensen RN, Gensel JC, Miller BA, Sun F, Beattie EC, Bresnahan JC, and Beattie MS. Cell death after spinal cord injury is exacerbated by rapid TNF alpha-induced trafficking of GluR2-lacking AMPARs to the plasma membrane. J Neurosci 28: 11391-11400, 2008. 33. Flood JF and Morley JE. Increased Food-Intake by Neuropeptide-Y Is Due to an Increased Motivation to Eat. Peptides 12: 1329-1332, 1991. 34. Floyd RA and Krueger JM. Diurnal variation of TNF alpha in the rat brain. Neuroreport 8: 915-918, 1997. 35. Frei K, Siepl C, Groscurth P, Bodmer S, Schwerdel C, and Fontana A. Antigen presentation and tumor cytotoxicity by interferon-gamma-treated microglial cells. European Journal of Immunology 17: 1271-1278, 1987. 36. Friedman JM. Obesity in the new millennium. Nature 404: 632-634, 2000.

48

37. Garcia MC, Wernstedt I, Berndtsson A, Enge M, Bell M, Hultgren O, Horn M, Ahren B, Enerback S, Ohlsson C, Wallenius V, and Jansson JO. Mature-onset obesity in interleukin-1 receptor I knockout mice. Diabetes 55: 1205-1213, 2006. 38. Gundersen HJ, Bagger P, Bendtsen TF, Evans SM, Korbo L, Marcussen N, Moller A, Nielsen K, Nyengaard JR, Pakkenberg B, and et al. The new stereological tools: disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis. APMIS : Acta Pathologica, Microbiologica, et Immunologica Scandinavica 96: 857-881, 1988. 39. Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, Burley SK, and Friedman JM. Weight-reducing effects of the plasma protein encoded by the obese gene. Science 269: 543-546, 1995. 40. Harris RB, Zhou J, Redmann SM, Jr., Smagin GN, Smith SR, Rodgers E, and Zachwieja JJ. A leptin dose-response study in obese (ob/ob) and lean (+/?) mice. Endocrinology 139: 8-19, 1998. 41. Harris RBS, Zhou J, Smagin GN, Smith SR, Rogers E, and Kuijper JL. A leptin infusion dose response study in lean and obese ob mice. Faseb Journal 11: 2037-2037, 1997. 42. Hotamisligil GS. The role of TNFalpha and TNF receptors in obesity and insulin resistance. Journal of Internal Medicine 245: 621-625, 1999. 43. Howard JK, Cave BJ, Oksanen LJ, Tzameli I, Bjorbaek C, and Flier JS. Enhanced leptin sensitivity and attenuation of diet-induced obesity in mice with haploinsufficiency of Socs3. Nature Medicine 10: 734-738, 2004. 44. Humphreys MH. The brain splits obesity and hypertension. Nature Medicine 17: 782-783, 2011. 45. Imeri L and Opp MR. How (and why) the immune system makes us sleep. Nature Reviews Neuroscience 10: 199-210, 2009. 46. Jeon BT, Jeong EA, Shin HJ, Lee Y, Lee DH, Kim HJ, Kang SS, Cho GJ, Choi WS, and Roh GS. Resveratrol attenuates obesity-associated peripheral and central inflammation and improves memory deficit in mice fed a high-fat diet. Diabetes 61: 1444-1454, 2012. 47. Kim F, Pham M, Maloney E, Rizzo NO, Morton GJ, Wisse BE, Kirk EA, Chait A, and Schwartz MW. Vascular inflammation, insulin resistance, and reduced nitric oxide production precede the onset of peripheral insulin resistance. Arteriosclerosis, Thrombosis, and Vascular Biology 28: 1982-1988, 2008. 48. Krogh-Madsen R, Plomgaard P, Moller K, Mittendorfer B, and Pedersen BK. Influence of TNF-alpha and IL-6 infusions on insulin sensitivity and expression of IL-18 in humans. American Journal of Physiology Endocrinology and Metabolism 291: E108-114, 2006. 49. Krueger JM. The role of cytokines in sleep regulation. Current Pharmaceutical Design 14: 3408-3416, 2008. 50. Ladyman SR and Grattan DR. JAK-STAT and feeding. Jak-Stat 2: e23675, 2013. 51. Ladyman SR and Grattan DR. Region-specific reduction in leptin-induced phosphorylation of signal transducer and activator of transcription-3 (STAT3) in the rat hypothalamus is associated with leptin resistance during pregnancy. Endocrinology 145: 3704-3711, 2004. 52. Langhans W. Anorexia of infection: Current prospects. Nutrition 16: 996-1005, 2000. 53. Langhans W and Hrupka B. Interleukins and tumor necrosis factor as inhibitors of food intake. Neuropeptides 33: 415-424, 1999. 54. Leon LR, Conn CA, Glaccum M, and Kluger MJ. IL-1 type I receptor mediates acute phase response to turpentine, but not lipopolysaccharide, in mice. The American Journal of Physiology 271: R1668-1675, 1996. 55. Leonoudakis D, Zhao P, and Beattie EC. Rapid tumor necrosis factor alpha-induced exocytosis of glutamate receptor 2-lacking AMPA receptors to extrasynaptic plasma membrane potentiates excitotoxicity. J Neurosci 28: 2119-2130, 2008.

49

56. Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R, Ranganathan S, and et al. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nature Medicine 1: 1155-1161, 1995. 57. Matthews VB, Allen TL, Risis S, Chan MH, Henstridge DC, Watson N, Zaffino LA, Babb JR, Boon J, Meikle PJ, Jowett JB, Watt MJ, Jansson JO, Bruce CR, and Febbraio MA. Interleukin-6-deficient mice develop hepatic inflammation and systemic insulin resistance. Diabetologia 53: 2431-2441, 2010. 58. McGillicuddy FC, Harford KA, Reynolds CM, Oliver E, Claessens M, Mills KH, and Roche HM. Lack of interleukin-1 receptor I (IL-1RI) protects mice from high-fat diet-induced adipose tissue inflammation coincident with improved glucose homeostasis. Diabetes 60: 1688-1698, 2011. 59. Molotkov A, Satoh M, and Tohyama C. Tumor growth and food intake in interleukin-6 gene knock-out mice. Cancer Lett 132: 187-192, 1998. 60. Mori H, Hanada R, Hanada T, Aki D, Mashima R, Nishinakamura H, Torisu T, Chien KR, Yasukawa H, and Yoshimura A. Socs3 deficiency in the brain elevates leptin sensitivity and confers resistance to diet-induced obesity. Nature Medicine 10: 739-743, 2004. 61. Munzberg H, Huo L, Nillni EA, Hollenberg AN, and Bjorbaek C. Role of signal transducer and activator of transcription 3 in regulation of hypothalamic proopiomelanocortin gene expression by leptin. Endocrinology 144: 2121-2131, 2003. 62. Myers MG, Jr., Leibel RL, Seeley RJ, and Schwartz MW. Obesity and leptin resistance: distinguishing cause from effect. Trends in Endocrinology and Metabolism: TEM 21: 643-651, 2010. 63. Nyffeler M, Meyer U, Yee BK, Feldon J, and Knuesel I. Maternal immune activation during pregnancy increases limbic GABAA receptor immunoreactivity in the adult offspring: implications for schizophrenia. Neuroscience 143: 51-62, 2006. 64. Ouchi N, Parker JL, Lugus JJ, and Walsh K. Adipokines in inflammation and metabolic disease. Nature Reviews Immunology 11: 85-97, 2011. 65. Pan H, Guo J, and Su Z. Advances in understanding the interrelations between leptin resistance and obesity. Physiology & Behavior, 2014. 66. Plata-Salaman CR. Cytokines and Feeding. News in physiological sciences : an international journal of physiology produced jointly by the International Union of Physiological Sciences and the American Physiological Society 13: 298-304, 1998. 67. Porter MH, Arnold M, and Langhans W. TNF-alpha tolerance blocks LPS-induced hypophagia but LPS tolerance fails to prevent TNF-alpha-induced hypophagia. The American Journal of Physiology 274: R741-745, 1998. 68. Pradhan AD, Manson JE, Rifai N, Buring JE, and Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA : the journal of the American Medical Association 286: 327-334, 2001. 69. Purkayastha S, Zhang G, and Cai DS. Uncoupling the mechanisms of obesity and hypertension by targeting hypothalamic IKK-beta and NF-kappa B. Nature Medicine 17: 883-U255, 2011. 70. Reed AS, Unger EK, Olofsson LE, Piper ML, Myers MG, Jr., and Xu AW. Functional role of suppressor of cytokine signaling 3 upregulation in hypothalamic leptin resistance and long-term energy homeostasis. Diabetes 59: 894-906, 2010. 71. Rizk NM, Joost HG, and Eckel J. Increased hypothalamic expression of the p75 tumor necrosis factor receptor in New Zealand obese mice. Hormone and metabolic research = Hormon- und Stoffwechselforschung = Hormones et Metabolisme 33: 520-524, 2001. 72. Santello M, Bezzi P, and Volterra A. TNFalpha controls glutamatergic gliotransmission in the hippocampal dentate gyrus. Neuron 69: 988-1001, 2011. 73. Schele E, Benrick A, Grahnemo L, Egecioglu E, Anesten F, Palsdottir V, and Jansson JO. Inter-relation between Interleukin (IL)-1, IL-6 and Body Fat Regulating Circuits of the Hypothalamic Arcuate Nucleus. Journal of Neuroendocrinology 25: 580-589, 2013.

50

74. Schobitz B, Dekloet ER, Sutanto W, and Holsboer F. Cellular-Localization of Interleukin-6 Messenger-Rna and Interleukin-6 Receptor Messenger-Rna in Rat-Brain. European Journal of Neuroscience 5: 1426-1435, 1993. 75. Schwartz MW, Woods SC, Porte D, Jr., Seeley RJ, and Baskin DG. Central nervous system control of food intake. Nature 404: 661-671, 2000. 76. Seeley RJ, Yagaloff KA, Fisher SL, Burn P, Thiele TE, van Dijk G, Baskin DG, and Schwartz MW. Melanocortin receptors in leptin effects. Nature 390: 349, 1997. 77. Sonti G, Ilyin SE, and Plata-Salaman CR. Anorexia induced by cytokine interactions at pathophysiological concentrations. The American Journal of Physiology 270: R1394-1402, 1996. 78. St-Pierre J and Tremblay ML. Modulation of leptin resistance by protein tyrosine phosphatases. Cell Metabolism 15: 292-297, 2012. 79. Steensberg A, van Hall G, Osada T, Sacchetti M, Saltin B, and Klarlund Pedersen B. Production of interleukin-6 in contracting human skeletal muscles can account for the exercise-induced increase in plasma interleukin-6. The Journal of Physiology 529 Pt 1: 237-242, 2000. 80. Stenlof K, Wernstedt I, Fjallman T, Wallenius V, Wallenius K, and Jansson JO. Interleukin-6 levels in the central nervous system are negatively correlated with fat mass in overweight/obese subjects. The Journal of Clinical Endocrinology and Metabolism 88: 4379-4383, 2003. 81. Swiergiel AH and Dunn AJ. Feeding, exploratory, anxiety- and depression-related behaviors are not altered in interleukin-6-deficient male mice. Behavioural Brain Research 171: 94-108, 2006. 82. Vandermeer MJM, Sweep CGJF, Pesman GJ, Borm GF, and Hermus ARMM. Synergism between Il-1-Beta and Tnf-Alpha on the Activity of the Pituitary-Adrenal Axis and on Food-Intake of Rats. Am J Physiol-Endoc M 268: E551-E557, 1995. 83. Viviani B, Bartesaghi S, Corsini E, Galli CL, and Marinovich M. Cytokines role in neurodegenerative events. Toxicol Lett 149: 85-89, 2004. 84. Vuillermot S, Joodmardi E, Perlmann T, Ogren SO, Feldon J, and Meyer U. Prenatal immune activation interacts with genetic Nurr1 deficiency in the development of attentional impairments. J Neurosci 32: 436-451, 2012. 85. Wahlsten D. Maternal Effects on Mouse-Brain Weight. Dev Brain Res 9: 215-221, 1983. 86. Wallenius V, Wallenius K, Ahren B, Rudling M, Carlsten H, Dickson SL, Ohlsson C, and Jansson JO. Interleukin-6-deficient mice develop mature-onset obesity. Nature Medicine 8: 75-79, 2002. 87. Wang X, Ge A, Cheng M, Guo F, Zhao M, Zhou X, Liu L, and Yang N. Increased hypothalamic inflammation associated with the susceptibility to obesity in rats exposed to high-fat diet. Experimental Diabetes Research 2012: 847246, 2012. 88. Wauman J and Tavernier J. Leptin receptor signaling: pathways to leptin resistance. Front Biosci (Landmark Ed) 16: 2771-2793, 2011. 89. Weingarten S, Savoldelli D, and Langhans W. Enhancement or Loss of the Hypophagic Effect of Interleukin-1 Upon Chronic Administration. Physiology & Behavior 52: 831-837, 1992. 90. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, and Ferrante AW, Jr. Obesity is associated with macrophage accumulation in adipose tissue. The Journal of Clinical Investigation 112: 1796-1808, 2003. 91. Wisse BE, Ogimoto K, Tang J, Harris MK, Jr., Raines EW, and Schwartz MW. Evidence that lipopolysaccharide-induced anorexia depends upon central, rather than peripheral, inflammatory signals. Endocrinology 148: 5230-5237, 2007. 92. Woods SC, Seeley RJ, Porte D, Jr., and Schwartz MW. Signals that regulate food intake and energy homeostasis. Science 280: 1378-1383, 1998. 93. Woods SC, Seeley RJ, Porte D, and Schwartz MW. Signals that regulate food intake and energy homeostasis. Science 280: 1378-1383, 1998.

51

94. Yabuuchi K, Minami M, Katsumata S, and Satoh M. Localization of type I interleukin-1 receptor mRNA in the rat brain. Brain research Molecular Brain Research 27: 27-36, 1994. 95. Yang ZJ, Koseki M, Meguid MM, Gleason JR, and Debonis D. Synergistic Effect of Rhtnf-Alpha and Rhil-1-Alpha in Inducing Anorexia in Rats. Am J Physiol-Reg I 267: R1056-R1064, 1994.

52

Chapter 4.

Effect of IL-6 on Central Food Inhibitory Effect of Leptin in Mice

Eugenia Mc Allister, Wolfgang Langhans

Physiology and Behavior Laboratory, ETH Zurich, Schwerzenbach, Switzerland

4.1. Introduction

Obesity is characterized by low grade chronic inflammation that may lead to type-2-diabetes (T2D) and

cardio-vascular diseases (33). In obesity, enlarged adipocytes and infiltrating macrophages secret pro-

inflammatory signals that are responsible for the observed inflammation (33). One of these signals is IL-

6, a cytokine that exerts both an inflammatory and anti-inflammatory function. The role of IL-6 in whole

body metabolism is controversial. On one hand, IL-6 is implicated in obesity-related insulin resistance

(37, 40). Its baseline plasma concentration has been shown to be positively correlated with body mass

index (BMI), fasting insulin levels, and the development of T2D (40). On the other hand, IL-6 deficient

mice develop late-onset obesity as well as disturbed glucose metabolism (5, 45, 46), indicating that IL-6

may protect from these metabolic derailments. Also, IL-6 is secreted by skeletal muscle in response to

exercise and can increase insulin sensitivity by triggering the secretion of the incretin glucagon-like

peptide-1 (GLP-1) (18, 34).

Circulating IL-6 is capable of crossing the blood-brain barrier (BBB), but the central actions of

blood-born IL-6 are presumably limited by the low levels that reach the brain (2). Yet, IL-6 can also be

synthesized within the brain (13, 51), and IL6Rα, the ligand binding part of the IL-6 receptor complex, is

expressed on NPY neurons in the hypothalamic arcuate nucleus (Arc) (44), one of the brain areas that

regulate energy homeostasis and control appetite (51). Nevertheless, the effect of IL-6 on food intake is

unclear. A few studies have reported that centrally administered IL-6 reduced food intake in rats fed a

high fat diet (HFD)(42, 53), while other studies have not found an eating-inhibitory effect of IL-6 (29, 38,

54).

Another adipokine, leptin, which is secreted from adipose tissue in proportion to adipocyte size,

targets the same brain areas as IL-6 (12, 30, 32). Activation of the leptin receptor (LepRb) initiates

catabolic and inhibits anabolic pathways, up-regulating energy expenditure and reducing food intake (9).

53

Nevertheless, plasma leptin levels positively correlate with adiposity, indicating a state of leptin

resistance in obese animals and humans, who have increased levels of leptin without anorectic response

(8, 23).

Both, leptin and IL-6 exert their actions via a common intracellular signaling cascade, the Jak2-

STAT3 pathway, indicating that the two adipokines may interact with respect to their centrally mediated

effects, including their effects on food intake. To test this hypothesis we measured food intake in C57B6

mice after the intra-third-ventricular (i3vt) administration of IL-6, leptin and their combination.

4.2. Materials and methods

4.2.1. Animals and housing

Male 12 week old C57BL/6J mice were bred locally with breeding pairs that were originally purchased

from Jackson Laboratories (Charles River Inc., Germany). The mice were single housed in Makrolon®

type III cages (Indulab, Gems, Switzerland) on pine wood chip bedding (Lignocel hygienic animal

bedding, IRS, Rosenberg, Germany). The animal room was kept under a controlled 12 h light/dark cycle

(lights on at 01:00 AM) with 220 lux (approx. 60-80 lux inside the animal cage) during the light phase and

1 lux during the dark phase. The temperature in the room was 23.0 ± 2 °C with 55% ± 5% humidity, and

10/h air exchanges. The mice were fed chow (12% of the energy from fat; extruded KLIBA 3436, Provimi

Kliba SA, Switzerland) with a caloric density of 3.1 kcal/g. After adaptation to handling the mice

underwent i3vt surgery at the age of 14 weeks (average body weight 25 g). All procedures were

approved by the Veterinary Office of the Canton of Zurich.

4.2.2. Diet / food intake and body weight measurements

For food intake measurements pre-weighed food (4 pellets, ~ 13g) was offered and food intake was

calculated by weighing the remaining food to the nearest 0.01 g using a digital balance (Mettler PM 460

Delta Range, Switzerland) at several time points. Individual food containers (50 ml Falcon; Sarstedt,

Germany) were assigned to each animal. Water was constantly available ad libitum. Body weight was

measured to the nearest 0.1 g using a digital balance (Mettler PM 3000 Delta Range, Switzerland).

4.2.3. Third ventricular surgery (i3vt)

A few hours before surgery, the mice received intraperitoneal (ip) injections of

trimethoprim/sulfadoxine (20 mg/kg, Borgal 24%, Intervet, Zurich, Switzerland) and carprofen (5 mg/kg,

Norocarp, Ufamed, Sursee, Switzerland). Inhalation anesthesia was induced with 5.0% / 600 cc/min O2

54

and maintained with 1.5 - 2.5% / 200 cc/min O2. The mice were placed in a digital stereotaxic frame (940

Kopf Instruments, USA) and a small incision was made to expose the sagittal, coronal, and lambdoid

sutures of the scull. The soft tissues were removed from the skull surface with a scraper and ethanol

(70%) soaked swabs. A guide cannula (C315GS-4-SPC; Plastic one Inc., USA) was implanted into the third

cerebral ventricle using the following coordinates from Bregma: - 0.82 mm posterior, 0.00 mm lateral, +

4.80 mm ventral. A screw (4.0 mm length / 0.85 mm diameter; FST, Germany) was implanted at the

following coordinates: - 2.8 mm posterior and + 2.1 mm lateral to support the guide cannula.

Cyanoacrylate glue and dental acrylic was used to securely fix the cannula and the screw to the scull

surface. The wound was closed with a skin suture (6-0 Vicryl; Ethicon, Germany).

Trimethoprim/sulfadoxine (20 mg/kg, Borgal 24%, Intervet, Switzerland) and carprofen (5 mg/kg,

Norocarp, Ufamed, Switzerland) were ip injected one day after the surgery. The post-surgery recovery

period was 7 days.

4.2.4. Intra-third-ventricular (i3vt) infusions

I3vt infusions were done in awake mice using a micro-syringe pump controller (Micro4TM, WPI, USA)

operating a 10 µl micro-syringe (Microliter TM #701, Hamilton Bonaduz, Switzerland) attached to a silicon

tube (0.212 × 0.024 mm; Connectors, Switzerland) with an injector cannula (IC315IS-4-SPC, 33GA; Plastic

one Inc., USA). The injector cannula protruded 1.5 mm beyond the tip of the guide cannula. The

infusion volume was always 1 µl with an infusion rate of 1 µl/30 sec.

We infused leptin (rec Leptin (mouse), Bachem AG, Switzerland), IL-6 (Recombinant Mouse IL-6,

Sigma-Aldrich, Switzerland), neuropeptide Y (NPY, Bachem AG, Switzerland), artificial cerebro-spinal

fluid (aCSF) saline (NaCl 0.9%; B. Braun) based, or bovine serum albumin (BSA, 0.01%; A-2153, Sigma-

Aldrich, Switzerland) dissolved in saline (2.0 µg BSA/10 µl). IL-6 was dissolved in saline (NaCl 0.9%)

containing 2.0 µg BSA/10 µl. The injector remained in the cannula for 30 s after the infusion to allow a

compound to defuse. All infusions were performed during the light phase, starting 3 or 4 h before the

onset of the dark phase.

4.2.5. Cannula placement verification: NPY functional test

One week after the i3vt cannula implantation, mice were adapted to the test regimen and i3vt infusion

procedure. On the test day, the mice received an i3vt infusion of NPY (1 µg/1µl) or aCSF (1 µl) after 2 h

food deprivation. The animals received food right after the drug administration, and 2 h cumulative

food intake was measured. The test was done in a counter-balanced manner during the light phase with

55

two intervening days between infusions. Animals whose food intake was ≥ 0.5 g were considered to

have correct cannula placement and were included in the study as previously described (9).

4.2.6. Experimental design

All experiments were performed after the adaptation to the experimental regimen and the i3vt infusion

procedure. In Experiments 1 and 2 food was removed 2 h prior to dark phase onset and body weight

was measured; the guide cannula cap was removed, washed in 70% ethanol and then in saline (NaCl

0.9%) to be placed back onto the guide cannula; 1µl of aCSF was infused i3vt. Pre-weighed food was

distributed 5 min before the onset of the dark phase. Food intake was measured at 2, 12, and 22 h. The

animals were considered to be adapted when their 2 h cumulative food intake after infusions was ≥

0.7g. The same procedures were followed during the intervening days.

In Experiment 3, on the experimental day, 1 h prior to the onset of the dark phase the food was

removed from the cages, body weight was measured, and the animals received the i3vt infusions of 1.0

mg MT-II or 1 µl vehicle. One and a half h after the onset of the dark phase, pre-weighed food cups

were distributed and cumulative food intake was measured at 2, 12, and 22 h.

4.2.7. Statistical analysis

The data were analyzed using SPSS for Windows (ver. 17.0; SPSS, Inc., US) and tested for normality using

the Shapiro-Wilcoxon test. The data are expressed as means ± SEM. Means were compared using a

Repeated Measures ANOVA, followed by the Bonferroni correction. Datasets that were not normally

distributed were analyzed using Friedman’s ANOVA. In this case the data are expressed as median with

25th/75th percentiles. Differences were considered significant when p ≤ 0.05.

4.3. Results

4.3.1. Experiment 1. Effect of i3vt administration of IL-6 on food intake in mice

Seventeen male mice received two doses of IL-6 (50.0, 100.0 ng) and vehicle (0.01% BSA in saline, aCSF)

in a within subjects design with 3 intervening days between trials. Fifty ng IL-6 increased cumulative

food intake at 22 h (Fig. 1). The dose of 100.0 ng IL-6 did not significantly affect food intake.

56

Figure 1. Effect of i3vt IL-6 infusion on mean (± SEM) chow intake. Mice (n = 17) received 50.0 and 100.0 ng IL-6 and vehicle (aCSF, BSA) with 3 intervening days between trials. *higher than Vehicle (p < 0.05).

4.3.2. Experiment 2. Effect of i3vt co-administration of IL-6 and leptin on food intake in

mice fed chow

Thirteen male mice received IL-6 (50.0 ng), leptin (0.4 µg), a combination of the same doses of IL-6 and

leptin, or vehicle (BSA, aCSF) in a counterbalanced manner with 3 intervening days between trials (Fig.

2). The combined i3vt administration of leptin and IL-6 reduced cumulative food intake compared with

IL-6 alone at all time points measured, and compared with leptin alone at 2 and 22 h (Fig. 2). Leptin and

IL-6 alone did not reduce cumulative food intake compared with vehicle.

2 h 1 2 h 2 2 h

0

1

2

3

4

5

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

veh

IL -6 5 0

IL -6 1 0 0

*

57

Figure 2. Effect of i3vt infusion of IL-6, leptin, and their combination on mean (± SEM) chow intake. Mice (n = 13) received 50.0 ng IL-6, 0.4 µg leptin, a combination (comb) of the same doses of IL-6 and leptin, and vehicle (aCSF, BSA) with 3 intervening days between trials. * lower than Vehicle (p ≤ 0.05); & lower than leptin (p < 0.05), # lower than IL-6 (p < 0.05).

4.3.3. Experiment 3. Effect of i3vt administration of MT-II on food intake in mice

Twelve mail mice received MT-II (1.0 ng) and vehicle (aCSF) in a within subjects design with 2 intervening

days between trials. One ng MT-II reduced cumulative food intake at 2 and 4 h (Fig. 3).

Figure 3. Effect of i3vt MT-II infusion on mean (± SEM) chow intake. Mice (n = 12) received 1.0 ng MT-II and vehicle (aCSF) with 2 intervening days between trials. * lower than Vehicle (p ≤ 0.05).

2 h 1 2 h 2 2 h

0

1

2

3

4

5

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

veh

lep

IL -6

c o m b

*& #

*& #

*#

2 h 4 h

0 .0

0 .5

1 .0

1 .5

Cu

mu

lati

ve

fo

od

in

tak

e (

g)

veh

M T -II

*

*

58

4.4. Discussion

IL-6 exerts pleiotropic effects on energy homeostasis. In our experiments, the administration of IL-6 into

the third cerebral ventricle increased food intake at 22 h after administration (Exp.1), and the co-

administration of IL-6 with leptin inhibited food intake from 2 to 22 h after administration in chow-fed

mice, whereas neither IL-6 nor leptin infused alone was effective in this experiment (Exp. 2). An effect of

IL-6 on eating behavior has been previously described in only a few studies, perhaps due to the small

behavioral response compared with other cytokines. The findings are contradictory. Some studies

showed a reduction of food intake in response to central administration of IL-6 in obese and HFD fed

rats (42, 53) whereas others report no change in food intake after central IL-6 administration in chow-

fed animals (29, 38, 54).

In our first experiment, IL-6 increased food intake only late after administration in the cycle (at

22 h). At present we do not have any plausible explanation for this delayed effect. Looking at the figures

it seems, however, as if the effect slowly accumulated until it finally reached significance because food

intake tended to be somewhat higher at 12 h after IL-6 than after vehicle administration, although at

that time the difference did not reach statistical significance yet. IL-6, secreted by adipocytes and

macrophages that infiltrated the adipose tissue, is one of the pro-inflammatory cytokines implied in the

development of the metabolic syndrome and peripheral insulin resistance (16, 41). The peripheral levels

of IL-6 are increased in obese subjects (19, 59), which may contribute to the reduction of insulin

signaling in the periphery and lead to chronic polyphagia. Adipocytes cannot store fat in the absence of

insulin signaling, leading to a reduction of the adiposity signals to the brain and a subsequent increase in

eating (3, 49). Also the diminished insulin signaling in the brain by IL-6 may increase food intake. The

reduced insulin signaling could convey the message about decreased adiposity, to which the brain might

respond by increasing food intake (10, 58).

In contrast with adipocyte-derived IL-6, IL-6 secreted from muscle during exercise increases

peripheral insulin sensitivity (18, 34-36, 50). Muscle-derived IL-6 is also implicated in leptin sensitivity.

Thus, exercised animals exhibited increased sensitivity to leptin’s eating-inhibitory effect, and anti-IL-6

pretreated animals did not reduce food intake in response to centrally administrated leptin (21). The

difference between the effects of muscle and adipocyte-derived IL-6 on sensitivity to leptin may depend

on the levels of suppressor of cytokine signaling-3 (SOCS3), a suppressor of leptin signaling. The

chronically increased levels of adipocyte-derived IL-6 result in an over-expression of SOCS3, which could

contribute to leptin intolerance. Perhaps the transient IL-6 secretion by muscle fails to significantly

59

activate SOCS3 enough to cause an inhibition of leptin signaling (34). In addition, IL-6 deficient mice (IL-

6-/-) develop late-onset obesity with increased levels of leptin and decreased sensitivity to leptin (54),

indicating leptin resistance. In the same study IL-6 administration partially reversed the increased body

adiposity. Therefore, the origin of IL-6 may define its effect on metabolic parameters and modulate the

sensitivity to leptin.

Our findings, obtained with a different approach, are consistent with a leptin-sensitizing effect

of IL-6. We observed that IL-6 and leptin co-administration into the third brain ventricle inhibited food

intake in chow-fed mice, whereas neither IL-6 nor leptin infused alone was effective in this experiment

(Exp. 2). IL-6 and the mRNA of its receptor are expressed by microglia, astrocytes, and neurons (6, 55,

57) in various brain areas (25) including the hypothalamus (24, 47). The neuronal pathways that

maintain the balance between energy intake and expenditure are tightly regulated (1, 4). The

hypothalamic melanocortin system expresses IL-6 and its receptor (IL6Rα) in the hypothalamic arcuate

(Arc)(44), ventromedial (VMH), and dorsomedial (DMH) nuclei (46, 48). The same nuclei express the

form of the leptin receptor associated with the catabolic effect of leptin (LepRb) (14, 31), indicating the

possibility of an interaction between leptin and IL-6. In addition, leptin treated IL-6-/- mice showed no

decrease in food intake or body weight compared with vehicle treated IL-6-/- (54), suggesting a critical

permissive role of IL-6 in the eating-inhibitory effect of leptin.

The intracellular mechanism leading to the central synergistic interaction between IL-6 and

leptin should be further investigated. Nevertheless, some studies imply that IL-6 exerts its effect on the

central pathway that control food intake by influencing leptin signaling. Thus, leptin-deficient (ob/ob)

mice and ob/ob mice with sustained circulating human IL-6 (hIL-6) were equally hyperphagic, whereas

transgenic mice with sustained circulating hIL-6 fed HFD had nearly normal expression of POMC and

AgRP and consumed less food (43). Another study showed an increase of the JAK2/STAT3 pathway

activation in the hypothalamus of exercised animals in response to central leptin (21), supporting the

presumption that increased circulating IL-6 modulates LepRb signaling.

Leptin is well known to inhibit food intake by activating hypothalamic leptin receptors (11, 28).

In our experiment, however, centrally administrated leptin did not reduce food intake. Leptin resistance

is often observed in obese and HFD fed animals (23, 32). Still, the mice in our experiment were kept on

standard chow with only 12% of calories accountable for by the fat content of the diet and they had a

body weight of a lean adult male mouse, i.e., 27 g on average. Therefore, the absence of the food intake

inhibition in our experiments can scarcely be associated with leptin resistance. In our earlier study leptin

60

reliably reduced food intake (see Chapter 3 of this thesis). This discrepancy between the present and our

previously obtained results may be explained by differences in compound activity, the doses or the

diluents, the handling of the animals, the habituation conditions, the stress level of the animals, and

other factors (56). In our case, a different experimental paradigm in both studies was presumably also

involved. In the present experiment, food was distributed at the same time throughout the experiment,

i.e., always 5 min before the onset of the dark phase. In previous experiments, food was distributed 1 h

after the onset of the dark phase on the days when leptin was administered. This shift in food

availability might have prevented a learning or conditioning effect. Learning based on the experience of

environmental factors may affect the behavioral response to exogenously administered compounds that

inhibit eating. Thus, when cholecystokinin (CCK) was administered on several consecutive days in rats,

the response to its satiating action was reduced due to learning (17, 26). Therefore, such learning

phenomena should be considered in designing experimental paradigms.

We could not replicate the delayed eating-stimulatory effect of IL-6 in our second experiment.

As these animals, however, were not naïve to the central administration of IL-6, it is possible that they

developed some kind of IL-6 tolerance. Cytokines such as IL-6 are known to induce tolerance after

repetitive administration. For example, pre-stimulation with IL-6 renders hepatocytes less sensitive to

further stimulation with the same cytokine (20). The anorexigenic effect of TNF-α administered

peripherally is also subject to a desensitization phenomenon (7, 15, 22, 27, 39, 52). In a previous study

we also found that after repetitive central infusions the anorexigenic potency of TNF-α was diminished

(unpublished data).

In summary, this study aimed at understanding the central effect of IL-6 on food intake and its

potential interaction with the adiposity signal leptin. We found that the IL-6 effect may differ depending

on whether it acts by itself or in combination with leptin. In any case, our findings support the

pleiotropic role of IL-6 in energy homeostasis regulation and food intake control.

61

4.5. References

1. Air EL, Benoit SC, Coolen LM, Strauss R, Jackman A, Seeley RJ, and Woods SC. Melanocortins mediate insulin-induced hypophagia. Diabetes 51: A40-A41, 2002. 2. Banks WA, Kastin AJ, and Gutierrez EG. Penetration of interleukin-6 across the murine blood-brain barrier. Neuroscience Letters 179: 53-56, 1994. 3. Begg DP and Woods SC. The endocrinology of food intake. Nat Rev Endocrinol 9: 584-597, 2013. 4. Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, and Woods SC. The catabolic action of insulin in the brain is mediated by melanocortins. Journal of Neuroscience 22: 9048-9052, 2002. 5. Benrick A, Schele E, Pinnock SB, Wernstedt-Asterholm I, Dickson SL, Karlsson-Lindahl L, and Jansson JO. Interleukin-6 gene knockout influences energy balance regulating peptides in the hypothalamic paraventricular and supraoptic nuclei. Journal of Neuroendocrinology 21: 620-628, 2009. 6. Benveniste EN, Sparacio SM, Norris JG, Grenett HE, and Fuller GM. Induction and regulation of interleukin-6 gene expression in rat astrocytes. Journal of Neuroimmunology 30: 201-212, 1990. 7. Bernstein IL, Taylor EM, and Bentson KL. TNF-induced anorexia and learned food aversions are attenuated by area postrema lesions. Am J Physiol 260: R906-910, 1991. 8. Bjorbaek C. Central leptin receptor action and resistance in obesity. Journal of investigative medicine : the official publication of the American Federation for Clinical Research 57: 789-794, 2009. 9. Brown LM, Clegg DJ, Benoit SC, and Woods SC. Intraventricular insulin and leptin reduce food intake and body weight in C57BL/6J mice. Physiology & Behavior 89: 687-691, 2006. 10. Bruning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, Klein R, Krone W, Muller-Wieland D, and Kahn CR. Role of brain insulin receptor in control of body weight and reproduction. Science 289: 2122-2125, 2000. 11. Campfield LA, Smith FJ, Guisez Y, Devos R, and Burn P. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 269: 546-549, 1995. 12. Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephens TW, Nyce MR, Ohannesian JP, Marco CC, McKee LJ, Bauer TL, and et al. Serum immunoreactive-leptin concentrations in normal-weight and obese humans. The New England Journal of Medicine 334: 292-295, 1996. 13. Conti B, Tabarean I, Andrei C, and Bartfai T. Cytokines and fever. Frontiers in Bioscience : a Journal and Virtual Library 9: 1433-1449, 2004. 14. Cowley MA, Smart JL, Rubinstein M, Cerdan MG, Diano S, Horvath TL, Cone RD, and Low MJ. Leptin activates anorexigenic POMC neurons through a neural network in the arcuate nucleus. Nature 411: 480-484, 2001. 15. Darling G, Fraker DL, Jensen JC, Gorschboth CM, and Norton JA. Cachectic effects of recombinant human tumor necrosis factor in rats. Cancer Res 50: 4008-4013, 1990. 16. Diamant M, Lamb HJ, van de Ree MA, Endert EL, Groeneveld Y, Bots ML, Kostense PJ, and Radder JK. The association between abdominal visceral fat and carotid stiffness is mediated by circulating inflammatory markers in uncomplicated type 2 diabetes. The Journal of Clinical Endocrinology and Metabolism 90: 1495-1501, 2005. 17. Duncan EA, Davita G, and Woods SC. Changes in the satiating effect of cholecystokinin over repeated trials. Physiology & Behavior 85: 387-393, 2005. 18. Ellingsgaard H, Hauselmann I, Schuler B, Habib AM, Baggio LL, Meier DT, Eppler E, Bouzakri K, Wueest S, Muller YD, Hansen AM, Reinecke M, Konrad D, Gassmann M, Reimann F, Halban PA, Gromada J, Drucker DJ, Gribble FM, Ehses JA, and Donath MY. Interleukin-6 enhances insulin secretion by increasing glucagon-like peptide-1 secretion from L cells and alpha cells. Nature Medicine 17: 1481-1489, 2011.

62

19. Esposito K, Pontillo A, Di Palo C, Giugliano G, Masella M, Marfella R, and Giugliano D. Effect of weight loss and lifestyle changes on vascular inflammatory markers in obese women: a randomized trial. JAMA : the journal of the American Medical Association 289: 1799-1804, 2003. 20. Fischer P, Lehmann U, Sobota RM, Schmitz J, Niemand C, Linnemann S, Haan S, Behrmann I, Yoshimura A, Johnston JA, Muller-Newen G, Heinrich PC, and Schaper F. The role of the inhibitors of interleukin-6 signal transduction SHP2 and SOCS3 for desensitization of interleukin-6 signalling. Biochemical Journal 378: 449-460, 2004. 21. Flores MB, Fernandes MF, Ropelle ER, Faria MC, Ueno M, Velloso LA, Saad MJ, and Carvalheira JB. Exercise improves insulin and leptin sensitivity in hypothalamus of Wistar rats. Diabetes 55: 2554-2561, 2006. 22. Fraker DL, Stovroff MC, Merino MJ, and Norton JA. Tolerance to tumor necrosis factor in rats and the relationship to endotoxin tolerance and toxicity. J Exp Med 168: 95-105, 1988. 23. Friedman JM and Halaas JL. Leptin and the regulation of body weight in mammals. Nature 395: 763-770, 1998. 24. Gadient RA and Otten U. Differential expression of interleukin-6 (IL-6) and interleukin-6 receptor (IL-6R) mRNAs in rat hypothalamus. Neuroscience Letters 153: 13-16, 1993. 25. Gadient RA and Otten U. Expression of interleukin-6 (IL-6) and interleukin-6 receptor (IL-6R) mRNAs in rat brain during postnatal development. Brain Research 637: 10-14, 1994. 26. Goodison T and Siegel S. Learning and Tolerance to the Intake Suppressive Effect of Cholecystokinin in Rats. Behavioral Neuroscience 109: 62-70, 1995. 27. Grunfeld C, Wilking H, Neese R, Gavin LA, Moser AH, Gulli R, Serio MK, and Feingold KR. Persistence of the hypertriglyceridemic effect of tumor necrosis factor despite development of tachyphylaxis to its anorectic/cachectic effects in rats. Cancer Res 49: 2554-2560, 1989. 28. Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, Burley SK, and Friedman JM. Weight-Reducing Effects of the Plasma-Protein Encoded by the Obese Gene. Science 269: 543-546, 1995. 29. Harden LM, du Plessis I, Poole S, and Laburn HP. Interleukin (IL)-6 and IL-1 beta act synergistically within the brain to induce sickness behavior and fever in rats. Brain Behavior and Immunity 22: 838-849, 2008. 30. Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, Fei H, Kim S, Lallone R, Ranganathan S, and et al. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nature Medicine 1: 1155-1161, 1995. 31. Mizuno TM, Kleopoulos SP, Bergen HT, Roberts JL, Priest CA, and Mobbs CV. Hypothalamic pro-opiomelanocortin mRNA is reduced by fasting and [corrected] in ob/ob and db/db mice, but is stimulated by leptin. Diabetes 47: 294-297, 1998. 32. Myers MG, Jr., Leibel RL, Seeley RJ, and Schwartz MW. Obesity and leptin resistance: distinguishing cause from effect. Trends in Endocrinology and Metabolism: TEM 21: 643-651, 2010. 33. Ouchi N, Parker JL, Lugus JJ, and Walsh K. Adipokines in inflammation and metabolic disease. Nature Reviews Immunology 11: 85-97, 2011. 34. Pal M, Febbraio MA, and Whitham M. From cytokine to myokine: the emerging role of interleukin-6 in metabolic regulation. Immunology and Cell Biology 92: 331-339, 2014. 35. Pedersen BK and Hoffman-Goetz L. Exercise and the immune system: regulation, integration, and adaptation. Physiological Reviews 80: 1055-1081, 2000. 36. Penkowa M, Keller C, Keller P, Jauffred S, and Pedersen BK. Immunohistochemical detection of interleukin-6 in human skeletal muscle fibers following exercise. The FASEB journal : official publication of the Federation of American Societies for Experimental Biology 17: 2166-2168, 2003. 37. Pickup JC, Chusney GD, Thomas SM, and Burt D. Plasma interleukin-6, tumour necrosis factor alpha and blood cytokine production in type 2 diabetes. Life Sciences 67: 291-300, 2000.

63

38. Plata-Salaman CR, Sonti G, Borkoski JP, Wilson CD, and French-Mullen JMb. Anorexia induced by chronic central administration of cytokines at estimated pathophysiological concentrations. Physiology & Behavior 60: 867-875, 1996. 39. Porter MH, Arnold M, and Langhans W. TNF-alpha tolerance blocks LPS-induced hypophagia but LPS tolerance fails to prevent TNF-alpha-induced hypophagia. The American Journal of Physiology 274: R741-745, 1998. 40. Pradhan AD, Manson JE, Rifai N, Buring JE, and Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. Jama-J Am Med Assoc 286: 327-334, 2001. 41. Pradhan AD, Manson JE, Rifai N, Buring JE, and Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA : the Journal of the American Medical Association 286: 327-334, 2001. 42. Ropelle ER, Flores MB, Cintra DE, Rocha GZ, Pauli JR, Morari J, de Souza CT, Moraes JC, Prada PO, Guadagnini D, Marin RM, Oliveira AG, Augusto TM, Carvalho HF, Velloso LA, Saad MJA, and Carvalheira JBC. IL-6 and IL-10 Anti-Inflammatory Activity Links Exercise to Hypothalamic Insulin and Leptin Sensitivity through IKK beta and ER Stress Inhibition. Plos Biol 8, 2010. 43. Sadagurski M, Norquay L, Farhang J, D'Aquino K, Copps K, and White MF. Human IL6 enhances leptin action in mice. Diabetologia 53: 525-535, 2010. 44. Schele E, Benrick A, Grahnemo L, Egecioglu E, Anesten F, Palsdottir V, and Jansson JO. Inter-relation between Interleukin (IL)-1, IL-6 and Body Fat Regulating Circuits of the Hypothalamic Arcuate Nucleus. Journal of Neuroendocrinology 25: 580-589, 2013. 45. Schele E, Fekete C, Egri P, Fuzesi T, Palkovits M, Keller E, Liposits Z, Gereben B, Karlsson-Lindahl L, Shao R, and Jansson JO. Interleukin-6 Receptor alpha is Co-localised with Melanin-Concentrating Hormone in Human and Mouse Hypothalamus. Journal of Neuroendocrinology 24: 930-943, 2012. 46. Schobitz B, Dekloet ER, Sutanto W, and Holsboer F. Cellular-Localization of Interleukin-6 Messenger-Rna and Interleukin-6 Receptor Messenger-Rna in Rat-Brain. European Journal of Neuroscience 5: 1426-1435, 1993. 47. Schobitz B, Voorhuis DA, and De Kloet ER. Localization of interleukin 6 mRNA and interleukin 6 receptor mRNA in rat brain. Neuroscience Letters 136: 189-192, 1992. 48. Shizuya K, Komori T, Fujiwara R, Miyahara S, Ohmori M, and Nomura J. The influence of restraint stress on the expression of mRNAs for IL-6 and the IL-6 receptor in the hypothalamus and midbrain of the rat. Life Sciences 61: PL 135-140, 1997. 49. Sindelar DK, Mystkowski P, Marsh DJ, Palmiter RD, and Schwartz MW. Attenuation of diabetic hyperphagia in neuropeptide Y--deficient mice. Diabetes 51: 778-783, 2002. 50. Steensberg A, van Hall G, Osada T, Sacchetti M, Saltin B, and Klarlund Pedersen B. Production of interleukin-6 in contracting human skeletal muscles can account for the exercise-induced increase in plasma interleukin-6. The Journal of Physiology 529 Pt 1: 237-242, 2000. 51. Stenlof K, Wernstedt I, Fjallman T, Wallenius V, Wallenius K, and Jansson JO. Interleukin-6 levels in the central nervous system are negatively correlated with fat mass in overweight/obese subjects. The Journal of Clinical Endocrinology and Metabolism 88: 4379-4383, 2003. 52. Stovroff MC, Fraker DL, Swedenborg JA, and Norton JA. Cachectin/tumor necrosis factor: a possible mediator of cancer anorexia in the rat. Cancer Res 48: 4567-4572, 1988. 53. Wallenius K, Wallenius V, Sunter D, Dickson SL, and Jansson JO. Intracerebroventricular interleukin-6 treatment decreases body fat in rats. Biochemical and Biophysical Research Communications 293: 560-565, 2002. 54. Wallenius V, Wallenius K, Ahren B, Rudling M, Carlsten H, Dickson SL, Ohlsson C, and Jansson JO. Interleukin-6-deficient mice develop mature-onset obesity. Nature Medicine 8: 75-79, 2002.

64

55. Wesselingh SL, Gough NM, Finlay-Jones JJ, and McDonald PJ. Detection of cytokine mRNA in astrocyte cultures using the polymerase chain reaction. Lymphokine Res 9: 177-185, 1990. 56. Woods SC and Langhans W. Inconsistencies in the assessment of food intake. American journal of physiology Endocrinology and Metabolism 303: E1408-1418, 2012. 57. Yan HQ, Banos MA, Herregodts P, Hooghe R, and Hooghe-Peters EL. Expression of interleukin (IL)-1 beta, IL-6 and their respective receptors in the normal rat brain and after injury. European Journal of Immunology 22: 2963-2971, 1992. 58. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, and Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 372: 425-432, 1994. 59. Ziccardi P, Nappo F, Giugliano G, Esposito K, Marfella R, Cioffi M, D'Andrea F, Molinari AM, and Giugliano D. Reduction of inflammatory cytokine concentrations and improvement of endothelial functions in obese women after weight loss over one year. Circulation 105: 804-809, 2002.

65

Chapter 5.

General Discussion

Pro-inflammatory cytokines such as IL-6 and TNF-α, major mediators of the acute phase response, are

also implicated in the low grade chronic inflammation of the metabolic syndrome (15). Although the

same mediators are involved in these two types of inflammatory processes, the chronic and acute

inflammations affect the organism differently. Inflammation in response to injury or pathogens, the

hallmarks of which are swelling, redness, pain, and fever (20), is a short-term adaptive response

designed to repair damaged tissue (15). Various cell-derived mediators, including IL-6 and TNF-α, are

up-regulated during this acute inflammatory response (10). The acute phase is then followed by the

process of the inflammatory resolution which is essential for the healing.

The chronic, low-grade up-regulation of the same cytokines in obesity and the metabolic

syndrome is implicated in the reduced sensitivity to adiposity signals insulin and leptin and considered

detrimental. In-vivo models of the metabolic syndrome are usually animals with diet-induced obesity.

Studies in these animals show that the inflammation also occurs in the CNS, particularly in the

hypothalamus (9, 24, 51), the major site of action of adiposity signals. This hypothalamic inflammation

occurs in obesity as well as in response to HFD-feeding independent of obesity (13, 25). HFD, in addition

to its caloric value, modulates gene expression in the hypothalamus. The long-chain saturated fatty

acids present in the diet are most harmful; they initiate the inflammatory response by microglia through

the activation of the toll-like receptor 4 (TLR4), resulting in the induction of pro-inflammatory cytokine

synthesis (9, 18, 24). Also, the TLR4 signaling may induce endoplasmic reticulum (ER) stress, which in

turn may enhance inflammation or induce apoptosis (50). The pro-apoptotic response may lead to the

loss of anorexigenic neurons in the Arc (25) and to hypothalamic resistance to anorexigenic adiposity

signals such as leptin and insulin.

Nevertheless, the cytokine action in the brain is divergent and may be pro- as well as anti-

inflammatory. Both TNF-α and IL-6 may exert protective or noxious functions depending on the

activated receptor, the target cell, the duration of action, the concentration, and, in the case of IL-6, the

type of tissue that secrets the cytokine (11, 36, 37, 41). For example, IL-6 triggers neuronal toxicity or

protection depending on whether the targets are glia cells or neurons (12). Regarding energy

66

homeostasis, the overexpression of IL-6 is implicated in an inhibition of the insulin action (21, 31). On

the other hand, overexpressed human IL-6 protects HFD-fed mice from obesity or glucose intolerance

(35). TNF-α secreted by glial cells increases synaptic efficacy by increasing the surface expression of

AMPA receptors (1). The CNS TNF-α also participates in the physiological and pathophysiological

regulation of food intake, sleep, and thermogenesis (22, 26, 33, 42).

The ambiguity of the central cytokine action raises the question of their impact on the function

of adiposity signals, considering the shared sites of action in the brain (2). Studying the interaction

between adiposity signals and pro-inflammatory cytokines is thus very relevant to obesity. In our

experiments we investigated the effect of insulin on food intake and the effect of the pro-inflammatory

cytokines IL-6 and TNF-α on the eating-inhibitory effect of leptin.

Insulin is known to reduce food intake when administered into the brain of an animal (8, 11, 43,

49). In our initial experiments we also saw a food intake reduction in response to the administration of

insulin into the third cerebral ventricle (i3vt) in mice. However, we could not reliably reproduce these

results, mostly because the baseline food intake decreased, i.e., because the food intake in response to

vehicle administration was often lower than usual. There were several possible reasons for this: First,

the i3vt infusions we used to deliver insulin in close proximity to the hypothalamus could have changed

some physiological parameters in the brain. For example, the i3vt infusions may change the volume of

CSF, the temperature, the pH, and introduce bacteria into the brain. This change may be perceived as a

neuronal insult and elicit astrogliosis followed by synthesis of pro-inflammatory mediators (28). The

induced mild inflammatory response may reduce the baseline food intake and thus obscure any small

eating-inhibitory effect of insulin. Secondly, the NPY test we used to assure the patency of the

implanted cannulas was done before the main experiments in order to guarantee an appropriate dose.

NPY is a potent orexigenic neuropeptide that acts on the same brain circuitries as insulin (23) and there

may be unforeseeable effects of the supra-physiological NPY doses on these neurons. Therefore to fully

reproduce the results, the NPY test should be done after the experiments, and in our experience this can

be done as our experiments showed about 90% success with respect to the accuracy of cannula

placement.

Last, but not least, the regimen we used in these insulin experiments was perhaps not optimal

to detect changes in eating behavior. Specifically, we distributed food at the same time on each day of

the experiment including the experimental days. This prior experience could have affected the eating

due to expectation or learning (48). For example, studies have shown that even an orexigenic peptide

67

such as NPY may or may not have an effect in the same animals due to prior experience with the

experimental paradigm (4, 6). The same may apply to the eating-inhibitory effect of insulin. Insulin and

leptin’s effects on food intake depend on melanocortins (3). The eating-inhibitory effect of

melanocortins may also be altered by prior experience with the experimental paradigm (4).

Interestingly, we also did not observe an eating-inhibitory effect of leptin in the last series of our

experiments (IL-6 / leptin experiments) in which we used the same paradigm as in the experiments with

insulin. In our second study (TNF-α / leptin experiments) we “disrupted” the routine regimen by shifting

the food restriction period one hour into the dark phase without increasing its duration. In this

experiment we observed the expected, reliable food intake reduction in response to leptin in addition to

the slightly increased food intake in response to vehicle administration compared with the insulin

experiments.

Considering the ability of leptin and TNF-α to reduce food intake when administered centrally

(29) and the notion that increased central TNF-α levels in obesity act on the same brain areas as leptin

(5, 33), the possible interaction between TNF-α and leptin may be relevant to obesity. Another aspect

that supports an interaction between leptin and TNF-α is the similarity of their receptors and signaling

pathways. Thus, in the second series of experiments we investigated the possible impact of TNF-α

action on the eating-inhibitory effect of leptin in the brain.

The long form of the leptin receptor that is expressed predominantly in the hypothalamus and

that mediates the weight regulatory effects of leptin is a member of the class I cytokine receptor

superfamily (17, 40). Receptors of this class are activated by ligand-induced receptor homo- or hetero-

dimerization and often require activation of the Janus kinases including JAK2. Among the downstream

targets of the JAK proteins are members of the STAT family such as STAT3 that is implicated in the

control of food intake (16, 19). Unlike the IL-6 receptor, which belongs to the same family, activation of

LepRb is independent of the signal transducing subunit of IL-6 type cytokine receptor accessory chain

gp130 (46). Nevertheless, the JAK/STAT pathway downstream of the leptin as well as IL-6 receptors

leads to the induction of the SOCS3, a negative regulator of STAT3 and leptin receptor signaling (7, 47).

TNF-α, by acting through the TNFR1, also transmits signals from the cell surface to the nucleus

through the JAK/STAT signaling pathway with activation of STAT3 (32). Moreover, a recent study

showed a 6-7 fold stimulation of STAT3 phosphorylation in the rat hypothalamus in response to both

leptin and TNF-α (32). The immunohistochemical analysis in our study showed, however, an additive

expression of pSTAT3 in the Arc in response to co-administration of TNF-α and leptin. The VMH, the

68

area that lies presumably downstream of the Arc with respect to the activation of pSTAT3, had almost

no expression of pSTAT3 in response to the TNF-α administration and, compared to the Arc, a reduced

overall pSTAT3 expression in response to leptin alone and in combination with TNF-α. The 15 min

interval between the administration of the compounds and the fixation of the brain tissue was perhaps

too short to allow for the full extent of pSTAT3 expression in the VMH. On the other hand, a prolonged

time period could lead to the overexpression of SOCS3, the negative feedback signal that may inhibit the

STAT3 phosphorylation and lead to a blunted expression of pSTAT in the Arc at the time point of the

tissue fixation, i.e. we would observe the post-incremental expression. A time-course study should be

performed and the time intervals between the compound administration and the tissue fixation should

be optimized to achieve more precise representation of pSTAT activation in different hypothalamic

areas.

The inhibitory effect of centrally administered leptin and TNF-α on food intake is described in

several studies (29, 30, 38, 39). Our results confirm and extend these findings. In our study TNF-α and

leptin reduced food intake when administered centrally in mice fed a low fat diet, and the co-

administration of leptin and TNF-α produced a greater food intake reduction when compared with the

effects of each compound administered alone. The food intake reduction by leptin and TNF-α was

reflected in the expression of pSTAT3 in Arc, supporting the catabolic role of pSTAT3 in energy

homeostasis regulation.

The third series of experiments addressed the possible interaction of IL-6 and leptin with respect

to their effects on food intake. The ability of IL-6 to enhance leptin’s action has been described in

studies with exercised animals, i.e., the related to muscle-derived IL-6 in response to exercise (27, 34).

Studies with IL-6 knockout mice also indicated an effect of IL-6 on leptin’s actions (45). Nevertheless, we

were the first to show that centrally co-administrated IL-6 and leptin has greater eating-inhibitory effect

than the sum of the effects of each peptide alone, indicating that there is a central interaction between

leptin and IL-6. The mechanisms of this interaction need further investigation. One study showed,

however, that leptin deficient ob/ob mice with increased circulating IL-6 levels reduced body weight and

food intake in response to a low dose of leptin, whereas normal mice did not (35). This indicates that IL-

6 does not exert a catabolic function by itself, but interacts with the leptin signaling and increases the

insulin sensitivity in the brain.

The central effect of IL-6 on food intake has been investigated in several studies (14, 30, 45).

Only in two of these studies, however, IL-6 exerted an effect on eating. In both these studies, central IL-

69

6 reduced food intake in obese or HFD-fed rats; in one IL-6 reduced 12 h food intake, in the other it

reduced average daily food intake measured throughout the study (34, 44). In contrast, in our study IL-6

increased food intake later in the light cycle (22hr) without affecting food intake at earlier time points of

measurement. The mechanisms that trigger food intake increase in the late phase of the circadian cycle

deserve further investigation.

Overall, our study showed that the central actions of cytokines may be different from the

peripheral actions. More specifically, cytokines may enhance the eating-inhibitory effect of adiposity

signals in the brain. Also, the origin of the cytokine may determine its action. Clearly, more studies are

needed to understand the mechanism of action of IL-6 and TNF-α with regards to the central regulation

of energy homeostasis. Knowing the inconsistent effect of insulin on eating, a study with an

experimental design that allows for avoiding the learning effect in animals can be used in investigations

of the possibility of interaction between insulin and TNF-α. The role of TNF-α in metabolism may be

investigated with the help of a more targeted approach. Considering that astrocytes play an important

role in the metabolism of neurons, a study using animal models with specific knockdown of TNF-α

receptor in astrocytes could elucidate its effects. Also, further experiment are needed to understand

the mechanisms behind the ability of IL-6 to increase food intake later in the cycle and, at the same

time, to enhance leptin’s eating-inhibitory effect in the brain.

70

5.1. References

1. Beattie EC, Stellwagen D, Morishita W, Bresnahan JC, Ha BK, Von Zastrow M, Beattie MS, and Malenka RC. Control of synaptic strength by glial TNFalpha. Science 295: 2282-2285, 2002. 2. Begg DP and Woods SC. The endocrinology of food intake. Nat Rev Endocrinol 9: 584-597, 2013. 3. Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, and Woods SC. The catabolic action of insulin in the brain is mediated by melanocortins. Journal of Neuroscience 22: 9048-9052, 2002. 4. Benoit SC, Clegg DJ, Barrera JG, Seeley RJ, and Woods SC. Learned meal initiation attenuates the anorexic effects of the melanocortin agonist MTII. Diabetes 52: 2684-2688, 2003. 5. Benoit SC, Clegg DJ, Seeley RJ, and Woods SC. Insulin and leptin as adiposity signals. Recent Progress in Hormone Research 59: 267-285, 2004. 6. Benoit SC, Clegg DJ, Woods SC, and Seeley RJ. The role of previous exposure in the appetitive and consummatory effects of orexigenic neuropeptides. Peptides 26: 751-757, 2005. 7. Borland G, Bird RJ, Palmer TM, and Yarwood SJ. Activation of protein kinase Calpha by EPAC1 is required for the ERK- and CCAAT/enhancer-binding protein beta-dependent induction of the SOCS-3 gene by cyclic AMP in COS1 cells. The Journal of Biological Chemistry 284: 17391-17403, 2009. 8. Brown LM, Clegg DJ, Benoit SC, and Woods SC. Intraventricular insulin and leptin reduce food intake and body weight in C57BL/6J mice. Physiology & Behavior 89: 687-691, 2006. 9. De Souza CT, Araujo EP, Bordin S, Ashimine R, Zollner RL, Boschero AC, Saad MJ, and Velloso LA. Consumption of a fat-rich diet activates a proinflammatory response and induces insulin resistance in the hypothalamus. Endocrinology 146: 4192-4199, 2005. 10. Eming SA, Krieg T, and Davidson JM. Inflammation in wound repair: Molecular and cellular mechanisms. J Invest Dermatol 127: 514-525, 2007. 11. Flores MB, Fernandes MF, Ropelle ER, Faria MC, Ueno M, Velloso LA, Saad MJ, and Carvalheira JB. Exercise improves insulin and leptin sensitivity in hypothalamus of Wistar rats. Diabetes 55: 2554-2561, 2006. 12. Gadient RA and Otten U. Expression of interleukin-6 (IL-6) and interleukin-6 receptor (IL-6R) mRNAs in rat brain during postnatal development. Brain Research 637: 10-14, 1994. 13. Guyenet SJ, Nguyen HT, Hwang BH, Schwartz MW, Baskin DG, and Thaler JP. High-fat diet feeding causes rapid, non-apoptotic cleavage of caspase-3 in astrocytes. Brain Research 1512: 97-105, 2013. 14. Harden LM, du Plessis I, Poole S, and Laburn HP. Interleukin (IL)-6 and IL-1 beta act synergistically within the brain to induce sickness behavior and fever in rats. Brain Behavior and Immunity 22: 838-849, 2008. 15. Hotamisligil GS. Inflammation and metabolic disorders. Nature 444: 860-867, 2006. 16. Ihle JN, Witthuhn BA, Quelle FW, Yamamoto K, Thierfelder WE, Kreider B, and Silvennoinen O. Signaling by the cytokine receptor superfamily: JAKs and STATs. Trends in Biochemical Sciences 19: 222-227, 1994. 17. Kishimoto T, Taga T, and Akira S. Cytokine Signal-Transduction. Cell 76: 253-262, 1994. 18. Kleinridders A, Schenten D, Konner AC, Belgardt BF, Mauer J, Okamura T, Wunderlich FT, Medzhitov R, and Bruning JC. MyD88 signaling in the CNS is required for development of fatty acid-induced leptin resistance and diet-induced obesity. Cell Metabolism 10: 249-259, 2009. 19. Ladyman SR and Grattan DR. JAK-STAT and feeding. Jak-Stat 2: e23675, 2013. 20. Larsen GL and Henson PM. Mediators of inflammation. Annual Review of Immunology 1: 335-359, 1983. 21. Lazar MA. How obesity causes diabetes: Not a tall tale. Science 307: 373-375, 2005.

71

22. Leon LR. Invited review: cytokine regulation of fever: studies using gene knockout mice. J Appl Physiol (1985) 92: 2648-2655, 2002. 23. Mercer RE, Chee MJ, and Colmers WF. The role of NPY in hypothalamic mediated food intake. Frontiers in Neuroendocrinology 32: 398-415, 2011. 24. Milanski M, Degasperi G, Coope A, Morari J, Denis R, Cintra DE, Tsukumo DM, Anhe G, Amaral ME, Takahashi HK, Curi R, Oliveira HC, Carvalheira JB, Bordin S, Saad MJ, and Velloso LA. Saturated fatty acids produce an inflammatory response predominantly through the activation of TLR4 signaling in hypothalamus: implications for the pathogenesis of obesity. J Neurosci 29: 359-370, 2009. 25. Moraes JC, Coope A, Morari J, Cintra DE, Roman EA, Pauli JR, Romanatto T, Carvalheira JB, Oliveira AL, Saad MJ, and Velloso LA. High-fat diet induces apoptosis of hypothalamic neurons. PloS one 4: e5045, 2009. 26. Opp MR. Cytokines and sleep. Sleep Medicine Reviews 9: 355-364, 2005. 27. Pal M, Febbraio MA, and Whitham M. From cytokine to myokine: the emerging role of interleukin-6 in metabolic regulation. Immunology and Cell Biology 92: 331-339, 2014. 28. Pekny M, Wilhelmsson U, and Pekna M. The dual role of astrocyte activation and reactive gliosis. Neuroscience Letters 565: 30-38, 2014. 29. Plata-Salaman CR. Food intake suppression by growth factors and platelet peptides by direct action in the central nervous system. Neuroscience Letters 94: 161-166, 1988. 30. Plata-Salaman CR, Sonti G, Borkoski JP, Wilson CD, and French-Mullen JMb. Anorexia induced by chronic central administration of cytokines at estimated pathophysiological concentrations. Physiology & Behavior 60: 867-875, 1996. 31. Pradhan AD, Manson JE, Rifai N, Buring JE, and Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. Jama-J Am Med Assoc 286: 327-334, 2001. 32. Rizk NM, Stammsen D, Preibisch G, and Eckel J. Leptin and tumor necrosis factor-alpha induce the tyrosine phosphorylation of signal transducer and activator of transcription proteins in the hypothalamus of normal rats in vivo. Endocrinology 142: 3027-3032, 2001. 33. Romanatto T, Cesquini M, Amaral ME, Roman EA, Moraes JC, Torsoni MA, Cruz-Neto AP, and Velloso LA. TNF-alpha acts in the hypothalamus inhibiting food intake and increasing the respiratory quotient--effects on leptin and insulin signaling pathways. Peptides 28: 1050-1058, 2007. 34. Ropelle ER, Flores MB, Cintra DE, Rocha GZ, Pauli JR, Morari J, de Souza CT, Moraes JC, Prada PO, Guadagnini D, Marin RM, Oliveira AG, Augusto TM, Carvalho HF, Velloso LA, Saad MJA, and Carvalheira JBC. IL-6 and IL-10 Anti-Inflammatory Activity Links Exercise to Hypothalamic Insulin and Leptin Sensitivity through IKK beta and ER Stress Inhibition. Plos Biol 8, 2010. 35. Sadagurski M, Norquay L, Farhang J, D'Aquino K, Copps K, and White MF. Human IL6 enhances leptin action in mice. Diabetologia 53: 525-535, 2010. 36. Santello M and Volterra A. TNFalpha in synaptic function: switching gears. Trends in Neurosciences 35: 638-647, 2012. 37. Scheller J, Chalaris A, Schmidt-Arras D, and Rose-John S. The pro- and anti-inflammatory properties of the cytokine interleukin-6. Biochimica et Biophysica Acta 1813: 878-888, 2011. 38. Seeley RJ, van Dijk G, Campfield LA, Smith FJ, Burn P, Nelligan JA, Bell SM, Baskin DG, Woods SC, and Schwartz MW. Intraventricular leptin reduces food intake and body weight of lean rats but not obese Zucker rats. Hormone and metabolic research = Hormon- und Stoffwechselforschung = Hormones et Metabolisme 28: 664-668, 1996. 39. Seeley RJ, Yagaloff KA, Fisher SL, Burn P, Thiele TE, van Dijk G, Baskin DG, and Schwartz MW. Melanocortin receptors in leptin effects. Nature 390: 349, 1997. 40. Tartaglia LA, Dembski M, Weng X, Deng NH, Culpepper J, Devos R, Richards GJ, Campfield LA, Clark FT, Deeds J, Muir C, Sanker S, Moriarty A, Moore KJ, Smutko JS, Mays GG, Woolf EA, Monroe CA,

72

and Tepper RI. Identification and expression cloning of a leptin receptor, OB-R. Cell 83: 1263-1271, 1995. 41. Thaler JP, Choi SJ, Schwartz MW, and Wisse BE. Hypothalamic inflammation and energy homeostasis: resolving the paradox. Frontiers in Neuroendocrinology 31: 79-84, 2010. 42. Vaisman N and Hahn T. Tumor necrosis factor-alpha and anorexia--cause or effect? Metabolism: Clinical and Experimental 40: 720-723, 1991. 43. Vogt MC and Bruning JC. CNS insulin signaling in the control of energy homeostasis and glucose metabolism - from embryo to old age. Trends in Endocrinology and Metabolism: TEM 24: 76-84, 2013. 44. Wallenius K, Wallenius V, Sunter D, Dickson SL, and Jansson JO. Intracerebroventricular interleukin-6 treatment decreases body fat in rats. Biochemical and Biophysical Research Communications 293: 560-565, 2002. 45. Wallenius V, Wallenius K, Ahren B, Rudling M, Carlsten H, Dickson SL, Ohlsson C, and Jansson JO. Interleukin-6-deficient mice develop mature-onset obesity. Nature Medicine 8: 75-79, 2002. 46. White DW, Kuropatwinski KK, Devos R, Baumann H, and Tartaglia LA. Leptin receptor (OB-R) signaling. Cytoplasmic domain mutational analysis and evidence for receptor homo-oligomerization. The Journal of Biological Chemistry 272: 4065-4071, 1997. 47. Williams JJ and Palmer TM. Unbiased identification of substrates for the Epac1-inducible E3 ubiquitin ligase component SOCS-3. Biochemical Society Transactions 40: 215-218, 2012. 48. Woods SC and Langhans W. Inconsistencies in the assessment of food intake. American Journal of Physiology Endocrinology and Metabolism 303: E1408-1418, 2012. 49. Woods SC, Lotter EC, McKay LD, and Porte D, Jr. Chronic intracerebroventricular infusion of insulin reduces food intake and body weight of baboons. Nature 282: 503-505, 1979. 50. Zhang K and Kaufman RJ. From endoplasmic-reticulum stress to the inflammatory response. Nature 454: 455-462, 2008. 51. Zhang X, Zhang G, Zhang H, Karin M, Bai H, and Cai D. Hypothalamic IKKbeta/NF-kappaB and ER stress link overnutrition to energy imbalance and obesity. Cell 135: 61-73, 2008.

73

List of Abbreviations

aCSF artificial cerebrospinal fluid AgRP agouti-related protein AMPA α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid Arc arcuate nucleus BBB blood-brain barrier BMI body mass index BSA bovine serum albumin CART cocaine and amphetamine-related transcript CCK cholecystokinin CNS central nervous system DMH dorsomedial hypothalamic nucleus GABAa γ-aminobutyric acid GLP-1 glucagon-like peptide-1 h hour HF high fat HFD high fat diet hIL-6 human interleukin-6 i3vt intra-third-cerebroventricular ICV intra-cerebroventricular IL-1 interleukin-1 IL-6 interleukin-6 IL-6-/- interleukin-6 deficient IL6Rα interleukin-6 receptor alpha ip intraperitoneal IU international unit JAK janus kinase signal transducer and activator of transcription STAT signal transducer and activator of transcription LepRb leptin receptor beta LPS lipopolysaccharide MC3-R, MC4-R melanocortin receptor-3, -4 mRNA messenger ribonucleic acid MT-II melanotan-II NA numerical aperture NPY neuropeptide Y ob/ob leptin deficient mouse POMC pro-opiomelanocortin PYY peptide YY SI system of units STAT3 signal transduction and activator of transcription-3 SOCS3 suppressor of cytokine signaling-3 T2D type-2-diabetes TNFKO tumor necrosis factor-alpha knock-out TNFR2 tumor necrosis factor-alpha type two receptor TNF-α tumor necrosis factor-alpha

74

U unit VMH ventromedial WT wild type α-MSH melanocyte-stimulating hormone